Morphology changes in the cochlea of impulse noise-induced hidden hearing loss

Abstract

Background

This study was focused on impulse noise induces hidden hearing loss.

Objectives

This study was designed to determine the morphology changes of noise-induced hidden hearing loss (NIHHL).

Method

Fifteen guinea pigs were divided into three groups: noise-induced hidden hearing loss (NIHHL) group, noise-induced hearing loss (NIHL) group, and normal control group. For the NIHHL group, guinea pigs were exposed to 15 times of impulse noise with peak intensity of 163 dB SPL at one time. For the NIHL group, animals were exposed to two rounds of 100 times impulse noise, and the time interval is 24 h. Auditory brain response (ABR) was tested before, immediately, 24 h, one week, and one month after noise exposure to evaluate cochlear physiology changes. One month after noise exposure, all guinea pigs in three groups were sacrificed, and basement membranes were carefully dissected immediately after ABR tests. The cochlea samples were observed by transmission electron microscopy (TEM) to find out the morphology changes.

Result

The ABR results showed that 15 times of impulse noise exposure could cause NIHHL in guinea pigs and 200 times could cause completely hearing loss. Impulse noise exposure could cause a dramatic increase of mitochondria in the inner hair cell. The structures of ribbon synapse and heminode were also obviously impaired compared to the normal group. The nerve fiber and myelin sheath remained intact after impulse noise exposure.

Conclusion

This research revealed that impulse noise could cause hidden hearing loss, and the changes in inner hair cells, ribbon synapse, and heminode all played a vital role in the pathogenesis of hidden hearing loss.

Chinese Abstract

背景：这项研究的重点是脉冲噪声引起的隐性听力损失。

目的：本研究旨在确定噪声引起的隐性听力损失（NIHHL）的形态变化。

方法：15只豚鼠分为三组：噪声引起的隐性听力损失(NIHHL) 组、噪声引起的听力损失 (NIHL) 组和正常对照组。对于 NIHHL组, 豚鼠暴露于 15 次脉冲噪声, 一次峰值强度为 163 dB SPL 。对于 NIHL 组, 动物暴露于两轮 100 次的脉冲噪声中, 时间间隔为 24 小时。听觉大脑反应 (ABR) 在噪声暴露之前、立即、24 小时、一周和一个月后进行测试, 以评估耳蜗生理变化。噪声暴露一个月后, 将三组豚鼠全部处死, ABR测试后立即仔细解剖基底膜。通过透射电子显微镜（TEM）观察耳蜗样品以找出形态变化。

结果：ABR结果显示, 15次脉冲噪声暴露可导致豚鼠的NIHHL, 200次可能导致完全听力损失。脉冲噪声暴露可能导致内毛细胞中线粒体的急剧增加。带状突触和半结的结构与正常组相比也明显受损。神经纤维和髓鞘脉在冲噪声暴露后保持完好。

结论：这项研究表明, 脉冲噪声可能导致隐性听力损失, 而内毛细胞、带状突触和半结的变化都在隐性听力损失发病机制中起重要作用。

Keywords:

• Hidden hearing loss
• inner hair cell
• ribbon synapse
• heminode
• impulse noise
• transmission electron microscopy

Introduction

Exposure to an episode of low intensity noise can damage the hearing that audiometry can't detect. This kind of silent damage to the acoustic system is called hidden hearing loss (HHL) [1]. Since the first time HHL was put forward, many studies on the pathogenesis have been reported. Among these studies, broadband noises (98-110dB, 8-32 kHz, 1–2 h exposure) were widely adopted [2–5]. Apart from broadband noise, impulse noise is another kind of noise that disturbed many people, especially those of certain occupations such as miners, building workers, and soldiers. At the same time, the study on damages caused by impulse noise makes sense for military medicine.

