Double-stranded DNA translocase activity of transcription factor … · XPB/Ssl2 attempts to track away from the PIC, but, because it is bound to the PIC via TFIIH, it instead feeds

Double-stranded DNA translocase activity oftranscription factor TFIIH and the mechanism of RNApolymerase II open complex formationJames Fishburna, Eric Tomkob, Eric Galburtb,1, and Steven Hahna,1

aDivision of Basic Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA 98109; and bDepartment of Biochemistry and Molecular Biophysics,Washington University in St. Louis, St. Louis, MO 63110

Edited by Robert G. Roeder, The Rockefeller University, New York, NY, and approved February 24, 2015 (received for review September 12, 2014)

Formation of the RNA polymerase II (Pol II) open complex (OC) re-quires DNA unwinding mediated by the transcription factorTFIIH helicase-related subunit XPB/Ssl2. Because XPB/Ssl2 bindsDNA downstream from the location of DNA unwinding, it cannotfunction using a conventional helicase mechanism. Here we showthat yeast TFIIH contains an Ssl2-dependent double-stranded DNAtranslocase activity. Ssl2 tracks along one DNA strand in the 5′ →3′ direction, implying it uses the nontemplate promoter strand toreel downstream DNA into the Pol II cleft, creating torsional strainand leading to DNA unwinding. Analysis of the Ssl2 and DNA-de-pendent ATPase activity of TFIIH suggests that Ssl2 has a proces-sivity of approximately one DNA turn, consistent with the lengthof DNA unwound during transcription initiation. Our results canexplain why maintaining the OC requires continuous ATP hydrolysisand the function of TFIIH in promoter escape. Our results also sug-gest that XPB/Ssl2 uses this translocase mechanism during DNA re-pair rather than physically wedging open damaged DNA.

transcription initiation | DNA unwinding | DNA helicase

For all multisubunit RNA polymerases (Pols), a universal stepin the transcription initiation pathway is formation of the

open complex (OC) (1, 2). The OC forms when Pol and its as-sociated transcription machinery bind to promoter DNA, gen-erating a series of conformational changes in both DNA andprotein, including the unwinding of ∼11 bp of DNA upstreamfrom the transcription start site (TSS). This open state is stabi-lized by interactions of Pol with the unwound strands of pro-moter DNA and by the binding of downstream double-strandedpromoter DNA to the Pol Cleft/Jaw domains (3–5). All testedmultisubunit Pols spontaneously form OCs, except for RNA PolII, where ATP and the general transcription factor TFIIH arerequired for DNA unwinding (6–8). For Pol II, this unwoundstate is unstable, decaying with a half-life of 30–60 s (8, 9).The general transcription factor TFIIH contains two subunits

with DNA-dependent ATPase activity, XPD/Rad3 and XPB/Ssl2(human/yeast proteins), which are members of the SF2 helicase-translocase family (10). The XPB/Ssl2 ATPase is required forDNA unwinding in the OC, whereas the XPD/Rad3 ATPaseactivity does not function in transcription (11–14). In DNAstrand displacement assays, the two isolated human subunitshave DNA helicase activity of opposite polarity with XPD having5′ → 3′ activity and XPB having 3′ → 5′ activity. XPD is at leasteightfold more active than XPB in strand displacement (15). Incontrast to the activity of purified XPB, the most purified prepa-rations of native or recombinant human TFIIH have only the XPD5′ → 3′ helicase activity, with no detectable 3′ → 5′ helicasefunction (14, 15). In addition to its role in transcription initiation,TFIIH can assist in Pol II promoter escape in an XPB-dependentmechanism (16, 17), and both the XPB and XPD subunits ofTFIIH play an essential role in general and transcription-couplednucleotide excision repair (NER) (18, 19).Mapping the location of XPB/Ssl2 in RNA Pol II preinitiation

complexes (PICs) revealed that this factor binds promoter DNA

downstream from the site of DNA unwinding in the OC (20–24).Therefore, XPB/Ssl2 cannot function as a conventional helicaseto promote OC formation (20). Three models have been pro-posed to explain the role of XPB/Ssl2 in transcription. First, itwas postulated that XPB acts as a molecular wrench, binding toits site on downstream DNA and using its ATPase to rotate up-stream DNA within the Pol II cleft (20). Because upstream DNAis constrained by TBP, TFIIB, and other factors, DNA rotationcould lead to DNA opening. Second, it was proposed that theXPB ATPase activity promoted DNA opening via a conforma-tional change in PIC components, leading to a structural rear-rangement of both protein and DNA, analogous to the ATP-dependent mechanism of OC formation in the bacterial σ54 system(25). Third, comparing structure models of the PIC and OC sug-gested that ∼15 bp of downstream promoter DNA inserts into thePol II cleft upon OC formation (22). Based on this and other data,it was proposed that XPB/Ssl2 functions as a double-strandedDNA (dsDNA) translocase (22, 23, 26). According to this model,XPB/Ssl2 attempts to track away from the PIC, but, because it isbound to the PIC via TFIIH, it instead feeds and rotates dsDNAinto the Pol II cleft, leading to DNA opening. However, a keyfeature of this latter model, whether XPB/Ssl2 has dsDNA trans-locase function, had not been tested.Here we show that yeast TFIIH has Ssl2-dependent dsDNA

translocase activity and that it primarily tracks along one DNAstrand in the 5′ → 3′ direction. The kinetic properties of theenzyme suggest a rate-limiting step between DNA binding andtranslocation and a processivity on the order of a turn of DNA,

