VIROLOGY 184, 695-706 (1991)

Two Mechanisms of Antigenic Diversification of Foot-and-Mouth Disease Virus

M . A. MARTINEZ,* t J . HERNANDEZ,* M. E. PICCONE,t 1 E . L. PALMA,#E . DOMINGO * .2 N. KNOWLES,t AND M . G. MATEU*

'Centro de Biologia Molecular (CSIC-UAM), UniversidadAutdnoma de Madrid, Canto Blanco, 28049 Madrid, Spain, tAFRC Institute for AnimalHealth, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey GU24 ONF, United Kingdom; and tinstituto de Biologia Molecular,

Centro de lnvestigaci6n Veterinaria, INTA-cc77, 1708 Moron, Buenos Aires, Argentina

Received April23, 1991; accepted July 3. 1991

The amino acid replacements that underlay the diversification of the main antigenic site A (VP1 residues 138 to 150)of foot-and-mouth disease virus (FMDV) of serotype C have been identified . Sixteen new VP1 sequences of isolatesfrom 1926 until 1989 belonging to subtypes C,, C 2 , C„ C„ Cs , and unclassified are reported . The reactivities inenzyme-linked immunoelectrotransfer blot assays of capsid protein VPi with a panel of neutralizing monoclonal anti-bodies that recognize sites A or C (the VP1 carboxy-terminus) have been correlated with the amino acid sequence atthe relevant epitopes . The analyses involving the immunodominant site A reveal two mechanisms of antigenic change .One is a gradual increase in antigenic distance brought about by accumulation of amino acid replacements at twohypervariable segments within site A- A second mechanism consists of an abrupt antigenic change manifested by lossof many epitopes, caused by one replacement at a critical position (particularly Ale (145) -• Val or His (146)-. Gin) . Theidentification of the amino acid substitutions responsible for such large antigenic changes provides new information forthe design of synthetic anti-FMD vaccines . However, the screening of isolates from six decades suggests that thevirus, even within the confines of a single serotype, has exploited a minimum of its potential for antigenic variation .9 1991 Academic Press, Inc .

INTRODUCTION

Foot-and-mouth disease virus (FMDV) is an aphtho-virus of the Picornaviridae family that causes an eco-nomicallyimportant disease of farm animals (reviews inBachrach, 1968 ; Pereira, 1981 ; Domingo et al., 1990) .There are seven serotypes (A, 0, C, Asia 1, SAT1,SAT2, SAT3), more than 65 subtypes (Pereira, 1977),and myriads of variant FMDVs. Even within one sero-type, differences in antigenicity among cocirculatingviruses has in some cases impaired vaccine efficacy(Hyslop at at, 1963 ; Martinez et at, 1988). Use ofmonoclonal antibodies (MAbs) documented an exten-sive antigenic heterogeneity of the neutralizing epi-topes among FMDVs of Europe and South Americabelonging to serotype C (Mateu et al., 1988). At least24 distinct antigenic groups were defined, and lack ofcorrespondence between antigenic composition andvirus origin-place and date of isolation-was notedin several instances . This suggested that rapid fluctua-tions in the antigenic behavior of FMDV occurred innature (Mateu at al., 1988), but the amino acid replace-ments underlying the antigenic changes were notknown .

' Present address : Department of Microbiology, Mount SinaiSchool of Medicine, New York, New York .

2 To whom requests for reprints should be addressed .

695

Two important antigenic sites are located in capsidprotein VPI (Strohmaier at al., 1982). Site C is withinthe carboxy-terminal segment of the protein . The im-munodominant site A (residues 138 to 150) of FMDVtype C includes multiple, overlapping, continuous epi-topes. They are present in isolated VP1 as well as insynthetic peptides representing the relevant se-quences (Mateu etat, 1989, 1990) . Site A is within theexposed GH loop (Acharya et al., 1989), also termedthe "FMDV loop ." Several MAb-resistant (MAR) mu-tants of FMDV type C have been isolated and se-quenced. Each included one amino acid substitution atthe corresponding epitope (Mateu et al ., 1989, 1990) .Different substitutions affected the reactivity of VP1(and of complete particles) with other MAbs to widelydifferent extents . While most substitutions altered oneor a few epitopes, a particular replacement [His(146) -> Arg] abolished all epitopes identified in site A(Mateu et al., 1990) . Thus, it became important to de-termine the replacements fixed during long-term evolu-tion of FMDV in the field and their effect on antigenicspecificity. In the present report we relate the aminoacid sequence of antigenic sites A and C of FMDVs ofserotype C with the reactivity of neutralizing MAbs thatrecognize different epitopes within these sites . The vi-ruses were isolated in Europe, South America, and thePhilippines over a six-decade period (1926 to 1989) .The results document two distinct mechanisms of an-

0042-6822/91 $3.00copyright (s' 1991 by Academic Press, Inc .All rights of reproduction in any form reserved .

696 MARTINEZ ET AL.

TABLE 1

FMDV FIELD ISOLATES OF SEROTYPE C USED IN THE PRESENT STUDY0

The three groups given correspond to isolates from Europe, South America, and The Philippines .CGC Ger/26 is, to our knowledge, the first FMDV described as serotype C (Waldmann and Trautwein, 1926). The stock from WRL was

received in 1933 from Inset Riems, Germany .` C 3 Pando Ur/44 and C 3 Pando Ur/45 may be different preparations from a single field isolate .CPFA, viruses from CPFA that received a code number and no further designation .

Virus designationIsolation date(monthlyear) Place of isolation

CGC Ger/26°C, 997 UK/53

circa 19261953

GermanyUnited Kingdom

C, Loupoigne Bel/53 5/1953 Loupoigne, BelgiumC, Vosges Fr/60 1960 Vosges, FranceC, Turup Den/61 1961 Turup, DenmarkC, Noville Sw/65 1965 Noville, SwitzerlandC, Corneze Fr/66 1966 Corneze, FranceC, Haute Loire Fr/69 1969 Haute Loire, FranceC, Santa Pau Sp/70 (C-S8c1) 2/1970 Santa Pau, Girona, SpainC, Bouches du Rhone Fr/76 1976 Bouches du Rhone, FranceC, Serra de Dar6 Sp/81 (C-Si 5) 1/1981 Serra de Dar6, Girona, SpainC Monte Pulciano It/88 6/1988 Monte Pulciano, Toscana, ItalyC Carpi It/88 1988 Carpi, ItalyC Brescia IV89 3/1989 Brescia, Lombardia, Italy

C 3 Pando Ur/44` 1944-45 Pence, UruguayC 3 Pando Ur/45 1944-45 Pando, UruguayC 3 Resende Br/55 5/1955 Resende, Rio de Janeiro, BrazilC, Tierra del Fuego Arg/66 12/1966 Rio Grande, Tierra del Fuego, ArgentinaC, Leticia Col/67 9/1967 Leticia, Leticia, ColombiaCPFA 8647 Br/68° 9/1968 Campos, Rio de Janeiro, BrazilCPFA 8945 Br/69d 2/1969 S. Jos6 dos Campos, Sao Paulo, BrazilCPFA 9094 Br/69° 5/1969 Guaiba, Rio Grande do Sul, BrazilC, Argentina/69 6/1969 Pehuaj6, Buenos Aires, ArgentinaC, Paraguay/69 7/1969 Central, ParaguayCPFA 9400 Arg/69° 10/1969 General L6pez, Santa Fe, ArgentinaCPFA 9639 Arg/70d 2/1970 Santa Fe, ArgentinaC 3 Leticia Col/70 8/1970 Leticia, Leticia, ColombiaCPFA 10083 Br/70" 9/1970 Barra do Ribeiro, Rio Grande do Sul, BrazilCPFA 10085 Br/70" 9/1970 Dom Pedrito, Rio Grande do Sul, BrazilCPFA 10091 Br/70° 911970 Uruguaiana, Rio Grande do Sul, BrazilC, Indaial Br/71 10/1971 Indaial, Santa Catarina, BrazilC, Chaco Par/74 4/1974 Chaco, Chaco, ParaguayC 3 Santa Fe Arg/75 1/1975 San Carlos, Santa Fe, ArgentinaC 3 Porto Alegre Br/82 9/1982 Porto Alegre, Rio Grande do Sul, BrazilC 3 Alegrete Br/82 9/1982 Alegrete, Rio Grande do Sul, BrazilCPFA 38817 Arg/83° 3/1983 Daireaux, Buenos Aires, ArgentinaC 3 Argentina/83 3/1983 Daireaux, Buenos Aires, ArgentinaCPFA 43643 Arg/84d 4/1984 Irigoyen, Buenos Aires, ArgentinaCPFA 43647 Arg/80 4/1984 Saavedra, Buenos Aires, ArgentinaCPFA 43652 Arg/84d 5/1984 San )usto, Santa Fe, ArgentinaC3 Argentina/84 9/1984 General Roca, Cordoba, ArgentinaCPFA 45892 Arg/84" 12/1984 Colon, Buenos Aires, ArgentinaC3 Argentina/85 12/1984 Marcos Ju6rez, Cordoba, ArgentinaCPFA 45913 Arg/85 2/1985 La Vi%ita, Catamarca, ArgentinaCPFA 45915 Arg/85 2/1985 Union, Cordoba, ArgentinaC3 Santa Marfa Br/87 8/1987 Santa Maria, Rio Grande do Sul, Brazil