According to the studies reported recently, the pathology in ribbon synapse that between inner hair cell and type II auditory never fiber, referred to as cochlear synaptopathy, have been considered the main pathogenesis for HHL [6]. Another research group revealed that transient auditory nerve demyelination could also cause HHL [7]. They found that transient auditory nerve demyelination can damage the normal structure of heminode, which is formed by the first Schwann cell on the auditory nerve fiber. Animal that recovered from transient auditory nerve demyelination have normal hearing threshold and reduced ABR peak I amplitude, which represents the summed electronic activity of the SGN. At the moment, the reduced suprathreshold amplitude of the ABR peak I has been widely adopted as the auditory diagnostic criteria for HHL [8–10].

Our previous study revealed that impulse noise could cause HHL in guinea pig [11]. However, we still don’t know its impact on inner hair cell (IHC), ribbon synapse, heminode, and auditory nerve fiber. Thus, in the present study, we used transmission electron microscopy (TEM) to observe the morphology changes in the structures mentioned above to shed some light on the pathogenesis of NIHHL.

Materials and methods

Animals

In total, 15 healthy adult albino guinea pigs (250–300 g) of either sex were used in this experiment. They all passed the Preyer reflex test and otoscopic exam. Simultaneously, all guinea pigs’ hearing statuses were evaluated by the auditory brainstem response (ABR) threshold, and those with abnormal results were abandoned. Fifteen guinea pigs were randomly assigned to three experiment groups: Group A: noise-induced hearing loss (NIHL), n = 5; Group B: noise-induced hidden hearing loss (NIHHL), n = 5; Group C: normal control, n = 5. All of the animals were bought from Vital River Laboratory Animal Technology Co. Ltd. (Beijing, China). Care and use of the albino guinea pig in this study were approved by the Institutional Animal Care and Use Committee of the Chinese PLA General Hospital.

Noise exposure and procedures

Impulse noise was adopted in this study. The impulse noise was generated by a special electronic device (BDMS1-040528) designed by the College of Architecture, Tsinghua University. The impulse noise was conducted by a mental barrel (length 25 cm, diameter 10 cm). Its peak sound intensity was 163 dB SPL, and the pulse width was 0.25 s. The interval between two consecutive impulse noises was 6.5 s. The acoustic spectrum of the impulse noise was shown in Figure 1. The high-intensity portion was mainly distributed between 1.6 and 20 kHz due to the generator's physical and electrical characteristics. The noise levels were calibrated by a 1/4 inch microphone (B&K type 4136) and a precision sound analyzer (Norsonic Nor140).

Figure 1. The sound intensity spectrum of impulse noise used in this study.

During the exposure, non-anesthetized animals were placed in a special restraint cage in the sound booth and the distance from the barrel to the ear was set at 5 cm. The animals’ heads were carefully fixed to ensure that they couldn’t move during the exposure and each ear received equal intensity of impulse noise. The animals in the NIHHL group were exposed to 15 times of impulse noise. The animals in the NIHL group were exposed to 200 times of impulse noise, which were equally divided into two rounds with a 24-hour interval.

Auditory brainstem response (ABR) measurements

Electrophysiological measurements were carried out in a double-walled sound booth. For ABR recordings, the animals were inhalational anesthetized with isoflurane (Keyue Life Science Co. Ltd. Beijing, China), and a thermostatic heating pad was used to maintain body temperature. Three subdermal needle electrodes were used to record ABRs. The reference electrode was inserted beneath the recording ear's pinna, the ground electrode beneath the opposite ear, and the recording electrode beneath the head skin at the middle point of two pinnae.

Stimulus generation and bio-signal acquisition were obtained with Tucker-Davis Technologies (TDT RZ6) hardware and BioSig software. Acoustic stimuli used in this study were: (1) tone bursts 4 kHz, 8 kHz, and 16 kHz of 4-ms duration, 0.5- ms rise/fall time; and (2) click of 10-ms duration with band-pass-filtered between 100 Hz and 3000 Hz. Stimulus was produced through an electronic speaker (MF1 SN: 1250), which was placed 0.5 cm to the recording ear’s external auditory canal. Evoked response was preamplified by a TDT RA4PA and repeated 1024 times. The threshold was detected from 90 dB SPL with a 5 dB SPL descending step and was determined as the lowest level that a repeatable wave III response could be obtained at each frequency. The sound level for the suprathreshold stimulus was set at 70 dB SPL.

