| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, Monash University, Monash, Australia
Submitted 18 August 2004; accepted in final form 22 November 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
). These pathways originate both ipsilateral [uncrossed OC pathways (UOCs)] and contralateral [crossed OC pathways (COCs)] to a cochlea, and both consist of a medial OC system (MOCS) that terminates on the cochlea's outer hair cells (OHCs) and a lateral OC system (LOCS) that terminates primarily on afferent dendrites. They exert a number of cochlear effects (Weiderhold 1986) including, of pertinence here, the modulation of loud sound-induced damage to cochlear hearing sensitivity (Cody and Johnstone 1982a; Fiorino et al. 1989; Handrock and Zeisberg 1982; Kujawa and Liberman 1997; LePrell et al. 2003; Liberman and Gao 1995; Luebke and Foster 2002; Maison and Liberman 2000; Maison et al. 2002; Rajan 1992, 1995a, b, 2001a, b, 2003; Rajan and Johnstone 1988; Robertson and Anderson 1994; Yamasoba and Dolan 1997; Zheng et al. 1997a, b).
Studies using different types of traumatic stimuli [generally relatively moderate, short-duration traumata that induce temporary threshold shifts (TTSs) in hearing] report that the effects of OC pathways on loud sound-induced TTSs depend on the type of traumatic sound. With loud pure tones, only the COC pathway [consisting almost exclusively of crossed MOCS (CMOCS)] modulates TTSs, and the outcome is solely a reduction in TTSs. With loud narrow band sounds, the COC pathway and the UOC pathways (consisting of the uncrossed MOCS, UMOCS, and the LOCS) modulate TTSs (Rajan 2001b). There is a greater modulation of TTSs than seen when using pure tones as the traumata. The net OC effect is to reduce TTSs, but this arises from a complex interplay between COC and UOC pathways: each exerts effects that reduce TTSs over one frequency range and exacerbate TTSs over another range. The fine details vary with different noise bands but, generally, the COC pathway reduces TTSs at frequencies within the noise band ("within-band" frequencies) but exacerbates TTSs at frequencies higher than the noise band ("high-side" frequencies). The UOC pathway has a small exacerbative effect or no effect at within-band frequencies, but reduces TTSs at high-side frequencies—i.e., the UOC pathway acts to reduce TTSs at the frequencies at which the COC pathway exacerbates TTSs (Rajan 2001b).
These results show that, depending on the type of acoustic trauma, there can be a complex interplay between different OC components and their TTS-reducing and TTS-exacerbating effects. Any hypothesis to account for OC actions at the cochlea must therefore also take into account the stimulus type eliciting OC effects. This view is reinforced by the demonstration that the background to a traumatic sound can modulate TTSs caused by that sound and the OC effects on TTSs in that condition (Rajan 2000). Thus when a loud tone is presented together with a background white noise (WN) that is in itself atraumatic, TTSs to the loud tone are exacerbated, almost exclusively at frequencies much higher than the tone. The OC modulation of pure tone-induced TTSs is also altered under these conditions (Rajan 2000). When a loud tone is presented by itself (in a background of silence), only the COC pathway modulates the tone-induced TTSs, and it reduces those TTSs. However, when the same tone is presented with background atraumatic WN, the COC and UOC pathways reduce TTSs (Rajan 2000). The COC pathway reduces TTSs over the same frequency range as it did in the absence of the background WN. Over this range of frequencies most affected by the loud tone, UOC pathways reduce TTSs only by a small amount. However, they strongly reduce TTSs at the high frequencies at which the WN background itself now exacerbates TTSs.
The observation that background atraumatic WN can exacerbate TTSs caused by a loud tone and that it can elicit expression of a broader range of, and more powerful, OC effects on TTSs, carries the physiological implication that activation of OC pathways and/or their effects at the cochlea depend on factors that modulate cochlear state. [This is also supported by studies indicating that OC pathways can have a net exacerbative effect on TTSs in the normal-hearing ears of animals with unilateral hearing losses. These ears have a lower than normal intrinsic susceptibility to loud sound, indicating a change in cochlear state in the normal-hearing ear consequent to the hearing loss in the other ear (Rajan 2001a, 2003).] This observation also has important functional implications since everyday workplaces (e.g., factories, ship-yards, lecture theaters) or recreational settings (e.g., rock music concerts, pubs, lecture theaters) where traumatic sound is likely to be an important issue are also likely to be noisy environments.
The various OC effects detailed above with different types of traumata suggested that the most potent activators of OC effects would be relatively broadband stimuli in a background noise environment and led to the specific working hypotheses that in such stimulus conditions 1) UOC and COC pathways act in opposition to maintain the basilar membrane in an optimal position for normal transduction, 2) the effects of UOC pathways are primarily targeted toward high frequencies outside the frequency band of a narrow band stimulus, and 3) UOC pathways damp cochlear vibration at the high-side frequencies (in keeping with our hypothesis that the exacerbation of TTSs by WN is caused by increasing cochlear vibration at high-side frequencies). This study was an attempt to test some of these predictions in developing a hypothesis as to how the OC pathways reduce TTSs. Hence here we manipulated the background in which a traumatic narrow band noise was presented and examined the consequences for TTSs caused by the traumatic sound and for OC effects on TTSs in such conditions. It was expected that 1) background WN would exacerbate narrow-band-sound-induced TTSs at high-side frequencies, confirming it's role in targeting mainly high-frequency areas of the cochlea, and 2) UOC pathways would be very powerfully active to reduce TTSs at these high-side frequencies, confirming their role in specifically acting to protect these cochlear regions. However, these expectations were confounded by the unexpected outcome that TTSs to the narrow band trauma were reduced in the presence of the atraumatic background WN and that, concomitantly, there was also a reduction in the TTS-modulating effects of the OC pathways.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Procedures were generally identical to those used in recent studies (e.g., Rajan 2001a–c, 2003). They conformed to the guidelines of the National Health and Medical Research Council of Australia and were approved by the Monash University Standing Committee on Ethics in Animal Experimentation. Adult cats (3–6 kg) were anesthetized (60 mg/kg, ip) and maintained with continuous intravenous pentobarbital sodium (2–3 mg/kg/h). [With this anesthetic, there are no effects of the middle ear muscles to attenuate sound input to the cochlea, even when using low frequency, high-intensity sounds, see Rajan (1995a).] Depth of anesthesia was monitored through continuous recording of rectal temperature, the ECG and EMG activity from forearm muscles, and regular hourly checks of response to strong noxious pinching of the forepaw, the presence of pupillary dilatation, and absence of corneal reflexes. Output from the ECG/EMG electrodes was displayed on an oscilloscope and fed into a speaker for continuous monitoring of the cat's condition and depth of anesthesia. Body temperature was maintained at 37.5 ± 0.5°C by a warming blanket, regulated by rectal probe feedback. Cats were tracheostomized and generally allowed to self-ventilate on room air, except in a few cats, in which respiration was shallow after initial anesthetization; artificial ventilation with room air was applied at 20–30 breaths/min depending on animal weight, with tidal volume set from normogram data for animal weight.
