Processing of Paired Click-Tone Stimulation in the Mice Inferior Colliculus

The inferior colliculus (IC) is known as a neuronal structure involved in the integration of acoustic information in the ascending auditory pathway. However, the processing of paired acoustic stimuli containing different sound types, especially when they are applied closely, in the IC remains poorly studied. We here firstly investigated the IC neuronal response to the paired stimuli comprising click and pure tone with different inter-stimulus (click-tone) intervals using in vivo loose-patch recordings in anesthetized BALB/c mice. It was found that the total acoustic evoked spike counts decreased under certain click-tone interval conditions on some neurons with or without click-induced supra-threshold responses. Application of click could enhance the minimum threshold of the neurons responding to the tone in a pair without changing other characteristics of the neuronal tone receptive fields. We further studied the paired acoustic stimuli evoked excitatory/inhibitory inputs, IC neurons received, by holding the membrane potential at -70/0 mV using in vivo whole-cell voltage-clamp techniques. The curvature and peak amplitude of the excitatory/inhibitory post-synaptic current (EPSC/IPSC) could be almost unchanged under different inter-stimulus interval conditions. Instead of showing the summation of synaptic inputs, most recorded neurons only had the EPSC/IPSC with the amplitude similar as the bigger one evoked by click or tone in a pair when the inter-stimulus interval was small. We speculated that the IC could inherit the paired click-tone information which had been integrated before reaching it.


INTRODUCTION
In the natural environments, acoustic stimuli are always performed in a temporal complicated context which is affecting their perception and processing in the brain. Paired sounds are commonly adopted to investigate the effect of temporal context on the neuronal responses. Temporal separation of two sounds in a pair could affect the perceptual grouping causing the auditory stream segregation (Bregman, 1990) as well as determine the response to the second sound, i.e., forward masking/suppression (Calford and Semple, 1995;Moore, 1995;Brosch and Schreiner, 1997;Wehr and Zador, 2005). Forward masking has been widely studied in the neuronal auditory pathway including the auditory nerve (Delgutte, 1980), the medial nucleus of the trapezoid body (Gao and Berrebi, 2016), the superior paraolivary nucleus (Gao et al., 2017), the inferior colliculus (IC) (Nelson et al., 2009) and the auditory cortex (Wehr and Zador, 2005). The suppression of neuronal response to the second sound in a pair increases as the temporal sound separation (from the offset of the first sound to the onset of the second one) decreases (Wehr and Zador, 2005;Gao and Berrebi, 2016). However, the neuronal processing and integration of paired acoustic information with different interstimulus intervals, especially being given nearly at the same time, are still unclear.
Although forward masking is found in many nuclei (Delgutte, 1980;Wehr and Zador, 2005;Nelson et al., 2009;Gao and Berrebi, 2016;Gao et al., 2017), we focused on the IC in this study considering its integrative function as a synaptic relay station for acoustic information in the neuronal auditory pathway (Malmierca and Hackett, 2010). Paired stimuli comprising of identical sounds are not suitable for identifying the sound processing, especially when they are applied closely, because of the superimposition and/or distortion. Therefore, we here adopted the paired sounds containing a short click and a long pure tone to study the basic features of the IC neuronal responses. The inter-stimulus (click-tone) interval was defined as the time window between the onsets of two sounds in a pair in the present study. Then, no matter how the recorded neurons responded to click, we studied the effect of click on the neuronal tone receptive fields under different inter-stimulus interval conditions.
Identifying the IC sub-threshold response features to the paired sounds will be crucial for understanding the mechanisms underlying the processing and integration of paired acoustic information. Summation of the synaptic inputs evoked by click and tone in a pair with short inter-stimulus intervals should be found on the IC neurons if these acoustic information were integrated on them. To testify this hypothesis, we attempted to investigate the post-synaptic responses induced by paired stimuli using in vivo whole-cell recording techniques. Excitatory and inhibitory post-synaptic currents (EPSC and IPSC) were separated by holding the membrane potential at −70 and 0 mV, respectively, in voltage-clamp configuration (Zhou et al., 2010). The peak amplitudes of EPSC/IPSC evoked by click and tone in a pair under different inter-stimulus interval conditions (especially when they were applied nearly at the same time) were analyzed to reveal the mechanisms underlying the processing and integration of paired sounds.