C Philippines/3/87 10/1987 Santa Cruz, Lubao, Pampanga, PhilippinesC Philippines/1/88 2/1988 Cotabato, Mindanao, Philippines

W

W

pne 1/53rIll

I~:.' ~1nt7~r~IT(1cc rC Breseb h/89

SITE A

pY`SEEMsunWNROSSDonsNMM

JEWS!Owes

mill

0MENman**

So

mammonoolOU∎~4assnu

usae~%soon%mooseWOMENOWNERson a

UEUEnnmuseum

FMOV MAb

ugi13&11'rerr.IllisSaC3Resende Br/55

C, Argentina/85r,

FT Pniklppines/3/87Ippines/1/88 IMMUNEwoman

ANTIGENIC DIVERSIFICATION OF FMDV

697

arr

Ac. 1 . Relative reactivity in EITB assays of field isolates of FMDVtype C with neutralizing MAbs directed to antigenic sites A or C ofVP1 . The MAbs that recognize distinct epitopes in the VP1 segment138 to 150 (antigenic site A) have been divided in C, (SD6, 5A2, 4G3 .4C4, 6D11) and C 3 (7CA11, 71D1, 7FC12, 7AH1, 7EE6, and 7AB5) .They were elicited against FMDV C, and C„ respectively (Mateu eta1., 1987, 1988, 1990) . MAb 7JA1 and 6EE2 were raised againstFMDV C 3 (Mateu et a/., 1988) and A12 (Robertson et al., 1984),respectively ; they recognize two epitopes within site C. For eachMAb, a relative reactivity (a) of any virus i with respect to the homolo-gous virus s (either C, Santa Pau Sp/70 or C 3 lndaial Br/71) is definedas the amount of product formed with virus i in the calorimetric assayrelative to that with virus s and is quantitated by densitometry of thebands. Reactivities are given as positive (a, >0 .5, •) , intermediate(0 .2 < a ; < 0.5, 0), very weak (0 .01 < a, < 0 .2, D) or negative (nosignal detected, a, = 0, 0) according to the criteria defined by Mateuet a!. (1988) . The place and date of isolation of the viruses are givenin Table 1 . The letters and the last two digits following the virus namecorrespond to the country and year of isolation . For seven isolates,more than one virus preparation was used, as indicated in the subdi-

tigenic diversification of FMDV and suggest that thevirus harbors considerable potential for further varia-tion .

MATERIALS AND METHODS

Viruses

The FMDV preparations used were those availableat the World Reference Laboratory (WRL), Pirbright,U .K . ; the Pan American Foot-and-Mouth DiseaseCenter (CPFA), Rio de Janeiro, Brazil ; Servicio Nacionalde Sanidad Animal (SENASA), Buenos Aires, Argen-tina; Rhone-Merieux (RM), Lyon, France ; and Labora-torios Sobrino S.A ., Olot, Girona, Spain. In somecases, two or more preparations of the same isolatediffering in passage history were analyzed . Differencesin reactivity with MAbs or in nucleotide sequenceamong such preparations have been indicated in thecorresponding figures and Tables. The place and dateof isolation of the FMDVs analyzed are indicated inTable 1 .

Monoclonal antibodies

All MAbs except 6EE2 were raised against FMDVtype C and have been described (Capucci et al., 1984 ;Mateu etal., 1987, 1988, 1990) . MAb 6EE2 was raisedagainst FMDV A12 (Robertson et a/., 1984), and itcross-reacts with FMDV type C (Grubman and Mor-gan, 1986 ; Mateu etal., 1987). Each MAb used recog-nizes a different continuous epitope in site A or C ofcapsid protein VP1 (listed in Fig . 1) (Mateu et al., 1989,1990; Baxt et al ., 1989). The terms antigenic site andepitope are used according to the definition of Wimmerand Jameson (1984) . Antigenic site describes a regionof a capsid protein that binds antibodies . Epitope is themolecular conformation of the site that an antibody willrecognize and bind . Thus, a single site may includemany epitopes .

Immunoassays

For the enzyme-linked immunoelectrotransfer blot(EITB) assays, viruses were concentrated (Mateu et al .,1987) and disrupted by boiling 2 min in 80 mM Tris-HCI (pH 6 .3), 10% SDS, 8 M urea, 1 .2 M 2-mercap-toethanol, 18% glycerol, 0 .02% bromophenol blue .Proteins were electrophoresed in a SDS-polyacryl-amide gel (Laemmli, 1970) containing 8 M urea . Then

~~.

"mil

visions within the list of FMDVs . Generally only minor differences ofreactivity were noted between different preparations of the sameisolate, and the origin of such differences was not investigated .About half of the reactivities have been reported previously fMateu etat, 1988, 1990) and are included here for completeness .

iiWEiiY1'Y,lii~

??F7d~1a~rlir

nowwasWallM: .

=amanSC FA 4 647 Arq

wink∎C, Argentina/84

fta'll•1118[f3:rFf.S1 mums

manwgikm7yi1 1~'Trio

hiss :CsArgentina/69

rid=EnnowMtaa aimtonNOONENMNNN

ttra;ai 1F1r7t1~ iiiC,Indaias Br/71

f%

a'7c' a J Oman

u.nnn#cV1

w, 'KulNNlow .IMtt~tatc'~tttutstni R IonRUM.∎

698

TABLE 2

CONSERVATION OF EPITOPES IN SITE A OF FMDV SEROTYPE C

Determined by comparing the reactivity in EITB assays for eachMAb-FMDV pair with the reactivity of the corresponding MAb witha reference antigen, either C, Santa Pau or C 3 Indaial (homologousreactivity ; black square in Fig . 1) . Values are the percentage of reac-tivities that coincided with the homologous reactivity according tothe criteria described in the legend for Fig . 1 . For the comparison,either individual MAbs, the subset of those raised against FMDV 0, ,(MAbs C,), against FMDV C3 (MAbs C 3), or all available MAbs (AI ))were considered, as indicated in the first column. Likewise, the typeC isolates have been divided in five subsets that, to facilitate compari-sons, have been grouped as All (all isolates considered), virusesfrom Europe or South America, and viruses belonging to subtypes C,or C3 , respectively .

the gel was equilibrated in 25 mM Tris-HCI (pH 8 .3),192 mM glycine, 20% methanol for 1 hr at room tem-perature, and transferred to nitrocellulose (Bio-Rad) inan electrotransfer apparatus at 250 mA for 5 hr at 4°Cin the same buffer (Towbin et al., 1979); 0.7 µg of VP1were used in each assay. Transfer to nitrocellulosewas monitored by Coomassie blue staining of the geland Amidoblack staining of a nitrocellulose strip . Toassay the reactivity with MAbs, the nitrocellulose sheetwas saturated with 3% BSA in phosphate buffered sa-line (PBS) and incubated 1 .5 hr at room temperaturewith the appropriate dilution of supernatant of hybrid-oma culture adjusted to 1% BSA . The nitrocellulosewas then washed twice with PBS-0 .05% Tween-20(used in all subsequent washings) and incubated for 1hr at room temperature with a 1 :1000 dilution of goatanti-mouse IgG coupled to peroxidase (Bio-Rad) . Afterextensive washing, color was developed by incubationwith 0 .015% H 2O2 , 0.5 mg/ml 4-chloro-1-naphtol, 20%methanol in PBS. Color development was stopped byextensive rinsing with water .

