Clinical and mutational spectrum of Mowat-Wilson syndrome.

Mowat-Wilson Syndrome is a recently delineated mental retardation syndrome usually associated with multiple malformations and a recognizable facial phenotype caused by defects of the transcriptional repressor ZFHX1B. To address the question of clinical and mutational variability, we analysed a large number of patients with suspected Mowat-Wilson Syndrome (MWS). Without prior knowledge of their mutational status, 70 patients were classified into "typical MWS", "ambiguous" and "atypical" groups according to their facial phenotype. Using FISH, qPCR and sequencing, ZFHX1B deletions, splice site or truncating mutations were detected in all 28 patients classified as typical MWS. No ZFHX1B defect was apparent in the remaining 15 cases with ambiguous facial features or in the 27 atypical patients. Genotype-phenotype analysis confirmed that ZFHX1B deletions and stop mutations result in a recognizable facial dysmorphism with associated severe mental retardation and variable malformations such as Hirschsprung disease and congenital heart defects. Our findings indicate that structural eye anomalies such as microphthalmia should be considered as part of the MWS spectrum. We also show that agenesis of the corpus callosum and urogenital anomalies (especially hypospadias) are significant positive predictors of a ZFHX1B defect. Based on our observation of affected siblings and the number of MWS cases previously reported, we suggest a recurrence risk of around 1%. The lack of missense mutations in MWS and MWS-like patients suggests there may be other, as yet unrecognized phenotypes, associated with missense mutations of this transcription factor.

Segmental haplosufficiency: transmitted deletions of 2p12 include a pancreatic regeneration gene cluster and have no apparent phenotypic consequences.

Segmental aneuploidy usually has phenotypic consequences but unbalanced rearrangements without phenotypic consequences have also been reported. In particular, harmless deletions of G-dark bands 5p14 and 16q21 have each been found in more than one independent family. Here, we report two families that were ascertained at prenatal diagnosis and had similar overlapping deletions that removed most of the gene poor G-dark band 2p12. PCR mapping showed that the deletions had a minimum size of 6.1 and 6.9 Mb with at least 13 hemizygous loci including a cluster of six pancreatic islet-regenerating genes. These deletions had no apparent phenotypic consequences in eight family members. In contrast, a third family was ascertained through a child with Wilm's tumour; both the child and his mother had more proximal deletions, developmental delay and some dysmorphic features. The deletion had a minimum size of 5.7 Mb and extended into the gene-rich area of 2p11.2. These results are consistent with the idea that there may be segments of the genome that are consistently haplosufficient. The introduction of higher resolution methods of dosage analysis into diagnostic laboratories is already revealing more transmitted abnormalities of uncertain significance. As a result, published cases of transmitted imbalances have been collected as a guide to the possible significance of such findings in the future (see the 'Chromosome Anomaly Collection' at www.som.soton.ac.uk/research/geneticsdiv).