General
Totally seventy-six female BALB/c mice (aged 4-6 weeks, weighing 16-18 g) without any hearing defects provided by the Experimental Animal Center of Southern Medical University, Guangzhou, China were adopted. Surgery procedures, acoustic stimulation, data acquisition and processing had been approved by the Animal Care and Use Committee of Southern Medical University.

Surgical Preparation
Atropine sulfate (0.25 mg/kg. Sigma-Aldrich, St. Louis, MO, United States) was injected subcutaneously to reduce tracheal mucous secretion. Fifteen minutes later, urethane (1.2 g/kg. i.p., Sigma-Aldrich, St. Louis, MO, United States) was adopted to anesthetize the animal. The following experiment was conducted on an anti-vibration table placed in a double-walled soundproof room (air temperature: 24-26 • C). We used a stereotaxic apparatus to fix the animal's head via a 1.5 cm long nail stuck to the dorsal skull surface with dental cement. For the recordings, a 2 × 2 mm area was opened on the skull and the dura was removed under a surgical microscope (WPI, Sarasota, FL, United States) to expose the IC. Vaseline was used to cover the exposed brain during the experiment. The external auditory meatus on the same side of the recording was sealed with dental cement while the pinna on the other side was maintained as in normal animals.

Acoustic Stimulation
Acoustic stimuli were generated using a TDT 3 (Tucker-Davis Technologies, Alachua, FL, United States) and delivered to the animals via a free-field loudspeaker (ES1, frequency range: 2-110 kHz). The loudspeaker, calibrated by a 1/8 or 1/4 inch microphone (Brüel and Kjaer 4138,4135,Naerum,Denmark) and an amplifier (Brüel and Kjaer 2610, Naerum, Denmark), was placed 10 cm away from the animal's head facing the unsealed ear. The frequency, intensity, duration, rise/fall time/function of the acoustic stimuli were controlled manually or automatically via a computer with BrainWare software (Version 9.21. Tucker-Davis Technologies, Alachua, FL, United States).
Broadband noise (frequency: 0-50 kHz; intensity: 90 dB SPL (sound pressure level); duration: 50 ms; rise/fall time: 5 ms; rise/fall function: linear) was firstly applied to detect the neuronal acoustic response. After an IC auditory neuron was found, a frequency-intensity scan was performed by applying pure tones (frequency: 2-64 kHz in 0.1 octave steps; intensity: 10-90 dB SPL in 10 dB steps; duration: 50 ms; rise/fall time: 5 ms; rise/fall function: linear) to the animals (1/s) randomly. The CF (characteristic frequency, the tone frequency at which a neuron responded to the lowest stimulus intensity), MT tone [minimum threshold to tone, the minimum tone intensity at CF eliciting a spike with probability of 0.1 (Keithley and Feldman, 1983)] and BW 10 (the bandwidth of the tones eliciting response at 10 dB above the MT tone ) were identified to evaluate the neuronal tone receptive fields. Then, clicks with varying intensities (10-90 dB SPL in 10 dB steps) generated by using 0.1 ms square pulses were delivered randomly (1/s) to test the neuronal response. In this study, a "click +" neuron was termed when it had clickevoked spikes while, otherwise, as a "click −" neuron. On the neurons having click responses, the minimum intensities of clicks eliciting a spike with probability of 0.1 were identified as the MT to click (MT click ).
To investigate the processing of two sounds on the same neurons, we adopted paired stimuli consisting of a click (80 dB SPL) and a pure tone (at CF, 70 dB SPL, other parameters were same as those used previously in this study). The interstimulus intervals were expressed as "delay click − delay tone " and the delay click/tone was calculated from the recording onset to the sound onset. We applied the paired sounds with the interstimulus intervals ranging from −50 to 100 ms in 10-20 ms steps or in 1-5 ms steps, when they were close, to the animals randomly. Furthermore, a frequency-intensity scan with a click (80 dB SPL) ahead of each tone [inter-stimulus interval as that causing total acoustic evoked spike count (SC total ) changes by 20%] was performed to reveal the effect of click on the neuronal tone receptive field.
The location of the recorded neuron was confirmed by iontophoretically applied biocytin (1-10 nA for 300 ms ON and 300 ms OFF for 30 min as previous studies; Pinault, 1994Pinault, , 1996 using microiontophoresis (Neurophore BH-2, Harvard, Holliston, MA, United States). The animal was sacrificed with an overdose of pentobarbital sodium (80 mg/kg, i.p.) and perfused transcardially with saline (0.9%) and fixative (4% paraformaldehyde in 0.1 M PBS, pH 7.4). We removed the brain and fixed it in the same fixative at 4 • C overnight, and then for cryoprotection, it was immersed in 20 and 30% sucrose overnight. The brain was cut coronally into 40 µm slices using a freezing microtome (Leica CM 1950, Nussloch, Germany). For staining, the slices were washed by 0.1M PBS (3 × 10 min) and permeabilized with Triton (0.3%) for 2 h to increase permeability of antibody on the cell membrane. After PBS rinsing (3 × 10 min), the slices were incubated with Molecular probes,United States) for 4 h at room temperature. After being washed and transferred to subbed slides, the sections were stained with 4 ,6-diamidino-2-phenylindole (DAPI, 0.25 µg/ml) to visualize the nucleus. Confocal microscopes (Nikon, A1R, Japan) were adopted to examine the location of the recorded neurons. Data from the neurons recorded outside the IC were discarded.

