Veterinary Immunology and lmmunopathology 55 (1996) 127-139

veterinary ilNlWnOlogy

and immunopathobgy

Mimicry of a 21.5 kJIa myelin basic protein peptide by a Maedi Visna virus polymerase peptide does

not contribute to the pathogenesis of encephalitis in sheep

J.M. Davies a* * , N.J. Watt b, S. Torsteinsdottir ‘, P.R. Carnegie a a Biotechnology Programme, School of Biological and Enuironmental Sciences, Murdoch University,

Murdoch 6150 Perth, WA, Australia b Department of Veterinary Pathology, University of Edinburgh, Edinburgh, Scotland, UK

’ Institute for Experimental Pathology. University of Iceland, Reykjavik, Iceland

Accepted 18 March 19%

Abstract

Epitope mimicry is the theory that an infectious agent such as a virus causes pathological effects via mimicry of host proteins and thus elicits a cross-reactive immune response to host tissues. Weise and Carnegie (1988) found a region of sequence similarity between the pal gene of

the Maedi Visna virus (MVV), which induces demyelinating encephalitis in sheep, and myelin basic protein (MBP), which is known to induce experimental allergic encephalitis (EAE) in laboratory animals. In this study, cross-reactions between sera raised in sheep against synthetic

peptides of MVV (TGKIPWILLPGR) and 21.5 kDa MBP (SGKVPWLKPGR) were demonstrated using enzyme-linked immunosorbant assay (ELISA) and thin layer chromatography (TLC) immunoprobing. The antibody responses of MVV-infected sheep were investigated using ELISA against the peptides, and MBP protein, immunoprobing of the peptides on TLC plates and Western blotting against MBP. Slight significant reactions to the 21.5 kDa MBP peptide (P < 0.001) and to a lesser extent sheep MBP (P < 0.004) were detected in ELISA. The MBP peptide evoked stronger responses from more sera than the MVV peptide on immunoprobed TLC

Abbreviations: CNS central nervous system; MBP myelin basic protein; MVV Maedi Visna virus; TLC thin layer chromatography

* Author to whom correspondence should be addressed at: Department of Biochemistry and Molecular Biology, Monash University, Clayton, Vie. 3168, Australia.

0165-2427/96/$15.00 Published by Elsevier Science B.V. PII SO165-2427(96)05619-X

128 J.M. Dauies et al./Veterinary Immunology and Immutwpathology 55 (1996) 127-139

plates. On the Western blots, eight of the 23 sheep with Visna had serum reactivity to MBP. This slight reaction to MBP in MVV-infected sheep is of interest because of the immune responses to MBP evident in multiple sclerosis and EAE, but its relevance in Visna is limited since no correlation with disease severity was observed. The cell-mediated immune responses of MVV-in- fected sheep against similar peptides was assessed. The peptides did not stimulate proliferation of peripheral blood lymphocytes of MVV-infected sheep. Since the MVV peptide was not recognised by antibodies or T lymphocytes from MVV-infected and encephalic sheep, it was concluded that epitope mimicry of this 2 1.5 kDa MBP peptide by the similar MVV pal peptide was not contributing to the immunopathogensis of Visna. The slight antibody response to MBP and the MBP peptide can be attributed to by-stander effects of the immunopathology of MVV-induced encephalitis.

1. Introduction

The encephalitic manifestation of Maedi Visna virus (MVV) in sheep resembles the pathology observed in the central nervous system (CNS) of multiple sclerosis (MS) patients (Haase, 1986), and hence it provides a natural animal model for some aspects of MS. In sheep with Visna there is a low level of virus antigen expression within cells in the CNS (Georgsson et al., 1989), but the immune response is thought to be directed against white matter in the CNS, in particular glial cells (Georgsson et al., 1990; Haase, 1990). The correlation between severity of Visna and the presence of the virus (Stowring et al., 1985) may not necessarily indicate a direct viral effect.

Pre-vaccination with MVV prior to infection worsens the course of the disease (Nathanson et al., 1981), and MVV-infected sheep with Visna lesions have a different antibody profile on Western blots than sheep without CNS pathology (Houwers and Nauta, 1989). Antibody and cell-mediated (Torsteinsdottir et al., 1992) immunity have roles in the development of CNS lesions.

It is possible that non-viral antigens, altered self-antigens due to the presence of the virus, or epitope mimicry (Oldstone, 1987) of host antigens may be responsible for the lymphocyte infiltration and inflammation. Since the level of viral antigen expression is low, the target antigen may conceivably be one that is mimicked on the host tissue and expressed at a higher level than the viral antigen. This would account for the develop- ment of disease in the CNS despite a low level of virus antigen expression.

