Peripheral Nerve Myelin Proteins Research Articles

Inherited demyelinating peripheral neuropathies: relating myelin packing abnormalities to P0 molecular defects.

P0-glycoprotein, the major integral membrane protein of peripheral nerve myelin, is thought to mediate myelination and membrane interactions via its extracellular domain (P0-ED). Molecular modeling of P0-ED has suggested which of its amino acid side-chains may be involved in heterophilic and homophilic adhesions. We previously showed that some of these amino acids are the same ones that are substituted or deleted due to mutations in the human gene for P0 (MPZ), which correlate with certain cases of demyelinating motor and sensory peripheral neuropathies. In the current study, high magnification electron microscopy was used to examine the myelin membrane packing in sural nerve biopsies from patients with MPZ mutations. We found that there were distinguishable ultrastructural phenotypes that could be explained by the alterations in P0-ED. These phenotypes, which were not observed in a control nerve, included widening or irregularity of the extracellular apposition alone (delta Ser34; Arg69Cys), widening at both the extracellular and cytoplasmic appositions (Arg69His), the presence of focal bridges in the widened extracellular space (Arg69His), and a diminished (Arg69Cys) or absence (Arg69His) of staining of the double intraperiod line. Our study, which suggests that the altered P0 is incorporated into the myelin sheath, provides a unique basis for further molecular/ultrastructural correlations between P0-ED structure and myelination irregularities.

Inherited demyelinating peripheral neuropathies: Relating myelin packing abnormalities to P0 molecular defects

P0-glycoprotein, the major integral membrane protein of peripheral nerve myelin, is thought to mediate myelination and membrane interactions via its extracellular domain (P0-ED). Molecular modeling of P0-ED has suggested which of its amino acid side-chains may be involved in heterophilic and homophilic adhesions. We previously showed that some of these amino acids are the same ones that are substituted or deleted due to mutations in the human gene for P0 (MPZ), which correlate with certain cases of demyelinating motor and sensory peripheral neuropathies. In the current study, high magnification electron microscopy was used to examine the myelin membrane packing in sural nerve biopsies from patients with MPZ mutations. We found that there were distinguishable ultrastructural phenotypes that could be explained by the alterations in P0-ED. These phenotypes, which were not observed in a control nerve, included widening or irregularity of the extracellular apposition alone (delta Ser34; Arg69Cys), widening at both the extracellular and cytoplasmic appositions (Arg69His), the presence of focal bridges in the widened extracellular space (Arg69His), and a diminished (Arg69Cys) or absence (Arg69His) of staining of the double intraperiod line. Our study, which suggests that the altered P0 is incorporated into the myelin sheath, provides a unique basis for further molecular/ultrastructural correlations between P0-ED structure and myelination irregularities.

Crystal Structure of the Extracellular Domain from P 0, the Major Structural Protein of Peripheral Nerve Myelin

P 0, the major protein of peripheral nerve myelin, mediates membrane adhesion in the spiral wraps of the myelin sheath. We have determined the crystal structure of the extracellular domain from P 0 (P 0ex) at 1.9 Å resolution. P 0ex is folded like a typical immunoglobulin variable-like domain; five residues at the C-terminus are disordered, suggesting a flexible linkage to the membrane. The requirements for crystallization of P 0ex are similar to those for maintaining the native extracellular spacing of adjacent myelin lamellae; thus, given the self-adhesive character of P 0ex, the crystal itself may reveal some of the natural interactions that occur between P 0 molecules in myelin. The structure leads to the suggestion that P 0 extracellular domains may emanate from the membrane surface as tetramers that link to tetramers on the opposing membrane surface, to result in the formation of networks of molecules. We report analytical ultracentrifugation data for P 0ex that support this idea.

Clinical Phenotypes of Different MPZ (P 0) Mutations May Include Charcot–Marie–Tooth Type 1B, Dejerine–Sottas, and Congenital Hypomyelination

Hereditary demyelinating peripheral neuropathies consist of a heterogeneous group of genetic disorders that includes hereditary neuropathy with liability to pressure palsies (HNPP), Charcot–Marie–Tooth disease (CMT), Dejerine–Sottas syndrome (DSS), and congenital hypomyelination (CH). The clinical classification of these neuropathies into discrete categories can sometimes be difficult because there can be both clinical and pathologic variation and overlap between these disorders. We have identified five novel mutations in the myelin protein zero ( MPZ) gene, encoding the major structural protein (P 0) of peripheral nerve myelin, in patients with either CMT1B, DSS, or CH. This finding suggests that these disorders may not be distinct pathophysiologic entities, but rather represent a spectrum of related “myelinopathies” due to an underlying defect in myelination. Furthermore, we hypothesize the differences in clinical severity seen with mutations in MPZ are related to the type of mutation and its subsequent effect on protein function (i.e., loss of function versus dominant negative).

