Development of a quantitative enzyme linked immunosorbent assay for monitoring the Enterovirus 71 vaccine manufacturing process

(1)

AUTHOR QUERY FORM

Journal: VIRMET

Please e-mail or fax your responses and any corrections to:

E-mail:

corrections.esch@elsevier.thomsondigital.com

Article Number: 11579

Fax: +353 6170 9272

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen

annotation in the PDF file) or compile them in a separate list. To ensure fast publication of your paper please return your

corrections within 48 hours.

For correction or revision of any artwork, please consult

http://www.elsevier.com/artworkinstructions

.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in

the proof. Click on the ‘

Q

’ link to go to the location in the proof.

Location in

Query / Remark:

click on the Q link to go

article

Please insert your reply or correction at the corresponding line in the proof

Reference(s) given here were noted in the reference list but are missing from the text – please position

each reference in the text or delete it from the list.

Q1

Uncited reference: This section comprises references that occur in the reference list but not in the body

of the text. Please position each reference in the text or, alternatively, delete it. Any reference not dealt

with will be retained in this section.

(2)

VIRMET115791

ARTICLE IN PRESS

G Model

JournalofVirologicalMethodsxx(2011)xxx–xxx

ContentslistsavailableatScienceDirect

Journal

of

Virological

Methods

j ourna l h o me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / j v i r o m e t

Highlights

JournalofVirologicalMethodsxx (2011)xxx–xxx

Developmentofaquantitativeenzymelinkedimmunosorbentassayfor monitoringtheEnterovirus71vaccinemanufacturingprocess

Chia-ChyiLiu,Hsuen-WenChang,GraceYang,Jen-RonChiang,Yen-HungChow,I-HsiSai,Jui-YuanChang,Sue-ChenLin,CharlesSia,

Chia-HsinHsiao,Ai-HsiangChou∗_,_Pele_Chong∗∗

AquantitativemeasurementofVP2subunitthroughoutthevaccineproductioncyclewasdeveloped.Theinfectiousparticlesdetermined byTCID50assaywasnotdirectlycorrelatedwithVP2content.TheVP2contentandthemagnitudeofneutralizingtiterswerefoundtobe dose-dependent.

(3)

JournalofVirologicalMethodsxxx (2011) xxx–xxx 1

ContentslistsavailableatScienceDirect

Journal

of

Virological

Methods

j o ur na l h o me p ag e :w ww . e l s e v i e r . c o m / l o ca t e / j v i r o m e t

Development

of

a

quantitative

enzyme

linked

immunosorbent

assay

for

monitoring

the

Enterovirus

_,

_Pele

_Chong

a,c,∗∗

3 4

a_Vaccine_R&D_Center,_Na_tional_Health_Research_Institutes,_Zhunan_Town,_Miaoli_County_350,_Taiwan

5

b_Vaccine_Center,_Taiwan_Centers_for_Diseases_Control,_Taipei,_Taiwan 6

c_Graduate_Institute_of_Immunology,_China_Medical_University,_Taichung_402,_Taiwan 7 8 9 Articlehistory: 10 Received7March2011 11

Receivedinrevisedform25May2011 12 Accepted1June2011 13 Available online xxx 14 Keywords: 15 Enterovirus71 16 ELISA 17 Viralantigen 18

Processanalyticaltechnology 19

Neutralizationepitope 20

a

b

s

t

r

a

c

t

Enterovirus71(EV71),theetiologicagentcausesoutbreakswithsignificantmortalityinyoungchildren inAsiaandcurrentlythereisnovaccineavailable.Inthisstudy,wereportaquantitativeenzymelinked immunosorbentassay(Q-ELISA)todeterminetheconcentrationoftheEV71VP2antigen.EV71 virus-likeparticles(VLPs)wereproducedinthebaculovirusexpressionsystemandusedastheEV71antigen referencestandard.AntiserafrombothEV71-immunizedchickensandrabbitswereveryefficientand usefulascaptureantibodiestobindvariousformsofEV71antigens,whereasacommercialVP2-specific virusneutralizingmonoclonalantibodyMAB979wasfoundtobesuitableforquantifyingtheamount ofVP2antigen.ThisQ-ELISAwasusedsuccessfullytodetermineVP2contentateachstageofEV71 vaccinemanufacturingprocess,particularlyduringtheupstreamharvest,downstreampurificationand viralinactivationsteps.TheamountofVP2antigenandthemagnitudeofneutralizingtiterswerefound tobedose-dependentinmiceimmunizedwithvaccinecandidates.TheseresultsindicatethatQ-ELISA couldprovideoff-linetimelyquantitativemeasurementsofVP2antigenthroughouttheproductioncycle toevaluatecriticalattributesandconditionsthatmayaffectvirusyieldsinculturemedia,thequalityof purificationmethods,thestabilityandpotencyoffinalvaccineformulations.

1. Introduction

21

Since 1969the ﬁrst case of Enterovirus 71 (EV71)infection

22

wasrecordedinCA,USA(Schmidtetal.,1974),EV71hasinfected

23

periodicallyyoungchildrenthroughouttheworld(forreview,see

24

LeeandChang,2010;Xuetal.,2010).Althoughthemajorityof

25

EV71infectionsinadultsareasymptomatic,EV71infection

man-26

ifestsmost frequentlyashand-foot-and-mouth disease(HFMD)

27

orhyperanginainyoungchildren,whoarepotentiallyatriskfor

28

severeneurological complications, includingaseptic meningitis,

29

cerebellaencephalitisandacuteﬂaccidparalysis(AFP),thatmay

30

leadtooccasionaldeaths(Chumakovetal.,1979;Alexanderetal.,

31

1994; Gilbert et al., 1988; Chang et al., 1988; AbuBakar et al.,

32

1999;Hoetal.,1999;McMinn,2003).EV71hasnowemergedas

33

∗ Correspondingauthor.Tel.:+88637246166;fax:+88637583009.

∗∗ Correspondingauthorat:VaccineR&DCenter,NationalHealthResearch

Insti-tutes,ZhunanTown,MiaoliCounty350,Taiwan.Tel.:+88637246166; fax:+88637583009.

E-mailaddresses:Hsiang@nhri.org.tw(A.-H.Chou),pelechong@nhri.org.tw

(P.Chong).

1 _These_authors_equally_contributed_to_this_work.

animportantneurotropicvirus,buttherearenoeffectivemed- 34

icationsand aprophylacticvaccine (McMinn,2002; Qiu,2008). 35

EV71vaccinedevelopmenteffortshavebeenundertakenmainlyin 36

Asia,andpromisingpreclinicalimmunogenicityresultshavebeen 37

obtainedusingseveralprototypevaccinecandidates.Theseinclude 38

ﬁndingsfromstudieswithsmallanimalsimmunizedwithchem- 39

icallyinactivatedEV71(LeeandChang,2010;Xuetal.,2010;Liu 40

etal.,2011),a formulationofviralcapsidproteinpeptides(Foo 41

etal.,2007),aDNAplasmidcarryingthemajorviralcapsidpro- 42

teinVP1(Tungetal.,2007),virus-like-particles(VLPs)formedby 43

thephysicalassociationofthefourEV71capsidproteinsVP1,VP2, 44

VP3 and VP4 (Chunget al., 2008), and VP1-enrichedmilk pro- 45

ducedbytransgenicmicecontainingtheVP1 gene(Chenetal., 46

2008).In light of thesuccessof the Salk-inactivatedpoliovirus 47

vaccine,theproductionofaninactivatedwholevirionEV71vac- 48

cineis feasible, andthe vaccinecouldbe licensedifan animal 49

potencyassayanda methodforquantifyingviralantigenswere 50

available.Apartfromvaccineformulationpurposes,determining 51

viralantigenyieldsincrudecellculturesupernatantsand puri- 52

ﬁedviruspreparationsisaveryimportantparametertoguidethe 53

scaling-upofthedownstreamprocess.Vaccinemanufacturersare 54

beingchallengedcurrentlytomodernizetheirscientiﬁcprocesses 55

for optimizing theﬁnal product quality and improvingprocess 56

(4)

Pleasecitethisarticleinpressas:Liu,C.-C.,etal.,Developmentofaquantitativeenzymelinkedimmunosorbentassayformonitoringthe Enterovirus71vaccinemanufacturingprocess.J.Virol.Methods(2011),doi:10.1016/j.jviromet.2011.06.001

ARTICLE IN PRESS

GModel VIRMET115791–9

2 C.-C.Liuetal./JournalofVirologicalMethodsxxx (2011) xxx–xxx

yields.In this report,wedescribea quantitativeenzyme-linked

57

immunosorbentassay(Q-ELISA)todeterminetheconcentration

58

ofEV71antigenusingdifferentin-houseEV71viruspreparations,

59

EV71-speciﬁcantisera and a commercially-available EV71

VP2-60

speciﬁcvirusneutralizingmonoclonalantibody.ThisQ-ELISAhas

61

beenusedsuccessfullytoassessvirusproductionyieldsand

mea-62

suretheamountofEV71VP2epitopeinculturesupernatants,the

63

upstreamharvest,downstreampuriﬁcationfractionsandvaccine

64

bulk.

