Changes in the virulence of myxoma virus strains in Britain.

113 Epidem. Inf. (1987) 98, 113-117 Printed in Great Britain Changes in the virulence of myxoma virus strains in Britain BY J. ROSS AND M. F. SANDERS M.A.F.F., Worplesdon Laboratory, Tangley Place, Worplesdon, Guildford, Surrey (Accepted 30 July 1986) SUMMARY National surveys of the virulence of field strains of myxoma virus were carried out in 1975 (128 virus strains) and in 1981 (123 strains), using the virulence testing method employed in a similar survey in 1962. Results showed that the virulence of field strains had increased between 1962 and 1975, and again between 1975 and 1981. The increases in virulence are thought to be a result of the development of resistance to myxomatosis in wild rabbit populations. The effects of the changes in virulence and resistance are discussed. INTRODUCTION Myxomatosis first appeared in Great Britain in late 1953 and, in the major epizootic that followed, it is estimated to have reduced the rabbit population by 99 % (Lloyd, 1970). Since then, despite the continuous presence of the disease, rabbit numbers have increased gradually although they are nationally still much lower than the pre-myxomatosis levels. The disease continues to be an important factor regulating rabbit numbers (Ross & Tittensor, 1981). The natural evolution of myxomatosis has been the subject of considerable research in both Australia and Great Britain. In particular, changes in the virulence of field strains of myxoma virus and the development of inheritable resistance to the disease have been monitored. In Great Britain inheritable resistance was first detected in rabbits from one site in England in the early 1970s (Ross & Sanders, 1977) and subsequently from several other sites, including one in Scotland (Ross & Sanders, 1984). It is reasonable to assume that a degree of resistance is common throughout Britain. Changes in the virulence of field strains of virus occurred quickly in Britain, as in Australia, with attenuated strains being detected within two years of the appearance of the disease (Hudson & Mansi, 1955). In 1962 a national survey of field strains was conducted and it was shown that fully virulent strains, similar to that which caused the first outbreaks, had almost disappeared. Virus strains of moderate virulence, killing 70-95 % of susceptible rabbits were found most commonly (Fenner & Chapple, 1965). It was shown that the predominance of such strains was due to more effective transmission by mosquitoes in Australia (Fenner, Day & Woodroofe, 1956) and by the rabbit flea in Britain (Mead-Briggs & Vaughan, 1975). Fenner & Ratcliffe (1965) postulated that, in resistant rabbits, strains of higher virulence would be more effectively transmitted and they predicted that the development of resistance 114 J. ROSS AND M. F. SANDERS in wild rabbits would lead to an increase in the virulence of field strains of virus. There is some evidence that such a change has occurred in Australia where virus strains collected in 1970-1974 from an area of Victoria with very resistant rabbits were of higher virulence than strains from other areas in Victoria where the rabbits were less resistant (Edmonds et al. 1975). To monitor changes in the virulence of field strains in Britain, national surveys (similar to the 1962 survey) were carried out in 1975 and in 1981, and the results of these surveys are reported here. METHODS Collection of samples During October to December 1975 and October to December 1981 two diseased rabbits were obtained from locations at least 10 miles apart within each 'old-style' county in Great Britain. The rabbits, showing obvious symptoms of myxomatosis, were killed and immediately placed in insect-proof linen bags and sent to the laboratory. Eyelid lesions were taken from each rabbit and stored dry at -70 °C until processed. Preparation of virus samples Virus was extracted from the eyelid lesions into 5 ml of phosphate buffered saline (pH 7-2) by Colworth Stomacher (Sharpe & Jackson, 1972). The suspension was centrifuged at 1000 g for 10 min to remove cell debris. Each virus sample was passaged in a New Zealand White rabbit using 0-1 ml of a 10-1 dilution of suspension. The rabbits were killed 14 days after infection and eyelid lesions taken. Virus was then extracted from the eyelid lesions of each of these rabbits as before, and the resulting suspension was titrated by injecting serial 10-fold dilutions on to the shaved flank of another white rabbit. Six intradermal injections of 0-1 ml of each dilution enabled the virus titre of the original suspension to be estimated using the method of Reed & Muench (1938). Virulence test The virulence of 128 virus samples collected in 1975 and 123 samples collected in 1981 was tested using the method of Fenner & Marshall (1957). For each virus sample, six (or occasionally five) New Zealand White rabbits were injected intradermally with 0-1 ml of a diluted suspension containing 10-50 rabbit infectious doses (RID50) and kept at a temperature of 17 + 4 'C. Development of symptoms and day of death were noted for each rabbit. The mean survival time (MST) was calculated for each virus strain and, from this, each strain was assigned to one of six virulence grades as described by Fenner & Marshall (1957). Statistical analysis The distributions of field strains in different virulence grades in 1962, 1975 and 1981 were compared by fitting a log-linear model using Poisson error distribution by means of generalized linear interactive modelling (GLIM) (Baker & Nelder, 1978). The virulence of myxoma virus strains 115 Table 1. The virulence of field strains of myxona virus in Britain in 1962, 1975 and 1981 (The number of strains falling into each virulence grade is expressed as a percentage of the total sample.) Virulence grade No. of _A__ strains _ V II I tested IV III a III b < 13 14-16 17-22 23-28 29-50 MST (days) ... > 99 95-99 90-95 70-90 50-70 < 50 % mortality ... Year 09 14-0 24 8 17-6 38-8 4-1 222 1962* 5*5 0-8 18 0 25-8 48-4 1-6 128 1975 1-6 0 81 35 8 54-5 0 123 1981 * From Fenner & Chapple (1965). RESULTS Calculation of mean survival time The distribution of times to death (survival times) of all rabbits that died after infection with all strains collected in the 1975 and 1981 surveys was skewed, indicating that transformation of the data is required before mean survival times can be calculated. The methods of Jeffers (1959) and of Kirk (1968) both indicated that a reciprocal transform would be most appropriate, but, in almost every case, this gave an identical virulence grade to that obtained by the modified logarithmic transform used by Fenner & Marshall (1957). For the sake of continuity therefore, the logarithmic transform, log [(ST-8)] (where ST = survival time in days) was used in this paper. When one or more of the six rabbits injected with each strain survived infection, Fenner & Marshall (1957) used the method of Sampford (1954) to calculate MST. However, in analysing the results of the (...truncated)