Hearing impairment is the fastest growing and one of the most prevalent chronic conditions today with over 5% of the world's population—360 million people have disabling hearing loss (328 million adults and 32 million children) and approximately one-third of individuals over 65 years are affected by disabling hearing loss (World Health Organization global estimates on prevalence of hearing loss, 2012). Furthermore, congenital hearing loss affects two to three infants per 1,000 live births (1). The cochlear implant has been successful at providing hearing to people with sever-to-profound hearing loss and as of 2012, an estimated 324,000 patients worldwide have received cochlear implants (National Institute on Deafness and Other Communication Disorders, NIDCD); however, the performance of cochlear implants is largely dependent on the degree of degeneration or loss of primary auditory neurons (PANs, spiral ganglion neurons). Once damaged or lost due to disease, noise, or aging, there is no effective way to regenerate PANs in humans or other mammals (for a comprehensive review on PANs see (2)). Although as low as 10% of surviving PANs is sufficient to send meaningful sound information to the brain (3), the number of surviving PANs is positively correlated with word recognition (4). Thus, PANs are a primary target for regeneration, and replacement therapy would have significant impact on research and advancement in cochlear implants and the amelioration of hearing loss.
The 2016 Symposium “Cochlear Implants Meet Regenerative Biology” provided a broad overview of the potential benefits of integrating regenerative biology with cochlear implants for the treatment of hearing loss and presented the most recent advances and key challenges in the field as well as the potential for an entirely biological solution. This review summarizes the presentations of auditory neuron research being conducted in the speakers’ laboratories: Drs. M. Charles Liberman, Harvard Medical School, MA; Gary Housley, The University of South Wales, Australia; Tobias Moser, University Medical Center Goettingen, Germany; Marcelo Rivolta, University of Sheffield, United Kingdom; and Alain Dabdoub, Sunnybrook Research Institute/University of Toronto, Canada.
NOISE-INDUCED AND AGE-RELATED NEURAL DEGENERATION
The loss of PANs has been considered secondary to hair cell; however, accumulating evidence unmistakably demonstrates that PANs can degenerate primarily due to aging and noise exposure with little or no effect on other cell types including mechanosensory hair cells. Thus, a significant number of people are suffering from primary PAN degeneration and the resulting hearing loss, which has remained hidden, since a large loss of PANs can be compensated for by increasing discharge rates in remaining fibers with low threshold and high spontaneous rates (SRs) without threshold elevation until the loss becomes extreme (5). This aspect partially benefits patients with few remaining PANs since cochlear implants (CIs) can still send sound information to the brain. However, fibers with high threshold and low-SRs are more vulnerable and play a role in supra-threshold coding responsible for speech discrimination in challenging noisy listening environments (6). The difficulty CI users encountered in discriminating speech in noise may partially come from dysfunction of supra-threshold coding. More than three decades ago it was demonstrated that primary loss of PANs occurred due to aging before hair cell loss (7,8). More recently, Sergeyenko et al. (9) demonstrated that loss of PANs parallels cochlear synaptic loss. Moreover, using high-powered confocal imaging of synapses, physiological tests such as auditory brainstem responses (ABRs) and distortion product otoacoustic emissions (DPOAEs) and conventional histology for quantification of the neuronal cells and hair cells, Liberman presented recent evidence from his group that clearly indicates that PANs can degenerate primarily as a result of moderate noise exposure with little or no effect on hair cells (5,6,10,11). Specifically, moderate acoustic overexposure causing reversible threshold elevation leaves cochlear hair cells intact but causes acute loss of the synapse between the type I PAN terminal and the inner hair cells, followed by delayed loss of PANs (5,10). The loss of synapses is highly biased towards the subset of PANs with high thresholds (10), which are responsible for processing supra-threshold sounds resulting in difficulties hearing in low signal-to-noise conditions such as hearing a conversation in a noisy environment. Therefore, hearing aids, cochlear implants, as well as normal hearing all require the presence of healthy PANs including more vulnerable low-SR neurons to transmit sound signals from the inner ear to the brain. Lastly, Suzuki et al. (12) presented strategies for neural regeneration in hidden hearing loss due to noise-induced synaptopathy: local delivery of neurotrophin 3 via the round window after acoustic overexposure regenerated cochlear synapses in mice coupled with a functional recovery, as seen in the supra-threshold amplitudes for ABR wave 1, which represents the summed activity of the PANs.