Stainless steel electrodes were implanted against the round window membrane of both cochleas (Rajan et al. 1991) to measure cochlear hearing sensitivity, which was assessed by measuring thresholds for the compound action potential (CAP) of the auditory nerve at frequencies from 1 to 40 kHz (tone bursts, 1 ms rise/fall times; 10-ms duration; 8–10 Hz). Only animals with bilaterally normal hearing sensitivity from 1 to 40 kHz were used (Rajan 1995a; Rajan et al. 1991).
Acoustic stimuli and trauma
Tone and noise stimuli to each ear were digitally generated independently (Tucker-Davis Technologies) by Fast Fourier Transform (FFT) and "brick-wall filtered." Noise output was flat at frequencies within the narrow band sound used as the acoustic trauma and frequencies within the WN used as the background atraumatic sound (see next paragraph but one). Output at frequencies outside the respective frequency bands, due to intermodulation distortion in the sound delivery system, was less than within-band output by >50 dB.
Tonal stimuli (to measure CAP thresholds) were gated under computer control and fed through separate attenuators into one of four channels of an electronic mixer. Cross-talk between the channels of the mixer box was better than –100 dB at 
10 kHz, –100 dB from 10–20 kHz, and declined thereafter to –95 dB at 40 kHz. Two output channels from the mixer box separately fed sound to one of two Sennheiser HD 535 speakers, each in specially designed housing leading out to a sound delivery tube placed in one external auditory meatus (Rajan 2000). Manual switches were used to control the delivery of stimuli to each of the two ears.
The traumatic stimulus was a narrow band sound (8–13 kHz) presented binaurally at 100 dB SPL for 15 min. This stimulus was one of two traumatic narrow band sounds used in a previous study (Rajan 2001b) examining the effects of narrow band traumata; as noted there, this particular band was chosen as the frequencies are from within the most sensitive part of the cat's CAP audiogram (Rajan et al. 1991), and frequencies from this region cause hearing damage more easily than do other frequencies (Rajan 1995b), as well as most readily activate cochlear effects of both COC and UOC efferent pathways (Rajan 2001b). In the test groups of this study, this traumatic sound was always presented simultaneous with a background of continuous white noise (WN: 0.5–40 kHz) at 60 or 80 dB SPL, also presented binaurally. The background WN was switched on 5 s before the traumatic sound; both sounds were switched off simultaneously by computer control after 15 min. Control animals, in which the same traumatic sound was presented with no other sound (i.e., in a background of silence) in similar OC status conditions (see next section), came from the previous study (Rajan 2001b).
CAP thresholds were remeasured 5 min after the traumatic sound, at frequencies from 6 to 32 kHz, in constant (but not linear) order. It took about 3 min to measure thresholds bilaterally at frequencies from 9 to 28 kHz—the frequencies most affected by the traumatic loud sound. As shown previously with other relatively moderate short-duration traumata, at 5 min after loud sound, the rate of recovery of threshold sensitivity is very slow (at most 5 dB/30 min; cf. Rajan 1988a; Fig. 2); hence, there were only minor changes in thresholds over the test frequency range in the time taken to measure thresholds. Frequency-specific hearing desensitizations (TTSs) were the difference between pre- and post-trauma thresholds. Two-way repeated measures ANOVAs compared TTSs between groups, with frequency constituting the within-subjects repeated-measures main factor, and either one or both of background WN level (no WN, WN at 60 dB SPL, or WN at 80 dB SPL) and status of OC pathways (see next section) being the between-subjects main factor(s). If there were significant between-subjects effects or interactions between main factors, post hoc analyses were conducted to determine which groups differed and which frequencies differed between groups (Tukey's HSD tests or pairwise tests based on estimated marginal means, with Bonferroni corrections for multiple comparisons). Note that TTSs always varied with frequency in the manner shown in the figures (e.g., Figs. 1, A and C, and 2, A–D), and the ANOVA F value for the frequency term was always significant at P < 0.0001 and is not reported further. For the post hoc tests, because of the number of groups and frequencies, P values for significant differences are not detailed; significance was always taken as P < 0.05. Statistical analyses were carried out using SPSS version 11.
|
|
Animals were exposed to binaural acoustic trauma with OC pathways intact bilaterally or after surgical lesions to cut various components of OC pathways to one or both cochleas. Lesions were made at the floor of the fourth ventricle, after removing the overlying cerebellum (Rajan 1995a). At this location, it is possible to lesion all OC fibers to one or both cochleas, crossed OC pathways to both cochleas, or to totally de-efferent one cochlea and only crossed pathways to the other ear (Warren and Liberman 1989). To totally de-efferent a cochlea, a lesion was made 1.5–2 mm lateral of the midline and on the brain stem side ipsilateral to that cochlea (Rajan 1995a, b). To cut only crossed pathways (bilaterally), a midline lesion was made (Rajan 1995a, b). Lesions were always 6–8 mm long, extending about the facial colliculi, identifiable on the floor of the fourth ventricle. Postmortem histology was used to confirm location of cuts (Rajan 1995a; Warren and Liberman 1989). Ears were grouped according to efferent status: all pathways intact (OC+ ears), all pathways cut (OC– ears), or only crossed pathways cut (COC–).