MARTINEZ ET AL .

Nucleotide sequencing

FMDV RNA was prepared and sequenced by primerextension and dideoxy chain termination as previouslydescribed (Zimmern and Kaesberg, 1978 ; Sobrino etal., 1986). VP1 RNA from FMDV CGC and C, HauteLoire were also amplified by using the polymerasechain reaction and sequenced (Jansen at at, 1990)without molecular cloning of the cDNA. Thus, in allcases the average nucleotide sequence present in thepopulation of RNA or cDNA molecules was deter-mined .

RESULTS

Reactivity with MAbs of FMDVs of serotype Cisolated from 1926 until 1989

EITB assays of 47 FMDV type C field isolates with 13neutralizing MAbs that recognize distinct, continuousepitopes on antigenic sites A and C (residues 138-150and the C-terminus, respectively) of capsid protein VP1produced 27 different patterns of reactivity (Fig . 1) . Theextent of conservation of the 11 epitopes located onsite A was expressed by comparing the reactivity ofeach MAb-VP1 pair (given in Fig . 1) with the reactivityof that MAb with VP1 of the eliciting virus (Table 2) . Asexpected, higher conservations were scored when theVP1s were probed with MAbs raised against the ho-mologous than with MAbs against the heterologousvirus subtype . Both highly variable and significantlyconserved epitopes were found within site A (Table 2),whereas the two epitopes located at site C were highlyconserved . FMDV C, Haute Loire, C.Argentina/69 andthe two Philippines isolates showed low or no reactivitywith all site A MAbs tested (Fig . 1) .

Correlation between antigenic and amino aciddivergences of sites A and C

To investigate the molecular basis of the antigenicvariation revealed by neutralizing MAbs directed tosites A or C (Fig . 1), the VP1 RNA segment of 16 newFMDV type C isolates of the five recognized subtypes(or unclassified) was sequenced . The deduced aminoacid sequences were aligned with those previously de-termined for eight other type C viruses (Fig . 2). These24 FMDV isolates represent most of the different anti-genic specificities shown in Fig . 1 . Replacements (Fig .2) are preferentially located in surface-exposed loopsand especially at sites A, C, and around the BC loop(approximately residues 42 to 48, probably part of avirion conformation-dependent antigenic site ; Kitson etal., 1990) .

The present study is centered on the analysis ofsites A and C. The possible implication in antigenic

Degree of antigenic conservation (%)among FMDV type C isolatesa

MAbElicitingsubtype All Europe

SouthAmerica C, C 3

SD6 C, 11 36 0 45 05A2 C, 13 43 0 54 07AB5 C, 27 7 35 0 387EE6 C, 58 7 80 0 69403 C, 29 86 0 100 07AH1 C 3 62 14 84 0 1007FC12 C 3 64 21 87 9 1004C4 C, 42 86 32 100 236D11 C, 55 93 55 100 617CA11 C, 91 93 90 100 10071D1 C 3 91 93 90 100 100

MAbs C, 33 68 17 80 6MAbs C3 65 39 77 35 87All 51 52 50 55 55

BC AC CD

ANTIGENIC DIVERSIFICATION OF FMDV

10

20

30

40

50

60

70

Bo

90

100

110

diversification of other variable residues in VP1, VP2

and VP3 requires delineating additional antigenic sites

in FMDV type C .

For this set of 24 FMDV isolates the variability index

of Wu and Kabat (1970) was calculated for each aminoacid around sites A and C (Fig . 3) . At site A, two hyper-variable stretches (residues 138 to 140 and 148 to

150) flank a conserved Arg-Gly-Asp-Leu-Ala sequence .

699

FIG . 2 . Alignment of amino acid sequences of capsid protein VPI of FMDVs of serotype C . Viruses are grouped according to geographicalorigin (Europe, South America, or the Philippines), and ordered chronologically within each group ; c indicates a plaque-purified virus . Eightsequences (of C, Oberbayern, C, Santa Pau (C-S8), C, Serra de Dar6 (C-Si 5), C, Barcelona (C-S30), C, Resende Br/55, CS Indaial Br/71, C3Indaial Br/71 (78), C, Argentina 84, C, Argentina 85) were reported previously (Makoff et al., 1982 ; Beck or al, 1983 ; Villanueva et a1., 1983 ;Cheung et al., 1983; Sobrino et al ., 1986; Martinez etch, 1988 ; Piccone et al., 1988; Gebauer at al., 1988) . The single letter amino acid code isused. Only amino acid changes relative to FMDV CGC Ger/26c sequence are indicated . For some isolates, preparations from different laborato-ries were sequenced . Symbols : dot, deletion ; asterisk, ambiguity in the sequencing gel ; dash, sequence not determined . Amino acid numberingis that of previous reports (Mateu e1 al., 1987, 1989, 1990) . Amino acid insertions between residues 48-49 and 140-141 (found only in someviruses) have not been numbered . On top of the first sequence, the secondary structure is indicated, assuming that it corresponds to theequivalent positions of FMDV 0, BFS (Acharya et al ., 1989) upon alignment ; symbols : thin line, N- and C-terminus and loops ; arrow, 0-sheet ;wavy line, a-helix . Nomenclature of secondary structure domains is according to Acharya et al. (1989).

At site C, variable positions are scattered along the24-amino acid domain . The index values show thateven at the most variable residues, a minimum of thepotential variation (that ranges from index 1 to 240) has

occurred (Fig . 3) .

To correlate the amino acid divergence at site A withantigenic divergence-defined as the number of dis-similar reactivities with the panel of 1 1 MAbs directed

4C Cer/26r TTTTCESAOMTNW1pf'EIGVQRRHHTWAFVLCNFVAVQV LD SNROS1V •ALLILUTYTFSOLl3NVAN7f1q,MVRA:APVSALL71'SNPPAYXAGRLTNLALPVCSC Gar/26 G • 1

A RAC2 997 UB/53 V V • PC1 Loc.oigne "1/53 A T V A N G • T PC1 unerbayern Get/60 T V A N C T PC3 Voagee Pr/64 A T V A N G T PC1 1Vnop Dan/6l A ' • V A N C T T PCl Noville SY/65 T V A N G T T PCl Naute Loire Fr/69 T V A N C S T T PCl Santa Pau Sp/70c T O v A N G T N T PVCl Serra de 0 r6 Sp/81c T T PC1 Barcelona SP/82 N

I

T V A I T T PC Nome Polciano 11/68 . -- . .___ ._ -.- ..c Brescia It/89 - . ----------- ... . . _ .C3 Pando Ur144 A ` N N V G T Q rPVC3 Resende Br/55 P CR 0

y V G I T T PC3 Reeende Br/55c12 N CR Q

•V V C I T Q T PC3 Resende Br/55C P 1M Q

V V -

- -

-CL Tierra del Puego Arg/66 A •

KA V V G I •' T • PCS Argentina /69 P G Q

V V C • I T T PC3 Indaial Br/71 V V C I T A Pci Indaial Br/71 (78) N V V G I T A PC3 Santa Fe arg/75C ----V V C L A I T • T -----------Ci Argentina /63c E

P pR Q V V G I T T PCI Argentina /01 P L71C Q

V V G I T L T PC3 Argentina /86C P LW Q

V V ---------------- --C3 Argentina /85 A T V A N A I T T PC3 Argentina /85c A V A N -C M,ilipines/3/07 • ' Y IP W Q