65

2. Materialsandmethods

66

2.1. Ethicsstatement

67

Allexperimentswereconductedinaccordancewiththe

guide-68

linesof theLaboratory AnimalCenter of NHRI.Theanimal use

69

protocolshave beenreviewedand approvedbytheNHRI

Insti-70

tutionalAnimalCareandUseCommittee(Approvedprotocolno.

71

NHRI-095054-A).

72

2.2. Cells,mediaandvirus

73

Africangreenmonkeykidney(Vero)cellswerekindlyprovided

74

bythe Taiwan Centers of Disease Control(Taiwan CDC) which

75

obtainedtheoriginalVerocellline(passage#125)fromthe

Ameri-76

canTypeCultureCollection(ATCC,Rockville,MD,USA).TheE59

77

strain(genotypeB4), theclinicalisolateoftheEV71virus, was

78

obtainedfromtheTaiwanCDC.EV71/E59virusstockswere

col-79

lectedfromthesupernatantsofinfectedVerocellsthreedayspost

80

infection(DPI).Thetitersof virusstocksweredeterminedby a

81

plaqueassay,andthesestockswerestoredat−80◦_C.

82

2.3. ProductionofEV71virus-likeparticle(VLP)usingthe

83

baculovirusexpressionsystem

84

EV71VLP wasproduced using therecombinant baculovirus

85

expressionsystemdescribedinChungetal.(2008),withasmall

86

modiﬁcation:theP1genewasderivedfromtheEV71E59virus

87

isolate.TotalproteinconcentrationwasdeterminedbytheBCA

88

proteinassay(Biorad,Hercules,CA,USA)andquantitativeamino

89

acidanalysisthatwasperformedbyUBI-Asia,Taiwan.

90

2.4. ProductionofEV71vaccinebulkusingrollerbottles

91

TheproductionofEV71/E59vaccinebulkwasperformedusing

92

theroller bottletechnology. In brief, Verocells were grown in

93

850cm2_roller_bottles_(Corning_Life_Science,_Corning,_NY,_USA)_in

94

200mLofVP-SFMmedium(GIBCO,Carlsband,CA,USA),andbottles

95

wererotatedat0.33rpmat37◦Cona100-bottlerollerrack.Each

96

rollerbottleculturewasinoculatedwith1.5–2×107_cells,_and_the

97

celldensityusuallyreached1.5–2×108_cells_after_six_days_of

culti-98

vation.Afterculturemediumreplacement,Verocellsineachbottle

99

wereinfectedwithEV71/E59atanm.o.iof10−5.Productionbatches

100

weretypicallyobtainedfrom200mL×200mLor100mL×400mL

101

rollerbottles in each run.EV71was collectedfromtheculture

102

supernatantofeachbottleattheﬁfthDPI.Celldebriswereremoved

103

by ﬁltration through a 0.65-␮m membrane (Sartorius Stedim

104

Biotech,Haywood,CA,USA),andthecrudevirusbulkwas20-to

105

40-foldconcentratedusinga100-kDacut-offdiaﬁltration

mem-106

branein atangentialﬂowﬁlter(TFF)cassette(Sartorius Stedim

107

Biotech).EV71waspuriﬁedusinganAKTAPilotliquid

chromatog-108

raphysystem(GE Healthcare,SaltLakeCity,UT,USA)equipped

109

withSepharoseFastFlow6gel.Thecolumn(200mm×900mm)

110

waspackedwith26Lofgel.PBSwasusedastheelutingbuffer

111

and theﬂow-rate wasset at80mLper min.Fractions (160mL

112

perfraction)werecollectedandanalyzedbyimmunoblotting,and

113

virusinfectivitywasmeasuredusingtheTCID50 assay.Fractions 114

containingtheviruswerepooled,concentratedfurtherandthen 115

inactivatedwith0.2%formalin(v/v).Thevaccinebulkwasobtained 116

aftersterileﬁltrationusinga0.22-␮mﬁlter.Thetotalproteincon- 117

centrationof thevaccine bulkwasalsodeterminedbytheBCA 118

proteinassay. 119

2.5. ProductionofEV71virusinbioreactor 120

The production of EV71/E59 virususing serum-freeVP-SFM 121

mediumina BIOFLO310bioreactor(NBS,New Jersey,NJ,USA) 122

wasbasedonthemicrocarriercellculturebioprocesspreviously 123

reportedbyChangetal.(2011).Thetotalproteinconcentrationof 124

thepuriﬁedvirusbulkwasdeterminedbytheBCAproteinassay. 125

HalfofthepuriﬁedEV71/E59bulk(10mL)wasstoredat−80◦_C ₁₂₆

in0.5-mL aliquots;theotherhalfwasinactivatedby0.2%(v/v) 127

formalinat37◦Cfor3daysandstoredat4◦C. 128

2.6. SDS-PAGEandWesternblotanalyses 129

SDS-PAGEandWesternblotanalysesofthepuriﬁedEV71/E59 130

bulkwereperformedaccordingtheprotocolreportedpreviously 131

byChangetal.(2011). 132

2.7. DensitometricanalysisofCoomassieblue-stainedEV71 133

proteinsresolvedbySDS-PAGE 134

A control calibrationcurve was established with increasing 135

amountsofbovinealbuminstandard(Biorad)usingSDS-PAGEand 136

Coomassiebluestaining.Inthesamegel,5,10and20␮Lofeither 137

baculovirus-expressedEV71VLPpreparedaccordingtoChungetal. 138

(2008)orasucrosegradient-puriﬁedEV71/E59bulkwereincluded 139

toestimatetheEV71proteinconcentration.Coomassiebrilliant 140

blueR-250-stainedalbuminandEV71proteinbandswerescanned 141

individuallyat150dotsperinch(DPI)settingusingtheHewlett 142

PackardScanJet4890toobtaintheirrespectiveintegratedopti- 143

cal density (IOD) values. IOD recorded for each of the protein 144

bandsrepresentedthetotalamountofproteinmoleculesbecause 145

Coomassiebluebindsstoichiometricallytoproteins(Hames,1988). 146

EV71proteinIODsobtainedfromthealbumincalibrationcurve 147

representedtheestimatedconcentrationoftheviralprotein. 148

2.8. Determinationofvirustiters 149

Virustitersweredeterminedusingthemedianendpointofthe 150

tissue culture’sinfectious dose(TCID50)asdescribed previously 151

byLiuetal.(2011)TheTCID50values werecalculatedusingthe 152

Reed–Muenchmethod(ReedandMuench,1938). 153

2.9. PreparationofinactivatedEV71/E59andanimal 154

immunogenicitystudies 155

The EV71/E59 bulk chromatographicallypuriﬁed from roller 156

bottlesupernatantswasinactivatedwitha37%formaldehydesolu- 157

tion(Merck,WestPoint,NY,USA)ataratioof4000:1(v/v)either 158

at4◦Cfor45daysor37◦Cfor3days.Differentconcentrationsof 159

inactivatedvirusproteinwereadsorbedon5mLofaluminumphos- 160

phate(1.5mgofaluminum)atroomtemperaturefor3hbefore 161

immunization.Agroupof sixfemaleBALB/cmice(18–25g,6–8 162

weeksold)wasimmunizedintramuscularly(i.m.)with0.2mLof 163

thealum-adsorbedinactivatedEV71immunogens,andtheywere 164

boostedwiththesamedoseeverytwoweeksafterpriming.Immu- 165

nizedmicewerebledtwoweeksaftertheboost,andtheserum 166

wascollectedforavirusneutralizationassaybasedontheTCID50 167

determination.Inparallel,tworabbitswereimmunizedi.m.three 168

(5)

