Molecular epidemiology of recent outbreaks of swine vesicular disease: two genetically and antigenically distinct variants in Europe, 1987-94

Epidemiol. Infect. (1997), 118, 51–61 Copyright # 1997 Cambridge University Press

Molecular epidemiology of recent outbreaks of swine

vesicular disease: two genetically and antigenically distinct

variants in Europe, 1987–94

E. BROCCHI

, G. ZHANG

*, N. J. KNOWLES

, G. WILSDEN

J. W. MCAULEY

†, O. MARQUARDT

, V. F. OHLINGER

*  F. DE SIMONE

Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey, GU24 0NF,

United Kingdom

Istituto Zooproﬁlattico Sperimentale della Lombardia e dell’Emilia ‘‘B. Ubertini’’, Via A. Bianchi 7, 25125

Brescia, Italy

Division of Molecular Biology, Institute for Animal Health, Compton Laboratory, Compton, Newbury,

Berkshire, RG20 7NN, United Kingdom

Bundesforschungsanstalt fu

r Viruskrankheiten der Tiere, P.O. Box 1149, D 72001, Tu

bingen, Germany

(Accepted 22 August 1996)

SUMMARY

Viruses from the recent epidemic of swine vesicular disease (SVD) in Europe have been isolated

and characterized by antigenic and genetic methods to examine the likely epidemiological

origins of the disease. Antigenic analysis was performed on 77 SVD viruses (SVDV) isolated in

Europe between 1966 and 1994 using two panels of monoclonal antibodies (MAb) in a

trapping ELISA. Genetic analysis of 33 of the SVD viruses by reverse transcription-polymerase

chain-reaction (RT-PCR) ampliﬁcation and nucleotide sequencing of the 1D (VP1) coding

region was also performed. Comparison of the nucleotide sequences with each other and with

three other previously published SVDV sequences revealed four distinct groups which

correlated exactly with the results of the pattern of reactivity with MAbs. The ﬁrst group

consisted solely of the earliest SVD virus isolated (ITL}1}66) while the second group

comprised viruses present in Europe and Japan between 1972 and 1981. The third group

consisted of viruses isolated from outbreaks of SVD in Italy between December 1988 and June

1992. Viruses isolated between 1987 and 1994 from Romania, the Netherlands, Italy and Spain

formed a fourth group. The genetic and antigenic similarity of the most recent virus isolates

from Western Europe to a virus isolated in Romania 5 years previously suggests that the

possible origin of the recent epidemic of swine vesicular disease in Western Europe was in

Eastern Europe.

INTRODUCTION

Swine vesicular disease (SVD) is a highly contagious

disease of pigs which is caused by a virus belonging to

* Present addresses: G.Z., Institute for Animal Health, Compton,

Newbury, Berks, RG20 7NN, United Kingdom ; V. F.O., State

Investigation Centre for Chemistry, Food and Animal Health, von

Esmarch Str. 12, D-48149 Munster, Germany.

† Author for correspondence: Dr J. W. McCauley, Institute for

Animal Health, Compton, Newbury, Berks, RG20 7NN, United

Kingdom.

the genus Enterovirus within the family Picornaviridae.

Swine vesicular disease virus (SVDV) is antigenically

closely related to the human pathogen coxsackievirus

B5 (CB5) [1, 2]. Recently the complete genome

sequences of three SVD viruses and a CB5 virus have

been described [3–6] and the relationship between

SVDV and CB5 virus was conﬁrmed but considerable

divergence of nucleotide sequence was seen through-

out the virus genome.

Swine vesicular disease was ﬁrst recorded in

52 E. Brocchi and others

October 1966 when a vesicular disease appeared

simultaneously on two farms in Lombardy, Italy

which had received pigs for fattening from a common

origin. The outbreak was diagnosed as foot-and-

mouth disease (FMD) on clinical grounds but the

failure to conﬁrm this diagnosis by laboratory tests

led to a study that showed the causative agent to be an

enterovirus [7]. No further cases of the disease

occurred in 1966.

Swine vesicular disease was next encountered in

Hong Kong in 1971 [8] and retrospective studies

revealed the presence of the virus as early as May 1970

(N. J. Knowles, unpublished observations). Out-

breaks in Hong Kong have continued almost annually

until 1991 [9].

In October and December 1972 a number of

outbreaks of SVD occurred in Italy: in Naples, Rome,

Latina, Matera and Ravenna between September and

December [10]. During December 1972 SVD was also

conﬁrmed in Austria [11], the United Kingdom [12],

France [13] and Poland [11]. In 1973 the ﬁrst outbreaks

were recorded in West Germany [14], Switzerland

[15], Belgium [10] and Japan [16]. Between 1975 and

1983 outbreaks of SVD appeared sporadically in

France (1975, 1982–83), the Netherlands (1975) [17],

Malta (1975) [18], Japan (1975) [19], Belgium (1979)

[21]. In Germany and UK occasional outbreaks

occurred almost annually until 1982 despite the

imposition of strict hygiene and control measures. In

Italy SVDV appeared to persist and small numbers of

outbreaks occurred nearly every year until 1992.

