So far, mutations in the human COL3A1 gene have been associated with the predominantly inherited Ehlers-Danlos syndrome (EDS), vascular type. Genotype -phenotype correlation perspectives collapsed, as haploinsufficiency, which was long suggested to confer a milder or unrecognized phenotype, was reported in four patients with a phenotype similar to that of vascular EDS. Here, we study a case of recessive EDS in a young consanguineous girl of healthy parents. She fulfilled the vascular EDS criteria for laboratory testing. Total sequencing of COL3A1 cDNA identified a homozygous nucleotide duplication (c.479dupT) resulting in a premature termination codon (p.Lys161GlnfsX45). Studies in genomic DNA showed that this mutation was inherited from each parent. The expression analysis (RT-PCR, quantitative-PCR, immunohistochemistry, WB) showed strong mRNA decay and an absence of type III collagen in the proband. The expected COL3A1 haploinsufficiency in her healthy ascendants did not lead to the manifestations of vascular EDS. This case provides evidence of a stochastic effect of COL3A1 haploinsufficiency in humans, which could be explained by the relation between nonsense-mediated mRNA decay efficiency and the resulting dominant-negative effect depending on the position of the mutation and/or modifying factors. It opens up new perspectives for the understanding of COL3A1 genotype -phenotype correlations, which is required while considering targeted therapy.

No abstract

Marfan syndrome (MFS) is a rare connective tissue disorder caused by mutation in the gene encoding the extracellular matrix protein fibrillin-1 (FBN1), leading to transforming growth factor-beta (TGF-b) signaling dysregulation. Although decreased axial and peripheral bone mineral density (BMD) has been reported in adults with MFS, data about the evolution of bone mass during childhood and adolescence are limited. The aim of the present study was to evaluate bone and muscle characteristics in children, adolescents, and young adults with MFS. The study population included 48 children and young adults (22 girls) with MFS with a median age of 11.9 years (range 5.3 to 25.2 years). The axial skeleton was analyzed at the lumbar spine using dual-energy X-ray absorptiometry (DXA), whereas the appendicular skeleton (hand) was evaluated using the BoneXpert system (with the calculation of the Bone Health Index). Muscle mass was measured by DXA. Compared with healthy age-matched controls, bone mass at the axial and appendicular levels and muscle mass were decreased in children with MFS and worsened from childhood to adulthood. Vitamin D deficiency (<50 nmol/L) was found in about a quarter of patients. Serum vitamin D levels were negatively correlated with age and positively correlated with lumbar spine areal and volumetric BMD. Lean body mass (LBM) Z-scores were positively associated with total body bone mineral content (TB-BMC) Z-scores, and LBM was an independent predictor of TB-BMC values, suggesting that muscle hypoplasia could explain at least in part the bone loss in MFS. Patients with a FBN1 premature termination codon mutation had a more severe musculoskeletal phenotype than patients with an inframe mutation, suggesting the involvement of TGF-b signaling dysregulation in the pathophysiologic mechanisms. In light of these results, we recommend that measurement of bone mineral status should be part of the longitudinal clinical investigation of MFS children.

Marfan syndrome (MFS) is a heritable connective tissue disorder (HCTD) caused by pathogenic variants in FBN1 that frequently occur de novo. Although individuals with somatogonadal mosaicisms have been reported with respect to MFS and other HCTD, the overall frequency of parental mosaicism in this pathology is unknown. In an attempt to estimate this frequency, we reviewed all the 333 patients with a disease-causing variant in FBN1. We then used direct sequencing, combined with High Resolution Melting Analysis, to detect mosaicism in their parents, complemented by NGS when a mosaicism was objectivized. We found that (1) the number of apparently de novo events is much higher than the classically admitted number (around 50% of patients and not 25% as expected for FBN1) and (2) around 5% of the FBN1 disease-causing variants were not actually de novo as anticipated, but inherited in a context of somatogonadal mosaicisms revealed in parents from three families. High Resolution Melting Analysis and NGS were more efficient at detecting and evaluating the level of mosaicism compared to direct Sanger sequencing. We also investigated individuals with a causal variant in another gene identified through our “aortic diseases genes” NGS panel and report, for the first time, on an individual with a somatogonadal mosaicism in COL5A1. Our study shows that parental mosaicism is not that rare in Marfan syndrome and should be investigated with appropriate methods given its implications in patient’s management.

BackgroundPathogenic SMAD3 variants are responsible for a cardiovascular phenotype, mainly thoracic aortic aneurysms and dissections. Precocious identification of the vascular risk such as aortic dilatation in mutated patients has a major impact in terms of management, particularly to avoid dissection and sudden death. These vascular damages are classically associated with premature osteoarthritis and skeletal abnormalities. However, variable expressivity and incomplete penetrance are common with SMAD3 variants.MethodsTo investigate the clinical variability observed within SMAD3 patients, we reviewed the phenotypic and genetic data of 22 new patients from our Centre and of 133 patients reported in the literature. From this cohort of 155 mutated individuals, we first aimed to delineate an estimated frequency of the main clinical signs associated with SMAD3 pathogenic variants and, then, to look for genotype‐phenotype correlations, mainly to see if the aortic phenotype (AP) could be predicted by the SMAD3 variant type.ResultsWe showed, herein, the absence of correlation between the SMAD3 variant type and the occurrence of an AP in patients.ConclusionTherefore, this report brings additional data for the genotype‐phenotype correlations of SMAD3 variants and the need to explore in more detail the effects of genetic modifiers that could influence the phenotype.