Glutathione peroxidase 1 protects against age-related but especially
noise-induced cochlear synaptopathy in high frequency cochlear areas
in C57BL/6.CAST-Cdh23
Ahl+mice
Olivier Teerling, 25-01-2020, Universitair Medisch Centrum Groningen Supervised by: Prof. Sonja Pyott at the department of Otorhinolaryngology Assessed by: Dr. H.J. Krugers at the faculty of Science of the Amsterdam University
Abstract
Next to our society being an aging society, we are also more and more exposed to noisy environments, be it for leisure activities or for work and living. Listeners with no apparent audiometric threshold shifts nor hair cell loss sometimes still report difficulty with speech perception and often show reduced temporal hearing resolution. Recently, this ‘hidden hearing loss’- phenomenon has been connected to widespread reductions of afferent synapses and degeneration of the cochlear nerve following both aging and noise-exposure. Aging and acoustic trauma are accompanied by the release of Reactive Oxygen Species and oxidative stress, which are countered by natural defense mechanisms like anti-oxidant defense mechanisms. Recently it has been found that for noise-induced hearing loss, glutathione peroxidase 1 (GPX1) plays an important protective role and limits the auditory threshold shifts and inner & outer hair cell loss. This study examined whether B6.CAST mice with a GPX1 knock-out also exhibited greater synapse loss at the cochlear inner hair cells as a consequence of aging and noise exposure. Synapses per inner hair cell were counted in, partly noised (98 dB SPL 8-16 kHz broadband noise) B6.CAST mice (6 – 96 weeks) with and without the GPX1 enzyme by immunolabeling the cochlear synapses against CTBP2. GPX1 KO mice showed overall lower synapse counts for all ages and also an increased age-related synapse loss, especially in the higher frequencies. However, GPX1 KO mice especially showed increased vulnerability to noise-exposure even at an older age (50 weeks). This suggests an important protective role for GPX1 for synapse survival in general, but especially for protection against noise-induced synapse loss.
Keywords: GPX1, Cochlea, inner and outer hair cells, inner ear, mice, presynaptic afferent ribbons, reactive oxygen species, free radical, antioxidant, knockout
1 | Introduction
1.1 | Hearing and hearing loss: significance and underlying causes
Our ability to hear is essential to interpret the rich sound landscape and interpret acoustic information necessary
for communication and survival. However, exposure to environmental sounds can also stress and even injure the auditory system. Even though not all of these environmental sounds lead to auditory stress or cochlear injury, it has been found that even constant background sounds, especially in urban
environments can lead to a form of hearing loss (Caciari et al., 2013; Krefis, Augustin, Schlünzen, Oßenbrügge, & Augustin, 2018). The prevalence of noise-induced hearing loss is illustrated by the fact that more than 11% of the world’s population (around 800 million people) suffers from a certain form of hearing loss, which makes hearing loss one of the five biggest sensory disorders worldwide (Cunningham & Tucci, 2017; Davis & Hoffman, 2019; Masterson, Themann, & Calvert, 2018). Besides noise-induced hearing loss (NIHL), hearing abilities also gradually decrease with age (Wang & Puel, 2018; Wong & Ryan, 2015). Not infrequently hearing loss leads to physical or emotional stress at home or at work and it often leads to distressing symptoms, like tinnitus and hyperacusis (Eggermont, 2017; Le, Straatman, Lea, & Westerberg, 2017). As our aging society is increasingly exposed to environmental noise, the prevalence of noise-induced hearing loss (NIHL), age-related hearing loss (ARHL) and their interaction will probably increase in the coming decades, making it a serious issue for public health (Anna R. Fetoni, Picciotti, Paludetti, & Troiani, 2011; Wong & Ryan, 2015).
When it comes to our hearing, all audible sounds are picked up by the cochlea. The cochlea consists of a collection of highly specialized structures including the organ of Corti (OC) and the stria vascularis (SV). The OC contains the specialized structures for sensorineural transduction. The SV is a vascular organ responsible for maintaining an ionic equilibrium of the fluid compartments necessary to provide metabolic support to the OC. The OC contains two sets of cells, which are arranged in rows: the outer-
and inner hair cells (OHC & IHC). The OHCs, arranged in three rows, are responsible for acoustic amplification of the sound, while the IHCs, a single row, have synaptic connections with afferent auditory nerve fibers (ANF). Pre-synaptic ribbons control exocytotic release of vesicles which release glutamate in the synaptic cleft. This process is crucial for the temporal precision of the subsequent sound-evoked spiral ganglion neuron (SGN) responses (Glowatzki & Fuchs, 2002; Moser, Neef, & Khimich, 2006; Moser, Predoehl, & Starr, 2013). These SGNs in turn connect to the ANFs which transfer the information of the acoustic environment to the central auditory system where it is processed in higher-order auditory areas. Not all sounds from the acoustic environment are audible. Normal hearing individuals hear all sounds within a 20 – 20.000 kHz frequency range, with changing hearing thresholds for different frequencies (for a more extensive description of the cochlea, see Wang & Puel, 2018).
Following noise exposure often a temporarily or permanent shift of the hearing threshold (TTS & PTS) is found. When it comes to daily non-traumatic sounds however, the thresholds often recover and stabilize over time. These threshold shifts are mostly found in the basal, high frequency, areas of the cochlea, often even more basal than the frequency of the original noise (Kujawa & Liberman, 2009). Auditory threshold shifts have been shown to reflect damage to OHCs and IHCs (Eileen, 2007; Le et al., 2017), which also shows a base-to-apex gradient of hair cell susceptibility to damage (Ryan, Kujawa, Hammill, Le Prell, & Kil, 2016; Sha & Schacht, 2017). While
threshold shifts are a good reflection of hair cell damage following noise exposure, threshold shifts do not fully indicate damage to the sensory spiral ganglion neurons (SGN) innervating the hair cells. Recent work has shown that within 24 hours after noise exposure, neural amplitudes for high cochlear frequencies are severely reduced and cochlear synapses between peripheral synapse terminals of the SGN and IHCs are lost (Fernandez, Jeffers, Lall, Liberman, & Kujawa, 2015; Jensen, Lysaght, Liberman, Qvortrup, & Stankovic, 2015; Sergeyenko, Lall, Liberman, & Kujawa, 2013). Further evidence shows that the damage is related to an increase in oxidative stress, which disturbs the redox balance between oxidative stressors and anti-oxidant defense mechanisms, with loss of IHC synaptic ribbons and degeneration of SGNs as a result (Anna Rita Fetoni et al., 2013; Kujawa & Liberman, 2009). This damage, also described as synaptopathy, occurs even when hair cells are not permanently damaged and auditory thresholds fully recover over time (Kujawa & Liberman, 2009; Lin, H, Furman, A, Kujawa, & Liberman, 2011). In ARHL, this pattern of synaptic damage preceding hair cell damage associated with auditory threshold shifts has also been observed (Sergeyenko et al., 2013). It is thus suggested that synaptopathy is one of the earliest markers of general hearing loss.
Currently, treatments for hearing loss are mostly based on an assessment of noise-induced threshold sensitivity changes. However, it is shown in animals and humans that threshold changes rely on hair cell injury, which, once lost, prove difficult, if not impossible, to regenerate (Brigande & Heller, 2009; Wang & Puel,
2018; Zhang et al., 2020). If synapse loss, which so far precedes hair cell injury and auditory threshold shifts in all available animal studies, also precedes these processes in humans, it might be wise to revise clinical protocols in order to detect the earliest cochlear injury, with the goal to prevent subsequent permanent injury. In order to contribute to this line of research, which is in it’s infancy, and improve our chances of finding adequate pharmacological treatment, it is important to fully grasp the cause of this early-stage synaptopathy in animals (M. C. Liberman & Kujawa, 2017).
