The rat as a model for studying noise injury and otoprotection

The use of laboratory rats (Rattus norvegicus) in hearing research dates back to the work of Henry on auditory and discriminative functions in the white rat (Henry, 1938; Firstova et al., 2012). Early studies describing the hearing sensitivity of the rat (0.20–85 kHz) reported the upper range of hearing as more than four times that found in people (Gould and Morgan, 1941). These early studies also included behavioral training (Gould and Morgan, 1941), thereby allowing psychophysical assessment of hearing and related pathological conditions. Since then, the laboratory rat has become a common and useful animal model for hearing research. This article will consider both the advantages and limitations of the use of rats to study noise-induced hearing loss (NIHL), including in the development of treatments.

Laboratory rats are relatively straightforward to work with since they tend to be non-aggressive, can be trained easily, and have a lifespan of between 2.5 and 3.5 years (Sharp, 2010) with female rats entering menopause around 18 months while menopause occurs in human females between 48 and 55 years of age. On the surface these facts suggest that ∼11.8 rat days equal one human year in adult rats (Andreollo et al., 2012). However, this does not seem to apply to physiological processes such as noise injury. For example, that calculation would imply that when a rat recuperates from a mild noise injury after three days a person would not recover from an equivalent noise until approximately 66 days. Adult male rats range in weight from 300 to 520 g, while adult female rats weigh between 250 and 300 g. Although they belong to the order of Rodentia, which also includes mice, beavers, hamsters, and porcupines, rats have a distinct hearing profile.

As alluded to earlier, while hearing in people ranges from 20 to 20 000 Hz, the rat has a right shifted (but expanded at higher frequencies) audiogram ranging from 200 to 90 000 Hz (Warfield, 1973; Fay, 1988). Humans are “precocial” in terms of hearing, with the ability to hear beginning in utero. Rats are “altricial” in nature, with cochlear development such that the ability to hear occurs postnatally and the ear canal does not open until the second postnatal week precluding the recording of auditory brain stem responses until that time (Romand and Anniko, 1983; Rybak et al., 1991; Romand, 1992; Jones and Jones, 2011; Romand and Varela-Nieto, 2014).

The organization of the rat cochlea is comparable to cochleae found in humans and other mammalians, with similar transduction mechanisms and synaptic connections. The rat cochlear spiral has 2.5 turns (Igarashi et al., 1968; Albuquerque et al., 2009), analogous to that found in people. There is “natural” variability in length even within inbred rat strains. The most basal region has a large bend that is not observed in the human cochlea; this bend is often termed the “hook” and is also found in the mouse cochlea (less so in guinea pig or chinchilla cochleae). The “hook” region in the rat is approximately 2 mm long, the middle portion is ∼4 mm long, and the apical portion is ∼3 mm long. Therefore, the usual length of the cochlear spiral in rat is reported to be ∼9 mm, compared to ∼7 mm in mouse, ∼20 mm in guinea pig and ∼30 mm in people. Hair cells are arranged tonotopically along the cochlear spiral from apex to base with hair cells responding best to low frequencies closer to the apex and those responding best to high frequencies closer to the base. The number of inner hair cells in rat cochlea is just under 1000 while the number of outer hair cells is approximately 3500. The number of spiral ganglion neurons is approximately 16 000 (Keithley and Feldman, 1979). The human cochlea has approximately 3500 inner hair cells, 12 500 outer hair cells, and 35 000 spiral ganglion cells. These characteristics have implications when determining the type of noise to deliver and the formulation of testable hypotheses when generating NIHL and related conditions.

Rats used for auditory research have primarily been of the outbred variety and are commercially available. The rats can be divided into two main groups, non-pigmented and pigmented. Wistar, Fischer 344, and Sprague-Dawley rats, which are all albino, are the most commonly used in NIHL-related research. Pigmented rats such as Long Evans, Fischer Brown Norway (FBN), and FBR (a cross between Fischer 344 and FBN) are also used frequently to study hearing related conditions such as tinnitus and aging. Pigment allows for easier visualization and dissection of the cochlear epithelium. Whether pigment changes susceptibility to NIHL has been the focus of several studies in rodents with varying results (Yanz et al., 1985; Conlee et al., 1986; Chen and Chen, 1990; Overbeck and Church, 1992). These models are interesting in light of studies that have suggested a positive correlation between people with more melanin and hearing acuity (Post, 1964; Anticaglia and Cohen, 1970; Karsai et al., 1972; Attias and Pratt, 1985). While people exhibiting albinism are reported to have reduced hearing sensitivity, albino rats appear to have fairly normal hearing sensitivity and sound localization (Kelly and Masterton, 1977; Creel et al., 1980; Garber et al., 1982; Heffner and Heffner, 1985; Syka et al., 1996).