Significance

How ATP hydrolysis is coupled to promoter DNA unwindingand open complex formation at RNA polymerase II (Poll II)promoters is a longstanding question. Of the multisubunit RNApolymerases, only Pol II requires ATP for DNA unwinding. Herewe show that the general transcription factor TFIIH subunitSsl2 is a double-stranded DNA translocase. These and otherdata suggest that Ssl2 promotes DNA opening by trackingalong the nontemplate promoter strand, rotating and insertingDNA into the Pol II active site cleft, leading to DNA unwinding.Our accompanying biochemical studies explain why the opencomplex is unstable and how TFIIH can promote Pol II escapefrom the promoter. Our findings also have important implica-tions for the mechanism of TFIIH-mediated DNA repair.

Author contributions: J.F., E.T., E.G., and S.H. designed research; J.F. and E.T. performedresearch; J.F., E.T., E.G., and S.H. analyzed data; and J.F., E.T., E.G., and S.H. wrotethe paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1To whom correspondence may be addressed. Email: [email protected] [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1417709112/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1417709112 PNAS | March 31, 2015 | vol. 112 | no. 13 | 3961–3966

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021

consistent with the amount of DNA thought to be unwound inthe OC. Our findings have important implications for the mech-anism of OC formation, the relative instability of the OC, and therole of XPB/Ssl2 in NER.

ResultsTFIIH Has Ssl2-Dependent dsDNA Translocase Activity. We first mea-sured DNA helicase activity of purified yeast TFIIH using asubstrate consisting of a 100-base oligonucleotide with a labeled25-base oligonucleotide annealed at either the 3′ or 5′ end(Fig. 1A). Consistent with the function of human TFIIH, we foundthat yeast TFIIH contains 5′→ 3′ helicase but not 3′→ 5′ helicasefunction (Fig. 1B). The 5′ → 3′ helicase activity requires bothhydrolysable ATP and the Rad3 ATPase, because a TFIIH de-rivative containing a mutation in the Rad3 Walker B motif (Rad3

E236Q) that is predicted to be defective in ATP hydrolysis (27)is defective in helicase function. In contrast, both WT and Rad3mutant TFIIH preparations are equally active in supportingtranscription with purified factors and Pol II from the yeastHIS4 promoter (Fig. S1).As an initial test for dsDNA translocase function, we assayed

whether TFIIH can displace a 22 nucleotide DNA triplex, a well-established assay for dsDNA translocase function (28–30). TheDNA triplex was formed by annealing a polypyrimidine triplex-forming oligonucleotide (TFO) to a 142-bp dsDNA containinga complementary polypurine sequence at one end (Fig. 1A). Thetriplex is disrupted by heating, but is stable at 26 °C with orwithout the addition of ATP (Fig. 1C, lanes 1–3). Incubation ofWT TFIIH, ATP, and the triplex substrate results in near com-plete dissociation of the TFO (Fig. 1C, lanes 4–6). The triplexdisplacement activity of TFIIH is unaffected by the Rad3 E236QATPase mutation but is abolished by the equivalent WalkerB ATPase mutation in Ssl2 (E489Q; Fig. 1C, lanes 7–11). Toassay whether TFIIH can simply bind and displace the triplexwithout translocation, we used a 22 nucleotide triplex substratethat did not contain additional dsDNA (Fig. 2A). The short triplexsubstrate is less stable than the longer triplex, likely due to the lackof stabilizing dsDNA. Nevertheless, as predicted for a dsDNAtranslocase, TFIIH cannot displace the TFO from this shortersubstrate lacking dsDNA (Fig. 2B). The TFIIH triplex displace-ment activity was not dependent on a free DNA end, becauseTFIIH can also displace the TFO annealed to a 3.2-kb double-stranded plasmid, although with much slower kinetics (Fig. 2C).Both ATP and dATP function to promote OC formation (7) and,as expected, both nucleotides can promote triplex displacement(compare Fig. 1C and Fig. 2C, Right). Our combined results areconsistent with the prediction that TFIIH contains an Ssl2-dependent dsDNA translocase activity.