Myelin protein zero (MPZ) gene mutations in nonduplication type 1 Charcot-Marie-Tooth disease.

The myelin protein zero gene (MPZ) maps to chromosome 1q22-q23 and encodes the most abundant peripheral nerve myelin protein. The Po protein functions as a homophilic adhesion molecule in myelin compaction. Mutations in the MPZ gene are associated with the demyelinating peripheral neuropathies Charcot-Marie-Tooth disease type 1B (CMT1B), and the more severe Dejerine-Sottas syndrome (DSS). We have surveyed a cohort of 70 unrelated patients with demyelinating polyneuropathy for additional mutations in the MPZ gene. The 1.5-Mb DNA duplication on chromosome 17p11.2-p12 associated with CMT type 1A (CMT1A) was not present. By DNA heteroduplex analysis, four base mismatches were detected in three exons of MPZ. Nucleotide sequence analysis identified a de novo mutation in MPZ exon 3 that predicts an Ile(135)Thr substitution in a family with clinically severe early-onset CMT1, and an exon 3 mutation encoding a Gly(137)Ser substitution was identified in a second CMT1 family. Each predicted amino acid substitution resides in the extracellular domain of the Po protein. Heteroduplex analysis did not detect either base change in 104 unrelated controls, indicating that these substitutions are disease-associated mutations rather than common polymorphisms. In addition, two polymorphic mutations were identified in MPZ exon 5 and exon 6, which do not alter the codons for Gly(200) and Ser(228), respectively. These observations provide further confirmation of the role of MPZ in CMT1B and suggest that MPZ coding region mutations may account for a limited percentage of disease-causing mutations in nonduplication CMT1 patients.

T cell antigenic and neuritogenic activity of recombinant human peripheral myelin P2 protein

The major neuritogenic protein of peripheral nerve myelin is the P2 protein. Human P2 is a candidate autoantigen in inflammatory demyelinating diseases of the peripheral nervous system. Since human P2 is not readily available, we produced full-length recombinant human P2 protein (rhP2) in Escherichia coli. RhP2 was recognized by neuritogenic rat T cell lines and induced experimental autoimmune neuritis in Lewis rats. Production of rhP2 allowed the generation of human T cell lines reactive to the autologous protein. Studies of human T cell autoreactivity as well as efforts to use hP2 as a tolerogen will be facilitated by the large-scale expression of rhP2.

Mutations in demyelinating peripheral neuropathies support molecular model of myelin P0-glycoprotein extracellular domain.

Homophilic interactions of the major integral membrane protein of peripheral nerve myelin, P0-glycoprotein, are thought to mediate membrane adhesion and compaction. Molecular modeling of its extracellular domain (P0-ED), based on its resemblance to an immunoglobulin variable domain and on X-ray diffraction measurements of inter-membrane spacings of myelin, has suggested which amino acid sidechains may be involved in the homophilic adhesion. Recently identified point-mutations in the human P0 gene result in amino acid substitutions in P0 protein and correlate with demyelinating motor and sensory neuropathies. The molecular model explains how these changes result in disrupted P0-P0 interactions; indicates how compensatory changes in amino acids, as occur in P0-ED of other species, preserve normal homophilic interactions; and predicts what other residue substitutions might underlie additional cases of demyelinating neuropathies.

