Pitt-Hopkins syndrome (PTHS), characterized by severe intellectual disability and typical facial gestalt, is part of the clinical spectrum of Rett-like syndromes. TCF4, encoding a basic helix-loop-helix (bHLH) transcription factor, was identified as the disease-causing gene with de novo molecular defects. While PTHS appears to be a recognizable clinical entity, it seems to remain underdiagnosed, especially when facial gestalt is less typical. With the aim to facilitate the diagnosis of PTHS and to increase its rate and specificity, we have investigated 33 novel patients and defined a Clinical Diagnosis Score. Analysis of 112 individuals (79 previously reported and 33 novel patients) allowed us to delineate the TCF4 mutational spectrum, with 40% point mutations, 30% small deletions/insertions, and 30% deletions. Most of these were private mutations and generated premature stop codons. Missense mutations were localized in the bHLH domain, which is a mutational hotspot. No obvious difference was observed between patients harboring truncating, missense mutations, or deletions, further supporting TCF4 haploinsufficiency as the molecular mechanism underlying PTHS. In this study, we have summarized the current knowledge of TCF4 molecular pathology, reported all the mutations in the TCF4 database (http://www.LOVD.nl/TCF4), and present a novel and comprehensive diagnostic strategy for PTHS.

GPR56 mutations cause an autosomal recessive polymicrogyria syndrome that has distinctive radiological features combining bilateral frontoparietal polymicrogyria, white matter abnormalities and cerebellar hypoplasia. Recent investigations of a GPR56 knockout mouse model suggest that bilateral bifrontoparietal polymicrogyria shares some features of the cobblestone brain malformation and demonstrate that loss of GPR56 leads to a dysregulation of the maintenance of the pial basement membrane integrity in the forebrain and the rostral cerebellum. In light of these findings and other data in the literature, this study aimed to refine the clinical features with the first description of a foetopathological case and to define the range of cobblestonelike features in GPR56 bilateral bifrontoparietal polymicrogyria in a sample of 14 patients. We identified homozygous GPR56 mutations in 14 patients from eight consanguineous families with typical bilateral bifrontoparietal polymicrogyria and in one foetal case, out of 30 patients with bifrontoparietal polymicrogyria referred for molecular screening. The foetal case, which was terminated at 35 weeks of gestation in view of suspicion of Walker Warburg syndrome, showed a cobblestone-like lissencephaly with a succession of normal, polymicrogyric and 'cobblestone-like' cortex with ectopic neuronal overmigration, agenesis of the cerebellar vermis and hypoplastic cerebellar hemispheres with additional neuronal overmigration in the pons and the cerebellar cortex. The 14 patients with GPR56 mutations (median 8.25 years, range 1.5-33 years) were phenotypically homogeneous with a distinctive clinical course characterized by pseudomyopathic behaviour at onset that subsequently evolved into severe mental and motor retardation. Generalized seizures (12/14) occurred later with onset ranging from 2.5 to 10 years with consistent electroencephalogram findings of predominantly anterior bursts of low amplitude a-like activity. Neuroimaging demonstrated a common phenotype with bilateral frontoparietally predominant polymicrogyria (13/13), cerebellar dysplasia with cysts mainly affecting the superior vermis (11/13) and patchy to diffuse myelination abnormalities (13/13). Additionally, the white matter abnormalities showed a peculiar evolution from severe hypomyelination at 4 months to patchy lesions later in childhood. Taken as a whole, these observations collectively demonstrate that GPR56 bilateral bifrontoparietal polymicrogyria combines all the features of a cobblestone-like lissencephaly and also suggest that GRP56-related defects produce a phenotypic continuum ranging from bilateral bifrontoparietal polymicrogyria to cobblestone-like lissencephaly.

