key: cord-0008505-8dsd5rag
authors: Wibberley, G.; Swallow, C.; Roberts, D.H.
title: Characterization of an influenza A (H3N2) virus isolated from pigs in England in 1987
date: 2007-11-19
journal: Br Vet J
DOI: 10.1016/0007-1935(88)90053-x
sha: 76325235ce5f04d0d8e9e973ce5ef5b4c9d09a84
doc_id: 8505
cord_uid: 8dsd5rag

This study describes the isolation and characterization of an influenza virus subtype H3N2 designated A/Swine/Weybridge/163266/87. The virus was isolated from a severe outbreak of respiratory disease in East Anglia. Haemagglutinin and neuraminidase characterization showed the virus to be very similar to H3N2 strains circulating in the human population during the years 1972–1975, and to H3N2 strains recently isolated from pigs in Belgium and France. A serological survey showed antibodies to the virus to be present in 31% of pigs tested, and reactors were detected on 43% of farms sampled.

Serological testing has shown H3N2 influenza A virus to be present in the pig population of several countries including Great Britian (Styk, Sabo & Blaskovic, 1971 ; Harkness et al., 1972 ; Kundin & Easterday, 1973 ; Chapman, Lamont & Harkness, 1978 ; Madec et al., 1984 ; Roberts, Cartwright & Wibberley, 1987) ; in some countries these findings have been supported by isolation of the virus (Kundin, 1970 ; Ottis et al, 1982 ; Madec et al, 1984 ; Haesebrouck et al, 1985) . In general it was felt that infection in pigs followed the appearance of infection in the human population (Harkness et al., 1972 ; Tumova, Stumpa & Mensik, 1980) , and with a few exceptions (Ottis et al., 1982) there was no evidence of clinical disease . In 1984 isolations of H3N2 influenza virus were made from pigs in France (Madec et al., 1984) and Belgium (Haesebrouck et al, 1985) showing signs of respiratory tract disease . In the French outbreak, reproductive failures such as abortions and returns to service were also described .

In 1987 a virus related to the human A/Port Chalmers/1/73 (H3N2) strain was isolated from pigs in England showing severe respiratory symptoms ; this isolate was designated A/Swine/Weybridge/163266/87 . This report describes the identification of the isolate, and the results of inoculating it into susceptible pigs .

Pritchard et al. (1987) described a severe outbreak of influenza in a pen of 22 pigs housed on a farm of 1600 animals . Infection spread rapidly to affect 60% of the total farm stock within 3 weeks ; seven pigs died . The pigs showed dullness, lethargy, pyrexia (up to 42 . 8°C) and a barking cough . Survivors recovered over a period of 1 week .

Samples of lung were taken from a slaughtered animal, as were nasal swabs from other affected animals, and examined for the presence of virus . Serum samples from survivors were examined for antibodies some 2 months later, using the haemagglutination inhibition (HI) test .

Ten per cent suspensions of lung were made by homogenizing in Eagle's MEM containing 5% lactalbumin hydrolysate and antibiotics ; swabs were placed in this same medium . Following clarifiction by centrifugation (2000 g), 9-day-old embryonated chicken eggs were inoculated either amniotically or allantoically with 0 . 2 ml of sample and were incubated at 33°C for 5 days . The harvested fluids were tested for haemagglutinin activity . The virus was identified by immunodiffusion and by HI tests with specific antisera using standard methods (US Department of Health, Education and Welfare, Immunology Series No . 6, 1975) .

Strains A/Port Chalmers/1/73 (H3N2), A/Bangkok/1/79 (H3N2), A/USSR/98/1/77 (HINI), A/Swine/Belgium/1/79 (HINI), A/Singapore/1/57 (H2N2) used in these studies were obtained from Dr G . C . Schild or Dr J . J . Skehel, National Institute for Medical Research, Mill Hill, London . Strain A/Swine/Weybridge/117316/86 (HIN1) was isolated from an outbreak of influenza in pigs in England in 1986 (Roberts et al., 1987 . The viruses were cultivated in the allantoic cavity of embryonated chicken eggs . Allantoic fluids were harvested after 5 days' incubation, clarified by centrifugation and held at -70°C . Antisera to the above influenza strains were prepared in chickens according to standard procedures (US Department of Health, Education and Welfare, Immunology Series No . 6, 1975).

These tests were performed in microtitre plates by standard methods (US Department of Health, Education and Welfare, Immunology Series No . 6, 1975) . To destroy nonspecific inhibitors, the sera were inactivated at 56°C for 30 min and treated with 25% kaolin overnight at 4°C . After absorption with a 10% suspension of chicken erythrocytes for 1 hour at 37°C, the sera were tested against four haemagglutinating units of virus using 1% chicken erythrocytes . Serum titres of 1/10 or greater were considered positive .

A 10-week-old pig was inoculated with 10 ml of allantoic fluid from embryonated fowl eggs infected with the isolate . This inoculum contained 10' egg infection doses (EID)/ml of virus and was injected by the intratracheal route whilst the pig was lightly sedated with Strensil (Janssen Animal Health, Oxford) . This animal was housed with an uninoculated littermate as an in-contact sentinel . Both were examined daily ; rectal temperatures and nasal swabs were also taken daily . The pigs were slaughtered after 4 weeks . 

