Blockade of cochlear NMDA receptors prevents long-term tinnitus during a brief consolidation window after acoustic trauma

Abstract

Tinnitus, the perception of sound in the absence of external acoustic stimulation, is a common and devastating pathology. It is often a consequence of acoustic trauma or drug toxicity. The neuronal mechanisms of tinnitus are neither yet fully understood nor are effective treatments available. Using a novel behavioral paradigm for measuring tinnitus in the rat based on tone-guided navigation, we show here that the development of long-term noise-induced tinnitus, the most prevalent and clinically important form of human tinnitus, can be abated by local administration of the NMDA antagonist "ifenprodil" into the cochlea in the first 4 days following the noise insult but not afterwards. This suggests that long-term tinnitus undergoes a consolidation-like process, resembling the ontogeny of items in long-term memory. Furthermore, this finding paves the way to potential therapeutic strategies for the prevention of chronic tinnitus once the noise insult had taken place.

Figures

Figure 1

The behavioral paradigm. (a) Schematic description of the task. Rats were conditioned to associate the location of a submerged platform in a water T-maze (WTM) with the absence or presence of a tone. When the tone was present, the platform was in one arm (tone arm), whereas when the tone was absent, the platform was in the opposite arm (no-tone arm). The sidearm-platform-tone permutations were counterbalanced between subjects within each treatment group to control for potential side preference. In the test, conducted 2 weeks after the end of training, the rat was placed in the maze for 100 seconds in the absence of the platform. The arm entered first by the rat and the time spent in each arm in the first and in the last 50 seconds of the test were recorded. (b) Lack of arm preference in the WTM test in the absence of conditioning. Animals were tested in the absence of the tone (upper panel, n = 8), in the presence of a 10 khz tone (middle panel, n = 8), or in the absence of the tone after salicylate treatment at a dose established to induce tinnitus (300 mg/kg/day for 4 days, tested 2 hours after the last administration, lower panel, n = 8). (c) Learning curves of the acquisition of the tone-platform association. Acquisition of the tone-platform association, expressed in time to reach the platform and in percent time spent in the correct arm, each averaged per training day (n = 124, ***P < .001 compared to the previous training day).

Figure 2

Induction of tinnitus by salicylate (sal), its blockade by NR2B antagonist (ifneprodil), and establishment of behavioral criteria for tinnitus. (a) Time spent in the tone arm during the test. The time spent by various treatment-groups (identified along the X-axis) in the tone arm during the first 50 seconds (left-hand bar in each pair) and the last 50 seconds of the test. Black bars refer to groups of animals which were tested after conditioning without any treatment. The test was performed two weeks after the end of conditioning. Rats tested in the presence of the tone displayed preference for the tone arm, whereas rats tested in the absence of the tone displayed preference for the no-tone arm. Salicylate-treated rats (300 mg/kg/day for 4 consecutive days, the last injection taking place 2 hours before the test) behaved as if they perceive a tone, spending most of their time in the tone arm though the tone was absent. Cochlear application of artificial perilymph (AP) had no effect, but cochlear application of ifenprodil reversed the behavior induced by salicylate (*P < .01 from animals tested in silence, # P < .01 from the first 50 seconds window of the same group (n = 8 each)). (b) Definition of the criteria used to designate rats as experiencing tinnitus. Rats were designated as perceiving a tone in its absence, that is, suffering from tinnitus, if in the WTM test in the absence of the tone they spent in the tone arm not less than 2 SD (P > .05) below the mean time spent by untreated conditioned rats, tested in the presence of the tone, in the tone arm. Similarly, treated rats were designated as lacking tinnitus if in the test they spent in the tone arm not more than 2 SD above the mean time spent by untreated conditioned rats, tested in the absence of the tone, in the tone arm (the separation of the two possible behavioral patterns generated according to the aforementioned criteria is emphasized by the grey zone). These criteria were used to designate rats as experiencing tinnitus in the noise-induced tinnitus experiments.

Figure 3

Induction of tinnitus by sound overexposure. The time spent in the tone arm by the conditioned rats tested in the absence of the tone, 2 weeks after the sound overexposure. Inset: raw data of WTM performance of the conditioned rats that underwent sound overexposure (n = 26). These rats segregated into two populations, based on the criteria depicted in Figure 2(b): in one population (n = 14), all the rats behave as if they are not experiencing the tone, while in the second (n = 12), as if they are experiencing the tone in its absence (for the grey zone, see Figure 2). Insert: the horizontal line in each group is the mean of that group; the individual data points are also displayed. Left and right panels: time spent by these two populations, respectively, in the tone arm during the first and the last 50 seconds of the test (*P < .01 compared to the other subpopulation, as well as to the conditioned untreated rats tested in the absence of the tone).

Figure 4

Effect on tinnitus of cochlear application of glutamatergic antagonists as a function of time after noise overexposure. Depicted is the proportion of animals experiencing tinnitus, as evident in the WTM test, 2 weeks after sound overexposure. Day 0 corresponds to drug application performed just before sound overexposure. (a) Effect of local cochlear application of ifenprodil as a function of time. When applied just before sound overexposure, none of the animals displayed tinnitus when tested 2 days later. Similar results were obtained when ifenprodil was applied 4 days after sound overexposure. However, when ifenprodil was applied 8 or 12 days after sound overexposure, about 50% of the rats experienced tinnitus (not different from the proportion observed in the group that underwent sound overexposure, chisquare test (*P < .05 compared to the group of animals receiving ifenprodil at day 0, Wilcoxon test)). (b) Effect of DNQX and mCPP. When applied just before sound overexposure, DNQX totally prevented the occurrence of tinnitus 2 weeks afterwards. However, when applied 4 days after sound overexposure, no protective effect was evident. Application of mCPP just before sound overexposure did not modify the proportion of rats experiencing tinnitus (n = 8 each, *P < .05 compared to the group of animals receiving ifenprodil at Day 0, Wilcoxon test).