Restricted hypotonic swelling of peripheral nerve myelin in streptozocin‐induced diabetic rats

Experimental diabetic neuropathy includes the non-enzymatic glycosylation (or glycation) of the major proteins of peripheral nerve myelin. We have used X-ray diffraction to determine whether such glycation affects myelin membrane structure and interactions in peripheral nerves from experimental diabetic rats. Streptozocin at 60 mg/kg was injected intraperitoneally to induce diabetes; controls were pair-fed and age-matched. Animals were sacrificed periodically from 2 weeks to after 1 year. The dissected sciatic nerves were tied off and incubated overnight at room temperature in hypotonic saline of defined pH and ionic strength or in distilled water. Such treatments have been shown to result in systematic changes of myelin period, which can be detected using X-ray diffraction, and which may indicate alterations in inter-membrane interactions owing to changes in composition. We observed no differences in repeat periods between control and diabetic nerves at pH 4.0 and 7.4, and ionic strength 0.01, 0.02, 0.06, 0.15, and 0.18; however, we did detect a significant difference (P < .02) in their maximum extent of swelling in distilled water: control nerves showed a period of 292 A (s.d. 23 A; n = 12) compared to 272 A (s.d. 19 A; n = 11) for diabetic nerves. To determine whether this difference in swelling was due to an alteration in the properties of the apposed, extracellular surfaces of the myelin membranes or to the connective tissue in peripheral nerve, we compared the X-ray patterns from peripheral nerve myelin isolated by sucrose density gradient centrifugation from sciatic nerves of diabetic and control rats. No difference in the patterns was observed.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for a recessive PMP22 point mutation in Charcot–Marie–Tooth disease type 1A

Charcot-Marie-Tooth disease type 1A (CMT1A) is an autosomal dominant neuropathy that can be caused by dominant point mutations in PMP22 which encodes a peripheral nerve myelin protein. Usually, CMT1A is caused by the duplication of a 1.5-megabase (Mb) region on chromosome 17p11.2-p12 containing PMP22. Deletion of a similar 1.5-Mb region is associated with hereditary neuropathy with liability to pressure palsies (HNPP), a clinically distinct neuropathy. We have identified a severely affected CMT1 patient who is a compound heterozygote for a recessive PMP22 point mutation, and a 1.5 Mb deletion in 17p11.2-p12. A son heterozygous for the PMP22 point mutation had no signs of neuropathy, while two others heterozygous for the deletion had HNPP, suggesting that point mutations in PMP22 can result in dominant and recessive alleles contributing to CMT1A.

Antibody responses to P 0 and P 2 myelin proteins in Guillain-Barré syndrome and chronic idiopathic demyelinating polyradiculoneuropathy

Immunisation with the peripheral nerve myelin proteins P 0 or P 2 induces inflammatory neuropathy in animals. We sought antibodies with an ELISA to these proteins in 38 patients with acute Guillain-Barré syndrome (GBS), 32 patients with chronic idiopathic demyelinating polyradiculoneuropathy (CIDP), 31 patients with other neuropathies (ONP) and 26 normal control (NC) subjects. We discovered IgM antibodies to human P 0 protein in the sera of 18.5% of the patients with GBS, 15.6% with CIDP, 6.4% with ONP and 3.8% of NC subjects. Of the sera which reacted with P 0, sera from 4 7 of GBS, 3 5 of CIDP, 1 2 of ONP patients and 0/1 of NC subjects reacted with a synthetic P 0 peptide representing residues 150–169 from the cytoplasmic portion of the molecule. IgG antibodies to P 0 were slightly less common than IgM antibodies, being present in only 7.9% of GBS, 0% of CIDP and 3% of ONP patients and 0% of NC subjects. We found antibodies to bovine P 2 protein more commonly than antibodies to P 0. IgM antibodies were present in 39.5% of GBS, 34.4% of CIDP, 16.1% of ONP patients and 15.4% of NC subjects. IgG antibodies were present in 18.4% of GBS, 12.5% of CIDP, 3.2% of ONP patients and 7.6% of NCs. Of the sera which contained antibodies to P 2 protein, only a few reacted with P 2 peptides 14–25 or 58–81, but without any consistent pattern of reactivity. We propose that IgM antibodies to P 0 may contribute to the demyelination in some cases of GBS and CIDP, whereas the more commom antibodies to P 2 may merely be markers of a simultaneous pathogenetic T cell-mediated response.

T cell responses to myelin proteins in Guillain-Barré syndrome

We have investigated the hypothesis that the pathogenesis of Guillain-Barré syndrome (GBS) involves an autoimmune T cell response to P 0 and P 2 proteins of peripheral nerve myelin. The proliferative responses of blood mononuclear cells (MNC) to myelin proteins and synthetic peptides derived from them were determined in patients with GBS and chronic idiopathic demyelinating polyradiculoneuropathy (CIDP), normal controls (NC) and patients with other neuropathies (ONP). Twelve out of 19 GBS patients responded to P 0 or P 2, 6 to P 0 and its peptides only, 3 to P 2 and its peptides only, and 3 to both P 0 and P 2 antigens. Responses to at least one of the antigens were also found in 6 13 of CIDP patients, but in only 4 17 NC and 2 6 ONP. Immune responses in GBS are heterogeneous. The early T cell responses to P 0 protein, described here for the first time, may be important in the pathogenesis of some cases.

