This manuscript describes an experimental protocol for evaluating the morphological characteristics and functional status of ribbon synapses in normal mice. The present model is also suitable for noise-induced and age-related cochlear synaptopathy-restricted models. The correlative results of previous mouse studies are also discussed.
Cochlear inner hair cells (IHCs) transmit acoustic signals to spiral ganglion neurons (SGNs) through ribbon synapses. Several experimental studies have indicated that hair cell synapses may be the initial targets in sensorineural hearing loss (SNHL). Such studies have proposed the concept of cochlear "synaptopathy", which refers to alterations in ribbon synapse number, structure, or function that result in abnormal synaptic transmission between IHCs and SGNs. While cochlear synaptopathy is irreversible, it does not affect the hearing threshold. In noise-induced experimental models, restricted damage to IHC synapses in select frequency regions is employed to identify the environmental factors that specifically cause synaptopathy, as well as the physiological consequences of disturbing this inner ear circuit. Here, we present a protocol for analyzing cochlear synaptic morphology and function at a specific frequency region in adult mice. In this protocol, cochlear localization of specific frequency regions is performed using place-frequency maps in conjunction with cochleogram data, following which the morphological characteristics of ribbon synapses are evaluated via synaptic immunostaining. The functional status of ribbon synapses is then determined based on the amplitudes of auditory brainstem response (ABR) wave I. The present report demonstrates that this approach can be used to deepen our understanding of the pathogenesis and mechanisms of synaptic dysfunction in the cochlea, which may aid in the development of novel therapeutic interventions.
Frequencies in the range of approximately 20‒20,000 Hz can be perceived as auditory stimuli by humans. Human hearing is normally most sensitive near 1,000 Hz, where average sound pressure level is 20 μPa in young adults (i.e., 0 decibels of sound pressure level [dB SPL]). In some pathological conditions, hearing loss is restricted to specific frequencies. For example, in the early stages of noise-induced hearing loss (NIHL), a “notch” (i.e., hearing threshold elevation) can be observed in the audiogram at 4 kHz1. Along the mammalian cochlear partition, its gradations of stiffness and mass produce an exponential frequency map, with high-frequency sound detection at the base of the cochlea and low-frequency detection at the apex2. Indeed, there is a cochlear place-frequency map along the basilar membrane, leading to what is known as tonotopic organization2,3. Each given place on the basilar membrane has the highest sensitivity to only one particular sound frequency, which is usually termed the characteristic frequency3,4, although responses to other frequencies can also be observed.
To date, various mouse models have been employed to investigate normal function, pathological processes, and therapeutic efficacy in the auditory system. Precise knowledge of physiological parameters in the mouse cochlea is a prerequisite for such studies of hearing loss. The mouse cochlea is anatomically divided into apical, middle, and basal turns, which correspond to different frequency regions. By labeling auditory nerve afferents at the cochlear nucleus to analyze their corresponding peripheral innervation sites in the cochlea, Müller et al. succeeded in establishing the cochlear place-frequency map in the normal mouse in vivo5. In the interval of 7.2–61.8 kHz, which corresponds to positions between 90% and 10% of the full length of the basilar membrane, the mouse cochlear place-frequency map can be described by a simple linear regression function, suggesting a relation between the normalized distance from the cochlear base and the logarithm of the characteristic frequency5. In laboratory mice, the place-frequency map can be used to explore the relationship between hearing thresholds within specific frequency ranges and cochleograms showing the numbers of missing hair cells in relative regions along the basilar membrane6. Importantly, the place-frequency map provides a positioning system for the investigation of minimal structural damage, such as damage to the ribbon synapses of hair cells at specific cochlear frequency locations in mice with peripheral auditory trauma7,8.
