Field and laboratory analysis of an outbreak of foot and mouth disease in Bulgaria in 1991

Transcription

1 Rev. sci. tech. Off. int. Epiz., 1993,12 (3), Field and laboratory analysis of an outbreak of foot and mouth disease in Bulgaria in 1991 A.R. SAMUEL, D.M. ANSELL, R.T. RENDLE, R.M. ARMSTRONG, F.L. DAVIDSON, N.J. KNOWLES and R.P. KITCHING * Summary: In July 1991, an outbreak of foot and mouth disease (FMD) occurred near Stefan Karadjovo village in Boliarovo (south-east Bulgaria, close to the Turkish border). The virus isolated was identified in Bulgaria as serotype O and this was subsequently confirmed by the World Reference Laboratory for Foot and Mouth Disease in Pirbright (United Kingdom). Serological studies using bovine sera and monoclonal antibody analysis were made. In addition, the sequence of approximately 170 nucleotides at the 3' end of the 1D gene was determined for the field isolate and for vaccine strains used in Bulgaria. These were compared with other sequences of type O FMD viruses from outbreaks in the Middle East. Serum samples were taken from domestic animals in the region close to the outbreak and examined for anti-fmd virus antibodies to assess the extent (if any) of spread of the virus before or after the outbreak. No evidence of infection was found in these animals. The virus involved in the Bulgarian outbreak was antigenically similar to the O1 vaccine strains but probably did not originate from these strains. The virus was closely related genetically to a group of viruses isolated in the Middle East since 1987, suggesting that it may have been introduced into Bulgaria from an area in the Middle East by unidentified means. KEYWORDS: Aphthovirus - Bulgaria - Foot and mouth disease - Monoclonal antibody - Nucleotide sequence analysis. INTRODUCTION On 29 July 1991, animal attendants in charge of a group of 99 cattle which had been grazing alongside 400 sheep, since March, in a forest 5 km east of the village of Stefan Karadjovo in Boliarovo municipality, Yambol district, south-east Bulgaria, reported that some of the animals were ill. The local and regional veterinary officers suspected foot and mouth disease (FMD) and although it was not possible to examine all the animals, due to inadequate handling facilities, eleven cattle with mild vesicular lesions were identified. Samples sent to the FMD laboratory in Sliven, Bulgaria, were identified by complement fixation as containing FMD virus (FMDV) serotype O. These results were confirmed by a laboratory in Sofia the following day and later at the Office International des Epizooties (OIE)/Food and Agriculture Organisation of the United * Agricultural and Food Research Council, Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Near Woking, Surrey, GU24 ONF, United Kingdom.

2 840 Nations (FAO) World Reference Laboratory for FMD (WRL) in Pirbright, United Kingdom. Prior to this outbreak, Bulgaria had been free of FMD since This paper presents field observations and results of serological and biochemical investigations which compared a virus isolate from the outbreak (O/BUL/1/91) with type O reference viruses and other field viruses. Comparisons between the Bulgarian isolate, current vaccine strains and other contemporary type O isolates from the Near and Middle East were of particular interest. Additionally, probang and serum samples were taken from domestic animals to assess the extent (if any) of spread of virus beyond the original site, which might have occurred before the outbreak was discovered or following discovery, due to an elevenday delay in slaughtering the infected herd. MATERIALS AND METHODS Viruses The virus isolates used in the study are listed in Table I. Most viruses were isolated on primary bovine thyroid (BTy) cells and subsequently adapted to IBRS (Instituto Biológico rim suino) 2 cells for further analysis. Enzyme-linked immunosorbent assay The viruses were compared by one-way liquid-phase blocking sandwich enzymelinked immunosorbent assay (ELISA) using sera from vaccinated cattle according to the method described by Kitching and colleagues (6). Briefly, the assay measures the residual antigen remaining after overnight reaction between dilutions of a reference antiserum and pre-titrated masses of antigen prepared from the field isolate and the homologous reference strain. The serum titre obtained at 50% of the predetermined virus concentration is used to calculate the relationship ("r") value. This method has been shown to correlate with results obtained using the conventional virus neutralisation (VN) test, "r" values were calculated as follows: r = serum titre against field isolate serum titre against vaccine strain The following criteria were used for interpretation of "r" values (12): represents a highly significant antigenic variation from the reference vaccine strain. Where possible, this situation is best countered by use of a vaccine strain with a closer relationship to the field virus. However, in an emergency, a potent vaccine more distantly related to the field virus may provide sufficient protection, especially if administered on more than one occasion represents an area of concern. These values show a significant difference from the reference strain, but protection may be satisfactory if using a potent vaccine represents values not significantly different from the reference vaccine strain as measured by the particular test system used.

