Mutational analysis of Kabuki Syndrome patients and functional dissection of KMT2D mutations

The discovery of histone methyltransferase KMT2D and demethylase KDM6A genetic alterations in Kabuki Syndrome (KS) expanded and highlighted the role of histone modifiers in causing congenital anomalies and intellectual disability syndromes. KS is a rare autosomal dominant condition characterized by facial features, various organ malformations, postnatal growth deficiency, and intellectual disability. Since 2011 we performed a mutational screening of our KS cohort, that includes now 505 KS patients, by Sanger sequencing and MLPA of KMT2D, followed by KDM6A analysis in those patients resulted as KMT2Dnegative. Of these 505 patients, we identified 196/505 (39%) patients with KMT2D variants and 208 different KMT2D variations; of them 37/208 (18%) never described before. The majority of KS patients carry nonsense and splicesite variants, suggesting the loss of function, and therefore haploinsufficiency, as the likely mechanism for the KS phenotype. RT-PCR and direct sequencing on cDNA from Kabuki patients carrying KMT2D splice site variants demonstrated that these cause aberrant splicing of the corresponding transcript, resulting in a truncating and not functional translated protein. Molecular assays also showed that KMT2D mRNAs bearing premature stop codon are degraded by the nonsense mediated mRNA decay, contributing to KMT2D protein haploinsufficiency. We hypothesized that KS patients may benefit from a readthrough therapy that mediates translational suppression of nonsense variants, restoring the physiologically levels of endogenous KMT2D protein. Fourteen KMT2D nonsense variants were tested for their response to readthrough treatment through an in vitro dual reporter luciferase vector system, identifying 11/14 variants that displayed high levels of readthrough in response to gentamicin treatment. Among our cohort we identified three new cases with a mosaic variants in KMT2D gene, consisting in single nucleotide change resulting in two already reported nonsense variants, the c.13450C=/>T (p.R4484X) and the c.15061C=/>T (p.R5021X) and in a new frameshift variant, the c.3596_3597=/del (p.L1199HfsX7) KMT2D, respectively. Moreover, relevant for diagnostic and counselling purposes, we implemented a number of bioinformatics tools to assess the pathogenicity of 69 KMT2D missense variants, found overall in our cohort of 505 KS patients, and for 14 of them we adopted a combination of biochemical and cellular approaches to investigate their role and characterize their functional impact in the pathogenesis of the disease. We found 9/14 missense variants showing altered H3K4 methylation activity. We additionally assessed the impact on complex formation with WRAD protein complex, and we found that the reduced methyltransferase activity could be a consequence of lack of interaction