NEUROTROPHIN GENE DELIVERY USING COCHLEAR IMPLANTS
The neural interface of cochlear implants has changed little over time (13), and despite the considerable advances in the speech processor capabilities, hearing quality with CIs is still far from natural (14). The current spread that occurs at the stimulus levels required to recruit PANs is such that position-based mapping is coarse, thus cochlear implant recipients rely on central auditory processing to interpret speech resulting in poor pitch perception. A key challenge is to use the linear electrode array comprising just a few platinum electrodes to selectively drive the PANs normally tonotopically mapped along the length of the cochlea. Central to this is closing the “neural gap” between the electrodes and the surviving PANs (13). While temporal treatment with brain-derived neurotrophic factor (BDNF) by osmotic pump had long-term effects on preserving deafness-induced degeneration of PANs in guinea pigs (15), gene therapy is still attractive for correct neurite extension toward an electrode array, since inner ear mesenchymal cells that are genetically modified to emit neurotrophins can lure neurites toward an electrode.
Pinyon et al. (16.) investigated the potential of the cochlear implant electrode arrays as gene electrotransfer platform for focal delivery of plasmid DNA with a BDNF expression cassette incorporating a green fluorescent protein reporter thus using electroporation as an alternative strategy for cochlear gene delivery. Voltage pulses are delivered through the CI array to electroporate mesenchymal cells in close proximity to the array. Using ex vivo and in vivo experiments with the guinea-pig cochleae, they identified that the wiring configuration of the CI implant array was a key factor in the efficiency of gene delivery. Mapping of the electric field around the array in different configurations showed that electric field focusing produced a compression of field strength which enables gene electrotransfer with applied voltages lower than that necessary for conventional “open-field” electroporation (17). The transfected mesenchymal cells expressed BDNF and within 7 days PAN neurite extension beyond the osseous spiral lamina and into scala tympani occurred. Functional assessment using bipolar stimulation via the CI for electrically-evoked ABR (eABR) in a deafened guinea pig model indicated that this gene therapy treatment had the desired effects of lowering threshold current levels for PAN recruitment and increasing the dynamic range of the input–output function, indicative of progressive recruitment into the local PAN population as the current levels were increased. These promising developments around local DNA-based gene electrotransfer in the cochlea support further studies and utilization of this process for targeted neurotrophin gene therapy during CI surgery to enhance PAN survival and optimize the neural interface.
FUTURE OPTICAL COCHLEAR IMPLANTS
When hearing fails, speech comprehension can be restored by auditory prostheses. However, as discussed above, sound coding with the current prostheses, based on electrical stimulation of PANs, has limited frequency resolution due to broad current spread. One-to-two dozen electrode contacts that electric CIs typically use do not fully make use of the remaining PANs, which perform fine tuning of frequency-specific auditory information. Thus, improving frequency and intensity resolution of CI are key objectives for restoring better hearing. Hernandez et al. (18) presented an innovative method to tackle current spread by using optical stimulation which can be spatially restricted, thus enhancing frequency resolution. Utilizing an optogenetic animal model in which auditory neurons were engineered to express light-sensitive ion channels (channelrhodopsins) and optical stimulation technology, they were able to effectively activate the auditory pathway. These studies demonstrate the clear potential for future applications in hearing research and restoration—see reviews by Moser (19) and Jeschke and Moser (20).
