High-resolution immunofluorescence imaging of mouse cochlear hair bundles

VASCilia (Vision Analysis StereoCilia): A Napari Plugin for Deep Learning-Based 3D Analysis of Cochlear Hair Cell Stereocilia Bundles

Cochlear hair cells are essential for hearing, and their stereocilia bundles are critical for mechanotransduction. However, analyzing the 3D morphology of these bundles can be challenging due to their complex organization and the presence of other cellular structures in the tissue. To address this, we developed VASCilia (Vision Analysis StereoCilia), a Napari plugin suite that automates the analysis of 3D confocal microscopy datasets of phalloidin-stained cochlear hair cell bundles. VASCilia includes five deep learning-based models that streamline the analysis process, including: (1) Z-Focus Tracker (ZFT) for selecting relevant slices in a 3D image stack; (2) PCPAlignNet (Planar Cell Polarity Alignment Network) for automated orientation of image stacks; (3) a segmentation model for identifying and delineating stereocilia bundles; (4) a tonotopic Position Prediction tool; and (5) a classification tool for identifying hair cell subtypes. In addition, VASCilia provides automated computational tools and measurement capabilities. Using VASCilia, we found that the total actin content of stereocilia bundles (as measured by phalloidin staining) does not necessarily increase with bundle height, which is likely due to differences in stereocilia thickness and number. This novel biological finding demonstrates the power of VASCilia in facilitating detailed quantitative analysis of stereocilia. VASCilia also provides a user-friendly interface that allows researchers to easily navigate and use the tool, with the added capability to reload all their analyses for review or sharing purposes. We believe that VASCilia will be a valuable resource for researchers studying cochlear hair cell development and function, addressing a longstanding need in the hair cell research community for specialized deep learning-based tools capable of high-throughput image quantitation. We have released our code along with a manually annotated dataset that includes approximately 55 3D stacks featuring instance segmentation (https://github.com/ucsdmanorlab/Napari-VASCilia). This dataset comprises a total of 502 inner and 1,703 outer hair cell bundles annotated in 3D. As the first open-source dataset of its kind, we aim to establish a foundational resource for constructing a comprehensive atlas of cochlea hair cell images. Ultimately, this initiative will support the development of foundational models adaptable to various species, markers, and imaging scales to accelerate advances within the hearing research community.

Morphological phenotyping of the aging cochlea in inbred C57BL/6N and outbred CD1 mouse strains

Morphological mouse phenotyping plays a pivotal role in the translational setting and even more in the area of auditory research, where mouse is a central model organism due to the evolutionary genetic relationship and morpho-functional analogies with the human auditory system. However, some results obtained in murine models cannot be translated to humans due to the inadequate description of experimental conditions underlying poor reproducibility. We approach the characterization of the aging process of the mouse cochlea in animals up to 18 months of age belonging to two of the most used outbred (CD1) and inbred (C57BL/6N) strains. Striving to reduce any environmental variable we performed our study compliantly to the ARRIVE guidelines. We integrated instrumental data (auditory brainstem response test), with morphological analyses to correlate functional discrepancies to morphological changes and track the differences in the evolution of sensorineural hearing loss in the two strains. We featured the localization of Gipc3, Myosin VIIa, and TMC1 in hair cells of the Corti organ as well as NF 200 and the density of type I neuron in the spiral ganglion. We outlined age-related hearing loss (ARHL) in both strains, and a clear drop in the selected marker localization. However, in CD1 we detected a different trend allowing the identification of potential strain-specific mechanisms, namely an increase in myosin VIIa in 6 months aging mice in comparison to 2 months old animals. Our findings represent an asset to investigate the strain-dependent physiological trigger of ARHL providing new insights in the translational area.

SUB-immunogold-SEM reveals nanoscale distribution of submembranous epitopes

Electron microscopy paired with immunogold labeling is the most precise tool for protein localization. However, these methods are either cumbersome, resulting in small sample numbers and restricted quantification, or limited to identifying protein epitopes external to the membrane. Here, we introduce SUB-immunogold-SEM, a scanning electron microscopy technique that detects intracellular protein epitopes proximal to the membrane. We identify four critical sample preparation factors contributing to the method's sensitivity. We validate its efficacy through precise localization and high-powered quantification of cytoskeletal and transmembrane protein distribution. We evaluate the capabilities of SUB-immunogold-SEM on cells with highly differentiated apical surfaces: (i) auditory hair cells, revealing the presence of nanoscale MYO15A-L rings at the tip of stereocilia; and (ii) respiratory multiciliate cells, mapping the distribution of the SARS-CoV-2 receptor ACE2 along the motile cilia. SUB-immunogold-SEM extends the application of SEM-based nanoscale protein localization to the detection of intracellular epitopes on the exposed surfaces of any cell.

