Comprehensive analysis of pathogenic deletion variants in Fanconi anemia genes.

Fanconi anemia (FA) is a rare recessive disease resulting from mutations in one of at least 16 different genes. Mutation types and phenotypic manifestations of FA are highly heterogeneous and influence the clinical management of the disease. We analyzed 202 FA families for large deletions, using high-resolution comparative genome hybridization arrays, single-nucleotide polymorphism arrays, and DNA sequencing. We found pathogenic deletions in 88 FANCA, seven FANCC, two FANCD2, and one FANCB families. We find 35% of FA families carry large deletions, accounting for 18% of all FA pathogenic variants. Cloning and sequencing across the deletion breakpoints revealed that 52 FANCA deletion ends, and one FANCC deletion end extended beyond the gene boundaries, potentially affecting neighboring genes with phenotypic consequences. Seventy-five percent of the FANCA deletions are Alu-Alu mediated, predominantly by AluY elements, and appear to be caused by nonallelic homologous recombination. Individual Alu hotspots were identified. Defining the haplotypes of four FANCA deletions shared by multiple families revealed that three share a common ancestry. Knowing the exact molecular changes that lead to the disease may be critical for a better understanding of the FA phenotype, and to gain insight into the mechanisms driving these pathogenic deletion variants.

Genetic subtyping of Fanconi anemia by comprehensive mutation screening.

Fanconi anemia (FA) is a recessively inherited syndrome with predisposition to bone marrow failure and malignancies. Hypersensitivity to cross-linking agents is a cellular feature used to confirm the diagnosis. The mode of inheritance is autosomal recessive (12 subtypes) as well as X-linked (one subtype). Most genetic subtypes have initially been defined as "complementation groups" by cell fusion studies. Here we report a comprehensive genetic subtyping approach for FA that is primarily based on mutation screening, supplemented by protein expression analysis and by functional assays to test for pathogenicity of unclassified variants. Of 80 FA cases analyzed, 73 (91%) were successfully subtyped. In total, 92 distinct mutations were detected, of which 56 were novel (40 in FANCA, eight in FANCC, two in FANCD1, three in FANCE, one in FANCF, and three in FANCG). All known complementation groups were represented, except D2, J, L, and M. Three patients could not be classified because proliferating cell cultures from the probands were lacking. In cell lines from the remaining four patients, immunoblotting was used to determine their capacity to monoubiquitinate FANCD2. In one case FANCD2 monoubiquitination was normal, indicating a defect downstream. In the remaining three cases monoubiquitination was not detectable, indicating a defect upstream. In the latter four patients, pathogenic mutations in a known FA gene may have been missed, or these patients might represent novel genetic subtypes. We conclude that direct mutation screening allows a molecular diagnosis of FA in the vast majority of patients, even in cases where growing cells from affected individuals are unavailable. Proliferating cell lines are required in a minority (<15%) of the patients, to allow testing for FANCD2 ubiquitination status and sequencing of FANCD2 using cDNA, to avoid interference from pseudogenes.