Morphology

After the electrophysiological measurement, the guinea pig was sacrificed, and its cochlea was used for transmission electron microscopy observation. The cochlea was perfused with 2.5% (vol/vol) glutaraldehyde and immersed in the fixative for 12 h at 4 °C. Then decalcification in 10% EDTA for 48 h. The basilar membrane was carefully dissected and fixed by 2.5% glutaraldehyde with Phosphate Buffer (PB) (0.1 M, pH 7.4), washed four times in PB. Then basilar membranes were first immersed in 1% (wt/vol) OsO4 and 1.5% (wt/vol) potassium ferricyanide aqueous solution at 4 °C for one h. After washing, they were incubated in filtered 1% thiocarbohydrazide (TCH) aqueous solutions (Sigma-Aldrich) at room temperature for 30 min, 1% unbuffered OsO4 aqueous solution at 4 °C for one h and 1% UA aqueous solution at 4 °C overnight. Then tissues were dehydrated through graded alcohol (30, 50, 70, 80, 90, 100%, 100%, 10 min each) into pure acetone (2 × 10min). Samples were infiltrated in graded mixtures (3:1, 1:1, 1:3) of acetone and SPI-PON812 resin (19.6 mL SPI-PON812, 6.6 mL DDSA, and 13.8 mL NMA), then changed pure resin. Finally, cells were embedded in pure resin with 1.5% BDMA and polymerized for 12 h at 45 °C, 48 h at 60 °C. The ultrathin sections (70 nm thick) were sectioned with a microtome (Leica EM UC6), stained by lead citrate, and examined by a transmission electron microscope (FEI Tecnai Spirit120kV).

Statistical analysis

All data were expressed as mean ± standard deviation (SD). Data were analyzed using unpaired t-test (Student’s t-test) for two groups and one-way ANOVA with Bonferroni post hoc tests for multiple comparisons. Graphpad prism 8 for Windows was used for statistical analyses and graphs. The significance level was indicated by the number of asterisks (*: p < .05, **: p < .01, and ***: p < .001, respectively).

Result

Auditory brainstem response

In this study, hearing thresholds were evaluated by click, 4k, 8k, and 16k evoked ABR. Before noise exposure, all guinea pigs passed ABR tests. Immediately after impulse noise exposure, the NIHHL group had an average ABR threshold increase of 15.5, 16.5, 17.5, and 8 dB at click, 4k, 8k, and 16k. The ABR threshold of guinea pigs in the NIHL group could not be detected at 90 dB. ABR threshold was then tested one day, one week, and one month after noise exposure subsequently. The results showed that hearing thresholds were gradually recovered in the NIHHL group. Compared with the normal group, the ABR threshold tested one month after noise showed no significant difference in all frequencies, which ‘traditionally’ represented a complete recovery of hearing. Simultaneously, ABR threshold still couldn’t be tested at all frequencies in the NIHL group even one month after exposure, which represented a complete hearing loss (Figure 2(a)).

Figure 2. Impulse noise resulted in permanent impairments of auditory function. ABR tests were performed before, immediately, 1 day, 1 week and 1 month after noise exposure. Results showed that immediately, 1 day and 1 week after noise exposure, thresholds of animals in NIHHL were increased. One month later, results were recovered to the normal level (a). At all-time points after noise in NIHHL group, ABR P1 supra-threshold latencies (b) are reduced; ABR P1 supra-threshold amplitudes (c) are reduced. Hearings of animals in NIHL group couldn’t be detected. Supra-threshold level was set at 70 dB SPL. NIHL: noise induced hearing loss group; NIHHL: noise induced hidden hearing loss group; Normal: normal group. *: p < .05; **: p < .01 and ***: p < .001 NIHHL versus Normal n = 10.

