Evidence that the Transmission of One Source of Scrapie Agent to Hamsters Involves Separation of Agent Strains from a Mixture

Transcription

1 J. gen. Virol. 0978), 39, Printed in Great Britain 487 Evidence that the Transmission of One Source of Scrapie Agent to Hamsters Involves Separation of Agent Strains from a Mixture By R. H. KIMBERLIN AND CAROL A. WALKER Agricultural Research Council, Institute for Research on Animal Diseases, Compton, Nr. Newbury, Berks. RGI6 onn (Accepted I6 December I977) SUMMARY A previous paper (Kimberlin & Walker, 1977) described an experimental model of scrapie in hamsters in which the incubation period decreased progressively over the first 4 passages before becoming stable at the 5th and subsequent passes. Studies have been made of some of the agent strains present in brains taken from the 2nd, 3rd, 4th and 6th hamster passes. The results indicate the presence of at least two strains of agent at the 3rd passage level. One of these (43IK) is highly pathogenic for mice and the other (263K) has an extremely low pathogenicity for mice. However only one of these strains (263K) is present in hamster brain after the 6th serial passage. It is suggested that the 'adaptation' of scrapie to hamsters may involve the selection, from a mixture, of a single strain which is highly pathogenic for hamsters. The possibility of modification of the properties of agent strains on passage discussed. INTRODUCTION A recent publication described a model of scrapie in golden hamsters in which the incubation period following intracerebral infection with high doses of agent was as low as 60 days (Kirnberlin & Walker, i977). However this short incubation period was not consistently obtained until after the 4th serial passage in hamsters. At the first pass, only a proportion of the injected hamsters developed scrapie and the incubation period was more than 300 days. At the second hamster passage, all the animals developed scrapie and the incubation time was halved to about 130 days, and then halved again during the next two passages to 60 to 70 days. This phenomenon is a good example of the species barrier effect that is commonly seen in scrapie. Some possible explanations for this effect have been discussed (Dickinson, I976 ) but there is very little published work on the nature of the underlying mechanisms. In one study (Kimberlin et al, I975) it was pointed out that experimental infection with scrapie necessarily involves injecting animals with crude inocula and that the species barrier may be due in part to a host response to the large amount of antigenically foreign material, leading to the passive removal of much of the injected agent. This possibility was investigated in some experiments involving the intraperitoneal infection of mice with hamster passaged scrapie. It was found that repeated intraperitoneal injections of mice with normal hamster brain, before infection with scrapie brain, impeded the transmission of scrapie from hamsters to mice (Kimberlin et al. I975). However, it seems unlikely that an enhanced inactivation of agent by the new host is the sole explanation of the species barrier effect. First, it was shown that at the 5th hamster 32-2

2 488 R.H. KIMBERLIN AND C. A. WALKER passage, the incubation period following the injection of I LD~0 (hamster intracerebral or intraperitoneai unit) rarely exceeded 2oo days. In contrast, the incubation period at the first passage in hamsters was more than 3oo days, well beyond the normal dose-response end point (Kimberlin & Walker, 1977). Secondly, we have unpublished data showing that the incubation periods of different strains of mouse passaged scrapie vary enormously at first passage in hamsters even though the nature and amount of host material in the inoculum is essentially the same. Thirdly, the present paper describes the transmission of one strain of agent (designated 43IK) from hamsters to mice with only a minimal species barrier effect. Dickinson (I976) has shown that many natural isolates of sheep scrapie contain mixtures of different strains. It is possible that the balance of agent strains maintained in one host species may alter considerably when passaged in a different species. The exaggerated incubation period at first pass could reflect the low pathogenicity for the new host of the preponderant strain in the original mixture, and the reduced incubation period at later passes could be due to the selective advantage which a minor strain gains due to its greater pathogenicity for the new host. Alternatively the adaptation of scrapie to a new species may involve an alteration in the properties of a single agent. We have examined these possibilities by comparing the properties of the agent after the 2rid, 3rd, 4th and 6th serial passage in hamsters when injected into mice. The results show that the reduction of incubation period in hamsters between the 2rid and 4th pass is accompanied by the selection of an agent strain (263K) which does not produce scrapie after intracerebral injection into mice and by the loss of a strain (43IK) which can produce scrapie in mice. METHODS Sources ofscrapie. All experiments were carried out with scrapie agent that was originally derived from the Compton 'drowsy goat' source (see Dickinson, 1976) and transmitted to mice by Chandler (1963). This material has been serially passaged intracerebrally (i.c.) as brain homogenates in Compton white mice for more than 15 years and for at least 35 passages (main lille Chandler agent; see Fig. I). Available evidence from studies of agent properties in mice of different sinc genotypes indicates that only a single agent is present after 2o mouse passes, designated' Chandler' or 139A (Dickinson, 1976; and unpublished results). Brain homogenates from the 3rd mouse passage were used to transmit the disease to rats (Chandler & Fisher, 1963). After approximately 12 serial i.c. passages ill rats, scrapie was transmitted to golden hamsters (Chandler & Turfrey, 1972; Kimberlin & Walker, 1977) as illustrated in Fig. I. Animals. Female, outbred golden hamsters were obtained from a local supplier and injected as adults. Except where otherwise stated, the mice used were female, adult, Compton albinos (genotype sinc s~) randomly bred under specific pathogen-free (SPF) conditions. In In some studies, female LAC/G mice (random-bred, SPF) and C57BL mice of both sexes (originally inbred but random-bred for current experiments) were used. All mice were injected as adults (6 to 9 weeks). Passage experiments. Full details of tissue storage, preparation of inocula, sterilization procedures and titration of infectivity are given elsewhere (Kimberlin & Walker, 1977). Briefly, all experiments (with specified exceptions in Results) were carried out with brain taken from animals with clinical signs of scrapie and stored at - 20 C. Whole brain homogenates were prepared in saline at a wet weight concentration of I or 5 ~- With the exceptions noted in Table 3, homogenates were prepared from pools of several brains. All passage

