Identification of rs6161, previously designated a benign variant of CYP11A1, as a recurrent, pathogenic splicing mutation.

A Maharaj*, F Buonocore*, E Meimaridou, T D Cheetham, C Brain, E Gray, J Suntharalingham, N Striglioni, H Spoudeas, M Donaldson, J C Achermann*, L A Metherell*

Background: The genetic variant chr15:74635368 in CYP11A1 (SNP ID rs6161 c.940G>A) has a minor allele frequency of 0.0024 (ExAC); At protein level (p.E314K) it is predicted to be benign with heterologous expression of the mutant protein showing no functional difference from wild-type. CYP11A1 encodes the P450 side chain cleavage enzyme which initiates the steroidogenic cascade by conversion of cholesterol to pregnenolone. Severe deficiency of this enzyme is characterised by disordered sexual differentiation in addition to adrenal and gonadal insufficiency but partial loss-of-function mutations can present with a milder phenotype.


Results: In ten patients, with variable presentations, sequencing and segregation analysis revealed the presence of the rs6161 variant in exon 5 of CYP11A1 in compound heterozygosity with deleterious mutations on the other allele. This rs6161 variant is enriched in our patient cohort with an allele frequency = 0.015 (p<0.01). We hypothesized that it affected the gene at the RNA level, perhaps altering splicing, or that it represented a marker for another intronic/regulatory change within the gene.  An in vitro splicing assay, utilising minigene constructs incorporating wild-type or variant exon 5 sequences, revealed markedly reduced splicing of variant exon 5. 


Conclusion: Previously considered a benign variant, rs6161 occurs more frequently than expected in patients, alters splicing and is pathogenic when in combination with a disruptive change on the other allele. It represents a partially inactivating mutation and can lead to mild phenotypic presentation.


Impact: With the advent of whole exome and genome sequencing the necessity of assessing whether a sequence variant is deleterious or not becomes key.  In silico prediction tools may not be adequate to assign causality and hence analysis of exonic and non-exonic variants will need to be performed in vitro not only considering the change at protein but also at nucleic acid level.


N.B. Local Ethical Committee approval obtained.