In early July 1992, SVD was reported in breeding

stock on two neighbouring farms in Ede, Netherlands

and more than 1000 pigs were destroyed [22]. Later in

the month SVD was detected on a third farm in the

same area. In October 1992, the disease was found on

three additional farms in the Netherlands, two of

which were close to those involved in the July

outbreaks. The sixth farm to be aﬀected was located

in Putten and the virus was thought to have been

introduced by a vehicle which had been associated

with two of the outbreaks at Ede. Nearly 2000 pigs

were destroyed in the latter three outbreaks. No new

outbreaks occurred until January and February 1994

when SVD reappeared in the Netherlands at a pig

collecting centre at Haarle, Overijssel.

In June 1992 ﬁve outbreaks of SVD occurred in

Italy. In August 1992 two outbreaks were recorded in

the Modena and Siena provinces : both involving

animals belonged to the same consignment of pigs

imported from the Netherlands. It was apparent (vide

infra) that these were antigenically distinct from

previously encountered viruses.

In October 1992, during a SVD serological survey

of pigs imported into Belgium from the Netherlands,

sera containing antibodies to SVDV were found in

fattening pigs originating from a farm in West

Flanders Province, Belgium. In early February 1993

SVDV antibodies were detected in pigs on a farm in

Antwerp Province, Belgium. Again the infection was

thought to have been introduced by the purchase of

piglets from the Netherlands. In both cases no clinical

disease was seen and no virus could be isolated.

An epidemiological survey of SVD in Spain was

undertaken and began in November 1992. In all, 94

farms in 39 diﬀerent towns had imported pigs from

the Netherlands and one farm, in Lerida Province,

had pigs with antibodies to SVDV.

In February 1993, an outbreak of a vesicular

condition was reported on a farm in the Lerida

Province, Spain, and the entire herd was slaughtered.

In April 1993, SVD was diagnosed at two other farms

in Huesca Province in north-eastern Spain and the

pigs were slaughtered (a total of 3278 head). No

outbreaks of SVD had occurred in Spain since April

1993 and more than 10000 pigs have since been tested

for SVDV antibodies [23, 24].

It had been suspected that SVD outbreaks occurred

during the 1970s in some Eastern European countries.

The ﬁrst outbreaks recorded in Austria in 1972 were

associated with the importation of pigs from Poland

where the virus was also isolated [11]. We have learned

of a number of outbreaks of SVD in the former

Eastern Bloc following improved communication with

these countries and virus from one outbreak in

Romania in 1987 has been analysed in this study.

These events are summarized in Table 1 which

shows the number of SVD outbreaks which have

occurred in Europe since 1987.

De Simone and colleagues [25], using two panels of

monoclonal antibodies, showed that it was possible to

adduce that SVDV fell into four antigenic groups :

group 1 consisted of the virus from the ﬁrst outbreak

(ITL}1}66); group 2 consisted of strains which

circulated in Europe in the early 1970s; group 3 was

composed of Italian strains from 1988 until December

1992, and group 4 was composed of viruses which had

been present in Europe since July 1992. This paper

reports the detailed antigenic and genetic analysis of

viruses isolated between 1987 and 1994 from epi-

demics of SVD in Europe and a comparison with

earlier strains. Four congruent groupings were found

53Molecular epidemiology of SVD

Table 1. Swine vesicular disease outbreaks in Europe,

1987–94*

Year

Country 1987 1988 1989 19901991 1992 1993 1994

Romania & 1†

Italy 1 1 6 31 11 30

Netherlands 6 3

Belgium 1‡ 1‡

Spain 3

* Data was derived from OIE Disease Information reports,

except:

† M. Danes, personal communication, 1994.

‡ Conﬁrmed by serological testing, no virus isolated.

in both the antigenic and genetic properties of the

virus.

MATERIALS AND METHODS

Viruses

The SVD virus isolates studied are listed in Table 2.

Monoclonal antibody analysis (IZSLE, Brescia)

Two panels of MAbs prepared against SVDV strains

R178 (isolated in Italy in 1973) and R1046 (ITL}2}91)

(Brocchi and colleagues, unpublished observations)

were used in a trapping ELISA as previously described

[26]. Essentially each MAb was reacted with virus

grown in IB-RS-2 cells which had been trapped using

a pre-titrated polyclonal rabbit SVDV antiserum. The

results were expressed as a percentage of the hom-

ologous reaction for each MAb after ﬁrst

standardizing against a polyclonal guinea-pig SVDV

antibody. Multivariate analysis was performed using

the Unistat4 Statistical Package version 1±2 (Unistat

Ltd., London, UK). This consisted of a hierarchical

cluster analysis using the average between groups.

Distance was measured up by the squared Euclid

method where:

Distance (x, y) ¯

®y

)

Nucleotide sequence determination (IAH, Pirbright)

Viruses examined in this study (ITL}1}66, ITL}2}73,

ITL}A}89, ITL}1-2}91, ITL}1}92, ITL}7-10}93,

ITL}12}93, ITL}1-8}94, NET}1}75, NET}1-3}92,

NET}1-4}94, ROM}1}87 and SPA}1}93; Table 2)

were grown in tissue culture ﬂasks containing approxi-

mately 5¬10

(

IB-RS-2 cells. Total RNA was

extracted from infected cells and ﬁrst strand cDNA

synthesis was carried out as described previously [6]

using the oligonucleotide GSVD-1 (Table 3) in a ﬁnal

reaction volume of 20 µl. The product was extracted

with phenol, precipitated with ethanol and

resuspended in 15 µl of water. A 5 µl sample was

ampliﬁed by PCR with Taq polymerase (Boehringer-

Mannheim, Germany), 200 n of the oligonucleotides

(GSVD-3 and NK44, Table 3), 200 µ deoxy-

nucleotides and buﬀer as supplied by the manu-

facturer, in a reaction volume of 50 µl. Cycling

conditions were: 95 °C, 4 min ; 50 °C, 90 s; 72 °C,

90 s; followed by 25 cycles of 94 °C, 60 s ; 50 °C, 60 s ;

and 72 °C, 60 s. An aliquot of the product was

analysed by electrophoresis through an agarose gel.