A monoclonal antibody against HSV type 1 ribonucleotide reductase cross-reacts with the P 0 protein of peripheral nerve myelin

An epitope on peripheral nerve myelin was detècted by the use of a mouse monoclonal antibody directed against the 38 kDa subunit of herpes simplex virus (HSV) type 1 ribonucleotide reductase. Immunohistochemistry showed reactivity solely in PNS myelin. In nerve roots there was a sharp border in transitional zones to the negative CNS myelin. The immunoreactivity was found in rat, guinea pig, bovine and human peripheral nerves. Western blot analysis of peripheral nerve myelin as well as purified P 0 revealed a distinctly stained band corresponding to a molecular weight of approximately 29 kDa. The present finding of a shared antigenic determinant between HSV ribonucleotide reductase and peripheral nerve P 0 may be of pathogenetic relevance in virus induced demyelinating diseases in the peripheral nervous system.

The role of complex carbohydrates in adhesion of the myelin protein, P0

The most abundant protein of peripheral nerve myelin, a glycoprotein termed P0, is believed to be involved in the compaction of the myelin sheath and is postulated to be the closest relative to the ancestral gene for the immunoglobulin superfamily. Recently, P0 has indeed been shown to behave like a homophilic adhesion molecule via interactions of its extracellular domains. Here we demonstrate the importance of the oligosaccharide moieties of P0 in its functioning as a homophilic adhesion molecule. Expression of the complex form of P0 glycoprotein in transfected Chinese hamster ovary cells greatly increased the adhesiveness of those cells, whereas expression of the high-mannose form of P0 glycoprotein did not. This is the first step in the dissection of P0-P0 interaction at the molecular level.

Comparative studies on myelin proteins in mammalian peripheral nerve

Myelin proteins in mammalian peripheral nerve were studied comparatively. 1. 1. While each content of P1 and P2 in the myelin varied among species, additional content of P1 and P2 are relatively constant. 2. 2. The antigenic determinants of P2 for induction of experimental allergic neuritis were reported. 3. 3. Amino acid sequence analysis of P0 revealed that P0 is conserved across species and belongs to the immunoglobulin superfamily. 4. 4. The characteristic carbohydrate chain of P0 containing sulfate and sialic acid showed a positive reaction to the molecule-related immunity and adhesion. 5. 5. Molecular architecture of the myelin is discussed.

Correction of Altered Metabolic Activities in Sciatic Nerves of Streptozocin-Induced Diabetic Rats: Effect of Ganglioside Treatment

The effect of ganglioside administration to nondiabetic and streptozocin-induced diabetic rats on sciatic nerve Na(+)-K(+)-ATPase, polyphosphoinositide (PPI) turnover, and protein phosphorylation was investigated. Gangliosides were injected (10 mg/kg body wt i.p.) for 10 or 30 days beginning 20 days after induction of diabetes. Na(+)-K(+)-ATPase activity was reduced nearly 50% in diabetic nerve and was restored to normal by both ganglioside treatments. The elevated levels of fructose and sorbitol and depressed content of myoinositol in diabetic nerve were unaffected by 30 days of ganglioside treatment, indicating that the restoration of Na(+)-K(+)-ATPase activity is not dependent on normal concentrations of these compounds. In the same nerves, 32P incorporation into phosphatidylinositol 4,5-bisphosphate and phosphatidylinositol 4-phosphate increased 73-76 and 39-53%, respectively, in diabetic compared with nondiabetic tissue. Ganglioside administration abolished the elevated labeling of PPIs after 30 days but was ineffective after only 10 days. Neither ganglioside regimen was able to reverse enhanced phosphorylation of the major peripheral nerve myelin protein P0. The finding that gangliosides can more quickly correct the effects of diabetes on Na(+)-K(+)-ATPase activity than on PPI turnover suggests that the mechanisms underlying these two phenomena are not closely related and are distinct from the sequence of events responsible for altered myelin protein phosphorylation.

