Just as our pupils dilate or shrink depending on the amount of light available to our eyes, our ears adjust their sensitivity based on the sound environment we encounter. Evidence suggests that a group of cells known as olivocochlear neurons (OCNs for short) may be involved in this process. These cells are located in the brainstem but project into the cochlea, the inner ear structure that converts sound waves into the electrical impulses relayed to the brain. OCNs may mediate how sounds are detected and encoded "at the source."

Historically, OCNs have been divided into two groups (medial or lateral OCNs) based on different morphologies and roles in hearing. For instance, medial OCNs are thought to protect our ears against loud sounds by sending molecular signals to the inner ear cells that amplify certain auditory signals. However, it remains difficult to disentangle the precise function of the different types of OCNs, in part because scientists still lack markers that would allow them to distinguish between medial and lateral cells simply based on genetic activity.

Frank et al. aimed to eliminate this bottleneck by identifying which genes were switched on and to what degree in individual mouse medial and lateral OCNs; this was done throughout development and after exposure to loud noises. The experiments uncovered a range of genetic markers for medial and lateral OCNs, showing that these cells switch on different sets of genes relevant to their role over development. This gene expression data also revealed that two distinct groups of lateral OCNs exist, one of which is characterised by the production of large amounts of neuropeptides, a type of chemical messenger that can modulate neural circuit activity.

Further work in both developing and adult mice showed that this production is shaped by the activity of the cells, with the neuropeptide levels increasing when the animals are exposed to damaging levels of noise. This change lasts for several days, suggesting that such an experience can have long-lasting effects on how the brain provides feedback to the ear.

Overall, the results by Frank et al. will help to better identify and characterize the different types of OCNs and the role that they have in hearing. By uncovering the chemical messengers that mediate the response to loud noises, this research may contribute to a better understanding of how to prevent or reduce hearing loss.

All sensory systems are modulated by feedback circuits that dynamically tune sensory information in response to both external experience and internal state. In the sense of vision, for instance, central efferent pathways mediate the reflexive restriction of the pupil in response to bright light and also modulate pupil diameter with changes in arousal state (McGinley et al., 2015). Analogously, a small group of several hundred olivocochlear neurons (OCNs) in the superior olivary complex (SOC) extend projections all the way to the peripheral hearing organ, the cochlea, enabling direct regulation of initial detection and encoding of auditory stimuli (Rasmussen, 1946; Rasmussen, 1953; Figure 1A and B). By integrating information from both ascending auditory pathways and descending central circuits, OCNs serve as a conduit for an animal’s experiences and needs to modulate sound information at the earliest stages of sensory processing.

Figure 1 with 2 supplements see all

Download asset Open asset

The impact of OCNs on cochlear function is poorly understood, as few markers for these cells have been identified and the population is difficult to target either genetically or physiologically. OCNs express a variety of neurotransmitters (Eybalin, 1993; Kitcher et al., 2022; Sewell, 2011) and can even change their neurotransmitter profiles in response to intense sounds (Niu and Canlon, 2002; Wu et al., 2020), making it difficult to identify and catalog cell types. Developmentally and evolutionarily, OCNs are closely related to the facial branchial motor neurons (FMNs; Frank and Goodrich, 2018; Fritzsch and Elliott, 2017). Both populations derive from common motor neuron progenitors and share expression of motor neuron markers such as Islet-1 (Frank and Goodrich, 2018; Fritzsch and Elliott, 2017), although OCNs are not considered motor neurons. Expression of the zinc-finger transcription factor GATA3 seems to be crucial for specifying an OCN fate, but little else is known about how OCNs and FMNs diverge, or how OCN subtypes differentiate from one another and acquire their specialized attributes (Pata et al., 1999; Karis et al., 2001). Thus, identifying markers that can distinguish OCN subtypes from one another and from neighboring FMNs is critical for understanding the anatomical and molecular logic of OCN connectivity.

OCNs fall into two major groups: medial olivocochlear neurons (MOCs), which sit in the ventral nucleus of the trapezoid body (VNTB), and lateral olivocochlear neurons (LOCs), which are intermingled with afferent principal neurons in the lateral superior olive (LSO; Figure 1A). Both MOCs and LOCs extend peripheral axons that spiral along the length of the cochlea, often spanning large frequency domains (Brown, 2011; Figure 1B). MOCs terminate mainly on outer hair cells (OHCs), a group of mechanosensory cells that influence cochlear responses to auditory stimuli through rapid, sound-induced cell-body movements (van der Heijden and Vavakou, 2022; Ashmore, 2019). It is widely accepted that MOCs affect cochlear gain through inhibitory, cholinergic synapses that reduce OHC motility (Mountain, 1980; Siegel and Kim, 1982; Blanchet et al., 1996; Dallos et al., 1997; Fuchs and Lauer, 2019), much as visual efferent pathways control gain by altering pupil diameter. As well as protecting cochlear circuitry from the impact of acoustic trauma (Reiter and Liberman, 1995; Taranda et al., 2009; Maison et al., 2013; Boero et al., 2018; Fuente, 2015; Rajan, 1995; Tong et al., 2013), MOCs are thought to contribute to the detection of signals in noise (Guinan, 2018; Kawase et al., 1993; Winslow and Sachs, 1987) and to tune sensory responses based on states of attention (Delano et al., 2007; Terreros et al., 2016; Oatman, 1976; Oatman, 1971). Consistent with this range of effects on the perception of sound, electrophysiological characterization suggests that MOCs are controlled both by inputs from the cochlear nucleus and by descending inputs from multiple sources (Romero and Trussell, 2022).

In contrast, we know little about the LOCs, as they are difficult to access for recording, stimulation, or surgical ablation. Anatomically, LOCs terminate on the peripheral processes of Type I spiral ganglion neurons (SGNs), primary sensory neurons that carry perceptual auditory information from mechanosensory inner hair cells (IHCs) into the brain. Functionally, the nature of communication between LOCs and SGNs remains opaque: although LOCs express acetylcholine, a variety of other signaling molecules have also been reported in these cells, including GABA, CGRP, dopamine, enkephalin, dynorphin, and urocortin (Ucn) (Eybalin, 1993; Kitcher et al., 2022; Sewell, 2011; Safieddine and Eybalin, 1992; Safieddine et al., 1997; Kaiser et al., 2011; Takeda et al., 1987; Adams et al., 1987). However, definitive evidence for many of these transmitters is lacking. Moreover, indirect activation of the auditory efferent system can elicit both excitatory and inhibitory effects on SGNs, hinting at the possibility of multiple LOC subtypes that can direct distinct effects on their downstream targets (Groff and Liberman, 2003). No specific role in hearing has been definitively linked to LOCs, although they have been implicated in sound localization (Darrow et al., 2006a; Larsen and Liberman, 2010; Irving et al., 2011) and appear to protect the ear from noise damage (Fuente, 2015; Darrow et al., 2007; Kujawa and Liberman, 1997). The mechanisms underlying this protective effect remain unclear. One promising model holds that dopamine—and potentially other neurotransmitters—released from LOCs inhibits SGN firing, thereby dampening excitotoxicity (Wu et al., 2020; Ruel et al., 2001). However, this model has yet to be tested directly, and other, as yet unidentified, pathways may contribute a protective role as well.

Here, we took a multi-pronged approach to investigate key molecular, anatomical, and physiological features of the auditory efferent system. We found many novel transcriptional features of MOCs and LOCs in both developing and mature OCNs, including cell-type specific markers, genes that could confer distinct physiological properties, and developmental gene-expression changes that highlight pathways involved in OCN maturation. In addition, we found that LOCs cluster into two molecularly and anatomically distinct subtypes based on expression of neuropeptides. However, regardless of the amount of neuropeptide Y in their pre-synaptic puncta, LOCs elaborate axon arborizations that vary extensively and terminate on multiple SGN subtypes. Further, LOCs change their neuropeptide expression profiles upon the onset of hearing and after acoustic trauma, indicating that variability in peptide expression may serve as a way to modulate sensory circuits based on prior experience.

To assess transcriptional variability within OCNs, we sequenced cell nuclei from individual cholinergic neurons, which were labeled using Chat^Cre to drive expression of a nuclear-localized GFP (Rosa26^Sun1-GFP) (Rossi et al., 2011; Mo et al., 2015). We enriched for OCNs and the closely related FMNs by dissecting a region of the ventral brainstem that includes the SOC as well as facial and trigeminal motor neurons. We then used fluorescence-activated cell-sorting (FACS) to isolate dissociated GFP+ cholinergic nuclei and employed the 10x Genomics platform to encapsulate nuclei and generate barcoded single-nucleus libraries (Figure 1C). To examine the maturation of these cell types and identify markers that are expressed consistently across postnatal development, nuclei were collected at two pre-hearing timepoints, postnatal day (P)1 (n=13 animals) and P5 (n=16 animals), the latter of which is an important time of synapse refinement in auditory circuits (Frank and Goodrich, 2018; Yu and Goodrich, 2014). We also collected nuclei at P26–P28 (n=32 animals), when auditory circuitry is grossly mature. After filtering out low-quality cells and infrequently expressed genes (see Materials and methods), our dataset includes 45,828 nuclei: 16,753 cells from P1 animals; 15,542 from P5 animals; and 13,533 cells from P26–P28 animals. Unsupervised, graph-based clustering analysis identified 56 clusters (Figure 1D). These clusters primarily consist of neurons, as indicated by the expression of neuronal markers like Snap25 (Figure 1E). Populations of oligodendrocyte precursors, Mog-positive oligodendrocytes (Figure 1F), and astrocytes were also identified based on expression of well-established markers (Zhang, 2001). Each neuronal cluster includes cells from all three timepoints (Figure 1—figure supplement 1A). Non-neuronal cells primarily originate from the adult dataset, consistent with prior reports showing an increase in myelination of auditory brainstem neurons and elevated expression of oligodendrocyte markers over the first few postnatal weeks (Long et al., 2018). Although cell types vary somewhat in the number of genes and unique molecular identifiers (UMI) detected, the structure of the data is not driven by these technical variables (Figure 1—figure supplement 1B–D). Cell types were also similar in the fraction of genes mapping to the mitochondrial genome, suggesting that clusters are not affected by differences in cell health (Ilicic et al., 2016; Luecken and Theis, 2019; Figure 1—figure supplement 1E). The neuronal clusters include several motor neuron clusters, as defined by co-expression of Isl1 (Figure 1G), Tbx20, and Phox2b. Among these, two OCN clusters were identified based on expression of Gata3, which is expressed in OCNs but not in canonical motor neuron populations (Figure 1H; Karis et al., 2001). LOC and MOC clusters were identified by expression of the LOC-specific peptide Urocortin (Ucn) (Kaiser et al., 2011; Figure 1I) and the Na,K-ATPase Atp1a3, which is expressed in MOCs but not LOCs in mature rats (McLean et al., 2009; Figure 1J).

Gene-expression profiles that distinguish LOCs and MOCs from FMNs and each other

LOCs and MOCs arise from a common progenitor pool with and develop alongside FMNs, raising the question of how OCNs deviate from a typical motor neuron fate to play such an unorthodox role in peripheral sensory modulation (Frank and Goodrich, 2018; Fritzsch and Elliott, 2017). To learn more about the transcriptional origins of their unique developmental and mature properties, we compared all OCNs to a cluster of FMNs, which was identified based on co-expression of Etv1 and Epha7 (Figure 1—figure supplement 2A, B; Tenney et al., 2019). Our analysis revealed substantial transcriptional differences between OCNs and FMNs that could contribute to their distinct functions and anatomy (Figure 1—figure supplement 2C). For instance, OCNs show selective expression of Cadps2, which encodes the calcium-dependent activator protein for secretion 2 (CADPS2, also known as CAPS-2; Figure 1K). CADPS2 is a member of the Munc13 and CAPS family of priming proteins (Imig et al., 2014; Jockusch et al., 2007; Nestvogel et al., 2020) and has been linked to exocytosis of dense-core vesicles (Tandon et al., 1998; Speidel et al., 2003; Grishanin et al., 2004; Renden et al., 2001), consistent with neuropeptide release from OCNs (Eybalin, 1993; Kitcher et al., 2022). Fluorescent in situ hybridization (FISH) in P27–P28 mice confirmed that Cadps2 is expressed in both MOCs and LOCs but is not detectable in FMNs (Figure 1L and M; n=3 animals). This finding validates our OCN and FMN clusters and establishes a new OCN-specific marker.

Despite their early shared origins, FMNs and OCNs express unique combinations of genes encoding transcription factors, adhesion molecules, and other receptors needed for proper neuronal migration, targeting, and synaptic specificity. These differences are particularly prominent at P1 and P5, while OCN axons are still growing to their final targets in the inner ear and before many developmentally salient molecules become downregulated (Figure 1—figure supplement 2D, E). Cell-type-specific transcription factor genes include Pbx3 in FMNs and Sall3 in OCNs (Figure 1—figure supplement 2G). Genes for many guidance and adhesion molecules are also differentially expressed, including Kirrel3 in OCNs and Pcdh15 in FMNs (Figure 1—figure supplement 2H). Consistent with their integration into fundamentally different circuitry, OCNs and FMNs also differ in the expression of molecules relevant to mature function (Figure 1—figure supplement 2F, I). In particular, OCNs express Gad2, a gene involved in GABA synthesis, whereas FMNs express the serotonin receptor gene Htr2c (Figure 1—figure supplement 2I).

Although OCNs share several properties that broadly distinguish them from FMNs, OCNs are themselves heterogeneous, with LOCs and MOCs playing distinct roles in auditory circuitry (Guinan, 2018). To date, however, few reliable markers for either population have been identified. For instance, although Atp1a3 is enriched in MOCs in adulthood, it is expressed in both LOCs and MOCs developmentally (Figure 2—figure supplement 1B, C). To address this gap, we identified several genes that are enriched in either the MOC or LOC cluster across all three timepoints and verified their expression in mature animals by FISH (Figure 2A–G, Figure 2—figure supplement 1A–G). We found that Col4a4 is expressed in virtually all LOCs in mature (P26–P28) animals, whereas Zfp804a is expressed in virtually all MOCs (Figure 2B–G). These findings further validate the identity of MOC and LOC clusters and present new markers to distinguish MOC and LOC neurons across development.

Figure 2 with 1 supplement see all

Download asset Open asset

In mature animals, LOCs and MOCs are integrated into distinct circuits, receiving input from multiple brain regions and responding to numerous neurotransmitters and modulatory substances (Brown, 2011; Romero and Trussell, 2022). Consistently, our data point to many subtype-specific transcription factors and adhesion molecules that could direct their unique developmental trajectories, including migration to different nuclei in the SOC and guidance of axons to different target cells in the cochlea (Figure 2H, I). In addition, genes for many physiologically relevant receptors and ion channels are differentially expressed (Figure 2J), including receptors and receptor subunits for neurotransmitters such as acetylcholine, glutamate, and orexin (Figure 2K–R). Differential expression analysis between neonatal and mature timepoints identified numerous genes, including several that encode neurotransmitter receptors and ion channels (Figure 2S, Figure 2—figure supplement 1H, I), supporting a recent report that mature physiological properties of LOCs emerge gradually (Hong et al., 2022). We also identified altered expression of transcription factor genes, like Auts2, that may orchestrate these developmental changes (Figure 2S).