Ssl2 Tracks in the 5′ → 3′ Direction Along the DNA Duplex. dsDNAtranslocases are thought to track along one strand of the DNAbackbone (31). If true for Ssl2, discerning the polarity of trans-location will reveal which promoter strand is used by Ssl2 duringDNA unwinding. As an initial test of polarity, we assayed triplexdisplacement using substrates where the TFO was annealedto either the top or bottom strand of the 142-bp duplex DNA(Fig. 3A). As shown above, TFIIH can readily displace the TFOfrom the bottom strand triplex substrate (Fig. 3B, lanes 4–6). Incontrast, the top strand triplex substrate is almost completelyresistant to TFO displacement (Fig. 3B, lanes 1–3). One modelconsistent with this result is that Ssl2 tracks along one strand ofduplex DNA in the 5′ → 3′ direction and that this tracking isblocked by the annealed TFO. As a further test of this model, weintroduced biotin as a tracking barrier on the triplex substrates(Fig. 3A). These substrates contain a single-strand DNA nickwith biotin attached to the 5′ end of one strand of duplex DNA.We found that the top-strand biotin was a much stronger blockto TFO displacement, with only one third of the TFO displacedafter 6 h (Fig. 3 C and D). Although the bottom-strand biotinmodestly inhibited TFO displacement, the TFO was >80% dis-placed after 6 h, similar to the nonbiotinylated template. As afurther test of whether the integrity of the top strand is the mostcritical for TFO displacement, we generated triplex substrateswith 5-bp single-stranded gaps on either the top or bottom du-plex strand, directly adjacent to the triplex (Fig. S2). Consistentwith the above results, we found that TFO displacement was mostinhibited by the gap on the top strand. Under conditions of theassay, 80–90% of the TFO was displaced from the nongappedtemplate, 45% displaced from the template with the gap on thebottom strand, and only 20–25% displaced from the templatewith the gap on the top strand. Our combined results stronglysuggest that Ssl2 primarily tracks along one strand of duplex DNAin the 5′ → 3′ direction.

Fig. 1. ATP-dependent helicase and translocase activities of TFIIH. (A) Heli-case and triple helix substrates. Helicase substrates are designed todistinguish 5′ → 3′ and 3′ → 5′ enzyme polarity. The short displaced oligo-nucleotide is fluorescently labeled for visualization. The triplex substrate isformed via Hoogsteen base pairs between the 22 nucleotide (nt) 32P-labeledtriplex forming oligo (TFO) and the purine rich strand of the PCR generatedduplex. (B) Helicase assay testing ATP-dependence and polarity of WT TFIIHor TFIIH containing the ATPase mutant Rad3-E236Q. (C) Triplex disruption as-say testing ATPase requirements for triplex displacement with either WT TFIIHor TFIIH containing the ATPase defective mutants Rad3-E236Q or Ssl2-E489Q.

3962 | www.pnas.org/cgi/doi/10.1073/pnas.1417709112 Fishburn et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021

Blocking Ssl2 Translocation Inhibits Transcription Initiation. Theabove results predict that blocking Ssl2 translocation should in-hibit transcription initiation. Mapping the location of Ssl2/XPB-DNA binding in PICs has suggested Ssl2 binds 30–36 (yeast) and40–50 bp (human) downstream from TATA (20–23). To test therole of Ssl2 translocase function in transcription initiation, wecreated three promoter derivatives with Cy3 dye inserted in thephosphodiester backbone of the nontemplate DNA strand 37, 41,or 46 bp downstream from the HIS4 TATA (Fig. 4A). Cy3 posi-tioned in the DNA backbone is a strong inhibitor of Ssl2 trans-location in the TFO displacement assay (Fig. S3).The modified and unmodified DNA templates were tested for

in vitro transcription activity using the reconstituted system (32),and transcription from these templates was completely TFIIHdependent (Fig. 4B, compare lanes 4–7 and 8–11). We found thatCy3 positioned on the nontemplate strand 41 and 46 bp fromTATA inhibited transcription approximately fivefold (Fig. 4B,lanes 4, 6, and 7), consistent with the dsDNA translocase model.Cy3 positioned 37 bp downstream from TATA inhibited tran-scription approximately twofold (Fig. 4B, lanes 4–5). We speculatethat transcription escaping this more upstream Cy3 insertion maybe due to inherent flexibility in the Ssl2–DNA interaction, allow-ing Ssl2 to escape the Cy3 block by binding DNA just downstreamof Cy3 37 in a position nearly equivalent to the observed XPB-human promoter interaction (23).To confirm that the observed transcription inhibition by Cy3

could not be explained by an inhibition of transcription elongation,

we created HIS4 promoter derivatives containing a 12 nucleotidesingle-strand bubble beginning 21 bp downstream from TATA(Fig. 4C). Pol II initiating from this bubble in the absence of anygeneral factors transcribed past the Cy3 block with 47–77% ef-ficiency compared with a bubble template lacking Cy3 (Fig. 4D,lanes 4–7). Taken together, our results show that the integrity ofthe nontemplate strand is important for transcription initiationand is consistent with the dsDNA translocase model for opencomplex formation.