3 Separation of scrapie strains on passage 489 experiments were carried out with undiluted homogenates. Hamsters received o'o5 ml of inoculum i.c. and mice o'o3 ml i.c. and either o'o3 ml or o.i ml intraperitoneally (i.p.). The length of incubation period was determined from the occurrence of clinical signs of scrapie as already described for hamsters (Kimberlin & Walker, I977) and mice (Kimberlin & MiUson, I976). RESULTS Altered pathogenicity for mice of agent passaged in hamsters The origin of hamster passaged scrapie agent is illustrated in Fig. I. It was isolated from the' drowsy goat' passage line into mice and serially passaged through rats. Pattison & Jones (I968) showed that after 5 passages in rats, the agent could still produce disease when injected i.c. into mice. The rat passaged agent was used after the Izth pass to transmit disease to hamsters. Half-way through the second pass in hamsters (at 66 and 77 days after i.e. injection) a small pool of hamsters brains was prepared. This pool had an infectivity titre of 6.o -logl0lds0 i.c. units/o.o 5 g brain when assayed in hamsters, i.e. the titre was about Ioo times lower than that found in the clinical stage of the disease (Kimberlin & Walker, I977). Even though the infectivity titre in this brain pool was relatively low, agent was still readily transmissible to mice; the i.c. injection of a 5 ~ homogenate of the pool gave I oo ~ scrapie cases in mice with an incubation period ofz7z _+ 4 days (mean _+ s.e. mean.). On completion of the second pass in hamsters, two brain pools were established and thereafter hamster scrapie was passaged in two separate lines (Kimberlin & Walker, I977; see Fig. I). Brain material from 2nd and 4th hamster passes produced disease in mice when injected i.c. although the mean incubation periods were variable (Table I). However, when brain material was taken from both lines at 6th passage in hamsters and injected i.e. into mice, no cases of scrapie developed even though the mice were observed for up to 735 days after injection (Table I). In further experiments, mice were injected i.c. with hamster brain from 6th pass, line I, and brains and spleens were removed from groups of two mice at intervals of I, 2, 4, 6, 9, I2, I6, 20, 24, 3 O, 35 and 4o weeks after injection. Tissue homogenates were prepared (5 ~ in saline) and injected i.c. into further groups of mice which were observed for a minimum of 48o days after injection. A total of 384 mice were injected but none developed scrapie indicating that the agent present in hamster brain at 6th passage is unable to replicate in mice for at least 4o weeks after infection, and probably not at all. This agent (designated strain 263 K) has now been passaged in hamsters Io times in line I and 9 times in line 2 and the last 3 and 4 passes (respectively) have been at limiting dilution (i.e. agent has been cloned). Strain z63k appears to have stable properties which can be summarized as follows: (a) the concentration of agent in brain in the clinical stage of the disease is ~ 8.o - logl0lds0 i.c. hamster units/o.o5 g, (b) the incubation period at successive passes is 6o to 7o days (I to 5 ~ brain suspensions injected i.c.) and (e) the agent has extremely low pathogenicity for mice (at least in the Compton stock which has the sine genotype s7s7; Kimberlin & Walker, I977; and unpublished data). Identification of two strains of serapie agent in hamster brains from third serial passage The data in Table I show that serial passage of scrapie in hamsters causes a loss of pathogenicity of agent for mice but it is not clear whether this change is due to a modification of the properties of a single agent on passage or to the selection of one strain from a preexisting mixture. If the second possibility is correct then one might expect to detect two or more separate strains of agent in hamster brain at early passage levels.

