Although consanguinity can facilitate the discovery of novel genes associated with many diseases, yet it challenges the concept of single causative genetic variant [3]. Interestingly, in this study we revealed a polygenic inheritance of NSHL with the liaison of two independent homozygous alterations in well-known HL genes. To the best of our knowledge, this is the first study to report the implication of a novel MITF variant in an NSHL case with an autosomal recessive mode of inheritance and a post-lingual onset.
MYO15A and MITF homozygous alterations: the dilemma of predictive tools?
MYO15A encodes for XVA myosin protein which plays a vital role in the elongation and development of stereocilia and actin filaments. More than forty MYO15 mutations have been reported in the motor domain of the protein with generally autosomal recessive HL impairment characterized by a profound phenotype at all frequencies [11]. The detected homozygous MYO15A mutation, p.V485A, was previously associated with a HL phenotype in an Iranian family[3]. Mutations in the N-terminal domain are thought to be associated with a milder form of HL since they can affect only one of the two major isoforms of the gene[12]. Although the p.V485A mutation is located within the N-terminal domain, our indexed patients suffer from a mild to severe phenotype. In addition, two healthy individuals from the Gnomad Exome database harbor this variant which argue against a major role for this mutation in the affected individuals. Accordingly, we postulate that other players might be linked, in collaboration or independent of MYO15A, to the underlying phenotype.
We therefore considered the second shared bi-allelic novel MITF gene mutation p. P338L between the two sisters. MITF encodes the melanocyte-specific promoter of microphthalmia-associated bHLH transcription factor. A total of more than forty MITF mutations have been verified to be disease-causing in patients with either the Waardenburg's syndrome type 2 )WS2) (OMIM#193510) or the Tietz syndrome (OMIM #103500)[13] . Both syndromes are autosomal dominant and are characterized by overlapping phenotypes that encompass HL and pigmentary abnormalities with variable penetrance. To the best of our knowledge, only 2 homozygous MITF cases were detected in WS2 and WS4 [14,15].. In the present study, the detected homozygous p.P338L missense mutation was neither reported in the dbSNP database, nor in the Gnomad Exome/Genome database. It was also absent from more than 300 Lebanese exomes. The heterozygous frequency of this variant is less than 0.00001 in these databases as it is only present in 3 individuals. Since the detected MITF misssense mutation is localized in the bHLH DNA-binding domain and since the in-silico analysis revealed a deleterious effect prediction, we accordingly hypothesize that this mutation is disease causing (Table-1). Thus, structural and functional assays are compulsory to assess the effect of this mutation on the ability of MITF to heterodimerize, bind DNA, and/or translocate to the nucleus.
Patients who previously presented with HL as the only phenotypic feature were thought to have NSHL. In consequence, only mutations in genes associated with this type of HL were investigated. On the other hand, some SHL cases require special confirmatory tests since the penetrance of secondary features is either incomplete or age dependent. One example is the Usher syndrome which is presented as an NSHL case early-on in life as the onset of the secondary symptom (retinitis pigmentosa) does not appear until puberty. This might cause a false clinical classification of some patients with SHL who can benefit from the appropriate implementation of visual rehabilitation at early stages [6]. Therefore, it is very critical to categorize genes and variants that are either specific to each type or involved in both forms of HL. Another example is the heterozygous MITF (p.R110X) variant that was specifically associated with SHL cases but was recently detected in an NSHL case that presented in the absence of WS2 common features ( no pigmentary changes in hair, eyes, or skin) [16]. Originally in-vivo studies on the phenotypic variation seen with the different alleles of the mouse MITF gene referred to as mi gene suggests that mutations in the human MITF gene may also manifest themselves in different ways. This proposed a possibility for detecting phenotypes different from the characteristic WS2 phenotype among patients with MITF mutations[17]. Combining these facts with our results, we propose expanding the implications of MITF variants from syndromic to non-syndromic HL cases while associating it with an autosomal recessive mode of inheritance.
Additionally, it is widely known that most mutations in autosomal dominant loci cause post-lingual hearing impairment (including MYO7A and MYH14) while mutations in autosomal recessive HL cases with delayed childhood onset are rare clinical findings [10]. Herein, we are the first to propose MITF and MYO15A variants as autosomal recessive loci causing stable post-lingual hearing impairment rather than a progressive pre-lingual one.
Polygenic inheritance
Although most genetic deafness cases result from mutations in a single gene, an emerging number of examples are being documented where recessive mutations at two loci are being involved. For example, the digenic interaction that underlies the cause of deafness in individuals carrying a single mutation at the GJB2 locus along with a deletion in the functionally related GJB6 gene[18]. Moreover, a study done by Legar.et al on twelve patients with MITF mutations demonstrated a large range of variability in phenotype among these patients which argue for the possible interaction with modifier loci[19]. Herein, we propose a polygenic form of inheritance mainly through the implication of both MITF and MYO15A variants coupled with two detected heterozygous variants in MYO7A and MYH14 genes. Different compound heterozygous or homozygous mutations related to MYO7A have been reported in a variety of autosomal recessive Usher Syndrome families[20]. However, mutations in MYH14 gene are associated with autosomal dominant hearing impairment[21]. Thus, we speculate an involvement of the detected MYH14 and MYO7A mutations in the observed phenotype but not as the direct independent cause of HL since the parents presented as healthy carriers. Further functional studies are needed to assess the independent and combined effect of these mutations on the development of HL.
Finally, we could not rule out other genetic/epigenetic modifiers that could be associated with the underlying phenotype, especially that a growing number of studies have showed that copy number variation (CNV) is widely encountered in syndromic and non-syndromic HL cases[22–24]. Such studies would require a case-control study with a substantial number of patients with SHL, NSHL, and controls.