Comparison and contrast of noise-induced hyperactivity in the dorsal cochlear nucleus and inferior colliculus

Abstract

Induction of hyperactivity in the central auditory system is one of the major physiological hallmarks of animal models of noise-induced tinnitus. Although hyperactivity occurs at various levels of the auditory system, it is not clear to what extent hyperactivity originating in one nucleus contributes to hyperactivity at higher levels of the auditory system. In this study we compared the time courses and tonotopic distribution patterns of hyperactivity in the dorsal cochlear nucleus (DCN) and inferior colliculus (IC). A model of acquisition of hyperactivity in the IC by passive relay from the DCN would predict that the two nuclei show similar time courses and tonotopic profiles of hyperactivity. A model of acquisition of hyperactivity in the IC by compensatory plasticity mechanisms would predict that the IC and DCN would show differences in these features, since each adjusts to changes of spontaneous activity of opposite polarity. To test the role of these two mechanisms, animals were exposed to an intense hyperactivity-inducing tone (10 kHz, 115 dB SPL, 4 h) then studied electrophysiologically at three different post-exposure recovery times (from 1 to 6 weeks after exposure). For each time frame, multiunit spontaneous activity was mapped as a function of location along the tonotopic gradient in the DCN and IC. Comparison of activity profiles from the two nuclei showed a similar progression toward increased activity over time and culminated in the development of a central peak of hyperactivity at a similar tonotopic location. These similarities suggest that the shape of the activity profile is determined primarily by passive relay from the cochlear nucleus. However, the absolute levels of activity were generally much lower in the IC than in the DCN, suggesting that the magnitude of hyperactivity is greatly attenuated by inhibition.

Introduction

Chronic tinnitus affects the quality of life of approximately 5–15% of the population of the United States (Roberts et al., 2010) and continues to be one of the major reasons for patient visits to the Otolaryngology clinic. Over the past decade, numerous animal models of the acute and chronic forms of tinnitus have been developed. These models are based on the use of manipulations which induce tinnitus in humans, such as exposure to intense noise and ototoxic drugs such as salicylate, quinine, cisplatin and carboplatin (Bauer, 2004; Eggermont and Roberts, 2004; Rachel et al., 2002; Kaltenbach et al., 2002). These inducers of tinnitus lead to increases in spontaneous activity (hyperactivity) in brainstem auditory nuclei as well as in primary and secondary auditory cortical areas. Noise-induced hyperactivity has been well characterized at the level of dorsal cochlear nucleus (DCN) (Zhang and Kaltenbach, 1998; Kaltenbach and Afman, 2000; Zacharek et al., 2002; Brozoski et al., 2002; Kaltenbach et al., 2002, 2004), ventral cochlear nucleus (VCN) (Vogler et al., 2011), inferior colliculus (IC) (Bauer et al., 2008; Mulders and Robertson, 2009; Dong et al., 2010a, 2010b) and auditory cortex (Norena et al., 2003; Seki and Eggermont, 2003). Hyperactivity has been detected in animals that show behavioral patterns consistent with tinnitus using various psychophysical tests (Brozoski et al., 2002; Kaltenbach et al., 2004; Wang et al., 2011; Longenecker and Galazyuk, 2011; Middleton et al., 2011; Dehmel et al., 2012).

The mechanisms underlying the induction of hyperactivity in the auditory brainstem have been investigated in numerous studies. The prevailing view is that hyperactivity is due to a shift in the balance of excitation and inhibition (Roberts et al., 2010). Several auditory brainstem nuclei show reductions of inhibition and/or increases in excitation, following noise exposure (see reviews of Kaltenbach, 2007, 2011). The most notable examples include the DCN, VCN and IC. Following noise exposure, these nuclei show evidence of decreased glycinergic and gabaergic neurotransmission (Abbott et al., 1999; Wang et al., 2009; Milbrandt et al., 2000; Dong et al., 2010a, 2010b) and upregulated glutamatergic and cholinergic transmission (Jin and Godfrey, 2006; Jin et al., 2005, 2006; Zeng et al., 2009; Muly et al., 2004; Dehmel et al., 2012).

A shift in the balance of excitation and inhibition is thought to be the result of synaptic plasticity triggered by injury to cochlear receptors. The resulting loss of normal input from the auditory nerve may trigger compensatory and non-compensatory adjustments in local circuitry. The compensatory adjustments can involve both anatomical and physiological changes. When there is loss of anatomical input, the lost synapses are replaced through the process of axonal sprouting and re-growth of new synapses (Benson et al., 1997; Bilak et al., 1997; Kim et al., 2004; Illing et al., 2005). When the loss of input is physiological, neurons can compensate homeostatically by scaling up the strength of weakened synapses (Desai et al., 2002; Turrigiano et al., 1998; Kim and Tsien, 2008; Stellwagen and Malenka, 2006; Knogler et al., 2010; Goel and Lee, 2007; O'Brien et al., 1998). Homeostatic changes have been implicated as a basis of tinnitus-related hyperactivity in the cochlear nucleus (Schaette and Kempter, 2006, 2008).

In addition to these compensatory mechanisms, deafferentation-induced hyperactivity may also be induced by non-compensatory mechanisms. For example, the spontaneous activity of neurons in the IC may be driven directly by passive relay of hyperactivity from the cochlear nucleus (CN). One of the afferent inputs to IC neurons comes directly from fusiform cells in the DCN. These cells have been shown to become hyperactive following intense noise exposure (Brozoski et al., 2002; Finlayson and Kaltenbach, 2009; Shore et al., 2008; Middleton et al., 2011). In addition, neurons in the VCN also become hyperactive (Vogler et al., 2011). Therefore, it is expected that a significant component of the hyperactivity induced in the IC after noise exposure might be inherited from these cochlear nucleus inputs.

These two responses to alterations of external inputs, (compensatory adjustments vs. passive relay) represent opposing forces and make fundamentally different predictions regarding the relationships between DCN and IC hyperactivity. If changes of activity in the DCN and IC are largely determined by compensatory adjustments in response to altered input, then the two nuclei should show very different levels and/or patterns of hyperactivity following cochlear injury. Because noise exposure causes decreases of spontaneous and stimulus-driven activity of auditory nerve fibers (Liberman and Kiang, 1978; Liberman and Dodds, 1984), a compensatory mechanism would trigger increases of activity in the cochlear nucleus, and the increases in spontaneous activity would exceed normal levels to compensate for lost stimulus-driven activity (Schaette and Kempter, 2006). However, at the IC level, the loss of normal stimulus-driven activity would be offset by increases of spontaneously activity at the CN level. The increased input to the IC from the CN would therefore be expected to trigger a weaker increase of activity or even a decrease relative to that in the CN. In contrast, if hyperactivity in the IC is determined largely by passive relay from the CN, then the IC and CN should show similar levels and tonotopic patterns of hyperactivity following noise exposure.

To gain insight into this issue, we compared the tonotopic patterns of hyperactivity in the DCN and IC following intense sound exposure. We mapped spontaneous activity in both structures as a function of location along the tonotopic axis. Since, the tonotopic profiles of hyperactivity have previously been shown to change over time, comparison of DCN activity profiles was also compared with those from the contralateral IC at three different post-exposure recovery periods. The results suggest two opposing response in the IC, one of which increases activity in conformance with a passive relay of afferent activity from the CN, the other of which leads to a global decrease in activity relative to that of its CN inputs.