Salicylate-induced auditory perceptual disorders and plastic changes in nonclassical auditory centers in rats

Abstract

Previous studies have shown that sodium salicylate (SS) activates not only central auditory structures, but also nonauditory regions associated with emotion and memory. To identify electrophysiological changes in the nonauditory regions, we recorded sound-evoked local field potentials and multiunit discharges from the striatum, amygdala, hippocampus, and cingulate cortex after SS-treatment. The SS-treatment produced behavioral evidence of tinnitus and hyperacusis. Physiologically, the treatment significantly enhanced sound-evoked neural activity in the striatum, amygdala, and hippocampus, but not in the cingulate. The enhanced sound evoked response could be linked to the hyperacusis-like behavior. Further analysis showed that the enhancement of sound-evoked activity occurred predominantly at the midfrequencies, likely reflecting shifts of neurons towards the midfrequency range after SS-treatment as observed in our previous studies in the auditory cortex and amygdala. The increased number of midfrequency neurons would lead to a relative higher number of total spontaneous discharges in the midfrequency region, even though the mean discharge rate of each neuron may not increase. The tonotopical overactivity in the midfrequency region in quiet may potentially lead to tonal sensation of midfrequency (the tinnitus). The neural changes in the amygdala and hippocampus may also contribute to the negative effect that patients associate with their tinnitus.

Figure 1

The recording electrodes in the brain. The drawings of the brain coronal section are from the rat brain atlas [72] and the inserts are photomicrographs of the brain showing DiI labeling of the recording electrodes (pointed by the arrows) in the Str and LA (a), the HC (b), and the Cg (c).

Figure 2

The effects of sodium salicylate (SS) injection on auditory perception. (a) Mean hearing thresholds for broadband noise bursts (n = 5). Baseline (black open bar), saline (blue shadowed bar), and salicylate (red filled bar) conditions are shown with standard error (SE) bars. Thresholds significantly increased by about 17 dB following salicylate administration; (b) salicylate-induced tinnitus. Rats (n = 3) were trained to respond to 3 stimuli. Quiet (red filled squares) and amplitude modulated (AM) (blue open circles) stimuli were paired with the right feeder while a narrowband noise (NBN) (black open triangles) was paired with the left feeder. Injection of saline (sal) showed no significant difference in responding during Quiet trials compared with baseline (no injection). However, an injection of 200 mg/kg SS showed a significant difference in response only during Quiet trials. This switch in response suggests that the rats perceived a steady state sound in the absence of an acoustic source. (c) Mean percentage (±SE) of startle amplitude relative to saline control startle amplitudes at 115 dB (marked with the star); note significantly increased startle amplitudes after salicylate injection, (d) Mean reaction time measures for broadband noise bursts (n = 7). Baseline (black circles), saline (blue triangles), and salicylate (red squares) are shown with standard error (SE) bars. Reaction times for 70, 80, and 90 dB SPL noise bursts decreased significantly with salicylate, suggesting an increased sensitivity to loud sounds.

Figure 3

The effects of SS-injection on sound-evoked local field potential (LFP) elicited from electrodes in the Str. (a) An example of LFP at 100 dB SPL recoded before treatment (black filled squares), after saline injection (blue open triangles), and after SS injection (red filled triangles), showing an enhanced response following SS-injection. (b) Mean RMSs of LFP (n = 32) in a time window of 100 ms as a function of stimulation level, showing progressive increase of LFP amplitude at high stimulation levels but a reduction at low stimulation levels. Acoustic stimulation: 50 ms noise burst; treatments: saline (5 mL/kg, i.p.) and SS (250 mg/kg, i.p.); the vertical bars are standard errors (SEs) and the ∗∗∗ means P < 0.001; the arrows indicate increase and decrease of LFP amplitude.

Figure 4

The effects of SS-injection on the tone-evoked LFP in the Str. (a) Mean RMSs of LFP (n = 64) to 12.1 kHz in a time window of 100 ms as a function of stimulation level, showing increase of LFP amplitude at high stimulation levels but a reduction at low stimulation levels. (b) Mean RMSs of LFP (n = 64) at 100 dB SPL as a function of stimulation frequency, showing significant enhancement at midfrequencies (3.5–18.3 kHz). Acoustic stimulation: 50 ms tone bursts at different frequencies; treatment: SS (250 mg/kg, i.p.); the vertical bars are SEs and the ∗∗∗ means P < 0.001; the arrows indicate increase and decrease of LFP amplitude.

Figure 5

The SS effects on unit activity of neurons in the Str. (a) Mean spontaneous discharge rates (n = 32) as a function of time showing significant decrease after SS injection (P < 0.001). (b) An example of peristimulus time histograms (PSTH) obtained before SS (blue) and after SS (red), showing SS-induced increase of discharge rate. (c) Averaged discharge rates of neurons in the Str (n = 32) in a time window of 100 ms, showing a similar effect of SS injection as the LFP recorded in the nucleus. The discharge rates of each neuron were normalized to that at 0 dB SPL. Acoustic stimulation: 50 ms noise burst; treatment: SS (250 mg/kg, i.p.); the vertical bars are SEs; the arrows indicate increase and decrease of unit activity; ***P < 0.001; **P < 0.01.

Figure 6

Averaged PSTHs of 26 units recorded in the Str showing significant effect (enhancement) in the midfrequency region (5.3, 8.0, and 12.1 kHz). The PSTHs were obtained before (blue) and after SS injection (red). Stimulation: 50 ms tone bursts at 100 dB SPL and at different frequencies; treatment: SS (250 mg/kg, i.p.). The vertical bars are SEs. **P < 0.01; *P < 0.05.

Figure 7

The effects of SS injection on auditory responses of the LA. (a) An example of LFP to noise burst at 100 dB SPL (black) showing an increase after SS injection (red). (b) Mean RMSs of LFP (n = 15) in a time window of 100 ms as a function of stimulation level, showing enhancement at high stimulation levels ≥60 dB SPL but reduction at low-stimulation levels <60 dB SPL. (c) Example of PSTHs in response to tones at 1.0 kHz (left) and 8.0 kHz (right) before and after SS injection showing greater increase after SS injection at the high-frequency; (d) SS-induced changes (%) of mean discharge rate in a time window of 100 ms showing SS-induced increase in the frequency range of 8–27.7 kHz. Stimulation: 50 ms noise or tone bursts; treatment: SS (250 mg/kg, i.p.). The vertical bars are SEs. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 8

The effect of SS injection on noise-burst-evoked LFP elicited from electrodes in the hippocampus. (a) Averaged LFP (n = 4 recordings in one rat) at 100 dB SPL recoded before treatment (black), after saline injection (blue), and after SS injection (red), showing a slight increase after SS injection. (b) Mean RMSs of LFP (n = 29) in a time window of 100 ms as a function of stimulation level, showing enhancement at high stimulation levels. Stimulation: 50 ms noise burst; treatment: SS (250 mg/kg, i.p.). The vertical bars are SEs and the ∗∗∗ means P < 0.001.

Figure 9

The effect of SS on noise-burst-evoked LFP elicited from electrodes in the cingulate cortex. (a) An example of LFP at 100 dB SPL recoded before treatment (black), after saline (blue), and after SS (red), showing no change during treatment. (b) Mean RMSs of LFP (n = 16) in a time window of 100 ms as a function of stimulation level, showing no significant change of the mean LFP. Stimulation: 50 ms noise burst; treatment: SS (250 mg/kg, i.p.). The vertical bars are SEs.