C.-C.Liuetal./JournalofVirologicalMethodsxxx (2011) xxx–xxx 3

Fig.1.CoomassiebluestainingwasusedtovisualizetheEV71/E59virusbulkseparatedonaSDS-PAGEgel(1A);twodifferentlotsofEV71VLPproteins(1B,lanes1and 2)andformalin-inactivatedEV71/E59(1B,lane3)wereanalyzedbySDS-PAGE;theviralproteinswereanalyzedbyWesternblot(1C)usingthecommercialEV71-speciﬁc monoclonalantibodyMAB979.

assay)perdose.Rabbitserawerecollectedtwoweeksaftereach

170

immunizationandusedforimmunologicalanalysis.Rabbitswere

171

bledoutatfourweeksaftertheﬁnalimmunization,andarabbit

172

polyclonalIgG(RPAb)fractionwaspuriﬁedbyProteinAafﬁnity

173

chromatographyofpooledhyper-immuneantiseraandstoredat

174

−20◦_C. _The_{concentration}_of _RPAb_was_determined_by _the_BCA

175

proteinassay.Thespeciﬁcityandanti-EV71neutralizationtiterof

176

RPAbweretestedusingWesternblotandvirusneutralizationassay,

177

respectively.

178

2.10. Virusneutralizationassay

179

Virusneutralizationtiterofeachsamplewasdeterminedusing

180

TCID50assayaccordingtotheprotocolreportedpreviouslybyLiu

181

etal.(2011).

182

2.11. VP2epitope-speciﬁcquantitativeenzyme-linked

183

immunosorbentassay(Q-ELISA)

184

The Q-ELISA was done by coating the wells of a 96-well

185

MaxisorbELISAplate(Nunc,Denmark)with1␮gofeitherrabbit

186

anti-EV71polyclonalIgGantibodies(RPAb)orPY10267in100␮L

187

of coating buffer (pH 9.6, Sigma) overnight at 4◦C. The wells

188

wereincubatedwith250␮Lof5%skimmilktoblocknon-speciﬁc

189

bindingoftheconjugatedsecondaryantibodies.Theplatewasleft

190

for2hatroomtemperaturebeforeitwaswashedthreetimes,each

191

timewith250␮Lofassaybuffer;then,100␮Loftwofolddilutions

192

ofeitherformalin-inactivated,chromatographically-puriﬁedEV71

193

bulk,ordifferentviruspreparationsinPBS,pH7.4,containing1%

194

BSA(Biorad), wasaddedtothewells. Afterincubationatroom

195

temperature for 2h, the wells were washed four times with

196

250␮Lofthewashbufferbefore100␮Lofassaybuffercontaining

197

0.1␮LoftheMAB979monoclonalantibodystock wasaddedto

198

eachtestwelltodetectthebindingofnativeordenaturedviral

199

proteinstoimmobilizedpolyclonal captureantibodies.Afterthe

200

1-h incubation at room temperature, wells were washed four

201

timeswith250␮Lofwashbuffer;100␮LofanHRP-conjugated

202

anti-mouseIgGantiserum(JacksonImmunoResearch,WestGrove, 203

PA,USA)dilutedat1:30,000intheassaybufferwerethenadded 204

to each test well. Binding wasallowed during 30min at room 205

temperaturebeforethe platewaswashed sixtimes withassay 206

bufferand blotteddry withapaper towel.Then,50␮LofTMB 207

peroxidase substrate (SureBlueTM_,_KPL, _{Gaithersburg,} _MD, _USA) ₂₀₈

wasadded,andthereactionoccurredinthedarkbecausetheplate 209

wascoveringwithaluminumfoilandleftinadrawerfor30min 210

atroomtemperature.Thereactionwasstoppedbyadding50␮L 211

of2NH2SO4toeachwell,andabsorbanceat450nmwasrecorded 212

withanELISAreader(SpectraMaxM2model,Sunnydale,CA,USA). 213

3. Results 214

3.1. Reagents 215

ReagentsandmaterialsusedintheQ-ELISAwereﬁrstlycharac- 216

terized.AnEV71/E59virusbulkwasproducedandpuriﬁedusing 217

serum-freebioreactorasdescribedpreviously(Changetal.,2011) 218

anditsproteinconcentrationwasfoundtobe21.5␮g/mLbythe 219

BCAassay.TheinfectivitywasmeasuredbytheTCID50assayand 220

foundtobe7.2×106_TCID

50permL.SDS-PAGEanalysisofthepuri- 221

ﬁedEV71/E59 antigenrevealedthree majorviralproteinbands 222

withestimatedmolecularweightsof28,34–36and38kDa(Fig.1A). 223

Asshown in Table1, the34–36kDa proteinbandmatched the 224

estimatedmolecularweightsoftheVP1subunitofEV71(Chung 225

etal.,2008).The28kDaproteinbandmight comprisetwo pro- 226

teinswithveryclosemolecularweights(MW),suchasVP2and 227

VP3,whichwereestimatedtohaveMWsof28and27kDa,respec- 228

tively(Table1).AsindicatedinTable1,the38kDaproteinband 229

correspondedtotheincompletelyprocessedprocapsidproteinVP0 230

(VP4–VP2).Thisis basedontryptic digestionandpeptidemap- 231

pingusingmassspectrometry,bothVP2andVP4proteinsequences 232

wereidentiﬁedinthe38kDaproteinband(Liuetal.,2011). 233

To characterize thespeciﬁcity of theantibodies used in the 234

Q-ELISA,immuno-dotblottingshowedthatboththechickenpoly- 235

(6)

ARTICLE IN PRESS

Table1

Summaryofpredictedandobservedmolecularweights(MW)ofEV71/E59viralproteinsandincompletelyprocessedviralpolypeptides.

EV71antigens Aminoacidsequences Residues PredictedMWa_(kDa) _Observed_MWb_(kDa) Viralantigeninvirion

VP4 1–69 69 7.49 8

VP2 70–323 254 27.78 28

VP3 324–565 242 26.52 27

VP1 566–862 297 32.73 34–36

Incompletelyprocessedviralpolypeptides(Fig.1B,lanes1&2)

VP0(VP4–VP2) 1–323 323 35.27 38

VP3-VP1 324–862 519 59.25 60

Viralantigenscross-linkedbyformalininactivationbasedonVP2recognizedbyMAB979(Fig.1C,lane3)

VP4–VP2 1–323 323 35.27 38

VP4–VP2–VP4 42.76 45–49

VP2–VP2 55.56 56–58

VP2–VP4–VP2 63.05 64–66

VP4–VP2–VP3–VP1 1–862 862 94.52 96–98

a_The_predicted_MW_is_based_on_the_{corresponding}_amino_acid_sequence_derived_from_the_EV71/E59_strain_genome. b _The_observed_MW_is_based_on_the_SDS-PAGE_and_Western_blot_analyses_of_the_EV71/E59_virus_bulk.

reactedwith puriﬁed EV71/E59 bulk. Both antisera had

ELISA-237

reactivetiters>50,000(datanotshown).Acommerciallyavailable

238

monoclonalantibodyMAB979wasfoundtohavea>1/64

neutral-239

izingtiternotonlyforEV71/E59(genotypeB4)butalsoforother

240

EV71isolatescurrentlycirculatinginTaiwan,suchasgenotypes

241

C4andB5(Liuetal.,2011).Inaddition,MAB979recognizedan

242

epitopewiththeAGGTGTEDSHPPYKQsequencethatcorresponded

243

toresidues136–150ofVP2(Liuetal.,2011).Anothercommercial

244

monoclonalantibodyM7064(Dako,Denmark)thathadno

reactiv-245

ityagainstEV71/E59proteins,wasservedasanegativecontrolinall

246

immunologicalassays.TheseresultssuggestedthataQ-ELISAcould

247

beestablishedusingapairofselectedantibodies,namelya

poly-248

clonalantibodyasacapturereagentandtheMAB979monoclonal

249

antibodyasareportingmarker.

