© 1994 Nature Publishing Group http://www.nature.com/naturegenetics • correspondence 334 both mothers were measured in duplicate on three occasions. Both showed a duplication of PLP (Fig. 1). No sequence changes were found when PCR products from each exon were screened using single strand conformation polymorphism analysis. We also found that the PLP genes were intact after duplication. A restriction map produced using PstI, EcoRI and XbaI (ref. 3) covering the entire coding region revealed no altered fragments in either individual or their mothers. Thus, the breakpoints fall outside the gene which must, therefore, be totally duplicated. One boy (JH) was more severely affected than the other. This may be due to the differing extent of the duplicated region which we have not yet been able to define. A suggestion that increased dosage of PLPmay cause PMD came from a patient with a large, cytogenetically visible, de novo duplication ofXq21- 22 (ref. 8). The boy showed multiple abnormalities: muscular hypotonia, growth retardation, cryptoorchidism and a severe generalized disorder of myelination suggestive of PMD at autopsy. Dosage studies showed PLP to be within this large duplicated region 9 • These cases form a strong parallel with CMTlA, which generally in- volves a 1.5 Mb duplication of DNA including PMP-22. Point mutations of PMP-22 have also been found in patients with CMTlA (refs 10,11) and decreased nerve conduction velocity, characteristic ofCMTl has been observed in three individuals with larger, cytogenetically visible duplications 12 -u. One of these patients 1 3, who had a complete trisomy for chromosome 17p resulting from an unbalanced trans- location t(14;17)(pll;pll), had clinical symptoms of a peripheral neuropathy as well as the reduced nerve conduction velocities. There are parallels, therefore, between CMTlA and PMD in all three respects: point mutations, dupli- cations and large cytogenetically visible duplications. Why do point mutations and increased dosage produce such similar phenotypes? Neither mother of the two boys with PMD has any symptoms despite an increased dosage themselves. This may be due to selective survival of cells in which the normal chromosome is active. A possible explanation for the similar phenotypes is that both proteins act as part of a multi-component unit within myelin and the stoichiometry between the components is critical. However, no interacting molecules have yet been identified. It would be most interesting to establish whether Chromosome 4p16 and osteochondroplasias Sir-We read with interest the recent articles in Nature Genetics by Velinov et aL I and Le Merrer et al. 2 establishing linkage of achondroplasia (ACH) 1 .2 andhypochondroplasia (HCH) 2 near the telomere of chromosome 4p. We proposed recentlya tentative location in 4pl6 or 4q13 of the gene(s) responsible for osteochondro- dysplasias3. This was based on our observation ofa pericentric inversion in chromosome 4, with breakpoints at pl6 and ql3.2, in a patient with characteristic skeletal and extra- skeletal manifestations ofalethal short rib-polydactyly syndrome (SRPS), one form of autosomal recessive osteochondrodysplasia. Rivas et al. 4 noted a similar chromosome 4 inversion (breakpoint at 4pl6) in a family with thanatophoric dysplasia, an autosomal dominant osteo- chondrodysplasia. Thus, chromo- some 4pl6 appears to be related to other osteochondrodysplasias in addition to ACH and HCH. These data raise important clinical and aetiologic questions about osteochondrodysplasias. The association of chromosomal re- arrangements involving 4pl6 with both SRPS 3 and thanatophoric dysplasia 4 raises the possibility of a close genetic basis for these two clinically distinct forms of osteochondrodysplasia. Evidence that ACH and HCH, also well distinguishable clinically, are due to defects in 4p 16 strongly supports this possibility.We suggest that the distal short arm of chromosome 4 may contain several genes for other similar neurological disorders arise from mutations of the other components. David Ellis Sue Malcolm Molecular Genetics Unit, Institute of Child Health, 30 Guilford Street, London WCIN lEH, UK 1. Auborg. P. Nature Genet. s, 106-106 (1993). 2. Pratt,V.M. eta/.Am. J.med. Genet. 38, 136-139 (1991). 3. Hudson, LD., Puckett, C., Berndt, J ., Chan, J. & Gencic, S. Proc. natn. Acad. Sci. U.S.A 88, 812&-8131 (1989). 4. Pham·Dinh, D. et a/. Proc. natn. Acad. Sci. U.S.A. 88, 7562-7566 (1991). 5. Timmerman, V. eta/. Nature Genet 1, 171-175 (1992). 6. Valentijn, LJ. et a/. Nature Genet. 1, 166-170 (1992). 7. Matsunaml, N. et al . Nature Genet. 1, 176-179 (1992). 8. Cremers, F.P.M . et a/. Hum. Genet. 71, 23-27 (1987). 9. Cremers, F.P.M. et Bl . Am. J. hum. Genet. 43, 452-461 (1988). 10. Valentijn, LJ. et al. Nature Genet. 2, 288-291 (1992). 11 . Roa, B. etal . lllewEng/. J. Med. 329,96-101 (1993). 12. Lupski, J.R. eta/. Nature Genet. 1, 29-33 (1992). 13. Chance, P.F. et al. Neurology 42, 2295-2299 (1992). 14. Upadhyaya, M. et al. Hum. Genet. 91. 392--394 (1993) . Acknowledgements We thank M. Baraitser, H. Hughes and A. Wilkie for referring patients and for helpful discussion. This work was supported by the Child Health Research Appeal Trust and the Research Trust for Metabolic Diseases in Children. osteochondrodysplasia, or (less likely) that these conditions result from different mutations of the same gene. Miguel Urioste Maria Luisa Martinez-Frias Eva Bermejo Amelia Villa Hospital Universitario San Carlos, Departamento de Farmacologia, ECEMC, Facultad de Medicina, Universidad Complutense, 28040 Madrid, Spain Nicolas Jimenez Dolores Romero Carmen Nieto Hospital General, Segovia, Spain 1. Vetinvov, M. et a/. Nature Genet. 8, 314-317 (1994). 2. Le Merrer, M. et a/. Nature Genet. 8, 318-321 (1994). 3. Urioste, M. et al . Am. J. med. Genet. 49, 94-97 (1994). 4. Rivas. F. et al. C//n. Genet. 31, 97-101 (1987). Nature Genetics volume 6 april 1994