PNS Myelin Proteins Research Articles

Myelin Structural Integrity in a Model for Human Early-Onset CMT1B

Charcot-Marie-Tooth disease type 1B (CMT1B), a peripheral neuropathy, is caused by mutations in MPZ, the gene encoding protein zero (P0), the major integral protein of PNS myelin. An adhesive protein, P0 plays a significant role during elaboration and maintenance of multilamellar myelin. P0 mutation Arg69Cys (R69C) causes a severe early-onset form of CMT1B. To elucidate the pathogenesis of this neuropathy, an Arg69Cys knock-in mouse was generated by targeting the Arg69Cys mutation to one MPZ allele by homologous recombination in ES cells. Here we report our x-ray diffraction (XRD) measurements on the periodicity, membrane structure, and amount of myelin in unfixed, freshly-dissected nerves from wildtype (WT or +/+), heterozygous (R69C/+), and homozygous (R69C/ R69C) mice. CNS myelin (optic nerve) was also examined. The diffraction patterns showed decreasing strength of scattering intensity from myelin: WT > R69C/+ > R69C/R69C, indicating decreasing relative amounts of myelin. By contrast, optic nerves exhibited no such differences among genotypes. From the positions of the reflections the myelin periods of sciatic but not optic nerves were found to differ among the genotypes: 177.0 ± 0.4 A for WT, 178.4 ± 0.5 A for R69C/+, and 193.1 ± 4.2 A for R69C/R69C. The calculated electron density profiles showed R69C/R69C's wider period derived from ∼20 A-swelling at the extracellular apposition. The extent of membrane packing distortion (Δ/d) in PNS myelin, calculated using Bragg order peak widths, was 25% greater in R69C/+ and doubled in R69C/R69C compared to WT. Differences in amount of myelin, period, and Δ/d among the genotypes were statistically significant at p < 0.001. Finally, comparison of R69C/+ with P0± and R69C/R69C with P0-/- suggested the small amount of mutant P0 that enters the myelin may detrimentally affect myelin-myelin interactions to produce less regular/unstable packing.

The different forms of PNS myelin P0 protein within and outside lipid rafts

It is now well established that plasma membranes, such as the myelin sheath, are made of different microdomains with different lipid and protein composition. Lipid rafts are made mainly of sphingolipids and cholesterol, whereas the non-raft regions are made mainly of phosphoglycerides. Most myelin proteins may distribute themselves in raft and non-raft microdomains but the driving force that gives rise to their different distribution is not known yet. In this paper, we have studied the distribution of protein zero (P0), the most representative protein of PNS myelin, in the membrane microdomains. To this end, we have purified P0 from both non-raft (soluble P0, P0-S) and raft (P0-R) regions of PNS. Purified proteins were analyzed by two-dimensional gel electrophoresis and identified and characterized by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. A detailed structural description of the two P0 forms is given in terms of amino acid sequence, post-translational modifications, and composition of associated lipids. Our findings suggest that structural differences between the two proteins, mainly related to the glycogroups, might be responsible for their different localization.

The SOX10 transcription factor: evaluation as a candidate gene for central and peripheral hereditary myelin disorders.

The SOX10 transcription factor is involved in development of neural crest derivatives and fate determination in glial cells. SOX10 mutations have been found in patients with intestinal aganglionosis and depigmentation with deafness (Waardenburg-Hirschsprung). Associated neurological signs have been reported in some cases, including a patient exhibiting a central and peripheral myelin deficiency. Therefore, we screened for SOX10 mutations in a large cohort of patients with peripheral and central myelin disorders. 56 were affected by classical demyelinating Charcot-Marie-Tooth disease without identified mutations in the genes encoding PNS myelin proteins (PMP22, P0), connexin 32 and the zinc-finger transcription factor, EGR2. 88 patients with undetermined leukodystrophy were selected from a large European prospective study. Associated clinical, magnetic resonance imaging and electrophysiological signs were consistent with a defect in CNS myelination in 83 and with an active degeneration of the CNS myelin in 5. No abnormalities in the proteolipid protein gene (PLP) were found. The absence of SOX100 mutation in this large cohort of patients suggests that this gene is not frequently involved in peripheral or central inherited myelin disorders.

Tyrosine phosphorylation of PNS myelin P(0) occurs in the cytoplasmic domain and is maximal during early development.

