Summary The inherited disorders of CNS myelin formation represent a heterogeneous group of leukodystrophies. The proteolipoprotein (PLP1) gene has been implicated in two X‐linked forms, Pelizaeus‐Merzbacher disease (PMD) and spastic paraplegia type 2, and the gap junction protein α12 (GJA12) gene in a recessive form of PMD. The myelin basic protein (MBP) gene, which encodes the second most abundant CNS myelin protein after PLP1, presents rearrangements in hypomyelinating murine mutants and is always included in the minimal region deleted in 18q‐ patients with an abnormal hypomyelination pattern on cerebral MRI. In this study, we looked at the genomic copy number at the Golli‐MBP locus in 195 patients with cerebral MRI suggesting a myelin defect, who do not have PLP1 mutation. Although preliminary results obtained by FISH suggested the duplication of Golli‐MBP in 3 out of 10 patients, no abnormal gene quantification was found using Quantitative Multiplex PCR of Short Fluorescent fragments (QMPSF), Multiplex Amplifiable Probe Hybridization (MAPH), or another FISH protocol using directly‐labelled probes. Pitfalls and interest in these different techniques to detect duplication events are emphasised. Finally, the study of this large cohort of patients suggests that Golli‐MBP deletion or duplication is rarely involved in inherited defects of myelin formation.

The proteolipid protein 1 (PLP1) gene is known to be mutated in the X-linked disorders of myelin formation Pelizaeus-Merzbacher disease (PMD) and spastic paraplegia type 2. The most commonly found PLP1 mutations are gene duplications (60-70%) and point mutations (20%). About 20% of patients with a PMD phenotype do not present identified PLP1 mutation, thus suggesting genetic heterogeneity and/or undetected PLP1 abnormalities. Except the recently described MLPA screening the seven exonic regions, the currently used techniques to quantify PLP1 gene copy number do not investigate small intragenic PLP1 rearrangements. Using the multiplex amplifiable probe hybridization (MAPH) technique, we looked simultaneously for intragenic rearrangements along the PLP1 gene (exonic and regulatory regions) and for rearrangements in the GPM6B candidate gene (a member of the proteolipid protein family). We tested 262 hypomyelinating patients: 56 PLP1 duplicated patients, 1 PLP1 triplicated patient, and 205 patients presenting a leukodystrophy of undetermined origin with brain MRI suggesting a defect in myelin formation. Our results show that MAPH is an alternative reliable technique for diagnosis of PLP1 gene copy number. It allows us (1) to demonstrate that all PLP1 duplications previously found encompass the whole gene, (2) to establish that copy number changes in GPM6B and intragenic duplications of PLP1 are very unlikely to be involved in the etiology of UHL, and (3) to identify one partial triplication and two partial deletions of PLP1 in patients presenting with a PMD phenotype.