TRF1 and TRF2 use different mechanisms to find telomeric ... Lin and countryman.pdf · the TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects

Post on 27-Nov-2018

220 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

Transcript

TRF1 and TRF2 use different mechanisms to findtelomeric DNA but share a novel mechanism tosearch for protein partners at telomeresJiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3

Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6

Patricia L Opresko2 Neil M Kad7 and Hong Wang1

1Physics Department North Carolina State University Raleigh NC 27695 USA 2Department of Environmentaland Occupational Health University of Pittsburgh Graduate School of Public Health Pittsburgh PA 15219USA 3Electric and Computer Engineering Department University of North Carolina at Charlotte Charlotte NC28223 USA 4Department of Industrial and System Engineering North Carolina State University Raleigh NC27695 USA 5Department of Cell Biology Center for Biologic Imaging University of Pittsburgh GraduateSchool of Public Health Pittsburgh PA 15219 USA 6Division of Biophysics Universitat OsnabruckBarbarstrasse 11 49076 Osnabruck Germany and 7School of Biological Sciences University of EssexColchester Essex CO4 3SQ UK

Received July 19 2013 Revised October 11 2013 Accepted October 23 2013

ABSTRACT

Human telomeres are maintained by the shelterinprotein complex in which TRF1 and TRF2 binddirectly to duplex telomeric DNA How theseproteins find telomeric sequences among agenome of billions of base pairs and how they findprotein partners to form the shelterin complexremains uncertain Using single-molecule fluores-cence imaging of quantum dot-labeled TRF1 andTRF2 we study how these proteins locate TTAGGG repeats on DNA tightropes By virtue of its basicdomain TRF2 performs an extensive 1D search onnontelomeric DNA whereas TRF1rsquos 1D search islimited Unlike the stable and static associationsobserved for other proteins at specific bindingsites TRF proteins possess reduced binding stabil-ity marked by transient binding (9ndash17 s) and slow1D diffusion on specific telomeric regions Theseslow diffusion constants yield activation energybarriers to sliding 28ndash36 iBT greater than thosefor nontelomeric DNA We propose that the TRFproteins use 1D sliding to find protein partners andassemble the shelterin complex which in turnstabilizes the interaction with specific telomericDNA This lsquotag-team proofreadingrsquo represents amore general mechanism to ensure a specific setof proteins interact with each other on long

repetitive specific DNA sequences without requiringexternal energy sources

INTRODUCTION

Telomeres play a crucial role in maintaining the stabilityof linear chromosomes (12) Loss of telomere functioncan activate DNA repair processes leading to nucleolyticdegradation of natural chromosome ends and their end-to-end fusion (3) Telomere dysfunction and associatedchromosomal abnormalities have been strongly associatedwith age-related degenerative diseases and cancer (45)In a typical human somatic cell the telomeric repeatsequence TTAGGG is 2ndash15 kb in length with a 30-over-hang of 100ndash200 nt (6) This 30-overhang serves as asubstrate for the reverse transcriptase telomerase whichreplicates the telomeric sequence by using an internalRNA subunit as a template to direct the DNA synthesis(17ndash9) A specialized protein complex shelterin (ortelosome) binds to and protects the chromosome ends(210) The shelterin complex in humans consists ofsix core proteins TRF1 TRF2 POT1 TIN2 TPP1 andRAP1 (111)TRF1 and TRF2 are the only proteins in the shelterin

complex that make high-affinity contact with double-stranded telomeric DNA (1213) TRF1 negatively regu-lates telomere length and promotes telomere replication(14) Whereas TRF2 caps and protects chromosomeends (11) in addition to regulating telomere length (15)

To whom correspondence should be addressed Tel +1 919 5137203 Fax +1 919 5156538 Email hong_wangncsueduCorrespondence may also be addressed to Neil M Kad Tel +44 1206 874403 Fax +44 1206 872592 Email nkadessexacuk

Nucleic Acids Research 2013 1ndash12doi101093nargkt1132

The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby30) whichpermits unrestricted reuse distribution and reproduction in any medium provided the original work is properly cited

Nucleic Acids Research Advance Access published November 22 2013 at D

H H

ill Library - A

cquis S on Novem

ber 27 2013httpnaroxfordjournalsorg

Dow

nloaded from

Removal of TRF2 from the telomeres results in loss of the30-overhang covalent fusion of telomeres and inductionof ATM and p53 dependent apoptosis (1617) BothTRF1 and TRF2 contain a TRFH domain thatmediates homodimerization and a Myb type domainthat sequence-specifically binds to telomeric DNA(Figure 1A) (12) However these two proteins differ attheir N-termini where TRF1 and TRF2 are rich inacidic and basic residues respectively Previous electronmicroscopy (EM) and atomic force microscopy (AFM)studies established that both TRF1 and TRF2 playimportant architectural roles at telomeres (18ndash21) TRF1forms protein filaments on longer telomeric repeats (27repeats) and promotes parallel pairing of telomeric tracts(19) In vitro TRF2 can remodel linear telomeric DNAinto T-loops (20)A previous cell-based study of TRF1 and TRF2 using

fluorescence recovery after photobleaching and fluores-cence loss in photobleaching suggested that TRF1 andTRF2 interact with telomeres in a dynamic fashion (22)Although TRF1 and TRF2 are proposed to have extra-telomeric functions they preferentially localize to the TTAGGG repeat sequences whether these target sites are atinterstitial regions or at chromosome ends (23ndash25) Oncetelomeric sequences are located TRF1 and TRF2 mustfind protein partners to form the shelterin complex andto regulate the functions of other DNA-binding proteinsat telomeres (26ndash28) Despite recent advancements in theunderstanding of functions of TRF1 and TRF2 it is stillunclear how TRF1 and TRF2 are able to find telomericsequences and protein partners in a genome of billions ofbase pairsAccumulating evidence suggests that a protein can use

one-dimensional (1D) sliding (correlated translocation

while maintaining continuous DNA contact) jumping(noncorrelated detachment and reattachment) orhopping (correlated detachment and reattachment) tonavigate through the vast excess of nonspecific DNA se-quences in vivo (29ndash32) Investigations of DNA-bindingdynamics on nonspecific DNA at the single-moleculelevel have significantly advanced our understanding ofhow proteins with diverse functions conduct their targetDNA search (313334) However the paradoxicalrequirements of rapid search at nonspecific sites andstability at target sites have been primarily investigatedin theoretical studies (35ndash38) direct comparisons of theprotein-binding energy landscape at nonspecific sites andtarget sites from single-molecule experimental data are stilllacking

Here we used single-molecule fluorescence imaging tostudy the dynamics of quantum dot (QD)-labeled TRF1and TRF2 proteins on DNA and DNA substrates con-taining alternating regions of telomeric and nontelomericsequences TRF1 appears to bind directly to telomeric se-quences with very little 1D searching through nontelo-meric DNA whereas TRF2 possesses a significantcomponent of 1D search Using a truncation mutant welocalized this 1D searching activity to the basic domain ofTRF2 On telomeric DNA both TRF1 and TRF2 diffuseslowly due to higher energy barriers to diffusion and theypossess longer attached lifetimes at telomeric repeatscompared with nontelomeric DNA sequences Theseobservations indicate that there is preferential binding totelomeric DNA but the affinity is not high enough toprevent TRF proteins from diffusing along TTAGGGrepeats We postulate that this allows TRF1 and TRF2to find their protein partners locally and that this is amore general mechanism for coupling the energy from

Figure 1 TRF1- and TRF2-QDs retain DNA-binding activity (A) Schematic representations of the domain structures of TRF1 and TRF2A Acidic domain B Basic domain M Myb type domain (B) Schematic representations of TRF1- and TRF2-QD conjugates (left) BTtris-NTAcompound (middle) and the DNA substrate (T270) with two tandem (TTAGGG)135 repeats connected by a short linker region (right 54 kb inlength) (CndashE) Representative AFM images of DNA in the presence of (C) only QDs and BTtris-NTA compound (D) TRF1-QDs or (E) TRF2-QDsThe scale bar is 200 nm White arrows point to QDs bound to DNA The numbers in (CndashE) indicate the percent of DNA molecules bound with QDsin each condition The total numbers of complexes analyzed were 200 250 and 250 for no protein TRF1-QDs and TRF2-QDs respectively

2 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

multiple weak DNA-binding components to ensure highbinding specificity on long repetitive sequences

MATERIALS AND METHODS

Protein purification

Recombinant N-terminal His6-tagged TRF1 and TRF2were purified using a baculovirusinsect cell expressionsystem and an AKTA Explorer FPLC (GE Healthcare)as described previously (39) TRF2B was purified usinga bacterial expression system (40) Protein concentrationswere determined using the Bradford assay Proteins usedin this study are gt90 pure based on SDSndashPAGE andCoomassie staining Proteins are active in binding to thetelomeric DNA substrate containing three TTAGGGrepeats based on electrophoresis mobility shift assays(EMSAs)

DNA substrates

DNA was purchased from New England BioLabs OtherDNA substrates used in this study are shown in Figure 1Band Supplementary Figure S1 pSXneo(T2AG3) plasmidDNA containing 270 TTAGGG repeats was a gift fromDr Peter Lansdorp (University of British Columbia) (41)pGTK4 plasmid-derived Tel10 plasmid is 5994-bp longand contains 10 TTAGGG repeats and was prepared asdescribed previously (42) To generate DNA fragmentscontaining TTAGGG repeats for AFM imaging digestionof T270 DNA (10 mg) was carried out at 37C for 4 h usingHpaI (130 U) in Buffer 4 (New England BioLabs) ForTel10 plasmid digestions were carried out using XbaI (100U) in Buffer 4 For fluorescence imaging linearizedplasmids were ligated to generate longer DNA substratesusing a Quick LigationTM Kit (New England BioLabs)The ligation reactions were done at room temperaturefor 15min The nontelomeric DNA substrate withoutthe (TTAGGG)270 sequence was gel purified after the di-gestion of pSXneo(T2AG3) with BglII and XbaI FinalDNA substrate purification was done using an illustraGFXTM PCR DNA and Gel Band Purification Kit (GEHealthcare)

ProteinndashQD conjugation

Streptavidin-conjugated QDs (Sav-QDs) were purchasedfrom Invitrogen Biotinylated multivalent chelatortris-nitrilotriacetic acid (BTtris-NTA) was preparedaccording to the previous reports (4344) The TRFndashDNA reaction buffer contains 50mM HEPES (pH 75)and varying concentrations of NaCl (25 50 75 and100mM) The total ionic strengths are 75 125 175 and225mM respectively (45)

For single color QD labeling of His6-tagged TRF1 orTRF2 1 ml of red QD (655 nm 1 mM Invitrogen hydro-dynamic radius 115 nm) was incubated with 1 ml ofBTtris-NTA (2mM) for 20min (46) An amount of 1 ml ofproteins (2mM) were then added to the QD-NTA solutionand incubated for additional 20min For dual-color QDlabeling 1 ml of red (1mM) and green QDs (565 nm 1 mMhydrodynamic radius 95 nm) were incubated with 1 ml of

BTtris-NTA (2mM) (46) TRF1 or TRF2 (1ml 2 mM) wasadded to the solution and incubated for additional 20minFor fluorescence imaging unless otherwise specifiedproteinndashNTA-QD solutions were diluted 200-fold beforebeing drawn into the flow cell using a syringe pump(model SP260p World Precision Instruments) at 300 mlml flow rate The final protein concentration was 33 nMfor both TRF1 and TRF2 Protein concentrations andionic strengths of the buffer used in this study are com-parable to physiological conditions (Supplementary Text)For AFM imaging of TRF2-QDs in the presence ofmonoclonal TRF2 antibody (Imagenex Corporation)the AbTRF2NTAQD ratio was 1121 or 5121 andreactions were carried out at room temperature for 30minafter the addition of antibodies

AFM imaging and image analysis

All DNA and protein samples were diluted 10-fold in 1AFM buffer [25mM NaOAc 25mM HEPESndashKOH (pH75) and 10mM Mg(OAc)2] before deposition onto afreshly cleaved mica (SPI Supply) The samples werethen washed with MilliQ water and dried under a streamof nitrogen gas All images were collected in tappingmode using a MFP-3D-Bio AFM (Asylum Research)Pointprobe PPP-FMR probes (Nanosensors) withspring constants at 28Nm (nominal value) were usedAll images were captured at a scan size of 1 mm 1 mm ascan rate of 1ndash2Hz and a resolution of 512 512 pixelsThe position of TRF proteins on DNA was analyzedusing the software from Asylum Research

Fluorescence imaging and analysis of fluorescencemicroscopy data

Fluorescence imaging was carried out with an invertedmicroscope (Nikon Ti-E) equipped with an encodedmotorized stage perfect focus system (PFS) and a Ti-TIRF E motorized illuminator unit Fluorescenceimaging was performed by excitation at 488 nm using asolid-state laser (20mW Sapphire DPSS) a 100 objectivewith a numerical aperture of 149 (APO TIRF Nikon)and 15 additional magnification The laser power wascontrolled by using neutral density filters The excitationbeam was reflected into the objective through a TIRFfilter set containing zt488rdc and ET500LP filters Forsimultaneous imaging of green (565 nm) and red(655 nm) QDs a dual view simultaneous imaging system(DV2 Photometrics) was used in combination with aT605LPXR dichroic beamsplitter (Chroma) and a band-pass filter ET65540m (Chroma) The images werecaptured using an electron multiplied (EM) CCDcamera (iXon DU897 Andor Technology) operated at60C with an EM gain of 250 and a frame rate of20Hz Construction of the flow cell was carried out ac-cording to a procedure described previously (334748)Silica beads (5 mm Polysciences) were first treated withpoly-L-lysine hydrobromide (2500 mgml MWgt 300KDa Wako Chemicals) DNA or ligated DNA sub-strate (5 mgml) were stretched unless otherwise specifiedunder hydrodynamic flow at 300 mlmin flow rate using asyringe pump Extended DNA strands anchored between

Nucleic Acids Research 2013 3

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

two poly-L-lysine-coated beads formed DNA tightropesAfter introducing the proteinndashQDs into the flow cell alldata collection was performed in the absence of anyfurther buffer flow The presence of YOYO-1 on DNAsignificantly reduced the diffusion constant a-factor andthe percentage of motile proteinndashQD complexes on DNAat certain salt conditions Consequently all data analysiswas done using movies collected from using unstainedDNA tightropes (Supplementary Text)

Statistical analysis

Single-factor ANOVA and Student-t tests were used forstatistical analysis

RESULTS

TRF1- and TRF2-QD conjugates are functional inDNA binding

Fluorescent labeling of TRF1 and TRF2 was achieved byconjugating 6 histidine (His6) tagged TRF1 and TRF2to streptavidin-conjugated QDs using the biotinylatedmultivalent chelator tris-nitrilotriacetic acid (BTtris-NTA)(44) (Figure 1B see lsquoMaterials and Methodsrsquo section) Themultiple Ni-NTAs on the circular scaffold of the tris-NTAadaptor bind the His-tag with subnanomolar affinity re-sulting in a bound lifetime in the range of hours (4344)Importantly we applied a previously established methodbased on AFM imaging to characterize the stoichiometryof QDndashTRF complexes (4950) AFM imaging revealedthat using TRF2 antibody marking the presence ofTRF2 (TRF2Ab=11 or 15) among the QDs displayedTRF2-Ab complexes (24) 90 (n=39) possessed onlyone TRF2ndashAb complex (Supplementary Figure S2)

QDs alone exhibited minimal nonspecific binding toDNA as confirmed by AFM (Figure 1C) As expectedaddition of QD-labeled TRF1 or TRF2 to DNA contain-ing two stretches of (TTAGGG)135 connected by a shortlinker region (T270 DNA Figure 1B see lsquoMaterialsand Methodsrsquo section) resulted in substantial binding(Figure 1D and E) Furthermore AFM image analysisrevealed that both TRF1- and TRF2-QDs bound prefer-entially to the telomeric DNA sequences on both the T270and Tel10 DNA substrates (Supplementary Figure S3)

TRF1 and TRF2 diffuse one-dimensionally onnontelomeric DNA

To study the dynamics of individual TRF1 and TRF2 mol-ecules on DNA using oblique-angle fluorescence micros-copy we applied a DNA tightrope assay (Figure 2A)(33) DNA strands are suspended between poly-L-lysinecoated microspheres at an elongation of 90 DNAcontour length using hydrodynamic flow (47) Thisprocess isolates DNA from the surface and does notrequire continuous buffer flow for the observation ofproteinndashDNA interactions QDs did not bind to DNAtightropes alone or in the presence of TRF proteinswithout BTtris-NTA However with both BTtris-NTAand His6-tagged TRF1 or TRF2 QDs were observed onDNA throughout the visual field (Figure 2B and C) BothTRF1- (Supplementary Movie S1) and TRF2-QDs(Supplementary Movie S2) showed clear 1D diffusion onDNA which was tracked by Gaussian fitting to kymo-graphs (particle position versus time plots SupplementaryData) (3347)

To determine whether TRF1 and TRF2 slide or hop weevaluated the effect of ionic conditions on the dynamic

Figure 2 DNA tightrope assay based oblique-angle fluorescence imaging of TRF1- and TRF2-QDs on DNA tightropes (A) A schematic drawingof the DNA tightropes (green lines) bound with QD (red ball)-labeled proteins (green balls) between silica beads (large white balls) The drawing isnot to scale (B and C) Representative fluorescence images of red (655 nm) QD-conjugated His6-TRF1 (B) and His6-TRF2 (C) on DNA (stainedwith YoYo1) (D) Classification of different types of proteinndashDNA interactions observed with TRF2-QDs on DNA for attached lifetimemeasurement

4 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

interactions between the QD-labeled TRF proteins andDNA Increasing the salt concentration should not affectthe diffusion constants of a sliding process but shouldelevate the diffusion constants of hopping (295152) Weperformed experiments at 75 125 175 and 225mM ionicstrengths (see lsquoMaterials and Methodsrsquo section) Thefraction of motile TRF1 proteins ranged from 15 to33 (Supplementary Figure S4A) and followed a trendof decreasing diffusion constants as the ionic strengthincreased (75ndash38 102mm2s) such that the differencebetween the highest and lowest salt was statistically signifi-cant (P=0017 Supplementary Figure S4B and Table 1)In contrast TRF2 was highly motile on DNA acrossall ionic strengths and showed no significant change indiffusion constant (84ndash95 102 mm2s SupplementaryFigure S4B and Table 1) TRF2 diffused substantiallyfaster than TRF1 at all ionic strengths showing statisticalsignificance at ionic strengths between 125 and 225mM

In addition to the diffusion constant we also measuredthe diffusive exponent (a-factor Supplementary Data)An a factor of 1 indicates an unbiased random walk gt1indicates directed motion and lt1 indicates periodsof pausing in the random walk (subdiffusion) (53)TRF1 showed a slight trend toward increasing a factorfrom 065 to 089 with increasing ionic strength(Supplementary Figure S4C and Table 1) this resultsuggests pausing at low ionic strength which is abrogatedby salt For TRF2 however the a factor was consistently1 and did not show any significant variation with ionicstrength suggesting an unbiased random walk Dual-colorlabeling of the TRF proteins allowed us to assesswhether protein hopping could enable bypass of otherDNA-bound proteins that act as diffusion barriers(Supplementary Figure S5) Neither TRF1 nor TRF2could bypass differentially labeled proteins of the samespecies on DNA which is consistent with a TRF2sliding mechanism and suggests that TRF1 also navigatesDNA by sliding (Supplementary Data)

Next we measured the attached lifetimes of proteinndashQD complexes on DNA First however we classified theproteinndashDNA interactions into four types based on howthey behaved during a movie Type I protein binds andthen releases Type II proteins binds and doesnrsquot leaveType III protein is bound at the beginning of the moviebut releases Type IV protein is bound from the beginningto end of the movie (Figure 2D and Supplementary Table

S1) Reliable attached lifetime measurements couldonly be obtained from analysis of the Type I interactionsThe lifetimes of both TRF1 and TRF2 on DNAdecreased with increasing ionic strength ranging from18 s (175mM) to 03 s (225mM) for TRF1 and from10 s (75mM) to 34 s (225mM) for TRF2 (Table 1 andSupplementary Figure S4D) These results are consistentwith salt-sensitive electrostatic interactions between TRFproteins and DNA and increased probability of dissoci-ation from DNA during sliding as the ionic strengthincreases (54)In summary these results demonstrate that both TRF1

and TRF2 slide on DNA in search of their target DNA-binding sites TRF2 is a canonical slider whereas TRF1alsoappears to slidebutmayalter its conformationwith salt

TRF1 and TRF2 bind specifically to telomeric sequenceson DNA tightropes

To examine the dynamics of TRF1 and TRF2 binding totelomeric DNA sequences we ligated linearized T270DNA to generate long DNA substrates with alternating(TTAGGG)270 telomeric and nontelomeric regions(Figure 3A) The lengths of these DNA tightropes rangedfrom 21 to 22 mm consistent with ligation of 2ndash12 of54 kb T270 DNA fragments (Supplementary Figure S6A)TRF1 and TRF2 bound to the ligated T270 DNA tight-ropes with regular spacing (Figure 3B SupplementaryMovies S3 and S4) For both TRF1 and TRF2 the distri-butions of the distances between adjacent binders fit wellto the sum of two Gaussian distribution functions centeredat 16 and 32 mm (Figure 3C) These findings are consist-ent with the expected spacing of the telomeric regions(Figure 3B) In contrast on the ligated nontelomericDNA the distribution of TRF2 spacing was broad(Figure 3C) and no examples of three or more boundproteinndashQDs on individual DNA tightropes with aspacing of 16 or 32 mm were observed for either TRF1or TRF2 As an additional control the telomeric repeatswere spaced further apart using a 599-kb long DNAsubstrate containing only 10 TTAGGG repeats (Tel10Supplementary Figure S7 and Supplementary Movie S5)and as expected adjacent bound TRF2 molecules werefurther apart (19 mm95 contour length) than on T270We also examined how far single molecules of TRF1

and TRF2 could slide on the ligated nontelomeric DNA

Table 1 Summary of the diffusion constant a factor and lifetime of Sav-QD (655 nm)-labeled TRF1 and TRF2 on DNA at different ionic

strengths

IonicStrength (mM)

TRF1 TRF2

D (102mm2s) a Factor Lifetime (s) D (102mm2s) a Factor Lifetime (s)

75 75plusmn12 (51) 065plusmn004 (51) ndash 89plusmn09 (59) 094plusmn005 (59) 10plusmn01 (104)125 55plusmn14 (37) 072plusmn005 (37) ndash 84plusmn09 (54) 095plusmn006 (54) 25plusmn01 (106)175 49plusmn10 (40) 072plusmn006 (40) 18plusmn01(63) 95plusmn01 (63) 082plusmn003 (63) 46plusmn01 (107)225 38plusmn12 (33) 089plusmn007(33) 03plusmn001(128) 95plusmn01 (66) 084plusmn004 (66) 34plusmn01 (95)125- TRF2B 91plusmn18 (21) 093plusmn004 (21) ndash

The numbers in the parentheses indicate the total number of complexes analyzed Lifetime was measured for complexes showing both protein bindingand release events within the video frame (Type I Figure 2D) Data are presented as meanplusmnstandard error

Nucleic Acids Research 2013 5

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

TRF1 and TRF2 exhibit slower dynamics on telomericDNA

To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

6 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

DNA Ionicstrength(mM)

TRF1 TRF2

Telomeric Nontelomeric Telomeric Nontelomeric

D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

Nucleic Acids Research 2013 7

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

The basic domain is essential for the 1D search by TRF2

The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

DISCUSSION

TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

8 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

TRF1 and TRF2 strike a balance between search andspecificity

TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

Nucleic Acids Research 2013 9

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Onlineincluding [65ndash73]

ACKNOWLEDGEMENTS

The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

FUNDING

The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

Conflict of interest statement None declared

REFERENCES

1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

10 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

Nucleic Acids Research 2013 11

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

12 Nucleic Acids Research 2013

at D H

Hill L

ibrary - Acquis S on N

ovember 27 2013

httpnaroxfordjournalsorgD

ownloaded from

SUPPLEMENTARY INFORMATION

For manuscript

TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

Germany

This supplementary information contains

Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

2

2322

412aNbs

N

as πσ ++

= Equation 1

where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

[ ]summinus

=++ minus+minus

minus=∆

nN

iiniini yyxx

nNtnMSD

1

22 )()(1)(

For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

lang1199092rang(119905) = 1198711199092

6minus 161198711199092

1205874sum 1

1198994119890119909119901 minus 1

2 (119899120587120590119909

119871119909)2119905infin

119899=1(119900119889119889) Equation 4

where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

1198631119888119886119897 = 119870119861 119879

61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

119896 = 2119863(119897119887119901)2 Equation 7

where 119896 is the stepping rate Stepss

The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

119870119899119900119899119905119890119897 Equation 11

where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

120591119899119900119899119905119890119897 Equation 12

where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

Mln2

the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

The scale bar is 1 μm

2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

The scale bar is 1 μm

3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

scale bar is 1 μm

4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

min and 16 s) The scale bar is 1 μm

5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

bar is 1 μm

SUPPLEMENTARY REFERENCES

65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

vivo stoichiometry of shelterin components The Journal of biological chemistry

285 1457-1467

66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

localization analysis for individual fluorescent probes Biophysical journal 82

2775-2783

67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

of the TRFH dimerization domain of the human telomeric proteins TRF1 and

TRF2 Molecular cell 8 351-361

68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

414

69 Hughes BD (1995) Random Walks Oxford University Press 1

70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

Relationship of lipid rafts to transient confinement zones detected by single

particle tracking Biophysical journal 82 274-284

71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

journal 69 389-398

72 Destainville N and Salome L (2006) Quantification and correction of systematic

errors due to detector time-averaging in single-molecule tracking experiments

Biophysical journal 90 L17-19

73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

confinement and jumps in single-molecule membrane trajectories Phys Rev E

Stat Nonlin Soft Matter Phys 73 011915

Table S1

Ionic Strength

DNA TRF1-QDs () TRF2-QDs ()

(mM) n I II III IV n I II III IV

75 λ DNA 179 5 22 168 76 278 374 83 165 378

125 λ DNA 63 79 16 111 794 171 62 41 152 187

175 λ DNA 210 30 48 105 547 177 604 34 136 226

225 λ DNA 291 443 17 62 478 154 617 52 117 214

125 T270 184 109 6 179 652 374 27 51 166 513

225 T270 110 45 55 273 627 216 435 32 241 292

Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

54 kb 16 kb 38 kb 16 kb 38 kb

54 kb

ligated non-telomeric DNA

linearized T270 (54 kb)

16 kb

19 kb 16 kb

(TTAGGG)135 - (TTAGGG)135

linearized Tel10 (599 kb)

16 kb 42

(TTAGGG)10

DNA substrates for AFM imaging

DNA substrates for fluorescence imaging

ligated T270

599 kb ligated Tel10

599 kb

19 kb

QD Ab

Ab

C

B

100 nm 100 nm 100 nm

nm

QD

TRF2

Antibody

A

20 nm

Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

nm

Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

D

B A

C

A B

C

Ionic Strength

TRF1-QDs TRF2-QDs

(mM) n n

75 84 27 166 97

125 87 33 142 95

175 116 30 154 95

225 135 15 158 89

D

Percentages of motile TRF1- and TRF2-QDs on λ DNA

Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

A

C

D 2

microm

10 s

TRF2 + λ DNA

TRF2 + λ DNA

B

2 microm

10 s

TRF1 + λ DNA

TRF1 + λ DNA

Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

10 s

Complexes TRF1-QDs TRF2-QDs

Total number 141 43

Dual color () 19 79

Red () 41 10

Green () 39 11

E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

A

Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

225 mM ionic strength

C D

B

3 min 0 min

0 min 63 s

A

B

TRF2 + Tel10 DNA

TRF2 + Tel10 DNA

Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

125 mM 225 mM

A

C D 125 mM 225 mM

Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

A B

C D

88 plusmn 33 times10-2 microm2sec (994)

A

B 027 plusmn 013 times10-2 microm2sec (1000)

027 plusmn 009 times10-2 microm2sec (1000)

48 plusmn 13 times10-2 microm2sec (433)

Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

Unlimited DNA length

Unlimited DNA length

16 kb DNA length

16 kb DNA length

B

Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

0 min 4 min

TRF2 + T270 DNA

2 min 0 min

TRF1

TRF1 + T270 DNA A

0 min

TRF2 + T270 DNA

4 min

4 min 0 min

5 min

TRF2 + T270 DNA

B

0 min 2 min

TRF2ΔB + λ DNA

D

C

0 min 3 min

TRF2ΔB + T270 DNA

Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

A

  • Lin and countryman et all 2013 NAR
  • Lin et al Supplement ALL 10-10-13
    • Lin et al Supplementary Information text 10-10-13
    • Lin et al supplementary figure 10-10-13
      • Slide Number 1
      • Slide Number 2
      • Slide Number 4
      • Slide Number 5
      • Slide Number 6
      • Slide Number 7
      • Slide Number 8
      • Slide Number 9
      • Slide Number 10
      • Slide Number 11
      • Slide Number 12

    Removal of TRF2 from the telomeres results in loss of the30-overhang covalent fusion of telomeres and inductionof ATM and p53 dependent apoptosis (1617) BothTRF1 and TRF2 contain a TRFH domain thatmediates homodimerization and a Myb type domainthat sequence-specifically binds to telomeric DNA(Figure 1A) (12) However these two proteins differ attheir N-termini where TRF1 and TRF2 are rich inacidic and basic residues respectively Previous electronmicroscopy (EM) and atomic force microscopy (AFM)studies established that both TRF1 and TRF2 playimportant architectural roles at telomeres (18ndash21) TRF1forms protein filaments on longer telomeric repeats (27repeats) and promotes parallel pairing of telomeric tracts(19) In vitro TRF2 can remodel linear telomeric DNAinto T-loops (20)A previous cell-based study of TRF1 and TRF2 using

    fluorescence recovery after photobleaching and fluores-cence loss in photobleaching suggested that TRF1 andTRF2 interact with telomeres in a dynamic fashion (22)Although TRF1 and TRF2 are proposed to have extra-telomeric functions they preferentially localize to the TTAGGG repeat sequences whether these target sites are atinterstitial regions or at chromosome ends (23ndash25) Oncetelomeric sequences are located TRF1 and TRF2 mustfind protein partners to form the shelterin complex andto regulate the functions of other DNA-binding proteinsat telomeres (26ndash28) Despite recent advancements in theunderstanding of functions of TRF1 and TRF2 it is stillunclear how TRF1 and TRF2 are able to find telomericsequences and protein partners in a genome of billions ofbase pairsAccumulating evidence suggests that a protein can use

    one-dimensional (1D) sliding (correlated translocation

    while maintaining continuous DNA contact) jumping(noncorrelated detachment and reattachment) orhopping (correlated detachment and reattachment) tonavigate through the vast excess of nonspecific DNA se-quences in vivo (29ndash32) Investigations of DNA-bindingdynamics on nonspecific DNA at the single-moleculelevel have significantly advanced our understanding ofhow proteins with diverse functions conduct their targetDNA search (313334) However the paradoxicalrequirements of rapid search at nonspecific sites andstability at target sites have been primarily investigatedin theoretical studies (35ndash38) direct comparisons of theprotein-binding energy landscape at nonspecific sites andtarget sites from single-molecule experimental data are stilllacking

    Here we used single-molecule fluorescence imaging tostudy the dynamics of quantum dot (QD)-labeled TRF1and TRF2 proteins on DNA and DNA substrates con-taining alternating regions of telomeric and nontelomericsequences TRF1 appears to bind directly to telomeric se-quences with very little 1D searching through nontelo-meric DNA whereas TRF2 possesses a significantcomponent of 1D search Using a truncation mutant welocalized this 1D searching activity to the basic domain ofTRF2 On telomeric DNA both TRF1 and TRF2 diffuseslowly due to higher energy barriers to diffusion and theypossess longer attached lifetimes at telomeric repeatscompared with nontelomeric DNA sequences Theseobservations indicate that there is preferential binding totelomeric DNA but the affinity is not high enough toprevent TRF proteins from diffusing along TTAGGGrepeats We postulate that this allows TRF1 and TRF2to find their protein partners locally and that this is amore general mechanism for coupling the energy from

    Figure 1 TRF1- and TRF2-QDs retain DNA-binding activity (A) Schematic representations of the domain structures of TRF1 and TRF2A Acidic domain B Basic domain M Myb type domain (B) Schematic representations of TRF1- and TRF2-QD conjugates (left) BTtris-NTAcompound (middle) and the DNA substrate (T270) with two tandem (TTAGGG)135 repeats connected by a short linker region (right 54 kb inlength) (CndashE) Representative AFM images of DNA in the presence of (C) only QDs and BTtris-NTA compound (D) TRF1-QDs or (E) TRF2-QDsThe scale bar is 200 nm White arrows point to QDs bound to DNA The numbers in (CndashE) indicate the percent of DNA molecules bound with QDsin each condition The total numbers of complexes analyzed were 200 250 and 250 for no protein TRF1-QDs and TRF2-QDs respectively

    2 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    multiple weak DNA-binding components to ensure highbinding specificity on long repetitive sequences

    MATERIALS AND METHODS

    Protein purification

    Recombinant N-terminal His6-tagged TRF1 and TRF2were purified using a baculovirusinsect cell expressionsystem and an AKTA Explorer FPLC (GE Healthcare)as described previously (39) TRF2B was purified usinga bacterial expression system (40) Protein concentrationswere determined using the Bradford assay Proteins usedin this study are gt90 pure based on SDSndashPAGE andCoomassie staining Proteins are active in binding to thetelomeric DNA substrate containing three TTAGGGrepeats based on electrophoresis mobility shift assays(EMSAs)

    DNA substrates

    DNA was purchased from New England BioLabs OtherDNA substrates used in this study are shown in Figure 1Band Supplementary Figure S1 pSXneo(T2AG3) plasmidDNA containing 270 TTAGGG repeats was a gift fromDr Peter Lansdorp (University of British Columbia) (41)pGTK4 plasmid-derived Tel10 plasmid is 5994-bp longand contains 10 TTAGGG repeats and was prepared asdescribed previously (42) To generate DNA fragmentscontaining TTAGGG repeats for AFM imaging digestionof T270 DNA (10 mg) was carried out at 37C for 4 h usingHpaI (130 U) in Buffer 4 (New England BioLabs) ForTel10 plasmid digestions were carried out using XbaI (100U) in Buffer 4 For fluorescence imaging linearizedplasmids were ligated to generate longer DNA substratesusing a Quick LigationTM Kit (New England BioLabs)The ligation reactions were done at room temperaturefor 15min The nontelomeric DNA substrate withoutthe (TTAGGG)270 sequence was gel purified after the di-gestion of pSXneo(T2AG3) with BglII and XbaI FinalDNA substrate purification was done using an illustraGFXTM PCR DNA and Gel Band Purification Kit (GEHealthcare)

    ProteinndashQD conjugation

    Streptavidin-conjugated QDs (Sav-QDs) were purchasedfrom Invitrogen Biotinylated multivalent chelatortris-nitrilotriacetic acid (BTtris-NTA) was preparedaccording to the previous reports (4344) The TRFndashDNA reaction buffer contains 50mM HEPES (pH 75)and varying concentrations of NaCl (25 50 75 and100mM) The total ionic strengths are 75 125 175 and225mM respectively (45)