4 490 R.H. KIMBERLIN AND C. A. WALKER DROWSY GOAT SOURCE OF SCRAPIE ] (3 1 M ICE 3rd pass I RATS 5th pass 27th 12th pass HAMSTERS 2nd pass Two pools Linel I ] Line2 HAMSTERS 3rd pass HAMSTERS 3rd pass 1 1 4th pass 4th pass 1 Cloned 5th pass 5th pass 1 Cloned 6th pass 6th pass Cloned by 3 passes at limiting dilution (10-7 or 10-3 ) 35th pass Main Line ] CHANDLER I Cloned ] CHANDLER Fig. I. Passage history of the 'drowsy goat' source of scrapie in various species of rodents at Compton. Boxes indicate isolates and agent strains. Serial passages were carried out by i.c. injection of brain homogenates in saline prepared fromanimals with clinical signs of scrapie. Passage numbers refer to the host indicated at the head of each passage line. (l) Chandler M.C. Clarke Chandler & Fisher Pattison & Jones Chandler & Turfrey Kimberlin & Walker (I977). In the course of other experiments a large pool of scrapie hamster brains was prepared from both lines at 3rd passage. This material produced scrapie in mice with an incubation time of only days after i.c. injection (Table 2). At 2nd i.c. passage in mice, the incubation period decreased slightly to I4o + e days and was essentially the same at 3rd passage (148_+ 3 days; Table 2). These results indicated the presence of an agent in this particular (3rd passage) hamster brain pool which could produce scrapie in mice with only

5 Separation of scrapie strains on passage 49I Table I. Incubation period of hamster passaged scrapie at first intracerebral passage in mice Passage number in hamsters Line I Scrap±e* cases Incubation period (days_+ s.c. mean.) I7/I7 316"+7 18/I8 248±7 o/19 (735)t 0/12~ (735) Line 2 2 9/ /20~ 325,+ 6 4 I4/I6 5o2_ o/14 (690) * Ratio of the number of eases of scrapie to the number of mice inoculated. t Observation period. :~ Female LAC/G mice used. This material had been twice cloned by passage at limiting dilution in hamsters. Table 2. Incubation period of scrapie from different hamster passages when serially passaged in miee intraeerebrally Passage and Incubation period [days _+ s.c. mean. of (n) animals] line number ~ - - ~ in hamsters 1st pass 2nd pass 3rd pass 4th pass 2, Line I 316,+7 (17) I64+ I (I2) I42,+ 1 (I1) 2, Line z 315"+7 (9) 163_+ I (17) I42~2 (I9) 3, Lines I and 2 168,+2 (16) 13o,+2 (I2) I48 3 (13) 4, Line I ) I46+2 (I8) - 4, Line 2 5o2_+ 36 (14) 166,+2 (62) - 136,+ I (16) a minimal species barrier effect. This agent is designated strain 43~K and it clearly differs from 263K in its pathogenicity for mice. Further transmission studies were carried out with homogenates of four individual brains and two small brain pools taken from the 3rd passage in hamsters and injected into mice. Table 3 shows that three of the single brains and one of the small brain pools contained an agent very similar to 263K, i.e. the concentration of the agent in brain was 7"8 to logl0lds0 i.c. hamster units/o.o5 g; the i.c. incubation period in hamsters at the next pass was 64 to 69 days; it showed extremely low pathogenicity for mice. The second brain pool also contained an agent which resembled 263K in having a very low pathogenicity for mice. However, one hamster brain contained an agent with quite different properties. The concentration of the agent in brain was ~ 5"2 -loga0lds0 i.c. hamster units/o-o5 g; the incubation period in hamsters at the next pass was 88 to 99 days; it was highly pathogenic for mice with only a small species barrier effect, the incubation period at first pass in mice being ~76_+ 2 days (Table 3) and at second pass I days (data not shown). In the last respect, the agent closely resembled strain 43~K. Possibility of other strains of agent in hamster passaged scrapie The present studies have relied mainly on relative pathogenicity for mice to distinguish between strains of agent. The strains 263K and 43~K possess the most divergent properties that can be detected by this method, namely, non-pathogenic for mice and highly pathogenic

