Volume 2: Science
3. The nature and cause of BSE
Infectivity and transmissibility of BSE
The measurement of infectivity of different cattle tissues
Summary

3.194 In this chapter, we have so far outlined the investigations undertaken between 1986 and the present day into the origin of BSE, its nature and how the disease spreads. As described, most people involved with the study of the disease, and the development and implementation of control measures, were working on the basis that the disease originated from scrapie in sheep.

3.195 Over the same period, other important questions about BSE had also been identified, and were addressed by scientific investigation. It was vital, for example, to address the issue of infectivity of BSE and the risk of transmissibility to other animals, including humans. These risks had been raised by several scientists and physicians outside government itself, for example Dr Tim Holt (June 1988), ¹ Dr Helen Grant (February 1989), ² and Dr Stephen Dealler and Professor Richard Lacey (June 1990). ³

3.196 We noted in paragraph 3.163 that the effective dose would depend both on the amount of material given and on its infectivity titre. This section will now look in detail at the experiments carried out to assess both the infectivity of tissues and the dose needed to cause disease.

3.197 As discussed in Chapter 2, various facts were known early on in the epidemic, from work looking at the infectivity and transmission of other TSEs. These formed the basis of the understanding of BSE in the early days following its emergence, and suggested several areas for further investigation:

1. Species barrier effect - it was known that the incubation period at first pass in a new species was significantly longer than those seen at later passes in the same species. ⁴ It had also been shown that the efficiency of transmission was higher within, rather than between, species.
2. Different routes of natural transmission existed - oral route in kuru and TME, maternal and lateral routes in scrapie, and a possible intradermal component to TME transmission.
3. Possibility for iatrogenic transmission existed - louping-ill vaccine had been responsible for an outbreak of scrapie, and neurosurgical instruments, corneal grafts and pituitary-derived hormones had been implicated in certain cases of CJD.
4. Different experimental inoculation routes varied in their efficiency - intracerebral had consistently been found to be more efficient than peripheral routes (oral, subcutaneous, intraperitoneal, intravenous), although the intravenous route had also been found to be relatively efficient. ⁵
5. Pathogenesis - studies of scrapie had identified the lymphoreticular system as the main area of subclinical infection, with the spleen having a particularly important role in agent replication. ⁶ When clinically ill, naturally infected sheep were studied, the agent was found to be widely dispersed throughout the central nervous tissue, with titres (concentrations of infectious agent) much higher than in non-neural tissues.
6. Species host range - scrapie, kuru and CJD had been shown to be experimentally transmissible to several species, including, in some cases, primates.
7. Disease susceptibility/incubation times - genetic factors had been shown to be important (sinc gene in mice, sip in sheep), ie, certain strains of mice with specific forms of the sinc gene were more susceptible than others.
8. Existence of different strains - different strains of scrapie and TME had been isolated, each showing different lesion profiles and incubation times. Similarly, different mutations of the prion gene in familial CJD were associated with different clinical presentations.

The measurement of infectivity of different cattle tissues

3.198 One of the first questions raised was which cattle tissues carried the BSE agent, and hence were infective. The question was crucial for determining appropriate measures to protect animal and human health. However, in the absence of a test for the BSE agent, and before results of tissue infectivity tests were available for BSE, any decision to exclude specific tissues from the human food chain could only be based on experience with other TSEs. This essentially meant that the justification for the SBO ban of November 1989 was based on work with scrapie. ⁷ The study by Hadlow of natural scrapie, published in 1982, had indicated that after brain and spinal cord the most infective tissues were those of the lymphoreticular system, specifically spleen, thymus, tonsils, lymph nodes and the lymphatic tissues (Peyer's patches) within the wall of the intestines (see paragraphs 2.162-2.163). ⁸ Hadlow showed that the level of scrapie infectivity of LRS tissues was similar in pre-clinical cases of scrapie aged between 10 and 25 months and adults with clinical disease aged between 34 and 57 months.