    For single color QD labeling of His6-tagged TRF1 orTRF2 1 ml of red QD (655 nm 1 mM Invitrogen hydro-dynamic radius 115 nm) was incubated with 1 ml ofBTtris-NTA (2mM) for 20min (46) An amount of 1 ml ofproteins (2mM) were then added to the QD-NTA solutionand incubated for additional 20min For dual-color QDlabeling 1 ml of red (1mM) and green QDs (565 nm 1 mMhydrodynamic radius 95 nm) were incubated with 1 ml of

    BTtris-NTA (2mM) (46) TRF1 or TRF2 (1ml 2 mM) wasadded to the solution and incubated for additional 20minFor fluorescence imaging unless otherwise specifiedproteinndashNTA-QD solutions were diluted 200-fold beforebeing drawn into the flow cell using a syringe pump(model SP260p World Precision Instruments) at 300 mlml flow rate The final protein concentration was 33 nMfor both TRF1 and TRF2 Protein concentrations andionic strengths of the buffer used in this study are com-parable to physiological conditions (Supplementary Text)For AFM imaging of TRF2-QDs in the presence ofmonoclonal TRF2 antibody (Imagenex Corporation)the AbTRF2NTAQD ratio was 1121 or 5121 andreactions were carried out at room temperature for 30minafter the addition of antibodies

    AFM imaging and image analysis

    All DNA and protein samples were diluted 10-fold in 1AFM buffer [25mM NaOAc 25mM HEPESndashKOH (pH75) and 10mM Mg(OAc)2] before deposition onto afreshly cleaved mica (SPI Supply) The samples werethen washed with MilliQ water and dried under a streamof nitrogen gas All images were collected in tappingmode using a MFP-3D-Bio AFM (Asylum Research)Pointprobe PPP-FMR probes (Nanosensors) withspring constants at 28Nm (nominal value) were usedAll images were captured at a scan size of 1 mm 1 mm ascan rate of 1ndash2Hz and a resolution of 512 512 pixelsThe position of TRF proteins on DNA was analyzedusing the software from Asylum Research

    Fluorescence imaging and analysis of fluorescencemicroscopy data

    Fluorescence imaging was carried out with an invertedmicroscope (Nikon Ti-E) equipped with an encodedmotorized stage perfect focus system (PFS) and a Ti-TIRF E motorized illuminator unit Fluorescenceimaging was performed by excitation at 488 nm using asolid-state laser (20mW Sapphire DPSS) a 100 objectivewith a numerical aperture of 149 (APO TIRF Nikon)and 15 additional magnification The laser power wascontrolled by using neutral density filters The excitationbeam was reflected into the objective through a TIRFfilter set containing zt488rdc and ET500LP filters Forsimultaneous imaging of green (565 nm) and red(655 nm) QDs a dual view simultaneous imaging system(DV2 Photometrics) was used in combination with aT605LPXR dichroic beamsplitter (Chroma) and a band-pass filter ET65540m (Chroma) The images werecaptured using an electron multiplied (EM) CCDcamera (iXon DU897 Andor Technology) operated at60C with an EM gain of 250 and a frame rate of20Hz Construction of the flow cell was carried out ac-cording to a procedure described previously (334748)Silica beads (5 mm Polysciences) were first treated withpoly-L-lysine hydrobromide (2500 mgml MWgt 300KDa Wako Chemicals) DNA or ligated DNA sub-strate (5 mgml) were stretched unless otherwise specifiedunder hydrodynamic flow at 300 mlmin flow rate using asyringe pump Extended DNA strands anchored between

    Nucleic Acids Research 2013 3

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    two poly-L-lysine-coated beads formed DNA tightropesAfter introducing the proteinndashQDs into the flow cell alldata collection was performed in the absence of anyfurther buffer flow The presence of YOYO-1 on DNAsignificantly reduced the diffusion constant a-factor andthe percentage of motile proteinndashQD complexes on DNAat certain salt conditions Consequently all data analysiswas done using movies collected from using unstainedDNA tightropes (Supplementary Text)

    Statistical analysis

    Single-factor ANOVA and Student-t tests were used forstatistical analysis

    RESULTS

    TRF1- and TRF2-QD conjugates are functional inDNA binding

    Fluorescent labeling of TRF1 and TRF2 was achieved byconjugating 6 histidine (His6) tagged TRF1 and TRF2to streptavidin-conjugated QDs using the biotinylatedmultivalent chelator tris-nitrilotriacetic acid (BTtris-NTA)(44) (Figure 1B see lsquoMaterials and Methodsrsquo section) Themultiple Ni-NTAs on the circular scaffold of the tris-NTAadaptor bind the His-tag with subnanomolar affinity re-sulting in a bound lifetime in the range of hours (4344)Importantly we applied a previously established methodbased on AFM imaging to characterize the stoichiometryof QDndashTRF complexes (4950) AFM imaging revealedthat using TRF2 antibody marking the presence ofTRF2 (TRF2Ab=11 or 15) among the QDs displayedTRF2-Ab complexes (24) 90 (n=39) possessed onlyone TRF2ndashAb complex (Supplementary Figure S2)

    QDs alone exhibited minimal nonspecific binding toDNA as confirmed by AFM (Figure 1C) As expectedaddition of QD-labeled TRF1 or TRF2 to DNA contain-ing two stretches of (TTAGGG)135 connected by a shortlinker region (T270 DNA Figure 1B see lsquoMaterialsand Methodsrsquo section) resulted in substantial binding(Figure 1D and E) Furthermore AFM image analysisrevealed that both TRF1- and TRF2-QDs bound prefer-entially to the telomeric DNA sequences on both the T270and Tel10 DNA substrates (Supplementary Figure S3)

    TRF1 and TRF2 diffuse one-dimensionally onnontelomeric DNA

    To study the dynamics of individual TRF1 and TRF2 mol-ecules on DNA using oblique-angle fluorescence micros-copy we applied a DNA tightrope assay (Figure 2A)(33) DNA strands are suspended between poly-L-lysinecoated microspheres at an elongation of 90 DNAcontour length using hydrodynamic flow (47) Thisprocess isolates DNA from the surface and does notrequire continuous buffer flow for the observation ofproteinndashDNA interactions QDs did not bind to DNAtightropes alone or in the presence of TRF proteinswithout BTtris-NTA However with both BTtris-NTAand His6-tagged TRF1 or TRF2 QDs were observed onDNA throughout the visual field (Figure 2B and C) BothTRF1- (Supplementary Movie S1) and TRF2-QDs(Supplementary Movie S2) showed clear 1D diffusion onDNA which was tracked by Gaussian fitting to kymo-graphs (particle position versus time plots SupplementaryData) (3347)

    To determine whether TRF1 and TRF2 slide or hop weevaluated the effect of ionic conditions on the dynamic

    Figure 2 DNA tightrope assay based oblique-angle fluorescence imaging of TRF1- and TRF2-QDs on DNA tightropes (A) A schematic drawingof the DNA tightropes (green lines) bound with QD (red ball)-labeled proteins (green balls) between silica beads (large white balls) The drawing isnot to scale (B and C) Representative fluorescence images of red (655 nm) QD-conjugated His6-TRF1 (B) and His6-TRF2 (C) on DNA (stainedwith YoYo1) (D) Classification of different types of proteinndashDNA interactions observed with TRF2-QDs on DNA for attached lifetimemeasurement

    4 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    interactions between the QD-labeled TRF proteins andDNA Increasing the salt concentration should not affectthe diffusion constants of a sliding process but shouldelevate the diffusion constants of hopping (295152) Weperformed experiments at 75 125 175 and 225mM ionicstrengths (see lsquoMaterials and Methodsrsquo section) Thefraction of motile TRF1 proteins ranged from 15 to33 (Supplementary Figure S4A) and followed a trendof decreasing diffusion constants as the ionic strengthincreased (75ndash38 102mm2s) such that the differencebetween the highest and lowest salt was statistically signifi-cant (P=0017 Supplementary Figure S4B and Table 1)In contrast TRF2 was highly motile on DNA acrossall ionic strengths and showed no significant change indiffusion constant (84ndash95 102 mm2s SupplementaryFigure S4B and Table 1) TRF2 diffused substantiallyfaster than TRF1 at all ionic strengths showing statisticalsignificance at ionic strengths between 125 and 225mM

    In addition to the diffusion constant we also measuredthe diffusive exponent (a-factor Supplementary Data)An a factor of 1 indicates an unbiased random walk gt1indicates directed motion and lt1 indicates periodsof pausing in the random walk (subdiffusion) (53)TRF1 showed a slight trend toward increasing a factorfrom 065 to 089 with increasing ionic strength(Supplementary Figure S4C and Table 1) this resultsuggests pausing at low ionic strength which is abrogatedby salt For TRF2 however the a factor was consistently1 and did not show any significant variation with ionicstrength suggesting an unbiased random walk Dual-colorlabeling of the TRF proteins allowed us to assesswhether protein hopping could enable bypass of otherDNA-bound proteins that act as diffusion barriers(Supplementary Figure S5) Neither TRF1 nor TRF2could bypass differentially labeled proteins of the samespecies on DNA which is consistent with a TRF2sliding mechanism and suggests that TRF1 also navigatesDNA by sliding (Supplementary Data)

    Next we measured the attached lifetimes of proteinndashQD complexes on DNA First however we classified theproteinndashDNA interactions into four types based on howthey behaved during a movie Type I protein binds andthen releases Type II proteins binds and doesnrsquot leaveType III protein is bound at the beginning of the moviebut releases Type IV protein is bound from the beginningto end of the movie (Figure 2D and Supplementary Table

    S1) Reliable attached lifetime measurements couldonly be obtained from analysis of the Type I interactionsThe lifetimes of both TRF1 and TRF2 on DNAdecreased with increasing ionic strength ranging from18 s (175mM) to 03 s (225mM) for TRF1 and from10 s (75mM) to 34 s (225mM) for TRF2 (Table 1 andSupplementary Figure S4D) These results are consistentwith salt-sensitive electrostatic interactions between TRFproteins and DNA and increased probability of dissoci-ation from DNA during sliding as the ionic strengthincreases (54)In summary these results demonstrate that both TRF1

    and TRF2 slide on DNA in search of their target DNA-binding sites TRF2 is a canonical slider whereas TRF1alsoappears to slidebutmayalter its conformationwith salt

    TRF1 and TRF2 bind specifically to telomeric sequenceson DNA tightropes

    To examine the dynamics of TRF1 and TRF2 binding totelomeric DNA sequences we ligated linearized T270DNA to generate long DNA substrates with alternating(TTAGGG)270 telomeric and nontelomeric regions(Figure 3A) The lengths of these DNA tightropes rangedfrom 21 to 22 mm consistent with ligation of 2ndash12 of54 kb T270 DNA fragments (Supplementary Figure S6A)TRF1 and TRF2 bound to the ligated T270 DNA tight-ropes with regular spacing (Figure 3B SupplementaryMovies S3 and S4) For both TRF1 and TRF2 the distri-butions of the distances between adjacent binders fit wellto the sum of two Gaussian distribution functions centeredat 16 and 32 mm (Figure 3C) These findings are consist-ent with the expected spacing of the telomeric regions(Figure 3B) In contrast on the ligated nontelomericDNA the distribution of TRF2 spacing was broad(Figure 3C) and no examples of three or more boundproteinndashQDs on individual DNA tightropes with aspacing of 16 or 32 mm were observed for either TRF1or TRF2 As an additional control the telomeric repeatswere spaced further apart using a 599-kb long DNAsubstrate containing only 10 TTAGGG repeats (Tel10Supplementary Figure S7 and Supplementary Movie S5)and as expected adjacent bound TRF2 molecules werefurther apart (19 mm95 contour length) than on T270We also examined how far single molecules of TRF1

    and TRF2 could slide on the ligated nontelomeric DNA

    Table 1 Summary of the diffusion constant a factor and lifetime of Sav-QD (655 nm)-labeled TRF1 and TRF2 on DNA at different ionic

    strengths

    IonicStrength (mM)

    TRF1 TRF2

    D (102mm2s) a Factor Lifetime (s) D (102mm2s) a Factor Lifetime (s)

    75 75plusmn12 (51) 065plusmn004 (51) ndash 89plusmn09 (59) 094plusmn005 (59) 10plusmn01 (104)125 55plusmn14 (37) 072plusmn005 (37) ndash 84plusmn09 (54) 095plusmn006 (54) 25plusmn01 (106)175 49plusmn10 (40) 072plusmn006 (40) 18plusmn01(63) 95plusmn01 (63) 082plusmn003 (63) 46plusmn01 (107)225 38plusmn12 (33) 089plusmn007(33) 03plusmn001(128) 95plusmn01 (66) 084plusmn004 (66) 34plusmn01 (95)125- TRF2B 91plusmn18 (21) 093plusmn004 (21) ndash

    The numbers in the parentheses indicate the total number of complexes analyzed Lifetime was measured for complexes showing both protein bindingand release events within the video frame (Type I Figure 2D) Data are presented as meanplusmnstandard error

    Nucleic Acids Research 2013 5

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

    were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

    Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

    TRF1 and TRF2 exhibit slower dynamics on telomericDNA

    To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

    Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

    6 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

    simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

    Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

    Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

    DNA Ionicstrength(mM)

    TRF1 TRF2

    Telomeric Nontelomeric Telomeric Nontelomeric

    D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

    Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

    Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

    Nucleic Acids Research 2013 7

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

    The basic domain is essential for the 1D search by TRF2

    The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

    three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

    DISCUSSION

    TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

    TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

    Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

    8 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

    Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

    We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

    behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

    4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

    TRF1 and TRF2 strike a balance between search andspecificity

    TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

    series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

    Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

    Nucleic Acids Research 2013 9

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

    AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

    TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

    tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

    a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

    SUPPLEMENTARY DATA

    Supplementary Data are available at NAR Onlineincluding [65ndash73]

    ACKNOWLEDGEMENTS

    The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

    FUNDING

    The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

    Conflict of interest statement None declared

    REFERENCES

    1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

    2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

    3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

    4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

    5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

    6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

    7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

    8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

    9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

    10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

    11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

    12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

    10 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

    14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

    15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

    16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

    17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

    18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

    19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

    20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

    21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

    22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

    23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

    24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

    25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

    26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

    27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

    28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

    29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

    30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

    31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

    32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

    33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

    34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

    35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

    36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

    37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

    38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

    39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

    40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

    41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

    42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

    43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

    44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

    45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

    46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

    47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

    48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

    49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

    50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

    51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

    52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

    53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

    54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

    Nucleic Acids Research 2013 11

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

    56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

    57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

    58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

    59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

    60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

    61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

    62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

    63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

    functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

    64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

    65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

    66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

    67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

    68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

    69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

    70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

    71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

    72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

    73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

    12 Nucleic Acids Research 2013

    at D H

    Hill L

    ibrary - Acquis S on N

    ovember 27 2013

    httpnaroxfordjournalsorgD

    ownloaded from

    SUPPLEMENTARY INFORMATION

    For manuscript

    TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

    Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

    1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

    Germany

    This supplementary information contains

    Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

    SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

    2

    2322

    412aNbs

    N

    as πσ ++

    = Equation 1

    where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

    are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

    TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

    [ ]summinus

    =++ minus+minus

    minus=∆

    nN

    iiniini yyxx

    nNtnMSD

    1

    22 )()(1)(

    For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

    lang1199092rang(119905) = 1198711199092

    6minus 161198711199092

    1205874sum 1

    1198994119890119909119901 minus 1

    2 (119899120587120590119909

    119871119909)2119905infin

    119899=1(119900119889119889) Equation 4

    where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

    119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

    where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

    Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

    Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

    1198631119888119886119897 = 119870119861 119879

    61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

    where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

    119896 = 2119863(119897119887119901)2 Equation 7

    where 119896 is the stepping rate Stepss

    The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

    119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

    119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

    ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

    120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

    119870119899119900119899119905119890119897 Equation 11

    where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

    120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

    120591119899119900119899119905119890119897 Equation 12

    where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

    regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

    programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

    Mln2

    the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

    Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

    Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

    Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

    1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

    The scale bar is 1 μm

    2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

    1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

    The scale bar is 1 μm

    3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

    on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

    scale bar is 1 μm

    4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

    TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

    min and 16 s) The scale bar is 1 μm

    5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

    on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

    bar is 1 μm

    SUPPLEMENTARY REFERENCES

    65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

    vivo stoichiometry of shelterin components The Journal of biological chemistry

    285 1457-1467

    66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

    localization analysis for individual fluorescent probes Biophysical journal 82

    2775-2783

    67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

    of the TRFH dimerization domain of the human telomeric proteins TRF1 and

    TRF2 Molecular cell 8 351-361

    68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

    absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

    414

    69 Hughes BD (1995) Random Walks Oxford University Press 1

    70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

    Relationship of lipid rafts to transient confinement zones detected by single

    particle tracking Biophysical journal 82 274-284

    71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

    journal 69 389-398

    72 Destainville N and Salome L (2006) Quantification and correction of systematic

    errors due to detector time-averaging in single-molecule tracking experiments

    Biophysical journal 90 L17-19

    73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

    confinement and jumps in single-molecule membrane trajectories Phys Rev E

    Stat Nonlin Soft Matter Phys 73 011915

    Table S1

    Ionic Strength

    DNA TRF1-QDs () TRF2-QDs ()

    (mM) n I II III IV n I II III IV

    75 λ DNA 179 5 22 168 76 278 374 83 165 378

    125 λ DNA 63 79 16 111 794 171 62 41 152 187

    175 λ DNA 210 30 48 105 547 177 604 34 136 226

    225 λ DNA 291 443 17 62 478 154 617 52 117 214

    125 T270 184 109 6 179 652 374 27 51 166 513

    225 T270 110 45 55 273 627 216 435 32 241 292

    Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

    Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

    Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

    54 kb 16 kb 38 kb 16 kb 38 kb

    54 kb

    ligated non-telomeric DNA

    linearized T270 (54 kb)

    16 kb

    19 kb 16 kb

    (TTAGGG)135 - (TTAGGG)135

    linearized Tel10 (599 kb)

    16 kb 42

    (TTAGGG)10

    DNA substrates for AFM imaging

    DNA substrates for fluorescence imaging

    ligated T270

    599 kb ligated Tel10

    599 kb

    19 kb

    QD Ab

    Ab

    C

    B

    100 nm 100 nm 100 nm

    nm

    QD

    TRF2

    Antibody

    A

    20 nm

    Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

    nm

    Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

    D

    B A

    C

    A B

    C

    Ionic Strength

    TRF1-QDs TRF2-QDs

    (mM) n n

    75 84 27 166 97

    125 87 33 142 95

    175 116 30 154 95

    225 135 15 158 89

    D

    Percentages of motile TRF1- and TRF2-QDs on λ DNA

    Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

    A

    C

    D 2

    microm

    10 s

    TRF2 + λ DNA

    TRF2 + λ DNA

    B

    2 microm

    10 s

    TRF1 + λ DNA

    TRF1 + λ DNA

    Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

    10 s

    Complexes TRF1-QDs TRF2-QDs

    Total number 141 43

    Dual color () 19 79

    Red () 41 10

    Green () 39 11

    E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

    A

    Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

    225 mM ionic strength

    C D

    B

    3 min 0 min

    0 min 63 s

    A

    B

    TRF2 + Tel10 DNA

    TRF2 + Tel10 DNA

    Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

    125 mM 225 mM

    A

    C D 125 mM 225 mM

    Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

    A B

    C D

    88 plusmn 33 times10-2 microm2sec (994)

    A

    B 027 plusmn 013 times10-2 microm2sec (1000)

    027 plusmn 009 times10-2 microm2sec (1000)

    48 plusmn 13 times10-2 microm2sec (433)

    Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

    Unlimited DNA length

    Unlimited DNA length

    16 kb DNA length

    16 kb DNA length

    B

    Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

    0 min 4 min

    TRF2 + T270 DNA

    2 min 0 min

    TRF1

    TRF1 + T270 DNA A

    0 min

    TRF2 + T270 DNA

    4 min

    4 min 0 min

    5 min

    TRF2 + T270 DNA

    B

    0 min 2 min

    TRF2ΔB + λ DNA

    D

    C

    0 min 3 min

    TRF2ΔB + T270 DNA

    Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

    A

    • Lin and countryman et all 2013 NAR
    • Lin et al Supplement ALL 10-10-13
      • Lin et al Supplementary Information text 10-10-13
      • Lin et al supplementary figure 10-10-13
        • Slide Number 1
        • Slide Number 2
        • Slide Number 4
        • Slide Number 5
        • Slide Number 6
        • Slide Number 7
        • Slide Number 8
        • Slide Number 9
        • Slide Number 10
        • Slide Number 11
        • Slide Number 12

      multiple weak DNA-binding components to ensure highbinding specificity on long repetitive sequences

      MATERIALS AND METHODS

      Protein purification

      Recombinant N-terminal His6-tagged TRF1 and TRF2were purified using a baculovirusinsect cell expressionsystem and an AKTA Explorer FPLC (GE Healthcare)as described previously (39) TRF2B was purified usinga bacterial expression system (40) Protein concentrationswere determined using the Bradford assay Proteins usedin this study are gt90 pure based on SDSndashPAGE andCoomassie staining Proteins are active in binding to thetelomeric DNA substrate containing three TTAGGGrepeats based on electrophoresis mobility shift assays(EMSAs)

      DNA substrates

      DNA was purchased from New England BioLabs OtherDNA substrates used in this study are shown in Figure 1Band Supplementary Figure S1 pSXneo(T2AG3) plasmidDNA containing 270 TTAGGG repeats was a gift fromDr Peter Lansdorp (University of British Columbia) (41)pGTK4 plasmid-derived Tel10 plasmid is 5994-bp longand contains 10 TTAGGG repeats and was prepared asdescribed previously (42) To generate DNA fragmentscontaining TTAGGG repeats for AFM imaging digestionof T270 DNA (10 mg) was carried out at 37C for 4 h usingHpaI (130 U) in Buffer 4 (New England BioLabs) ForTel10 plasmid digestions were carried out using XbaI (100U) in Buffer 4 For fluorescence imaging linearizedplasmids were ligated to generate longer DNA substratesusing a Quick LigationTM Kit (New England BioLabs)The ligation reactions were done at room temperaturefor 15min The nontelomeric DNA substrate withoutthe (TTAGGG)270 sequence was gel purified after the di-gestion of pSXneo(T2AG3) with BglII and XbaI FinalDNA substrate purification was done using an illustraGFXTM PCR DNA and Gel Band Purification Kit (GEHealthcare)

      ProteinndashQD conjugation

      Streptavidin-conjugated QDs (Sav-QDs) were purchasedfrom Invitrogen Biotinylated multivalent chelatortris-nitrilotriacetic acid (BTtris-NTA) was preparedaccording to the previous reports (4344) The TRFndashDNA reaction buffer contains 50mM HEPES (pH 75)and varying concentrations of NaCl (25 50 75 and100mM) The total ionic strengths are 75 125 175 and225mM respectively (45)

      For single color QD labeling of His6-tagged TRF1 orTRF2 1 ml of red QD (655 nm 1 mM Invitrogen hydro-dynamic radius 115 nm) was incubated with 1 ml ofBTtris-NTA (2mM) for 20min (46) An amount of 1 ml ofproteins (2mM) were then added to the QD-NTA solutionand incubated for additional 20min For dual-color QDlabeling 1 ml of red (1mM) and green QDs (565 nm 1 mMhydrodynamic radius 95 nm) were incubated with 1 ml of

      BTtris-NTA (2mM) (46) TRF1 or TRF2 (1ml 2 mM) wasadded to the solution and incubated for additional 20minFor fluorescence imaging unless otherwise specifiedproteinndashNTA-QD solutions were diluted 200-fold beforebeing drawn into the flow cell using a syringe pump(model SP260p World Precision Instruments) at 300 mlml flow rate The final protein concentration was 33 nMfor both TRF1 and TRF2 Protein concentrations andionic strengths of the buffer used in this study are com-parable to physiological conditions (Supplementary Text)For AFM imaging of TRF2-QDs in the presence ofmonoclonal TRF2 antibody (Imagenex Corporation)the AbTRF2NTAQD ratio was 1121 or 5121 andreactions were carried out at room temperature for 30minafter the addition of antibodies

      AFM imaging and image analysis

      All DNA and protein samples were diluted 10-fold in 1AFM buffer [25mM NaOAc 25mM HEPESndashKOH (pH75) and 10mM Mg(OAc)2] before deposition onto afreshly cleaved mica (SPI Supply) The samples werethen washed with MilliQ water and dried under a streamof nitrogen gas All images were collected in tappingmode using a MFP-3D-Bio AFM (Asylum Research)Pointprobe PPP-FMR probes (Nanosensors) withspring constants at 28Nm (nominal value) were usedAll images were captured at a scan size of 1 mm 1 mm ascan rate of 1ndash2Hz and a resolution of 512 512 pixelsThe position of TRF proteins on DNA was analyzedusing the software from Asylum Research

      Fluorescence imaging and analysis of fluorescencemicroscopy data

      Fluorescence imaging was carried out with an invertedmicroscope (Nikon Ti-E) equipped with an encodedmotorized stage perfect focus system (PFS) and a Ti-TIRF E motorized illuminator unit Fluorescenceimaging was performed by excitation at 488 nm using asolid-state laser (20mW Sapphire DPSS) a 100 objectivewith a numerical aperture of 149 (APO TIRF Nikon)and 15 additional magnification The laser power wascontrolled by using neutral density filters The excitationbeam was reflected into the objective through a TIRFfilter set containing zt488rdc and ET500LP filters Forsimultaneous imaging of green (565 nm) and red(655 nm) QDs a dual view simultaneous imaging system(DV2 Photometrics) was used in combination with aT605LPXR dichroic beamsplitter (Chroma) and a band-pass filter ET65540m (Chroma) The images werecaptured using an electron multiplied (EM) CCDcamera (iXon DU897 Andor Technology) operated at60C with an EM gain of 250 and a frame rate of20Hz Construction of the flow cell was carried out ac-cording to a procedure described previously (334748)Silica beads (5 mm Polysciences) were first treated withpoly-L-lysine hydrobromide (2500 mgml MWgt 300KDa Wako Chemicals) DNA or ligated DNA sub-strate (5 mgml) were stretched unless otherwise specifiedunder hydrodynamic flow at 300 mlmin flow rate using asyringe pump Extended DNA strands anchored between

      Nucleic Acids Research 2013 3

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      two poly-L-lysine-coated beads formed DNA tightropesAfter introducing the proteinndashQDs into the flow cell alldata collection was performed in the absence of anyfurther buffer flow The presence of YOYO-1 on DNAsignificantly reduced the diffusion constant a-factor andthe percentage of motile proteinndashQD complexes on DNAat certain salt conditions Consequently all data analysiswas done using movies collected from using unstainedDNA tightropes (Supplementary Text)

      Statistical analysis

      Single-factor ANOVA and Student-t tests were used forstatistical analysis

      RESULTS

      TRF1- and TRF2-QD conjugates are functional inDNA binding

      Fluorescent labeling of TRF1 and TRF2 was achieved byconjugating 6 histidine (His6) tagged TRF1 and TRF2to streptavidin-conjugated QDs using the biotinylatedmultivalent chelator tris-nitrilotriacetic acid (BTtris-NTA)(44) (Figure 1B see lsquoMaterials and Methodsrsquo section) Themultiple Ni-NTAs on the circular scaffold of the tris-NTAadaptor bind the His-tag with subnanomolar affinity re-sulting in a bound lifetime in the range of hours (4344)Importantly we applied a previously established methodbased on AFM imaging to characterize the stoichiometryof QDndashTRF complexes (4950) AFM imaging revealedthat using TRF2 antibody marking the presence ofTRF2 (TRF2Ab=11 or 15) among the QDs displayedTRF2-Ab complexes (24) 90 (n=39) possessed onlyone TRF2ndashAb complex (Supplementary Figure S2)

      QDs alone exhibited minimal nonspecific binding toDNA as confirmed by AFM (Figure 1C) As expectedaddition of QD-labeled TRF1 or TRF2 to DNA contain-ing two stretches of (TTAGGG)135 connected by a shortlinker region (T270 DNA Figure 1B see lsquoMaterialsand Methodsrsquo section) resulted in substantial binding(Figure 1D and E) Furthermore AFM image analysisrevealed that both TRF1- and TRF2-QDs bound prefer-entially to the telomeric DNA sequences on both the T270and Tel10 DNA substrates (Supplementary Figure S3)

      TRF1 and TRF2 diffuse one-dimensionally onnontelomeric DNA

      To study the dynamics of individual TRF1 and TRF2 mol-ecules on DNA using oblique-angle fluorescence micros-copy we applied a DNA tightrope assay (Figure 2A)(33) DNA strands are suspended between poly-L-lysinecoated microspheres at an elongation of 90 DNAcontour length using hydrodynamic flow (47) Thisprocess isolates DNA from the surface and does notrequire continuous buffer flow for the observation ofproteinndashDNA interactions QDs did not bind to DNAtightropes alone or in the presence of TRF proteinswithout BTtris-NTA However with both BTtris-NTAand His6-tagged TRF1 or TRF2 QDs were observed onDNA throughout the visual field (Figure 2B and C) BothTRF1- (Supplementary Movie S1) and TRF2-QDs(Supplementary Movie S2) showed clear 1D diffusion onDNA which was tracked by Gaussian fitting to kymo-graphs (particle position versus time plots SupplementaryData) (3347)

      To determine whether TRF1 and TRF2 slide or hop weevaluated the effect of ionic conditions on the dynamic

      Figure 2 DNA tightrope assay based oblique-angle fluorescence imaging of TRF1- and TRF2-QDs on DNA tightropes (A) A schematic drawingof the DNA tightropes (green lines) bound with QD (red ball)-labeled proteins (green balls) between silica beads (large white balls) The drawing isnot to scale (B and C) Representative fluorescence images of red (655 nm) QD-conjugated His6-TRF1 (B) and His6-TRF2 (C) on DNA (stainedwith YoYo1) (D) Classification of different types of proteinndashDNA interactions observed with TRF2-QDs on DNA for attached lifetimemeasurement

      4 Nucleic Acids Research 2013

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      interactions between the QD-labeled TRF proteins andDNA Increasing the salt concentration should not affectthe diffusion constants of a sliding process but shouldelevate the diffusion constants of hopping (295152) Weperformed experiments at 75 125 175 and 225mM ionicstrengths (see lsquoMaterials and Methodsrsquo section) Thefraction of motile TRF1 proteins ranged from 15 to33 (Supplementary Figure S4A) and followed a trendof decreasing diffusion constants as the ionic strengthincreased (75ndash38 102mm2s) such that the differencebetween the highest and lowest salt was statistically signifi-cant (P=0017 Supplementary Figure S4B and Table 1)In contrast TRF2 was highly motile on DNA acrossall ionic strengths and showed no significant change indiffusion constant (84ndash95 102 mm2s SupplementaryFigure S4B and Table 1) TRF2 diffused substantiallyfaster than TRF1 at all ionic strengths showing statisticalsignificance at ionic strengths between 125 and 225mM

      In addition to the diffusion constant we also measuredthe diffusive exponent (a-factor Supplementary Data)An a factor of 1 indicates an unbiased random walk gt1indicates directed motion and lt1 indicates periodsof pausing in the random walk (subdiffusion) (53)TRF1 showed a slight trend toward increasing a factorfrom 065 to 089 with increasing ionic strength(Supplementary Figure S4C and Table 1) this resultsuggests pausing at low ionic strength which is abrogatedby salt For TRF2 however the a factor was consistently1 and did not show any significant variation with ionicstrength suggesting an unbiased random walk Dual-colorlabeling of the TRF proteins allowed us to assesswhether protein hopping could enable bypass of otherDNA-bound proteins that act as diffusion barriers(Supplementary Figure S5) Neither TRF1 nor TRF2could bypass differentially labeled proteins of the samespecies on DNA which is consistent with a TRF2sliding mechanism and suggests that TRF1 also navigatesDNA by sliding (Supplementary Data)

      Next we measured the attached lifetimes of proteinndashQD complexes on DNA First however we classified theproteinndashDNA interactions into four types based on howthey behaved during a movie Type I protein binds andthen releases Type II proteins binds and doesnrsquot leaveType III protein is bound at the beginning of the moviebut releases Type IV protein is bound from the beginningto end of the movie (Figure 2D and Supplementary Table

      S1) Reliable attached lifetime measurements couldonly be obtained from analysis of the Type I interactionsThe lifetimes of both TRF1 and TRF2 on DNAdecreased with increasing ionic strength ranging from18 s (175mM) to 03 s (225mM) for TRF1 and from10 s (75mM) to 34 s (225mM) for TRF2 (Table 1 andSupplementary Figure S4D) These results are consistentwith salt-sensitive electrostatic interactions between TRFproteins and DNA and increased probability of dissoci-ation from DNA during sliding as the ionic strengthincreases (54)In summary these results demonstrate that both TRF1

      and TRF2 slide on DNA in search of their target DNA-binding sites TRF2 is a canonical slider whereas TRF1alsoappears to slidebutmayalter its conformationwith salt

      TRF1 and TRF2 bind specifically to telomeric sequenceson DNA tightropes

      To examine the dynamics of TRF1 and TRF2 binding totelomeric DNA sequences we ligated linearized T270DNA to generate long DNA substrates with alternating(TTAGGG)270 telomeric and nontelomeric regions(Figure 3A) The lengths of these DNA tightropes rangedfrom 21 to 22 mm consistent with ligation of 2ndash12 of54 kb T270 DNA fragments (Supplementary Figure S6A)TRF1 and TRF2 bound to the ligated T270 DNA tight-ropes with regular spacing (Figure 3B SupplementaryMovies S3 and S4) For both TRF1 and TRF2 the distri-butions of the distances between adjacent binders fit wellto the sum of two Gaussian distribution functions centeredat 16 and 32 mm (Figure 3C) These findings are consist-ent with the expected spacing of the telomeric regions(Figure 3B) In contrast on the ligated nontelomericDNA the distribution of TRF2 spacing was broad(Figure 3C) and no examples of three or more boundproteinndashQDs on individual DNA tightropes with aspacing of 16 or 32 mm were observed for either TRF1or TRF2 As an additional control the telomeric repeatswere spaced further apart using a 599-kb long DNAsubstrate containing only 10 TTAGGG repeats (Tel10Supplementary Figure S7 and Supplementary Movie S5)and as expected adjacent bound TRF2 molecules werefurther apart (19 mm95 contour length) than on T270We also examined how far single molecules of TRF1

      and TRF2 could slide on the ligated nontelomeric DNA

      Table 1 Summary of the diffusion constant a factor and lifetime of Sav-QD (655 nm)-labeled TRF1 and TRF2 on DNA at different ionic

      strengths

      IonicStrength (mM)

      TRF1 TRF2

      D (102mm2s) a Factor Lifetime (s) D (102mm2s) a Factor Lifetime (s)