The remainder was puriﬁed by adsorption to and

elution from a silica matrix (Magic PCR preps4,

Promega, WI). Sequencing was done essentially as

described in the fmol4 sequencing kit (Promega, WI)

using approximately 100 mg of template DNA and

oligonucleotides labelled at the 5« terminus with

P-

γATP (Table 3).

Nucleotide sequence determination (BFAV, Tu

bingen)

Each virus examined (D6494}73, D6533}75,

D6629}81, R1004, R1071 (ITL}2}92), NL 92; Table

2) was passaged once on porcine secondary embryonic

kidney (PSEK) cells and then used to infect a 25 cm

ﬂask of cells. Total RNA was extracted when CPE

was observed, as described previously [27, 28]. Fol-

lowing hybridization of 20 µg of infected cell RNA

with 50 ng of primer 3-R (Table 3), ﬁrst strand cDNA

synthesis was carried out with 10U of avian myelo-

blastoma virus reverse transcriptase (Stratagene, CA).

Ampliﬁcation was carried out with primers 1-F and 1-

R (Table 3). Taq polymerase (Promega, WI), in a Bio-

Med Thermocycler 60 (Bachofer, Germany). Thirty

cycles of 1 min at 93 °C, 2 min at 55 °C and 5 min at

72 °C were performed. The PCR ampliﬁcation

products were analysed by agarose gel electrophoresis.

The products were then treated with Klenow enzyme

and T4 polynucleotide kinase, puriﬁed with Geneclean

II (Bio-101, La Jolla, CA) and concentrated ﬁvefold.

Sequencing was performed using a commercially

available sequencing kit (USB, CA). PCR ampli-

ﬁcation products exhibiting signiﬁcant numbers of

54 E. Brocchi and others

Table 2. Designation and origin of the swine vesicular disease viruses studied by MAb analysis and nucleotide