In the mammalian cochlea, ribbon synapses are comprised of a presynaptic ribbon, an electron-dense projection that tethers a halo of release-ready synaptic vesicles containing glutamate within the IHC, and a postsynaptic density on the nerve terminal of the SGN with glutamate receptors9. During cochlear sound transduction, deflection of the hair cell bundle results in IHC depolarization, which leads to glutamate release from IHCs onto the postsynaptic afferent terminals, thereby activating the auditory pathway. Activation of this pathway leads to the transformation of sound-induced mechanical signals into a rate code in the SGN10. Indeed, the IHC ribbon synapse is highly specialized for indefatigable sound transmission at rates of hundreds of Hertz with high temporal precision, and is of critical importance for presynaptic mechanisms of sound encoding. Previous studies have revealed that ribbon synapses vary greatly in size and number at different frequency regions in the adult mouse cochlea11,12, likely reflecting structural adaptation to the particular sound coding for survival needs. Recently, experimental animal studies have demonstrated that cochlear synaptopathy contributes to multiple forms of hearing impairments, including noise-induced hearing loss, age-related hearing loss, and hereditary hearing loss13,14. Thus, methods for identifying correlated changes in synaptic number, structure, and function at specific frequency regions have been increasingly employed in studies of auditory development and inner ear disease, using models generated via experimental manipulation of genetic or environmental variables15,16,17.
In the current report, we present a protocol for analyzing the synaptic number, structure, and function at a specific frequency region of the basilar membrane in adult mice. Cochlear frequency localization is performed using a given place-frequency map in combination with a cochleogram. The normal morphological characteristics of cochlear ribbon synapses are evaluated via presynaptic and postsynaptic immunostaining. The functional status of cochlear ribbon synapses is determined based on the suprathreshold amplitudes of ABR wave I. With minor alterations, this protocol can be used to examine physiological or pathological conditions in other animal models, including rats, guinea pigs, and gerbils.
All procedures were carried out in accordance with the NRC/ILAR Guide for the Care and Use of Laboratory Animals (8th Edition). The study protocol was approved by the Institutional Animal Care and Use Committee of Capital Medical University, Beijing, China.
1. Animal Selection
2. Hearing Assessment
3. Cochlear Tissue Processing
4. Immunofluorescence Staining
5. Morphological Evaluation of Cochlear Ribbon Synapses
6. Functional Evaluation of Cochlear Ribbon Synapses
ABR hearing tests were performed for 10 C57BL/6J mice (8 weeks of age) under anesthesia. ABRs were elicited using tone burst stimuli at 4, 8, 16, 32, and 48 kHz. The hearing threshold of each animal was visually detected by distinguishing at least one clear waveform in the ABR. All mice exhibited ABR thresholds in response to tone bursts, ranging between 25 and 70 dB SPL depending on the frequency of the stimulus. Our results indicated that the hearing threshold was lowest at 16 kHz (Figure 1), corresponding to approximately 43% distance from the cochlear apex (Figure 2), suggesting that acoustic sensitivity is significantly reduced in other cochlear regions.
Cochlear whole-mounts were isolated from temporal bones in adult mice under a stereo dissection microscope (Figure 2A). Whole-mounts of the auditory epithelium were dissected into three pieces, the lengths of which were measured and eventually converted into percent distance from the cochlear apex. The frequency location on the basilar membrane of each cochlear turn was calculated using a logarithmic function, as previously described5,6 (Figure 2B).
To evaluate the morphological characteristics of cochlear ribbon synapses, antibodies against CtBP2 and GluR2 were used to label presynaptic and postsynaptic structures, respectively. In normal ears of adult mice, immunostaining revealed juxtaposed pairs of synaptic ribbons and glutamate receptor patches studding the surface of the basolateral membrane of IHCs, with 8–20 pairs per IHC (Figure 3A). Although the vast majority of puncta appeared as juxtaposed pairs in normal ears, orphan ribbons could be observed rarely at high magnification (Figure 3B). Counts of IHC ribbon synapses (immunopositive spots for both CtBP2 and GluR2) were highest at the 16 kHz region, significantly decreasing as the distance from this location increased (Figure 3C). The synaptic counts determined based on confocal projections provide an estimate of the maximum number of auditory nerve fibers that transmit information from the cochlea to the brain29.