250

3.2. PreparationofpuriﬁedEV71/E59bulkandVLPasantigen

251

referencestandards

252

Thequalityofonepotentialreferencestandard(EV71/E59bulk)

253

derived from the 5L bioreactor run has been described above

254

andshowninFig.1A.EV71virus-likeparticles(VLPs)asanother

255

potentialreferencestandardwaspreparedfromthebaculovirus

256

expressionsystemaccordingtoChungetal.(2008).Asshownby 257

SDS-PAGE(Fig.1B,lanes1and2),twomajorproteinbands(60 258

and38kDa)wereobservedinVLPpreparations.The60kDapro- 259

teinbandmaycorrespondtotheincompletelyprocessedVP1–VP3 260

polypeptidesthatcontainedVP1(34–36kDa)andVP3(27kDa)(see 261

Table1).This60kDaproteinbandisbeingcharacterizedcurrently. 262

The38kDaproteinwastheVP0procapsidproteindescribedpre- 263

viously.The38kDaband(VP0)wasrecognizedbyMAB979inthe 264

Westernblotanalyses(Fig.1C,lanes1and2).Itwasa surprise 265

toobservea28-kDaprotein(VP2)bandreactedwithMAB979in 266

Fig.1C,lane1sincenoproteinbandwasdetectedintheSDS-PAGE 267

analysis(Fig.1B,lane1). 268

TheCoomassieblue-stainedviralproteinbandsresolvedinthe 269

SDS-PAGEcouldbescanneddensitometricallytoobtainintegrated 270

opticaldensity (IOD)values thatallowedfor thedetermination 271

oftheirrespectiveconcentrationsfromaknownproteinstandard 272

curveasshowninFig.2A.TheviralantigensintwopotentialEV71 273

antigenreferencestandards(VLPsandEV71/E59bulk)separatedby 274

SDS-PAGEasshowninFigs.1Band2AwerescannedandtheirIOD 275

valueswererecorded.Thealbuminproteinstandardcurveobtained 276

fromthegelwasfoundtobelinear(R2₌_0.9935,_Fig.₂_A)._Based

277

onthealbumincalibrationcurve,theconcentrationofthe38kDa 278

Fig.2. DensitometricdeterminationofEV71viralantigenconcentrations40␮LofapuriﬁedEV71/E59virusbulk,andvariousknownconcentrationsofalbuminstandard (thenumbersabovetheindividualalbuminbands)wereseparatedona10.0%SDS-PAGEgel(A).Theindividualdye-boundalbuminand38-kDabandswerescannedwith theHewlettPackardScanJet4890toobtaintheirrespectiveIODs.TheIODvaluesobtainedfortheindividualalbuminbandswereplotted(B).

(7)

C.-C.Liuetal./JournalofVirologicalMethodsxxx (2011) xxx–xxx 5 1 6 1.8 1 0 1.2 1.4 1.6 50nm y = 0 526Ln(x) + 2 5839 0 4 0.6 0.8 1.0 OD 4 y =0.526Ln(x)+2.5839 R2 = 0.9988 0.0 0.2 0.4 0 01 0 1 1 0.01 0.1 1

EV71/E59 bulk concentration (µg/mL)

A

B

C

D

2.5 50nm 1.5 2.0 OD 4 y= 0.5677Ln(x)+ 2.6555 0 5 1.0 y 0.5677Ln(x) 2.6555 R2_{= 0.988}₃ 0.0 0.5 0 01 0 1 1 0.01 0.1 1

EV71/E59 bulk concentration (µg/mL)

0 8 0.9 50nm _{0 5}0.6 0.7 0.8 OD 4 y = 0 5364x + 0 1561 0 2 0.3 0.4 0.5 y=0.5364x +0.1561 R2 = 0.9984 0.0 0.1 0.2 0 0 5 1 0 0.5 1

Heat-inactivated EV71/E59 bulk concentration (µg/mL) 3.5 50nm 2 0 2.5 3.0 OD 4 y = 0 7389Ln(x) + 1 3906 1.0 1.5 2.0 y =0.7389Ln(x)+1.3906 R2_{= 0.99}₂₂ -0.5 0 1 1 0 10 0 0.1 1.0 10.0

EV71 VLP concentration (µg/mL)

Fig.3.SensitivityandspecificityofthePY-10267/MAB979sandwichELISAwerecalibratedwithpurifiedEV71/E59bulk.TheQ-ELISAwasconductedbycoatingthewells with0.1␮gofcapturePY-10267polyclonalantibodies.ThesecondaryantibodyMAB979in(A)wasaddedat0.1␮Lfromthesuppliedstockperwell.(B)and(C)showthe sensitivityandspecificityoftheRPAb/MAB979sandwichELISAformatcalibratedwithpurifiedEV71/E59virusbulkandtheVLPLot#2,respectively.(D)showsthesensitivity andspecificityofthePY-10267/MAB979sandwichELISAformatcalibratedwithpurified,heat-inactivatedEV71/E59bulk.

(VP2–VP4),36kDa(VP1)and27k-Da(VP3)proteinsinEV71/E59

279

virus bulk (Fig. 2B) were calculated tobe 5, 5.5 and 5␮g/mL,

280

respectively. Therefore, theapproximate concentration ofEV71

281

viralantigensinthepuriﬁedvirusbulk(10mL)was15.5␮g/mL

282

(thesumofVP0,VP1 andVP3),closetothevalue(21.5_␮g/mL)

283

obtainedbytheBCAproteinassay.Usingthesamemethod,the

284

arbitraryconcentrationsofthe60kDaand38kDaproteinbandsin

285

VLPLot#2weredeterminedtobe50and75␮g/mL,respectively.

286

Thesumofthesetwoconcentrations(125␮g/mL)wasconsistent

287

withthevalueofthetotalproteinconcentrationdeterminedbythe

288

BCAassay(135␮g/mL).

289

3.3. Q-ELISAdevelopment

290

TwoassayformatswereevaluatedtoestablishtheQ-ELISA.

PY-291

10267polyclonalantibodieswereimmobilizedtocapturevarying

292

amountsofviralantigensfromthepuriﬁed EV71/E59bulk.The

293

MAB979monoclonalantibodywasusedtodetectVP2inpuriﬁed

294

virusbulkcapturedbyPY-10267antibodies.Resultsshowedthat

295

thisassayformatcoulddeterminetheVP2epitopeinEV71/E59bulk

296

inadose-dependentmanner(Fig.3A).Similarresults(Fig.3B)were

297

obtainedfromthesecondQ-ELISAformatcombiningrabbit

anti-298

EV71polyclonalIgGantibody(RPAb,preparedin-houseandfound

299

tohavea>1000EV71/E59virusneutralizationtiter)andMAB979.

300

Excellentlinearﬁts(R2 _values_0.98_–0.99)_were_obtained_for_each

301

Q-ELISAformat(Fig.3AandB).Basedonthecurrentresults,the

302

bestrangeofdetectionwasfoundtobe0.05–0.16␮g/mLofthe

303

EV71/E59bulk.