3 841 TABLE I Foot and mouth disease virus type O strains used in the investigation Country and year of origin WRL Ref. No. Date collected Comments 0,/Lausanne/Switzerland /Lausanne I 165 vaccine strain 0,/Lausanne/Switzerland ,/Laus/BVS Bulgarian vaccine strain 0,/BFS 1860/UK1967 0,/BFS /11/67 vaccine strain 0,/Sharquia/Egypt ,/Sharquia vaccine strain 0,/Manisa/Turkey ,/Manisa 01/04/69 vaccine strain O/Israel 1985 O/ISR/2/85 /05/85 vaccine strain 0/194/USSR1958 0/194/USSR / /58 Russian vaccine strain O/Bahrain 1988 O/BAR/9/88 / /88 O/Bahrain 1991 O/BAR/2/91 11/03/91 O/Bulgaria 1991 O/BUL/1/91 /07/91 O/Iran 1987 O/IRN/3/87 08/12/87 O/Israel 1988 O/ISR/1/88 24/06/88 O/Israel 1989 O/ISR/6/89 22/12/89 O/Israel 1991 O/ISR/1/91 / /91 O/Jordan 1988 O/JOR/1/88 16/06/88 O/Jordan 1989 O/JOR/1/89 / /89 O/Kuwait 1988 O/KUW/3/88 28/04/88 O/North Yemen 1987 O/NYE/10/87 23/07/87 0/Oman 1991 O/OMN/58/91 01/06/91 O/Saudi Arabia 1988 O/SAU/8/88 28/09/88 O/Saudi Arabia 1988 O/SAU/33/88 15/12/88 O/Saudi Arabia 1989 O/SAU/3/89 / /89 O/Saudi Arabia 1990 O/SAU/26/90 26/11/90 O/Saudi Arabia 1990 O/SAU/35/90 05/12/90 O/Saudi Arabia 1991 O/SAU/3/91 19/02/91 O/Saudi Arabia 1991 O/SAU/7/91 05/01/91 O/Syria 1987 O/SYR/1/87 01/03/87 O/Syria 1989 O/SYR/1/89 20/02/89 O/Syria 1991 O/SYR/1/91 10/05/91 O/Turkey 1987 O/TUR/2/87 18/05/87 O/Turkey 1988 O/TUR/2/88 06/01/88 O/Turkey 1988 O/TUR/8/88 11/05/88 O/Turkey 1989 O/TUR/6/89 20/03/89 O/Turkey 1990 O/TUR/5/90 08/05/90 O/Turkey 1990 O/TUR/21/90 24/09/90 O/Turkey 1991 O/TUR/1/91 14/01/91 O/Turkey 1991 O/TUR/2/91 17/01/91 O/Turkey 1991 O/TUR/11/91 29/04/91 O/Turkey 1991 O/TUR/13/91 28/06/91 WRL: World Reference Laboratory for Foot and Mouth Disease, Pirbright, United Kingdom