In all animals with brain stem lesions, prelesion checks were made of the CAP audiogram, heart rate, ECG waveform, and body temperature. The ECG was monitored through a speaker throughout the lesioning procedure, and all four prelesion parameters were rechecked immediately postlesion. The lesion never caused any large or systematic changes in these parameters of animal and hearing status. Only then was the animal presented with the binaural noise band trauma.
At the end of experimentation, while animals were still under deep anesthesia, they were killed with an overdose of pentobarbital sodium, and ECG/EMG measures were monitored until death ensued. As required, brains were fixed, and histology was performed subsequently, as detailed previously (Rajan 1988a).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In cochleas that were totally de-efferented (OC– ears) to remove any centrifugal influences, it was shown (Rajan 2000) that WN at 60 and 80 dB SPL were themselves atraumatic, but exacerbated TTSs caused by a loud pure tone (13 kHz at 100 dB SPL for 15 min). Unexpectedly, in similar OC– ears, the same atraumatic WN backgrounds significantly reduced TTSs to a loud narrow band sound compared with the OC– group given the same trauma without background WN (Fig. 1A; WN level: F = 26, df = 2, P < 0.0001), with effects varying with frequency (WN level x frequency: F = 5.7, df = 36, P < 0.0001). There were no significant differences (Tukey's HSD; P = 0.72) between TTSs in the two groups with background WN (at 60 or 80 dB SPL), but TTSs in each of these with WN background OC– groups differed from those in the no-WN background OC– group (all P < 0.0001). Compared with the no-WN background group, background WN at 60 dB SPL reduced TTSs at frequencies from 7 to 9 kHz and 15 to 30 kHz (Fig. 1B), and background WN at 80 dB SPL reduced TTS at 9 kHz and from 13–24 kHz. Thus the predominant effect of both background WN levels was to reduce TTSs at frequencies higher than the traumatic sound (high-side frequencies as defined by Rajan 2001b). The effect was established with background WN at 60 dB SPL, and the higher background WN level of 80 dB SPL did not further reduce TTSs over most (9–26 kHz) of the trauma-affected range.
Figure 1B contrasts the WN-induced reductions in TTSs to the loud narrow band sound against the WN-induced exacerbations in TTSs to a traumatic tone (Rajan 2000). Two common features are that the major changes induced by WN were always at frequencies higher than the frequency content of the traumatic stimulus and that there are two peaks of effects separated by a region of null effects. However, in addition to the direction of effects (exacerbation vs. reduction), there are other distinct differences: 1) the exacerbative effects on TTSs are shifted overall to higher frequencies than the TTS reductions (e.g., peak TTS exacerbation is at a higher frequency than peak TTS reduction, exacerbative effects spread to higher frequencies than do TTS reductions, and the low-frequency exacerbation peak occurs at the null point of the TTS reductions), and 2) exacerbative effects are generally graded to WN level, whereas TTS reductions did not differ substantially between the two WN levels. Some of the differences may be due to differences in the profile of TTSs (Fig. 1C) caused by the two traumata: with the loud narrow band sound, TTSs extend to lower frequencies than they do with the loud tone. However, it is also evident that the effects of background WN are most different at the high frequencies where the TTSs to the two traumata are relatively similar. Thus differences in the TTS profiles alone do not account for differences in the pattern of exacerbative effects versus TTS reductions.
Intact OC pathways can further reduce the desensitization caused by loud narrow band sound in a background of atraumatic WN
In ears with intact OC pathways (OC+ ears; Fig. 2A), the loud narrow band sound in either with-WN background condition resulted in lower TTSs than in the OC+ ears in the without-WN background group given the same traumatic sound in a background of silence (WN level: F = 21.1, df = 2, P < 0.0001). Background WN at 60 dB SPL resulted in lower TTSs at high-side frequencies 
16 kHz, as well as at 12 and 13 kHz (within-band frequencies—frequencies within the traumatic band). With WN at 80 dB SPL, most trauma-affected frequencies (7, 8, and from 11 to 26 kHz) showed lower TTSs. Effects between the two WN levels differed (Tukey's HSD; P = 0.019) at high-side frequencies from 14 to 20 kHz, where TTS reductions were greater with WN at 80 dB SPL than at 60 dB SPL.
Given that the WN itself could reduce TTSs to the loud narrow band sound (OC– ears), the lower TTSs in with-background WN OC+ ears compared with without background WN OC+ ears may reflect only the effect of the background WN. To determine this, OC+ ears and OC– ears from the two groups with background WN were compared (Fig. 2, B and C) to determine whether, in the OC+ ears, there were any OC effects additional to the effects due to the WN. The status of OC pathways was a significant main factor (OC status: F = 18.7, df = 1, P < 0.0001), indicating that for both test WN levels, intact OC pathways produced effects over and above those due to the background WN itself (OC– ears). Differences between OC+ and OC– ears varied in a frequency-specific manner that depended on WN level (significant interactions between all main factors: frequency x WN level; frequency x OC status; WN level x OC status; frequency x WN level x OC status; all P_< 0.001, generally <0.0001). Hence separate analyses were made for each with-WN background condition to compare OC+ against OC– ears. With WN at 60 dB SPL (Fig. 2B), there was a significant reduction in TTSs at 10–14 kHz in the OC+ ears but also a significant increase in TTSs at 17–24 kHz (OC status: F = 1.96, df = 1, P = 0.19; but frequency x OC status: F = 27.3, df = 18, P < 0.0001). With WN at 80 dB SPL (Fig. 2C), TTSs from 10 to 15 kHz in OC+ ears were reduced compared with OC– ears (OC status: F = 17.5, df =1 , P = 0.002; frequency x OC status: F = 15.6, df = 18, P < 0.0001).
