Leber congenital amaurosis - a model for efficient genetic testing of heterogeneous disorders: LXIV Edward Jackson Memorial Lecture

Abstract

Purpose: To critically evaluate our experience in molecular testing of Leber congenital amaurosis (LCA) and to use this information to devise a general approach to heterogeneous recessive disorders. Careful clinical and molecular characterization of large cohorts of patients affected with inherited eye diseases will be an essential step in the development of effective therapy for these diseases, especially when the therapy involves gene replacement.

Design: A molecular genetic case-control study.

Methods: Six hundred forty-two unrelated individuals with the clinical diagnosis of LCA and 200 unrelated control individuals were screened for disease-causing sequence variations in eight genes using various combinations of single-strand conformational polymorphism analysis (SSCP), automated DNA sequencing, multiplex allele-specific ligation analysis (SNPlex), and high-density solid-phase single nucleotide polymorphism genotyping.

Results: Four hundred forty instances of 189 different disease-causing sequence variations were observed in this study, 98 of which have not been previously reported. One hundred forty-six of the 189 variations (77%) were observed in only a single individual. The observed variations were not evenly distributed among the LCA patients or among the eight genes. Empirical analysis of this uneven distribution was used to devise a multi-platform mutation detection strategy that is four times more efficient than a more conventional strategy of completely sequencing all of the coding regions of all LCA genes in all subjects. Hardy-Weinberg analysis of the observed mutations suggests that these eight genes are collectively responsible for about 70% of the cases of LCA in North America. The carrier frequency of the most common LCA allele (an intron 26 variation in CEP290) was found to be 2/3,248, which suggests that the overall prevalence of LCA in this population is about 1/81,000. An allele-specific ligation assay (SNPlex) was designed to detect 68 of the most common LCA-causing alleles, and semi-quantitative analysis of the data from this assay also revealed examples of gene deletion and isodisomy in the cohort.

Conclusions: The data demonstrates that a tiered screening strategy combining allele-specific detection with automated DNA sequencing can increase the efficiency of autosomal recessive mutation detection four-fold when compared with DNA sequencing alone. However, the very high rate of unique mutations observed in this study (77%) suggests that DNA sequencing will remain an important part of the overall strategy if high sensitivity is to be achieved.