Recently, pathogenic variants in the MLL2 gene were identified as the most common cause of Kabuki (Niikawa-Kuroki) syndrome (MIM#147920). To further elucidate the genotype-phenotype correlation, we studied a large cohort of 86 clinically defined patients with Kabuki syndrome (KS) for mutations in MLL2. All patients were assessed using a standardized phenotype list and all were scored using a newly developed clinical score list for KS (MLL2-Kabuki score 0-10). Sequencing of the full coding region and intron-exon boundaries of MLL2 identified a total of 45 likely pathogenic mutations (52%): 31 nonsense, 10 missense and four splice-site mutations, 34 of which were novel. In five additional patients, novel, i.e. non-dbSNP132 variants of clinically unknown relevance, were identified. Patients with likely pathogenic nonsense or missense MLL2 mutations were usually more severely affected (median 'MLL2-Kabuki score' of 6) as compared to the patients without MLL2 mutations (median 'MLL2-Kabuki score' of 5), a significant difference (p < 0.0014). Several typical facial features such as large dysplastic ears, arched eyebrows with sparse lateral third, blue sclerae, a flat nasal tip with a broad nasal root, and a thin upper and a full lower lip were observed more often in mutation positive patients.

Complex cortical malformations associated with mutations in tubulin genes are commonly referred to as “Tubulinopathies”. To further characterize the mutation frequency and phenotypes associated with tubulin mutations, we studied a cohort of 60 foetal cases. Twenty-six tubulin mutations were identified, of which TUBA1A mutations were the most prevalent (19 cases), followed by TUBB2B (6 cases) and TUBB3 (one case). Three subtypes clearly emerged. The most frequent (n = 13) was microlissencephaly with corpus callosum agenesis, severely hypoplastic brainstem and cerebellum. The cortical plate was either absent (6/13), with a 2–3 layered pattern (5/13) or less frequently thickened (2/13), often associated with neuroglial overmigration (4/13). All cases had voluminous germinal zones and ganglionic eminences. The second subtype was lissencephaly (n = 7), either classical (4/7) or associated with cerebellar hypoplasia (3/7) with corpus callosum agenesis (6/7). All foetuses with lissencephaly and cerebellar hypoplasia carried distinct TUBA1A mutations, while those with classical lissencephaly harbored recurrent mutations in TUBA1A (3 cases) or TUBB2B (1 case). The third group was polymicrogyria-like cortical dysplasia (n = 6), consisting of asymmetric multifocal or generalized polymicrogyria with inconstant corpus callosum agenesis (4/6) and hypoplastic brainstem and cerebellum (3/6). Polymicrogyria was either unlayered or 4-layered with neuronal heterotopias (5/6) and occasional focal neuroglial overmigration (2/6). Three had TUBA1A mutations and 3 TUBB2B mutations. Foetal TUBA1A tubulinopathies most often consist in microlissencephaly or classical lissencephaly with corpus callosum agenesis, but polymicrogyria may also occur. Conversely, TUBB2B mutations are responsible for either polymicrogyria (4/6) or microlissencephaly (2/6).Electronic supplementary materialThe online version of this article (doi:10.1186/2051-5960-2-69) contains supplementary material, which is available to authorized users.

EXTL3 regulates the biosynthesis of heparan sulfate (HS), important for both skeletal development and hematopoiesis, through the formation of HS proteoglycans (HSPGs). By whole-exome sequencing, we identified homozygous missense mutations c.1382C>T, c.1537C>T, c.1970A>G, and c.2008T>G in EXTL3 in nine affected individuals from five unrelated families. Notably, we found the identical homozygous missense mutation c.1382C>T (p.Pro461Leu) in four affected individuals from two unrelated families. Affected individuals presented with variable skeletal abnormalities and neurodevelopmental defects. Severe combined immunodeficiency (SCID) with a complete absence of T cells was observed in three families. EXTL3 was most abundant in hematopoietic stem cells and early progenitor T cells, which is in line with a SCID phenotype at the level of early T cell development in the thymus. To provide further support for the hypothesis that mutations in EXTL3 cause a neuro-immuno-skeletal dysplasia syndrome, and to gain insight into the pathogenesis of the disorder, we analyzed the localization of EXTL3 in fibroblasts derived from affected individuals and determined glycosaminoglycan concentrations in these cells as well as in urine and blood. We observed abnormal glycosaminoglycan concentrations and increased concentrations of the non-sulfated chondroitin disaccharide D0a0 and the disaccharide D0a4 in serum and urine of all analyzed affected individuals. In summary, we show that biallelic mutations in EXTL3 disturb glycosaminoglycan synthesis and thus lead to a recognizable syndrome characterized by variable expression of skeletal, neurological, and immunological abnormalities.