1.2 | Linking cochlear damage to formation of reactive oxygen species (ROS)
One line of evidence indicates that cochlear synaptopathy is caused by increasing amounts of reactive oxygen species (ROS). Increased levels of ROS have been found after acoustic trauma and with aging (Morioka et al., 2018; Kevin K. Ohlemiller, Wright, & Dugan, 1999; Wu, Xiong, & Sha, 2020). While acoustic trauma can have different causes, exposure to (loud) noises is one of the most common (Ohlemiller et al., 1999; Yamane et al., 1995; McFadden et al., 2001). ROS can damage cellular processes, attack lipids and proteins, and damage cells in various other ways. There are several types of (noise-induced) ROS. ROS often arise from natural cellular processes, like aerobic cellular respiration, phagocyte activity and other processes. During these processes O2
-, a mildly toxic ROS by itself, is produced. Besides being inherently toxic it can also contribute to the formation of even more toxic ROS, like
H2O2 or OH of which the latter is the most toxic. There are multiple ways through which O2
can induce toxic ROS formation of which the most important is the Fenton reaction where ferric iron (Fe3+
) is reduced back to ferrous iron (Fe2+
) (Evans & Halliwell, 1999; Koppenol, 2001). Ferrous iron plays an important role in the production of toxic OH. How ROS is formed in the ear during and after noise exposure is not entirely clear, but multiple hypotheses have been proposed. The leading hypothesis suggests that, when under metabolic stress, mitochondria increase their activity to meet the increased energy demands in the OC and SV (McFadden et al., 2001). In addition, phagocyte activity at the injured tissue can also generate ROS (see for a more detailed description: McFadden et al., 2001).
1.3 | The role of glutathione peroxidase 1 in protection from ROS
Under normal physiological conditions the body is protected from base levels of ROS production through a set of various defense mechanisms. The most prominent of these mechanisms is an elaborate antioxidant defense mechanism, with the predominant cytosolic selenium-dependent isoform of glutathione peroxidase (GPX1) as one of its most important antioxidants. (Brigelius-Flohe, 1999; Pierson & Gray, 1982; Rita et al., 2019). Antioxidants prevent ROS formation or remove ROS after they are formed. GPX1 converts H2O2 to water (H2O), with which it prevents toxic ROS formation. During this process glutathione reductase (GSR) convert oxidized
glutathione (GSSG) back to reduced glutathione. GPX1 is one of the major anti-oxidants in the cochlea and appears to be severely down-regulated when under acoustic stress (Kil, Pierce, Tran, Gu, & Lynch, 2007; Liang et al., 2019), leading to a weaker defense mechanism against ROS.
In order to investigate the role of GPX1 in the defense against ROS, scientists have also developed GPX1 knock-out (KO) mice. When using these mice to investigate the role of GPX1 in NIHL, GPX1-deficiency was found to increase the susceptibility to noise-induced ROS damage, illustrated by higher threshold elevation and greater OHC loss in C57BL/6 GPX1 KO compared to C57BL/6 wildtype mice (K. K. Ohlemiller, McFadden, Ding, Lear, & Ho, 2000; McFadden et al., 2001). Moreover, studies that tried to mimic the GPX1 enzyme, by making use of an Ebselen-treatment, found upregulation of GPX1 levels both in-vivo (Kil et al., 2007) and in-vitro (Liang et al., 2019), leading to protection against NIHL, as evidenced by reduced OHC loss, reduced auditory threshold shifts and reduced swelling of the SV. Together, these findings implicate an important role for the GPX1 enzyme in the protection of the cochlea against noise-induced, and possibly age-related oxidative stress.
As post-exposure hair cell regeneration remains difficult, pre-exposure anti-oxidant supplementation may be one of the most promising therapeutic directions for treating hearing loss (Brigande & Heller, 2009; Wong & Ryan, 2015; Zhang et al., 2020). Several antioxidant cocktails, including one containing L-cysteine-glutathione, have been shown to reduce age-related
threshold shifts in C57BL/6 (B6) mice (Heman-Ackah et al., 2010). These antioxidant treatments have also already been trialed on humans (Choi et al., 2014). To contribute to this line of therapeutic research, it is essential to fully grasp the working of antioxidant defense mechanisms and in this case GPX1 in the inner ear, so that antioxidant-based treatments can be optimized and hopefully be implemented in a therapeutic way.
1.4 | Is Glutathione peroxidase 1 also involved in protection against noise-induced afferent synapse loss?
Although these previous studies indicate a role for GPX1 in protection against (especially outer-) hair cell loss, the role of GPX1 in protecting against synapse loss, an early event in noise-induced and age-related hearing loss, remains elusive. One recent study has found an important defensive role for the GPX1 enzyme against electrically-induced damage to the cochlear synapses in vitro (Liang et al., 2019), but no work has investigated the role of GPX1 in auditory synaptopathy in response to noise exposure and aging in vivo. Therefore, the goal of this study is to investigate the role of the GPX1 enzyme in noise-induced and age-related afferent synapse loss (NISL) in C57BL/6.CAST-Cdh23Ahl+
mice and
measure afferent synapse loss in mice of different ages and following noise exposure .
This study will use C57BL/6.CAST-Cdh23Ahl+
mice with and without the GPX1 gene. The C57BL/6.CAST-Cdh23Ahl+ (B6.CAST) mice are a crossing of C57BL/6 (B6) mice and CAST/Ei (CAST) mice, with the goal of obtaining a homozygous strain for a recessive locus called the ‘age-related hearing loss’-locus. Previous studies have shown that this locus leads to early-onset ARHL in regular B6 mice. By crossing B6 with CAST mice, this deficiency should be overcome. This claim will be tested by comparing results gathered for this project to results from a previous project where regular B6 mice were used.
A successful delay in the onset of ARSL was indeed observed in B6.CAST mice compared to B6 mice. Besides, overall lower synapse counts were found for GPX1 KO mice for all ages. Increased age-related synapse loss, especially in the higher frequency areas was observed. However, the GPX1 KO mice were especially vulnerable for noise-exposure, even at an older age (50 weeks). This could suggest an important protective role for the GPX1 enzyme for synapse survival in general, but especially for protection against noise-induced synapse loss.
2 | Material and Methods
2.1 | Animals
In sake of completeness of description, all methodological steps of this project will be described. It will also be described which parts of the project have
been conducted by myself and which parts are relevant for the interpretation of the current results and discussion.
The experiments of the current project were done in collaboration with the University of Oldenburg in Germany and the University Medical Center Groningen in the Netherlands. Experiments with live animals, including behavioral assessments, auditory brainstem response measurements (ABRs), and cochlear dissections, were performed at the University of Oldenburg. Subsequent dissection and scanning was subsequently performed at the University Medical Centre Groningen. All experiments with live animals were in accordance with the procedures of animal experimentation approved by the Government of Lower Saxony, Germany and in accordance with EU Directive 2010/63/EU.
2.2 | Noising paradigm
Three strains of mice have been used for the current project. The first group consisted of 34 regular B6 mice of which the experiments were already previously conducted. For the current project 70 B6.CASTmice were sacrificed. These mice were subdivided in four groups: 45 B6.CAST mice with the GPX1 enzyme and 25 B6.CAST mice where the GPX1 enzyme has been knocked out, called GPX1 KO hereafter. In both groups a part of the mice were noised, thus leading to a total of four different experimental groups. Four B6.CAST mice were noised with a 98 dB SPL 8-16 kHz broadband noise and two B6.CAST mice were noised with a 104 dB SPL broadband noise (8-16 kHz). Of the 39 B6.CAST mice
that were not noised, four were excluded as no correct scan of the cochlea could be made. Four GPX1 KO mice were noised with a 98 dB SPL 8- 16 kHz broadband noise, while three GPX1 KO mice were noised with a 104 dB SPL sound. One GPX1 KO mouse was noised with a 115 dB SPL 8-16 kHz broadband noise. Due to the time constraint of this project not all cochleae of all mice could be analyzed, resulting in missing data in all groups.
2.3 | Experimental set-up
Mice in all groups were roughly divided in five age groups: Six weeks, which served as baseline counts, seventeen weeks, 43 weeks, 69 weeks and 96 weeks. Noised mice were always age-matched.