In MS and experimental allergic encephalitis (EAE), both antibodies and T cells with specificity for myelin basic protein (MBP) and other myelin antigens play a role in the CNS pathology (Oksenberg et al., 1993). Since MBP has long been known to induce encephalitis in laboratory animals and is a key antigen in MS, it is possible that epitope mimicry of MBP by the MVV contributes to the pathology of encephalitis.

Previous attempts to find antibody responses to MBP in MVV-infected sheep have failed (Dr. M. Brahic, personal communication, 1990). Experimental infection of sheep with MVV produced no detectable T cell response to myelin antigens early in the disease (Panitch et al., 1976). However, injection of sheep with MBP in complete Freund’s adjuvant induces an EAE-type pathology that histologically resembles lesions in Visna and in which the sheep had slight T cell responses to MBP (Panitch et al., 1976). Weise and Carnegie (1988) have identified a region of the 2 1.5 kDa isoform of

J.M. Davies et al./Veterimry Immunology and Immunopathology 5.5 (1996) 127-139 129

MBP (SGKVPWLKPGR) that is expressed in adult sheep and which shares sequence similarity with the pal protein of the MVV (TGKIPWILLPGR). The amino acid differences in the two peptides are predicted to be conservative substitutions (Geysen et al., 1988). Investigation of the hypothesis that epitope mimicry of MBP by the MVV peptide is involved in the pathogenesis of the encephalitic condition in MVV-infected sheep is reported here.

The immunogenicity of the similar peptides from MVV and 21.5 kDa MBP was examined in sheep, and the ability of the antibodies to the MVV peptide to cross-react with MBP was assessed. Both the serological and cell-mediated responses of MVV-in- fected sheep were assessed in order to examine epitope mimicry of MBP by MVV.

2. Methods

2.1. Peptides

The peptides from the pol gene of the MVV strain K15 14 (amino acids 521-532) (Sonigo et al., 1985) and f rom the 21.5 kDa isoform of human MBP (amino acids 56-66) (Kamholz et al., 1986) were identified from the Protein identification Resourse data base by the FASTP program (Weise and Carnegie, 1988). Compositional analysis of peptides from sheep MBP confirmed that the sequences of human and sheep MBP are highly conserved in the region of similarity (Carnegie and Dowse, 1984). These peptides were synthesised by F-mot chemistry on a RAMPS system (DuPont, CA, USA) according to the manufacturer’s specifications, with modifications in the coupling procedure for certain amino acids (Hudson, 1988) and the cleavage mixture (King et al., 1990).

The purity of synthesised peptides was assessed using reverse phase high perfor- mance liquid chromatography in a manner similar to that described by Potter and Lewis (1986) using a gradient of isopropanol from 5 to 60% in 0.1% trifluoroacetic acid. Samples of the synthetic peptides (20 Fg) were analyzed on cellulose thin layer chromatography (TLC) plates (Merck, cat number 5577) by ascending capillary action in the presence of solvent (butanol:distilled water:acetic acid:pyridine 12:8:0.1: 10 (BWAP)). Peptides were purified by preparative paper chromatography according to the procedure of Dunkley and Carnegie (1974) using BWAP. TLC and paper chromatographs were dipped in 0.2% w/v ninhydrin and developed at 80°C. Hydrolysates of the peptides were prepared and analyzed for amino acid composition by Dr. E. Rasins at the Botany Department, University of Western Australia.

The peptide SPLLGCIGSTCAE from the S protein of the mouse hepatitis virus (846-858) (MHV peptide) was kindly supplied by Dr. M. Koolen (Utrecht University, The Netherlands) for use as a negative control peptide.

Purified sheep MBP, highly enriched in the 21.5 kDa isoform but also containing the 18.5 kDa isoform, was supplied by C. Dowse (Biotechnology, Murdoch University). Despite the use of several separation methods, it has not been possible to obtain 2 1.5 kDa MBP free of 18.5 kDa MBP (P.R. Carnegie, personal communication, 1989).

130 J.M. Davies et al./ Veterinary Immunology and Immwwpathology 55 (1996) 127-139

2.2. Sheep serum and CSF samples

The MVV and MBP peptides were coupled to keyhole limpet haemocyanin (KLH) in a 1:1 weigh& ratio using 2% glutaraldehyde (Reichlin, 1980). Hyper-immune sera were raised in 18-month-old female Suffolk-Merino sheep with a primary injection of 500 Fg peptide/ pg KLH in 1 ml and three subsequent injections of 0.5 ml of the suspension at weekly intervals using Freund’s complete (for the first injection) or incomplete adjuvant.

