A novel class of inferior colliculus principal neurons labeled in vasoactive intestinal peptide-Cre mice — TEI-tagged view

GoyerpersName, SilveirapersName, GeorgepersName, BeebepersName, EdelbrockpersName, MalinskipersName, SchofieldpersName, RobertspersName “A novel class of inferior colliculus principal neurons labeled in vasoactive intestinal peptide-Cre mice.” eLifetitle·j.

Object of study — whether a molecular markerterm can carve a classterm out of a structure whose neurons HAVE PROVEN DIFFICULT TO DELINEATE INTO DISTINCT CLASSESterm

Research question — how to identify neuron classesterm where every existing criterion has failed - PROBING NEURAL CIRCUITS REQUIRES THE ABILITY TO IDENTIFY AND MANIPULATE SPECIFIC CLASSESterm OF NEURONS, BUT IC NEURONS HAVE PROVEN DIFFICULT TO DELINEATE INTO DISTINCT CLASSESterm

Contribution — BY COMBINING MOLECULAR, MORPHOLOGICAL, AND PHYSIOLOGICAL ANALYSES, WE IDENTIFY VASOACTIVE INTESTINAL PEPTIDE (VIPterm) NEURONS AS A NOVEL CLASSterm OF IC PRINCIPAL NEURONS - GLUTAMATERGICterm STELLATEterm CELLS WITH SUSTAINED FIRING PATTERNSterm·def

Full text — tagged

A novel classterm of inferior colliculus principal neurons labeled in vasoactive intestinal peptide-Cre mice eLife 8 :e43770.

https://doi.org/10.7554/eLife.43770 Copy to clipboard Download BibTeX Download .RIS Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Article and author information Metrics Located in the midbrain, the inferior colliculus (IC) is the hub of the central auditory system. Although the IC plays important roles in speech processing, sound localization, and other auditory computations, the organization of the IC microcircuitry remains largely unknown. Using a multifaceted approach in mice, we have identified vasoactive intestinal peptide (VIPterm) neurons as a novel classterm of IC principal neurons. VIP neuronsterm·def are glutamatergicterm stellateterm cells with sustained firing patternsterm·def. Their extensive axons project to long-range targets including the auditory thalamus, auditory brainstem, superior colliculus, and periaqueductal gray. Using optogenetic circuit mapping, we found that VIP neuronsterm·def integrate input from the contralateral IC and the dorsal cochlear nucleus. The dorsal cochlear nucleus also drove feedforward inhibition to VIP neuronsterm·def, indicating that inhibitory circuits within the IC shape the temporal integration of ascending inputs. Thus, VIP neuronsterm·def are well-positioned to influence auditory computations in a number of brain regions.

https://doi.org/10.7554/eLife.43770.001 Our brains help us make sense of our surroundings. Our sense organs, for example the ears, receive signals from the environment, which are passed on to specialized parts of the brain, called nuclei. There, neurons process the information and send impulses to other areas in the brain, which eventually help us decide how to respond.

A nucleus called the inferior colliculus is one of the major hearing centres in the mammalian brain. In humans, it is vital for recognizing speech and pinpointing the location of sounds. It contains several different types of neurons, and links with many other areas of the brain.

While the inferior colliculus is crucial for our sense of hearing, little is known about the specific properties of the neurons within it. In particular, these cells are difficult to divide into well-defined groups. Even less is understood about the precise nature of the connections between these neurons, which likely underpin the computational power of the inferior colliculus. GoyerpersName et al. therefore set out to identify a specific classterm of neurons in this region and map the circuits they formed.

Experiments were conducted on genetically modified mice whose neurons were only ‘glowing’ if they had a gene called VIPterm switched on. Detailed examination of the shape of the ‘VIPterm cells’, as well as their chemical and electrical properties, confirmed that they were indeed a distinct classterm of neurons.

Another set of experiments relied on a method that uses light to control the activity of brain cells. This showed that VIPterm cells received signals both from other neurons in the inferior colliculus, and from another hearing center in the brain. In turn, VIPterm cells sent signals over long distances to many other parts of the brain that handle sound signals. This suggests that VIP neuronsterm·def have a wide-ranging influence on our brains’ ability to process sound.

The work by GoyerpersName et al. has, for the first time, reliably identified specific circuits in a brain region essential for our sense of hearing. By knowing more about how the brain’s hearing centers are connected to each other, it may become possible to understand their roles in hearing loss. In this effort, the inferior colliculus may become a target for treatments for patients with hearing difficulties.

https://doi.org/10.7554/eLife.43770.002 The inferior colliculus (IC) is the hub of the central auditory pathway. Nearly all ascending output from the lower auditory brainstem and a large descending projection from the auditory cortex converge in the IC ( Adams, 1979 ; Glendenning and Masterton, 1983 ; Oliver, 1987 ; Oliver, 1984 ; Winer et al., 1998 ). In turn, the IC provides the main auditory input to the thalamocortical system ( Calford and Aitkin, 1983 ). Neurons in the IC exhibit selective responses to the spectral and temporal content of sounds and perform computations important for sound localization and the identification of speech and other communication sounds ( Felix et al., 2018 ; Winer and Schreiner, 2005 ). Despite these critical functions, we have limited knowledge about the organization and function of neural circuits in the IC. This is because probing neural circuits requires the ability to identify and manipulate specific classesterm of neurons, but IC neurons have proven difficult to delineate into distinct classesterm.

Anatomical studies have shown that IC neurons have disc-shapedterm or stellateterm morphologies ( Malmierca et al., 1993 ; Meininger et al., 1986 ; Oliver and Morest, 1984 ). Disc-shapedterm neurons maintain their dendritic arbors within isofrequency lamina and make up the majority of neurons in the tonotopically organized central nucleus of the IC (ICcterm). Stellateterm neurons in the ICcterm extend their dendritic arbors across laminae and are therefore thought to integrate information across sound frequencies ( Oliver et al., 1991 ). Both disc-shapedterm and stellateterm cells can be glutamatergicterm or GABAergicterm, an indication that each of these morphological groups consists of at least two neuron types ( Oliver et al., 1994 ). Based on soma size and extracellular markersterm, IC GABAergicterm neurons have been divided into four classesterm ( BeebepersName et al., 2016 ). Among these, ‘large GABAergicterm’ neurons are the one consistently identified neuron type in the IC ( Geis and Borst, 2013 ; Ito et al., 2015 ; Ito et al., 2009 ; Ito and Oliver, 2012 ). However, there are currently no known molecular markersterm specific for large GABAergicterm neurons ( SchofieldpersName and BeebepersName, 2019 ).

Defining IC neuron types based on physiology has also proven difficult. IC neurons exhibit diverse responses to tones, but a comprehensive study showed that these responses form a continuumterm and cannot be used on their own to define functionally significant groups of neurons ( Palmer et al., 2013 ). In addition, disc-shapedterm neurons could not be divided into distinct groups by matching their morphologyterm with their in vivo physiology ( Wallace et al., 2012 ). Similarly, GABAergicterm and glutamatergicterm IC neurons exhibit overlapping and equally diverse responses to sounds ( Ono et al., 2017 ). In vitro recordings have shown that IC neurons exhibit diverse firing patternsterm, but these firing patternsterm do not correlate with neuronal morphologyterm or neurotransmitter phenotype ( Ono et al., 2005 ; Peruzzi et al., 2000 ; Reetz and Ehret, 1999 ; Sivaramakrishnan and Oliver, 2001 ).

In many brain regions, a multidimensional analysis that includes molecular markersterm has proven key to identifying neuron classesterm ( Ascoli et al., 2008 ; Tremblay et al., 2016 ; Zeng and Sanes, 2017 ). Here, by combining molecular, morphological, and physiological analyses, we identify vasoactive intestinal peptide (VIPterm) neurons as a novel classterm of IC principal neurons. Our results show that VIP neuronsterm·def are glutamatergicterm stellateterm neurons and represent approximately 20% of the stellateterm neurons in the ICcterm. VIP neuronsterm·def are labeled in the VIP-IRES-Cre mouse line and are present in the major subdivisions of the IC, with a higher prevalence in caudal regions of the ICcterm, where most ascending input originates from monaural brainstem nuclei. Using viral tract tracing, we found that VIP neuronsterm·def project to multiple auditory and non-auditory areas, demonstrating that a single neuron classterm can participate in most of the major projection pathways out of the IC. Using Channelrhodopsin-assisted circuit mapping (CRACM)term·def, we found that VIP neuronsterm·def integrate input from the contralateral IC and the auditory brainstem. Input from the auditory brainstem also drove local, feedforward inhibition onto VIP neuronsterm·def. Thus, our data reveal a novel circuit motif that may control the temporal summation of ascending input to the IC. Together, these results represent a critical step toward determining how defined neural circuits in the IC support sound processing.

By crossing VIP-IRES-Cre mice with Ai14 reporter mice, we obtained mice in which VIPterm + neurons expressed the fluorescent protein tdTomato. Distribution of VIPterm + neurons throughout the brain matched the description provided by Taniguchi et al. (2011) . There were very few or no VIP-expressing neurons in auditory centers outside of the IC (including cochlear nucleus, superior olivary complex, nuclei of the lateral lemniscus, nucleus of the brachium of the IC and medial geniculate nucleus), with the exception of auditory cortex, where sparse neurons matching descriptions of VIP-expressing interneuronsterm were labeled. The present report focuses on the IC, which contained many VIP-expressing neurons. Figure 1 shows the distribution of VIPterm + neurons (magenta) in transverse sections through the IC. Labeled neurons were present throughout most of the rostro-caudal extent of the IC, including in the ICcterm, ICd, and IClc, but were most numerous in the caudal regions. Labeled neurons were rare or absent in the IC rostral pole and intercollicular tegmentum.

Figure 1 Download asset Open asset VIP neuronsterm·def are distributed across multiple subdivisions of the IC.

Photomicrographs of transverse sections through the IC ranging from caudal (left-most) to rostral (right-most). VIP-expressing cells (labeled with tdTomato) are shown in magenta, and GAD67 staining is shown in green to show the border of the IC. VIP-expressing cells are present in multiple subdivisions of the IC, but are most prominent in caudal and dorsal parts of the IC. Scale = 1 mm. Cb (cerebellum), ICcterm, ICd, IClc (central nucleus, dorsal cortex and lateral cortex of the inferior colliculus), ICt (intercollicular tegmentum), IV (fourth ventricle), PAG (periaqueductal gray).

https://doi.org/10.7554/eLife.43770.003 Previous studies have shown that IC neurons are either glutamatergicterm or GABAergicterm ( Merchán et al., 2005 ; Oliver et al., 1994 ). To investigate the neurotransmitter content of VIP neuronsterm·def, we performed immunohistochemical staining against GAD67, a markerterm for GABAergicterm neurons, in brain slices from three VIP-IRES-Cre x Ai14 animals, aged P58 ( Figure 2A ). We then counted tdTomato + neurons and GAD67-labeled cell bodies in one caudal and one rostral IC slice per animal. Because there were no regional differences in GAD67 staining among VIP neuronsterm·def, neurons located in the ICcterm, ICd, and IClc were combined for this analysis. Across 793 tdTomato + neurons, only 10 neurons co-labeled with GAD67 (1.3%, Table 1 ). We suspect the 10 tdTomato + neurons that stained for GAD67 were non-specifically labeled or represent rare cases of tdTomato expression in non-VIP neurons. These data suggest that VIP neuronsterm·def are a subgroup of glutamatergicterm neurons in the ICcterm, ICd, and IClc.

Table 1 VIP neuronsterm·def are glutamatergicterm.

Across three mice, an average of 1.3% of tdTomato + neurons were labeled with an antibody against GAD67.

https://doi.org/10.7554/eLife.43770.004 Animal Slice # # tdTomato + # GAD67 + # co-labeled % tdTomato + co-labeled P58 female, #1 1 (caudal) 210 184 3 1.4 2 (middle) 172 65 2 1.2 Total 382 249 5 1.3 P58 male 1 (caudal) 151 152 2 1.3 2 (middle) 46 212 2 4.3 Total 197 364 4 2.0 P58 female, #2 1 (caudal) 161 137 0 0.0 2 (middle) 53 187 1 1.9 Total 214 324 1 0.5 Grand total 793 937 10 1.3 Average across three mice (mean ± SD) 1.3 ± 0.8% Figure 2 Download asset Open asset VIP neuronsterm·def are glutamatergicterm and represent 3.5% of neurons in the ICcterm.

( A ) Confocal z-stack projections showing IC VIP neuronsterm·def (magenta, left), GAD67 staining (cyan, middle), and an overlay (right). White arrowheads mark VIP neuronsterm·def, yellow arrows GABAergicterm cell bodies. There was virtually no overlap between VIP neuronsterm·def and GABAergicterm neurons (right). ( B ) Confocal z-stack projections showing VIP neuronsterm·def (magenta, left), NeuN staining (cyan, middle), and an overlay (right). White arrowheads mark VIP neuronsterm·def labeled by NeuN. Scale bar applies to A and B.

https://doi.org/10.7554/eLife.43770.005 To determine the percentage of neurons in the ICcterm and IC shell (ICd plus IClc) that are VIP neuronsterm·def, we performed immunostaining with anti-NeuN, a neuron-selective antibody previously shown to label most or all neurons in the IC ( BeebepersName et al., 2016 ; Foster et al., 2014 ; Mellott et al., 2014 ) ( Figure 2B ), and anti-bNOS, a markerterm that differentiates the ICcterm from the IC shell regions ( Coote and Rees, 2008 ). Coronal IC sections from two VIP-IRES-Cres x Ai14 mice were stained with anti-NeuN and anti-bNOS. Three sections per mouse were analyzed: one caudal, one middle, and one rostral. To ensure unbiased counting of neurons, we applied the optical fractionator method, a design-based stereology approachterm·stat (see Materials and methods; West et al., 1991 ). Accordingly, we used systematic random sampling to collect confocal image stacks with a 63x objective at evenly spaced intervals from each IC section. Each image stack was inspected to determine the boundaries of the slice, and guard zones were set at the top and bottom of the slice to delineate a central, 15 µm-thick section of the slice for subsequent analysis. Within this 15 µm-thick region, we separately marked NeuN + neurons and tdTomato + neurons, then overlaid the NeuN and tdTomato images to determine the number of double-marked cells. We also collected tile scan images of each IC section analyzed using a 20x objective and used bNOS staining to determine the border separating the ICcterm from the IC shell. The 63x z-stacks were aligned to these tile scans, and counted neurons were assigned to the ICcterm or IC shell.

The results of the stereological analysis revealed that VIP neuronsterm·def represented a larger portion of neurons in the ICcterm (3.5 ± 1.0%) than in the IC shell (1.5 ± 0.2%; two-tailed t-testterm·stat, t (79) = 2.86, p = 0.005; Table 2 ). In addition, the prevalence of VIP neuronsterm·def was highest in caudal regions of the ICcterm and the IC shell and tended to decrease in a caudal to rostral gradient. This trend was significant when comparing the caudal ICcterm to the rostral ICcterm (one-way ANOVAterm·stat, F (2,37) = 7.27, p = 0.002, Tukey’s post hoc, p=0.001) and the caudal IC shell to the middle IC shell (one-way ANOVAterm·stat, F (2,38) = 3.88, p = 0.03, Tukey’s post hoc, p = 0.03). These results suggest that VIP neuronsterm·def are not evenly distributed along the rostral-caudal extent of the isofrequency lamina of the ICcterm and, according to the functional domain hypothesis, may be more likely to receive input from ascending afferents that preferentially target the caudal ICcterm ( Cant and Benson, 2006 ; Loftus et al., 2010 ; Oliver et al., 1997 ). The functional implications of this anatomical arrangement will be addressed in more detail in the Discussion. Overall, a combined count of VIP neuronsterm·def from the ICcterm and IC shell showed that VIP neuronsterm·def represent 2.3 ± 0.3% of the total population of neurons in the mouse IC (208 of 9304 neurons, n = 69 systematic random samples).

Table 2 VIP neuronsterm·def represent 3.5% of ICcterm neurons, 1.5% of IC shell neurons, and are present at a higher density in the caudal ICcterm and IC shell.