The functional status of cochlear ribbon synapses was investigated in all adult mice based on ABR wave I amplitudes, which provide information regarding the functional integrity of auditory nerve fibers29,31. ABR wave I amplitudes at each frequency of the stimulus presented at a sound pressure level of 90 dB were measured from peak to the following trough, as shown in Figure 4A. ABR wave I amplitude was highest at a frequency of 16 kHz, corresponding to the lowest hearing threshold, and amplitude values significantly decreased as distance from this location increased (Figure 4B). This result is consistent with the observed alterations in ribbon synapse counts, indicating that synapses within this cochlear region may exhibit the most vivid synaptic function. Furthermore, in previous mouse studies of noise-induced and age-related cochlear neurodegeneration, suprathreshold amplitudes of ABR wave I decreased in proportion to ribbon loss, indicating that ABR wave I amplitude is highly correlated with the degree of cochlear synaptopathy29,31.
Figure 1: Hearing Assessment. ABR threshold comparisons among different frequencies of tone burst stimuli demonstrated that the hearing threshold was lowest at 16 kHz in 10 adult C57BL/6J mice. The ABR response was significantly elevated at other frequency regions (one-way ANOVA with Dunnett’s multiple comparison post hoc test; *: P < 0.01, n = 20 ears). Data are expressed as the mean ± SEM. Please click here to view a larger version of this figure.
Figure 2: Cochlear frequency localization in mice. (A) A representative image of the full explanted cochlea, which has been dissected out of the temporal bone under a stereo dissection microscope. (B) The mouse cochlea is divided into apical, middle, and basal turns, where the cochlear lateral wall is removed. Red circles on the fragments of the cochlear basilar membrane indicate the frequency locations and their corresponding normalized positions in the cochlea (0% refers to the cochlear apex, 100% to the cochlear base). Scale bar = 250 μm. Please click here to view a larger version of this figure.
Figure 3: Confocal analysis of cochlear ribbon synapses in mice. (A) Representative images of ribbon synapses for the 4, 8, 16, 32, and 48 kHz regions, immunostained for presynaptic ribbons (CtBP2, red) and postsynaptic structures (GluR2, green). White dashed lines are used to outline inner hair cells for reference. Scale bar = 10 μm. (B) High-power thumbnails from confocal z-stacks show that these ribbon synapses appeared as closely juxtaposed pairs of CtBP2-positive (red) and GluR2-positive (green) puncta (left and middle), whereas orphan ribbons (right) lacking postsynaptic glutamate receptor patches were very rare. Scale bar = 0.5 μm. (C) Quantitative analysis of paired ribbon puncta with all presynaptic and postsynaptic elements revealed that the number of synaptic puncta per inner hair cell was significantly higher at the 16 kHz region than in other frequency regions (one-way ANOVA with Dunnett’s multiple comparison post hoc test; *: P < 0.01, n=6 ears). Data are expressed as the mean ± SEM. Please click here to view a larger version of this figure.
Figure 4: Analysis of ABR wave I amplitudes in mice. (A) Representative ABR waveform from an 8 weeks old C57BL/6J mouse exposed to a 16 kHz pure tone stimulus at an intensity of 90 dB (stimulus onset at 0 ms). Roman numerals mark the peaks of the ABR waves. Dotted lines mark the wave I peak and trough, indicating the amplitude. (B) Quantitative analysis of average wave I amplitudes in response to the 4, 8, 16, 32, and 48 kHz stimuli presented at a sound pressure level of 90 dB. ABR wave I amplitudes were highest at the frequency of 16 kHz and significantly lower at other frequencies (one-way ANOVA with Dunnett’s multiple comparison post hoc test; *: P < 0.01, n = 20 ears). Data are expressed as the mean ± SEM. Please click here to view a larger version of this figure.
Since cochlear synaptopathy was first characterized in adult mice with a temporary threshold shift (TTS) induced by 8‒16 kHz octave band noise at 100 dB SPL for 2 h31, researchers have increasingly investigated the effects of synaptopathy in various mammals, including monkeys and humans32,33. In addition to noise exposure, several other conditions have been associated with cochlear synaptopathy (e.g., aging, the use of ototoxic drugs, and genetic mutations), leading to short-term disruption of suprathreshold audition, followed by irreversible degeneration of the auditory nerve. In the early phase of cochlear insult, synaptopathy often occurs at a specific frequency location, especially in experimental models in which damage to IHC synapses is restricted to select frequency regions24,31. Therefore, our protocol is significant in that it enables the investigation of synaptic morphology and function at a specific frequency region.