3.199 Kimberlin used the observations of Hadlow and others to group the tissues of clinically affected sheep into four categories of relative infectivity according to the observed infectivity titres as measured in the mouse bioassay (see Table 3.2, below).

Table 3.2: Categories of tissue infectivity

3.200 In lambs up to 1 year of age there is virtually no infectivity in category I and titres in category II are about ten times lower than at the clinical stage. Between 1 year and the onset of disease the category II levels are similar to those in the clinical stage:

. . . pre-clinical animals older than one year are regarded as having category II and category IV titres the same as at the clinical stage. Category I titres would be progressively lower the further the animals are away from clinical disease. But there is no way of estimating these values and so a 'worst case' assumption has to be made. ⁹

3.201 These considerations provided the basis for the human and animal SBO bans introduced in 1989 and 1990 respectively.

3.202 In order to confirm that this was the correct approach to take, studies of the infectivity of tissues from BSE-affected cows were needed. Experiments took place at both the NPU and the CVL.

3.203 The first successful BSE transmission studies, carried out at the NPU in November 1987, involved the intracerebral inoculation of mice with brain homogenates from BSE-affected cattle. By September 1988 positive results were obtained. ¹⁰ This was important in two respects:

1. it established the transmissible nature of BSE, and showed that brain tissue was infective in BSE-affected cattle; and
2. it showed that mice could be used as an experimental model for the disease - hence the establishment of the mouse bioassay.

3.204 Investigations of cattle tissues other than the brain were initiated at the NPU in 1989, after both the Southwood Working Party and the Tyrrell Committee identified the need for further work. ¹¹ The next question, however, was how to measure infectivity.

3.205 As mentioned in Chapter 2, work prior to 1986 on the experimental transmissibility of scrapie and CJD had shown the importance of the concept of the species barrier - the barrier that had to be crossed before infection occurred. ¹² The species barrier was measured largely in terms of the difference in incubation periods on transmission between animals of the same species and animals of different species. After a successful transmission between two different species, the incubation period was invariably much shorter during subsequent passage from one animal to another within the same species.

3.206 These observations indicated that the most sensitive method for determining the infectivity of cattle tissues would be the experimental inoculation of tissue samples into unaffected calves, since no species barrier would have to be breached. However, titrations (experiments carried out to estimate infectivity) in cattle were regarded as prohibitively expensive and lengthy, and reliance was instead placed on the mouse bioassay. ¹³ This meant the presence of a species barrier and the inevitable loss in sensitivity. However, the Hadlow studies had successfully used the mouse assay to demonstrate the infectivity of a wide range of nervous and lymphoreticular tissues in sheep affected with scrapie. ¹⁴ It was therefore expected that the same assay would be suitable for use in assessing infectivity of tissues in BSE-affected cattle. As discussed below, the sensitivity of the bioassay was later to be questioned. Independent scientists, including Dr Dealler and Professor Lacey, also criticised the reliance upon this assay, and raised the issue with Mr Bradley and Mr Wilesmith (of the CVL) and Dr Tyrrell (the SEAC chairman) at a meeting at Stoneleigh National Agricultural Showground in the summer of 1993. ¹⁵

Bioassay and infectivity criteria

3.207 The method used for assaying the infectivity of a tissue has been described in Chapter 1. Briefly, the bioassay is a titration experiment whereby infective tissue, ground up in salt solution, is sequentially diluted and inoculated into experimental animals. The titre of the material can be calculated based on the dilution at which approximately 50 per cent of the experimental animals succumb to disease. There are two units to define amount of titre in a given tissue: the LD₅₀ (lethal dose) unit per gram and the ID₅₀ (infective dose) unit per gram. One LD₅₀ unit is defined as the dose necessary to give a 50 per cent probability of killing an animal, while one ID₅₀ unit is the dose necessary to give a 50 per cent chance of infecting an animal. Since infection with TSEs invariably leads to death, these are effectively the same thing for these diseases. The bioassay is used to compare the amounts of infectivity in different tissues and at different times after infection.