A haemagglutinating agent was recovered from one piece of lung from the herd outbreak ; this was isolated using the amniotic cavity route, then subcultured using the allantoic cavity route . No agents were isolated from any other samples . The isolate, designated A/Swine/Weybridge/163266/87 (H3N2) was identified as an influenza A virus by immunodiffusion tests involving antisera to the appropriate matrix protein antigen .

Specific antisera were prepared in chickens and ferrets, then tested against a range of HINT, H2N2 and H3N2 influenza A viruses . The results are shown in Table I , and indicate a low level of reactivity between the isolate and the human H3N2 strain A/Port Chalmers/1/73 . Further studies by Dr J . Skehel, Mill Hill, are listed in Table II . These show a relationship to H3N2 strains present in the human population in 1972 (A/ England/42/72), 1973 (A/Port Chalmers/1/73) and 1975 (A/Victoria/3/75) . Although there is close similarity between the isolate and recent H3N2 strains found in European pigs (Table II) , they are distinguishable . There is also a clear difference between this virus and a swine H3N2 isolated in Northern Ireland in 1986 (Dr J . Skehel, personal communication) .

Neuraminidase-inhibition tests carried out by Dr Skehel showed a close relationship between the neuraminidase of A/Port Chalmers/1/73 and that of the isolate ; this test also shows the neuraminidase to be very similar to that of an isolate from Europe, SW/ OMS/3633/84 (Table III) .

There was no clinical reaction seen in either the pig inoculated with the isolate, or in the sentinel, other than slight lethargy . The pigs were slaughtered 4 weeks after inoculation and no macroscopic lesions were detected . Virus was recovered from nasal swabs from the inoculated pig 1 and 3 days after inoculation, and from the sentinel at 7 days (strain A/Swine/Weybridge/163266/87) . At 14 days after inoculation, the HI antibody titres were 1/80 and 1/40 for the sentinel and inoculated animals respectively ; these had dropped to 1/40 and 1/20 at slaughter 2 weeks later . Serum samples taken before inoculation had shown no HI antibodies at a dilution of 1/10 .

A preliminary survey of 834 adult pigs from 47 farms has shown that 255 pigs (31%) have antibody titres of 1/10 or higher, 140 (17%) have titres of 1/40 or higher . Twenty premises (43%) have pigs with titres of 1/40 or higher . 

Although the presence of H3N2 influenza virus in the British pig population had been indicated by serological surveys over several years, clinical disease had not been a problem, nor had any virus been isolated . In the autumn of 1985 and spring 1986 outbreaks of respiratory disease occurred in Yorkshire and Suffolk, which were associated with HIN1 influenza virus infection (Roberts et al., 1987) . An agent serologically identical to transmissible gastroenteritis (TGE) of pigs was also found at this time to be associated with respiratory disease (Brown & Cartwright, 1986 . Outbreaks of respiratory disease occurred that could not be attributed to either H1N1 of TGE-like coronavirus infections, and from one of these outbreaks a haemagglutinating agent was isolated . The agent was found to be very similar to H3N2 strains that circulated in the human population from 1972 to 1975, and closely related to viruses isolated from pigs in Belgium and France since 1984 . The clinical pictures in Belgium, France and England were very similar, with pyrexia, anorexia, lethargy, some respiratory distress and some mortality . Recovery by survivors was fairly rapid over a 7-day period .

The degree of severity of the disease apparently associated with H3N2 virus in the field was not reproduced experimentally . Similar lack of clinical response in pigs inoculated with H3N2 virus has been observed by others (Styk et al., 1971 ; Pospisil et al., 1973 ; Haesebrouck et al, 1985) . The current study showed that pigs had been infected, virus was recovered from the nasal cavity, and antibody response observed . Other factors such as environment and the presence of other pathogens play an important part in any outbreak of respiratory disease . It is interesting that the new TGE-like coronavirus was detected at a similar time to the appearance of an apparently virulent H3N2 influenza virus in both north-western Europe and in England (Brown & Cartwright, 1986 ; Pensaert, Callebaut & Vergote, 1986) .

The preliminary serological survey indicates a higher number of English pigs to have antibodies to H3N2 influenza virus in 1987 (31%) than did at the last survey in 1983 (17 . 3%) (Roberts et al., 1987) . It would appear that the virus is already widespread .

There was an outbreak of influenza-like illness in stockmen at the premises where the original isolation was made . Haemagglutination inhibition tests performed on serum samples gave negative results (Dr P . Chakraverty, 1987, personal communication) ; the illness was probably coincidental .

The isolate A/Swine/Weybridge/163266/87 is antigenically different from current H3N2 influenza strains in the human population . Commercial vaccines containing only current human strains as the H3N2 component may not be fully protective for pigs (Haesebrouck, Pensaert & Wyffels, 1987) . There are many agents at present involved with respiratory illness in pigs in Britian ; H3N2 influenza would appear to be one of these agents .

Bulletin of the World Health Organisation 47, 489

Zentralblatt für Veterinärmedizin B 20

Veterinary Record 121, 548

Zentralblatt für veterinärmedizin B27, 601

The authors would like to thank Mrs Ruth Manvell for carrying out the type specificity immunodiffusion tests, and Miss Michaela Williams for technical assistance .