6 - - - I R.H. KIMBERLIN AND C. A. WALKER Table 3. Incubation periods in hamsters and mice injected with different brain samples taken after third serial passage of scrapie in hamsters Incubation period [days 4- s.e. mean of (n) animals] Brain Titre of sample agent in Hamsters Mice Mice number brain* i.c. i.c. i.p. Line 1 I 8" (6) -- (72o)~ " (32) 2:~ ~ 7"8 644-o(3) 518 (I)II (665) (I6) (7Io) (43) Lines I and 2 (16) (28) Line 2 I* ~ 5" (5) (I6) (15) (4) 2:~ 8"2 654-I (3) 46I+22(4) (665) (I6) 3~: 8.z 67+_1 (3) 482 (1)/! (665) (16) * Titre in hamsters expressed as -logl0ld~0 i.c. units/o.o5 g brain.? Period of observation in which no cases of scrapie were seen. Single hamster brains; othelwise pools from two or more hamsters used. Repeat on same sample stored as a frozen homogenate for 16 months. ]1 7 ~ and 29 ~ of mice developed scrapie. Table 4- Comparison of the incubation periods produced in two mouse strains following intraeerebral or intraperitoneal infection with the 'Chandler' strain of agent or isolate 3ozK Incubation period (days) (mean 4- s.e. mean)* Strain Sex,. Route r ~- of of of ' Chandler' ' Chandler' Isolate mouse mouse inoculation cloned? main liner 302K~ Compton F i.c I i.p I P/C ratio... 1'37 1"41 1'38 C57BL F i.c i.p. 19o+2 I P/C ratio... 1"4o r C57BL M i.c. I i.p I9I +4 P/C ratio... 1"45 I "4o 1.43 * Incubation periods are the mean values of I1 to 12 Compton white mice or 5 to 6 C57BL mice. t Both sources of Chandler scrapie were routinely passaged in Compton white mice and passaged once (i.e.) in female C57BL before injection into the mouse strains shown above. + Isolate 3o2K was obtained from the 2nd hamster pass, line I (Tables 1 and 2) and passaged (i.c.) twice in Compton white mice and once in female C57BL mice. Ratio of incubation periods following i.p. or i.c. infection with o'o3 ml of I ~ scrapie brain suspensions. with minimal species barrier effect. There could be other strains present which have intermediate properties. For example an isolate (designated 3oaK) was made from and hamster passage, line I (see Fig. I) which was pathogenic for mice. This isolate showed a large species barrier effect at Ist passage in mice (Tables i and a). A similar isolate from line 2 had identical properties. However these isolates have not yet been cloned in mice and the properties of isolate 3oaK at Ist pass in mice could be due to a mixture of strains z63 K and 43IK, with selection against 263 K at later mouse passes. Further studies with isolate 3o2K have been carried out because it is clearly different from strain 263K but its relationship with strain 43rK is not yet known.

