Secondary Logo

Journal Logo

AUDITORY AND VESTIBULAR SCIENCE: Edited by Robert S. Hong and Zhengqing Hu

Subtype maturation of spiral ganglion neurons

Sun, Shuohao; Siebald, Caroline; Müller, Ulrich

Author Information
Current Opinion in Otolaryngology & Head and Neck Surgery: October 2021 - Volume 29 - Issue 5 - p 391-399
doi: 10.1097/MOO.0000000000000748
  • Open

Abstract

INTRODUCTION

Hearing starts in the inner ear with the conversion of sound-induced vibration into electrical signals. This conversion process is carried out by mechanosensory hair cells within the cochlear sensory epithelium (Fig. 1A). Three rows of outer hair cells (OHCs) amplify sound-induced vibrations that reach the cochlea. These input signals lead to the activation of inner hair cells (IHCs), which transmit the sound information to the central nervous system (CNS) [1,2]. Hair cells are innervated by spiral ganglion neurons (SGNs) that can be divided into two types. So-called type I SGNs constitute 95% of all SGNs, whereas the remaining 5% are called type II SGNs [3–5]. Type I SGNs innervate single IHCs, whereas type II SGNs extend long projections and innervate several dozen OHCs (Fig. 1A). Type I SGNs transmit sound information to the CNS, whereas type II SGNs appear to have a role in damage perception and pain signaling [6,7]. Although type I SGNs typically receive input from only one IHC, each IHC is innervated by 5 - 30 type I SGNs [8]. This intriguing synaptic organization raises the question of whether type I SGNs contacting the same IHC encode different aspects of auditory information or just transmit copies of the same signal. 

FB1
Box 1:
no caption available
F1
FIGURE 1:
Identification of type I SGN subtypes by single-cell RNA sequencing. (a) Diagram of the inner ear showing on the right a cross-section through the organ of Corti highlighting inner hair cells (IHCs), outer hair cells (OHCs), type I SGNs innervating IHCs and type II SGNs innervating OHCs. SGNs are shown in different colors to highlight that they are a diverse population of neurons. (b) Diagram indicating type I SGNs with low, medium and high spontaneous rates (SRs) that show distinct innervation patterns onto IHCs. (c) Heat maps showing standardized expression of the top 10 differentially expressed genes between type Ia, Ib and Ic SGNs (P < 0.01 from pairwise comparisons, highest average log-fold change). For each subtype, expression is shown for four mice. Expression was averaged across all single cells from an individual mouse for each cluster. Rows are genes, columns are averages for each mouse, grouped by cluster. Data are from [29]. SGN, spiral ganglion neuron.

ANATOMICAL AND PHYSIOLOGICAL HETEROGENEITY OF TYPE I SPIRAL GANGLION NEURONS

Although type I SGNs appear at first glance to be a homogenous cell population with similarly sized cell bodies and bipolar morphology, physiological studies have revealed differences between them [9]. Type I SGNs have variable spontaneous rates (SRs) that are inversely correlated with their threshold to sound and dynamic range [10,11]. Discharge rates ranging from near 0 up to 120 spikes per second have been observed in cats, mice, rats and rabbits and even higher rates in other species [9]. The distribution of SGNs with different SRs is largely bimodal with peaks at low rates (≤ 1 spike/sec) and high rates (60–70 spikes/sec) in several species [10,12,13], but not in rodents where nerve fibers appear to show a gradual increase from low to high SR [14–17]. Based on studies in cats, a classification of SGNs into three subgroups has been proposed: high-SR (>18 spikes/sec), medium-SR (0.5–18 spikes/sec), and low-SR (<0.5 spikes/sec) fibers. In cats, low- and high-SR fibers preferentially contact the modiolar and pillar sides of IHCs, respectively (Fig. 1B) [18]. Studies in rodents support the view that SGNs might similarly be divided into subgroups. Accordingly, single IHCs in rodents are innervated by fibers with different SRs [19] and intensity thresholds [20], and synapses formed by SGNs on the pillar versus modiolar sides of IHCs differ in size and alpha-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid receptor expression levels [21,22]. Differences in voltage dependence and release-site coupling of Ca2+ channels between modiolar and pillar synapses have been reported in rodents [23▪▪], and analysis of excitatory postsynaptic currents suggests that rodent modiolar and pillar synapses have different operating points [24▪▪]. Thus, morphological and physiological differences between SGNs innervating IHCs on the pillar versus modiolar side are likely evolutionarily conserved features.