4 842 Monoclonal antibody analysis A panel of monoclonal antibodies (MAbs) raised against the reference strain 01/Lausanne/Switzerland/1965 (1, 2, 9, 10) and characterised by sequencing neutralisation escape mutant viruses raised against individual MAbs (7, 15) was used in a trapping ELISA (13) to obtain antigenic profiles. Briefly, viruses to be tested were reacted with a panel of MAbs and the reactivity compared to that of the homologous virus. Results were expressed as a percentage of the homologous reaction. Nucleotide sequence analysis The sequence of approximately 170 nucleotides at the 3' end of the 1D (virus protein 1: VP1) gene was determined for the O/194/USSR vaccine strain, O1/Lausanne (Bulgarian vaccine strain) and the O/BUL/1/91 field isolate by directly sequencing ribonucleic acid (RNA) obtained from semi-purified virus harvests using the primer extension sequence method described by Sanger and colleagues (14), with modifications (8). Serum and probang survey Samples were collected between 16 October and 6 November Serum samples were collected from non-vaccinated pigs in Stefan Karadjovo and Goliamo Keushevo, the two villages closest to the site of the outbreak. Samples were collected from nonvaccinated sheep in twenty-three locations in surrounding areas and extending between the outbreak site and the Turkish border. A total of 530 serum samples were collected and examined at the WRL to determine levels of antibody against FMD virus type 01, using a liquid-phase blocking sandwich ELISA (4, 5). Three serum samples were obtained from wildlife in the area (one from a deer, two from wild pigs). Probang samples (116 in total) were collected from vaccinated sheep in Stefan Karadjovo and from vaccinated cattle in villages closest to the outbreak site. Duplicate serum and probang samples were made available to the Bulgarian Veterinary Authorities. Primary BTy tissue culture tubes inoculated with probang samples were examined after 24 h and 48 h. Cultures showing a cytopathic effect (CPE) were then screened for FMDV by sandwich ELISA (11). Serum samples were titrated against the 01/Manisa vaccine strain. RESULTS AND DISCUSSION The aim of these studies was to provide data from which it might be possible to determine the most likely origin of the virus involved. A number of possibilities were considered, including: i) vaccine virus ii) aerosol from an FMD laboratory iii) the outbreak in cattle was secondary to an unidentified primary outbreak iv) indirect transmission via humans or animals from other countries.

5 The Bulgarian outbreak strain, the vaccine strain produced at the plant in Sliven and the Russian vaccine strain O/194/USSR used to perform ring vaccination were reacted in a liquid-phase blocking sandwich ELISA with bovine post-vaccination sera prepared using the European vaccine strains 0,/BFS 1860 and O1/Lausanne, and the Middle East vaccine strain 01/Manisa. Results are expressed as "r" values (Table II), and interpreted as described above (in "Materials and Methods"). The results show that the O1 vaccine strains 01/Lausanne/BVS and O/194/USSR were antigenically similar to the outbreak strain. The results also suggest that a potent vaccine formulated from the Middle East 843 TABLE II Comparison of results of serological analysis by enzyme-linked immunosorbent assay of a foot and mouth disease virus type O isolate (O/Bulgaria/I/91) and the vaccine virus strains in use Viruses O/BFS 1860 Bovine post-vaccinal sera O1/Manisa O1/Lausanne O/BUL/1/91 0.8* ,/Lausanne/BVS 1.0 >1.0 >1.0 0/194/USSR 1.0 > * "r" values vaccine strain O1/Manisa could be expected to provide adequate protection. Table III shows the reactivity of the Bulgarian outbreak strain and vaccine strains with a panel of anti O1/Lausanne MAbs. The antigenic profile obtained showed that the field isolate O/BUL/1/91 had a profile which was similar to the 01/Lausanne vaccine strain except for the low reactivity with MAb C6. This epitope has been mapped onto the 1B (VP2) capsid protein (7, 15) and reactivity of MAb C6 has been shown to vary when the homologous virus has been subjected to passage in different cell cultures (3). The vaccine strain O/194/USSR differed at the C6 and G5 sites. No FMD virus was detected in any of the probang samples. The pig sera and wildlife TABLE III Results of monoclonal antibody (MAb) analysis of a foot and mouth disease virus type O isolate (O/Bulgaria/1/91) and the vaccine virus strains in use O1/Lausanne/Switzerland/65 MAbs Viruses B2 C6 C8 C9 D9 A8 G5 O/BUL/1/91 118* 5** /Lausanne/BVS /194/USSR ** * mean percentage reactivity as compared to the homologous * significantly different at P = :