      75 75plusmn12 (51) 065plusmn004 (51) ndash 89plusmn09 (59) 094plusmn005 (59) 10plusmn01 (104)125 55plusmn14 (37) 072plusmn005 (37) ndash 84plusmn09 (54) 095plusmn006 (54) 25plusmn01 (106)175 49plusmn10 (40) 072plusmn006 (40) 18plusmn01(63) 95plusmn01 (63) 082plusmn003 (63) 46plusmn01 (107)225 38plusmn12 (33) 089plusmn007(33) 03plusmn001(128) 95plusmn01 (66) 084plusmn004 (66) 34plusmn01 (95)125- TRF2B 91plusmn18 (21) 093plusmn004 (21) ndash

      The numbers in the parentheses indicate the total number of complexes analyzed Lifetime was measured for complexes showing both protein bindingand release events within the video frame (Type I Figure 2D) Data are presented as meanplusmnstandard error

      Nucleic Acids Research 2013 5

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

      were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

      Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

      TRF1 and TRF2 exhibit slower dynamics on telomericDNA

      To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

      Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

      6 Nucleic Acids Research 2013

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

      simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

      Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

      Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

      DNA Ionicstrength(mM)

      TRF1 TRF2

      Telomeric Nontelomeric Telomeric Nontelomeric

      D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

      Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

      Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

      Nucleic Acids Research 2013 7

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

      The basic domain is essential for the 1D search by TRF2

      The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

      three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

      DISCUSSION

      TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

      TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

      Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

      8 Nucleic Acids Research 2013

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

      Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

      We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

      behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

      4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

      TRF1 and TRF2 strike a balance between search andspecificity

      TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

      series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

      Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

      Nucleic Acids Research 2013 9

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

      AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

      TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

      tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

      a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

      SUPPLEMENTARY DATA

      Supplementary Data are available at NAR Onlineincluding [65ndash73]

      ACKNOWLEDGEMENTS

      The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

      FUNDING

      The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

      Conflict of interest statement None declared

      REFERENCES

      1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

      2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

      3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

      4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

      5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

      6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

      7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

      8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

      9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

      10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

      11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

      12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

      10 Nucleic Acids Research 2013

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

      14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

      15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

      16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

      17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

      18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

      19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

      20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

      21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

      22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

      23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

      24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

      25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

      26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

      27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

      28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

      29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

      30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

      31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

      32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

      33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

      34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

      35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

      36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

      37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

      38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

      39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

      40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

      41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

      42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

      43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

      44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

      45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

      46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

      47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

      48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

      49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

      50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

      51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

      52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

      53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

      54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

      Nucleic Acids Research 2013 11

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

      56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

      57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

      58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

      59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

      60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

      61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

      62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

      63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

      functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

      64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

      65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

      66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

      67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

      68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

      69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

      70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

      71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

      72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

      73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

      12 Nucleic Acids Research 2013

      at D H

      Hill L

      ibrary - Acquis S on N

      ovember 27 2013

      httpnaroxfordjournalsorgD

      ownloaded from

      SUPPLEMENTARY INFORMATION

      For manuscript

      TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

      Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

      1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

      Germany

      This supplementary information contains

      Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

      SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

      2

      2322

      412aNbs

      N

      as πσ ++

      = Equation 1

      where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

      are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

      TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

      [ ]summinus

      =++ minus+minus

      minus=∆

      nN

      iiniini yyxx

      nNtnMSD

      1

      22 )()(1)(

      For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

      lang1199092rang(119905) = 1198711199092

      6minus 161198711199092

      1205874sum 1

      1198994119890119909119901 minus 1

      2 (119899120587120590119909

      119871119909)2119905infin

      119899=1(119900119889119889) Equation 4

      where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

      119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

      where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

      Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

      Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

      1198631119888119886119897 = 119870119861 119879

      61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

      where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

      119896 = 2119863(119897119887119901)2 Equation 7

      where 119896 is the stepping rate Stepss

      The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

      119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

      119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

      ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

      120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

      119870119899119900119899119905119890119897 Equation 11

      where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

      120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

      120591119899119900119899119905119890119897 Equation 12

      where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

      regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

      programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

      Mln2

      the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

      Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

      Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

      Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

      1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

      The scale bar is 1 μm

      2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

      1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

      The scale bar is 1 μm

      3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

      on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

      scale bar is 1 μm

      4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

      TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

      min and 16 s) The scale bar is 1 μm

      5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

      on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

      bar is 1 μm

      SUPPLEMENTARY REFERENCES

      65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

      vivo stoichiometry of shelterin components The Journal of biological chemistry

      285 1457-1467

      66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

      localization analysis for individual fluorescent probes Biophysical journal 82

      2775-2783

      67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

      of the TRFH dimerization domain of the human telomeric proteins TRF1 and

      TRF2 Molecular cell 8 351-361

      68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

      absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

      414

      69 Hughes BD (1995) Random Walks Oxford University Press 1

      70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

      Relationship of lipid rafts to transient confinement zones detected by single

      particle tracking Biophysical journal 82 274-284

      71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

      journal 69 389-398

      72 Destainville N and Salome L (2006) Quantification and correction of systematic

      errors due to detector time-averaging in single-molecule tracking experiments

      Biophysical journal 90 L17-19

      73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

      confinement and jumps in single-molecule membrane trajectories Phys Rev E

      Stat Nonlin Soft Matter Phys 73 011915

      Table S1

      Ionic Strength

      DNA TRF1-QDs () TRF2-QDs ()

      (mM) n I II III IV n I II III IV

      75 λ DNA 179 5 22 168 76 278 374 83 165 378

      125 λ DNA 63 79 16 111 794 171 62 41 152 187

      175 λ DNA 210 30 48 105 547 177 604 34 136 226

      225 λ DNA 291 443 17 62 478 154 617 52 117 214

      125 T270 184 109 6 179 652 374 27 51 166 513

      225 T270 110 45 55 273 627 216 435 32 241 292

      Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

      Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

      Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

      54 kb 16 kb 38 kb 16 kb 38 kb

      54 kb

      ligated non-telomeric DNA

      linearized T270 (54 kb)

      16 kb

      19 kb 16 kb

      (TTAGGG)135 - (TTAGGG)135

      linearized Tel10 (599 kb)

      16 kb 42

      (TTAGGG)10

      DNA substrates for AFM imaging

      DNA substrates for fluorescence imaging

      ligated T270

      599 kb ligated Tel10

      599 kb

      19 kb

      QD Ab

      Ab

      C

      B

      100 nm 100 nm 100 nm

      nm

      QD

      TRF2

      Antibody

      A

      20 nm

      Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

      nm

      Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

      D

      B A

      C

      A B

      C

      Ionic Strength

      TRF1-QDs TRF2-QDs

      (mM) n n

      75 84 27 166 97

      125 87 33 142 95

      175 116 30 154 95

      225 135 15 158 89

      D

      Percentages of motile TRF1- and TRF2-QDs on λ DNA

      Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

      A

      C

      D 2

      microm

      10 s

      TRF2 + λ DNA

      TRF2 + λ DNA

      B

      2 microm

      10 s

      TRF1 + λ DNA

      TRF1 + λ DNA

      Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

      10 s

      Complexes TRF1-QDs TRF2-QDs

      Total number 141 43

      Dual color () 19 79

      Red () 41 10

      Green () 39 11

      E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

      A

      Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

      225 mM ionic strength

      C D

      B

      3 min 0 min

      0 min 63 s

      A

      B

      TRF2 + Tel10 DNA

      TRF2 + Tel10 DNA

      Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

      125 mM 225 mM

      A

      C D 125 mM 225 mM

      Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

      A B

      C D

      88 plusmn 33 times10-2 microm2sec (994)

      A

      B 027 plusmn 013 times10-2 microm2sec (1000)

      027 plusmn 009 times10-2 microm2sec (1000)

      48 plusmn 13 times10-2 microm2sec (433)

      Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

      Unlimited DNA length

      Unlimited DNA length

      16 kb DNA length

      16 kb DNA length

      B

      Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

      0 min 4 min

      TRF2 + T270 DNA

      2 min 0 min

      TRF1

      TRF1 + T270 DNA A

      0 min

      TRF2 + T270 DNA

      4 min

      4 min 0 min

      5 min

      TRF2 + T270 DNA

      B

      0 min 2 min

      TRF2ΔB + λ DNA

      D

      C

      0 min 3 min

      TRF2ΔB + T270 DNA

      Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

      A

      • Lin and countryman et all 2013 NAR
      • Lin et al Supplement ALL 10-10-13
        • Lin et al Supplementary Information text 10-10-13
        • Lin et al supplementary figure 10-10-13
          • Slide Number 1
          • Slide Number 2
          • Slide Number 4
          • Slide Number 5
          • Slide Number 6
          • Slide Number 7
          • Slide Number 8
          • Slide Number 9
          • Slide Number 10
          • Slide Number 11
          • Slide Number 12

        two poly-L-lysine-coated beads formed DNA tightropesAfter introducing the proteinndashQDs into the flow cell alldata collection was performed in the absence of anyfurther buffer flow The presence of YOYO-1 on DNAsignificantly reduced the diffusion constant a-factor andthe percentage of motile proteinndashQD complexes on DNAat certain salt conditions Consequently all data analysiswas done using movies collected from using unstainedDNA tightropes (Supplementary Text)

        Statistical analysis

        Single-factor ANOVA and Student-t tests were used forstatistical analysis

        RESULTS

        TRF1- and TRF2-QD conjugates are functional inDNA binding

        Fluorescent labeling of TRF1 and TRF2 was achieved byconjugating 6 histidine (His6) tagged TRF1 and TRF2to streptavidin-conjugated QDs using the biotinylatedmultivalent chelator tris-nitrilotriacetic acid (BTtris-NTA)(44) (Figure 1B see lsquoMaterials and Methodsrsquo section) Themultiple Ni-NTAs on the circular scaffold of the tris-NTAadaptor bind the His-tag with subnanomolar affinity re-sulting in a bound lifetime in the range of hours (4344)Importantly we applied a previously established methodbased on AFM imaging to characterize the stoichiometryof QDndashTRF complexes (4950) AFM imaging revealedthat using TRF2 antibody marking the presence ofTRF2 (TRF2Ab=11 or 15) among the QDs displayedTRF2-Ab complexes (24) 90 (n=39) possessed onlyone TRF2ndashAb complex (Supplementary Figure S2)

        QDs alone exhibited minimal nonspecific binding toDNA as confirmed by AFM (Figure 1C) As expectedaddition of QD-labeled TRF1 or TRF2 to DNA contain-ing two stretches of (TTAGGG)135 connected by a shortlinker region (T270 DNA Figure 1B see lsquoMaterialsand Methodsrsquo section) resulted in substantial binding(Figure 1D and E) Furthermore AFM image analysisrevealed that both TRF1- and TRF2-QDs bound prefer-entially to the telomeric DNA sequences on both the T270and Tel10 DNA substrates (Supplementary Figure S3)

        TRF1 and TRF2 diffuse one-dimensionally onnontelomeric DNA

        To study the dynamics of individual TRF1 and TRF2 mol-ecules on DNA using oblique-angle fluorescence micros-copy we applied a DNA tightrope assay (Figure 2A)(33) DNA strands are suspended between poly-L-lysinecoated microspheres at an elongation of 90 DNAcontour length using hydrodynamic flow (47) Thisprocess isolates DNA from the surface and does notrequire continuous buffer flow for the observation ofproteinndashDNA interactions QDs did not bind to DNAtightropes alone or in the presence of TRF proteinswithout BTtris-NTA However with both BTtris-NTAand His6-tagged TRF1 or TRF2 QDs were observed onDNA throughout the visual field (Figure 2B and C) BothTRF1- (Supplementary Movie S1) and TRF2-QDs(Supplementary Movie S2) showed clear 1D diffusion onDNA which was tracked by Gaussian fitting to kymo-graphs (particle position versus time plots SupplementaryData) (3347)

        To determine whether TRF1 and TRF2 slide or hop weevaluated the effect of ionic conditions on the dynamic

        Figure 2 DNA tightrope assay based oblique-angle fluorescence imaging of TRF1- and TRF2-QDs on DNA tightropes (A) A schematic drawingof the DNA tightropes (green lines) bound with QD (red ball)-labeled proteins (green balls) between silica beads (large white balls) The drawing isnot to scale (B and C) Representative fluorescence images of red (655 nm) QD-conjugated His6-TRF1 (B) and His6-TRF2 (C) on DNA (stainedwith YoYo1) (D) Classification of different types of proteinndashDNA interactions observed with TRF2-QDs on DNA for attached lifetimemeasurement

        4 Nucleic Acids Research 2013

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        interactions between the QD-labeled TRF proteins andDNA Increasing the salt concentration should not affectthe diffusion constants of a sliding process but shouldelevate the diffusion constants of hopping (295152) Weperformed experiments at 75 125 175 and 225mM ionicstrengths (see lsquoMaterials and Methodsrsquo section) Thefraction of motile TRF1 proteins ranged from 15 to33 (Supplementary Figure S4A) and followed a trendof decreasing diffusion constants as the ionic strengthincreased (75ndash38 102mm2s) such that the differencebetween the highest and lowest salt was statistically signifi-cant (P=0017 Supplementary Figure S4B and Table 1)In contrast TRF2 was highly motile on DNA acrossall ionic strengths and showed no significant change indiffusion constant (84ndash95 102 mm2s SupplementaryFigure S4B and Table 1) TRF2 diffused substantiallyfaster than TRF1 at all ionic strengths showing statisticalsignificance at ionic strengths between 125 and 225mM

        In addition to the diffusion constant we also measuredthe diffusive exponent (a-factor Supplementary Data)An a factor of 1 indicates an unbiased random walk gt1indicates directed motion and lt1 indicates periodsof pausing in the random walk (subdiffusion) (53)TRF1 showed a slight trend toward increasing a factorfrom 065 to 089 with increasing ionic strength(Supplementary Figure S4C and Table 1) this resultsuggests pausing at low ionic strength which is abrogatedby salt For TRF2 however the a factor was consistently1 and did not show any significant variation with ionicstrength suggesting an unbiased random walk Dual-colorlabeling of the TRF proteins allowed us to assesswhether protein hopping could enable bypass of otherDNA-bound proteins that act as diffusion barriers(Supplementary Figure S5) Neither TRF1 nor TRF2could bypass differentially labeled proteins of the samespecies on DNA which is consistent with a TRF2sliding mechanism and suggests that TRF1 also navigatesDNA by sliding (Supplementary Data)

        Next we measured the attached lifetimes of proteinndashQD complexes on DNA First however we classified theproteinndashDNA interactions into four types based on howthey behaved during a movie Type I protein binds andthen releases Type II proteins binds and doesnrsquot leaveType III protein is bound at the beginning of the moviebut releases Type IV protein is bound from the beginningto end of the movie (Figure 2D and Supplementary Table

        S1) Reliable attached lifetime measurements couldonly be obtained from analysis of the Type I interactionsThe lifetimes of both TRF1 and TRF2 on DNAdecreased with increasing ionic strength ranging from18 s (175mM) to 03 s (225mM) for TRF1 and from10 s (75mM) to 34 s (225mM) for TRF2 (Table 1 andSupplementary Figure S4D) These results are consistentwith salt-sensitive electrostatic interactions between TRFproteins and DNA and increased probability of dissoci-ation from DNA during sliding as the ionic strengthincreases (54)In summary these results demonstrate that both TRF1

        and TRF2 slide on DNA in search of their target DNA-binding sites TRF2 is a canonical slider whereas TRF1alsoappears to slidebutmayalter its conformationwith salt

        TRF1 and TRF2 bind specifically to telomeric sequenceson DNA tightropes

        To examine the dynamics of TRF1 and TRF2 binding totelomeric DNA sequences we ligated linearized T270DNA to generate long DNA substrates with alternating(TTAGGG)270 telomeric and nontelomeric regions(Figure 3A) The lengths of these DNA tightropes rangedfrom 21 to 22 mm consistent with ligation of 2ndash12 of54 kb T270 DNA fragments (Supplementary Figure S6A)TRF1 and TRF2 bound to the ligated T270 DNA tight-ropes with regular spacing (Figure 3B SupplementaryMovies S3 and S4) For both TRF1 and TRF2 the distri-butions of the distances between adjacent binders fit wellto the sum of two Gaussian distribution functions centeredat 16 and 32 mm (Figure 3C) These findings are consist-ent with the expected spacing of the telomeric regions(Figure 3B) In contrast on the ligated nontelomericDNA the distribution of TRF2 spacing was broad(Figure 3C) and no examples of three or more boundproteinndashQDs on individual DNA tightropes with aspacing of 16 or 32 mm were observed for either TRF1or TRF2 As an additional control the telomeric repeatswere spaced further apart using a 599-kb long DNAsubstrate containing only 10 TTAGGG repeats (Tel10Supplementary Figure S7 and Supplementary Movie S5)and as expected adjacent bound TRF2 molecules werefurther apart (19 mm95 contour length) than on T270We also examined how far single molecules of TRF1

        and TRF2 could slide on the ligated nontelomeric DNA

        Table 1 Summary of the diffusion constant a factor and lifetime of Sav-QD (655 nm)-labeled TRF1 and TRF2 on DNA at different ionic

        strengths

        IonicStrength (mM)

        TRF1 TRF2

        D (102mm2s) a Factor Lifetime (s) D (102mm2s) a Factor Lifetime (s)

        75 75plusmn12 (51) 065plusmn004 (51) ndash 89plusmn09 (59) 094plusmn005 (59) 10plusmn01 (104)125 55plusmn14 (37) 072plusmn005 (37) ndash 84plusmn09 (54) 095plusmn006 (54) 25plusmn01 (106)175 49plusmn10 (40) 072plusmn006 (40) 18plusmn01(63) 95plusmn01 (63) 082plusmn003 (63) 46plusmn01 (107)225 38plusmn12 (33) 089plusmn007(33) 03plusmn001(128) 95plusmn01 (66) 084plusmn004 (66) 34plusmn01 (95)125- TRF2B 91plusmn18 (21) 093plusmn004 (21) ndash

        The numbers in the parentheses indicate the total number of complexes analyzed Lifetime was measured for complexes showing both protein bindingand release events within the video frame (Type I Figure 2D) Data are presented as meanplusmnstandard error

        Nucleic Acids Research 2013 5

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

        were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

        Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

        TRF1 and TRF2 exhibit slower dynamics on telomericDNA

        To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

        Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

        6 Nucleic Acids Research 2013

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

        simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

        Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

        Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

        DNA Ionicstrength(mM)

        TRF1 TRF2

        Telomeric Nontelomeric Telomeric Nontelomeric

        D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

        Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

        Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

        Nucleic Acids Research 2013 7

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

        The basic domain is essential for the 1D search by TRF2

        The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

        three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

        DISCUSSION

        TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

        TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

        Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

        8 Nucleic Acids Research 2013

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

        Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

        We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

        behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

        4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

        TRF1 and TRF2 strike a balance between search andspecificity

        TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

        series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

        Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

        Nucleic Acids Research 2013 9

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

        AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

        TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

        tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

        a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

        SUPPLEMENTARY DATA

        Supplementary Data are available at NAR Onlineincluding [65ndash73]

        ACKNOWLEDGEMENTS

        The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

        FUNDING

        The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

        Conflict of interest statement None declared

        REFERENCES

        1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

        2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

        3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

        4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

        5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

        6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

        7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

        8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

        9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

        10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

        11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

        12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

        10 Nucleic Acids Research 2013

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

        14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

        15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

        16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

        17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

        18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

        19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

        20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

        21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

        22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

        23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

        24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

        25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

        26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

        27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

        28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

        29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

        30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

        31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

        32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

        33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

        34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

        35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

        36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

        37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

        38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

        39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

        40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

        41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

        42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

        43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

        44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

        45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

        46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

        47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

        48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

        49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

        50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

        51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

        52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

        53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

        54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

        Nucleic Acids Research 2013 11

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

        56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

        57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

        58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

        59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

        60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

        61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

        62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

        63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

        functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

        64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

        65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

        66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

        67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

        68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

        69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

        70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

        71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

        72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

        73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

        12 Nucleic Acids Research 2013

        at D H

        Hill L

        ibrary - Acquis S on N

        ovember 27 2013

        httpnaroxfordjournalsorgD

        ownloaded from

        SUPPLEMENTARY INFORMATION

        For manuscript

        TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

        Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

        1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

        Germany

        This supplementary information contains

        Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

        SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

        2

        2322

        412aNbs

        N

        as πσ ++

        = Equation 1

        where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

        are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

        TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

        [ ]summinus

        =++ minus+minus

        minus=∆

        nN

        iiniini yyxx

        nNtnMSD

        1

        22 )()(1)(

        For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

        lang1199092rang(119905) = 1198711199092

        6minus 161198711199092

        1205874sum 1

        1198994119890119909119901 minus 1

        2 (119899120587120590119909

        119871119909)2119905infin

        119899=1(119900119889119889) Equation 4

        where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

        119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

        where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

        Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

        Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

        1198631119888119886119897 = 119870119861 119879

        61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

        where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

        119896 = 2119863(119897119887119901)2 Equation 7

        where 119896 is the stepping rate Stepss

        The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

        119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

        119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

        ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

        120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

        119870119899119900119899119905119890119897 Equation 11

        where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

        120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

        120591119899119900119899119905119890119897 Equation 12

        where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

        regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

        programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

        Mln2

        the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

        Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

        Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

        Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

        1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

        The scale bar is 1 μm

        2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

        1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

        The scale bar is 1 μm

        3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

        on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

        scale bar is 1 μm

        4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

        TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

        min and 16 s) The scale bar is 1 μm

        5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

        on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

        bar is 1 μm

        SUPPLEMENTARY REFERENCES

        65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

        vivo stoichiometry of shelterin components The Journal of biological chemistry

        285 1457-1467

        66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

        localization analysis for individual fluorescent probes Biophysical journal 82

        2775-2783

        67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

        of the TRFH dimerization domain of the human telomeric proteins TRF1 and

        TRF2 Molecular cell 8 351-361

        68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

        absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

        414

        69 Hughes BD (1995) Random Walks Oxford University Press 1

        70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

        Relationship of lipid rafts to transient confinement zones detected by single

        particle tracking Biophysical journal 82 274-284

        71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

        journal 69 389-398

        72 Destainville N and Salome L (2006) Quantification and correction of systematic

        errors due to detector time-averaging in single-molecule tracking experiments

        Biophysical journal 90 L17-19

        73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

        confinement and jumps in single-molecule membrane trajectories Phys Rev E

        Stat Nonlin Soft Matter Phys 73 011915

        Table S1

        Ionic Strength

        DNA TRF1-QDs () TRF2-QDs ()

        (mM) n I II III IV n I II III IV

        75 λ DNA 179 5 22 168 76 278 374 83 165 378

        125 λ DNA 63 79 16 111 794 171 62 41 152 187

        175 λ DNA 210 30 48 105 547 177 604 34 136 226

        225 λ DNA 291 443 17 62 478 154 617 52 117 214

        125 T270 184 109 6 179 652 374 27 51 166 513

        225 T270 110 45 55 273 627 216 435 32 241 292

        Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

        Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

        Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

        54 kb 16 kb 38 kb 16 kb 38 kb

        54 kb

        ligated non-telomeric DNA

        linearized T270 (54 kb)

        16 kb

        19 kb 16 kb

        (TTAGGG)135 - (TTAGGG)135

        linearized Tel10 (599 kb)

        16 kb 42

        (TTAGGG)10

        DNA substrates for AFM imaging

        DNA substrates for fluorescence imaging

        ligated T270

        599 kb ligated Tel10

        599 kb

        19 kb

        QD Ab

        Ab

        C

        B

        100 nm 100 nm 100 nm

        nm

        QD

        TRF2

        Antibody

        A

        20 nm

        Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

        nm

        Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

        D

        B A

        C

        A B

        C

        Ionic Strength

        TRF1-QDs TRF2-QDs

        (mM) n n

        75 84 27 166 97

        125 87 33 142 95

        175 116 30 154 95

        225 135 15 158 89

        D

        Percentages of motile TRF1- and TRF2-QDs on λ DNA

        Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

        A

        C

        D 2

        microm

        10 s

        TRF2 + λ DNA

        TRF2 + λ DNA

        B

        2 microm

        10 s

        TRF1 + λ DNA

        TRF1 + λ DNA

        Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

        10 s

        Complexes TRF1-QDs TRF2-QDs

        Total number 141 43

        Dual color () 19 79

        Red () 41 10

        Green () 39 11

        E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

        A

        Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

        225 mM ionic strength

        C D

        B

        3 min 0 min

        0 min 63 s

        A

        B

        TRF2 + Tel10 DNA

        TRF2 + Tel10 DNA

        Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

        125 mM 225 mM

        A

        C D 125 mM 225 mM

        Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

        A B

        C D

        88 plusmn 33 times10-2 microm2sec (994)

        A

        B 027 plusmn 013 times10-2 microm2sec (1000)

        027 plusmn 009 times10-2 microm2sec (1000)

        48 plusmn 13 times10-2 microm2sec (433)

        Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

        Unlimited DNA length

        Unlimited DNA length

        16 kb DNA length

        16 kb DNA length

        B

        Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

        0 min 4 min

        TRF2 + T270 DNA

        2 min 0 min

        TRF1

        TRF1 + T270 DNA A

        0 min

        TRF2 + T270 DNA

        4 min

        4 min 0 min

        5 min

        TRF2 + T270 DNA

        B

        0 min 2 min

        TRF2ΔB + λ DNA

        D

        C

        0 min 3 min

        TRF2ΔB + T270 DNA

        Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

        A

        • Lin and countryman et all 2013 NAR
        • Lin et al Supplement ALL 10-10-13
          • Lin et al Supplementary Information text 10-10-13
          • Lin et al supplementary figure 10-10-13
            • Slide Number 1
            • Slide Number 2
            • Slide Number 4
            • Slide Number 5
            • Slide Number 6
            • Slide Number 7
            • Slide Number 8
            • Slide Number 9
            • Slide Number 10
            • Slide Number 11
            • Slide Number 12

          interactions between the QD-labeled TRF proteins andDNA Increasing the salt concentration should not affectthe diffusion constants of a sliding process but shouldelevate the diffusion constants of hopping (295152) Weperformed experiments at 75 125 175 and 225mM ionicstrengths (see lsquoMaterials and Methodsrsquo section) Thefraction of motile TRF1 proteins ranged from 15 to33 (Supplementary Figure S4A) and followed a trendof decreasing diffusion constants as the ionic strengthincreased (75ndash38 102mm2s) such that the differencebetween the highest and lowest salt was statistically signifi-cant (P=0017 Supplementary Figure S4B and Table 1)In contrast TRF2 was highly motile on DNA acrossall ionic strengths and showed no significant change indiffusion constant (84ndash95 102 mm2s SupplementaryFigure S4B and Table 1) TRF2 diffused substantiallyfaster than TRF1 at all ionic strengths showing statisticalsignificance at ionic strengths between 125 and 225mM

          In addition to the diffusion constant we also measuredthe diffusive exponent (a-factor Supplementary Data)An a factor of 1 indicates an unbiased random walk gt1indicates directed motion and lt1 indicates periodsof pausing in the random walk (subdiffusion) (53)TRF1 showed a slight trend toward increasing a factorfrom 065 to 089 with increasing ionic strength(Supplementary Figure S4C and Table 1) this resultsuggests pausing at low ionic strength which is abrogatedby salt For TRF2 however the a factor was consistently1 and did not show any significant variation with ionicstrength suggesting an unbiased random walk Dual-colorlabeling of the TRF proteins allowed us to assesswhether protein hopping could enable bypass of otherDNA-bound proteins that act as diffusion barriers(Supplementary Figure S5) Neither TRF1 nor TRF2could bypass differentially labeled proteins of the samespecies on DNA which is consistent with a TRF2sliding mechanism and suggests that TRF1 also navigatesDNA by sliding (Supplementary Data)

          Next we measured the attached lifetimes of proteinndashQD complexes on DNA First however we classified theproteinndashDNA interactions into four types based on howthey behaved during a movie Type I protein binds andthen releases Type II proteins binds and doesnrsquot leaveType III protein is bound at the beginning of the moviebut releases Type IV protein is bound from the beginningto end of the movie (Figure 2D and Supplementary Table

          S1) Reliable attached lifetime measurements couldonly be obtained from analysis of the Type I interactionsThe lifetimes of both TRF1 and TRF2 on DNAdecreased with increasing ionic strength ranging from18 s (175mM) to 03 s (225mM) for TRF1 and from10 s (75mM) to 34 s (225mM) for TRF2 (Table 1 andSupplementary Figure S4D) These results are consistentwith salt-sensitive electrostatic interactions between TRFproteins and DNA and increased probability of dissoci-ation from DNA during sliding as the ionic strengthincreases (54)In summary these results demonstrate that both TRF1

          and TRF2 slide on DNA in search of their target DNA-binding sites TRF2 is a canonical slider whereas TRF1alsoappears to slidebutmayalter its conformationwith salt

          TRF1 and TRF2 bind specifically to telomeric sequenceson DNA tightropes

          To examine the dynamics of TRF1 and TRF2 binding totelomeric DNA sequences we ligated linearized T270DNA to generate long DNA substrates with alternating(TTAGGG)270 telomeric and nontelomeric regions(Figure 3A) The lengths of these DNA tightropes rangedfrom 21 to 22 mm consistent with ligation of 2ndash12 of54 kb T270 DNA fragments (Supplementary Figure S6A)TRF1 and TRF2 bound to the ligated T270 DNA tight-ropes with regular spacing (Figure 3B SupplementaryMovies S3 and S4) For both TRF1 and TRF2 the distri-butions of the distances between adjacent binders fit wellto the sum of two Gaussian distribution functions centeredat 16 and 32 mm (Figure 3C) These findings are consist-ent with the expected spacing of the telomeric regions(Figure 3B) In contrast on the ligated nontelomericDNA the distribution of TRF2 spacing was broad(Figure 3C) and no examples of three or more boundproteinndashQDs on individual DNA tightropes with aspacing of 16 or 32 mm were observed for either TRF1or TRF2 As an additional control the telomeric repeatswere spaced further apart using a 599-kb long DNAsubstrate containing only 10 TTAGGG repeats (Tel10Supplementary Figure S7 and Supplementary Movie S5)and as expected adjacent bound TRF2 molecules werefurther apart (19 mm95 contour length) than on T270We also examined how far single molecules of TRF1

          and TRF2 could slide on the ligated nontelomeric DNA

          Table 1 Summary of the diffusion constant a factor and lifetime of Sav-QD (655 nm)-labeled TRF1 and TRF2 on DNA at different ionic

          strengths

          IonicStrength (mM)

          TRF1 TRF2

          D (102mm2s) a Factor Lifetime (s) D (102mm2s) a Factor Lifetime (s)

          75 75plusmn12 (51) 065plusmn004 (51) ndash 89plusmn09 (59) 094plusmn005 (59) 10plusmn01 (104)125 55plusmn14 (37) 072plusmn005 (37) ndash 84plusmn09 (54) 095plusmn006 (54) 25plusmn01 (106)175 49plusmn10 (40) 072plusmn006 (40) 18plusmn01(63) 95plusmn01 (63) 082plusmn003 (63) 46plusmn01 (107)225 38plusmn12 (33) 089plusmn007(33) 03plusmn001(128) 95plusmn01 (66) 084plusmn004 (66) 34plusmn01 (95)125- TRF2B 91plusmn18 (21) 093plusmn004 (21) ndash

          The numbers in the parentheses indicate the total number of complexes analyzed Lifetime was measured for complexes showing both protein bindingand release events within the video frame (Type I Figure 2D) Data are presented as meanplusmnstandard error

          Nucleic Acids Research 2013 5

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

          were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

          Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

          TRF1 and TRF2 exhibit slower dynamics on telomericDNA

          To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

          Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

          6 Nucleic Acids Research 2013

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

          simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

          Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

          Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

          DNA Ionicstrength(mM)

          TRF1 TRF2

          Telomeric Nontelomeric Telomeric Nontelomeric

          D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

          Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

          Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

          Nucleic Acids Research 2013 7

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

          The basic domain is essential for the 1D search by TRF2

          The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

          three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

          DISCUSSION

          TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

          TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

          Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

          8 Nucleic Acids Research 2013

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

          Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

          We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

          behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

          4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

          TRF1 and TRF2 strike a balance between search andspecificity

          TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

          series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

          Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

          Nucleic Acids Research 2013 9

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

          AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

          TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

          tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

          a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

          SUPPLEMENTARY DATA

          Supplementary Data are available at NAR Onlineincluding [65ndash73]

          ACKNOWLEDGEMENTS

          The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

          FUNDING

          The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

          Conflict of interest statement None declared

          REFERENCES

          1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

          2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

          3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

          4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

          5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

          6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

          7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

          8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

          9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

          10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

          11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

          12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

          10 Nucleic Acids Research 2013

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

          14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

          15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

          16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

          17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

          18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

          19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

          20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

          21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

          22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

          23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

          24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

          25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

          26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

          27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

          28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

          29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

          30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

          31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

          32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

          33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

          34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

          35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

          36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

          37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

          38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

          39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

          40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

          41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

          42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

          43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

          44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

          45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

          46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

          47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

          48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

          49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

          50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

          51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

          52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

          53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

          54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

          Nucleic Acids Research 2013 11

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

          56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

          57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

          58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

          59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

          60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

          61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

          62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

          63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

          functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

          64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

          65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

          66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

          67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

          68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

          69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

          70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

          71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

          72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

          73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

          12 Nucleic Acids Research 2013

          at D H

          Hill L

          ibrary - Acquis S on N

          ovember 27 2013

          httpnaroxfordjournalsorgD

          ownloaded from

          SUPPLEMENTARY INFORMATION

          For manuscript

          TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

          Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

          1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

          Germany

          This supplementary information contains

          Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

          SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

          2

          2322

          412aNbs

          N

          as πσ ++

          = Equation 1

          where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

          are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

          TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

          [ ]summinus

          =++ minus+minus

          minus=∆

          nN

          iiniini yyxx

          nNtnMSD

          1

          22 )()(1)(

          For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

          lang1199092rang(119905) = 1198711199092

          6minus 161198711199092

          1205874sum 1

          1198994119890119909119901 minus 1

          2 (119899120587120590119909

          119871119909)2119905infin

          119899=1(119900119889119889) Equation 4

          where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

          119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

          where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

          Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

          Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

          1198631119888119886119897 = 119870119861 119879

          61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

          where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

          119896 = 2119863(119897119887119901)2 Equation 7

          where 119896 is the stepping rate Stepss

          The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

          119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

          119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

          ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

          120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

          119870119899119900119899119905119890119897 Equation 11

          where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

          120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

          120591119899119900119899119905119890119897 Equation 12

          where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

          regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

          programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

          Mln2

          the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

          Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

          Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

          Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

          1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

          The scale bar is 1 μm

          2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

          1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

          The scale bar is 1 μm

          3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

          on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

          scale bar is 1 μm

          4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

          TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

          min and 16 s) The scale bar is 1 μm

          5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

          on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

          bar is 1 μm

          SUPPLEMENTARY REFERENCES

          65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

          vivo stoichiometry of shelterin components The Journal of biological chemistry

          285 1457-1467

          66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

          localization analysis for individual fluorescent probes Biophysical journal 82

          2775-2783

          67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

          of the TRFH dimerization domain of the human telomeric proteins TRF1 and

          TRF2 Molecular cell 8 351-361

          68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

          absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

          414

          69 Hughes BD (1995) Random Walks Oxford University Press 1

          70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

          Relationship of lipid rafts to transient confinement zones detected by single

          particle tracking Biophysical journal 82 274-284

          71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

          journal 69 389-398

          72 Destainville N and Salome L (2006) Quantification and correction of systematic

          errors due to detector time-averaging in single-molecule tracking experiments

          Biophysical journal 90 L17-19

          73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

          confinement and jumps in single-molecule membrane trajectories Phys Rev E

          Stat Nonlin Soft Matter Phys 73 011915

          Table S1

          Ionic Strength

          DNA TRF1-QDs () TRF2-QDs ()