Reconstitution into liposomes of highly purified PO glycoprotein from avian peripheral nerve myelin

To explore the adhesive, possibly fusogenic properties of the major protein of peripheral nerve myelin (PO protein), three methods were employed for the reconstitution of this glycoprotein into liposomes of various lipid compositions. The most successful technique for the preparation of proteoliposomes was a detergent solubilization that employed Bio-Beads for the removal of Triton X-100 from the PO protein – lipid – detergent mixture. This method yielded a similar level of PO protein relative to lipids in the liposomes to that found in myelin. The protein to lipid ratio of the reconstituted proteoliposomes was only 1.4-fold lower than the initial protein to lipid ratio employed to prepare the liposomes. In contrast, the solvent evaporation and reverse-phase evaporation methods provided liposomes with 30 times lower protein to lipid ratio when compared with PO protein to lipid ratio initially used for the preparation of proteoliposomes. The incorporation of PO protein was dependent on the lipid composition of the vesicles and on the degree of purification of the protein. Proteolytic enzyme treatment showed that proteoliposomes containing the highly purified form of PO protein were partially or completely sensitive to trypsin treatment, whereas in liposomes containing a partially purified form of PO protein, a 22.4-kilodalton fragment was completely protected from trypsin digestion, indicating a transmembrane orientation of the partially purified polypeptide. Endoglycosidase F treatment of the liposomes established that all of the oligosaccharide moieties were removed from the liposome-associated protein, indicating the oligosaccharide moiety was exposed to the outside surface of the liposomes. Taken together, the results of limited proteolysis and endoglycosidase treatments did not provide strong evidence for the transmembrane orientation of highly purified PO in the liposomes, but were consistent with a surface of partial hairpin configuration, where the extracellular domains are facing the outside milieu. Electron microscopic examinations revealed highly convoluted, elongated shapes arid extensive fusion of PO liposomes containing lipids capable of adopting the hexagonal (H11) phase. Proteoliposomes prepared with the simple lipid composition of dipalmitoyl phosphatidylcholine -cholesterol (10:1 molar ratio) yielded stable, unilamellar liposomes of 80 nm average diameter, that encapsulated water-soluble drugs efficiently into the intravesicular compartment.Key words: liposomes, PO glycoprotein, myelin, peripheral nerve.

Analysis of myelin proteins in sympathetic peripheral nerve of adult rats

Biochemical analyses of myelin proteins in rat sympathetic peripheral nerve were correlated with morphological observations. Myelin proteins in superior cervical ganglia (SCG) and the paravertebral (thoraco-lumbar) chain of ganglia were quantitated by immunoassays and examined qualitatively by Western blotting. The results were compared to those obtained on sciatic nerves from the same animals. In rats aged one year, the concentrations of PO glycoprotein and myelin basic protein (MBP) in SCG were about 1% of those in sciatic nerve, consistent with the relatively low numbers of myelinated fibers in sympathetic nerve. The relative concentration of myelin-associated glycoprotein (MAG) was higher, being 6.7% of that in sciatic nerve. The latter finding is probably due to the greater proportion of MAG-containing membranes (periaxonal, paranodal, and Schmitt-Lanterman incisures) in myelinated fibers of the SCG, in which the internodes are both short and thinly myelinated. The proportion of the 21 kDa, 18 kDa and 17 kDa forms of MBP relative to the 14 kDa form was much higher in SCG than in sciatic nerve, probably reflecting the fact that myelin formation continues actively during adult life in the ganglia, whereas the deposition of myelin is complete at a much earlier age in somatic nerves. The levels of myelin proteins were 2- to 3-fold higher in the paravertebral chain ganglia. These studies indicate that quantitation of myelin proteins in sparsely myelinated sympathetic nerve tissue is feasible and provide a baseline for further studies on the control of myelination in sympathetic nerve during adult life.

A Phorbol Ester‐Sensitive Kinase Catalyzes the Phosphorylation of P0 Glycoprotein in Myelin