Ssl2 Translocates with Low Processivity. To further characterize theSsl2 motor within the TFIIH complex, we next investigated stim-ulation of the ATPase by nucleic acid. Measurements of how thesteady-state rate of ATP hydrolysis depends on DNA templatelength and concentration can be used to assay for translocationand to evaluate different models of translocation (33–35). For

Fig. 2. Template requirements for triplex disruption by TFIIH. (A) Triplehelix substrates used to monitor TFIIH translocation. TFO22 was generatedby annealing the 22-nt TFO to the 22-bp triplex target sequence. This smallertriplex was somewhat less stable than the 142-bp triplex DNA, likely due tono additional dsDNA. To make the 142-bp and circular triplex templates, theTFO was annealed to a PCR-generated 142-bp duplex or a 3.2-kb plasmidcontaining the triplex target sequence. (B) Triplex disruption assay using theTFO22 triple helix template and the TFIIH derivative Rad3 E236Q. (C) Timecourse of triplex disruption comparing the circular 3.2-kb plasmid triplex andthe linear 142-bp triplex templates. Intact triplex was measured at each timepoint and quantified by comparison with a standard curve. The percentdisruption of each triplex is indicated. ATP and dATP both function in theTFO displacement assay (compare with Fig. 1C) and in OC formation (7).

Fig. 3. Polarity of Ssl2 catalyzed TFIIH translocation. (A) Triple helix sub-strates designed to test the polarity of Ssl2 translocation. Top and bottomstrand triplexes are annealed as shown using PCR-generated duplex DNAand the TFO. PAGE-purified oligonucleotides were annealed with the TFO togenerate the biotin-containing triplexes where a 5′-biotin labeled oligonu-cleotide is positioned immediately upstream of the triplex on the top orbottom duplex strand. (B) Triplex disruption assay comparing templates withthe TFO annealed to either the top or bottom strand of the duplex. EitherWT or Rad3 E236Q TFIIH was used as indicated. (C) Time course of triplexdisruption from biotin templates by TFIIH (Rad3-E236Q). Reactions wereincubated for the indicated times with 1 mM dATP at 26 °C and quantifiedfor intact triplex remaining. (D) Quantitation of results in C.

Fishburn et al. PNAS | March 31, 2015 | vol. 112 | no. 13 | 3963

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021

these measurements, we used the TFIIH preparation contain-ing Rad3 E236Q so that Ssl2 was the only functional ATPase.TFIIH and DNA were preincubated for 40 min, and then ATPwas added, and phosphate release was quantitated at differenttimes (1–20 min). The steady-state rate of ATP hydrolysis wasdetermined by a linear fit of these data (Fig. 5A). ATP hydrolysisin the absence of nucleic acid was undetectable, and dsDNAconsistently stimulated ATPase activity four- to fivefold higherthan that of single-stranded DNA (Table 1), consistent with theenzyme being a dsDNA translocase.The steady-state rate of ATP hydrolysis was measured as a

function of DNA concentration and DNA template lengthover a range of DNA lengths from 30 to 80 bp and on a circularplasmid representing an infinitely long template (Fig. 5 A andB). As the dependence of rate on DNA concentration follows

a Michaelis–Menten curve, we were able to determine both aMichaelis constant (KM) and a maximal velocity (Vmax) for eachlength of DNA (Fig. 5B). Although KM does not show a lengthdependence (Fig. 5C), Vmax is clearly sensitive to DNA length(Fig. 5D). The dependence of Vmax on template length is a clearindication that Ssl2 is a dsDNA translocase as nontranslocatingenzymes do not show this behavior (33–35) (Fig. S4A).Models of translocation can be used to fit the dependence of

Vmax on DNA length to extract quantitative values for parame-ters such as occluded site size, kinetic step size, and motor pro-cessivity, i.e., the probability that the motor takes a step forwardinstead of dissociating (Materials and Methods) (33–35). In thesimplest model for translocation, the motor binds DNA at any siteand steps forward coupled to ATP hydrolysis. At each site, themotor may either hydrolyze ATP and take a step forward or dis-sociate (Fig. S4B). At the end of the template, the motor eitherdissociates or translocates off the end of the DNA. This modelpredicts that KM will vary with the length of the DNA template,whereas Vmax will not (Fig. S4B). This behavior is inconsistent withour observed results (Fig. 5 C and D).However, in contrast to the simple translocation model above,

models that incorporate a slow kinetic step between unboundand the active DNA-bound motor states result in a length-dependent Vmax and a length-independent KM (33–35). Two lim-iting cases of this behavior are (i) a slow step after binding andbefore translocation or (ii) a slow step before dissociation fromthe end of the DNA (Fig. S4C). These two cases predict different

Fig. 4. In vitro transcription from promoters with DNA backbone blocks onthe nontemplate DNA strand. (A) HIS4 promoter derivatives with Cy3 DNAbackbone insertions. DNAs were constructed from synthetic oligonucleotidesand contained Cy3 positioned 37, 41, or 46 bp downstream from HIS4 TATA.Transcription was assayed by primer extension using the lacI oligonucleotideas shown. (B) In vitro transcription using the reconstituted yeast Pol II systemand the promoters in A. Lanes 1–4 contain the indicated amounts of atranscription reaction used to generate a standard for quantitation of tran-scription signals relative to the unmodified template. Lanes 5–7 are tran-scription reactions using the indicated Cy3 templates. Percent transcriptionrelative to the unblocked template is indicated. No transcription is observedwhen TFIIH is omitted (lanes 8–11). (C) HIS4 promoter derivatives identical tothose in A except that they contain the 12 nucleotide single-stranded DNAbubble as shown. This bubble allows transcription initiation by Pol II in theabsence of other general factors (33). (D) In vitro transcription using purifiedyeast Pol II on the bubble templates and assayed by primer extension. Lanes1–4 contain the indicated amounts of a transcription reaction using the non-Cy3 bubble template. Lanes 5–7 are transcription reactions using the Cy3-modified bubble templates. Lanes 8–9 are mock transcription reactions lackingPol II and assayed by primer extension. The products marked by * are due toprimer extension of the DNA template, which is blocked by Cy3. These blockedproducts are not visible in B, lanes 9–11, as they are significantly longer thanthe RNA products initiated at the HIS4 TSS.