All mice were bred and held individually in 1284L Eurostandard type II L cages at the University of Oldenburg. Mice were held on a 12 h light/dark cycle (lights on at 7 a.m.) in a climatized environment. A running plate and plastic igloo (Zoonlab GmbH, Castrop-Rauxel, Germany) served as environmental enrichment. Water was available ad libitum, but food was restricted in order to ensure motivation to perform a behavioral task performed in a subset of this study. It was made sure that the body weight of food restricted animals did not drop below 15% of the body weight with food being available ad libitum. Transportation of the animals from their home cage to the experimental set-up happened in a small transfer cage.
2.4 | Behavioral assessment
This part of the project has been conducted by K. Bleckmann who is a collaborator in Oldenburg. Data of the behavioral assessment will not be described in this report. A brief methodological explanation is given in the appendix.
2.5 | Auditory Brainstem Response measurements
This part of the project has been conducted by K. BLeckmann, who is a collaborator in Oldenburg. Data of the behavioral assessment will not be described in this report. A brief methodological explanation is given in the appendix.
2.4 | Immunostaining
The immunostaining has been conducted by my supervisor Sonja Pyott either before and during the 6 months of this project.
The OC dissection was performed as described earlier (Reijntjes et al., 2019). Isolated cochleae were immediately
immersed in ice-cold 4%
paraformaldehude (PFA) in phosphate buffered saline (PBS) for 1 hour, after which they were stored in ice-cold PBS until further processing. OC were dissected from the cochleae and placed into blocking buffer (PBS with 5% normal goat serum, 4% Triton X-100, and 1% saponin) for at least 1 h after which they were incubated overnight with the primary antibodies diluted in blocking buffer. These primary antibodies existed of mouse monoclonal (IgG1) anti-CTBP2 (BD Biosciences, 612044, 1:500) and mouse monoclonal (IG2a) anti-GluA2 (Millipore, MAP397; 1:300). All samples were rinsed three times for 10 min with PBS with 0.6% Triton-X 100 (PBT) after which they were incubated with the secondary antibodies for at least 4 h. Secondary antibodies included AlexaFluor 488 goat anti-mouse (IgG1, ThermoFisher, A-21121) and AlexaFluor 647 goat anti-mouse (IgG2a, ThermoFisher, A-21241). All secondary antibodies were diluted 1:500 with blocking buffer. All samples were rinsed three times for 10 min with PBS with 0.6% Triton-X 100 (PBT) and once with PBS before they were mounted. All incubation and rinse steps were performed on a rocking platform at room temperature. The mounted samples were stored at 4C.
2.5 | Microscopy and image analysis
Cochlear place frequency maps were constructed for each OC in order to identify the tonotopic regions. Using a Leica DM 4000b microscope, 5x magnified micrographs were collected. Sometimes, the OCs were too big to capture in one image and a ‘stitching’ procedure was performed with the panorama editor Hugin (hugin-2020.0.0). Because OCs
were often collected in individual pieces, micrographs were then montaged using Fiji ImageJ . Coordinates of previously determined place frequency maps were used to create the current tonotopic maps in Fiji ImageJ (Muller, Hunerbein, Hoidis, & Smolders, 2005). The determined frequency locations were 5, 10, 20, 30 and 40 kHz. In the rare case the exact location didn’t have viable tissue, surrounding tissue was photographed. This step
required a plug-in which can be found at
http://www.masseyeandear.org
/research/otolaryngology/
eaton-peabody-laboraties/histology-core).
Using a Leica TCS SPB8 confocal microscope mounted with a 63X oil immersion objective with a resolution of 1024 x 1024 pixels at a speed of 400 Hz, high resolution image z-stacks were collected from the above mentioned frequency regions of the organ of Corti. The z-step (optical section thickness) was optimized to around 0.3 µm. Z-stacks varied in thickness in order to entail the whole synaptic pole of the sensory inner hair cells. The laser power gain and offset adjustments were set in such a way that the dynamic range of the measured intensity values was optimized. Results in this report are presented as a z-projection through the collected individual optical stacks.
Using Imaris v7.6 (Bitplane Inc.), three-dimensional (3D) quantative image analysis was conducted. The Imaris “spots” function was used to detect and quantify the amount, shape and location of CTBP2-positive presynaptic ribbons and GluA2-positive postsynaptic glutamate receptor clusters. The total amount of CTBP2-positive immunopuncta was then divided by the total number of CTBP2 immunolabeled inner hair cell nuclei in the same field of view. The IHC nuclei were identified by their bigger size, higher fluorescence and proximity to the CTBP2-positive afferent immunopuncta.
2.6 | Statistical analyses
Due to the time constraint of this project, unfortunately not all samples could be scanned. This leads to missing
data and in many cases to a not high enough n-value to do reliable statistical comparison. This section will describe how the stats wil be done if we reach the designated n-values for all grousp.
Statistical analyses will be performed with Rstudio (Rstudio, 2021). A general linear mixed model (GLMM) / two-way repeated measures analyses of variance (ANOVA) will be used to test for within-subject effects of frequency (5, 10, 20, 30 and 40) and the between-subject effect of genotype on the number of CTBP2-positive immunopuncta per frequency region. Bonferroni corrections will be applied for post-hoc comparisons. A P-value of <0.05 (*) represents significant differences, while a P-vlaue of <0.01 (**) represents highly significant differences. The results per group are represented as the mean of that group ± SEM. The ‘n’ value represents the amount of individual samples.
3 | Results
The goal of this project was to identify the role of the GPX1 enzyme in protection against age-related and noise-induced synapse loss (ARSL & NISL) in B6.CAST mice. Therefore, we quantified the amount of afferent CTBP2-positive synapses (immunopuncta) in the sensory inner hair cells of the Organs of Corti of B6.CAST mice with and without the GPX1 KO. We immunolabeled for CTBP2 as this protein is highly expressed in the presynaptic afferent ribbons (Barone et al., 2019; Khimich et al., 2005). A part of the mice from both groups were noised with a primarily 98 dB SPL, 8-16 kHz broadband noise. Comparisons between these four groups were made by
calculating a ‘ribbon density’ value, where the amount of ribbons was divided by the amount of IHC.
As explained earlier, in contrast to many previous studies (Colzato, Barone, Sellaro, & Hommel, 2017; Hequembourg & Liberman, 2001; Kevin K. Ohlemiller, Jones, & Johnson, 2016), this study used B6.CAST mice which are a crossing of normal B6 mice with CAST mice, resulting in B6.CAST mice with a homozygous Ahl+ locus, which should protect mice against early-onset ARHL. In previous projects, colleagues have looked at CTBP2-positive ribbon density in B6 mice. Therefore, this section will first compare data from B6 and B6.CAST mice in order to confirm the protective role of the Cdh23Ahl+
locus. Secondly, this section will show and compare ARSL in B6.CAST mice with and without the GPX1 enzyme. Lastly, this section will show and compare synapse loss in noised and not noised B6.CAST and GPX1 KO mice.
A grand overview of the absolute values of afferent synapse per IHC counts is provided below in table 1. It shows the number of afferent synapses per IHC for all genotypes, ages and frequency regions. This table also contains the amount of counted cochlear frequency regions. Every cochlea is counted as one and in some cochlea not all frequency regions were countable due to missing synapses, missing hair cells or damaged tissue. Unfortunately not all groups had a robust n-value, nor was the gender of the animal standardized. Besides, the age groups in B6.CAST mice and GPX1 KO mice was not completely similar and thus not all age groups could be compared. Seen that no reliable statistical analysis could be done, the results described below should be interpreted as trends, even though some results seem to trend towards significance.