sequencing

Virus designation

Date Genetic MAb

WRL ref. no.* Other number† Geographic origin collected group group

West Germany

None D6494}73 Krefeld, Northrhine-Westphalia September 73 II nd

None D6533}75 Bersenbru

ck, Lower-Saxony August 75 II nd

None D6629}81 Freudenstadt, Baden-Wu

rttemberg September 81 II nd

Italy

ITL}1}66 Italy}66 Lombardy October 66 I 1

ITL}2}73 None Rome, Lazio December 72 II nd

None R178 Padova, Veneto 06}04}73 nd 2

ITL}A}89 R967 Campodoro, Padova, Veneto 23}12}88 III 3a

None R1004 Portici, Naples, Campania 26}05}89 III 3b

ITL}1}91 R1042 Messina, Sicily 30}01}91 III 3c

ITL}2}91 R1046 Agerola, Naples, Campania 15}04}91 III 3c

None R1053 Alfonsine, Ravenna, Emilia Romagna 09}06}92 nd 3c

None R1056 Ischie, Naples, Campania 16}06}92 nd 3c

None R1057 Riolo Terme, Ravenna, Emilia Romagna 17}06}92 nd 3c

ITL}1}92 R1058 Campogalliano, Modena, Emilia Romagna 27}06}92 III 3c

None R1061 Castevetro, Modena, Emilia Romagna 13}08}92 nd 4c

None R1062 Montepulciano, Siena, Tuscany 14}08}92 nd 4c

None R1063 Montepulciano, Siena, Tuscany 14}08}92 nd 4c

None R1064 Spilamberto, Modena, Emilia Romagna 14}08}92 nd 4c

None R1065 Spilamberto, Modena, Emilia Romagna 14}08}92 nd 4c

None R1068 Chiamicelle, Siena, Tuscany 24}08}92 nd 4c

None R1069 Poggio D., Siena, Tuscany 24}08}92 nd 4c

None R1070 Castiglione del Lago, Perugia, Umbria 28}08}92 nd 4c

ITL}2}92 R1071 Fano, Pesaro, Marche 02}09}92 IV 4c

None R1072 Ripat., Chieti, Abruzzo 11}09}92 nd 4c

None R1073 Castiglione del Lago, Perugia, Umbria 17.09}92 nd 4c

None R1076 Pozzilli, Isernia, Molise 20}09}92 nd 4c

None R1079 S. Giovanni Incarico, Frosinone, Lazio 10}10}92 nd 3e

None R1080 Centallo, Cuneo, Piemonte 01}10}92 nd 4c

None R1083 Erchie, Brindisi, Puglia 22}10}92 nd 3d

None R1086 Luzzara, Reggio Emilia, Emilia Romagna 05}11}92 nd 4c

None R1087 Cusano Mutri, Benevento, Campania 16}11}92 nd 4c

None R1092 Serradifalco, Caltanissetta, Sicily 07}12}92 nd 3d

None R1093 Condofuri, Reggio di Calabria, Calabria 07}12}92 nd 4c

None R1094 Luzzara, Reggio Emilia, Emilia Romagna 11}12}92 nd 4c

None 284}93 Nola Naples, Campania 01}02}93 nd 4c

ITL}7}93 534}93 Nola, Naples, Campania 22}02}93 IV nd

ITL}8}93 R1098 Potenza, Basilicata 04}03}93 IV 4c

None R1105 Tursi, Matera, Basilicata 12}03}93 nd 4c

ITL}9}93 R1120 Bertinoro, Forli, Emilia Romagna 18}03}93 IV 4c

None R1150 Matera, Basilicata March 93 nd 4c

ITL}10}93 R1159 Barano d’Ischia, Naples, Campania 10}04}93 IV 4c

ITL}12}93 None Ariccia, Rome, Lazio May 93 IV 4c

ITL}13}93 R1164 Ariccia, Rome, Lazio 01}06}93 nd 4c

None R1168 Cosenza, Calabria 23}06}93 nd 4c

None R1172 Davoli, Cosenza, Calabria 18}11}93 nd 4c

None R1173 Mescia R., Foggia, Puglia 28}12}93 nd 4c

ITL}1}94 R1174 S. Felice Rubiera, Reggio Emilia, Emilia 24}01}94 IV 4c

Romagna

None R1175 Albano L, Potenza, Basilicata 02}02}94 nd 4c

None R1176 Potenza, Basilicata 02}02}94 nd 4c

None R1177 Campomaggiore, Potenza, Basilicata 02}02}94 nd 4c

55Molecular epidemiology of SVD

Table 2—(cont.)

Virus designation

Date Genetic MAb

WRL ref. no.* Other number† Geographic origin collected group group

None R1178 Grottole, Matera, Basilicata 10}02}94 nd 4c

ITL}2}94 R1179 S. Giorgio Lucano, Matera, Basilicata 10}02}94 IV 4c

None R1180 Castiglione del Lago, Perugia, Umbria 14}02}94 nd 4c

None R1181 Pisticci, Matera, Basilicata 17}02}94 nd 4c

None R1182 Pianopoli, Cosenza, Calabria 16}02}94 nd 4c

None R1183 Nemoli, Potenza, Basilicata 26}02}94 nd 4c

None R1184 Lattarico, Cosenza, Calabria 05}03}94 nd 4c

ITL}3}94 R1185 Bettona, Perugia, Umbria 05}03}94 IV 4c

None R1186 S. Giorgio Lucano, Matera, Basilicata 08}03}94 nd 4c

ITL}4}94 R1187 Soncino, Cremona, Lombardy 13}04}94 IV 4c

None R1188 Torre Pallavicina, Bergamo, Lombardy 13}04}94 nd 4c

ITL}5}94 R1189 Salsomaggiore, Parma, Emilia Romagna 29}04}94 IV 4c

None R1190 Fidenza, Palma, Emilia Romagna 05}05}94 nd 4c

None R1191 Tricarico, Matera, Basilicata 05}05}94 nd 4c

ITL}6}94 R1193 Mirandola, Modena, Emilia Romagna 11}05}94 IV 4c

ITL}7}94 R1194 Amantea, Cosenza, Calabria 15}05}94 IV 4c

ITL}8}94 R1195 Macomer, Nuore, Sardinia 18}05}94 IV 4c

None R1196 Bardi, Palma, Emilia Romagna 24}05}94 nd 4c

None R1197 Poggio Rusco, Mantova, Lombardy 25}05}94 nd 4c

None R1198 Pavullo, Modena, Emilia Romagna 03}06}94 nd 4c

None R1200 Agro di Bella, Potenza, Basilicata 29}09}94 nd 4c

None R1201 Cavriana, Mantova, Lombardy 22}11}94 nd 4d

None R1202 Guidizzolo, Mantova, Lombardy 07}12}94 nd 4d

None R1203 Guidizzolo, Mantova, Lombardy 07}12}94 nd 4d

None R1204 Bozzolo, Mantova, Lombardy 08}12}94 nd 4d

Japan

None J1«73‡ Ibaraki November 73 II nd

None H3«76‡ Hokkiado February 76 II nd

Netherlands

NET}1}75 Uden, North Brabant Province 17}11}75 II 2

NET}1}92 1992-1 Ede, Gelderland 03}07}92 IV 4b

NET}2}92 1992-2 Ede, Gelderland 03}07}92 IV 4c

NET}3}92 1992-3 Ede, Gelderland 04}08}92 IV 4c

None NL 92 Ede, Gelderland 1992 IV 4c

NET}1}94 1994-1 Haarle, Overijssel 28}01}94 IV 4c

NET}2}94 1994-2 Haarle, Overijssel February 94 IV 4c

NET}3}94 1994-3 March 94 IV 4c

NET}4}94§ Rosmalen February 94 IV nd

Romania

ROM}1}73 1973 II 2

ROM}1}87 1987 IV 4a

Spain

SPA}1}93 Vallfogona de Balaguer, Lerida February 93 IV 4c

United Kingdom

UKG}27}72‡ Staﬀordshire 11}12}72 II 2

* World Reference Laboratory for Foot-and-Mouth Disease reference number.

† Reference number used in the country of origin.

‡ Sequences previously published [3–5].

All viruses were isolated in their country of origin except § isolated in WRL, Pirbright, UK.

