Fig 1.
Noise exposure results in elevated expression of proinflammatory cytokines in AI.
A. Monaural exposure to loud noises caused long-lasting ABR threshold increases of up to 50 dB in the exposed left ear and temporary threshold increases in the protected right ear that recovered to less than 10 dB in 10 days. B. TNF-α mRNA level increased rapidly within 12 hours of noise exposure, with a stronger ipsilateral than contralateral increase. The increase was also significant at 1 day and 10 days post noise exposure. C–D. IL-1β and IL-18 mRNA levels both increased in bilateral AI 10 days post noise exposure. E. NLRP3 mRNA level showed biphasic increases at 12 hours and 10 days post noise exposure. F. TNF-α protein levels increased in AI of the right hemisphere 1 day after noise exposure, and the increase persisted to at least 10 days post exposure. Data associated with this figure can be found in S1 Data. Error bars represent SEM. * depicts P < 0.01 compared to control; # indicates P < 0.01 comparing left and right hemispheres. n = 4 mice for each time point in B–F. ABR, auditory brainstem response; AI, primary auditory cortex; IL, interleukin; NLRP3, nod-like receptor protein 3; TNF-α, tumor necrosis factor alpha.
Fig 2.
Microglial deramification in AI after noise exposure.
A. Representative images of IBA1 antibody–stained microglia in AI of control and noise-exposed mice. Microglia in control mice showed ramified morphology, indicative of their resting state. A proportion of the microglia became activated and transitioned into nonramified and amoeboid shapes (arrows and arrowheads) 5 days after noise exposure. Lower panels show morphologies of ramified (control and Noise_3 days) and deramified (Noise_5 days) microglia. B. The soma-to-whole cell size ratio of microglia was used to measure microglial morphological change as an index of microglial activation. There was a significant increase in the microglial activation 5 days after noise exposure, and the activation was stronger for the right than left side. n ≥ 8 for each group. C. No significant increase in microglial activation index was observed in the right visual cortex 5 days after noise exposure. n = 17 for control and 22 for 5-day groups. Data associated with this figure can be found in S1 Data. Error bars represent SEM. * depicts P < 0.01 comparing left and right sides. AI, primary auditory cortex; IBA1, ionized calcium binding adaptor molecule 1.
Fig 3.
TNF-α KO mice do not show microglial deramification or tinnitus after noise exposure.
A. Example images of microglia stained with IBA1 antibody in TNF-α KO mice with (Noise_5d, 5 days after noise exposure) or without noise exposure (Control). Inserts show enlarged images of microglial morphology. B. Noise-induced microglial activation 5 days after noise exposure, as quantified with the soma-to-whole cell size ratio, was absent in TNF-α KO mice. n ≥ 12 for each group. C–D. Wild-type but not TNF-α KO mice showed noise-induced tinnitus. Tinnitus and hearing were assessed with the gap detection and PPI, respectively. While gap detection was impaired in wild-type mice 10 days after noise exposure, their PPI was improved at a certain frequency. Neither gap detection nor PPI was altered in TNF-α KO mice by noise exposure. E. Wild-type and TNF-α KO mice showed similar levels of noise-induced increases in ABR thresholds. Data associated with this figure can be found in S1 Data. Error bars represent SEM. * and ** depict P < 0.05 and P < 0.01, respectively. ABR, auditory brainstem response; IBA1, ionized calcium binding adaptor molecule 1; KO, knockout; PPI, prepulse inhibition; TNF-α, tumor necrosis factor alpha.
Fig 4.
TNF-α infusion in AI results in tinnitus in normal-hearing wild-type and TNF-α KO mice.
A. Behavioral evidence of tinnitus was assessed with gap detection. B. The ability to hear tones was assessed with PPI. TNF-α infusion resulted in impaired gap detection in both wild-type and TNF-α KO mice but did not alter their performances in PPI. Albumin infusion did not affect gap detection or PPI in either wild-type or TNF-α KO mice. All mice had normal hearing. Data associated with this figure can be found in S1 Data. All results are presented as mean ± SEM and * indicates P < 0.05. AI, primary auditory cortex; KO, knockout; PPI, prepulse inhibition; TNF-α, tumor necrosis factor alpha.
Fig 5.
Microglial depletion down-regulates TNF-α expression and prevents noise-induced tinnitus.