The significance of these physiological differences between type I SGNs for the processing of auditory signals is not known, although it has been proposed to be important for intensity coding. In addition, it is unclear whether the physiological heterogeneity of type I SGNs is determined by their intrinsic properties or by external factors. Though intrinsic biophysical properties of SGNs are known to correlate to their in vivo activation thresholds [20], differences in presynaptic vesicle release [23▪▪,25] and efferent regulation [26,27,28▪▪] have also been suggested to contribute to SGN physiological diversity.

MOLECULAR HETEROGENEITY AMONG SPIRAL GANGLION NEURONS

To understand the intrinsic heterogeneity among SGNs and how intrinsic differences may contribute to the morphological, physiological and potential functional diversity of SGNs, recent studies have used single-cell ribonucleic acid (RNA) sequencing (scRNAseq) to comprehensively characterize the transcriptome of developing and mature SGNs in the mouse [29–31]. Not surprisingly, type I and type II SGNs exhibit pronounced differences in gene expression programs, consistent with their dramatic morphological, physiological, and functional divergences. More than 1,600 genes are differentially expressed between type I and type II SGNs, including transcription factors, structural proteins and ion channel components. Novel marker genes for type II SGNs, such as Ngfr, the low-affinity cell surface receptor for nerve growth factor, have been uncovered and validated [29].

When delving into type I SGNs, three independent studies have demonstrated that type I SGNs can be divided into three molecularly distinct subtypes, named type Ia, type Ib and type Ic (Fig. 1C) [29–31], which constitute around 40%, 30%, and 30% of type I SGNs, respectively. These subtypes are characterized by the expression of distinct transcription factors, cell adhesion molecules, ion channels, neurotransmitter receptors, etc. [29–31]. Subtype-specific expression was remarkably consistent between studies, especially for highly expressed genes which can function as reliable subtype markers for further analyses. For genes with low expression levels or expression in only a small fraction of cells, determination of possible subtype specificity is particularly difficult, resulting in slight discrepancies of reported markers [29–31]. Tools to reliably identify and isolate SGN type I subtypes for further study will help clarify these results.

Physiologically, types Ia, Ib and Ic have been proposed to correspond to SGNs with high, medium and low SRs respectively, as Ia fibers preferentially contact the pillar side and Ic the modiolar side of IHCs [29–31]. Consistently, type Ic, similar to previously described low SR fibers, appeared to be preferentially affected by aging [30]. Yet further experiments are needed to confirm the physiological differences between subtypes. In addition, SGNs across tonotopic locations seem to exhibit molecular heterogeneity, though to a lesser extent than the distinctions among subtypes [30].

FROM MOLECULAR HETEROGENEITY TO PHYSIOLOGICAL DIVERSITY

Although anatomical evidence suggests that type Ia, Ib, and Ic SGNs may correspond to SGNs with high, medium and low SRs, respectively, direct experimental validation of this hypothesis is still outstanding. However, the analysis of differences in gene expression programs of SGNs provides some clues as to potential differences in their physiological properties.

Voltage-gated ion channels set the excitability of neurons. The genes encoding the voltage-gated sodium channel subunits SCN1B and SCN4B are preferentially expressed in type Ia SGNs, whereas higher expression levels of Scn2b are seen in type Ic SGNs [29,31]. Genes encoding voltage-gated calcium channel components CACNA1B and CACNB4 are detected mainly in type Ia and type Ib SGNs, respectively [29–31]. Voltage-gated potassium channels play an important role in setting the firing rates of neurons [32]. Notably, several voltage-gated potassium channel subunit genes are differentially expressed among subtypes. Kcnc1, Kcnd2, Kcnh2, Kcns1 and Kcns3 are preferentially expressed by type Ib SGNs, whereas higher expression levels of Kcna1, Kcna2, Kcnc3 are detected in type Ic SGNs [29,31]. Interestingly, auxiliary units for voltage-gated potassium channels (Kcnip1, Kcnip2 and Kcnab2) are expressed at higher levels in type Ic SGNs [29,31] (Fig. 2).

F2
FIGURE 2:
Differential expression of ion channels and neurotransmitter receptors in type I SGN subtypes. Dot plots showing differentially expressed genes (log fold-change >0.1 and P < 0.01 from pairwise comparisons) in type I SGN subtypes grouped by functional category; color scale: average expression of all single cells in each cluster; dot size: percentage of single cells with detectable expression (>1 transcript). Data are from [29]. SGN, spiral ganglion neuron.