Table 1 | Number of afferent synapses per IHC per genotype, age and frequency region
Genotype (age*) Frequency (kHz) Ribbons / IHC ± St. Deviation** (n***)
B6 (6w) 8 16 15,4 ± 1,2 (n = 6) 16,6 ± 1,5 (n = 6) 32 18,9 ± 2,5 (n = 6) B6.CAST (6w) 5 10 14,6 ± 1,7 (n = 7) 18,2 ± 0,9 (n = 7) 20 19,9 ± 1,2 (n = 6) 30 40 19,2 ± 1,2 (n = 7) 18,7 ± 0,9 (n = 7) GPX1 KO (6w) 5 10 14,7 ± 1,8 (n = 4) 17,0 ± 0,8 (n = 4) 20 19,9 ± 1,1 (n = 3) 30 40 18,3 ± 0,8 (n = 4) 17,1 ± 1,4 (n = 4) B6 (12 + 24w) 8 16 14,5 ± 1,7 (n = 15) 16,5 ± 1,8 (n = 15) 32 17,0 ± 2,5 (n = 15) B6.CAST (17w) 5 10 15,2 ± 2,3 (n = 6) 18,3 ± 1,6 (n = 6) 20 18,7 ± 0,7 (n = 6) 30 17,6 ± 1,2 (n = 7)
40 15,8 ± 5,0 (n = 4) GPX1 KO (12w) 5 10 10,6 ± 1,3 (n = 2) 16,6 ± 0,1 (n = 2) 20 19,5 ± 0,5 (n = 2) 30 40 17,7 ± 0,0 (n = 1) 18,6 ± 1,1 (n = 2) B6 (36w) 8 16 11,1 ± 2,5 (n = 6) 14,9 ± 2,1 (n = 6) 24 8,6 ± 2,9 (n = 6) B6.CAST (43w) 5 10 13,6 ± 1,9 (n = 4) 14,8 ± 1,3 (n = 4) 20 19,3 ± 0,8 (n = 3) 30 40 16,8 ± 1,2 (n = 3) 14,5 ± 3,2 (n = 3) GPX1 KO (37w) 5 10 11,7 ± 1,5 (n = 6) 15,4 ± 1,5 (n = 7) 20 17,2 ± 1,0 (n = 5) 30 40 16,9 ± 1,7 (n = 5) 12,1 ± 3,4 (n = 6) B6 (72w) 8 16 12,1 ± 1,6 (n = 3) 14,3 ± 0,3 (n = 3) 32 11,3 ± 2,2 (n = 3) B6.CAST (69w) 5 10 10,8 ± 2,5 (n = 6) 13,5 ± 2,5 (n = 6) 20 15,4 ± 2,1 (n = 6) 30 40 16,9 ± 4,0 (n = 6) 15,5 ± 2,1 (n = 5) GPX1 KO (56w) 5 10 10,2 ± 0,4 (n = 3) 14,3 ± 2,5 (n = 3) 20 16,3 ± 3,2 (n = 3) 30 40 16,0 ± 0,8 (n = 3) 14,5 ± 1,1 (n = 3) B6 (104w) 8 16 9,8 ± 1,4 (n = 4) 9,4 ± 1,8 (n = 4) 32 4,3 ± 2,8 (n = 3) B6.CAST (96w) 5 10 8,1 ± 2,8 (n = 7) 11,5 ± 1,2 (n = 6) 20 13,9 ± 6,0 (n = 5) 30 40 14,5 ± 3,5 (n = 7) 12,5 ± 3,9 (n = 6) GPX1 KO (64w) 5 10 9,6 ± 0,9 (n = 2) 13,1 ± 2,1 (n = 3) 20 14,5 ± 3,9 (n = 3) 30 40 14,5 ± 0,6 (n = 2) 11,4 ± 3,4 (n = 2)
3.1 | Age-related synapse loss
3.1.1 | The Cdh-23A hl+gene appears to reduce early-onset of age-related synapse loss in B6.CAST mice compared to regular B6 mice, especially in the higher cochlear frequency areas.
In order to test the hypothesis that the Cdh23Ahl+gene protects against early-onset ARSL, afferent ribbon density calculations from B6 and B6.CAST mice were compared (figure 1). Afferent ribbon density was measured in both groups on different, but comparable ages. In B6 mice measurements took place at 6, 17 (12 + 24), 36, 72 and 104 weeks old, while in B6.CAST mice measurements took place at 6, 17, 43, 69 and 96 weeks old. Results are shown in figure 1 for three
comparable cochlear frequency regions: ‘Low’, ‘Middle’ and ‘High’. ‘Low’ stands for 8- (B6) & 10 kHz (B6.CAST), ‘Middle’ stands for 16- (B6) & 20 kHz (B6.CAST) and ‘High’ stands for 32- (B6) & 30 kHz (B6.CAST).
The quantification shows that the number of afferent ribbons per IHC was greater in B6.CAST mice for all frequency regions and for all ages. Specifically, when comparing B6 to B6.CAST, a greater amount of ribbons per IHC was found at 6 & 17 weeks in the Low frequency region and a greater amount of ribbons per IHC was found at 6 and ~40 weeks in the Middle frequency region. However, especially age-related synapse loss in the high frequency area decreased, which is shown with the greater amount of afferent ribbons per IHC at ~40, 70 & 100 weeks old.
Figure 1. Measurement of the average amount of CTBP2-positive synapses per IHC for both aging B6 and aging B6.CAST in three different cochlear regions. Age-groups are color-paired and “Low”
stands for 8- (B6) & 10 kHz (B6.CAST), Middle stands for 16- (B6) & 20 (B6.CAST) and High stands for 32- kHz (B6) & 30 kHz (B6.CAST). N-number and exact values can be found in Table 1
In order to visualize the difference in amount of afferent ribbons per IHC between B6 and B6.CAST mice, the B6 numbers were subtracted from the B6.CAST numbers, which is shown in figure 2. This graph clearly visualises the higher amount of synapses per IHC for B6.CAST mice in all ages and for all frequencies, Especially it shows a greater
amount of synapses per IHC at 6, 17 and ~40 weeks in the Low frequency region; a greater amount of synapses per IHC at 6, ~40 and ~100 weeks in the Middle frequency region. However, it especially shows a greater amount of synapses per IHC at ~40, ~70 and ~100 weeks in the High frequency region.
Figure 2. Difference measurement of the average amount of CTBP2-positive synapses per IHC for both aging B6 and aging B6.CAST in three different cochlear regions. Age-groups are color-paired and “Low” stands for 8 (B6) & 10 kHz (B6.CAST), Middle stands for 16 (B6) & 20 (B6.CAST) and High stands for 32 kHz (B6) & 30 kHz (B6.CAST). N-number and exact values can be found in Table 1
3.1.2 | Age related synapse loss is apparent in both B6.CAST and GPX1 KO mice not exposed to noise
In order to test the hypothesis that both B6.CAST and GPX1 KO mice show ARSL,
this paragraph will show the amount of synapses per IHC for B6.CAST (figure 3) and for GPX1 KO (figure 4) mice. These mouse-lines have been tested on various ages and for various cochlear frequency regions. B6.CAST mice have been tested
at 6, 17, 43, 69 and 96 weeks, while GPX1 KO mice have been tested at 6, 12, 37, 56 and 64 weeks old. Therefore, figure 3 & 4 don’t have completely comparable age groups. Both mouse-lines have been tested at 5, 10, 20, 30 and 40 kHz.
Figure 3 shows that for B6.CAST mice, a general decrease in the amount of synapses per IHC is found with age for all frequencies. Besides it confirms that also B6.CAST have most synapses per IHC at their most sensitive frequency (~ 20 kHz), to then steadily drop towards higher frequencies, which is in line with previous results from other strains (Barone et al.,
2019; M. C. Liberman, 2017; Sergeyenko et al., 2013). The 5 kHz region compared to the higher frequency areas shows an overall lower amount of synapses, which is also in line with previous research (Barone et al., 2019; M. C. Liberman, 2017; Sergeyenko et al., 2013). Besides, a lower amount of synapses per IHC is seen at 43, 69 and 96 weeks at 5, 10 and 20 kHz. A lower amount of synapses is also seen at 96 weeks, compared to other ages in the 30 kHz region. Lastly, a lower amount of synapses are seen at 43, 69 and 96 weeks in the 40 kHz region.
Figure 3. Measurement of the average amount of synapses / IHC for aging B6.CAST mice in five different frequency regions. N-number and exact values can be found in Table 1.
Figure 4 shows that for GPX1 KO mice, a general decrease in the amount of synapses per IHC is found with age for all frequencies. In more detail, the 5 kHz region compared to the higher frequency areas shows an overall lower amount of synapses. Moreover, in the 5 kHz region the amount of synapses per IHC quickly decreases at 12, 37, 56 and 64 weeks compared to 6 weeks. In the 10 kHz region there is a lower amount of
synapses for 64 weeks compared to the younger ages. In the 20 kHz region a lower amount of synapses is found for 37, 56 and 64 weeks. In the 30 kHz region a lower amount of synapses per IHC is found at 56 and 64 weeks. However, especially apparent is the lower amount of synapses per IHC at 37, 56 and 64 weeks in the 40 kHz region compared to younger ages.