The cochlear place-frequency map can be used to discriminate normal and abnormal auditory function, further reflecting normal and abnormal area in the inner ear. Since surface preparation techniques were first used to plot intact and missing hair cells in the different cochlear turns, the cochleogram has become a routine method for quantifying hair cell loss6. Therefore, to correlate morphological insults in the cochlea to relevant physiological changes, it is reasonable to include a place-frequency map on the cochleogram. This method allows one to determine the number and structure of synapses based on their positions along the cochlear basilar membrane in relation to established cochlear place-frequency maps, thereby providing sufficient information for detailed comparisons between histological and physiological outcomes. However, some studies assess synaptopathy based on cochlear segment or distance along the cochlear duct, which does not allow for direct comparisons among individual cochleae due to intra-species variations in basilar membrane length. Thus, it is particularly important to standardize the cochleogram by converting individual cochlear lengths from millimeters to a relative percentage.
Synaptopathy can be confirmed via visualization of immunostaining for CtBP2 protein (the synaptic isoform “ribeye”) in the basolateral regions of IHCs. These CtBP2-positive patches can serve as a structural marker for quantification of presynaptic ribbons, and the presence of fewer patches usually indicates presynaptic loss. Although previous studies have reported that 98% of presynaptic ribbons are paired with post-synaptic terminals in the normal ear30, the simple number of CtBP2-positive patches may not be accurate for quantitative analysis of complete synapses, as this may lead to overestimation of synapse counts in the injured ear by involving “orphans” (presynaptic ribbons unpaired with postsynaptic terminals). To improve the accuracy of estimating synapse number, additional antibodies against postsynaptic structure such as GluA2, GluA2/3, or PSD-95 are used. The postsynaptic density protein PSD-95, a membrane-associated guanylate kinase (MAGUK) scaffolding protein, can be labeled using an antibody against PSD-95, which is mostly observed at the contacts between hair cells and SGN fiber endings34,35. However, the antibody against GluR2 is more specific for AMPA-type ionotropic glutamate receptors in the postsynaptic membrane, which can more reliably identify ribbon synapses (juxtaposed pairs of immunofluorescent puncta of presynaptic CtBP2 and postsynaptic GluR2)15. In addition to morphological analysis, histological analysis of complete synapses can be used as a functional indicator for synaptopathy. In adult mice with cochlear synaptopathy, reductions in ABR wave I amplitude following the presentation of moderate- to high-level tone stimuli occur at frequencies tonotopically related to regions of synaptic loss26,29. Synapse counts obtained using this method are limited in that it takes approximately 10 minutes to scan one site on the basilar membrane. Furthermore, to evaluate spatial variations in synaptic number and morphology based on IHC location, it is necessary to accurately identify the boundary of the ICH body (e.g., via myosin VIIa staining) and to perform particular image-processing steps26. Using such methods, previous studies have confirmed that synaptic loss is greater on the modiolar side of the IHC than on the pillar side in animal models of noise-induced degeneration26.
Previous studies have demonstrated that low spontaneous-rate, high-threshold fibers are more susceptible to noise damage than high spontaneous-rate, low-threshold fibers in animal models of restricted synaptopathy26,30. These studies provide the rationale for using suprathreshold amplitudes of ABR wave I (measured using moderate- to high-level tone stimuli) to assess synaptic function in animal models with normal ABR threshold and distortion product otoacoustic emissions (DPOAEs). Because wave I of the ABR reflects the summed neural responses of auditory nerve fibers, when synaptopathy is assessed via ABR wave I amplitude, DPOAE testing should also be performed to exclude OHC damage, which also can reduce ABR amplitudes due to the disruption of mechanoelectric transduction. Although the first wave of the compound action potential (CAP), which is measured from round-window electrodes, also represents summed activity of the cochlear nerve, this method is more invasive and complex than ABR measurements. If DPOAE responses return to normal after temporary threshold shifts in noise-exposed mice31 or they have not yet deteriorated in aging mice29, the suprathreshold amplitude of ABR wave I can strongly predict the degree of cochlear synaptopathy, since affected neurons are silenced when their synaptic connections to IHCs are disrupted. However, if cochlear sensitivity is reduced due to various factors (e.g., OHC dysfunction), decreases in ABR wave I amplitude can no longer be attributed solely to synaptic loss because they will reflect the combination of all cochlear damage. Therefore, controlled experimental conditions for synaptopathy-restricted model preparation are required to ensure that cochlear sensitivity is not compromised by other factors. Unfortunately, although ABR wave I amplitude provides an objective measure of auditory nerve fiber loss in animals, it is difficult to measure in humans. Moreover, mixed pathologies involving synaptopathy, loss of hair cells, and the presence of other abnormalities in the cochlea may co-occur in humans, limiting the use of ABR wave I amplitude measurements in clinical settings. ABR wave latency is calculated as the time in milliseconds from the onset of the stimulus to the positive peak of each wave, providing insight into the transmission times along the auditory pathway. No significant changes in wave I latency have been observed in mouse models of noise-induced or age-related cochlear synaptopathy36. However, some evidence suggests that the effects of masking noise on ABR wave V latency can be used to diagnose cochlear synaptopathy in humans37.
