The dual molecular identity of vestibular kinocilia bridges structural and functional traits of primary and motile cilia
Figures
Single-cell transcriptional atlas of cochlear and vestibular cells.
(A) Schematic drawing of the organ of Corti (top panel) and representative images of IHCs and OHCs from adult mouse cochleae. (B) Schematic drawing of the utricle (top panel) and confocal images of the utricle prepared from an adult mouse. HCs are stained with MYO6, SOX2, and DAPI. SOX2-positive cells include type II HCs and SCs underneath HCs. (C) Representative images of type I and II HCs from utricular and saccular maculae as well as crista ampullaris from adult mice. (D, E) tSNE plots of distinct cell types detected in the adult CBA mouse cochlea (D) and utricle, saccule, and crista. Different cell types are color-coded. (F) Feature plots of the expression six marker genes in different HC populations. (G) Dot plot heatmap of average expression and cellular detection rate of 28 representative marker genes in different HC types in cochleae and vestibular end organs. Abbreviations: IHC (inner HC); OHC (outer HC); SGN (spiral ganglion neuron); SC (Schwann cell)/SGC (satellite glial cell); DC (Deiters’ cell)/PC (pillar cell); IPhC (inner phalangeal cell)/IB (inner border cell); ISC (inner sulcus cell); HeC (Hensen’s cell); SpC (spindle cell)/RC (root cell); MC (marginal cell)/IC (intermediate cell); BaC (basal cell); MP (macrophage); RM (Reissner’s membrane); BC (B cell); TC (T cell); Gran (granulocyte); Mono (monocyte); RBC (red blood cell); OSC (outer sulcus cell); TBC (tympanic border cell). Type I HC (type I HC); Type II HC (type II HC); SESC (sensory epithelial SC); TEC (transitional epithelial cell); PTEC (peripheral transitional epithelial cell); ERC (epithelial roof cell). Neurons (vestibular neurons). For those cells whose definite identities cannot be annotated, the top expressed genes were used for identification and annotation. These cells include Tectb+, Mgp+, Coch+, Prl+, Otos+, Emilin2+/Cavin2+, Fxyd2+/Kcnk2+, Prtn3+/Ccl5+, and Ptgds+/Coch+.
Similarity and difference among different HC types and biological processes enriched in cochlear and vestibular HCs.
(A) Principal component analysis (PCA) plot showing similarity based on pseudo bulk RNA-seq data from four HC types. (B) PCA plot showing similarity based on individual HC gene expression among the four HC types. (C) Venn diagram depicting the number of expressed genes (RPKM >0) in four HC types. (D) Volcano plot showing differentially expressed genes between different HC types. Red dots indicate differentially expressed genes with p-value <10e−5 and log2 fold change >1. Only the top 20 differentially expressed genes are labeled. (E) Biological processes enriched in vestibular HCs compared to cochlear HCs. Biological processes related to motile cilia are enriched. (F) Biological processes enriched in type I HCs compared to type II HCs. (G) Biological processes enriched in type II HCs compared to type I HCs.
© 2010, Schwander et al. Figure 5A was reprinted with permission from Figure 2 of Schwander et al. 2010, which was published under a CC BY-NC SA 3.0 license. Further reproductions must adhere to the terms of this license.
Shared and unique genes expressed in cochlear and vestibular HCs.
(A) Violin plots of the expression of 72 genes in four different HC types. (B) Validation of differential expressions of nine genes (with underline in A) in cochlear and vestibular HCs in thin section. Bar: 10 μm for all images in B. (C) Confocal images of expression of DNM1, SLC7A14, and TJAP1 in cochlear and vestibular HCs. Bar: 10 μm for all images in C.
Expression of genes related to HC specialization.
(A) Heatmap showing expression of genes related to stereocilia bundles, mechanotransduction, ion channels, and synaptic structure. (B) Validation of gene expression using single-molecule fluorescent in situ hybridization (smFISH). Bar: 10 μm.
Cilia-related genes detected in cochlear and vestibular HCs.
(A) Schematic illustration of HC hair bundle (adapted from Schwander et al., 2010). (B) Venn diagram of the number of genes in each database and the cilia-related genes detected in HC transcriptomes. (C) Schematic illustration of primary and motile cilia, highlighting 9 + 0 or 9 + 2 arrangement of microtubules for primary and motile cilia, respectively: radial spokes (RS), central pair complex (CPC), nexin–dynein regulatory complex (N-DRC), microtubule inner proteins (MIPs), inner and outer dynein arms (IDA and ODA). (D) Expression of top 50 cilia-related genes and genes related to IFT in the four types of HCs. (E) Immunostaining of IFT172 and CLUAP1 expression in vestibular kinocilia. Bar: 5 µm. (F) Violin plots showing aggregated expression of genes associated with 96 nm repeat. Expression value of these genes is based on Supplementary file 1. (G) Heatmaps of comparison of gene expressions related to motile cilia machinery in cochlear and vestibular HCs. Red asterisks indicate the genes whose encoded proteins are expressed in both cilia and cytoplasm or are multifunctional.
© 2025, BioRender Inc. Panel C was created with BioRender and is published under a CC BY 4.0 license. Further reproductions must adhere to the terms of this license.
Gene ontology (GO) analysis of shared genes in the current study.
(A) Cellular localization, (B) molecular function, and (C) biological processes. The analysis reveals enrichment in terms related to cilia organization, assembly, maintenance, intracellular transport, and microtubule dynamics, particularly those associated with motile cilia.
Primary cilia-related genes in four HC subtypes.