In terms of low frequencies, the non-pigmented Fischer 344 rat demonstrates better hearing by 20 dB. While for high frequencies (32 kHz), the pigmented FBN rat displays comparatively enhanced hearing, by approximately 20 dB (Turner et al., 2005). Generally, the Fischer Brown Norway rats are most often used in auditory studies related to aging and appear more vulnerable to noise insult (Syka, 2010). In addition to the commonly used strains mentioned above, a plethora of rat strains with diseases, conditions, and disorders are available and many have been used in noise susceptibility studies.

Now that the rat genome has been sequenced, testing for changes in gene expression, comparisons of genetic differences across strains, and manipulation of genes to create knock-out, knock-in models can be used to study the role of heritable and non-heritable genes on susceptibility to NIHL. With the availability of viral vectors and genetic editing tools such as CRISPR, modulating gene expression has become even more accessible in the rat. This allows exploitation of some of the advantages of using the rat for gene expression studies in noise injury models, including the larger brain and more cochlear epithelium when compared to the widely used mouse models. Consequently, more genetic material can be obtained from fewer subjects. While even larger animals exist with larger brains and cochleae, their genomes have not been studied as widely as the rat.

The audiogram is a feature that can be used when selecting an animal model for research related to hearing injury. When comparing the audiogram across laboratory animals used in pre-clinical studies (e.g., chinchilla, gerbil, rabbit, guinea pig, mouse, and rat), the chinchilla and gerbil are most closely matched to human audiograms (Fay, 1988). However, gerbils demonstrate a central auditory system degenerative condition with microcysts and vacuoles in the cochlear nerve root and cochlear nucleus that spread throughout the auditory pathway with age. The upper limit of hearing in mice is 85 kHz while the lower limit of mouse hearing is ∼2.35 kHz. While rats hear the upper limit in rats is fairly high, ∼70 kHz, the lower limit is 0.520 kHz. Human hearing ranges from an upper limit of 20 kHz to a lower limit of 20 Hz making the rat audiogram closer to the human audiogram than that of the mouse. In addition, many strains of mice demonstrate genetic-related hearing loss after two months of age (ahl) and are more susceptible to NIHL, while rats do not appear to have such a loss (Turner et al., 2005; Willott, 2007).

In addition, Sprague-Dawley rats do not appear to show elevated levels of aggressive behavior after salicylate administration or noise exposure (Lauer et al., 2018), unlike mice and Wistar rats (Guitton, 2009; Zheng et al., 2011).

In both people and rats, the degree of noise-induced injury is often assessed based upon hearing sensitivity after the trauma. The degree of recovery from hearing loss is typically used to classify subjects into two categories. Under the first condition, subjects sustain a short-term elevation in hearing thresholds, a so-called temporary threshold shift (TTS) following exposure to loud sound that returns to normal within hours to days. The second condition, however, results in an irreversible elevation in hearing thresholds after experiencing loud sounds, termed as a permanent threshold shift (PTS). During the noise exposure there is a “critical point” at which the resulting injury changes from a TTS to a PTS. This could be due to many factors, including age, health condition, or genetics. For example, chinchillas are less susceptible to solvent facilitated NIHL when compared to rats, mice, and humans. In mice, the noise dose response curve is significantly different across strains (Davis et al., 1999). In the rat, young animals (less than seven weeks of age) are more susceptible to NIHL (Rybalko and Syka, 2001) with little variability from animal to animal and no effect of strain (Borg, 1982). However, in older animals, strain is a factor in high frequency NIHL. The length of time the noise exposure lasts, the intensity of the noise exposure, and the type of noise all contribute to this crucial, but as yet to be identified, critical point in the injury process. And as suggested by the aforementioned studies, the associated genetics of a given strain is a crucial determinant of degree of injury and recovery.

