B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Metabotropic glutamate receptors modulate glutamatergic andGABAergic synaptic transmission in the central nucleus of theinferior colliculus

Rasoul Farazifard, Shu Hui Wu⁎

Institute of Neuroscience, Department of Psychology, Carleton University, Ottawa, Ontario, Canada K1S 5B6

A R T I C L E I N F O

⁎ Corresponding author. 335 Life Sciences ReseOntario, Canada K1S 5B6. Fax: +1 613 520 405

E-mail address: shwu@ccs.carleton.ca (S.HAbbreviations: ACSF, artificial cerebral spin

of the inferior colliculus

0006-8993/$ – see front matter © 2010 Elsevidoi:10.1016/j.brainres.2010.02.021

A B S T R A C T

Article history:Accepted 4 February 2010Available online 11 February 2010

Fast glutamatergic andGABAergic transmission in the central nucleus of the inferior colliculus(ICC), a major auditory midbrain structure, is mediated respectively by alpha-amino-3-hydroxy-5-methylisoxazole-4 propionic acid (AMPA) and γ-aminobutyric acid (GABA)Areceptors. In this study, we used whole-cell patch clamp recordings in brain slices toinvestigate the effects of activation ofmetabotropic glutamate receptors (mGluRs) on synapticresponses mediated by AMPA and GABAA receptors in ICC neurons of young rats. Excitatoryand inhibitory postsynaptic currents (EPSCs and IPSCs) mediated respectively by AMPAand GABAA receptors were elicited by stimulation of the lateral lemniscus, the majorafferentpathway to the ICC.Theagonists for groups I and IImGluRs, (±)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (ACPD), and for group III mGluRs, L-2-amino-3-hydroxypropanoicacid 3-phosphate (L-SOP), did not affect intrinsicmembrane properties of the ICC neurons. Theagonist for group IImGluRs, (1R,4R,5S,6R)-4-amino-2-oxabicyclo[3.1.0] hexane-4,6-dicarboxylicacid (LY379268), significantly reduced the AMPA receptor-mediated EPSCs and GABAA

receptor-mediated IPSCs. The effects were reversed by the group II mGluR antagonist, (2S)-2-amino-2-[(1S,2S)-2-carboxycycloprop-1-yl]-3-(xanth-9-yl) propanoic acid (LY341495). Theagonists for groups I and III, (RS)-3,5-dihydroxyphenylglycine (DHPG) and L-SOP, respectively,did not affect AMPA or GABAA receptor-mediated responses. The reduction of the synapticresponses by LY379268 was accompanied by a substantial increase in a ratio of the second tothe first AMPA receptor-mediated EPSCs and GABAA receptor-mediated IPSCs to paired-pulsestimulation. The results suggest that group II mGluRs regulate both fast glutamatergic andGABAergic synaptic transmission in the ICC, probably through a presynaptic mechanism dueto reduction of transmitter release.

© 2010 Elsevier B.V. All rights reserved.

Keywords:HearingCentral auditory systemPatch clamp recordingMetabotropic glutamate receptorAMPA receptorGABAA receptor

arch Building, Institute of Neuroscience, Carleton University, 1125 Colonel By Drive, Ottawa,2.. Wu).al fluid; mGluR, metabotropic glutamate receptor; IC, inferior colliculus; ICC, central nucleus

er B.V. All rights reserved.

29B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

1. Introduction

Along the auditory pathway, the inferior colliculus (IC) is animportant midbrain integration center and plays a significantrole in auditory processing. Auditory information from thelower brainstem is analyzed and transformed by cells in the ICand is then sent to the auditory thalamus and further to theauditory cortex. Electrophysiological studies show that theneurons in the IC respond not only to simple acoustic stimuli,i.e., tones, but also to spectrally and temporally complexacoustic stimuli (Portfors and Sinex, 2005). The responses ofthe IC neurons to the specific features of acoustic stimuli mayunderlie processing of biologically significant sounds (Portforset al., 2009).

A better understanding of the mechanisms of auditoryprocessing in the IC can be obtained from studies of synaptictransmission and modulation. Early in vivo physiological andpharmacological studies showed that glutamate plays amajor role in evoking excitatory responses to sounds andγ-aminobutyric acid (GABA) mediates most inhibitoryresponses (Faingold et al., 1991). Later studies further demon-strated that the early and fast responses of the IC neuronsto contralateral acoustic stimulation are mediated by iono-tropic α-amino-3-hydroxy-5-methylisoxazole-4 propionic acid(AMPA) receptors (ZhangandKelly 2001, 2003).Manyaspects ofauditory processing in the IC, including frequency tuning,selectivity of frequencymodulation sweeps, binaural time, andintensity coding, and sound duration tuning, depend on thesynaptic inhibition mediated by GABAA receptors (D'Angelo,et al., 2005; Fuzessery and Hall, 1996; Ingham and McAlpine,2005; Klug et al., 1995; Le Beau et al., 1996, 2001; Koch andGrothe, 1998; Nataraj and Wenstrup, 2005, 2006; Palombi andCaspary, 1996; Pollak et al., 2002; Vater et al., 1992; Wang et al.,2007; Wu and Jen, 2006; Yin et al., 2008).

