Ultrastructural protein zero expression in Charcot-Marie-Tooth type 1B Disease

ABSTRACT: Charcot–Marie–Tooth type 1B (CMT 1B) disease, an inherited demyelinating peripheral neuropathy, results from different point mutations located in the P0 gene on chromosome 1 q21–23. We have quantified, at the ultrastructural level, the immunocytochemical expression of the P0 protein in two unrelated CMT 1B patients with mutations (Ser 78 to Leu and Asn 122 to Ser) located in two different exons in the extracellular domain of the protein. A twofold decrease in P0 expression was observed in compact myelin in each case, compared with age-matched controls. The severity of the phenotypes showed no direct relationship to the levels of P0 protein expression in these 2 patients. © 1999 John Wiley & Sons, Inc. Muscle Nerve 22: 99–104, 1999

ULTRASTRUCTURAL PROTEIN ZERO EXPRESSION IN CHARCOT–MARIE– TOOTH TYPE 1B DISEASE PHILIPPE SINDOU, PhD,1 JEAN-MICHEL VALLAT, MD,1 FRANC ¸ OISE CHAPON, MD,2 JUAN-JOSE´ ARCHELOS, MD,3 FRANC ¸ OIS TABARAUD, MD,1 THIERRY ANANI, MD,1 KYLE G. BRAUND, PhD,4 THIERRY MAISONOBE, MD,5 JEAN-JACQUES HAUW, MD, PhD,5 and ANTOON VANDENBERGHE, PhD6 1

Department of Neurology, University Hospital, 2 Avenue Martin Luther King, 87042 Limoges, France 2 Department of Neurology, University Hospital, Caen, France 3 Department of Neurology and Multiple Sclerosis Research Group, Julius-Maximilians-Universita¨t, Wu¨rzburg, Germany 4 Scott-Ritchey Research Center, College of Veterinary Medicine, Auburn, Alabama, USA 5 Salpe´trieˆre Hospital, Paris, France 6 Department of Molecular Neurogenetics, Claude Bernard University, Lyon I, France Accepted 14 August 1998

The most common inherited demyelinating peripheral neuropathy in humans is Charcot–Marie–Tooth disease type 1 (CMT 1), with an incidence of 1 in 2,500.26 Patients exhibit distal muscle weakness, pes cavus, atrophy, mild sensory loss, and reduced nerve conduction velocities. Histological findings show a demyelinating process and typical onion-bulb formations8 in nerve biopsies.The autosomal dominant transmission in CMT 1 patients can be classified into Abbreviations: Arg, arginine; Asn, aspartic acid; CMT, Charcot–Marie– Tooth disease; Cys, cysteine; Ig, immunoglobulin; Leu, leucine; mab, monoclonal antibody; PCR, polymerase chain reaction; PMP 22, peripheral myelinic protein 22; P0, protein zero; P0-ED, extracellular domain of the protein zero; Ser, serine; SSCP, single strand conformational polymorphism Key words: Charcot–Marie–Tooth disease 1B; immunocytochemical quantification; protein zero; ultracryomicrotomy Correspondence to: Dr. J.M. Vallat CCC 0148-639X/99/010099-06 © 1999 John Wiley & Sons, Inc.

P0 Expression in CMT 1B Disease

different subtypes depending on the genetic defect responsible for the disease.29 These autosomal dominant forms are referred to as CMT 1A, CMT 1B,7 and CMT 1C.6 CMT 1A derives from the duplication of the major locus mapped to chromosome 17 p112-12.32 Most patients present this duplication for the peripheral region containing the gene coding for the PMP 22 protein.15 The CMT 1B locus maps to the centromeric region of chromosome 1q21–23.3,18 This minor locus results from a mutation in the gene encoding P0.13,16 CMT 1B is shown to result from various point mutations of the P0 gene located on human chromosome 1, as is reviewed elsewhere.17 In the present study, we quantify the expressivity of the P0 protein in nerve biopsies processed for ultrastructural immunocytochemical assays, and obtained from 2 unrelated patients with CMT 1B disease. We employ the same qualitative method as that

MUSCLE & NERVE

January 1999

99

used in a previous study on the expression of PMP 22 protein in CMT 1A and hereditary neuropathy to pressure palsies.31 PATIENTS AND METHODS Clinical and Electrophysiological Examinations.