The frequency of the noise exposure impacts the location and distribution of noise-induced effects along the cochlear spiral and within the brain. The greatest influence of a narrow frequency band noise exposure is often above the frequency of the noise, but this can be dependent upon the sound pressure level (SPL) and duration of the noise. Noise exposure from a broad band of noise frequencies can be less predictable in terms of the expected hearing loss, hair cell loss, and central changes.

While early reports on rat auditory sensitivity were often limited in terms of the ability to measure very low or very high frequencies, those studies were able to determine that rat hearing is less sensitive at lower frequencies and more sensitive at higher frequencies (⟨8 kHz⟩) when compared to people (Gould and Morgan, 1941). As the sensitivity of equipment has improved, the assessment of the range of hearing in rats has become more accurate, especially for higher frequencies (Gould and Morgan, 1941; Gourevitch and Hack, 1966; Kelly and Masterton, 1977).

Hearing loss can be measured in rats using methods similar to those used in people, including behavioral paradigms and/or physiological assessments using auditory brain stem responses (ABRs) to determine hearing thresholds and otoacoustic emissions (OAEs) to assess hair cell function. Using animal models for cochlear assessment also allows for the possibility of collecting physiological measures not easily collected in people. For example, the cochlear microphonic (Wever and Bray, 1930; Adrian, 1931), summating potential (SP) and compound action potential (CAP) are electrical recordings (electrocochleography) from the cochlea and the vestibulo-cochlear nerve (CN VIII). The recordings are collected under general anesthesia from the tympanic membrane (Salomon and Elberling, 1971), or with a needle electrode placed through the tympanic membrane on the round window, or the cochlear promontory wall (Eggermont and Odenthal, 1974). The CM and SP provide assessment of all cochlear hair cells. While the CAP electrical activity (same as ABR wave I activity) allows assessment of the integrity of CN VIII. The advantage of CAP is that the evoked potential measured at the cochlear promontory can be 100 times greater than that measured at the mastoid—as with ABRs. Many types of cochlear responses have been collected from rats (Davis and Cleveland, 1934; Tasaki, 1954) and their origin ascertained (Goldstein and Kiang, 1958; Dallos and Cheatham, 1976; Durrant et al., 1998). Using these cochlear responses, certain cochlear dysfunction states can be ascertained. An SP/CAP ratio of more than 30% is usually indicative of endolymphatic hydrops associated with Menière's disease. When the CM is normal, but the CAP is absent, then auditory neuropathy is the diagnosis.

In addition to cochlear responses, the size of the rat skull and brain allows for implantation of electrode arrays into specific brain regions for longitudinal assessment of neuronal activity and gene expression before and after acoustic trauma and/or deafness (Holt et al., 2006; Otazu et al., 2009; von der Behrens et al., 2009; Fyk-Kolodziej et al., 2011; Fyk-Kolodziej et al., 2015; Holt et al., 2016). Being able to collect cochlear responses and perform other invasive physiological recordings and molecular assessments reflective of cochlear or auditory related brain function is a great benefit of using the rat.

Animal studies, including those in rat, suggest there is not just one underlying cause for a noise-induced TTS and the subsequent recovery. The effects of noise can be “mechanical” or “metabolic.” The mechanical effects of noise can include diminished stereocilia stiffness, loss of tip-links between hair cell stereocilia or effects on other components of the mechano-transduction complex. Mild or moderate noise can cause loss of either tip links between hair cells or disruption of actin organization in stereocilia, both of which can re-form or re-organize and restore hearing.

In terms of metabolic effects, noise-induced oxidative stress can disrupt subsequent phases of transduction resulting in diminished hair cell function and transmitter release. Repair and recovery is possible if the noise exposure is not beyond the “critical point.”

The degree and duration of the TTS can also vary depending on length of time and the intensity and type of the noise exposure. The frequencies and range of TTS will correlate with the frequencies and range of the noise overstimulation (e.g., Fredelius et al., 1987).