Table shows results from stereological analysis of the percentage of neurons (NeuN + ) in the ICcterm and IC shell (ICd + IClc) that express tdTomato in VIP-IRES-Cre x Ai14 mice. Values indicate mean ± SEM, (#tdTomato + neurons / #NeuN + neurons), and number of systematic random samples analyzed from each slice.

https://doi.org/10.7554/eLife.43770.006 ICcterm Coronal slice P54 male 1 P54 male 2 Per slice plane Grand average Caudal 3.1 ± 0.9% (12/503) five samples 8.4 ± 1.2% (26/338) four samples 5.8 ± 2.7% (38/841) Middle 2.4 ± 0.8% (20/741) eight samples 3.9 ± 0.9% (44/1173) eight samples 3.2 ± 0.7% (64/1914) Rostral 1.9 ± 0.6% (21/929) eight samples 1.2 ± 0.4% (12/1024) seven samples 1.5 ± 0.4% (33/1953) Per mouse 2.5 ± 0.3% (53/2173) 4.5 ± 2.1% (82/2535) 3.5 ± 1.0% (135/4708) IC shell Coronal slice P54 male 1 P54 male 2 Per slice plane Grand average Caudal 2.9 ± 0.8% (35/1092) 10 samples 1.9 ± 0.8% (9/615) five samples 2.4 ± 0.5% (44/1707) Middle 0.9 ± 0.6% (4/534) six samples 0.9 ± 0.6% (10/944) eight samples 0.9 ± 0.0% (14/1478) Rostral 1.2 ± 0.4% (10/842) eight samples 1.2 ± 0.6% (5/569) four samples 1.2 ± 0.0% (15/1411) Per mouse 1.6 ± 0.6% (49/2468) 1.3 ± 0.3% (24/2128) 1.5 ± 0.2% (73/4596) Table 2—source data 1 Percentages of VIP neuronsterm·def in ICcterm and IC shell.

https://doi.org/10.7554/eLife.43770.007 Download elife-43770-table2-data1-v2.xlsx Next, we investigated the firing patternterm and intrinsic physiologyterm·var of VIP neuronsterm·def by targeting whole cell patch clamp recordings to tdTomato + neurons in brain slices from VIP-IRES-Cre x Ai14 mice. For the majority of neurons, neuronal location relative to the IC subdivisions was determined post hoc, during retrieval of neuronal morphologyterm (see below and Materials and methods). Recordings made from the ICcterm, ICd, and IClc were lumped together for this experiment because there were no clear differences in VIP neuronterm·def physiology across these subdivisions of the IC. VIP neuronsterm·def had a resting membrane potential of −69.5 mV ± 4.4 mV (n = 216, corrected for liquid junction potential). In response to a current step protocol with hyperpolarizing and depolarizing current injections, VIP neuronsterm·def showed minimal to no voltage sag to hyperpolarizing current steps and a sustained firing patternterm·def of action potentials to depolarizing current steps ( Figure 3A 1 , A 2 ). Neurons were classified as sustained if their spike frequency adaptation ratioterm·def (SFAterm) was less than 2 ( Peruzzi et al., 2000 ). The SFAterm ratio was calculated by dividing the last interspike interval by the first for a depolarizing current step that elicited ~10 spikes. 90.3% (214 of 237) of patched VIP neuronsterm·def exhibited a sustained firing patternterm·def, 8.4% (20 of 237) showed an adapting firing patternterm (SFAterm ratio >= 2), and 1.3% (3 of 237) of VIP neuronsterm·def had a transient firing patternterm (firing stopped before the end of the current step).

Figure 3 Download asset Open asset VIP neuronsterm·def have sustained firing patternsterm·def and moderate membrane properties.

( A ) VIP neuronsterm·def exhibited a regular, sustained firing patternterm·def in response to depolarizing current steps, while hyperpolarizing current steps elicited minimal voltage sag (A 1 ). As the amplitude of depolarizing current steps was increased, VIP neuronsterm·def increased their firing rate while keeping their sustained firing patternterm·def (A 2 ). Example firing versus input (FI) curves from 15 VIP neuronsterm·def show that firing rate increased in a mostly linear fashion over a broad range of current step amplitudes (A 3 ). ( B ) Intrinsic physiologyterm·var of VIP neuronsterm·def is statistically different from the general population of IC neurons for all parameters tested. On average, VIP neuronsterm·def had a significantly higher peak input resistance (R pk ) and steady-state input resistance (R ss ), a lower rheobase, a longer membrane time constantterm·var, a smaller and less variable voltage sag (I h ) at −91 mV, and a markedly small and highly invariable spike frequency adaptation ratioterm·def (SFAterm). Boxplots show median, 25 th and 75 th percentile (box), and 9 th and 91 th percentile (whiskers).

https://doi.org/10.7554/eLife.43770.008 Figure 3—source data 1 Intrinsic physiologyterm·var of VIP neuronsterm·def and from non-targeted recordings in mouse IC.

https://doi.org/10.7554/eLife.43770.009 Download elife-43770-fig3-data1-v2.xlsx To compare the physiology of VIP neuronsterm·def to that of the general population of neuronsterm·var in the IC, we patched neurons in IC slices of C57BL/6J mice in a random, non-targeted approach as a control. Neurons patched with the non-targeted (NTterm) approach showed a higher diversity in firing patternsterm, with a higher proportion of transient neurons (21.8%, 12 out of 55) and adapting neurons (16.4%, 9 out of 55) when compared to VIP neuronsterm·def. Sustained firing neurons were the most prevalent group in NTterm recordings (58.2%, 32 out of 55). Additionally, 3.6% (2 of 55) of randomly patched IC neurons fired only one spike at the onset of the depolarizing current step. This firing patternterm was never observed in VIP neuronsterm·def. The intrinsic physiologyterm·var of VIP neuronsterm·def also differed significantly from the general population of IC neurons (see Figure 3B ). VIP neuronsterm·def on average had a higher peak input resistance ( R pk ) than non-targeted neurons (mean ± SD: VIPterm 242.1 ± 139.4 MΩ vs NTterm 191.1 ± 161.4 MΩ, p = 0.0003, Wilcoxon rank sum testterm·stat), a higher steady-state input resistance ( R ss ) (mean ± SD: VIPterm 239.7 ± 170.7 MΩ vs NTterm 153.4 ± 153.2, p = 0.0001, Wilcoxon rank sum testterm·stat), a slower membrane time constantterm·var (mean ± SD: VIPterm 15.0 ± 8.8 ms vs NTterm 9.7 ± 7.6 MΩ, p = 6.8*10 −7 , Wilcoxon rank sum testterm·stat), lower rheobase values (mean ± SD: VIPterm 67.8 ± 96.2 pA vs NTterm 120.0 ± 100.9 pA, p = 0.013, Wilcoxon rank sum testterm·stat), and a much less pronounced voltage sag (mean ± SD: VIPterm 0.87 ± 0.16 vs NTterm 0.75 ± 0.21, p = 0.0003, Wilcoxon rank sum testterm·stat). Most striking, the SFAterm ratio of VIP neuronsterm·def was tightly clustered at 1.47 ± 1.62, whereas SFAterm of NTterm neurons showed a significantly higher value and spread (3.62 ± 5.43, mean ± SD, p = 2.07*10 −7 ).

Although statistically different from the general neuronal population in the IC and showing a tightly clustered SFAterm ratio, the intrinsic physiologyterm·var of VIP neuronsterm·def still showed some level of variability. In the lower auditory brainstem, it has been found that the intrinsic physiologyterm·var of some neurons varies along the tonotopicterm axis ( Baumann et al., 2013 ; Hassfurth et al., 2009 ). We therefore hypothesized that the intrinsic physiologyterm·var of VIP neuronsterm·def in the ICcterm varied along the tonotopicterm axis of the ICcterm. During patch clamp experiments, VIP neuronsterm·def were passively filled with biocytin via the internal solution. Slices were fixed and stained post hoc with a streptavidin-Alexa Fluor conjugate (see Materials and methods). We then used confocal imaging to map the location of the recorded neurons relative to a two-dimensional (medial-lateral and dorsal-ventral) coordinate system superimposed on the left IC (n = 61 neurons; Figure 4A ). Correlations between intrinsic physiologyterm·var and location in the ICcterm·var were tested by fitting a plane to scatter plots of intrinsic parameters versus medial-lateral and dorsal-ventral coordinates ( Figure 4B ). Because the ICd and IClc are not tonotopically organized, only neurons located in the ICcterm were included in this analysis.

Figure 4 Download asset Open asset Intrinsic physiologyterm·var of VIP neuronsterm·def in the ICcterm varies along the tonotopicterm axis.

( A ) A 2D coordinate system was fit to every IC slice a VIP neuronterm·def was recorded from. The medial-lateral axis runs from the midline (zero) to the lateral edge (max) of the slice, the dorsal-ventral axis from the dorsal edge of the slice (zero) to the ventral border of the IC (max). For illustrative purposes, the approximate position along the tonotopicterm axis of the ICcterm is color-coded from blue (low frequency) to yellow (high frequency). ( B ) Correlation of measured intrinsic parameters with recording location. Black dots represent physiological parameters of individual VIP neuronsterm·def (z-axis, left) mapped to their recording location (x- and y-axes, bottom). Planes show Levenberg-Marquardt least squares fits, color-coded from low z-axis values (blue) to high z-axis values (yellow). Asterisks indicate statistical significance of fit.

https://doi.org/10.7554/eLife.43770.010 Figure 4—source data 1 Intrinsic physiologyterm·var of VIP neuronsterm·def matched to their location in the ICcterm·var.

https://doi.org/10.7554/eLife.43770.011 Download elife-43770-fig4-data1-v2.xlsx We found that variability in the intrinsic physiologyterm·var of VIP neuronsterm·def was at least partially correlated to their location within the coronal plane of the ICcterm. This was particularly true for the voltage sag ratios, which measure hyperpolarization-activated cation current ( I h ). Approximately one quarter of the variability in sag ratios was explained by location in the ICcterm·var (sag ratio at −91 mV: R = 0.536, R 2 adj = 0.262, p = 1.24×10 −05 , n = 60, Figure 4B 4 ; sag ratio at −111 mV: R = 0.516, R 2 adj = 0.233, p = 0.0002, n = 47, Figure 4B 5 ). A significant but smaller portion of the variability in membrane time constantterm·var was explained by location in the ICcterm·var (τ: R = 0.343, R 2 adj = 0.088, p = 0.007, n = 61; Figure 4B 3 ). There was also a significant relationship between the steady-state input resistance of VIP neuronsterm·def and location in the ICcterm·var and a trend toward a relationship between peak input resistance and location ( R ss : R = 0.328, R 2 adj = 0.076, p = 0.011, n = 60; R pk : R = 0.227, R 2 adj = 0.018, p = 0.084, n = 60; Figures 4B1 and 2 ). The tonotopicterm axis of the ICcterm runs along a dorsolateral (low frequency) to ventromedial (high frequency) axis ( Malmierca et al., 2008 ; Portfors et al., 2011 ; Stiebler and Ehret, 1985 ; Willott and Urban, 1978 ). For each of the above intrinsic parameters, values tended to be lower, indicating faster membrane properties, at more dorsolateral locations and higher, indicating slower membrane properties, at more ventromedial locations. Combined, these results suggest that variability in the intrinsic physiologyterm·var of VIP neuronsterm·def is at least in part due to their localization along the tonotopicterm axis of the ICcterm and that the membrane properties of VIP neuronsterm·def tend to be faster in lower frequency regions of the ICcterm.

The streptavidin staining of biocytin-filled VIP neuronsterm·def allowed for a detailed analysis of morphologyterm. In total, we recovered the morphologyterm of 55% of patched VIP neuronsterm·def (n = 100 of 183). Nearly all (81/86 = 94.2%) VIP neuronsterm·def had spinyterm dendrites ( Figure 5A,B insets). This contrasts sharply with the 28% (12 of 43 neurons) of neurons that had spinyterm dendrites in non-targeted recordings from C57BL/6J mice ( Figure 5C ), suggesting that VIP neuronsterm·def represent a subset of IC cells. Nonetheless, dendritic spines can be present on a variety of cell types, including stellateterm and disc-shapedterm cells ( Herrera and Correa, 1988 ; Paloff et al., 1992 ; Willard and Ryugo, 1983 ). On average, VIP neuronsterm·def had five primary dendrites (mean ± SD: 4.77 ± 1.38) that spread out in all directions from the soma, consistent with a stellateterm morphologyterm. This is unsurprising in the ICd and IClc, where stellateterm morphologyterm predominates, but warranted further analysis in the ICcterm, where stellateterm cells can have oriented dendritic trees but are outnumbered by the more highly oriented disc-shapedterm cells ( Malmierca et al., 1993 ; Oliver and Morest, 1984 ; Willard and Ryugo, 1983 ). There were no obvious differences between the morphologyterm of VIP neuronsterm·def in the ICcterm versus the ICd or IClc, but this question warrants more detailed analysis in a future study.

Figure 5 Download asset Open asset VIP neuronsterm·def in the ICcterm are a classterm of stellateterm cells and most VIP neuronsterm·def have dendritic spines.

( A, B ) Maximum-intensity projections of confocal z-stacks showing streptavidin-Alexa Fluor-stained VIP neuronsterm·def from the ICcterm. Insets: enlarged views of dendritic segments show dendritic spines. ( C ) 94% of VIP neuronsterm·def across all IC subdivisions had spinyterm dendrites vs 28% of neurons from non-targeted recordings in C57BL/6J animals. ( D ) Representative reconstructions of the morphologyterm of 12 VIP neuronsterm·def from the ICcterm. Neurons are oriented as if in the left ICcterm. Gray lines were drawn at a 45° angle to illustrate the general orientation of the laminae. Solid gray lines are spaced 200 µm apart, dashed lines and solid lines are spaced 100 µm apart. ( E ) Orientation of the dendritic fields of VIP neuronsterm·def from the ICcterm. Combined: Orientation of all reconstructed VIP neuronsterm·def from the ICcterm (n = 42). Blue lines denote the orientation of the longest axis (first principal direction) found for each neuron using 2D PCA. Caudal, middle, rostral: Orientation of dendritic fields separated according to position along the rostro-caudal axis of the ICcterm. Blue lines show longest axis, perpendicular red lines show second longest axis (second principal direction) of each neuron as defined by 2D PCA. ( F ) Angular orientation of the long axis for every reconstructed VIP neuronterm·def within the ICcterm. Angles indicate counter-clockwise rotation relative to the medial-lateral (horizontal) axis. ( G ) Spread of the dendritic arbors of ICcterm VIP neuronsterm·def measured perpendicular to a predicted 45° isofrequency plane. The dendrites of 83% of VIP neuronsterm·def extended more than 100 µm across the laminar plane. ( H ) Dendritic arbor length to width ratio for all reconstructed VIP neuronsterm·def from the ICcterm (n = 42). 93% of VIP neuronsterm·def had a length to width ratio <3, indicating that they are stellateterm cells. The orientation of length and width axes was determined using 3D PCA.

https://doi.org/10.7554/eLife.43770.012 Figure 5—source data 1 Morphometric analysis of VIP neuronsterm·def in the ICcterm.

https://doi.org/10.7554/eLife.43770.013 Download elife-43770-fig5-data1-v2.xlsx Figure 5D shows the variability in the morphologyterm of VIP neuronsterm·def located in the ICcterm. Neurons are displayed as they would appear in a coronal slice of the left IC viewed from a caudal perspective. Oliver and colleagues distinguished disc-shapedterm from stellateterm neurons by calculating the length to width ratio of the dendritic arbor: neurons with a ratio <3 were stellateterm and those with a ratio >= 3 were disc-shapedterm ( Oliver et al., 1991 ). Applying this classification to our sample, 93% of VIP neuronsterm·def in the ICcterm (39 of 42 neurons) had a length to width ratio <3, therefore being classified as stellateterm ( Figure 5H ). Only three VIP neuronsterm·def from the ICcterm had a length to width ratio >3. These results demonstrate that the dendritic arbors of VIP neuronsterm·def are not as highly oriented as disc-shapedterm neurons, again consistent with the hypothesis that VIP neuronsterm·def are a classterm of stellateterm neurons.