Figure 4. Measurement of the average amount of synapses / IHC for aging GPX1 KO mice in five different frequency regions. (be aware that these ages don’t completely align with those in figure 3). N-number and exact values can be found in Table 1
3.1.3 | The Glutathione Peroxidase 1 enzyme appears to reduce age-related synapse loss
In order to test the hypothesis that the GPX1 enzyme protects against ARSL,
results from B6.CAST and GPX1 KO mice are compared (figure 5 & 6). As not all ages of both groups were comparable, the graph only shows synapses per IHC for three age groups, notably: ‘Young’, ‘Adult’ and ‘Old’. ‘Young’ stands for 6 weeks old
(B6.CAST & GPX1 KO). ‘Adult’ stands for 43 weeks old (B6.CAST) & 37 weeks old (GPX1 KO). ‘Old’ stands for 69 weeks old (B6.CAST) & 64 weeks old (GPX1 KO).
Figure 5 shows a color-paired bar graph of the amount of synapses per IHC for B6.CAST and GPX1 KO mice tested for 5, 10, 20, 30 & 40 kHz. It is apparent that in almost all cochlear frequency regions and all ages the B6.CAST mice have a higher amount of synapses per IHC than
GPX1 KO mice. Apparent differences are visible at 20 kHz where 37 weeks old GPX1 KO mice have a lower amount of synapses than 43 weeks old B6.CAST mice. Also at 30 kHz, 64 weeks old GPX1 KO mice seem to have severely reduced numbers of synapses per IHC compared to 69 weeks old B6.CAST mice. Finally, especially in the 40 kHz region, GPX1 KO mice see to have severely reduced numbers of synapses per IHC for all ages.
Figure 5. Measurement of the average amount of synapses per IHC for aging B6.CAST & aging GPX1 KO mice in five different frequency regions. N-number and exact values can be found in Table 1
In order to visualize the difference in synapses per IHC between B6.CAST and GPX1 KO mice, the GPX1 KO values were subtracted from the B6.CAST values, which is shown in figure 6. This is, as far as
I know, the first study looking at the effect of GPX1 (KO) on ARSL. Studies so far have found noise-induced hearing loss and electrically-induced synapse loss in GPX1 mice compared to WT, but never was the
role of the GPX1 enzyme in aging determined. In figure 5 it is clear that mice lacking the GPX1 enzyme are more vulnerable for ARSL, especially in the higher frequency areas. Previous studies did report on noise-induced elevated
hearing thresholds in young GPX1 KO mice. This could mean that the GPX1 enzyme serves a similar role in protection against NRHL and ARHL.
.
Figure 6. Difference measurement between the average amount of synapses per IHC for aging B6.CAST compared to aging GPX1 KO mice (B6.CAST – GPX1 KO) in three different frequency regions. “Young” stands for 6 weeks old in both mouse lines, “Middle” stands for 43- (B6.CAST) & 37 weeks (GPX1 KO) and “Old” stands for 69 (B6.CAST) & 64 weeks (GPX1 KO). N-number and exact values can be found in Table 1
3.2 | Noise-related synapse loss
3.2.1 | The Glutathione Peroxidase 1 enzyme appears to reduce noise-induced loss of CTBP2-positive synapses
In order to test the hypothesis that the GPX1 enzyme protects against NRSL,
results from noised and not noised B6.CAST and GPX1 KO mice of similar ages (~52 weeks) were compared. As mentioned in the methods, mice were noised with a primarily 98 dB SPL 8-16 kHz broadband noise.
Figure 7 shows the amount of synapses per IHC per cochlear frequency region for 50 weeks old not noised and
noised B6.CAST mice and for 56 weeks old not noised and 52 weeks old noised GPX1 KO mice. The graph shows a smaller amount of synapses per IHC for the B6.CAST mice at 5, 30 and 40 kHz, but higher synapses per IHC at 10 & 20 kHz. For the GPX1 KO mice, there is a clear
smaller amount of synapses per IHC at 20, 30 and 40 kHz, but a higher or equal amount of synapses per IHC at 5 and 10 kHz. Besides, a clear lower amount of synapses per IHC are visible for noised GPX1 KO versus noised B6.CAST mice.
Figure 7. Measurement of the average amount of synapses for B6.CAST and GPX1 KO mice either exposed or not exposed to noise. N-number and exact values can be found in Table 1
In order to visualize these differences in loss of synapses per IHC in the five cochlear frequency regions between not noised and noised B6.CAST and GPX1 KO mice, the noised values were subtracted from the not noised values for both B6.CAST and GPX1 KO mice and compared to each other, which is shown in figure 8. Even though studies have shown increased noise-induced
hearing loss in GPX1 mice already more than two decades ago, no follow up has been done about the dynamics of synapse loss following noise exposure in GPX1 mice. The results in figure 6 and 7 show synapse loss in the higher cochlear frequency regions, strikingly parallel to the NIHL found previously (K. K. Ohlemiller et al., 2000; Ohlemiller et al., 2001). Ohlemiller (2000) found aggravating
hearing loss starting at 20 kHz. Ohlemiller’s study used 8 weeks old mice, while the current study used 50 weeks old mice. Interestingly, previous research found a decreased vulnerability for noise with age (L. D. Liberman & Liberman,
2015; Möhrle et al., 2016). It seems like mice lacking the GPX1 enzyme do not comply with this finding. This could mean that the GPX1 enzyme plays an especially important role for protection against noise damage.
Figure 8. Difference measurement of the average amount of synapses for un-noised minus noised mice for both B6.CAST (50w old, both groups) and GPX1 KO mice (56 & 52w old, respectively). N-number and exact values can be found in Table 1
4 | Discussion
4.1 | Overview
This study aimed to (1) confirm the reduced early-onset of ARSL in B6.CAST mice compared to B6 mice and (2) investigate the protective role of the GPX1 enzyme against ROS-induced synapse loss in ARSL and NISL. This study
contributes to the growing body of research which currently suggests that in ARHL and NIHL, synaptopathy precedes the more traditionally studied auditory threshold shifts and (inner and) outer hair cell loss, which are traditionally observed with hearing loss. Immunofluorescence and quantitative image analysis was used to count the amount of CTBP2-positive pre-synaptic immunopuncta. A density
calculation was made by dividing this amount by the amount of viable IHC found in the same part of the cochlea.
Firstly, it is found that B6.CAST mice have a higher amount of synapses per IHC, especially at higher ages and higher frequencies than B6 mice (figure 1 & 2). Secondly, B6.CAST mice do show ARSL, but with a later onset and at a slower pace, especially for the higher frequency areas (figure 1 & 2). Thirdly, the GPX1 KO mice have overall reduced synapses per IHC compared to the B6.CAST and aggravated ARSL (figure 5 & 6). Lastly, and most promising, the GPX1 enzyme appears to protect adult mice more against NISL than ARSL, especially in the higher frequency regions. Depending on the exact relation between synapse loss and hearing loss, which remains yet to be further determined, these results could suggests a therapeutic role for the GPX1 enzyme against age-related but especially noise-induced hearing loss.
4.2 Ahl+ protection against early-onset age-related synapse loss in B6.CAST mice
For a long time, studies in the field of hearing loss have made use of the common B6 mice. However, it was found that this strain suffers from early age-related hearing loss (Henry, Kenneth & Chole, Richard, 1980; Mikaelian, 1979). Therefore researchers have back-crossed this strain with the CAST-Ei mouse line. This led to a similar (to B6), but 3-6 months delayed pattern of hearing loss, where hearing loss developed more gradually over age and with frequency (Keithley, Canto, Zheng, Fischel-Ghodsian, & Johnson, 2004). Moreover,
the B6.CAST strain is thought not to have the same degeneration of the basal (but not apical- and middle-) turn as is the case in the B6 line. The differences between these strains on the level of synapses wasn’t studied yet. Earlier research reported generally increasing age-related hearing threshold shifts, with the highest hearing loss at higher frequencies (Keithley et al., 2004). This study partly confirmed these results as generally increasing age-related synapse loss was indeed found. However, the slowest ARSL was observed in the highest frequencies as opposed to the lower frequencies (see figure 1 & 2). This result could be related to the better preservation of the basal turn, in contrast to the apical and middle turn in aged B6.CAST mice, but to what extend this is the case can not be determined from the current results.