Whole-Cell Recordings
In vivo whole-cell recordings were adopted to reveal the acoustic evoked excitatory/inhibitory synaptic inputs to the IC neurons. Glass microelectrodes containing a cesium-based solution (in mM: 125 Cs-gluconate, 8 phosphocreatine-2Na, 5 TEA-Cl, 4 MgATP, 0.3 GTP, 10 HEPES, 10 EGTA, 1 QX-314, and 2 CsCl) and 0.5% biocytin (pH 7.3, impedance: 4-7 M ) were used. By holding the membrane potential at −70 and 0 mV, respectively (Zhou et al., 2010), we separated the excitatory and inhibitory inputs. The IC neuronal activity was recorded for 50 ms before the acoustic stimulation was applied to obtain the baseline. And the baselines obtained from the same neuron for five representations of the identical acoustic stimulation were averaged (expressed as Mean ± SD). EPSC/IPSC was determined when the post-synaptic current changed from the Mean baseline by more than 2 SD.
Paired click-tone stimuli were adopted to explore the relationship between the synaptic inputs induced by different acoustic stimuli. The intensity of click (70 dB) and tone (at CF, 80 dB) in each paired stimulation were fixed while their intervals (delay click -delay tone ) changed from −100 to 100 ms in 10-20 ms steps or in 1-5 ms steps when they were close. The recorded neurons were labeled with biocytin and data from neurons within the IC were further analyzed.

Data Processing
The waveforms of the neuronal spikes were collected and stored as data sets. The IC neuronal responses recorded extracellularly to five presentations of identical acoustic stimulation (including paired stimuli) were displayed as post-stimulus time histograms (PSTH). The acoustic evoked SCs were calculated within a 100 ms time window started at the stimulus onset (Liang et al., 2011).
To investigate the processing of click and tone in a pair on the same neurons, we plotted the SC total as a function of inter-stimulus intervals and compared the SC total s evoked by paired sounds with three inter-stimulus intervals (−40, 1 and 40 ms). Then, the CF, MT and BW 10 of the IC neuronal tone receptive field with and without a click before each tone with a certain inter-stimulus interval (causing SC total changes by 20%) were compared to evaluate the effects of clicks on the neuronal receptive fields to tones. Finally, we analyzed the relationship between excitatory/inhibitory inputs (EPSCs/IPSCs) evoked by paired click-tone stimulations with different interstimulus intervals.
We used Excel 2007 and OriginPro 7.5 to calculate the values of relevant parameters and data fitting and plotting. One-way ANOVA was used to compare means and LSD's test was used for multiple comparisons. Paired-samples T Test was also adopted to compare means of paired samples in this study. P < 0.05 was concerned as significant difference.

IC Neuronal Responses to Clicks and/or Pure Tones
Totally one hundred and eight IC neurons recorded in anesthetized mice ( Figure 1A) had responses to the pure tones in this study and showed as disk-shaped cells (Oliver et al., 1991) sampled as in Figure 1B. The recording depth ranged from 232 to 1580 µm beneath the brain surface. The CF, MT tone and BW 10 were 2 -42.22 (13.92 ± 7.46) kHz, 10 -60 (22.04 ± 12.13) dB SPL and 0.59 -21.30 (5.43 ± 4.86) kHz, respectively.