          (mM) n I II III IV n I II III IV

          75 λ DNA 179 5 22 168 76 278 374 83 165 378

          125 λ DNA 63 79 16 111 794 171 62 41 152 187

          175 λ DNA 210 30 48 105 547 177 604 34 136 226

          225 λ DNA 291 443 17 62 478 154 617 52 117 214

          125 T270 184 109 6 179 652 374 27 51 166 513

          225 T270 110 45 55 273 627 216 435 32 241 292

          Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

          Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

          Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

          54 kb 16 kb 38 kb 16 kb 38 kb

          54 kb

          ligated non-telomeric DNA

          linearized T270 (54 kb)

          16 kb

          19 kb 16 kb

          (TTAGGG)135 - (TTAGGG)135

          linearized Tel10 (599 kb)

          16 kb 42

          (TTAGGG)10

          DNA substrates for AFM imaging

          DNA substrates for fluorescence imaging

          ligated T270

          599 kb ligated Tel10

          599 kb

          19 kb

          QD Ab

          Ab

          C

          B

          100 nm 100 nm 100 nm

          nm

          QD

          TRF2

          Antibody

          A

          20 nm

          Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

          nm

          Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

          D

          B A

          C

          A B

          C

          Ionic Strength

          TRF1-QDs TRF2-QDs

          (mM) n n

          75 84 27 166 97

          125 87 33 142 95

          175 116 30 154 95

          225 135 15 158 89

          D

          Percentages of motile TRF1- and TRF2-QDs on λ DNA

          Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

          A

          C

          D 2

          microm

          10 s

          TRF2 + λ DNA

          TRF2 + λ DNA

          B

          2 microm

          10 s

          TRF1 + λ DNA

          TRF1 + λ DNA

          Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

          10 s

          Complexes TRF1-QDs TRF2-QDs

          Total number 141 43

          Dual color () 19 79

          Red () 41 10

          Green () 39 11

          E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

          A

          Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

          225 mM ionic strength

          C D

          B

          3 min 0 min

          0 min 63 s

          A

          B

          TRF2 + Tel10 DNA

          TRF2 + Tel10 DNA

          Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

          125 mM 225 mM

          A

          C D 125 mM 225 mM

          Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

          A B

          C D

          88 plusmn 33 times10-2 microm2sec (994)

          A

          B 027 plusmn 013 times10-2 microm2sec (1000)

          027 plusmn 009 times10-2 microm2sec (1000)

          48 plusmn 13 times10-2 microm2sec (433)

          Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

          Unlimited DNA length

          Unlimited DNA length

          16 kb DNA length

          16 kb DNA length

          B

          Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

          0 min 4 min

          TRF2 + T270 DNA

          2 min 0 min

          TRF1

          TRF1 + T270 DNA A

          0 min

          TRF2 + T270 DNA

          4 min

          4 min 0 min

          5 min

          TRF2 + T270 DNA

          B

          0 min 2 min

          TRF2ΔB + λ DNA

          D

          C

          0 min 3 min

          TRF2ΔB + T270 DNA

          Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

          A

          • Lin and countryman et all 2013 NAR
          • Lin et al Supplement ALL 10-10-13
            • Lin et al Supplementary Information text 10-10-13
            • Lin et al supplementary figure 10-10-13
              • Slide Number 1
              • Slide Number 2
              • Slide Number 4
              • Slide Number 5
              • Slide Number 6
              • Slide Number 7
              • Slide Number 8
              • Slide Number 9
              • Slide Number 10
              • Slide Number 11
              • Slide Number 12

            versus ligated T270 DNA (Figure 4 SupplementaryMovies S3 and S4) On T270 DNA TRF1 displayed onemajor population with diffusion ranges centered on038mm (Figure 4D) TRF2 exhibited two distinct popu-lations centered on 05 and 12 mm at 125mM ionicstrength and on 05 and 15 mm at 225mM ionicstrength (Figure 4 and Supplementary Figure S6B) Buton nontelomeric DNA no clear peak was evident (whitebars Figure 4D) Approximately 90 (n=29) of TRF1and 73 (n=30) of TRF2 diffused in a short range(lt850 nm) The diffusion range was invariant across alltime windows (10ndash100 s Supplementary Figure S8)ruling out the possibility that the short range diffusionobserved was due to shorter video lengths Instead thisfinding suggests that once the molecules are within a telo-meric region they tend to remain there We explored thepossibility that short range diffusion was caused bymultiple proteins binding to the same telomeric regionand restricting 1D sliding However at a lower TRF2concentration the short diffusion range did not change(compare Figure 4D and Supplementary Figure S6C)Therefore the two diffusion range populations couldbe assigned to diffusion of TRF proteins over the (TTAGGG)270 telomeric regions (05 mm 90 contourlength) and the nontelomeric spacers (12 mm 90contour length) respectively (Figure 3B) For TRF2 tran-sitions were observed between telomeric and nontelomericregions or even between two adjacent T270 repeats which

            were more frequent at 225mM ionic strength (whitearrows Figure 4C) These events provided the peak withdiffusion range centered at 15mm (SupplementaryFigure S6B)

            Taken together the regular spacing betweenQD-labeled TRFs demonstrated that TRF1 and TRF2bind specifically to the telomeric regions on both T270and Tel10 DNA substrates These results also showedthat compared with TRF2 TRF1 undergoes a greaternumber of direct binding events from solution to the (TTAGGG)270 region forgoing a 1D search (Figure 4D)

            TRF1 and TRF2 exhibit slower dynamics on telomericDNA

            To quantify the diffusion constants at the (TTAGGG)270telomeric region we selectively analyzed TRF1 and TRF2on the ligated T270 DNA tightropes with at least three ormore proteinndashQDs in a row spaced at the length ofnontelomeric spacers (15ndash17 mm Figures 3 and 4)TRF1 and TRF2 diffused at 015ndash022 102 mm2s and027ndash029 102 mm2s at the (TTAGGG)270 region re-spectively These rates are 17- to 37- and 30-foldslower for TRF1 and TRF2 respectively comparedwith those on DNA at the same ionic strength(Tables 1 and 2) We noted that in many cases TRFproteins binding to telomere repeats would be confinedto diffuse within this region due to the higher affinity for

            Figure 3 TRF1- and TRF2-QDs bind specifically to telomeric sequences on DNA tightropes (A) A representative fluorescence image of DNAtightropes formed using ligated linear T270 DNA containing telomeric sequences (stained with YoYo1) (B) A schematic drawing of the ligated T270DNA substrate (top) and representative fluorescence images of dual color (655 and 565 nm)-labeled TRF1- (middle) and TRF2-QDs (bottom) on theligated T270 DNA substrate (C) Measured distances between two adjacent TRF1- (n=96 top) and TRF2-QDs (bottom n=96) on the ligatedT270 substrate (blue bars) and between TRF2-QDs on the nontelomeric DNA substrate (bottom white bars n=204) The lines in the top andbottom panels are double Gaussian fits to the data which have R2 of 099 and 095 respectively

            6 Nucleic Acids Research 2013

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

            simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

            Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

            Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

            DNA Ionicstrength(mM)

            TRF1 TRF2

            Telomeric Nontelomeric Telomeric Nontelomeric

            D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

            Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

            Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

            Nucleic Acids Research 2013 7

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

            The basic domain is essential for the 1D search by TRF2

            The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

            three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

            DISCUSSION

            TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

            TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

            Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

            8 Nucleic Acids Research 2013

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

            Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

            We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

            behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

            4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

            TRF1 and TRF2 strike a balance between search andspecificity

            TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

            series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

            Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

            Nucleic Acids Research 2013 9

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

            AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

            TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

            tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

            a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

            SUPPLEMENTARY DATA

            Supplementary Data are available at NAR Onlineincluding [65ndash73]

            ACKNOWLEDGEMENTS

            The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

            FUNDING

            The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

            Conflict of interest statement None declared

            REFERENCES

            1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

            2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

            3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

            4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

            5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

            6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

            7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

            8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

            9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

            10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

            11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

            12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

            10 Nucleic Acids Research 2013

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

            14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

            15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

            16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

            17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

            18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

            19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

            20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

            21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

            22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

            23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

            24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

            25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

            26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

            27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

            28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

            29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

            30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

            31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

            32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

            33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

            34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

            35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

            36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

            37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

            38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

            39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

            40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

            41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

            42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

            43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

            44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

            45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

            46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

            47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

            48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

            49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

            50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

            51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

            52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

            53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

            54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

            Nucleic Acids Research 2013 11

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

            56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

            57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

            58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

            59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

            60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

            61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

            62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

            63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

            functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

            64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

            65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

            66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

            67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

            68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

            69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

            70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

            71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

            72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

            73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

            12 Nucleic Acids Research 2013

            at D H

            Hill L

            ibrary - Acquis S on N

            ovember 27 2013

            httpnaroxfordjournalsorgD

            ownloaded from

            SUPPLEMENTARY INFORMATION

            For manuscript

            TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

            Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

            1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

            Germany

            This supplementary information contains

            Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

            SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

            2

            2322

            412aNbs

            N

            as πσ ++

            = Equation 1

            where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

            are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

            TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

            [ ]summinus

            =++ minus+minus

            minus=∆

            nN

            iiniini yyxx

            nNtnMSD

            1

            22 )()(1)(

            For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

            lang1199092rang(119905) = 1198711199092

            6minus 161198711199092

            1205874sum 1

            1198994119890119909119901 minus 1

            2 (119899120587120590119909

            119871119909)2119905infin

            119899=1(119900119889119889) Equation 4

            where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

            119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

            where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

            Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

            Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

            1198631119888119886119897 = 119870119861 119879

            61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

            where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

            119896 = 2119863(119897119887119901)2 Equation 7

            where 119896 is the stepping rate Stepss

            The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

            119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

            119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

            ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

            120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

            119870119899119900119899119905119890119897 Equation 11

            where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

            120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

            120591119899119900119899119905119890119897 Equation 12

            where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

            regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

            programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

            Mln2

            the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

            Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

            Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

            Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

            1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

            The scale bar is 1 μm

            2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

            1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

            The scale bar is 1 μm

            3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

            on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

            scale bar is 1 μm

            4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

            TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

            min and 16 s) The scale bar is 1 μm

            5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

            on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

            bar is 1 μm

            SUPPLEMENTARY REFERENCES

            65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

            vivo stoichiometry of shelterin components The Journal of biological chemistry

            285 1457-1467

            66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

            localization analysis for individual fluorescent probes Biophysical journal 82

            2775-2783

            67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

            of the TRFH dimerization domain of the human telomeric proteins TRF1 and

            TRF2 Molecular cell 8 351-361

            68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

            absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

            414

            69 Hughes BD (1995) Random Walks Oxford University Press 1

            70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

            Relationship of lipid rafts to transient confinement zones detected by single

            particle tracking Biophysical journal 82 274-284

            71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

            journal 69 389-398

            72 Destainville N and Salome L (2006) Quantification and correction of systematic

            errors due to detector time-averaging in single-molecule tracking experiments

            Biophysical journal 90 L17-19

            73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

            confinement and jumps in single-molecule membrane trajectories Phys Rev E

            Stat Nonlin Soft Matter Phys 73 011915

            Table S1

            Ionic Strength

            DNA TRF1-QDs () TRF2-QDs ()

            (mM) n I II III IV n I II III IV

            75 λ DNA 179 5 22 168 76 278 374 83 165 378

            125 λ DNA 63 79 16 111 794 171 62 41 152 187

            175 λ DNA 210 30 48 105 547 177 604 34 136 226

            225 λ DNA 291 443 17 62 478 154 617 52 117 214

            125 T270 184 109 6 179 652 374 27 51 166 513

            225 T270 110 45 55 273 627 216 435 32 241 292

            Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

            Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

            Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

            54 kb 16 kb 38 kb 16 kb 38 kb

            54 kb

            ligated non-telomeric DNA

            linearized T270 (54 kb)

            16 kb

            19 kb 16 kb

            (TTAGGG)135 - (TTAGGG)135

            linearized Tel10 (599 kb)

            16 kb 42

            (TTAGGG)10

            DNA substrates for AFM imaging

            DNA substrates for fluorescence imaging

            ligated T270

            599 kb ligated Tel10

            599 kb

            19 kb

            QD Ab

            Ab

            C

            B

            100 nm 100 nm 100 nm

            nm

            QD

            TRF2

            Antibody

            A

            20 nm

            Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

            nm

            Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

            D

            B A

            C

            A B

            C

            Ionic Strength

            TRF1-QDs TRF2-QDs

            (mM) n n

            75 84 27 166 97

            125 87 33 142 95

            175 116 30 154 95

            225 135 15 158 89

            D

            Percentages of motile TRF1- and TRF2-QDs on λ DNA

            Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

            A

            C

            D 2

            microm

            10 s

            TRF2 + λ DNA

            TRF2 + λ DNA

            B

            2 microm

            10 s

            TRF1 + λ DNA

            TRF1 + λ DNA

            Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

            10 s

            Complexes TRF1-QDs TRF2-QDs

            Total number 141 43

            Dual color () 19 79

            Red () 41 10

            Green () 39 11

            E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

            A

            Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

            225 mM ionic strength

            C D

            B

            3 min 0 min

            0 min 63 s

            A

            B

            TRF2 + Tel10 DNA

            TRF2 + Tel10 DNA

            Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

            125 mM 225 mM

            A

            C D 125 mM 225 mM

            Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

            A B

            C D

            88 plusmn 33 times10-2 microm2sec (994)

            A

            B 027 plusmn 013 times10-2 microm2sec (1000)

            027 plusmn 009 times10-2 microm2sec (1000)

            48 plusmn 13 times10-2 microm2sec (433)

            Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

            Unlimited DNA length

            Unlimited DNA length

            16 kb DNA length

            16 kb DNA length

            B

            Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

            0 min 4 min

            TRF2 + T270 DNA

            2 min 0 min

            TRF1

            TRF1 + T270 DNA A

            0 min

            TRF2 + T270 DNA

            4 min

            4 min 0 min

            5 min

            TRF2 + T270 DNA

            B

            0 min 2 min

            TRF2ΔB + λ DNA

            D

            C

            0 min 3 min

            TRF2ΔB + T270 DNA

            Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

            A

            • Lin and countryman et all 2013 NAR
            • Lin et al Supplement ALL 10-10-13
              • Lin et al Supplementary Information text 10-10-13
              • Lin et al supplementary figure 10-10-13
                • Slide Number 1
                • Slide Number 2
                • Slide Number 4
                • Slide Number 5
                • Slide Number 6
                • Slide Number 7
                • Slide Number 8
                • Slide Number 9
                • Slide Number 10
                • Slide Number 11
                • Slide Number 12

              telomeric sequences (Figure 4 and Supplementary FigureS8) To ensure that this confinement would not artificiallyreduce the apparent diffusion constant we simulated 1Ddiffusion of proteins on a linear DNA lattice of unlimitedlength versus a 16 kb total length which mimics the (TTAGGG)270 region (Supplementary Text) These simulationsrevealed that confinement within 16 kb DNA does notsignificantly reduce the observed diffusion constant atthe (TTAGGG)270 region (Supplementary Figure S9) Inaddition camera-based time-averaging was not a majorcontributor to the observed slower diffusion constants atthe telomeric region under these experimental conditions(Supplementary Data) An alternative fitting method to

              simultaneously determine the diffusion constant andconfined DNA length also provided similar results(Supplementary Data) (55) Furthermore the diffusionconstants of TRF2-QDs (031plusmn0003 102mm2sn=37) on DNA tightropes formed under a 12 slowerflow rate (25 mlmin) are not significantly different fromthose on DNA tightropes stretched at a higher flow rate(300mlmin) (Table 2) Under this condition DNA tight-ropes were under less tension with final extension to only88 of DNA contour length (Supplementary FigureS6D) These results suggest that under these conditionsdiffusion constants of TRF2 do not vary significantly withthe amount of tension on dsDNA tightropes

              Figure 4 TRF1 and TRF2 show different diffusional properties over telomeric region versus nontelomeric regions (AndashC) Kymographical analysis ofdual color (655 and 565 nm)-labeled TRF1 (A 125mM ionic strength) and TRF2 (B125 and C225mM ionic strengths) on the ligated T270 DNAThe panel left to the vertical white line shows a schematic drawing of the ligated T270 substrate with telomeric (purple) and nontelomeric sequences(blue) and a fluorescence image of the DNA with proteinndashQDs The horizontal white lines indicate the estimated center of the telomeric region basedon the spacing between adjacent QDs The white arrows in (C) indicate TRF2 diffusing between two adjacent telomeric sequences (D) The diffusionrange distributions of TRF1-(top n=29) and TRF2-QDs (bottom n=28) on the ligated T270 substrate (blue bars) and TRF2-QDs on thenontelomeric DNA (bottom white bars n=77) Diffusion ranges below and beyond 850 nm are categorized into short (telomeric) and long range(nontelomeric) respectively The lines in the top and bottom panels of (D) are single and double Gaussian fits to the data respectively which haveR2 of 090 and 096 respectively

              Table 2 Summary of the diffusion constant and lifetime of TRF1- and TRF2-QDs on the ligated T270 DNA substrates

              DNA Ionicstrength(mM)

              TRF1 TRF2

              Telomeric Nontelomeric Telomeric Nontelomeric

              D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s) D 102mm2s Lifetime (s)

              Tel270 125 015plusmn002 (22) 173plusmn02 (15) 10plusmn02 (8) 59plusmn02 (6) 027plusmn004 (22) 148plusmn02 (105) 30plusmn05 (8) 44plusmn01 (53)Tel270 225 022plusmn004 (21) 92plusmn02 (9) 18plusmn07 (5) 57plusmn03 (2) 029plusmn004 (34) 103plusmn06 (115) 99plusmn30 (7) 38plusmn03 (41)Tel10 125 ndash 67plusmn04 (50) ndash 33plusmn02 (51)

              Proteins were labeled with equal molar amount of red (655 nm) and green (565 nm) QDs The numbers in the parentheses indicate the total numberof complexes analyzed Lifetime was measured for complexes showing both protein binding and release events within the video frame (Type IFigure 2D) Data are presented as meanplusmnstandard error

              Nucleic Acids Research 2013 7

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

              The basic domain is essential for the 1D search by TRF2

              The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

              three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

              DISCUSSION

              TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

              TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

              Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

              8 Nucleic Acids Research 2013

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

              Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

              We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

              behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

              4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

              TRF1 and TRF2 strike a balance between search andspecificity

              TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

              series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

              Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

              Nucleic Acids Research 2013 9

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

              AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

              TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

              tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

              a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

              SUPPLEMENTARY DATA

              Supplementary Data are available at NAR Onlineincluding [65ndash73]

              ACKNOWLEDGEMENTS

              The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

              FUNDING

              The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

              Conflict of interest statement None declared

              REFERENCES

              1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

              2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

              3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

              4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

              5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

              6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

              7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

              8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

              9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

              10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

              11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

              12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

              10 Nucleic Acids Research 2013

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

              14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

              15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

              16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

              17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

              18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

              19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

              20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

              21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

              22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

              23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

              24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

              25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

              26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

              27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

              28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

              29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

              30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

              31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

              32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

              33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

              34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

              35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

              36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

              37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

              38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

              39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

              40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

              41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

              42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

              43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

              44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

              45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

              46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

              47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

              48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

              49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

              50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

              51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

              52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

              53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

              54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

              Nucleic Acids Research 2013 11

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

              56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

              57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

              58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

              59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

              60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

              61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

              62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

              63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

              functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

              64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

              65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

              66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

              67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

              68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

              69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

              70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

              71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

              72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

              73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

              12 Nucleic Acids Research 2013

              at D H

              Hill L

              ibrary - Acquis S on N

              ovember 27 2013

              httpnaroxfordjournalsorgD

              ownloaded from

              SUPPLEMENTARY INFORMATION

              For manuscript

              TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

              Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

              1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

              Germany

              This supplementary information contains

              Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

              SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

              2

              2322

              412aNbs

              N

              as πσ ++

              = Equation 1

              where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

              are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

              TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

              [ ]summinus

              =++ minus+minus

              minus=∆

              nN

              iiniini yyxx

              nNtnMSD

              1

              22 )()(1)(

              For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

              lang1199092rang(119905) = 1198711199092

              6minus 161198711199092

              1205874sum 1

              1198994119890119909119901 minus 1

              2 (119899120587120590119909

              119871119909)2119905infin

              119899=1(119900119889119889) Equation 4

              where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

              119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

              where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

              Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

              Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

              1198631119888119886119897 = 119870119861 119879

              61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

              where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

              119896 = 2119863(119897119887119901)2 Equation 7

              where 119896 is the stepping rate Stepss

              The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

              119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

              119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

              ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

              120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

              119870119899119900119899119905119890119897 Equation 11

              where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

              120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

              120591119899119900119899119905119890119897 Equation 12

              where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

              regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

              programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

              Mln2

              the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

              Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

              Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

              Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

              1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

              The scale bar is 1 μm

              2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

              1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

              The scale bar is 1 μm

              3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

              on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

              scale bar is 1 μm

              4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

              TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

              min and 16 s) The scale bar is 1 μm

              5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

              on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

              bar is 1 μm

              SUPPLEMENTARY REFERENCES

              65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

              vivo stoichiometry of shelterin components The Journal of biological chemistry

              285 1457-1467

              66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

              localization analysis for individual fluorescent probes Biophysical journal 82

              2775-2783

              67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

              of the TRFH dimerization domain of the human telomeric proteins TRF1 and

              TRF2 Molecular cell 8 351-361

              68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

              absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

              414

              69 Hughes BD (1995) Random Walks Oxford University Press 1

              70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

              Relationship of lipid rafts to transient confinement zones detected by single

              particle tracking Biophysical journal 82 274-284

              71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

              journal 69 389-398

              72 Destainville N and Salome L (2006) Quantification and correction of systematic

              errors due to detector time-averaging in single-molecule tracking experiments

              Biophysical journal 90 L17-19

              73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

              confinement and jumps in single-molecule membrane trajectories Phys Rev E

              Stat Nonlin Soft Matter Phys 73 011915

              Table S1

              Ionic Strength

              DNA TRF1-QDs () TRF2-QDs ()

              (mM) n I II III IV n I II III IV

              75 λ DNA 179 5 22 168 76 278 374 83 165 378

              125 λ DNA 63 79 16 111 794 171 62 41 152 187

              175 λ DNA 210 30 48 105 547 177 604 34 136 226

              225 λ DNA 291 443 17 62 478 154 617 52 117 214

              125 T270 184 109 6 179 652 374 27 51 166 513

              225 T270 110 45 55 273 627 216 435 32 241 292

              Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

              Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

              Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

              54 kb 16 kb 38 kb 16 kb 38 kb

              54 kb

              ligated non-telomeric DNA

              linearized T270 (54 kb)

              16 kb

              19 kb 16 kb

              (TTAGGG)135 - (TTAGGG)135

              linearized Tel10 (599 kb)

              16 kb 42

              (TTAGGG)10

              DNA substrates for AFM imaging

              DNA substrates for fluorescence imaging

              ligated T270

              599 kb ligated Tel10

              599 kb

              19 kb

              QD Ab

              Ab

              C

              B

              100 nm 100 nm 100 nm

              nm

              QD

              TRF2

              Antibody

              A

              20 nm

              Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

              nm

              Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

              D

              B A

              C

              A B

              C

              Ionic Strength

              TRF1-QDs TRF2-QDs

              (mM) n n

              75 84 27 166 97

              125 87 33 142 95

              175 116 30 154 95

              225 135 15 158 89

              D

              Percentages of motile TRF1- and TRF2-QDs on λ DNA

              Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

              A

              C

              D 2

              microm

              10 s

              TRF2 + λ DNA

              TRF2 + λ DNA

              B

              2 microm

              10 s

              TRF1 + λ DNA

              TRF1 + λ DNA

              Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

              10 s

              Complexes TRF1-QDs TRF2-QDs

              Total number 141 43

              Dual color () 19 79

              Red () 41 10

              Green () 39 11

              E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

              A

              Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

              225 mM ionic strength

              C D

              B

              3 min 0 min

              0 min 63 s

              A

              B

              TRF2 + Tel10 DNA

              TRF2 + Tel10 DNA

              Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

              125 mM 225 mM

              A

              C D 125 mM 225 mM

              Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

              A B

              C D

              88 plusmn 33 times10-2 microm2sec (994)

              A

              B 027 plusmn 013 times10-2 microm2sec (1000)

              027 plusmn 009 times10-2 microm2sec (1000)

              48 plusmn 13 times10-2 microm2sec (433)

              Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

              Unlimited DNA length

              Unlimited DNA length

              16 kb DNA length

              16 kb DNA length

              B

              Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

              0 min 4 min

              TRF2 + T270 DNA

              2 min 0 min

              TRF1

              TRF1 + T270 DNA A

              0 min

              TRF2 + T270 DNA

              4 min

              4 min 0 min

              5 min

              TRF2 + T270 DNA

              B

              0 min 2 min

              TRF2ΔB + λ DNA

              D

              C

              0 min 3 min

              TRF2ΔB + T270 DNA

              Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

              A

              • Lin and countryman et all 2013 NAR
              • Lin et al Supplement ALL 10-10-13
                • Lin et al Supplementary Information text 10-10-13
                • Lin et al supplementary figure 10-10-13
                  • Slide Number 1
                  • Slide Number 2
                  • Slide Number 4
                  • Slide Number 5
                  • Slide Number 6
                  • Slide Number 7
                  • Slide Number 8
                  • Slide Number 9
                  • Slide Number 10
                  • Slide Number 11
                  • Slide Number 12

                We observed that TRF1 and TRF2 can directly dissoci-ate from telomeric regions or through nontelomericregions (Supplementary Figure S10) Overall we foundthat the relative proportions of Type I (protein bindsand releases) and Type IV (protein is bound from thebeginning to end of the movie) proteinndashDNA interactionsobserved during the experimental time course dependedon the DNA substrate (Supplementary Table S1) ForTRF1 on T270 DNA the vast majority of moleculeswere Type IV indicating a considerably longer attach-ment Consistent with this result the average lifetime ofType I TRF1 bound to the telomeric sequences on T270DNA was 31-fold longer than that for DNA (92 sversus 03 s 225mM ionic strength Tables 1 and 2)TRF2 behaved quite differently showing a lesspronounced difference between the proportions of TypeI and Type IV complexes on DNA and T270 DNAFurthermore the attached lifetimes for Type I TRF2complexes was only 3-fold longer at the telomericregions on T270 DNA compared with DNA (103 sversus 34 s Tables 1 and 2) It is worth noting that thelifetimes of TRF proteins on DNA are longer than the QDblinking rate (56) ruling out artifacts from QD blinkingin the lifetime measurement In summary compared withbinding to nontelomeric DNA both TRF1 and TRF2possess distinctly slower detachment and diffusionaldynamics on the telomeric DNA

                The basic domain is essential for the 1D search by TRF2

                The basic domain at the N-terminus of TRF2 permits itsbinding to model replication forks and four-way junctionsindependent of telomere sequences (57) In addition theabsence of this domain leads to a diminished ability ofTRF2 to localize to model telomere ends and to facilitateT-loop formation (57) We created and imaged a basicdomain deletion mutant of TRF2 (TRF2B) on DNAand the ligated T270 (Supplementary Figure S11)Compared with full-length TRF2 TRF2B-QDs havehigher specificity for the telomeric sequences on T270DNA substrate and lower affinity to DNA ends (compareSupplementary Figures S11A and S3B) Furthermorerelative to the full-length TRF2 the fraction of motileproteinndashDNA complexes decreased by 15-fold forTRF2B (Supplementary Figure S11 legend)Interestingly the diffusion constant (91plusmn18102 mm2s) and a-factor (093plusmn004) of TRF2B on DNA were not significantly different from those of full-length TRF2 (Table 1) However the percentage ofcomplexes undergoing long-range diffusion (10 at125mM ionic strength) was significantly lower (P=001)than for full-length protein (27) at the same ionicstrength (Supplementary Figure S11D) On T270 DNAmajority of motile TRF2B (90) was found with adiffusing range consistent with length of the telomericregion on T270 DNA suggesting that TRF2B directlyassociates with telomeric DNA from solution and not bydiffusion from a nontelomeric region Since the frequencyof TRF2B DNA binding was lower than the full-lengthprotein (11 versus 38 moleculesbead pair) it was notpossible to restrict the analysis to those tightropes with

                three adjacent bound molecules Therefore we treated allshort range diffusion (lt850 nm) by TRF2B on theligated T270 as diffusion over the telomeric region Thedynamics of TRF2B over the (TTAGGG)270 regionwere similar to those of full-length TRF2 with a similardiffusion range (047plusmn003 mm Supplementary FigureS11D) and diffusion constant (027plusmn001 102 mm2sat 125mM and 026plusmn001 102mm2s at 225mM)These observations suggest that the basic domain ofTRF2 normally facilitates its 1D search on nontelomericDNA The reduced degree of TRF2 localization to thetelomeric region due to deletion of the basic domain dem-onstrates the importance of 1D diffusion in the TRF2 telo-meric target site search (Supplementary Figure S11)

                DISCUSSION

                TRF1 and TRF2 are the only scaffolding shelterinproteins that bind directly to duplex telomeric DNAThe results presented here from single-molecule imagingof TRF1 and TRF2 dynamics on telomeric and nontelo-meric DNA provide for the first time a fundamentalunderstanding of the mechanisms that drive the dynamicsof shelterin assemblydisassembly at telomeres

                TRF2 performs 1D searching more effectively than TRF1to find telomeric sequences

                Rotational tracking along DNA during which a proteinfollows a helical track along the DNA to maintain optimalcontact has been inferred for several DNA-bindingproteins (58) The measured diffusion constants forTRF1 and TRF2 obtained using the DNA tightropeassay were consistent with rotational tracking of theDNA helix (Table 1 and Supplementary Text) althoughslightly higher than the predicted upper limit for thismotion (21 102 mm2s Supplementary Text) This dis-crepancy could be due to the flexible linkage mediated bythe His-tag and BTtris-NTA between TRF proteins andQDs (59) The measured diffusion constants togetherwith the lack of observed barrier bypass events in dualcolor experiments (Supplementary Figure S5)demonstrated that both TRF1 and TRF2 track theDNA helix to maintain optimum contact between theirDNA-binding surfaces and the DNA (Figure 5A)However the attached lifetime of Type I TRF1 at thenontelomeric region was 10-fold shorter than that ofTRF2 (03 s versus 34 s at 225mM Table 1) Theseresults are consistent with a significantly lower percentof TRF1 molecules exhibiting long range diffusioncompared with TRF2 (Figure 4) This differencebetween TRF1 and TRF2 is partly due to the sequencesat the N-termini of TRF proteins (Figure 1A) For TRF2this region contains a basic domain the deletion of which(TRF2B) led to a clear reduction in the percentage ofmotile protein complexes on DNA Importantly it wasobserved that 90 of TRF2B molecules underwentshort-range diffusion consistent with the length of thetelomeric regions (Supplementary Figure S11) Thisresult suggests that the majority of the TRF2B mol-ecules found the telomeric region directly from solution

                8 Nucleic Acids Research 2013

                at D H

                Hill L

                ibrary - Acquis S on N

                ovember 27 2013

                httpnaroxfordjournalsorgD

                ownloaded from

                forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

                Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

                We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

                behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

                4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

                TRF1 and TRF2 strike a balance between search andspecificity

                TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

                series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

                Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

                Nucleic Acids Research 2013 9

                at D H

                Hill L

                ibrary - Acquis S on N

                ovember 27 2013

                httpnaroxfordjournalsorgD

                ownloaded from

                energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

                AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

                TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

                tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

                a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

                SUPPLEMENTARY DATA

                Supplementary Data are available at NAR Onlineincluding [65ndash73]

                ACKNOWLEDGEMENTS

                The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

                FUNDING

                The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

                Conflict of interest statement None declared

                REFERENCES

                1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

                2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

                3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

                4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

                5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

                6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

                7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

                8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

                9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

                10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

                11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

                12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

                10 Nucleic Acids Research 2013

                at D H

                Hill L

                ibrary - Acquis S on N

                ovember 27 2013

                httpnaroxfordjournalsorgD

                ownloaded from

                13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

                14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

                15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

                16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

                17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

                18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

                19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

                20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

                21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

                22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

                23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

                24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

                25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

                26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

                27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

                28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

                29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

                30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

                31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

                32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

                33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

                34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

                35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

                36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

                37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

                38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

                39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

                40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

                41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

                42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

                43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

                44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

                45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

                46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

                47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

                48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

                49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

                50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

                51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

                52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

                53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

                54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

                Nucleic Acids Research 2013 11

                at D H

                Hill L

                ibrary - Acquis S on N

                ovember 27 2013

                httpnaroxfordjournalsorgD

                ownloaded from

                55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

                56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

                57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

                58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

                59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

                60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

                61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

                62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

                63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

                functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

                64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

                65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

                66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

                67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

                68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

                69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

                70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

                71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

                72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

                73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

                12 Nucleic Acids Research 2013

                at D H

                Hill L

                ibrary - Acquis S on N

                ovember 27 2013

                httpnaroxfordjournalsorgD

                ownloaded from

                SUPPLEMENTARY INFORMATION

                For manuscript

                TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                Germany

                This supplementary information contains

                Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                2

                2322

                412aNbs

                N

                as πσ ++

                = Equation 1

                where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                [ ]summinus

                =++ minus+minus

                minus=∆

                nN

                iiniini yyxx

                nNtnMSD

                1

                22 )()(1)(

                For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                lang1199092rang(119905) = 1198711199092

                6minus 161198711199092

                1205874sum 1

                1198994119890119909119901 minus 1

                2 (119899120587120590119909

                119871119909)2119905infin

                119899=1(119900119889119889) Equation 4

                where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                1198631119888119886119897 = 119870119861 119879

                61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                119896 = 2119863(119897119887119901)2 Equation 7

                where 119896 is the stepping rate Stepss

                The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                119870119899119900119899119905119890119897 Equation 11

                where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                120591119899119900119899119905119890119897 Equation 12

                where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                Mln2

                the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                The scale bar is 1 μm

                2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                The scale bar is 1 μm

                3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                scale bar is 1 μm

                4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                min and 16 s) The scale bar is 1 μm

                5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                bar is 1 μm

                SUPPLEMENTARY REFERENCES

                65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                vivo stoichiometry of shelterin components The Journal of biological chemistry

                285 1457-1467

                66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                localization analysis for individual fluorescent probes Biophysical journal 82

                2775-2783

                67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                TRF2 Molecular cell 8 351-361

                68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                414

                69 Hughes BD (1995) Random Walks Oxford University Press 1

                70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                Relationship of lipid rafts to transient confinement zones detected by single

                particle tracking Biophysical journal 82 274-284

                71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                journal 69 389-398