Two unrelated patients from different families with a history of CMT were studied. Case 1. A 60-year-old woman with a past history of delayed ability to walk complained of a longstanding impairment of movement of the lower limbs. This had worsened over the last 5 years and she was now unable to walk without the aid of canes. She had experienced muscle cramps and distal paresthesias in the legs for several years. There was no objective sensory deficit, but the leg muscles were moderately atrophic and dorsiflexion at the ankles was weak. Pes cavus and tendon areflexia were also found on clinical examination. The disease was of autosomal dominant transmission. Case 2. A 53-year-old woman5 had a 9-year history of a disorder of gait. Examination revealed a moderate motor deficit in all four limbs. Sensation was intact. There was no pes cavus or distal wasting of the lower limbs. The tendon reflexes were absent. She had three children with no clinical or electrophysiological signs of the disease. One of her children had died several years previously, however, with a similar symptomatology to the mother, but this had never been verified by clinical examination. Both patients exhibited decreased nerve conduction velocities (Table 1). The diagnosis was confirmed by nerve biopsy after informed consent. Samples of sural nerves from each patient were divided into two sections, one of which was then embedded in a hydrophobic resin (Epon). Semithin 1–2-µm-thick sections were cut and stained

Nerve Biopsy.

in 1% paraphenylenediamine solution. Microscopic examination demonstrated a loss of myelinated fibers associated with Schwann cell proliferation in onion-bulb formations around the demyelinated or partially remyelinated axons. In both cases, we observed an abnormal number of tomaculous lesions. Electron microscopic examination of biopsies from both cases did not reveal the defective compaction of myelin reported recently.10 Molecular Analysis. Screening of the genomic DNA with probes for duplication (or deletion) of the PMP 22 gene region of chromosome 17 was negative. However, examination of the P0 gene by polymerase chain reaction (PCR), single strand conformational polymorphism (SSCP), 23 and direct sequencing analysis revealed a point mutation in each patient. The first patient (case 1) had a Leu instead of Ser at position 78 in exon 2, and the second patient (case 2) a Ser instead of Asn 122 in exon 3 at the single glycosylation site.5 Both mutations were located in the extracellular domain of the P0 protein. Both patients were heterozygous for these mutations. Ultrastructural Immunocytochemistry. As the nerve biopsies were obtained several years earlier, nerves were not fixed under ideal conditions for ultracryomicrotomy. 31 The remaining sections of nerve biopsies were embedded in paraffin. Blocks from the 2 biopsied patients and from 2 normal nerves taken from age-matched control patients (who were later diagnosed with amyotrophic lateral sclerosis) were deparaffinized under identical conditions. Nerves had first been embedded in paraffin after a 10% fixative formaldehyde procedure. They were deparaffinized in successive toluene and alcohol baths and then processed for ultracryomi-

Table 1. Electrophysiological examination in the 2 CMT 1B patients for different nerves tested. Motor nerve Peroneal nerve MNCV (m/s)

DML (ms)

F (ms)

Sensory nerve Median nerve

A of CMAP (mV)

MNCV (m/s)

DML (ms)

F (ms)

Sural nerve A of CMAP (mV)

SCV (m/s)

A of SNAP (µV)

Median nerve SCV (m/s)

A of SNAP (µV)