Overlap of shared genes involved in cilia maintenance, including microtubule-associated proteins, BBSome family members, and components of the transition zone and primary cilia signaling pathways associated with primary cilia and their basal body.
Comparison of vestibular HC single-cell RNA-sequencing (scRNA-seq) data with proteomics datasets from multiple tissues and species containing motile cilia.
(A–D) Cross-species comparisons of motile cilia proteomes from different tissues, each compared with VHC scRNA-seq data: (A) Human trachea, oviduct, and sperm. (B) Bovine trachea, oviduct, and sperm. (C) Porcine sperm and ventricle. (D) Mouse sperm. (E–G) Tissue-specific comparisons across species, each compared with VHC scRNA-seq data: (E) Sperm proteomes from bovine, mouse, human, and porcine. (F) Tracheal proteomes from human and bovine. (G) Oviduct proteomes from human and bovine. (H) Number of proteins identified in each dataset included in the comparisons.
Enhanced accessibility of motile cilia-associated gene loci in adult vestibular HCs based on published ATAC-seq data (Jen et al., 2019).
Expression of motile cilia-related genes/proteins in the vestibular HCs.
(A) Expression of motile cilia-related genes in zebrafish, mouse, and human vestibular HCs. Expression values of these genes are based on Supplementary file 1 (mouse), Data Citation 4 (Barta et al., 2018; GSE101693, zebrafish), single-cell RNA-sequencing (scRNA-seq) dataset (Wang et al., 2024, GSE207817, human). Mouse gene nomenclature is used in heatmaps. (B) Confocal images of the expression of key motile cilia-related proteins. Scale bars represent 5 µm. (C) SEM micrograph of hair bundles of OHCs from P2 cochlea. Kinocilia (in magenta) are still present at this age. Bar: 2.5 μm. (D) Comparison of expression of motile cilia-related genes between P2 cochlear and vestibular HCs. Gene expression values are based on HC transcriptomic dataset by Burns et al., 2015. Red asterisks mark the genes whose encoded proteins are expressed in both cilia and cytoplasm or multifunctional. Red arrows indicate Dnah5 and Dnah6, which were not detected in P2 cochlear HCs. (E) Confocal images of expression of CCDC39, CCDC40, and DNAH6 in P2 vestibular and cochlear HCs. CCDC39, CCDC40, and DNAH6 were not expressed in cochlear HCs at P2. Bar: 5 μm.
Kinocilia morphology and motility.
(A) Transmission electron microscopy (TEM) images of stereocilia and kinocilium from bullfrog crista HCs. Different regions of the kinocilium in higher magnification are also shown. Long black arrows indicate where the magnified images were taken. Bars: 250 nm. Red arrow indicates two central microtubule singlets. Short black arrows mark the absence of central microtubule singlets in the distal regions near the tip of kinocilium and transition zone. (B) Images captured from in vitro live imaging of kinocilium and bundle motion of a bullfrog crista HC. The images were captured at a speed of 15 frames per second. Black arrows indicate kinocilium. (C) Representative waveforms of spontaneous cilia motion from middle ear tissue. The FFT analysis of cilia motion is also shown. (D) Three representative waveforms of spontaneous motion of hair bundles. The response waveform in blue was taken from a hair bundle with no spontaneous motility. FFT analysis of bundle motion is shown. Response waveforms and spectra are color-coded and -matched.
Predicted models of the molecular architecture of 96 nm axonemal repeat of vestibular kinocilia.
Longitudinal and cross-sectional views of the doublet microtubule (DMT) and associated structure in 96 nm repeat, derived from combining cryo-electron microscopy (cryo-EM) data and single-cell transcriptomic analysis from human respiratory cilia (A) and bovine sperm flagella (B). Key axonemal motile-machinery components are color-coded: ODA (Indian red), IDA (cyan), N-DRC (green), MIPs (orchid), RS (purple), and external coiled-coils (blue). Radial spoke 3 (RS3) has not been resolved to atomic resolution, but its shorter form (RS3s) is depicted. DMTs are represented in gray. Regions highlighted in gold indicate the absence of corresponding transcripts in our mouse transcriptomic data. (C) Genes which are not detected in mouse and human vestibular HC transcriptomes and related to motility-relevant compartments are listed in the table. The roles of these genes in the 96 nm repeat module and cilia motility and ciliopathy are also included.
Videos
Bullfrog kinocilia motility.
Bullfrog kinocilia motility.
Predicted structures models of molecular composition of the vestibular kinocilium based on the structural frameworks derived from human (Video 3) respiratory (PDB: 8J07) (Walton et al., 2023) and bovine (Video 4) sperm (PDB: 9FQR) (Leung et al., 2025) axonemes.
Predicted structures models of molecular composition of the vestibular kinocilium based on the structural frameworks derived from bovine sperm.
Additional files
-
Supplementary file 1
Transcriptomes of four HC types.
- https://cdn.elifesciences.org/articles/108071/elife-108071-supp1-v1.xlsx
-
Supplementary file 2
Genes/gene products candidates associated with axonemal components in vestibular kinocilia derived from the 96 nm repeat structures of the human respiratory axoneme (PDB: 8J07).
Missing genes in mouse vestibular hair cells are in red.
- https://cdn.elifesciences.org/articles/108071/elife-108071-supp2-v1.xlsx
-
Supplementary file 3
Microtubule inner proteins and microtubule-associated proteins absent from mouse HC scRNA-seq data.
- https://cdn.elifesciences.org/articles/108071/elife-108071-supp3-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/108071/elife-108071-mdarchecklist1-v1.docx