                72 Destainville N and Salome L (2006) Quantification and correction of systematic

                errors due to detector time-averaging in single-molecule tracking experiments

                Biophysical journal 90 L17-19

                73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                confinement and jumps in single-molecule membrane trajectories Phys Rev E

                Stat Nonlin Soft Matter Phys 73 011915

                Table S1

                Ionic Strength

                DNA TRF1-QDs () TRF2-QDs ()

                (mM) n I II III IV n I II III IV

                75 λ DNA 179 5 22 168 76 278 374 83 165 378

                125 λ DNA 63 79 16 111 794 171 62 41 152 187

                175 λ DNA 210 30 48 105 547 177 604 34 136 226

                225 λ DNA 291 443 17 62 478 154 617 52 117 214

                125 T270 184 109 6 179 652 374 27 51 166 513

                225 T270 110 45 55 273 627 216 435 32 241 292

                Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                54 kb 16 kb 38 kb 16 kb 38 kb

                54 kb

                ligated non-telomeric DNA

                linearized T270 (54 kb)

                16 kb

                19 kb 16 kb

                (TTAGGG)135 - (TTAGGG)135

                linearized Tel10 (599 kb)

                16 kb 42

                (TTAGGG)10

                DNA substrates for AFM imaging

                DNA substrates for fluorescence imaging

                ligated T270

                599 kb ligated Tel10

                599 kb

                19 kb

                QD Ab

                Ab

                C

                B

                100 nm 100 nm 100 nm

                nm

                QD

                TRF2

                Antibody

                A

                20 nm

                Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                nm

                Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                D

                B A

                C

                A B

                C

                Ionic Strength

                TRF1-QDs TRF2-QDs

                (mM) n n

                75 84 27 166 97

                125 87 33 142 95

                175 116 30 154 95

                225 135 15 158 89

                D

                Percentages of motile TRF1- and TRF2-QDs on λ DNA

                Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                A

                C

                D 2

                microm

                10 s

                TRF2 + λ DNA

                TRF2 + λ DNA

                B

                2 microm

                10 s

                TRF1 + λ DNA

                TRF1 + λ DNA

                Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                10 s

                Complexes TRF1-QDs TRF2-QDs

                Total number 141 43

                Dual color () 19 79

                Red () 41 10

                Green () 39 11

                E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                A

                Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                225 mM ionic strength

                C D

                B

                3 min 0 min

                0 min 63 s

                A

                B

                TRF2 + Tel10 DNA

                TRF2 + Tel10 DNA

                Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                125 mM 225 mM

                A

                C D 125 mM 225 mM

                Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                A B

                C D

                88 plusmn 33 times10-2 microm2sec (994)

                A

                B 027 plusmn 013 times10-2 microm2sec (1000)

                027 plusmn 009 times10-2 microm2sec (1000)

                48 plusmn 13 times10-2 microm2sec (433)

                Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                Unlimited DNA length

                Unlimited DNA length

                16 kb DNA length

                16 kb DNA length

                B

                Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                0 min 4 min

                TRF2 + T270 DNA

                2 min 0 min

                TRF1

                TRF1 + T270 DNA A

                0 min

                TRF2 + T270 DNA

                4 min

                4 min 0 min

                5 min

                TRF2 + T270 DNA

                B

                0 min 2 min

                TRF2ΔB + λ DNA

                D

                C

                0 min 3 min

                TRF2ΔB + T270 DNA

                Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                A

                • Lin and countryman et all 2013 NAR
                • Lin et al Supplement ALL 10-10-13
                  • Lin et al Supplementary Information text 10-10-13
                  • Lin et al supplementary figure 10-10-13
                    • Slide Number 1
                    • Slide Number 2
                    • Slide Number 4
                    • Slide Number 5
                    • Slide Number 6
                    • Slide Number 7
                    • Slide Number 8
                    • Slide Number 9
                    • Slide Number 10
                    • Slide Number 11
                    • Slide Number 12

                  forgoing the 1D component of the search (Figure 5A)These results support the notion that domain B facilitatesthe association of TRF2 to nonspecific DNA and thisresults in sliding subsequently However the diffusionconstant and a factor of TRF2B were not significantlydifferent from the full-length protein (Table 1) We specu-late that TRF2B containing the Myb-type domain hasweak DNA-binding affinity for nontelomeric DNA Onnontelomeric DNA the DNA-binding energy land-scapes are similar for full-length TRF2 and TRF2Bleading to similar diffusion constants However it isunclear whether in full-length TRF2 nonspecific DNAbinding is solely dependent of domain B or combinationof this domain and the Myb domain TRF1 behaved simi-larly to TRF2B perhaps as a consequence of alsolacking the basic domain Therefore unlike TRF1TRF2 can bind to nontelomeric sequences and use a 1Dsearch to more efficiently locate telomeric DNA

                  Comparing the 1D diffusion of TRF1 and TRF2 onnontelomeric and telomeric DNA

                  We found that in general TRF2 slides faster than TRF1at nontelomeric sequences (Supplementary Figure S4Band Table 1) The diffusive exponent was lt1 only forTRF1 at lower ionic strengths consistent with subdiffu-sive motion or pausing during diffusion (SupplementaryFigure S4C and Table 1) Together these observationsindicate that TRF2rsquos diffusion is consistent with thecanonical description of sliding However TRF1rsquos

                  behavior changed with salt in a manner that was incon-sistent with a solely electrostatic-mediated proteinndashDNAinteraction (60) and suggesting a possible conformationalrearrangement induced by salt at the DNA-bindinginterface This rearrangement could lead to obstacles todiffusion andor traps within the binding energy landscapeor escape time (53)The Myb type DNA-binding domain of TRF2 has a

                  4-fold weaker DNA-binding affinity than the Mybdomain in TRF1 (equilibrium dissociation constants Kd750 versus 200 nM respectively) (61) The diffusionconstant of TRF1 was 2-fold slower than that ofTRF2 within telomeric repeats (125mM ionic strengthTable 2) This result is equivalent to 06 kBT increasein the roughness of the DNA-binding landscape or 2-fold change in affinity While these results are consistentwith the stronger binding to the telomeric sequences byTRF1 Myb domain other domains on TRF proteinscould also indirectly influence the DNA-bindingdynamics of these two proteins over the telomericregions Furthermore the difference in the dynamics ofthe TRF proteins between telomeric DNA and nontelo-meric DNA is due to inherent sequence effects and there-fore likely represents the situation in vivo This is furthersupported by the ionic conditions used in our experimentswhich were chosen to represent those encountered in vivo(Supplementary Text)

                  TRF1 and TRF2 strike a balance between search andspecificity

                  TRF proteins face a unique challenge They must findboth their cognate sites and protein partners to form theshelterin complex and to regulate the functions of amyriad of proteins involved in telomere maintenanceand cell-cycle progression (26) For example TRF1 andTRF2 both bind to TIN2 to form a ternary complex ofTRF1 TRF2 and TIN2 (2728) Importantly TRF2 isa protein hub interacting with several DNA-bindingproteins that play important roles in DNA repair includ-ing WRN Ku70-Ku80 and ERCC1-XPF (26396263)This requires that TRF proteins retain specificity fortheir DNA target site but also the ability to slide withinthe telomeric regions to encounter protein partners toform protein complexesThe binding energy of a protein along DNA contains a

                  series of local energy minima separated by energy barriersProtein sliding on DNA has been modeled as a particlediffusing along a rough potential energy landscape Theroughness of the landscape reduces the diffusion constantfrom the theoretical maximum determined by solution vis-cosity We found that the diffusion of TRF1 and TRF2was 17- to 37-fold slower at telomeric regions comparedwith nontelomeric DNA corresponding to 28ndash36kBT increase in the roughness of the energy landscape(Supplementary Text and Figure 5B) Also the TRF1and TRF2 attached lifetimes within telomeric sequenceswere 31- and 3-fold longer respectively comparedwith those on DNA (225mM ionic strength Table 2)These differences correspond to an increase of 34 kBT(for TRF1) and 11kBT (for TRF2) in relative binding

                  Figure 5 TRF1 and TRF2 strike a balance between target search andspecificity (A) TRF1 and TRF2 can undertake a 1D search on DNAconsistent with rotation-coupled diffusion along the DNA helix Thesmall ovals represent the basic and acidic domains of TRF1 and TRF2The blue and purple lines represent nontelomeric and telomeric DNArespectively TRF1 relies more on 3D search and majority of theTRF2B molecules bind to the telomeric region directly fromsolution forgoing the 1D component of the search (B) The energylandscape along the positions at telomeric and nontelomeric sequencesThe diffusion constant and lifetime measurements are consistent with28ndash36 kBT higher energy barriers to diffusion at the telomericsequences in comparison with nontelomeric sequences (Tables 1 and2) The additional energy barrier at the nontelomeric and telomericjunction represents the activation energy needed for conformationalchangeDNA-binding domain switching on proteins to achievespecific binding

                  Nucleic Acids Research 2013 9

                  at D H

                  Hill L

                  ibrary - Acquis S on N

                  ovember 27 2013

                  httpnaroxfordjournalsorgD

                  ownloaded from

                  energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

                  AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

                  TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

                  tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

                  a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

                  SUPPLEMENTARY DATA

                  Supplementary Data are available at NAR Onlineincluding [65ndash73]

                  ACKNOWLEDGEMENTS

                  The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

                  FUNDING

                  The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

                  Conflict of interest statement None declared

                  REFERENCES

                  1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

                  2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

                  3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

                  4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

                  5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

                  6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

                  7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

                  8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

                  9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

                  10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

                  11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

                  12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

                  10 Nucleic Acids Research 2013

                  at D H

                  Hill L

                  ibrary - Acquis S on N

                  ovember 27 2013

                  httpnaroxfordjournalsorgD

                  ownloaded from

                  13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

                  14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

                  15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

                  16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

                  17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

                  18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

                  19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

                  20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

                  21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

                  22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

                  23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

                  24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

                  25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

                  26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

                  27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

                  28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

                  29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

                  30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

                  31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

                  32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

                  33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

                  34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

                  35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

                  36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

                  37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

                  38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

                  39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

                  40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

                  41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

                  42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

                  43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

                  44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

                  45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

                  46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

                  47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

                  48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

                  49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

                  50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

                  51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

                  52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

                  53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

                  54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

                  Nucleic Acids Research 2013 11

                  at D H

                  Hill L

                  ibrary - Acquis S on N

                  ovember 27 2013

                  httpnaroxfordjournalsorgD

                  ownloaded from

                  55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

                  56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

                  57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

                  58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

                  59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

                  60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

                  61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

                  62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

                  63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

                  functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

                  64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

                  65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

                  66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

                  67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

                  68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

                  69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

                  70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

                  71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

                  72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

                  73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

                  12 Nucleic Acids Research 2013

                  at D H

                  Hill L

                  ibrary - Acquis S on N

                  ovember 27 2013

                  httpnaroxfordjournalsorgD

                  ownloaded from

                  SUPPLEMENTARY INFORMATION

                  For manuscript

                  TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                  Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                  1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                  Germany

                  This supplementary information contains

                  Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                  SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                  2

                  2322

                  412aNbs

                  N

                  as πσ ++

                  = Equation 1

                  where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                  are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                  TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                  [ ]summinus

                  =++ minus+minus

                  minus=∆

                  nN

                  iiniini yyxx

                  nNtnMSD

                  1

                  22 )()(1)(

                  For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                  lang1199092rang(119905) = 1198711199092

                  6minus 161198711199092

                  1205874sum 1

                  1198994119890119909119901 minus 1

                  2 (119899120587120590119909

                  119871119909)2119905infin

                  119899=1(119900119889119889) Equation 4

                  where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                  119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                  where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                  Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                  Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                  1198631119888119886119897 = 119870119861 119879

                  61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                  where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                  119896 = 2119863(119897119887119901)2 Equation 7

                  where 119896 is the stepping rate Stepss

                  The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                  119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                  119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                  ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                  120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                  119870119899119900119899119905119890119897 Equation 11

                  where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                  120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                  120591119899119900119899119905119890119897 Equation 12

                  where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                  regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                  programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                  Mln2

                  the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                  Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                  Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                  Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                  1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                  The scale bar is 1 μm

                  2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                  1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                  The scale bar is 1 μm

                  3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                  on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                  scale bar is 1 μm

                  4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                  TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                  min and 16 s) The scale bar is 1 μm

                  5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                  on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                  bar is 1 μm

                  SUPPLEMENTARY REFERENCES

                  65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                  vivo stoichiometry of shelterin components The Journal of biological chemistry

                  285 1457-1467

                  66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                  localization analysis for individual fluorescent probes Biophysical journal 82

                  2775-2783

                  67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                  of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                  TRF2 Molecular cell 8 351-361

                  68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                  absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                  414

                  69 Hughes BD (1995) Random Walks Oxford University Press 1

                  70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                  Relationship of lipid rafts to transient confinement zones detected by single

                  particle tracking Biophysical journal 82 274-284

                  71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                  journal 69 389-398

                  72 Destainville N and Salome L (2006) Quantification and correction of systematic

                  errors due to detector time-averaging in single-molecule tracking experiments

                  Biophysical journal 90 L17-19

                  73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                  confinement and jumps in single-molecule membrane trajectories Phys Rev E

                  Stat Nonlin Soft Matter Phys 73 011915

                  Table S1

                  Ionic Strength

                  DNA TRF1-QDs () TRF2-QDs ()

                  (mM) n I II III IV n I II III IV

                  75 λ DNA 179 5 22 168 76 278 374 83 165 378

                  125 λ DNA 63 79 16 111 794 171 62 41 152 187

                  175 λ DNA 210 30 48 105 547 177 604 34 136 226

                  225 λ DNA 291 443 17 62 478 154 617 52 117 214

                  125 T270 184 109 6 179 652 374 27 51 166 513

                  225 T270 110 45 55 273 627 216 435 32 241 292

                  Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                  Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                  Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                  54 kb 16 kb 38 kb 16 kb 38 kb

                  54 kb

                  ligated non-telomeric DNA

                  linearized T270 (54 kb)

                  16 kb

                  19 kb 16 kb

                  (TTAGGG)135 - (TTAGGG)135

                  linearized Tel10 (599 kb)

                  16 kb 42

                  (TTAGGG)10

                  DNA substrates for AFM imaging

                  DNA substrates for fluorescence imaging

                  ligated T270

                  599 kb ligated Tel10

                  599 kb

                  19 kb

                  QD Ab

                  Ab

                  C

                  B

                  100 nm 100 nm 100 nm

                  nm

                  QD

                  TRF2

                  Antibody

                  A

                  20 nm

                  Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                  nm

                  Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                  D

                  B A

                  C

                  A B

                  C

                  Ionic Strength

                  TRF1-QDs TRF2-QDs

                  (mM) n n

                  75 84 27 166 97

                  125 87 33 142 95

                  175 116 30 154 95

                  225 135 15 158 89

                  D

                  Percentages of motile TRF1- and TRF2-QDs on λ DNA

                  Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                  A

                  C

                  D 2

                  microm

                  10 s

                  TRF2 + λ DNA

                  TRF2 + λ DNA

                  B

                  2 microm

                  10 s

                  TRF1 + λ DNA

                  TRF1 + λ DNA

                  Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                  10 s

                  Complexes TRF1-QDs TRF2-QDs

                  Total number 141 43

                  Dual color () 19 79

                  Red () 41 10

                  Green () 39 11

                  E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                  A

                  Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                  225 mM ionic strength

                  C D

                  B

                  3 min 0 min

                  0 min 63 s

                  A

                  B

                  TRF2 + Tel10 DNA

                  TRF2 + Tel10 DNA

                  Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                  125 mM 225 mM

                  A

                  C D 125 mM 225 mM

                  Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                  A B

                  C D

                  88 plusmn 33 times10-2 microm2sec (994)

                  A

                  B 027 plusmn 013 times10-2 microm2sec (1000)

                  027 plusmn 009 times10-2 microm2sec (1000)

                  48 plusmn 13 times10-2 microm2sec (433)

                  Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                  Unlimited DNA length

                  Unlimited DNA length

                  16 kb DNA length

                  16 kb DNA length

                  B

                  Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                  0 min 4 min

                  TRF2 + T270 DNA

                  2 min 0 min

                  TRF1

                  TRF1 + T270 DNA A

                  0 min

                  TRF2 + T270 DNA

                  4 min

                  4 min 0 min

                  5 min

                  TRF2 + T270 DNA

                  B

                  0 min 2 min

                  TRF2ΔB + λ DNA

                  D

                  C

                  0 min 3 min

                  TRF2ΔB + T270 DNA

                  Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                  A

                  • Lin and countryman et all 2013 NAR
                  • Lin et al Supplement ALL 10-10-13
                    • Lin et al Supplementary Information text 10-10-13
                    • Lin et al supplementary figure 10-10-13
                      • Slide Number 1
                      • Slide Number 2
                      • Slide Number 4
                      • Slide Number 5
                      • Slide Number 6
                      • Slide Number 7
                      • Slide Number 8
                      • Slide Number 9
                      • Slide Number 10
                      • Slide Number 11
                      • Slide Number 12

                    energy at the telomeric regions (Supplementary Data)Taken together the relative activation energy barriersbased on the diffusion constants and lifetimes are notonly consistent with each other but also close to theestimated minimal roughness of the energy landscape atspecific binding sites (66 kBT) for a genome size of3 109bp (Supplementary Data) (35)Interestingly the percentage of TRF2 arriving at the (TT

                    AGGG)270 region (73 125mM Figure 4D) was lowerthan the simulated equivalent situation assuming TRF2first binds to the nontelomeric spacer (98 n=500)This discrepancy is consistent with an additional activationenergy barrier between telomeric and nontelomericregions likely due to a switch within TRF2 from a non-specific binding mode to a specific recognition mode(Figure 5B) (64) Noticeably for TRF2 this energybarrier was lower at 225mM ionic strength than at125mM since more proteins arrived at telomeric regionsfrom the nontelomeric spacers (Figure 4D and Supple-mentary Figure S6B) consistent with the desolvation ofelectrostatic residues required for DNA bindingIn contrast to the metastable and dynamic nature of the

                    TRF protein binding to telomeric sequences (Figure 5)other systems characterized by single-molecule imagingshow long-lived stable binding to specific sequences Forexample the mismatch repair protein MutSa binds to amismatch (+ADP) with a half-life of 96plusmn15min (36)and the average lifetime of the Type III restriction enzymeEcoP15I on DNA with specific binding sites was 180 s(38) The primary differences between these systems arethe target DNA sites For TRF proteins the target is along repetitive sequence whereas for other systems targetsites consist of much shorter nonrepetitive DNA Wepropose that TRF proteins utilize the combined freeenergy of binding from the association of multiple TRFproteins in the same region to increase binding specificityand stability For example TRF1 and TRF2 linked byTIN2 would increase the total affinity for telomericsequences by summing the interaction energies of TRF1and TRF2 We postulate that in vivo the diffusionalproperties of TRF proteins at the telomeric regionsenable these proteins to search for their protein partnerssuch as another TRFndashTIN2 complex to assemble stableshelterin complexes on telomeric substrates In thisputative model of partner search we expect that longdistance searching is unlikely due to DNA-bound obs-tacles such as nucleosomes and other DNA-bindingproteins Rather 1D diffusion represents a relativelylocal search mechanism which increases the probabilityof partner encounter during the attached period In cellsthe intrinsic dynamics of TRF1 and TRF2 could poten-tially be important for regulating the assembly and disas-sembly of shelterin complexes and switching betweendifferent telomere structures (capped and uncappedstates)In summary using QD-conjugated proteins DNA

                    tightropes embedded with site-specific sequences AFMand fluorescence imaging we reveal that TRF1 andTRF2 use different mechanisms to find telomeric DNAbut share a novel mechanism to search for proteinpartners at telomeres Based on these results we postulate

                    a general mechanism for how multiprotein complexesstrike a balance between achieving specificity and targetsearch in a process we define as lsquotag-team proofreadingrsquoIn this model proteins first form weak transientcomplexes with their cognate DNA sequences and thenrely on the additive energies of binding provided bypartner proteins to generate higher specificity

                    SUPPLEMENTARY DATA

                    Supplementary Data are available at NAR Onlineincluding [65ndash73]

                    ACKNOWLEDGEMENTS

                    The authors would like to thank R Keller and MNazareth at the University of Essex Y Jin at theZhejiang University Drs K Weninger R Riehn SFLim K Daniels at the North Carolina State Universityand D Erie at the University of North Carolina at ChapelHill for providing critical comments and technicalsupport

                    FUNDING

                    The BBSRC [BBI0034601 to NMK] NationalInstitutes of Health [ES0515052 to PLO and4R00ES016758 to HW] Funding for open accesscharges National Institutes of Health [4R00ES016758]

                    Conflict of interest statement None declared

                    REFERENCES

                    1 CechTR (2004) Beginning to understand the end of thechromosome Cell 116 273ndash279

                    2 PalmW and de LangeT (2008) How shelterin protectsmammalian telomeres Annu Rev Genet 42 301ndash334

                    3 drsquoAdda di FagagnaF ReaperPM Clay-FarraceL FieglerHCarrP Von ZglinickiT SaretzkiG CarterNP andJacksonSP (2003) A DNA damage checkpoint response intelomere-initiated senescence Nature 426 194ndash198

                    4 SahinE and DePinhoRA (2012) Axis of ageing telomeres p53and mitochondria Nature reviews Mol Cell Biol 13 397ndash404

                    5 ArmaniosM and BlackburnEH (2012) The telomeresyndromes Nat Rev Genet 13 693ndash704

                    6 WrightWE TesmerVM HuffmanKE LeveneSD andShayJW (1997) Normal human chromosomes have long G-richtelomeric overhangs at one end Genes Dev 11 2801ndash2809

                    7 BlackburnEH (2005) Telomeres and telomerase theirmechanisms of action and the effects of altering their functionsFEBS Lett 579 859ndash862

                    8 NandakumarJ and CechTR (2013) Finding the endrecruitment of telomerase to telomeres Nat Rev Mol Cell Biol14 69ndash82

                    9 HengesbachM AkiyamaBM and StoneMD (2011)Single-molecule analysis of telomerase structure and functionCurr Opin Chem Biol 15 845ndash852

                    10 VerdunRE and KarlsederJ (2007) Replication and protectionof telomeres Nature 447 924ndash931

                    11 SfeirA and de LangeT (2012) Removal of shelterin reveals thetelomere end-protection problem Science 336 593ndash597

                    12 CourtR ChapmanL FairallL and RhodesD (2005) How thehuman telomeric proteins TRF1 and TRF2 recognize telomericDNA a view from high-resolution crystal structures EMBORep 6 39ndash45

                    10 Nucleic Acids Research 2013

                    at D H

                    Hill L

                    ibrary - Acquis S on N

                    ovember 27 2013

                    httpnaroxfordjournalsorgD

                    ownloaded from

                    13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

                    14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

                    15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

                    16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

                    17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

                    18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

                    19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

                    20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

                    21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

                    22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

                    23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

                    24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

                    25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

                    26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

                    27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

                    28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

                    29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

                    30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

                    31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

                    32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

                    33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

                    34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

                    35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

                    36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

                    37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

                    38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

                    39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

                    40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

                    41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

                    42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

                    43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

                    44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

                    45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

                    46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

                    47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

                    48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

                    49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

                    50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

                    51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

                    52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

                    53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

                    54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

                    Nucleic Acids Research 2013 11

                    at D H

                    Hill L

                    ibrary - Acquis S on N

                    ovember 27 2013

                    httpnaroxfordjournalsorgD

                    ownloaded from

                    55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

                    56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

                    57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

                    58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

                    59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

                    60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

                    61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

                    62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

                    63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

                    functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

                    64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

                    65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

                    66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

                    67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

                    68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

                    69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

                    70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

                    71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

                    72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

                    73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

                    12 Nucleic Acids Research 2013

                    at D H

                    Hill L

                    ibrary - Acquis S on N

                    ovember 27 2013

                    httpnaroxfordjournalsorgD

                    ownloaded from

                    SUPPLEMENTARY INFORMATION

                    For manuscript

                    TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                    Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                    1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                    Germany

                    This supplementary information contains

                    Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                    SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                    2

                    2322

                    412aNbs

                    N

                    as πσ ++

                    = Equation 1

                    where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                    are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                    TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                    [ ]summinus

                    =++ minus+minus

                    minus=∆

                    nN

                    iiniini yyxx

                    nNtnMSD

                    1

                    22 )()(1)(

                    For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                    lang1199092rang(119905) = 1198711199092

                    6minus 161198711199092

                    1205874sum 1

                    1198994119890119909119901 minus 1

                    2 (119899120587120590119909

                    119871119909)2119905infin

                    119899=1(119900119889119889) Equation 4

                    where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                    119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                    where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                    Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                    Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                    1198631119888119886119897 = 119870119861 119879

                    61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                    where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                    119896 = 2119863(119897119887119901)2 Equation 7

                    where 119896 is the stepping rate Stepss

                    The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                    119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                    119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                    ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                    120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                    119870119899119900119899119905119890119897 Equation 11

                    where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                    120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                    120591119899119900119899119905119890119897 Equation 12

                    where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                    regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                    programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                    Mln2

                    the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                    Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                    Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                    Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                    1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                    The scale bar is 1 μm

                    2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                    1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                    The scale bar is 1 μm

                    3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                    on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                    scale bar is 1 μm

                    4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                    TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                    min and 16 s) The scale bar is 1 μm

                    5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                    on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                    bar is 1 μm

                    SUPPLEMENTARY REFERENCES

                    65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                    vivo stoichiometry of shelterin components The Journal of biological chemistry

                    285 1457-1467

                    66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                    localization analysis for individual fluorescent probes Biophysical journal 82

                    2775-2783

                    67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                    of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                    TRF2 Molecular cell 8 351-361

                    68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                    absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                    414

                    69 Hughes BD (1995) Random Walks Oxford University Press 1

                    70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                    Relationship of lipid rafts to transient confinement zones detected by single

                    particle tracking Biophysical journal 82 274-284

                    71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                    journal 69 389-398

                    72 Destainville N and Salome L (2006) Quantification and correction of systematic

                    errors due to detector time-averaging in single-molecule tracking experiments

                    Biophysical journal 90 L17-19

                    73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                    confinement and jumps in single-molecule membrane trajectories Phys Rev E

                    Stat Nonlin Soft Matter Phys 73 011915

                    Table S1

                    Ionic Strength

                    DNA TRF1-QDs () TRF2-QDs ()

                    (mM) n I II III IV n I II III IV

                    75 λ DNA 179 5 22 168 76 278 374 83 165 378

                    125 λ DNA 63 79 16 111 794 171 62 41 152 187

                    175 λ DNA 210 30 48 105 547 177 604 34 136 226

                    225 λ DNA 291 443 17 62 478 154 617 52 117 214

                    125 T270 184 109 6 179 652 374 27 51 166 513

                    225 T270 110 45 55 273 627 216 435 32 241 292

                    Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                    Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                    Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                    54 kb 16 kb 38 kb 16 kb 38 kb

                    54 kb

                    ligated non-telomeric DNA

                    linearized T270 (54 kb)

                    16 kb

                    19 kb 16 kb

                    (TTAGGG)135 - (TTAGGG)135

                    linearized Tel10 (599 kb)

                    16 kb 42

                    (TTAGGG)10

                    DNA substrates for AFM imaging

                    DNA substrates for fluorescence imaging

                    ligated T270

                    599 kb ligated Tel10

                    599 kb

                    19 kb

                    QD Ab

                    Ab

                    C

                    B

                    100 nm 100 nm 100 nm

                    nm

                    QD

                    TRF2

                    Antibody

                    A

                    20 nm

                    Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                    nm

                    Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                    D

                    B A

                    C

                    A B

                    C

                    Ionic Strength

                    TRF1-QDs TRF2-QDs

                    (mM) n n

                    75 84 27 166 97

                    125 87 33 142 95

                    175 116 30 154 95

                    225 135 15 158 89

                    D

                    Percentages of motile TRF1- and TRF2-QDs on λ DNA

                    Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                    A

                    C

                    D 2

                    microm

                    10 s

                    TRF2 + λ DNA

                    TRF2 + λ DNA

                    B

                    2 microm

                    10 s

                    TRF1 + λ DNA

                    TRF1 + λ DNA

                    Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                    10 s

                    Complexes TRF1-QDs TRF2-QDs

                    Total number 141 43

                    Dual color () 19 79

                    Red () 41 10

                    Green () 39 11

                    E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                    A

                    Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                    225 mM ionic strength

                    C D

                    B

                    3 min 0 min

                    0 min 63 s

                    A

                    B

                    TRF2 + Tel10 DNA

                    TRF2 + Tel10 DNA

                    Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                    125 mM 225 mM

                    A

                    C D 125 mM 225 mM

                    Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                    A B

                    C D

                    88 plusmn 33 times10-2 microm2sec (994)

                    A

                    B 027 plusmn 013 times10-2 microm2sec (1000)

                    027 plusmn 009 times10-2 microm2sec (1000)

                    48 plusmn 13 times10-2 microm2sec (433)

                    Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                    Unlimited DNA length

                    Unlimited DNA length

                    16 kb DNA length

                    16 kb DNA length

                    B

                    Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                    0 min 4 min

                    TRF2 + T270 DNA

                    2 min 0 min

                    TRF1

                    TRF1 + T270 DNA A

                    0 min

                    TRF2 + T270 DNA

                    4 min

                    4 min 0 min

                    5 min

                    TRF2 + T270 DNA

                    B

                    0 min 2 min

                    TRF2ΔB + λ DNA

                    D

                    C

                    0 min 3 min

                    TRF2ΔB + T270 DNA

                    Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                    A

                    • Lin and countryman et all 2013 NAR
                    • Lin et al Supplement ALL 10-10-13
                      • Lin et al Supplementary Information text 10-10-13
                      • Lin et al supplementary figure 10-10-13
                        • Slide Number 1
                        • Slide Number 2
                        • Slide Number 4
                        • Slide Number 5
                        • Slide Number 6
                        • Slide Number 7
                        • Slide Number 8
                        • Slide Number 9
                        • Slide Number 10
                        • Slide Number 11
                        • Slide Number 12

                      13 WalkerJR and ZhuXD (2012) Post-translational modificationsof TRF1 and TRF2 and their roles in telomere maintenanceMech Ageing Dev 133 421ndash434

                      14 SfeirA KosiyatrakulST HockemeyerD MacRaeSLKarlsederJ SchildkrautCL and de LangeT (2009) Mammaliantelomeres resemble fragile sites and require TRF1 for efficientreplication Cell 138 90ndash103

                      15 SmogorzewskaA van SteenselB BianchiA OelmannSSchaeferMR SchnappG and de LangeT (2000) Control ofhuman telomere length by TRF1 and TRF2 Mol Cell Biol 201659ndash1668

                      16 van SteenselB SmogorzewskaA and de LangeT (1998) TRF2protects human telomeres from end-to-end fusions Cell 92401ndash413

                      17 KarlsederJ BroccoliD DaiY HardyS and de LangeT(1999) p53- and ATM-dependent apoptosis induced by telomereslacking TRF2 Science 283 1321ndash1325

                      18 BianchiA StanselRM FairallL GriffithJD RhodesD andde LangeT (1999) TRF1 binds a bipartite telomeric site withextreme spatial flexibility EMBO J 18 5735ndash5744

                      19 GriffithJ BianchiA and de LangeT (1998) TRF1 promotesparallel pairing of telomeric tracts in vitro J Mol Biol 27879ndash88

                      20 GriffithJD ComeauL RosenfieldS StanselRM BianchiAMossH and de LangeT (1999) Mammalian telomeres end in alarge duplex loop Cell 97 503ndash514

                      21 PouletA PisanoS Faivre-MoskalenkoC PeiB TauranYHaftek-TerreauZ BrunetF Le BihanYV LeduMHMontelF et al (2012) The N-terminal domains of TRF1 andTRF2 regulate their ability to condense telomeric DNANucleic Acids Res 40 2566ndash2576

                      22 MatternKA SwiggersSJ NiggAL LowenbergBHoutsmullerAB and ZijlmansJM (2004) Dynamics of proteinbinding to telomeres in living cells implications for telomerestructure and function Mol Cell Biol 24 5587ndash5594

                      23 SimonetT ZaragosiLE PhilippeC LebrigandKSchoutedenC AugereauA BauwensS YeJ SantagostinoMGiulottoE et al (2011) The human TTAGGG repeat factors 1and 2 bind to a subset of interstitial telomeric sequences andsatellite repeats Cell Res 21 1028ndash1038

                      24 YangD XiongY KimH HeQ LiY ChenR andSongyangZ (2011) Human telomeric proteins occupy selectiveinterstitial sites Cell Res 21 1013ndash1027

                      25 BoscoN and de LangeT (2012) A TRF1-controlled commonfragile site containing interstitial telomeric sequencesChromosoma 121 465ndash474

                      26 de LangeT (2010) Telomere biology and DNA repair enemieswith benefits FEBS Lett 584 3673ndash3674

                      27 HoughtalingBR CuttonaroL ChangW and SmithS (2004)A dynamic molecular link between the telomere length regulatorTRF1 and the chromosome end protector TRF2 Curr Biol 141621ndash1631

                      28 YeJZ DonigianJR van OverbeekM LoayzaD LuoYKrutchinskyAN ChaitBT and de LangeT (2004) TIN2 bindsTRF1 and TRF2 simultaneously and stabilizes the TRF2 complexon telomeres J Biol Chem 279 47264ndash47271

                      29 BergOG WinterRB and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 1Models and theory Biochemistry 20 6929ndash6948

                      30 von HippelPH and BergOG (1989) Facilitated target locationin biological systems J Biol Chem 264 675ndash678

                      31 GormanJ and GreeneEC (2008) Visualizing one-dimensionaldiffusion of proteins along DNA Nat Struct Mol Biol 15768ndash774

                      32 TafviziA MirnyLA and van OijenAM (2011) Dancing onDNA kinetic aspects of search processes on DNAChemphyschem 12 1481ndash1489

                      33 KadNM WangH KennedyGG WarshawDM and VanHoutenB (2010) Collaborative dynamic DNA scanning bynucleotide excision repair proteins investigated by single- moleculeimaging of quantum-dot-labeled proteins Mol Cell 37 702ndash713

                      34 KadNM and Van HoutenB (2012) Dynamics of lesionprocessing by bacterial nucleotide excision repair proteinsProg Mol Biol Transl Sci 110 1ndash24

                      35 SlutskyM and MirnyLA (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

                      36 GormanJ WangF ReddingS PlysAJ FazioT WindSAlaniEE and GreeneEC (2012) Single-molecule imagingreveals target-search mechanisms during DNA mismatch repairProc Natl Acad Sci USA 109 E3074ndashE3083

                      37 LeithJS TafviziA HuangF UspalWE DoylePSFershtAR MirnyLA and van OijenAM (2012) Sequence-dependent sliding kinetics of p53 Proc Natl Acad Sci USA109 16552ndash16557

                      38 SchwarzFW TothJ van AelstK CuiG ClausingSSzczelkunMD and SeidelR (2013) The helicase-like domains oftype III restriction enzymes trigger long-range diffusion alongDNA Science 340 353ndash356

                      39 OpreskoPL von KobbeC LaineJP HarriganJ HicksonIDand BohrVA (2002) Telomere-binding protein TRF2 binds toand stimulates the Werner and Bloom syndrome helicases J BiolChem 277 41110ndash41119