Fig. 5. Dependence of the ATP hydrolysis rate on DNA length. In all reac-tions, TFIIH (Rad3 E236Q) was preincubated with DNA for 40 min before ATPaddition. (A) The phosphate released (Pi) as a function of time for a range ofDNA lengths shows a linear dependence that can be fit to give a steady-staterate of ATP hydrolysis. (B) The steady-state rate of ATPase hydrolysis asa function of DNA length and DNA concentration. Concentration is plottedas micromolar-base pairs (μMbp), and the data are fit to Michaelis–Mentencurves, allowing for the extraction of KM and Vmax parameters. The Vmax ofthe plasmid data and its associated error are shown by two dashed hori-zontal lines. (C) KM as a function of DNA length. (D) Vmax as a function of DNAlength. Fit shown using analytical expression for the dependence of Vmax onDNA length for the model described in the text with a step size of 1 bp anda processivity of 10 bp (SI Materials and Methods).

3964 | www.pnas.org/cgi/doi/10.1073/pnas.1417709112 Fishburn et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021

pre–steady-state behavior in the ATPase rate (34). A slow stepat the end of the DNA predicts a burst of rapid ATP hydrolysison ATP addition compared with the steady-state rate. Thisburst phase arises because initial rounds of hydrolysis do notencounter the rate-limiting step that takes place at the DNA end.In contrast, a slow step only between DNA binding and trans-location predicts a lag in ATP hydrolysis rate. This lag phase arisesbecause the rate-limiting step occurs before translocation andbefore ATP hydrolysis occurs. Our data are consistent with thelatter case, as there is a lag in ATP hydrolysis after adding ATPto TFIIH prebound to DNA (Fig. 5A). The observed lag suggeststhat Ssl2, in complex with TFIIH, first binds DNA in an inactiveconformation and then, after ATP addition, isomerizes to theactive state before translocation.To extract quantitative measures of translocase parameters,

we analyzed the DNA length dependence of the Ssl2 ATPaseVmax with an expression derived from the isomerization model(SI Materials and Methods) (34). Assuming a translocation stepsize of 1 bp and a contact size on the DNA of 16 bp (Table S1 andMaterials and Methods), we determined a processivity of Ssl2 in thecontext of TFIIH of 0.90 with a fit error of 0.01 corresponding toan average of 10.0 ± 0.9 bp translocated before dissociation fromthe DNA. Because we have no independent measure of step size,we note that a larger step size would lead to a lower processivity.For example, analyzing the same data with a step size of 2 bpwould correspond to 5.3 ± 0.3 bp translocated before dissociationfrom the DNA.

DiscussionOur results on the translocase activity of Ssl2 in the context of theTFIIH complex are completely consistent with the translocasemodel of OC formation and explain how TFIIH might generatetorsional stress in promoter DNA, leading to unwinding. Ourfinding that Ssl2 tracks in the 5′ → 3′ direction implies that Ssl2uses the nontemplate strand of promoter DNA as it attempts totrack away from the PIC but instead reels downstream dsDNAtoward the PIC, creating open DNA via torsional or mechanicalstress that is then fed into the Pol II cleft (Fig. 6). Interestingly,this is the opposite direction to the characterized helicase activityof the Ssl2 subunit alone, suggesting that the enzyme does not usethe same motor mechanism for helicase and dsDNA translocaseactivities. Our conclusion that the processivity of Ssl2 translocationis similar to the amount of DNA unwound in the OC can explainwhy the Pol II OC is unstable. We speculate that DNA openingrequires continuous ATP hydrolysis because Ssl2 likely dissociatesfrom the DNA after translocating a short distance (while re-maining in complex with TFIIH and the PIC), resetting the initi-ation complex to its closed state. In contrast to the situation withother prokaryotic and eukaryotic RNA Pols, it seems likely that

the contacts between Pol II and unwound DNA are not strongenough to stabilize the fully unwound state in the absence of con-tinual Ssl2 translocation activity.This mechanism predicts multiple cycles of DNA opening and

closing during the initial stages of transcription. It was found thatthe Pol II OC is stabilized by a four-nucleotide RNA (8), so thismay be the minimum length of transcript to prevent reversion tothe PIC dsDNA state. However, translocase function may be nec-essary to assist in DNA opening until a 7- to 8-base RNA:DNAhybrid is formed. This model can explain the observation that TFIIHstimulates promoter escape from templates with a short stretch ofpremelted DNA from −9 to −1 with respect to the TSS (17).Finally, our results have important implications for the action of