A novel, anatomically segregated subtype of LOCs is defined by neuropeptide expression

One challenge in understanding LOC function has been the presence of additional heterogeneity within this population (Brown, 2011; Romero and Trussell, 2022). Therefore, we sub-clustered mature OCNs, focusing on differences among LOCs. This analysis revealed five OCN clusters (Figure 3A). Quality-control metrics—including the number of genes detected and originating batch—were similar among all clusters (Figure 3—figure supplement 1A–D). Based on expression of MOC and LOC markers, we identified a large cluster of MOCs (230 cells) (Figure 3B) and two main clusters of LOCs (242 LOC1 cells, 75 LOC2 cells; Figure 3C). This analysis also revealed two smaller clusters with only 21 cells each. One of these clusters (MOC2) expressed MOC markers, including Zfp804a; the other appears to consist of miscellaneous cells (Misc), as these cells did not express either MOC or LOC markers. We identified only a small number of differentially expressed genes in the MOC2 and Misc clusters (Figure 3—figure supplement 1E, F), possibly due to low cell counts. We were therefore unable to definitively identify either cluster and did not pursue them further.

Figure 3 with 1 supplement see all

Download asset Open asset

Comparison of the two LOC subtypes revealed that LOC2s are distinguished by upregulation of genes for the neuropeptides CGRP (Calca), CGRP-II (Calcb), NPY, and Ucn (Figure 3D and J-M). LOC1s and LOC2s also differ in expression of the genes for the transcription factor Meis1 and for guidance and adhesion molecules, such as Tenm2, Tenm3, Unc5c, and Epha6 (Figure 3E–I), suggesting that these clusters indeed represent distinct cell types rather than reflecting the same pool of cells in different states. Consistent with this idea, LOC2s are confined to the medial wing of the LSO (n=4 animals of either sex, P28–P32), as revealed by immunostaining for CGRP and Ucn in tissue from an Npy-GFP reporter mouse (Figure 3O). Peptide expression is heterogeneous: although all adult LOCs express some level of CGRP, as previously reported (Brown, 2011; Wu et al., 2018; Maison et al., 2003a), LOC2s did not uniformly express both NPY and Ucn (Figure 3N; Supplementary file 1). NPY seems to be a particularly reliable marker for LOC2s, as Ucn was not always expressed in NPY+ cells and very few cells expressed only Ucn. Furthermore, an intersectional genetic approach in which Npy^FlpO; Chat^Cre drives expression of the dual recombinase Rosa26^FLTG reporter recapitulates the pattern of NPY+ LOC restriction to the medial wing of the LSO (Figure 3P), without labeling MOCs in the VNTB (arrowheads, Figure 3P) or the lateral wing of the LSO (n=4 animals of either sex, P28–P30). Thus, we propose that LOC2s comprise a distinct cell type that is poised to release multiple neuropeptides in the cochlea.

Previous studies have indicated that LOCs produce an array of signaling molecules, including acetylcholine, GABA, CGRP, and opioids (Eybalin, 1993; Safieddine and Eybalin, 1992). Having confirmed expression of Chat, Gad2, and Calca across OCNs (Figure 1—figure supplement 2F, Figure 3—figure supplement 1G), we sought to verify whether transcripts for any opioid precursors were detected in either LOC subtype. However, we were unable to observe expression of either Penk or Pdyn in any LOC subtype in either our sequencing data or via FISH (Figure 3—figure supplement 1G, H). The lack of transcripts implies that neither LOC subtype produces opioid peptides, although we cannot rule out the presence of a minority population or that expression is transient.

Anatomically diverse OCNs broadly innervate the cochlea

Anatomical studies suggest that the frequency distribution of LOC axons is organized in a stereotyped manner, with medial neurons projecting to higher-frequency regions of the cochlea and lateral neurons projecting to lower-frequency regions (Brown, 2011; Robertson et al., 1987; Guinan et al., 1984). These projections are consistent with the tonotopic organization of LSO principal neurons, which are arranged from low-to-high frequencies along a lateral-to-medial axis (Kandler et al., 2009). The confinement of LOC2s to the medial LSO therefore raises the possibility of tonotopically restricted effects in the cochlea. However, although the density of presumptive LOC2 axons, labeled by Npy^FlpO; Chat^Cre; Rosa26^FLTG, decreases from high-frequency (basal) to low-frequency (apical) regions, NPY+ LOC axons span the entirety of the base-to-apex length (Figure 3Q, left to right).

Because OCN axons can arborize extensively along the length of the cochlea, the broad distribution of LOC2 processes observed in Npy^FlpO; Chat^Cre; Rosa26^FLTG cochleae might obscure meaningful differences in synaptic connectivity. Indeed, a wide variety of LOC morphologies has been described (Brown, 2011; Warr et al., 1997; Brown, 1987), raising the possibility that LOC2s correspond to one anatomically distinct group. To analyze the morphology and synaptic connectivity of individual OCN axons, we leveraged our discovery that Ret is expressed by all OCNs from early in development (Figure 5—figure supplement 1A, B) and used low doses of tamoxifen in Ret^CreER mice to drive expression of the Cre-dependent Igs7^GFP reporter in a small number of OCNs (Figure 5—figure supplement 1C, D; Luo et al., 2009). Labeled OCNs colocalize with ChAT immunolabeling in the brainstem (Figure 5—figure supplement 1C, D), and a subset of LOCs are also positive for NPY (Figure 5—figure supplement 1C’, closed arrowhead), confirming that this approach captures both LOC1s and LOC2s. Sparsely labeled OCN axons display peripheral innervation patterns consistent with what has been described (Brown, 2011; Warr et al., 1997; Brown, 1987; Figure 5—figure supplement 1E–E”). To better define the connectivity of individual Ret^CreER; Igs7^GFP-labeled MOC and LOC axons, cochleae were also stained for the pre-synaptic marker synaptophysin (Syp) to mark putative OCN synapses and calbindin-2 (CALB2, also known as calretinin) to label IHCs and Type I SGN subtypes (Shrestha et al., 2018; Sun et al., 2018; Petitpré et al., 2018; Figure 4A). Putative pre-synaptic sites in each axon were isolated using the fluorescence signal from labeled OCN axons to mask the Syp signal (Figure 4A’, Figure 4—video 1, Figure 5—video 1).

Figure 4 with 1 supplement see all

Download asset Open asset

Analysis of MOC innervation patterns confirmed that this sparse labeling approach recapitulates known morphologies and synaptic distributions. Sparsely labeled MOC axons (n=40 axons, N=9 animals of both sexes) have terminal morphologies consistent with prior dye-labeling studies (Brown, 2011; Brown, 1987; Robertson and Gummer, 1985; Brown, 2014), including relatively simple axons with little-to-no branching beneath the OHCs (Figure 4A” and B); branching in the OSL (Figure 4C); complex branching beneath IHCs and among OHCs (Figure 4D) or branching in a goal-post fashion (Figure 4E). Labeled MOCs show a slight preference for innervating the third row of OHCs, with 21.4% of Syp puncta in row 1, 31.6% in row 2, and 47% in row 3. Additionally, 45% of the reconstructed MOC axons (18/40) have Syp puncta in the inner spiral bundle (ISB), localized among SGN peripheral fibers, either en passant (Figure 4A”) or, less frequently, at terminal endings (Figure 4F). This observation confirms previous reports that a subset of MOC terminal axons innervate SGN processes, although the number of putative pre-synaptic sites within a given MOC axon are somewhat lower than what was identified in EM reconstructions (Robertson and Gummer, 1985; Hua et al., 2021). This suggests that our approach is reliable and, if anything, may undercount pre-synaptic sites.

Because the tamoxifen induction approach labeled multiple OCNs in a given animal, we could not reliably associate individual terminal axons to a particular labeled soma in the brainstem. Instead, our analyses consider each terminal axon independently. However, in one animal, a single MOC soma was labeled in the brainstem. We were therefore able to confidently associate multiple terminal axons in the corresponding cochlea to the same MOC cell body (Figure 5—figure supplement 1E–E”). Six terminal fibers stemming from this single MOC innervated discrete patches along a~27 kHz swath of the cochlea (Figure 4A’’, Figure 5—figure supplement 1F–I). Within this one MOC, some (shown in Figure 4A–A”), but not all (Figure 5—figure supplement 1F–I), of its terminal fibers contained Syp puncta in the ISB. This suggests that there is not a discrete morphological class of MOCs that makes branches that consistently contact both SGN peripheral fibers and OHCs. Likewise, when considering the entire dataset, there were no significant differences in the frequency of innervation (Figure 4G) or number of branch points (Figure 4H) between terminal MOC axons that did and did not contain Syp puncta in the ISB. Collectively, these observations indicate that individual MOCs can produce projections with variable morphologies and synaptic targets.

NPY-positive LOC axons exhibit highly variable patterns of connectivity

We next asked whether LOC2s, as identified by NPY expression, exhibit any stereotyped axon morphologies or patterns of connectivity with different SGN subtypes in the cochlea. There are three physiologically and molecularly distinct Type I SGN subtypes whose terminals are arranged along the base of IHCs in a stereotyped manner (Figure 5A; Shrestha et al., 2018; Sun et al., 2018; Petitpré et al., 2018; Liberman, 1978; Liberman et al., 2011; Liberman, 1982; Petitpré et al., 2020). This diversity is thought to enable the encoding of a wide range of sound intensities (Petitpré et al., 2020). Type Ia SGNs express high levels of CALB2 and form synapses on the pillar side of IHCs, nearer the Tunnel of Corti and OHCs (Figure 5A, dark blue). On the other hand, Type Ic SGNs express low levels of CALB2 and synapse on the modiolar side of IHCs, nearer the SGN cell bodies (Figure 5A, light blue). Type Ib SGNs are defined by moderate levels of CALB2, likely filling in the space between the Ia and Ic subtypes along the base of the IHC. In Npy^FlpO; Chat^Cre; Rosa26^FLTG cochlea, genetically labeled LOC2 axons intermingle closely with parvalbumin (PV)+ SGN peripheral processes on both sides of the inner hair cell, although there was a qualitative bias to the modiolar side (Figure 5B), similar to what has been reported for LOCs in general (Hua et al., 2021; Liberman et al., 1990). This distribution confirms that Type I SGN processes are the primary targets of LOCs and argues against selective modulation of specific Type I SGN subtypes by LOC2s.

Figure 5 with 2 supplements see all

Download asset Open asset

To quantify the relationship between LOC subtypes and connectivity, we reconstructed individual terminal LOC axons using the same sparse labeling approach that we validated with MOC axon reconstructions (Figure 4). Cochlear wholemounts from Ret^CreER; Igs7^GFP animals were immunolabeled for GFP to visualize overall OCN axon morphology, NPY to assess peptidergic identity, Syp to identify putative pre-synaptic sites, and CALB2 to label IHCs and identify Type I SGN subtypes (Figure 5C–H, Figure 5—video 1). As with the MOC analysis, we were unable to assign projections to specific LOC neurons, and therefore considered each projection independently. Most of the GFP-labeled LOC axons were located in the middle or base of the cochlea, with axons entering the ISB at frequencies ranging from 18 to 69 kHz. This analysis therefore covers much of the ethologically relevant frequency range in mice (Turner et al., 2005). NPY distribution was punctate and heterogeneous not just between axons, but also within individual axons, raising the possibility that individual LOC fibers may vary in the signaling molecules released at specific synaptic sites. To assess each LOC axon’s potential to release NPY, we used the surface of the axon to mask the Syp and NPY channels (Figure 5F–H, Figure 5—video 1) and created reconstructions of Syp puncta labeled as either NPY+ or NPY- (open and closed arrowheads, respectively, Figure 5G and H). We also identified whether each Syp puncta was in contact with CALB2 immunofluorescence signal (Figure 5C–E) to ascertain if either LOC subtype forms selective relationships with Type I SGN subtypes.

Consistent with the known distribution of Type I SGN subtype peripheral processes along the base of the IHC and the corresponding gradient of CALB2 expression (Figure 5A), putative synaptic puncta on CALB2+ targets are predominantly positioned on the pillar side of IHCs, and those that do not contact CALB2+ targets are positioned on the modiolar side (Figure 5E, Figure 5—video 1). Because CALB2 also labels IHCs, it is possible that some Syp puncta contacting CALB2 signal could be LOC contacts onto IHCs. Efferent contacts on mature IHCs have been reported in cat and rat cochleae, although at least in rodents this phenomenon appears to be associated with pathology (Liberman et al., 1990; Liberman, 1980; Lauer et al., 2012; Zachary and Fuchs, 2015). However, since most of the LOC Syp puncta in our dataset are positioned below the IHCs, most On-CALB2 Syp puncta are contacting Ia or Ib SGN peripheral processes. Conversely, we can conclude that most Off-CALB2 Syp puncta contact Ic SGN peripheral processes, although a few branches extend away from the SGN peripheral processes and may terminate near supporting cells. While intriguing, interactions with non-SGN targets were too rare for further quantification or analysis.

Our analysis revealed a high degree of variability in LOC axon morphology and connectivity, with no evidence for stereotyped classes. LOC axons turn towards the base, apex, or are bifurcated (indicated by colored arrows in Figure 5I–O), and can extend either short, simple branches with few Syp puncta (Figure 5J and M–O) or long, complex tangles of branches that form dense forests of Syp puncta (Figure 5I, K, and L). Syp puncta were not evenly distributed—some LOC axons can extend upwards of 1 kHz with no Syp puncta (Figure 5K). Prior studies of LOC subtypes in rats and guinea pigs suggested bifurcating axons cover longer stretches of the cochlea than axons turning in a single direction (Warr et al., 1997; Brown, 1987). However, we did not identify any relationship between LOC turn direction and any of the morphological attributes we examined, including the percent of NPY+ Syp puncta, cumulative axon length, tonotopic span, stem frequency, total number of Syp puncta, or the selectivity for CALB2+ SGNs (Figure 5P, Figure 5—figure supplement 1J–N). Turn direction therefore appears to be a poor predictor for morphological attributes of LOC axons.

Furthermore, we found that no morphological features of the axon—such as its length or branchiness—correlated with its NPY status. Far from falling into distinct categories, the majority of reconstructed LOC axons had both NPY+ and NPY- Syp puncta that were present both on and off CALB2+ targets (e.g. Figure 5I, K, and L). Indeed, individual LOC axons can weave from the pillar to modiolar side and extend branches that make pre-synaptic contacts on SGN peripheral processes on both sides of the IHC (Figure 5E, Figure 5—video 1). In some cases, the Syp puncta along an individual branch preferentially innervate one type of target (e.g. all puncta on CALB2+ targets on the right branch, Figure 5L). However, On- and Off-CALB2 Syp puncta were also observed intermingled along the length of an axon (Figure 5I–K; Figure 5L, left branch). We did identify some axons that had only On-CALB2+ or Off-Calb2+ Syp puncta, but they also showed no relationship with turn direction or the number of NPY+ Syp puncta (Figure 5M–O). These terminal axons, although relatively short in length, also varied in the percent of Syp puncta that were NPY+. This was also true of the LOC axon population as a whole, which showed no strong correlation between the percent of NPY+ Syp puncta in an individual LOC and any metric we examined (Figure 5P–U, Figure 5—figure supplement 1O–Q), including the frequency where the LOC axon enters the ISB (Figure 5Q), its tonotopic span (Figure 5R), number of branch points (Figure 5S), total number of Syp puncta (Figure 5T), selectivity for CALB2+ targets (Figure 5U), branch depth (Figure 5—figure supplement 1O), fiber density (Figure 5—figure supplement 1P), or density of Syp puncta (Figure 5—figure supplement 1Q). Although some of these measures are weakly correlated with the fraction of NPY+ Syp puncta, NPY identity does not appear to be predictive of LOC morphology or innervation patterns, illustrated most vividly by examining axons with either 0% NPY or 100% NPY+ Syp puncta. On the contrary, even individual LOC axons are poised to have heterogeneous and widespread effects in the cochlea.

