Saima Nawaz, Gregory E D Mullen, Kavitha Sunassee ...10.1186...1 Supplementary data for: Simple, mild, one-step labelling of proteins with 68Ga using a tris(hydroxypyridinone) bifunctional

1

Supplementarydatafor:Simple,mild,one-steplabellingofproteinswith68Gausingatris(hydroxypyridinone)

bifunctionalchelator:68Ga-THP-scFvtargetingtheprostatespecificmembraneantigen

SaimaNawaz,GregoryEDMullen,KavithaSunassee,JayantaBordoloi,PhilipJBlower,and

JamesRBallinger

2

Methods

Proteinpurificationandanalysis

Sizeexclusionchromatography:PurificationofJ591c-scFvanditsTHP-malconjugatewas

performedusingFPLCsizeexclusioncolumnchromatographyusingaSuperdex7510/300

GLcolumn(GE),elutingwithphosphatebufferedsaline(pH7)ataflowrateof0.5ml/min

andmaximumpressureof1.8MPa.

PolyacrylamidegelelectrophoresisandWesternblottingwereusedtoevaluatepurityofthe

size-exclusion-purifiedJ591c-scFv,usingpre-castpolyacrylamidegels(NuPAGE12%Novex

Bis-TrisminigelswithMESbuffer,LifeTechnologies)andWesternblotting.Forthe

preparationofsamples,10µlaliquotsofJ591-scFvwereplacedinmicrocentrifugetubes

and5µlofNuPAGELDSloadingbuffer(4x)wasadded.Forreducedsamples,thereducing

agent(NuPAGEreducingagentofstrength10x,2µl)wasaddedalongwithNuPAGELDS

bufferof4xconcentration(5µl).Sampleswereplacedonaheatingblockfor5minat95oC

andthencooledonicefor1min.TheNuPAGEchamberswerefilledwiththeMESbuffer(50

mlof20xMESbuffer+950mlofdeionisedwater).Novexsharppre-stainedprotein

standardwasloadedinoneofthewells.Thegelwasrunfor40minatanominalvoltageof

200Vwithconstantcurrentof125mA,developedwithSimplyBlueSafeStain(Thermo

FisherScientific)for2handdestainedwithwaterovernight.Inthecaseofradiolabelled

proteins,beforestaining,activitymarkerswereplacedonthegelswhichwerethenimaged

withaphosphorimager.

ForWesternblotting,the12%gelwaspreparedasabove;insteadofstaining,thegelwas

loadedonanitrocellulosetransfermembraneanddevelopedinnon-reducingtransfer

buffer(NuPAGEtransferbuffer(20x)50ml+methanol(100ml)+deionisedwater(850ml))

for1hatconstant30Vandcurrentof170-130mA.After1hthetransfermembranewas

3

removedandplacedinablockingagent(4%milkin50mlofPBSwith0.05%Tween20)for

90min,washedbrieflywithPBS,thenplacedinasolutioncontainingantibodyagainstHis-

tag(6.3µlofantibodydiluted(1:4000dilution)byadding25mlofPBSwith0.5%BSA,which

is0.125g/25mlofBSA)overnightat4oCwithshaking.Thetransfermembranewaswashed

fourtimeswith10mlofPBST(1lofphosphatebufferedsalinewith0.05%Tween20)before

incubatingwithsecondaryantibody(goatanti-mousehorseradishperoxidase)for60min

(1:5000dilution,approximately4µlofHRPin20,000µlofPBS).Thetransfermembranewas

washedwithPBSTfourtimesbeforeimmersingitinthesolutionofsubstrateDAB[(3,3-

diaminobenzidine)PeroxidaseSubstrateTabletSetfromSigmaAldrich](oneDABtabletand

oneUreaHydrogenPeroxidetabletdissolvedin15mlofultrapurewater).Thetransfer

membranewasdevelopedinthedarkuntilthebrown–blackprecipitationwasvisible.

Instantthinlayerradiochromatography(ITLC)

Proteinlabellingwithgallium-68wasassessedbyinstantthinlayerchromatography(ITLC)

usingITLC-SA(Agilent)developedwith0.1Msodiumcitratebuffer(pH6)asthemobile

phase.Undertheseconditionsunboundgallium-68movedwiththesolventfront(Rf=1)

andradiolabelledconjugatedproteinremainedattheorigin(Rf=0).