                      40 NoraGJ BuncherNA and OpreskoPL (2010) Telomericprotein TRF2 protects Holliday junctions with telomeric armsfrom displacement by the Werner syndrome helicase NucleicAcids Res 38 3984ndash3998

                      41 HanishJP YanowitzJL and de LangeT (1994) Stringentsequence requirements for the formation of human telomeresProc Natl Acad Sci USA 91 8861ndash8865

                      42 DamerlaRR KnickelbeinKE KepchiaD JacksonAArmitageBA EckertKA and OpreskoPL (2010) Telomericrepeat mutagenicity in human somatic cells is modulated byrepeat orientation and G-quadruplex stability DNA Repair(Amst) 9 1119ndash1129

                      43 LataS ReichelA BrockR TampeR and PiehlerJ (2005)High-affinity adaptors for switchable recognition of histidine-tagged proteins J Am Chem Soc 127 10205ndash10215

                      44 ReichelA SchaibleD Al FuroukhN CohenM SchreiberGand PiehlerJ (2007) Noncovalent site-specific biotinylationof histidine-tagged proteins Anal Chem 79 8590ndash8600

                      45 KochaniakAB HabuchiS LoparoJJ ChangDJCimprichKA WalterJC and van OijenAM (2009)Proliferating cell nuclear antigen uses two distinct modes to movealong DNA J Biol Chem 284 17700ndash17710

                      46 ArnspangEC BrewerJR and LagerholmBC (2012)Multi-color single particle tracking with quantum dots PloS One7 e48521

                      47 DunnAR KadNM NelsonSR WarshawDM andWallaceSS (2011) Single Qdot-labeled glycosylase molecules usea wedge amino acid to probe for lesions while scanning alongDNA Nucleic Acids Res 39 7487ndash7498

                      48 HughesCD WangH GhodkeH SimonsM TowheedAPengY Van HoutenB and KadNM (2013) Real-timesingle-molecule imaging reveals a direct interaction betweenUvrC and UvrB on DNA tightropes Nucleic Acids Res 414901ndash4912

                      49 WangH TessmerI CroteauDL ErieDA and Van HoutenB(2008) Functional characterization and atomic force microscopyof a DNA repair protein conjugated to a quantum dot NanoLett 8 1631ndash1637

                      50 TessmerI KaurP LinJ and WangH (2013) Investigatingbioconjugation by atomic force microscopy J Nanobiotechnol11 25

                      51 Komazin-MeredithG MirchevR GolanDE van OijenAMand CoenDM (2008) Hopping of a processivity factor on DNArevealed by single-molecule assays of diffusion Proc Natl AcadSci USA 105 10721ndash10726

                      52 GormanJ PlysAJ VisnapuuML AlaniE and GreeneEC(2010) Visualizing one-dimensional diffusion of eukaryotic DNArepair factors along a chromatin lattice Nat Struct Mol Biol17 932ndash938

                      53 SaxtonMJ and JacobsonK (1997) Single-particle trackingapplications to membrane dynamics Annu Rev Biophys BiomolStruct 26 373ndash399

                      54 HalfordSE and MarkoJF (2004) How do site-specificDNA-binding proteins find their targets Nucleic Acids Res 323040ndash3052

                      Nucleic Acids Research 2013 11

                      at D H

                      Hill L

                      ibrary - Acquis S on N

                      ovember 27 2013

                      httpnaroxfordjournalsorgD

                      ownloaded from

                      55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

                      56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

                      57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

                      58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

                      59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

                      60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

                      61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

                      62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

                      63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

                      functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

                      64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

                      65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

                      66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

                      67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

                      68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

                      69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

                      70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

                      71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

                      72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

                      73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

                      12 Nucleic Acids Research 2013

                      at D H

                      Hill L

                      ibrary - Acquis S on N

                      ovember 27 2013

                      httpnaroxfordjournalsorgD

                      ownloaded from

                      SUPPLEMENTARY INFORMATION

                      For manuscript

                      TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                      Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                      1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                      Germany

                      This supplementary information contains

                      Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                      SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                      2

                      2322

                      412aNbs

                      N

                      as πσ ++

                      = Equation 1

                      where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                      are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                      TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                      [ ]summinus

                      =++ minus+minus

                      minus=∆

                      nN

                      iiniini yyxx

                      nNtnMSD

                      1

                      22 )()(1)(

                      For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                      lang1199092rang(119905) = 1198711199092

                      6minus 161198711199092

                      1205874sum 1

                      1198994119890119909119901 minus 1

                      2 (119899120587120590119909

                      119871119909)2119905infin

                      119899=1(119900119889119889) Equation 4

                      where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                      119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                      where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                      Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                      Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                      1198631119888119886119897 = 119870119861 119879

                      61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                      where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                      119896 = 2119863(119897119887119901)2 Equation 7

                      where 119896 is the stepping rate Stepss

                      The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                      119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                      119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                      ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                      120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                      119870119899119900119899119905119890119897 Equation 11

                      where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                      120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                      120591119899119900119899119905119890119897 Equation 12

                      where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                      regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                      programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                      Mln2

                      the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                      Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                      Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                      Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                      1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                      The scale bar is 1 μm

                      2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                      1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                      The scale bar is 1 μm

                      3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                      on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                      scale bar is 1 μm

                      4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                      TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                      min and 16 s) The scale bar is 1 μm

                      5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                      on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                      bar is 1 μm

                      SUPPLEMENTARY REFERENCES

                      65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                      vivo stoichiometry of shelterin components The Journal of biological chemistry

                      285 1457-1467

                      66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                      localization analysis for individual fluorescent probes Biophysical journal 82

                      2775-2783

                      67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                      of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                      TRF2 Molecular cell 8 351-361

                      68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                      absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                      414

                      69 Hughes BD (1995) Random Walks Oxford University Press 1

                      70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                      Relationship of lipid rafts to transient confinement zones detected by single

                      particle tracking Biophysical journal 82 274-284

                      71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                      journal 69 389-398

                      72 Destainville N and Salome L (2006) Quantification and correction of systematic

                      errors due to detector time-averaging in single-molecule tracking experiments

                      Biophysical journal 90 L17-19

                      73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                      confinement and jumps in single-molecule membrane trajectories Phys Rev E

                      Stat Nonlin Soft Matter Phys 73 011915

                      Table S1

                      Ionic Strength

                      DNA TRF1-QDs () TRF2-QDs ()

                      (mM) n I II III IV n I II III IV

                      75 λ DNA 179 5 22 168 76 278 374 83 165 378

                      125 λ DNA 63 79 16 111 794 171 62 41 152 187

                      175 λ DNA 210 30 48 105 547 177 604 34 136 226

                      225 λ DNA 291 443 17 62 478 154 617 52 117 214

                      125 T270 184 109 6 179 652 374 27 51 166 513

                      225 T270 110 45 55 273 627 216 435 32 241 292

                      Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                      Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                      Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                      54 kb 16 kb 38 kb 16 kb 38 kb

                      54 kb

                      ligated non-telomeric DNA

                      linearized T270 (54 kb)

                      16 kb

                      19 kb 16 kb

                      (TTAGGG)135 - (TTAGGG)135

                      linearized Tel10 (599 kb)

                      16 kb 42

                      (TTAGGG)10

                      DNA substrates for AFM imaging

                      DNA substrates for fluorescence imaging

                      ligated T270

                      599 kb ligated Tel10

                      599 kb

                      19 kb

                      QD Ab

                      Ab

                      C

                      B

                      100 nm 100 nm 100 nm

                      nm

                      QD

                      TRF2

                      Antibody

                      A

                      20 nm

                      Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                      nm

                      Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                      D

                      B A

                      C

                      A B

                      C

                      Ionic Strength

                      TRF1-QDs TRF2-QDs

                      (mM) n n

                      75 84 27 166 97

                      125 87 33 142 95

                      175 116 30 154 95

                      225 135 15 158 89

                      D

                      Percentages of motile TRF1- and TRF2-QDs on λ DNA

                      Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                      A

                      C

                      D 2

                      microm

                      10 s

                      TRF2 + λ DNA

                      TRF2 + λ DNA

                      B

                      2 microm

                      10 s

                      TRF1 + λ DNA

                      TRF1 + λ DNA

                      Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                      10 s

                      Complexes TRF1-QDs TRF2-QDs

                      Total number 141 43

                      Dual color () 19 79

                      Red () 41 10

                      Green () 39 11

                      E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                      A

                      Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                      225 mM ionic strength

                      C D

                      B

                      3 min 0 min

                      0 min 63 s

                      A

                      B

                      TRF2 + Tel10 DNA

                      TRF2 + Tel10 DNA

                      Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                      125 mM 225 mM

                      A

                      C D 125 mM 225 mM

                      Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                      A B

                      C D

                      88 plusmn 33 times10-2 microm2sec (994)

                      A

                      B 027 plusmn 013 times10-2 microm2sec (1000)

                      027 plusmn 009 times10-2 microm2sec (1000)

                      48 plusmn 13 times10-2 microm2sec (433)

                      Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                      Unlimited DNA length

                      Unlimited DNA length

                      16 kb DNA length

                      16 kb DNA length

                      B

                      Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                      0 min 4 min

                      TRF2 + T270 DNA

                      2 min 0 min

                      TRF1

                      TRF1 + T270 DNA A

                      0 min

                      TRF2 + T270 DNA

                      4 min

                      4 min 0 min

                      5 min

                      TRF2 + T270 DNA

                      B

                      0 min 2 min

                      TRF2ΔB + λ DNA

                      D

                      C

                      0 min 3 min

                      TRF2ΔB + T270 DNA

                      Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                      A

                      • Lin and countryman et all 2013 NAR
                      • Lin et al Supplement ALL 10-10-13
                        • Lin et al Supplementary Information text 10-10-13
                        • Lin et al supplementary figure 10-10-13
                          • Slide Number 1
                          • Slide Number 2
                          • Slide Number 4
                          • Slide Number 5
                          • Slide Number 6
                          • Slide Number 7
                          • Slide Number 8
                          • Slide Number 9
                          • Slide Number 10
                          • Slide Number 11
                          • Slide Number 12

                        55 KusumiA SakoY and YamamotoM (1993) Confined lateraldiffusion of membrane receptors as studied by single particletracking (nanovid microscopy) Effects of calcium-induceddifferentiation in cultured epithelial cells Biophys J 652021ndash2040

                        56 NirmalM DabbousiBO BawendiMG MacklinJJTrautmanJK HarrisTD and brusLE (1996) Fluorescenceintermittency in single cadmium selenide nanocrystals Nature383 802ndash804

                        57 FoucheN CesareAJ WillcoxS OzgurS ComptonSA andGriffithJD (2006) The basic domain of TRF2 directs binding toDNA junctions irrespective of the presence of TTAGGG repeatsJ Biol Chem 281 37486ndash37495

                        58 BlaineyPC LuoG KouSC MangelWF VerdineGLBagchiB and XieXS (2009) Nonspecifically bound proteins spinwhile diffusing along DNA Nat Struct Mol Biol 16 1224ndash1229

                        59 BonnetI and DesbiollesP (2011) The diffusion constant of alabeled protein sliding along DNA Eur Phys J E Soft Matter34 1ndash10

                        60 WinterRB BergOG and von HippelPH (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids 3The Escherichia coli lac repressorndashoperator interaction kineticmeasurements and conclusions Biochemistry 20 6961ndash6977

                        61 HanaokaS NagadoiA and NishimuraY (2005) Comparisonbetween TRF2 and TRF1 of their telomeric DNA-boundstructures and DNA-binding activities Protein Sci 14 119ndash130

                        62 ZhuXD NiedernhoferL KusterB MannMHoeijmakersJH and de LangeT (2003) ERCC1XPF removesthe 30 overhang from uncapped telomeres and represses formationof telomeric DNA-containing double minute chromosomesMol Cell 12 1489ndash1498

                        63 KimH LeeOH XinH ChenLY QinJ ChaeHKLinSY SafariA LiuD and SongyangZ (2009) TRF2

                        functions as a protein hub and regulates telomere maintenanceby recognizing specific peptide motifs Nat Struct Mol Biol 16372ndash379

                        64 MarcovitzA and LevyY (2011) Frustration in protein-DNAbinding influences conformational switching and target searchkinetics Proc Natl Acad Sci USA 108 17957ndash17962

                        65 TakaiKK HooperS BlackwoodS GandhiR andde LangeT (2010) In vivo stoichiometry of shelterin componentsJ Biol Chem 285 1457ndash1467

                        66 ThompsonRE LarsonDR and WebbWW (2002) Precisenanometer localization analysis for individual fluorescent probesBiophys J 82 2775ndash2783

                        67 FairallL ChapmanL MossH de LangeT and RhodesD(2001) Structure of the TRFH dimerization domain of the humantelomeric proteins TRF1 and TRF2 Mol Cell 8 351ndash361

                        68 SchurrJM (1979) The one-dimensional diffusion coefficient ofproteins absorbed on DNA Hydrodynamic considerationsBiophys Chem 9 413ndash414

                        69 HughesBD (1995) Random Walks Vol 1 Oxford UniversityPress USA

                        70 DietrichC YangB FujiwaraT KusumiA and JacobsonK(2002) Relationship of lipid rafts to transient confinement zonesdetected by single particle tracking Biophys J 82 274ndash284

                        71 SaxtonMJ (1995) Single-particle tracking effects of corralsBiophys J 69 389ndash398

                        72 DestainvilleN and SalomeL (2006) Quantification andcorrection of systematic errors due to detector time-averagingin single-molecule tracking experiments Biophys J 90L17ndashL19

                        73 MeilhacN Le GuyaderL SalomeL and DestainvilleN (2006)Detection of confinement and jumps in single-moleculemembrane trajectories Phys Rev E Stat Nonlin Soft MatterPhys 73 011915

                        12 Nucleic Acids Research 2013

                        at D H

                        Hill L

                        ibrary - Acquis S on N

                        ovember 27 2013

                        httpnaroxfordjournalsorgD

                        ownloaded from

                        SUPPLEMENTARY INFORMATION

                        For manuscript

                        TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                        Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                        1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                        Germany

                        This supplementary information contains

                        Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                        SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                        2

                        2322

                        412aNbs

                        N

                        as πσ ++

                        = Equation 1

                        where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                        are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                        TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                        [ ]summinus

                        =++ minus+minus

                        minus=∆

                        nN

                        iiniini yyxx

                        nNtnMSD

                        1

                        22 )()(1)(

                        For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                        lang1199092rang(119905) = 1198711199092

                        6minus 161198711199092

                        1205874sum 1

                        1198994119890119909119901 minus 1

                        2 (119899120587120590119909

                        119871119909)2119905infin

                        119899=1(119900119889119889) Equation 4

                        where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                        119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                        where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                        Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                        Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                        1198631119888119886119897 = 119870119861 119879

                        61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                        where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                        119896 = 2119863(119897119887119901)2 Equation 7

                        where 119896 is the stepping rate Stepss

                        The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                        119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                        119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                        ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                        120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                        119870119899119900119899119905119890119897 Equation 11

                        where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                        120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                        120591119899119900119899119905119890119897 Equation 12

                        where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                        regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                        programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                        Mln2

                        the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                        Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                        Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                        Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                        1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                        The scale bar is 1 μm

                        2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                        1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                        The scale bar is 1 μm

                        3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                        on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                        scale bar is 1 μm

                        4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                        TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                        min and 16 s) The scale bar is 1 μm

                        5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                        on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                        bar is 1 μm

                        SUPPLEMENTARY REFERENCES

                        65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                        vivo stoichiometry of shelterin components The Journal of biological chemistry

                        285 1457-1467

                        66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                        localization analysis for individual fluorescent probes Biophysical journal 82

                        2775-2783

                        67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                        of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                        TRF2 Molecular cell 8 351-361

                        68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                        absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                        414

                        69 Hughes BD (1995) Random Walks Oxford University Press 1

                        70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                        Relationship of lipid rafts to transient confinement zones detected by single

                        particle tracking Biophysical journal 82 274-284

                        71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                        journal 69 389-398

                        72 Destainville N and Salome L (2006) Quantification and correction of systematic

                        errors due to detector time-averaging in single-molecule tracking experiments

                        Biophysical journal 90 L17-19

                        73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                        confinement and jumps in single-molecule membrane trajectories Phys Rev E

                        Stat Nonlin Soft Matter Phys 73 011915

                        Table S1

                        Ionic Strength

                        DNA TRF1-QDs () TRF2-QDs ()

                        (mM) n I II III IV n I II III IV

                        75 λ DNA 179 5 22 168 76 278 374 83 165 378

                        125 λ DNA 63 79 16 111 794 171 62 41 152 187

                        175 λ DNA 210 30 48 105 547 177 604 34 136 226

                        225 λ DNA 291 443 17 62 478 154 617 52 117 214

                        125 T270 184 109 6 179 652 374 27 51 166 513

                        225 T270 110 45 55 273 627 216 435 32 241 292

                        Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                        Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                        Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                        54 kb 16 kb 38 kb 16 kb 38 kb

                        54 kb

                        ligated non-telomeric DNA

                        linearized T270 (54 kb)

                        16 kb

                        19 kb 16 kb

                        (TTAGGG)135 - (TTAGGG)135

                        linearized Tel10 (599 kb)

                        16 kb 42

                        (TTAGGG)10

                        DNA substrates for AFM imaging

                        DNA substrates for fluorescence imaging

                        ligated T270

                        599 kb ligated Tel10

                        599 kb

                        19 kb

                        QD Ab

                        Ab

                        C

                        B

                        100 nm 100 nm 100 nm

                        nm

                        QD

                        TRF2

                        Antibody

                        A

                        20 nm

                        Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                        nm

                        Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                        D

                        B A

                        C

                        A B

                        C

                        Ionic Strength

                        TRF1-QDs TRF2-QDs

                        (mM) n n

                        75 84 27 166 97

                        125 87 33 142 95

                        175 116 30 154 95

                        225 135 15 158 89

                        D

                        Percentages of motile TRF1- and TRF2-QDs on λ DNA

                        Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                        A

                        C

                        D 2

                        microm

                        10 s

                        TRF2 + λ DNA

                        TRF2 + λ DNA

                        B

                        2 microm

                        10 s

                        TRF1 + λ DNA

                        TRF1 + λ DNA

                        Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                        10 s

                        Complexes TRF1-QDs TRF2-QDs

                        Total number 141 43

                        Dual color () 19 79

                        Red () 41 10

                        Green () 39 11

                        E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                        A

                        Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                        225 mM ionic strength

                        C D

                        B

                        3 min 0 min

                        0 min 63 s

                        A

                        B

                        TRF2 + Tel10 DNA

                        TRF2 + Tel10 DNA

                        Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                        125 mM 225 mM

                        A

                        C D 125 mM 225 mM

                        Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                        A B

                        C D

                        88 plusmn 33 times10-2 microm2sec (994)

                        A

                        B 027 plusmn 013 times10-2 microm2sec (1000)

                        027 plusmn 009 times10-2 microm2sec (1000)

                        48 plusmn 13 times10-2 microm2sec (433)

                        Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                        Unlimited DNA length

                        Unlimited DNA length

                        16 kb DNA length

                        16 kb DNA length

                        B

                        Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                        0 min 4 min

                        TRF2 + T270 DNA

                        2 min 0 min

                        TRF1

                        TRF1 + T270 DNA A

                        0 min

                        TRF2 + T270 DNA

                        4 min

                        4 min 0 min

                        5 min

                        TRF2 + T270 DNA

                        B

                        0 min 2 min

                        TRF2ΔB + λ DNA

                        D

                        C

                        0 min 3 min

                        TRF2ΔB + T270 DNA

                        Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                        A

                        • Lin and countryman et all 2013 NAR
                        • Lin et al Supplement ALL 10-10-13
                          • Lin et al Supplementary Information text 10-10-13
                          • Lin et al supplementary figure 10-10-13
                            • Slide Number 1
                            • Slide Number 2
                            • Slide Number 4
                            • Slide Number 5
                            • Slide Number 6
                            • Slide Number 7
                            • Slide Number 8
                            • Slide Number 9
                            • Slide Number 10
                            • Slide Number 11
                            • Slide Number 12

                          SUPPLEMENTARY INFORMATION

                          For manuscript

                          TRF1 and TRF2 use different mechanisms to find telomeric DNA but share a novel mechanism to search for protein partners at telomeres

                          Jiangguo Lin1 Preston Countryman1 Noah Buncher2 Parminder Kaur1 Longjiang E3 Yiyun Zhang4 Greg Gibson5 Changjiang You6 Simon C Watkins5 Jacob Piehler6 Patricia L Opresko2 Neil M Kad7 Hong Wang1

                          1Physics Department 4Industry and System Engineering North Carolina State University Raleigh North Carolina NC 27695 USA 2Department of Environmental and Occupational Health 5Center for Biologic Imaging University of Pittsburgh Graduate School of Public Health Pittsburgh Pennsylvania 15219 USA 3Electric and Computer Engineering University of North Carolina at Charlotte Charlotte North Carolina 28223 USA 6 Division of Biophysics Universitaumlt Osnabruumlck Barbarstrasse 11 49076 Osnabruumlck

                          Germany

                          This supplementary information contains

                          Supplementary Text Supplementary Legends for Movies S1ndash5 Supplementary References Supplementary Table S1 Supplementary Figures S1ndash11

                          SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                          2

                          2322

                          412aNbs

                          N

                          as πσ ++

                          = Equation 1

                          where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                          are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                          TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                          [ ]summinus

                          =++ minus+minus

                          minus=∆

                          nN

                          iiniini yyxx

                          nNtnMSD

                          1

                          22 )()(1)(

                          For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                          lang1199092rang(119905) = 1198711199092

                          6minus 161198711199092

                          1205874sum 1

                          1198994119890119909119901 minus 1

                          2 (119899120587120590119909

                          119871119909)2119905infin

                          119899=1(119900119889119889) Equation 4

                          where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                          119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                          where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                          Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                          Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                          1198631119888119886119897 = 119870119861 119879

                          61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                          where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                          119896 = 2119863(119897119887119901)2 Equation 7

                          where 119896 is the stepping rate Stepss

                          The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                          119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                          119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                          ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                          120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                          119870119899119900119899119905119890119897 Equation 11

                          where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                          120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                          120591119899119900119899119905119890119897 Equation 12

                          where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                          regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                          programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                          Mln2

                          the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                          Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                          Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                          Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                          1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                          The scale bar is 1 μm

                          2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                          1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                          The scale bar is 1 μm

                          3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                          on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                          scale bar is 1 μm

                          4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                          TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                          min and 16 s) The scale bar is 1 μm

                          5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                          on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                          bar is 1 μm

                          SUPPLEMENTARY REFERENCES

                          65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                          vivo stoichiometry of shelterin components The Journal of biological chemistry

                          285 1457-1467

                          66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                          localization analysis for individual fluorescent probes Biophysical journal 82

                          2775-2783

                          67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                          of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                          TRF2 Molecular cell 8 351-361

                          68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                          absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                          414

                          69 Hughes BD (1995) Random Walks Oxford University Press 1

                          70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                          Relationship of lipid rafts to transient confinement zones detected by single

                          particle tracking Biophysical journal 82 274-284

                          71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                          journal 69 389-398

                          72 Destainville N and Salome L (2006) Quantification and correction of systematic

                          errors due to detector time-averaging in single-molecule tracking experiments

                          Biophysical journal 90 L17-19

                          73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                          confinement and jumps in single-molecule membrane trajectories Phys Rev E

                          Stat Nonlin Soft Matter Phys 73 011915

                          Table S1

                          Ionic Strength

                          DNA TRF1-QDs () TRF2-QDs ()

                          (mM) n I II III IV n I II III IV

                          75 λ DNA 179 5 22 168 76 278 374 83 165 378

                          125 λ DNA 63 79 16 111 794 171 62 41 152 187

                          175 λ DNA 210 30 48 105 547 177 604 34 136 226

                          225 λ DNA 291 443 17 62 478 154 617 52 117 214

                          125 T270 184 109 6 179 652 374 27 51 166 513

                          225 T270 110 45 55 273 627 216 435 32 241 292

                          Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                          Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                          Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                          54 kb 16 kb 38 kb 16 kb 38 kb

                          54 kb

                          ligated non-telomeric DNA

                          linearized T270 (54 kb)

                          16 kb

                          19 kb 16 kb

                          (TTAGGG)135 - (TTAGGG)135

                          linearized Tel10 (599 kb)

                          16 kb 42

                          (TTAGGG)10

                          DNA substrates for AFM imaging

                          DNA substrates for fluorescence imaging

                          ligated T270

                          599 kb ligated Tel10

                          599 kb

                          19 kb

                          QD Ab

                          Ab

                          C

                          B

                          100 nm 100 nm 100 nm

                          nm

                          QD

                          TRF2

                          Antibody

                          A

                          20 nm

                          Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                          nm

                          Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                          D

                          B A

                          C

                          A B

                          C

                          Ionic Strength

                          TRF1-QDs TRF2-QDs

                          (mM) n n

                          75 84 27 166 97

                          125 87 33 142 95

                          175 116 30 154 95

                          225 135 15 158 89

                          D

                          Percentages of motile TRF1- and TRF2-QDs on λ DNA

                          Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                          A

                          C

                          D 2

                          microm

                          10 s

                          TRF2 + λ DNA

                          TRF2 + λ DNA

                          B

                          2 microm

                          10 s

                          TRF1 + λ DNA

                          TRF1 + λ DNA

                          Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                          10 s

                          Complexes TRF1-QDs TRF2-QDs

                          Total number 141 43

                          Dual color () 19 79

                          Red () 41 10

                          Green () 39 11

                          E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                          A

                          Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                          225 mM ionic strength

                          C D

                          B

                          3 min 0 min

                          0 min 63 s

                          A

                          B

                          TRF2 + Tel10 DNA

                          TRF2 + Tel10 DNA

                          Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                          125 mM 225 mM

                          A

                          C D 125 mM 225 mM

                          Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                          A B

                          C D

                          88 plusmn 33 times10-2 microm2sec (994)

                          A

                          B 027 plusmn 013 times10-2 microm2sec (1000)

                          027 plusmn 009 times10-2 microm2sec (1000)

                          48 plusmn 13 times10-2 microm2sec (433)

                          Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                          Unlimited DNA length

                          Unlimited DNA length

                          16 kb DNA length

                          16 kb DNA length

                          B

                          Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                          0 min 4 min

                          TRF2 + T270 DNA

                          2 min 0 min

                          TRF1

                          TRF1 + T270 DNA A

                          0 min

                          TRF2 + T270 DNA

                          4 min

                          4 min 0 min

                          5 min

                          TRF2 + T270 DNA

                          B

                          0 min 2 min

                          TRF2ΔB + λ DNA

                          D

                          C

                          0 min 3 min

                          TRF2ΔB + T270 DNA

                          Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                          A

                          • Lin and countryman et all 2013 NAR
                          • Lin et al Supplement ALL 10-10-13
                            • Lin et al Supplementary Information text 10-10-13
                            • Lin et al supplementary figure 10-10-13
                              • Slide Number 1
                              • Slide Number 2
                              • Slide Number 4
                              • Slide Number 5
                              • Slide Number 6
                              • Slide Number 7
                              • Slide Number 8
                              • Slide Number 9
                              • Slide Number 10
                              • Slide Number 11
                              • Slide Number 12

                            SUPPLEMENTARY TEXT Comparison between in vivo and in vitro conditions Previously HeLa cell lines (volume ~ 2000 μm3) examined by Takai et al contained approximately 04 to 1 x 105 molecules of TRF1 and TRF2 proteins per cell corresponding to ~32 to 83 nM concentrations inside cells (65) However given non-telomeric roles of TRF1 and TRF2 the numbers of TRF1 and TRF2 binding to the telomeric region are uncertain The final concentrations of TRF proteins in our flow cells is 33 nM which is at the upper limit of the concentration that we can use without generating too much background signal The physiological ionic strength is between 100 to 200 mM KCl or NaCl In this study we used total ionic strengths at 75 mM (with 25 mM NaCl) to 225 mM (with 100 mM NaCl) for investigating the dynamics of TRF proteins on telomeric and non-telomeric DNA Single-particle tracking The resolution of tracking of proteins on DNA was determined to be 16 nm using the data from static TRF2-QDs binding to the (TTAGGG)10 regions on ligated Tel10 DNA tightropes (Supplementary Figure S7) based on the following equation (66)

                            2

                            2322

                            412aNbs

                            N

                            as πσ ++

                            = Equation 1

                            where N a b and s are the numbers of photons the effective pixel size (110 nm) standard deviation of the background signal and the standard deviation of 2D Gaussian fit respectively The mean square displacement (MSD) as a function of time interval is given by Equation 2 where N is the total number of frames in the trajectory n is the number of frames for different time intervals ∆t is the time between frames and xi and yi

                            are the positions of the protein-QD in the frame i The 1-D diffusion constant (D) and diffusion exponent (α factor) were analyzed by a custom routine developed in LabView Software based on the following equation (53) 119872119878119863 = 2119863119905120572 Equation 3

                            TRF2 binding at (TTAGGG)10 sequence on the ligated Tel10 DNA substrate displayed an apparent diffusion constant of 25 (plusmn005) times10-4 μm2s (n = 10) This provides a baseline for the characterization of static complexes on DNA A protein was categorized as being motile if the diffusion constant is greater than 5 times10-4 μm2s and R2 value from data fitting using Equation 3 is greater than 08

                            [ ]summinus

                            =++ minus+minus

                            minus=∆

                            nN

                            iiniini yyxx

                            nNtnMSD

                            1

                            22 )()(1)(

                            For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                            lang1199092rang(119905) = 1198711199092

                            6minus 161198711199092

                            1205874sum 1

                            1198994119890119909119901 minus 1

                            2 (119899120587120590119909

                            119871119909)2119905infin

                            119899=1(119900119889119889) Equation 4

                            where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                            119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                            where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                            Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                            Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                            1198631119888119886119897 = 119870119861 119879

                            61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                            where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                            119896 = 2119863(119897119887119901)2 Equation 7

                            where 119896 is the stepping rate Stepss

                            The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                            119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                            119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                            ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                            120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                            119870119899119900119899119905119890119897 Equation 11

                            where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                            120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                            120591119899119900119899119905119890119897 Equation 12

                            where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                            regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                            programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                            Mln2

                            the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                            Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                            Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                            Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                            1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                            The scale bar is 1 μm

                            2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                            1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                            The scale bar is 1 μm

                            3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                            on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                            scale bar is 1 μm

                            4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                            TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                            min and 16 s) The scale bar is 1 μm

                            5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                            on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                            bar is 1 μm

                            SUPPLEMENTARY REFERENCES

                            65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                            vivo stoichiometry of shelterin components The Journal of biological chemistry

                            285 1457-1467

                            66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                            localization analysis for individual fluorescent probes Biophysical journal 82

                            2775-2783

                            67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                            of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                            TRF2 Molecular cell 8 351-361

                            68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                            absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                            414

                            69 Hughes BD (1995) Random Walks Oxford University Press 1

                            70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                            Relationship of lipid rafts to transient confinement zones detected by single

                            particle tracking Biophysical journal 82 274-284

                            71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                            journal 69 389-398

                            72 Destainville N and Salome L (2006) Quantification and correction of systematic

                            errors due to detector time-averaging in single-molecule tracking experiments

                            Biophysical journal 90 L17-19

                            73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                            confinement and jumps in single-molecule membrane trajectories Phys Rev E

                            Stat Nonlin Soft Matter Phys 73 011915

                            Table S1

                            Ionic Strength

                            DNA TRF1-QDs () TRF2-QDs ()

                            (mM) n I II III IV n I II III IV

                            75 λ DNA 179 5 22 168 76 278 374 83 165 378

                            125 λ DNA 63 79 16 111 794 171 62 41 152 187

                            175 λ DNA 210 30 48 105 547 177 604 34 136 226

                            225 λ DNA 291 443 17 62 478 154 617 52 117 214

                            125 T270 184 109 6 179 652 374 27 51 166 513

                            225 T270 110 45 55 273 627 216 435 32 241 292

                            Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                            Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                            Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                            54 kb 16 kb 38 kb 16 kb 38 kb

                            54 kb

                            ligated non-telomeric DNA

                            linearized T270 (54 kb)

                            16 kb

                            19 kb 16 kb

                            (TTAGGG)135 - (TTAGGG)135

                            linearized Tel10 (599 kb)

                            16 kb 42

                            (TTAGGG)10

                            DNA substrates for AFM imaging

                            DNA substrates for fluorescence imaging

                            ligated T270

                            599 kb ligated Tel10

                            599 kb

                            19 kb

                            QD Ab

                            Ab

                            C

                            B

                            100 nm 100 nm 100 nm

                            nm

                            QD

                            TRF2

                            Antibody

                            A

                            20 nm

                            Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                            nm

                            Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                            D

                            B A

                            C

                            A B

                            C

                            Ionic Strength

                            TRF1-QDs TRF2-QDs

                            (mM) n n

                            75 84 27 166 97

                            125 87 33 142 95

                            175 116 30 154 95

                            225 135 15 158 89

                            D

                            Percentages of motile TRF1- and TRF2-QDs on λ DNA

                            Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                            A

                            C

                            D 2

                            microm

                            10 s

                            TRF2 + λ DNA

                            TRF2 + λ DNA

                            B

                            2 microm

                            10 s

                            TRF1 + λ DNA

                            TRF1 + λ DNA

                            Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                            10 s

                            Complexes TRF1-QDs TRF2-QDs

                            Total number 141 43

                            Dual color () 19 79

                            Red () 41 10

                            Green () 39 11

                            E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                            A

                            Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                            225 mM ionic strength

                            C D

                            B

                            3 min 0 min

                            0 min 63 s

                            A

                            B

                            TRF2 + Tel10 DNA

                            TRF2 + Tel10 DNA

                            Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                            125 mM 225 mM

                            A

                            C D 125 mM 225 mM

                            Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                            A B

                            C D

                            88 plusmn 33 times10-2 microm2sec (994)

                            A

                            B 027 plusmn 013 times10-2 microm2sec (1000)

                            027 plusmn 009 times10-2 microm2sec (1000)

                            48 plusmn 13 times10-2 microm2sec (433)

                            Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                            Unlimited DNA length

                            Unlimited DNA length

                            16 kb DNA length

                            16 kb DNA length

                            B

                            Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                            0 min 4 min

                            TRF2 + T270 DNA

                            2 min 0 min

                            TRF1

                            TRF1 + T270 DNA A

                            0 min

                            TRF2 + T270 DNA

                            4 min

                            4 min 0 min

                            5 min

                            TRF2 + T270 DNA

                            B

                            0 min 2 min

                            TRF2ΔB + λ DNA

                            D

                            C

                            0 min 3 min

                            TRF2ΔB + T270 DNA

                            Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                            A

                            • Lin and countryman et all 2013 NAR
                            • Lin et al Supplement ALL 10-10-13
                              • Lin et al Supplementary Information text 10-10-13
                              • Lin et al supplementary figure 10-10-13
                                • Slide Number 1
                                • Slide Number 2
                                • Slide Number 4
                                • Slide Number 5
                                • Slide Number 6
                                • Slide Number 7
                                • Slide Number 8
                                • Slide Number 9
                                • Slide Number 10
                                • Slide Number 11
                                • Slide Number 12

                              For confined diffusion at the telomere sequences an alternative equation was used to fit the MSD vs ∆t to determine the diffusion constant and confined length (55)

