At chromosome 11p15.5, the imprinting centre 1 (IC1) controls the parent of origin-specific expression of the IGF2 and H19 genes. The 5 kb IC1 region contains multiple target sites (CTS) for the zinc-finger protein CTCF, whose binding on the maternal chromosome prevents the activation of IGF2 and allows that of H19 by common enhancers. CTCF binding helps maintaining the maternal IC1 methylation-free, whereas on the paternal chromosome gamete-inherited DNA methylation inhibits CTCF interaction and enhancer-blocking activity resulting in IGF2 activation and H19 silencing. Maternally inherited 1.4–2.2 kb deletions are associated with methylation of the residual CTSs and Beckwith–Wiedemann syndrome, although with different penetrance and expressivity. We explored the relationship between IC1 microdeletions and phenotype by analysing a number of previously described and novel mutant alleles. We used a highly quantitative assay based on next generation sequencing to measure DNA methylation in affected families and analysed enhancer-blocking activity and CTCF binding in cultured cells. We demonstrate that the microdeletions mostly affect IC1 function and CTCF binding by changing CTS spacing. Thus, the extent of IC1 inactivation and the clinical phenotype are influenced by the arrangement of the residual CTSs. A CTS spacing similar to the wild-type allele results in moderate IC1 inactivation and is associated with stochastic DNA methylation of the maternal IC1 and incomplete penetrance. Microdeletions with different CTS spacing display severe IC1 inactivation and are associated with IC1 hypermethylation and complete penetrance. Careful characterization of the IC1 microdeletions is therefore needed to predict recurrence risks and phenotypical outcomes.

The parent of origin-dependent expression of the IGF2 and H19 genes is controlled by the imprinting centre 1 (IC1) consisting in a methylation-sensitive chromatin insulator. Deletions removing part of IC1 have been found in patients affected by the overgrowth- and tumour-associated Beckwith-Wiedemann syndrome (BWS). These mutations result in the hypermethylation of the remaining IC1 region, loss of IGF2/H19 imprinting and fully penetrant BWS phenotype when maternally transmitted. We now report that 12 additional cases with IC1 hypermethylation have a similar clinical phenotype but showed neither a detectable deletion nor other mutation in the local vicinity. Likewise, no IC1 deletion was detected in 40 sporadic non-syndromic Wilms' tumours. A detailed analysis of the BWS patients showed that the hypermethylation variably affected the IC1 region and was generally mosaic. We observed that all these cases were sporadic and in at least two families affected and unaffected members shared the same maternal IC1 allele but not the abnormal maternal chromosome epigenotype. Furthermore, the chromosome with the imprinting defect derived from either the maternal grandfather or maternal grandmother. Overall, these results indicate that methylation-imprinting defects at the IGF2-H19 locus can result from inherited mutations of the IC and have high recurrence risk or arise independently from the sequence context and generally not transmitted to the progeny. Despite these differences, the epigenetic abnormalities are usually present in the patients in the mosaic form and probably acquired by post-zygotic de novo methylation. Distinguishing between these two groups of cases is important for genetic counselling.

Type I disorders of glycosylation (CDG), the most frequent of which is phosphomannomutase 2 (PMM2-CDG), are a group of diseases causing the incomplete N-glycosylation of proteins. PMM2-CDG is an autosomal recessive disease with a large phenotypic spectrum, and is associated with mutations in the PMM2 gene. The biochemical analysis of mutants does not allow a precise genotype–phenotype correlation for PMM2-CDG. PMM2 is very tolerant to missense and loss of function mutations, suggesting that a partial deficiency of activity might be beneficial under certain circumstances. The patient phenotype might be influenced by variants in other genes associated with the type I disorders of glycosylation in the general population.

BackgroundFabry disease is a rare disorder caused by a large variety of mutations in the gene encoding lysosomal alpha-galactosidase. Many of these mutations are unique to individual families. Fabry disease can be treated with enzyme replacement therapy, but a promising novel strategy relies on small molecules, so called "pharmacological chaperones", which can be administered orally. Unfortunately only 42% of genotypes respond to pharmacological chaperones.ResultsA procedure to predict which genotypes responsive to pharmacological chaperones in Fabry disease has been recently proposed. The method uses a position-specific substitution matrix to score the mutations. Using this method, we have screened public databases for predictable responsive cases and selected nine representative mutations as yet untested with pharmacological chaperones. Mutant lysosomal alpha galactosidases were produced by site directed mutagenesis and expressed in mammalian cells. Seven out of nine mutations responded to pharmacological chaperones. Nineteen other mutations that were tested with pharmacological chaperones, but were not included in the training of the predictive method, were gathered from literature and analyzed in silico. In this set all five mutations predicted to be positive were responsive to pharmacological chaperones, bringing the percentage of responsive mutations among those predicted to be positive and not used to train the classifier to 86% (12/14). This figure differs significantly from the percentage of responsive cases observed among all the Fabry mutants tested so far.ConclusionsIn this paper we provide experimental support to an "in silico" method designed to predict missense mutations in the gene encoding lysosomal alpha galactosidase responsive to pharmacological chaperones. We demonstrated that responsive mutations can be predicted with a low percentage of false positive cases. Most of the mutations tested to validate the method were described in the literature as associated to classic or mild classic phenotype. The analysis can provide a guideline for the therapy with pharmacological chaperones supported by experimental results obtained in vitro. We are aware that our results were obtained in vitro and cannot be translated straightforwardly into benefit for patients, but need to be validated by clinical trials.