A. Treatment with PLX3397 suppressed the TNF-α mRNA level in AI and prevented noise-induced increases in TNF-α expression 12 hours after noise exposure. Control mice (n = 4) that had undergone PLX3397 treatment and sham exposure had a lower TNF-α mRNA level compared to naïve mice (n = 7). In PLX 3397-treated mice (n = 4), noise exposure did not significantly increase TNF-α expression in AI of either hemisphere. * indicates P < 0.05 compared to naïve. B. Behavioral evidence of tinnitus was evaluated with gap detection. Behavioral tests were conducted at four time points: 1) before PLX 3397 treatment (naïve), 2) after 21 days of PLX3397 treatment (PLX), 3) after noise exposure and still with PLX 3397 treatment (PLX/post noise), and 4) after 7 days of washout of PLX3397 (post noise/PLX washout). Microglial depletion improved basal-level gap detection and prevented noise exposure–induced impairment in gap detection as observed in wild-type mice in Fig 3C. *, **, and *** indicate P < 0.05, P < 0.01, and P < 0.001, respectively. C. Animals’ ability to hear tones, which was measured with PPI, was not altered by PLX treatment or noise exposure (treatment effect, P = 0.15). Data associated with this figure can be found in S1 Data. Error bars represent SEM. PPI, prepulse inhibition; TNF-α, tumor necrosis factor alpha.
Fig 6.
Treatment with dTT prevents noise-induced TNF-α expression and microglial deramification 5 days after the exposure.
A. Treatment with dTT completely prevented TNF-α mRNA increase in AI area 5 days after noise exposure. Mice in the Noise group were exposed to noise. Mice in the Noise/dTT group underwent noise exposure followed by 5 days of dTT treatment. n = 7 mice for control, 6 for noise, and 4 for noise/dTT groups. B. Example images of microglia in AI visualized with IBA1 antibody staining. C. Noise exposure–induced microglial activation, as quantified with the soma-to-whole cell size ratio, was completely blocked by dTT treatment. Error bars represent SEM. n = 12 sections for control, 16 for noise, and 24 for noise/dTT groups. Data associated with this figure can be found in S1 Data. * depicts P < 0.05 and ** indicates P < 0.01 compared to control. AI, primary auditory cortex; dTT, 3,6′-dithiothalidomide; IBA1, ionized calcium binding adaptor molecule 1; TNF-α, tumor necrosis factor alpha.
Fig 7.
Treatment with dTT prevents noise-induced proinflammatory cytokine expression and tinnitus 10 days after the exposure.
A–D. mRNA levels of proinflammatory proteins TNF-α, NLRP3, IL-1β, and IL-18 in AI were up-regulated 10 days after noise exposure, and the up-regulations were completely blocked by dTT treatment. n ≥ 4 mice for each group. E. Treatment with dTT completely prevented noise exposure–induced tinnitus, as assessed with gap detection. F. The ability to detect tones, as assessed with PPI, was not altered by noise exposure and dTT treatment. Data associated with this figure can be found in S1 Data. Error bars represent SEM. * depicts P < 0.05 and ** indicates P < 0.01. AI, primary auditory cortex; dTT, 3,6′-dithiothalidomide; IL, interleukin; NLRP3, nod-like receptor protein 3; PPI, prepulse inhibition; TNF-α, tumor necrosis factor alpha.
Fig 8.
Treatment with dTT prevents noise-induced excitatory-to-inhibitory synaptic imbalance.
A. An image of a recorded pyramidal neuron. B–C. Example traces of mIPSCs and mEPSCs recorded from pyramidal neurons from control, noise exposed or noise-exposed and dTT-treated animals. D–G. Frequency and amplitude of mIPSCs and mEPSCs. Noise exposure resulted in a reduction of mIPSC frequency and an increase of mEPSC frequency, both of which were completely blocked by dTT treatment. Error bars depict SEM. n = 16, 10, and 19 for mIPSC of the control, noise, and noise/dTT groups and n = 6, 9, and 5 for mEPSC of the control, noise, and noise/dTT groups. Data associated with this figure can be found in S1 Data. * depicts P < 0.05 and ** indicates P < 0.01. dTT, 3,6′-dithiothalidomide; mEPSC, miniature excitatory synaptic current; mIPSC, miniature inhibitory synaptic current.
Fig 9.
Treatment with dTT alleviates noise-induced tinnitus measured with a conditioned lick suppression paradigm.
Licking rates during sound and silent trials were measured before and after noise exposure with dTT or vehicle injections. Both vehicle- and dTT-injected animals had similar baseline licking rates during silent and sound trials. Following noise exposure, all groups significantly increased their licking rates during silent trials, suggesting that they perceived tinnitus and increased licking in order to receive presumed water rewards. However, dTT-treated animals licked significantly less during silent trials compared to vehicles. This suggests that dTT suppressed tinnitus perception. Importantly, sound trial licking remained consistent following noise exposure and dTT or vehicle injections, indicating that manifestation and suppression of tinnitus-like behavior, or silent trial licking, was not due to overall changes in behavior. The shaded area represents the band of noise exposure (8–16 kHz). Data associated with this figure can be found in S1 Data. Error bars depict SEM. ** and *** indicate P < 0.01 and P < 0.001, respectively. dTT, 3,6′-dithiothalidomide.
Table 1.
Primers for RT-PCR.