304

SimilarlyVLPLot#2(thesecondantigenreferencestandard)

305

wastestedbybothQ-ELISAformatsandtheR2_values_were_found

306

tobeconsistent(0.99).Atypicalplotofabsorbancevs.VLPantigen

307

generatedfromtheRPAb/MAB979formatisshowninFig.3C.Based

308

onthecurrentresults,theQ-ELISAusingVLPLot#2asanantigen 309

workingstandardwasfoundtobe10-foldlesssensitiveandhad 310

adetectionrangeof0.5–1.6␮g/mLoftheVP0antigen.Sincethe 311

RPAb/MAB979Q-ELISAformatwasestablishedandcharacterized, 312

itwasselectedforallfuturestudies. 313

3.4. Stabilityofthepotentialantigenreferencestandards 314

Because the sensitivity of the assay obtained with VLP Lot 315

#2was10-foldless than thatobtainedwithpuriﬁed EV71/E59 316

bulk,itwasofinteresttodeterminethestabilityofbothVLPLot 317

#2and puriﬁed EV71/E59 aslong-termantigen referencestan- 318

dardsforEV71vaccinedevelopment.Thestabilityproﬁlesofthese 319

antigen referencestandards storedat −80◦_C _for _one_and _three ₃₂₀

months were evaluated by Q-ELISA. The VLP Lot #2 stored at 321

−80◦_C_for_one_and_three_months_was_found_to_be_relatively_sta- ₃₂₂

bleandyieldedstandardcurvessimilartothoseobtainedbefore 323

storage. In contrast, puriﬁed EV71/E59 required two to three 324

times more frozenmaterialsto generate standard curves simi- 325

lartothoseobtainedbeforestorage.Otherstorageconditionsfor 326

puriﬁed EV71/E59bulk werestudied.Unfortunately, ﬁvetoten 327

timesmorematerialoftheheat-treatedEV71-E59(56◦Cfor3h)or 328

formalin-inactivatedEV71/E59bulk(3daysat37◦C)wasneeded 329

togenerateastandardcurvesimilartothatobtainedwithfreshly 330

prepared EV71/E59 bulk (Fig. 3D). The current resultsindicate 331

that both theEV71/E59 bulk and VLP Lot #2couldbe usedas 332

theantigenreferencestandard. Butthestability proﬁleand the 333

concentrationofantigendeterminedbySDS-PAGEandBCAassay 334

showedVLP Lot #2tobe betterand correlated (125␮g/mL vs. 335

135_␮g/mL).Forthesereasons,VLPLot#2andtheRPAb/MAB979 336

format were selected as reagents for subsequent Q-ELISA 337

(8)

ARTICLE IN PRESS

Table2

Q-ELISAdeterminationofVP2-speciﬁcepitopecontentintheserum-freeculturesupernatants,virusconcentrates,chromatographically-puriﬁedvirusbulks,and formalin-inactivatedvaccinebulksproducedatvariouskeystepsoftheEV71vaccinemanufacturingprocess.

Lot# Process step Total volume(L) TCID50a (×106_/mL) Totalproteinb (␮g/mL) VP2epitopec (Unit/mL) Recoveryd_(%) _VP2_{epitope/total} protein(Unit/␮g) Lot#1 Harvest 39.0 5.4 1342±46 7.1±0.4 100 ND Dif/Cone _1.65 _77.5 ₅₂₀_±₁₀ ₁₆₆_±_4.1 _99.3 _ND LCf _1.40 _ND _28.6_±_2.9 ₁₁₀_±_14.9 _55.8 _3.8 Lot #2 Harvest 39.1 1.2 1534±27 6.8±0.5 100 ND Dif/Con 1.75 40 358±9.7 228±7.8 151 ND LC 1.50 ND 21.5±2.5 113±4.2 64.2 5.3 Lot#3 Harvest 41.0 2.3 1460±5.5 11.9±0.8 100 ND Dif/Con 1.10 57 430±6.6 475±21 107 ND LC 0.75 ND 66.7±2.2 225±16.7 34.6 3.4 Formalin inactivatedg 0.75 53.0±1.5 198±21.3 30.5 3.7 ND:notdetermined. a_TCID

50isthemediantissuecultureinfectivedoseofEV71thatproducespathologicalcytopathiceffects(CPE)in50%ofinoculatedcellcultures. b _Total_protein_{concentration}_was_determined_by_the_BCA_protein_assay.

c _{VP2-speciﬁc}_epitope_content_was_determined_by_Q-ELISA_using_VLP_Lot_#2_as_the_antigen_working_standard_and_the_RPAb/MAB979_ELISA_format. d _Recovery_was_calculated_using_the_total_VP2_epitope_content_at_each_step_divided_by_the_total_VP2_epitope_at_the_harvest_stage.

e_Dif/Con_is_the_{diaﬁltration/concentration}_step.

f _LC_is_the_{gel-ﬁltration}_{chromatography}_{puriﬁcation.}

g_Vaccine_bulk_was_obtained_by_formalin_inactivation_of_virus_bulk.

3.5. TheVP2epitopecontentinaformalin-inactivatedEV71-E59

338

vaccinebulkmeasuredbytheQ-ELISA

339

TodeterminewhethertheVP2epitopecontentina

formalin-340

inactivated EV71-E59 vaccine bulk could be measured by the

341

Q-ELISA,VLPLot#2wasusedastheantigenreferencestandard.

342

TheVP2epitopecontentintheformalin-inactivatedEV71/E59

vac-343

cinebulkwasfoundtobeequalto45.7␮gofVP0permL,which

344

wasgreater than the valueobtained by theBCA protein assay

345

(18␮g/mL).Itwasinterestingtoknowthattheratiobetweenthe

346

VP2epitopecontentdeterminedbyQ-ELISAandthetotalprotein

347

measuredbytheBCAassaywas2.53(45.7/18).Thisdiscrepancy

348

maybeduetothefactthattheVP2epitopecouldexistin

differ-349

entformsandconformations(VP2subunit,virion-associatedVP2,

350

incompletely processedVP0 and otherVP2 associated antigens

351

cross-linkedbyformalin).Therefore, arbitraryQ-ELISAunitsare

352

preferabletodescribethecontentofVP2epitopeinagivensample.

353

Tothisend,oneQ-ELISAVP2epitopeunit/mLwasdeﬁnedas1␮g

354

ofVP0(38-kDaproteinband)permLofVLPLot#2determinedby

355

theRPAb/MAB979Q-ELISAformat.Inthisexample,45unitsofVP2

356

epitopepermLwereintheformalin-inactivatedEV71/E59vaccine

357

bulk.

358

3.6. DeterminationofVP2epitopecontentinEV71/E59vaccine

359

producedfromtherollerbottlemanufacturingprocess

360

Culture supernatants collected from each 40-L roller bottle

361

manufacturing runs were assayed for their titers of infectious

362

virusparticlesusingtheTCID50assayandVP2epitope contents

363

by Q-ELISA. Table 2 shows that the titers of infectious virus

364

particlesinpilot-scalecrudesupernatantsharvested fromthree

365

individual runswere 5.4, 1.2 and 2.3×106 _TCID

50 units/mL in

366

runs #1, 2 and 3, respectively. VP2 epitope contents in

cul-367

turesupernatants of these batches were measuredby Q-ELISA

368

andfoundtobe7.1, 6.7and 11.9units/mLin runs#1,2and 3,

369

respectively.ItwasofinteresttoknowwhethertheVP2epitope

370

contentandthetiterofinfectiousviralparticleswerecorrelated.

371

Q-ELISAunitsoftheVP2 epitopecontentper106 _TCID 50

infec-372

tiveunits were found tobe 1.3 (7.1/5.4), 5.6 (6.8/1.2) and 5.2

373

(11.9/2.3) for run #1, 2 and 3, respectively. Based on the

cur-374

rentlimited number of experiments, nodirect correlationwas

375

established.