Although less oriented than disc-shapedterm cells, the VIPterm dendrites tended to show some orientation that could influence the range of frequencies that converge on the cell. To measure the orientation of ICcterm VIP neuronsterm·def in relation to the isofrequency laminae, which in mouse run in a ~45° angle through the ICcterm ( Stiebler and Ehret, 1985 ), we identified the longest and second longest axis of each neuron through principal component analysisterm·stat and plotted the orientation of these axes on a standardized model of the IC ( Figure 5E ). No preferred orientation was apparent ( Figure 5E , combined ). Only 17% of VIP neuronsterm·def (7 of 42) had their longest axis oriented within ± 15° of the 45° laminar plane, indicating that the dendritic arbors of most VIP neuronsterm·def (83%, 35 of 42) may be positioned to cross one or more isofrequency laminae in the ICcterm ( Figure 5F ). To quantify this, we calculated the length the dendritic arbor extended perpendicular to a 45° laminar plane. The dendritic arbors of 83% of ICcterm VIP neuronsterm·def (35 of 42) spread more than 100 µm perpendicular to the laminar plane, and more than 36% (15 of 42) spread more than 200 µm across the laminar plane ( Figure 5G ). Previous work in rats has shown that the isofrequency laminae have a center-to-center distance that ranges from 90 to 150 µm, while neurons contained within a lamina had a thickness ranging from 30 to 70 µm ( Malmierca et al., 1993 ). If we assume that laminar dimensions in mouse are no thicker than those in rats, our results indicate that the dendritic fields of VIP neuronsterm·def usually extend beyond at least one isofrequency lamina, consistent with the conclusion that VIP neuronsterm·def in the ICcterm are a classterm of stellateterm neurons. Previous estimates indicate that approximately 15 – 20% of ICcterm neurons are stellateterm ( Oliver et al., 1991 ; Oliver and Morest, 1984 ). Since our stereological analysis showed that VIP neuronsterm·def represent 3.5% of ICcterm neurons, our results indicate that VIP neuronsterm·def account for ~18 – 23% of ICcterm stellateterm neurons.

Injections of an AAV encoding a Cre-dependent eGFP construct, AAV1.CAG.FLEX.eGFP.WPRE.bGH, led to eGFP expression in VIPterm + IC cells. Figure 6A shows a representative deposit site, with eGFP-positive neurons (yellow) located among a population of VIPterm + cells (magenta, labeled by cross-breeding the VIP-IRES-Cre mice with Ai14 reporter mice). Neurons that expressed eGFP routinely co-expressed tdTomato, confirming VIPterm expression by those neurons ( Figure 6B ). Many tdTomato + neurons did not express the eGFP, despite their intermingling with many virally-labeled neurons. eGFP-labeled axons were prominent within the injected IC, where the labeled boutons were located in the neuropil or in close apposition to IC somas, suggesting extensive contributions to local circuits ( Figure 6C ). In addition, eGFP-labeled axons were present in several fiber tracts carrying projections from the IC, including the brachium of the IC, the commissure of the IC and the lateral lemniscus. Labeled axons and boutons were found in numerous auditory nuclei, including the contralateral IC, medial geniculate body and superior olivary complex ( Figure 6D–F ). Additional targets included the nucleus of the brachium of the IC, the periaqueductal gray, and the superior colliculus (not shown; details of termination patterns and terminal axon morphologyterm will be described in a subsequent report). These data indicate that VIPterm + IC neurons contribute to ascending, commissural and descending pathways from the IC.

Figure 6 Download asset Open asset VIP neuronsterm·def project to multiple local and long-range targets.

( A ) Photomicrograph of an AAV deposit site in the right IC. AAV-infected, VIP-expressing cells are labeled with eGFP (yellow), while all VIP-expressing cells are labeled with tdTomato (magenta). Cells expressing both fluorescent proteins appear white. Scale = 500 µm. ( B ) High magnification photomicrographs showing labeled cells in the AAV deposit site. The field shows four tdTomato-expressing cells (magenta), two of which (white arrows) were also AAV-infected and expressed eGFP (yellow). Scale = 20 µm. ( C–F ) High magnification photomicrographs showing eGFP-labeled collicular axons (yellow) terminating in the ipsilateral IC ( C ), the contralateral IC ( D ), the medial geniculate body ( E ), or the ventral nucleus of the trapezoid body in the superior olivary complex ( F ) after an AAV injection in the IC. The white dashed box in each image identifies an area enlarged in the inset to show details of labeled axons and boutons. A fluorescent Nissl stain (cyan) shows that boutons are located in close association with cell bodies as well as in the intervening neuropil. Scale = 20 µm.

https://doi.org/10.7554/eLife.43770.014 In addition to axonal projection patterns, the sources of synaptic input to a neuron classterm are an important predictor of neuronal function. Anatomical studies have shown that the IC receives ascending, descending, and commissural input, but, with the exception of large GABAergicterm neurons ( Ito et al., 2015 ; Ito and Oliver, 2014 ), it has rarely been possible to identify the sources and physiology of synaptic input to a specific classterm of neurons in the IC. This is largely because axons from multiple sources overlap in the IC, making it difficult to use electrical stimulation to selectively activate axons from specific presynaptic sources. In addition, electrical stimulation of commissural projections cannot differentiate between axons originating in the ipsilateral and contralateral IC. This is because electrical stimulation evokes both orthodromic and antidromic spikes, and, since most IC neurons have local axon collaterals ( Oliver et al., 1991 ), antidromic spikes can lead to synaptic release from the local collaterals of neurons ipsilateral to the recording site. Furthermore, the commissure contains axons from sources other than IC cells, including cells in the auditory cortex, sagulum, nuclei of the lateral lemniscus and superior paraolivary nucleus (reviewed by Saldaña and Merchán, 2005 ); electrical stimulation in the commissure could also activate these axons. To overcome these obstacles, we turned to Channelrhodopsin-assisted circuit mapping (CRACM)term·def ( Petreanu et al., 2007 ). With CRACM, it is possible to selectively activate a single population of presynaptic neurons by anatomically and/or molecularly restricting the expression of an optogenetic protein. In our experiments, we used stereotaxic coordinates and intracranial virus injections to anatomically restrict Chronos expression. Epi-fluorescence imaging confirmed that Chronos-GFP expression was always clearly limited to the right IC ( Figure 7B ) or the right DCN ( Figure 8A ).

Figure 7 Download asset Open asset VIP neuronsterm·def in the ICcterm receive excitatory and inhibitory synaptic input from the contralateral IC.

( A ) Experimental setup. An AAV encoding Chronos-GFP was injected into the right IC. Three weeks later, light-evoked postsynaptic potentials were recorded from VIP neuronsterm·def in the left ICcterm. ( B ) Image of a coronal slice of the IC. Injection sites and Chronos expression were validated through Chronos-GFP fluorescence. ( C ) Optogenetically-evoked IPSPs recorded from VIP Neuronsterm·def in the ICcterm contralateral to the AAV injection site. IPSPs were evoked by 2–5 ms blue light flashes (left), while EPSPs were blocked with NBQX and AP5. IPSPs were abolished by gabazine (right). ( D ) Optogenetically-evoked EPSPs recorded from VIP neuronsterm·def in the ICcterm contralateral to the AAV injection site. EPSPs were evoked by 2–5 ms blue light flashes (left), while IPSPs were blocked with strychnine and gabazine. Wash-in of AP5 significantly reduced the halfwidthterm·var and decay time constantterm·var of light-evoked EPSPs (middle). Wash-in of NBQX abolished the remaining EPSP (right). ( E ) Population data showing amplitude and kinetics of optogenetically-evoked IPSPs. ( F ) Population data showing amplitude and kinetics of optogenetically-evoked EPSPs. The significant reduction of EPSP halfwidthterm·var by AP5 indicates that NMDA receptor activation prolonged EPSP duration.

https://doi.org/10.7554/eLife.43770.015 Figure 7—source data 1 EPSP and IPSP analysis of commissural inputs to VIP neuronsterm·def.

https://doi.org/10.7554/eLife.43770.016 Download elife-43770-fig7-data1-v2.xlsx Figure 8 Download asset Open asset VIP neuronsterm·def in the ICcterm receive direct synaptic input from the DCN and feedforward inhibition driven by DCN afferents.

( A ) Experimental setup. An AAV encoding Chronos-GFP was injected into the right DCN. For every experiment, the injection site and Chronos-GFP expression were confirmed through GFP fluorescence. Current clamp recordings were made from VIP neuronsterm·def in the ICcterm contralateral to the injection site. ( B ) With inhibition blocked by gabazine and strychnine, 2–5 ms blue light flashes evoked EPSPs (left). AP5 did not significantly reduce EPSP halfwidthterm·var or decay time constantterm·var. Subsequent addition of NBQX abolished the EPSP. ( C ) Population data showing amplitude and kinetics for EPSPs elicited by activation of DCN synapses onto VIP neuronsterm·def in the ICcterm. The absence of a significant effect of AP5 indicates that NMDA receptors did not make a significant contribution to EPSPs. ( D ) In several recordings made in the absence of inhibitory blockers, EPSP duration was limited through GABAergicterm feedforward inhibition (left; n = 5). In these instances, gabazine wash-in increased EPSP halfwidthterm·var to values similar to those in ( B ). ( E ) Population data for feedforward inhibition to VIP neuronsterm·def. Washing in gabazine increased EPSP halfwidthterm·var in 5 out of 5 tested connections. ( F ) Population data comparing amplitude, halfwidthterm·var, and decay time constantterm·var of EPSPs from DCN inputs recorded in VIP neuronsterm·def (VIPterm) and non-VIP neurons (NTterm, non-targeted recording). Halfwidthterm·var and decay time constantsterm·var in VIP neuronsterm·def showed a trend to cluster more tightly that in non-targeted recordings.

https://doi.org/10.7554/eLife.43770.017 Figure 8—source data 1 Analysis of EPSPs from DCN inputs in VIP neuronsterm·def and non-targeted recordings.

https://doi.org/10.7554/eLife.43770.018 Download elife-43770-fig8-data1-v2.xlsx Numerically, the contralateral IC provides the largest single source of input to the IC ( Moore, 1988 ). We therefore first used CRACM to test whether VIP neuronsterm·def in the ICcterm receive commissural input. Using stereotaxic, intracranial virus injections with AAV1.Syn.Chronos-GFP.WPRE.bGH, we drove expression of GFP-tagged Chronos, a fast Channelrhodopsin variant ( Klapoetke et al., 2014 ), in the right IC. Note that we did not attempt to limit Chronos expression to a particular subdivision of the right IC. Visual inspection of Chronos-GFP fluorescence suggested that most of the somata labeled in the right IC were located in the ICcterm, but labeled somata were often also present in the ICd and occasionally the IClc.

We then targeted recordings to VIP neuronsterm·def in the contralateral (left) ICcterm ( Figure 7A ). In each experiment, we used GFP fluorescence to verify transfection of the right IC ( Figure 7B ) and restricted our recordings to VIP neuronsterm·def in areas of the left ICcterm where GFP-labeled axons were visible. Because commissural projections are a mixture of glutamatergicterm and GABAergicterm projections ( González-Hernández et al., 1996 ; Hernández et al., 2006 ; Nakamoto et al., 2013 ; Saint Marie, 1996 ), we used pharmacology to isolate EPSPs and IPSPs. Indeed, without adding receptor antagonists to the bath, postsynaptic potentials often were mixtures of IPSPs and EPSPs (data not shown). In the presence of AMPA and NMDA receptor antagonists (10 µM NBQX and 50 µM D-AP5, bath application), 2 – 5 ms flashes of blue light elicited IPSPs in 6 out of 12 neurons tested ( Figure 7C , left ). IPSPs were completely abolished by the GABA A receptor antagonist gabazine (5 µM, Figure 7C , right; n = 6; amplitude, p = 0.006; rise time, p = 0.003; halfwidthterm·var, p = 0.001; membrane time constantterm·var, p = 0.003, paired t-testterm·stat). On average (n = 6), commissural IPSPs in ICcterm VIP neuronsterm·def were small (−1.53 mV ± 0.96 mV) and had moderate 10 – 90% rise times (7.8 ms ± 2.1 ms), halfwidthsterm·var (15.1 ms ± 6.8 ms) and decay time constantsterm·var (32.4 ms ± 17.0 ms) ( Figure 7E ).

To investigate excitatory commissural inputs to ICcterm VIP neuronsterm·def, recordings were carried out in the presence of GABA A and glycine receptor antagonists (5 µm gabazine and 1 µM strychnine, bath application, Figure 7D ). We found that 2 – 5 ms flashes of blue light elicited EPSPs in 11 out of 27 neurons ( Figure 7D , left). On average (n = 6), commissural EPSPs in VIP neuronsterm·def were small (1.52 mV ± 1.08 mV) and had moderate 10 – 90% rise times (8.3 ms ± 4.3 ms), halfwidthsterm·var (19.6 ms ± 7.6 ms) and decay time constantsterm·var (43.5 ms ± 16.8 ms) ( Figure 7F ). Adding the NMDA receptor antagonist D-AP5 to the bath significantly reduced the halfwidthterm·var of EPSPs (14.3 ms ± 4.7 ms, p = 0.006) and revealed a trend toward reducing the rise time (6.3 ms ± 1.6 ms, p = 0.09) and decay time constantterm·var (30.6 ms ± 7.3 ms, p = 0.06) of EPSPs (ANOVAterm·stat for repeated measurements with Tukey post-hoc test). The remainder of the EPSP was completely blocked by the AMPA receptor antagonist NBQX ( Figure 7F ). These results suggest that VIP neuronsterm·def in the ICcterm receive excitatory commissural input and express both AMPA and NMDA receptors at excitatory commissural synapses. Interestingly, commissural input activated NMDA receptors even though there was 1 mM Mg 2+ in the bath and the somatic membrane potential was at or near the resting membrane potential throughout the recording. This may indicate that commissural synapses are located on the distal dendrites and/or the dendritic spines of VIP neuronsterm·def.

Combined, the commissural CRACM experiments show that VIP neuronsterm·def in the ICcterm receive excitatory and inhibitory synaptic input from the contralateral IC. Surprisingly, although GABAergicterm neurons make up <= 20% of commissural projections ( Hernández et al., 2006 ; Nakamoto et al., 2013 ), we found a higher connection probability for inhibitory commissural projections (50%; 6 out of 12 recordings performed in the presence of NBQX and D-AP5), than for excitatory connections (41%; 11 out of 27 recordings performed in the presence of gabazine and strychnine).

Due to the kinetics of optogenetic proteins and the possibility of activating optogenetic proteins in synaptic terminals, optogenetically-evoked PSPs may have slower kinetics than PSPs evoked by electrical stimulation ( Jackman et al., 2014 ; Zhang and Oertner, 2007 ). To test whether the kinetics of Chronos-evoked commissural PSPs were similar to those of electrically-evoked PSPs, we electrically stimulated the IC commissure while recording PSPs from VIP neuronsterm·def in the contralateral IC. As indicated above, this raises the complication that electrically-evoked EPSPs were probably elicited both from the desired contralateral projections and from antidromic stimulation of ipsilateral axons. Compared to optogenetically-evoked EPSPs, electrically-evoked EPSPs (n = 6) had similar amplitudes and rise times, but had smaller halfwidthsterm·var and trended toward faster decay time constantsterm·var (amplitude: 2.4 mV ± 1.7 mV, p = 0.37; rise time: 6.28 ms ± 6.3 ms, p = 0.40; halfwidthterm·var: 10.3 ms ± 4.8 ms, p = 0.008*; decay time constantterm·var 28.4 ms ± 10.0 ms, p = 0.016, two tailed t-testterm·stat, *critical p value with Bonferroni correction for multiple comparisonsterm·stat = 0.0125). Electrically-evoked IPSPs (n = 6) showed no significant difference in amplitude, halfwidthterm·var or decay time constantterm·var, but trended toward faster rise times (amplitude: −1.7 mV ± 1.3 mV, p = 0.78; rise time: 4.2 ms ± 2.3 ms, p = 0.025; halfwidthterm·var: 8.8 ms ± 7.3 ms, p = 0.16; decay time constantterm·var 36.8 ms ± 17.3 ms, p = 0.67, two tailed t-testterm·stat, critical p value with Bonferroni correction for multiple comparisonsterm·stat = 0.0125). It should be noted that to position the stimulus electrode in the IC commissure contralateral to the recording site, these recordings had to be biased to VIP neuronsterm·def in more rostral parts of the IC where the commissure was clearly visible, a bias that was not present in the CRACM recordings. Also, we cannot exclude that antidromic spikes may have influenced PSP analysis. Both may account for some of the observed differences in PSP kinetics. Overall, however, these data suggest that the kinetics of Chronos-evoked PSPs were generally similar to those of electrically-evoked PSPs.

