The hair cell mechanotransduction (MET) channel complex resides at the tips of the second and shorter rows of stereocilia in the hair bundle, which are both dynamic and prolate. In addition to TMC1 and TMC2, other transmembrane proteins have recently been identified as essential for normal MET, including transmembrane inner ear (TMIE). MET currents were reported to be abolished in the absence of TMIE, and Tmie mutations cause deafness in mice and humans. Yet, the spatiotemporal expression of TMIE in stereocilia during development and its copy number relative to other MET components are not known. To address these questions, we generated CRISPR/Cas9-mediated knockin mice expressing GFP-tagged TMIE. We found that, in organ of Corti hair cells, TMIE localizes along the stereocilia during early bundle development and becomes more restricted to stereocilia tips as the bundle matures. TMIE is highly localized at the MET site in most vestibular hair cells but shows a dispersed localization in a sub-population with shorter hair bundles. Further, we found a large pool of TMIE-GFP in cytoplasmic membrane compartments of hair cells, suggesting continuous traffic and turnover. These findings provide new directions to investigate roles of TMIE in hair cells. Although both TMC1 and TMIE are essential for normal MET function, it is not known if and how they interact prior to targeting the MET site. To determine the extent of any interaction or co-traffic in stereocilia, we crossed mice expressing TMIE-GFP with those expressing TMC1-mCherry and assessed the relative expression profiles of both proteins along stereocilia. We found that TMIE-GFP and TMC1-mCherry co-localize at the MET site at stereocilia-tips of mature hair cells. However, during development TMIE-GFP also showed a broader distribution along the length of stereocilia and the two proteins did not co-localize outside the MET site. To verify their interdependence in mice we crossed Tmie-/- mice with mice expressing TMC1-mCherry and TMC2-GFP and observed that the TMCs failed to localize to stereocilia-tips without TMIE. Conversely, TMIE localization was not affected in stereocilia of Tmc1-/- mice. Thus, although TMIE and TMC1 traffic to the MET site independently of each other, TMIE is essential to stabilize TMC1 and TMC2 locally at the MET site. Analysis of stereocilia bundle morphology in the organ of Corti of mice with mosaic expression of TMC1-mCherry and TMC2-GFP on a Tmc1/:Tmc2/ background revealed that either TMC1 or TMC2 was necessary for normal bundle development, supporting the hypothesis that there is a degree of compensation between the two. Because TMIE is essential for TMC1 and TMC2 localization at the MET site and for normal MET function, we hypothesized that its absence would also affect bundle development. We observed that as late as P7, inner hair cells (IHCs) lacking TMIE had multiple rows of stereocilia in a pyramidal organization, reminiscent of Tmc1/:Tmc2/ bundles. This observation further supports the hypothesis that the processes underlying onset of MET and stereocilia development are interconnected and highlight a link between TMIE and TMC1 and TMC2 expression, MET activity, and stereocilia regulation. Stereocilia-tips where the MET channel complex is located, have prolate tips. However, the molecular basis of membrane curvature sensing and remodeling at stereocilia-tips is not known. Recently the gene encoding the I-BAR protein BAIAP2L2 was reported to be associated with hearing loss. We discovered that BAIAP2L2 localizes to stereocilia prolate tips, and the rise of its expression levels coincides with the onset of MET and the expression of stereocilia actin-regulatory proteins. Strikingly, BAIAP2L2 resides in a distinct spatial compartment, between the membrane and actin-regulatory machinery, providing new exciting evidence for the stratified organization of the protein complex(es) at stereocilia tips. Finally, we found, using a heterologous co-transfection assay, that BAIAP2L2 self-organizes into dense molecular aggregates and binds to multiple components of the stereocilia MET complex, as well as MYO3A/B- and MYO15A-based actin-regulatory proteins. We propose that BAIAP2L2 forms a scaffold that helps sculpt the membrane at stereocilia prolate tips and integrates MET and actin-regulatory protein complexes. We previously showed that MYO3A and MYO3B and their cargo ESPN1 localize to stereocilia distal tips, the sites of actin polymerization. In collaboration with Peter Barr-Gillespie (Oregon Health & Science University), we also identified a second MYO3A/3B cargo protein, ESPNL (espin-like), which appears early in stereocilia development and showed that that ESPN1 and ESPNL interact differently with MYO3A/B to regulate stereocilia staircase step-size. We next generated Espn1-/-:EspnL-/- mice to ascertain the degree of compensation between the two cargo proteins in determining stereocilia staircase morphology. We found that in mice lacking both ESPN1 and ESPNL the growth of the shorter rows of stereocilia begins to slow at P3, and almost all are lost completely by P12. Correspondingly, these mice have severe hearing loss, a complete loss of DPOAEs (as a measure of OHC function), and a 90 % reduction in OHC MET currents. Also, MYO3B, which we previously showed requires ESPN1 or ESPNL to localize to stereocilia-tips, is no longer able to do so. The stereocilia phenotype in mice lacking ESPN1 and/or ESPNL suggests some compensation between them in generating the stereocilia staircase and highlights the complex crosstalk between MYO3A/3B and their cargos in this process. In a previous study we demonstrated that the deletion of the polycistronic miRNA-183 cluster inhibited stereocilia elongation and global bundle maturation. We hypothesize that the miR-183 cluster coordinates the expression of multiple stereocilia actin-regulatory proteins via selective blocking of mRNA translation, and that this regulatory pathway provides an opportunity to design and test miR-based therapeutic approaches to rescue or regrow stereocilia. We used an in-silico approach to predict miR-183 cluster gene targets that overlap with stereocilia-specific mRNAs in the RNA-Seq repository database (gEAR Portal). The gene targets identified include Myo3b, Eps8, Triobp, which are known to be involved in regulation of stereocilia bundle structure and anchoring to the cuticular plate. We are currently using immunofluorescence to examine and quantify the changes in the stereocilia expression of these proteins in absence of the miR183 cluster. We previously showed that the appearance of MYO3A in cochlear stereocilia coincides with the onset of MET, and a recent report suggests that MYO3A is involved in transport of the PCDH15-CD2 isoform. We also observed that in long vestibular stereocilia, MYO7A shows a base-to-tip gradient of distribution, similar to MYO3A, and consistent with tipward translocation and dynamic accumulation. Based on these observations, we hypothesize that MYO3A and MYO7A transport components of the MET complex to stereocilia tips with the potential for complementary function and/or redundancy. Using a heterologous expression system, we observed that while MYO3A transports to filopodia tips only the PCDH15-CD2 isoform, MYO7A can transport all three (CD1, CD2, and CD3) PCDH15 isoforms. In a collaboration with Jung-Bum Shin (UVA), we helped demonstrate that the hair cell protein LIM only protein 7 (LMO7) localizes to the cuticular plate and to hair cell junctions. We showed that LMO7 forms actin networks via two putative actin binding domains and that Lmo7-/- mice suffer multiple cuticular plate deficiencies, including reduced actin density and abnormal stereociliary rootlets. Together, these defects affect cochlear tuning and sensitivity in mice lacking LMO7.
Showing the most recent 10 out of 60 publications
