Targeted next-generation sequencing identifies a homozygous nonsense mutation in ABHD12, the gene underlying PHARC, in a family clinically diagnosed with Usher syndrome type 3

Usher syndrome (USH) is an autosomal recessive disorder manifesting in about 10% of children with congenital sensorineural hearing impairment, characterized by additional retinitis pigmentosa (RP). The clinical subtype USH1 presents with severe to profound hearing loss, vestibular impairment and early RP, whereas USH2 is characterized by moderate to severe hearing impairment and RP in adolescence. USH3 is very rare and variable; it may resemble USH1 or USH2, and vestibular dysfunction may be present. Ten genes (including a digenic contributor and modifier, PDZD7) have been implicated in USH[1‐3]. Clarin-1 (CLRN1) has been the only known USH3 gene to date (locus: USH3A) until recently, when HARS, encoding histidyl-tRNA synthetase, has been proposed as a novel USH3 gene[4]. Moreover, Dad et al. have mapped a condition with clinical overlap to USH3 (RP, progressive hearing impairment, vestibular dysfunction, and congenital cataract) to chromosome 15q22.2-23[5].

We have identified a consanguineous Lebanese family with an USH3-like phenotype. The patients exhibited sensorineural hearing loss, RP and cataracts, an ocular complication USH patients are predisposed for. Our study aimed at identifying the causative mutation – and potentially the second USH3 gene – in this family by homozygosity mapping and targeted next-generation sequencing (NGS). After the identification of a homozygous truncating mutation in a known disease gene, ABHD12, one patient was re-examined and found to display ataxia, reversing the diagnosis to a neurodegenerative disease, PHARC, that is characterized by p olyneuropathy, h earing loss, a taxia, r etinitis pigmentosa, and early-onset c ataract.

Recent studies indicate a population-prevalence for USH of 1/6000[14]. Children with USH initially have non-syndromic hearing loss (NSHL): RP manifests in the first (USH1) or second (USH2, USH3) decade of life. At least 10% of the hearing-impaired children carry mutations in USH genes, making USH an important differential diagnosis. Molecular genetic testing can confirm or exclude USH at an early time point, even before the onset of visual problems, and may help limiting detailed ophthalmological follow-up in deaf children to those with USH-causing mutations. Moreover, because certain mutations in USH genes cause hearing impairment without retinal degeneration, USH-causing mutations will be unevitably identified in children with apparently NSHL by massively parallel NGS of all known deafness genes – an approach that will become a diagnostic routine within the next few years.

Mutations in the known USH genes account for 72 – 86% of cases[15, 16]. The remainder may be due to mutations far outside the coding regions[17] or large structural rearrangements[18] in these genes that escape detection by genomic DNA amplification and sequencing of the coding exons, and to further genetic heterogeneity. Moreover, USH may be mimicked by clinically overlapping conditions, such as Alström syndrome, or by the co-occurrence of non-syndromic deafness and RP in the same individual[19]. Therefore, the awareness of potential differential diagnoses is important when seeking molecular verification of the clinical diagnosis.

Both patients from our family have deafness, RP and cataracts, all symptoms compatible with USH. Linkage analysis excluded all loci for known USH genes, and targeted NGS revealed an ABHD12 nonsense mutation segregating with disease in the family. Truncating ABHD12 mutations have been shown to cause PHARC, a neurodegenerative disease with p olyneuropathy, h earing loss, a taxia, R P, and early c ataract[20]. II:1 therefore underwent neurological examination which revealed ataxia – a symptom that may be present in USH1 and USH3 due to the affection of vestibular hair cells. Most patients with PHARC and confirmed ABHD12 mutations had ataxia, and these patients had cerebellar atrophy or peripheral polyneuropathy or both[20]. There were no obvious signs of polyneuropathy in patient II:1, and no indication of cerebellar atrophy in the cranial CT scan. However, her balance problems could result from polyneuropathy that remained undetected because detailed investigation of the peripheral nerves had been omitted as long as USH was the clinical diagnosis. II:4 did not complain about balance problems, indicating that ataxia was either very mild or not present. Unfortunately, this patient was not available anymore for detailed clinical follow-up. However, given the homozygous ABHD12 nonsense mutation segregating with the disease, the phenotype in our family can be considered a variant of PHARC.

In retinal photoreceptor cells, the USH protein interactome presumably plays a role in transport, trafficking and synaptic function; in the inner ear, the USH protein interactions are important for hair cell development, maintenance, and for tip link formation and hence mechanotransduction[21‐23]. The only known USH3 protein known, clarin-1, has recently been shown to be part of this interactome as well[24]. The PHARC protein ABHD12 hydrolyzes 2-arachidonoyl glycerol, an endocannabinoid lipid transmitter that acts on cannabinoid receptors CB1 and CB2. The pathway affected in PHARC is not yet known to be related to the USH protein complex. As can be expected because of the RP component in PHARC, ABHD12 is expressed in the retina[20], but no details are available on inner ear expression. Further studies are needed to determine the expression of ABHD12 on the cellular and subcellular level in both sensory systems. The investigation of potential interactions of ABHD12 with the known USH proteins will be crucial to find out if the clinical overlap of PHARC and USH is based on a functional relationship between these proteins.

We suggest that PHARC should be taken into account as differential diagnosis for USH, especially in “USH3-like” patients with progressive hearing loss and balance problems. Comprehensive genetic analysis, mainly by NGS-based approaches, will increasingly be helpful in correcting the diagnosis (reverse phenotyping)[25] and thereby improve patient management.