The DCN provides one of the major sources of excitatory input to the IC ( Adams, 1979 ; Brunso-Bechtold et al., 1981 ; Oliver, 1984 ; Osen, 1972 ; Ryugo et al., 1981 ). A previous study has shown that DCN afferents synapse onto glutamatergicterm and GABAergicterm neurons in the IC ( Chen et al., 2018 ), but it is not known which specific classesterm of IC neurons receive input from the DCN. In addition, the physiology of DCN afferent synapses remains unknown. To test whether VIP neuronsterm·def receive synaptic input from DCN principal neurons, we injected the right DCN with the AAV1.Syn.Chronos-GFP.WPRE.bGH viral vector and recorded from VIP neuronsterm·def in the contralateral (left) ICcterm ( Figure 8A , left). To confirm selective transfection of the DCN, we sliced the brainstem of every animal used and determined whether GFP expression was present and limited to the right DCN. If there was no transfection or if there was considerable expression of GFP in the auditory nerve or VCN, no recordings were performed. In most cases, GFP expression was limited to the DCN ( Figure 8A , right) and GFP-labeled axons were present in the left ICcterm.

To block spontaneous IPSPs, we performed DCN CRACM experiments with GABAergicterm and glycinergic blockers in the bath (5 µm gabazine and 1 µM strychnine, Figure 8B ). We found that 2–5 ms pulses of blue light elicited EPSPs in 19 of 25 neurons tested, confirming that ICcterm VIP neuronsterm·def receive synaptic input from the DCN. Light-evoked EPSPs had moderate amplitudes (2.85 mV ± 2.98 mV) and relatively slow rise times (4.2 ms ± 1.3 ms), halfwidthsterm·var (20.6 ms ± 14.4 ms) and decay time constantsterm·var (22.0 ms ± 6.7 ms) (n = 6 cells, Figure 8B , left and 8C). Because EPSP kinetics were relatively slow, we hypothesized that DCN synapses activate NMDA receptors on VIP neuronsterm·def. Interestingly, D-AP5 had no significant effect on any of the measured properties (amplitude: 2.80 mV ± 2.54 mV, p = 0.96, rise time: 4.4 ms ± 1.7 ms, p = 0.95, halfwidthterm·var: 15.4 ms ± 6.1 ms, p = 0.37, decay time constantterm·var: 29.0 ms ± 7.1 ms, p = 0.16, two tailed t-testterm·stat, critical p value with Bonferroni correction for multiple comparisonsterm·stat = 0.0125) ( Figure 8B , middle, and 8C). In contrast to commissural inputs where D-AP5 had a significant effect on EPSP kinetics, this suggests that DCN inputs to ICcterm VIP neuronsterm·def do not activate NMDA receptors under resting conditions. Subsequent addition of NBQX completely abolished DCN-evoked EPSPs ( Figure 8B , right), confirming that DCN synapses activate AMPA receptors on ICcterm VIP neuronsterm·def. Since fusiform cells in the DCN often fire at rates exceeding 100 Hz ( Davis et al., 1996 ; Ma and Brenowitz, 2012 ; Nelken and Young, 1994 ; Spirou and Young, 1991 ; Young and Brownell, 1976 ), the slow kinetics of DCN-evoked EPSPs suggests that these EPSPs undergo temporal summation in VIP neuronsterm·def.

To test whether Chronos-elicited synaptic release from DCN terminals was driven by action potentials as opposed to direct activation of DCN terminals by light, we repeated the DCN CRACM experiments in the presence of 1 µM tetrodotoxin (TTX). After TTX wash-in, EPSPs elicited by blue light flashes were completely and reversibly abolished in 3 of 3 tested VIP neuronsterm·def (data not shown). This suggests that synaptic release in CRACM experiments was driven by action potentials, or was at least dependent on sodium channel activation in synaptic terminals.

To determine whether PSPs evoked by DCN inputs to VIP neuronsterm·def differed from those evoked in non-VIP neurons, we recorded from non-fluorescent neurons in the ICcterm of VIP-IRES-Cre x Ai14 mice in which the DCN was transfected with Chronos. Recordings were targeted to neurons in regions of the ICcterm where Chronos-GFP-labeled axons were clearly visible. We found that 8 out of 12 non-VIP neurons received input from the DCN. EPSPs evoked by DCN inputs to non-VIP neurons (n = 8) had similar amplitudes, rise time, halfwidthterm·var and decay times (amplitude: 1.7 mV ± 0.6 mV, p = 0.34; rise time: 4.7 ms ± 1.6 ms, p = 0.52; halfwidthterm·var: 16.5 ms ± 9.6 ms, p = 0.77; decay time constantterm·var 25.9 ms ± 17.8 ms, p = 0.72; two tailed t-testterm·stat, critical p value with Bonferroni correction for multiple comparisonsterm·stat = 0.0125) compared to those recorded from VIP neuronsterm·def. Although the means were not statistically different, DCN-evoked EPSPs in VIP neuronsterm·def showed less variability in halfwidthterm·var and decay time than EPSPs in non-VIP neurons ( Figure 8F ). These results suggest that ICcterm neurons located near DCN afferents have a high probability of receiving input from the DCN, consistent with predictions made from anatomical studies ( Oliver, 1984 ; Oliver et al., 1997 ).

Together, these results identify VIP neuronsterm·def in the ICcterm as a distinct postsynaptic target of DCN afferents. Given the number of conditions that must be met for a long-range CRACM experiment to succeed, our observation that the connection probability for DCN to VIPterm projections was 76% (19 of 25 neurons) suggests that most VIP neuronsterm·def in the ICcterm receive input from the DCN.

We next repeated the DCN-CRACM experiments without GABA A and glycine receptor antagonists in the bath. Under these conditions, we observed that an IPSP was elicited 2–3 ms after the onset of the light-evoked EPSP ( Figure 8D ). This IPSP could vary in strength between recorded VIP neuronsterm·def. In some instances, the IPSP slightly altered the halfwidthterm·var and decay time constantterm·var of the EPSP. In other cases, the IPSP strongly limited the EPSP duration and generated a hyperpolarization after the EPSP ( Figure 8D , left). Washing in gabazine restored the EPSP to values comparable to the EPSPs recorded in the presence of inhibitory receptor antagonists (compare Figure 8B,C ). Washing out gabazine restored the IPSP and limited EPSP halfwidthterm·var again ( Figure 8D , right). Across five ICcterm VIP neuronsterm·def, the IPSP significantly shortened the halfwidthterm·var of the elicited EPSP (p = 0.048, paired t-testterm·stat, Figure 8E ). Halfwidthterm·var reductions ranged from 22% to 73%, with a median reduction of 36%. Because DCN to IC projections are glutamatergicterm ( Ito and Oliver, 2010 ; Oliver, 1984 ), and the periolivary nuclei and nuclei of the lateral lemniscus, the sources of ascending GABAergicterm input to the IC ( González-Hernández et al., 1996 ), were not present in the brain slice, this inhibition must be due to DCN afferents activating a local feedforward inhibitory circuit within the IC. The latency to the IPSP onset also supports the theory of a disynaptic, local inhibitory circuit, as the IPSP always succeeded the initial EPSP. Thus, these results indicate that ascending input from the DCN activates a feedforward inhibitory circuit within the IC and that this circuit regulates the duration of DCN-evoked excitation in ICcterm VIP neuronsterm·def.

By combining molecular genetics with physiological and anatomical approaches, we identified VIP neuronsterm·def as a novel classterm of glutamatergicterm principal neurons in the auditory midbrain. In contrast to the diverse properties present in the broader population of IC neurons, VIP neuronsterm·def consistently exhibited sustained firing patternsterm·def, spinyterm dendrites, and stellateterm morphologyterm. Surprisingly, the identification of VIP neuronsterm·def revealed that a molecularly-defined classterm of IC neurons can broadcast the output of the IC to multiple auditory and non-auditory brain regions ( Figure 9A ). The identification of VIP neuronsterm·def also enabled us to identify microcircuit motifs within the IC. In particular, we found that VIP neuronsterm·def receive ascending input from the contralateral DCN and commissural input from the contralateral IC ( Figure 9B ). That VIP neuronsterm·def receive input from the DCN was consistent with the distribution of VIP neuronsterm·def in the ICcterm, which was biased toward more caudal regions where DCN afferents are prominent. Input from the DCN also drove feedforward inhibition that could sculpt the effects of excitatory inputs to VIP neuronsterm·def. Thus, feedforward inhibitory circuits within the IC may regulate the temporal summation of ascending inputs. By integrating multiple sources of input and participating in most of the major projections from the IC, VIP neuronsterm·def are well-positioned to broadly influence auditory computations in numerous brain regions.

Figure 9 Download asset Open asset Summary of inputs and projection targets of VIP neuronsterm·def in mouse IC.

( A ) Summary of the major projection targets of VIP neuronsterm·def identified by axonal tract tracing: MGB (medial geniculate body), SC (superior colliculus), contralateral IC, PAG (periaqueductal gray), SOC (superior olivary complex). ( B ) Summary of the sources of input to VIP neuronsterm·def identified by CRACM experiments.

https://doi.org/10.7554/eLife.43770.019 It has long been argued that the classification of neurons requires a multifaceted analysis of morphological and physiological features ( Tyner, 1975 ). More recent efforts have emphasized the importance of combining these features with molecular markersterm ( Ascoli et al., 2008 ; Tremblay et al., 2016 ; Zeng and Sanes, 2017 ). This combination has proven to be particularly effective for unambiguously classifying neuron types. In the cerebral cortexterm, a multifaceted classification scheme including molecular markersterm has enabled investigations into how specific classesterm of interneuronsterm shape circuit computations, sensory processing, and animal behavior ( Cichon et al., 2017 ; Kato et al., 2017 ; Kato et al., 2015 ; Kuchibhotla et al., 2017 ; Lee et al., 2013 ; Milstein et al., 2015 ; Pfeffer et al., 2013 ; Pi et al., 2013 ). Similar approaches have succeeded in the amygdala, hypothalamus, basal ganglia, and other brain regions where it was previously difficult to identify neuron types ( Campbell et al., 2017 ; Capogna, 2014 ; Wallace et al., 2017 ).

In the auditory midbrain, previous efforts to identify neuron classesterm relied on in vitro physiology, in vivo physiology, morphologyterm, neurochemical markersterm, or some combination of these ( BeebepersName et al., 2016 ; Fujimoto et al., 2017 ; Malmierca et al., 1993 ; Oliver and Morest, 1984 ; Ono et al., 2005 ; Palmer et al., 2013 ; Peruzzi et al., 2000 ; Ramachandran et al., 1999 ; SchofieldpersName and BeebepersName, 2019 ; Sivaramakrishnan and Oliver, 2001 ). Despite these and other attempts, the only neuron classterm that has been consistently identified in the IC are the large GABAergicterm neurons ( BeebepersName et al., 2016 ; Geis and Borst, 2013 ; Ito et al., 2015 ; Ito et al., 2009 ; Ito and Oliver, 2014 ). However, there are currently no known molecular markersterm for large GABAergicterm neurons, making it difficult to test the role of these neurons in auditory computations ( SchofieldpersName and BeebepersName, 2019 ).

Using a multifaceted approach, we identified VIP neuronsterm·def as a novel classterm of IC neurons. VIP neuronsterm·def share a common set of molecular, neurochemical, morphological, and physiological features. VIP neuronsterm·def are labeled in the VIP-IRES-Cre mouse line and are glutamatergicterm. Ninety-three percent of the ICcterm VIP neuronsterm·def in our data set had a stellateterm morphologyterm and 94% of all IC VIP neuronsterm·def had dendritic spines. Similarly, 90% of VIP neuronsterm·def had sustained firing patternsterm·def. Although the input resistance, membrane time constantterm·var, and expression of I h varied within the population of VIP neuronsterm·def, for VIP neuronsterm·def in the ICcterm, we found that a portion of this variability reflected their location along the tonotopicterm axis of the ICcterm. Tonotopicterm variation in the intrinsic physiologyterm·var of VIP neuronsterm·def was consistent with VIP neuronsterm·def having faster membrane properties and therefore better temporal coding potential in lower frequency regions of the ICcterm. This parallels an in vivo study that found a gradient in the temporal coding capacity of cat ICcterm neurons that varied from better temporal coding in low frequency regions to worse temporal coding in high frequency regions ( Rodríguez et al., 2010 ).

It is important to note that it would not be possible to identify VIP neuronsterm·def based on their morphologyterm or physiology alone. VIP neuronsterm·def are not the only stellateterm neurons in the IC, nor are they the only neurons with sustained firing patternsterm·def or dendritic spines. These results provide insight to why it has been difficult to classify neuron types in the IC. We propose that a multidimensional approach incorporating molecular markersterm will be essential to identifying the remaining neuron classesterm in the IC.

We found that VIP neuronsterm·def project not only to MGB and contralateral IC, the most common recipients of IC projections, but also to the nucleus of the brachium of the IC, superior colliculus, periaqueductal gray, superior olivary complex, and ipsilateral IC ( Figure 6 ). The number of extrinsic targets reached by VIPterm axons was a surprise given the relatively small population of VIP neuronsterm·def. Do individual VIP neuronsterm·def project to multiple targets? Previous retrograde labeling studies suggest that some patterns of collateral projection are more common than others for IC cells. IC cells that project to the contralateral thalamus appear to quite commonly have a collateral projection to the ipsilateral thalamus ( Mellott et al., 2019 ). In contrast, very few IC neurons project to the thalamus and the cochlear nucleus ( Coomes and SchofieldpersName, 2004 ; Hashikawa and Kawamura, 1983 ; Okoyama et al., 2006 ), or to the left and right cochlear nuclei ( SchofieldpersName, 2001 ). Whether IC commissural cells can have collateral projections to the thalamus has been supported ( González-Hernández et al., 1996 ) or denied ( Okoyama et al., 2006 ). Because retrograde tracing studies underestimate collateral projections ( SchofieldpersName et al., 2007 ), such studies may have missed VIP neuronsterm·def with collateral projections.

Alternatively, individual VIP neuronsterm·def might project to one or a few targets. It would then be possible to subdivide VIP neuronsterm·def based on their axonal projections. This would parallel the cerebral cortexterm, where the major classesterm of interneuronsterm contain subclasses that often differ in their axonal targeting ( Tremblay et al., 2016 ). If such is the case, then the unifying feature of VIP neuronsterm·def might be that they perform similar computational roles within circuits, even when the circuits themselves are involved in different functions. In any event, the axonal projection patterns of individual VIP neuronsterm·def, to extrinsic targets and within the IC, will be important features for further characterizing VIPterm subclasses and their functional roles.

Our results show that VIP neuronsterm·def receive input from at least four sources: principal neurons in the DCN, local inhibitory neurons, and excitatory and inhibitory neurons in the contralateral IC ( Figure 8G ). In multiple instances, we observed that VIP neuronsterm·def received input from the DCN and a local inhibitory neuron or a combination of excitatory and inhibitory commissural input. Given that optogenetic circuit mapping experiments underestimate connection probabilities (not all synapses are transfected by the virus), these results suggest that many individual VIP neuronsterm·def integrate input from ascending, local, and commissural sources. This is consistent with previous studies showing that individual IC neurons can integrate input from numerous sources ( Ito et al., 2015 ; Ito and Oliver, 2014 ).