                              lang1199092rang(119905) = 1198711199092

                              6minus 161198711199092

                              1205874sum 1

                              1198994119890119909119901 minus 1

                              2 (119899120587120590119909

                              119871119909)2119905infin

                              119899=1(119900119889119889) Equation 4

                              where Lx is the confined length and σ = 2D Data fitting using this model provides similar diffusion constants of 022 (plusmn005) times10-2 microm2s and 037 (plusmn005) times10-2 microm2s for TRF1- and TRF2-QDs respectively (combined 125 and 225 mM ionic strength data) This data fitting scheme also provides confined DNA lengths of 061 (plusmn009) and 04 (plusmn009) μm for TRF2 and TRF1 respectively The lifetime of protein attachment on DNA was determined by the length of the streak in the kymographic analysis Only streaks that began and ended in a movie (Type I Figure 2D) were analyzed which may lead to a small systematic underestimation of the attached lifetime The lifetime data were plotted as cumulative frequency (CF) histograms and fitted to

                              119862119865 = 119873(1 minus 119890minus119896119905)(1 minus 119890minus119896119905119898119886119909) Equation 5

                              where N is the number of observed points t the bin tmax the maximum bin size and k the reciprocal of the dwell time

                              Due to the brightness variation of QDs we were not be able to directly correlate the QD brightness and number of protein molecules on DNA

                              Prediction of diffusion constants and stepping rates based on Stokes-Einstein relation The hydrodynamic radii of red (655 nm) and green Sav-QDs (565 nm) are assumed to be 115 and 95 nm respectively based on recent measurements (46) The estimated radii of free TRF1 and TRF2 are 10 nm based on the crystal structures of the Myb type and dimerization domains (1267) The expected upper limit for diffusion constants for a single red QD labeled TRF1 or TRF2 sliding on DNA is 178 microm2s This is 187- to 468- fold higher than the measured diffusion constants for TRF proteins (between 38 times10-2 and 95 times10-2 microm2s Table 1) Assuming protein rotating around DNA helix the expected upper limits for diffusion constants are based on the modified version of the Stokes-Einstein relation (68)

                              1198631119888119886119897 = 119870119861 119879

                              61205871205781198861+43(2120587)2 11988634times10minus92 Equation 6

                              where η is viscosity of the medium 119886 the radius of the particle 119870119861 the Boltzmann constant T the temperature The stepping rate can be calculated by assuming the diffusion constant to occur as a series of steps of a single base pair using the following relationship (69)

                              119896 = 2119863(119897119887119901)2 Equation 7

                              where 119896 is the stepping rate Stepss

                              The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                              119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                              119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                              ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                              120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                              119870119899119900119899119905119890119897 Equation 11

                              where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                              120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                              120591119899119900119899119905119890119897 Equation 12

                              where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                              regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                              programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                              Mln2

                              the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                              Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                              Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                              Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                              1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                              The scale bar is 1 μm

                              2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                              1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                              The scale bar is 1 μm

                              3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                              on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                              scale bar is 1 μm

                              4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                              TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                              min and 16 s) The scale bar is 1 μm

                              5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                              on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                              bar is 1 μm

                              SUPPLEMENTARY REFERENCES

                              65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                              vivo stoichiometry of shelterin components The Journal of biological chemistry

                              285 1457-1467

                              66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                              localization analysis for individual fluorescent probes Biophysical journal 82

                              2775-2783

                              67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                              of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                              TRF2 Molecular cell 8 351-361

                              68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                              absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                              414

                              69 Hughes BD (1995) Random Walks Oxford University Press 1

                              70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                              Relationship of lipid rafts to transient confinement zones detected by single

                              particle tracking Biophysical journal 82 274-284

                              71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                              journal 69 389-398

                              72 Destainville N and Salome L (2006) Quantification and correction of systematic

                              errors due to detector time-averaging in single-molecule tracking experiments

                              Biophysical journal 90 L17-19

                              73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                              confinement and jumps in single-molecule membrane trajectories Phys Rev E

                              Stat Nonlin Soft Matter Phys 73 011915

                              Table S1

                              Ionic Strength

                              DNA TRF1-QDs () TRF2-QDs ()

                              (mM) n I II III IV n I II III IV

                              75 λ DNA 179 5 22 168 76 278 374 83 165 378

                              125 λ DNA 63 79 16 111 794 171 62 41 152 187

                              175 λ DNA 210 30 48 105 547 177 604 34 136 226

                              225 λ DNA 291 443 17 62 478 154 617 52 117 214

                              125 T270 184 109 6 179 652 374 27 51 166 513

                              225 T270 110 45 55 273 627 216 435 32 241 292

                              Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                              Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                              Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                              54 kb 16 kb 38 kb 16 kb 38 kb

                              54 kb

                              ligated non-telomeric DNA

                              linearized T270 (54 kb)

                              16 kb

                              19 kb 16 kb

                              (TTAGGG)135 - (TTAGGG)135

                              linearized Tel10 (599 kb)

                              16 kb 42

                              (TTAGGG)10

                              DNA substrates for AFM imaging

                              DNA substrates for fluorescence imaging

                              ligated T270

                              599 kb ligated Tel10

                              599 kb

                              19 kb

                              QD Ab

                              Ab

                              C

                              B

                              100 nm 100 nm 100 nm

                              nm

                              QD

                              TRF2

                              Antibody

                              A

                              20 nm

                              Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                              nm

                              Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                              D

                              B A

                              C

                              A B

                              C

                              Ionic Strength

                              TRF1-QDs TRF2-QDs

                              (mM) n n

                              75 84 27 166 97

                              125 87 33 142 95

                              175 116 30 154 95

                              225 135 15 158 89

                              D

                              Percentages of motile TRF1- and TRF2-QDs on λ DNA

                              Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                              A

                              C

                              D 2

                              microm

                              10 s

                              TRF2 + λ DNA

                              TRF2 + λ DNA

                              B

                              2 microm

                              10 s

                              TRF1 + λ DNA

                              TRF1 + λ DNA

                              Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                              10 s

                              Complexes TRF1-QDs TRF2-QDs

                              Total number 141 43

                              Dual color () 19 79

                              Red () 41 10

                              Green () 39 11

                              E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                              A

                              Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                              225 mM ionic strength

                              C D

                              B

                              3 min 0 min

                              0 min 63 s

                              A

                              B

                              TRF2 + Tel10 DNA

                              TRF2 + Tel10 DNA

                              Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                              125 mM 225 mM

                              A

                              C D 125 mM 225 mM

                              Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                              A B

                              C D

                              88 plusmn 33 times10-2 microm2sec (994)

                              A

                              B 027 plusmn 013 times10-2 microm2sec (1000)

                              027 plusmn 009 times10-2 microm2sec (1000)

                              48 plusmn 13 times10-2 microm2sec (433)

                              Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                              Unlimited DNA length

                              Unlimited DNA length

                              16 kb DNA length

                              16 kb DNA length

                              B

                              Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                              0 min 4 min

                              TRF2 + T270 DNA

                              2 min 0 min

                              TRF1

                              TRF1 + T270 DNA A

                              0 min

                              TRF2 + T270 DNA

                              4 min

                              4 min 0 min

                              5 min

                              TRF2 + T270 DNA

                              B

                              0 min 2 min

                              TRF2ΔB + λ DNA

                              D

                              C

                              0 min 3 min

                              TRF2ΔB + T270 DNA

                              Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                              A

                              • Lin and countryman et all 2013 NAR
                              • Lin et al Supplement ALL 10-10-13
                                • Lin et al Supplementary Information text 10-10-13
                                • Lin et al supplementary figure 10-10-13
                                  • Slide Number 1
                                  • Slide Number 2
                                  • Slide Number 4
                                  • Slide Number 5
                                  • Slide Number 6
                                  • Slide Number 7
                                  • Slide Number 8
                                  • Slide Number 9
                                  • Slide Number 10
                                  • Slide Number 11
                                  • Slide Number 12

                                The expected upper limits for diffusion constants and stepping rates for TRF proteins (with one red QD) with rotation-coupled diffusion are 0021 μm2s and 365492 stepss respectively These numbers correspond to a diffusion rate of 042 μm2s and a stepping rate of 7315292 stepss without QDs Prediction of additional energy barriers at telomeric sequences The estimated minimal roughness of the energy landscape at specific binding sites is (35) σ = 119896 BT = 66 119896 BT Equation 8 for a genome size of M = 3 x 109 bp The activation energy barriers (EA) to protein diffusion can be calculated from stepping rates using Arrhenius relationship

                                119896 = 119890minus119864119860 119896119861119879 Equation 9 119864119860 = minusln (119896) ∙ 119896119861119879 where 119896 is the stepping rate Stepss Then the additional energy barrier at the telomeric region compared with the non-telomeric region can be calculated as

                                119864119860119905119890119897 minus 119864119860119899119900119899119905119890119897 = minusln ( 119896119905119890119897119896119899119900119899119905119890119897

                                ) ∙ 119896119861119879 Equation 10 In addition the relative free binding energy at the telomeric regions with respect to non-telomeric regions can be defined as

                                120549120549119866119887119894119899119889 = 119896119861119879119897119899 119870119905119890119897

                                119870119899119900119899119905119890119897 Equation 11

                                where 119870119905119890119897 and 119870119899119900119899119905119890119897 are equilibrium association constants at the telomeric and non-telomeric regions respectively Assuming the association constants are the same at the telomeric and non-telomeric regions Equation 11 can be expressed in terms of lifetimes of a protein on DNA (τ)

                                120549120549119866119887119894119899119889 = 119896119861119879119897119899 120591119905119890119897

                                120591119899119900119899119905119890119897 Equation 12

                                where τtel and τnontel are the lifetimes of a protein at the telomeric and non-telomeric

                                regions respectively Computer simulations of 1-D diffusion Computer simulations of 1-D diffusion of proteins were carried out using PythonTM

                                programming language 330 (Supplementary Figure S9) Nonspecific λ DNA was represented by a 1-D lattice without confinement Since TRF2 and TRF1 have higher affinity for the telomeric sequences we used 16 kb DNA with totally reflecting diffusion barriers at two ends to simulate TRF2 over the 16 kb (TTAGGG)270 telomeric sequence flanked by non-telomeric sequences In our simulations we treated the telomeric region as a transient confinement zone a concept developed for studying lipid rafts (70) The 1-D random walker starts at random locations inside the confined zone To approximate

                                Mln2

                                the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                                Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                                Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                                Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                                1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                                The scale bar is 1 μm

                                2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                                1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                                The scale bar is 1 μm

                                3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                                on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                                scale bar is 1 μm

                                4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                                TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                                min and 16 s) The scale bar is 1 μm

                                5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                                on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                                bar is 1 μm

                                SUPPLEMENTARY REFERENCES

                                65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                                vivo stoichiometry of shelterin components The Journal of biological chemistry

                                285 1457-1467

                                66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                                localization analysis for individual fluorescent probes Biophysical journal 82

                                2775-2783

                                67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                                of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                                TRF2 Molecular cell 8 351-361

                                68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                                absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                                414

                                69 Hughes BD (1995) Random Walks Oxford University Press 1

                                70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                                Relationship of lipid rafts to transient confinement zones detected by single

                                particle tracking Biophysical journal 82 274-284

                                71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                                journal 69 389-398

                                72 Destainville N and Salome L (2006) Quantification and correction of systematic

                                errors due to detector time-averaging in single-molecule tracking experiments

                                Biophysical journal 90 L17-19

                                73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                                confinement and jumps in single-molecule membrane trajectories Phys Rev E

                                Stat Nonlin Soft Matter Phys 73 011915

                                Table S1

                                Ionic Strength

                                DNA TRF1-QDs () TRF2-QDs ()

                                (mM) n I II III IV n I II III IV

                                75 λ DNA 179 5 22 168 76 278 374 83 165 378

                                125 λ DNA 63 79 16 111 794 171 62 41 152 187

                                175 λ DNA 210 30 48 105 547 177 604 34 136 226

                                225 λ DNA 291 443 17 62 478 154 617 52 117 214

                                125 T270 184 109 6 179 652 374 27 51 166 513

                                225 T270 110 45 55 273 627 216 435 32 241 292

                                Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                                Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                                Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                54 kb 16 kb 38 kb 16 kb 38 kb

                                54 kb

                                ligated non-telomeric DNA

                                linearized T270 (54 kb)

                                16 kb

                                19 kb 16 kb

                                (TTAGGG)135 - (TTAGGG)135

                                linearized Tel10 (599 kb)

                                16 kb 42

                                (TTAGGG)10

                                DNA substrates for AFM imaging

                                DNA substrates for fluorescence imaging

                                ligated T270

                                599 kb ligated Tel10

                                599 kb

                                19 kb

                                QD Ab

                                Ab

                                C

                                B

                                100 nm 100 nm 100 nm

                                nm

                                QD

                                TRF2

                                Antibody

                                A

                                20 nm

                                Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                nm

                                Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                D

                                B A

                                C

                                A B

                                C

                                Ionic Strength

                                TRF1-QDs TRF2-QDs

                                (mM) n n

                                75 84 27 166 97

                                125 87 33 142 95

                                175 116 30 154 95

                                225 135 15 158 89

                                D

                                Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                A

                                C

                                D 2

                                microm

                                10 s

                                TRF2 + λ DNA

                                TRF2 + λ DNA

                                B

                                2 microm

                                10 s

                                TRF1 + λ DNA

                                TRF1 + λ DNA

                                Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                10 s

                                Complexes TRF1-QDs TRF2-QDs

                                Total number 141 43

                                Dual color () 19 79

                                Red () 41 10

                                Green () 39 11

                                E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                A

                                Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                225 mM ionic strength

                                C D

                                B

                                3 min 0 min

                                0 min 63 s

                                A

                                B

                                TRF2 + Tel10 DNA

                                TRF2 + Tel10 DNA

                                Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                125 mM 225 mM

                                A

                                C D 125 mM 225 mM

                                Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                A B

                                C D

                                88 plusmn 33 times10-2 microm2sec (994)

                                A

                                B 027 plusmn 013 times10-2 microm2sec (1000)

                                027 plusmn 009 times10-2 microm2sec (1000)

                                48 plusmn 13 times10-2 microm2sec (433)

                                Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                Unlimited DNA length

                                Unlimited DNA length

                                16 kb DNA length

                                16 kb DNA length

                                B

                                Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                0 min 4 min

                                TRF2 + T270 DNA

                                2 min 0 min

                                TRF1

                                TRF1 + T270 DNA A

                                0 min

                                TRF2 + T270 DNA

                                4 min

                                4 min 0 min

                                5 min

                                TRF2 + T270 DNA

                                B

                                0 min 2 min

                                TRF2ΔB + λ DNA

                                D

                                C

                                0 min 3 min

                                TRF2ΔB + T270 DNA

                                Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                A

                                • Lin and countryman et all 2013 NAR
                                • Lin et al Supplement ALL 10-10-13
                                  • Lin et al Supplementary Information text 10-10-13
                                  • Lin et al supplementary figure 10-10-13
                                    • Slide Number 1
                                    • Slide Number 2
                                    • Slide Number 4
                                    • Slide Number 5
                                    • Slide Number 6
                                    • Slide Number 7
                                    • Slide Number 8
                                    • Slide Number 9
                                    • Slide Number 10
                                    • Slide Number 11
                                    • Slide Number 12

                                  the positional averaging from the camera integration time required for imaging the simulated position was averaged for each 50 ms period

                                  Accuracy of determining diffusion constant (D) and confinement length (L) values with camera time averaging

                                  Camera time averaging has been a concern when trying to obtain diffusion constants for confined diffusion because a particle may diffuse a significant fraction of the corral size in time between consecutive measurements (71) However previous theoretical modeling revealed that when τ the dwell time to explore a 1-D lattice with a length of L is large compared to the exposure time ΔT (50 ms) then the estimation of τ L (confinement length) and D (diffusion constant) remain accurate (72) The accuracy of measuring diffusion constants from single-molecule experiment remains accurate for confinements with dwell times down to τ = ΔT3 (72) The dwell time to explore a linear lattice with a confinement length of L is τ = L2(π2D) (73) In the case of TRF1 and TRF2 over the telomeric region τ would be 100 s for a case with a diffusion constant of 03 times 10-2 um2s and L of 055 μm corresponding to τ = 200ΔT gt ΔT3 These parameters support the assumption that the measurement of diffusion constants in this study remains accurate even with camera time averaging

                                  Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                                  1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                                  The scale bar is 1 μm

                                  2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                                  1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                                  The scale bar is 1 μm

                                  3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                                  on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                                  scale bar is 1 μm

                                  4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                                  TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                                  min and 16 s) The scale bar is 1 μm

                                  5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                                  on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                                  bar is 1 μm

                                  SUPPLEMENTARY REFERENCES

                                  65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                                  vivo stoichiometry of shelterin components The Journal of biological chemistry

                                  285 1457-1467

                                  66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                                  localization analysis for individual fluorescent probes Biophysical journal 82

                                  2775-2783

                                  67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                                  of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                                  TRF2 Molecular cell 8 351-361

                                  68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                                  absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                                  414

                                  69 Hughes BD (1995) Random Walks Oxford University Press 1

                                  70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                                  Relationship of lipid rafts to transient confinement zones detected by single

                                  particle tracking Biophysical journal 82 274-284

                                  71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                                  journal 69 389-398

                                  72 Destainville N and Salome L (2006) Quantification and correction of systematic

                                  errors due to detector time-averaging in single-molecule tracking experiments

                                  Biophysical journal 90 L17-19

                                  73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                                  confinement and jumps in single-molecule membrane trajectories Phys Rev E

                                  Stat Nonlin Soft Matter Phys 73 011915

                                  Table S1

                                  Ionic Strength

                                  DNA TRF1-QDs () TRF2-QDs ()

                                  (mM) n I II III IV n I II III IV

                                  75 λ DNA 179 5 22 168 76 278 374 83 165 378

                                  125 λ DNA 63 79 16 111 794 171 62 41 152 187

                                  175 λ DNA 210 30 48 105 547 177 604 34 136 226

                                  225 λ DNA 291 443 17 62 478 154 617 52 117 214

                                  125 T270 184 109 6 179 652 374 27 51 166 513

                                  225 T270 110 45 55 273 627 216 435 32 241 292

                                  Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                                  Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                                  Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                  54 kb 16 kb 38 kb 16 kb 38 kb

                                  54 kb

                                  ligated non-telomeric DNA

                                  linearized T270 (54 kb)

                                  16 kb

                                  19 kb 16 kb

                                  (TTAGGG)135 - (TTAGGG)135

                                  linearized Tel10 (599 kb)

                                  16 kb 42

                                  (TTAGGG)10

                                  DNA substrates for AFM imaging

                                  DNA substrates for fluorescence imaging

                                  ligated T270

                                  599 kb ligated Tel10

                                  599 kb

                                  19 kb

                                  QD Ab

                                  Ab

                                  C

                                  B

                                  100 nm 100 nm 100 nm

                                  nm

                                  QD

                                  TRF2

                                  Antibody

                                  A

                                  20 nm

                                  Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                  nm

                                  Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                  D

                                  B A

                                  C

                                  A B

                                  C

                                  Ionic Strength

                                  TRF1-QDs TRF2-QDs

                                  (mM) n n

                                  75 84 27 166 97

                                  125 87 33 142 95

                                  175 116 30 154 95

                                  225 135 15 158 89

                                  D

                                  Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                  Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                  A

                                  C

                                  D 2

                                  microm

                                  10 s

                                  TRF2 + λ DNA

                                  TRF2 + λ DNA

                                  B

                                  2 microm

                                  10 s

                                  TRF1 + λ DNA

                                  TRF1 + λ DNA

                                  Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                  10 s

                                  Complexes TRF1-QDs TRF2-QDs

                                  Total number 141 43

                                  Dual color () 19 79

                                  Red () 41 10

                                  Green () 39 11

                                  E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                  A

                                  Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                  225 mM ionic strength

                                  C D

                                  B

                                  3 min 0 min

                                  0 min 63 s

                                  A

                                  B

                                  TRF2 + Tel10 DNA

                                  TRF2 + Tel10 DNA

                                  Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                  125 mM 225 mM

                                  A

                                  C D 125 mM 225 mM

                                  Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                  A B

                                  C D

                                  88 plusmn 33 times10-2 microm2sec (994)

                                  A

                                  B 027 plusmn 013 times10-2 microm2sec (1000)

                                  027 plusmn 009 times10-2 microm2sec (1000)

                                  48 plusmn 13 times10-2 microm2sec (433)

                                  Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                  Unlimited DNA length

                                  Unlimited DNA length

                                  16 kb DNA length

                                  16 kb DNA length

                                  B

                                  Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                  0 min 4 min

                                  TRF2 + T270 DNA

                                  2 min 0 min

                                  TRF1

                                  TRF1 + T270 DNA A

                                  0 min

                                  TRF2 + T270 DNA

                                  4 min

                                  4 min 0 min

                                  5 min

                                  TRF2 + T270 DNA

                                  B

                                  0 min 2 min

                                  TRF2ΔB + λ DNA

                                  D

                                  C

                                  0 min 3 min

                                  TRF2ΔB + T270 DNA

                                  Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                  A

                                  • Lin and countryman et all 2013 NAR
                                  • Lin et al Supplement ALL 10-10-13
                                    • Lin et al Supplementary Information text 10-10-13
                                    • Lin et al supplementary figure 10-10-13
                                      • Slide Number 1
                                      • Slide Number 2
                                      • Slide Number 4
                                      • Slide Number 5
                                      • Slide Number 6
                                      • Slide Number 7
                                      • Slide Number 8
                                      • Slide Number 9
                                      • Slide Number 10
                                      • Slide Number 11
                                      • Slide Number 12

                                    Supplementary Movie Legends 1 Movie S1 TRF1-QD on λ DNA A movie demonstrating real-time observation of

                                    1-D sliding of a TRF1-QD on λ DNA at 75 mM ionic strength (real time 34 s)

                                    The scale bar is 1 μm

                                    2 Movie S2 TRF2-QD on λ DNA A movie demonstrating real-time observation of

                                    1-D sliding of a TRF2-QD on λ DNA at 75 mM ionic strength (real time 36 s)

                                    The scale bar is 1 μm

                                    3 Movie S3 TRF1-QDs on the ligated T270 DNA Dual-color labeled TRF1-QDs

                                    on the ligated T270 DNA at 125 mM ionic strength (real time 1 min 26 s) The

                                    scale bar is 1 μm

                                    4 Movie S4 TRF2-QDs on the ligated T270 DNA Dual-color labeled

                                    TRF2-QDs on the ligated T270 DNA at 125 mM ionic strength (real time 1

                                    min and 16 s) The scale bar is 1 μm

                                    5 Movie S5 TRF2-QDs on the ligated Tel10 DNA Dual-color labeled TRF2-QDs

                                    on the ligated Tel10 DNA at 125 mM ionic strength (real time 49 s) The scale

                                    bar is 1 μm

                                    SUPPLEMENTARY REFERENCES

                                    65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                                    vivo stoichiometry of shelterin components The Journal of biological chemistry

                                    285 1457-1467

                                    66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                                    localization analysis for individual fluorescent probes Biophysical journal 82

                                    2775-2783

                                    67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                                    of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                                    TRF2 Molecular cell 8 351-361

                                    68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                                    absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                                    414

                                    69 Hughes BD (1995) Random Walks Oxford University Press 1

                                    70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                                    Relationship of lipid rafts to transient confinement zones detected by single

                                    particle tracking Biophysical journal 82 274-284

                                    71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                                    journal 69 389-398

                                    72 Destainville N and Salome L (2006) Quantification and correction of systematic

                                    errors due to detector time-averaging in single-molecule tracking experiments

                                    Biophysical journal 90 L17-19

                                    73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                                    confinement and jumps in single-molecule membrane trajectories Phys Rev E

                                    Stat Nonlin Soft Matter Phys 73 011915

                                    Table S1

                                    Ionic Strength

                                    DNA TRF1-QDs () TRF2-QDs ()

                                    (mM) n I II III IV n I II III IV

                                    75 λ DNA 179 5 22 168 76 278 374 83 165 378

                                    125 λ DNA 63 79 16 111 794 171 62 41 152 187

                                    175 λ DNA 210 30 48 105 547 177 604 34 136 226

                                    225 λ DNA 291 443 17 62 478 154 617 52 117 214

                                    125 T270 184 109 6 179 652 374 27 51 166 513

                                    225 T270 110 45 55 273 627 216 435 32 241 292

                                    Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                                    Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                                    Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                    54 kb 16 kb 38 kb 16 kb 38 kb

                                    54 kb

                                    ligated non-telomeric DNA

                                    linearized T270 (54 kb)

                                    16 kb

                                    19 kb 16 kb

                                    (TTAGGG)135 - (TTAGGG)135

                                    linearized Tel10 (599 kb)

                                    16 kb 42

                                    (TTAGGG)10

                                    DNA substrates for AFM imaging

                                    DNA substrates for fluorescence imaging

                                    ligated T270

                                    599 kb ligated Tel10

                                    599 kb

                                    19 kb

                                    QD Ab

                                    Ab

                                    C

                                    B

                                    100 nm 100 nm 100 nm

                                    nm

                                    QD

                                    TRF2

                                    Antibody

                                    A

                                    20 nm

                                    Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                    nm

                                    Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                    D

                                    B A

                                    C

                                    A B

                                    C

                                    Ionic Strength

                                    TRF1-QDs TRF2-QDs

                                    (mM) n n

                                    75 84 27 166 97

                                    125 87 33 142 95

                                    175 116 30 154 95

                                    225 135 15 158 89

                                    D

                                    Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                    Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                    A

                                    C

                                    D 2

                                    microm

                                    10 s

                                    TRF2 + λ DNA

                                    TRF2 + λ DNA

                                    B

                                    2 microm

                                    10 s

                                    TRF1 + λ DNA

                                    TRF1 + λ DNA

                                    Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                    10 s

                                    Complexes TRF1-QDs TRF2-QDs

                                    Total number 141 43

                                    Dual color () 19 79

                                    Red () 41 10

                                    Green () 39 11

                                    E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                    A

                                    Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                    225 mM ionic strength

                                    C D

                                    B

                                    3 min 0 min

                                    0 min 63 s

                                    A

                                    B

                                    TRF2 + Tel10 DNA

                                    TRF2 + Tel10 DNA

                                    Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                    125 mM 225 mM

                                    A

                                    C D 125 mM 225 mM

                                    Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                    A B

                                    C D

                                    88 plusmn 33 times10-2 microm2sec (994)

                                    A

                                    B 027 plusmn 013 times10-2 microm2sec (1000)

                                    027 plusmn 009 times10-2 microm2sec (1000)

                                    48 plusmn 13 times10-2 microm2sec (433)

                                    Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                    Unlimited DNA length

                                    Unlimited DNA length

                                    16 kb DNA length

                                    16 kb DNA length

                                    B

                                    Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                    0 min 4 min

                                    TRF2 + T270 DNA

                                    2 min 0 min

                                    TRF1

                                    TRF1 + T270 DNA A

                                    0 min

                                    TRF2 + T270 DNA

                                    4 min

                                    4 min 0 min

                                    5 min

                                    TRF2 + T270 DNA

                                    B

                                    0 min 2 min

                                    TRF2ΔB + λ DNA

                                    D

                                    C

                                    0 min 3 min

                                    TRF2ΔB + T270 DNA

                                    Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                    A

                                    • Lin and countryman et all 2013 NAR
                                    • Lin et al Supplement ALL 10-10-13
                                      • Lin et al Supplementary Information text 10-10-13
                                      • Lin et al supplementary figure 10-10-13
                                        • Slide Number 1
                                        • Slide Number 2
                                        • Slide Number 4
                                        • Slide Number 5
                                        • Slide Number 6
                                        • Slide Number 7
                                        • Slide Number 8
                                        • Slide Number 9
                                        • Slide Number 10
                                        • Slide Number 11
                                        • Slide Number 12

                                      SUPPLEMENTARY REFERENCES

                                      65 Takai KK Hooper S Blackwood S Gandhi R and de Lange T (2010) In

                                      vivo stoichiometry of shelterin components The Journal of biological chemistry

                                      285 1457-1467

                                      66 Thompson RE Larson DR and Webb WW (2002) Precise nanometer

                                      localization analysis for individual fluorescent probes Biophysical journal 82

                                      2775-2783

                                      67 Fairall L Chapman L Moss H de Lange T and Rhodes D (2001) Structure

                                      of the TRFH dimerization domain of the human telomeric proteins TRF1 and

                                      TRF2 Molecular cell 8 351-361

                                      68 Schurr JM (1979) The one-dimensional diffusion coefficient of proteins

                                      absorbed on DNA Hydrodynamic considerations Biophysical chemistry 9 413-

                                      414

                                      69 Hughes BD (1995) Random Walks Oxford University Press 1

                                      70 Dietrich C Yang B Fujiwara T Kusumi A and Jacobson K (2002)

                                      Relationship of lipid rafts to transient confinement zones detected by single

                                      particle tracking Biophysical journal 82 274-284

                                      71 Saxton MJ (1995) Single-particle tracking effects of corrals Biophysical

                                      journal 69 389-398

                                      72 Destainville N and Salome L (2006) Quantification and correction of systematic

                                      errors due to detector time-averaging in single-molecule tracking experiments

                                      Biophysical journal 90 L17-19

                                      73 Meilhac N Le Guyader L Salome L and Destainville N (2006) Detection of

                                      confinement and jumps in single-molecule membrane trajectories Phys Rev E

                                      Stat Nonlin Soft Matter Phys 73 011915

                                      Table S1

                                      Ionic Strength

                                      DNA TRF1-QDs () TRF2-QDs ()

                                      (mM) n I II III IV n I II III IV

                                      75 λ DNA 179 5 22 168 76 278 374 83 165 378

                                      125 λ DNA 63 79 16 111 794 171 62 41 152 187

                                      175 λ DNA 210 30 48 105 547 177 604 34 136 226

                                      225 λ DNA 291 443 17 62 478 154 617 52 117 214

                                      125 T270 184 109 6 179 652 374 27 51 166 513

                                      225 T270 110 45 55 273 627 216 435 32 241 292

                                      Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                                      Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                                      Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                      54 kb 16 kb 38 kb 16 kb 38 kb

                                      54 kb

                                      ligated non-telomeric DNA

                                      linearized T270 (54 kb)

                                      16 kb

                                      19 kb 16 kb

                                      (TTAGGG)135 - (TTAGGG)135

                                      linearized Tel10 (599 kb)

                                      16 kb 42

                                      (TTAGGG)10

                                      DNA substrates for AFM imaging

                                      DNA substrates for fluorescence imaging

                                      ligated T270

                                      599 kb ligated Tel10

                                      599 kb

                                      19 kb

                                      QD Ab

                                      Ab

                                      C

                                      B

                                      100 nm 100 nm 100 nm

                                      nm

                                      QD

                                      TRF2

                                      Antibody

                                      A

                                      20 nm

                                      Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                      nm

                                      Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                      D

                                      B A

                                      C

                                      A B

                                      C

                                      Ionic Strength

                                      TRF1-QDs TRF2-QDs

                                      (mM) n n

                                      75 84 27 166 97

                                      125 87 33 142 95

                                      175 116 30 154 95

                                      225 135 15 158 89

                                      D

                                      Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                      Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                      A

                                      C

                                      D 2

                                      microm

                                      10 s

                                      TRF2 + λ DNA

                                      TRF2 + λ DNA

                                      B

                                      2 microm

                                      10 s

                                      TRF1 + λ DNA

                                      TRF1 + λ DNA

                                      Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                      10 s

                                      Complexes TRF1-QDs TRF2-QDs

                                      Total number 141 43

                                      Dual color () 19 79

                                      Red () 41 10

                                      Green () 39 11

                                      E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                      A

                                      Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                      225 mM ionic strength

                                      C D

                                      B

                                      3 min 0 min

                                      0 min 63 s

                                      A

                                      B

                                      TRF2 + Tel10 DNA

                                      TRF2 + Tel10 DNA

                                      Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                      125 mM 225 mM

                                      A

                                      C D 125 mM 225 mM

                                      Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                      A B

                                      C D

                                      88 plusmn 33 times10-2 microm2sec (994)

                                      A

                                      B 027 plusmn 013 times10-2 microm2sec (1000)

                                      027 plusmn 009 times10-2 microm2sec (1000)

                                      48 plusmn 13 times10-2 microm2sec (433)

                                      Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                      Unlimited DNA length

                                      Unlimited DNA length

                                      16 kb DNA length

                                      16 kb DNA length

                                      B

                                      Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                      0 min 4 min

                                      TRF2 + T270 DNA

                                      2 min 0 min

                                      TRF1

                                      TRF1 + T270 DNA A

                                      0 min

                                      TRF2 + T270 DNA

                                      4 min

                                      4 min 0 min

                                      5 min

                                      TRF2 + T270 DNA

                                      B

                                      0 min 2 min

                                      TRF2ΔB + λ DNA

                                      D

                                      C

                                      0 min 3 min

                                      TRF2ΔB + T270 DNA

                                      Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                      A

                                      • Lin and countryman et all 2013 NAR
                                      • Lin et al Supplement ALL 10-10-13
                                        • Lin et al Supplementary Information text 10-10-13
                                        • Lin et al supplementary figure 10-10-13
                                          • Slide Number 1
                                          • Slide Number 2
                                          • Slide Number 4
                                          • Slide Number 5
                                          • Slide Number 6
                                          • Slide Number 7
                                          • Slide Number 8
                                          • Slide Number 9
                                          • Slide Number 10
                                          • Slide Number 11
                                          • Slide Number 12

                                        Table S1

                                        Ionic Strength

                                        DNA TRF1-QDs () TRF2-QDs ()

                                        (mM) n I II III IV n I II III IV

                                        75 λ DNA 179 5 22 168 76 278 374 83 165 378

                                        125 λ DNA 63 79 16 111 794 171 62 41 152 187

                                        175 λ DNA 210 30 48 105 547 177 604 34 136 226

                                        225 λ DNA 291 443 17 62 478 154 617 52 117 214

                                        125 T270 184 109 6 179 652 374 27 51 166 513

                                        225 T270 110 45 55 273 627 216 435 32 241 292

                                        Percentages of four types of protein-DNA interactions for TRF1 and TRF2 on λ DNA and the ligated T270 DNA substrate at different ionic strengths

                                        Note The examples of each type of kymograph are shown in Fig 2D The total video length was 2 minutes for λ DNA and 4 mins for the ligated T270 DNA substrate Increasing the video length to 4 mins for λ DNA does not change the lifetime measurement for the Type I complexes

                                        Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                        54 kb 16 kb 38 kb 16 kb 38 kb

                                        54 kb

                                        ligated non-telomeric DNA

                                        linearized T270 (54 kb)

                                        16 kb

                                        19 kb 16 kb

                                        (TTAGGG)135 - (TTAGGG)135

                                        linearized Tel10 (599 kb)

                                        16 kb 42

                                        (TTAGGG)10

                                        DNA substrates for AFM imaging

                                        DNA substrates for fluorescence imaging

                                        ligated T270

                                        599 kb ligated Tel10

                                        599 kb

                                        19 kb

                                        QD Ab

                                        Ab

                                        C

                                        B

                                        100 nm 100 nm 100 nm

                                        nm

                                        QD

                                        TRF2

                                        Antibody

                                        A

                                        20 nm

                                        Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                        nm

                                        Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                        D

                                        B A

                                        C

                                        A B

                                        C

                                        Ionic Strength

                                        TRF1-QDs TRF2-QDs

                                        (mM) n n

                                        75 84 27 166 97

                                        125 87 33 142 95

                                        175 116 30 154 95

                                        225 135 15 158 89

                                        D

                                        Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                        Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                        A

