Volume 2: Science
3. The nature and cause of BSE
Transmission of BSE to sheep
Summary

3.229 At the time of publication of the Southwood Report in 1989, it was widely believed that the BSE epidemic was due to infection of cattle with the unmodified scrapie agent. Since scrapie had been endemic in sheep for centuries, with no evidence of any risk to humans, the likelihood that BSE had been passed back to sheep through MBM was considered of little significance.

3.230 However, by 1989 it had become apparent that the BSE agent was a single strain, ¹ and that its range of transmission to other species was different from that of the usual strains of scrapie (see paragraphs 3.51-3.53). A project to test the transmission of BSE to sheep and goats (SE1402) had started at the NPU in 1988. ² Intracerebral transmission to negative-line sheep was achieved by November 1989, ³ and oral transmission by February 1992. ⁴ The ability to transmit to negative-line sheep was significant, in that these animals were not susceptible to scrapie, and thus represented yet another difference between scrapie and BSE. The oral transmission to sheep was achieved by a tiny amount of BSE-infected brain (0.5 g). The implications of this were not recognised at the time.

3.231 This result added further doubt to the scrapie origin theory that had been relied upon earlier. The particular concern with BSE transmission to sheep was that, if it occurred naturally, BSE might become endemic in sheep through maternal and lateral transmission, as with scrapie. Mutton and lamb entering the human food chain could therefore pose the same risk to humans as beef.

3.232 By January 1994 a further important observation had been made: the transmission of BSE to mice from the spleen of a negative-line sheep affected with BSE. ⁵ This indicated that the pattern of tissue infectivity of BSE in sheep was similar to the pattern found in scrapie, and different from that in cattle. The fact that BSE in sheep behaved more like scrapie further raised the concern that the BSE agent might become endemic.

3.233 As BSE transmitted to mice was known from work in the late 1980s and early 1990s to have a characteristic incubation period and lesion profile different from scrapie, ⁶ the mouse bioassay had the potential to test whether or not the BSE agent had passed back into sheep. Indeed, this assay was used in the work described above showing the BSE agent to be a single strain. ⁷ In June 1994 a proposal to determine whether BSE had become endemic in the British sheep population was agreed within MAFF, ⁸ and the project (SE1919) was started at the CVL in April 1995.

3.234 However, the progress of this project appears to have been hampered by difficulties in the collection of sheep brains from suspect cases of scrapie. ⁹ A major difficulty in the design of the project relates to the decision to pool sheep brains before assay in mice. It was already known that when mixtures of scrapie strains were injected into mice, there was competition between the strains, so that the strain most compatible with the mouse genotype was the one that was amplified. The BSE strain would therefore only be recognised in a sheep brain pool if it happened to be the strain that was most compatible. The project is ongoing, and no results are available. In another project (SE1423) being carried out at the NPU, individual strain-typing is the approach adopted. This avoids the loss of sensitivity introduced by brain pooling, since it looks at strains from individual sheep. Results are not due until 2003.

3.235 In March 1996 SEAC reviewed its approach over several years to BSE in sheep, and concluded that although the possibility of sustained BSE infection in the sheep population could not be theoretically ruled out, there did not appear to be any evidence so far to suggest that such sustained infection had actually taken place. However, this assessment appeared to change soon afterwards, and by May 1996 it was acknowledged that various further research was urgently needed in the form of:

1. better scrapie surveillance;
2. more extensive strain-typing of isolates;
3. research into the routes of transmission; and
4. research into the behaviour of BSE infectivity in sheep.

3.236 In 1988 a study of the incidence of scrapie in the UK, based on the results of two anonymous self-administered questionnaires to farmers, gave estimates of 0.5 per cent and 1.1 per cent incidence in affected flocks of over 100 sheep (see paragraphs 2.7-2.9). The prevalence of affected flocks in the two surveys was 17 per cent and 34 per cent respectively, and these were identified in 35 counties. There appear to be no other surveys of flocks which might indicate trends in incidence of the disease before and after the emergence of BSE. However, the numbers of cases diagnosed annually by the Veterinary Investigation Centres (VICs) from referrals are provided by the CVL in their VIDA publications. The numbers are small, indicating that it is likely that only a minority of scrapie cases are investigated at the VICs. Nonetheless, they provide useful information on trends in incidence over time. Table 3.3 shows that the numbers of cases remained more or less constant between 1981 and 1987, ranging between 100 and 176 per annum. From 1988 to 1991, there was an annual rise in numbers from 211 to 906, followed by a decline to 235 in 1994.