Excitatory postsynaptic responses in IC neurons often involve activation of NMDA receptors ( Ma et al., 2002 ; Wu et al., 2004 ). Under in vivo conditions, the activation of NMDA receptors can influence how IC neurons respond to tones ( Sanchez et al., 2007 ). In VIP neuronsterm·def, we found that excitatory commissural synapses activated AMPA and NMDA receptors, while synaptic input from DCN afferents activated only AMPA receptors. The activation of NMDA receptors occurred even though our ACSF contained 1 mM Mg 2+ and neurons were at their resting membrane potential. Interestingly, previous studies have shown that NMDA receptors in some IC neurons can be activated under similar conditions, even when AMPA are receptors blocked ( Ma et al., 2002 ; Sivaramakrishnan and Oliver, 2006 ). The activation of NMDA receptors in our recordings may indicate that excitatory commissural synapses tend to be located on the distal dendrites or dendritic spines of VIP neuronsterm·def, where the local membrane potential might be sufficiently depolarized by activation of AMPA receptors to remove Mg 2+ block of NMDA receptors. A distal location would be consistent with the proposed modulatory role for commissural inputs ( Orton et al., 2016 ; Orton and Rees, 2014 ). Similarly, the lack of NMDA receptor activation by DCN afferents might indicate that DCN synapses are located on proximal dendrites or possibly on the soma itself, or that these synapses lack NMDA receptors. These synaptic arrangements may have important implications for auditory coding and synaptic plasticity mechanisms in VIP neuronsterm·def.

In many brain regions, feedforward inhibitory circuits control the time window for temporal integration of synaptic input ( Gabernet et al., 2005 ; Pouille and Scanziani, 2001 ; RobertspersName et al., 2013 ; Stokes and Isaacson, 2010 ). In the ICcterm, it was recently shown that GABAergicterm neurons provide local inhibitory input mainly to neurons in the same isofrequency lamina ( Sturm et al., 2014 ). However, the conditions that recruit local inhibition have remained unclear. Our data provide direct evidence that activation of DCN afferents can elicit both direct excitatory input and disynaptic feedforward inhibition to VIP neuronsterm·def. Feedforward inhibition can dramatically reduce EPSP halfwidthterm·var, suggesting that local feedforward inhibition regulates the temporal summation of synaptic inputs. In addition, while DCN afferents elicited modest EPSPs in VIP neuronsterm·def, DCN input presumably drove spiking in the GABAergicterm neurons that provided feedforward inhibition. It will be important for future studies to identify this population of GABAergicterm neurons and the extent of their influence on auditory computations in VIPterm and other IC neurons.

Ascending projections from most auditory brainstem nuclei restrict their synapses to subregions of the ICcterm, dividing the ICcterm into functional zones that may be specialized for processing specific classesterm of auditory cues ( Brunso-Bechtold et al., 1981 ; Cant, 2013 ; Cant and Benson, 2006 ; Loftus et al., 2010 ; Oliver, 2005 ; Oliver et al., 1997 ). We found that VIP neuronsterm·def in the ICcterm were distributed with a significant bias toward the caudal ICcterm ( Table 2 ) and a tendency to be enriched in medial and dorsal regions of the ICcterm ( Figure 1 ). Intriguingly, this distribution overlaps with the monaural domain of the ICcterm, an area that predominantly receives input from monaural nuclei, including the contralateral DCN and AVCN and the ipsilateral ventral nucleus of the lateral lemniscus (VNLL) ( Cant and Benson, 2006 ; Loftus et al., 2010 ). In addition, a recent study in mice showed that excitatory ICcterm neurons that receive input from the DCN are unlikely to receive input from other ascending auditory sources but likely to receive input from the ipsilateral auditory cortex ( Chen et al., 2018 ). These results lead us to hypothesize that VIP neuronsterm·def are specialized for processing monaural cues, and in particular, monaural cues from the DCN. In future studies, it will be important to determine whether ascending input to VIP neuronsterm·def is predominantly from the DCN, from a combination of monaural brainstem nuclei, or from both monaural and binaural sources. It will also be important to assess whether VIP neuronsterm·def receive descending input from the auditory cortex and how commissural and cortical inputs shape sound processing in VIP neuronsterm·def.

An important question is whether IC VIP neuronsterm·def use VIPterm signaling to modulate the activity of their postsynaptic targets. In cerebral cortexterm, 99.1% of neurons labeled in the VIP-IRES-Cre mouse immunostained with an antibody against VIPterm, suggesting that Cre expression in the VIP-IRES-Cre mouse is well correlated with the production of VIPterm ( Prönneke et al., 2015 ). There is therefore a strong likelihood that VIP neuronsterm·def in the IC produce VIPterm. VIPterm acts by binding to one of three G-protein coupled receptors, VPAC1, VPAC2, and PAC1 ( Dickson and Finlayson, 2009 ; Vaudry et al., 2009 ). Anatomical studies suggest that all three VIPterm receptors are expressed in the MGB and SOC ( Joo et al., 2004 ; Sheward et al., 1995 ), both of which are postsynaptic targets of VIP neuronsterm·def ( Figure 6 ). Activation of VIPterm receptors can have significant effects on neuronal excitability. In somatosensory thalamus, activation of VIPterm receptors increased the activation of HCN channels, depolarizing thalamocortical neurons and inducing them to switch from burst firing to tonic firing ( Lee and Cox, 2003 ; Sun et al., 2003 ). These data suggest that IC VIP neuronsterm·def may potently modulate the activity of MGB neurons and other postsynaptic target neurons through VIPterm signaling. Thus, VIP neuronsterm·def in the IC may exert an outsize influence on their long-range postsynaptic targets.

Key resources table Reagent type (species) or resource Designation Source or reference Identifiers Additional information Strain, strain background ( Mus musculus ) C57BL/6J The Jackson Laboratory JAX:000664 Genetic reagent ( Mus musculus ) VIP-IRES-Cre The Jackson Laboratory JAX:010908 Genetic reagent ( Mus musculus ) Ai14 The Jackson Laboratory JAX:007914 Antibody anti-GAD67 (mouse monoclonal) Millipore RRID: AB_2278725 Cat#:MAB5406 IHC (1:1000) Antibody anti-NeuN (rabbit polyclonal) Millipore RRID: AB_10807945 Cat#:ABN78 IHC (1:500) Antibody anti-bNOS (mouse monoclonal) Sigma-Aldrich RRID: AB_260754 Cat#:N2280 IHC (1:1000) Antibody anti-mouse IgG Alexa Fluor 488 (donkey polyclonal) ThermoFisher RRID: AB_141607 Cat#:A-21202 IHC (1:500) Antibody anti-rabbit IgG Alexa Fluor 488 (donkey polyclonal) ThermoFisher RRID: AB_2535792 Cat#:A-21206 IHC (1:500) Recombinant DNA reagent AAV1.Syn.Chronos-GFP.WPRE.bGH University of Pennsylvania Vector Core/Addgene Addgene:59170-AAV1 RRID: Addgene_59170 http://n2t.net/addgene:59170 Recombinant DNA reagent AAV1.CAG.FLEX.eGFP.WPRE.bGH University of Pennsylvania Vector Core/Addgene Addgene:51502-AAV1 RRID: Addgene_51502 http://n2t.net/addgene:51502 Chemical compound, drug gabazine Hello Bio Cat#:HB0901 also called SR95531 hydrobromide Chemical compound, drug strychnine hydrochloride Sigma-Aldrich Cat#:S8753 Chemical compound, drug D-AP5 Hello Bio Cat#:HB0225 Chemical compound, drug NBQX disodium salt Hello Bio Cat#:HB0443 Software, algorithm Igor Pro 7 and 8 Wavemetrics RRID: SCR_000325 Software, algorithm MATLAB R2018a and R2018b Mathworks RRID: SCR_001622 Software, algorithm Neurolucida MBF Bioscience RRID: SCR_001775 Software, algorithm Neurolucida 360 MBF Bioscience RRID: SCR_016788 All experiments were approved by the University of Michigan Institutional Animal Care and Use Committee and were in accordance with NIH guidelines for the care and use of laboratory animals. Animals were kept on a 12 hour day/night cycle with ad libitum access to food and water. VIP-IRES-Cre mice ( Vipterm tm1(cre)Zjh /J, Jackson Laboratory, stock # 010908) ( Taniguchi et al., 2011 ) were crossed with Ai14 reporter mice (B6.Cg- Gt(ROSA)26Sor tm14(CAG-tdTomato)Hze /J, Jackson Laboratory, stock # 007914) ( Madisen et al., 2010 ) to yield F1 offspring that expressed the fluorescent protein tdTomato in VIP neuronsterm·def. For control experiments, C57BL/6J mice (Jackson Laboratory, stock # 000664) were used. Because mice on the C57BL/6 background undergo age-related hearing loss, experiments were restricted to an age range where hearing loss should be minimal or not present ( Zheng et al., 1999 ). This age range was P70 or less for all mice except for three C57BL/6J mice used for electrophysiologyterm·mth experiments that were aged P89, P90, and P113.

Mice were deeply anesthetized and perfused transcardially with 0.1 M phosphate-buffered saline (PBS), pH 7.4, for 1 min and then with a 10% buffered formalin solution (Millipore Sigma, cat# HT501128) for 10 min. Brains were collected and post-fixed in the same fixative for 2 hr and cryoprotected overnight at 4°C in 0.1 M PBS containing 20% sucrose. Brains were cut into 40 μm sections on a vibratome or freezing microtome. Sections were rinsed in 0.1 M PBS, and then treated with 10% normal donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) and 0.3% Triton X-100 for 2 hr. Slides were incubated overnight at 4 °C in mouse anti-GAD67 (1:1000; Millipore Sigma, cat# MAB5406), rabbit anti-NeuN (1:500; Millipore Sigma, cat# ABN78), or mouse anti-bNOS (1:1000; Millipore Sigma, cat# N2280). The next day, sections were rinsed in 0.1 M PBS and incubated in Alexa Fluor 488-tagged donkey anti-mouse IgG or donkey anti-rabbit IgG (1:500, Thermo Fisher, cat# A-21202 and A-21206) for 1 hr at room temperature. Sections were then mounted on gelatin-subbed slides (SouthernBiotech, cat# SLD01-BX) and coverslipped using Fluoromount-G (SouthernBiotech, cat# 0100–01). Images were collected using a 1.30 NA 40x oil-immersion objective or a 1.40 NA 63x oil-immersion objective on a Leica TCS SP8 laser scanning confocal microscope.

The mouse monoclonal anti-GAD67 antibody (Millipore Sigma, cat# MAB5406) was raised against the 67 kDA isoform of glutamic acid decarboxylase (GAD). The manufacturer reports that Western blot analysis shows no cross-reactivity with the 65 kDa isoform of GAD (GAD65). This antibody has been previously used to identify GABAergicterm cells in the IC ( BeebepersName et al., 2016 ; Ito et al., 2009 ; Mellott et al., 2014 ). The mouse monoclonal anti-nitric oxide synthase-brain (bNOS) (Sigma, cat# N2280) was raised against the IgG1 isotype from the NOS-B1 hybridoma. The manufacturer reports that anti-bNOS reacts specifically with nitric oxide synthase (NOS), derived from brain (bNOS, 150 – 160 kDa). This antibody has been previously used, in guinea pig and mouse, to delineate the borders of the IC ( Coote and Rees, 2008 ; Keesom et al., 2018 ). To perform NeuN staining, we used a rabbit polyclonal antibody (Millipore Sigma, cat# ABN78). The manufacturer reports that anti-NeuN specifically recognizes the DNA-binding, neuron-specific protein NeuN, which is present in most central and peripheral neuronal cell types of all vertebrates tested. Previous studies reported the use of this antibody to label neurons in the IC ( BeebepersName et al., 2016 ; Foster et al., 2014 ; Mellott et al., 2014 ).

Images from representative sections of the IC (n = 3 animals, two sections per animal, one caudal and one middle) were collected at 2 µm depth intervals with a 1.30 NA 40x oil-immersion objective on a Leica TCS SP8 laser scanning confocal microscope. Images were analyzed using Fiji software ( Rueden et al., 2017 ; Schindelin et al., 2012 ). Consistent with previous studies, we found that the anti-GAD67 antibody did not penetrate the entire depth of the tissue sections ( BeebepersName et al., 2016 ; Mellott et al., 2014 ). We therefore restricted our analysis to the top 10 – 12 µm of each section, where the antibody was fully penetrant. Within this region, we manually marked every GAD67 + cell body and every tdTomato + cell body in the left IC. The green (GAD67) and red (tdTomato) color channels were analyzed separately, so that labeling in one channel did not influence analysis of the other channel. After cells were marked, the GAD67 and tdTomato color channels were merged, and every instance where a cell body contained markersterm for both GAD67 and tdTomato was counted. The number of double-labeled cells was compared to the total number of tdTomato + neurons to determine the percentage of tdTomato + neurons that were GAD67 + .

A design-based stereology approach was used to estimate the numbers of NeuN + and tdTomato + neurons in anti-NeuN stained sections ( Schmitz and Hof, 2005 ). To collect systematic random samples, a virtual 370 µm x 370 µm grid was overlaid on the IC section. The starting coordinates for the grid were set using the Mersenne Twister random number generator in Igor Pro 7 or 8 (WaveMetrics Inc). Images were then collected at coordinates determined by the upper-left intersection of each grid-square that fell over the left IC. Each image consisted of a 184 µm x 184 µm Z-stack collected at 1 µm depth intervals with a 1.40 NA 63x oil immersion objective on a Leica TCS SP8 confocal microscope. Six to sixteen images were collected per slice. Three slices were analyzed per mouse, with slices from each mouse evenly distributed along the rostral-caudal axis of the IC. Images were imported to Neurolucida 360 (MBF Bioscience), where neurons were counted using the optical fractionator approach ( West et al., 1991 ). In this approach, we determined the image planes corresponding to the top, center, and bottom of the slice in each image stack. Top and bottom regions of each slice (≥2 µm thick) were treated as guard zones and discarded from subsequent analysis. Removal of guard zones left a 15 µm-thick region at the center of the slice for subsequent analysis. Neurons within this region were counted by making a single mark at the top of each cell. Cells crossing the right and top borders of the image stack were counted, whereas those crossing the left and bottom borders were not. The green (NeuN) and red (tdTomato) color channels were analyzed separately, so that labeling in one channel did not affect analysis of the other. Next, the color channels were merged and cells with both NeuN and tdTomato markersterm were counted. In every instance, tdTomato + cells were also NeuN + (208/208 cells). The total number of double-labeled (tdTomato + /NeuN + ) cells was then compared to the total number of NeuN + cells.

Following transcardial perfusion as described previously, brains from three VIP-IRES-Cre x Ai14 mice were frozen and sectioned on a sliding microtome. Brains were cut into 40 µm sections, one each in the transverse, sagittal, and horizontal planes, and sections were collected in three series. The distribution of tdTomato-expressing (VIPterm + ) cells in one series from each case was analyzed using a Neurolucida system (MBF Bioscience, Williston, VT) attached to a Zeiss Axioimager.Z1 fluorescence microscope. Major IC subdivisions, including the central nucleus (ICcterm), dorsal cortex, (ICd) and lateral cortex (IClc), were identified by comparing bNOS and GAD67 immunostains with previous studies of mouse IC ( Dillingham et al., 2017 ; Meininger et al., 1986 ; Ono et al., 2016 ; Willard and Ryugo, 1983 ). Neurolucida Explorer was used to export drawings to Adobe Illustrator for figure preparation.

Mice of either sex were used, aged postnatal day (P) 21 to P70 for VIP-IRES-Cre x Ai14 crosses and P21 to P113 for C57BL/6J animals. Mice were deeply anesthetized with isoflurane, decapitated, and the brain was dissected quickly in ~34 °C artificial cerebrospinal fluid (ACSF) containing (in mM): 125 NaCl, 12.5 Glucose, 25 NaHCO 3 , 3 KCl, 1.25 NaH 2 PO 4 , 1.5 CaCl 2 , 1 MgSO 4 , bubbled to a pH of 7.4 with 5% CO 2 in 95% O 2 . Chemicals were obtained from Fisher Scientific or Millipore Sigma unless stated otherwise. Coronal or parasagittal slices (200 – 250 µm) containing the IC were cut in ~34 °C ACSF with a vibrating microtome (VT1200S, Leica Biosystems) and incubated at 34 °C for 30 min in a holding chamber filled with ACSF and bubbled with 5% CO 2 in 95% O 2 . After incubation, slices were stored at room temperature until used for recordings.