P(0), the major protein of PNS myelin, is considered to play a critical role in the compaction and stabilization of myelin lamellae. The protein undergoes extensive posttranslational modifications, including phosphorylation at multiple serine moieties in the cytoplasmic region. Recently, we demonstrated that P(0) is phosphorylated on one or more tyrosine residues in rat nerve homogenates after incubation. In this study, we show that P(0) phosphorylated on tyrosine is also present in the intact animal. The proportion of P(0) molecules phosphorylated on tyrosine is highest during the first postnatal week, a period that coincides with the most rapid period of myelin deposition in the PNS. A peptide that constitutes the cytoplasmic domain was isolated from purified P(0) and shown by immunochemical and chemical means to be phosphorylated on the tyrosine corresponding to Y(191) in the intact protein. No evidence was obtained supporting the possibility that P(0) is phosphorylated on other tyrosine residues. The sequence of amino acids surrounding Y(191) resemble known substrate phosphorylation sites for some nonreceptor cytoplasmic tyrosine kinases, as well as tyrosine-based recognition signals associated with clathrin vesicle-mediated cndocytosis.

Schwann cell myelination requires timely and precise targeting of P(0) protein.

This report investigated mechanisms responsible for failed Schwann cell myelination in mice that overexpress P(0) (P(0)(tg)), the major structural protein of PNS myelin. Quantitative ultrastructural immunocytochemistry established that P(0) protein was mistargeted to abaxonal, periaxonal, and mesaxon membranes in P(0)(tg) Schwann cells with arrested myelination. The extracellular leaflets of P(0)-containing mesaxon membranes were closely apposed with periodicities of compact myelin. The myelin-associated glycoprotein was appropriately sorted in the Golgi apparatus and targeted to periaxonal membranes. In adult mice, occasional Schwann cells myelinated axons possibly with the aid of endocytic removal of mistargeted P(0). These results indicate that P(0) gene multiplication causes P(0) mistargeting to mesaxon membranes, and through obligate P(0) homophilic adhesion, renders these dynamic membranes inert and halts myelination.

Control of cholesterol biosynthesis in Schwann cells.

Cholesterol accounts for over one-fourth of total myelin lipids. We found that, during development of the rat sciatic nerve, expression of mRNA for hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in cholesterol biosynthesis, was up-regulated in parallel with mRNA for P0, the major structural protein of PNS myelin, and with ceramide galactosyltransferase (CGT), the rate-limiting enzyme in cerebroside biosynthesis. To help establish the nature of this coordinate regulation of myelin-related genes, we examined their steady-state mRNA levels in cultured primary Schwann cells. We also assayed synthesis of cholesterol and cerebroside to distinguish how much control of synthetic activity for these two myelin lipids involved mRNA levels for HMG-CoA reductase and CGT, and how much involved post-mRNA control mechanisms. Addition of forskolin to cells cultured in media supplemented with normal calf serum resulted in up-regulation of P0 and CGT mRNA expression and cerebroside synthesis, without corresponding increases in HMG-CoA reductase mRNA or cholesterol synthesis. Cholesterol synthesis increased approximately threefold in Schwann cells cultured with lipoprotein-deficient serum, without any increase in HMG-CoA reductase mRNA. Furthermore, addition of either serum lipoproteins or 25-hydroxycholesterol decreased cholesterol synthesis without altering HMG-CoA reductase mRNA levels. We conclude that, as in other tissues, cholesterol synthesis in Schwann cells is regulated primarily by intracellular sterol levels. Much of this regulation occurs at posttranscriptional levels. Thus, the in vivo coordinate up-regulation of HMG-CoA reductase gene expression in myelinating Schwann cells is secondary to intracellular depletion of cholesterol, as it is compartmentalized within the myelin. It is probably not due to coordinate control at the level of mRNA expression.

Macrophage recruitment in different models of nerve injury: lysozyme as a marker for active phagocytosis.