In summary, in the groups with background WN, intact OC pathways reduce TTSs only at within-band and just-adjacent high-side frequencies, but not at more distant high-side frequencies (16 kHz); in fact, with background WN at 60 dB SPL, intact OC pathways actually exacerbated TTSs at 17–24 kHz compared with the OC– group with background WN at 60 dB SPL. This must mean, when comparing the with-WN and without-WN background groups where both factors were exerting effects (i.e., OC+ ears; Fig. 2A), that 1) TTS reductions at high-side frequencies 16 kHz in the with-WN background groups were due to background WN rather than any OC effects, 2) at within-band frequencies from 11 to 13 kHz, TTS reductions in the with-WN background groups were due to OC effects beyond any effects of the background WN, and 3) at just-adjacent high-side frequencies of 14 and 15 kHz, both OC effects and background WN effects were responsible for TTS reductions. These frequency-related patterns of OC effects and background WN effects are examined in greater detail in the next section.
The effects of OC pathways in the with-WN background groups were generally similar to the effects of OC pathways in equivalent groups given the loud sound without background WN (Fig. 2D). In the control condition, OC pathways significantly influenced TTSs in a frequency-dependent manner (OC status: F = 14.6, df = 1, P = 0.002; OC status x frequency: F = 19.2, df = 18, P < 0.0001); post hoc tests showed that with intact OC+ pathways, there was a significant reduction in TTSs from 7 to 16 kHz, with no effects at higher frequencies. Thus as in the with-WN background groups, in the groups without WN background, OC pathways reduced TTSs only at within-band and just-adjacent high-side frequencies, but not at more distant high-side frequencies. However, in the with-WN background groups, the effects occur across a more restricted frequency range (10 to 
15 kHz) than in the control group (7–16 kHz).
Total effects of OC pathways and of atraumatic WN backgrounds
In previous studies, the difference in TTSs in OC– and OC+ groups was used to measure the total effects of OC pathways at the cochlea (Rajan 2001b, 2003). The same calculation was used here to determine the total effect of all OC pathways on TTSs in each with-WN background group and in the without-WN background group (latter data presented previously in Rajan 2001b). This total OC effect in the three conditions is shown in Fig. 3A.
|
), the main OC effect at the cochlea was to reduce TTSs, with a broad peak effect from 7 to 16 kHz (within-band and just-adjacent high-side frequencies), and with effects decreasing at lower and higher frequencies. There was also a small, but nonsignificant, exacerbation of TTSs at frequencies from 19 to 26 kHz.
In the with-WN background groups, there was a marked diminution of TTS reductions at frequencies <11 kHz, producing a more focal peak of TTS reduction at 12 kHz. At higher frequencies, there were changes with background WN at 60 dB SPL (Fig. 3C; a diminution of TTS reductions and even an increase in TTS exacerbations at frequencies >16 kHz), but not with background WN at 80 dB SPL (Fig. 3D). Note that at frequencies 16 kHz, OC effects in the two test groups ranged by <5 dB on either side of control OC effects, suggesting that these changes in OC effects in test ears were small variations within the normal (control) range. Figures 1A and 2A lend support to this view, since they show no systematic difference in TTSs at these high-side frequencies in the two test OC– groups (Fig. 1A), and in OC+ groups (Fig. 2A), TTSs decrease with WN level at high-side frequencies 20 kHz with no systematic difference at higher frequencies. Thus the "raw" TTSs indicate no systematic change in OC effects at the high frequencies in the with-WN background groups compared with the without-WN background group.
Thus in the with-WN background groups, two effects modulate TTSs to the loud narrow band sound: effects due to the WN background and effects due to OC pathways. These two sets of effects are shown in Fig. 4, A (WN at 60 dB SPL) and B (WN at 80 dB SPL). In general, across both background WN levels, OC effects are exercised predominantly at within-band frequencies (with peak effect at 12 kHz), where they act to reduce TTSs. In this same frequency range, the effects of background WN are weak or absent—in fact, the peak total OC effect is always at a frequency in a range with null effects of background WN. At high-side frequencies 15 kHz, the effects of background WN reduce TTSs; total OC effects are either mildly exacerbative over a broad range of high-side frequencies (Fig. 4A) or nearly absent (Fig. 4B). As noted above, OC effects in the with-WN background groups at high-side frequencies are likely small variations about the normal (control) total OC modulation of TTSs. If so, the dominant modulation of TTSs at these high-side frequencies is exercised by the background WN rather than by OC pathways.
|
16 kHz, the dominant modulator of TTSs appears to be WN background not the OC pathways. Crossed OC pathways seem solely responsible for all OC effects with loud narrow band sound in a background of atraumatic WN
A second issue addressed here was the contribution of different OC components to the total OC modulatory effect on TTSs. In two sets of animals tested with WN background, lesions were made at the brain stem midline to cut only the COC–, leaving the UOC pathways intact, before applying the test conditions. Figure 5, A and B, compares TTSs in the with-WN background COC intact groups to TTSs in the OC+ and OC intact groups in the same test conditions of trauma + background WN.
|
With background WN at 80 dB SPL (Fig. 5B), there were generally similar effects: TTSs in COC– and OC– ears were similar (Tukey's HSD with Bonferroni corrections; COC– vs. OC–: P = 0.39) except for lower TTSs in COC– ears at the very high frequencies of 22–30 kHz. Compared with OC+ ears, TTSs in the COC– ears were significantly higher at 11–15 kHz but significantly lower at 20–30 kHz. Thus as with the lower WN level, cutting only the COC pathway abolished the OC-induced reduction in TTSs at 11–15 kHz as well as lowered TTSs at the very high frequencies. Note that at high-side frequencies from 20 to 30 kHz, there were no differences in TTSs in the OC+ and the totally de-efferented (OC–) conditions, suggesting that OC pathways did not exacerbate TTSs at these frequencies. Nevertheless, at these frequencies, cutting the COC pathway resulted in similar effects in both WN backgrounds—a decrease in TTSs compared with when all OC components were intact.