376

To monitor the consistency of the EV71 vaccine manufac- 377

turingprocess, theVP2 epitope contentin in-processmaterials 378

obtainedfromboththediaﬁltration/concentrationandliquidchro- 379

matographypuriﬁcationstepswereanalyzedbyQ-ELISA.Afterthe 380

diaﬁltration/concentrationstep,asshowninTable2,Q-ELISAunits 381

ofVP2epitopeper106 _TCID

50infectiousunitswerefoundtobe 382

2.2(166.5/77.5),5.7(228/40)and8.3(475/57)forruns#1,2and3, 383

respectively.Again,theVP2epitopecontentmeasuredbyQ-ELISA 384

didnotcorrelatewellwithTCID50values.VP2epitope/totalprotein 385

ratioswerecalculatedtobe3.9(110/28.6),5.2(113/21.5)and3.4 386

(225/66.7)forruns#1,2and3,respectively(Table2).Theseresults 387

suggestthatthereisroomforimprovementinthechromatogra- 388

phypuriﬁcationprocess, particularlyin establishingcriteria for 389

poolingelutedfractions.Infact,basedontheVP2epitopedeter- 390

minationbyQ-ELISAtheoverallrecoveryyieldsofvirusfromthe 391

threerunswerefoundtobe55.8%,64.2%and34.6%forruns#1,2 392

and3,respectively(Table2). 393

Thechromatographically-puriﬁed EV71/E59 bulk wasinacti- 394

vatedwithformalin at37◦C for three days,and after 0.22-␮m 395

sterileﬁltration,theVP2epitopecontentdeterminedbyQ-ELISA 396

andthetotal proteinconcentrationmeasuredbytheBCAassay 397

were found to be 198.4units/mL and 53.0␮g/mL, respectively 398

(Table2).Therewereprotein(from66.7to53.0␮g/mL)andVP2 399

epitopecontent(from224.7to198.4units/mL)lossesduringthe 400

inactivationandsterileﬁltrationsteps.Nevertheless,theQ-ELISA 401

unitwasfoundtobe3.7VP2epitopesper␮goftotalprotein,a 402

valueclosetothatinvirusbulk(3.4).Aclinicallot(#23-04-0001) 403

wasrecentlymanufacturedusingthesameproductionprocess,and 404

thenumberofVP2Q-ELISAunitsper␮goftotalproteinwasfound 405

tobe3.9(datanotshown).Theseresultssuggestedthatformalin 406

inactivationdidnotmodifytheVP2epitopecontentasdetermined 407

byQ-ELISA. 408

3.7. MouseimmunogenicitystudieswithdifferentEV71/E59 409

vaccinecandidates 410

The amount of VP2 epitopes in three different formalin- 411

inactivatedEV71wholevirionsvaccinesdeterminedbytheQ-ELISA 412

wereusedforthedosageformulationandmouseimmunogenicity 413

studies.ThequalityofEV71/E59vaccinebulkproducedfromVero 414

cellgrownintheBioreactorwithserum-freemediumhadbeen 415

(9)

Fig.4. TheEV71viralantigensinthevaccinebulkproducedfromserum-freeculture medium(A,lane1);andfromserum-containingmedium(lane3)wereseparated bySDS-PAGEandsliverstained.Theviralproteinsrecognizedbythecommercially availableEV71-speciﬁcmonoclonalantibodyMAB979intheWesternblot(B).

cellgrowninthepresenceofserumorserum-freemediumwas

417

analyzedbySDS-PAGEandWesternblots,andshowninFig.4.The

418

vaccinebulkproducedinthepresenceofserumwasfoundto

con-419

tainfewerhighMWproteinsthatwerenotrecognizedbyMAB979

420

(Fig.4B,lane3).Thismayexplainthevaccinebulkproducedin

421

thepresenceofserum(Lot#07EV03inTable3)haslowerQ-ELISA

422

unitofVP2epitopeper␮goftotalproteinascomparedtothose

423

obtainedfromtheserum-free.

424

Table3

MouseimmunogenicitystudieswiththreedifferentEV71/E59vaccinecandidates. Vaccinecandidatesa _Protein

(␮g)/doseb

Q-ELISAunitofVP2 epitope/dosec

GMTNTtitersd (std.dev.) 1.4-Lbioreactor(2.5Q-ELISAunitsofVP2epitopeper␮goftotalprotein)

D16 0.5 0.75 1600(499)

D16 2.5 6.25 2099(1211)

Roller-bottle/serumcontainingmedium(0.41Q-ELISAunitofVP2epitope per␮goftotalprotein)

07EV03 0.15 0.06 <20

07EV03 0.71 0.29 199(70)

07EV03 3.5 1.45 447(220)

07EV03 7.1 2.91 707(511)

Rollerbottle/serum-freemedium(3.7Q-ELISAunitofVP2epitopeper␮g oftotalprotein)

Lot#3 0.01 0.05 28(40)

Lot#3 0.07 0.26 100(72)

Lot#3 0.34 1.29 854(476)

Lot#3 0.68 2.58 911(532)

a_EV71/E59_vaccine_candidates_were_prepared_under_various_culture_conditions usingthreedifferentdownstreampuriﬁcationmethods.Theimmunizationprotocol wasdescribedinSection 2.

b_The_protein_{concentration}_of_each_vaccine_candidate_was_measured_by_the_BCA proteinassay.Vaccineswereformulatedwithalumasanadjuvant.

c _{VP2-speciﬁc}_epitope_content_was_determined_by_the_Q-ELISA_described_in

Table2.

d _Virus_{neutralization}_titers_were_determined_using_the_TCID

50inhibitionassayas describedinSection2andarereportedasgeometricmeantiters(GMT).