XPB during NER. In the general genome repair pathway, TFIIHis recruited to DNA lesions bound by factor XPC, where it opensan asymmetric ∼27-bp DNA bubble surrounding the lesion (18,19). Both XPB and XPD are required for this DNA unwinding,although XPB does not seem to promote unwinding via a con-ventional helicase mechanism. Mutations that abolish the XPBATPase abolish DNA unwinding activity during NER, but twoXPB mutations with reduced 3′ → 5′ helicase function were re-portedly active for NER (13). Based on these results, it was pro-posed that XPB functions indirectly in DNA opening by usingits ATPase function to promote a conformational change in theXPC–DNA–TFIIH complex, physically wedging open 5 bp of theDNA duplex, and positioning the XPD helicase to open the DNAsurrounding the DNA lesion (10, 18, 19). Given our results, itseems more likely that XPB opens 5 bp of DNA using a dsDNAtranslocase mechanism similar to that in OC formation. By gen-erating torsional strain in rotationally fixed damaged DNA, theXPB subunit of TFIIH can lead to initial unwinding, generatinga substrate for XPD to generate the fully opened 27-bp asym-metric bubble surrounding DNA lesions.

Materials and MethodsDNA Helicase Assay. TFIIH helicase activity was monitored using a fluorescentdye-labeled oligonucleotide (IR700-TTCACCAGTGAGACGGGCAACAGCC) an-nealed to PAGE-purified 100-base oligonucleotides, with the resulting templateshaving 5′ or 3′ overhangs of 75 bases (Fig. 1). The assay was performed aspreviously described (36), with the following modifications: 10-μL reactionscontained 10 mM Hepes (pH 7.6), 100 mM potassium glutamate, 10 mM mag-nesium acetate, 3.5% (vol/vol) glycerol, 1 mMDTT, 1 μg BSA, 60 fmol holo-TFIIH,and 30 fmol template DNA. ATP or ATPγS was added to 1 mM, and reactionswere incubated 1–2 h at 26 °C. Controls received TE buffer (10 mM Tris, pH 7.5,1 mM EDTA) in lieu of ATP or were heated to 95 °C for 30 s. Reactions werequenched by the addition of 10 μL 33% glycerol, 40 mM EDTA, and 0.5% SDSand analyzed by PAGE using 10% acrylamide gels in 0.5× TBE buffer (134 mMTris, 44 mM boric acid, 2.5 mM EDTA, pH 8.8) plus 0.1% SDS. Following elec-trophoresis, gels were visualized using an Odyssey IR scanner (LI-COR).

Fig. 6. dsDNA translocase model for open complex formation. The Ssl2subunit of TFIIH tracks in the 5′ → 3′ direction on the nontemplate promoterDNA strand (red). Because TFIIH movement is constrained due to interactionwith other PIC components, translocation results in insertion and rotation ofpromoter DNA into the Pol II cleft, leading to DNA unwinding (right arrowsindicate rotation and direction of dsDNA movement). The short persistencelength of Ssl2/TFIIH predicts that the OC state is unstable, in agreement withexperimental observations (9, 10).

Table 1. Vmax of the Ssl2 ATPase with single- or double-stranded DNA

DNA Length [bp (ds) or nt (ss)] Vmax (ATP/Ssl2 s)

ds 30 5.19ds 40 6.41ds 60 8.82ds 80 9.49ss 30 1.03ss 40 1.20ss 60 2.46ss 80 2.58

Reactions with single-stranded DNA were carried out as described inMaterials and Methods and contained 12.5 μM nt of each template. Theaverage Vmax values from two independent experiments are given. For com-parison, the predicted Vmax from an analogous ds DNA template is shown(data from Fig. 4D).

Fishburn et al. PNAS | March 31, 2015 | vol. 112 | no. 13 | 3965

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021

In Vitro Transcription Assay. In vitro transcription using recombinant andpurified factors was performed similarly to previously described assays (32).See SI Materials and Methods for additional information.

TFIIH Purification.WT TFIIH was purified from strain SHY869 (RAD3-(HA)1-TAPtag, tfb6Δ) as previously described (32) except that, following the ultra-centrifugation step, potassium acetate was added to the extract to a finalconcentration of 0.6 M before binding to IgG-Sepharose (GE Healthcare).TFIIH with ATPase-defective Rad3 was purified by the same method fromstrain SHY887 (leu2Δ rad3Δ::KanMX, tfb6Δ::HPH) carrying plasmid pJF82 [arscen LEU2 RAD3 (E236Q)-(HA)1-TAP tag]. Because the Ssl2 E489Q mutation islethal, this TFIIH derivative was purified from a WT strain containing theTap-tagged Ssl2 mutation on a plasmid. Strain SHY861 (leu2Δ SSL2) carryingplasmid pJF62 [ars cen LEU2 ssl2 (E489Q)-(FLAG)1-TAP tag] was grown inglucose complete media lacking leucine, and TFIIH was purified by themethod described above.

