Native Hair Cells Research Articles

Precise genetic control of ATOH1 enhances maturation of regenerated hair cells in the mature mouse utricle

Vestibular hair cells are mechanoreceptors critical for detecting head position and motion. In mammals, hair cell loss causes vestibular dysfunction as spontaneous regeneration is nearly absent. Constitutive expression of exogenous ATOH1, a hair cell transcription factor, increases hair cell regeneration, however, these cells fail to fully mature. Here, we profiled mouse utricles at 14 time points, and defined transcriptomes of developing and mature vestibular hair cells. To mimic native hair cells which downregulate endogenous ATOH1 as they mature, we engineered viral vectors carrying the supporting cell promoters GFAP and RLBP1. In utricles damaged ex vivo, both CMV-ATOH1 and GFAP-ATOH1 increased regeneration more effectively than RLBP1-ATOH1, while GFAP-ATOH1 and RLBP1-ATOH1 induced hair cells with more mature transcriptomes. In utricles damaged in vivo, GFAP-ATOH1 induced regeneration of hair cells expressing genes indicative of maturing type II hair cells, and more hair cells with bundles and synapses than untreated organs. Together our results demonstrate the efficacy of spatiotemporal control of ATOH1 overexpression in inner ear hair cell regeneration.

AAV-mediated Gene Cocktails Enhance Supporting Cell Reprogramming and Hair Cell Regeneration.

Mammalian cochlear hair cells (HCs) are essential for hearing, and damage to HCs results in severe hearing impairment. Damaged HCs can be regenerated by neighboring supporting cells (SCs), thus the functional regeneration of HCs is the main goal for the restoration of auditory function in vivo. Here, cochlear SC trans-differentiation into outer and inner HC by the induced expression of the key transcription factors Atoh1 and its co-regulators Gfi1, Pou4f3, and Six1 (GPAS), which are necessary for SCs that are destined for HC development and maturation via the AAV-ie targeting the inner ear stem cells are successfully achieved. Single-cell nuclear sequencing and lineaging tracing results showed that the majority of new Atoh1-derived HCs are in a state of initiating differentiation, while GP (Gfi1, Pou4f3) and GPS (Gfi1, Pou4f3, and Six1) enhanced the Atoh1-induced new HCs into inner and outer HCs. Moreover, the patch-clamp analysis indicated that newborn inner HCs induced by GPAS forced expression have similar electrophysiological characteristics to those of native inner HCs. Also, GPAS can induce HC regeneration in the HC-damaged mice model. In summary, the study demonstrates that AAV-mediated co-regulation of multiple genes, such as GPAS, is an effective means to achieve functional HC regeneration in the mouse cochlea.

The role of Espin in the stereocilia regeneration and protection in Atoh1-overexpressed cochlear epithelium.

Hair cells (HCs) in mammals cannot spontaneously regenerate after damage. Atoh1 overexpression can promote HC regeneration in the postnatal cochlea, but the regenerated HCs do not possess the structural and functional characteristics of HCs in situ. The stereocilia on the apical surface of HCs are the first-level structure for sound conduction, and regeneration of functional stereocilia is the key basis for the reproduction of functional HCs. Espin, as an actin bundling protein, plays an important role in the development and structural maintenance of the stereocilia. Here, we found that the upregulation of Espin by AAV-ie was able to induced the aggregation of actin fibres in Atoh1-induced HCs in both cochlear organoids and explants. In addition, we found that persistent Atoh1 overexpression resulted in impaired stereocilia in both endogenous and newly formed HCs. In contrast, the forced expression of Espin in endogenous and regenerative HCs was able to eliminate the stereocilia damage caused by persistent Atoh1 overexpression. Our study shows that the enhanced expression of Espin can optimize the developmental process of stereocilia in Atoh1-induced HCs and can attenuate the damage to native HCs induced by Atoh1 overexpression. These results suggest an effective method to induce the maturation of stereocilia in regenerative HCs and pave the way for functional HC regeneration via supporting cell transdifferentiation.

Generation of innervated cochlear organoid recapitulates early development of auditory unit.