To further evaluate the impact of impulse noise on suprathreshold hearing function, we tested ABR peak I latency and amplitude, which represents the conduction velocity of the auditory nerve and the summed electronic activity of the spiral ganglion neurons (SGN) [7]. After impulse noise exposure, suprathreshold sound-evoked ABR peak I latency and amplitude were significantly affected in the NIHHL group at all tested frequencies. The results showed that immediately after impulse noise, the latency of ABR P I was significantly prolonged, and P I amplitude was significantly reduced. Unlike the ABR threshold, the P I latency (Figure 2(b)) and amplitude (Figure 2(c)) of guinea pigs in the NIHHL group showed no recovery until one month after noise exposure.

Inner hair cell

One month after noise exposure, the number of mitochondria in inner hair cell was significantly changed in the NIHL and NIHHL groups (p < .01). Compared with the normal group, the number of mitochondria significantly increased in the NIHL group (p < .01) and the NIHHL group (p < .01) (Figure 3(d)). Simultaneously, mitochondria were equally distributed in normal guinea pigs’ inner hair cells, while in the NIHL group and NIHHL group, they were bipolar distributed (Figure 3(a–c)).

Figure 3. TEM images of IHCs in genuine pigs one month after noise exposure. Impulse noise resulted in increase and redistribution of chondriosomes in IHC. One month after impulse noise exposure, chondriosomes of animals in NIHHL and NIHL were significantly increased (d) and obviously bipolar distributed. NIHL: noise induced hearing loss group; NIHHL: noise induced hidden hearing loss group; Normal: normal group. #: p < .05, NIHHL versus Normal; ***: p < .001 NIHL versus NIHHL versus Normal n = 5.

Ribbon synapse

The synapse between inner hair cell and type I spiral ganglion neuron is called ribbon synapse, which is characterized by the presynaptic ribbon-like structure. This research mainly focused on the low spontaneous rate (Low-SR) ribbon synapse, which is located in the modiolar side of the inner hair cell and is more vulnerable to noise. In normal group, the typical ribbon synapse contains a rugby-ball-like ribbon structure, around which anchored synapse vesicles contain glutamate (Figure 4(a)). One month after impulse noise exposure, part of the Low-SR ribbon synapses in the NIHHL group were impaired (Figure 4(f)). The typical ribbon structure was replaced by a bulk of irregular-shaped electron-dense substance on the presynaptic membrane, with synapse vesicles anchored loosely around it. At the same time, some ribbon synapses remained intact (Figure 4(b, e)). Compared with the normal group, no difference was found in the post-synapse structure. In the NIHL group, the ribbon synapse was more seriously damaged. Though typical post synapse structure can be found in the modiolar side of the inner hair cell, no ribbon structure was found through transmission electron microscopy observation of consecutive sections in adjacent areas, which means high-intensive impulse noise exposure can destroy the presynaptic ribbon structure (Figure 4(c, g)).

Figure 4. TEM images of Low-SR ribbon synapses in genuine pigs one month after noise exposure. Compared with normal group (a), ribbon synapses in NIHHL group showed different degrees of impairment (b). Some ribbon synapses are still normal (e) and some are obviously damaged (f). In the NIHL group, the pre-synaptic ribbon structures were disappeared (c). NIHL: noise induced hearing loss group; NIHHL: noise induced hidden hearing loss group; Normal: normal group.