Outer hair cell function is more susceptible to noise disruption than inner hair cells in both rats and humans (although inner hair cell synapses may be even more susceptible—see Sec. VI). The effect of loss of OHCs on threshold shift varies across the cochlear spiral, ranging between 40 and 60 dB in several animal models including the rat (Stebbins et al., 1979; Hill et al., 2016). Therefore, a threshold shift of 55 dB or below suggests that outer hair cell function is primarily impacted, while threshold shifts of 65 dB and above suggest inner hair cell involvement as well. Assessment of distortion product otoacoustic emissions (DPOAEs) can be used to directly determine outer hair cell function and possible loss.

Noise can induce a mix of TTS and PTS, with some function returning over time leading to partial but not complete recovery of hearing metrics. There is no firm answer as to how long one must wait to differentiate the TTS and PTS components, and similar to work in mice (Jensen et al., 2015), rats can have continued pathological changes following noise for months after a noise exposure (Muca et al., 2018). However, the majority of hair cell loss occurs between three and 21 days following a noise and this range has been used in rats to differentiate TTS and PTS components of NIHL (Yamashita et al., 2004; Sha and Schacht, 2017).

As with TTS, noise-induced PTS can result from different causes and underlying mechanisms, the most common cause being loss of hair cells. As with TTS the causes of auditory pathway damage can be divided into “mechanical” and “metabolic” categories. Mechanical forces resulting from loud noise can disrupt cellular junctions, produce leakage of fluids and generate hair cell death by ionic poisoning (e.g., Ahmad et al., 2003). Mechanical overstimulation of individual cells produces bursting and physical rupturing of whole cells or disruption of specific cellular components, such as F-actin in the cuticular plate (Hu and Henderson, 1997). Massive overstimulation of stereocilia can cause breakage leading to cell death. Mechanical cell death can be by apoptosis or necrosis. Necrosis is often accompanied by inflammation. Noise-induced metabolic causes include excessive free radical formation leading to oxidative stress, ER stress, as well as energy and mitochondrial changes (Steinkamp and Schachtschabel, 2001; Wang et al., 2018). Cell death from metabolic changes is most often by apoptosis. This can begin within minutes following noise but can also be delayed, occurring days or weeks after the noise exposure. As with TTS, OHCs are most susceptible to noise-induced PTS. When testing for NIHL in animal models such as rat, the most common approach is to perform both histological assessment (examining hair cell loss along the cochlear spiral) and functional assessment (determining changes in hearing sensitivity and outer hair cell function). There is often a correlation between hair cell damage in specific locations along the cochlear spiral and threshold shifts at specific frequencies. The correlation between the amount of auditory hair cell loss and degree of threshold shift in the rat has been reported to be inconsistent (Chen and Fechter, 2003). Hair cells in the base of the rat cochlea show a linear relationship between measurable ABR thresholds and loss. However, apical auditory hair cells show no such relationship in the rat cochlea. This could result because although hair cells are present, they may not be functional and therefore threshold shifts would be greater than the loss of hair cells. In this way the rat cochlea is similar to the human cochlea, with basal auditory hair cells showing more vulnerability to damage (Johnsson and Hawkins, 1972). Alternatively, these hair cells might have a delayed cell death or even eventually recover (Yamashita et al., 2004). There could also be more hair cell loss than threshold shift, since ABRs are measures of total auditory nerve responses and there could be sufficient auditory neural response from remaining inner hair cells so that some loss is not noticeable (Chen et al., 2008). There is literature available for different animal models including rat and the review by Viberg and Canlon (2004) summarizes these and provides useful mapping of frequencies along the rat cochlear spiral.