6 844 sera were negative for anti-fmdv type O antibodies. Some of the sheep sera (approximately 1%) had low positive titres. These were mainly from individual animals under one year of age and were probably due to persisting maternal antibodies inherited from vaccinated dams. All of these positive sera were negative for antibodies against virus infection-associated (VIA) antigen (J. Ivanov, D. Mackay, personal communications, 1992). In conclusion, the serum and probang survey did not reveal any evidence of active infection in the animals tested. A dendrogram was constructed to show the genetic relationship between selected field and reference/vaccine strains of FMDV (Fig. 1). Comparison of the nucleotide sequence of the Bulgarian isolate (O/BUL/1/91) with that of other recent type O strains Percentage nucleotide difference (positions of VP1) FiG.l Dendrogram depicting the genetic relationship between foot and mouth disease type O virus isolates from the Middle East and reference/vaccine strains The dendrogram is interpreted by reading the percentage nucleotide difference on the scale at the point where the furthest vertical line joining the two viruses being compared occurs

7 from the Middle East and the vaccine strains, showed that the Bulgarian outbreak strain was not closely related to either the O1/Lausanne vaccine strain used in Bulgaria or the Russian type O vaccine strain (0/194/USSR). However, it was closely related to a group of FMDV type O strains which have been found in parts of the Middle East. This group consists of isolates from Turkey in 1987 to 1991, Israel in 1991, Syria in 1987 and a single isolate from an outbreak in Saudi Arabia in Other type O isolates of FMDV from Saudi Arabia which have been sequenced belong to another distinct group (A.R. Samuel and colleagues, unpublished findings). Although the genomic region sequenced in the Bulgarian strain cannot be distinguished from O/TUR/2/87 and is very closely related to the O/TUR/13/91 and O/SAU/35/90 isolates, it cannot be assumed that these viruses are the direct ancestors of the Bulgarian strain. In the case of the O/TUR/2/87 isolate, slightly fewer nucleotides were compared with the O/BUL/1/91 isolate. This was due to two factors: a) a smaller number of nucleotides were determined b) a number of ambiguities were present in the O/TUR/2/87 sequence. However, it can be concluded from these results that the Bulgarian outbreak did not originate from the vaccines which were either made or stored in Bulgaria, nor from other vaccine strains used in the Middle East (i.e. O1/Manisa) but is related to other viruses recently isolated in the Middle East. A possible scenario for the source of the Bulgarian outbreak is that the cattle were infected via sheep, pigs or deer (either domestic or wild). However, infection from domestic pigs is unlikely, the nearest herds being situated between 5 km and 7 km from the outbreak (climatic conditions were not conducive to virus survival and aerosol spread). Further outbreaks would have been expected and there was no traceable contact between the cattle attendants and pigs. Also, no anti-fmdv type O antibodies were detected in the samples from domestic pigs which were tested. The epidemiology and results of the serum survey would thus rule out the involvement of pigs. Samples from domestic sheep and goats in villages close to the outbreak were collected a few days after they had been vaccinated (it was expected that they would not yet have developed a serological response to vaccination). These samples were found to be negative for antibody against VIA antigen, indicating that they were not the source of infection. Deer and wild pigs could be considered a possible source of infection. Movement of deer during the mating season can be considerable and the possible area of infection could be very large, even extending across the border into Turkey. Wild pigs would be likely to infect a more localised area, although at various times of the year they invade maize fields to feed. However, large-scale collection of samples from wild animals was not possible, due to the economic importance which hunting plays in the local economy. The three samples from wild animals which were tested for antibody all gave negative results. Regular serological surveys in Bulgaria, Greece and Turkey (Thrace) over the next two to three years may help to identify any potential reservoirs of infection in the wild animal population. Alternatively, the infection may have been introduced from the Middle East by a route which has yet to be determined, e.g. indirect transmission from people or from movement of animals. The threat of this type of outbreak is damaging to the economy of the country concerned. This is due to the ban on imports enforced by other countries until the threat has subsided. The possibility of the recurrence of this type of incident remains a continuing threat, particularly now that vaccination against FMDV in Europe has ceased. 845