The most frequent disorder of glycosylation is due to mutations in the gene encoding phosphomannomutase2 (PMM2-CDG). For this disease, which is autosomal and recessive, there is no cure at present. Most patients are composite heterozygous and carry one allele encoding an inactive mutant, R141H, and one encoding a hypomorphic mutant. Phosphomannomutase2 is a dimer. We reproduced composite heterozygosity in vitro by mixing R141H either with the wild type protein or the most common hypomorphic mutant F119L and compared the quaternary structure, the activity and the stability of the heterodimeric enzymes. We demonstrated that the activity of R141H/F119L heterodimers in vitro, which reproduces the protein found in patients, has the same activity of wild type/R141H, which reproduces the protein found in healthy carriers. On the other hand the stability of R141H/F119L appears to be reduced both in vitro and in vivo. These findings suggest that a therapy designed to enhance protein stability such as those based on pharmacological chaperones or modulation of proteostasis could be beneficial for PMM2-CDG patients carrying R141H/F119L genotype as well as for other genotypes where protein stability rather than specific activity is affected by mutations.

The parent-of-origin-dependent expression of IGF2 and H19 is controlled by the imprinting center 1 (IC1) consisting of a methylation-sensitive chromatin insulator. IC1 is normally methylated on the paternal chromosome and nonmethylated on the maternal chromosome. We found that 22 cases in a large cohort of patients affected by Beckwith-Wiedemann syndrome (BWS) had IC1 methylated on both parental chromosomes, resulting in biallelic activation of IGF2 and biallelic silencing of H19. These individuals had marked macrosomia and high incidence of Wilms' tumor. A subset of these patients had 1.4- to 1.8-kb deletions with hypermethylation of the remaining IC1 region and fully penetrant BWS phenotype when transmitted maternally. Another subset of individuals with IC1 hypermethylation had a similar clinical phenotype but no mutation in the local vicinity. All these cases were sporadic and in at least two families affected and unaffected members shared the same maternal IC1 allele but not the abnormal maternal epigenotype. Similarly, no IC1 deletion was detected in 10 nonsyndromic Wilms' tumors with IC1 hypermethylation. In conclusion, methylation defects at the IGF2-H19 locus can result from inherited mutations of the imprinting center and have high recurrence risk or arise independently from the sequence context and not transmitted to the progeny.

Silver-Russell syndrome (SRS) is a heterogeneous disorder characterized by intrauterine and post-natal growth retardation, dysmorphic facial features and body asymmetry. About 50% of the patients carry (epi)genetic alterations involving chromosomes 7 or 11.The high proportion of patients with unidentified molecular etiology suggests the involvement of other genes. Interestingly, SRS patients share clinical features with the 12q14 microdeletion syndrome, characterized by several deletions with a 2.6 Mb region of overlap. Among the genes present in this interval, high mobility AT-hook 2 (HMGA2) appears to be the most likely cause of the growth deficiency, due to its described growth control function. To define the role of HMGA2 in SRS, we looked for 12q14 chromosome imbalances and HMGA2 mutations in a cohort of 45 patients with growth retardation and SRS-like phenotype but no 11p15 (epi)mutations or maternal uniparental disomy of chromosome 7 (matUPD7). We identified a novel 7 bp intronic deletion in HMGA2 present in heterozygosity in the proband and her mother both displaying the typical features of SRS. We demonstrated that the deletion affected normal splicing, indicating that it is a likely cause of HMGA2 deficiency. This study provides the first evidence that a loss-of-function mutation of HMGA2 can be associated with a familial form of SRS. We suggest that HMGA2 mutations leading to haploinsufficiency should be investigated in the SRS patients negative for the typical 11p15 (epi)mutations and matUPD7.

Fabry disease is caused by mutations in the GLA gene and is characterized by a large genotypic and phenotypic spectrum. Missense mutations pose a special problem for graduating diagnosis and choosing a cost-effective therapy. Some mutants retain enzymatic activity, but are less stable than the wild type protein. These mutants can be stabilized by small molecules which are defined as pharmacological chaperones. The first chaperone to reach clinical trial is 1-deoxygalactonojirimycin, but others have been tested in vitro. Residual activity of GLA mutants has been measured in the presence or absence of pharmacological chaperones by several authors. Data obtained from transfected cells correlate with those obtained in cells derived from patients, regardless of whether 1-deoxygalactonojirimycin was present or not. The extent to which missense mutations respond to 1-deoxygalactonojirimycin is variable and a reference table of the results obtained by independent groups that is provided with this paper can facilitate the choice of eligible patients. A review of other pharmacological chaperones is provided as well. Frequent mutations can have residual activity as low as one-fourth of normal enzyme in vitro. The reference table with residual activity of the mutants facilitates the identification of non-pathological variants.