Macrophages play critical roles in both degenerative and regenerative processes following peripheral nerve injury. These include phagocytosis of debris, stimulation of Schwann cell dedifferentiation and proliferation, and salvage of myelin lipids for reutilization during regeneration. To better define the role of macrophages, we studied models of primary demyelination (tellurium intoxication) and secondary demyelination (nerve crush and cut). Sections of paraformaldehyde-fixed rat sciatic nerves at various stages of demyelination were stained with monoclonal antibody ED1, a standard macrophage marker, and a polyclonal antiserum specific for lysozyme (LYS). Near the peak of demyelination in all three models, LYS immunoreactivity colocalized with ED1 staining. Macrophages present in nerve after the period of maximal phagocytosis of myelin were much less immunoreactive for LYS. These results suggest LYS is a good marker for macrophages which are active in phagocytosis. Tellurium intoxication, which causes synchronous demyelination and subsequent remyelination of only about 25% of myelin internodes, recruited more macrophages (and induced more lysozyme expression) than either nerve crush or cut, which cause demyelination of all internodes distal to the injury site. This suggests that Schwann cells may recruit macrophages soon after metabolic insult and prior to actual demyelination. The final signal for macrophage recruitment is not directly related to the amount of damaged myelin. In the models listed above, steady state mRNA levels for apolipoprotein E (ApoE; possible mediator of cholesterol salvage), LYS, and P0 (major structural protein of PNS myelin), were analyzed by Northern blot analysis. LYS mRNA levels peaked sharply in all models, with a temporal pattern consistent with the expected presence of activated, phagocytic macrophages.(ABSTRACT TRUNCATED AT 250 WORDS)

A novel role for myelin-associated glycoprotein as an inhibitor of axonal regeneration

Following nerve injury, axons in the CNS do not normally regenerate. It has been shown that CNS myelin inhibits neurite outgrowth, though the nature of the molecules responsible for this effect are not known. Here, we demonstrate that the myelin-associated glycoprotein (MAG), a transmembrane protein of both CNS and PNS myelin, strongly inhibits neurite outgrowth from both developing cerebellar and adult dorsal root ganglion (DRG) neurons in vitro. This inhibition is reversed by an antiMAG antibody. In contrast, MAG promotes neurite outgrowth from newborn DRG neurons. These results suggest that MAG may be responsible, in part, for the lack of CNS nerve regeneration in vivo and may influence, both temporally and spatially, regeneration in the PNS.

Identification of the palmitoylation site in rat myelin P0 glycoprotein.

P0 glycoprotein, the major protein of PNS myelin, contains approximately 1 mol of covalently bound long-chain fatty acids. To determine the chemical nature of the fatty acid-protein linkage, P0 was labeled in rat sciatic nerve slices with [3H]palmitic acid and subsequently treated with various reagents. The protein-bound palmitate was released by incubation with the reducing agents dithiothreitol and 2-mercaptoethanol, and with 1 M hydroxylamine at pH 7.5. In addition, P0 was deacylated by treatment with 10 mM NaBH4 with the concomitant production of [3H]hexadecanol, indicating that the fatty acid is bound in a thioester linkage. This conclusion was supported further by the fact that deacylation with hydroxylamine generated free thiol groups, which were titrated with [14C]-iodoacetamide. To identify the cysteine residue involved in the thioester linkage, [14C]carboxyamidomethylated P0 was digested with trypsin and the resulting peptides analyzed by reversed-phase HPLC. Identification of the radioactive protein fragments by amino acid analysis and amino-terminal peptide sequencing revealed that Cys153 in rat P0 glycoprotein is the acylation site. The acylated cysteine is located at the junction of the putative transmembrane and cytoplasmic domains. This residue is also present in the P0 glycoprotein of other species, including human, bovine, mice, and chicken.

Tellurium‐Induced Alterations in 3‐Hydroxy‐3‐Methylglutaryl‐CoA Reductase Gene Expression and Enzyme Activity: Differential Effects in Sciatic Nerve and Liver Suggest Tissue‐Specific Regulation of Cholesterol Synthesis

The demyelination of peripheral nerves that results from exposure of developing rats to tellurium is due to inhibition of squalene epoxidase, a step in cholesterol biosynthesis. In sciatic nerve, cholesterol synthesis is greatly depressed, whereas in liver, some compensatory mechanism maintains normal levels of cholesterol synthesis. This tissue specificity was further explored by examining, in various tissues, gene expression and enzyme activity of 3-hydroxy-3-methylglutaryl-CoA reductase, the rate-limiting enzyme in cholesterol biosynthesis. Exposure to tellurium resulted in pronounced increases in both message levels and enzyme activity in liver, the expected result consequent to up-regulation of this enzyme in response to decreasing levels of intracellular sterols. In contrast to liver, levels of mRNA and enzyme activity in sciatic nerve were both decreased during the tellurium-induced demyelinating period. The temporal pattern of changes in 3-hydroxy-3-methylglutaryl-CoA reductase message levels in sciatic nerve seen following exposure to tellurium was similar to the down-regulation seen for mRNA specific for PNS myelin proteins. Possible mechanisms for differential control of cholesterol biosynthesis in sciatic nerve and liver are discussed.