L V G IN T PC PAfllpfnea/1/80 • •

Y IP IS Q

L V C • I N T P

015 GH

H170

113i

0120 130

160 150

160 100 190 200 209coo Cer/26cC4'C Gar/26

TAMIXVLATVTWTTAY'T sw CpWMV797ARHLPPSn4roAVKAETITH.LvRM tAELYCPRPILPVQPIGMUI1cQPLIAPAxQLL

A

R I • SC2 997 UK/53 A T T ACl LcwOigne Bel/53 A T A A I VC1 OOertayern Cer/60 G T A TAR G I VCl Voagea Fr/60 A T A TA I VC1 7Vrup gm/61 A T A TA I VCl Noville by/65 A T A TA I • VCl haute Loire Fr/69 A T A V A VCI Santa Pau Sp/70C A T A A TT VCf Serra de Qard sip/B1, A T T ITA VC1 Barcelona Sp/82 A T T . WA VC None Pulcfano It/BB TA • VC Breads lt/89 A 'T A TA VC2 Pando Ur/6f A T T A C A R V I EC3 Reaende Bt/55 T T T A TA I 5C3 Reseede Br/55c12 T T • T A U TA V SC3 Resende Br/55c T T T A TA L SC6 Tlerra del Neg. /rg/66 T T T A A QR I A ECS Argentina /69 R T T A Q AA IC3 Indaial Br/73 A 7GV AA V AC3 Indaial Br/11 (78) T A A AA V0 Santa Fe Arg/75c T T V V AC3 Argentina /BIc T T T A TA V 5C3 Argentina /86 T T T A TA V 5 5C3 Argentine /Nc T T T A TA VCI Argentina /85 T ACA AA V YC3 Arganrin /85C T BACV AA V GC PAlllplnea/3/87 •

A T T . . Q TS R D V

• L •V E • 5C Pllilipinea/1/88 E •A T T . Q TS R Q V L •V B' 5

700

1

to site A-two consensus sequences were defined forVP 1 residues 138 to 150, one for the European iso-lates (CE) and another for the S. American isolates (CA ) .Then, site A sequences were ordered by their progres-sive divergence relative to the corresponding consen-sus (Fig . 4) . It is clear from such an alignment that bothlimited or extensive changes in epitopic compositioncan be brought about by few amino acid replacements .To measure if an increase in the number of amino acidsubstitutions led to an increase in antigenic diver-gence, for each pair of viruses the number of dissimilarreactivities with MAbs directed to site A was plottedagainst the number of amino acid differences withinsite A (Fig . 5). This analysis was carried out for all vi-ruses listed in Fig . 4 (Fig . 5A) . From this analysis it wasobserved that the points that showed the poorestcorrelation corresponded to viruses C, Haute Loire Fr/69, Ca Argentina/69 and C Philippines. When these

MARTINEZ ET AL .

FIG. 3 . Amino acid variability at antigenic sites A and C of 24 representative isolates of FMOV type C . The sequence of one preparation of eachdifferent virus isolate in Fig . 2 has been used for the calculation . The number of times that an amino acid is found at any given position (countedfrom the alignment of Fig . 2) is indicated . Undefined amino acids (asterisks in Fig . 2) were not considered in the calculations . The variability indexis the ratio between the number of different amino acids found at the considered position and the frequency of the most common amino acid atthe same position (Wu and Kabat, 1970) . Site A as defined in Mateu et al. (1990) is delimited by discontinuous lines . The tripeptide Arg-Gly-Asp(highly conserved in FMDV and proposed as apart of the receptor binding site) is boxed . The range of possible values for the variability index is 1(only one amino acid found) to 240 (four different amino acids occurring twice and 18 found once in the 24 isolates compared) .

viruses were excluded (Fig. 513) a linear regressioncould be approximated and a highly significative corre-lation coefficient was obtained . The same was truewhen the analyses involved virus subsets C E (Fig . 5C)or CA (Fig . 5D) . Thus, as a general trend, the increase inthe antigenic divergence paralleled the accumulationof amino acid replacements . This analysis also showsthat CE and CA viruses constitute two genetically andantigenically discrete clusters (Fig . 5B). CGC Ger/26,one of the earliest FMDVs available, appears to be in-termediate both in amino acid sequence and in reactiv-ity with MAbs between groups C E and CA (Figs . 4 and58) . This point is under study by defining the phylogen-etic position of the entire P1 segment of CGC in rela-tion to other isolates .

Site C shows higher conservation than site A regard-ing both reactivity with MAbs (Fig . 1) and amino acidsequence (Figs . 2 and 3) . The lack of reactivity of the C

-

II

I

III

I

SITE A II

I

I

II

T24 T23 T20 Y23 T24; T13 S20 112 R12

'12V2

L24 A23 H20 L22 A15 A17 IR23 624 021 T 16IH24 A24 R20 H19

A4 I A1o G2 A9 G1 V1 Q3 I, T9 16

V1

V1

A6

S2

1

1

I

at R5

1 •21

133 134 135 136 137 1 136 139 140 140a 141VPI

142POSITION

143 144 145 146 147 148 149 150 P51I

152 153 154

SITE

r

C

P24 h0V4

L24 P24 hoV4

0-24 P23 T17 6 24 On R23

A3

S1

S1

N22

Y1K24 (124P22

L21_21 V13 A24 P24

I1lA20E4

K20 0.21 L24 L15

S6

156 167 166 169 190 191 192 193 194 195 196VP1

197 196 199 200 201 202 203 204 205 206 207 208 209POSITION

L C

C A

ANTIGENIC DIVERSIFICATION OF FMDV

FMDV FMOVLOOPSITE A

I))

131141NOa

105%% 1150

3t

Consensus Europe C6

TTTYT

ARM

CI Vosges Fr/60C1 Turup Den/61C, Noville Sw/65C Monte Pulciano IF/88C Brescia It/89C, Loupoigne Bet/53C, Haute Loire Fr/69

C, Santa Pau Sip/70cC, Serra de Dard Sp/61c

C GC Ger/26

TTAYT

Consensus South America CA TTTYT

ARM

C5 Argentina/69C3 Resende Br/55C) Argentina/83r

C3 Argentina/84CA Tierra del Fuego Arg/66

ORC2 997 UK/53

RC1 Pando Ur/44

RC3 Indaial Br/71

AC3 Santa Fe Arg/75c

AC3 Argentina/85

A

C Philippines/3/87

TTTYTC Philippines/1/88

Philippines isolates with MAbs 71A1 and 6EE2 and thediminished binding of 6EE2 to some isolates correlatewith replacements Pro (200) - Leu and Ile (202) yVal, respectively . (Fig . 2) .

Two mechanisms of antigenic diversification ofFMDV of serotype C

Substitutions at the two hypervariable domains ofsite A had a minor influence on the antigenic specificityof this site in most viruses within each of the twogroups defined by consensus C E and CA (Fig . 4). Thedifferent spectrum of reactivities with MAbs depictedby FMDV groups C E and CA could be due to the addi-

701

FIG . 4. Alignment of the amino acids of site A (138 to 150) of FMDV type C and reactivity with neutralizing MAbs directed to this site . Isolatesfrom Europe and S . America are ordered according to increasing divergence from the respective consensus sequence Cc or CA . C 2 997 UK/53has been included among the C A viruses because of the close relatedness of its VP1 sequence with those of S . American isolates . Amino acidsthat differ between the two consensus are written in boldface letters . Forviruses C, C,,, and C Philippines 1/88, only amino acids that differ fromthe respective consensus sequences or from C Philippines 3/87, respectively, are indicated . CGC shows a sequence intermediate between Ccand C„ (see text) . Amino acid numbering is as in Fig . 2 . Note that Arg 140a is absent in one of the Philippines isolates . Vertical lines delimit themore variable positions within site A . Dashes above and below the CGC Ger/26 and C Philippines sequences indicate differences of theseisolates relative to consensus C r and CA , respectively. Other symbols are as in Fig . 2. The amino acid divergence--number of amino aciddifferences-of each isolate with respect to the two consensus C E and CA is given . Reactivities with MAbs are those indicated in Fig . 1 . Notethat for some viruses showing identical sequence at residues 133 to 154 small differences of reactivity with some MAbs were recorded . Thiscould be due to minor effects of distant residues in refolded VP1 or, more likely, to differences in the mutant spectra of the virus populations, asdocumented previously (Mateu et al., 1989) ; such differences were not investigated .

