Multiplex Ligation-dependent Probe Amplification analysis of Copy Number Variants in mentally retarded patients

The aetiology of mental retardation and dysmorphism is still poorly understood in over half of the affected individuals. Recent studies have shown that genomic mutations such as Copy Number Variants (CNV) can be important factors predisposing to the highly heterogeneous clinical conditions observed in affected individuals. Genome−wide array−Comparative Genomic Hybridization (aCGH) is the best technology available so far for a first screening of CNVs in genomic DNA in patients and control individuals. However, CNVs detected by aCGH need validation by an independent method such as Multiplex Ligation-dependent Probe Amplification (MLPA). In this report we present the results of a MLPA analysis performed on 120 mentally retarded patients and their parents and 400 samples from a reference population. Patients who underwent MLPA were previously scored positive to aCGH. Our results show that, overall, from the 150 CNVs identified by aCGH, 123 were confirmed by our MLPA analysis and overall account for 65 gains (size range: from 3,6 Kb to 3,0 Mb) and 58 losses (size range: from 2,2 Kb to 1,9 Mb). At least 16 CNVs are de novo (11 loss, 5 gain). Inheritance from a healthy parent could be established only for 83 CNVs. For 24 CNVs it was not possible to establish if they were de novo or inherited because one or both parents were unavailable for the analysis. In the majority of cases, a given CNV appear to be unique to a particular patient as far their chromosomal location and size is concerned. In contrast five 3 CNVs are shared each by two patients. One CNV_loss in chromosome 15q11.2 was shared by six unrelated patients. To investigate on the possible pathogenic role played by CNVs in this phenotypically and genetically heterogeneous population of mentally retarded patients and thus select the strongest candidate CNVs we have applied a multi-step strategy. Briefly, we first excluded CNVs also occurring in the general population. To this end, we screened by MLPA 400 neurotypical geographically- and ethnically-matched individuals. Second only CNVs occurring as de novo events were considered. Third, we queried a large body of data stored in the literature and in several databases to exclude CNVs previously associated to mental retardation (MR) or described as phenotypically unexpressed polymorphisms in the general population. The best candidate CNVs selected as just described were then analyzed for their gene content and to assess by in silico functional annotation analysis the role of these genes in nervous system. This approach led in some cases to the identification of several CNV, genes and statistically significant ontologies related to nervous system thus supporting their pathogenic role. A result of special interest was the identification in one MR patient with a complex phenotype of a duplicated microRNA (hsa-miR-150) overlapping a CNV on chromosome 19. The functions and diseases associated to the individual target genes/mRNA regulated by this microRNA appear to be highly correlated to the clinical phenotypes displayed 4 by this patient and suggest the occurrence of a new MR syndrome caused by a duplicated microRNA. In summary, the results of our study: (i) emphasize the importance of MLPA to exclude false positives generated by aCGH, (ii) describe the identification of novel CNVs and genes potentially implicated in MR. We believe that the results of this study will significantly contribute to the discovery of new microdeletion syndromes and a more precise correlation of the mutant genotype with the highly variable phenotype displayed by mentally retarded patients.