The NPU tissue infectivity study

3.208 The study at the NPU involved the combined intracerebral and intraperitoneal inoculation of mice with a large number of non-CNS tissues from BSE-affected cattle. Spleen and placental tissue were administered orally in addition to the other routes. ¹⁶ Preliminary results were published in 1992 and showed that bioassays of spleen, lymph node, semen, buffy coat (laboratory term for white blood cells) and muscle all failed to produce disease in the mouse strain most susceptible to BSE infection. ¹⁷ A separate study, carried out at the Royal Veterinary College and published in 1993, likewise showed absence of infectivity, with milk, udder, supramammary lymph nodes, placenta and foetal membranes fed to experimental mice. ¹⁸

3.209 When spleens from the mice inoculated intracerebrally with BSE brain were inoculated into further mice, infectivity was found, with disease developing after incubation times shorter than in the original inoculation. This was significant in that infectivity of spleen tissue with BSE agent could be shown once the species barrier had been passed. ¹⁹

3.210 However, the negative results described in paragraph 3.208 did not necessarily indicate the absence of infectivity. Possibilities were initially raised that infectivity was confined to the brain, or that infectivity in peripheral organs may have peaked and fallen before infectivity assays were carried out. ²⁰ Concern then quite quickly shifted to the validity of the use of the mouse bioassay for BSE. ²¹

3.211 Further results from the NPU, reported in 1993 and published in 1994, revealed that BSE had still not been transmitted to mice directly from cerebrospinal fluid (CSF), peripheral nervous tissue, spleen or lymphoid tissue from cattle, even though infectivity had been shown in the same tissues from scrapie-affected sheep. ²²

The pathogenesis study

3.212 At the same time as the NPU work was being carried out, the pathogenesis of BSE (temporal spread of infectivity around the body) was also being investigated. The pathogenesis study, carried out at the CVL, was initiated in December 1991, after it was suggested in 1990 that there was a need to 'experimentally justify the . . . offals ban'. ²³ There had been suggestions that, since no infectivity had been found in extra-neural tissues, the pattern of pathogenesis might not be the same as that seen in sheep. ²⁴

3.213 Cattle were given a single oral dose of 100 g of BSE-affected cattle brain and slaughtered at intervals thereafter in order to follow the route of infection to the brain. Forty-four tissues were studied for infectivity, again using the mouse bioassay. ²⁵ In June 1994 infectivity was reported in the ileum of cattle slaughtered 6 months after oral dosing. ²⁶ This result was significant in that, with scrapie, no infectivity was seen in this organ until 10 months after inoculation. It also represented a difference from the results of the NPU study, in that infectivity of BSE had now been detected outside the brain and spinal cord. It led to the 1994 amendment to the SBO Order banning the use of intestine from all bovine animals except those that had died aged 2 months or under.

3.214 Further results from the NPU were reported in 1995, ²⁷ from assays with milk from BSE-affected cows. Certain groups of weanling mice received combined intracerebral (0.02 ml) and intraperitonal (0.1 ml) injections, while others received an oral dose of 300 ml. No infectivity was found. Negative results were also reported for tracheal tissue. ²⁸

3.215 A 1998 update of the pathogenesis study revealed that infectivity had been demonstrated in brain, spinal cord, peripheral nervous system, cervical and thoracic dorsal root ganglia, the trigeminal ganglia, retina, bone marrow and ileum. ²⁹ This was the first time that dorsal root ganglia had been found to be infective in mice, even though vacuolation of this tissue had been found in 1987. ³⁰ Spleen and other lymphoreticular tissues have consistently given negative results in the mouse bioassay.

3.216 Infectivity of the dorsal root ganglia was significant, in that these cells represent a possible route from the lymphoreticular system to the central nervous system (see paragraph 2.169).

