Volume 2: Science
3. The nature and cause of BSE
Other theories about the nature and cause of BSE
Organophosphates
Autoimmune theory
Summary

3.76 This part of the chapter examines some of the alternative hypotheses that have been proposed for the nature and cause of BSE. The Inquiry has received information on many such theories, including the use of organophosphates, an autoimmune reaction, endocrine poisoning and methyl bromide poisoning. We have considered all the possible explanations offered to us. However, our in-depth review of other possible explanations has focused on the few which appear to be supported by scientific data. These have been examined in the light of the overwhelming evidence of the accumulation of PrP^Sc in degenerating nerve cells in affected animals, which indicate that all plausible theories must accommodate a central role for PrP in TSEs.

Organophosphates

3.77 From the late 1930s, organophosphorous (OP) compounds were widely used as pesticides for the treatment of cattle for warble fly infestations. Warble fly was of major economic importance, representing a significant source of loss to the livestock, meat processing and tanning industries (because of the damage from flies burrowing out of the hides). Consequently, legislation was introduced at various times from 1936 onwards to control and eradicate this parasite. ¹ Initially, a natural insecticide derived from plants (Derris) was used, though its non-systemic action meant that it had no effect on larvae imbedded beneath the animal's hide. Pour-on OP formulations were introduced in 1960 and offered an effective systemic treatment which could be easily applied. ² Several compounds were available and licensed for use at various times from 1960 onwards, each containing different active ingredients. Phosmet was more widely used than other formulations since it had a high pesticide activity and was metabolised quickly, thus reducing the time required to withhold potentially contaminated milk from public consumption. ³

3.78 Mr Mark Purdey, an organic farmer from Somerset, was actively involved from the early 1980s in the campaign against the widespread use of OPs in agriculture. Following the identification of BSE, he drew a link between the use of OPs and BSE incidence on the basis of a survey of five farms in Devon on which cases of BSE had been observed. He noted that use of insecticides containing the OP phosmet, in conjunction with a brand of feedstuff compounded from insecticide-treated raw materials, was a common denominator on these affected farms. Mr Purdey concluded that either or both of these factors could trigger BSE in genetically susceptible cattle. Further, he noted the similarity between the pathological observations seen in BSE-affected cattle and those seen in the nerve tissue of laboratory animals exposed to known OP neurotoxic agents. These initial observations were published in a letter to Farmer's Weekly in January 1988. ⁴

3.79 By 1992 MBM was generally accepted to be the vector of BSE. Mr Purdey did not accept this but correlated BSE incidence with 'warble fly eradication zones' - areas where compulsory treatment of affected cattle was enforced by the Warble Fly (England and Wales) Order 1982. ⁵ He claimed that there was no incidence of BSE in home-reared cattle that had been fed commercial concentrates on registered organic farms where OP substances were not used, and so rejected the feedstuffs link. ⁶ Furthermore, he said that BSE-like disorders in domestic and zoo animals had resulted from exposure to OPs in flea collars and wormers, rather than from ingestion of infected feed.

3.80 Counter arguments were put forward by MAFF and others. ⁷ Mr Purdey still maintained that phosmet usage caused BSE. On the basis of literature reviews and limited trials in cattle, he put forward several potential mechanisms by which OPs might exert their effects. Initially, he proposed that the compounds were directly responsible for neurodegeneration in BSE cattle, since the anticholinesterase activity of OPs was known to cause excitotoxicity ⁸ and cell death at toxic doses. ⁹ However, as it became accepted that BSE was a prion disorder caused by the abnormal prion protein, PrP^Sc, Mr Purdey modified his hypotheses and proposed several metabolic processes involving PrP that could be affected by OPs in BSE disease progression. ¹⁰ The linchpin of his argument was that OPs can cause the conversion of PrP^C to PrP^Sc, either through interference with auxiliary proteins (molecular chaperones) involved in protein folding, or by causing a 'frameshift' of the PrP messenger RNA, the translation molecule from which proteins are assembled, resulting in the production of an abnormally folded protein. These processes were postulated to occur both in the adult cow and in the embryo in utero.

Experimental evidence for OP involvement in BSE aetiology

3.81 The first experiment to assess the susceptibility of PrP to reacting covalently (ie, by chemical reaction) with an OP compound as an initial step in Mr Purdey's hypothesis was carried out at the Medical Research Council (MRC) in 1995. OPs were not shown specifically to bind prion protein in these experiments. ¹¹ Mr Purdey refuted these results, claiming that the OP compound used (Diisopropylfluorophosphate, chosen for its high reactivity with a wide range of proteins) was not sufficiently similar to phosmet, the compound which he blamed for triggering BSE, for the results to be meaningful.

