Origin, functional role, and clinical impact of Fanconi anemia FANCA mutations

Abstract

Fanconi anemia is characterized by congenital abnormalities, bone marrow failure, and cancer predisposition. To investigate the origin, functional role, and clinical impact of FANCA mutations, we determined a FANCA mutational spectrum with 130 pathogenic alleles. Some of these mutations were further characterized for their distribution in populations, mode of emergence, or functional consequences at cellular and clinical level. The world most frequent FANCA mutation is not the result of a mutational "hot-spot" but results from worldwide dissemination of an ancestral Indo-European mutation. We provide molecular evidence that total absence of FANCA in humans does not reduce embryonic viability, as the observed frequency of mutation carriers in the Gypsy population equals the expected by Hardy-Weinberg equilibrium. We also prove that long distance Alu-Alu recombination can cause Fanconi anemia by originating large interstitial deletions involving FANCA and 2 adjacent genes. Finally, we show that all missense mutations studied lead to an altered FANCA protein that is unable to relocate to the nucleus and activate the FA/BRCA pathway. This may explain the observed lack of correlation between type of FANCA mutation and cellular phenotype or clinical severity in terms of age of onset of hematologic disease or number of malformations.

Figure 1

Localization of mutations found in FANCA gene in Spanish patients. Exons and introns are illustrated by gray shadowed boxes and lines, respectively. Exons are represented approximately at scale. Point mutations shown as open circles and large deletions shown as discontinuous lines correspond to mutations belonging to Spanish Gypsy patients. The rest of point mutations and deletions were all found in white patients.

Figure 2

Strategy designed for optimization of FANCA mutation screening based on the Spanish mutational spectrum. It consists of 4 rounds of exon sequencing and an MLPA (included in first round). (Upper panel) Proportion of mutations identified in each round, in Spanish FA-A white population. (Lower panel) Percentage of worldwide mutations from the FA Mutation Database that would be identified following the same strategy.

Figure 3

Haplotype associated with mutation c.3788_3790delTCT. (A) Haplotype determined by SNPs (top panel) and VNTRs (bottom panel) in patients from different populations. Countries represented in this analysis are colored (gray or black) depending on the haplotype, as determined by SNP analysis. Common VNTRs in all patients analyzed are highlighted in gray. n.a. indicates not analyzed. (B) Position of SNPs and VNTRs analyzed relative to FANCA.

Figure 4

PCR amplification of a fragment containing the breakpoint of ex1-20del. (A) Position of primers used, exons (black boxes), and Alu sequences (white boxes) found in regions flanking both breakpoints. (B) Fragments amplified by PCR in 2 patient carriers of the deletion (FA145 and FA58) and a control (not carrying this deletion) using 2 sets of primers. (C) Sequence of the 0.9-kb fragment containing the breakpoint (only relevant parts are shown). Sequence (bottom line) is aligned with FANCA and TCF25 (top line). (D) Scheme of the region affected by the deletion in chromosome 16q and genes included.

Figure 5

Functional analysis of missense and other nontruncating FANCA mutations leading to an altered FANCA protein. (A) Analysis of FANCA expression by Western blot of patient-derived LCLs. Two cell lines with 2 truncating mutations (FA55 and FA178) and a cell line with a large deletion and a splicing mutation (FA145) are included as negative controls. A total of 40 μg of total extract was loaded per lane, except for lane 2, where 2 μg of wild-type (WT) cell extract was loaded to detect a minimum amount of FANCA protein. (B) Quantification of FANCA expression by densitometry, relative to the WT. (C) Analysis of FANCD2 monoubiquitinylation in patients' LCLs treated with 2mM HU for 24 hours. (D) FANCA subcellular localization analysis by sample fractionation and Western blot after treatment with 2mM HU for 24 hours. NF indicates nuclear fraction; and CF, cytoplasmic fraction. Cell-equivalent volumes from each cell fraction were loaded. ORC2 detection is included as a nuclear protein control and GAPDH as a cytoplasmic control. (E) Analysis of FANCA subcellular localization analysis and foci formation by immunohistochemistry after HU treatment. A wild-type and a representative cell line with missense mutations (FA170) are shown. Images were observed with an Axio Observer A1 microscope (Zeiss) using a 100×/1.3 oil objective. The slides were mounted with Vectashield (Vector Laboratories) and were captured with an AxioCam MRm camera (Zeiss). Digital images were aquired with Axiovision 4.6 software.

Figure 6

Analysis of the effect of type of mutation on chromosomal instability and clinical severity of patients' phenotype. Box plots are shown representing median ± maximum-minimum values. (A) Spontaneous chromosome fragility indicated by Spontaneous Chromosome Fragility Index (SCFI = breaks per cell × percentage cells with breaks). (B) DEB-induced chromosome fragility as indicated by Chromosome Fragility Index (CFI = breaks per multiaberrant cell × percentage of cells with breaks). (C) Age in years at onset of hematologic disease. (D) Number of congenital malformations.