Physiological heterogeneity does not correlate with variability in peptide expression

Given the lack of obvious differences in their innervation patterns, we next asked whether LOC subtypes differ physiologically. Like NPY and Ucn, CGRP (encoded by the Calca gene) is expressed as a gradient in LOCs, with high CGRP levels in the medial LSO correlated with expression of NPY and Ucn (Figure 3O). We therefore used GFP fluorescence intensity in Calca-GFP mice as a proxy for peptidergic identity as a whole. This strategy allowed us to distinguish Calca+ LOC neurons from neighboring, Calca- LSO principal neurons while also evaluating variability across LOC subtypes. Whole-cell patch-clamp recordings were performed from GFP+ LOC neurons in 200 µm-thick brain slices from 20 P16–P19 Calca-GFP mice (Gong et al., 2003). At these ages, the intensity of native GFP fluorescence varies in LOC neurons (Figure 6A), consistent with differences in Calca RNA and CGRP protein levels (Figure 3J and O). Although peptide-enriched LOC2s are restricted to the medial wing, cells with variable levels of GFP are also present here (Figure 6A’), consistent with the presence of both NPY+ and NPY- LOCs in this region (Figure 5—figure supplement 1C). Therefore, cells in the central region of the medial LSO were used for all analyses to reduce the confounding effects of differences that might be related to tonotopic position. Voltage-clamp and current-clamp recordings were assessed for a panel of measures correlating with cell activity and compared across fluorescence intensities of GFP+ cells.

Figure 6

Download asset Open asset

Although the population as a whole was heterogeneous, we observed no systematic differences in the physiological properties of LOCs expressing different levels of GFP (Figure 6, Supplementary file 3). The input resistance, which was measured during a step from a holding voltage of –66 mV to –76 mV, did not correlate with GFP intensity (Figure 6I, Supplementary file 3), nor did membrane capacitance measured in voltage-clamp recordings (Figure 6J, Supplementary file 3). Input resistances were consistent with published work in rat LOCs and putative LOCs in mice (Sterenborg et al., 2010; Fujino et al., 1997). Voltage-gated currents were assessed by stepping the membrane from a holding potential of –60 mV to voltages between –96 and +34 mV, in 10 mV increments (Figure 6B–D). GFP intensity also did not correlate with the magnitude of the outward potassium current measured 485ms following the voltage step to –46 mV or with the steady-state inward current evoked by a step to –96 mV (Supplementary file 3). There was heterogeneity in the patterns of potassium currents at steady-state, with some cells exhibiting a slowly increasing outward current during the largest step to +34 mV (Figure 6B), some cells having a slowly inactivating outward current (Figure 6C), and others exhibiting a fast-onset, non-inactivating outward current yielding a flat waveform (Figure 6D). However, all of these groups had similar average GFP intensities (slow increase outward current: 62.71±19.08, n=17; sustained: 67.47±18.36, n=11; decrease: 72.22±25.57, n=14; one-way ANOVA DF = 41, F=0.77, p=0.47; mean ± SD).

In some experiments, a 100ms pre-pulse to –96 mV was applied prior to voltage steps that revealed fast-inactivating, likely A-type potassium currents in all of the neurons tested (representative trace, Figure 6D’’), consistent with enrichment of the voltage-gated potassium channel gene Kcnd2 in OCNs (Figure 1—figure supplement 2F). Following a voltage step from –96 to 24 mV, there was no correlation between A-type potassium current amplitude from the peak to plateau and GFP intensity (Supplementary file 3). The decay of the fast-inactivating current was fit to a double exponential, with the fast component likely representing the A-type potassium current. The fast component of the time constant of decay of the fast-inactivating (likely A-type) potassium current did not correlate with GFP intensity (Supplementary file 3). Consistent with previous reports, currents indicative of HCN channels were not observed in any LOC neurons (Sterenborg et al., 2010). Thus, although LOCs exhibited differences that are consistent with previous characterizations, none of the properties assessed by voltage-clamp correlated with peptidergic identity.

Current-clamp experiments also revealed no correlations between firing properties and GFP intensity (Figure 6E–H and K–L). The majority of neurons exhibited action potentials at rest (13 of 23 neurons), some with oscillating patterns as recently reported in LOC neurons (Hong et al., 2022). Action potential rates did not correlate with GFP intensity (Supplementary file 3). When the membrane was held at ~–63 mV to suppress spontaneous action potentials, there was no correlation between GFP intensity and rheobase, the action potential threshold of the first evoked spike, the amplitude from baseline of the first spike evoked at rheobase, or the number of action potentials evoked by a 10 pA current injection (Supplementary file 3). The number of spikes evoked at different depolarizing injection steps between 10 and 60 pA was plotted for each cell to generate input-output curves that indicate the increase in cell activity with increased depolarization. The majority (16 of 19) of cells exhibited multiple spikes in response to the depolarizing steps. There was no relationship between GFP intensity and the slope of the input-output curves generated from each cell (Figure 6L, Supplementary file 3). A voltage sag in response to hyperpolarizing current steps which indicates I_h currents was not observed in any cells. In summary, although many measures of neuron activity varied among LOC neurons, no measures associated with neuron function correlate with the intensity of GFP and hence LOC subtype identity.

LOC properties emerge during postnatal development and are modulated by hearing

Since peptide expression seems to be the defining feature of different LOC subtypes, we asked when these differences arise and how they are influenced by hearing. In examining developmental changes in our sequencing data, we found that Calca, Calcb, Ucn, and NPY all increase in LOCs during the first four postnatal weeks (Figure 7A–D, Supplementary file 4). Consistently, analysis of Npy-GFP animals (n=3–5 per timepoint) revealed that Npy-GFP+ LOC2s are present and localized to the medial wing by at least P5 (Figure 7E), with a qualitative increase in the number of NPY+ LOC2s between P5 and P28 (Figure 7E, 7—figure supplement 1C–F). Although there is some variation, cells with the highest Npy-GFP levels were always situated in the medial wing of the LSO. In contrast, expression of Calca is spatially and temporally dynamic in LOCs across postnatal development, as shown by FISH (2–3 animals per timepoint per probe) (Figure 7F). At P5/6, Calca levels are low and biased medially. Expression increased dramatically with the onset of hearing, with high levels of transcript detected even in the lateral LSO at P12. By P28, high Calca levels are again restricted medially (Figure 7F). Previous work identified a similar, transient increase in the expression of Ucn and Calca in gerbils and hamsters, respectively, indicating that this phenomenon is conserved across species (Kaiser et al., 2011; Simmons et al., 1997). Similarly, work in the Xenopus lateral line found that peripheral responses to exogenous CGRP increase during development, suggesting that the developmental onset of CGRP expression in efferents may be matched by a gradual increase in expression of CGRP receptors in downstream targets (Bailey and Sewell, 2000a). Thus, the LOC2 population seems to be established early in development, but the overall pattern of neuropeptide gene expression is not mature until after the onset of hearing. This finding raises the possibility that peptide status depends on auditory experience.

Figure 7 with 1 supplement see all

Download asset Open asset

To assess the impact of hearing on peptide expression in LOCs, we examined expression of Ucn, NPY, tyrosine hydroxylate (TH), and CGRP in the brainstems of constitutively deaf Tmc1^-/-; Tmc2^-/- double-knockout mice (Tmc1/2^dKO) (Kawashima et al., 2011) alongside hearing littermate controls (Tmc1^+/-; Tmc2^-/-), both before (P7–P8) and after (P27–P28) the onset of hearing. CGRP expression was comparable between deaf and hearing animals at P7–P8 (Figure 7G, I, and K; n=11 LSOs from 6 control animals, 10 LSOs from 5 Tmc1/2^dKO animals), assessed both by quantifying the fraction of CGRP+ LOCs and the slope of variation in neurotransmitter levels from the medial to lateral LSO (Figure 7—figure supplement 1A). The fraction of NPY-expressing LOCs was also similar between Tmc1/2^dKO animals and controls at P7–P8 (Figure 7—figure supplement 1C, D; n=11 LSOs from 6 control animals, 10 LSOs from 5 Tmc1/2^dKO animals). LOCs with the highest level of NPY expression were found in the medial LSO at this timepoint, as well (Figure 7G, Figure 7—figure supplement 1C), although the slope of NPY expression was quantitatively shallower in Tmc1/2^dKO animals than controls (Figure 7—figure supplement 1D).

After the onset of hearing, however, the gradient of CGRP expression was significantly perturbed in deaf mice, with high CGRP levels reaching much farther into the lateral LSO (Figure 7H, J, and L; n=12 LSOs from 6 control animals, 10 LSOs from 5 Tmc1/2^dKO animals). Ucn expression was also shifted, reaching farther into the lateral LSO in deaf animals compared to hearing controls (Figure 7H and M, Figure 7—figure supplement 1I; n=7 control LSOs from 4 animals, 9 Tmc1/2^dKO LSOs from 5 animals). The Ucn antibody was not reliable at P7, so we were unable to evaluate earlier timepoints. In contrast, although NPY levels were qualitatively lower in deaf animals, there were no significant differences in the fraction or slope of NPY expression in adult animals (Figure 7—figure supplement 1E, F; n=12 LSOs from 6 control animals, 10 LSOs from 5 Tmc1/2^dKO animals). Differential effects on NPY, Ucn, and CGRP fit with our original observation that peptide expression is heterogeneous in mature LOCs (Figure 3N and O, Supplementary file 1). TH expression is also reduced in deaf animals compared to controls, consistent with the observation that TH levels can be elevated with sound exposure (Wu et al., 2020; Figure 7—figure supplement 1B, G, H; n=7 LSOs from 4 control animals, 9 LSOs from 5 Tmc1/2^dKO animals).

These results suggest that peptide expression in LOCs is malleable during postnatal development. To examine whether this flexibility is preserved into adulthood, we exposed 7-week-old C57BL/6J mice to 110 dB 8–16 kHz octave-band noise for 2 hours, a stimulus previously shown to induce TH expression in mouse LOCs (Wu et al., 2020) and induce permanent threshold shifts (Liberman and Kujawa, 2017). As expected, the fraction of ChAT+ LOCs expressing TH was elevated 9 days after sound exposure (Figure 8A–C; n=4 exposed and 4 unexposed animals of either sex). In addition, we observed an even larger increase in the fraction of LOCs expressing Ucn or NPY (Figure 8A and D–G). Because all LOCs already express CGRP, there was no increase in the fraction of CGRP+ LOCs, although the levels of CGRP increased qualitatively throughout the LSO (Figure 8A, H, I). Across all neuropeptides, the slope of peptide expression remained unaltered after sound exposure, with higher levels largely constrained to the medial wing (Figure 8). In contrast, TH levels increased throughout the LSO, suggesting that the mechanisms driving LOCs to express TH may be different from those that govern neuropeptide expression (Figure 8A–C). Thus, hearing shapes the signaling repertoire of a subset of LOC neurons, both developmentally and in adults.

Figure 8

Download asset Open asset

Some variant of the olivocochlear efferent system exists for all hair cell sensory systems: groups of feedback cells project to auditory, vestibular, and lateral line neurons in all classes of vertebrates, and even some invertebrates (Roberts and Meredith, 1992). The wide conservation of this circuitry across evolutionary time suggests that OCNs play a crucial functional role. Here, we examined the transcriptional, physiological, and anatomical properties of mammalian OCNs to clarify longstanding questions about their identities and attributes. In doing so, we identified a new, peptide-enriched LOC subtype and found that peptide expression is modulated by hearing, both developmentally and upon acoustic overexposure in adults. Induced peptide expression persists for over a week after noise exposure (Figure 8), presenting a mechanism by which LOCs can modulate downstream targets on the timescale of days. Collectively, our findings suggest that a diverse population of OCNs influences the initial detection of sound in a wide-ranging manner that occurs across timescales and changes in response to auditory experience.

Single-cell data collected in this study is available on GEO, accession number GSE214027.

1. 1. Frank MM
   2. Goodrich LV

   (2022) NCBI Gene Expression Omnibus

   ID GSE214027. Single-nucleus sequencing data from developing and adult mouse olivocochlear neurons.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE214027

1. 1. Adams JC
   2. Mroz EA
   3. Sewell WF

   (1987) A possible neurotransmitter role for CGRP in a hair-cell sensory organ

   Brain Research 419:347–351.

   https://doi.org/10.1016/0006-8993(87)90606-8

   • PubMed
   • Google Scholar
2. 1. Ashmore J

   (2019) Outer hair cells and electromotility

   Cold Spring Harbor Perspectives in Medicine 9:a033522.

   https://doi.org/10.1101/cshperspect.a033522

   • PubMed
   • Google Scholar
3. 1. Assas BM
   2. Pennock JI
   3. Miyan JA

   (2014) Calcitonin gene-related peptide is a key neurotransmitter in the neuro-immune axis

   Frontiers in Neuroscience 8:23.

   https://doi.org/10.3389/fnins.2014.00023

   • PubMed
   • Google Scholar
4. 1. Bailey GP
   2. Sewell WF

   (2000a) Pharmacological characterization of the CGRP receptor in the lateral line organ of Xenopus laevis

   Journal of the Association for Research in Otolaryngology 1:82–88.

   https://doi.org/10.1007/s101620010007

   • PubMed
   • Google Scholar
5. 1. Bailey GP
   2. Sewell WF

   (2000b) Calcitonin gene-related peptide suppresses hair cell responses to mechanical stimulation in the Xenopus lateral line organ

   The Journal of Neuroscience 20:5163–5169.

   https://doi.org/10.1523/JNEUROSCI.20-13-05163.2000

   • PubMed
   • Google Scholar
6. 1. Basappa J
   2. Turcan S
   3. Vetter DE

   (2010) Corticotropin-Releasing factor-2 activation prevents gentamicin-induced oxidative stress in cells derived from the inner ear

   Journal of Neuroscience Research 88:2976–2990.

   https://doi.org/10.1002/jnr.22449

   • PubMed
   • Google Scholar
7. 1. Blanchet C
   2. Eróstegui C
   3. Sugasawa M
   4. Dulon D

   (1996) Acetylcholine-Induced potassium current of guinea pig outer hair cells: its dependence on a calcium influx through nicotinic-like receptors

   The Journal of Neuroscience 16:2574–2584.

   https://doi.org/10.1523/JNEUROSCI.16-08-02574.1996

   • PubMed
   • Google Scholar
8. 1. Boero LE
   2. Castagna VC
   3. Di Guilmi MN
   4. Goutman JD
   5. Elgoyhen AB
   6. Gómez-Casati ME

   (2018) Enhancement of the medial olivocochlear system prevents hidden hearing loss

   The Journal of Neuroscience 38:7440–7451.

   https://doi.org/10.1523/JNEUROSCI.0363-18.2018

   • PubMed
   • Google Scholar
9. 1. Brown MC

   (1987) Morphology of labeled efferent fibers in the guinea pig cochlea

   The Journal of Comparative Neurology 260:605–618.

   https://doi.org/10.1002/cne.902600412

   • PubMed
   • Google Scholar
10. Book

    1. Brown MC

    (2011) Anatomy of olivocochlear neurons

    In: Ryugo DK, Fay RR, editors. Auditory and Vestibular Efferents. Springer. pp. 17–37.

    https://doi.org/10.1007/978-1-4419-7070-1

    • Google Scholar
11. 1. Brown MC

    (2014) Single-unit labeling of medial olivocochlear neurons: the cochlear frequency map for efferent axons

    Journal of Neurophysiology 111:2177–2186.

    https://doi.org/10.1152/jn.00045.2014

    • PubMed
    • Google Scholar
12. 1. Bruce LL
    2. Kingsley J
    3. Nichols DH
    4. Fritzsch B

    (1997) The development of vestibulocochlear efferents and cochlear afferents in mice

    International Journal of Developmental Neuroscience 15:671–692.

    https://doi.org/10.1016/s0736-5748(96)00120-7

    • PubMed
    • Google Scholar
13. 1. Chen W
    2. Liu Y
    3. Liu W
    4. Zhou Y
    5. He H
    6. Lin S