IC Neuronal Responses to Paired Click-Tone Stimulation
Effect of the Inter-Stimulus Interval of Paired Stimuli on the IC Neuronal Response In the present study, we adopted paired stimuli comprising a click (80 dB SPL) and a tone (at CF, 70 dB SPL) with different intervals (delay click -delay tone ) to identify the basic features of the processing of them on the IC neurons.
On the IC neurons showing spikes evoked by both click and tone in a pair ("click +" neurons: n = 77, e.g., Figures 1C,D), it is hard to separate the click-evoked spikes from those induced by tones when the inter-stimulus intervals were small. Therefore, on them, we calculated the total SC (SC total ) evoked by paired click and tone. The SC total obtained when the click was 50 ms earlier than the tone in a pair was adopted to normalize those under other inter-stimulus interval conditions by dividing them. In this study, the significant changes of neuronal responses were considered as that the evoked SC total varied by more than 20% of the reference.
The normalized SC total decreased when the paired sounds were given to the animal with short inter-stimulus intervals on 35.06% of the recorded "click +" neurons (n = 27, Figures 2A,B,E). On these neurons, the normalized SC total evoked by paired sounds with −40, 1, and 40 ms clicktone intervals were compared (Figure 2E'. One-way ANOVA, p = 0.011; LSD's test, −40_1: p = 0.007; −40_40: p = 0.017 and 1_40: p = 0.337). It indicated that the processing of click and tone in a pair on these IC neurons had interactions when they were close. No significant difference was found in the CFs of the "click +" neurons with (n = 27, 13.25 ± 7.15 kHz) and without (n = 50, 14.79 ± 7.27 kHz) the interactions between the click and tone responses (One-way ANOVA, p = 0.375).
We used the same procedure to test if it is also the case on the neurons without any spikes induced by clicks ("click −" neurons: n = 31). Since these neurons only had tone-evoked spikes, SC total equaled SC tone . The SC total decreased significantly when paired sounds had small click-tone intervals on seventeen "click −" neurons (54.84%, Figures 2C,D,F. Comparison of normalized SC total with −40, 1 and 40 ms inter-stimulus intervals: Oneway ANOVA, p = 0.013; LSD's test, −40_1: p = 0.004; −40_40: p = 0.382 and 1_40: p = 0.018, Figure 2F'). These results suggested that click-evoked sub-threshold inputs should affect the responses of the "click −" neurons to tones. Similar as the "click +" neurons, there was no significant difference in the CFs between the "click −" neurons with (n = 17, 14.00 ± 7.39 kHz) and without (n = 14, 12.00 ± 9.14 kHz) the effects of clicks on neuronal tone responses (One-way ANOVA, p = 0.504).
The power spectrum of the click used in this study was majorly less than 10 kHz (inserted in Figure 1C). So, "10 kHz" was adopted as a cutoff to categorize the recorded IC neurons based on their receptive frequency ranges at 80 dB SPL ( Table 1). The "click +/−" neurons (n = 77/31) with the receptive ranges higher than, lower than or cross 10 kHz were categorized, and so did the neurons showing interactions between the processing of the paired click-tone stimulation (n = 27/17). We found that, even on the neurons having receptive ranges higher than 10 kHz, the processing of click and tone in a pair had interactions ("click +" neurons: 6 out of 23; "click −" neurons: 4 out of 6). It indicated that the clicks might widely provide synaptic inputs to the IC neurons, even to those with receptive fields located apart from the major click frequency range.

Effects of Click on the IC Neuronal Receptive Field to Pure Tone
To reveal the mechanism underlying the affection of the clicktone interval on the neuronal response, we investigated the effect of click on the IC neuronal receptive field to pure tone by performing two frequency-intensity scans with or without a given click (80 dB SPL) before each tone. The inter-stimulus intervals were set as those causing the acoustic evoked SC total changes by 20% in the decreasing phase. The CF, MT tone and BW 10 were adopted to evaluate the location and shape of neuronal tone receptive fields in the coordinate system.
Both CF and BW 10 were not affected by the click application on some neurons with [n = 15, Figures 3A,C. Paired-samples T-test (two-tailed), p = 0.388, 0.942] or without [n = 13, Figures 3D,F, Paired-samples T-test (two-tailed), p = 0.656, 0.167] click evoked supra-threshold responses. It indicated that the clicks could have no effects on the horizontal location and shape of the IC neuronal tone receptive fields in the coordinate system. Interestingly, clicks could shift the tone receptive fields upward vertically by enhancing the MT tone on these "click +" [n = 15, Figure 3B, Paired-samples T-test (two-tailed), p = 0.001] or "click −" neurons [n = 13, Figure 3E, Paired-samples T-test (two-tailed), p = 0.001]. To sum up, the neuronal responses to the tones with different frequencies could be widely inhibited by the application of clicks by showing corresponding increased MTs. However, the corresponding synaptic mechanisms underlying this inhibition is still unclear.