The division between TTS and PTS has become more complicated recently with studies in animal models showing there can be a permanent noise-induced “hidden hearing loss” embedded in the TTS. Noise overstimulation will induce a loss of synaptic connections between inner hair cells (IHCs) and the auditory nerve (AN), termed a “synaptopathy.” These IHC–AN synapses can be sub-divided based on their response properties with some becoming active even at low sound levels (high sensitivity–high spontaneous activity), others becoming active only with very loud sounds (low sensitivity–low spontaneous activity) and others with responses in between these. This provides “dynamic range” to auditory nerve activity. Studies show that with many noise exposure conditions, the low sensitivity–low spontaneous rate connections are most likely to be lost (Furman et al., 2013). This can occur even under “TTS” conditions where ABR thresholds recover. There is some capacity for re-connection of lost IHC–AN synapses and this was once suggested as one mechanism for TTS in the guinea pig model (Pujol and Puel, 1999). Subsequent studies in mouse and rat, however, have shown that such re-connection following noise is either inefficient or does not occur at all, leading to a permanent hearing loss. If there is loss of the low sensitivity IHC–AN synaptic connections, then the ability to have increased AN responses to very loud sounds is diminished resulting in reduced dynamic range. While this can be determined using ABR assessment with suprathreshold measures, this is not part of typical ABR and audiological assessments and so the phenomenon has been termed hidden hearing loss. Until recently the majority of synaptopathy studies have been conducted in mouse and guinea pig models of NIHL. However, several studies have now demonstrated synaptopathy in rat models as well (Singer et al., 2013; Altschuler et al., 2016; Hickox et al., 2017; Altschuler et al., 2019).

Tinnitus, a ringing sound experienced in the absence of an external stimulus, is often experienced either acutely or chronically following exposure to therapeutic drugs (e.g., salicylate and cisplatin), traumatic brain injury, or loud noise. Behavioral studies in animals for determining tinnitus status were first performed in the rat (Jastreboff et al., 1988a; Jastreboff et al., 1988b). Since then a myriad of behavioral studies have continued to be performed to assess tinnitus in the rat and across many species (Table I).

Rats have traditionally been used for most behavioral studies when compared to mice based upon empirical experiences suggesting that mice were difficult to work with (temperament) and posed a challenge in terms of training (Kurt and Ehret, 2010; Steimer, 2011; Teegarden, 2012). Many recent tinnitus studies have made use of the Gap detection paradigm (acoustic startle reflex), which is not dependent upon cortical processing or behavioral training and may help to alleviate some of these behavioral based concerns. Nonetheless, to date, the majority of behavioral related tinnitus data have been collected in the rat. Similar to people, the tinnitus in rats is often reported to be at frequencies higher than the region of highest sensitivity (3–4 kHz in people and 8 kHz in rat). However, in the mouse the tinnitus is reported at the region of highest sensitivity, suggesting the rat as a better model for comparison to people. The aforementioned studies report that after exposure to loud noise producing either a TTS or a PTS, rats demonstrate psychophysical or startle response (gap detection) deficits that correlate with tinnitus. These altered physical responses can last from hours to months (Turner et al., 2006; Heffner, 2011) and are sensitive to hearing loss. In most cases the rat needs to have hearing in at least one ear for the test results to be meaningful (Rybalko and Syka, 2005; Turner et al., 2006; Knipper et al., 2013; Hayes et al., 2014; Hickox and Liberman, 2014; Galazyuk and Hebert, 2015; Muca et al., 2018). Loud noise can result in seizures in specific genetic models, suggesting increased excitability in specific populations of neurons (Skradski et al., 2001; Faingold, 2002; De Sarro et al., 2017). A similar change in neuronal excitability is often observed in models of noise induced tinnitus, with rats showing changes in neuronal activity (Milbrandt et al., 1996; Kaltenbach, 2006; Brozoski et al., 2007a; Dong et al., 2010; Sanes and Kotak, 2011; Wang et al., 2011; Muca et al., 2018), as well as in the expression of genes and level of proteins involved in plasticity (Tan et al., 2007; Kraus et al., 2011; Ruttiger et al., 2013), homeostasis, and amino acid-related neurotransmission (Wang et al., 2009; Luo et al., 2014; Holt et al., 2016).

The type of noise can also influence the characteristics of hearing loss. Impulse noises are rapid, near instantaneous, short duration bursts of sound. These sounds can be clicks, bangs or pops such as those associated with electrical interference, gunfire, and explosions. Understanding gunfire-related and blast-induced auditory injury has become an area of intense research considering the significant implications for both civilian and military personnel. Several rat models have been developed to assess auditory pathology following impulse noise (small arms fire and blast). Rats exposed to a single blast overpressure of 22 psi were shown to develop hearing loss, tinnitus and spontaneous firing in the dorsal cochlear nucleus (Luo et al., 2014). The same group further reported spontaneous firing activity in auditory cortex (AC) following a single blast exposure of 22 psi with increased bursting activity in low frequency regions. They suggest that blast generated bursting activity in the AC may play a role in chronic tinnitus (Luo et al., 2017). Chronic tinnitus was also reported in rats subjected to significantly lower blast overpressure levels of 14 psi. Elevated hearing thresholds, as measured by ABR, have also been observed after laser-induced shock wave. The dysfunction was attributed to stereocilia disruption in the outermost row of OHCs (Niwa et al., 2016). This implies a need for more impulse noise studies, particularly blast, to understand the range of overpressure levels that impact hearing related conditions. There is also a need for thorough assessment of the spectral characteristics of the blast to better understand the impact of blast on hearing thresholds as well as peripheral and central injury (Ouyang et al., 2017).