Triplex Disruption Assay. Triplex DNA template formation and disruptionreactions were performed as previously described (28), with the followingmodifications: templates were assembled from duplex DNA containinga triplex target sequence (AAGAAAAGAAAGAAGAAAGAAA) and a fluores-cent or 32P-labeled TFO (TTCTTTTCTTTCTTCTTTCTTT). The DNA and TFO werecombined at 1 μM final concentration in 25 mM Mes (pH 5.5) and 10 mMMgCl2 and heated to 57 °C for 15 min and then cooled at 1 °C/min over 35 minto allow annealing. Triplex DNA was stored at -20 °C and diluted in 50 mMTris·HCl, pH 8.0, 10 mM MgCl2, and 1 mM DTT before the assay. Ten-microliter reactions contained 10 mM Hepes (pH 7.6), 100 mM potassiumglutamate, 10 mM magnesium acetate, 3.5% glycerol, 1 mM DTT, 1 μg BSA,0.01% Nonidet P-40, 15 fmol holo-TFIIH, and 30 fmol triplex DNA. ATP ordATP was added to 1 mM, and reactions were incubated at 26 °C for the

indicated times before stopping with 2.5 μL 5× GSMB (15% glucose, 3% SDS,250 mM Mops, pH 5.5, and 0.04% bromophenol blue). The reactions wereanalyzed by PAGE using 6% acrylamide gels with buffer: 40 mM Tris-acetate(pH 5.5), 5 mM sodium acetate, and 1 mM magnesium chloride. Gels werevisualized using either an Odyssey IR scanner (LI-COR) or dried and visualizedby PhosphorImager (Molecular Dynamics).

ATPase Assay. DNA-dependent ATPase activity of TFIIH (Rad3-E236Q) wasmeasured using a colorimetric assay kit (Innova; 601-0120). Forty-microliterreactions contained 10 mM Hepes (pH 7.6), 100 mM potassium glutamate,10 mMmagnesium acetate, 3.5% glycerol, 1 mMDTT, 4 μg BSA, 40 fmol holo-TFIIH, and 1.25 nM to 1.5 μM template DNA. After 40 min at room tem-perature, purified ATP was added to 0.5 mM, and reactions were incubated1–20 min at 26 °C. Reactions were stopped by the addition of 10 μL gold mixand, after 4 min, 4 μL stabilizer 2. After 30 min at room temperature, ab-sorbance was measured at 635 nm and plotted against DNA concentration.A standard curve was established for every experiment using the kit-includedphosphate standard and used to determine TFIIH catalyzed ATP hydrolysis.Templates from 30 to 80 bp were tested at nine concentrations in triplicate,spanning a 1,200-fold range of DNA concentration. See SI Materials andMethods for information on extracting translocation kinetic parameters fromthe ATPase data.

ACKNOWLEDGMENTS. We thank members of the S.H. laboratory and TedYoung for comments and suggestions during the course of this workand S. Grünberg for Fig. 6 and comments on the manuscript. This workwas supported by National Institute of General Medical Sciences Grant2RO1GM053451 (to S.H.) and National Science Foundation-Molecular andCellular Biosciences Grant 1243918 (to E.G.).

1. Saecker RM, Record MT, Jr, Dehaseth PL (2011) Mechanism of bacterial transcription

initiation: RNA polymerase - promoter binding, isomerization to initiation-competent

open complexes, and initiation of RNA synthesis. J Mol Biol 412(5):754–771.2. Grünberg S, Hahn S (2013) Structural insights into transcription initiation by RNA

polymerase II. Trends Biochem Sci 38(12):603–611.3. Cheung ACM, Sainsbury S, Cramer P (2011) Structural basis of initial RNA polymerase II

transcription. EMBO J 30(23):4755–4763.4. Sainsbury S, Niesser J, Cramer P (2013) Structure and function of the initially tran-

scribing RNA polymerase II-TFIIB complex. Nature 493(7432):437–440.5. Zhang Y, et al. (2012) Structural basis of transcription initiation. Science 338(6110):

1076–1080.6. Bunick D, Zandomeni R, Ackerman S, Weinmann R (1982) Mechanism of RNA poly-

merase II—specific initiation of transcription in vitro: ATP requirement and uncapped

runoff transcripts. Cell 29(3):877–886.7. WangW, Carey M, Gralla JD (1992) Polymerase II promoter activation: Closed complex

formation and ATP-driven start site opening. Science 255(5043):450–453.8. Holstege FC, Fiedler U, Timmers HT (1997) Three transitions in the RNA polymerase II

transcription complex during initiation. EMBO J 16(24):7468–7480.9. Dvir A, et al. (1996) A role for ATP and TFIIH in activation of the RNA polymerase II

preinitiation complex prior to transcription initiation. J Biol Chem 271(13):7245–7248.10. Compe E, Egly J-M (2012) TFIIH: When transcription met DNA repair. Nat Rev Mol Cell

Biol 13(6):343–354.11. Feaver WJ, et al. (1993) Dual roles of a multiprotein complex from S. cerevisiae in

transcription and DNA repair. Cell 75(7):1379–1387.12. Guzder SN, et al. (1994) DNA repair gene RAD3 of S. cerevisiae is essential for tran-

scription by RNA polymerase II. Nature 367(6458):91–94.13. Oksenych V, Bernardes de Jesus B, Zhovmer A, Egly J-M, Coin F (2009) Molecular insights

into the recruitment of TFIIH to sites of DNA damage. EMBO J 28(19):2971–2980.14. Tirode F, Busso D, Coin F, Egly JM (1999) Reconstitution of the transcription factor

TFIIH: Assignment of functions for the three enzymatic subunits, XPB, XPD, and cdk7.