The proposed structural protein of peripheral nerve myelin, P0, has been shown to have several covalent modifications. In addition to being glycosylated, sulfated, and acylated, P0 is phosphorylated, with the intracellular site of this latter addition being in question. By employing nerve injury models that exhibit different levels of P0 biosynthesis in the absence and presence of myelin assembly, we have examined the cellular location of P0 phosphorylation. It is demonstrated that there is comparable P0 phosphorylation in both normal and crush-injured adult rat sciatic nerves, although the level of biosynthesis of P0 differs between these myelin maintaining and actively myelinating nerve models, respectively. The glycoprotein does not appear to be phosphorylated readily in the transected adult sciatic nerve, a preparation in which P0 biosynthesis is observed but that lacks myelin membrane. These observations suggest that the modification is not associated with the biosynthesis or maturation of P0 in the endoplasmic reticulum or Golgi, but that it instead occurs after myelin assembly. That P0 phosphorylation occurs in the normal nerve even when translation is inhibited by cycloheximide treatment lends further support to this conclusion. P0 is shown to be phosphorylated on one or more serine residues, with all or most of the phosphate group(s) being labile as evidenced by pulse-chase analysis. Addition of a biologically active phorbol ester, 12-O-tetradecanoylphorbol-13-acetate or 4 beta-phorbol 12,13-dibutyrate, substantially increases the extent of [32P]orthophosphate incorporation into the glycoprotein of normal and crushed nerve but not transected nerve. Biologically inactive 4 alpha-phorbol 12,13-didecanoate has no effect on P0 phosphorylation. Similarly, the addition of the cyclic AMP analog 8-bromo-cyclic AMP causes no appreciable changes in P0 labeling. These findings indicate that the phorbol ester-sensitive enzyme, protein kinase C, may be responsible for the phosphorylation of P0 within the myelin membrane.

A 42 K protein of chick sciatic nerve is immunologically related to PO protein of peripheral nerve myelin

The major protein (PO) in PNS myelin is an integral membrane glycoprotein with a molecular weight of about 30 K. The level of PO protein in the developing sciatic nerve of the chicken was monitored by a solid-phase immunoassay and densitometry of Coomassie blue stained polyacrylamide gels. The most rapid rate of accumulation of PO protein occurred after 16 days of embryonic development. In addition to the 30 K PO protein, a number of higher molecular weight proteins could be distinctly detected by immunoblotting. Amongst these high molecular weight proteins, a species with an apparent molecular weight of 42 K was specifically immunostained with epitope-selected polyclonal antibodies against PO protein. This 42 K protein could be first detected after 16 days of embryonic development and increased rapidly following the pattern of myelination in the sciatic nerve. The enzyme endoglycosidase F, which specifically removes N-asparagine linked high mannose and complex carbohydrates from glycoproteins, converted the PO and 42 K proteins to lower molecular weight forms, which could be specifically immunostained by epitope selected polyclonal antibodies to the PO protein. Subcellular fractionation of the 17-day embryonic nerve demonstrated that the 42 K protein was enriched in myelin and microsomal subfractions relative to the total homogenate. These results indicate that the 42 K immuno-crossreactive protein might be chemically and functionally related to the PO protein of the PNS myelin.

Induction of experimental allergic neuritis in the BN rat: P2 protein-specific T cells overcome resistance to actively induced disease.

T lymphocyte lines specific for the peripheral nerve myelin protein P2 were selected from the lymph nodes of Brown Norway (BN) rats immunized with bovine P2 protein in complete Freund's adjuvant. These T cells expressed the W3/25+, OX8-phenotype and responded specifically to bovine P2 protein, but not to PPD or bovine basic protein, in T cell proliferation assays. When injected i.v. into syngeneic recipients, BN P2-specific T cell lines induced both clinical and histologic signs of experimental allergic neuritis (EAN), overcoming the resistance of this rat strain to actively induced EAN. Although the histopathology of the disease was indistinguishable from that seen in T cell-mediated EAN in the Lewis rat, disease onset was considerably later, 7 to 8 days after cell transfer, as opposed to 4 days in Lewis. This lag phase between inoculation and disease onset could not be further reduced even by raising the cell dose to 50 X 10(6) cells/host. The fine specificity of the T cell response to P2 differs between Lewis- and BN-derived T cell lines. At least one neuritogenic epitope for each strain was present in the cyanogen bromide-derived peptide CB2 (residues 21-113), as shown by the ability of CB2-specific T cell lines derived from each strain to transfer EAN to the appropriate host strain. However, neuritogenic BN T lines fail to mount a response to the sequence 53-78 (SP4), which encompasses an epitope that is neuritogenic for Lewis rats. These results demonstrate that the resistance of BN rats to actively induced EAN is not due to the lack of appropriate P2-specific autoreactive T cell clones in the normal T repertoire. Furthermore, the results suggest that two distinct epitopes of P2 are responsible for EAN in Lewis and BN rats.