MouseimmunogenicitystudiessummarizedinTable3revealed 425

thatEV71vaccineswerehighlypotentinelicitingvirusneutral- 426

izationtiter.InadditionEV71vaccineproducedintheserum-free 427

medium (one VP2 epitope unit formulated withalum as adju- 428

vant)elicitedslightlybetterEV71neutralizingantibodyresponses 429

(GMT=854)thanthoseobtainedwiththeEV71vaccineproduced 430

inthepresenceofserum(GMT=447).However,thedifferencewas 431

foundtobeinsigniﬁcantbytheStudentt-test(datanotshown). 432

ThenumberofVP2epitopeunitsandthemagnitudeofneutralizing 433

titerswerefoundtobedose-dependentinmiceimmunizedwith 434

vaccinecandidatesproducedinrollerbottleseitherinthepresence 435

orabsenceofseruminculturemedia.Theseresultsindicatedthat 436

theamountofVP2Q-ELISAunitspresentinthevaccinebulkscould 437

beusedtoreﬂectthepotencyofthevaccine. 438

4. Discussion 439

Thisstudyaimedtodevelopanassaythatmeasuresthecon- 440

centrationofEV71virusand/orviralantigenscriticaltovaccine 441

potencyateachstepofthevaccineproductionprocess.Therefore, 442

theassayshouldhaveexcellentsensitivityandspeciﬁcity,andbe 443

suitableforapplicationthroughoutthewholevaccinedevelopment 444

cycleduringwhichvirusesareculturedinlivecells,harvested,puri- 445

ﬁed,inactivatedand sterileﬁlteredbeforeproductformulation. 446

Theplaqueassayisusednormallytoestimatelivevirustiters(i.e., 447

theconcentrationofinfectiousvirusparticles)duringtheculture 448

processwasnotapplicableorsuitableformeasuringandcharacter- 449

izingtheﬁnalinactivatedproduct.Previousstudies(Chungetal., 450

2008)hadshownthattheconcentrationofEV71antigenscouldbe 451

obtainedbyeithertheBCAproteinassayorSDS-PAGEanalysisifthe 452

EV71vaccinepreparationwasrelativelypure.However,theseanal- 453

ysesarenotsuitablefordeterminingtheantigencontentsinculture 454

supernatantsandvirusconcentratesduringtheupstreamvaccine 455

manufacturingprocessstepsbecauseoflowantigenconcentrations 456

insupernatantsandhostcelland/orserumproteincontamination. 457

Consideringallthesefactors,aQ-ELISAthatcoulddeterminethe 458

contentofEV71viralantigenswasinvestigatedanddeveloped.The 459

ﬁnalobjectivewastodevelopanassaymuchliketheSRIDassay 460

usedformeasuringthehemagglutinin(HA)contentintheseasonal 461

ﬂuvaccineQCpotencytest.Tovalidatetheusefulnessandaccuracy 462

ofaQ-ELISAinEV71vaccinedevelopment,weselectedreagentsfor 463

theirspeciﬁcity,stability,efﬁciencyandsuitability.Inadditionto 464

thein-house-producedEV71-speciﬁcrabbitantisera,a commer- 465

cialchicken polyclonal EV71-speciﬁcantiserumPY-102676 was 466

evaluatedand foundtobeusefulasa captureantibodyfordif- 467

ferenttypesofsamplesrangingfromculturesupernatants,virus 468

concentrates,chromatographically-puriﬁedvirusandformalin-or 469

heat-inactivatedvirus.UsingSDS-PAGEandgel-scanninganalyses, 470

we estimatedtheproteinconcentrations oftwo potentialEV71 471

viralantigen reference standards,namely puriﬁed VLPsand an 472

EV71/E59 bulk produced from Verocells grownon microcarri- 473

ersinabioreactor.TheQ-ELISAwas10-foldmoresensitivewhen 474

theEV71/E59virusbulk(0.05–0.16␮g/mL)wasusedasarefer- 475

encestandard,whilethebiosafetyissuesandtheinstabilityofthe 476

EV71/E59virusbulkduringstorageat−80◦_C_made_it_unsuitable_for ₄₇₇

Q-ELISAdevelopment.Thecurrentresultsshowedthatthefrozen 478

puriﬁedEV71/E59requiredtwotothreetimesmorematerialsto 479

generatestandard curves similartothoseobtainedbeforestor- 480

age.TooptimizethestorageconditionforpuriﬁedEV71/E59virus 481

bulk,wefoundthattentimesmorematerialoftheheat-treated 482

EV71-E59orformalin-inactivatedEV71/E59virionwasneededto 483

generatea standard curvesimilartothatobtainedwithfreshly 484

preparedEV71/E59virusbulk.Therearetwopossibilities:(1)the 485

heat-treatmentand/orformalin-inactivationhavealteredtheviral 486

(10)

ARTICLE IN PRESS

antibodies;(2)theepitopeofVP2ismodiﬁedbytheheat-treatment

488

orformalin,andhaslessbindingafﬁnitytoMAB979.Thedirect

489

ELISAexperimentswereperformedtoclarifythisissue.Rabbit

anti-490

EV71antibodieshadlessbindingtitertochemical-inactivatedEV71

491

bulkascomparedtothoseobtainedfromfreshlypreparedEV71

492

virus(datanotshown).Inaddition,thechemical-inactivatedEV71

493

vaccinebulkhad thesimilarbindingafﬁnity toMAB979asthe

494

freshlypreparedEV71bulk.Furtherstudyhasshownthe

formalin-495

inactivatedEV71bulktobestableat4◦C,soitcanbeconsidered

496

asareferencestandardforQ-ELISAinfuture.

497

Basedonthecurrentstabilityandimmunologicalstudies,EV71

498

VLPsprovidedlesssensitiverange(0.2–1.6␮g/mL),butbetterand

499

moreconsistentresultsthanthoseobtainedwiththeEV71/E59

500

virusbulk.VLPsconsistofassembledviralproteinsthattogether

501

createdamoreauthenticstructureandconformationofviral

anti-502

genstomimicanativevirus.Therefore,EV71VLPswereusedasan

503

antigenreferencestandardfortheQ-ELISAdevelopment.In

addi-504

tion,thecurrentVLPstandardonlycontainsVP0(VP4–VP2)asthe

505

viralantigenrecognizedbyMAB979(Fig.1C,lane2).Thus,the

bind-506

ingactivity(VP0/MAB979)orantigenicityvalueoftheVP2antigen

507

couldbeuseddirectlyasanindexorarbitraryunitofVP2epitope

508

concentration.Infact,oneQ-ELISAVP2epitopeunit/mLcannow

509

bedeﬁnedas1␮gofVP0(38kDaproteinband)permLofVLPLot

510

#2determinedbytheRPAb/MAB979Q-ELISA.

511

Fromabioprocessdevelopmentpointofview,estimatingthe

512

concentrationofvirusespresentinsequentialsamplesisessential

513

tomonitorthelossofproductateachstepofthemanufacturingand

514

puriﬁcationprocessandtooptimizethebioprocessconditionsto

515

increaseproductyields.TheVP2Q-ELISAwasusedtoevaluatethree

516

40-LbatchesofEV71/E59producedinserum-freerollerbottles

cul-517

turesystem.Theupstreambioprocess(culturesupernatantharvest,

518

diaﬁltrationandconcentration)stepswerefoundtobevery

efﬁ-519

cientbecausetheVP2epitoperecoverythroughoutthesestepswas

520

100%(Table2).Becauseofvariousamountsofnon-infectious

par-521

ticlespresentintheculturesupernatantharvest,theconcentration

522

ofinfectiousparticlesdeterminedbyTCID50assaywasnot

corre-523

lateddirectlywiththeVP2epitopecontentmeasuredbyQ-ELISA.

524

Thedownstreamgel-ﬁltrationchromatographypuriﬁcationstep,

525

theyieldsfortheﬁrsttwobatcheswerefoundtobebetween56

526

and64%.Thethirdbatchwasfoundtohave2–3-foldmoreVP2

epi-527

topecontentinthevirusconcentratesthanthoseobservedinthe

528

ﬁrst2batches,butthechromatographyyielddecreasedto35%.The

529

currentresultssuggestedthattheconditionsusedingel-ﬁltration

530

chromatographyshouldbeinvestigatedand optimized,

particu-531

larlyinestablishingcriteriaforpoolingelutedfractions.Basedon

532

theVP2Q-ELISAresults,formalin-inactivationandsterileﬁltration

533

stepsdidreduceproductyields,lessthan5%showninTable2which

534

isacceptable.

535

ItwasinterestingtoknowthattheratiobetweentheVP2

epi-536

topecontent in the EV71vaccine bulk determined by Q-ELISA

537

and the total protein measured by the BCA assay was ranged

538

2.5–5.2.Thisdiscrepancymaybedueto(i)thepresentofvarious

539

amountsofhostproteincontaminantsand(ii)thefactthatVP2

540

epitopemayexistindifferentformsandconformations(VP2

sub-541

unit,virion-associatedVP2,incompletelyprocessedVP0andother

542

VP2associatedantigenscross-linkedbyformalin).Thesevarious

543

formsofVP2wereshownandidentiﬁedintheWesternblot

anal-544

ysisofformalin-inactivatedEV71/E59vaccinebulk(Fig.1C,lane

545

3)butwerenotobservedontheSDS-PAGEgel(Fig.1B,lane3).

546

Inaddition, thebindingafﬁnityof MAB979for itscognateVP2

547

epitopeinVLPLot#2maybeweakerthanthatintheEV71/B59

548

virion(nativeorformalin-inactivated).Thishypothesiswas

sup-549

portedbythefactthatVLPLot#2required10-foldmoreantigens

550

(1.3_␮ginFig.3C)toobtainanabsorbancevalue(1.4ELISAOD)