    (2020) Neuropeptide Y is an immunomodulatory factor: direct and indirect

    Frontiers in Immunology 11:1–14.

    https://doi.org/10.3389/fimmu.2020.580378

    • Google Scholar
14. 1. Cheng S
    2. Butrus S
    3. Tan L
    4. Xu R
    5. Sagireddy S
    6. Trachtenberg JT
    7. Shekhar K
    8. Zipursky SL

    (2022) Vision-dependent specification of cell types and function in the developing cortex

    Cell 185:311–327.

    https://doi.org/10.1016/j.cell.2021.12.022

    • PubMed
    • Google Scholar
15. 1. Choi HMT
    2. Schwarzkopf M
    3. Fornace ME
    4. Acharya A
    5. Artavanis G
    6. Stegmaier J
    7. Cunha A
    8. Pierce NA

    (2018) Third-generation in situ hybridization chain reaction: multiplexed, quantitative, sensitive, versatile, robust

    Development 145:dev165753.

    https://doi.org/10.1242/dev.165753

    • PubMed
    • Google Scholar
16. 1. Daigle TL
    2. Madisen L
    3. Hage TA
    4. Valley MT
    5. Knoblich U
    6. Larsen RS
    7. Takeno MM
    8. Huang L
    9. Gu H
    10. Larsen R
    11. Mills M
    12. Bosma-Moody A
    13. Siverts LA
    14. Walker M
    15. Graybuck LT
    16. Yao Z
    17. Fong O
    18. Nguyen TN
    19. Garren E
    20. Lenz GH
    21. Chavarha M
    22. Pendergraft J
    23. Harrington J
    24. Hirokawa KE
    25. Harris JA
    26. Nicovich PR
    27. McGraw MJ
    28. Ollerenshaw DR
    29. Smith KA
    30. Baker CA
    31. Ting JT
    32. Sunkin SM
    33. Lecoq J
    34. Lin MZ
    35. Boyden ES
    36. Murphy GJ
    37. da Costa NM
    38. Waters J
    39. Li L
    40. Tasic B
    41. Zeng H

    (2018) A suite of transgenic driver and reporter mouse lines with enhanced brain-cell-type targeting and functionality

    Cell 174:465–480.

    https://doi.org/10.1016/j.cell.2018.06.035

    • Google Scholar
17. 1. Dallos P
    2. He DZ
    3. Lin X
    4. Sziklai I
    5. Mehta S
    6. Evans BN

    (1997) Acetylcholine, outer hair cell electromotility, and the cochlear amplifier

    The Journal of Neuroscience 17:2212–2226.

    https://doi.org/10.1523/JNEUROSCI.17-06-02212.1997

    • PubMed
    • Google Scholar
18. 1. Darrow KN
    2. Maison SF
    3. Liberman MC

    (2006a) Cochlear efferent feedback balances interaural sensitivity

    Nature Neuroscience 9:1474–1476.

    https://doi.org/10.1038/nn1807

    • PubMed
    • Google Scholar
19. 1. Darrow KN
    2. Simons EJ
    3. Dodds L
    4. Liberman MC

    (2006b) Dopaminergic innervation of the mouse inner ear: evidence for a separate cytochemical group of cochlear efferent fibers

    The Journal of Comparative Neurology 498:403–414.

    https://doi.org/10.1002/cne.21050

    • PubMed
    • Google Scholar
20. 1. Darrow KN
    2. Maison SF
    3. Liberman MC

    (2007) Selective removal of lateral olivocochlear efferents increases vulnerability to acute acoustic injury

    Journal of Neurophysiology 97:1775–1785.

    https://doi.org/10.1152/jn.00955.2006

    • PubMed
    • Google Scholar
21. 1. Delano PH
    2. Elgueda D
    3. Hamame CM
    4. Robles L

    (2007) Selective attention to visual stimuli reduces cochlear sensitivity in chinchillas

    The Journal of Neuroscience 27:4146–4153.

    https://doi.org/10.1523/JNEUROSCI.3702-06.2007

    • PubMed
    • Google Scholar
22. 1. Eybalin M

    (1993) Neurotransmitters and neuromodulators of the mammalian cochlea

    Physiological Reviews 73:309–373.

    https://doi.org/10.1152/physrev.1993.73.2.309

    • PubMed
    • Google Scholar
23. 1. Frank MM
    2. Goodrich LV

    (2018) Talking back: development of the olivocochlear efferent system

    Wiley Interdisciplinary Reviews. Developmental Biology 7:1–20.

    https://doi.org/10.1002/wdev.324

    • PubMed
    • Google Scholar
24. 1. Fritzsch B
    2. Elliott KL

    (2017) Evolution and development of the inner ear efferent system: transforming a motor neuron population to connect to the most unusual motor protein via ancient nicotinic receptors

    Frontiers in Cellular Neuroscience 11:114.

    https://doi.org/10.3389/fncel.2017.00114

    • PubMed
    • Google Scholar
25. 1. Fuchs PA
    2. Lauer AM

    (2019) Efferent inhibition of the cochlea

    Cold Spring Harbor Perspectives in Medicine 9:a033530.

    https://doi.org/10.1101/cshperspect.a033530

    • PubMed
    • Google Scholar
26. 1. Fuente A

    (2015) The olivocochlear system and protection from acoustic trauma: a mini literature review

    Frontiers in Systems Neuroscience 9:94.

    https://doi.org/10.3389/fnsys.2015.00094

    • PubMed
    • Google Scholar
27. 1. Fujino K
    2. Koyano K
    3. Ohmori H

    (1997) Lateral and medial olivocochlear neurons have distinct electrophysiological properties in the rat brain slice

    Journal of Neurophysiology 77:2788–2804.

    https://doi.org/10.1152/jn.1997.77.5.2788

    • PubMed
    • Google Scholar
28. 1. Gong S
    2. Zheng C
    3. Doughty ML
    4. Losos K
    5. Didkovsky N
    6. Schambra UB
    7. Nowak NJ
    8. Joyner A
    9. Leblanc G
    10. Hatten ME
    11. Heintz N

    (2003) A gene expression atlas of the central nervous system based on bacterial artificial chromosomes

    Nature 425:917–925.

    https://doi.org/10.1038/nature02033

    • PubMed
    • Google Scholar
29. 1. Graham CE
    2. Basappa J
    3. Turcan S
    4. Vetter DE

    (2011) The cochlear CRF signaling systems and their mechanisms of action in modulating cochlear sensitivity and protection against trauma

    Molecular Neurobiology 44:383–406.

    https://doi.org/10.1007/s12035-011-8203-3

    • PubMed
    • Google Scholar
30. 1. Graham CE
    2. Vetter DE

    (2011) The mouse cochlea expresses a local hypothalamic-pituitary-adrenal equivalent signaling system and requires corticotropin-releasing factor receptor 1 to establish normal hair cell innervation and cochlear sensitivity

    The Journal of Neuroscience 31:1267–1278.

    https://doi.org/10.1523/JNEUROSCI.4545-10.2011

    • PubMed
    • Google Scholar
31. 1. Grishanin RN
    2. Kowalchyk JA
    3. Klenchin VA
    4. Ann K
    5. Earles CA
    6. Chapman ER
    7. Gerona RRL
    8. Martin TFJ

    (2004) Caps acts at a prefusion step in dense-core vesicle exocytosis as a PIP2 binding protein

    Neuron 43:551–562.

    https://doi.org/10.1016/j.neuron.2004.07.028

    • PubMed
    • Google Scholar
32. 1. Groff JA
    2. Liberman MC

    (2003) Modulation of cochlear afferent response by the lateral olivocochlear system: activation via electrical stimulation of the inferior colliculus

    Journal of Neurophysiology 90:3178–3200.

    https://doi.org/10.1152/jn.00537.2003

    • PubMed
    • Google Scholar
33. 1. Guinan JJ
    2. Warr WB
    3. Norris BE

    (1984) Topographic organization of the olivocochlear projections from the lateral and medial zones of the superior olivary complex

    The Journal of Comparative Neurology 226:21–27.

    https://doi.org/10.1002/cne.902260103

    • PubMed
    • Google Scholar
34. 1. Guinan JJ

    (2018) Olivocochlear efferents: their action, effects, measurement and uses, and the impact of the new conception of cochlear mechanical responses

    Hearing Research 362:38–47.

    https://doi.org/10.1016/j.heares.2017.12.012

    • PubMed
    • Google Scholar
35. 1. Hafemeister C
    2. Satija R

    (2019) Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression

    Genome Biology 20:296.

    https://doi.org/10.1186/s13059-019-1874-1

    • PubMed
    • Google Scholar
36. 1. Highstein SM
    2. Baker R

    (1985) Action of the efferent vestibular system on primary afferents in the toadfish, Opsanus tau

    Journal of Neurophysiology 54:370–384.

    https://doi.org/10.1152/jn.1985.54.2.370

    • PubMed
    • Google Scholar
37. 1. Hoeffel G
    2. Debroas G
    3. Roger A
    4. Rossignol R
    5. Gouilly J
    6. Laprie C
    7. Chasson L
    8. Barbon P-V
    9. Balsamo A
    10. Reynders A
    11. Moqrich A
    12. Ugolini S

    (2021) Sensory neuron-derived TAFA4 promotes macrophage tissue repair functions

    Nature 594:94–99.

    https://doi.org/10.1038/s41586-021-03563-7

    • PubMed
    • Google Scholar
38. 1. Hong H
    2. Zeppenfeld D
    3. Trussell LO

    (2022) Electrical signaling in cochlear efferents is driven by an intrinsic neuronal oscillator

    PNAS 119:e2209565119.

    https://doi.org/10.1073/pnas.2209565119

    • PubMed
    • Google Scholar
39. 1. Hua Y
    2. Ding X
    3. Wang H
    4. Wang F
    5. Lu Y
    6. Neef J
    7. Gao Y
    8. Moser T
    9. Wu H

    (2021) Electron microscopic reconstruction of neural circuitry in the cochlea

    Cell Reports 34:108551.

    https://doi.org/10.1016/j.celrep.2020.108551

    • PubMed
    • Google Scholar
40. 1. Ilicic T
    2. Kim JK
    3. Kolodziejczyk AA
    4. Bagger FO
    5. McCarthy DJ
    6. Marioni JC
    7. Teichmann SA

    (2016) Classification of low quality cells from single-cell RNA-seq data

    Genome Biology 17:29.

    https://doi.org/10.1186/s13059-016-0888-1

    • PubMed
    • Google Scholar
41. 1. Imig C
    2. Min S-W
    3. Krinner S
    4. Arancillo M
    5. Rosenmund C
    6. Südhof TC
    7. Rhee J
    8. Brose N
    9. Cooper BH

    (2014) The morphological and molecular nature of synaptic vesicle priming at presynaptic active zones

    Neuron 84:416–431.

    https://doi.org/10.1016/j.neuron.2014.10.009

    • PubMed
    • Google Scholar
42. 1. Irving S
    2. Moore DR
    3. Liberman MC
    4. Sumner CJ

    (2011) Olivocochlear efferent control in sound localization and experience-dependent learning

    The Journal of Neuroscience 31:2493–2501.

    https://doi.org/10.1523/JNEUROSCI.2679-10.2011

    • PubMed
    • Google Scholar
43. 1. Jockusch WJ
    2. Speidel D
    3. Sigler A
    4. Sørensen JB
    5. Varoqueaux F
    6. Rhee J-S
    7. Brose N

    (2007) Caps-1 and CAPS-2 are essential synaptic vesicle priming proteins

    Cell 131:796–808.

    https://doi.org/10.1016/j.cell.2007.11.002

    • PubMed
    • Google Scholar
44. 1. Kaiser A
    2. Alexandrova O
    3. Grothe B

    (2011) Urocortin-expressing olivocochlear neurons exhibit tonotopic and developmental changes in the auditory brainstem and in the innervation of the cochlea

    The Journal of Comparative Neurology 519:2758–2778.

    https://doi.org/10.1002/cne.22650

    • PubMed
    • Google Scholar
45. 1. Kandler K
    2. Clause A
    3. Noh J

    (2009) Tonotopic reorganization of developing auditory brainstem circuits

    Nature Neuroscience 12:711–717.

    https://doi.org/10.1038/nn.2332

    • PubMed
    • Google Scholar
46. 1. Karis A
    2. Pata I
    3. van Doorninck JH
    4. Grosveld F
    5. de Zeeuw CI
    6. de Caprona D
    7. Fritzsch B

    (2001) Transcription factor GATA-3 alters pathway selection of olivocochlear neurons and affects morphogenesis of the ear

    The Journal of Comparative Neurology 429:615–630.

    https://doi.org/10.1002/1096-9861(20010122)429:4<615::aid-cne8>3.0.co;2-f

    • PubMed
    • Google Scholar
47. 1. Kawase T
    2. Delgutte B
    3. Liberman MC

    (1993) Antimasking effects of the olivocochlear reflex. II. enhancement of auditory-nerve response to masked tones

    Journal of Neurophysiology 70:2533–2549.

    https://doi.org/10.1152/jn.1993.70.6.2533

    • PubMed
    • Google Scholar
48. 1. Kawashima Y
    2. Géléoc GSG
    3. Kurima K
    4. Labay V
    5. Lelli A
    6. Asai Y
    7. Makishima T
    8. Wu DK
    9. Della Santina CC
    10. Holt JR
    11. Griffith AJ

    (2011) Mechanotransduction in mouse inner ear hair cells requires transmembrane channel-like genes

    The Journal of Clinical Investigation 121:4796–4809.

    https://doi.org/10.1172/JCI60405

    • PubMed
    • Google Scholar
49. 1. Kim D-W
    2. Yao Z
    3. Graybuck LT
    4. Kim TK
    5. Nguyen TN
    6. Smith KA
    7. Fong O
    8. Yi L
    9. Koulena N
    10. Pierson N
    11. Shah S
    12. Lo L
    13. Pool A-H
    14. Oka Y
    15. Pachter L
    16. Cai L
    17. Tasic B
    18. Zeng H
    19. Anderson DJ

    (2019) Multimodal analysis of cell types in a hypothalamic node controlling social behavior

    Cell 179:713–728.

    https://doi.org/10.1016/j.cell.2019.09.020

    • PubMed
    • Google Scholar
50. 1. Kitcher SR
    2. Pederson AM
    3. Weisz CJC

    (2022) Diverse identities and sites of action of cochlear neurotransmitters

    Hearing Research 419:108278.

    https://doi.org/10.1016/j.heares.2021.108278

    • PubMed
    • Google Scholar
51. 1. Korada S
    2. Schwartz IR

    (1999) Development of GABA, glycine, and their receptors in the auditory brainstem of gerbil: a light and electron microscopic study

    The Journal of Comparative Neurology 409:664–681.

    https://doi.org/10.1002/(sici)1096-9861(19990712)409:4<664::aid-cne10>3.0.co;2-s

    • PubMed
    • Google Scholar
52. 1. Kotak VC
    2. Korada S
    3. Schwartz IR
    4. Sanes DH

    (1998) A developmental shift from GABAergic to glycinergic transmission in the central auditory system

    The Journal of Neuroscience 18:4646–4655.

    https://doi.org/10.1523/JNEUROSCI.18-12-04646.1998

    • PubMed
    • Google Scholar
53. 1. Kujawa SG
    2. Liberman MC

    (1997) Conditioning-related protection from acoustic injury: effects of chronic deefferentation and sham surgery

    Journal of Neurophysiology 78:3095–3106.

    https://doi.org/10.1152/jn.1997.78.6.3095

    • PubMed
    • Google Scholar
54. 1. Larsen E
    2. Liberman MC

    (2010) Contralateral cochlear effects of ipsilateral damage: no evidence for interaural coupling