tional Arg at position 140a present only in C A and CGCFMDVs and/or to multiple replacements at the anti-genic domain . In contrast to the limited effects of mostamino acid substitutions, a drastic influence on anti-genicity was traced to some single amino acid replace-ments . Comparison of C, Haute Loire Fr/69 and C,Loupoigne Bel/53 shows that substitution Ala (145) -Val diminished or abolished the binding of all MAbstested (Figs . 4 and 6) . Position 146 appeared as criticalfor antigenic specificity, as shown by the lack of reactiv-ity of FMDV C 5 Argentina/69, of MAb-resistant mutantsand of variants rescued from persistent infections ofcattle (compare Fig . 6) . His (found in all viruses used to

AMINO ACIDDIVERGENCERELATIVECONSENSUSC

TO

EA

REACTIVITY WITH MAbsCI C)

0 5

0 50 5 6f loot0 5 .///0 5 N long0 5 's 'own1 4 noun2 52 5 NON2 5 00

U. X O

5 0

6 1 nommemosson6 1 NONE 11%

6 1 some ON6 1 ∎Eon lC4 1 U.. N4 1 now s5 2 O•.s s6 2 U. ..4 2 Bongo5 2 .me..

6 6 unn.nn.nn..7 5 ∎.n•unnnnnn

702

9

B

A

C

a 3u :m

.: .

. ..

i

Fwi :c

B

D

r map

IA >R . xr . .

'F

111'

5 6 7 0 1 2 3 4 5 6 7AMINO ACID DIVERGENCE

FIG . 5 . Correlation between antigenic and amino acid divergencesof site A for any pair of FMDV type C isolates . Antigenic divergence isgiven as the number of dissimilar reactivities with the panel of 11MAbs directed to site A (Fig . 1) . Amino acid divergence is the num-ber of amino acid differences within VP1 positions 138 to 150 (Figs .2 and 4) . The viruses compared are : (A) All possible pairs among theisolates listed in Fig . 4 . (8) Same as (A), except that C, Haute Loire,C, Argentina, and C Philippines are excluded . (C) All possible pairsamong the C E viruses listed in Fig . 4 . (D) All possible pairs among theCA viruses listed in Fig . 4 . Symbols for panel (B) : (0) both viruses ofthe pair were from the same group (C E or C A as indicated in Fig . 4);(0) each virus of a pair was from a different group (C E and C A) ; (X)pairs involving FMDV CGC . Symbols for (C) : (0) all pairs except thoseincluding FMDV C, Haute Loire Fr/69 ; (a) pairs that included FMDVC, Haute Loire . Symbols for (D) : (•) all pairs except those includingFMDV C5 Argentina/69 ; (o) pairs that included FMDV C 5 Argentina/69 . The correlation coefficients are : (A) r = 0 .73 ; (8) r = 0 .85 ; (C) r- 0 .68 (this value becomes r = 0 .72 if C, Haute Loire Fr/69 is ex-cluded) ; (D) r = 0 .33 (this value becomes r= 0 .58 if C5 Argentina/69is excluded). In C, D the regression lines shown were calculatedexcluding C, Haute Loire Fr/69 and C 5 Argentina/69 .

raise the MAbs) or Asp at position 146 yielded positivereactivity with all MAbs, whereas Arg, Gln or Leu at thesame position drastically diminished or abolished MAbreactivity (Figs. 4 and 6) .

Thus, during the evolution of FMDV serotype C inthe field, the antigenic diversification of the main anti-genic site A occurred by two distinct mechanisms .One involved gradual modification of epitopes broughtabout by a parallel accumulation of amino acid replace-ments, and the other consisted in abrupt changes ofantigenic specificity caused by single replacements atcritical positions .

MARTINEZ ET AL .

DISCUSSION

Relevance of site A on the antigenic diversificationof FMDV of serotype C

Residue 146 of VP1 has a dominant influence on theantigenic specificity of site A of FMDV serotype C (Fig .6) . It could be argued that in the EITB assays somerefolding of VP1 may occur that would allow an influ-ence on site A of residues distant in the primary struc-ture. However, a replacement His (146) -• Arg was theonly difference in VP1 between C, Santa Pau and MARmutants SD6-28 and 4G3-2 that did not react with anysite A MAb. Similarly, His (146) -. Leu was the onlysubstitution between CGC and its MAR mutant deriva-tive 7A85-1 (Fig . 6) . Also, the MAbs reacted to a similarextent with whole virus and with synthetic peptidescomprising VP1 residues 138 to 156 suggesting thatthe corresponding epitopes are entirely included withinthis segment. Moreover, peptides mimicked the effectof substitutions found in FMDV variants on the reactiv-ity with those MAbs (Mateu at al., 1989, 1990 andunpublished results), Accordingly, a peptide with His-146 was reactive with the MAbs directed to site A,whereas a peptide that included Arg-146 as the onlydifference, was not reactive (D . Andreu, E . Giralt andM . G. Mated, unpublished experiments) . No variantswith replacements at positions 145 or 146 that deviatefrom the reactivity profiles indicated here (Fig . 6) havebeen found, and no other virus of serotype C testeddisplayed such a divergent antigenic specificity withour panel of MAbs . In contrast to our observations,replacement His (146) Arg was not scored as anti-genically relevant when introduced into an hexapep-tide with residues 143 to 148 identical to those of VP1of FMDV C, Santa Pau Sp/70 (Geysen et al., 1985). Weattribute this discrepancy to the fact that hexapeptidesare insufficient to detect most of the epitopes of site A,that are nevertheless mimicked with longer peptides(Mateu et al ., 1989, 1990) . An additional problem wehave encountered in the use of replacement sets isthat the effect of an amino acid substitution may becritically dependent on neighboring residues (Mateu etal., unpublished results), as recently described for aT-cell determinant (Boyer et al ., 1990). We are nowstudying if positions 145 and 146 are points of contactwith antibodies or if certain substitutions at those sitesinduce a global structural change of site A .

Several independent antigenic sites (Thomas at al .,1988 ; Kitson et al., 1990) may contribute in a highlysignificant way to the overall antigenicity of FMDV . Thefollowing evidence suggest that site A, either alone orin combination with other capsid domains, (Parry et al.,1989, 1990), is one of the immunodominant regions ofFM DV: (i) Specific cleavage of the viral capsid at sites A

FIG . 6 . Effect of replacements at VP1 positions 145 and 146 on the antigenicity of site A of FMDV type C . In addition to isolates described inthis report, the first column lists the following variant FMDVs : MAR CSP, MAR mutants of FMDV C, Santa Pau Sp/70 selected with MAbs SD6 or4G3 (Mateu et al., 1989, 1990); MAR C GC, a MAR mutant of FMDV C GC Ger/26 resistant to neutralization by MAb 7AB5 (Hern6ndez et at,unpublished result) ; C785/63 and C788/63, variants from cattle persistently infected with C . Resende c 12 (Gebauer et al., 1988) . Symbols are asin Fig . 2 . S/R indicates a double band in the corresponding position of the sequencing gel . Reactivities with MAbs are as in Fig . 1 . ND, notdetermined . For each virus the same preparations was used to determine nucleotide sequence and reactivity with MAbs . Site A and the criticalpositions 145 and 146 are boxed .

or C greatly diminished the antigenicity of the particles(Wild at al ., 1969; Strohmaier at al ., 1982); (ii) Syntheticpeptides representing site A adsorbed most of the neu-tralizing antibodies present in animal sera (quoted byPfaff at at, 1982 and Brown, 1990); such peptides,alone or in combination with site C peptides, inducedneutralizing and protective responses in animals (Pfaffat al., 1982 ; Bittle et al., 1982 ; Di Marchi et al ., 1986 ;Clarke et al., 1987) ; (iii) . Substituted peptides mimickedserotype and subtype specificity (Clarke at at, 1983),as well as the behavior of variant viruses with polyclo-nal sera (Rowlands at al., 1983); (iv) Many neutralizingMAbs raised against FMDV are directed against site A(Mated et al., 1987, 1988, 1990 ; Bolwell at al., 1989a) .

