Transmembrane channel-like isoform 1 and 2 (TMC1 and TMC2) are essential components of stereocilia mechanoelectrical transduction (MET) channel complex and are required for normal hearing. We examined the relative contributions of TMC1 and TMC2 to the hair bundle morphology development and tonotopic conductance gradients along the Organ of Corti. During post-natal development, TMC2 is progressively excluded, and in older animals, only TMC1 is present in outer (OHCs) and inner (IHCs) hair cells. We performed quantitative analysis of the expression of TMC1 and TMC2 at the sites of mechanotransduction using transgenic mice expressing fluorescently tagged TMC1 and TMC2. At postnatal day 6 (P6), the number of TMC1 molecules expressed at each MET channel site had a small increase from apex to base in IHCs but showed a 3-fold increase in OHCs. The fluorescence intensity of TMC2 exhibited a significant change in the opposite direction to TMC1 gradient in both IHCs and OHCs. The graded expression of TMC1 in OHCs was maintained through adulthood with the average number of TMC1 molecules calculated to be 8 at apex to 20 at base, while the expression of TMC2 decline from base to apex around P6 and by P10 can only be detected in apical IHCs. This data reveals a correlation between the number of TMC1 molecules and the tonotopic gradient in single channel conductance in OHCs. As part of a collaboration with the Fettiplace laboratory, single-channel conductance measurements were performed in OHCs and IHCs after brief treatment with sub-micromolar calcium. We observed that in OHCs, only TMC1-dependent channels support a tonotopic apex-to-base gradient in single-channel conductance. This gradient is absent in IHCs. These results are consistent with TMC1 being a component of the MET channel pore. Moreover, they suggest a varying number of channels per MET complex, each requiring multiple TMC1 molecules, and together operating in a coordinated or cooperative manner. Morphological analysis of the stereocilia development in mice exhibiting mosaic expression of fluorescently tagged TMC1 and TMC2 in a TMC1/TMC2 null background showed that the presence of either TMC1 or TMC2 was required to produce and sustain normal stereociliary bundle development. At P6, hair cells lacking both TMCs had immature phenotype characterized by multiple rows of stereocilia in IHCs and rounded bundles in OHCs. Compared to hair cells expressing both TMCs, a significant upregulation of one isoform was detected in stereocilia in the absence of the other. Moreover, we observed an upregulated TMC2 expression in the absence of TMC1 during hair cell development, suggesting that the removal of TMC2 may be facilitated by the increased level of TMC1. Our data suggests that while both TMC1 and TMC2 expression is required for normal stereocilia bundle morphology, the two isoforms likely make differential contributions towards MET channel properties as well as the complex mechanisms that regulate the slope of the stereocilia staircase and overall bundle development. We generated knockin mice expressing fluorescently-tagged TMC4 and TMC5 and observed that these proteins are not expressed in the stereocilia of hair cells but are present on the plasma membrane of microvilli of supporting cells. We examined other tissues from the TMC4-GFP and TMC5-mCherry-expressing mice and found that both proteins localize at the tips of microvilli of various non-sensory epithelial cells. This result is particularly intriguing given the structural and molecular similarities between hair cell stereocilia and intestinal brush-border microvilli. Hair cell stereocilia and brush-border microvilli share the same, or homologous, actin bundling proteins, membrane-cytoskeletal tether proteins (e.g., myosin1 and myosin 7 isoforms), as well as protocadherins. We are now using intestinal microvilli as an accessible and robust model system to understand the role of TMC4 and TMC5 and their possible functional similarities to TMC1 and TMC2. We are also examining the localization and copy number of TMIE molecules at the stereocilia along the Organ of Corti functional tonotopic gradient, using a mouse line expressing TMIE-EGFP. We confirmed the expression of TMIE-EGFP at the MET sites in Organ of Corti and vestibular hair cells. TMIE-EGFP is also found in abundant levels along the length of stereocilia in the inner and outer hair cells of the Organ of Corti. Contrary to what has been previously reported, we observed that mice lacking TMIE showed a residual MET current consistent with an accessory role in MET. Furthermore, mice expressing TMIE-GFP showed a reduction in MET currents and a significant hearing loss suggesting that the GFP tag at the C-terminus of the protein affects either its function or its localization in the Organ of Corti but not in the vestibular hair cells. We are currently examining the coordinated traffic and turnover of TMC1 and TMIE in the hair cells. MYO7A is an Usher type 1 protein that has long been thought to function as an actin-based molecular motor implicated in stereocilia cargo transport. Early work from our lab showed that MYO7A co-localizes with three other Usher type 1 proteins USH1C, SANS, and CDH23 at the upper tip link density of the MET apparatus. We also showed that MYO7A co-localizes with the scaffolding protein PDZD7 at the pivoting, tapered base of the stereocilia. We have now observed that the GIP3, a protein linked to DFNB72 deafness, also co-localizes with MYO7A at the stereocilia tapered base. We assessed the localization and behavior of these proteins in a heterologous expression system and showed a dynamic tripartite interaction between MYO7A, GIPC3 and the retrograde motor protein MYO6, which is also implicated in DFNB37 recessive forms of hearing loss. Our results suggest that MYO7A in conjunction with MYO6 transports and compartmentalizes GIPC3 at the stereocilia tapered region, where it likely complexes with other actin-binding components of the stereocilia taper complex such as CLIC4, TPRN, and RDX. Stereocilin (STRC) is expressed in the OHC and is implicated in anchoring the stereocilia to the overlaying tectorial membrane. The mechanism for STRC localization of is unknown, but active transport is likely necessary. STRC secondary structure is only predicted, and its tertiary structure is still unresolved which makes it challenging to predict interaction domains with known ligands or molecular transporters. Using RT-PCR on mouse inner ear mRNA preparations we identified two STRC isoforms: a previously described long isoform, and a novel shorter splice variant. Using immunofluorescence, we screened stereocilia myosins for their ability to transport STRC. Our results indicate that MYO15A transports STRC independently of WHRN and EPS8, although each of these proteins were found to increase STRC abundance at filopodia tips, likely by stabilizing the MYO15A/STRC complex at that location. We found that in mice lacking whirlin, STRC localizes to the tips of the tallest stereocilia, although TM anchoring imprints appeared abnormal, likely due to instability of the anchoring complex. In mice lacking MYO15A, where the tallest stereocilia did not exceed 0.5 m in length, STRC was seen at the apical plasma membrane surface but with no abundance at the tips of the tallest stereocilia and no tectorial membrane anchoring imprints. We argue that MyoXVA transports STRC and WHRN and EPS8 help stabilize the complex at stereocilia tips.
Showing the most recent 10 out of 60 publications