In addition to voltage-gated ion channels, neurotransmitter receptors may contribute to the physiological and functional heterogeneity of subtypes. IHCs use glutamate as a neurotransmitter [33], and elevated expression levels of genes for several glutamate receptor components, including Grin1, Grik4 and Grm8 have been detected in type Ic neurons [29–31] (Fig. 2). The gene for the metabotropic glutamate receptor mGluR8 (Grm8) is exclusively expressed in type Ic neurons [29–31]; mGluR8 is Gi-coupled and generates inhibitory effects upon glutamate release [34], consistent with Ic having low spontaneous firing rates. Efferent innervation likely also plays a critical role in shaping the physiological properties of SGNs. Genes for the gamma-aminobutyric acid A receptor subunit Gabrb3 and the GABAB receptor subunit Gabbr1 are preferentially expressed in type Ic and type Ib SGNs, respectively [29]. These findings suggest that intrinsic molecular differences between SGN subtypes contribute to physiological heterogeneity, however, genetic perturbation experiments will be necessary to establish causal effect and examine the contribution of intrinsic versus extrinsic factors in generating this diversity.

MATURATION OF SPIRAL GANGLION NEURON SUBTYPES

Neuronal differentiation depends on genetically hardwired and activity-dependent mechanisms [35]. At present, the question of how these mechanisms contribute to the diversification of SGNs into molecular and physiological subtypes remains unanswered. SGNs in mice are born from progenitors in the otic vesicle and delaminate around embryonic day (E) 10 to form the spiral ganglion. Postmitotic SGNs elaborate projections to hair cells already at birth and refine them during the first few postnatal weeks [3,36–40]. Mechanotransduction currents in hair cells are detected at birth and mature over the next 5–10 days [41–43]. Spontaneous activity is present before hearing onset around P12 [44] and synapse formation between SGNs and hair cells progresses up to P28.

Genes with both developmentally stable and dynamic expression patterns were identified with bulk RNA sequencing studies on SGNs across pre and postnatal developmental stages [45▪▪]. Studies using scRNAseq [29–31] and single-cell qPCR [46▪] have demonstrated that molecular differences in SGNs can already be observed shortly after birth, suggesting that some aspects of the diversification of SGNs are genetically hardwired. However, subtype specification as characterized by molecular diversification is only completed postnatally over the period when neurons experience spontaneous and sensory-driven activity (Fig. 3A) [29,30]. The importance of activity-dependent mechanisms for SGN subtype diversification has been investigated with the use of genetic mouse models that disrupt activity patterns. Mutations that abolish hair cell mechanotransduction or glutamatergic communication between hair cells and SGNs disrupt normal spontaneous activity patterns in the prehearing period and input-driven activity at the onset of hearing [29]. These mutations also affect SGN subtype specification [29,30]. However, blocking glutamatergic vesicle release in Vglut3 knock-out mice affects subtypes in a way distinct from mechanotransduction mutants. Type Ic neurons specifically degenerate in Vglut3 knock-outs but not in mechanotransduction mutants (Fig. 3B) [29,30], suggesting that glutamate signaling may play a critical role in the survival of type Ic neurons, consistent with the specific expression of Grm8 and genes for other glutamate receptor subunits in the type Ic SGNs. Related mechanism may also contribute to the vulnerability of type Ic SGNs in pathological conditions, such as aging and ototoxicity [30,47].

F3
FIGURE 3:
Temporal specification of SGN subtypes and specification defects in mutant mice. (a) Diagram showing sequential steps during the maturation of the peripheral auditory sense organ and the maturation of SGN subtypes. (b) Venn diagrams showing the relative proportions of type Ia, Ib, and Ic neurons in adult mice as defined by the expression of molecular markers. In wildtype mice, a small proportion of neurons co-express markers for different subtypes. In mice with mutations in genes affecting mechanotransduction by hair cells or glutamate release from IHCs, the segregation of type I SGNs into subtypes is affected, with increased numbers of neurons co-expressing markers for subtypes. In glutamate release mutants, the pool of type Ic SGNs is also diminished, suggesting increased cell death of this specific neuronal subtype. Data are from [29]. IHCs, inner hair cells; SGN, spiral ganglion neuron.