    Hearing Research 260:70–80.

    https://doi.org/10.1016/j.heares.2009.11.011

    • PubMed
    • Google Scholar
55. 1. Lauer AM
    2. Fuchs PA
    3. Ryugo DK
    4. Francis HW

    (2012) Efferent synapses return to inner hair cells in the aging cochlea

    Neurobiology of Aging 33:2892–2902.

    https://doi.org/10.1016/j.neurobiolaging.2012.02.007

    • PubMed
    • Google Scholar
56. 1. Le Prell CG
    2. Hughes LF
    3. Dolan DF
    4. Bledsoe SC

    (2021) Effects of calcitonin-gene-related-peptide on auditory nerve activity

    Frontiers in Cell and Developmental Biology 9:752963.

    https://doi.org/10.3389/fcell.2021.752963

    • PubMed
    • Google Scholar
57. 1. Liberman MC

    (1978) Auditory‐nerve response from cats raised in a low‐noise chamber

    The Journal of the Acoustical Society of America 63:442–455.

    https://doi.org/10.1121/1.381736

    • PubMed
    • Google Scholar
58. 1. Liberman MC

    (1980) Efferent synapses in the inner hair cell area of the cat cochlea: an electron microscopic study of serial sections

    Hearing Research 3:189–204.

    https://doi.org/10.1016/0378-5955(80)90046-5

    • PubMed
    • Google Scholar
59. 1. Liberman MC

    (1982) Single-neuron labeling in the cat auditory nerve

    Science 216:1239–1241.

    https://doi.org/10.1126/science.7079757

    • PubMed
    • Google Scholar
60. 1. Liberman MC
    2. Dodds LW
    3. Pierce S

    (1990) Afferent and efferent innervation of the cat cochlea: quantitative analysis with light and electron microscopy

    The Journal of Comparative Neurology 301:443–460.

    https://doi.org/10.1002/cne.903010309

    • PubMed
    • Google Scholar
61. 1. Liberman LD
    2. Wang H
    3. Liberman MC

    (2011) Opposing gradients of ribbon size and AMPA receptor expression underlie sensitivity differences among cochlear-nerve/hair-cell synapses

    The Journal of Neuroscience 31:801–808.

    https://doi.org/10.1523/JNEUROSCI.3389-10.2011

    • PubMed
    • Google Scholar
62. 1. Liberman MC
    2. Kujawa SG

    (2017) Cochlear synaptopathy in acquired sensorineural hearing loss: manifestations and mechanisms

    Hearing Research 349:138–147.

    https://doi.org/10.1016/j.heares.2017.01.003

    • PubMed
    • Google Scholar
63. 1. Linkert M
    2. Rueden CT
    3. Allan C
    4. Burel J-M
    5. Moore W
    6. Patterson A
    7. Loranger B
    8. Moore J
    9. Neves C
    10. MacDonald D
    11. Tarkowska A
    12. Sticco C
    13. Hill E
    14. Rossner M
    15. Eliceiri KW
    16. Swedlow JR

    (2010) Metadata matters: access to image data in the real world

    Journal of Cell Biology 189:777–782.

    https://doi.org/10.1083/jcb.201004104

    • PubMed
    • Google Scholar
64. 1. Liu H
    2. Chen L
    3. Giffen KP
    4. Stringham ST
    5. Li Y
    6. Judge PD
    7. Beisel KW
    8. He DZZ

    (2018) Cell-specific transcriptome analysis shows that adult pillar and deiters’ cells express genes encoding machinery for specializations of cochlear hair cells

    Frontiers in Molecular Neuroscience 11:356.

    https://doi.org/10.3389/fnmol.2018.00356

    • PubMed
    • Google Scholar
65. 1. Long P
    2. Wan G
    3. Roberts MT
    4. Corfas G

    (2018) Myelin development, plasticity, and pathology in the auditory system

    Developmental Neurobiology 78:80–92.

    https://doi.org/10.1002/dneu.22538

    • PubMed
    • Google Scholar
66. 1. Luecken MD
    2. Theis FJ

    (2019) Current best practices in single-cell RNA-seq analysis: a tutorial

    Molecular Systems Biology 15:e8746.

    https://doi.org/10.15252/msb.20188746

    • PubMed
    • Google Scholar
67. 1. Luo W
    2. Enomoto H
    3. Rice FL
    4. Milbrandt J
    5. Ginty DD

    (2009) Molecular identification of rapidly adapting mechanoreceptors and their developmental dependence on RET signaling

    Neuron 64:841–856.

    https://doi.org/10.1016/j.neuron.2009.11.003

    • PubMed
    • Google Scholar
68. 1. Madisen L
    2. Zwingman TA
    3. Sunkin SM
    4. Oh SW
    5. Zariwala HA
    6. Gu H
    7. Ng LL
    8. Palmiter RD
    9. Hawrylycz MJ
    10. Jones AR
    11. Lein ES
    12. Zeng H

    (2010) A robust and high-throughput CRE reporting and characterization system for the whole mouse brain

    Nature Neuroscience 13:133–140.

    https://doi.org/10.1038/nn.2467

    • PubMed
    • Google Scholar
69. 1. Maison SF
    2. Adams JC
    3. Liberman MC

    (2003a) Olivocochlear innervation in the mouse: immunocytochemical maps, crossed versus uncrossed contributions, and transmitter colocalization

    The Journal of Comparative Neurology 455:406–416.

    https://doi.org/10.1002/cne.10490

    • PubMed
    • Google Scholar
70. 1. Maison SF
    2. Emeson RB
    3. Adams JC
    4. Luebke AE
    5. Liberman MC

    (2003b) Loss of alpha CGRP reduces sound-evoked activity in the cochlear nerve

    Journal of Neurophysiology 90:2941–2949.

    https://doi.org/10.1152/jn.00596.2003

    • PubMed
    • Google Scholar
71. 1. Maison SF
    2. Usubuchi H
    3. Liberman MC

    (2013) Efferent feedback minimizes cochlear neuropathy from moderate noise exposure

    The Journal of Neuroscience 33:5542–5552.

    https://doi.org/10.1523/JNEUROSCI.5027-12.2013

    • PubMed
    • Google Scholar
72. 1. Mathews MA
    2. Camp AJ
    3. Murray AJ

    (2017) Reviewing the role of the efferent vestibular system in motor and vestibular circuits

    Frontiers in Physiology 8:552.

    https://doi.org/10.3389/fphys.2017.00552

    • PubMed
    • Google Scholar
73. 1. McGinley MJ
    2. Vinck M
    3. Reimer J
    4. Batista-Brito R
    5. Zagha E
    6. Cadwell CR
    7. Tolias AS
    8. Cardin JA
    9. McCormick DA

    (2015) Waking state: rapid variations modulate neural and behavioral responses

    Neuron 87:1143–1161.

    https://doi.org/10.1016/j.neuron.2015.09.012

    • PubMed
    • Google Scholar
74. 1. McLean WJ
    2. Smith KA
    3. Glowatzki E
    4. Pyott SJ

    (2009) Distribution of the Na, K-ATPase alpha subunit in the rat spiral ganglion and organ of Corti

    Journal of the Association for Research in Otolaryngology 10:37–49.

    https://doi.org/10.1007/s10162-008-0152-9

    • PubMed
    • Google Scholar
75. 1. Mo A
    2. Mukamel EA
    3. Davis FP
    4. Luo C
    5. Henry GL
    6. Picard S
    7. Urich MA
    8. Nery JR
    9. Sejnowski TJ
    10. Lister R
    11. Eddy SR
    12. Ecker JR
    13. Nathans J

    (2015) Epigenomic signatures of neuronal diversity in the mammalian brain

    Neuron 86:1369–1384.

    https://doi.org/10.1016/j.neuron.2015.05.018

    • PubMed
    • Google Scholar
76. 1. Mountain DC

    (1980) Changes in endolymphatic potential and crossed olivocochlear bundle stimulation alter cochlear mechanics

    Science 210:71–72.

    https://doi.org/10.1126/science.7414321

    • PubMed
    • Google Scholar
77. 1. Nestvogel DB
    2. Merino RM
    3. Leon-Pinzon C
    4. Schottdorf M
    5. Lee C
    6. Imig C
    7. Brose N
    8. Rhee J-S

    (2020) The synaptic vesicle priming protein CAPS-1 shapes the adaptation of sensory evoked responses in mouse visual cortex

    Cell Reports 30:3261–3269.

    https://doi.org/10.1016/j.celrep.2020.02.045

    • PubMed
    • Google Scholar
78. 1. Niu X
    2. Canlon B

    (2002) Activation of tyrosine hydroxylase in the lateral efferent terminals by sound conditioning

    Hearing Research 174:124–132.

    https://doi.org/10.1016/s0378-5955(02)00646-9

    • PubMed
    • Google Scholar
79. 1. Oatman LC

    (1971) Role of visual attention on auditory evoked potentials in unanesthetized cats

    Experimental Neurology 32:341–356.

    https://doi.org/10.1016/0014-4886(71)90003-3

    • PubMed
    • Google Scholar
80. 1. Oatman LC

    (1976) Effects of visual attention on the intensity of auditory evoked potentials

    Experimental Neurology 51:41–53.

    https://doi.org/10.1016/0014-4886(76)90052-2

    • PubMed
    • Google Scholar
81. 1. Pata I
    2. Studer M
    3. van Doorninck JH
    4. Briscoe J
    5. Kuuse S
    6. Engel JD
    7. Grosveld F
    8. Karis A

    (1999) The transcription factor GATA3 is a downstream effector of HOXB1 specification in rhombomere 4

    Development 126:5523–5531.

    https://doi.org/10.1242/dev.126.23.5523

    • PubMed
    • Google Scholar
82. 1. Paul A
    2. Crow M
    3. Raudales R
    4. He M
    5. Gillis J
    6. Huang ZJ

    (2017) Transcriptional architecture of synaptic communication delineates gabaergic neuron identity

    Cell 171:522–539.

    https://doi.org/10.1016/j.cell.2017.08.032

    • PubMed
    • Google Scholar
83. 1. Peng H
    2. Xie P
    3. Liu L
    4. Kuang X
    5. Wang Y
    6. Qu L
    7. Gong H
    8. Jiang S
    9. Li A
    10. Ruan Z
    11. Ding L
    12. Yao Z
    13. Chen C
    14. Chen M
    15. Daigle TL
    16. Dalley R
    17. Ding Z
    18. Duan Y
    19. Feiner A
    20. He P
    21. Hill C
    22. Hirokawa KE
    23. Hong G
    24. Huang L
    25. Kebede S
    26. Kuo H-C
    27. Larsen R
    28. Lesnar P
    29. Li L
    30. Li Q
    31. Li X
    32. Li Y
    33. Li Y
    34. Liu A
    35. Lu D
    36. Mok S
    37. Ng L
    38. Nguyen TN
    39. Ouyang Q
    40. Pan J
    41. Shen E
    42. Song Y
    43. Sunkin SM
    44. Tasic B
    45. Veldman MB
    46. Wakeman W
    47. Wan W
    48. Wang P
    49. Wang Q
    50. Wang T
    51. Wang Y
    52. Xiong F
    53. Xiong W
    54. Xu W
    55. Ye M
    56. Yin L
    57. Yu Y
    58. Yuan J
    59. Yuan J
    60. Yun Z
    61. Zeng S
    62. Zhang S
    63. Zhao S
    64. Zhao Z
    65. Zhou Z
    66. Huang ZJ
    67. Esposito L
    68. Hawrylycz MJ
    69. Sorensen SA
    70. Yang XW
    71. Zheng Y
    72. Gu Z
    73. Xie W
    74. Koch C
    75. Luo Q
    76. Harris JA
    77. Wang Y
    78. Zeng H

    (2021) Morphological diversity of single neurons in molecularly defined cell types

    Nature 598:174–181.

    https://doi.org/10.1038/s41586-021-03941-1

    • PubMed
    • Google Scholar
84. 1. Petitpré C
    2. Wu H
    3. Sharma A
    4. Tokarska A
    5. Fontanet P
    6. Wang Y
    7. Helmbacher F
    8. Yackle K
    9. Silberberg G
    10. Hadjab S
    11. Lallemend F

    (2018) Neuronal heterogeneity and stereotyped connectivity in the auditory afferent system

    Nature Communications 9:3691.

    https://doi.org/10.1038/s41467-018-06033-3

    • PubMed
    • Google Scholar
85. 1. Petitpré C
    2. Bourien J
    3. Wu H
    4. Diuba A
    5. Puel J-L
    6. Lallemend F

    (2020) Genetic and functional diversity of primary auditory afferents

    Current Opinion in Physiology 18:85–94.

    https://doi.org/10.1016/j.cophys.2020.09.011

    • Google Scholar
86. 1. Plummer NW
    2. Evsyukova IY
    3. Robertson SD
    4. de Marchena J
    5. Tucker CJ
    6. Jensen P

    (2015) Expanding the power of recombinase-based labeling to uncover cellular diversity

    Development 142:4385–4393.

    https://doi.org/10.1242/dev.129981

    • PubMed
    • Google Scholar
87. 1. Rajan R

    (1995) Involvement of cochlear efferent pathways in protective effects elicited with binaural loud sound exposure in cats

    Journal of Neurophysiology 74:582–597.

    https://doi.org/10.1152/jn.1995.74.2.582

    • PubMed
    • Google Scholar
88. 1. Rasmussen GL

    (1946) The olivary peduncle and other fiber projections of the superior olivary complex

    The Journal of Comparative Neurology 84:141–219.

    https://doi.org/10.1002/cne.900840204

    • PubMed
    • Google Scholar
89. 1. Rasmussen GL

    (1953) Further observations of the efferent cochlear bundle

    The Journal of Comparative Neurology 99:61–74.

    https://doi.org/10.1002/cne.900990105

    • PubMed
    • Google Scholar
90. 1. Reiter ER
    2. Liberman MC

    (1995) Efferent-mediated protection from acoustic overexposure: relation to slow effects of olivocochlear stimulation

    Journal of Neurophysiology 73:506–514.

    https://doi.org/10.1152/jn.1995.73.2.506

    • PubMed
    • Google Scholar
91. 1. Renden R
    2. Berwin B
    3. Davis W
    4. Ann K
    5. Chin CT
    6. Kreber R
    7. Ganetzky B
    8. Martin TF
    9. Broadie K

    (2001) Drosophila caps is an essential gene that regulates dense-core vesicle release and synaptic vesicle fusion

    Neuron 31:421–437.

    https://doi.org/10.1016/s0896-6273(01)00382-8

    • PubMed
    • Google Scholar
92. Book

    1. Roberts BL
    2. Meredith GE

    (1992) The efferent innervation of the ear: variations on an enigma

    In: Webster DB, Popper AN, Fay RR, editors. The Evolutionary Biology of Hearing. Springer. pp. 185–210.

    https://doi.org/10.1007/978-1-4612-2784-7_16

    • PubMed
    • Google Scholar
93. 1. Robertson D
    2. Gummer M

    (1985) Physiological and morphological characterization of efferent neurones in the guinea pig cochlea

    Hearing Research 20:63–77.

    https://doi.org/10.1016/0378-5955(85)90059-0

    • PubMed
    • Google Scholar
94. 1. Robertson D
    2. Anderson CJ
    3. Cole KS

    (1987) Segregation of efferent projections to different turns of the guinea pig cochlea

    Hearing Research 25:69–76.

    https://doi.org/10.1016/0378-5955(87)90080-3

    • PubMed
    • Google Scholar
95. 1. Romero GE
    2. Trussell LO

    (2022) Central circuitry and function of the cochlear efferent systems

    Hearing Research 425:108516.