Thus, the two modes of antigenic variation identifiedin site A probably contribute substantially to the overallantigenic diversification of FMDV . This provides an ex-ample of a type of antigenic "shift"-as opposed toantigenic "drift"-due to critical amino acid replace-ments, as first suggested by Houghten et al. (1986) .

Potential for further antigenic variation andimplications for vaccine design

Our analysis included the most representative typeC isolates from Europe and South America diagnosed

at the WRL and CPFA during several decades (Table 1) .This permits addressing the question whether FMDVtype C, during its natural evolution, has exploited mostof its potential for antigenic variation or if additionalchange is possible and likely . The results suggest thelatter alternative for the following reasons . Most posi-tions at site A have used a restricted subset of all possi-ble amino acids that would be generated by a pointmutation from any parental nucleotide sequence (un-published compilations). That many of such substitu-tions would lead to viable FMDV is shown by their oc-currence in variants selected in cell culture (Domingoat al., 1990). In addition, during several episodes ofFMD in Spain from 1970 until 1982, amino acid His-146 remained invariant . However, FMDV with the anti-genically critical His (146) --.. Arg replacement was iso-lated from vesicular fluids of swine, without passage ofthe virus in cell culture (Carrillo at al., 1990), indicatingthat the variant can replicate in swine . These observa-tions suggest that FMDV, as evolved during six de-cades, still embodies high potential for further anti-genic variation. Obviously, we cannot exclude thatother replacements in site A may be restricted by theiradverse effect on viral fitness .

Two levels of antigenic heterogeneity are relevant forvaccine design: (i) Variants within one isolate (Row-

ANTIGENIC DIVERSIFICATION OF FMDV 703

FMOV VARIANTS FMDV LOOP REACTIVITY WITH MAbsSITE A

1331

138

145146 158 151I

N~~~$r ~ yx~1

I

I

SELECTION C, Santa Pau Sp/70c T T T Y T A S A • R G D L A H

R

R

L 1 T T H A R H ONMENEMEMOF MAR

MAR CSP S06-28MUTANTSMAR CSP 463-2

CGC Ger/26MAR AC 7AB5-1

T T A Y T .STRRG0LA H

L

L A V T H A R H an

TTTYTA Y

7 S A R R G D L A Da

L A T A H A R HPERSISTENT C3 Resende c12INFECTION C 785/63

T T T Y T T S A R R G 0 L A H L A A A H A R H salt\C 788/63

ACUTE

Cs Argentina 69 aINFECTIONS

1 3 Resende Br/55 T T T Y T T S A R R G 0 L A H L A T A H A R H flu. 1%

C Philippines 1/88 T a S R

C Philippines 3/87 T • a S R

C, Loupoigne Bet 53C, Haute Loire Fr 69

T T T Y T AST •R G0L H L A A T H A R H

704

MARTINEZ ET AL .

lands et al., 1983; Bolwell et al., 1989b ; Mateu et at,

REFERENCES1989 ; Diez et al ., 1989, 1990) . (ii) Antigenic differencesbetween isolates, with the occasional appearance ofhighly divergent viruses such as C, Haute Loire Fr/69,C a Argentina/69 and C Philippines in our survey (Fig . 4) .Level (i) poses the problem that minor variants may beselected as a result of poor immune responses in ani-mals. Level (ii) requires that different protein se-quences be included in vaccine formulations . Indeed aMAb-resistant mutant with replacement His (146) -*Arg was neutralized with a 10-fold lower efficiency thanits parental FMDV by sera from vaccinated swine (Ma-teu et al ., unpublished results) . Thus, a synthetic vac-cine formulation intended to control FMDV type C, mayrequire the inclusion of peptides of the two most dis-tant antigenic specificities defined by our analyses(Figs . 4 and 6) . It is noteworthy that field isolate C 5

Argentina/69 was classified by WRL as a new, widelydivergent subtype (Arrowsmith, 1975) . Since VP1 ofthis isolate differs from that of C 3 Resende-thecurrent type C vaccine strain at that time-in only fiveamino acids, including His (146) --. GIn (Fig . 2), it islikely that this replacement had a major effect on itsdiffering antigenicity (compare Fig . 6) . Sequence hy-pervariability and its effects on antigenic diversity havebeen documented for immunodominant sites of otherRNA viruses, including the V3 loop of human immuno-deficiency virus (Coffin, 1986 ; Modrow et al., 1987 ;Rusche at al., 1988; Palker et al., 1988 ; Meloen et al.,1989; Neurath and Strick, 1990 ; Simmonds at at,1990; Zwart etal., 1991 ; recent review in Kurstak et al.,1990) . Since antigenic variation affects many RNA vi-ruses, molecular epidemiological surveys of the typereported here are important for the design of vaccinesintended to control RNA viruses .

ACKNOWLEDGMENTS

We are indebted to M . Lombard (Rhone Merieux, Lyon, France), A .Alonso and H . Barahona (Pan American Foot-and-Mouth DiseaseCenter, Rio de Janeiro, Brazil), G . Fern6ndez and E. FernBndez (Servi-cio Nacional de Sanidad Animal, Buenos Aires, Argentina), E . Broc-chi and L. Capucci (Istituto Zooprofilattico Sperimentale delta Lom-bardia a dell'Emilia, Brescia, Italy), D . Morgan (Plum Island AnimalDisease Center, Greenport, U.S.A.), and 1 . Plana (Laboratorios So-brine, Clot, Girona, Spain) for the generous supply of some viralisolates and MAbs . We thank F . Sobrino and E . Martinez-Salas forvaluable discussions. Work at CBM was supported by CICYT, Fondode Investigaciones Sanitarias, Consejo Superior de InvestigacionesCientfficas (Spain) Fundaci6n Ram6n Areces, and the EuropeanCommunity. Work at WRL was supported by MAFF (U .K .) . Work atINTA was supported by SECYT (Secretana de Ciencia yT6cnica) andCONICET (Argentina) . M .A.M. visited WRL supported by @ fellowshipfrom the European Community .

AOHARYA, R ., FRY, E ., STUART, D ., Fox, G ., ROWLANDS, D ., and BROWN,F. (1969) . The three-dimensional structure of foot-and-mouth dis-ease virus at 2 .9 A resolution . Nature (London) 337, 709-716 .

ARROWSMITH, A . E . M . (1975) . Further data on FMD type C in Jersey .In "Report of the twenty-first session of the European Commissionforthe control of foot-and-mouth disease ." Appendix Ii : WRL infor-mation sheet No . 24, pp . 65-66. FAO, Rome .

BACHRACH, H . L . (1968) . Foot-and-mouth disease virus . Annu. Rev.Microbiol. 22, 201-244 .

BAx-r, B ., VAKHARIA, V ., MOORE, D . M., FRANKE, A. J ., and MORGAN,D. 0, (1989) . Analysis of neutralizing antigenic sites on the surfaceof type A, z foot-and-mouth disease virus . J. Virol. 63, 2143-2151 .

BECK, E ., FEIL, G ., and STROHMAIER, K . (1983) . The molecular basis ofthe antigenic variation of foot-and-mouth disease virus . EMSOJ.. 2,555-559 .

BITTLE, J . L ., HOUGHTEN, R . A., ALEXANDER, H ., SHINNICK, T . M ., SUT-CLIFFE, J . G ., LERNER, R . A ., ROWLANDS, D . J ., and BROWN, F . (1982) .Protection against foot-and-mouth disease by immunization with achemically synthesized peptide predicted from the viral nucleotidesequence . Nature (London) 298, 30-33 .

BOLWELL, C ., CLARKE, B . E ., PARRY, N . R ., OULDRIDGE, E . J ., BROWN, F .,and ROWIANDS, D . ) . (1989a) . Epitope mapping of foot-and-mouthdisease virus with neutralizing monoclonal antibodies . J. Gen .Virol . 70, 59-68 .