Functional cochlear hair cells (HCs) innervated by spiral ganglion neurons (SGNs) are essential for hearing, whereas robust models that recapitulate the peripheral auditory circuity are still lacking. Here, we developed cochlear organoids with functional peripheral auditory circuity in a staging three-dimensional (3D) co-culture system by initially reprogramming cochlear progenitor cells (CPCs) with increased proliferative potency that could be long-term expanded, then stepwise inducing the differentiation of cochlear HCs, as well as the outgrowth of neurites from SGNs. The function of HCs and synapses within organoids was confirmed by a series of morphological and electrophysiological evaluations. Single-cell mRNA sequencing revealed the differentiation trajectories of CPCs toward the major cochlear cell types and the dynamic gene expression during organoid HC development, which resembled the pattern of native HCs. We established the cochlear organoids with functional synapses for the first time, which provides a platform for deciphering the mechanisms of sensorineural hearing loss.

Mechanical gating of the auditory transduction channel TMC1 involves the fourth and sixth transmembrane helices.

The transmembrane (TM) channel-like 1 (TMC1) and TMC2 proteins play a central role in auditory transduction, forming ion channels that convert sound into electrical signals. However, the molecular mechanism of their gating remains unknown. Here, using predicted structural models as a guide, we probed the effects of 12 mutations on the mechanical gating of the transduction currents in native hair cells of Tmc1/2-null mice expressing virally introduced TMC1 variants. Whole-cell electrophysiological recordings revealed that mutations within the pore-lining TM4 and TM6 helices modified gating, reducing the force sensitivity or shifting the open probability of the channels, or both. For some of the mutants, these changes were accompanied by a change in single-channel conductance. Our observations are in line with a model wherein conformational changes in the TM4 and TM6 helices are involved in the mechanical gating of the transduction channel.

Developmental and Functional Hair Cell-Like Cells Induced by Atoh1 Overexpression in the Adult Mammalian Cochlea In Vitro.

Hair cells (HCs) in the mammalian cochleae cannot spontaneously regenerate once damaged, resulting in permanent hearing loss. It has been shown that Atoh1 overexpression induces hair cell-like cells (HCLCs) in the cochlea of newborn rodents, but this is hard to achieve in adult mammals. In this study, we used a three-dimensional cochlear culture system and an adenoviral-mediated delivery vector to overexpress Atoh1 in adult mouse cochleae. HCLCs were successfully induced from 3 days after virus infection (3 DVI) in vitro, and the number increased with time. HCLCs were myosin7a positive and distinguishable from remnant HCs in a culture environment. Meanwhile, patch-clamp results showed that noninactive outward potassium currents (sustained outward potassium currents) could be recorded in HCLCs and that their magnitude increased with time, similar to normal HCs. Furthermore, transient HCN currents were recorded in some HCLCs, indicating that the HCLCs experienced a developmental stage similar to normal HCs. We also compared the electrophysiological features of HCLCs from adult mice with native HCs and found the HCLCs gradually matured, similar to the normal HCs. Meanwhile, HCLCs from adult mice possessed the same bundles as developmental HCs. However, these HCLCs did not express prestin, which is a special marker for outer hair cells (OHCs), even at 13 DVI. These results demonstrate that Atoh1 overexpression induces HCLC formation in the adult mammalian cochlea and that these HCLCs were functional and experienced a developmental process similar to that of normal HCs.

Probing the Structure and Function of TMC1 in Sensory Hair Cells using Mutagenesis and Cysteine Modification

Transmembrane channel-like genes 1 and 2 (Tmc1 and Tmc2) have recently been implicated as components of the sensory transduction channel in mammalian hair cells. Mice deficient in both Tmc1 and Tmc2 have complete loss of hearing and balance function and lack sensory transduction, despite intact hair bundles and tip-links (Kawashima et al., 2011). Mice that express only TMC1 or TMC2 have distinct single-channel conductances and calcium selectivity. A methionine-to-lysine substitution at position 412 in TMC1 reduces the single-channel current amplitude and calcium permeability of transduction (Pan et al., 2013). These results suggest that TMCs participate as essential components of the sensory transduction channel in auditory and vestibular hair cells, but their exact role is not yet clear. They may form a vestibule at the mouth of the pore, the pore of the ion channel itself or both (Holt et al., 2014). Alternatively, TMCs may function as non-essential accessory subunits (Beurg et al., 2014). Progress toward understanding the structure and function of TMC proteins has been limited by the lack of homology to known genes or domains (Kurima et al., 2003); the low quantity of native protein (50-100 functional channels/cell); and poor membrane localization in heterologous cells. To circumvent these limitations, we combined electrophysiological recording with mutagenesis and cysteine modification in native hair cells. To probe TMC1 structure and function, mutant Tmc1 sequences were introduced, via viral transfection, into organotypic cultures harvested from Tmc1/Tmc2 doubly-deficient mice. We find that the mutant constructs restore sensitivity to hair bundle deflections in virally-transfected hair cells and that acute application of cysteine modification reagents alters the biophysical properties of sensory transduction within seconds. The data further support a direct role for TMC1 in sensory transduction in mammalian hair cells.