Sera were obtained from 13 Scottish sheep naturally infected with MVV. Ten serum and four cerebrospinal fluid (CSF) samples were obtained from Icelandic sheep with Visna that had been infected with MVV strain K15 14, K1552 or K1772.

2.3. Antibody immunoassays

The protocol used for peptide ELISA has been described previously (Davies et al., 1992). Griener F high-binding flat-bottom plates (Griener, Austria) were used with a total of 0.5 p,g of antigen. The stock solutions of the purified peptides were made at 4 mg ml -’ in 40% dimethylsulphoxide in tris buffered saline (TBS). Sera were tested for response to the peptides at dilutions from l/40 to l/5120. Peroxidase conjugated rabbit anti-goat antibody (Dakopatts, Denmark) was used at l/2000 to detect antibody recognition of the antigen with 2,2-azinobis(2-ethylbenzthiazoline) sulphonic acid as a chromogen for development. Differences in group mean responses were assessed by Student’s f-test.

A 200 ~1 sample of sheep MBP, enriched for 21.5 kDa, 1 mg ml- ’ was elec- trophoresed across a 12 cm 10% SDS-polyacrylamide slab gel and blotted onto nitrocellulose. Blocking, primary antibody and secondary antibody incubations were preformed for 1 h, 4 h and 2 h, respectively, using the buffers described for ELISA. The sheep sera or CSF were incubated at l/20 or l/200, and as a positive control sheep anti-human MBP antibody was incubated l/1000. A biotinylated multispecific anti- goat/mouse/rabbit antibody (Dakopatts, Denmark) was used at l/4000 and followed by a 1 h incubation with peroxidase-bound streptavidin (Zymed Laboratories, USA) at l/8000. Blots were developed with a combination of chloronaphthol and diaminobenzi- dine as described by Young (1989).

The method for immunoprobing peptides on TLC plates was an adaptation of that described by Dowse et al. (1987). The same reagents, times and development methods were used as for Western blotting. The experimental anti-peptide antisera was used at a dilution of l/200 and the test sheep sera and CSF were used at a dilution of l/50.

3. Results

3.1. Experimental cross-reaction of anti-MW peptide sera

Antibodies were raised against the MVV peptide and the 21.5 kDa MBP peptide in sheep and used in an ELISA assay to determine cross-reactivity of the sera to MBP

J.M. Davies et al./Veterinary Immunology and Immunopathology 55 (1996) 127-139 131

*.51 coating antigen

n 21.5kDa MBP peptide l MW peptide

E C unrelated peptide 8 1.0

E $ 8 9 0.5

Dilution of anti-MVV peptide serum

Fig. I. Response of antisera raised in sheep against the MVV peptide. The data are presented as an average of

two replicates. The whole MBP antigen is the 21.5 kDa-enriched sheep MBP (visualised in Fig. 5).

peptide and MBP protein. Fig. 1 depicts the titration of the anti-MVV peptide serum response to the MVV peptide, MBP peptide, MBP and an unrelated control peptide. There was no recognition of an unrelated control peptide, nor did the pre-immune serum

respond to the test antigens. The antiserum raised against the MVV peptide was able to react with the MBP peptide and sheep MBP protein in the ELISA (Fig. 1). Sheep MBP

was not recognised by the anti-peptide sera on Western blots. Sera raised against the MVV and MBP peptides cross-reacted with the MVV peptide

and the similar MBP peptide, but not with an unrelated peptide on TLC plates (Fig. 2al. Normal sheep serum did not react to the MVV or MBP peptides. Combined with the

ELISA data, for which purified peptide was used, the TLC data show that sheep are able to recognise these peptides as antigens following immunization with peptide-KLH conjugates, and that the peptides are sufficiently similar to induce cross-reactive antibody responses.