EMG Tibialis anterior muscle

Case 1 Right NO 15 5.4 NO 2 NO NO Fibrillation Left NO 17 5.6 NO 1.5 NO NO potentials Case 2 Right 23 5 NO 2 29 4.4 41.5 3 25 0.5 30 3.5 No fibrillation Left 21.5 5.3 NO 0.7 28 4.5 42 2.5 25 1 29.5 4.5 potentials Normal values* 52 ± 5 5.2 ± 0.5 45.3 ± 5.4 5.2 ± 1.5 54.7 ± 4.2 3.7 ± 0.3 26.6 ± 3 12 ± 2 49.5 ± 4.5 15 ± 2 50.7 ± 4 15.2 ± 3 MNCV, motor nerve conduction velocity; DML, distal motor latency; F, F-wave latency; A of CMAP, amplitude of compound motor action potential; A of SNAP, amplitude of sensory nerve action potential; SCV, sensory conduction velocity; EMG, electromyogram; NO, not obtained. The sensory potentials were antidromic for the sural nerve and orthodromic for the median nerve. *mean ± standard deviation.

100

P0 Expression in CMT 1B Disease

MUSCLE & NERVE

January 1999

crotomy as described elsewhere.30 Myelin proteins had fortunately been conserved as indicated by the presence of reactions with specific antibodies. Immunogold Procedure. P0-7 monoclonal antibody (mab) of the immunoglobulin (Ig)G1 class at 1 : 75,000 dilution (kindly provided by Dr Archelos, Wu¨rzburg, Germany) and PMP 22 polyclonal antibody (kindly provided by Dr Snipes, Montreal, Canada, at 1:1,000 dilution) were used and revealed by IgG-coated colloidal gold particles (12 nm in diameter at 1 :25 dilution). P0-7 mab was characterized by enzyme-linked immunosorbent assay, immunohistochemistry, and Western blotting. It reacted with whole native rat and human P0 and recognized the extracellular domain of the P0 protein (P0-ED).2 PMP 22 antibody was raised against sequence 117 to 132 of the human PMP 22 protein.27 Quantitative Ultrastructural Analysis. P0 and PMP 22 protein immunoreactivity was observed under a CM 10 Philips transmission electron microscope (×28,500 or ×39,000). More than 10 photomicrographs were taken for each patient and analyzed automatically using the Comptex program (specially developed by Klein Industrie, France) and the Inspector program for morphological analysis. The number of gold particles within several regions (areas) of the myelin sheaths was counted automatically and the software calculated the mean particle density per µm2 of tissue after subtracting background values (calculated within the nonmyelinic compartments). The photomicrographs for the particle counts were selected at random. The gold particle density for each CMT 1B patient was compared to those obtained for the control patients (normal nerves) using the Mann–Whitney test. RESULTS Electrophysiology. In both cases, the relatively homogeneous reductions in nerve conduction velocities (Table 1) were indicative of a demyelinating sensorimotor neuropathy and consistent with a genetic origin.21 One case had more severe impairment, so that no responses were recorded in the lower limbs and there was evidence of secondary axonal involvement (fibrillation potentials). The motor nerve conduction velocities were lower in this patient than in the other (15 m/s versus 29 m/s), both markedly lower than those obtained from normal individuals (54.7 ± 4.2 m/s).

We calculated the fiber density and the number of tomaculous fibers in the

Nerve Fiber Quantification.

P0 Expression in CMT 1B Disease

biopsy specimens from both patients. The loss of myelinated fibers led to a significant and similar decrease in fiber density per mm2 (3,426 for case 1 and 3,933 for case 2 versus 7,000–8,000 for age-matched controls). The percentage of tomaculous fibers was 9.5% in the first case and 3% in the second. Immunogold Staining. No statistically significant differences were observed in PMP 22 antibody reactions for either case compared with controls. However, with the P0-7 mab, we noted a marked decrease in P0 expression in both cases (Table 2). Ratios, calculated from the gold particle mean densities and compared to those obtained for each control patient, were 0.60 (P < 0.05) and 0.56 (P < 0.001), respectively (Table 3). The significant decreases in P0 expressivity on ultrathin cryosections of myelin sheaths immunostained with the P0-7 mab in the CMT 1B patients compared with a normal nerve are illustrated in Figure 1. No altered distribution of P0 protein was found in the disease samples. DISCUSSION