    https://doi.org/10.1016/j.heares.2022.108516

    • PubMed
    • Google Scholar
96. 1. Rossi J
    2. Balthasar N
    3. Olson D
    4. Scott M
    5. Berglund E
    6. Lee CE
    7. Choi MJ
    8. Lauzon D
    9. Lowell BB
    10. Elmquist JK

    (2011) Melanocortin-4 receptors expressed by cholinergic neurons regulate energy balance and glucose homeostasis

    Cell Metabolism 13:195–204.

    https://doi.org/10.1016/j.cmet.2011.01.010

    • PubMed
    • Google Scholar
97. 1. Ruel J
    2. Nouvian R
    3. Gervais d’Aldin C
    4. Pujol R
    5. Eybalin M
    6. Puel JL

    (2001) Dopamine inhibition of auditory nerve activity in the adult mammalian cochlea

    The European Journal of Neuroscience 14:977–986.

    https://doi.org/10.1046/j.0953-816x.2001.01721.x

    • PubMed
    • Google Scholar
98. 1. Safieddine S
    2. Eybalin M

    (1992) Triple immunofluorescence evidence for the coexistence of acetylcholine, enkephalins and calcitonin gene-related peptide within efferent (olivocochlear) neurons of rats and guinea-pigs

    The European Journal of Neuroscience 4:981–992.

    https://doi.org/10.1111/j.1460-9568.1992.tb00124.x

    • PubMed
    • Google Scholar
99. 1. Safieddine S
    2. Prior AM
    3. Eybalin M

    (1997) Choline acetyltransferase, glutamate decarboxylase, tyrosine hydroxylase, calcitonin gene-related peptide and opioid peptides coexist in lateral efferent neurons of rat and guinea-pig

    The European Journal of Neuroscience 9:356–367.

    https://doi.org/10.1111/j.1460-9568.1997.tb01405.x

    • PubMed
    • Google Scholar
100. 1. Schindelin J
     2. Arganda-Carreras I
     3. Frise E
     4. Kaynig V
     5. Longair M
     6. Pietzsch T
     7. Preibisch S
     8. Rueden C
     9. Saalfeld S
     10. Schmid B
     11. Tinevez J-Y
     12. White DJ
     13. Hartenstein V
     14. Eliceiri K
     15. Tomancak P
     16. Cardona A

     (2012) Fiji: an open-source platform for biological-image analysis

     Nature Methods 9:676–682.

     https://doi.org/10.1038/nmeth.2019

     • PubMed
     • Google Scholar
101. 1. Sewell WF
     2. Starr PA

     (1991) Effects of calcitonin gene-related peptide and efferent nerve stimulation on afferent transmission in the lateral line organ

     Journal of Neurophysiology 65:1158–1169.

     https://doi.org/10.1152/jn.1991.65.5.1158

     • PubMed
     • Google Scholar
102. Book

     1. Sewell WF

     (2011) Pharmacology and neurochemistry of olivocochlear efferents

     In: Ryugo DK, Fay RR, Popper AN, editors. Auditory and Vestibular Efferents. Springer. pp. 83–101.

     https://doi.org/10.1007/978-1-4419-7070-1_4

     • Google Scholar
103. 1. Shrestha BR
     2. Chia C
     3. Wu L
     4. Kujawa SG
     5. Liberman MC
     6. Goodrich LV

     (2018) Sensory neuron diversity in the inner ear is shaped by activity

     Cell 174:1229–1246.

     https://doi.org/10.1016/j.cell.2018.07.007

     • PubMed
     • Google Scholar
104. 1. Siegel JH
     2. Kim DO

     (1982) Efferent neural control of cochlear mechanics? olivocochlear bundle stimulation affects cochlear biomechanical nonlinearity

     Hearing Research 6:171–182.

     https://doi.org/10.1016/0378-5955(82)90052-1

     • PubMed
     • Google Scholar
105. 1. Simmons DD
     2. Raji-Kubba J
     3. Popper P
     4. Micevych PE

     (1997)

     Developmentally regulated expression of calcitonin gene-related peptide in the superior olive

     The Journal of Comparative Neurology 377:207–216.

     • PubMed
     • Google Scholar
106. 1. Slominski AT
     2. Zmijewski MA
     3. Zbytek B
     4. Tobin DJ
     5. Theoharides TC
     6. Rivier J

     (2013) Key role of CRF in the skin stress response system

     Endocrine Reviews 34:827–884.

     https://doi.org/10.1210/er.2012-1092

     • PubMed
     • Google Scholar
107. 1. Speidel D
     2. Varoqueaux F
     3. Enk C
     4. Nojiri M
     5. Grishanin RN
     6. Martin TFJ
     7. Hofmann K
     8. Brose N
     9. Reim K

     (2003) A family of Ca2+-dependent activator proteins for secretion: comparative analysis of structure, expression, localization, and function

     The Journal of Biological Chemistry 278:52802–52809.

     https://doi.org/10.1074/jbc.M304727200

     • PubMed
     • Google Scholar
108. 1. Sterenborg JC
     2. Pilati N
     3. Sheridan CJ
     4. Uchitel OD
     5. Forsythe ID
     6. Barnes-Davies M

     (2010) Lateral olivocochlear (Loc) neurons of the mouse LSO receive excitatory and inhibitory synaptic inputs with slower kinetics than LSO principal neurons

     Hearing Research 270:119–126.

     https://doi.org/10.1016/j.heares.2010.08.013

     • PubMed
     • Google Scholar
109. 1. Stringer C
     2. Wang T
     3. Michaelos M
     4. Pachitariu M

     (2021) Cellpose: a generalist algorithm for cellular segmentation

     Nature Methods 18:100–106.

     https://doi.org/10.1038/s41592-020-01018-x

     • PubMed
     • Google Scholar
110. 1. Stuart T
     2. Butler A
     3. Hoffman P
     4. Hafemeister C
     5. Papalexi E
     6. Mauck WM
     7. Hao Y
     8. Stoeckius M
     9. Smibert P
     10. Satija R

     (2019) Comprehensive integration of single-cell data

     Cell 177:1888–1902.

     https://doi.org/10.1016/j.cell.2019.05.031

     • PubMed
     • Google Scholar
111. 1. Sun S
     2. Babola T
     3. Pregernig G
     4. So KS
     5. Nguyen M
     6. Su S-SM
     7. Palermo AT
     8. Bergles DE
     9. Burns JC
     10. Müller U

     (2018) Hair cell mechanotransduction regulates spontaneous activity and spiral ganglion subtype specification in the auditory system

     Cell 174:1247–1263.

     https://doi.org/10.1016/j.cell.2018.07.008

     • PubMed
     • Google Scholar
112. 1. Takeda N
     2. Doi K
     3. Mori N
     4. Yamazaki H
     5. Tohyama M
     6. Matsunaga T

     (1987)

     Localization and fine structure of calcitonin gene-related peptide (CGRP) -like immunoreactive nerve fibres in the organ of Corti of guinea pigs by immunohistochemistry

     Acta Oto-Laryngologica 103:567–571.

     • PubMed
     • Google Scholar
113. 1. Tandon A
     2. Bannykh S
     3. Kowalchyk JA
     4. Banerjee A
     5. Martin TF
     6. Balch WE

     (1998) Differential regulation of exocytosis by calcium and caps in semi-intact synaptosomes

     Neuron 21:147–154.

     https://doi.org/10.1016/s0896-6273(00)80522-x

     • PubMed
     • Google Scholar
114. 1. Taranda J
     2. Maison SF
     3. Ballestero JA
     4. Katz E
     5. Savino J
     6. Vetter DE
     7. Boulter J
     8. Liberman MC
     9. Fuchs PA
     10. Elgoyhen AB

     (2009) A point mutation in the hair cell nicotinic cholinergic receptor prolongs cochlear inhibition and enhances noise protection

     PLOS Biology 7:e18.

     https://doi.org/10.1371/journal.pbio.1000018

     • PubMed
     • Google Scholar
115. 1. Tenney AP
     2. Livet J
     3. Belton T
     4. Prochazkova M
     5. Pearson EM
     6. Whitman MC
     7. Kulkarni AB
     8. Engle EC
     9. Henderson CE

     (2019) Etv1 controls the establishment of non-overlapping motor innervation of neighboring facial muscles during development

     Cell Reports 29:437–452.

     https://doi.org/10.1016/j.celrep.2019.08.078

     • PubMed
     • Google Scholar
116. 1. Terreros G
     2. Jorratt P
     3. Aedo C
     4. Elgoyhen AB
     5. Delano PH

     (2016) Selective attention to visual stimuli using auditory distractors is altered in alpha-9 nicotinic receptor subunit knock-out mice

     The Journal of Neuroscience 36:7198–7209.

     https://doi.org/10.1523/JNEUROSCI.4031-15.2016

     • PubMed
     • Google Scholar
117. 1. Tong H
     2. Kopp-Scheinpflug C
     3. Pilati N
     4. Robinson SW
     5. Sinclair JL
     6. Steinert JR
     7. Barnes-Davies M
     8. Allfree R
     9. Grubb BD
     10. Young SM Jr
     11. Forsythe ID

     (2013) Protection from noise-induced hearing loss by Kv2.2 potassium currents in the central medial olivocochlear system

     The Journal of Neuroscience 33:9113–9121.

     https://doi.org/10.1523/JNEUROSCI.5043-12.2013

     • PubMed
     • Google Scholar
118. 1. Torres Cadenas L
     2. Fischl MJ
     3. Weisz CJC

     (2020) Synaptic inhibition of medial olivocochlear efferent neurons by neurons of the medial nucleus of the trapezoid body

     The Journal of Neuroscience 40:509–525.

     https://doi.org/10.1523/JNEUROSCI.1288-19.2019

     • PubMed
     • Google Scholar
119. 1. Turner JG
     2. Parrish JL
     3. Hughes LF
     4. Toth LA
     5. Caspary DM

     (2005)

     Hearing in laboratory animals: strain differences and nonauditory effects of noise

     Comparative Medicine 55:12–23.

     • PubMed
     • Google Scholar
120. 1. van den Pol AN
     2. Yao Y
     3. Fu L-Y
     4. Foo K
     5. Huang H
     6. Coppari R
     7. Lowell BB
     8. Broberger C

     (2009) Neuromedin B and gastrin-releasing peptide excite arcuate nucleus neuropeptide Y neurons in a novel transgenic mouse expressing strong Renilla green fluorescent protein in NPY neurons

     The Journal of Neuroscience 29:4622–4639.

     https://doi.org/10.1523/JNEUROSCI.3249-08.2009

     • PubMed
     • Google Scholar
121. 1. van der Heijden M
     2. Vavakou A

     (2022) Rectifying and sluggish: outer hair cells as regulators rather than amplifiers

     Hearing Research 423:108367.

     https://doi.org/10.1016/j.heares.2021.108367

     • PubMed
     • Google Scholar
122. 1. Vega R
     2. García-Garibay O
     3. Soto E

     (2022) Opioid receptor activation modulates the calcium current in the cochlear outer hair cells of the rat

     The European Journal of Neuroscience 56:3543–3552.

     https://doi.org/10.1111/ejn.15682

     • PubMed
     • Google Scholar
123. 1. Vetter DE
     2. Adams JC
     3. Mugnaini E

     (1991) Chemically distinct rat olivocochlear neurons

     Synapse 7:21–43.

     https://doi.org/10.1002/syn.890070104

     • PubMed
     • Google Scholar
124. 1. Vetter DE
     2. Mugnaini E

     (1992) Distribution and dendritic features of three groups of rat olivocochlear neurons

     Anatomy and Embryology 185:1–16.

     https://doi.org/10.1007/BF00213596

     • PubMed
     • Google Scholar
125. 1. Waltman L
     2. van Eck NJ

     (2013) A smart local moving algorithm for large-scale modularity-based community detection

     The European Physical Journal B 86:471.

     https://doi.org/10.1140/epjb/e2013-40829-0

     • Google Scholar
126. 1. Warr WB
     2. Boche JB
     3. Neely ST

     (1997) Efferent innervation of the inner hair cell region: origins and terminations of two lateral olivocochlear systems

     Hearing Research 108:89–111.

     https://doi.org/10.1016/s0378-5955(97)00044-0

     • PubMed
     • Google Scholar
127. 1. Winslow RL
     2. Sachs MB

     (1987) Effect of electrical stimulation of the crossed olivocochlear bundle on auditory nerve response to tones in noise

     Journal of Neurophysiology 57:1002–1021.

     https://doi.org/10.1152/jn.1987.57.4.1002

     • PubMed
     • Google Scholar
128. 1. Wu JS
     2. Vyas P
     3. Glowatzki E
     4. Fuchs PA

     (2018) Opposing expression gradients of calcitonin-related polypeptide alpha (calca/cgrpα) and tyrosine hydroxylase (th) in type II afferent neurons of the mouse cochlea

     The Journal of Comparative Neurology 526:425–438.

     https://doi.org/10.1002/cne.24341

     • PubMed
     • Google Scholar
129. 1. Wu JS
     2. Yi E
     3. Manca M
     4. Javaid H
     5. Lauer AM
     6. Glowatzki E

     (2020) Sound exposure dynamically induces dopamine synthesis in cholinergic loc efferents for feedback to auditory nerve fibers

     eLife 9:e52419.

     https://doi.org/10.7554/eLife.52419

     • PubMed
     • Google Scholar
130. 1. Yu WM
     2. Goodrich LV

     (2014) Morphological and physiological development of auditory synapses

     Hearing Research 311:3–16.

     https://doi.org/10.1016/j.heares.2014.01.007

     • PubMed
     • Google Scholar
131. 1. Zachary SP
     2. Fuchs PA

     (2015) Re-Emergent inhibition of cochlear inner hair cells in a mouse model of hearing loss

     Journal of Neuroscience 35:9701–9706.

     https://doi.org/10.1523/JNEUROSCI.0879-15.2015

     • PubMed
     • Google Scholar
132. 1. Zeng H

     (2022) What is a cell type and how to define it?

     Cell 185:2739–2755.

     https://doi.org/10.1016/j.cell.2022.06.031

     • PubMed
     • Google Scholar
133. 1. Zhang SC

     (2001) Defining glial cells during CNS development

     Nature Reviews. Neuroscience 2:840–843.

     https://doi.org/10.1038/35097593

     • PubMed
     • Google Scholar
134. 1. Zhang SX
     2. Lutas A
     3. Yang S
     4. Diaz A
     5. Fluhr H
     6. Nagel G
     7. Gao S
     8. Andermann ML

     (2021) Hypothalamic dopamine neurons motivate mating through persistent cAMP signalling

     Nature 597:245–249.

     https://doi.org/10.1038/s41586-021-03845-0

     • PubMed
     • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Experience-dependent flexibility in a molecularly diverse central-to-peripheral auditory feedback system" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Catherine Emily Carr as Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Barbara Shinn-Cunningham as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Zhiyong Liu (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) Please modify your title, along the lines of perhaps along the lines of " Single cell transcriptomic profiling reveals xxxxx"

2) Follow the reviewer's suggestions to improve clarity.

Reviewer #2 (Recommendations for the authors):

Frank et al. in this study thoroughly characterized the molecular, physiological, and morphological properties of the brainstem olivocochlear neurons (OCNs). First, by using the ChATCre+; Rosa26-Sun1-GFP/+ and 10x single cell RNA-Seq platform, they identified two Gata3+ cell clusters that are defined as OCNs: one is Ucn+ LOC, and the other is Atp1a3+ MOC. By further comparing the gene profiles between OCNs and the lineage-related FMNs, several previously unknown OCN markers such as Cadps2 are revealed. Furthermore, they also noticed that OCNs have high internal heterogeneity and they further found LOC and MOC-enriched genes, including but not limited to Col4a4 and Zfp804a, respectively. More importantly, two LOC subtypes, LOC-1 and LOC-2, are identified, and especially the LOC-2 sub-cluster cells, which are located in the medial wing of the LSO, can enrich many neuropeptide genes, CGRP, Calcb, NPY, Ucn and so on. The NPY expression in LOC-2 is further validated by the well-designed genetic fate mapping approach, which is convincing. Finally, the author showed that the neuropeptide expression patterns did not significantly change between wild-type and Tmc1/2 knockout mice at the pre-hearing ages, but either the neuropeptide expression levels or distribution patterns did change manifestly after hearing onset. Besides, they also showed that noise exposure can alter the neuropeptide patterns.