8 846 ACKNOWLEDGEMENTS The authors wish to thank Dr E. Brocchi of the Istituto Zooprofilattico in Brescia, Italy, for the use of the anti-0,/lausanne/switzerland/65 monoclonal antibodies, and also the countries which submitted the samples to the WRL which were used in this investigation, particularly the Bulgarian Veterinary Authorities. This work was partly funded by the United Kingdom Ministry of Agriculture, Fisheries and Food. * * * ANALYSE SUR LE TERRAIN ET EN LABORATOIRE D'UNE ÉPIDÉMIE DE FIÈVRE APHTEUSE SURVENUE EN BULGARIE, A.R. Samuel, D.M.Ansell, R.T. Rendle, R.M. Armstrong, F.L. Davidson, N.J. Knowles et R.P. Kitching. Résumé : En juillet 1991, une épidémie de fièvre aphteuse est survenue près du village de Stefan Karadjovo dans le Boliarovo (sud-est de la Bulgarie, près de la frontière turque). Le virus en cause était de sérotype O, d'après l'identification faite en Bulgarie puis confirmée par le Laboratoire mondial de référence pour la fièvre aphteuse de Pirbright (Royaume-Uni). Des études utilisant des sérums de bovins et une analyse par anticorps monoclonaux ont été réalisées. Après détermination de la séquence d'environ 170 nucléotides en position terminale 3' du gène 1D, pour l'isolat de terrain comme pour les souches vaccinales employées en Bulgarie, on a procédé à leur comparaison avec d'autres séquences de virus de la fièvre aphteuse de type O, provenant de foyers du Moyen-Orient. Des prélèvements de sérum, effectués sur des animaux domestiques dans une région voisine du foyer de la maladie, ont fait l'objet de recherches d'anticorps contre le virus de la fièvre aphteuse pour évaluer l'ampleur éventuelle de la propagation de ce virus, avant et après le déclenchement de la maladie. Aucun signe d'infection n'a été décelé chez ces animaux. Le virus à l'origine du foyer de Bulgarie présentait des caractères antigéniques similaires à ceux des souches vaccinales 01 mais il n'en était probablement pas issu. Le virus était, en revanche, étroitement lié, d'un point de vue génétique, à un groupe de virus isolés au Moyen-Orient en On peut donc supposer qu'il provenait de cette région et qu'il a été introduit en Bulgarie par une voie inconnue. MOTS-CLÉS : Analyse d'une séquence de nucleotides - Anticorps monoclonaux - Bulgarie - Fièvre aphteuse - Virus de la fièvre aphteuse. * *

9 ANÁLISIS EN EL CAMPO Y EN LABORATORIO DE UN BROTE DE FIEBRE AFTOSA ACAECIDO EN BULGARIA EN A.R. Samuel, D.M. Ansell, R.T. Rendle, R.M. Armstrong, EL. Davidson, N.J. Knowles y R.P. Kitching. 847 Resumen: En julio de 1991 se declaró un brote de fiebre aftosa cerca del pueblo de Stefan Karadjovo, situado en Boliarovo, sudeste de Bulgaria, cerca de la frontera con Turquía. El virus se aisló e identificó en Bulgaria como de serotipo O, lo que después fue confirmado en el Laboratorio de referencia mundial para la fiebre aftosa de Pirbright, Reino Unido. Se realizaron estudios usando sueros de bovinos y un análisis por anticuerpos monoclonales. Por otra parte, se determinó la secuencia de unos 170 nucleótidos en posición terminal 3 ' del gene 1D, tanto para el aislado de campo como para las cepas vacunales usadas en Bulgaria, y se los comparó con otras secuencias del virus de la fiebre aftosa tipo O provenientes de brotes existentes en el Medio Oriente. Se investigó asimismo la presencia de anticuerpos contra el virus de la fiebre aftosa en muestras de suero recogidas de animales domésticos de una región cercana a la del brote de la enfermedad para evaluar la amplitud de propagación posible del virus antes y después de la aparición del brote, sin hallarse en estas muestras ningún signo de la enfermedad. El virus causante del brote presentaba características antigénicas similares a las de las cepas vacunales O,, pero probablemente no procedía de ellas. En la medida en que estaba estrechamente relacionado, desde un punto de vista genético, con un grupo de virus aislados en el Medio Oriente desde 1987, se puede suponer que provenía de esa región y que fue introducido en Bulgaria por vía aún desconocida. PALABRAS CLAVE: Análisis de secuencia de nucleótidos - Anticuerpos monoclonales - Bulgaria - Fiebre aftosa - Virus de la fiebre aftosa. * * * REFERENCES 1. BROCCHI E., CIVARDI A., SIMONE F. DE & PANINA G.F. (1983). - Characterisation of foot and mouth disease virus antibodies. In Proc. 20th Congress of the Italian Society of Microbiology. Gardone, Italy. Atti Soc. ital. Sci. vet., 36, CAPUCCI L., BROCCHI E., SIMONE F. DE & PANINA CF. (1984). - Characterisation of monoclonal antibodies produced against foot and mouth disease viruses. In Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot and Mouth Disease, Brescia, Italy. Appendix 6, CROWTHER J.R. & SAMUEL A.R. (1987). - Monoclonal antibodies and foot and mouth disease. In Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot and Mouth Disease, Lyons, France. FAO, Rome. Appendix 13, HAMBLIN C, BARNETT I.T.R. & HEDGER R.S. (1986). - A new enzyme-linked immunosorbent assay for the detection of antibodies against foot and mouth disease virus. I. Development and method of ELISA. J. immunol. Meth., 93,