To make recordings, slices were placed in a recording chamber under a fixed stage upright microscope (BX51WI, Olympus Life Sciences) and were constantly perfused with 34 °C ACSF at ~2 ml/min. All recordings were conducted near physiological temperature (34 °C). IC neurons were patched under visual control using infrared Dodt gradient contrast and epifluorescence imaging. Recordings were performed with a BVC-700A patch clamp amplifier (Dagan Corporation). Data were low pass filtered at 10 kHz, sampled at 50 kHz with a National Instruments PCIe-6343 data acquisition board, and acquired using custom written algorithms in Igor Pro. For every recording, series resistance and pipette capacitance were corrected using the bridge balance circuitry of the BVC-700A. Recordings with a series resistance above 25 MΩ were discarded. All membrane potentials have been corrected for a liquid junction potential of 11 mV.

Electrodes were pulled from borosilicate glass (outer diameter 1.5 mm, inner diameter 0.86 mm, Sutter Instrument) to a resistance of 3.5 – 4.5 MΩ using a P-1000 microelectrode puller (Sutter Instrument). The electrode internal solution contained (in mM): 115 K-gluconate, 7.73 KCl, 0.5 EGTA, 10 HEPES, 10 Na 2 phosphocreatine, 4 MgATP, 0.3 NaGTP, supplemented with 0.1% biocytin (w/v), pH adjusted to 7.3 with KOH and osmolality to 290 mmol/kg with sucrose.

Input resistance was determined by delivering a series of 100 ms hyperpolarizing current steps incremented to elicit hyperpolarization ranging from just below the resting membrane potential to < −110 mV. For each response, the amplitudes of the peak (most negative value) and steady-state (average of last 10 ms of response) hyperpolarization were measured relative to the resting potential. Voltage versus current plots were prepared, and input resistance was determined from the slopes of lines fit to the peak ( R pk ) and steady-state ( R ss ) data for current steps that achieved a peak hyperpolarization between 0 and −15 mV relative to rest. Membrane time constantterm·var was determined by delivering 50 current steps at an amplitude that hyperpolarized the membrane potential by 1–3 mV. Current step duration was set to ensure that the membrane potential achieved a steady-state value before the end of the current step. An exponential function was then fit to onset of each response and the median time constant determined.

For electrical stimulation of commissural inputs, a glass electrode filled with ACSF was placed into the commissure near the midline, connected to a DS3 constant current, isolated stimulator (Digitimer Ltd.). PSPs were elicited with 100 µs electric shocks, ranging from 45 µA to 320 µA.

To isolate or manipulate synaptic events, the following pharmacological agents were used, all diluted in standard ACSF: 5 µM SR95531 (gabazine, GABA A receptor antagonist, Hello Bio), 1 µM strychnine hydrochloride (glycine receptor antagonist, Millipore Sigma), 50 µM D-AP5 (NMDA receptor antagonist, Hello Bio), 10 µM NBQX disodium salt (AMPA receptor antagonist, Hello Bio), 1 µM TTX (voltage dependent sodium channel blocker, Hello Bio).

Data analysis was performed using custom written algorithms in Igor Pro or MATLAB (Mathworks). Statistical tests were performed in R Studio (R Studio, Boston) for R 3.5.1 (The R Project for Statistical Computing, The R Foundation).

After recordings, the electrode was removed slowly to allow the cell membrane to reseal, and the slice was fixed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB, pH 7.4) for 12 – 24 hr. Slices were then washed in 0.1 M PB and stored in 0.1 M PB for up to three weeks. Recorded neurons were stained using fluorescent biocytin-streptavidin histochemistry. In brief, slices were washed in 0.1 M PB three times for 10 min (3 × 10 min in PB), permeabilized in 0.2% Triton X-100 in 0.1 M PB for 2 hr, washed 3 × 10 min in PB, and stained at 4 °C for 48 hr with streptavidin-Alexa Fluor 488 or 647, diluted 1:1000 in 0.1 M PB. Slices were then washed 3 × 10 min in PB and mounted on Superfrost Plus microscope slides in anti-fade media (Fluoromount-G). Z-stack images of streptavidin-Alexa Fluor labeled cells were obtained with a Leica TCS SP8 laser scanning confocal microscope using a 1.40 NA 63x oil-immersion objective. Three-dimensional reconstructions of neuronal morphologyterm and quantitative analyses of soma and dendrite shape were performed on image stacks imported into Neurolucida 360 (MBF Bioscience). To facilitate comparisons of neuronal morphologyterm, all reconstructed neurons are displayed as if they were in the left IC as viewed from a caudal perspective. Reconstructions of neurons that were located in the right IC were flipped along the dorsal-ventral axis so that they appear as if they were in the left IC.

Following biocytin-streptavidin histochemistry, tile scan images of the entire IC were collected using a 20x objective on a Leica TCS SP8 confocal microscope. These images were then used to determine medial-lateral and dorsal-ventral coordinates of recorded neurons. The medial-lateral coordinate was measured as the distance of the soma from the medial axis (midline) of the IC slice (x axis). The dorsal-ventral coordinate was measured as the distance of the soma from the dorsal-most edge of the IC slice (y axis). Neurons in the right IC were combined with those from the left IC by multiplying the medial-lateral coordinate of neurons from the right IC by −1. Neurons were assigned to IC subdivisions, including the central nucleus (ICcterm), dorsal cortex, (ICd) and lateral cortex (IClc), by comparing neuron location in 20x tile scans to IC subdivision borders determined from standard series of IC sections immunostained for bNOS or GAD67 ( Choy Buentello et al., 2015 ; Coote and Rees, 2008 ). Neuron coordinates were then compared to physiological parameters obtained during whole cell recordings. To test for correlations, data were fit with a plane using the Levenberg-Marquardt least squares methodterm·stat in Igor Pro. Fit quality was assessed with Pearson’s correlation coefficient and the adjusted R 2 . Fit significance (p value) was calculated based on the chi-squared statistic from the fit and the chi-squared cumulative distribution function.

Neuron morphologyterm was reconstructed as described above. To determine the ‘length’ and ‘width’ axes of the dendritic arbors, the set of coordinates describing the morphologyterm of the dendritic arbor of each neuron was exported from Neurolucida 360 (MBF Bioscience). Coordinates were imported to Igor Pro, where principal components analysis (PCA)term·stat was performed on either the x and y coordinates (2D PCA) or the x, y, and z coordinates (3D PCA). The orientation of the length and width axes was then derived from the first and second principal directions of the resulting eigenvector matrices. 2D PCA was used to determine the orientation of neurons within the coronal plane. 3D PCA was used to determine the axes to use for measuring dendritic arbor length/width ratios. For this, the spread of the dendritic arbor along the first and second principal directions was determined by rotating each morphologyterm coordinate set according to its eigenvector matrix, then calculating the range from the minimum to maximum coordinates along the x (length, first principal direction) and y (width, second principal direction) axes.

Intracranial virus injections were performed on mice age P21 – P35 using standard aseptic techniques. Throughout the procedure, mice were anesthetized with isoflurane and their body temperature maintained with a homeothermic heating pad. An injection of the analgesic carprofen (5 mg/kg, CarproJect, Henry Schein Animal Health) was delivered subcutaneously. The scalp was shaved and a rostro-caudal incision was made along the midline to expose the skull. Injection sites were mapped using stereotaxic coordinates relative to the lambda suture. A single craniotomy was performed using a micromotor drill (K.1050, Foredom Electric Co.) with a 0.5 mm burr (Fine Science Tools).

Viral constructs were injected with a NanoJect III nanoliter injector (Drummond Scientific Company) connected to a MP-285 micromanipulator (Sutter Instruments). Glass injection pipettes were prepared by pulling capillary glass (Drummond Scientific Company) with a P-1000 microelectrode puller (Sutter Instrument). The injector tip was cut to an opening of ~20 µm and beveled at 30° with a BV-10 pipette beveller (Sutter Instrument). Injectors were back-filled with mineral oil and then front-filled with virus. AAV1.Syn.Chronos-GFP.WPRE.bGH (University of Pennsylvania Vector Core, Lot# CS1027L, 2.986e13 genome copies (GC)/ml) was used for CRACM experiments. AAV1.CAG.FLEX.eGFP.WPRE.bGH (Allen Institute 854, University of Pennsylvania Vector Core, Lot# CS0922, 4.65e13 GC/ml) was used for axonal tract tracing. For CRACM experiments, the IC was injected via two penetrations. Virus deposits were made at 250 µm intervals along the dorsal-ventral axis, resulting in four deposits in penetration one and three deposits in penetration 2. At each depth, 20 nl of virus was deposited, resulting in seven virus deposits and a total load of 140 nl virus per injected IC. DCN injections were limited to two deposits of 20 nl virus. For axonal tracing studies, viral load was reduced to 40 nl total (20 nl per site) to achieve sparser labeling of neurons. Injections were made at the coordinates shown in Table 3 .

After injections were completed, the scalp was sutured with Ethilon 6–0 (0.7 metric) nylon sutures (Ethicon USA LLC), and the wound was treated with 0.5 – 1 ml 2% Lidocaine hydrochloride jelly (Akorn Inc). Once mice were ambulatory, they were returned to the vivarium where they were monitored daily until sutures fell out and the wound was completely healed.

Table 3 Stereotaxic coordinates for virus injections. All coordinates are relative to the lambda suture.

https://doi.org/10.7554/eLife.43770.020 Location X coordinate (caudal) Y coordinate (lateral) Z coordinates (depth) Right IC penetration 1 (CRACM) -900 µm 1000 µm 2250 - 1500 µm, 250 µm interval Right IC penetration 2 (CRACM) -900 µm 1250 µm 2250 - 1750 µm, 250 µm interval Right IC penetration 1 (axonal tracing) -900 µm 1000 µm 1850 µm, 2000 µm Right DCN -1325 µm 2150 µm 4750 µm, 4550 µm After allowing 3 – 4 weeks for Chronos expression, animals were used for in vitro slice electrophysiologyterm·mth experiments as described above, with the exception that after decapitation all steps were performed in red light and recordings were conducted in darkness or red light to limit Chronos activation. For standard CRACM experiments, recordings were targeted to VIP neuronsterm·def. In additional control experiments, recordings were targeted to non-fluorescent IC neurons in VIP-IRES-Cre x Ai14 crosses. During whole cell recordings, Chronos was activated by brief pulses of 470 nm light emitted by a blue LED coupled to the epi-fluorescence port of the microscope and delivered to the brain slice through a 0.80 NA 40x water immersion objective with a field number of 26.5 mm. Accordingly, the blue light spot had an area of 0.345 mm 2 . In all CRACM experiments, the soma of the recorded neuron was present in the field of view. Blue light flashes were 2 to 5 ms long, illuminated the entire field of a 0.80 NA 40x objective, and yielded optical power densities that ranged from 6 to 48 mW/mm². Optical power was set using a minimal stimulation protocol. In general, the shortest stimulus duration that elicited a PSP was chosen, combined with 120% of the optical power that was determined as the threshold to elicit PSPs. Yet, when using a 5 ms pulse with maximum optical power, EPSP kinetics did not change significantly (8 of 8 recorded neurons, data not shown) compared to the minimal stimulation paradigm. This suggests that light pulses elicited action potentials with consistent durations, regardless of optical power. Recording sweeps with light flashes were repeated 20 to 50 times in 0.5 – 1 s intervals to obtain average PSP values. During experiments to investigate receptor contribution to PSPs, drugs were washed in for at least 10 min before recording under the new condition. For each receptor antagonist, 7 – 8 washout experiments were conducted. In each case, drug effects reversed after washout (data not shown).

The right IC of VIP-IRES-Cre x Ai14 mice was injected with AAV1.CAG.FLEX.eGFP.WPRE.bGH and transcardially perfused 3 – 4 weeks later, as described above. Brains were frozen and sectioned at 40 µm on a sliding microtome. For some brains, sections were collected serially, and for others, sections were collected in three series. The brains were examined for eGFP-labeled axons and boutons, which were interpreted as VIPterm + projections originating in the IC. Some sections were counterstained with a fluorescent Nissl stain (Neurotrace 640/660, ThermoFisher, cat# N21483). Injection sites comprised a collection of eGFP-labeled cell bodies. Cases were included for analysis only if the eGFP labeled cell bodies were restricted to the IC. Images were collected on a Zeiss AxioImager.Z2 microscope. High magnification images were collected as z-stacks using a 1.40 NA 63X oil-immersion objective and structured illumination (Zeiss Apotome 2) for optical sectioning at 0.5 µm intervals. Images shown are maximum projections of collected stacks. Adobe Photoshop was used to colorize images, to globally adjust levels, and to add scale bars.

Add a comment + Open annotations. The current annotation count on this page is being calculated .

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 3, 4, 5, 7, and 8 and Table 2.

Adams JC (1979) Ascending projections to the inferior colliculus The Journal of Comparative Neurology 183 :519–538.

https://doi.org/10.1002/cne.901830305 PubMed Google Scholar Ascoli GA Alonso-Nanclares L Anderson SA Barrionuevo G Benavides-Piccione R Burkhalter A Buzsáki G Cauli B Defelipe J Fairén A Feldmeyer D Fishell G Fregnac Y Freund TF Gardner D Gardner EP Goldberg JH Helmstaedter M Hestrin S Karube F Kisvárday ZF Lambolez B Lewis DA Marin O Markram H Muñoz A Packer A Petersen CC Rockland KS Rossier J Rudy B Somogyi P Staiger JF Tamas G Thomson AM Toledo-Rodriguez M Wang Y West DC Yuste R Petilla Interneuronterm Nomenclature Group (2008) Petilla terminology: nomenclature of features of GABAergicterm interneuronsterm of the cerebral cortexterm Nature Reviews Neuroscience 9 :557–568.

https://doi.org/10.1038/nrn2402 PubMed Google Scholar Baumann VJ Lehnert S Leibold C Koch U (2013) Tonotopicterm organization of the hyperpolarization-activated current (Ih) in the mammalian medial superior olive Frontiers in Neural Circuits 7 :117.

https://doi.org/10.3389/fncir.2013.00117 PubMed Google Scholar BeebepersName NL Young JW Mellott JG SchofieldpersName BR (2016) Extracellular molecular markersterm and soma size of inhibitory neurons: evidence for four subtypes of GABAergicterm cells in the inferior colliculus The Journal of Neuroscience 36 :3988–3999.

https://doi.org/10.1523/JNEUROSCI.0217-16.2016 PubMed Google Scholar Brunso-Bechtold JK Thompson GC Masterton RB (1981) HRP study of the organization of auditory afferents ascending to central nucleus of inferior colliculus in cat The Journal of Comparative Neurology 197 :705–722.

https://doi.org/10.1002/cne.901970410 PubMed Google Scholar Calford MB Aitkin LM (1983) Ascending projections to the medial geniculate body of the cat: evidence for multiple, parallel auditory pathways through thalamus The Journal of Neuroscience 3 :2365–2380.

https://doi.org/10.1523/JNEUROSCI.03-11-02365.1983 PubMed Google Scholar Campbell JN Macosko EZ Fenselau H Pers TH Lyubetskaya A Tenen D Goldman M Verstegen AM Resch JM McCarroll SA Rosen ED Lowell BB Tsai LT (2017) A molecular census of arcuate hypothalamus and median eminence cell types Nature Neuroscience 20 :484–496.

https://doi.org/10.1038/nn.4495 PubMed Google Scholar Cant NB (2013) Patterns of convergence in the central nucleus of the inferior colliculus of the mongolian gerbil: organization of inputs from the superior olivary complex in the low frequency representation Frontiers in Neural Circuits 7 :29.

https://doi.org/10.3389/fncir.2013.00029 PubMed Google Scholar Cant NB Benson CG (2006) Organization of the inferior colliculus of the gerbil (Meriones unguiculatus): differences in distribution of projections from the cochlear nuclei and the superior olivary complex The Journal of Comparative Neurology 495 :511–528.

https://doi.org/10.1002/cne.20888 PubMed Google Scholar Capogna M (2014) GABAergicterm cell type diversity in the basolateral amygdala Current Opinion in Neurobiology 26 :110–116.