How genetically hardwired and activity-dependent mechanisms interact to regulate the diversification of SGNs is the subject of much current research. In the following, we discuss the possible contributions of transcription factors, guidance molecules, and spontaneous activity.

TRANSCRIPTION FACTORS AND SPIRAL GANGLION NEURON SUBTYPES

Transcription factors with subtype-specific expression could contribute to the developmental emergence of molecular subtype identities and projection patterns. The gene for the pituritary-specific Pit-1 and octamer factor proteins Oct-1 and Oct-2 (POU) domain transcription factor Pou4f1 is expressed in adult mice in the type Ic population [29–31], whereas Pou4f2 is detected at high levels in type Ia SGNs, and at medium and low levels in type Ib and Ic SGNs, respectively (Fig. 4) [29,31]. Notably, Pou4f1 is broadly expressed in all SGNs at birth, and its expression becomes refined to the type Ic SGN population postnatally by mechanisms that are disrupted when hair cell mechanotransduction or glutamate release from hair cells is affected [29,30]. This suggests that the refinement of Pou4f1 gene expression to SGN subtypes is activity-dependent and that POU4F1 alone does not specify differences between type I SGNs, although it could act in concert with other transcription factors that exhibit SGN subtype-specific expression. Germline deletion of the Pou4f1 gene leads to reduction in neuronal size, pathfinding defects and loss of about 30% of SGNs [48,49], which is equivalent to the proportion of type Ic SGNs [29–31]. Yet, perinatal lethality of straight knock-outs prevents the detailed examination of its role in subtype maturation. Conditional deletion of Pou4f1 in SGNs at E14.5 reveals a grossly normal cochlea with no significant neuronal loss or innervation defects [50]. Consistently, auditory thresholds as measured by ABR are not affected in the mutant mice. Presynaptic voltage sensitivity and active zone properties, however, are altered in the conditional mutants, suggesting that Pou4f1 plays an instructive role in presynaptic Ca2+ signaling in IHCs.

F4
FIGURE 4:
Differential expression of transcription factors and axonal guidance molecules in type I SGN subtypes. Dot plots showing differentially expressed genes (log fold-change >0.1 and P < 0.01 from pairwise comparisons) in type I SGN subtypes grouped by functional category; color scale: average expression of all single cells in each cluster; dot size: percentage of single cells with detectable expression (>1 transcript). Data are from [29]. SGN, spiral ganglion neuron.

Several other transcription factors show differences in expression levels between SGN subtypes (Fig. 4). Expression of the gene for runt-related transcription factor 1 (RUNX1), is mainly detected in the type Ib and Ic SGN populations [29–31]. Runx1 is required for the differentiation and maturation of nonpeptidergic nociceptors in the dorsal root ganglion [51,52]. In the inner ear, Runx1 inactivation leads to the loss of a subset of vestibulocochlear ganglion neurons [53]. The early lethality of the Runx1 mutant, however, has prevented detailed analysis of its function in the postnatal maturation of SGNs. The gene for RXRG, a retinoic acid receptor, is selectively expressed by mature type Ia SGNs [29,30]. Retinoic acid is important for the development of tonotopy [54]. It would be interesting to see if these transcription factors participate in the differentiation and maturation of type I subtypes. Furthermore, Maf family, Myocyte enhancer factor-2 (Mef2) family, Basic helix-loop-helix and Zas family transcription factors also demonstrate subtype-specific expression patterns. Maf and Mafb are expressed at elevated levels in type Ia and type Ic SGNs, whereas Mef2a and Mef2c are preferentially expressed in type Ia and type Ic SGNs, respectively [29,31]. Mafb in particular is required for the normal differentiation of the postsynaptic density at the ribbon synapse between IHCs and type I SGNs [55]. In addition, Bhlhe40 and Bhlhe22 are expressed at higher levels in type Ib and type Ic SGNs, whereas Zas family transcription factors Hivep1 and Hivep2 are selectively expressed in type Ib and type Ic SGNs [29,30]. Further studies are necessary to determine the developmentally regulated expression patterns of these transcription factors and the extent to which they contribute to SGN subtype specification.