Hernandez et al. (18) has established optogenetic stimulation of the auditory pathway in rodents using virus-mediated expression of channelrhodopsins to render PANs light-sensitive. Optogenetic stimulation of PANs activated the auditory pathway, as demonstrated by recordings of single neuron and neuronal population responses at various stages along the auditory system. They approximated the spatial spread of cochlear excitation by recording local field potentials in the inferior colliculus in response to supra-threshold optical and electrical stimuli, which suggested a better frequency resolution for optogenetic than for electrical stimulation. Moreover, they were able to restore auditory activity in deaf mice. In a collaborative effort they developed and characterized flexible μLED-based multichannel intracochlear stimulators. Optogenetic stimulation of the auditory nerve is feasible and holds great potential for future application in research and hearing restoration. Lastly, Moser (19) mentioned three challenges for optical cochlear implants as discussed in Jeschke and Moser (20): for efficient, reliable, and safe expression of light-sensitive opsins in PANs they have used replication deficient adeno-associated virus (AAV)-2/6 and observed no side effects upon transfection (21); the optical stimulation device needs to be small enough for implantation and needs to emit sufficient light power to stimulate PANs; and finally, they need to demonstrate an advantage of optical stimulation over electrical stimulation. Theoretically, optical stimulation has an advantage in that it has the increased spatial resolution with more stimulation channels and future psychophysical experiments can test the advantage in practical use.
DEVELOPING MODELS TO STUDY COCHLEAR IMPLANTS AND STEM CELLS
The development of animal models for human deafness continues to be central for the exploration of potential therapies. The gerbil is an ideal model for cochlear implantation for several reasons including the similarity of their auditory frequency range to humans, their relatively large cochlear size, and their robust recovery postsurgery. Chen et al. (22) have previously shown that human embryonic stem cell derived-otic neuroprogenitors can be transplanted in a gerbil model of auditory neuropathy and elicit a functional recovery. As a next step in developing a cell-based therapy for hearing loss, they investigated whether stem cells could be employed together with a cochlear implant.
They initially developed a reliable protocol for hair cell damage with preservation of PANs in the gerbil using kanamycin and furosemide (23), which is an ideal model for CI studies as it mirrors the situation often found in human patients. Allowing a week for recovery, animals then received a custom-built cochlear implant via a round window cochleostomy. In other groups, an implant was inserted in untreated, naïve animals. The custom-built electrodes were designed to be stimulated by an electro-magnetic field. Their initial data showed that both a behavioral response in a waking animal and an eABR wave in an anesthetized animal can be induced by stimulating the implant's chip. This is analogous to the functioning of the speech processor unit of a medical implant. These responses can be seen in both naïve and hair-cell ablated animals. In the latter group, they have also shown that a single, systemic administration of kanamycin and furosemide is not followed by secondary degeneration of the PANs in the gerbil animal model. This is similar to the situation seen in humans with aminoglycoside-induced ototoxicity, where functioning neurons can persist after hair cell loss (24), contrasting with the rapid secondary degeneration observed in other species, such as guinea pigs (25), rats (26), and chinchilla (27); thus they anticipated that the implants would continue to function long-term in gerbils whose hair cells were ablated.
They are currently exploring hair cell ablation in combination with the selective ablation of PANs using ouabain (28), to generate a “double knock-down” model. Topical application of ouabain, which blocks Na+–K+–ATPase, induces apoptosis of type I PANs (28). They aim to replace hair cells by a cochlear implant (29) and the PANs with stem cells (22), in what should be a true “bionic” prosthesis.
REPROGRAMING ENDOGENOUS CELLS INTO INDUCED AUDITORY NEURONS
Cellular reprograming, the direct conversion of one cell type to another, is an emerging area of regenerative medicine (30). The use of transcription factors to transdifferentiate somatic cells into terminally differentiated cell types, for example, fibroblasts into muscle cells (31) and more recently fibroblasts into induced neurons (32), has opened the door for potential use in clinical applications. In terms of neuron induction, the basic helix-loop-helix transcription factor Ascl1 has emerged as the main reprograming factor due to its ability to bind closed chromatin and promote accessibility at the regulatory regions of its targets (33); thus Ascl1 is considered a pioneer factor.
Similarly, Puligilla et al. (34) have demonstrated that the proneural basic helix-loop-helix transcription factors Neurog1 and NeuroD1, which are required for the formation and survival, respectively, of the PANs (35–37), converted cochlear non-sensory epithelial cells into neuronal cells. This research approach also demonstrated that the high-mobility group (HMG) box domain transcription factor Sox2 induced neurons. The neuronal conversion occurred at early embryonic stages with a progressive decrease in the percentage of cells that developed as neurons at older stages.