https://doi.org/10.1016/j.conb.2014.01.006 PubMed Google Scholar Chen C Cheng M Ito T Song S (2018) Neuronal organization in the inferior colliculus revisited with Cell-Type-Dependent monosynaptic tracing The Journal of Neuroscience 38 :3318–3332.

https://doi.org/10.1523/JNEUROSCI.2173-17.2018 PubMed Google Scholar Choy Buentello D Bishop DC Oliver DL (2015) Differential distribution of GABA and glycine terminals in the inferior colliculus of rat and mouse Journal of Comparative Neurology 523 :2683–2697.

https://doi.org/10.1002/cne.23810 PubMed Google Scholar Cichon J Blanck TJJ Gan WB Yang G (2017) Activation of cortical somatostatin interneuronsterm prevents the development of neuropathic pain Nature Neuroscience 20 :1122–1132.

https://doi.org/10.1038/nn.4595 PubMed Google Scholar Coomes DL SchofieldpersName BR (2004) Separate projections from the inferior colliculus to the cochlear nucleus and thalamus in guinea pigs Hearing Research 191 :67–78.

https://doi.org/10.1016/j.heares.2004.01.009 PubMed Google Scholar Coote EJ Rees A (2008) The distribution of nitric oxide synthase in the inferior colliculus of guinea pig Neuroscience 154 :218–225.

https://doi.org/10.1016/j.neuroscience.2008.02.030 PubMed Google Scholar Davis KA Ding J Benson TE Voigt HF (1996) Response properties of units in the dorsal cochlear nucleus of unanesthetized decerebrate gerbil Journal of Neurophysiology 75 :1411–1431.

https://doi.org/10.1152/jn.1996.75.4.1411 PubMed Google Scholar Dickson L Finlayson K (2009) VPAC and PAC receptors: from ligands to function Pharmacology & Therapeutics 121 :294–316.

https://doi.org/10.1016/j.pharmthera.2008.11.006 PubMed Google Scholar Dillingham CH Gay SM Behrooz R Gabriele ML (2017) Modular-extramodular organization in developing multisensory shell regions of the mouse inferior colliculus Journal of Comparative Neurology 525 :3742–3756.

https://doi.org/10.1002/cne.24300 PubMed Google Scholar Felix RA Gourévitch B Portfors CV (2018) Subcortical pathways: towards a better understanding of auditory disorders Hearing Research 362 :48–60.

https://doi.org/10.1016/j.heares.2018.01.008 PubMed Google Scholar Foster NL Mellott JG SchofieldpersName BR (2014) Perineuronal nets and GABAergicterm cells in the inferior colliculus of guinea pigs Frontiers in Neuroanatomy 7 :53.

https://doi.org/10.3389/fnana.2013.00053 PubMed Google Scholar Fujimoto H Konno K Watanabe M Jinno S (2017) Late postnatal shifts of parvalbumin and nitric oxide synthase expression within the GABAergicterm and glutamatergicterm phenotypes of inferior colliculus neurons Journal of Comparative Neurology 525 :868–884.

https://doi.org/10.1002/cne.24104 PubMed Google Scholar Gabernet L Jadhav SP Feldman DE Carandini M Scanziani M (2005) Somatosensory integration controlled by dynamic thalamocortical Feed-Forward inhibition Neuron 48 :315–327.

https://doi.org/10.1016/j.neuron.2005.09.022 Google Scholar Geis HR Borst JG (2013) Large GABAergicterm neurons form a distinct subclass within the mouse dorsal cortex of the inferior colliculus with respect to intrinsic properties, synaptic inputs, sound responses, and projections Journal of Comparative Neurology 521 :189–202.

https://doi.org/10.1002/cne.23170 PubMed Google Scholar Glendenning KK Masterton RB (1983) Acoustic chiasm: efferent projections of the lateral superior olive The Journal of Neuroscience 3 :1521–1537.

https://doi.org/10.1523/JNEUROSCI.03-08-01521.1983 PubMed Google Scholar González-Hernández T Mantolán-Sarmiento B González-González B Pérez-González H (1996) 3.0.CO;2-E" classterm="reference__title">Sources of GABAergicterm input to the inferior colliculus of the rat The Journal of Comparative Neurology 372 :309–326.

https://doi.org/10.1002/(SICI)1096-9861(19960819)372:2<309::AID-CNE11>3.0.CO;2-E PubMed Google Scholar Hashikawa T Kawamura K (1983) Retrograde labeling of ascending and descending neurons in the inferior colliculus. A fluorescent double labelling study in the cat Experimental Brain Research 49 :457–461.

https://doi.org/10.1007/BF00238787 PubMed Google Scholar Hassfurth B Magnusson AK Grothe B Koch U (2009) Sensory deprivation regulates the development of the hyperpolarization-activated current in auditory brainstem neurons European Journal of Neuroscience 30 :1227–1238.

https://doi.org/10.1111/j.1460-9568.2009.06925.x PubMed Google Scholar Hernández O Rees A Malmierca MS (2006) A GABAergicterm component in the commissure of the inferior colliculus in rat NeuroReport 17 :1611–1614.

https://doi.org/10.1097/01.wnr.0000236857.70715.be PubMed Google Scholar Herrera M Correa J (1988) Stellateterm cells and their axonal patterns in the central nucleus of the inferior colliculus of the cat (Felis domesticus) Journal FüR Hirnforschung 29 :393–402.

Google Scholar Ito T Bishop DC Oliver DL (2009) Two classesterm of GABAergicterm neurons in the inferior colliculus Journal of Neuroscience 29 :13860–13869.

https://doi.org/10.1523/JNEUROSCI.3454-09.2009 Google Scholar Ito T Hioki H Sohn J Okamoto S Kaneko T Iino S Oliver DL (2015) Convergence of lemniscal and local excitatory inputs on large GABAergicterm tectothalamic neurons Journal of Comparative Neurology 523 :2277–2296.

https://doi.org/10.1002/cne.23789 PubMed Google Scholar Ito T Oliver DL (2010) Origins of glutamatergicterm terminals in the inferior colliculus identified by retrograde transport and expression of VGLUT1 and VGLUT2 genes Frontiers in Neuroanatomy 4 :135.

https://doi.org/10.3389/fnana.2010.00135 PubMed Google Scholar Ito T Oliver DL (2012) The basic circuit of the IC: tectothalamic neurons with different patterns of synaptic organization send different messages to the thalamus Frontiers in Neural Circuits 6 :48.

https://doi.org/10.3389/fncir.2012.00048 PubMed Google Scholar Ito T Oliver DL (2014) Local and commissural IC neurons make axosomatic inputs on large GABAergicterm tectothalamic neurons Journal of Comparative Neurology 522 :3539–3554.

https://doi.org/10.1002/cne.23623 PubMed Google Scholar Jackman SL Beneduce BM Drew IR Regehr WG (2014) Achieving high-frequency optical control of synaptic transmission Journal of Neuroscience 34 :7704–7714.

https://doi.org/10.1523/JNEUROSCI.4694-13.2014 PubMed Google Scholar Joo KM Chung YH Kim MK Nam RH Lee BL Lee KH Cha CI (2004) Distribution of vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide receptors (VPAC1, VPAC2, and PAC1 receptor) in the rat brain The Journal of Comparative Neurology 476 :388–413.

https://doi.org/10.1002/cne.20231 PubMed Google Scholar Kato HK Gillet SN Isaacson JS (2015) Flexible sensory representations in auditory cortex driven by behavioral relevance Neuron 88 :1027–1039.

https://doi.org/10.1016/j.neuron.2015.10.024 PubMed Google Scholar Kato HK Asinof SK Isaacson JS (2017) Network-Level control of frequency tuning in auditory cortex Neuron 95 :412–423.

https://doi.org/10.1016/j.neuron.2017.06.019 PubMed Google Scholar Keesom SM Morningstar MD Sandlain R Wise BM Hurley LM (2018) Social isolation reduces serotonergic fiber density in the inferior colliculus of female, but not male, mice Brain Research 1694 :94–103.

https://doi.org/10.1016/j.brainres.2018.05.010 PubMed Google Scholar Klapoetke NC Murata Y Kim SS Pulver SR Birdsey-Benson A Cho YK Morimoto TK Chuong AS Carpenter EJ Tian Z Wang J Xie Y Yan Z Zhang Y Chow BY Surek B Melkonian M Jayaraman V Constantine-Paton M Wong GK Boyden ES (2014) Independent optical excitation of distinct neural populations Nature Methods 11 :338–346.

https://doi.org/10.1038/nmeth.2836 PubMed Google Scholar Kuchibhotla KV Gill JV Lindsay GW Papadoyannis ES Field RE Sten TA Miller KD Froemke RC (2017) Parallel processing by cortical inhibition enables context-dependent behavior Nature Neuroscience 20 :62–71.

https://doi.org/10.1038/nn.4436 PubMed Google Scholar Lee S Kruglikov I Huang ZJ Fishell G Rudy B (2013) A disinhibitory circuit mediates motor integration in the somatosensory cortex Nature Neuroscience 16 :1662–1670.

https://doi.org/10.1038/nn.3544 PubMed Google Scholar Lee SH Cox CL (2003) Vasoactive intestinal peptide selectively depolarizes thalamic relay neurons and attenuates intrathalamic rhythmic activity Journal of Neurophysiology 90 :1224–1234.

https://doi.org/10.1152/jn.00280.2003 PubMed Google Scholar Loftus WC Bishop DC Oliver DL (2010) Differential patterns of inputs create functional zones in central nucleus of inferior colliculus Journal of Neuroscience 30 :13396–13408.

https://doi.org/10.1523/JNEUROSCI.0338-10.2010 PubMed Google Scholar Ma CL Kelly JB Wu SH (2002) AMPA and NMDA receptors mediate synaptic excitation in the rat’s inferior colliculus Hearing Research 168 :25–34.

PubMed Google Scholar Ma WL Brenowitz SD (2012) Single-neuron recordings from unanesthetized mouse dorsal cochlear nucleus Journal of Neurophysiology 107 :824–835.

https://doi.org/10.1152/jn.00427.2011 PubMed Google Scholar Madisen L Zwingman TA Sunkin SM Oh SW Zariwala HA Gu H Ng LL Palmiter RD Hawrylycz MJ Jones AR Lein ES 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 Malmierca MS Blackstad TW Osen KK Karagülle T Molowny RL (1993) The central nucleus of the inferior colliculus in rat: a Golgi and computer reconstruction study of neuronal and laminar structure Journal of Comparative Neurology 333 :1–27.

https://doi.org/10.1002/cne.903330102 PubMed Google Scholar Malmierca MS Izquierdo MA Cristaudo S Hernández O Pérez-González D Covey E Oliver DL (2008) A discontinuous tonotopicterm organization in the inferior colliculus of the rat Journal of Neuroscience 28 :4767–4776.

https://doi.org/10.1523/JNEUROSCI.0238-08.2008 PubMed Google Scholar Meininger V Pol D Derer P (1986) The inferior colliculus of the mouse. A Nissl and Golgi study Neuroscience 17 :1159–1179.

https://doi.org/10.1016/0306-4522(86)90085-0 PubMed Google Scholar Mellott JG Foster NL Nakamoto KT Motts SD SchofieldpersName BR (2014) Distribution of GABAergicterm cells in the inferior colliculus that project to the thalamus Frontiers in Neuroanatomy 8 :17.

https://doi.org/10.3389/fnana.2014.00017 PubMed Google Scholar Mellott JG BeebepersName NL SchofieldpersName BR (2019) Bilateral projections to the thalamus from individual neurons in the inferior colliculus The Journal of Comparative Neurology 527: :1118 -–1126.

https://doi.org/10.1002/cne.24600 PubMed Google Scholar Merchán M Aguilar LA Lopez-Poveda EA Malmierca MS (2005) The inferior colliculus of the rat: quantitative immunocytochemical study of GABA and glycine Neuroscience 136 :907–925.

https://doi.org/10.1016/j.neuroscience.2004.12.030 PubMed Google Scholar Milstein AD Bloss EB Apostolides PF Vaidya SP Dilly GA Zemelman BV Magee JC (2015) Inhibitory gating of input comparison in the CA1 microcircuit Neuron 87 :1274–1289.

https://doi.org/10.1016/j.neuron.2015.08.025 PubMed Google Scholar Moore DR (1988) Auditory brainstem of the ferret: sources of projections to the inferior colliculus The Journal of Comparative Neurology 269 :342–354.

https://doi.org/10.1002/cne.902690303 PubMed Google Scholar Nakamoto KT Sowick CS SchofieldpersName BR (2013) Auditory cortical axons contact commissural cells throughout the guinea pig inferior colliculus Hearing Research 306 :131–144.

https://doi.org/10.1016/j.heares.2013.10.003 PubMed Google Scholar Nelken I Young ED (1994) Two separate inhibitory mechanisms shape the responses of dorsal cochlear nucleus type IV units to narrowband and wideband stimuli Journal of Neurophysiology 71 :2446–2462.

https://doi.org/10.1152/jn.1994.71.6.2446 PubMed Google Scholar Okoyama S Ohbayashi M Ito M Harada S (2006) Neuronal organization of the rat inferior colliculus participating in four major auditory pathways Hearing Research 218 :72–80.

https://doi.org/10.1016/j.heares.2006.04.004 PubMed Google Scholar Oliver DL (1984) Dorsal cochlear nucleus projections to the inferior colliculus in the cat: a light and electron microscopic study The Journal of Comparative Neurology 224 :155–172.

https://doi.org/10.1002/cne.902240202 PubMed Google Scholar Oliver DL (1987) Projections to the inferior colliculus from the anteroventral cochlear nucleus in the cat: possible substrates for binaural interaction The Journal of Comparative Neurology 264 :24–46.

https://doi.org/10.1002/cne.902640104 PubMed Google Scholar Oliver DL Kuwada S Yin TC Haberly LB Henkel CK (1991) Dendritic and axonal morphologyterm of HRP-injected neurons in the inferior colliculus of the cat The Journal of Comparative Neurology 303 :75–100.

https://doi.org/10.1002/cne.903030108 PubMed Google Scholar Oliver DL Winer JA Beckius GE Saint Marie RL (1994) Morphologyterm of GABAergicterm neurons in the inferior colliculus of the cat The Journal of Comparative Neurology 340 :27–42.

https://doi.org/10.1002/cne.903400104 PubMed Google Scholar Oliver DL Beckius GE Bishop DC Kuwada S (1997) 3.0.CO;2-6" classterm="reference__title">Simultaneous anterograde labeling of axonal layers from lateral superior olive and dorsal cochlear nucleus in the inferior colliculus of cat The Journal of Comparative Neurology 382 :215–229.

https://doi.org/10.1002/(SICI)1096-9861(19970602)382:2<215::AID-CNE6>3.0.CO;2-6 PubMed Google Scholar Book Oliver DL (2005) Neuronal Organization in the Inferior Colliculus In: Winer J. A, Schreiner C. E, editors. The Inferior Colliculus . Springer. pp. 66–114.

Google Scholar Oliver DL Morest DK (1984) The central nucleus of the inferior colliculus in the cat The Journal of Comparative Neurology 222 :237–264.

https://doi.org/10.1002/cne.902220207 PubMed Google Scholar Ono M Yanagawa Y Koyano K (2005) GABAergicterm neurons in inferior colliculus of the GAD67-GFP knock-in mouse: electrophysiological and morphological properties Neuroscience Research 51 :475–492.

https://doi.org/10.1016/j.neures.2004.12.019 PubMed Google Scholar Ono M Bishop DC Oliver DL (2016) Long-Lasting Sound-Evoked afterdischarge in the auditory midbrain Scientific Reports 6 :20757.

https://doi.org/10.1038/srep20757 PubMed Google Scholar Ono M Bishop DC Oliver DL (2017) Identified GABAergicterm and glutamatergicterm neurons in the mouse inferior colliculus share similar response properties The Journal of Neuroscience 37 :8952–8964.

https://doi.org/10.1523/JNEUROSCI.0745-17.2017 PubMed Google Scholar Orton LD Papasavvas CA Rees A (2016) Commissural gain control enhances the midbrain representation of sound location Journal of Neuroscience 36 :4470–4481.

https://doi.org/10.1523/JNEUROSCI.3012-15.2016 PubMed Google Scholar Orton LD Rees A (2014) Intercollicular commissural connections refine the representation of sound frequency and level in the auditory midbrain eLife 3 :e03764.

https://doi.org/10.7554/eLife.03764 Google Scholar Osen KK (1972) Projection of the cochlear nuclei on the inferior colliculus in the cat The Journal of Comparative Neurology 144 :355–371.

https://doi.org/10.1002/cne.901440307 PubMed Google Scholar Palmer AR Shackleton TM Sumner CJ Zobay O Rees A (2013) Classification of frequency response areas in the inferior colliculus reveals continua not discrete classesterm The Journal of Physiology 591 :4003–4025.

https://doi.org/10.1113/jphysiol.2013.255943 PubMed Google Scholar Paloff AM Usunoff KG Hinova-Palova DV (1992) Ultrastructure of Golgi-impregnated and gold-toned neurons in the central nucleus of the inferior colliculus in the cat Journal FüR Hirnforschung 33 :361–407.