Thus in both test cases where the acoustic trauma was combined with the WN background, lesioning only the crossed OC pathway abolished any reduction in TTSs obtained when all OC pathways were intact. Additionally, at high-side frequencies well displaced from the frequency content of the traumatic narrow band sound, lesioning the COC pathway also reduced TTSs compared with TTSs seen when all OC pathways were intact. In the case with WN at 60 dB SPL, this reduction resulted in loss of the TTS exacerbation in the OC+ condition compared with the OC– condition. In the case with WN at 80 dB SPL, there was no TTS exacerbation in the OC+ condition compared with the OC– condition, and the reduction in TTSs obtained with lesioning the COC pathway resulted in TTSs being lower in COC– cases compared with either the OC+ or OC– conditions.
These effects on TTSs in the with-WN background groups can be compared with effects in the without-WN background groups with equivalent OC manipulations (Fig. 5C; data from Rajan 2001b). As reported before, there were significant differences between the without-WN background groups varying in OC status (OC status: F = 8.713, df = 2, P = 0.002; OC status x frequency: F = 24.9, df = 36, P < 0.0001). Intact OC pathways resulted in significant reduction in TTSs at within-band and just adjacent frequencies from 7 to 16 kHz (OC+ vs. OC– conditions: Tukey's HSD with Bonferroni corrections; P < 0.05), but without significant changes at higher frequencies. When only the COC pathway was cut (COC–), leaving UOC pathways intact, TTS reductions seen in OC+ ears at within-band frequencies from 9 to 13 kHz were now completely abolished. However, in the COC– ears, there was a significant reduction in TTSs at high-side frequencies from 14 to 30 kHz compared with OC– ears; in fact, TTSs from 17 to 30 kHz in COC– ears were even significantly lower than in OC+ ears.
Overall, there are strong similarities between effects in the without-WN background and the with-WN background groups: in both cases, lesioning only the COC pathway abolished any TTS reductions at within-band frequencies otherwise seen when all OC pathways were intact. In all three backgrounds, it also reduced the TTSs at high-side frequencies, resulting in these TTSs in the COC– condition being lower than in the OC+ condition. The size and extent of the high-side reductions in TTS varied with the level of background WN, being largest in the without-WN background condition (15- to 20-dB reductions comparing COC– and OC+ conditions in the without-WN background condition vs. 
10-dB reductions in the with-WN background conditions) as well as most extensive (without-WN background, OC+ vs. COC–: significant differences from 17 to 30 kHz) and less so with the WN backgrounds (WN at 60 dB SPL, OC+ vs. COC–: significant differences from 17 to 24 kHz; WN at 80 dB SPL, OC+ vs. COC–: significant differences from 20 to 30 kHz).
Finally, Fig. 5D compares the three COC– groups. There were significant differences as a function of background WN level (F = 7.2, df = 2, P = 0.007) with a marked dependency of effects on frequency (WN level x frequency interaction: F = 6.3, df = 36, P < 0.0001). Post hoc analyses (Tukey's HSD) found no differences between the two with-WN background groups with background noise but each of these groups differed from the without-WN background group at frequencies from 7 to 12 kHz, almost the full range of within-band frequencies. The import of these results is discussed in the next section.
Separating the effects of COC and UOC pathways
In previous studies, the three OC status conditions (e.g., OC+, OC–, and COC–) were used to differentiate cochlear effects of COC and UOC pathways (Rajan 2001a–c, 2003). The same analyses were used here. For each test condition (without-WN background condition, background WN at 60 dB SPL, or background WN at 80 dB SPL), effects of the crossed OC pathway alone were calculated as the frequency-specific differences between the OC+ group (all OC pathways intact) and the COC– group (only COC pathway cut). In a similar manner, the effects of UOC pathways alone were calculated as the frequency-specific differences between the COC– group (COC pathway cut, UOC pathways intact) and the OC– group (all OC pathways cut). These calculated effects of COC and UOC pathways in the three groups here are shown in the bottom panels in Fig. 5 and are collected together in Fig. 6 for each OC component.
|
5 dB) were largest and most extensive (16–30 kHz) in the without-WN background group, decreased in size in the with-WN background group with background WN at 60 dB SPL, and both decreased further in size and occurred over a much smaller, very-high-frequency range in the with-WN background group with background WN at 80 dB SPL. The UOC pathways seem to exercise almost no effect, or only small (and almost negligible) effects in the two groups with WN background (Fig. 6B). This can be contrasted against UOC effects in the without-WN background condition, where UOC effects always appear to be the mirror image of COC effects and are quite strong at high-side frequencies from 15 to 26 kHz. There are also systematic, albeit small, exacerbative UOC effects at within-band frequencies.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Modulatory effects of background WN vary with the type of acoustic trauma
The TTSs caused by short-duration moderate-intensity traumata, like that used here, are generally due to changes in cochlear mechanics (Cooper and Rhode 1992; Fridberger et al. 2002a,b; Patuzzi et al. 1984; Ruggero et al. 1993, 1996). These changes seem to be a decrease in the output of the reverse transduction electromotile "active" process in OHCs, and this may (Patuzzi 1992; Patuzzi et al. 1989) or may not (Fridberger et al. 2002a, b; Zhang and Zwislocki 1995) be caused by temporary inactivation of mechanosensitive transduction channels. The result that background WN reduced TTSs suggests it modulated cochlear mechanics in such a way as to "protect" the gain of the OHC active process. [With respect to the middle ear muscles that could attenuate input to the cochlea, under the anesthetic conditions used here, these pathways are not operational (Rajan 1995), even when using low-frequency sound trauma presented for much longer than the trauma in this study.] The effects of the background WN cannot be accounted for by two-tone suppression (2TS), whereby a secondary stimulus can reduce basilar membrane (BM) vibration to a primary stimulus (Robles and Ruggero 2001) since 2TS decreases with increasing intensity of the primary stimulus, and Robles and Ruggero (2001) show that even with a 70 dB SPL secondary tone, suppressive effects on BM vibration are present only for primary tones <70 dB SPL. In this study, WN at 60 dB SPL reduced TTSs induced by narrow band trauma at 100 dB SPL.