BOLWELL, C ., BROWN, A. L ., BARNETT, P . V ., CAMPBELL, R . 0 ., CLARKE,B . E ., PARRY, N . R ., OULDRIDGE, E . J ., BROWN, F., and ROWLANDS,D . J . (1989b). Host cell selection of antigenic variants of foot-and-mouth disease virus . J. Gen . Viral. 70, 45-57 .

BOYER, M ., NOVAK, Z ., FOTEDAR, A., FRAGA, E ., and SINGH, B . (1990) .Critical role of an amino acid residue in a T cell determinant is dueto its interaction with a neighboring non-critical residue . Eur. J .Immunol. 20,2145-2148 .

BROWN, F . (1990) . Picornaviruses . /n "Immunochemistry of virusesII" (M . H . V. Van Regennortel and A . R . Neurath, Eds), pp. 153-169. Elsevier, Amsterdam .

CApucci, L ., BROCCHI, E ., DE SIMONE, F ., and PANINA, G . F . (1984) .Characterization of monoclonal antibodies produced against foot-and-mouth disease virus . In ''Report of a Session of the ResearchGroup of the Standing Technical Committee of the EuropeanCommission for the Control of Foot-and-Mouth Disease" pp . 32-39. Brescia : Food and Agriculture organization of the United Na-tions .

CARRILLO, C ., PLANA, J ., MASCARELLA, R ., BERGADA, J ., and SOBRINO, F .(1990) . Genetic and phenotypic variability during replication offoot-and-mouth disease virus in swine . Virology 179, 890-892 .

CHEUNG, A., DELAMARTER, J ., WEISS, S ., and KOPPER, H . (1983) . Com-parison of the major antigenic determinants of different serotypesof foot-and-mouth disease virus . l. Virol. 48, 451-459 .

CLARKE, B . E ., CARROLL, A . R., ROWLANDS, D . J ., NICHOLSON, B . H .,HOUGHTEN, R . A ., LERNER, R . A., and BROWN, F . (1983) . Syntheticpeptides mimic subtype specificity of foot-and-mouth diseasevirus. FEBS Lett. 157, 261--264 .

CLARKE, B . E., NEWTON, S . E., CARROLL, A. R., FRANCIS, M . J .,APPLEYARD, G ., SYRED, A . D ., HIGHFIELD, P . E ., ROWLANDS, D . J ., andBROWN, F . (1987) . Improved immunogenicity of a peptide epitopeafter fusion to hepatitis B core protein . Nature (London) 330, 381-384 .

COFFIN, J . M . (1986) . Genetic variation in AIDS viruses . Cell 46, 1-4 .Di MARCH . R., BROOKE, G ., GALE, C ., CRACKNELL, V ., DOEL, T ., andMowAT, N . (1986) . Protection of cattle against foot-and-mouth dis-ease by a synthetic peptide . Science 232, 639-641 .

DIEz, J ., MATEU, M . G., and DOMINGO, E . (1989) . Selection of anti-genic variants of foot-and-mouth disease virus in the absence ofantibodies, as revealed by an in situ assay . J. Gen Virol. 70, 3281-3289 .

DIEZ, J ., DAVILA, M ., E50ARMIS, C ., MATED, M . G ., DOMINGUEZ, 1 .,PEREZ, J . J ., GIRALT, E ., MELERO, J . A., and DOMINGO, E . (1990) .Unique amino acid substitutions in the capsid proteins of foot-and-mouth disease virus from a persistent infection in cell culture . I.Virol. 64, 5519-5528 .

DOMINGO, E., MATED, M . G ., MARTINEZ, M . A., DoPAzo, J ., MovA, A .,and SoBRINO, F . (1990) . Genetic variability and antigenic diversityof foot-and-mouth disease virus . In "Applied Virology Research,"Virus variability, epidemiology and control (E . Kurstak, R . G . Maru-syk, F . A . Murphy, and M . H . V . Van Regenmortel, Eds .), Vol . 2 . pp .233-266 . Plenum Press, New York .

GEBAUER, F ., DE LA TORRE, J . C ., COMES, I ., MATED, M . G., BARAHONA,H., TIRABOSCHI, B ., BERGMANN, I ., AUGE DE MELLO, P., and Do-MINGO, E . (1988) . Rapid selection of genetic and antigenic variantsof foot-and-mouth disease virus during persistence in cattle . J.Virol. 62, 2041-2049 .

GEYSEN, H . M ., BARTELING, S . J ., and MELOEN, R . H . (1985) . Smallpeptides induce antibodies with a sequence and structural require-ments for binding antigen comparable to antibodies raised againstthe native protein . Proc . Nat!. Acad. Sci. USA 82, 178-182 .

GRUBMAN, M . J ., and MORGAN, D . 0 . (1986) . Antigenic comparison offoot-and-mouth disease virus serotypes with monoclonal antibod-ies . Virus Res. 6, 33-43 .

HOUGHTEN, R . A ., HOFFMANN, S . R ., and NIMAN, H . L . (1986) . Relativeimportance of position and individual amino acid residues in pep-tide antigen-antibody interactions : Implications in the mechanismof antigenic drift and antigenic shift. In ''Vaccines'86" (F . Brown,R. M. Chanook, and R . A. Lerner, Eds.), pp. 21-25 . Cold SpringHarbor Laboratory, New York .

HYSLOP, N .STG ., DAVIE, J ., and CARTER, S . P . (1963) . Antigenic differ-ences between strains of foot-and-mouth disease virus of typeSAT1 .J. Hyg. Camb . 61, 217-230 .

JANSEN, R . W ., SIEGL, G ., and LEMON, S . (1990) . Molecular epidemiol-ogy of human hepatitis A virus defined by an antigen-capture poly-morass chain reaction method . Proc. Nail. Acad. Sci. USA. 87,2867-2871 .

KITSON, J . D . A., MCCAHON, D ., and BELSHAM, G . J . (1990) . Sequenceanalysis of monoclonal antibody resistant mutants of type 0 toot-and-mouth disease virus : evidence for the involvement of thethree surface exposed capsid proteins in four antigenic sites . Virol-ogy 179, 26 34 .

KURSTAK, E ., MARUSYK, R . G ., MURPHY, F . A., and VAN REGENMORTEL,M . H . V., Eds. (1990) . "Applied Virology Research ." Vol . 2 . Virusvariability, epidemiology and control . Plenum, New York .

LAEMMLI, U . K . (1970) . Cleavage of structural proteins during theassembly of the head of bacteriophage T4 . Nature (London) 227,680-685 .

MAKOFF, A. J ., PAYNTER, C . A., ROWLANDS, D . J ., and BOOTHROYD, J . C .(1982) . Comparison of the amino acid sequence of the major im-munogen from three serotypes of foot-and-mouth disease virus .Nucleic Acids Res. 10, 8285-8295 .

MARTINEZ, M . A ., CARRILLO, C ., PLANA, J ., MASCARELLA, R., BERGADA,1 ., PAL MA, E . L ., DOMINGO, E ., and SOBRINO, F . (1988) . Genetic andimmunogenic variations among closely related isolates of foot-and-mouth disease virus . Gene 62, 75-84 .

MATEU, M . G ., ROCHA, E ., VICENTE, 0 ., VAYREDA, F ., NAVALPOTRO, C .,ANDREU, A ., PEDROSO, F ., GIRALT, E ., ENJUANES, L ., and DOMINGO,E. (1987) . Reactivity with monoclonal antibodies of viruses from anepisode of foot-and-mouth disease . Virus Res . 8, 261-274 .

ANTIGENIC DIVERSIFICATION OF FMDV

705

MATEu, M . G ., DA SILVA, J . L ., ROCHA, E ., DE BRUM, D . L., ALONSO, A .,ENJUANES, L., DOMINGO, E ., and BARAHONA, H. (1988) . Extensiveantigenic heterogeneity of foot-and-mouth disease virus of sero-type C . Virology 167, 113-124 .