AXONAL GUIDANCE MOLECULES AND TARGET INNERVATION

Axonal guidance molecules ensure proper wiring of neural circuits. A range of guidance molecules are differentially expressed in SGN type I subtypes (Fig. 4), which could contribute to their distinct innervation patterns underneath the IHCs as well as in the cochlear nucleus. Members of the semaphorin family are known to refine wiring in the cochlea. Sema3F restricts type I projections to the IHC region [56], whereas Sema5B/Plexin A1 signaling regulates terminal branch numbers of SGNs [57]. Sema7a, Sema3e/3d and Sema3f expression are elevated in type Ia, Ib and Ic, respectively [29,31], which warrants further investigations of their roles in the distinct innervation patterns of SGN subtypes. Eph/ Ephrin signaling is also critical in the development of cochlea [58]. EphA7 regulates neurite outgrowth from SGNs [59]. EphA4 and Ephrin A5 control type I afferent targeting [60]. EphA4 and Ephrin B2 contribute to the patterning of cochlea [61]. It would be interesting to see if EphA4 and Ephrin B2, for which genes are preferentially expressed by type Ia and Ib SNGs [29,31], also participate in the formation of subtype-specific innervations. Similarly, Nrxn1, Nrxn2 and Nrxn3 are enriched in type Ic, Ib and Ia SGNs, respectively [29], but their role in the regulation of hair cell innervation by type I SGNs still needs to be explored.

THE CONTRIBUTION OF SPONTANEOUS ACTIVITY

Recent studies have begun to address the role of spontaneous activity in the diversification of SGNs. Spontaneous activity is initiated in the prehearing phase with adenosine 5’-triphosphate (ATP) release from inner supporting cells (ISCs), resulting in the depolarization of IHCs, the release of glutamate from IHCs and subsequent depolarization of SGNs [62]. ATP release induces crenations in supporting cells surrounding hair cells [63]. Crenations could provide a mechanical stimulus to hair cells, leading to the activation of mechanotransduction channels in the prehearing period, and amplifying the direct depolarization of SGNs. This could provide a plausible explanation for the distinct effects of mechanotransduction mutants and glutamatergic transmission mutants on spontaneous activity. Consistent with the role of spontaneous activity in subtype maturation, mechanotransduction mutants [29] and glutamatergic transmission mutants [29,30] affect subtype specification. Mutations that disrupt mechanotransduction by murine cochlear hair cells reduce overall spontaneous activity in SGNs, whereas a mutation that affects glutamate release from IHC onto SGNs affects mini-burst firing without reducing overall spontaneous activity [29]. These findings suggest that the specific patterns of spontaneous activity may be important for SGN subtype specification. To further test the role of spontaneous activity in SGN subtype specification, it will be interesting to analyze mice with mutations that target receptors and ion channels linked to the generation of spontaneous activity. ATP release from ISCs and activation of purinergic autoreceptor P2RY1 is critical to the coordinated excitation of neurons. Blocking P2RY1 receptors dramatically reduced spontaneous activity [64], but how this affects SGN subtype specification is unclear. Another purinergic receptor, P2RX3 is required for spontaneous Ca2+ dependent refinement of type II afferents in early postnatal development [65]. P2RX3 is transiently expressed by type I SGNs and hair cells before postnatal day 6, and P2rx3-null mice show transient phenotypes affecting SGN type I branching. Interestingly, a transient increase in the percentage of type Ia SGNs relative to other type I SGN subtypes is also observed in P2rx3 mutant mice [66], indicating a potential involvement of purinergic signaling in the process of SGN subtype maturation.

Increase of intracellular calcium levels during spontaneous activity could drive the induction of immediate early genes, many of which are transcription factors that can regulate subsequent waves of late response genes to mediate long-term changes of neurons, such as dendritic growth, spine maturation and synapse elimination [67]. However, the detailed mechanisms by which activity consolidates the segregation of SGN cell types remain open to investigation. Critical period defines a time window during which neural circuits are influenced by activity. It warrants further investigation whether there is a ‘critical period’ for the activity-dependent specification of SGN subtypes.

CONCLUSION

The identification of molecular subclasses of SGNs has opened new opportunities to study the development and function of auditory afferent neurons. Using molecular tools such as conditional mutant mice targeting specific subtypes of SGNs, it will now be possible to define the relationship between molecularly and physiologically defined subtypes, to probe the mechanisms by which genetically hardwired and activity-dependent mechanisms contribute to SGN diversification, and to elucidate how different SGN subtypes contribute to the encoding of sound.

Acknowledgements

This work was supported by the NIH (U.M., RO1DC005965; C.S. and U.M., RO1DC016960; S.S K99DC018574) and the David M. Rubenstein Fund for Hearing Research. U.M. is a Bloomberg Distinguished Professor. U.M. is a co-founder of Decibel Therapeutics.