In the central nucleus of the IC (ICC), one of the majorsubdivisions of the IC, individual neurons receive excitatory aswell as inhibitory inputs (KellyandCaspary, 2005;Maetal., 2002;Wu et al., 2004). Balanced excitatory and inhibitory activity inICC neurons is important for them to integrate inputs andprocess temporal information accurately. One of the neuronalmechanisms to maintain the balance of excitation and inhibi-tion is synaptic modulation (Lu, 2007). It is well known thatglutamate acts not only on the ionotropic receptors but also onthe metabotropic receptors (mGluRs) and that activation of themGluRs results in a large number of diverse cellular actions inthe central nervoussystem (Anwyl, 1999; Cartmell andSchoepp,2000). The mGluRs are classified into three groups (group I:mGluR1 and 5; group II: mGluR2 and 3; and group III: mGluR4, 6,7, and 8) based on sequence similarity, second messengersystem involvement, and relative pharmacology (Conn and Pin,1997; Kew and Kemp, 2005). Immunocytochemical and molec-ular biological studies demonstrate that all three groups of themGluRs are present in the IC (Ferraguti and Shigemoto, 2006;Ohishi et al., 1993b; Petralia et al., 1996; Romano et al., 1995;Shigemoto et al., 1993; Tamaru et al., 2001).

The physiological role of the mGluRs has been investigatedin several structures along the central auditory pathway,including the cochlear nucleus, superior olivary complex,medial geniculate nucleus, and auditory cortex (Bandrowski

et al., 2001, 2002; Barnes-Davies and Forsythe, 1995; Kotak andSanes, 1995; Kudoh et al., 2002; Lu, 2007; Lu and Rubel, 2005;Nishimaki et al., 2007; Sanes et al., 1998; Schwarz et al., 2000;Takahashi et al., 1996; Wu and Fu, 1998). However, little isknown about the physiological function of the mGluRs in theIC. There has been only one previous study by Voytenko andGalazyuk (2008) demonstrating that the mGluRs contribute tothe encoding of temporally complex sounds. However, thedetails of the neural mechanism of action of the mGluRs onthe IC neurons are not known. Therefore, in this study, weused whole-cell patch clamp recordings in brain slice pre-parations to investigate whether the mGluRs play a role insynaptic transmission in the ICC. We examined if intrinsicmembrane properties and synaptic responses mediated byAMPA and GABAA receptors of ICC neurons were affected bymGluR agonists. Furthermore, we determined which group ofthe mGluRs influenced the synaptic transmission in the ICC.Finally, we explored the underlying mechanism of synapticmodulation by the mGluRs.

2. Results

Whole-cell patch clamp recordings were obtained from 150ICC neurons in this study. We recorded responses from onlyone neuron in each ICC slice to avoid interaction of pharma-cological agents applied to different neurons in the same slice.We examined intrinsic membrane properties of 16 neuronsunder current clamp mode and synaptic responses of 134neurons under voltage clamp mode. The resting potential ofthese neurons was stable throughout the recording. Themeanresting potential was −61.3±0.3 mV (range:−57.0 to −65.5 mV,n=16). Spontaneous firing and synaptic activity were very rarein these neurons.

2.1. Effects of activation of mGluR on intrinsic membraneproperties

In current clamp mode, responses of 16 ICC neurons todepolarizing and hyperpolarizing current injections wereexamined. Those neurons producedmultiple action potentialswhen membrane depolarization was higher than the thresh-old level for generation of action potentials. In response tohyperpolarizing current injection, some neurons (11/16, 69%)showed a sag of the membrane potential at high levels ofcurrent injection and generated a rebound depolarizationimmediately after hyperpolarization. The sag was alwaysaccompanied by the rebound in these cells. Figs. 1A and Cillustrate the responses of two representative neurons topositive and negative current injections. Both neurons had asustained firing pattern in response to positive currentinjection. However, these two neurons responded to hyper-polarizing current differently. The neuron in Fig. 1C had arebound, whereas the one in Fig. 1A did not.

Seven ICC neurons with known intrinsic membraneproperties were labeled with Lucifer Yellow. Four out of theseven neurons showed similar morphology and were classi-fied as bipolar cells (Figs. 1B and D). Bipolar neurons had a cellbody with fusiform shape. Their primary dendrites emergedfrom opposite ends of the cell body. Their dendritic branches

Fig. 1 – Effects of ACPD and L-SOP on themembrane properties of ICC neurons. (A) Responses of a neuron to positive (120 pA) andnegative (−40 to −200 pA in 40 pA/step) current injections in ACSF (left panel) and ACPD (right panel). This cell was a non-reboundcell. (B) A neuron was labeled with Lucifer Yellow and classified as a bipolar cell. This neuron was a non-rebound neuron.(C) Responses of a neuron to the current injection of 100, 0, −40, −80, −120, −160, and −200 pA in ACSF (left panel) and ACPD (rightpanel). This neuron showed a sag of the membrane potential during hyperpolarization and a rebound depolarization (*) afterhyperpolarization. (D) A neuron was labeled with Lucifer Yellow and classified as a multipolar cell. This neuron was a reboundneuron. * in B and D indicates the location of the neuron in the ICC. Scale=20μm in B and D. (E) The current–voltage relationshipderived from responses to hyperpolarizing current injections before and after application of ACPD. (F) The current–voltagerelationship derived from responses to hyperpolarizing current injections before and after application of L-SOP.