MATED, M . G ., MARTINEZ, M . A ., ROCHA, E ., ANDREU, D ., PAREJO, J .,GIRALT, E ., SOBRINO, F ., and DOMINGO, E . (1989) . Implications of aquasispecies genome structure : effect of frequent, naturally occur-ring amino acid substitutions on the antigenicity of foot-and-mouthdisease virus . Proc. Natl. Acad. Sci. USA 86, 5883-5887 .

MATED, M . G., MARTINEZ, M . A ., CAPuccl, L ., ANDRED, D ., GIRALT, E .,SOBRINO, F ., BROCCHI, E ., and DOMINGO, E . (1990) . A single aminoacid substitution affects multiple overlapping epitopes in the majorantigenic site of foot-and-mouth disease virus of serotype C . J.Gen. Virol. 71, 629-637 .

MELOEN, R . H ., LI5I AMP, R . M ., and GOUDSMIT, J . (1989) . Specificityand function of the individual amino acids of an important determi-nant of human immunodeficiency virus type 1 that induces neutral-izing activity . J. Gen. Virol . 70, 1505-1512 .

MODROW, S ., HAHN, B . H ., SHAW, G . M . . GALLO, R . C . . WONG-STAAL,F., and WOLF, H . (1987) . Computer-assisted analysis of envelopeprotein sequences of seven human immunodeficiency virus iso-lates: prediction of antigenic epitopes in conserved and variableregions . J. Virol. 61, 570-578 .

NEURATH, A . R ., and STRICK, N . (1990) . Confronting the hypervariabil-ity of an immunodominant epitope eliciting virus neutralizing anti-bodies from the envelope glycoprotein of the human immunodefi-ciency virus type 1 (HIV-1) . Mol. Immunol. 27, 539 -549 .

PALKER, T . I . . CLARK, M . E ., LANGLOIS, A . J ., MATTHEWS, T . J ., WEN-

HOLD, K . J ., RANDALL, R. R., BOLOGNESI, D ., and HAYNES, B . F .(1988) . Type-Specific neutralization of the human immunodefi-ciency virus with antibodies to env-encoded synthetic peptides .Proc. Natl. Acad. Sci. USA 85, 1932-1936 .

PARRY, N . R ., BARNETT, P . V ., OULDRIDGE, E . 1 ., ROWLANDS, D . J ., andBROWN, F . (1989) . Neutralizing epitopes of type 0 foot-and-mouthdisease virus . II . Mapping three conformational sites with syn-thetic peptide reagents . J. Gen . Virol. 70, 1493-1503 .

PARRY, N ., Fox, G ., ROWLANDS, D ., BROWN, F ., FRY, E ., AcHARYA, R .,LOGAN, D ., and STUART, D . (1990) . Structural and serological evi-dence for a novel mechanism of antigenic variation in foot-and-mouth disease virus . Nature (London) 347, 569--572 .

PEREIRA, H . G . (1977) . Subtyping of foot-and-mouth disease virus . !n''Developments in Biological Standardization ." (C . Mackowiakand R . H. Regamey, Eds .), Vol . 35, pp 167-174 . Karger, Basel .

PEREIRA, H . G . (1981). Foot-and-mouth disease . In "Virus Diseasesof Food Animals" (E . P . J . Gibbs, Ed .), Vol . 2 . pp 333-363 . Aca-demic Press, New York .

PFAFF, E ., MussGAY, M ., BOEHM, H . 0 ., SCHULZ, G . E ., and SCHALLER,H . (1982) . Antibodies against a preselected peptide recognize andneutralize foot-and-mouth disease virus . EMBOJ. 1, 869 874 .

PICCONE, M . E ., KAPLAN, G ., GIAVEDONI, L ., DOMINGO, E ., and PALMA,E . L. (1988) . VP1 of serotype C foot-and-mouth disease virus.Long-term conservation of sequences . J. Virol. 62, 1469-1473 .

ROBERTSON, B . H ., MORGAN, D . 0., and MOORE, D . M . (1984) . Loca-tion of neutralizing epitopes defined by monoclonal antibodiesgenerated against the outer capsid polypeptide, VP1, foot-and-mouth disease virus A, 2 . Virus Res . 1, 489-500 .

ROWLANDS, D . J ., CLARKE, B . E ., CARROLL, A . R ., BROWN, F ., NICHOL-SON, B . H ., BITTLE, J . L ., HOUGHTEN, R . A, and LERNER, R . A . (1983) .Chemical basis of antigenic variation in foot-and-mouth diseasevirus . Nature (London) 306, 694-697 .

RUSCHE, J . R ., JAVAHERIAN, K ., McDANAL, C ., PETRO, J . . LYNN, D . L.,GRIMAILA, R., LANGLOIS, A., GALLO, R . C., ARTHUR, L . 0 ., FISCH-INGER, P . J ., BOLOGNESI, D . P ., PUTNEY, S . D., and MATTHEWS, T . J .(1988) . Antibodies that inhibit fusion of human immunodeficiency

706

virus-infected cells bind to a 24-amino acid sequence of the viralenvelope, gpl20. Proc. Nat!. Aced. Sci. USA 85, 3198-3202 .

SIMMONDS, P., BALFE, P., LUDLAM, C . A., BISHOP, J . 0 ., and BROWN,A . J . L. (1990) . Analysis of sequence diversity in hypervariable re-gions of the external glycoprotein of human immunodeficiencyvirus type 1 . J. Viral. 64, 5840-5850 .

SOBRINO, F ., PALMA, E . L ., BECK, E ., DAVILA, M ., DE LA TORRE, J . C .,NEGRO, P ., VILLANUEVA, N ., ORTIN, J ., and DOMINGO, E . (1986) . Fixa-tion of mutations in the viral genome during an outbreak of foot-and-mouth disease : heterogeneity and rate variations . Gene 50,149-159 .

STROHMAIER, K ., FRANZE, R ., and ADAM, K . H . (1982) . Location andcharacterization of the antigenic portion of the FMDV immunizingprotein . J. Gen . Viral. 59, 295-306 .

THOMAS, A . A . M ., WOORTMEIJER, R . J ., PUIJK, W ., and BARTELING, S . J .(1988) . Antigenic sites on foot-and-mouth disease virus type A, 0 .J. Virol. 62, 2782-2789 .

TOWBIN, H ., STAEHELIN, T ., and GORDON, J . (1979). Electrophoretictransfer of proteins from polyacrylamide gels to nitrocellulosesheets : procedure and some applications . Proc. Natl. Aced. Sci .USA 76,4350-4354 .

VILLANUEVA, N., DAVILA, M ., ORTIN, J ., and DOMINGO, E . (1983) . Molec-

MARTINEZ ET AL .

ular cloning of cDNA from foot-and-mouth disease virus C,-SantaPau (C-S8) . Sequence of protein VP1-coding segment . Gene 23,185-194 .

WALDMANN, 0., and TRAUTWEIN, K . (1926) . Experimentelle unter-suchungen [)bar die pluralitat der maul-und-klauenseuchevirus .Berlin . Tierartzl. Wochschr. 42, 569-570.

WILD, T. F ., BURROUGHS, 1 . N., and BROWN, F . (1969) . Surface struc-ture of foot-and-mouth disease virus . L Gen, Viral. 4, 313-320 .

WIMMER, E ., and JAMESON, B . A . (1984) . Poliovirus antigenic sites andvaccines. Nature (London) 308, 19 .

WE, T. T ., and KABAr, E . A . (1970) . An analysis of the sequences ofthe variable regions of Bence-Jones proteins and myeloma lightchains and their implications for antibody complementarity . J.. Exp .Med. 132, 211-250 .

ZIMMERN, D., and KAESBERG, P . (1978) . 3'-Terminal nucleotide se-quence of encephalomycarditis virus RNA determined by reversetranscriptase and chain-terminating inhibitors . Proc. Natl. Acad.Sci. USA 75, 4257-4261 .

ZWART, G ., LANGEDIJK, H ., VAN DER HOEK, L ., DE JONG, J . J ., WOLFS,T. F. W., RAMAUTARSING, C ., BAKKER, M ., DE RONDE, A ., and GOUD-sMlr, J . (1991) . Immunodominance and antigenic variation of theprincipal neutralization domain of HIV-1 . Virology 181, 481-489 .