Treatment to prevent NIHL has largely focused on the “metabolic” causes rather than “mechanical” causes. Such metabolic consequences include oxidative stress, ER stress, and mitochondrial stress as well as inflammatory response. All of these can lead to a temporary sensorineural dysfunction or progress to cell death. There are treatment strategies, often pharmacological, that can reduce these stresses. Strategies that have been tested can involve strengthening or enhancing protective mechanisms, blocking harmful mechanisms, and sometimes a combination of the two. This can prevent the harmful effects of noise or sufficiently reduce them to allow recovery. Since regeneration strategies for hair cell loss are not yet available, repair for this facet of NIHL is not yet possible.

Anti-oxidants can be given systemically (sometimes in the diet) prior to noise exposure and reduce noise-induced oxidative stress and the subsequent TTS and PTS (Ada et al., 2017). Carnosine (beta-alanyl-L-histidine), ebselen, and N-acetylcysteine (NAC) are anti-oxidants/free radical scavengers that have been shown to reduce noise-induced TTS and PTS (Duan et al., 2004; Lynch et al., 2004; Zhuravskii et al., 2004; Lorito et al., 2006). Combination treatment of NAC/disodium 2,4-disulfophenyl-N-tert-butylnitrone (HPN-07), a free radical spin trap reagent (Lu et al., 2014), also resulted in reduced noise-induced TTS and PTS, as well as reduced c-fos positive neuronal cells in the cochlear nucleus. Furthermore, NAC treatment was also shown to maintain vertical and symmetrical alignment of stereocilia following acoustic trauma in rats (Ada et al., 2017). The role of melatonin, another antioxidant, was also shown to reduce noise-induced OHC loss and improve hearing. Variations in doses used in these studies highlight the need for arriving at an optimal anti-oxidant procedure for protection and recovery in NIHL. The harmful effects of oxidative stress can begin at various times after the noise exposure and continue for many days. Therefore, one barrier to effective treatment with anti-oxidants has been determining the timing and the localization of the oxidative stress after the noise. The development of methodologies such as QUEST MRI for detection of noise-induced excessive free radical production (Berkowitz, 2018; Kühl et al., 2019) could be beneficial in providing targeted treatments for reducing oxidative stress. Reducing oxidative stress can also be effective for enhancing recovery and reducing PTS (Sha and Schacht, 2017).

Noise can induce changes in cochlear blood flow that can potentiate other consequences of noise. Postnoise exposure treatment with hyperbaric oxygen (90 min daily for 10 days) reduced hair cell loss from impulse noise (Kuokkanen et al., 2000). In cochlear cell cultures subjected to hypoxia and treated with MK801 or magnesium, significant reductions in outer and inner hair cell damage were reported, indicating protective effects of magnesium against impaired oxygenation which may be induced by noise exposure (Konig et al., 2003). Adding magnesium (to enhance blood flow) increased the efficacy of an anti-oxidant mixture of Vitamins A, C, and E in reducing NIHL (Le Prell et al., 2007).

In a recent investigation by Jahani et al. (2016), elevated ABR amplitudes (higher than TTS) were observed 72 h after noise exposure in rats pretreated with low doses of Atorvastatin (5 mg/kg daily) (Jahani et al., 2016). Treatment with resveratrol, another statin, was also shown to offer protection against acoustic trauma in rats as evidenced by well-preserved cochlear structure and normal appearing stereocilia and hair cells (Hanci et al., 2016).