Financial support and sponsorship

None.

Conflicts of interest

There are no conflicts of interest.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

▪ of special interest

▪▪ of outstanding interest

REFERENCES

1. Gillespie PG, Müller U. Mechanotransduction by hair cells: models, molecules, and mechanisms. Cell 2009; 139:33–44.
2. Meyer AC, Moser T. Structure and function of cochlear afferent innervation. Curr Opin Otolaryngol Head Neck Surg 2010; doi:10.1097/MOO.0b013e32833e0586.
3. Perkins RE, Morest DK. A study of cochlear innervation patterns in cats and rats with the Golgi method and Nomarski optics. J Comp Neurol 1975; doi:10.1002/cne.901630202.
4. Ryugo DK. The auditory nerve: peripheral innervation, cell body morphology, and central projections. 1992; New York, NY: Springer, 23–65.
5. Spoendlin H. Anatomy of cochlear innervation. Am J Otolaryngol Neck Med Surg 1985; 6:453–467.
6. Flores EN, Duggan A, Madathany T, et al. A noncanonical pathway from cochlea to brain signals tissue-damaging noise. Curr Biol 2015; 25:606–612.
7. Liu C, Glowatzki E, Fuchs PA. Unmyelinated type II afferent neurons report cochlear damage. Proc Natl Acad Sci USA 2015; 112:14723–14727.
8. Eybalin M. Neurotransmitters and neuromodulators of the mammalian cochlea. Physiol Rev 1993; 73:309–374.
9. Heil P, Peterson AJ. Basic response properties of auditory nerve fibers: a review. Cell Tissue Res 2015; doi:10.1007/s00441-015-2177-9.
10. Liberman MC. Auditory-nerve response from cats raised in a low-noise chamber. J Acoust Soc Am 1978; 63:442–455.
11. Sachs MB, Abbas PJ. Rate versus level functions for auditory-nerve fibers in cats: Tone-burst stimuli. J Acoust Soc Am 1974; 56:1835–1847.
12. Borg E, Engström B, Linde G, Marklund K. Eighth nerve fiber firing features in normal-hearing rabbits. Hear Res 1988; 36:191–201.
13. Furman AC, Kujawa SG, Charles Liberman M. Noise-induced cochlear neuropathy is selective for fibers with low spontaneous rates. J Neurophysiol 2013; 110:577–586.
14. Barbary A. El. Auditory nerve of the normal and jaundiced rat. I. Spontaneous discharge rate and cochlear nerve histology. Hear Res 1991; 54:75–90.
15. Schmiedt RA. Spontaneous rates, thresholds and tuning of auditory-nerve fibers in the gerbil: comparisons to cat data. Hear Res 1989; 42:23–35.
16. Taberner AM, Liberman MC. Response properties of single auditory nerve fibers in the mouse. J Neurophysiol 2005; 93:557–569.
17. Wu JS, Young ED, Glowatzki E. Maturation of spontaneous firing properties after hearing onset in rat auditory nerve fibers: Spontaneous rates, refractoriness, and interfiber correlations. J Neurosci 2016; 36:10584–10597.
18. Liberman MC. Single-neuron labeling in the cat auditory nerve. Science 1982; 216:1239–1241.
19. Wu Z, Muller U. Molecular identity of the mechanotransduction channel in hair cells: not quiet there yet. J Neurosci 2016; 36:10927–10934.
20. Markowitz AL, Kalluri R. Gradients in the biophysical properties of neonatal auditory neurons align with synaptic contact position and the intensity coding map of inner hair cells. Elife 2020; 9:1–33.
21. Liberman LD, Wang H, Liberman MC. Opposing gradients of ribbon size and AMPA receptor expression underlie sensitivity differences among cochlear-nerve/hair-cell synapses. J Neurosci 2011; 31:801–808.
22. Liberman MC, Epstein MJ, Cleveland SS, et al. Toward a differential diagnosis of hidden hearing loss in humans. PLoS One 2016; 11:e0162726.
23▪▪. Özçete ÖD, Moser T. A sensory cell diversifies its output by varying Ca 2+ influx-release coupling among active zones. EMBO J 2021; 40:e106010.
24▪▪. Niwa M, Young ED, Glowatzki E, Ricci AJ. Functional subgroups of cochlear inner hair cell ribbon synapses differently modulate their EPSC properties in response to stimulation. J Neurophysiol 2021; doi:10.1152/jn.00452.2020.