More recently, to improve neuron induction efficiency and induce neurons at older stages, Nishimura et al. (38) used Ascl1. The pioneering factor reprogrammed cochlear non-sensory epithelial cells into neurons at embryonic, postnatal, and juvenile stages. Induced neurons were defined by morphology, expression of neuronal markers, synaptic proteins, and the generation of action potentials. Moreover, a combination of Ascl1 and NeuroD1 induced neurons, which had a trend of being electrophysiologically more mature resembling endogenous PANs.
The ultimate goal is to induce PANs in human inner ears and to enable functional innervation to both the cochlear nucleus in the central nervous system and hair cells or cochlear implants in the periphery. Spiral ganglion non-neural cells, comprised mainly of glia, are an excellent target cell population for replacing damaged neurons by cellular reprograming as these glial cells are resident within the modiolus and survive after neuron loss. Dabdoub also presented unpublished work using Ascl1 to reprogram spiral ganglion glial cells into neurons observing similarities between induced neurons and endogenous PANs in neuronal protein expression. Moreover, they are now testing the hypothesis that PAN developmental factors such as Neurog1 and NeuroD1 in combination with Ascl1 will induce neurons which more closely resemble endogenous neurons.
Although inner ear regenerative studies are mostly conducted in vitro or in some animal models, continued success and advancements in regenerative biology will provide the necessary platform for translational work and clinical application. This combined with technological advances and a rising effort to address the ever growing and prevalent chronic condition of hearing loss, will enable researchers to tackle the significant challenges to one day realize a therapeutic solution for the treatment of hearing impairment. Furthermore, to achieve more immediate goals for cochlear implant users, to understand speech in noise and to appreciate music, the importance of healthy PANs as well as next-generation stimulation technologies and improvements of the neural interface cannot be overemphasized.
The authors thank our invited speakers: Drs. M. C. Liberman, G. Housley, M. Rivolta, and T. Moser for their excellent presentations and for their comments on this manuscript. We also thank Dr. E. Keithley for her comments on the manuscript.
1. Vohr B. Overview: Infants and children with hearing loss-part I. Ment Retard Dev Disabil Res Rev
2. Dabdoub A, Fritzsch B, Popper AN, et al. The Primary Auditory Neurons of the Mammalian Cochlea. New York: Springer; 2016.
3. Fayad JN, Linthicum FH Jr. Multichannel cochlear implants: relation of histopathology to performance. Laryngoscope
4. Seyyedi M, Viana LM, Nadol JB Jr. Within-subject comparison of word recognition and spiral ganglion
cell count in bilateral cochlear implant
recipients. Otol Neurotol
5. Lin HW, Furman AC, Kujawa SG, et al. Primary neural degeneration in the Guinea pig cochlea after reversible noise-induced threshold shift. J Assoc Res Otolaryngol
6. Kujawa SG, Liberman MC. Synaptopathy in the noise-exposed and aging cochlea: Primary neural degeneration in acquired sensorineural hearing loss. Hear Res
7. Keithley EM, Feldman ML. Spiral ganglion
cell counts in an age-graded series of rat cochleas. J Comp Neurol
8. Keithley EM, Feldman ML. Hair cell counts in an age-graded series of rat cochleas. Hear Res
9. Sergeyenko Y, Lall K, Liberman MC, et al. Age-related cochlear synaptopathy: an early-onset contributor to auditory functional decline. J Neurosci
10. Furman AC, Kujawa SG, Liberman MC. Noise-induced cochlear neuropathy is selective for fibers with low spontaneous rates. J Neurophysiol
11. Kujawa SG, Liberman MC. Adding insult to injury: cochlear nerve degeneration after “temporary” noise-induced hearing loss. J Neurosci