We found an approximately twofold decrease in the P0 immunoreactivity of the two CMT 1B patients with respect to age-matched controls. The clinical, electrophysiological, histological, and quantitative ultrastructural immunocytochemical findings revealed a more severe impairment in case 1 than in case 2. For each patient, quantified P0 immunoreactivity correlated with the results of the genetic tests, which identify a point mutation in the myelin P0 gene located on chromosome 1q21–23 within the region corresponding to the extracellular domain (P0-ED). These P0 mutations are autosomal dominant. Each patient presents a normal and a mutant allele, which may account for the approximately 50% reduction in quantified P0 immunoreactivity compared to control nerves, although not for the difference in severity of condition between the two patients. Several possibilities could account for this Table 2. Gold particles density per µm2 using P0 monoclonal and PMP 22 polyclonal antibodies for case 1 and 2 and their age-matched controls. Immunogold density per µm2

Case 1 Control 1 Case 2 Control 2

P0 antibody

PMP 22 antibody

79.53 ± 23.53 131.41 ± 49.20 88.68 ± 31.37 156.97 ± 29.92

3.20 ± 1.67 3.88 ± 1.62 9.12 ± 5.68 12.13 ± 4.01

Values were obtained by an automated analysis and expressed as the mean values ± standard deviation.

MUSCLE & NERVE

January 1999

101

Table 3. Quantitative ultrastructural immunocytochemical study of P0 and PMP 22 proteins in 2 patients affected by a mutation in the extracellular domain (P0-ED) of P0. Patients

P0-ED mutations Exons P0-7 PMP 22

1

2

Ser 78 → Leu 2 0.60* 0.82 NS

Asn 122 → Ser 3 0.56† 0.75 NS

Ratios were calculated from the gold particle densities (in µm2) compared to a normal nerve. NS, no significant result. *P < 0.05. †P < 0.001.

discrepancy. One possibility is that the mutations change the epitopes of the protein, so that the antibody only detects a part of the P0 protein, coded by the normal allele. Another possibility derives from the effects of the P0 mutations. P0 is an integral membrane protein and a major structural protein of the peripheral nervous system. The Ig-like structure of P0-ED is thought to account for its homophilic adhesion properties,9,20,35 and its role in the compaction of peripheral myelin.19 This molecule is also reported to possess neurite outgrowth-promoting properties leading to heterophilic interactions.24 These mutations could lead to conformational changes in P0-ED, which in turn would affect the maturation of the protein28 and so alter homophilic interactions leading to the disrupted compaction of myelin observed in experimental models.11 In our study, the clinical, electrophysiological, and neuropathological differences in severity could not be attributed to differences in ages of the patients, nor to differences in the site and the nature of the mutations. The two mutations are located in two different exons (exon 2 and exon 3), but which code for P0-ED,19 which is believed to confer the adhesive function of the protein 34 by homophilic adhesion. Some mutations leading to a systematic substitution by a cysteine amino acid are marked by a more severe phenotype,22 which could be due to a ‘‘dominant negative effect.’’ This negative effect could be explained by the formation of abnormal P0 complexes with disulfide-bridged aggregates in the extracellular domain of the protein. However, the mutations leading to substitutions like Ser to Leu or Asn to Ser, as observed in our cases, are thought to give rise to a less severe phenotype due to ‘‘loss-offunction allelles.’’33 These mutations could disrupt interactions between P0 proteins, affecting formation of the homotetramer complex,25 which is involved in the compaction of myelin. However, even