3.217 The results of the two studies revealed that the pathogenesis of BSE varied markedly from that of scrapie in sheep. In this context, rather than suggesting a lack of infectivity in the tissues tested, the negative NPU results raised the possibility that tissues which failed to transmit across the species barrier to mice might still contain sufficient infectivity to transmit BSE to cattle. ³¹ The failure to detect infectivity in peripheral tissues, particularly those of the lymphoreticular system, was consistently seen as surprising. A 1994 MAFF summary report concluded that this 'show[ed] a major difference from scrapie and other homologous diseases', which needed to be resolved. The report suggested that in order to confirm the absence of agent in spleen or lymphoid tissue, the sensitivity of the mouse assay needed to be assessed against a bovine assay. ³²

The comparative bioassay

3.218 As reports of the negative tissue infectivity results continued, concern about the possible lack of sensitivity of the mouse bioassay led to a further project, the comparative bioassay, which started in January 1993. This new study aimed to compare infectivity following the simultaneous inoculation of mice and calves, and hence give a measure of the relative sensitivity of the two species as bioassays. ³³ Samples of pooled BSE-affected brain tissue, spleen and pooled lymph nodes were inoculated into the brains of calves and mice. ³⁴ By November 1995 it was clear that the sensitivity to infectivity from infected brain tissue was 1,000 times greater in the cattle assay compared with that in mice. ³⁵ Although not yet complete, interim results published in 1998 have supported this assessment. ³⁶

3.219 As suspected, the results from the comparative bioassay cast doubt over the negative findings obtained in the mouse bioassay for non-neural tissues, and indicated the need to confirm the results in the more expensive assay in cattle. So far, spleen and pooled lymph nodes are the only non-neural tissues from affected animals to have been tested in cattle, and neither has shown infectivity as yet. These results were reported as unpublished observations in 1999. ³⁷

3.220 Recent work on the production of transgenic mice could provide a useful alternative to direct assay on calves. ³⁸ It has proved possible by genetic manipulation to produce mice in which the mouse prion genes have been deletedand replaced by normal bovine genes. ³⁹ Preliminary studies indicate that BSE can be transmitted to these mice much more easily and with much shorter incubation times. In effect, the transgenic procedure has eliminated the species barrier, and provides a more sensitive bioassay.

Infectivity of bovine materials used in medicinal products and the importance of inoculation route

3.221 The risk from infectivity present in medicinal products was considered by the Southwood Working Party. They noted that 'the greatest risk . . . would be from the parenteral injection of material derived from bovine brain or lymphoid tissue'. ⁴⁰ (As described previously, it was generally accepted that the oral route was considerably less efficient than the parenteral route. ⁴¹)

3.222 In reality, different routes exist within the parenteral category - intracerebral, intraperitoneal, intramuscular, intravenous, intraspinal and subcutaneous. Experiments in 1978 looking at several of these routes found the efficiency between them to vary. Intracerebral and intraspinal were generally the most efficient, followed by intravenous, intraperitoneal and then subcutaneous. ⁴² The fact that certain medicinal products could be injected directly into the body (most commonly intramuscularly) meant that in theory they would pose a greater risk than beef products in food.

3.223 Various cattle tissues were of relevance to medicinal products, including insulin, heparin, surgical catgut sutures and serum. The consideration given to these materials prior to March 1996 is addressed in vol. 7: Medicines and Cosmetics.

Summary

3.224 Decisions made early in the BSE epidemic about the likely infectivity of cattle tissues were based upon previous experiences with other TSEs. Studies were therefore initiated to confirm these assumptions. Experiments to determine the infectivity of cattle tissues in the mouse bioassay suggested that several non-neural tissues known to be infective for scrapie were not infective for BSE. Furthermore, the pathogenesis study, which aimed to repeat studies previously performed for scrapie whereby disease was tracked around the body, showed infectivity in the ileum four months sooner than in scrapie, as well as in the dorsal root ganglia, which had not been found to be infective in scrapie. In combination, these results suggested that the mouse bioassay was possibly not sufficiently sensitive to detect BSE infectivity across the species barrier, and that BSE pathogenesis differed markedly from scrapie pathogenesis. The comparative bioassay has indeed shown that the mouse bioassay is approximately 1,000 times less sensitive than assaying the material in calves. The true infectivity of cattle tissues remains to be determined.