Elementary properties of Kir2.1, a strong inwardly rectifying K + channel expressed by pigeon vestibular type II hair cells

By using the patch-clamp technique in the cell-attached configuration, we have investigated the single-channel properties of an inward rectifier potassium channel (Kir) expressed by pigeon vestibular type II hair cells in situ. In high-K + external solution with 2 mM Mg 2+, Kir inward current showed openings to at least four amplitude levels. The two most frequent open states (L2 and L3) had a mean slope conductance of 13 and 28 pS, respectively. L1 (7 pS) and L4 (36 pS) together accounted for less than 6% of the conductive state. Closed time distributions were fitted well using four exponential functions, of which the slowest time constant (τ C4) was clearly voltage-dependent. Open time distributions were fitted well with two or three exponential functions depending on voltage. The mean open probability (P O) decreased with hyperpolarization (0.13 at −50 mV and 0.03 at −120 mV). During pulse-voltage protocols, the Kir current-decay process (inactivation) accelerated and increased in extent with hyperpolarization. This phenomenon was associated with a progressive increase of the relative importance of τ C4. Kir inactivation almost disappeared when Mg 2+ was omitted from the pipette solution. At the same time, P O increased at all membrane voltages and the relative importance of L4 increased to a mean value of 47%. The relative importance of τ C4 decreased for all open states, while L4 only showed a significantly longer open time constant. The present work provides the first detailed quantitative description of the elementary properties of the Kir inward rectifier in pigeon vestibular type II hair cells and specifically describes the Kir gating properties and the molecule's sensitivity to extracellular Mg 2+ for all subconductance levels. The present results are consistent with the Kir2.1 protein sustaining a strong inwardly rectifying K + current in native hair cells, characterized by rapid activation time course and slow partial inactivation. The longest closed state (τ C4) appears as the main parameter involved in time- and Mg 2+-dependent decay. Finally, in contrast to Kir2.1 results described so far for mammalian cells, external Mg 2+ had no effect on channel conductance.

Kvβ1.1 associates with Kvα1.4 in chinese hamster ovary cells and pigeon type II vestibular hair cells and enhances the amplitude, inactivation and negatively shifts the steady-state inactivation range

Although A-type potassium currents are found in type II hair cells in the inner ear of most species, the molecular mechanisms for activation and inactivation of the A-type potassium current (I A) remain unknown. In frog semicircular canal hair cells, for example, there appear to be two classes of currents having either fast or slow inactivation [Norris CH, Ricci AJ, Housley GD, Guth PS (1992) The inactivating potassium currents of hair cells isolated from the crista ampullaris of the frog. J Neurophysiol 68:1642–1653; Russo G, Calzi D, Martini M, Rossi ML, Fesce R, Prigioni I (2007) Potassium currents in the hair cells of vestibular epithelium: position-dependent expression of two types of A channels. Eur J Neurosci 25:695–704]. It has been suggested that somehow the “ball and chain” mechanism (NH 3 (N) terminus motif) is modified by alternative splicing to account for the two classes of inactivation. To examine other possibilities, we cloned alpha and beta subunits that comprise the A-type potassium channel complex in adult and embryonic pigeon brain, cochlea and labyrinth. By sequence homology, we concluded that the subunits present were Kvα1.4 and Kvβ1.1. The sequence of the open reading frame for Kvα1.4 contained the N-terminus, pore and COOH (C) terminus motifs for N-and C-type inactivation. The sequence for Kvβ1.1 displayed amino acids consistent with assembly and association with Kvα1.4 alpha subunits. Kvα1.4 and Kvβ1.1 were transfected either singly or in combination into Chinese hamster ovary (CHO) cells. These cells and native hair cells from the pigeon utricle were patch clamped and the inactivation properties of the A-type current were studied. In the native hair cells, the A-type current was identified by its pharmacological (4-aminopyridine (4-AP); IC 50=11 μM) and voltage dependent inactivation properties. A comparison of the mean time constants from best-fitted single exponential and sum of two exponential equations to the ionic current inactivation revealed the following. In CHO cells when Kvα1.4 was expressed alone, the mean time constant (τ 1=107 ms±19, N=32) was significantly ( P<0.001) longer and the mean peak amplitude (2.28 nA±0.39, N=32) was smaller than when Kvα1.4 and Kvβ1.1 were expressed in CHO cells. Moreover, the co-transfection of Kvα1.4 and Kvβ1.1 into CHO cells caused a shift in the steady state inactivation curve parameter Vo 30 mV in the hyperpolarized direction relative to CHO cells expressing only Kvα1.4. Similarly, Kvα1.4-transfected CHO cells produced longer time constants and smaller amplitudes than those found for native utricular hair cells. These data lead us to conclude that while the amino acid motifs are present in Kvα1.4 and Kvβ1.1 to suggest N-and C-type inactivation, co-assembly and association of Kvα1.4 and Kvβ1.1 may also produce changes in the time dependent inactivation properties of vestibular hair cells.