7 Separation of scrapie strains on passage (a) b tt./ L o 1 3 i i (b) ~".,..., " "., ~. & / I I i ' I *! i I Grey White matter matter Position in brain Fig. 2. Lesion profiles (in nine grey matter and three white matter areas of brain) of main line 'Chandler' and cloned 'Chandler' agent and of isolate 3o2K. The passage history of these agents is given in the footnote to Table 4- Lesion scores were determined in the brains of female Compton white mice killed in the clinical stage of scrapie after infection by i.c. (a) and i.p. (b) routes of inoculation. Scoring was carried out on coded samples by Dr H. Fraser according to published methods (Fraser & Dickinson, I968; Fraser, 1976). Profiles were constructed from the mean scores of Io to 12 individual brains ---, Main line ' Chandler' ;... cloned ' Chandler'; --, 3o2K isolate. Comparison of the properties of isolate 3o2K with the ' Chandler' strain of scrapie agent Fig. I illustrates that the source of hamster scrapie used in these studies was the same as the 'Chandler' strain of mouse passaged scrapie which has been used at Compton for many years. In view of this relationship, it was of interest to investigate the similarities between the main line' Chandler' strain of agent and isolate 3ozK, which was only obtained in mice after passage in rats and hamsters. Table 4 shows no differences between the two agents in their incubation period properties in two strains of mice with the sinc genotype s7s7. A standard method for identifying different strains of scrapie agent is by measuring the severity and distribution of vacuolar lesions produced in anatomically defined areas of the brain, to give a 'lesion profile' (see Fraser & Dickinson, ~968; Fraser, I976). The lesion profiles of the 'Chandler' and 3ozK sources of agent were identical when compared in two strains of mice infected by the i.c. or i.p. route of inoculation. The sensitivity of this technique is illustrated by the difference in profiles produced by the two routes of inoculation (Fig. 2; see also Fraser, I976). However, very marked differences between agent sources were seen when they were injected into hamsters. The incubation period of isolate 3o2K at first pass in hamsters was i86 to I95 days whereas the incubation period of the 'Chandler' strain was very much longer, 359 to 4o9 days (Table 5).

8 494 R.H. KIMBERLIN AND C. A. WALKER Table 5. Comparison of the incubation periods at first pass in golden hamsters of the 'Chandler' strain of scrapie and isolate 3o2K Experiment Incubation period (days) number* Strain of agent [mean + s.e. mean of (n) hamsters] I Main line 'Chandler' 359 q- 12 (IZ) 2 Main line ' Chandler' (5) 2 Cloned ' Chandler' 378 "-}- I I (5) 2 3o2K I95 _+ 6 (6) 3 3ozK I86_+ t6 (6) 4 3o2K I93 +_ 3 (9) * Expt. I main line ' Chandler' agent passaged only in Compton white mice. Expt. 2: both 'Chandler' lines passaged in Compton white mice and then once in C57BL females. Isolate 3o2K passaged twice in Compton white mice and once in C57BL females. Expt. 3 : isolate 3o2K passaged three times in Compton white mice. Expt. 4: isolate 3o2K passaged twice in Compton white mice. DISCUSSION The present work shows that at least two different strains of agent are present at the early passage levels of scrapie in hamsters. Strain 263K appeared to be present in brains taken from the 3rd hamster pass and was found at the 6th pass as the dominant strain of hamster scrapie. This strain is interesting because it does not appear to replicate in mice. However, it is known that agent replication in mice is controlled by the sinc gene which has two alleles, designated s7 and P7 (Dickinson et al. ~969; Dickinson & Fraser, I969; Dickinson & Meikle, I971 ). The mice used in the present study were of the genotype sincl Experiments are in progress to see if strain 263K can replicate and produce disease in mice of the genotype sinc ft. Strain 43IK was isolated from the 3rd hamster pass and differs from z63k in that it readily produces disease in mice with only a minimal species barrier effect after i.c. injection The properties of this agent strain clearly demonstrate that scrapie can be transmitted experimentally from one species to another without any apparent need for 'adaptation' to the new host. The main finding of this study is that the decrease in i.c. incubation period of scrapie in hamsters between 2nd and 4th passage is associated with the selection of strain 263 K and the loss of strain 431K. We suggest that a selection of agent strains from a mixture underlies the large species barrier effect seen at the early passage levels in this model of hamster scrapie. Unfortunately the agent composition of the original inoculum used to transmit scrapie to hamsters is unknown and it is impossible to analyse in detail the changing population of agent strains at all passage levels. However it is plausible to suggest that 263K was a minor component in the original inoculum. The very long incubation periods at Ist and 2nd passage (compared to later passes) in hamsters could be a reflection of at least four major factors: the relative amounts of each agent strain originally present; modification of pathogenesis due to association of agents with alien tissue; the relative pathogenicities of these strains for the new host; and perhaps most important of all, interactions between these strains. Competition between strains of scrapie agent for a limited number of replication sites is a well established phenomenon when different strains are injected into mice at different times (Dickinson et al. I972, ~975). However there is no published evidence that agents can block one another when injected at the same time and experiments are underway to investigate this possibility.