56 E. Brocchi and others

Table 3. Designation, sequence and location of the oligonucleotide primers used for PCR ampliﬁcation and

sequencing of the SVDV isolates studied

Oligonucleotide Position on the

name Sequence (5« ! 3«) SVDV genome* Used for

IAH, Pirbright

GSVD-1 GTCTGCTGGGGGTGTTGAT 3D 5979–5960 RT of RNA

GSVD-3 ACACCCTTTATAAAACAGG 1C 2414–2432 PCRsequencing

NK44 CCACACACAGTTTTGCCAGTC 2A 3394–3374 PCR-sequencing

NK45 GCCAACGTACACGGCACC 2A 3334–3317 Sequencing

GSVD-5 AACATGCTGTATGCGTTGCCTAT 1D 3027–3005 Sequencing

GSVD-6 GTCAAACCTGGCCCACCCGTCATA 1D 3052–3029 Sequencing

BFAV, Tu

bingen

3-R TCAAATGTGACTGGATAGTGCTT 2A 3525–3503 RT of RNA

1-F CGACAACTTCGCCTACTGGGT 1D 2704–2724 PCR-sequencing

1-R CTTCCCACACACAGTTTTGCCAGTC 2A 3398–3374 PCRsequencing

* Numbering according to the sequence of SVDV}UKG}27}72 [4].

ambiguous bases were cloned in E. coli using the

pSPT18 vector as previously described [29].

Phylogenetic analysis

Nucleotide sequences were analysed on an IBM

compatible personal computer using programmes

written by one of the authors (NJK). All pairwise

comparisons were performed by giving each base

substitution equal statistical weight (ambiguities were

ignored). A binary tree was constructed according to

sequence relatedness across the interval of nucleotides

315 to 504 of the 1D (VP1) gene (genomic bases 2762

to 2951) using the NEIGHBOR-JOINING method as

implemented in the computer program NEIGHBOR

and a dendrogram plotted using the program

DRAWGRAM both from the PHYLIP version 3.5c

phylogeny package [30].

RESULTS

Antigenic analysis

Antigenic analysis was conducted on 81 SVD viruses

from Europe. The analysis was performed using a

trapping ELISA with 2 panels, each of 10 monoclonal

antibodies, raised against 2 viruses from Italy (R178

and R1046 [ITL}2}91]). The panels contained

neutralizing and non-neutralizing antibodies. A sum-

mary proﬁle of reactivity with the panels of antibodies

is shown in Figure 1. Four of the non-neutralizing

antibodies (3F5, 2B1, 5B8 and 4E8) show variable

reactions with tissue culture supernatants of some

viruses which were dependent on the integrity of the

virus particle. As previously indicated [25], four

distinct patterns of antigenicity could be demon-

strated. It was apparent that a new virus population

had appeared in Italy in August 1992 which had a

similar proﬁle of reaction with monoclonal antibodies

to the proﬁle displayed by viruses from the

Netherlands earlier in June of the same year.

Hierarchical cluster analysis (average between groups)

conﬁrmed that there were four antigenic groups and

these are shown in Figure 2. The clusters consisted of

one group with only member (ITL}1}66), the second

group consisted of viruses from the early 1970s, the

third group contained viruses from Italy isolated

between 1988–91, and the fourth group consisted of

viruses isolated from Western Europe between July

1992 and the end of 1994 and a Romanian virus

isolated in 1987.

Nucleotide sequence analysis by RT-PCR and direct

sequencing

The nucleotide sequences of the SVDV isolates have

been submitted to the EMBL database at the

European Bioinformatics Institute, Hinxton, UK

(World Wide Web address: http:}}www.ebi.ac.uk}).

Complete 1D gene sequences were not obtained for all

SVDV isolates which was probably due to the

speciﬁcity of the internal sequencing primers. Ad-

ditionally the nucleotide identity at some positions

was unreadable, either due to premature polymerase

termination or to the presence of more than one equal

intensity band, possibly indicating a polymorphism.

Notwithstanding these reservations, phylogenetic

analysis of the nucleotide sequences of those viruses

57Molecular epidemiology of SVD

Group

Neutralization

25 25

0>50 0>50

25 25

000

100

100 25

0>50

100

50 50

100 100

5A10

++++

3B85D94H9

100 100

100

4H3

1H1

5B7

3F5 2B1 5B8 2G7 1G1 6E1 2D2 3G8 2H11 4C5 1A2 2H8 4E8

100100100

100

100100100

100100

100100100100

100

100 100 100 100

Monoclonal antibidies raised against SVDV Italy/73 Monoclonal antibidies raised against SVDV Italy/71

100

0>25

100

0000

100

25 25>50

0 ITL/1/66

R967 (Italy 1988)

R1004 (Italy 1989)

R1092 (Italy 1992)

ROM/1/87

NET/1/92100*

100*

R1079, R1083 (Italy 1992)

Europe 1992-94 (59 isolates; see Table 2)

R1201, R1202, R1203, R1204 (Italy 1994)

ITL/1/91, ITL/2/91, ITL/1/92, R1053, R1056, R1057

R178 (Italy 1973), UK/27/72, NET/1/75, ROM/1/73

Viruses

––––– –– – – –

+/–

100*

100

100*

100

100100100

0 50

Fig. 1. Reactivity patterns of SVD viruses with two panels of anti-SVDV monoclonal antibodies. Reactivity of virus with antibody was measured by a capture ELISA

as described by Samuel and colleagues [26]. Virus reactivity was grouped according to the percentage reactivity of antibody with the sample virus compared with the

reactivity to the virus against which antibody was raised. An asterisk following 100% indicates that 100 % reactivity was seen only following treatment of virus at 56 °C

for 30 min. Values indicated 0 " 25 and 0 " 50 indicate that variable reactivities of some viruses within a sub-group between 0 and either 25% or 50% of the homologous

reactivity were observed. The virus designation is shown in Table 2.

for which complete (or nearly complete) 1D sequences

were obtained was undertaken and the results are

shown in Figure 3a. Additionally a dendrogram was

constructed for a shorter region of the 1D genes

(nucleotides 315–504 ; genome sequence 2761–2950)

for which a larger number of sequences were available

(Fig. 3 b). Although this tree suggests that viruses

belonging to group II could be further subdivided,

this is not supported by the analysis of either the

longer sequences (Fig. 3a) or another study using a

larger data set (G. Zhang, D. T. Haydon, N. J.