In general, the paper is clearly written and the data are well presented. The logic is straightforward, and in most cases, the data are compelling to support the conclusions. I only have 2 main comments below:

1) In my point of view, it should be a resource paper rather than a research article because the current version did not connect any specific marker pattern with the biological properties. For example, the Npy identity does not appear to be predictive of LOC morphology. It remains unknown whether the neuropeptide changes are just a final outcome or really an important cellular event. Convincing evidence is lacking to address this question.

2) The author claimed that opioid (such as Pdyn and Penk) expression is not detected. I feel the authors tend to say it is a surprising observation. In my mind, it needs more caution, and it is better to confirm this negative observation by fate mapping analysis with Pdyn-cre (it is available). Fate mapping analysis is more solid to make a negative conclusion especially when we do not know the exact gene expression patterns. The samples should be analyzed at adult ages.

Reviewer #3 (Recommendations for the authors):

I strongly support publication in eLife given the indicated revisions.

I would like to congratulate the authors on this great study! Below please find some suggestions that might be helpful to further improve this manuscript.

Abstract:

"through unclear mechanisms" is inappropriate here given the amount of work and evidence on candidate mechanisms. This needs to be changed/weakened.

"we discovered a neuropeptide-enriched LOC subtype" given that LOCs have generally been thought to use neuropeptides and the analysis, as is, focuses on RNAseq I found this statement very strong (too strong), if not backed by immunohistochemistry in the cochlea.

Introduction:

There are some issues with references, which I tried to help with below. In general, according to DORA it would be preferred to cite original publications rather than reviews. However, there are cases that are well served by reviews: such as outer hair cell function.

Yet, the review of van der Heijden and Vavakou actually questions the cycle-by-cycle amplification by outer hair cells (OHCs) based on their OCT recordings (Vavakou et al., 2021). Perhaps cite a review of Dallos or others on electromotility and reference van der Heijden and Vavakou in the kind of (but see..) manner. Moreover, Glowatzki and Fuchs 2001 is on inner hair cells rather than OHCs. Also, the choice of the two Oatman papers might not be so helpful. Finally, the phrase "the detection of signals in noise26-28 and attention29-32." might be reconsidered (e.g. "effects of attention"?).

"No specific role in hearing has been definitively linked to LOCs, although they appear 78 to protect the ear from noise damage through unknown mechanisms34,35." While the shortage of functional data certainly applies, this statement and the references are falling quite short. I think you should tone down and consider citing Ruel et al., 2004 as well as Darrow et al., 2006.

Results:

"we adopted a single-nucleus sequencing strategy" While the approach will be clear to most of us it might still help clarify that this means cell nuclei rather than brainstem nuclei.

"Cell types were also similar in the fraction of mitochondrial genes detected," maybe I am missing something, but since there is mitochondrial DNA and nuclear sequencing was applied this should probably specify "nuclear-encoded mitochondrial genes" or the like. Moreover, the link between similarity in n-mt genes and homogeneous cell health might not be so obvious without a reference.

"OCNs but not in other motor neuron" this and later reads like you say OCNs are motor neurons: perhaps clarify in the intro that OCNs and "motor neurons" have common progenitors?

"Na,K-ATPase Atp1a3, which is selectively expressed in MOCs in mature rats45".

As shown by McLean et al., 2009 and other studies, in the cochlea, ATP1a3 is primarily expressed in type I SGNs, so this statement needs to be reconsidered.

"CADPS2 is linked to exocytosis of dense-core vesicles47-49",

I suggest mentioning the synonymous name CAPS-2, which is more frequently used in synaptic neuroscience, and to put its function into the context of vesicle priming (e.g. " a member of the Munc13 and CAPS family of priming proteins") potentially also quoting relevant neuronal work such as Jockusch et al., 2007, Vogl et al., 2015, Nestvogel et al., 2020.

Figure 1L and M: I found it difficult to detect any green signal in the respective left panels (also applies to other FISH analyses).

"Consistent with their integration into fundamentally different circuitry, OCNs and FMNs also differ in the expression of molecules relevant to mature function (Figures S2F, I). In particular, only OCNs express Gad2, a gene involved in GABA synthesis, whereas only FMNs express the serotonin receptor gene Htr2c (Figure S2I)." This is only one of the instances where the question of what predictive power RNA analysis has for protein expression is relevant. There is nothing wrong with the statement in principle, but is seems overtly simple to conclude this.

Figure 2, legend: "gene Zfp804a is expressed in MOCs (G) but is absent from most LOCs (F)."

"Both MOCs and LOCs have the ability to respond to numerous neurotransmitters, including 172 GABA (K), Glycine (L), acetylcholine (M), and glutamate (O)." another instance where care should be taken not to overinterpret the data.

Figure 3

It would be helpful to quantitatively see correlations between CGRP, NPY, and Ucn expression levels.

Figure 4

Please provide more explanation for how the masking of synaptophysin immunofluorescence was done.

"Strikingly, some (shown in Figure 4A-A"), but not all (Figure S5F-I), of these terminal fibers contained Syp puncta in the ISB." What precisely did you find striking here?

Please introduce CALB2 (calretinin), and also introduce the possibility of LOC-IHC contacts (e.g. Liberman et al., 1990, Lauer et al., 2012), which could confound the on-CALB2 analysis.

Figure 5

It looks like the majority of fibers that were analyzed have a considerable amount of NPY-positive terminals. However, the LOC2 cluster size seems to be smaller compared to LOC1, both from RNA sequencing data and qualitative immunolabeling. Is it possible that it has to do with Ret expression in different LOC subtypes? Please comment on how you reconcile these findings.

The section in Figure 5P is quite dense and a bit cryptic. I generally appreciate the effort to quantify potential relationships as done here but then again worry about how conclusive this is.

Legend to Figure 5: "High CALB2 fibers correspond to low threshold Type Ia SGNs; low CALB2 fibers correspond to high threshold Type Ic SGNs." To my knowledge, the correspondence of the molecular profiles and functional properties has yet to be tested. Please rephrase.

"….directly contacted the CALB2 surface ("On CALB2" in magenta) or not ("Off CALB2" in green).."

"…oriented with cochlear base to.."

The authors claim that there is no distinct pattern of NPY positive terminal connectivity along the pillar-modiolar axis. It seems that this statement is based on qualitative observations and the percentages of NPY+ terminals occurring on Syp+ puncta that are on or off CALB2. The authors could support this claim further, for instance, by a plot showing how many NPY+ Syp puncta occur on the pillar vs modiolar side.

Physiological characterization:

If analysis focused on the medial wing of LSO: how would you have a chance to pick up LOC1 given the data provided earlier in the manuscript?

When/how was the GFP signal recorded: prior to break-in (avoid loss via diffusion into the pipette?)?

Discussion:

"Molecular LOC subtypes are morphologically and physiologically diverse" I don't think the claim of physiological diversity is sufficiently supported by the data. Consider removing it from the section title and mark as important speculation and objective for future studies.

Reviewer #2 (Recommendations for the authors):

Frank et al. in this study thoroughly characterized the molecular, physiological, and morphological properties of the brainstem olivocochlear neurons (OCNs). First, by using the ChATCre+; Rosa26-Sun1-GFP/+ and 10x single cell RNA-Seq platform, they identified two Gata3+ cell clusters that are defined as OCNs: one is Ucn+ LOC, and the other is Atp1a3+ MOC. By further comparing the gene profiles between OCNs and the lineage-related FMNs, several previously unknown OCN markers such as Cadps2 are revealed. Furthermore, they also noticed that OCNs have high internal heterogeneity and they further found LOC and MOC-enriched genes, including but not limited to Col4a4 and Zfp804a, respectively. More importantly, two LOC subtypes, LOC-1 and LOC-2, are identified, and especially the LOC-2 sub-cluster cells, which are located in the medial wing of the LSO, can enrich many neuropeptide genes, CGRP, Calcb, NPY, Ucn and so on. The NPY expression in LOC-2 is further validated by the well-designed genetic fate mapping approach, which is convincing. Finally, the author showed that the neuropeptide expression patterns did not significantly change between wild-type and Tmc1/2 knockout mice at the pre-hearing ages, but either the neuropeptide expression levels or distribution patterns did change manifestly after hearing onset. Besides, they also showed that noise exposure can alter the neuropeptide patterns.

In general, the paper is clearly written and the data are well presented. The logic is straightforward, and in most cases, the data are compelling to support the conclusions.

Thank you for the kind comments.

I only have 2 main comments below:

1) In my point of view, it should be a resource paper rather than a research article because the current version did not connect any specific marker pattern with the biological properties. For example, the Npy identity does not appear to be predictive of LOC morphology. It remains unknown whether the neuropeptide changes are just a final outcome or really an important cellular event. Convincing evidence is lacking to address this question.

We agree with the reviewer that the significance of NPY expression remains to be determined. However, the research finding of note is the discovery of a population of LOCs marked by NPY – these NPY-positive LOCs are molecularly distinct and anatomically segregated, but they do not exhibit stereotyped morphologies or patterns of connectivity. We provide multiple types of evidence supporting this conclusion, including single nucleus RNA-seq data, immunohistochemistry, genetic fate mapping, and 3D reconstructions of individual axonal arbors in the cochlea. Moreover, a growing body of literature suggests that cellular subtype identity is rarely predictive of cellular morphology, as described in the Discussion (see e.g. Kim, …, Anderson 2019; Peng, …, Zeng 2021; general lack of correlation between transcriptional subtypes and morphological characteristics reviewed in Zeng 2022, Cell). Thus, we feel that the lack of correlation between NPY expression and cellular morphology is a biologically relevant observation that is best shared in the form of a research article. To make this point more clearly, we expanded the discussion of LOC heterogeneity (pg. 11, lines 504-514).

With respect to the role of NPY itself, we feel that the data do indeed provide compelling evidence that NPY (and other peptides) are induced by acoustic trauma, a finding that impacts how we think about LOC function more broadly, as presented in the Discussion (p. 20-21, “Functional consequences of dynamic neuropeptide expression”). Our work on neuropeptide induction is consistent with work in other peripheral systems, including neurons innervating the skin (Hoeffel, …, Ugolini 2021) and gut (Yang, …, Chiu 2022; see Discussion). Given the consistency of these phenotypes, the effect size (~4-5x increase in Ucn- or NPY-expressing LOCs, p<0.001), and agreement with the available literature in other sensory systems, it is reasonable to conclude that peptide induction in olivocochlear neurons is a genuine phenomenon that could play a similar role to that seen elsewhere in the body. Investigating the impact of this peptide induction is an important goal for the future, but is beyond the scope of the current study. This point has been added to the Discussion (pg. 13, lines 589-591).

2) The author claimed that opioid (such as Pdyn and Penk) expression is not detected. I feel the authors tend to say it is a surprising observation. In my mind, it needs more caution, and it is better to confirm this negative observation by fate mapping analysis with Pdyn-cre (it is available). Fate mapping analysis is more solid to make a negative conclusion especially when we do not know the exact gene expression patterns. The samples should be analyzed at adult ages.

The reviewer is correct that a negative result detecting Pdyn and Penk transcription is not conclusive evidence that LOC neurons never use these peptides. Although we include images of Pdyn- and Penk-expressing positive control cells, we did not definitively image the entire LOC across sections, so we cannot rule out the possibility that we may have missed expression in a subset of LOCs. We have adjusted our language in the text accordingly (pg. 5, lines 201-202). We agree that fate mapping using Pdyn-Cre could be interesting, but these experiments would not address the central question of whether LOC neurons signal with opioid peptides in adult animals, as there are several circuits where neurons have a history of neurotransmitter expression that is not retained in adults, including cells in the auditory system (reviewed in Spitzer 2017).

Reviewer #3 (Recommendations for the authors):

I strongly support publication in eLife given the indicated revisions.

I would like to congratulate the authors on this great study! Below please find some suggestions that might be helpful to further improve this manuscript.

Abstract:

"through unclear mechanisms" is inappropriate here given the amount of work and evidence on candidate mechanisms. This needs to be changed/weakened.

We agree and have removed the phrase accordingly (pg. 1).

"we discovered a neuropeptide-enriched LOC subtype" given that LOCs have generally been thought to use neuropeptides and the analysis, as is, focuses on RNAseq I found this statement very strong (too strong), if not backed by immunohistochemistry in the cochlea.

We appreciate the reviewer’s caution in drawing strong conclusions about the discovery of an LOC subtype. Examining expression of proteins other than NPY (Figure 5 and its supplement) or CGRP (e.g. Wu et al. 2017) in the cochlea would be an interesting and worthwhile set of experiments that could inform our understanding of how peptide release might influence cochlear function. However, we feel that our current approach is better suited to support the claim we make here, namely that neuropeptide-enriched LOCs exist. We focused on the cell bodies of LOCs in the brainstem as that is the only location in which we can reliably distinguish individual cells, given the extensive branching of their axons in the ear. The brainstem is therefore a more reliable place to look for variability in peptide expression between individual LOC neurons than the cochlea. In addition, we based our conclusion that there is a neuropeptide-enriched LOC subtype not only on RNAseq (Figure 3), but also on quantitative immunohistochemistry (Figure 3, 7, 7—figure supplement 1, and 8), in situ hybridization (Figure 3), fate mapping (Figure 3), and expression of promoter-based fluorescent reporter lines for NPY and CGRP (Figure 3, 6). All of these methods yield consistent results, indicating that bias in peptide expression arises in a stereotyped manner at the level of promoters, transcripts, and protein. Finally, although LOCs have indeed long been thought to signal through peptides, earlier work has not thoroughly examined the possibility of distinct cohorts of LOCs with varying peptide expression. Nonetheless, prior work in other species does recapitulate our finding of a medial bias in peptide expression across the LSO at the level of protein expression (Kaiser et al. 2011, Vetter et al. 1991). Thus, although additional immunohistochemistry in the cochlea would certainly be of interest, those results would not affect our central conclusion of variability in peptide expression across individual LOCs. We have added a sentence to the Discussion to better explain the evidence for a neuropeptide-enriched LOC subtype (pg. 11, lines 504-514).

Introduction:

There are some issues with references, which I tried to help with below. In general, according to DORA it would be preferred to cite original publications rather than reviews. However, there are cases that are well served by reviews: such as outer hair cell function.

Yet, the review of van der Heijden and Vavakou actually questions the cycle-by-cycle amplification by outer hair cells (OHCs) based on their OCT recordings (Vavakou et al., 2021). Perhaps cite a review of Dallos or others on electromotility and reference van der Heijden and Vavakou in the kind of (but see..) manner. Moreover, Glowatzki and Fuchs 2001 is on inner hair cells rather than OHCs. Also, the choice of the two Oatman papers might not be so helpful. Finally, the phrase "the detection of signals in noise26-28 and attention29-32." might be reconsidered (e.g. "effects of attention"?).

Thank you for your careful attention to detail in checking these references! We have added an additional review on outer hair cell electromotility (pg. 2, line 53) and replaced the Glowatzki and Fuchs 2001 reference with papers discussing the OHC-MOC synapse (Blanchet et al. 1996 and Dallos et al. 1997; pg. 2, line 54). We also revised the phrasing in line 56-57. Changes to the references are highlighted in yellow.