                                        C

                                        D 2

                                        microm

                                        10 s

                                        TRF2 + λ DNA

                                        TRF2 + λ DNA

                                        B

                                        2 microm

                                        10 s

                                        TRF1 + λ DNA

                                        TRF1 + λ DNA

                                        Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                        10 s

                                        Complexes TRF1-QDs TRF2-QDs

                                        Total number 141 43

                                        Dual color () 19 79

                                        Red () 41 10

                                        Green () 39 11

                                        E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                        A

                                        Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                        225 mM ionic strength

                                        C D

                                        B

                                        3 min 0 min

                                        0 min 63 s

                                        A

                                        B

                                        TRF2 + Tel10 DNA

                                        TRF2 + Tel10 DNA

                                        Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                        125 mM 225 mM

                                        A

                                        C D 125 mM 225 mM

                                        Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                        A B

                                        C D

                                        88 plusmn 33 times10-2 microm2sec (994)

                                        A

                                        B 027 plusmn 013 times10-2 microm2sec (1000)

                                        027 plusmn 009 times10-2 microm2sec (1000)

                                        48 plusmn 13 times10-2 microm2sec (433)

                                        Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                        Unlimited DNA length

                                        Unlimited DNA length

                                        16 kb DNA length

                                        16 kb DNA length

                                        B

                                        Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                        0 min 4 min

                                        TRF2 + T270 DNA

                                        2 min 0 min

                                        TRF1

                                        TRF1 + T270 DNA A

                                        0 min

                                        TRF2 + T270 DNA

                                        4 min

                                        4 min 0 min

                                        5 min

                                        TRF2 + T270 DNA

                                        B

                                        0 min 2 min

                                        TRF2ΔB + λ DNA

                                        D

                                        C

                                        0 min 3 min

                                        TRF2ΔB + T270 DNA

                                        Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                        A

                                        • Lin and countryman et all 2013 NAR
                                        • Lin et al Supplement ALL 10-10-13
                                          • Lin et al Supplementary Information text 10-10-13
                                          • Lin et al supplementary figure 10-10-13
                                            • Slide Number 1
                                            • Slide Number 2
                                            • Slide Number 4
                                            • Slide Number 5
                                            • Slide Number 6
                                            • Slide Number 7
                                            • Slide Number 8
                                            • Slide Number 9
                                            • Slide Number 10
                                            • Slide Number 11
                                            • Slide Number 12

                                          Figure S1 DNA substrates used for AFM and fluorescence imaging (not including λ DNA) The purple and bluegreen regions represent telomeric and non-telomeric sequences respectively Ligated DNA substrates have heterogeneous lengths

                                          54 kb 16 kb 38 kb 16 kb 38 kb

                                          54 kb

                                          ligated non-telomeric DNA

                                          linearized T270 (54 kb)

                                          16 kb

                                          19 kb 16 kb

                                          (TTAGGG)135 - (TTAGGG)135

                                          linearized Tel10 (599 kb)

                                          16 kb 42

                                          (TTAGGG)10

                                          DNA substrates for AFM imaging

                                          DNA substrates for fluorescence imaging

                                          ligated T270

                                          599 kb ligated Tel10

                                          599 kb

                                          19 kb

                                          QD Ab

                                          Ab

                                          C

                                          B

                                          100 nm 100 nm 100 nm

                                          nm

                                          QD

                                          TRF2

                                          Antibody

                                          A

                                          20 nm

                                          Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                          nm

                                          Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                          D

                                          B A

                                          C

                                          A B

                                          C

                                          Ionic Strength

                                          TRF1-QDs TRF2-QDs

                                          (mM) n n

                                          75 84 27 166 97

                                          125 87 33 142 95

                                          175 116 30 154 95

                                          225 135 15 158 89

                                          D

                                          Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                          Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                          A

                                          C

                                          D 2

                                          microm

                                          10 s

                                          TRF2 + λ DNA

                                          TRF2 + λ DNA

                                          B

                                          2 microm

                                          10 s

                                          TRF1 + λ DNA

                                          TRF1 + λ DNA

                                          Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                          10 s

                                          Complexes TRF1-QDs TRF2-QDs

                                          Total number 141 43

                                          Dual color () 19 79

                                          Red () 41 10

                                          Green () 39 11

                                          E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                          A

                                          Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                          225 mM ionic strength

                                          C D

                                          B

                                          3 min 0 min

                                          0 min 63 s

                                          A

                                          B

                                          TRF2 + Tel10 DNA

                                          TRF2 + Tel10 DNA

                                          Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                          125 mM 225 mM

                                          A

                                          C D 125 mM 225 mM

                                          Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                          A B

                                          C D

                                          88 plusmn 33 times10-2 microm2sec (994)

                                          A

                                          B 027 plusmn 013 times10-2 microm2sec (1000)

                                          027 plusmn 009 times10-2 microm2sec (1000)

                                          48 plusmn 13 times10-2 microm2sec (433)

                                          Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                          Unlimited DNA length

                                          Unlimited DNA length

                                          16 kb DNA length

                                          16 kb DNA length

                                          B

                                          Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                          0 min 4 min

                                          TRF2 + T270 DNA

                                          2 min 0 min

                                          TRF1

                                          TRF1 + T270 DNA A

                                          0 min

                                          TRF2 + T270 DNA

                                          4 min

                                          4 min 0 min

                                          5 min

                                          TRF2 + T270 DNA

                                          B

                                          0 min 2 min

                                          TRF2ΔB + λ DNA

                                          D

                                          C

                                          0 min 3 min

                                          TRF2ΔB + T270 DNA

                                          Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                          A

                                          • Lin and countryman et all 2013 NAR
                                          • Lin et al Supplement ALL 10-10-13
                                            • Lin et al Supplementary Information text 10-10-13
                                            • Lin et al supplementary figure 10-10-13
                                              • Slide Number 1
                                              • Slide Number 2
                                              • Slide Number 4
                                              • Slide Number 5
                                              • Slide Number 6
                                              • Slide Number 7
                                              • Slide Number 8
                                              • Slide Number 9
                                              • Slide Number 10
                                              • Slide Number 11
                                              • Slide Number 12

                                            QD Ab

                                            Ab

                                            C

                                            B

                                            100 nm 100 nm 100 nm

                                            nm

                                            QD

                                            TRF2

                                            Antibody

                                            A

                                            20 nm

                                            Figure S2 Stoichiometry of TRF2-QDs revealed by AFM imaging (A) A schematic representation of TRF2 primary antibody binding to a TRF2-QD (B) AFM images of TRF2-QDs in the presence of TRF2 primary antibody (TRF2Ab = 11) White arrows point to TRF2-QDs with a single antibody (C) Cross section analysis of a TRF2-QD complex binding to TRF2 antibody The section analysis on the right is from the path drawn in the AFM image on the left (red line)

                                            nm

                                            Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                            D

                                            B A

                                            C

                                            A B

                                            C

                                            Ionic Strength

                                            TRF1-QDs TRF2-QDs

                                            (mM) n n

                                            75 84 27 166 97

                                            125 87 33 142 95

                                            175 116 30 154 95

                                            225 135 15 158 89

                                            D

                                            Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                            Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                            A

                                            C

                                            D 2

                                            microm

                                            10 s

                                            TRF2 + λ DNA

                                            TRF2 + λ DNA

                                            B

                                            2 microm

                                            10 s

                                            TRF1 + λ DNA

                                            TRF1 + λ DNA

                                            Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                            10 s

                                            Complexes TRF1-QDs TRF2-QDs

                                            Total number 141 43

                                            Dual color () 19 79

                                            Red () 41 10

                                            Green () 39 11

                                            E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                            A

                                            Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                            225 mM ionic strength

                                            C D

                                            B

                                            3 min 0 min

                                            0 min 63 s

                                            A

                                            B

                                            TRF2 + Tel10 DNA

                                            TRF2 + Tel10 DNA

                                            Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                            125 mM 225 mM

                                            A

                                            C D 125 mM 225 mM

                                            Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                            A B

                                            C D

                                            88 plusmn 33 times10-2 microm2sec (994)

                                            A

                                            B 027 plusmn 013 times10-2 microm2sec (1000)

                                            027 plusmn 009 times10-2 microm2sec (1000)

                                            48 plusmn 13 times10-2 microm2sec (433)

                                            Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                            Unlimited DNA length

                                            Unlimited DNA length

                                            16 kb DNA length

                                            16 kb DNA length

                                            B

                                            Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                            0 min 4 min

                                            TRF2 + T270 DNA

                                            2 min 0 min

                                            TRF1

                                            TRF1 + T270 DNA A

                                            0 min

                                            TRF2 + T270 DNA

                                            4 min

                                            4 min 0 min

                                            5 min

                                            TRF2 + T270 DNA

                                            B

                                            0 min 2 min

                                            TRF2ΔB + λ DNA

                                            D

                                            C

                                            0 min 3 min

                                            TRF2ΔB + T270 DNA

                                            Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                            A

                                            • Lin and countryman et all 2013 NAR
                                            • Lin et al Supplement ALL 10-10-13
                                              • Lin et al Supplementary Information text 10-10-13
                                              • Lin et al supplementary figure 10-10-13
                                                • Slide Number 1
                                                • Slide Number 2
                                                • Slide Number 4
                                                • Slide Number 5
                                                • Slide Number 6
                                                • Slide Number 7
                                                • Slide Number 8
                                                • Slide Number 9
                                                • Slide Number 10
                                                • Slide Number 11
                                                • Slide Number 12

                                              Figure S3 TRF1- and TRF2-QDs bind specifically to the telomeric sequences (A) Position distributions of TRF1-QDs (black bars n = 103) and unlabeled TRF1 (gray bars n = 83) on the T270 telomeric substrate and TRF1-QDs on the non-telomeric DNA substrate (white bars n = 13) (B) Position distributions of TRF2-QDs (black bars n = 106) unlabeled TRF2 (gray bars n = 76) on the T270 telomeric substrate and TRF2-QDs on the non-telomeric DNA substrate (white bars n = 100) The (TTAGGG)270 sequence makes up approximately 30 of the total DNA length and is located in the middle of the linearized T270 DNA (between 35 to 50 from one end) Among the protein-QDs on DNA ~406 TRF1-QD and ~526 of TRF2-QD bound to the telomeric regions (excluding complexes bound right at the end of the DNA) The small peak at 24 is consistent with the locations of previously discovered preferred TRF1 and TRF2 binding sequence (CCATTC) over the non-telomeric region (C and D) Position distributions of TRF1-QDs (C n = 67) and TRF2-QDs (D n=51) on linear Tel10 DNA substrate with (TTAGGG)10 sequence located at 42 from one DNA end

                                              D

                                              B A

                                              C

                                              A B

                                              C

                                              Ionic Strength

                                              TRF1-QDs TRF2-QDs

                                              (mM) n n

                                              75 84 27 166 97

                                              125 87 33 142 95

                                              175 116 30 154 95

                                              225 135 15 158 89

                                              D

                                              Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                              Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                              A

                                              C

                                              D 2

                                              microm

                                              10 s

                                              TRF2 + λ DNA

                                              TRF2 + λ DNA

                                              B

                                              2 microm

                                              10 s

                                              TRF1 + λ DNA

                                              TRF1 + λ DNA

                                              Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                              10 s

                                              Complexes TRF1-QDs TRF2-QDs

                                              Total number 141 43

                                              Dual color () 19 79

                                              Red () 41 10

                                              Green () 39 11

                                              E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                              A

                                              Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                              225 mM ionic strength

                                              C D

                                              B

                                              3 min 0 min

                                              0 min 63 s

                                              A

                                              B

                                              TRF2 + Tel10 DNA

                                              TRF2 + Tel10 DNA

                                              Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                              125 mM 225 mM

                                              A

                                              C D 125 mM 225 mM

                                              Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                              A B

                                              C D

                                              88 plusmn 33 times10-2 microm2sec (994)

                                              A

                                              B 027 plusmn 013 times10-2 microm2sec (1000)

                                              027 plusmn 009 times10-2 microm2sec (1000)

                                              48 plusmn 13 times10-2 microm2sec (433)

                                              Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                              Unlimited DNA length

                                              Unlimited DNA length

                                              16 kb DNA length

                                              16 kb DNA length

                                              B

                                              Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                              0 min 4 min

                                              TRF2 + T270 DNA

                                              2 min 0 min

                                              TRF1

                                              TRF1 + T270 DNA A

                                              0 min

                                              TRF2 + T270 DNA

                                              4 min

                                              4 min 0 min

                                              5 min

                                              TRF2 + T270 DNA

                                              B

                                              0 min 2 min

                                              TRF2ΔB + λ DNA

                                              D

                                              C

                                              0 min 3 min

                                              TRF2ΔB + T270 DNA

                                              Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                              A

                                              • Lin and countryman et all 2013 NAR
                                              • Lin et al Supplement ALL 10-10-13
                                                • Lin et al Supplementary Information text 10-10-13
                                                • Lin et al supplementary figure 10-10-13
                                                  • Slide Number 1
                                                  • Slide Number 2
                                                  • Slide Number 4
                                                  • Slide Number 5
                                                  • Slide Number 6
                                                  • Slide Number 7
                                                  • Slide Number 8
                                                  • Slide Number 9
                                                  • Slide Number 10
                                                  • Slide Number 11
                                                  • Slide Number 12

                                                A B

                                                C

                                                Ionic Strength

                                                TRF1-QDs TRF2-QDs

                                                (mM) n n

                                                75 84 27 166 97

                                                125 87 33 142 95

                                                175 116 30 154 95

                                                225 135 15 158 89

                                                D

                                                Percentages of motile TRF1- and TRF2-QDs on λ DNA

                                                Figure S4 Diffusional properties of TRF1 and TRF2 on λ DNA (A) Percentages of motile TRF1- and TRF2-QDs on λ DNA n is the number of complexes analyzed TRF proteins were labeled with red QDs (655 nm) (B) Diffusion constants of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths (C) Alpha factors of TRF1- (white bars) and TRF2-QDs (black bars) at different ionic strengths The numbers of the complexes analyzed were 51 37 40 and 33 for TRF1-QDs and 59 54 63 and 66 for TRF2-QDs at 75 125 175 225 mM ionic strengths respectively (D) The effect of ionic strength on the lifetimes of TRF1- (white bars) and TRF2-QDs (Black bars) on λ DNA The numbers of the complexes analyzed were 63 and 128 for TRF1-QDs and 104 106 107 and 95 for TRF2-QDs with increasing ionic strengths Across all ionic strengths compared to TRF2 TRF1 showed lower affinity for nonspecific λ DNA indicated by lower average numbers of protein-QDs (11 plusmn 01 vs 133 plusmn 22) in each field of view (27 microm x 55 microm) and lower average numbers of TRF1-QDs on DNA tightrope(s) between two beads (11 plusmn 01 vs 38 plusmn 03)

                                                A

                                                C

                                                D 2

                                                microm

                                                10 s

                                                TRF2 + λ DNA

                                                TRF2 + λ DNA

                                                B

                                                2 microm

                                                10 s

                                                TRF1 + λ DNA

                                                TRF1 + λ DNA

                                                Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                                10 s

                                                Complexes TRF1-QDs TRF2-QDs

                                                Total number 141 43

                                                Dual color () 19 79

                                                Red () 41 10

                                                Green () 39 11

                                                E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                                A

                                                Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                                225 mM ionic strength

                                                C D

                                                B

                                                3 min 0 min

                                                0 min 63 s

                                                A

                                                B

                                                TRF2 + Tel10 DNA

                                                TRF2 + Tel10 DNA

                                                Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                                125 mM 225 mM

                                                A

                                                C D 125 mM 225 mM

                                                Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                                A B

                                                C D

                                                88 plusmn 33 times10-2 microm2sec (994)

                                                A

                                                B 027 plusmn 013 times10-2 microm2sec (1000)

                                                027 plusmn 009 times10-2 microm2sec (1000)

                                                48 plusmn 13 times10-2 microm2sec (433)

                                                Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                Unlimited DNA length

                                                Unlimited DNA length

                                                16 kb DNA length

                                                16 kb DNA length

                                                B

                                                Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                0 min 4 min

                                                TRF2 + T270 DNA

                                                2 min 0 min

                                                TRF1

                                                TRF1 + T270 DNA A

                                                0 min

                                                TRF2 + T270 DNA

                                                4 min

                                                4 min 0 min

                                                5 min

                                                TRF2 + T270 DNA

                                                B

                                                0 min 2 min

                                                TRF2ΔB + λ DNA

                                                D

                                                C

                                                0 min 3 min

                                                TRF2ΔB + T270 DNA

                                                Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                A

                                                • Lin and countryman et all 2013 NAR
                                                • Lin et al Supplement ALL 10-10-13
                                                  • Lin et al Supplementary Information text 10-10-13
                                                  • Lin et al supplementary figure 10-10-13
                                                    • Slide Number 1
                                                    • Slide Number 2
                                                    • Slide Number 4
                                                    • Slide Number 5
                                                    • Slide Number 6
                                                    • Slide Number 7
                                                    • Slide Number 8
                                                    • Slide Number 9
                                                    • Slide Number 10
                                                    • Slide Number 11
                                                    • Slide Number 12

                                                  A

                                                  C

                                                  D 2

                                                  microm

                                                  10 s

                                                  TRF2 + λ DNA

                                                  TRF2 + λ DNA

                                                  B

                                                  2 microm

                                                  10 s

                                                  TRF1 + λ DNA

                                                  TRF1 + λ DNA

                                                  Figure S5 Dual color labeling of TRF1 and TRF2 A representative fluorescence image (A) and kymograph (B) of dual-color labeled TRF1-QDs on λ DNA A representative ORF image (C) and kymograph (D) of dual-color labeled TRF2-QDs on λ DNA TRF1 or TRF2 was incubated with equal molar of red (655 nm) and green (565 nm) QDs The kymographs shown in (B) and (D) are from the boxed regions in (A) and (C) respectively Among the observed TRF2-QD complexes encountering another protein on DNA (n = 26) we did not observe any protein barrier bypass events for TRF2 Due to the lower binding affinity of TRF1 on DNA with non-telomeric sequences even though we did not observe bypass when TRF1-QD complexes encountered barriers posed by other proteins (n = 3 out of 87 binding events) the number of events was not sufficient to make any statistical comparisons (E) Percentages of dual color QD labeled TRF1 and TRF2

                                                  10 s

                                                  Complexes TRF1-QDs TRF2-QDs

                                                  Total number 141 43

                                                  Dual color () 19 79

                                                  Red () 41 10

                                                  Green () 39 11

                                                  E Percentages of dual color labeled TRF1 and TRF2 on λ DNA

                                                  A

                                                  Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                                  225 mM ionic strength

                                                  C D

                                                  B

                                                  3 min 0 min

                                                  0 min 63 s

                                                  A

                                                  B

                                                  TRF2 + Tel10 DNA

                                                  TRF2 + Tel10 DNA

                                                  Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                                  125 mM 225 mM

                                                  A

                                                  C D 125 mM 225 mM

                                                  Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                                  A B

                                                  C D

                                                  88 plusmn 33 times10-2 microm2sec (994)

                                                  A

                                                  B 027 plusmn 013 times10-2 microm2sec (1000)

                                                  027 plusmn 009 times10-2 microm2sec (1000)

                                                  48 plusmn 13 times10-2 microm2sec (433)

                                                  Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                  Unlimited DNA length

                                                  Unlimited DNA length

                                                  16 kb DNA length

                                                  16 kb DNA length

                                                  B

                                                  Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                  0 min 4 min

                                                  TRF2 + T270 DNA

                                                  2 min 0 min

                                                  TRF1

                                                  TRF1 + T270 DNA A

                                                  0 min

                                                  TRF2 + T270 DNA

                                                  4 min

                                                  4 min 0 min

                                                  5 min

                                                  TRF2 + T270 DNA

                                                  B

                                                  0 min 2 min

                                                  TRF2ΔB + λ DNA

                                                  D

                                                  C

                                                  0 min 3 min

                                                  TRF2ΔB + T270 DNA

                                                  Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                  A

                                                  • Lin and countryman et all 2013 NAR
                                                  • Lin et al Supplement ALL 10-10-13
                                                    • Lin et al Supplementary Information text 10-10-13
                                                    • Lin et al supplementary figure 10-10-13
                                                      • Slide Number 1
                                                      • Slide Number 2
                                                      • Slide Number 4
                                                      • Slide Number 5
                                                      • Slide Number 6
                                                      • Slide Number 7
                                                      • Slide Number 8
                                                      • Slide Number 9
                                                      • Slide Number 10
                                                      • Slide Number 11
                                                      • Slide Number 12

                                                    A

                                                    Figure S6 Additional information on the length of ligated T270 DNA tightropes the diffusion range distributions and distances between two adjacent TRF2-QDs on the ligated T270 substrate (A) Measured ligated T270 DNA lengths between two anchoring silica beads (n = 163) (B) The diffusion range distribution of TRF2-QDs at 225 mM ionic strength (n = 4 1) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 05 and 15 μm (C) The diffusion range of TRF2-QDs on ligated T270 DNA at 068 nM TRF2 and 033 nM QD concentrations (n = 33) The line in the plot is single Gaussian fit to the data with R2 of 097 and the peak centered at 05 μm Individual DNA tightropes with one or two QDs (no contact during recording) were analyzed (D) The distances between adjacent TRF2-QDs on DNA tightropes formed at 25 μlml flow rate (n = 121) The line in the plot is double Gaussian fit to the data with R2 of 097 and peaks centered at 15 and 27 μm The spacing between adjacent TRF2-QDs indicate that DNA was stretched to ~88 of its contour length

                                                    225 mM ionic strength

                                                    C D

                                                    B

                                                    3 min 0 min

                                                    0 min 63 s

                                                    A

                                                    B

                                                    TRF2 + Tel10 DNA

                                                    TRF2 + Tel10 DNA

                                                    Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                                    125 mM 225 mM

                                                    A

                                                    C D 125 mM 225 mM

                                                    Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                                    A B

                                                    C D

                                                    88 plusmn 33 times10-2 microm2sec (994)

                                                    A

                                                    B 027 plusmn 013 times10-2 microm2sec (1000)

                                                    027 plusmn 009 times10-2 microm2sec (1000)

                                                    48 plusmn 13 times10-2 microm2sec (433)

                                                    Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                    Unlimited DNA length

                                                    Unlimited DNA length

                                                    16 kb DNA length

                                                    16 kb DNA length

                                                    B

                                                    Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                    0 min 4 min

                                                    TRF2 + T270 DNA

                                                    2 min 0 min

                                                    TRF1

                                                    TRF1 + T270 DNA A

                                                    0 min

                                                    TRF2 + T270 DNA

                                                    4 min

                                                    4 min 0 min

                                                    5 min

                                                    TRF2 + T270 DNA

                                                    B

                                                    0 min 2 min

                                                    TRF2ΔB + λ DNA

                                                    D

                                                    C

                                                    0 min 3 min

                                                    TRF2ΔB + T270 DNA

                                                    Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                    A

                                                    • Lin and countryman et all 2013 NAR
                                                    • Lin et al Supplement ALL 10-10-13
                                                      • Lin et al Supplementary Information text 10-10-13
                                                      • Lin et al supplementary figure 10-10-13
                                                        • Slide Number 1
                                                        • Slide Number 2
                                                        • Slide Number 4
                                                        • Slide Number 5
                                                        • Slide Number 6
                                                        • Slide Number 7
                                                        • Slide Number 8
                                                        • Slide Number 9
                                                        • Slide Number 10
                                                        • Slide Number 11
                                                        • Slide Number 12

                                                      3 min 0 min

                                                      0 min 63 s

                                                      A

                                                      B

                                                      TRF2 + Tel10 DNA

                                                      TRF2 + Tel10 DNA

                                                      Figure S7 Dynamics of TRF2-QDs on the ligated Tel10 DNA (A) Kymographs of TRF2-QDs on the ligated Tel10 DNA On this substrate there were two populations static complexes with apparent diffusion constant of 25 (plusmn005) times10-4 microm2s (n = 10) and complexes with long diffusion range (gt850 nm white arrows) (B) Measured distance between two adjacent TRF2-QDs on ligated T10 DNA substrate (n = 35) The line in (B) is Gaussian fit to the data which has R2 of 098 Consistent with a larger spacing (599 kb) between two adjacent telomeric regions on the Tel10 DNA substrate static TRF2-QDs were spaced at longer distances (19 plusmn 0049 microm) compared to that on the ligated T270 substrate (16 plusmn 001 microm Fig 3)

                                                      125 mM 225 mM

                                                      A

                                                      C D 125 mM 225 mM

                                                      Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                                      A B

                                                      C D

                                                      88 plusmn 33 times10-2 microm2sec (994)

                                                      A

                                                      B 027 plusmn 013 times10-2 microm2sec (1000)

                                                      027 plusmn 009 times10-2 microm2sec (1000)

                                                      48 plusmn 13 times10-2 microm2sec (433)

                                                      Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                      Unlimited DNA length

                                                      Unlimited DNA length

                                                      16 kb DNA length

                                                      16 kb DNA length

                                                      B

                                                      Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                      0 min 4 min

                                                      TRF2 + T270 DNA

                                                      2 min 0 min

                                                      TRF1

                                                      TRF1 + T270 DNA A

                                                      0 min

                                                      TRF2 + T270 DNA

                                                      4 min

                                                      4 min 0 min

                                                      5 min

                                                      TRF2 + T270 DNA

                                                      B

                                                      0 min 2 min

                                                      TRF2ΔB + λ DNA

                                                      D

                                                      C

                                                      0 min 3 min

                                                      TRF2ΔB + T270 DNA

                                                      Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                      A

                                                      • Lin and countryman et all 2013 NAR
                                                      • Lin et al Supplement ALL 10-10-13
                                                        • Lin et al Supplementary Information text 10-10-13
                                                        • Lin et al supplementary figure 10-10-13
                                                          • Slide Number 1
                                                          • Slide Number 2
                                                          • Slide Number 4
                                                          • Slide Number 5
                                                          • Slide Number 6
                                                          • Slide Number 7
                                                          • Slide Number 8
                                                          • Slide Number 9
                                                          • Slide Number 10
                                                          • Slide Number 11
                                                          • Slide Number 12

                                                        125 mM 225 mM

                                                        A

                                                        C D 125 mM 225 mM

                                                        Figure S8 Short diffusion ranges of TRF1- and TRF2-QDs at (TTAGGG)270 telomeric regions are narrowly distributed independent of the video length The diffusion range over time for TRF1-QDs at 125 mM ionic strength (A) and 225 mM strength (B) and TRF2-QDs at 125 mM ionic strength (C) and 225 mM strength (D) on λ DNA (red dots) and the ligated T270 DNA (blue squares) The numbers of complexes plotted for TRF1 are 37 and 33 for λ DNA 26 and 21 for T270 DNA at 125 and 225 mM ionic strength respectively The numbers of complexes plotted for TRF2 are 48 and 66 for λ DNA 22 and 34 for T270 DNA at 125 and 225 mM ionic strength respectively On λ DNA the diffusion ranges are widely distributed between approximately 05 to 9 microm The time scale (from 10 to 120 s) is from the cropped videos used for the diffusion constant and diffusion range analysis

                                                        A B

                                                        C D

                                                        88 plusmn 33 times10-2 microm2sec (994)

                                                        A

                                                        B 027 plusmn 013 times10-2 microm2sec (1000)

                                                        027 plusmn 009 times10-2 microm2sec (1000)

                                                        48 plusmn 13 times10-2 microm2sec (433)

                                                        Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                        Unlimited DNA length

                                                        Unlimited DNA length

                                                        16 kb DNA length

                                                        16 kb DNA length

                                                        B

                                                        Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                        0 min 4 min

                                                        TRF2 + T270 DNA

                                                        2 min 0 min

                                                        TRF1

                                                        TRF1 + T270 DNA A

                                                        0 min

                                                        TRF2 + T270 DNA

                                                        4 min

                                                        4 min 0 min

                                                        5 min

                                                        TRF2 + T270 DNA

                                                        B

                                                        0 min 2 min

                                                        TRF2ΔB + λ DNA

                                                        D

                                                        C

                                                        0 min 3 min

                                                        TRF2ΔB + T270 DNA

                                                        Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                        A

                                                        • Lin and countryman et all 2013 NAR
                                                        • Lin et al Supplement ALL 10-10-13
                                                          • Lin et al Supplementary Information text 10-10-13
                                                          • Lin et al supplementary figure 10-10-13
                                                            • Slide Number 1
                                                            • Slide Number 2
                                                            • Slide Number 4
                                                            • Slide Number 5
                                                            • Slide Number 6
                                                            • Slide Number 7
                                                            • Slide Number 8
                                                            • Slide Number 9
                                                            • Slide Number 10
                                                            • Slide Number 11
                                                            • Slide Number 12

                                                          88 plusmn 33 times10-2 microm2sec (994)

                                                          A

                                                          B 027 plusmn 013 times10-2 microm2sec (1000)

                                                          027 plusmn 009 times10-2 microm2sec (1000)

                                                          48 plusmn 13 times10-2 microm2sec (433)

                                                          Figure S9 Computer simulations of diffusion by modeling random walk of proteins on a 1-D DNA lattice using PythonTM programming language The diffusion constants used for simulation were that of TRF2-QDs on λ DNA (A) and on the telomeric DNA (B) at 125 mM ionic strength A plot for MSD vs Δt obtained and an example of a trajectory of a protein (insert) are presented for (left A) DNA with unlimited length and (right A) DNA with 16 kb length at 1460436 stepss stepping rate (corresponding to TRF2-QDs on λ DNA) A plot for MSD vs Δt and an example of a trajectory of a protein (insert) are also presented for (left B) DNA with unlimited length and (right B) DNA with 16 kb length at 46713 stepss (corresponding to the rate of TRF2-QDs at the telomeric region) The fitting parameters were constructed such that only the initial linear portion of the MSD vs Δt plots was used for calculating the diffusion constant For diffusion with confinement a protein walks along a 1-DNA lattice with two totally reflecting barriers The numbers in each plot are the mean and standard deviation of the simulated data The number of particles simulated for each case is indicated in the parentheses

                                                          Unlimited DNA length

                                                          Unlimited DNA length

                                                          16 kb DNA length

                                                          16 kb DNA length

                                                          B

                                                          Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                          0 min 4 min

                                                          TRF2 + T270 DNA

                                                          2 min 0 min

                                                          TRF1

                                                          TRF1 + T270 DNA A

                                                          0 min

                                                          TRF2 + T270 DNA

                                                          4 min

                                                          4 min 0 min

                                                          5 min

                                                          TRF2 + T270 DNA

                                                          B

                                                          0 min 2 min

                                                          TRF2ΔB + λ DNA

                                                          D

                                                          C

                                                          0 min 3 min

                                                          TRF2ΔB + T270 DNA

                                                          Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                          A

                                                          • Lin and countryman et all 2013 NAR
                                                          • Lin et al Supplement ALL 10-10-13
                                                            • Lin et al Supplementary Information text 10-10-13
                                                            • Lin et al supplementary figure 10-10-13
                                                              • Slide Number 1
                                                              • Slide Number 2
                                                              • Slide Number 4
                                                              • Slide Number 5
                                                              • Slide Number 6
                                                              • Slide Number 7
                                                              • Slide Number 8
                                                              • Slide Number 9
                                                              • Slide Number 10
                                                              • Slide Number 11
                                                              • Slide Number 12

                                                            B

                                                            Figure S10 Dissociation events of TRF1 and TRF2-QDs on the ligated T270 DNA Representative kymographs of TRF1- (A) and TRF2-QDs (B) on the ligated T270 DNA at 125 mM ionic strength The scale bar is 1 μm The yellow arrows point to dissociation events either directly from the telomeric region or through the non-telomeric region The purple arrows point to the transient dissociation and rebinding at the same telomeric region

                                                            0 min 4 min

                                                            TRF2 + T270 DNA

                                                            2 min 0 min

                                                            TRF1

                                                            TRF1 + T270 DNA A

                                                            0 min

                                                            TRF2 + T270 DNA

                                                            4 min

                                                            4 min 0 min

                                                            5 min

                                                            TRF2 + T270 DNA

                                                            B

                                                            0 min 2 min

                                                            TRF2ΔB + λ DNA

                                                            D

                                                            C

                                                            0 min 3 min

                                                            TRF2ΔB + T270 DNA

                                                            Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                            A

                                                            • Lin and countryman et all 2013 NAR
                                                            • Lin et al Supplement ALL 10-10-13
                                                              • Lin et al Supplementary Information text 10-10-13
                                                              • Lin et al supplementary figure 10-10-13
                                                                • Slide Number 1
                                                                • Slide Number 2
                                                                • Slide Number 4
                                                                • Slide Number 5
                                                                • Slide Number 6
                                                                • Slide Number 7
                                                                • Slide Number 8
                                                                • Slide Number 9
                                                                • Slide Number 10
                                                                • Slide Number 11
                                                                • Slide Number 12

                                                              B

                                                              0 min 2 min

                                                              TRF2ΔB + λ DNA

                                                              D

                                                              C

                                                              0 min 3 min

                                                              TRF2ΔB + T270 DNA

                                                              Figure S11 Dynamics of TRF2ΔB-QDs on λ DNA and the ligated T270 DNA (A) Position distributions of TRF2ΔB-QDs on the linear T270 DNA substrate (n = 51) Among the protein-QDs on DNA ~66 TRF2ΔB-QDs bound to the telomeric regions (35 to 50 from DNA ends) Kymographs of TRF2ΔB-QDs on λ DNA (B) and the ligated T270 DNA (C) The scale bar is 1 μm Protein-QD-DNA reactions were carried out at 125 ionic strength (D) Diffusion range distributions of TRF2ΔB-QDs on λ DNA (white bars n = 21) and the ligated T270 DNA (stripped bars n = 30) The binding affinity of TRF2ΔB to λ DNA tightropes was significantly lower as indicated by lower average numbers of TRF2ΔB-QDs on DNA in the field of view (14 plusmn 02 vs 133 plusmn 22) and lower average numbers of protein-QDs on DNA tightropes between two beads (11 plusmn 01 vs 38 plusmn 03) Dual color QD labeling confirmed that TRF2ΔB can form dimers or higher order oligomeric species but at a significantly lower percent (14) compared with the full length protein (79) The fraction of motile protein-DNA complexes decreased from 95 for full length TRF2 to 65 (n = 40) for TRF2ΔB at 125 mM and from 89 to 74 (n = 23) at 225 mM ionic strength

                                                              A

                                                              • Lin and countryman et all 2013 NAR
                                                              • Lin et al Supplement ALL 10-10-13
                                                                • Lin et al Supplementary Information text 10-10-13
                                                                • Lin et al supplementary figure 10-10-13
                                                                  • Slide Number 1
                                                                  • Slide Number 2
                                                                  • Slide Number 4
                                                                  • Slide Number 5
                                                                  • Slide Number 6
                                                                  • Slide Number 7
                                                                  • Slide Number 8
                                                                  • Slide Number 9
                                                                  • Slide Number 10
                                                                  • Slide Number 11
                                                                  • Slide Number 12

                                                                top related