An interesting, and possibly related phenomenon, is that reported by Cody and Johnstone (1982b): that the TTSs to a primary 16-kHz traumatic tone were reduced by the addition of a secondary loud tone at 5 kHz. However, this phenomenon cannot account for the fact that the same background WN as used here exacerbates TTSs to a loud pure tone (Rajan 2000; 13 kHz in that study). Both effects are exerted by atraumatic WN and occur in the absence of any descending influences (i.e., in OC– ears). Thus in both cases, the WN must alter the intrinsic susceptibility of the cochlea to the acoustic trauma, albeit in opposite ways with the two types of traumata. Previously, it was suggested by Rajan (2000) that the atraumatic WN background might exacerbate loud tone-induced TTS by biasing the cochlear partition more or for longer in the vibration direction in which TTSs occur during loud sound (Patuzzi and Rajan 1990). This argument can apply to the results here if the reverse effect is hypothesized to occur, i.e., the WN may reduce loud narrow band sound-induced TTSs by biasing the cochlear partition away from the direction of sound-induced vibration in which TTSs occur. This change in the direction of the effects of the background WN in the two types of traumata must be linked to the broader bandwidth of the narrow band sound trauma compared with the pure tone trauma, but currently there is no study of cochlear mechanics that could confirm or rebut this speculation.
Despite the different direction of effects of background WN in the two types of trauma, there were some common features: the major effects were on TTSs at frequencies higher than the frequency content of the traumatic stimulus, and there were two peaks of effects separated by a region of null effects (Fig. 1B). The overall profile of WN-induced changes in TTSs to narrow band sound trauma was shifted to slightly lower frequencies compared with the WN-induced changes in TTSs to pure tone trauma, and this can be linked to the frequency profile of TTSs caused by the two traumata. With narrow band sound trauma by itself (Fig. 1C), the profile of TTSs is broadened much more toward low frequencies, in keeping with the bandwidth increase to low frequencies compared with the pure tone trauma. Similarly, the modulatory effects of WN (Fig. 1B) on the narrow band sound trauma are shifted toward low frequencies compared with the modulatory effects of WN on the pure tone trauma.
OC pathways involved in modulating cochlear desensitization
In addition to modulating TTSs to the narrow band sound, the background WN also altered the way in which the extrinsic pathways modulated TTSs. The latter modulation can be attributed to the OC pathways, both for theoretical reasons and on the basis of the effects of the brain stem lesions. It could not be due to the middle ear muscles that can attenuate input to the cochlea, since it has been shown directly (Rajan 1995a) that the pathways to these muscles are not operational under the anesthetic conditions used here. Autonomic pathways to the cochlea will also not be involved in the effects reported here since they would not be affected by the brain stem lesions. Additionally, in this study, it was found that in the animals without any brain stem lesions, TTSs to the narrow band sound trauma by itself or in a background of WN are less than in animals given the equivalent test conditions after the brain stem lesions were placed. In other words, the brain stem lesion removes something that protects the cochlea, whereas studies of the autonomic pathways (e.g., Borg 1982; Hildesheimer et al. 1991) show that these pathways increase the threshold losses caused by loud sound.
The OC pathways consist of the MOCS and LOCS. The COC pathway in cats consists almost totally of MOCS fibers (CMOCS), whereas the uncrossed OC pathway consists of MOCS fibers (UMOCS) terminating on OHCs and LOCS fibers terminating predominantly, but not exclusively, on dendrites of afferent neurons (Liberman 1980; Warr et al. 1986). In previous studies using narrow band sound trauma (Rajan 2001b) or pure tone trauma (Rajan 2000, 2001a), it was argued that COC- and UOC-induced modulation of TTSs to short-duration, moderate-intensity traumata were due only to the MOCS, i.e., by the CMOCS and UMOCS, respectively. In support of this proposition, LePrell et al. (2003) have shown that the LOCS has no influence on pure tone-induced losses in threshold sensitivity. However, this study and previous studies with narrow band traumata (Rajan 2001b) or pure tone sound in background WN (Rajan 2000) show that more complex test conditions than pure tone trauma in a background of silence lead to a wider range of OC effects on TTSs. These may include LOCS effects. An anonymous reviewer to this manuscript also noted the following arguments that make it hard to exclude an involvement of the LOCS in UOC-mediated effects seen here: 1) MOCS fibers are part of a loop that includes the inner hair cells and the type 1 afferent fibers, and since LOCS fibers synapse on the latter elements, they could modulate drive to the MOCS neurons, and thereby MOCS effects, at the cochlea; 2) LOCS fibers also form (minor) synapses on MOCS fibers in the tunnel of Corti (Liberman 1980) and thus could influence MOCS fibers; and 3) there may be direct or indirect connections between MOCS and LOCS neurons in the brain stem or between elements of their reflex pathways [e.g., MOCS fibers send collaterals into the cochlear nucleus (Brown et al. 1988), and these may affect MOCS or LOCS pathways], and these complex CNS interconnections may be interrupted by the brain stem cuts made in this study.
For these reasons, COC effects will be treated as CMOCS effects, but the effects of UOC pathways will be referred to as UOC pathway effects because they may involve both UMOCS and LOCS neurons, the latter through the modulation of MOCS pathways or effects.
With respect to how OC pathways modulate TTSs, there is evidence that the MOCS component of OC pathways reduces the damaging effects of loud sound through effects exerted on OHCs via the 
9 nicotinic acetylcholine receptor subunit (Luebke and Foster 2002; Maison et al. 2002). The MOCS is known to reduce the output of the OHC active process (Dolan et al. 1997; Murugasu and Russell 1996; Patuzzi and Rajan 1990) but it is still unclear how, at the trauma intensity when cochlear responses are determined solely by passive cochlear mechanics, the MOCS protects the OHC active process, preventing it from being desensitized by acoustic trauma. It is likely that actions through the same receptor subunit may also be responsible for the previously described exacerbative effects of the MOCS (Rajan 2001a, b, 2003), but this still does not elucidate how the MOCS could modulate TTSs. There is no data available currently on how the LOCS modulates TTSs.