Heminode

About 20 μm away from the Habenula Perforata (HP), Schwann cells begin to surround the auditory nerve fibers and form the myelin sheath. Different from the Ranvier's node that formed by the adjacent two Schwann cells, heminode is a special structure formed by the first Schwann cell on the auditory nerve axon. In the normal group, the ladder-like structure around the auditory axon was heminode (Figure 5(a, d)). The major structure of the heminode was composed of regularly arranged myelin lamellae, and these myelin lamellae were separated by an almost equal gap and formed loops spiral around the auditory nerve axon. One month after impulse noise exposure, we found obvious impairments in the animal’s heminode in the NIHHL group (Figure 5(b, e)). Though we can find the typical heminode structure around the auditory nerve axon, a portion of myelin lamellae appeared wrinkles. At the same time, abnormal vacuole-like substances were seen between the myelin lamellae, which compressed and disturbed the parallel structure of the myelin lamellae in adjacent areas. Compared with the NIHHL group, we saw severer damage in the NIHL group (Figure 5(c, f)). Around some auditory nerve axons, heminode completely disappeared. As to the rest, a greater portion of myelin lamellae appeared wrinkles. What’s more, larger-scale vacuole-like substances were seen between those myelin lamellae, causing greater disturbances to lager ranges of adjacent areas. To compare the degree of damage in three groups, we quantified the impairment by counting the number of normal myelin lamellae and the whole myelin lamellae in one heminode. By calculating the normal ratio, we found significant differences between the three groups (Figure 5(g)). (p < .01)

Figure 5. TEM images of heminodes in genuine pigs one month after noise exposure. In normal group, we can observe heminodes formed by regularly arranged myelin lamellae (a, d). In NIHHL group, parts of myelin lamellae appear wrinkles, and abnormal vacuole-like substances were seen between the myelin lamellae (b, e). In NIHL group, some heminodes were more sever damaged(c, f), and some were disappeared ( ). There existed a significant difference between three groups in normal ration of myelin lamellae in heminodes (g). NIHL: noise induced hearing loss group; NIHHL: noise induced hidden hearing loss group; Normal: normal group. #: p < .05, NIHHL versus Normal; ***: p < .001 NIHL versus NIHHL versus Normal n = 5.

Auditory nerve fiber

One month after impulse noise exposure, we found no morphology difference in auditory nerve fiber between the three groups (Figure 6(a–c)). To further evaluate the impact that impulse noise produced on the myelin sheath, we measured the diameters of the auditory nerve axon and the nerve fiber (axon plus myelin sheath). The quotient of the two diameters, which was named g-ratio, can reflect the myelin sheath’s thickness. By comparing the three groups’ results, we found no significant difference (Figure 6(d)), which indicates that impulse noise cannot impair myelin sheath.

Figure 6. TEM images of nerve fibers in genuine pigs one month after noise exposure. No obvious morphological differences were observer between three groups (a, b, c). Statistical analysis of g-ratio showed no significant differences as well (d). NIHL: noise induced hearing loss group; NIHHL: noise induced hidden hearing loss group; Normal: normal group.

Discussion

Hidden hearing loss is a hearing deficit featured by the normal threshold, reduced ABR P I amplitude, and prolonged P I latency. As a result, HHL can’t be detected by standard audiometry. People suffering from HHL usually have difficulties in speech discrimination and temporal processing, particularly in a noisy environment [1]. In the present study, we proved for the first time that impulse noise could cause HHL in guinea pig. According to the criteria mentioned above, all guinea pigs in the NIHHL group could be diagnosed with HHL, and there was no sign of recovery till one month later.

A recent study showed that HHL resulted from noise exposure is closely related to synapse loss and cochlea synaptopathy [6]. Type I spiral neural fibers can be divided into two kinds according to their SR: Low-SR and High-SR. Low-SR fiber makes synaptic contact with inner hair cells in the modiolar side, and the synapse is characterized by a relatively larger ribbon and smaller postsynaptic structure. At the same time, low-SR fiber is more vulnerable to noise exposure [12]. Compared with the reported studies [6,13], our research found similar morphologic impairments in low-SR ribbon synapses, which indicated that impulse noise shares a similar mechanism with stationary noise in causing hearing loss. The normal temporal resolution of the cochlea, which is critical for rapid response to sound stimulation, is closely related to ribbon synapse [14]. As the first speed limiting site for the auditory signal pathway, ribbon synapse can impact the impulse conduction velocity, which can be reflected by the latency of ABR peak I. One month after impulse noise exposure, we found obvious impairments in ribbon synapses of the NIHHL group. The results of suprathreshold ABR peak I latency were significantly prolonged at the same time point, just consistent with the monographic changes. Therefore, it is logical to infer that the deficits of temporal coding and speech discrimination in HHL suffers are attributed to cochlea synaptopathy.