102

P0 Expression in CMT 1B Disease

with a detailed electron microscopic study, anomalies of the myelin compaction were not observed. Meijerink et al.22 propose a role for arginine 69 in the formation and compaction of myelin sheaths, as 2 of their patients with a point mutation (Arg 69 → His and Arg 69 → Cys) in exon 3 had uncompacted myelin, and arginine 69 is conserved in a large number of species.12 By contrast, tomaculae were observed in significant numbers. Such lesions were reported by Gabree¨ls-Festen et al.10 from the biopsies of their cases 5, 6, and 7. Ikegami et al.14 report a severe phenotype in children evaluated for Dejerine–Sottas syndrome. These patients are homozygous for a mutation of the myelin P0 gene (Phe 64 deletion). Our patients were heterozygous for the mutations (two substitutions), which could account for the difference in severity. The clinical severity and the number of tomaculae in our patients could not be related directly to the effects of these two mutations. In this respect, mutations in P0-ED have been shown to have different pathological effects.10 The existence of two divergent pathological phenotypes may indicate that some mutations act differently on P0 protein formation or function than others. This is likely to be determined by the site and nature of the mutation in the P0 gene. Interestingly, Bird et al.4 reported variability in severity within the same family, even though each affected person had the same mutation. In view of the absence of any quantitative alteration in the expression of the PMP 22 gene in our 2 patients, PMP 22 is unlikely to be involved in the formation of the tomaculae that are commonly observed in CMT 1B patients. A mechanism for the formation of tomaculae has been proposed by Adlkofer et al.1 in PMP 22-deficient mice. Quantification of P0 immunoreactivity on nerve sections with different anti-P0 antibodies raised against different epitopes of the protein in a larger number of CMT patients should help define the relationship between genotype and phenotype. Correlations between P0 immunoreactivity with the phenotype and the genotype should help elucidate the role of the P0 protein in the pathophysiology of CMT 1 disease. This study was supported by grants from I.N.S.E.R.M. (CRE No. 930809) and Association Franc¸ aise contre les Myopathies (A.F.M.). We thank Dr. Philippe Couratier and Dr. Pierre-Marie Preux for helpful discussions, Laurence Richard and Martine Diot for technical assistance, Dr. S. Jarman for reading the English version of the manuscript, and Nathalie Couade for secretarial assistance.

MUSCLE & NERVE

January 1999

FIGURE 1. Ultrathin cryosections of myelin sheaths immunostained with P0-7 monoclonal antibody and immunogold procedures in the control patient with a normal nerve (A) and a CMT 1B patient (B). Statistically significant differences in immunolabeling were observed. C, collagen; m, myelin. Holes are probably induced by deparaffinization. Bar scale: 0.13 µm.

REFERENCES 1. Adlkofer K, Martini R, Aguzzi A, Zielasek J, Toyka KV, Suter U. Hypermyelination and demyelinating peripheral neuropathy in PMP 22-deficient mice. Nat Genet 1995;11:274–280. 2. Archelos JJ, Roggenbuck K, Schneider-Schaulies J, Linington C, Toyka KV, Hartung H-P. Production and characterization of monoclonal antibodies to the extracellular domain of P0. J Neurosci Res 1993;35:46–53. 3. Bird TD, Ott J, Giblett ER. Evidence for linkage of CharcotMarie-Tooth neuropathy to the Duffy locus on chromosome 1. Am J Hum Genet 1982;34:338–349. 4. Bird TD, Kraft GH, Lipe HP, Kenney KL, Sumi SM. Clinical and pathological phenotype of the original family with Charcot-Marie-Tooth type 1B: a 20-year study. Ann Neurol 1997; 41:463–469. 5. Blanquet-Grossard F, Pham-Dinh D, Dautigny A, Latour P, Bonnebouche C, Diraison P, Chapon F, Chazot G, Vandenberghe A. Charcot-Marie-Tooth type 1B neuropathy: a mutation at the single glycosylation site in the major peripheral myelin glycoprotein P0. Hum Mut 1996;8:185–186. 6. Chance PF, Bird TD, O’Connel P, Lipe H, Lalouel JM, Leppert M. Genetic linkage and heterogeneity in type I CharcotMarie-Tooth disease (hereditary motor and sensory neuropathy I). Am J Hum Genet 1990;47:915–925. 7. Chance PF, Fischbeck KH. Molecular genetics of CharcotMarie-Tooth disease and related neuropathies. Hum Mol Genet 1994;3:1603–1607. 8. Dyck PJ, Chance P, Lebo R, Carney JA. Hereditary motor and