Highperformanceradiochromatography

Theradiolabelledproteinconjugateswereanalysedusingasize-exclusioncolumn(SEC-

2000),samplevolume20µl,elutingwithphosphatebufferedsaline(pH7),flowrate1

ml/min,UVdetectionat280nmandgammadetectionwithasodiumiodidedetector.

Serumstability

THP-J591c-scFv(200µl,0.4µg/µl)wasradiolabelledbyincubationwith300µlof68Ga

generatoreluate(60MBq)for20minatroomtemperatureandlabellingefficiencyfoundto

be>99%byITLC.Withoutfurtherpurificationormodification,theradiolabelledconjugated

4

J591c(100µl)wasmixedwithserum(200µl)andphysiologicalsaline(100µl).Themixture

wasincubatedat37oCandsamples(5µleach)werecollectedat30,60,120,180,240and

300minandstoredat-80oCuntilallsampleshadbeencollected.Thebehaviourof

unchelatedGa-68inserumwasdeterminedbyincubating10µlofammoniumacetate

buffered68Gageneratoreluatewithserumasdescribedabove,incubatingforupto60min

at37oC,taking5µlsamplesandstoringthemat-80oCuntilallthetimepointswere

collectedforgelelectrophoresisanalysis.68Ga-THP-mal-J591c-scFvincubatedinsalineand

sampledsimilarlywasusedasareference.ForanalysisbySDS-PAGE,thesamples(5µl)

weremixedwith10µlofLDSbuffer(lithiumdodecylsulfate,pH8.4)andappliedonthegel

(10µl).ThegelwasdevelopedinMESbuffer(2-ethanesulfonicacid)for40minat200Vand

120-130A.Afterwardsthegelwasremoved,activitymarkerswereplacedanditwas

analysedbyphosphorimagerandthenstainedwithCoomassieblue.

PETimagingandquantification

Dynamicimaging(n=1eachgroup)wasperformedoverthetimeperiodof3hfromthe

timeofinjection,25daysaftertumourinoculation,usingaBioScannanoPET-CTPLUS

(Mediso)scannerusingtheirproprietaryacquisitionsoftware(Nucline,version2.00).CTwas

performedwithanx-raytubevoltageof45kVp,600msofexposuretime,and360

projections.Thisscantook10mintoobtain.DynamicPETscanswereacquiredwithina

94.7-mmfieldofviewfrom0to190minaftertailveininjectionofthetracer.Acquisition

tookplacein1–5coincidencemodewithacoincidencewindowof5nsanda400-to600-

keVenergywindow.ThedynamicPETdatawerereconstructedusingNuclinesoftware

(version2.00).Imageswereconstructedbasedon0.4mm3voxelsforPETand0.21mm3for

CT.ImageprocessingandanalysiswereperformedusingVivoquantsoftware(version1.23).

Beforeanalysis,bothCTandPETimageswererealignedandprocessedtoavoxelsizeof

5

0.21mm3andthePEToutputcalibratedtodisplayMBqpervoxel.Regionsofinterest(ROIs)

foreachdatafilewereproducedusingfreehandsegmentation.SUVmaxvalueswere

obtainedfromregionsofinterestnearthecentreofeachselectedorgan/tumour,usingthe

totalactivitywithintheimage(excludingactivitywithinthetail)asthetotalinjecteddose.

Time–activitycurveswereproducedfrom15-minbins.

Results

SupplementaryTableS1.DeconvolutedelectrospraymassspectraofJ591c-scFvpre-and

post-reductionandconjugationwithTHP-mal.

Sample Deconvolutedmass Assignment

J591c-scFvpre-TCEPtreatment 27923 J591c-scFvcysteinedisulfideconjugate

J591c-scFvpost-TCEPtreatment 27804 J591c-scFvfreethiolform

THP-mal-J591c-scFv 28724 THP-mal-J591c-scFv

THP-mal 920.4 THP-mal

6

SupplementaryFigureS1.ElectrospraymassspectraofJ591c-scFvpre-TCEPtreatment(top,27923.2correspondstodisulfideformedfromJ591clinkedtocysteineviaadisulfidebond);J591c-scFvpost-TCEPtreatment(middle,27803.9correspondstoJ591-scFv),andTCEP-treatedJ591-scFvafterincubationwithTHP-mal(bottom,28723.7correspondstoTHP-mal-J591-scFvadduct).