Knowles, J. W. McCauley, unpublished obser-

vations). The trees were not corrected for multiple

substitutions since their eﬀects are minimal at this

level of sequence diversity. In both trees there are four

distinct branches (clades): one which consists of the

ﬁrst virus isolated in Italy in 1966 (group I), one of

viruses from 1972–82 (group II), one of viruses from

1988–92 (group III) and one of viruses from Western

Europe between 1992 to 1994 and the single virus

from Romania isolated in 1987 (group IV). For some

other viruses only partial sequences of the 1D coding

region were obtained, however, these all fell into one

of the four genetic groups (see Table 2). The results

show that the viruses from Western Europe between

1988 and 1994 formed two quite separate groupings

with a wide sequence divergence and it is considered

very unlikely that a virus from group III was the

progenitor of viruses of group IV.

Nucleotide sequence analysis following cloning

In some cases nucleotide sequences were determined

by preparing cloned PCR products. Several plasmids

containing inserts were obtained and sequenced. Two

virus isolates were sequenced in this manner

(D6629}81 and R1071 [ITL}2}92]) and some het-

erogeneity between clones of the same virus was

detected. The sequence of D6629}81 consisted of two

sequences (varying by 2±5%) which might be due to

the error-prone polymerase activities of reverse trans-

criptase and the DNA polymerase from Thermus

aquaticus used to produce the clones or alternatively

because two closely related viruses were present in an

outbreak. However, the analysis of R1071 (ITL}2}92)

showed a wider diversity with two distinct SVDV

sequences: one was encountered in two clones and

was similar to the group III sequences and the other

was found seven times and was related to the group IV

sequences. The sequences were approximately 15 %

diﬀerent and would be very unlikely to be due to the

58 E. Brocchi and others

11109876543210

Distance ¬0·0001

R1064

R1065

R1061

R1068

R1069

NET/3/92

R1063

R1093

R1062

R1120 (ITL/9/93)

284/93 (Italy)

NET/2/92

R1150

ITL/12/93

R1159 (ITL/10/93)

R1076

R1080

R1105

R1071 (ITL/2/92)

R1073

R1087

R1094

R1070

R1086

R1168

R1164 (ITL/13/93)

SPA/1/93

ROM/1/87

ITL/1/66

ROM/1/73

NET/1/75

R178 (Italy 1973)

R967 (ITL/A/89)

R1004

R1057

R1056

R1053

R1046 (ITL/2/91)

R1042 (ITL/1/91)

R1092

R1058 (ITL/1/92)

R1083

R1079

Group 3e

Group 3d

Group 3b

Group 3a

Group 2

Group 1

Group 4a

Group 4c

Fig. 2. Dendrogram of the antigenic relationships between the SVD viruses examined using a MAb ELISA. The dendrogram

was prepared as described in the materials and methods. Virus designation is shown in Table 2.

use of error-prone polymerases. It is diﬃcult to

exclude the possibility of laboratory contamination

but it is also feasible that the pig was infected with two

diﬀerent viruses or that the epithelial tissue from

which the virus was isolated had become contami-

nated at the abattoir during collection. Monoclonal

antibody proﬁles of uncloned virus however revealed

only the presence of virus from antigenic group 4. In

the case of D6629}81, in which one sequence

determined from a cDNA clone was identical to that

of a virus isolated in Germany in 1975 (D6533}75), we

cannot rule out laboratory contamination. Due to the

reservations we hold about these two viruses, they

have not been included in the dendrograms shown in

Figure 3 but their omission does not signiﬁcantly

inﬂuence the above conclusions.

DISCUSSION

From 1972–94 SVDV outbreaks were reported in

Europe, and viruses isolated from many of these

outbreaks have been subjected to analysis with

monoclonal antibodies and by sequencing a region of

the virus RNA genome. Two sequences of SVDV

from Japan were also included in the genetic com-

parison.

A comparison of the SVDV nucleotide sequences in

the region of the genome that encodes the virus capsid

59Molecular epidemiology of SVD

Group I

Group II

Group III

Group IV

ITL/1/66

UKG/27/72

NET/1/75

J1’73 (Japan)

H3’76 (Japan)

ITL/2/73 (1972)

ITL/1/92

ITL/2/91

ITL/1/91

ITL/A/89 (1988)

SPA/1/93

ROM/1/87

ITL/10/93

ITL/9/93

ITL/8/93

1 %

(a)

Group I

Group II

Group III

Group IV

ITL/1/66

UKG/27/72

NET/1/75

H3’76 (Japan)

D6629/81 (Germany)

ITL/2/73 (1972)

ITL/1/92

ITL/2/91

ITL/1/91

ITL/A/89 (1988)

SPA/1/93

ROM/1/87

ITL/10/93

NET/3/92

ITL/8/93

1 %

(b)