Modulatory effects of OC pathways on cochlear desensitization change in parallel with the effects of WN on cochlear desensitization
Compared with the no-background WN condition, when the narrow band sound trauma was presented with a WN background, OC effects on TTSs were modulated as follows: 1) at lower within-band frequencies (8–10 kHz), OC-induced reductions in TTSs were markedly diminished; 2) at higher within-band frequencies (11–13 kHz, the within-band frequencies suffering the largest TTSs and the largest OC effects in the control condition), there was a small enhancement in OC-induced TTS reductions; and 3) at high-side frequencies (14 kHz), there were variable changes in OC-induced TTS exacerbations, which were enhanced when background WN was 60 dB SPL but were reduced with background WN at 80 dB SPL.
A detailed comparison of the effects of COC and UOC pathways (Fig. 6) revealed that the above-detailed changes in OC modulation of TTSs in the presence of background WN were due to changes in the way in which both OC pathways modulated TTSs. There was a total absence of UOC pathway effects at all frequencies (Fig. 6B) and the following changes in CMOCS effects: a significant compression of effects at low within-band frequencies, a small decrease in the peak TTS reduction (at high within-band frequencies), and a larger reduction in the high-side TTS exacerbations. In essence there seemed to be an overall reduction in the TTS-modulating effects of both OC pathways (Fig. 6).
The reduction in the TTS-modulating effects of OC pathways may have been caused by the WN reducing the OC drive to the cochlea. This seems unlikely given that OC-induced reductions in TTSs to a pure tone trauma in the presence of the same atraumatic WN background are much greater than to the same trauma without any background sound (Rajan 2000); this is hardly consistent with the background WN reducing OC drive to the cochlea. Furthermore, studies of single efferent neurons have shown that their responses to tones are facilitated by noise (Liberman 1988). Also, a study (Kawase et al. 1993) of the "anti-masking" function of the MOCS indicates that the anti-masking effect likely increases with increasing background noise, within some range of noise levels.
Alternatively the reduction in the TTS-modulating effects of the two OC pathways may have been caused by the WN directly or indirectly interfering with expression of these OC effects at the cochlea. This hypothesis derives from the fact that the WN itself reduced TTSs to the narrow band sound trauma. Hence it is proposed that both the WN background and the OC pathways acted on the same site/process. The dominant effect was that exerted by the WN and it reduced the damage occurring at this site/process. Previous studies of OC-induced (CMOCS-induced in those cases) reductions in TTSs to pure tones have shown that 1) there is no OC protection when there are relatively low levels of TTSs and 2) that, above some "threshold" level of TTSs, OC pathways reduce TTSs, with the amount of OC-induced TTS reduction being related to the amount of TTS that would otherwise occur. This has been shown with OC protection elicited with either direct electrical stimulation of cochlear efferents (Rajan 1988b; Rajan and Johnstone 1988a) or binaural acoustic stimulation (Rajan 1992, 1995b; Rajan and Johnstone 1988b); none of these manipulations evoke protection from TTSs to loud sounds at levels that cause small TTSs (Rajan 1995a, b; Rajan and Johnstone 1988a, b), even if they strongly drive efferent neurons (Liberman 1988; Liberman and Brown 1986; Robertson and Gummer 1985). The same manipulations do result in COC pathway-mediated protection at higher exposure levels that produce larger TTSs. These effects suggest that OC-mediated protection is targeted to some particular component or process within OHCs, and this component/process is affected only when TTSs above some "threshold" level are produced. Above this "threshold" TTS level, increasing TTSs reflect increasing damage to this component or process that can be modulated by the OC pathways. (Note that this relationship has been shown specifically for the CMOCS and it is not known if such a relationship holds true for TTS modulation by UOC pathways.) The hypothesis is consistent with one previously advanced to account for the absence of CMOCS effects on pure tone-induced TTSs in the normal ears of animals with unilateral hearing losses (Rajan 2001a): that the absence of CMOCS effects there reflected the reduced susceptibility to loud sound of a particular cochlear element or process specifically targeted by the CMOCS.
This hypothesis can be applied here by proposing that, at the lower within-band frequencies and at the high-side frequencies (frequencies beyond the high-frequency edge of the noise band), the WN background had reduced TTSs to a lower level and thereby lowered the amount of TTS reduction produced by the CMOCS or had reduced TTSs to below the "threshold" for CMOCS modulation of TTSs. The minor change in TTS reductions at the higher (peak-affected) within-band frequencies was consonant with the negligible WN effects at these frequencies. Thus at within-band frequencies, expression of OC modulation of TTSs seems contingent on the effects of background WN on TTSs. Previous studies show that the modulatory effects of OC pathways on TTSs produced by pure tone or narrow band sound trauma can differ markedly even if involving the same acoustic trauma but in ears with different intrinsic susceptibility to that trauma (bilaterally normal animals vs. normal ears of animals with unilateral hearing losses; cf., Rajan 2001a, 2003). These cases show that OC modulation of TTSs is dependent on the cochlea's intrinsic susceptibility to trauma. Background WN could then modulate the expression of OC modulation of TTSs by modulating the intrinsic susceptibility of the cochlea to TTSs.