Apoptosis is a non-inflammatory process involving the systematic breakdown of cells, packaging the products into membrane bound structures, and disposed of via phagocytosis. Following exposure to loud noise and subsequent to the activation of oxidative stress, apoptotic pathways are initiated by caspase intrinsic (ROS and cytochrome C) and extrinsic (TNFα) pathways. There are also non-caspase dependent apoptotic processes that can be induced. There are many studies that have explored the progression of apoptotic pathways following noise-induced injury, some of which utilize rat models (e.g., Bodmer et al., 2002; Tabuchi et al., 2007). One set of strategies for preventing hair cells from undergoing noise-induced cell death is to block downstream apoptotic pathways.

Excitotoxicity contributes to the noise-induced loss of synaptic connections between inner hair cells and the auditory nerve (IHC–AN), and strategies for prevention target this loss. Blocking the glutamate receptor will prevent this synaptopathy (Le Prell and Bao, 2012) however, since it will also block hearing, the usefulness of this approach can be restrictive. Glutamate receptor blockage will also have toxic consequences if not restricted to the cochlea. Blocking the NMDA receptor, however, can also reduce excitotoxic effects without toxic risk or loss of hearing, and rats receiving injection of (+)-MK-801 hydrogen maleate (1 mg/kg), an NMDA receptor antagonist, were shown to be protected against permanent CAP threshold elevation following noise (Chen et al., 2001). In rats, treatment with a combination of anti-excitotoxic agents, a dopamine agonist (Piribedil) and an NMDA channel receptor blocker (Memantine), reduced noise-induced loss of synaptic ribbons following exposure to 4 kHz octave band noise at 117 dB SPL for 3 h (Altschuler et al., 2016). A combination treatment that added ACEMg to Memantine and Piribedil reduced noise-induced loss of synaptic ribbons from exposure to a small arms fire-like impulse noise (Altschuler et al., 2019). Recent studies have also shown that, unlike loss of hair cells, loss of IHC–AN connections can be repaired. Intrascalar administration of the neurotrophin NT-3 (from poloxamer) on the round window given one day after a noise results in a large regeneration of the lost IHC-AN synapses (Wang and Green, 2011; Wan et al., 2014; Cunningham and Tucci, 2015; Suzuki et al., 2016).

Finding an effective treatment for managing or mitigating blast-induced auditory trauma has become an urgent priority considering the large number of affected service members and veterans. Accordingly, several studies have been undertaken using rats to address blast-induced auditory trauma. Rats exposed to three consecutive blast overpressures of 14 psi were shown to develop permanent hearing loss when tested 21 days after exposure. In rats a combination treatment of the anti-oxidants 2,4-disulfonyl α-phenyl tertiary butyl nitrone (HPN-07) and N-acetylcysteine (NAC) was shown to offer improvement by decreasing the amount of temporary and permanent threshold shift by 10 and 20 dB, respectively. Furthermore, a reduction in OHC loss also was observed following this combination anti-oxidant treatment (Ewert et al., 2012). Besides blast noise-induced auditory pathology, NAC also was shown to improve NIHL in streptozotocin-induced diabetic rats (Wu et al., 2010).

Several studies were also undertaken to assess the effects of low level laser radiation in NIHL. Unilateral trans-tympanic low level irradiation with an energy output of 100–165 mW/cm was shown to result in reduced hearing thresholds in rats subjected to noise exposure (Rhee et al., 2012). Another investigation suggests that low level laser therapy exerts cytoprotective effects via inhibition of inducible nitic oxide synthase and apoptosis (Tamura et al., 2015). Using bilateral trans tympanic laser therapy in rats subjected to narrow band noise (115 dB SPL for 6 h), hearing thresholds were improved with faster functional recovery (Lee et al., 2016).

The goal of the current article is to provide information to be used for assessing the value of rat models for understanding and repairing the impact of noise injury in future studies. While more rat models are needed to compare gender differences (Lauer and Schrode, 2017), the rat remains an excellent model for studying noise injury and otoprotection. The current article provides evidence for the power of rat models to offer a comprehensive view of the morphology, behavior, peripheral and central function, neuronal activity, and gene and protein levels associated with noise injury. As work with rat models continues, these pre-clinical models will yield more mechanisms and methods, the consequence of which will be treatments and diagnostic tools for prevention of and rescue from NIHL.