"No specific role in hearing has been definitively linked to LOCs, although they appear 78 to protect the ear from noise damage through unknown mechanisms34,35." While the shortage of functional data certainly applies, this statement and the references are falling quite short. I think you should tone down and consider citing Ruel et al., 2004 as well as Darrow et al., 2006.

Thank you for the suggestion. We have rephrased this sentence and added references to Ruel et al. 2001 and Wu et al. 2020 (concerning the role of dopamine in tuning SGN excitability) and Darrow et al. 2006, Larsen and Liberman 2010, and Irving et al. 2011 (concerning the possibility of OCNs in sound localization circuitry). We also added a reference to an additional review article (Fuente et al. 2015) describing possible mechanisms mediating LOC protection from acoustic trauma for readers interested in exploring the topic beyond the scope of this paper (pg. 2, lines 69-74). Changes to the references are highlighted in yellow.

Results:

"we adopted a single-nucleus sequencing strategy" While the approach will be clear to most of us it might still help clarify that this means cell nuclei rather than brainstem nuclei.

We have rewritten the text for clarity (pg. 2, lines 87-88).

"Cell types were also similar in the fraction of mitochondrial genes detected," maybe I am missing something, but since there is mitochondrial DNA and nuclear sequencing was applied this should probably specify "nuclear-encoded mitochondrial genes" or the like. Moreover, the link between similarity in n-mt genes and homogeneous cell health might not be so obvious without a reference.

We appreciate the reviewer’s careful attention to important analytic details. The fraction of mitochondrial genes detected is a standard metric to assess the quality of single-cell or single-nucleus sequencing data. This analysis specifically includes genes mapping to the mitochondrial genome, not nuclear-encoded mitochondrial genes. As described in the Methods, the presence of genes mapping to the mitochondrial genome likely indicates contaminants from outside the nucleus or other issues with sample quality. We have revised this phrasing for clarity and added appropriate references in both the Results and the Methods (pg. 3, line 108-109; pg. 15, lines 675-679). Changes to the references are highlighted in yellow.

"OCNs but not in other motor neuron" this and later reads like you say OCNs are motor neurons: perhaps clarify in the intro that OCNs and "motor neurons" have common progenitors?

This is a helpful suggestion, and we have added this information to the section of the Introduction that discussed the motor neuron origin of OCNs (pg. 1, lines 39-41).

"Na,K-ATPase Atp1a3, which is selectively expressed in MOCs in mature rats45".

As shown by McLean et al., 2009 and other studies, in the cochlea, ATP1a3 is primarily expressed in type I SGNs, so this statement needs to be reconsidered.

We agree that the use of the word “selectively” requires more explanation. We were referring specifically to preferential ATP1a3 expression in OCN cell bodies, so additional expression in SGNs or other cell types does not affect its utility as a distinguishing marker between MOCs and LOCs. We have updated this phrasing for clarity (pg. 3, line 114).

"CADPS2 is linked to exocytosis of dense-core vesicles47-49",

I suggest mentioning the synonymous name CAPS-2, which is more frequently used in synaptic neuroscience, and to put its function into the context of vesicle priming (e.g. " a member of the Munc13 and CAPS family of priming proteins") potentially also quoting relevant neuronal work such as Jockusch et al., 2007, Vogl et al., 2015, Nestvogel et al., 2020.

This is a great point. We adopted this phrasing and added additional references related to CAPS function (pg. 3, lines 125-127). Changes to the references are highlighted in yellow.

Figure 1L and M: I found it difficult to detect any green signal in the respective left panels (also applies to other FISH analyses).

We agree that it can be difficult to interpret the FISH analyses. We have tried our best to choose colors that are visible in the overlay. However, given the size of the puncta, there are limits to how much we can do to make the green points visible. For this reason, we also show an inverted image of the single-channel data with the mRNA puncta labeled in black. We feel that this is the best way to represent the data. We have also made slight adjustments to the overlay images shown in Figures 1L, 2C, and 2G to make the overlapping signals easier to see. There is no green signal visible in Figure 1M because the Cadps2 transcript is not expressed in the facial nucleus, which is more clearly seen in the single-channel panel on the right of 1M.

"Consistent with their integration into fundamentally different circuitry, OCNs and FMNs also differ in the expression of molecules relevant to mature function (Figures S2F, I). In particular, only OCNs express Gad2, a gene involved in GABA synthesis, whereas only FMNs express the serotonin receptor gene Htr2c (Figure S2I)." This is only one of the instances where the question of what predictive power RNA analysis has for protein expression is relevant. There is nothing wrong with the statement in principle, but is seems overtly simple to conclude this.

Figure 2, legend: "gene Zfp804a is expressed in MOCs (G) but is absent from most LOCs (F)."

"Both MOCs and LOCs have the ability to respond to numerous neurotransmitters, including 172 GABA (K), Glycine (L), acetylcholine (M), and glutamate (O)." another instance where care should be taken not to overinterpret the data.

We did not mean to imply that RNA analyses perfectly predict protein expression or function. We have made a number of changes throughout the text to better indicate that we are referring to genes. We also added a sentence to the Discussion indicating that additional confirmation is needed (pg. 12, lines 533-535).

Figure 3

It would be helpful to quantitatively see correlations between CGRP, NPY, and Ucn expression levels.

We agree that this would be a useful analysis. However, given the limitations of single-cell sequencing data (and in particular the v2 10x kit we used for collecting the adult nuclei in this study), we do not think these results are reliable enough to include here. Specifically, droplet-based single-cell sequencing approaches are known to have issues with dropouts and false zero counts due to stochasticity in transcript detection, an effect that is more pronounced for kits with older chemistries, such as the 10x v2 kit (see e.g. Hicks, …, Irazarry 2017). Thus, although Ucn and NPY expression in adult LOCs is tightly correlated with an adjusted R² of 0.91 for cells in which both Ucn and NPY were detected, this analysis only includes 5 cells. Droplet-based sequencing platforms (like 10x) are also limited in their ability to detect the dynamic range of transcript expression, especially for lowly expressed transcripts, like peptides (e.g. Wang, … Zhang 2021), making correlation analyses based on transcript expression levels even more unreliable. We therefore used an orthogonal method to qualitatively examine co-expression of peptides at the protein and promoter levels, as shown in Figure 3N, O and Supplementary File 1. Because commercially available antibodies for both NPY and Ucn are raised in rabbit, we were unable to quantitatively compare NPY and Ucn protein expression in the same cells.

Figure 4

Please provide more explanation for how the masking of synaptophysin immunofluorescence was done.

Additional details have been added to the Methods (pg. 18, lines 804-808).

"Strikingly, some (shown in Figure 4A-A"), but not all (Figure S5F-I), of these terminal fibers contained Syp puncta in the ISB." What precisely did you find striking here?

What stood out to us is that there do not appear to be two types of MOCs – those that form putative synapses with Type I SGN peripheral fibers and those that do not. Instead, individual MOC axons can extend some terminal branches that form synaptophysin puncta in the ISB while other terminal branches from the same axon bypass the ISB and exclusively terminate on OHCs. We edited the text to better communicate what is interesting about this observation (pg. 6, lines 250-252).

Please introduce CALB2 (calretinin), and also introduce the possibility of LOC-IHC contacts (e.g. Liberman et al., 1990, Lauer et al., 2012), which could confound the on-CALB2 analysis.

Although CALB2 is a reliable way to distinguish Ia vs Ib/c SGN subtypes (Shrestha…Goodrich, 2018), we agree that its expression in IHCs can be problematic. To make this point more clearly, we have re-written the text about why CALB2 was used and to acknowledge that we cannot rule out the possibility of LOC-IHC contacts. If such contacts exist, though, they are likely to be a minority, as most Syp puncta were well away from the IHCs. In the revised text, CALB2 is still introduced as a marker for both SGNs and IHCs when we describe the overall approach on pg. 5 (lines 224-227) and p. 6 (lines 262-268), and then we expanded the discussion of possible caveats on pg. 7, lines 293-301.

Figure 5

It looks like the majority of fibers that were analyzed have a considerable amount of NPY-positive terminals. However, the LOC2 cluster size seems to be smaller compared to LOC1, both from RNA sequencing data and qualitative immunolabeling. Is it possible that it has to do with Ret expression in different LOC subtypes? Please comment on how you reconcile these findings.

The reviewer raises an interesting and important point. Indeed, LOC axons with high NPY content are likely over-represented in our morphological analyses. It is possible that our tamoxifen dosing scheme may have enriched for LOC neurons with medium-to-high NPY load in their axons. In adult animals, Ret expression does not appear to be higher in LOC2s than LOC1s. (Ret was detected in 24% of LOC1s and 20% of LOC2s, with an average normalized expression of 2.1 and 1.9 in Ret+ LOC1s and LOC2s, respectively. Note that the low percentage of cells with Ret detected likely reflects limitations of droplet-based single-cell sequencing methods, and does not imply that only ~20% of mature LOCs are producing Ret transcripts). When we give high doses of tamoxifen to mature (~P20–P28) Ret^CreER;Igs7^GFP mice, MOCs and LOCs are labeled extensively throughout the base-to-apex extent of the cochlea. However, we found that we were better able to reliably titrate sparse labeling in the Ret^CreER line with embryonic tamoxifen doses than with postnatal or adult administration. It is possible that administering tamoxifen during late embryogenesis captured a developmentally distinct subpopulation of LOCs. Because many components of the auditory system develop from high-to-low frequencies, our embryonic injections could have captured an LOC population biased to the medial LSO, which would preferentially target NPY-high LOCs.

However, we feel confident in our experimental approach given the variety of LOC axonal morphologies we found as well as the large number of LOC axons with low or no NPY load. Our goal was to determine whether fibers with high NPY expression exhibited stereotyped morphologies in the cochlea, and the range of morphologies we identified indicates that they do not. Similarly, we found that axons with 0% NPY/Syp+ puncta do not exhibit any stereotyped properties, reinforcing our conclusion that peptide expression does not dictate axonal morphology. We have added a summary of our reasoning as well as potential caveats to the text (pg. 11, lines 504-514).

The section in Figure 5P is quite dense and a bit cryptic. I generally appreciate the effort to quantify potential relationships as done here but then again worry about how conclusive this is.

We agree that this section contains a lot of information that can be hard to digest. The point we are trying to make is that peripheral axon morphological features do not correlate with peptidergic content, which requires walking through a wide variety of parameters. We have rewritten this entire section to make the main point clear – namely, that LOCs are a diverse population of neurons containing two molecularly distinct subtypes, but whose innervation patterns in the cochlea suggest that they are poised to provide widespread and broad feedback to the cochlea. These changes are throughout the section entitled “NPY-positive LOC axons exhibit highly variable patterns of connectivity”, starting on p. 6.

Legend to Figure 5: "High CALB2 fibers correspond to low threshold Type Ia SGNs; low CALB2 fibers correspond to high threshold Type Ic SGNs." To my knowledge, the correspondence of the molecular profiles and functional properties has yet to be tested. Please rephrase.

We apologize for the confusing use of the word “correspond”, which is appropriate for the molecular subtypes but not for their functional properties. We have removed this sentence from the figure caption (pg. 30-31).

"….directly contacted the CALB2 surface ("On CALB2" in magenta) or not ("Off CALB2" in green).."

"…oriented with cochlear base to.."

We have made both of these wording changes to the figure legend (pg. 30-31).

The authors claim that there is no distinct pattern of NPY positive terminal connectivity along the pillar-modiolar axis. It seems that this statement is based on qualitative observations and the percentages of NPY+ terminals occurring on Syp+ puncta that are on or off CALB2. The authors could support this claim further, for instance, by a plot showing how many NPY+ Syp puncta occur on the pillar vs modiolar side.

As the reviewer suggests, we used the On and Off CALB2 contact as a proxy for pillar and modiolar position of Syp puncta (with or without NPY). Although correlative, this is a reasonable readout of modiolar-pillar position that adequately serves the purpose of our analysis (i.e., to determine if there are clear biases in the localization of peptidergic or non-peptidergic LOC Syp puncta). The stereotypy of CALB2 expression gradients among the SGN peripheral processes along the modiolar-pillar axis of the base of the IHC have been shown previously by multiple groups (e.g. Shrestha…Goodrich, 2018; Sun…Mueller, 2018) and, as illustrated in the orthogonal view in Figure 5E, we found the alignment between modiolar/pillar position and On and Off CALB2 to be quite reliable. While we agree it would be better to directly assess position, it would be a considerable amount of additional work to code the modiolar/pillar position of each Syp puncta in Imaris. Given the lack of qualitative evidence that there are ever any CALB2-on Syp+ puncta on the modiolar side, we feel it would not substantially add meaningful insight to our conclusions about the lack of an anatomical bias in the position of NPY+ Syp+ puncta. To make this logic clear, we have rewritten the text in this section (pg. 6, lines 262-268, 270-271; pg. 7, lines 293-301).

Physiological characterization:

If analysis focused on the medial wing of LSO: how would you have a chance to pick up LOC1 given the data provided earlier in the manuscript?

Although peptide-enriched LOC2s are restricted to the medial wing, cells with variable levels of CGRP-GFP are also present in the medial half of the LSO (Figure 6A’), consistent with the presence of both NPY+ and NPY- LOCs in this region (Figure S5C). We therefore chose to focus our analysis on cells in the mid-central LSO so that we could capture a range of CGRP intensities while limiting potential confounds associated with other types of variability along the length of the LSO (especially well-documented tonotopic variability). We have updated the text to make this logic clearer (pg. 8, lines 346-350).

When/how was the GFP signal recorded: prior to break-in (avoid loss via diffusion into the pipette?)?

GFP signal was measured prior to breaking into the cell. We have updated this detail in the Methods (pg. 20, lines 899-901).

Discussion:

"Molecular LOC subtypes are morphologically and physiologically diverse" I don't think the claim of physiological diversity is sufficiently supported by the data. Consider removing it from the section title and mark as important speculation and objective for future studies.

Although we did not map LOC subtypes to specific physiological properties, we did find a range of physiological properties in the LOC neurons we recorded from (Figure 6). However, we agree that the initial heading implies that there are physiological subtypes of LOC neurons that correspond to molecular subtypes, which is not reflected in our data. We have updated this header accordingly (pg. 11, line 469).

Author details

1. Michelle M Frank

   Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6613-8251

2. Austen A Sitko

   Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7601-6143

3. Kirupa Suthakar

   Section on Neuronal Circuitry, National Institute on Deafness and Other Communication Disorders, Bethesda, United States

   Contribution

   Investigation, Formal analysis, Writing – review and editing

   Competing interests

   No competing interests declared

4. Lester Torres Cadenas

   Section on Neuronal Circuitry, National Institute on Deafness and Other Communication Disorders, Bethesda, United States

   Contribution

   Investigation, Formal analysis, Writing – review and editing

   Competing interests

   No competing interests declared

5. Mackenzie Hunt

   Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   Investigation, Writing – review and editing, Formal analysis

   Competing interests

   No competing interests declared

6. Mary Caroline Yuk

   Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   Formal analysis, Writing – review and editing

   Competing interests

   No competing interests declared

7. Catherine JC Weisz

   Section on Neuronal Circuitry, National Institute on Deafness and Other Communication Disorders, Bethesda, United States

   Contribution

   Formal analysis, Supervision, Writing – original draft, Writing – review and editing, Funding acquisition

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2595-835X

8. Lisa V Goodrich

   Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   Lisa_Goodrich@hms.harvard.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3331-8600

• 828

  Page views

• 160

  Downloads

• 3

  Citations

Article citation count generated by polling the highest count across the following sources: PubMed Central, Crossref, Scopus.