The loss or absence of the effects of UOC pathway on TTSs are explicable on considering the types of UOC pathway effects (or absence of effects) on TTSs seen in animals with normal hearing sensitivity. First, during pure tone traumata by themselves, there is no UOC pathway effect (Rajan 1995a, 2000). Second, during pure tone trauma in background atraumatic WN (Rajan 2000), the UOC pathway reduces TTSs almost exclusively at frequencies much higher than the pure tone trauma, at which the WN background itself exacerbates TTSs. Third, during narrow band sound traumata (one being the same as used here; Rajan 2001b), the UOC pathway almost exclusively reduces TTSs at the high-side frequencies at which the CMOCS exacerbates TTSs. Thus in respect of TTSs, it seems that UOC pathway effects in normal-hearing animals are only expressed when a broad cochlear range is affected (narrow band sound trauma or pure tone trauma in background WN). Then the dominant effect of the UOC pathway is at high-side frequencies well displaced from the frequency content of the trauma; at those frequencies, the action of the UOC pathway prevents the exacerbation of TTSs (note: not prevent TTSs). When there is no exacerbation of TTSs at these frequencies, there is no overt UOC pathway effect on TTSs. In this study, the presence of background WN to narrow band sound trauma results in a marked diminution in the TTS-exacerbating effects of the CMOCS at high-side frequencies; additionally, the background WN exerts a powerful TTS-reducing effect at the high-side frequencies. These two factors then seem to prevent any UOC pathway effect being evinced on TTSs at high-side frequencies.
At within-band frequencies, in the control condition (no WN background), the UOC pathway caused a small exacerbation in TTSs, whereas the CMOCS causes a very large reduction in TTSs (Fig. 5C) (Rajan 2001b). With background WN, there was a diminution in the CMOCS-induced within-band TTS reductions (this was WN level–independent); conjointly, there was total absence of UOC pathway effects at these frequencies. Again this may be related to the fact that, at these within-band frequencies, although there is a large reduction (at the lower within-band frequencies) of CMOCS-induced reductions in TTSs, there is also a strong reduction in TTSs due to effects of the WN background. Thus with a dominant WN effect that partly reduces the CMOCS within-band effect, there is a loss of the small UOC pathway–mediated within-band effect that normally opposes the CMOCS effect.
Finally, it is notable that despite the reduction in the range of frequencies with CMOCS-induced reductions in TTSs and total abolition/absence of any UOC pathway–induced reductions in TTSs (or any UOC pathway–induced effects on TTSs) in the WN background, the conjoint effect of the OC pathways and the background WN resulted in enhanced reductions in TTSs, especially at higher frequencies (12 kHz), compared with the group given the loud sound in a background of silence. At high-side frequencies (14 kHz), this enhancement increased with increasing background WN level. These effects suggest that in the test conditions here both the WN background and the MOCS pathways acted on the same site to reduce the desensitization of the active process. It is intriguing that this joint promotion of a protective end-effect on TTSs occurs to narrow band sound trauma, but that with pure tone trauma the background noise exacerbates (high-side) TTSs while the UOC pathway reduces these TTSs. The latter is also consistent with the hypothesis advanced here, namely that both factors (WN and the OC pathways involved in TTS modulation) act at the same site, although clearly in opposing ways in that instance. Thus we conclude that both factors act at the same cochlear site or process to modulate cochlear state and that the modulatory effects of the OC on TTSs, at least at the high-side frequencies, are dependent on cochlear state.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: R. Rajan, Dept. of Physiology, Monash Univ., Monash, Victoria 3800, Australia (E-mail: ramesh.rajan{at}med.monash.edu.au)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Brown MC, Berglund AM, Kiang NY, and Ryugo DK. Central trajectories of type II spiral ganglion neurons. J Comp Neurol 278: 591–603, 1988.[CrossRef][ISI][Medline]
Cody AR and Johnstone BM. Temporary threshold shift modified by binaural acoustic stimulation. Hearing Res 6: 199–205, 1982a.[CrossRef][ISI][Medline]
Cody AR and Johnstone BM. Reduced temporary and permanent hearing losses with multiple tone exposures. Hearing Res 6: 291–301, 1982b.[CrossRef][ISI][Medline]
Cooper NP and Rhode WS. Basilar membrane mechanics in the hook region of cat and guinea-pig cochleae: sharp tuning and non-linearity in the absence of baseline position shifts. Hearing Res 63: 163–190, 1992.[CrossRef][ISI][Medline]
Dolan DF, Guo MH, and Nuttall AL. Frequency-dependent enhancement of basilar membrane velocity during olivocochlear bundle stimulation. J Acoust Soc Am 102: 3587–3596, 1997.[CrossRef][ISI][Medline]
Fiorino FG, Gratton MA, Subramaniam M, Bianchi L, and Henderson D. Physiological mechanism underlying the progressive resistance to noise induced hearing loss. Il Valsalva 54(1 Suppl): 36–41, 1989.
Fridberger A, Zheng J, and Nuttall A. Alterations of basilar membrane response phase and velocity after acoustic overstimulation. Hearing Res 167: 214–222, 2002a.[CrossRef][ISI][Medline]
Fridberger A, Zheng J, Parthasarati A, Ren T, and Nuttall A. Loud sound-induced changes in cochlear mechanics. J Neurophysiol 88: 2341–2348, 2002b.
Handrock M and Zeisberg J. The influence of the efferent system on adaptation, temporary and permanent threshold shift. Arch Otorhionlaryngol 234: 191–195, 1982.
Hildesheimer M, Sharon R, Muchnik C, Sahartov E, and Rubinstein M. The effect of bilateral sympathectomy on noise induced temporary thres
) and with-background WN animals (trauma + background WN: 
, 60 dB SPL; 
, 80 dB SPL). B–D: comparison of TTSs in efferent intact (OC+, 
) or cut (OC–, 
; data from Rajan 2000
, background WN at 80 dB SPL. Data were calculated as the sum of the effects of each modulatory system calculated separately in A and B. The sign (i.e., positive or negative) of the frequency-specific values calculated in A and B for each modulatory system was maintained in the summing. (The same result is obtained if the combined modulatory effect is calculated from the difference between TTSs in the OC+ group in a specific with-background WN condition and the OC– group in the without-background WN condition, since TTSs in the former group were subject to modulation by OC pathways and WN background and TTSs in the latter group were affected by neither system.) For comparison, the full line without symbols is from 
), in the groups with background WN (A: background WN at 60 dB SPL; B: background WN at 80 dB SPL) or without background WN (C). D: comparison of TTSs in the without-background WN condition and the with-background WN condition with only crossed OC pathways cut (COC–; intact UOC pathways). 