12. Suzuki J, Corfas G, Liberman MC. Round-window delivery of neurotrophin
3 regenerates cochlear synapses after acoustic overexposure. Sci Rep
13. O’Leary SJ, Richardson RR, McDermott HJ. Principles of design and biological approaches for improving the selectivity of cochlear implant
electrodes. J Neural Eng
14. Li X, Nie K, Imennov NS, et al. Improved perception of music with a harmonic based algorithm for cochlear implants. IEEE Trans Neural Syst Rehabil Eng
15. Ramekers D, Versnel H, Strahl SB, et al. Temporary neurotrophin
treatment prevents deafness-induced auditory nerve degeneration and preserves function. J Neurosci
16. Pinyon JL, Tadros SF, Froud KE, et al. Close-field electroporation
gene delivery using the cochlear implant
electrode array enhances the bionic ear. Sci Transl Med
17. Browne CJ, Pinyon JL, Housley DM, et al. Mapping of bionic array electric field focusing in plasmid DNA-based gene electrotransfer. Gene Ther
18. Hernandez VH, Gehrt A, Reuter K, et al. Optogenetic stimulation of the auditory pathway. J Clin Investig
19. Moser T. Optogenetic stimulation of the auditory pathway for research and future prosthetics. Curr Opin Neurobiol
20. Jeschke M, Moser T. Considering optogenetic stimulation for cochlear implants. Hear Res
21. Yin H, Kanasty RL, Eltoukhy AA, et al. Non-viral vectors for gene-based therapy. Nat Rev Genet
22. Chen W, Jongkamonwiwat N, Abbas L, et al. Restoration of auditory evoked responses by human ES-cell-derived otic progenitors. Nature
23. Abbas L, Rivolta MN. Aminoglycoside ototoxicity and hair cell ablation in the adult gerbil: A simple model to study hair cell loss and regeneration. Hear Res
24. Nadol JB Jr. Patterns of neural degeneration in the human cochlea and auditory nerve: implications for cochlear implantation. Otolaryngol Head Neck Surg
25. Versnel H, Agterberg MJ, de Groot JC, et al. Time course of cochlear electrophysiology and morphology after combined administration of kanamycin and furosemide. Hear Res
26. Ma L, Yi HJ, Yuan FQ, et al. An efficient strategy for establishing a model of sensorineural deafness in rats. Neural Regen Res
27. McFadden SL, Ding D, Jiang H, et al. Time course of efferent fiber and spiral ganglion
cell degeneration following complete hair cell loss in the chinchilla. Brain Res
28. Schmiedt RA, Okamura HO, Lang H, et al. Ouabain application to the round window of the gerbil cochlea: a model of auditory neuropathy and apoptosis. J Assoc Res Otolaryngol
29. Millard RE, Shepherd RK. A fully implantable stimulator for use in small laboratory animals. J Neurosci Methods
30. Srivastava D, DeWitt N. In vivo cellular reprogramming: the next generation. Cell
31. Davis RL, Weintraub H, Lassar AB. Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell
32. Vierbuchen T, Ostermeier A, Pang ZP, et al. Direct conversion of fibroblasts to functional neurons by defined factors. Nature
33. Wapinski OL, Vierbuchen T, Qu K, et al. Hierarchical mechanisms for direct reprogramming of fibroblasts to neurons. Cell
34. Puligilla C, Dabdoub A, Brenowitz SD, et al. Sox2 induces neuronal formation in the developing mammalian cochlea. J Neurosci
35. Kim WY, Fritzsch B, Serls A, et al. NeuroD-null mice are deaf due to a severe loss of the inner ear sensory neurons during development. Development
36. Ma Q, Chen Z, del Barco Barrantes I, et al. neurogenin1 is essential for the determination of neuronal precursors for proximal cranial sensory ganglia. Neuron
37. Ma Q, Anderson DJ, Fritzsch B. Neurogenin 1 null mutant ears develop fewer, morphologically normal hair cells in smaller sensory epithelia devoid of innervation. J Assoc Res Otolaryngol
38. Nishimura K, Weichert RM, Liu W, et al. Generation of induced neurons by direct reprogramming in the mammalian cochlea. Neuroscience