-3.2. Antibody response of MVV-infected sheep to the peptides in ELBA

The sera of 23 MVV-infected sheep were used in an ELISA to assess their reactivity

to the MVV peptide, MBP peptide and MBP in comparison with the unrelated MHV peptide. The responses of sera from infected sheep to each antigen were compared with

the responses of a group of uninfected sheep (Fig. 3). Sera from the infected group showed slight but significantly higher responses to the MBP peptide than sera from the uninfected group. The sero-negative group also showed low responses to the negative control MHV peptide. The differences in the groups’ responses to the MBP peptide (P < 0.001) and MBP (P < 0.004) but not the MVV peptide were significant. No

132 J.M. Davies er al,/ Vererinaty Immunology and Immunopathology 55 (1996) 127-139

(4 FR

McG Me0

xc

dyes H M V H M V

1 2

UN v>

M>

<M

m MN3 r&o

xc

N-A! M-Y M M V dve% 1 2 3 4

Fig. 2. Reaction of sheep sera on TLC of MVV and similar 21.52 kDa peptides. FR, Fuchsin Red; H, mouse

hepatitis virus peptide; M, 21.5 kDa MBP peptide; MeG, methyl green; MeO, methyl orange; V, MVV

peptide; XC, xylene cyanol. (a) Lane I; anti-MVV peptide serum. Lane 2; anti-MBP peptide serum. (b) Lane

1; MVV-uninfected sheep serum. Lane 2; MVV-infected cerebrospinal fluid. Lane 3; MVV-infected sheep

serum. Lane 4; peptide stained with ninhydrin.

differences were observed between MVV-infected sheep with neurological symptoms and those with only respiratory symptoms, but the small numbers (n = 4) in the asymptomatic subgroup precludes a definitive relationship.

J.M. Davies et al./ Veterinary Immunology and Immunopathology 55 (1996) 127-139 133

0- 1

MW infection + - + - + - +

coating antigen 21.5kDa MBP MW MHV 21.5kDa enriched peptide peptide peptide sheep MBP

statistical difference p<O.OOl - p<o.o04

Fig. 3. Distribution of serum antibody responses of MVV-infected and -unifected sheep to MVV and MBP

antigens. The shaded boxes encompass the responses of 50% of the sheep in each group at a dilution of l/80,

with the mean marked across the shaded section. The points of the two highest and lowest responses are given,

and the extended bars mark the range of 95% of responses.

3.3. Antibody responses of MVV-infected sheep to peptides on TLC plates

The antibody responses from serum and CSF of MVV-infected sheep to the MBP and MVV peptide on TLC plates was assessed (Fig. 2b). Some sera from MVV-negative sheep had slight reactions with the MBP peptide on the immunoprobed TLC plates, and these have been recorded in Table 1. The responses of all infected and uninfected sheep to each peptide were graded from - to + + + (Table 1). The data show that the MBP peptide was bound more strongly than the MVV peptide by antibodies from MVV-in- fected sheep sera. Both peptides were present in similar amounts (20 p,g); hence the differences in reaction are not due to antigen concentration effects. On TLC, more

Table 1 Number of MVV-infected vs. -uninfected sheep with antibody response to the 21.5 kDa MBP and MVV

peptides on an immunoprobed TLC plate

MVV status TLC reaction level 2 1.5 kDa MBP peptide MVV peptide

Infected +++ 4 1

++ 2 3

+ 11 12

* 2 2

- 2 4

Uninfected + 0 0

* 2 0

_ 4 6

134 J.M. Davies et al,/ Veterinary Immunology and Immunopathology 55 (1996) 127-139

(n = 4) MVV-infected sheep had strong responses to the MBP peptide than the MVV peptide (n = 1).

3.4. Antibody responses of MW-infected sheep to MBP

Titration of the MVV sheep sera against the peptide antigens and MBP demonstrated that four out of 13 MVV-infected sheep with Visna reacted with MBP but not the MVV peptide in the ELISA. One example is shown in Fig. 4. This finding supports the conclusion that a response to the MVV peptide did not induce a cross-reactive antibody response to the MBP peptide or MBP because there is no response to the MVV peptide in these sheep.

Several sheep with Visna reacted to 18.5 kDa and 21.5 kDa MBP on Western blot, but this reaction was not strong or widespread and reproducibility was poor (Fig. 5). The sera (n = 23) and CSF (n = 4) samples from the MVV-infected sheep were tested at l/20 and l/200 on up to six occasions each. In total, antibodies from eight MVV-in- fected sheep reacted with MBP on Western blot and a further three had slight reactions.

The clinical status of the sheep was compared to the responses to the peptides on TLC and to MBP on Western blot. There appeared to be no relationship between the severity of Visna and the response to the MBP or MVV peptide. The MBP peptide was recognised to a greater extent than the MVV peptide. If mimicry was inducing a cross-reactive response to the MBP peptide in MVV-infected sheep, then the MVV peptide ought to elicit the greatest response. No correlation between the severity of the encephalitic manifestation of Visna and the response to MBP on Western blot was observed.

