Background: TMC1, which encodes transmembrane channel-like protein 1, forms the mechanoelectrical transduction (MET) channel in auditory hair cells, necessary for auditory function. TMC1 variants are known to cause autosomal dominant (DFNA36) and autosomal recessive (DFNB7/11) non-syndromic hearing loss, but only a handful of TMC1 variants underlying DFNA36 have been reported, hampering analysis of genotype-phenotype correlations.

Methods: In this study, we retrospectively reviewed 338 probands in an in-house database of genetic hearing loss, evaluating the clinical phenotypes and genotypes of novel TMC1 variants associated with DFNA36. To analyze the structural impact of these variants, we generated two structural models of human TMC1, utilizing the Cryo-EM structure of C. elegans TMC1 as a template and AlphaFold protein structure database. Specifically, the lipid bilayer-embedded protein database was used to construct membrane-embedded models of TMC1. We then examined the effect of TMC1 variants on intramolecular interactions and predicted their potential pathogenicity.

Results: We identified two novel TMC1 variants related to DFNA36 (c.1256T > C:p.Phe419Ser and c.1444T > C:p.Trp482Arg). The affected subjects had bilateral, moderate, late-onset, progressive sensorineural hearing loss with a down-sloping configuration. The Phe419 residue located in the transmembrane domain 4 of TMC1 faces outward towards the channel pore and is in close proximity to the hydrophobic tail of the lipid bilayer. The non-polar-to-polar variant (p.Phe419Ser) alters the hydrophobicity in the membrane, compromising protein-lipid interactions. On the other hand, the Trp482 residue located in the extracellular linker region between transmembrane domains 5 and 6 is anchored to the membrane interfaces via its aromatic rings, mediating several molecular interactions that stabilize the structure of TMC1. This type of aromatic ring-based anchoring is also observed in homologous transmembrane proteins such as OSCA1.2. Conversely, the substitution of Trp with Arg (Trp482Arg) disrupts the cation-π interaction with phospholipids located in the outer leaflet of the phospholipid bilayer, destabilizing protein-lipid interactions. Additionally, Trp482Arg collapses the CH-π interaction between Trp482 and Pro511, possibly reducing the overall stability of the protein. In parallel with the molecular modeling, the two mutants degraded significantly faster compared to the wild-type protein, compromising protein stability.

Conclusions: This results expand the genetic spectrum of disease-causing TMC1 variants related to DFNA36 and provide insight into TMC1 transmembrane protein-lipid interactions.