25. Frank T, Khimich D, Neef A, Moser T. Mechanisms contributing to synaptic Ca2+ signals and their heterogeneity in hair cells. Proc Natl Acad Sci USA 2009; 106:4483–4488.
26. Liberman MC. Morphological differences among radial afferent fibers in the cat cochlea: An electron-microscopic study of serial sections. Hear Res 1980; 3:45–63.
27. Guinan JJ. Olivocochlear efferents: their action, effects, measurement and uses, and the impact of the new conception of cochlear mechanical responses. Hear Res 2018; 362:38–47.
28▪▪. Hua Y, Ding X, Wang H, et al. Electron microscopic reconstruction of neural circuitry in the cochlea. Cell Rep 2021; 34:1–15.
29. Sun S, Babola T, Pregernig G, et al. Hair cell mechanotransduction regulates spontaneous activity and spiral ganglion subtype specification in the auditory system. Cell 2018; 174:1247–1263. e15.
30. Shrestha BR, Chia C, Wu L, et al. Sensory neuron diversity in the inner ear is shaped by activity. Cell 2018; 174:1229–1246. e17.
31. Petitpré C, Wu H, Sharma A, et al. Neuronal heterogeneity and stereotyped connectivity in the auditory afferent system. Nat Commun 2018; 9:3691.
32. Johnston J, Forsythe ID, Kopp-Scheinpflug C. Going native: voltage-gated potassium channels controlling neuronal excitability. J Physiol. 2010:588; 3187–3200.
33. Seal RP, Akil O, Yi E, et al. Sensorineural deafness and seizures in mice lacking vesicular glutamate transporter 3. Neuron 2008; 57:263–275.
34. Niswender CM, Conn PJ. Metabotropic glutamate receptors: Physiology, pharmacology, and disease. Annu Rev Pharmacol Toxicol 2010; 50:295–322.
35. Wamsley B, Fishell G. Genetic and activity-dependent mechanisms underlying interneuron diversity. Nat Rev Neurosci 2017; 18:299–309.
36. Echteler SM. Developmental segregation in the afferent projections to mammalian auditory hair cells. Proc Natl Acad Sci USA 1992; 89:6324–6327.
37. Huang LC, Thorne PR, Housley GD, Montgomery JM. Spatiotemporal definition of neurite outgrowth, refinement and retraction in the developing mouse cochlea. Development 2007; 134:2925–2933.
38. Huang LC, Barclay M, Lee K, et al. Synaptic profiles during neurite extension, refinement and retraction in the developing cochlea. Neural Dev 2012; doi:10.1186/1749-8104-7-38.
39. Koundakjian EJ, Appler JL, Goodrich LV. Auditory neurons make stereotyped wiring decisions before maturation of their targets. J Neurosci 2007; doi:10.1523/JNEUROSCI.3765-07.2007.
40. Sobkowicz HM, Rose JE, Scott GL, Levenick CV. Distribution of synaptic ribbons in the developing organ of Corti. J Neurocytol 1986; doi:10.1007/BF01625188.
41. Kim KX, Fettiplace R. Developmental changes in the cochlear hair cell mechanotransducer channel and their regulation by transmembrane channel-like proteins. J Gen Physiol 2013; 141:141–148.
42. Lelli A, Asai Y, Forge A, et al. Tonotopic gradient in the developmental acquisition of sensory transduction in outer hair cells of the mouse cochlea. J Neurophysiol 2009; doi:10.1152/jn.00136.2009.
43. Pan B, Géléoc GS, Asai Y, et al. TMC1 and TMC2 are components of the mechanotransduction channel in hair cells of the mammalian inner ear. Neuron 2013; doi:10.1016/j.neuron.2013.06.019.
44. Mikaelian D, Ruben RJ. Development of hearing in the normal cba-j mouse: correlation of physiological observations with behavioral responses and with cochlear anatomy. Acta Otolaryngol 1965; 59:451–461.
45▪▪. Li C, Li X, Bi Z, et al. Comprehensive transcriptome analysis of cochlear spiral ganglion neurons at multiple ages. Elife 2020; 9:1–25.
46▪. Grandi FC, De Tomasi L, Mustapha M. Single-Cell RNA analysis of type I spiral ganglion neurons reveals a lmx1a population in the cochlea. Front Mol Neurosci 2020; 13:83.
