Unraveling haplotype errors in the DFNA33 locus

Abstract

Genetic heterogeneity makes it difficult to identify the causal genes for hearing loss. Studies from previous decades have mapped numerous genetic loci, providing critical supporting evidence for gene discovery studies. Despite widespread sequencing accessibility, many historically mapped loci remain without a causal gene. The DFNA33 locus was mapped in 2009 and coincidentally contains ATP11A, a gene recently associated with autosomal dominant hearing loss and auditory neuropathy type 2. In a rare opportunity, we genome-sequenced a member of the original family to determine whether the DFNA33 locus may also be assigned to ATP11A. We identified a deep intronic variant in ATP11A that showed evidence of functionally normal splicing. Furthermore, we re-assessed haplotypes from the originally published DFNA33 family and identified two double recombination events and one triple recombination event in the pedigree, a highly unlikely occurrence, especially at this scale. This brief research report also serves as a call to the community to revisit families who have previously been involved in gene mapping studies, provide closure, and resolve these historical loci.

Keywords: ATP11A; DFNA33; genome sequencing; haplotype analysis; hereditary hearing loss; unresolved deafness loci.

FIGURE 1

Updated pedigree and clinical information from the individual whose genome was subjected to sequencing. (A). Updated pedigree with an emphasis on the third generation. The individual who was sequenced is marked with a black arrow. Individuals who were not included in the Bönsch et al. (2009) publication are marked with an asterisk to the upper right of each pedigree symbol. Those included in the original pedigree are noted with their designated number from Bönsch et al. (2009). Haplotypes originally described in Bönsch et al. (2009) are included with the double and triple recombination events marked with red arrows. It should be noted that individual 407 was originally denoted as a male individual in the original pedigree. Upon re-review, an error was identified, and this has been updated and corrected as a female individual. (B). Pure-tone audiograms from the individual who was subjected to genome sequencing. Pure-tone audiogram from the right (left) and left (right) ears at the age of 66.9 years. Air conduction thresholds in the dB hearing level for the right and left ears are represented with circles and crosses, respectively. Bone conduction measurements are available for the right ear and shown by >.

FIGURE 2

In vitro splicing assay and in silico prediction of the ATP11A c.725 + 737C>T deep intronic variant. (A). Electrophoretic visualization of cDNA RT-PCR products amplified from the constructs, following transfection into HEK 293T cells. Amplicons were resolved on 1.5% agarose gel. Wild-type splicing yields a 455-bp amplicon and is composed of exons A and B (vector) and exons 8–10 that were amplified from the patient and control DNA samples. (B) Analysis with ESEfinder and RESCUE-ESE reveals the splicing sequence landscape for the wild-type (upper panel) and mutated (lower panel) human sequence at c.725 + 737. The nucleotide at the c.725 + 737 position is outlined in red. ESE hits are displayed above and below each sequence. The green boxes represent RESCUE-ESE hexamers. The c.725 + 737C>T variant is predicted to induce an ESE hexamer that is shown by the string of green boxes in the bottom sub-panel. (C) The vector construct of the in vitro splicing assay illustrates the wild-type or mutant amplicons inserted between exons A and B of the pSPL3 vector with a normal splicing result (upper and mid panels). The lower panel shows the sequence of each exon–exon junction for wild-type and mutant constructs, revealing an identical result.