P0 Expression in CMT 1B Disease

9.

10.

11.

12.

13.

14.

sensory neuropathies. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF, editor. Peripheral neuropathy, 3rd ed. Philadelphia: Saunders; 1993. p 1094–1136. Filbin MT, Walsh FS, Trapp BD, Pizzey JA, Tennekoon GI. Role of myelin P0 protein as a homophilic adhesion molecule. Nature 1990;344:871–872 Gabreëls-Festen AAWM, Hoogendijk JE, Meijerink PHS, Gabreëls FJM, Bolhuis PA, van Beersum S, Kulkens T, Nelis E, Jennekens FGI, de Visser M, van Engelus BGM, Van Broeckhoven C, Mariman ECM. Two divergent types of nerve pathology in patients with different P0 mutations in Charcot-MarieTooth disease. Neurology 1996;47:761–765. Giese KP, Martini R, Lemke G, Soriano P, Schachner M. Mouse P0 gene disruption leads to hypomyelination, abnormal expression of recognition molecules and degeneration of myelin and axons. Cell 1992;71:565–576. Hayasaka K, Nanao K, Tahara M, Sato W, Takada G, Miura M, Uyemura K. Isolation and sequence determination of cDNA encoding the major structural protein of human peripheral myelin. Biochem Biophys Res Commun 1991;180:515–518. Hayasaka K, Himoro M, Sato W, Takada G, Uyemura K, Shimizu N, Bird TD, Conneally PH, Chance PF. CharcotMarie-Tooth neuropathy type1B is associated with mutations of the myelin P0 gene. Nat Genet 1993;5:31–34. Ikegami T, Nicholson G, Ikeda H, Ishida A, Johnston H, Wise G, Ouvrier R, Hayasaka K. A novel homozygous mutation of the myelin P0 gene producing Dejerine-Sottas disease (hereditary motor and sensory neuropathy type III). Biochem Biophys Res Commun 1996;222:107–110.

