Abstract
Purpose: :
Pre-mRNA splicing plays a significant role in the pathogenesis of inherited retinal degeneration. Up to 50% of all point mutations responsible for genetic diseases cause aberrant splicing. Bardet-Biedl syndrome (BBS: OMIM 209900) is a genetically heterogeneous disorder characterized by retinal degeneration, polydactyly, early-onset obesity, cognitive impairment, genitourinary anomalies and renal dysfunction. To date, 12 BBS genes have been identified. The aim of this study was to model the pathological impact of known mutations on the splicing process in genes associated with BBS.
Methods: :
Computational methods were applied ‘in-silico’ to prioritise and direct subsequent laboratory workup of mutations in BBS genes to establish which, if any, might have a pathogenic effect on splicing. Transcripts were classified as vulnerable to aberrant or pathological splicing if sequence variations alter splice site (SS) strength or splice site definition via splicing enhancers and silencers. SS score was calculated on 190 exons in 12 BBS genes using the following scoring method(s): Consensus Sequence Weighted Matrices, Neural Network, Information Theory and Maximum Entropy. Possible mutational effects on exonic splicing enhancers and silencers were modelled using the following servers: ESEfinder, FAS-ESS, PESXs, and Rescue-ESE. Evidence of AS in each exon was also obtained using the EST databases: Aceview and AltSplice. Computational predictions of aberrant splicing were validated in vitro using a minigene system following in vitro mutagenesis of target constructs.
Results: :
290 mutations were modelled and 21% identified in silico as potential mis-splicing mutations. BBS9, BBS10 and BBS5 genes contained the greatest percentage of possible splicing mutations with 75%, 33% and 33% of mutations modeled affecting either SS score or splicing enhanchers respectively. A number of predicted aberrant splicing events were modeled with a minigene system in HEK293 and RB1 cell lines to validate predictions.
Conclusions: :
Traditionally, mutation screening is based on genomic DNA analysis and the effect of a mutation on the mRNA or protein is usually predicted from the primary genomic sequence, as opposed to direct experimental evaluation by determining mRNA expression and splicing patterns. Here, we present an approach to predict and model aberrant splicing in mutant BBS genes.
Keywords: computational modeling • retinal degenerations: hereditary • genetics