Folding and function of the myelin proteins from primary sequence data

To explain how the myelin proteins are involved in the organization and function of the myelin sheath requires knowing their molecular structures. Except for P2 basic protein of PNS myelin, however, their structures are not yet known. As an aid to predicting their molecular folding and possible functions, we have developed a FORTRAN program to analyze the primary sequence data for proteins, and have applied this to the myelin proteins in particular. In this program, propensities for the secondary structure conformations as well as physical-chemical parameters are assigned to the amino acids and the pattern of these parameters is examined by calculating their average values, autocorrelation functions and Fourier transforms. To compare two proteins, their sequences are aligned using a unitary scoring matrix, and homologies are searched by plotting a two-dimensional map of the correlation coefficients. Comparison of the corresponding myelin basic proteins (MBP) and P0 glycoproteins (P0) for rodent and shark showed that the conserved residues included most of the amino acids which were predicted to form the alpha or beta conformations, while the altered residues were mainly in the hydrophilic and turn or coil regions. In both rodent and shark the putative extracellular domain of P0 glycoprotein displayed consecutive peaks of beta propensity similar to that for the immunoglobulins, while the cytoplasmic domain showed alpha-beta-alpha folding. To trace the immunoglobulin fold along the P0 sequence, we compared the beta propensity curve of P0 with that of the immunoglobulin M603, whose three-dimensional structure has been determined. We propose that the flat beta-sheets of P0 are orientated parallel to the membrane surface to facilitate their homotypic interaction in the extracellular space. An extra beta-fold in the extracellular domain of shark P0 compared with rodent P0 was found, and this may result in a greater attraction between the apposed extracellular surfaces and may account for a smaller extracellular space as measured by x-ray diffraction. A computer search of the myelin protein sequences for functional motifs revealed sites for N-glycosylation, phosphorylation, nucleotide binding, and certain enzyme activities. We note especially that there are potential nucleotide binding sites in proteolipid protein (PLP), MBP and 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNP). This is consistent with the experimental observations that PLP acts like an ionophore or proton channel when reconstituted into planar lipid bilayers, MBP binds GTP, and CNP catalyzes in vitro the hydrolysis of 2',3'-nucleotides into corresponding 2'-nucleotides.

Developmental Expression of the P0 Glycoprotein and Basic Protein mRNAs in Normal and Trembler Mutant Mice

Mice affected by the autosomal dominant Trembler mutation exhibit a severe hypomyelinization of the PNS. Previous biochemical studies have shown that the accumulation of the major PNS myelin proteins, P0 and myelin basic protein (MBP), is strongly diminished in Trembler sciatic nerves during postnatal development. We performed Northern blots which showed that the size of mRNA species for P0 and MBP in normal and mutant mice are indistinguishable. Densitometric analysis of Northern blots showed that, in normal mice, the proportion of P0 mRNA increases up to the 12th day, then decreases slowly. At day 40, the proportion is 60% of the maximal value. In the mutant, the proportion of P0 mRNA increases up to the 12th day and then decreases much faster than in the control. At days 12 and 40, the P0 mRNA proportion measured in Trembler sciatic nerves represents only 40% and 7%, respectively, of the proportion measured in control littermates. The MBP mRNA proportion in the normal mice increases up to the 16th day, and then decreases to attain 45% of the maximum level at day 40. In the Trembler mouse, there is a maximum level at day 12, representing 25% of the normal level, but the MBP mRNA is barely detectable at days 8 or 40. Thus, these data seem to indicate that in the Trembler sciatic nerves, the proportions of P0 and MBP mRNAs are too small to allow the synthesis of normal levels of the corresponding proteins.