10 HAMBLIN C, BARNETT I.T.R. & CROWTHER J.R. (1986). - A new enzyme-linked immunosorbent assay for the detection of antibodies against foot and mouth disease virus. II. Application.J. Immunol. Meth., 93, KITCHING R.R, RENDLE R. & FERRIS N.R (1988). - Rapid correlation between field isolates and vaccine strains of foot and mouth disease virus. Vaccine, 6, KITSON J.D.A., MCCAHON D. & BELSHAM G.J. (1990). - Sequence analysis of monoclonal antibody resistant mutants of type O foot and mouth disease virus: evidence for the involvement of the three surface exposed capsid proteins in four antigenic sites. Virology, 179, KNOWLES N.J. (1990). - A method for direct nucleotide sequencing of foot and mouth disease virus RNA for epidemiological studies. In Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot and Mouth Disease, Lindholm, Denmark, June. FAO, Rome. Appendix 19: MCCULLOUGH K.C., CROWTHER J.R., BUTCHER R.N., CARPENTER W.C., BROCCHI E., CAPUCCI L. & SIMONE F. DE (1986). - Immune protection against foot and mouth disease virus studied using virus neutralising and non-neutralising concentrations of monoclonal antibodies. Immunology, 58, MCCULLOUGH K.C., CROWTHER J.R., CARPENTER W.C., BROCCHI E., CAPUCCI L., SIMONE F. DE, XIE Q. & MCCAHON D. (1987). - Epitopes on foot and mouth disease virus particles. I. Topology. Virology, 157, ROEDER P.L. & LE BLANC SMITH P.M. (1987). - Detection and typing of foot and mouth disease virus by enzyme-linked immunosorbent assay: a sensitive, rapid and reliable technique for primary diagnosis. Res. vet. Sci, 43, SAMUEL A.R., OULDRIDGE E.J., ARROWSMITH A.E.M., KITCHING R.P. & KNOWLES N.J. (1990). - Antigenic analysis of serotype O foot and mouth disease virus isolates from the Middle East, Vaccine, 8, SAMUEL A.R., KNOWLES N.J., SAMUEL G.D. & CROWTHER J.R. (1991). - Evaluation of a trapping ELISA for the differentiation of foot and mouth disease virus using monoclonal antibodies. Biologicals, 19, SANGER F., NICKLEN S. & COULSON A.R. (1977). - DNA sequencing with chainterminating inhibitors. Proc. natl Acad. Sci., USA, 74, XIE Q.C, MCCAHON D., CROWTHER J.R., BELSHAM G.J. & MCCULLOUGH K.C. (1987). - Neutralisation of foot and mouth disease virus can be mediated through any of at least three separate antigenic sites. J. gen. Virol, 68,