Google Scholar Peruzzi D Sivaramakrishnan S Oliver DL (2000) Identification of cell types in brain slices of the inferior colliculus Neuroscience 101 :403–416.

https://doi.org/10.1016/S0306-4522(00)00382-1 PubMed Google Scholar Petreanu L Huber D Sobczyk A Svoboda K (2007) Channelrhodopsin-2-assisted circuit mapping of long-range callosal projections Nature Neuroscience 10 :663–668.

https://doi.org/10.1038/nn1891 PubMed Google Scholar Pfeffer CK Xue M He M Huang ZJ Scanziani M (2013) Inhibition of inhibition in visual cortex: the logic of connections between molecularly distinct interneuronsterm Nature Neuroscience 16 :1068–1076.

https://doi.org/10.1038/nn.3446 PubMed Google Scholar Pi HJ Hangya B Kvitsiani D Sanders JI Huang ZJ Kepecs A (2013) Cortical interneuronsterm that specialize in disinhibitory control Nature 503 :521–524.

https://doi.org/10.1038/nature12676 PubMed Google Scholar Portfors CV Mayko ZM Jonson K Cha GF RobertspersName PD (2011) Spatial organization of receptive fields in the auditory midbrain of awake mouse Neuroscience 193 :429–439.

https://doi.org/10.1016/j.neuroscience.2011.07.025 PubMed Google Scholar Pouille F Scanziani M (2001) Enforcement of temporal fidelity in pyramidal cells by somatic feed-forward inhibition Science 293 :1159–1163.

https://doi.org/10.1126/science.1060342 PubMed Google Scholar Prönneke A Scheuer B Wagener RJ Möck M Witte M Staiger JF (2015) Characterizing VIP neuronsterm·def in the barrel cortex of VIPcre/tdTomato mice reveals Layer-Specific differences Cerebral Cortexterm 25 :4854–4868.

https://doi.org/10.1093/cercor/bhv202 Google Scholar Ramachandran R Davis KA May BJ (1999) Single-unit responses in the inferior colliculus of decerebrate cats. I. classification based on frequency response maps Journal of Neurophysiology 82 :152–163.

https://doi.org/10.1152/jn.1999.82.1.152 PubMed Google Scholar Reetz G Ehret G (1999) Inputs from three brainstem sources to identified neurons of the mouse inferior colliculus slice Brain Research 816 :527–543.

https://doi.org/10.1016/S0006-8993(98)01230-X PubMed Google Scholar RobertspersName MT Seeman SC Golding NL (2013) A mechanistic understanding of the role of feedforward inhibition in the mammalian sound localization circuitry Neuron 78 :923–935.

https://doi.org/10.1016/j.neuron.2013.04.022 PubMed Google Scholar Rodríguez FA Read HL Escabí MA (2010) Spectral and temporal modulation tradeoff in the inferior colliculus Journal of Neurophysiology 103 :887–903.

https://doi.org/10.1152/jn.00813.2009 PubMed Google Scholar Rueden CT Schindelin J Hiner MC DeZonia BE Walter AE Arena ET Eliceiri KW (2017) ImageJ2: imagej for the next generation of scientific image data BMC Bioinformatics 18 :529.

https://doi.org/10.1186/s12859-017-1934-z PubMed Google Scholar Ryugo DK Willard FH Fekete DM (1981) Differential afferent projections to the inferior colliculus from the cochlear nucleus in the albino mouse Brain Research 210 :342–349.

https://doi.org/10.1016/0006-8993(81)90907-0 PubMed Google Scholar Saint Marie RL (1996) 3.0.CO;2-2" classterm="reference__title">Glutamatergicterm connections of the auditory midbrain: selective uptake and axonal transport of D-[3H]aspartate The Journal of Comparative Neurology 373 :255–270.

https://doi.org/10.1002/(SICI)1096-9861(19960916)373:2<255::AID-CNE8>3.0.CO;2-2 PubMed Google Scholar Book Saldaña E Merchán MA (2005) Intrinsic and commissural connections of the inferior colliculus In: Winer J. A, Schreiner C, editors. The Inferior Colliculus: With 168 Illustrations . New York, NY: Springer. pp. 155–181.

Google Scholar Sanchez JT Gans D Wenstrup JJ (2007) Contribution of NMDA and AMPA receptors to temporal patterning of auditory responses in the inferior colliculus Journal of Neuroscience 27 :1954–1963.

https://doi.org/10.1523/JNEUROSCI.2894-06.2007 Google Scholar Schindelin J Arganda-Carreras I Frise E Kaynig V Longair M Pietzsch T Preibisch S Rueden C Saalfeld S Schmid B Tinevez JY White DJ Hartenstein V Eliceiri K Tomancak P 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 Schmitz C Hof PR (2005) Design-based stereology in neuroscience Neuroscience 130 :813–831.

https://doi.org/10.1016/j.neuroscience.2004.08.050 PubMed Google Scholar SchofieldpersName BR (2001) 3.0.CO;2-X" classterm="reference__title">Origins of projections from the inferior colliculus to the cochlear nucleus in guinea pigs The Journal of Comparative Neurology 429 :206–220.

https://doi.org/10.1002/1096-9861(20000108)429:2<206::AID-CNE3>3.0.CO;2-X PubMed Google Scholar SchofieldpersName BR SchofieldpersName RM Sorensen KA Motts SD (2007) On the use of retrograde tracers for identification of axon collaterals with multiple fluorescent retrograde tracers Neuroscience 146 :773–783.

https://doi.org/10.1016/j.neuroscience.2007.02.026 PubMed Google Scholar SchofieldpersName BR BeebepersName NL (2019) Subtypes of GABAergicterm cells in the inferior colliculus Hearing Research 376 :1–10.

https://doi.org/10.1016/j.heares.2018.10.001 PubMed Google Scholar Sheward WJ Lutz EM Harmar AJ (1995) The distribution of vasoactive intestinal peptide2 receptor messenger RNA in the rat brain and pituitary gland as assessed by in situ hybridization Neuroscience 67 :409–418.

https://doi.org/10.1016/0306-4522(95)00048-N PubMed Google Scholar Sivaramakrishnan S Oliver DL (2001) Distinct K currents result in physiologically distinct cell types in the inferior colliculus of the rat The Journal of Neuroscience 21 :2861–2877.

https://doi.org/10.1523/JNEUROSCI.21-08-02861.2001 PubMed Google Scholar Sivaramakrishnan S Oliver DL (2006) Neuronal responses to lemniscal stimulation in laminar brain slices of the inferior colliculus Journal of the Association for Research in Otolaryngology 7 :1–14.

https://doi.org/10.1007/s10162-005-0017-4 PubMed Google Scholar Spirou GA Young ED (1991) Organization of dorsal cochlear nucleus type IV unit response maps and their relationship to activation by bandlimited noise Journal of Neurophysiology 66 :1750–1768.

https://doi.org/10.1152/jn.1991.66.5.1750 PubMed Google Scholar Stiebler I Ehret G (1985) Inferior colliculus of the house mouse. I. A quantitative study of tonotopicterm organization, frequency representation, and tone-threshold distribution The Journal of Comparative Neurology 238 :65–76.

https://doi.org/10.1002/cne.902380106 PubMed Google Scholar Stokes CC Isaacson JS (2010) From dendrite to soma: dynamic routing of inhibition by complementary interneuronterm microcircuits in olfactory cortex Neuron 67 :452–465.

https://doi.org/10.1016/j.neuron.2010.06.029 PubMed Google Scholar Sturm J Nguyen T Kandler K (2014) Development of intrinsic connectivity in the central nucleus of the mouse inferior colliculus Journal of Neuroscience 34 :15032–15046.

https://doi.org/10.1523/JNEUROSCI.2276-14.2014 PubMed Google Scholar Sun QQ Prince DA Huguenard JR (2003) Vasoactive intestinal polypeptide and pituitary adenylate cyclase-activating polypeptide activate hyperpolarization-activated cationic current and depolarize thalamocortical neurons in vitro The Journal of Neuroscience 23 :2751–2758.

https://doi.org/10.1523/JNEUROSCI.23-07-02751.2003 PubMed Google Scholar Taniguchi H He M Wu P Kim S Paik R Sugino K Kvitsiani D Kvitsani D Fu Y Lu J Lin Y Miyoshi G Shima Y Fishell G Nelson SB Huang ZJ (2011) A resource of cre driver lines for genetic targeting of GABAergicterm neurons in cerebral cortexterm Neuron 71 :995–1013.

https://doi.org/10.1016/j.neuron.2011.07.026 PubMed Google Scholar Tremblay R Lee S Rudy B (2016) GABAergicterm Interneuronsterm in the Neocortex: From Cellular Properties to Circuits Neuron 91 :260–292.

https://doi.org/10.1016/j.neuron.2016.06.033 PubMed Google Scholar Tyner CF (1975) The naming of neurons: applications of taxonomic theory to the study of cellular populations Brain, Behavior and Evolution 12 :75–96.

https://doi.org/10.1159/000124141 PubMed Google Scholar Vaudry D Falluel-Morel A Bourgault S Basille M Burel D Wurtz O Fournier A Chow BK Hashimoto H Galas L Vaudry H (2009) Pituitary adenylate cyclase-activating polypeptide and its receptors: 20 years after the discovery Pharmacological Reviews 61 :283–357.

https://doi.org/10.1124/pr.109.001370 PubMed Google Scholar Wallace MN Shackleton TM Palmer AR (2012) Morphological and physiological characteristics of laminar cells in the central nucleus of the inferior colliculus Frontiers in Neural Circuits 6 :.

https://doi.org/10.3389/fncir.2012.00055 PubMed Google Scholar Wallace ML Saunders A Huang KW Philson AC Goldman M Macosko EZ McCarroll SA Sabatini BL (2017) Genetically distinct parallel pathways in the entopeduncular nucleus for limbic and sensorimotor output of the basal ganglia Neuron 94 :138–152.

https://doi.org/10.1016/j.neuron.2017.03.017 PubMed Google Scholar West MJ Slomianka L Gundersen HJ (1991) Unbiased stereological estimation of the total number of neurons in thesubdivisions of the rat hippocampus using the optical fractionator The Anatomical Record 231 :482–497.

https://doi.org/10.1002/ar.1092310411 PubMed Google Scholar Book Willard FH Ryugo DK (1983) Anatomy of the central auditory system In: Willott J. F, editors. The Auditory Psychobiology of the Mouse . Springfield, Ill: C.C. Thomas. pp. 201–304.

Google Scholar Willott JF Urban GP (1978) Response properties of neurons in nuclei of the mouse inferior colliculus Journal of Comparative Physiology ? A 127 :175–184.

https://doi.org/10.1007/BF01352302 Google Scholar Winer JA Larue DT Diehl JJ Hefti BJ (1998) 3.0.CO;2-9" classterm="reference__title">Auditory cortical projections to the cat inferior colliculus The Journal of Comparative Neurology 400 :147–174.

https://doi.org/10.1002/(SICI)1096-9861(19981019)400:2<147::AID-CNE1>3.0.CO;2-9 PubMed Google Scholar Book Winer JA Schreiner C (2005) The Inferior Colliculus New York, NY: Springer.

Google Scholar Wu SH Ma CL Kelly JB (2004) Contribution of AMPA, NMDA, and GABA(A) receptors to temporal pattern of postsynaptic responses in the inferior colliculus of the rat The Journal of Neuroscience 24 :4625–4634.

https://doi.org/10.1523/JNEUROSCI.0318-04.2004 PubMed Google Scholar Young ED Brownell WE (1976) Responses to tones and noise of single cells in dorsal cochlear nucleus of unanesthetized cats Journal of Neurophysiology 39 :282–300.

https://doi.org/10.1152/jn.1976.39.2.282 PubMed Google Scholar Zeng H Sanes JR (2017) Neuronal cell-type classification: challenges, opportunities and the path forward Nature Reviews Neuroscience 18 :530–546.

https://doi.org/10.1038/nrn.2017.85 PubMed Google Scholar Zhang YP Oertner TG (2007) Optical induction of synaptic plasticity using a light-sensitive channel Nature Methods 4 :139–141.

https://doi.org/10.1038/nmeth988 PubMed Google Scholar Zheng QY Johnson KR Erway LC (1999) Assessment of hearing in 80 inbred strains of mice by ABR threshold analyses Hearing Research 130 :94–107.

https://doi.org/10.1016/S0378-5955(99)00003-9 PubMed Google Scholar Author details David GoyerpersName Kresge Hearing Research Institute, Department of Otolaryngology – Head and Neck Surgery, University of Michigan, Ann Arbor, United States Contribution Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Writing—original draft, Writing—review and editing Competing interests No competing interests declared "This ORCID iD identifies the author of this article:" 0000-0001-7547-8285 Marina A SilveirapersName Kresge Hearing Research Institute, Department of Otolaryngology – Head and Neck Surgery, University of Michigan, Ann Arbor, United States Contribution Conceptualization, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing Competing interests No competing interests declared Alexander P GeorgepersName Kresge Hearing Research Institute, Department of Otolaryngology – Head and Neck Surgery, University of Michigan, Ann Arbor, United States Contribution Formal analysis, Investigation, Methodology, Writing—review and editing Competing interests No competing interests declared Nichole L BeebepersName Department of Anatomy and Neurobiology, Northeast Ohio Medical University, Rootstown, United States Contribution Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing Competing interests No competing interests declared Ryan M EdelbrockpersName Department of Anatomy and Neurobiology, Northeast Ohio Medical University, Rootstown, United States Contribution Formal analysis, Investigation, Writing—review and editing Competing interests No competing interests declared Peter T MalinskipersName Kresge Hearing Research Institute, Department of Otolaryngology – Head and Neck Surgery, University of Michigan, Ann Arbor, United States Contribution Formal analysis, Investigation, Writing—review and editing Competing interests No competing interests declared Brett R SchofieldpersName Department of Anatomy and Neurobiology, Northeast Ohio Medical University, Rootstown, United States Contribution Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Writing—original draft, Writing—review and editing Competing interests No competing interests declared "This ORCID iD identifies the author of this article:" 0000-0002-0875-7759 Michael T RobertspersName Kresge Hearing Research Institute, Department of Otolaryngology – Head and Neck Surgery, University of Michigan, Ann Arbor, United States Contribution Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Writing—original draft, Writing—review and editing For correspondence microb@umich.edu Competing interests No competing interests declared "This ORCID iD identifies the author of this article:" 0000-0003-2835-8752 David GoyerpersName Michael Thomas RobertspersName Michael Thomas RobertspersName Brett R SchofieldpersName The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

We thank Gabriel Corfas and Carol F Elias for helpful discussions and advice. This work was supported by a Deutsche Forschungsgemeinschaft Research Fellowship (GO 3060/1–1, project number 401540516, to DG), an Emerging Research Grant from the Hearing Health Foundation (MTR), and National Institutes of Health Grants R56 DC016880 (MTR) and R01 DC004391 (BRS).

Animal experimentation: All experiments were approved by the University of Michigan Institutional Animal Care and Use Committee (protocol # PRO00006642 and PRO00008773) and were in accordance with NIH guidelines for the care and use of laboratory animals.

© 2019, GoyerpersName et al.

This article is distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use and redistribution provided that the original author and source are credited.