Cyclic AMP regulation of P0 glycoprotein and myelin basic protein gene expression in semi-differentiated peripheral neurinoma cell line D6P2T

We have analyzed the effects of cAMP-elevating drugs (dibutyryl cAMP, forskolin, and isobutyl methylxanthine) on growth properties and myelin-specific gene expression in the peripheral neurinoma cell line D6P2T. The steady-state levels of RNA and polypeptide for the two major PNS myelin proteins, P0 glycoprotein (P0) and myelin basic protein (MBP), were measured by Northern blotting and immunoblotting, respectively. The levels of the two RNAs in individual cells were examined by in situ hybridization. The transcriptional activities of the P0 and MBP genes were analyzed by nuclear run-off experiments. Treatment with cAMP-elevating agents caused cell aggregation and dose-dependent increase in growth control. Expression of P0 RNA was constitutive in untreated cells and was repressed at high doses. Expression of MBP RNA was induced at low doses and repressed at higher doses. For both MBP and P0 the effects on gene expression were first detected after a lag of approximately 6 h, were manifested in all cells and were mediated, at least in part, at the transcriptional level. The level of P0 polypeptide was proportional to the level of P0 RNA, but MBP polypeptide was not detectable even under conditions where MBP RNA was induced. The results with this clonal model suggest that cAMP plays a pivotal role in regulation of growth and gene expression during Schwann cell differentiation.

Spasmodic, a mutation on chromosome 11 in the mouse.

A new recessive mutation, spasmodic (spd), producing behavior that mimics that of the neurological mutation spastic (spa) with rapid tremors, stiff posture, and difficulty in righting, arose spontaneously in strain A/HeJ at the Jackson Laboratory in 1979. It is not an allele of spa and linkage tests show that this mutation is located close to vestigial tail (vt) near the center of chromosome 11. Additional genetic tests show that it is not an allele of trembler (Tr), shaker-2 (sh-2), nor vibrator (vb), all neurological mutations located in the same region of chromosome 11. No differences were observed in the levels of the major CNS and PNS myelin proteins or lipids of spd/spd mice versus littermate controls, suggesting that, unlike several closely linked mutations, the spd mutation does not affect myelination. Pharmacological studies reported here show that aminooxyacetic acid improves the behavioral abnormalities of affected spd/spd mice in the same way it improves the behavior of affected spa/spa mice. However, unlike the spa/spa mice, there are no changes in the postsynaptic receptors for glycine, GABA, or benzodiazepines in spd/spd mice.

Antigenic sites common to major fish myelin glycoproteins (IP) and to major tetrapod PNS myelin glycoprotein (Po) reside in the amino acid chains

The major myelin glycoproteins in the CNS and PNS of trout (IP) were enzymatically deglycosylated with endoglycosidase F (Endo F) and examined by electro-immunoblotting. Following carbohydrate removal and loss of concanavalin A affinity each of the four IP components underwent a similar reduction in molecular size, corresponding to approximately 3,000 daltons. Immunological cross-reactivities with anti-bovine Po or anti-trout IP2 antibodies, were however fully retained by the Endo F cleavage products. This strongly implies that the antigenic sites shared by the mammalian Po protein and the various intermediate glycoproteins of trout CNS and PNS are located in the protein portion. Immunoblot analysis of the PNS myelin proteins from various species of the major vertebrate classes with anti-trout IP2 antiserum revealed striking differences in the immunological properties of the individual Po components which were not detected when anti-bovine Po antiserum was used as a probe.

Characterization of the anti-neural antibodies in the Sera of leprosy patients

Sera from 43 leprosy patients were tested for antibodies that bound to normal human nerve. Thirty-eight percent showed positive staining as demonstrated by indirect immunofluorescence. Only 1 out of 30 control sera tested displayed similar staining. Western blots of myelin and neural intermediate filament (IF) proteins were tested with patient sera. Two of the anti-neural antibody (ANeAb)-positive leprosy sera bound to the P0 protein of PNS myelin. All 17 ANeAb-positive leprosy sera displayed 2 or more bands in the molecular weight range of Mr 45 000-55 000, when tested against IF proteins. One explanation for these findings is that leprosy patients produce antibodies to intermediate filament (IF) proteins released subsequent to the bacterial invasion of the peripheral nerves. The importance of these autoantibodies in the pathogenesis of leprosy is discussed.