LOXHD1 is indispensable for maintaining TMC1 auditory mechanosensitive channels at the site of force transmission

Hair cell bundles consist of stereocilia arranged in rows of increasing heights, connected by tip links that transmit sound-induced forces to shorter stereocilia tips. Auditory mechanotransduction channel complexes, composed of proteins TMC1/2, TMIE, CIB2, and LHFPL5, are located at the tips of shorter stereocilia. While most components can interact with the tip link in vitro, their ability to maintain the channel complexes at the tip link in vivo is uncertain. Return, using mouse models, we show that an additional component, LOXHD1, is essential for keeping TMC1-pore forming subunits at the tip link but is dispensable for TMC2. Using SUB-immunogold-SEM, we showed that TMC1 localizes near the tip link but mislocalizes without LOXHD1. LOXHD1 selectively interacts with TMC1, CIB2, LHFPL5, and tip-link protein PCDH15. Our results demonstrate that TMC1-driven mature auditory channels require LOXHD1 to stay connected to the tip link and remain functional, while TMC2-driven developmental channels do not.

SUB-Immunogold-SEM reveals nanoscale distribution of submembranous epitopes

Electron microscopy paired with immunogold labeling is the most precise tool for protein localization. However, these methods are either cumbersome, resulting in small sample numbers and restricted quantification, or limited to identifying protein epitopes external to the membrane. Here, we introduce SUB-immunogold-SEM, a scanning electron microscopy technique that detects intracellular protein epitopes proximal to the membrane. We identified four critical sample preparation factors that contribute to the method's sensitivity and validate its efficacy through precise localization and high-powered quantification of cytoskeletal and transmembrane proteins. We evaluated the capabilities of SUB-immunogold-SEM on cells with highly differentiated apical surfaces: (i) auditory hair cells, revealing the presence of nanoscale Myosin rings at the tip of stereocilia; and (ii) respiratory multiciliate cells, mapping the distribution of the SARS-CoV-2 receptor ACE2 along the motile cilia. SUB-immunogold-SEM provides a novel solution for nanoscale protein localization at the exposed surface of any cell.

LOXHD1 is indispensable for coupling auditory mechanosensitive channels to the site of force transmission

Hearing is initiated in hair cells by the mechanical activation of ion channels in the hair bundle. The hair bundle is formed by stereocilia organized into rows of increasing heights interconnected by tip links, which convey sound-induced forces to stereocilia tips. The auditory mechanosensitive channels are complexes containing at least four protein-subunits - TMC1/2, TMIE, CIB2, and LHFPL51-16 - and are located at the tips of shorter stereocilia at a yet-undetermined distance from the lower tip link insertion point17. While multiple auditory channel subunits appear to interact with the tip link, it remains unknown whether their combined interaction alone can resist the high-frequency mechanical stimulations owing to sound. Here we show that an unanticipated additional element, LOXHD1, is indispensable for maintaining the TMC1 pore-forming channel subunits coupled to the tip link. We demonstrate that LOXHD1 is a unique element of the auditory mechanotransduction complex that selectively affects the localization of TMC1, but not its close developmental paralogue TMC2. Taking advantage of our novel immunogold scanning electron microscopy method for submembranous epitopes (SUB-immunogold-SEM), we demonstrate that TMC1 normally concentrates within 100-nm of the tip link insertion point. In LOXHD1's absence, TMC1 is instead mislocalized away from this force transmission site. Supporting this finding, we found that LOXHD1 interacts selectively in vitro with TMC1 but not with TMC2 while also binding to channel subunits CIB2 and LHFPL5 and tip-link protein PCDH15. SUB-immunogold-SEM additionally demonstrates that LOXHD1 and TMC1 are physically connected to the lower tip-link complex in situ. Our results show that the TMC1-driven mature channels require LOXHD1 to stay coupled to the tip link and remain functional, but the TMC2-driven developmental channels do not. As both tip links and TMC1 remain present in hair bundles lacking LOXHD1, it opens the possibility to reconnect them and restore hearing for this form of genetic deafness.