9 Separation of scrapie strains on passage 495 Two lines of evidence suggest that this particular example of a species barrier effect is not associated with a host induced modification of a single agent strain. First, the incubation period of scrapie in hamsters was not stable until after the 4th serial passage (Kimberlin & Walker, I977). Secondly, two different strains of agent (263K and 43IK) were identified in different brains at the 3rd hamster pass. These observations are consistent with a ' host permitted' selection of agent strains from a mixture. However this does not exclucle the possibility that host induced modification of agent properties may be a factor in the transmission of scrapie to new hosts. It is interesting to point out that strain 263K and isolate 3o2K (which may or may not contain 43IK) are different from the main line of 'Chandler', mouse-passaged scrapie agent even though they all originated from a common source. One could argue that different agent strains arose by host induced modifications of the 'Chandler' strain during serial passage through rats. The problem here is that there is no information on agent strain composition at the early mouse passages of 'Chandler' scrapie from which the rat and hamster passage lines were derived. However since 263K does not appear to replicate in mice it is difficult to see how it could have been isolated into mice from the 'drowsy goat' line unless one postulates the existence of strains of scrapie agent which can replicate in mice but are completely non-pathogenic in this host. Other possibilities are that 263K arose by host induced modification during passage in rats, by mutation or even contamination. The question of host induced modification of agent strains is an important one to answer from the point of view of the epidemiology of scrapie but it is clear from this discussion that it can only be pursued by studying the effects of passage in different hosts of well characterized, single strains of agent. Experiments on these lines are in progress with several mouse passaged strains of scrapie injected into hamsters. We would like to thank Dr H. Fraser and Mrs Patricia Robertson for kindly determining the histological lesion profiles of some sources of scrapie agent and we gratefully acknowledge the expert care of the animals provided by Janice A. Robinson. REFERENCES CHANDLER, R. L. (I963). Experimental scrapie in the mouse. Research in Veterinary Science 4, CHANDLER, R. L. a FISHER, J. (1963). Experimental transmission of scrapie to rats. Lancet ii, I I65. CnANDLER, a. L. & TtJRFREY, B. g. (I972)- Inoculation of voles, Chinese hamsters, gerbils and guinea pigs with scrapie brain material. Research in Veterinary Science x3, DICKINSON, A. G. (I976). In Slow Virus Diseases of Animals and Man. Frontiers of Biology series, vol. 44, chapter Io, pp. 2o9-24I. Edited by R. H. Kimberlin. Amsterdam and New York: North Holland. DICKINSON, A. G. & FRASER, H. 0969). Genetical control of the concentration of ME7 scrapie agent in mouse spleen. Journal of Comparative Pathology 79, DICKINSON, A. G., FRASER, H., McCONNELL, I., OUTRAM, G. W., SALES, D. I. & TAYLOR, D. M. 0975). Extraneural competition between different scrapie agents leading to loss of infectivity. Nature, London 253, 556. DICKINSON, A. G., FRASER, n., MEIKLE, V. M. H. & OUTRAM, G. W. (I972). Comlzetition between different scrapie agents in mice. Nature, New Biology 237, DICKINSON, A. G. & MEIKLE, V. M. H. (I97I). Host-genotype and agent effects in scrapie incubation: change in allelic interaction with different strains of agent. Molecular and General Genetics xx2, DICKINSON, A.G., MEIKLE, V. M. H. & FRASER, H. (I969). Genetical control of the concentration of ME7 scrapie agent in the brain of mice. Journal of Comparative Pathology 79, I5-22. FRASER, n. (I976). In Slow Virus Diseases of Animals andman. Frontiers of Biology series, vol. 44, chapter I2, pp Edited by R. H. Kimberlin. Amsterdam and New York: North Holland. FRASER, H. & DICKINSON, A. G. (I968). The sequential development of the brain lesions of scrapie in three strains of mice. Journal of Comparative Pathology 78, 3oI-3 H. KIMBERLIN, R. H. & M1LLSON, G. C. 0976). The effects of cuprizone toxicity on the incubation period of scrapie in mice. Journal of Comparative Pathology 86,

10 496 R.H. KIMBERLIN AND C.A. WALKER KIMBERLIN, R.H. & WALKER, C.A. (1977). Characteristics of a short incubation model of scrapie in the golden hamster. Journal of General Virology 34, 295-3o4. KIMBERLIN, R. H., WALKER, C. A. & MILLSON, G. C. (1975). Interspecies transmission of serapie-like diseases. Lancet ii, I3O9-13IO. PATTISON, I. H. & JONES, K. M. (1968). Modification of a strain of mouse-adapted scrapie by passage through rats. Research in Veterinary Science 9, 4o8-41o. (Received 2 November 1977)