Appearance of Myelin proteins during vertebrate evolution

Myelin, defined as an arrangement of spirally fused unit membranes, is an acquisition of vertebrates and first appeared during evolution in Gnathostomata. In all species studied PNS and CNS myelins contain the myelin-associated glycoprotein (MAG) and the myelin basic protein (MBP). Throughout phylogeny PNS myelin is characterized by the major P(0) glycoprotein which is called IP in fishes. The PNS myelin proteins did not evolve further except for the addition of P(2) protein from reptiles onward. In Elasmobranchii and Chondrostei, PNS and CNS myelin proteins are similar. CNS myelin of actinopterygian fishes possesses a 36,000 Da protein (36K) in addition to P(0)-like IP glycoproteins. In tetrapod CNS myelin, P(0) is replaced by the proteolipid protein (PLP) and the Wolfgram protein (WP). Of particular interest in a transitional phylogenetic sense are the lungfish Protopterus, carrying glycosylated PLP (g-PLP) but no P(0), 36K or WP, and the bichir Polypterus, showing simultaneous presence of P(0), 36K and PLP. These results indicate that myelin proteins could be valuable molecular markers in establishing vertebrate phylogenetic relationships and in reconstructing the fish-tetrapod transition.

Specificity of mouse and human monoclonal antibodies to myelin-associated glycoprotein.

Some patients with neuropathy have IgM M-proteins that bind to myelin and to myelin-associated glycoprotein (MAG). We compared the binding properties of a human anti-MAG M-protein with three mouse monoclonal anti-MAG antibodies (GEN-S1, GEN-S3, GEN-S8) and with a mouse monoclonal antibody (HNK-1) that binds to both MAG and to human natural killer cells. The antibodies GEN-S1, GEN-S3, and GEN-S8 bound to different epitopes in the polypeptide portion of MAG as shown by peptide mapping, deglycosylation and competitive binding studies. The M-protein and HNK-1 bound to both CNS and PNS MAG and to several additional protein bands of 70K, 30K, 26K, and 23K daltons in peripheral, but not in central myelin; they did not bind to deglycosylated MAG. The M-protein and HNK-1 immunostained myelin diffusely, whereas GEN-S8 immunostained only the periaxonal and outer regions of myelin sheath, and there was no staining with GEN-S1 or GEN-S3. The human M-proteins probably bind to a carbohydrate moiety in MAG that is also present in other PNS myelin proteins. This may explain the observed differences in immunostaining and the sparing of the CNS in patients with anti-MAG M-proteins.

Cyclic AMP-stimulated protein kinase activity in rabbit peripheral myelin.

Cyclic AMP-sensitive protein kinase activity has been found in suspensions of purified rabbit peripheral myelin. The enzyme phosphorylated the P0, "Y", X, P1, and P2 myelin proteins. Kinase activity, which was maximal at physiological pH, 2.5 mM Mg2+, and 2 microM cAMP, was stimulated three-fold over basal levels by cyclic AMP. Addition of calcium or EGTA had no effect on the enzyme activity in the presence or absence of cyclic AMP. Cyclic GMP also did not stimulate endogenous or exogenous protein phosphorylation. Theophylline, an inhibitor of 3',5'-cyclic nucleotide phosphodiesterase activity, increased protein kinase activity in the presence of cyclic AMP. These data show that PNS myelin proteins can be phosphorylated in situ by a protein kinase system whose activity is stimulated selectively by cyclic AMP.

Myelin-associated glycoprotein is the antigen for a monoclonal IgM in polyneuropathy.

Recent studies show that IgM monoclonal antibody from patients with IgM paraproteinemia and peripheral neuropathy reacts with a protein component of human PNS myelin and an analogous component or components of human CNS myelin. We have now demonstrated that the antigen for this antibody is a specific glycoprotein component of myelin, referred to as myelin-associated glycoprotein (MAG). Human PNS and CNS myelin proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on pore-gradient slabs, and MAG was identified by the immuno-electroblot procedure with rabbit anti-MAG (rat). The identical band(s) were stained by an analogous procedure with patient serum as the first antibody. Human PNS MAG had an apparent molecular weight of 107,000. Human CNS MAG appeared as three bands: 113,000, 107,000, and 92,000. Passage of myelin proteins through a concanavalin A-Sepharose column removed the staining component. Purified patient IgM, added to a lithium diiodosalicylate extract of myelin, immunoprecipitated MAG. This antibody also cross-reacted with MAG from bovine CNS, but not from rabbit, rat, or mouse.