When it comes to the relation between synapse loss and elevation of hearing thresholds, previous studies have found that hearing thresholds, measured with Distortion Product Otoacoustic Emissions (DPOAE), remained stable up until at least 52 weeks for mid- to high frequencies (Vázquez, Jimenez, Martin, Luebke, & Lonsbury-Martin, 2004). Overall, the current study finds similar results when it comes to synapses per IHC. However, some differences are notable. A decrease in synapses per IHC is visible at 43 weeks at 10 kHz and at 40 kHz a trend of decrease is visible from 17 weeks already. This might be explained by the idea that loss of synapses, or synaptopathy occurs prior to hearing threshold reduction (Sergeyenko et al., 2013)
In summary, this study confirms the protective role of the homozygous Ahl+
locus for early-onset ARHL and contributes to this claim by showing that synapse counts are higher for B6.CAST for all ages at all frequencies compared to B6 mice (see figure 1 & 2). However, gradual decrease of synapses per IHC is visible for all ages (17-96 weeks) at all frequencies (see figure 3). The exact relation between synapse loss and hearing threshold loss remains yet to be further investigated.
4.3 Age-related synapse loss is apparent in B6.CAST and GPX1 KO mice
The previous section already described the gradual, age-related decrease in synapses per IHC at all measured frequencies for B6.CAST mice. This section will look at ARSL in B6.CAST and GPX1 KO mice in more detail.
Synaptopathy has not been studied much yet in B6.CAST mice, but previous studies did find a gradual age-related decrease in IHC ribbon survival in CBA/CaJ mice, especially around 10-16 kHz, with higher survival rates at higher frequencies (see figure 5A in Sergeyenko et al., 2013). The current results show a surprisingly similar patterns for older ages, but not for younger ages. Figure 1 in section 3.1.2 in the appendix, shows a bar graph representation of the normalized survival fraction of synapses per IHC in B6.CAST mice. The values in this graph are normalized to 6 week B6.CAST baseline counts, while the survival rates in the study of Sergeyenko (2013) were normalized to 4 week old CBA-CaJ baseline counts.
In comparison, the baseline values of the current study are generally higher. This is probably because of the Ahl+ genotype described in section 4.2.
Nevertheless, there are clear similarities and differences. It for example, becomes clear that the survival rate at 17 weeks differs, in the sense that at higher frequencies B6.CAST mice seem to have a lower synaptic survival rate than CBA-CAJ mice (81% versus 94% respectively). Comparing 43 week old B6.CAST mice to 32 week old CBA/CaJ mice, it becomes clear that at 5 and 20 kHz, B6.CAST mice seem to have a higher survival rate, while the other frequencies look similar. 69 weeks old B6.CAST mice have strikingly similar survival patterns as 64 weeks old CBA-CaJ mice. 96 weeks old B6.CAST mice seem to have a lower ribbon survival rate at 5 kHz than CBA/CaJ (55% versus 65%) , but at other frequencies the survival rate is similar. Even though comparing the average values between species is interesting, it should be done with caution as individual genetic and environmental factors can always skew the nature and progress of ARSL.
4.4 Noise-related synapse loss is aggravated in GPX1 KO mice in the higher cochlear frequencies
Previous studies have used various noising paradigms, for various strains and at various ages in order to study threshold shifts, hair cell loss and, to a lower extend, noise-induced synaptopathy. This methodological inconsistency between studies makes it tricky to compare results. However, some of the result found in this study align well with previously reported results.
Previous studies have found that depending on the pressure level either temporary threshold shifts (TTS) or permanent threshold shifts (PTS) are
induced. For TTS studies, noising paradigms equivalent to the noising paradigm in this study are used (L. D. Liberman & Liberman, 2015), while for PTS, noising levels of 100 dB or higher are used (Möhrle et al., 2016). Besides, it is found that vulnerability for noise decreases with age, as a 100 dB pressure level led to PTS in 8 weeks old CBA-CaJ mice but to TTS in 16 weeks old CBA-CaJ mice (Kujawa & Liberman, 2006; Möhrle et al., 2016). The current study noised 50 weeks old mice with a TTS-like noising paradigm. This could explain why no apparent differences were found between noised and not noised 59 weeks old B6.CAST mice in this study. As two B6.CAST mice were exposed to a 104 dB rather than a 98 dB pressure level, it was interesting to see that these two mice, in contrast to the three other mice, had the lowest synapses per IHC values for especially the frequencies above 20 kHz. This could imply that increasing the pressure level is indeed necessary to induce noise-related hearing and synapse loss at older ages. Future studies can look into the relation between pressure level, age and also strain, as these all seem to be related when it comes to noise-induced cochlear damage.
It is thus even more interesting to see that in the GPX1 KO mice the effect of the current noising paradigm is so severe. While the current results confirm a lack of noise-induced apical synapse loss (Fernandez et al., 2015), noise-exposed GPX1 KO mice did show reduced synapses per IHC at mid-basal frequencies (≥20 kHz) compared to unexposed GPX1 KO mice, exposed- and unexposed B6.CAST mice. Apparently, the reduced age-related vulnerability to noise which is seen
in previous studies, does not occur for B6.CAST mice who miss the GPX1 enzyme. Synapse loss started to become clear from 20 kHz onwards, which is a similar starting point of the previously found auditory threshold shifts in GPX1 KO mice (K. K. Ohlemiller et al., 2000). It will thus be highly interesting to include the gathered ABR data of the current study to see if the mice from all groups have ABR wave 1 amplitudes that parallel the (loss of) cochlear synapses, especially as the first ABR wave is a representation of all activity of the ANFs contacting the IHCs (Buchwald and Huang, 1975).
When comparing the current result to result from Liberman & Liberman (2015), it becomes clear that the 50 weeks old regular noised B6.CAST mice don’t show the same dynamics as the 16- or 8 weeks old noised CBA-CaJ mice. Rather, it’s the 50 weeks old GPX1 KO mice that show comparable dynamics with the 16 weeks old CBA-CaJ mice. This once again shows the strain-, age- and noise- dependency of these kind of studies. It will be highly valuable for future studies to find more consensus about standardizing these variables, so that results can be more reliably compared.
For future studies it will be interesting to extend on this seemingly protective role of the GPX1 enzyme against noise-induced synaptopathy. Just like in previous studies (Fernandez et al., 2015; Furman, Kujawa, & Liberman, 2020; Jensen et al., 2015), it will be interesting to also look at post-synaptic glutamate patches, as these seem to be internalized following noise-exposure. Besides, it seems that the noise-induced cochlear synaptopathy is selective for ANFs with low spontaneous rates. Following up on
this with the GPX1 KO model could contribute to this body of research by further identifying the role of this protective enzyme. As synapses don’t seem to regenerate after they are lost, much can be won by preventing synapse loss (Brigande & Heller, 2009; Zhang et al., 2020). Obviously the best way to reduce cochlear damage, is by reducing the chances of being exposed to enduring and/or high intensity noisy environments, but as this seems challenging in the current society, it will be worthwhile further investigating potential therapeutic roles of anti-oxidants, especially as the anti-oxidant defence mechanism is a natural defense mechanism in the human cochlea.
4.5 & 4.6 | Qualitative findings and caveats
While quantitative analysis can expose group differences based on averaged numbers, it does not show individual differences amongst all the unique samples looked at in every unique study. Therefore, this section will describe qualitative findings which have been found during the data gathering process and which I ought worth mentioning, as it might help future studies understand and contextualize individual differences between samples, which are not represented in the averaged quantitative data.