3.82 Further work was not carried out until 1998, when Whatley and others tested the effect of phosmet on neuroblastoma cells in culture. Exposure of the cells to phosmet resulted in increased levels of PrP on the cell surface, though levels of PrP mRNA did not change. ¹² Although these results did not show a direct interaction between PrP and phosmet, and therefore could not support the hypothesis that phosmet directly caused the emergence of BSE by producing PrP^Sc, they did suggest that phosmet could modify the susceptibility of cells to the prion disease agent. Whether these results represent a common effect of the class of organophosphate pesticides remains to be established, as does the relevance of these findings to the emergence of BSE in the UK.

3.83 While there have been no studies showing that conversion of PrP^C to PrP^Sc can be induced experimentally, in vitro experiments carried out in 1999 demonstrated that PrP^C, which has a structure rich in -helices, could be converted to a structure rich in -sheets by breaking a single disulphide bond under acidic conditions. ¹³ This -PrP was soluble and showed marked resistance to protease digestion. In many ways it seems an intermediate to PrP^Sc, which also has a marked -sheet structure. However, complete conversion of -PrP to the disease-producing PrP isoform has not been achieved. These results provide the possibility that chemical modification could have been a trigger for the conversion of PrP^C to PrP^Sc, though infectivity has not been demonstrated from chemically modified PrP. ¹⁴

3.84 Recent research has revealed that prion protein might play a role in the regulation of copper metabolism. ¹⁵ Prion protein was shown to bind to copper at the octapeptide repeat region, and it was found that mice lacking prion protein had a deficiency in copper at the point of contact between nerves. This deficiency might be responsible for subtle neurological changes in these mice. Copper bound to prion protein was thought to function, like superoxide dismutase, as an antioxidant (deactivating harmful oxygen free radicals). PrP^Sc might act, in part, by inhibiting this antioxidant function.

3.85 Further studies by the same researchers demonstrated that manganese can also bind to PrP. PrP within cells grown in the presence of high manganese and low copper can be converted into a protease-resistant form. However, it is not yet known if the protease-resistant prion is similar in other respects to PrP^Sc and, in particular, whether it can be infective and lead to disease in experimental mice. Mr Mark Purdey has drawn attention to evidence that suggests that areas with high prevalence of scrapie, Chronic Wasting Disease and CJD are, respectively, regions where the soil is low in copper and high in manganese. ¹⁶ It remains to be determined whether PrP bound to manganese can trigger the conversion of PrP^C to PrP^Sc.

Autoimmune theory

3.86 As described in Chapter 1, autoimmune diseases occur when an individual's immune system initiates a cytotoxic destructive reaction to a protein in its own tissues to which the immune system does not normally react. It is believed that some autoimmune diseases are caused by the immune system producing antibodies to a foreign bacterial protein, which resembles the host protein to such an extent that the antibodies then cross-react with the host protein and lead to inflammation and destruction of host tissues.

3.87 The possibility that BSE was an autoimmune disease was not considered until late in the epidemic. In 1996 a publication by Taylor et al. showed that mice lacking a functional immune system did not develop disease following exposure to scrapie via skin scarification. ¹⁷ This suggested that the cells of the immune system are required to carry the infectious agent from the peripheral site of inoculation to the brain. However, these results were interpreted differently by Professor Alan Ebringer and Professor John Pirt of King's College London, who proposed that the immune system itself was responsible for the disease, ie, that scrapie was an autoimmune disease.

3.88 To test this theory, they looked for evidence of molecular similarities between bacteria likely to be present in cattle feed and basic myelin protein, and identified a likely sequence in the bacterium Acinetobacter calcoaceticus. ¹⁸ A sequence in the bacterium Escherichia coli (E. coli) was also identified which was similar to the prion protein. Both of these bacteria were likely to be found in 'green offal' (in general, this is gastro-intestinal material including the faeces).

3.89 The autoimmune hypothesis appeared to be supported when antibodies toA. calcoaceticus were found to be increased in cattle affected by BSE when compared with normal controls. ¹⁹ Raised levels of the antibody were also found in two cases of vCJD, and in a proportion of patients with multiple sclerosis, but not in patients up to 12 months following a stroke, nor in patients with viral encephalitis or normal controls. ²⁰ Levels of E. coli antibodies were not significantly different from those in the controls.

3.90 On the basis of these results, it was postulated that BSE is an autoimmune disease similar to EAE ( allergic encephalomyelitis), caused by antibodies produced by the cattle against A. calcoaceticus. It was also suggested that infection with A. calcoaceticus might play a role in the pathogenesis of multiple sclerosis and vCJD.

Evidence to suggest BSE is not an autoimmune disease

3.91 The autoimmune hypothesis has not been widely accepted as a credible explanation for the BSE epidemic, since there are a number of observations that seem inconsistent with it. The assertion that EAE and BSE are similar is unfounded since the pathology of BSE and EAE are so different; EAE is characterised by an acute inflammatory response early in infection, whereas no such response occurs with BSE. In addition, the main feature of EAE is demyelination (degeneration of the myelin sheath which encases nerve fibres), which does not occur in BSE.