47. Kang KW, Pangeni R, Park J, et al. Selective loss of calretinin-poor cochlear afferent nerve fibers in streptozotocin-induced hyperglycemic mice. J Nanosci Nanotechnol 2020; 20:5515–5519.
48. McEvilly RJ, Erkman L, Luo L, et al. Requirement for Brn-3.0 in differentiation and survival of sensory and motor neurons. Nature 1996; 384:574–577.
49. Huang EJ, Liu W, Fritzsch B, et al. Brn3a is a transcriptional regulator of soma size, target field innervation and axon pathfinding of inner ear sensory neurons. Development 2001; 128:2421–2432.
50. Sherrill HE, Jean P, Driver EC, et al. Pou4f1 defines a subgroup of type i spiral ganglion neurons and is necessary for normal inner hair cell presynaptic Ca2+ signaling. J Neurosci 2019; 39:5284–5298.
51. Chen CL, Broom DC, Liu Y, et al. Runx1 determines nociceptive sensory neuron phenotype and is required for thermal and neuropathic pain. Neuron 2006; 49:365–377.
52. Huang S, O’Donovan KJ, Turner EE, et al. Extrinsic and intrinsic signals converge on the Runx1/CBFβ transcription factor for nonpeptidergic nociceptor maturation. Elife 2015; 4:
53. Theriault FM, Roy P, Stifani S. AML1/Runx1 is important for the development of hindbrain cholinergic branchiovisceral motor neurons and selected cranial sensory neurons. Proc Natl Acad Sci USA 2004; 101:10343–10348.
54. Thiede BR, Mann ZF, Chang W, et al. Retinoic acid signalling regulates the development of tonotopically patterned hair cells in the chicken cochlea. Nat Commun 2014; 5:1–13.
55. Yu WM, Appler JM, Kim YH, et al. A Gata3-Mafb transcriptional network directs postsynaptic differentiation in synapses specialized for hearing. Elife 2013; 2013:
56. Coate TM, Spita NA, Zhang KD, et al. Neuropilin-2/semaphorin-3F-mediated repulsion promotes inner hair cell innervation by spiral ganglion neurons. Elife 2015; 4:1–24.
57. Jung JS, Zhang KD, Wang Z, et al. Semaphorin-5B controls spiral ganglion neuron branch refinement during development. J Neurosci 2019; 39:6425–6438.
58. Defourny J. Eph/ephrin signalling in the development and function of the mammalian cochlea. Dev Biol 2019; 449:35–40.
59. Kim YJ, Ibrahim LA, Wang SZ, et al. EphA7 regulates spiral ganglion innervation of cochlear hair cells. Dev Neurobiol 2016; 76:452–469.
60. Defourny J, Peuckert C, Kullander K, Malgrange B. EphA4-ADAM10 interplay patterns the cochlear sensory epithelium through local disruption of adherens junctions. iScience 2019; 11:246–257.
61. Defourny J, Poirrier AL, Lallemend F, et al. Ephrin-A5/EphA4 signalling controls specific afferent targeting to cochlear hair cells. Nat Commun 2013; 4:1–12.
62. Wang HC, Lin CC, Cheung R, et al. Spontaneous activity of cochlear hair cells triggered by fluid secretion mechanism in adjacent support cells. Cell 2015; 163:1348–1359.
63. Tritsch NX, Yi E, Gale JE, et al. The origin of spontaneous activity in the developing auditory system. Nature 2007; 450:50–55.
64. Babola TA, Li S, Wang Z, et al. Purinergic signaling controls spontaneous activity in the auditory system throughout early development. J Neurosci 2021; 41:594–612.
65. Ceriani F, Hendry A, Jeng J, et al. Coordinated calcium signalling in cochlear sensory and nonsensory cells refines afferent innervation of outer hair cells. EMBO J 2019; 38:
66. Wang Z, Jung JS, Inbar TC, et al. The purinergic receptor p2rx3 is required for spiral ganglion neuron branch refinement during development. eNeuro 2020; 7:1–21.
67. Yap EL, Greenberg ME. Activity-regulated transcription: bridging the gap between neural activity and behavior. Neuron 2018; 100:330–348.
Keywords:

deafness; hearing; neuronal differentiation; single-cell RNA sequencing; spiral ganglion neurons

Copyright © 2021 The Author(s). Published by Wolters Kluwer Health, Inc.