Hair cell regeneration: Look to the future

Less than 2 decades ago it was discovered that birds can regenerate hair cells in the auditory and vestibular parts of the inner ear after the native hair cells are destroyed by exposure to excessive noise or by mechanical trauma of aminoglycoside antibiotics. This discovery issued in a new era of hearing research—it suggested that some day it may be possible to actually restore hearing in people with congenital or acquired hearing loss due to the degeneration of sensory cells or supporting cells in the inner ear. Fifteen years is a very short time in the history of science. Consider the fact that we have actively sought chemical treatments to prevent or cure cancers for well over a half century and the ‘‘war on Cancer,’’ resulted in enormous public and private support. Progress has been great, and some forms of cancer can be treated with great success, but the overall 5-year survival rates have only risen from about 50% to 63%. Progress will continue and many more forms of cancer will be cured and prevented during the next half century. Similarly, during the first 15 years of hair cell regeneration research enormous progress has been made, and we now know that postnatal mammalian ears have the capacity to produce new hair cells. We are indeed a long way from restoring hearing through hair cell regeneration, but the future is pretty clear. I will review the progress of this field with an eye toward the future and what it means for treatments of today. In particular, I will address the potential cost versus benefits of bilateral implantation when applied to babies and young children.

Avian cochlear hair cell regeneration: stereological analyses of damage and recovery from a single high dose of gentamicin

Hair cell regeneration after acoustic trauma has been conclusively documented in birds. Previous studies of aminoglycoside ototoxicity have typically used 5–10 day courses of drug to damage the cochlea and trigger regeneration. This long-term lesion prevented analysis of the early events of regeneration. We set out to determine how much damage would occur and how recovery would proceed after a single high-dose injection of the aminoglycoside gentamicin. White Leghorn chicks were given a single high dose of gentamicin (100 mg/kg). Three post-injection survival groups with age-matched controls were studied: short-term (3–5 days), intermediate-term (2 weeks) and long-term (5 weeks). After sacrifice, cochleae were dissected and processed for scanning electron microscopy. Using stereological techniques, a quantitative analysis of cochlear hair cell counts along the proximal 50% of the cochlea was performed from scanning electron micrographs on 4–7 chicks from each group. Variable degrees of damage were seen 3–5 days after the drug injection. All hair cells were lost from the proximal 20% of the cochlea in all chicks. This complete hair cell loss could extend to 50% of the cochlea. Immature appearing hair cells could be first identified by their immature stereocilia at 3 days. Immature appearing hair cells were present in greatest number in regions which had been denuded of native hair cells and in regions where partial loss occurred. Interestingly, immature appearing hair cells also occasionally appeared in adjacent areas in which there was no apparent loss of native hair cells. Two-week survivors showed an elevation in hair cell number compared to controls in regions which had sustained damage and immediately adjacent regions. This elevation implies that an overproduction of hair cells might occur as part of the regeneration response. By 5 weeks after damage hair cell numbers approximated controls.