Mol Cell 3(1):87–95.15. Coin F, et al. (1998) Mutations in the XPD helicase gene result in XP and TTD phe-

notypes, preventing interaction between XPD and the p44 subunit of TFIIH. Nat

Genet 20(2):184–188.16. Dvir A, Conaway RC, Conaway JW (1997) A role for TFIIH in controlling the activity

of early RNA polymerase II elongation complexes. Proc Natl Acad Sci USA 94(17):

9006–9010.17. Moreland RJ, et al. (1999) A role for the TFIIH XPB DNA helicase in promoter escape by

RNA polymerase II. J Biol Chem 274(32):22127–22130.18. Egly J-M, Coin F (2011) A history of TFIIH: Two decades of molecular biology on

a pivotal transcription/repair factor. DNA Repair (Amst) 10(7):714–721.

19. Fuss JO, Tainer JA (2011) XPB and XPD helicases in TFIIH orchestrate DNA duplexopening and damage verification to coordinate repair with transcription and cellcycle via CAK kinase. DNA Repair (Amst) 10(7):697–713.

20. Kim TK, Ebright RH, Reinberg D (2000) Mechanism of ATP-dependent promotermelting by transcription factor IIH. Science 288(5470):1418–1422.

21. Miller G, Hahn S (2006) A DNA-tethered cleavage probe reveals the path for promoterDNA in the yeast preinitiation complex. Nat Struct Mol Biol 13(7):603–610.

22. Grünberg S, Warfield L, Hahn S (2012) Architecture of the RNA polymerase II pre-initiation complex and mechanism of ATP-dependent promoter opening. Nat StructMol Biol 19(8):788–796.

23. He Y, Fang J, Taatjes DJ, Nogales E (2013) Structural visualization of key steps inhuman transcription initiation. Nature 495(7442):481–486.

24. Murakami K, et al. (2013) Architecture of an RNA polymerase II transcription pre-initiation complex. Science 342(6159):1238724.

25. Lin YC, Choi WS, Gralla JD (2005) TFIIH XPB mutants suggest a unified bacterial-likemechanism for promoter opening but not escape. Nat Struct Mol Biol 12(7):603–607.

26. Goel S, Krishnamurthy S, Hampsey M (2012) Mechanism of start site selection by RNApolymerase II: Interplay between TFIIB and Ssl2/XPB helicase subunit of TFIIH. J BiolChem 287(1):557–567.

27. Lammens A, Schele A, Hopfner K-P (2004) Structural biochemistry of ATP-driven di-merization and DNA-stimulated activation of SMC ATPases. Curr Biol 14(19):1778–1782.

28. Firman K, Szczelkun MD (2000) Measuring motion on DNA by the type I restrictionendonuclease EcoR124I using triplex displacement. EMBO J 19(9):2094–2102.

29. Saha A, Wittmeyer J, Cairns BR (2002) Chromatin remodeling by RSC involves ATP-dependent DNA translocation. Genes Dev 16(16):2120–2134.

30. Whitehouse I, Stockdale C, Flaus A, Szczelkun MD, Owen-Hughes T (2003) Evidencefor DNA translocation by the ISWI chromatin-remodeling enzyme. Mol Cell Biol 23(6):1935–1945.

31. Lohman TM, Tomko EJ, Wu CG (2008) Non-hexameric DNA helicases and translocases:Mechanisms and regulation. Nat Rev Mol Cell Biol 9(5):391–401.

32. Fishburn J, Hahn S (2012) Architecture of the yeast RNA polymerase II open complexand regulation of activity by TFIIF. Mol Cell Biol 32(1):12–25.

33. Young MC, Kuhl SB, von Hippel PH (1994) Kinetic theory of ATP-driven translocaseson one-dimensional polymer lattices. J Mol Biol 235(5):1436–1446.

34. Fischer CJ, Saha A, Cairns BR (2007) Kinetic model for the ATP-dependent trans-location of Saccharomyces cerevisiae RSC along double-stranded DNA. Biochemistry46(43):12416–12426.

35. Khaki AR, et al. (2010) The macroscopic rate of nucleic acid translocation by hepatitisC virus helicase NS3h is dependent on both sugar and base moieties. J Mol Biol 400(3):354–378.

36. Sung P, Prakash L, Matson SW, Prakash S (1987) RAD3 protein of Saccharomycescerevisiae is a DNA helicase. Proc Natl Acad Sci USA 84(24):8951–8955.

3966 | www.pnas.org/cgi/doi/10.1073/pnas.1417709112 Fishburn et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

2, 2

021