1

Fig. 4. Titration of serum from a MVV-infected sheep with CNS lesions. The coating antigens were the 21.5 kDa MBP peptide ( W), MVV peptide (0). MBP (0) and the negative control MHV peptide (0). The serum

dilutions are shown on the x-axis and the absorbance at 405 nm is on the y-axis.

J.M. Davies et al./ Veterinary Immunology and lmmwwpathology 55 (1996) 127-139 135

21.5kDa

18.5kDa

lane number 12345

Fig. 5. Antibody response of sheep with Visna to MBP on Western blots. Approximately 6 p,g per lane of

purified sheep MBP was loaded onto a 10% SDS-PAGE gel, electrophoresed and transfered to nitrocellulose.

Lane 1, strip stained with Ponceau S; Lane 2, anti-human MBP sera at l/ IOOO; Lane 3, MVV-negative sheep

sera at l/200; Lane 4, CSF of Visna sheep at l/200; Lane 5, Visna sheep serum at l/200.

3.5. T cell responses of MW-infected sheep

Initially, lymphocyte proliferation in response to the MVV virus antigen could not be detected in the flock of experimentally infected Icelandic sheep. Five sheep infected with MVV were then challenged with the virus 4 weeks prior to demonstrating lymphocyte proliferation in response to virus antigen. At the same time responses to the MVV peptide, the MBP peptide and a negative control peptide of sequence KNPRRPSC were tested at 15.6-2000 ng ml-‘. No proliferation of lymphocytes from these sheep was detected in response to these antigens.

Testing the cell-mediated responses of Scottish sheep naturally infected with MVV did not reveal any recognition of either the MVV and MBP peptides or sheep MBP. Recently, evidence of T cell responses to other MVV antigens from infected sheep in this flock have been published (Reybum et al., 1992).

4. Discussion

The impetus to study this example of peptide similarity was because the target host protein MBP is important in the development of MS and EAE which Visna in sheep

136 J.M. Davies et al./ Veterinary Immunology and Immutwpathalogy 55 (1996) 127-139

resembles. It was considered that the MVV peptide with sequence similarity to 21.5 kDa MBP, which is an isoform expressed in adult sheep (Carnegie and Dowse, 1984), may play a role in the immunopathology of encephalitis in MVV-infected sheep.

This study demonstrates that it is possible to induce antibodies against the MVV peptide that could recognise the similar MBP peptide and MBP (Fig. 1). The clinical relevance of this finding was sought by determining whether these peptides were recognised by antibodies and T cells from MVV-infected sheep.

Finding a response to MBP in MVV-infected sheep is in contrast with other research to date in which no antibody response to MBP has been found (M. Brahic, personal communication, 1990). Since the response to MBP was slight in ELISA and restricted to a few sheep, it is not surprising that responses to MBP have not been reported previously in MVV-infected sheep. The low level of response may be because a large proportion of the immunoglobulin reactive to MBP is bound to the antigen in situ and thus cannot be assayed in serum or CSF samples. These anti-MBP antibodies could be pathogenic given the role of antibodies to MBP and other myelin antigens in MS (Scolding and Compston, 1991). Antibodies were present in the brain of 12 of 17 MS patients at the time of autopsy (Bernard et al., 19811, and B cells producing antibodies reactive to MBP have been shown to persist in the CSF of MS patients (Link et al., 1990).

If epitope mimicry was involved in the pathogenesis of Visna, one would expect a strong response to the MVV peptide. The slight response to the MBP peptide seen in the Visna sheep is most likely due to a by-stander effect (Duke, 1989). Anti-MBP responses in some sheep have previously been reported (Carnegie et al., 1983). The encephalitic process itself could be sufficient for the development of antibodies to MBP in MVV-in- fected sheep. At least one of the sheep that had an antibody response to MBP in ELISA had primary demyelinating plaques when examined at necropsy. The pathology of the CNS lesions in this sheep (No. 19117) has been presented previously (Georgsson et al., 1982). Reactivity to MBP could be due to subclinical CNS pathology since 10% of the sheep actually have histological signs of Visna in the CNS at necropsy (Dr. G. Georgsson, personal communication, 1990). It is reasonable to assume that the antibody responses to MBP and the MBP peptide could have arisen as by-stander effects rather than be instigating the pathogenesis since there was no pattern of response associated with severity of disease. It was concluded that the presence of anti-MBP antibodies in MVV-infected sheep is most probably unrelated to the patbogenesis of disease.