As these results are not yet analyzed to the same extend as the above described results, this section has been taken up in the appendix. Besides, the appendix also holds a section that further elaborates on the current studies caveats.
4.7 | Conclusion
The rationale of this kind of research is the understanding of the process behind ARHL and NIHL, with the eventual goal to improve the protection against hearing loss, or even to find ways to promote recovery from hearing loss. While regenerative methods for hair cell regeneration and synapse regeneration have not really taken off yet, the current study provides interesting insights in the role of one of the natural defense mechanism enzymes, GPX1. As GPX1 showed mild protection against ARHL and strong protection against NIHL in especially the higher cochlear frequencies, it could be worthwhile to dive deeper into the protective effects of the GPX1 anti-oxidant enzyme. However, as GPX1 is only one of many enzymes and the ant-oxidant defense mechanism is not the only protective process in the cochlea, these claims should be treated with caution and further investigation is required. Nevertheless, as anti-oxidant supplements are well known and widely used for various health benefits, the GPX1 enzyme could potentially become a good therapeutic target for noise-protective anti-oxidant treatments.
5 | Acknowledgements
I would like to thank my supervisor Sonja Pyott for giving me the chance to work on this interesting project during an especially difficult period. I felt recognised as an almost-finished master student and the independency that I received during this project was absolutely great. I also like to thank Harm Krugers for fulfilling the role as assessor and being flexible with the
deadline time. I was lucky to be able to conduct my project without major restrictions. For this I need to thank Klaas Sjollema, the head of the University Microscopy and Imaging Center, who found the time to train me and help me with lab-related questions. I’d like to also thank Pim van Dijk for being open to my questions when I was still in the phase of deciding where to do my internship. During the same phase Vincent Tijms also helped me a lot with finding the right people who worked in the field I was interested in. Lastly, I’d like to thank my colleagues for helping me with science-related questions but also for the infrequent lunches and bouldering sessions we had. I especially want to thank Nick Schubert for helping me with my results.
6 | Literature
Barone, C. M., Douma, S., Reijntjes, D. O. J., Browe, B. M., Köppl, C., Klump, G., … Pyott, S. J. (2019). Altered cochlear innervation in developing and
mature naked and Damaraland mole rats. Journal of Comparative
Neurology, 527(14), 2302–2316. https://doi.org/10.1002/cne.24682 Brigande, J. V., & Heller, S. (2009). Quo
vadis, hair cell regeneration? Nature Neuroscience, 12(6), 679–682. https://doi.org/10.1038/nn.2311 Brigelius-Flohe, R. (1999). Tissue-Specific
Functons of Individual Glutathione Peroxidases. Free Radical Biology and Medicine, 27(99), 951–965. Caciari, T., Rosati, M. V., Casale, T., Loreti,
B., Sancini, A., Riservato, R., … Tomei,
G. (2013). Noise-induced hearing loss in workers exposed to urban stressors. Science of the Total Environment, 463–464, 302–308. https://doi.org/10.1016/j.scitotenv.2 013.06.009
Colzato, L. S., Barone, H., Sellaro, R., & Hommel, B. (2017). More attentional focusing through binaural beats: evidence from the global–local task. Psychological Research, 81(1), 271– 277.
https://doi.org/10.1007/s00426-015-0727-0
Cunningham, L. L., & Tucci, D. L. (2017). Hearing loss in adults. New England Journal of Medicine, 377(25), 2465– 2473.
https://doi.org/10.1056/NEJMra1616 601
Davis, A. C., & Hoffman, H. J. (2019). Hearing loss: Rising prevalence and impact. Bulletin of the World Health Organization, 97(10), 646-646A. https://doi.org/10.2471/BLT.19.2246 83
Eggermont, J. J. (2017). Tinnitus and Stress. https://doi.org/10.1007/978-3-319-58397-6
Eileen, D. (2007). Noise and hearing loss: A review. Journal of School Health, 77(5), 225–231.
Evans, P., & Halliwell, B. (1999). Free Radicals and Hearing. Annals of the New York Academy of Sciences, 19– 40.
Fernandez, K. A., Jeffers, P. W. C., Lall, K., Liberman, M. C., & Kujawa, S. G. (2015). Aging after noise exposure:
Acceleration of cochlear
synaptopathy in “recovered” ears. Journal of Neuroscience, 35(19), 7509–7520.
https://doi.org/10.1523/JNEUROSCI. 5138-14.2015
Fetoni, Anna R., Picciotti, P. M., Paludetti, G., & Troiani, D. (2011).
Pathogenesis of presbycusis in animal models: A review.
Experimental Gerontology, 46(6), 413–425.
https://doi.org/10.1016/j.exger.2010. 12.003
Fetoni, Anna Rita, De Bartolo, P., Eramo, S. L. M., Rolesi, R., Paciello, F., Bergamini, C., … Troiani, D. (2013). Noise-induced hearing loss (NIHL) as a target of oxidative stress-mediated damage: Cochlear and cortical responses after an increase in antioxidant defense. Journal of Neuroscience, 33(9), 4011–4023. https://doi.org/10.1523/JNEUROSCI. 2282-12.2013
Furman, A. C., Kujawa, S. G., & Liberman, M. C. (2020). Noise-induced
cochlear neuropathy is selective for fibers with low spontaneous rates, 577–586.
https://doi.org/10.1152/jn.00164.201 3
Glowatzki, E., & Fuchs, P. A. (2002). Transmitter release at the hair cell ribbon synapse. Nature
Neuroscience, 5(2), 147–154. https://doi.org/10.1038/nn796 Henry, Kenneth, R., & Chole, Richard, A.
(1980). genotypic differences in behavioral, physiological and
anatomical expressions of Age-Related Hearing Loss in the Laboratory Mouse. Audiology, 19, 369–383.
Hequembourg, S., & Liberman, M. C. (2001). Spiral ligament pathology: A major aspect of age-related
cochlear degeneration in C57BL/6 mice. JARO - Journal of the Association for Research in Otolaryngology, 2(2), 118–129. https://doi.org/10.1007/s101620010 075
Jensen, J. B., Lysaght, A. C., Liberman, M. C., Qvortrup, K., & Stankovic, K. M. (2015). Immediate and delayed cochlear neuropathy after noise exposure in pubescent mice. PLoS ONE, 10(5), 1–17.
https://doi.org/10.1371/journal.pone .0125160
Keithley, E. M., Canto, C., Zheng, Q. Y., Fischel-Ghodsian, N., & Johnson, K. R. (2004). Age-related hearing loss and the ahl locus in mice. Hearing Research, 188(1–2), 21–28.
https://doi.org/10.1161/CIRCULATIO NAHA.110.956839
Khimich, D., Nouvian, R., Pujol, R., Dieck, T. S., Egner, A., Gundelfinger, E., & Moser, T. (2005). Hair cell synaptic ribbons are essential for
synchronous auditory signalling. Nature, 434(7035), 889–893.
https://doi.org/10.1038/nature03393 Kil, J., Pierce, C., Tran, H., Gu, R., & Lynch,
E. D. (2007). Ebselen treatment reduces noise induced hearing loss via the mimicry and induction of glutathione peroxidase. Hearing
Research, 226(1–2), 44–51.
https://doi.org/10.1016/j.heares.200 6.08.006
Koppenol, W. H. (2001). The Haber-Weiss cycle - 70 years later. Redox Report, 6(4), 229–234.
https://doi.org/10.1179/1351000011 01536373
Krefis, A., Augustin, M., Schlünzen, K., Oßenbrügge, J., & Augustin, J. (2018). How Does the Urban
Environment Affect Health and Well-Being? A Systematic Review. Urban Science, 2(1), 21.
https://doi.org/10.3390/urbansci201 0021
Kujawa, S. G., & Liberman, M. C. (2006). Acceleration of age-related hearing loss by early noise exposure:
Evidence of a misspent youth. Journal of Neuroscience, 26(7), 2115–2123.
https://doi.org/10.1523/JNEUROSCI. 4985-05.2006
Kujawa, S. G., & Liberman, M. C. (2009). Adding Insult to Injury : Cochlear Nerve Degeneration after “
Temporary ” Noise-Induced Hearing Loss, 29(45), 14077–14085.