ITL/2/92

ITL/12/93

NET/1/94

NET/2/94

NET/3/94

ITL/7/93

ITL/9/93

SPA/1/93

R1004 (Italy 1989)

D6533/75 (Germany)

D6494/73 (Germany)

J1’73 (Japan)

Fig. 3. Phylogenetic tree derived from VP1 RNA sequences from SVDV. The designation of the viruses is shown in Table

2. The tree reconstructions were done using PHYLIP version 3.5 [30] as described in the materials and methods using the

program NEIGHBOR. ITL}1}66 was used as an outgroup to root the trees. (a) NEIGHBOR-JOINING tree of viruses for

which the complete VP1 sequence was determined. (b) NEIGHBOR-JOINING tree of SVD viruses reconstructed from VP1

bases 315–504 (genome sequence 2761–2950). Where WRL reference numbers do not match the year of isolation, the year

is included in parentheses.

60 E. Brocchi and others

polypeptide 1D was made and a dendrogram of the

relationships was constructed (Fig. 3). Similar dendro-

grams were made from the data from monoclonal

antibody analysis (Fig. 2). The analysis of the

European and Japanese SVD viruses showed that

they fell into four distinct groups. The earliest SVD

virus isolated, ITL}1}66, was distinct from all the

other viruses examined, although there was less

genetic distance to the viruses from the 1970s than to

those isolated later. SVD viruses which predominated

in Europe during the 1970s, when infection was

widespread, were eliminated from most countries by

rigorous control measures. During the late 1980s and

1990s two groups emerged: one present in Italy from

1988 to June 1992 (group III) and another in

Netherlands, Italy and Spain from July 1992 to 1993

(group IV). It seems unlikely that viruses typical of

group III (1988–92) were the ﬁrst ancestors of those

belonging to the later group (IV) because of their wide

sequence divergence. The monoclonal antibody den-

drogram has four groups, albeit group 3 can be easily

divided into ﬁve subgroups and group 4 into four

subgroups (Figs. 1 and 2). The ability to subdivide

these two groups may be a reﬂection of the selection

of the strain of virus against which the two panels of

MAbs used were raised. Although the relationships

between monoclonal antibody proﬁles can be used to

group antigenic variants, it is less evident that they

can be used to interpret population genetic and

evolutionary trends. Nevertheless, our observation

that virtually congruent virus groups resulted from

using the two methods reinforces our conclusions that

two populations of SVDV, with distinct origins, have

recently been circulating in Europe.

Neither the virus that seeded the outbreaks nor the

geographical location of the initial focus of infection

can be ﬁrmly established. Epidemiological data

suggest that the SVD viruses in group IV ﬁrst

appeared in Western Europe in the Netherlands.

During 1993 seven consignments of pigs received for

slaughter in Italy were found to be positive for SVD

virus; all originated in the Netherlands [31]. However,

it seems likely that the origin of viruses of group IV

was Eastern Europe. A strain of SVDV belonging to

this group was present in Romania in 1987. It seems

unlikely that SVDV had remained undetected in

Western Europe because recent serological surveys in

all EU countries have only detected the presence of

SVDV antibodies in Italy [32]. SVDV has been present

in Eastern Europe : it was known that outbreaks of

SVD occurred in Poland during 1972 and 1973 [11]

and it has also been recently reported that outbreaks

occurred in USSR in 1973 and 1975 (V. V. Drygin,

personal communication, 1994), Bulgaria in 1971 (Y.

Ivanov, personal communication, 1994) and Romania

in 1973 (M. Danes, personal communication, 1994).

Whether the virus lineage of genetic group IV evolved

from SVDV previously introduced into Eastern

European or is a recent introduction into that area

from either Hong Kong or China has not been

resolved.

In summary and conclusion: genetic and antigenic

data has shown that two lineages of SVDV have been

present in Western Europe since 1988. The nucleotide

sequence diversity of the earlier one (group III) makes

it unlikely that it gave rise to the later one. We submit

that the origin for this later lineage (group IV) may be

Eastern Europe and that it probably evolved subse-

quently in Italy.

ACKNOWLEDGEMENTS

We would like to thank Dr R. Ahl, BFAV, Tu

bingen,

Germany, Dr C. Terpstra, ID-DLO, Lelystad,

Netherlands, Dr Alberto San Gabriel, Laboratorio

De Sanidad Y Produccion Animal Ministerio De

Agricultural Pesca Y Alimentacion, Barcelona, Spain

and Dr Mihai Danes, National Institute for Vet-

erinary Medicine, Bucharest, Romania for supplying

SVDV isolates from their respective countries.

REFERENCES

1. Graves JH. Serological relationship of swine vesicular

disease virus and coxsackie B5 virus. Nature 1973 ; 245 :

314–5.

2. Brown F, Talbot P, Burrows R. Antigenic diﬀerences

between isolates of swine vesicular disease virus and

their relationship to coxsackie B5 virus. Nature 1973 ;

245: 315–6.

3. Inoue T, Suzuki T, Sekiguchi K. The complete

nucleotide sequence of swine vesicular disease virus. J

Gen Virol 1989; 70: 919–34.

4. Seechurn P, Knowles NJ, McCauley JW. The complete

nucleotide sequence of a pathogenic swine vesicular

disease virus. Virus Res 1990; 16: 255–74.