Synaptic Inputs Evoked by Click and Tone in a Pair to the IC Neurons
We recorded the synaptic inputs evoked by paired acoustic stimuli to the IC neurons by using in vivo whole-cell recording techniques (in the voltage-clamp configuration). The excitatory and inhibitory post-synaptic currents were isolated by holding the membrane potential at −70 and 0 mV, respectively (Wu et al., 2008). In this study, we totally recorded the paired click-and/or tone-induced PSCs on eleven IC neurons (CF: 16.64 ± 8.22 kHz). And only on six of them (CF: 15.61 ± 4.30 kHz), both acoustic evoked EPSCs and IPSCs were recorded. We attempted to reveal the relationship between the EPSCs/IPSCs evoked by the sounds in a pair with different intervals on these six neurons.
PSC peak amplitudes for five presentations of identical acoustic stimulation were averaged and the maximal amplitude of completely overlapped PSC was set as a reference to normalize the PSCs obtained under different click-tone intervals by dividing them with it. The normalized peak amplitudes were plotted as a function of the inter-stimulus interval of the paired sounds (Figures 5A,B, the same neurons as in Figure 4). On the same neurons, the relationship between the peak amplitudes of EPSC and IPSC could be consistent (n = 3. e.g., Unit15-0909-002, the tone-evoked PSCs were higher than those induced by clicks, One-way ANOVA, p < 0.001 for both EPSC and IPSC, Figures 4A2,B2, 5E,F), or not (n = 3., e.g., Unit15-0908-002, click evoked higher EPSC and similar IPSC as tone, One-way ANOVA, p < 0.001 and p = 0.967, respectively, Figures 4A1,B1, 5C,D).
To reveal the relationship between PSCs evoked by paired stimuli, we compared the peak amplitudes of PSCs induced by click (Figures 5G-J, 6B-I) or tone (Figures 5K-N, 6B-I) in a pair  (17) 108 (44) Tone responsive neurons recorded extracellularly with (n = 77) or without (n = 31) click-evoked spikes were categorized based on their frequency receptive fields at 80 dB SPL. Three categories: more than, less than, and cross 10 kHz. Neurons having interactions between the processing of click and tone (n = 44) were also categorized (cell numbers were shown in brackets).
These results suggested that some IC neurons only responded to a single acoustic stimulus (click or tone) in a pair which induced the relative larger synaptic inputs when the paired sounds were applied nearly at the same time.

Properties of Neuronal Click Perceptions in the IC
Tonotopic organization and sharp frequency tuning in the IC have been widely studied in the cat (Aitkin et al., 1975;Semple and Aitkin, 1979;Serviere and Webster, 1981;Brown et al., 1997;Schreiner and Langner, 1997), gerbil (Ryan et al., 1982;Grana et al., 2017), monkey (FitzPatrick, 1975), bat (Zook et al., 1985;Poon et al., 1990), rat (Clopton and Winfield, 1973;Huang and Fex, 1986;Malmierca et al., 2008), mouse (Stiebler and Ehret, 1985), and human (De Martino et al., 2013;Ress and Chandrasekaran, 2013). As a powerful acoustic stimulation, click could generally evoked the responses on the IC neurons whose receptive frequency range at 80 dB had intersections with the click power spectrum (majorly < 10 kHz, Figure 1 and Table 1). Interestingly, even on those neurons whose receptive frequency ranges (at 80 dB) were away from the click power spectrum, clicks could also provide synaptic inputs ( Table 1) by showing supra-threshold spikes ("click +" neuron, Figure 1) or sub-threshold inputs affecting the neuronal tone response ("click −" neuron, Figures 1, 2C,D,F). We therefore speculate that, in addition to the tonotopic organized lemniscal pathway, the click information might reach the IC via the non-lemniscal pathway.