MUSCLE & NERVE

January 1999

103

15. Ionasescu VV. Charcot-Marie-Tooth neuropathies: from clinical description to molecular genetics. Muscle Nerve 1995;18: 267–275. 16. Kulkens T, Bolhuis PA, Wolterman RA, Kemps S, te Nijenhuis S, Valentijn LJ, Hensel GW, Jennekens FG, de Visser M, Hoogendijk JE, Baas F. Deletion of the serine 34 codon from the major peripheral myelin protein P0 gene in CharcotMarie-Tooth disease type 1b. Nat Genet 1993;5:35–39. 17. Latour P, Blanquet F, Nelis E, Bonnebouche C, Chapon F, Diraison P, Ollagnon E, Dautigny A, Pham-Dinh D, Chazot G, Boucherat M, Van Broeckhoven C, Vandenberghe A. Mutations in the myelin protein zero gene associated with CharcotMarie-Tooth disease type 1B. Hum Mutat 1995;6:50–54. 18. Lebo RV, Chance PF, Dyck PJ, Redila-Flores MY, Lynch ED, Golbus MS, Bird TD, King MC, Anderson LA, Hall J, Wiegant J, Jiang Z, Dazin PF, Lovelace R, Conneally PM. Chromosome 1 Charcot-Marie-Tooth disease (CMT 1b) locus in the Fcg receptor gene region. Hum Genet 1991;88:1–12. 19. Lemke G, Lamar E, Patterson J. Isolation and analysis of the gene encoding peripheral myelin protein zero. Neuron 1988; 1:73–83. 20. Lemke G. Myelin and myelination. In: Hall ZW, editor. An introduction to molecular neurobiology. Sunderland, MA: Sinauer Associates; 1992. p 281–309. 21. Lewis RA, Summer AJ. The electrodiagnostic distinctions between chronic familial and acquired demyelinating neuropathies. Neurology 1982;32:592–596. 22. Meijerink PHS, Hoogendijk JE, Gabree¨ls-Festen AAWM, Zorn I, Veldman H, Baas F, De Visser M, Bolhuis PA. Clinically distinct codon 69 mutations in major myelin protein zero in demyelinating neuropathies. Ann Neurol 1996;40:672–675. 23. Nelis E, Timmerman V, De Jonghe P, Vandenberghe A, Pham-Dinh D, Dautigny A, Martin TJ, Van Broeckhoven C. Rapid screening of myelin genes in CMT 1 patients by SSCP analysis: identification of new mutations and polymorphisms in the P0 gene. Hum Genet 1994;94:653–657. 24. Schneider-Schaulies J, von Brunn A, Schachner M. Recombinant peripheral myelin protein P0 confers both adhesion and

104

P0 Expression in CMT 1B Disease

25.

26. 27. 28. 29. 30. 31.

32.

33.

34. 35.

neurite outgrowth-promoting properties. J Neurosci Res 1990;27:286–297. Shapiro L, Doyle JP, Hensley P, Colman DR, Hendrickson WA. Crystal structure of the extracellular domain from P0, the major structural protein of peripheral nerve myelin. Neuron 1996;17:435–449. Skre H. Genetic and clinical aspects of Charcot-Marie-Tooth disease. Clin Genet 1974;6:98–118. Snipes GJ, Suter U, Shooter EM. Human peripheral myelin protein-22 carries the L2/HNK1 carbohydrate adhesion epitope. J Neurochem 1993;61:1961–1964. Snipes GJ, Suter U. Molecular basis of common hereditary motor and sensory neuropathies in humans and in mouse models. Brain Pathol 1995;5:233–247. Suter U, Snipes GJ. Biology and genetics of hereditary motor and sensory neuropathies. Annu Rev Neurosci 1995;18:45–75. Tokuyasu KT. A technique for ultracryotomy of cell suspensions and tissues. J Cell Biol 1973;57:551–565. Vallat JM, Sindou P, Preux PM, Tabaraud F, Milor AM, Couratier P, LeGuern E, Brice A. Ultrastructural PMP 22 expression in inherited demyelinating neuropathies. Ann Neurol 1996;39:813–817. Vance JM, Barker D, Yamaoka LH, Stajich JM, Loprest L, Hung WY, Fischbeck K, Roses AD, Pericak-Vance MA. Localization of Charcot-Marie-Tooth disease type 1a (CMT 1a) to chromosome 17p11.2. Genomics 1991;9:623–628. Warner LE, Hilz MJ, Appel SH, Killian JM, Kolodny EH, Karpati G, Carpenter S, Watters GV, Wheeler C, Witt D, Bodell A, Nelis E, Van Broeckhoven CV, Lupski JR. Clinical phenotypes of different MPZ (P0) mutations may include Charcot-Marie-Tooth type 1B, Dejerine-Sottas, and congenital hypomyelination. Neuron 1996;17:451–460. Williams AF, Barclay AN. The immunoglobulin super familydomains for cell-surface recognition. Annu Rev Immunol 1988;6:381–405. Zhang K, Filbin MT. Formation of a disulfide bond in the immunoglobulin domain of the myelin P0 protein is essential for its adhesion. J Neurochem 1994;623:367–370.

MUSCLE & NERVE

January 1999