5. Inoue T, Yamaguchi S, Kanno T, Sugita S, Saeki T.

The complete nucleotide sequence of a pathogenic

swine vesicular disease virus isolated in Japan (J173)

and phylogenetic analysis. Nucleic Acids Res 1993 ; 21 :

3896.

6. Zhang G, Wilsden G, Knowles NJ, McCauley JW.

Complete nucleotide sequence of a coxsackie B5 virus

and its relationship to swine vesicular disease virus. J

Gen Virol 1993; 74: 845–53.

61Molecular epidemiology of SVD

7. Nardelli L, Lodetti E, Gualandi GL, et al. A foot-and-

mouth disease syndrome in pigs caused by an entero-

virus. Nature 1968; 219: 1275–6.

8. Mowat GN, Darbyshire JH, Huntley JF. Diﬀeren-

tiation of a vesicular disease of pigs in Hong Kong from

foot-and-mouth disease. Vet Rec 1972; 90: 618–21.

9. Knowles NJ, Sellers RF. Swine vesicular disease. In :

Beran GW, ed. Handbook of zoonoses, Section B :

Viral. 2nd edn. Boca Raton : CRC Press Inc., 1994 :

437–44.

10. OIE: Report of the Director. Swine vesicular disease.

Bull Oﬀ Int Epiz 1973; 79: 879–84.

11. Kubin G. Appearance of swine vesicular disease in

Austria. Wiener tiera

rztl Mschr 1973; 60: 283–8.

12. Dawe PS, Forman AJ, Smale CJ. A preliminary

investigation of the swine vesicular disease epidemic in

Britain. Nature 1973; 241: 540–2.

13. Dhennin L, Dhennin L. La maladie ve

siculeuse du porc

son apparition en France. Bull Acad Vet Fr 1973; 46 :

47–51.

14. Vittoz R. Swine vesicular disease. Reveta Vet Venez

1973; 34: 289–90.

15. Pohlenz J, Williams DM, Keller H. Die

Vesikula

rkrankheit des Schweines bei ihrem Auftreten

in der Schweiz. Schweizer Archiv Tierheilk 1974 ; 116 :

413–22.

16. Ohashi Y. The ﬁrst case of swine vesicular disease in

Japan. Bull Oﬀ Int Epiz 1974; 81: 839–43.

17. Franssen PGJ. Een geval van vesiculaire varkensziekte

(swine vesicular disease – S.V.D.) in Nederland. A case

of swine vesicular disease (SVD) in the Netherlands.

Tijdschr Diergeneeskunde 1975; 100: 1325–7.

18. Ford GW. The outbreaks of foot-and-mouth disease

and swine vesicular disease in Malta. World Anim Rev

1976; 20: 42–4.

19. Kodama M. Outline of studies on swine vesicular

disease in Japan. Bull Oﬀ Int Epiz 1976; 86: 423–32.

20. Vittoz R. Swine vesicular disease. Regional evolution.

Report of the Director-General on the Scientiﬁc and

Technical activities of the Oﬃce International des

Epizooties between May 1979 and May 1980. 48th Gen.

Sess. OIE Comm., Paris, May 1980. 1980: 89–92.

21. Dimitriades J, Pappous C, Brovas D. A case of swine

vesicular disease in Greece. Bull Hellen Vet Med Soc

1979; 30: 265–76.

22. Terpstra C. Swine vesicular disease in the Netherlands.

Tijdschr Diergeneeskunde 1992; 117: 623–6.

23. Espuna E, Alemany R, Riera P, et al. Isolation of the

swine vesicular disease virus in Spain. Medna Vet,

Valladolid 1993; 10: 657–62.

24. Kaman S. Swine vesicular disease (SVD) in Spain.

USDA Foreign Animal Disease Report 1993; 21: 12.

25. De Simone F, Brocchi E, Bugnetti M, Fusi P, Pacciarini

ML. Impiego di reagenti e metodi innovativi nella

diagnosi di malattie vescicolari. 1° Conferenza

Nazionale su: Stato dell’arte delle ricerche Italiane nel

Settore delle biotecnologie applicate alle scienze

veterinarie e zootecniche, Italy 1993: 27–33.

26. Samuel AR, Knowles NJ, Samuel GD, Crowther JR.

Evaluation of a trapping ELISA for the diﬀerentiation

of foot-and-mouth disease virus strains using mono-

clonal antibodies. Biologicals 1991; 19: 299–310.

27. Marquardt O, Adam K-H. Foot-and-mouth disease

virus subtyping by sequencing VP1 genes. Vet Microbiol

1990; 23: 175–83.

28. Marquardt O, Ohlinger VF. Diﬀerential diagnosis and

genetic analysis of the antigenically related swine

vesicular disease virus and coxsackie viruses. J Virol

Meth 1995; 53: 189–99.

29. Leister D, Adam K-H, Marquardt O. Co-replication of

several isotypes of foot-and-mouth disease virus. J Gen

Virol 1993; 74: 2753–7.

30. Felsenstein J. PHYLIP (Phylogeny Inference Package).

3.5c ed. Seattle: Department of Genetics. University of

Washington, 1993.

31. Terpstra C, Dekker A, Reek FH, Chenard G. Swine

vesicular disease: threat or challenge for the Dutch pig

industry? Lelystad, The Netherlands : Annual Report

of the Central Veterinary Institute (CDI-DLO), 1993 :

15–18.

32. Mackay DKJ, Armstrong RM, Kilner CG. Serological

survey for swine vesicular disease in the UK. Vet Rec

1995; 136: 248–9.