Properties of Click Affecting the IC Neuronal Response to the Tone in a Pair
We found in this study that even on those IC neurons only having the supra-threshold responses to tones (i.e., "click −" neurons), the application of click could depress their tone responses (Figures 2C,D,F,F'). And these effects were depending on the inter-stimulus intervals, several to tens of milliseconds, between the click and tone in a pair no matter which sound was applied early (Figures 2F,F'). It should be the sub-threshold response evoked by click causing these changes of the IC tone response by enhancing their MT tone s On a neuron with similar (Figures 3D-F).
Although the SCs evoked by click and tone in a pair were adopted to evaluate the interactions between them on the "click +" neurons, the changes of SC total were similar as those on the "click −" neurons (Figure 2). The clicks enhanced the FIGURE 4 | Synaptic inputs induced by paired stimulation with different inter-stimulus intervals (Examples: 1: Unit15-0908-002 and 2: Unit15-0909-002). Five repeats of EPSCs (holding at −70 mv, A,C,E,G,I,K,M) and IPSCs (holding at 0 mv, B,D,F,H,J,N) were averaged. Click and tone evoked EPSCs A1 and IPSCs (B1/2,N1/2) were apart from each other when the inter-stimulus intervals (delay click -delay tone ) were large (Unit15-0908-002: −50 ms, A1/B1, 50 ms, M1/N1; Unit15-0909-002: -100 ms, A2/B2, 100 ms, M2/N2). And they were partially overlapped when the click was moved towards (C1/2-F1/2) or away from the tone in a pair (G1/2-L1/2). When the click and tone in a pair were applied to the animal with a short interval or at the same time (0 ms: G1/H1 and −5 ms: G2/H2), the PSCs evoked by tone or click were completely overlapped by showing only one PSC.
(F) The completely overlapped IPSC amplitudes were similar as the larger ones evoked by clicks (n = 1, One-way ANOVA, click: p = 0.071, tone: p < 0.001). (G) On the neurons with similar IPSCs induced by the paired stimuli (n = 3), all the IPSCs obtained under different inter-stimulus intervals had no significant differences in the peak amplitudes (One-way ANOVA, click: p = 0.961, tone: p = 0.511). (H) The completely overlapped IPSC amplitudes fell in between those induced by click and tone (One-way ANOVA, click: p = 0.347, tone: p = 0.156) on the neuron with larger tone responses (the same neuron as in Figures 5J,N). (I) Although the completely overlapped IPSC amplitudes were similar as the larger ones evoked by tones (One-way ANOVA, click: p = 0.007, tone: p = 0.553), the inter-stimulus intervals had complicated effects on the tone-evoked IPSCs (LSD's test, before_during: p = 0.189; during_after: p = 0.003; before_after: p = 0.009).

Mechanism Underlying the Processing of Paired Click-Tone Stimulation on the IC Neurons
Although previous studies had indicated that a delayed, longlasting GABAergic inhibition should be involved in the forward masking formation in the sensory cortex (Swadlow, 2003;Wehr and Zador, 2005;Isaacson and Scanziani, 2011), the integrative patterns of the excitatory synaptic inputs induced by paired sounds remains unclear, and neither does the inhibitory inputs. In this study, by holding membrane potentials at different levels (−70 and 0 mV, respectively), we separated the acoustic (click/tone) evoked EPSC and IPSC focusing on their peak amplitude changes along with the inter-stimulus intervals of paired sounds. If the integration of paired acoustic information with relative short inter-stimulus intervals takes place in the IC, the summation of the sub-threshold responses (EPSC or IPSC) should occur on the recorded IC neurons by showing the changes in the curvature and peak amplitude as a function of the inter-stimulus intervals. In another word, the PSCs induced by the two sounds in a pair should be superimposed with each other when the inter-stimulus intervals are short. However, in spite of the small sample size (n = 6, Figure 6), our present in vivo whole-cell recording results indicate that it is not the case on most recorded IC neurons by showing relative unchanged EPSC (n = 5, Figures 4, 5G,K,I,M, 6B-D) or IPSC (n = 4, Figure 4, 5H,L, 6F,G) under different click-tone interval conditions. Although the paired click and tone could evoke PSCs with different peak amplitudes, when the paired sounds are given closely, the bigger EPSC/IPSC induced by click or tone completely overlapped the relative small one without obvious superimposition (Figures 4, 5G-I,K-M, 6B-D,F,G). Only a few exceptions for the complicated changes in EPSC (n = 1, Figure 6E) or IPSC (n = 2, Figures 6H,I) evoked by paired sounds with different click-tone intervals were found. Therefore, we have reasons to believe that, at least, part of the whole-cell recorded neurons in this study inherit the paired acoustic information, especially when they are relatively close to each other, integrated before it reaches the IC. Basilar membrane and auditory nuclei in the ascending auditory pathway below the IC might be the candidates for the acoustic information integration which needs further investigations.

DATA AVAILABILITY
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.