Several recent studies have supported the notion that cochlear synaptopathy is the primary initial event associated with hidden hearing loss, tinnitus, and hyperacusis. Although the concept of cochlear synaptopathy in inner ear diseases has now been firmly established, its detailed impact on hearing ability remains unknown. The protocol presented in the current study enables the investigation of morphology and function in cochlear ribbon synapses within a specific frequency region. Thus, this protocol can be used to investigate cochlear synaptopathy, its underlying mechanisms, and the efficacy of potential therapeutic interventions in various experimental animal models.
The authors have nothing to disclose.
This work was supported by the National Natural Science Foundation of China (81770997, 81771016, 81830030); the joint funding project of Beijing Natural Science Foundation and Beijing Education Committee (KZ201810025040); the Beijing Natural Science Foundation (7174291); and the China Postdoctoral Science Foundation (2016M601067).
Ketamine hydrochloride | Gutian Pharmaceutical Co., Ltd., Fujian, China | H35020148 | 100mg/kg |
Xylazine hydrochloride | Sigma-Aldrich, St. Louis, MO, USA | X-1251 | 10mg/kg |
TDT physiology apparatus | Tucker-Davis Technologies, Alachua, FL, USA | Auditory Physiology System III | |
SigGen/BioSig software | Tucker-Davis Technologies, Alachua, FL, USA | Auditory Physiology System III | |
Electric Pad | Pet Fun | 11072931136 | |
Dumont forceps 3# | Fine Science Tools, North Vancouver, B.C., Canada | 0203-3-PO | |
Dumont forceps 5# | Fine Science Tools, North Vancouver, B.C., Canada | 0209-5-PO | |
Stereo dissection microscope | Nikon Corp., Tokyo, Japan | SMZ1270 | |
Goat serum | ZSGB-BIO, Beijing,China | ZLI-9021 | |
Anti-glutamate receptor 2, extracellular, clone 6C4 | Millipore Corp., Billerica, MA, USA | MAB397 | mouse |
Purified Mouse Anti-CtBP2 | BD Biosciences, Billerica, MA, USA | 612044 | mouse |
Alexa Fluor 568 goat anti-mouse IgG1antibody | Thermo Fisher Scientific Inc., Waltham, MA, USA | A21124 | goat |
Alexa Fluor 488 goat anti-mouse IgG2a antibody | Thermo Fisher Scientific Inc., Waltham, MA, USA | A21131 | goat |
Mounting medium containing DAPI | ZSGB-BIO, Beijing,China | ZLI-9557 | |
Confocal fluorescent microscopy | Leica Microsystems, Wetzlar, Germany | TCS SP8 II | |
Image Pro Plus software | Media Cybernetics, Bethesda, MD, USA | version 6.0 | |
Professional diagnostic pocket otoscope | Lude Medical Apparatus and Instruments Trade Co., Ltd., Shanghai,China | HS-OT10 | |
Needle electrode | Friendship Medical Electronics Co., Ltd., Xi'an,China | 1029 | 20 mm, 28 G |
Closed-field speaker | Tucker-Davis Technologies, Alachua, FL, USA | CF1 |