At the SGN peripheral processes, the initial sections tightly coupled with IHC by ribbon synapses are unmyelinated. As the afferent terminals exit HP, Schwann cells gathered around the axon and formed the myelin sheath. Compared with Ranvier's node that is composed of two adjacent Schwann cells, heminonde is a special structure formed by the first Schwann cell. Apart from ribbon synapse, heminode is another key structure along the auditory signal pathway. A recent study proved that disrupted heminode caused by transient loss of cochlear Schwann cells could cause HHL [7]. Animals suffering from transient auditory nerve demyelination can have complete remyelination 16 weeks later, which leaves the heminode zone unrepaired. This kind of impairment in heminode and the hearing deficits showed no sign of recovery until one year later. The same study also proved that NIHHL couldn’t cause a similar impairment in the heminode zone. To further research the relationship between impulse noise exposure and heminode, we used TEM. Results showed that though NIHHL can’t cause a complete loss of heminode, its microstructures were disturbed to some extent.

Heminode is the generator of spiral ganglion cells, and it is the origin of electrical impulses along the auditory axon [15,16]. Heminode plays a vital role in clustering ion channels, the regulation of axon diameter, and axonal energy metabolism. In type I ganglion neuron, Nav⁺1.6 ion channels are gathered in heminode and Ranvier nodes. In auditory nerves, action potentials (AP) are generated by various voltage-dependent Na⁺ channels, and Nav⁺1.6 is the most important one [17,18]. In the present study, myelin lamellae in the heminode, where clustered Nav⁺1.6 ion channels, were significantly disturbed. This impairment resulted in the reduction of ABR peak I amplitude. Attenuated electrical signals along the auditory nerves can affect hearing in the cerebral auditory cortex and may eventually result in deficits in speech discrimination and intelligibility.

The excitatory electrical activity along myelinated auditory nerve fiber is conducted in a jump manner between Ranvier’s adjacent nodes. Its speed is closely related to the myelin sheath’s thickness and nerve fiber diameter [18]. In our study, the inner diameter and outer diameter of the type I spiral nerve fibers were measured before and after noise exposure. The value of g-ratio was calculated to reflect the thickness of the myelin sheath. No statistical difference was found between the three groups, indicating that impulse noise would not affect the myelin sheath.

Previously, studies on the mechanisms of NIHHL were all based on stationary noise. This study revealed for the first time that impulse noise could induce HHL as well. This is a potent supplementary to the cause of HHL. Cochlea synaptopathy and deficit in heminode after transient auditory nerve demyelination were two widely accepted pathogeneses, and they were additive. Our study indicated that impulse noise could injure the ribbon synapses and impair the microstructures of the heminode. The finding that the two pathogeneses co-exist after impulse noise exposure may shed light on further research of preventive and therapeutic methods for NIHHL.

Acknowledgments

We would like to thank the Center for Biological Imaging (CBI), Institute of Biophysics, Chinese Academy of Sciences for the transmission electric microscope (TEM, FEI Tecnai Spirit120kV), and we are grateful to Professor Xixia Li for her help in making TEM samples and analyzing images.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This article was funded by the Active Health Project of the Ministry of Science and Technology [2020YFC2004001], NSFC grant [81470700], Special Youth Project of the PLA General Hospital [QNC19051], and Beijing Natural Science Foundation [7222185].