4.1. Assessing cell-mediated responses of W-infected sheep

Although there was difficulty in demonstrating responses of MVV-infected sheep to MVV antigen preparations, sheep which showed good T cell responses to MVV were eventually identified or stimulated by new immunizations with the virus (Reybum et al., 1992; Torsteinsdottir et al., 1992). Ideally the CNS T cells, or lines of them, taken at the time of initial histological signs of white matter lesions would need to be tested to determine if there are T cells of specificity for the MVV peptide and the MBP peptide. Given the time-consuming and extensive investigations undertaken to study recognition of MBP epitopes in MS using T cell lines (Vandenbark et al., 1989) or analysis of T cell

J.M. Davies et al./Veterinury Immunology and Immunopathnlogy 55 (1996) 127-139 137

receptor transcripts in the demyelination plaques in MS brains (Oksenberg et al., 19931, it is unlikely that such efforts can be made in MVV-infected sheep using CNS T cells from early stages of the disease. Early diagnosis of Visna in sheep would be difficult and hence the stage at which epitope mimicry induces disease could be missed. Given the heterogeneity of the al3 T cell receptor repertoire in MS (Richert et al., 19911, it is unlikely that proliferation of one particular clone within the CNS actually initiated the pathological process or is the stimulus by which the endothelial blood brain barrier was crossed (Raine et al., 1990). In Visna it would be difficult to determine which was the initial clone or specificity of T cell that crossed the blood brain barrier first and for which antigen it had specificity: virus or CNS antigen. Cells within the CNS have been shown to possess antigen-presenting functions in human brain (Lee et al., 1990) and the vascular endothelial of MVV-infected sheep has been shown to be positive for MHC class II antigens (Torsteinsdottir et al., 1992). Thus it is possible that the pathology seen in Visna CNS lesions is one of a cell-mediated immune response such as is seen in MS (Cross and Raine, 1991).

There was no evidence of a T cell response to sheep MBP in the Scottish sheep infected with MVV, which confirms the data of Panitch et al. (1976) who also found no T cell responses to MBP in experimentally infected sheep. This lack of response may be because the primary clinical manifestations of Scottish sheep are arthritis and mastitis (Watt et al., 1992). Little was known about the clinical status of the CNS of the sheep from Scotland that were used in the T cell testing. The fact that there was no T cell recognition of the MBP antigens in this study implies that it is not likely to be the triggering antigen of Visna encephalitis. Epitope mimicry is unlikely to be a factor in involved in pathogenesis of Visna in MVV-infected sheep.

The demonstration of cross-reactivity of antibodies raised against the MVV peptide with the MBP peptide and MBP revealed nothing about the clinical status of antibodies to these peptides. This study refutes the belief of Dyrberg and Oldstone (1986) that in order to examine the significance of sequence similarities, experimental testing of the immunogenicity and cross-reactivity of synthetic peptides is necessary. From the results of this paper it can be recommended that to identify cases of mimicry or find the significance of sequence similarity, it would be best to examine serological and T cell responses of clinically affected individuals. As shown by Wucherpfennig and Stro- minger (19951, many examples of sequence similarities that can be identified by database searches for similarity with key epitopes have little physiological significance.

Acknowledgements

Funding for this project was provided by the National Multiple Sclerosis Society of Australia and the National Health and Medical Research Council of Australia. Work using the MVV-infected sheep sera was carried out at the Central Veterinary Institute (Lelystad, The N th 1 d ) e er an s and at the Immunology Department of the Veterinary School, Utrecht University (The Netherlands). The arrangements for this collaboration were made by Dr. Dirk Houwers, Utrecht University. We thank Dr. Stuart Bradley of

138 J.M. Davies et al./ Veterinary Immunology and Immunopathology 55 (19961127-139

the School of Biological and Environmental Sciences, Murdoch University, for statisti- cal analysis of the ELISA data.

References

Bernard, C.C.A., Randell, V.B., Horvath, L.B., Carnegie, P.R. and Mackay, I.R., 1981. Antibody to myelin

basic protein in extracts of multiple sclerosis brain. Immunology, 43: 447-457.

Carnegie, P.R. and Dowse, C.A., 1984. Partial characterization of 21SK myelin basic protein from sheep

brain. Science, 223: 936-938.

Carnegie, P.R., Lim, C.F. and Bernard, C.C.A., 1983. Role of viruses in the induction of antibodies to myelin

basic protein. In: Molecular Aspects of Neurological Disorders. Academic Press, Sydney, Australia, pp

257-264.