3.92 There are also several physical characteristics that cast doubt on the proposed autoimmune mechanism for BSE. Like most bacteria, A. calcoaceticus should be killed by ordinary pasteurisation treatment. ²¹ Furthermore, like salmonella andE. coli, it cannot survive any rendering procedure. If present in cattle feed it is more likely to be a contaminant of the cereal component than the MBM. However, it is a ubiquitous organism present in grass, hay and silage, and is likely to be a common contaminant of all commercial cattle feed. Thus, samples of cattle feed would be expected to have similar quantities of the bacterium both before and after the ruminant feed ban, in which case removal of MBM from cattle rations would not be expected to halt the epidemic.

3.93 It is also suggested that the relevant antigen in A. calcoaceticus, which shows sequence similarity to bovine myelin protein, is a sequence present in the enzyme 4-carboxy-muconolactone decarboxylase. ²² Independent research shows that the bovine myelin protein is readily cleaved by proteolytic enzymes. ²³ TheA. calcoaceticus antigen is therefore likely to be destroyed by digestion in the cattle gut.

3.94 Professors Ebringer and Pirt have claimed to demonstrate antibodies toA. calcoaceticus in BSE-affected cattle. ²⁴ However, given that the A. calcoaceticus antigen and bovine myelin protein have been shown to exhibit molecular mimicry, it is possible, in BSE-affected cattle whose neural proteins are being broken down, that it is actually antibodies to degrading neural tissue rather than toA. calcoaceticus that are being identified.

3.95 The autoimmune theory of BSE also seems incompatible with what is known about the prion hypothesis for TSEs. Specific mutation of the prion gene causes familial CJD, inherited as a classic autosomal dominant trait. ²⁵ Transgenic mice engineered to have a mutation of the prion gene similar to that for familial CJD develop spongiform encephalopathy, which can then be transmitted by infection to other mice. ²⁶

3.96 CJD belongs to a group of human neurodegenerative diseases in which the destruction of nerve cells is associated with the slow accumulation of insoluble aggregates of protein within these cells, without an inflammatory reaction. Several are genetically determined, like familial CJD, and are due to specific gene mutations (see paragraphs 2.21-2.27). Aggregations of abnormal proteins in nerve cells also cause a number of similar diseases including Alzheimer's and Parkinson's disease (see paragraphs 2.39-2.47). As the same basic mechanism seems to occur in TSEs as in these other neurodegenerative conditions, it seems likely that the prion hypothesis provides a more acceptable explanation for the presence of insoluble protein aggregates than the autoimmune hypothesis.

3.97 Perhaps more damaging to the autoimmune theory is the observation that mice lacking a functional immune system cannot be infected by mouse-adapted BSE inoculated peripherally, though they do develop disease if the mouse-adapted BSE agent is inoculated directly into the brain. ²⁷ The mouse-adapted disease cannot therefore be an autoimmune disease as it can be transmitted to mice lacking an immune system. It is noted that intracerebral transmission direct from brain tissue of a BSE-affected cow into immunodeficient mice does not occur. While this might appear consistent with the auto-immune hypothesis, work by Fraser and others suggests that this is because of the species barrier and because the infective agent has not replicated in the murine lymphoreticular system (LRS). ²⁸ This experiment indicates that in TSEs the LRS has to be intact in order to replicate and transport the infective agent.

3.98 Professors Ebringer and Pirt explained the results of transmission experiments in mice and other species using BSE-infected brain tissue as the production of EAE by an immune response to bovine myelin protein, and not to the BSE prion protein. However, it is known that it is difficult to induce EAE by this means in experimental animals without using adjuvants. Moreover, BSE has been transmitted using non-neural tissue such as ileum.

Summary

3.99 Theories which fail to acknowledge a place for PrP^Sc in the causation of TSEs remain unconvincing. Thus a direct toxic effect of OPs on nerve cells seems an unlikely explanation for these diseases. Experimental conversion of PrP^C to PrP^Sc by OPs has not been achieved experimentally, although there is some evidence that OPs may increase the expression of PrP^C on the cell membranes of neuroblastoma cells. Recent studies have shown partial conversion of PrP^C to a -PrP precursor in vitro. The molecular structure of -PrP is rich in beta-sheets, and shows resistance to protease digestion but, unlike PrP^Sc, is soluble. In separate studies PrP has been shown to have antioxidant properties when bound to copper. Replacement of copper by manganese converts the PrP to a protease-resistant form. Neither this form of PrP nor the -PrP precursor have yet been shown to cause disease in experimental animals. However, the finding that PrP can be modified in vitro by appropriate chemistry raises the possibility that similar reactions might be induced by environmental factors in vivo.

3.100 The evidence described in paragraphs 3.91-3.98 suggests that the autoimmune theory is based on a number of false premises. The theory is inconsistent with many of the observed features of BSE and the BSE epidemic, and therefore does not challenge our current understanding of the cause of the disease.