7

SupplementaryFigureS2.TheeffectofdifferentmolarexcessesofTCEPondimerisationof

J591-scFv.Lanes:1:molecularweightmarkers;2:non-reducedprotein;3:reducedprotein

(NuPAGEreducingagent);4:0.4:1molarratioTCEP:protein;5:1:1molarratio;6:2:1molar

ratio;7:5:1molarratio;8:10:1molarratio;9:15:1molarratio;10:20:1molarratio;11:

30:1molarratio.

SupplementaryFigureS3.FPLCpurificationofTHP-mal-J591-scFvconjugateonSuperdex

75HR10/30sizeexclusioncolumnelutedwithphosphate-bufferedsalineat0.5ml/min.

ConjugateelutesbeforefreeTHP-mal.FreeTHP-malcanbeseenelutingat15-18min,

monomericTHP-conjugatedproteinat10min,anddimericproteinat8min.

60KDa

30KDa

Free THP-mal

THP-mal-J591c-scFv

8

SupplementaryFigureS4.Exemplardatashowingradiolabellingefficiency(%proteinbound,determinedbyITLC)atdifferentproteinconjugateconcentrationsandtimes.Atconcentrationsof0.25mg/mlorhigher,labellingefficiencywas100%atalltimepointsfrom10sonward(thefirstdatapointineachseriesrepresentsasampletakenafter10sincubation).

0

20

40

60

80

100

120

0 500 1000 1500 2000 2500 3000 3500 4000

Timeandconcentrationdependenceofradiolabelling

%labelling, 0.25mg/ml %labelling, 0.125mg/ml

9

SupplementaryFigureS5.SDS-PAGEanalysisofserumstabilityof68Ga-THP-mal-J591-scFv.

A:radioactivegelanalysedwithCyclonephosphorimager.B:gelstainedwithCoomassie

blue.Lanes:1:molecularweightmarkers;2:humanserumincubatedwithammonium

acetatebuffered68Gaeluate;3:Radiolabelled68Ga-THP-mal-J591-scFvconjugatecontrol

(withoutserumincubation);4:conjugateincubatedwithhumanserumfor1min;5:30min;

6:60min;7:120min;8:180min;9:240min;10:360min.

SupplementaryFigureS6.Time course of 68Ga-THP-mal-J591c-scFv activity (SUVmax) in

DU145-PSMA xenografts derived from serial images in a single mouse determined by PET

imaging. A region of interest near the centre of each organ was drawn from which SUVmax

was determined in each 15 min bin.

0 50 100 150 200 2500

2

4

6

791113151719

Urinary BladderTumour

Lt. kidney

LiverRt. Kidney

DU145-PSMA

min

SUV m

ax

10

PSMA- PSMA- PSMA+ PSMA+organ ave SD ave SDblood 1.3 0.2 0.8 0.4tumour 0.5 0.2 5.4 0.5stomach 0.2 0.1 0.5 0.3smallintestine 0.4 0.1 0.6 0.2largeintestine 0.3 0.2 0.8 0.2spleen 3.0 1.4 3.4 0.7liver 3.4 1.5 4.9 1.1kidneys 41.5 12.7 89.3 10.0heart 0.6 0.1 1.5 0.3lungs 3.4 1.9 4.4 1.5trachea&thyroid 0.7 0.3 1.6 0.5salivarygland 0.4 0.2 0.9 0.2muscles 0.2 0.1 0.4 0.2bone 0.3 0.1 1.7 2.0reproductiveorgan 0.4 0.2 0.7 0.2tail 11.4 13.6 6.9 9.8skin&fur 0.6 0.3 0.8 0.2SupplementaryTableS2.Exvivobiodistributiondata(%ID/g)for68Ga-THP-mal-J591-scFvinmicebearingDU145(PSMA-)andDu145-PSMA(PSMA+)(meanandstandarddeviationareshown;n=4ineachgroup).

11

SupplementaryFigureS7.Exvivobiodistributiondatafor68Ga-THP-mal-J591-scFvinmicebearingDU145(PSMA-,blue)andDU145-PSMA(PSMA+,red),90minpost-injection.Errorbarsrepresentstandarddeviation(n=4pergroup).Dataarethesameasthoseshowninthemainmanuscript(Fig.5)butexpandedtoincludekidneys.

0

20

40

60

80

100

120Exvivobiodistribution

PSMA- PSMA+