https://doi.org/10.1523/JNEUROSCI. 2845-09.2009
Le, T. N., Straatman, L. V., Lea, J., & Westerberg, B. (2017). Current insights in noise-induced hearing loss: a literature review of the underlying mechanism,
pathophysiology, asymmetry, and management options. Journal of Otolaryngology - Head and Neck Surgery, 46(1), 1–15.
https://doi.org/10.1186/s40463-017-0219-x
Liang, Q., Shen, N., Lai, B., Xu, C., Sun, Z., & Wang, Z. (2019). Electrical
Stimulation Degenerated Cochlear Synapses Through Oxidative Stress in Neonatal Cochlear Explants, 13(October), 1–13.
https://doi.org/10.3389/fnins.2019.0 1073
Liberman, L. D., & Liberman, M. C. (2015). Dynamics of cochlear synaptopathy after acoustic overexposure. JARO - Journal of the Association for
Research in Otolaryngology, 16(2), 205–219.
https://doi.org/10.1007/s10162-015-0510-3
Liberman, M. C. (2017). Noise-induced and age-related hearing loss: New perspectives and potential therapies. F1000Research, 6(0), 1–11.
https://doi.org/10.12688/f1000resea rch.11310.1
Liberman, M. C., & Kujawa, S. G. (2017). Cochlear synaptopathy in acquired sensorineural hearing loss:
Manifestations and mechanisms. Hearing Research, 349, 138–147. https://doi.org/10.1016/j.heares.201 7.01.003
Lin, H, W., Furman, A, C., Kujawa, S. G., & Liberman, M. C. (2011). Primary Neural Degeneration in the Guinea Pig Cochlea After Reversible Noise-Induced Threshold Shift. JARO - Journal of the Association for Research in Otolaryngology, 616(12), 605–616.
https://doi.org/10.1007/s10162-011-0277-0
Masterson, E. A., Themann, C. L., & Calvert, G. M. (2018). Prevalence of Hearing Loss Among Noise-Exposed Workers Within the Health Care and Social Assistance Sector. Journal of Occupational and Environmental Medicine, 60(9), e511.
https://doi.org/10.1097/jom.000000 0000001435
Mikaelian, D. O. (1979). development and degeneration of hearing in the C57/BL6 mouse: relation of
electrophysiologic responses from the round window and cochlear nucleus to cochlear anatomy and behavioral responses.
Möhrle, D., Ni, K., Varakina, K., Bing, D., Lee, S. C., Zimmermann, U., … Rüttiger, L. (2016). Loss of auditory sensitivity from inner hair cell synaptopathy can be centrally compensated in the young but not old brain. Neurobiology of Aging, 44, 173–184.
https://doi.org/10.1016/j.neurobiola ging.2016.05.001
Morioka, S., Sakaguchi, H., Yamaguchi, T., Ninoyu, Y., Mohri, H., Nakamura, T., … Ueyama, T. (2018). Hearing vulnerability after noise exposure in a mouse model of reactive oxygen species overproduction. Journal of Neurochemistry, 146(4), 459–473. https://doi.org/10.1111/jnc.14451 Moser, T., Neef, A., & Khimich, D. (2006).
Mechanisms underlying the
temporal precision of sound coding at the inner hair cell ribbon synapse. Journal of Physiology, 576(1), 55–62.
https://doi.org/10.1113/jphysiol.200 6.114835
Moser, T., Predoehl, F., & Starr, A. (2013). Review of hair cell synapse defects in sensorineural hearing impairment. Otology and Neurotology, 34(6), 995–1004.
https://doi.org/10.1097/MAO.0b013 e3182814d4a
Muller, M., Hunerbein, K. Von, Hoidis, S., & Smolders, J. W. T. (2005). A
physiological place – frequency map of the cochlea in the CBA / J mouse. Hearing Research, 202, 63–73. https://doi.org/10.1016/j.heares.200 4.08.011
Ohlemiller, K. K., McFadden, S. L., Ding, D. L., Lear, P. M., & Ho, Y. S. (2000). Targeted mutation of the gene for cellular glutathione peroxidase (Gpx1) increases noise-induced hearing loss in mice. JARO - Journal of the Association for Research in Otolaryngology, 1(3), 243–254. https://doi.org/10.1007/s101620010 043
Ohlemiller, Kevin K., Jones, S. M., & Johnson, K. R. (2016). Application of Mouse Models to Research in Hearing and Balance. JARO - Journal of the Association for Research in Otolaryngology (Vol. 17). https://doi.org/10.1007/s10162-016-0589-1
Ohlemiller, Kevin K., Wright, J. S., & Dugan, L. L. (1999). Early elevation of cochlear oxygen species following reactive noise exposure. Audiology and Neuro-Otology, 4(5), 229–236. https://doi.org/10.1159/000013846
Pierson, M. G., & Gray, B. H. (1982). Superoxide dismutase activity in the cochlea. Hearing Research, 6(2), 141–151.
https://doi.org/10.1016/0378-5955(82)90050-8
Rita, A., Paciello, F., Rolesi, R., Paludetti, G., Troiani, D., Policlinico, F., & Gemelli, U. A. (2019). Free Radical Biology and Medicine Targeting dysregulation of redox homeostasis in noise-induced hearing loss : Oxidative stress and ROS signaling. Free Radical Biology and Medicine, 135(February), 46–59.
https://doi.org/10.1016/j.freeradbio med.2019.02.022
Ryan, A. F., Kujawa, S. G., Hammill, T., Le Prell, C., & Kil, J. (2016). Temporary and Permanent Noise-induced Threshold Shifts: A Review of Basic and Clinical Observations. Otology and Neurotology, 37(8), e271–e275. https://doi.org/10.1097/MAO.00000 00000001071
Sergeyenko, Y., Lall, K., Liberman, M. C., & Kujawa, S. G. (2013). Age-Related Cochlear Synaptopathy : An Early-Onset Contributor to Auditory Functional Decline, 33(34), 13686– 13694.
https://doi.org/10.1523/JNEUROSCI. 1783-13.2013
Sha, S. H., & Schacht, J. (2017). Emerging therapeutic interventions against noise-induced hearing loss. Expert Opinion on Investigational Drugs, 26(1), 85–96.
https://doi.org/10.1080/13543784.2 017.1269171
Vázquez, A. E., Jimenez, A. M., Martin, G. K., Luebke, A. E., & Lonsbury-Martin, B. L. (2004). Evaluating cochlear function and the effects of noise exposure in the B6.CAST + Ahl mouse with distortion product otoacoustic emissions. Hearing Research, 194(1–2), 87–96.
https://doi.org/10.1016/j.heares.200 4.03.017
Wang, J., & Puel, J. L. (2018). Toward cochlear therapies. Physiological Reviews, 98(4), 2477–2522.
https://doi.org/10.1152/physrev.000 53.2017
Wong, A. C. Y., & Ryan, A. F. (2015). Mechanisms of sensorineural cell damage, death and survival in the cochlea. Frontiers in Aging
Neuroscience, 7(APR), 1–15.
https://doi.org/10.3389/fnagi.2015.0 0058
Wu, F., Xiong, H., & Sha, S. (2020). Noise-induced loss of sensory hair cells is mediated by ROS/AMPKα pathway. Redox Biology, 29(December 2019), 101406.
https://doi.org/10.1016/j.redox.2019. 101406
Yamane, H., Nakai, Y., Takayama, M., Iguchi, H., Nakagawa, T., & Kojima, A. (1995). Appearance of free radicals in the guinea pig inner ear after noise-induced acoustic trauma. European Archives of Oto-Rhino-Laryngology, 252(8), 504–508. https://doi.org/10.1007/BF02114761 Zhang, S., Qiang, R., Dong, Y., Zhang, Y., Chen, Y., Zhou, H., … Chai, R. (2020). Hair cell regeneration from inner ear
progenitors in the mammalian cochlea. American Journal of Stem Cells, 9(3), 25–35. Retrieved from http://www.ncbi.nlm.nih.gov/pubme d/32699655%0Ahttp://www.pubmed central.nih.gov/articlerender.fcgi?arti d=PMC7364385