551

similartothatobservedwith0.13␮gofpuriﬁedEV71/E59 bulk

552

(Fig.3A).ThebindingafﬁnityofMAB979toVP2epitopemayvary

553

becausetheVP2antigencouldexistas(i)asubunit(28kDa)inlive 554

EV71particles,whereas thecurrent VLPantigenstandard could 555

bemostlyformedbyincompletelyprocessedVP1–VP3and VP0 556

(Fig.1B,lanes1and2);(ii)avirion-associatedcomponent(98kDa); 557

(iii)acomponentofanincompletelyprocessedVP0(38kDa);(iv)in 558

otherpossibleforms(45and64kDainTable1);and(v)invarious 559

conformations. Therefore, this Q-ELISAcould beused toevalu- 560

atecriticalfactorsinﬂuencingviruscultureconditions,upstream 561

harvest,down-streampuriﬁcationprocesses andthestabilityof 562

vaccinebulk. 563

WhentheQ-ELISAwasusedtoanalyzethequalityofvaccine 564

bulksproducedinrollerbottlesinthepresenceofserum,theVP2 565

epitopeunitsper ␮goftotal proteinratiowasfoundtobe10- 566

foldlowerthanthoseobtainedforthevaccinebulkproducedin 567

theabsenceofserum(Tables2and3).Thequalityofthevaccine 568

bulksproducedinthepresenceofserumandanalyzedbySDS-PAGE 569

andWesternblotusingMAB979wasfoundtocontainfewerhigh 570

MWviralantigens(Fig.4B,lane3).Mouseimmunogenicitystudies 571

summarizedinTable3revealedthatEV71vaccinesproducedin 572

serum-freemedium(oneVP2epitopeunitformulatedwithalum 573

as adjuvant) elicited slightly better EV71neutralizing antibody 574

responses(GMT=854)thanthoseobtainedwiththeEV71vaccine 575

producedinthepresenceofserum(GMT=447).Thedose–response 576

immunogenicitystudiesalsosuggestedthatthenumberofVP2epi- 577

topeunitsinvaccinebulksproducedineitherserum-freemediaor 578

serum-containingmediacouldreﬂectthepotencyofthevaccine. 579

5. Conclusion 580

To complywith process developmentand validation guide- 581

linesenforcedbyregulatoryagencies,vaccinemanufacturershave 582

an interest in developing assays that can monitorand charac- 583

terize the purity and yield of viral preparations at sequential 584

key stages ofthemanufacturing processes.In addition,vaccine 585

manufacturers require assays to determine whether the con- 586

centration of viral antigen(s) correlates with vaccine potency. 587

In this study, a Q-ELISA was successfully developed to pro- 588

videoff-line timelyquantitativemeasurementsofanimportant 589

epitope of the EV71 VP2 subunit throughout the vaccine pro- 590

ductioncycle(cellculturesupernatants,diaﬁltrates/concentrates, 591

chromatographically-puriﬁedorsucrose-gradientcentrifugation- 592

puriﬁed virus particles, and formalin-inactivated virions). This 593

Q-ELISAinfuturecouldserveasanassaytoevaluatecriticalfac- 594

tors inﬂuencing virusculture conditions,upstream harvestand 595

down-streampuriﬁcationprocesses.Themostimportantﬁndings 596

fromthecurrentanalyseswerethefollowing:(i)theconcentra- 597

tionofinfectiousparticles determinedbyTCID50assaywasnot 598

directlycorrelatedwiththecross-neutralizationVP2epitopecon- 599

tentmeasuredbyQ-ELISA;and(ii)theamountofVP2Q-ELISAunits 600

presentintheEV71vaccinebulksmayreﬂectthepotencyofthe 601

vaccinebecausethenumberofVP2epitope unitsandthemag- 602

nitudeofneutralizingtiterswerefoundtobedose-dependentin 603

mouseimmunogenicitystudies. 604

Uncitedreference Q1 605

Wuetal.(2004). 606

Acknowledgements 607

Thisworkwassupportedbyagrant(number–98A1-VCSP01- 608

014) from the National Science Council (NSC) of Taiwan. We 609

thankDrs.MichelKleinforreviewingthemanuscript;theTaiwan 610

CDCforprovidingtheEV71/E59virusstrainandperformingthe 611

(11)

Dr.Y.C.HuoftheNationalTsingHauUniversityforprovidingthe

613

baculovirusexpressionsystem.

614

References

615

AbuBakar,S.,Chee,H.Y.,Al-Kobaisi,M.F.,Xiaoshan,J.,Chua,K.B.,Lam,S.K.,1999. 616

Identiﬁcationofenterovirus71isolatesfromanoutbreakofhand,footand 617

mouthdisease(HFMD)withfatalcasesofencephalomyelitisinMalaysia.Virus 618

Res.61,1–9. 619

AlexanderJr.,J.P.,Baden,L.,Pallansch,M.A.,Anderson,L.J.,1994.Enterovirus71 620

infectionsandneurologicdisease(UnitedStates),1977–1991.J.Infect.Dis.169, 621

905–908. 622

Chang,L.Y.,Huang,Y.C.,Lin,T.Y.,1988.Fulminantneurogenicpulmonaryoedema 623

withhand,foot,andmouthdisease.Lancet352,367–368. 624

Chang,H.W.,Liu,C.C.,Lin,M.H.,etal.,2011.Isolationandcharacterizationofmurine 625

monoclonalantibodiesgeneratedagainstaclinicalisolateofenterovirus71.J.

626

Virol.Methods173,189–195.

627

Chen,H.L.,Huang,J.Y.,Chu,T.W.,Tsai,T.C.,Hung,C.M.,Lin,C.C.,etal.,2008. Expres-628

sionofVP1proteininthemilkoftransgenicmice:apotentialoralvaccine 629

protectsagainstenterovirus71infection.Vaccine26,2882–2889. 630

Chumakov,M.,Voroshilova,M.,Shindarov,L.,Lavrova,I.,Gracheva,L.,Koroleva,G., 631

etal.,1979.Enterovirus71isolatedfromcasesofepidemicpoliomyelitis-like 632

diseaseinBulgaria.Arch.Virol.60,329–340. 633

Chung,Y.C.,Ho, M.S.,Wu,J.C.,Chen,W.J.,Huang, J.H.,Chou,S.T.,etal.,2008. 634

Immunizationwithvirus-likeparticlesofenterovirus71elicitspotentimmune 635

responses and protects mice against lethal challenge. Vaccine 26,1855– 636

1862. 637

Foo,D.G.,Alonso,S.,Phoon,M.C.,Ramachandran,N.P.,Chow,V.T.,Poh,C.L.,2007. 638

IdentiﬁcationofneutralizinglinearepitopesfromtheVP1capsidproteinof 639

Enterovirus71usingsyntheticpeptides.VirusRes.125,61–68.

Gilbert,G.L.,Dickson,K.E.,Waters,M.J.,Kennett,M.L.,Land,S.A.,Sneddon,M.,1988. 640 Outbreakofenterovirus71infectioninVictoria,Australia,withahighincidence 641 ofneurologicinvolvement.Pediatr.Infect.Dis.J.7,484–488. 642 Hames,B.D.,1988.GelElectrophoresisofProteins:APracticalApproach,thirded. 643

OxfordUniversityPress,NewYork. 644

Ho,M.,Chen,E.R.,Hsu,K.H.,Twu,S.J.,Chen,K.T.,Tsai,S.F.,etal.,1999.Anepidemic 645 ofenterovirus71infectioninTaiwan.TaiwanEnterovirusEpidemicWorking 646

Group.N.Engl.J.Med.341,929–935. 647

Lee,M.S.,Chang,L.Y.,2010.Developmentofenterovirus71vaccines.ExpertRev. 648

Vaccines9,149–156. 649

Liu, C.C., Chou, A.H., Lien, S.P., Lin, H.Y., Liu, S.J., et al., 2011. Iden- 650 tiﬁcation of a cross-neutralization epitope of Enterovirus 71. Vaccine, 651

doi:10.1016/j.vaccine.2011.04.010. 652

McMinn,P.C.,2002.Anoverviewoftheevolutionofenterovirus71anditsclinical 653 andpublichealthsigniﬁcance.FEMSMicrobiol.Rev.26,91–107. 654 McMinn,P.C.,2003.Enterovirus71intheAsia-Paciﬁcregion:anemergingcauseof 655 acuteneurologicaldiseaseinyoungchildren.Neurol.J.SoutheastAsia8,57–63. 656 Qiu,J.,2008.Enterovirus71infection:anewthreattoglobalpublichealth?Lancet 657

Neurol.7,868–869. 658

Reed,L.J.,Muench,H.,1938.Asimplemethodofestimating50percentend-points. 659

Am.J.Hyg.27,493–497. 660

Schmidt,N.J.,Lennette,E.H.,Ho,H.H.,1974.Anapparentlynewenterovirusisolated 661 frompatientswithdiseaseofthecentralnervoussystem.J. Infect.Dis.129, 662

304–309. 663

Tung,W.S.,Bakar,S.A.,Sekawi,Z.,Rosli,R.,2007.DNAvaccineconstructsagainst 664 enterovirus71elicitimmuneresponseinmice.Genet.VaccinesTher.5,6. 665 Wu,S.C.,Liu,C.C.,Lian,W.C.,2004.Optimizationofmicrocarriercellculturepro- 666 cessfortheinactivatedenterovirustype71vaccinedevelopment.Vaccine22, 667

3858–3864. 668

Xu,J.,Qian,Y.,Wang,S.,Serrano,J.M.G.,Li,W.,Huang,Z.,Lu,S.,2010.EV71: 669

an emerging infectious disease vaccine target in the Far East? Vaccine, 670