Table 3.3: Cases of scrapie in sheep and goats, 1980-98

3.237 It is assumed that scrapie was maintained at a constant level in the natural flock by a combination of maternal and lateral transmission and, in part, by the scrapie agent in MBM fed as part of sheep rations. The latter possibility does not seem to have been formally tested. Assuming that it played a role in maintaining scrapie incidence, the introduction of the ruminant feed ban would have been expected to lead to a reduction in scrapie incidence. Instead, the number of scrapie cases diagnosed by the VICs increased from 1988 to a peak in 1991. Undoubtedly part of the increase in 1991 can be attributed to additional cases recruited by MAFF to provide material for research projects on scrapie inoculation and strain-typing. During 1991-92 MAFF offered farmers £15 per head for scrapie-affected sheep which were referred to the VICs for confirmation of the diagnosis. Although this practice ceased in 1993, in the same year scrapie became a 'notifiable disease', which required farmers to notify all their affected animals while referring only the first case in each flock for pathological confirmation. The numbers submitted in 1994 and 1995 dropped to 235 and 254 per annum respectively, rising in 1996 and 1997 to 459 and 508, then falling slightly in 1998 to 499. Despite these fluctuations, the overall annual numbers were maintained at a higher level than those recorded in the early 1980s, suggesting a trend towards increasing numbers since 1988 despite the introduction of the ruminant feed ban.

3.238 Regulations banning the use of sheep offal in human food were introduced in 1996, although this legislation probably did not affect the numbers of scrapie cases referred for diagnosis. More reliable figures on future scrapie incidence are expected following the slaughter and compensation policy for suspect sheep which was introduced in July 1998. ¹⁰

3.239 Since 1996 funding has been made available for various projects to investigate further the possibility of BSE passing to sheep.

3.240 The work of Professor Collinge's group, published in 1996, showed that BSE and vCJD could be distinguished from previous known forms of CJD by a consistent pattern on western blots with respect to ratios of PrP^Sc glycoforms following proteinase K digestion of infected brain homogenates (see paragraph 4.8). ¹¹ It was proposed that such analysis could be used to examine if BSE had been transmitted to, and was surviving in, the sheep population.

3.241 Later work, published in 1998, examined both PrP conformation and glycosylation ratio of several scrapie strains as well as experimental BSE in sheep. ¹² The glycosylation ratios and PrP fragment patterns of scrapie strains following proteinase K digestion of infected brain homogenate showed considerable diversity, consistent with the expected degree of known diversity in scrapie strain types. When compared with sheep BSE, some scrapie strains had similar glycoform ratios, but quite different fragment sizes. Thus a BSE-type pattern was not found in either archival or recent scrapie cases. Such analysis should allow a distinction to be made between scrapie in sheep and BSE in sheep. By 1998, 16 contemporary cases of sheep TSE had been examined and compared with 18 known scrapie cases; all 16 had scrapie glycosylation patterns. It is understood that this form of screening is continuing with much larger numbers, though results have not yet been published. ¹³

3.242 However, recent work by Hope et al. has found that BSE in sheep could not be easily distinguished from all strains of scrapie on the basis of glycoform analysis alone. ¹⁴ One strain, CH1641, was found to have a similar (but not identical) glycosylation pattern to BSE, although it differed in incubation period. It also differed from BSE in that it was difficult to transmit to mice. A similar result had been reported previously, when a mouse-passaged strain of scrapie (22A) was found to have similar molecular characteristics to a mouse-passaged strain of BSE (301V), although its lesion profile in the brain was different. ¹⁵

Summary

3.243 Currently, it is not established whether or not BSE has passed to sheep. Western blotting studies in a limited number of sheep affected with scrapie have not identified characteristic BSE patterns, and it is understood that strain-typing experiments using approximately 30 samples of sheep scrapie brains in experimental mice have not demonstrated lesion profiles characteristic of BSE. It therefore cannot be assumed that the apparent increase in scrapie numbers is due to infection with BSE agent.