Cross, A.H. and Raine, C.S., 1991. Central nervous system endothelial cell-polymorphonuclear cell interac-

tions during autoimmune demyelination. Am. J. Pathol., 139: 1401- 1409.

Davies, J.M., Sonoda, S., Yashiki, S., Osame, M. and Carnegie, P.R., 1992. Mimicry between HTLV-I and

myelin basic protein: No response in HTLV-I-associated myelopathy patients. J. Neuroimmunol., 41:

239-244.

Dowse, C.A., Carnegie, P.R., Kemp, B.E., Sheng, H.Z., Grgacic, E.V. and Bernard, C.C.A., 1987. Rapid

characterization of protein epitopes recognized by monoclonal antibodies using direct probing on thin-layer

and paper chromatograms. J. Immunol. Methods, 97: 229-235.

Duke, R.C., 1989. Self recognition by T cells. I. Bystander killing of target cells bearing syngeneic MHC

antigens. J. Exp. Med., 170: 59-71s.

Dunkley, P.R. and Carnegie, P.R., 1974. Amino acid sequence of the smaller basic protein from rat brain

myelin. B&hem. J., 141: 243-255.

Dyrberg, T. and Oldstone, M.B.A., 1986. Peptides as probes to study molecular mimicry and virus-induced

autoimmunity. Curr. Top. Microbial. Immunol., 130: 25-37.

Georgsson, G.. Martin, J.R., Klein, J., Palsson, A., Nathanson, N. and Petursson, G., 1982. Primary

demyelination in Visna: An ultrastructural study of Icelandic sheep with clinical signs following experi-

mental infection. Acta Neuropathol., 57: 171-178.

Georgsson, G., Houwers, D.J., Palsson, P.A. and Petursson, G., 1989. Expression of viral antigens in the

central nervous system of visna-infected sheep: An immunohistochemical study on experimental visna

induced by virus strains of increased neurovirulence. Acta Neuropathol., 77: 199-306.

Georgsson, G., Palsson, P.A. and Petursson, G., 1990. Some comparative aspects of Visna and AIDS. In: P.

Racz, A. Hasse and J. Gluckman (Editors), Modem Pathology of AIDS and Other Retroviral Infections.

Karger, Basel, pp. 82-98.

Geysen, H.M., Mason, T.J. and Rodda, S.J., 1988. Cognitive features of continuous antigenic determinants. J.

Mol. Recog., 1: 32-41.

Haase, A.T., 1986. The pathogenesis of slow virus infections: Molecular analyses. J. Inf. Dis., 153: 44-447.

Haase, A.T., 1990. Analysis of lentivirus infections by in situ hybridization: Introduction to lentiviruses. In: P.

Racz, A. Hasse and J. Gluckman (Editors), Modem Pathology of AIDS and Other Retroviral Infections.

Karger, Basel, pp. 1-16.

Houwers, D.J. and Nauta, I.M., 1989. Immunoblot analysis of the antibody response to ovine lentivirus

infections. Vet. Microbial., 19: 127- 139.

Hudson, D., 1988. Methodological implications of simultaneous solid-phase peptide synthesis. 1. I. Compari-

son of different coupling procedures. J. Org. Chem., 53: 617-624.

Kamholz, J., de Ferra, F., Puckett, C. and Lazzarini, R., 1986. Identification of three forms of human myelin

basic protein by cDNA cloning. Neurobiology, 83: 4962-4966.

King, D.S., Fields, C.G. and Fields, G.B., 19%. A cleavage method which minimizes side reactions following

Fmoc solid phase peptide synthesis. Int. J. Pept. Protein Res., 36: 225-266.

Lee, S.C., Moore, G.R.W., Golenwsky, G. and Raine, C.S., 19%. Multiple sclerosis: A role for astroglia in active demyelination suggested by class II MHC expression and ultrastructural study. J. Neuropathol. Exp.

Neural., 49: 122- 136.

J.M. Davies et al./Veterinary Immunology and Immunopathology 55 11996) 127-139 139

Link, H., Baig, S., Olsson, 0.. Yu-Ping, J., Hojeberg, B. and Olsson, T., 1990. Persistent anti-myelin basic

protein IgG antibody response in multiple sclerosis cerebrospinal fluid. J. Neuroimmunol., 28: 237-248.

Nathanson, N., Martin, J.R., Georgsson, G., Palsson, P.A., Lutley, R.E. and Petursson, G., 1981. The effect of

post-infection immunization on the severity of experimental visna. J. Comp. Pathol., 91: 18S- 191.

Oksenberg, J.R., Pazaram, M.A., Begovich, A.B., Mitchell, D., Erlich, H.A., Murray, RX, Shimonkevitz, R.,

She&t, M., Rothbard, R., Bernard, C.C.A. and Steinman, L., 1993. Selection for T-cell receptor

Vh- Db-Jb gene rearrangements with specificity for a myelin basic protein peptide in brain lesions of

multiple sclerosis. Nature, 362: 68-70.

Oldstone, M.B.A., 1987. Molecular mimicry and autoimmune disease. Cell, 50: 819-820.

Panitch, H., Petursson, G., Georgsson, G., Palsson, P.A. and Nathanson, N., 1976. Pathogenesis of visna. III.

Immune responses to central nervous system antigens in experimental allergic encephalomyelitis and

Visna. Lab. Invest., 35: 452-460.

Potter, R.L. and Lewis, R.V., 1986. Reversed-phase chromatography of proteins and nucleic acids: Practical

considerations. High-Performance Liq. Chromatogr., 4: l-45.

Raine, C.S., Cannella, B., Duijvestijn, A.M. and Cross, A.H., 1990. Homing to central nervous system

vasculature by antigen-specific lymphocytes. II. Lymphocyte/endotbeliaI cell adhesion during the initial

stages of autoimmune demyelination. Lab. Invest.. 63: 476-489.

Reichlin, M., 1980. Use of glutaraldehyde as a coupling agent for proteins and peptides. Methods Enzymol.,

70: 159-165.

Reybum, H.T., Roy, D.J., Blacklaws, B.A., Sargan, D.R., Watt, N.J. and McConnell, I., 1992. Characteristics

of the T cell-mediated immune response to Maedi-Visna virus. Virology, 191: 1009- 1012.

Richer-t, J.R., Robinson, E.D., Johnson, A.H., Bergman, C.A., Dragovic, L.J., Reidsmoen, N.L., Katovich, C.

and Hurley, C., 1991. Heterogeneity of the T-cell receptor b gene rearrangements generated in myelin

basic protein-specific T-cell clones isolated from a patient with multiple sclerosis. Ann. Neurol., 29:

299-306.

Scolding, N.J. and Compston, D.A.%, 1991. Oligodendrocyte-macrophage interactions in vitro triggered by

specific antibodies. Immunology, 72: 127- 132.

Sonigo, P., Alizon, M., Staskus, K., Klatzmann, D., Cole, S., Danos, O., Retzel, E., Toillais, P., Haase. A. and

Wain-Hobson, S., 1985. Nucleotide sequence of the Visna lentivirus: Relationship to the AIDS virus. Cell,

42: 369-382.

Stowing, L., Hasse, A.T., Petursson, G., Georgsson, G., Palsson, P., Lutley, R., Roos, R. and Szuchet, S.,

1985. Detection of Visna virus antigens and RNA in glial cells in foci of demyelination. Virology, 141:

311-318.

Torsteinsdottir, S., Georgsson, G., Gisladottir, E., Rafnar, B.. Palsson, P.A. and Petursson, G., 1992.

Pathogenesis of central nervous system lesions in visna: Cell-mediated immunity and lymphocyte subsets

in blood, brain and cerebrospinal fluid. J. Neuroimmunol., 41: 149- 158.

Vandenbark. A.A., Chou, Y.K., Bourdette, D., Whitham, R., Chilgren, J., Cbou, C.-H.J., Konat, G., Hashim,

G., Vainiene, M. and Offner, H., 1989. Human T lymphocyte response to myelin basic protein: Selection

of T lymphocyte lines from MBP-responsive donors. J. Neurosci. Res., 23: 21-30.

Watt, N.J., MacIntyre, N., Collie, D., Sargan, D. and McConneIl, I., 1992. Phenotypic analysis of lymphocyte

populations in the lungs and regional lymphoid tissue of sheep naturally infected with maedi visna virus. Clin. Exp. Immunol., 90: 204-208.

Weise, M.J. and Carnegie, P.R., 1988. An approach to searching protein sequences for superfamily relation-

ships or chance similarities relevant to the molecular mimicry hypothesis: Application to the basic protein of myelin. J. Neurochem., 51: 1267-1273.

WucherpfeMig, K.W. and Strominger, J.L., 1995. Molecular mimicry in T cell-mediated autoimmunity: Viral

peptides activate human T cell clones specific for myelin basic protein. Cell, 80: 695-705.

Young, P.R., 1989. An improved method for the detection of peroxidase-conjugated antibodies on im- munoblots. J. Virol. Methods, 24: 227-236.