30 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

extended in a direction roughly parallel to the long axis of thecell body as well as the fibrodendritic laminae in the ICC.Among the four bipolar cells, one was non-rebound and threewere rebound cells. The other three cells were classified asmultipolar cells (data not shown, but similar to cells in Figs. 2Dand 4D). These cells had a cell body withmultipolar shape andtheir primary dendrites emerged radially from the cell body.The most dendritic branches were oriented across thefibrodendritic laminae. Among the three multipolar cells,one was non-rebound and the other two were rebound cells.

The agonist for mGluR I and II, (±)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (ACPD), at 100 μMand the agonist formGluR III, L-2-amino-3-hydroxypropanoic acid 3-phosphate (L-SOP), at 100 μM were used to test whether intrinsic membrane

properties of the ICC neurons were affected by activation of themGluR. Application of ACPDdid not affect the firing pattern andfrequency in either rebound or non-rebound neurons (Figs. 1AandC, right panels).ACPDalsodidnothaveanyeffect on the sagand rebound (Fig. 1C, right panel). Current–voltage relationshipsof the ICCneuronswerenot alteredbyACPD (n=8, Fig. 1E). Otherintrinsic membrane characteristics, i.e., the resting membranepotential, input resistance, threshold, and amplitude of actionpotentials, were also not significantly affected by ACPD (p>0.05,Table 1).

Similar to ACPD, L-SOP had no effect on the firing patternand frequency or the sag and rebound (data not shown). Therewere no significant changes in current–voltage relationshipsof the ICC neurons after administration of L-SOP (n=8, Fig. 1F).

Table 1 – Effects of ACPD and L-SOP on restingmembranepotential (Vm), input resistance (Rin), action potential (AP)threshold, and AP amplitude of ICC neurons.

Cell no. Vm Rin (MΩ) APthreshold

(mV)

APamplitude

(mV)

ACSF n=8 −60.5±0.9 300±39 −40.7±2.6 54.8±3.9ACPD −60.6±0.9 305±32 −41.1±2.5 55.3±3.8ACSF n=8 −62.0±0.6 288±42 −42.4±2.7 58.6±3.6L-SOP −62.4±0.3 290±40 −41.0±3.0 59.7±3.5

31B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

Nor did L-SOP affect other intrinsic membrane properties ofthe ICC neurons (p>0.05, Table 1).

2.2. Effects of activation of mGluR on synaptic responsesmediated by AMPA receptors

Electrical stimulation of the lateral lemniscus (LL), the majorascending afferent pathway to the ICC, usually inducedmixedexcitatory (AMPA and NMDA receptor-mediated) and in-hibitory (GABAergic and glycinergic) responses in ICC neurons(Ma et al., 2002; Sun et al., 2006). Therefore, for this set ofexperiments, the AMPA receptor-mediated excitatory post-synaptic currents (EPSCs) were pharmacologically isolated bysuppressing GABAA-, glycine-, and NMDA receptor-mediatedresponses with their specific receptor antagonists (see detailsin Section 4). We then determined the effects of mGluR I, II,and III agonists on AMPA receptor-mediated responses. Bathapplication of ACPD (50 μM) reduced the amplitude of theAMPA receptor-mediated EPSCs reversibly (Fig. 2A1). We alsoapplied (RS)-3,5-dihydroxyphenylglycine (DHPG), the agonistfor the group I mGluRs at 100 μM, and L-SOP at 100 μM. Neitherof these drugs had an effect on the AMPA receptor-mediatedEPSCs (Figs. 2A2 and A3). These results suggest that the mGluRII is likely involved in the regulation of synaptic transmissionmediated by the AMPA receptors. To confirm further whetherthe AMPA receptor-mediated responses were modulated bythe group II mGluRs, we applied (1R,4R,5S,6R)-4-amino-2-oxabicyclo[3.1.0] hexane-4,6-dicarboxylic acid (LY379268), themGluR II agonist, at 20 nM. This drug substantially reduced theamplitude of the AMPA receptor-mediated responses (Fig. 2B,middle trace). The effect of LY379268 was dose-dependent(n=10, Fig. 2E). The concentration range that produced 50%suppression of the EPSCs was 10–20 nM. After application ofLY379268 (20 nM) for 10 min, the AMPA receptor-mediatedresponsewas reduced to 54.2±4.1% and returned to 80.6±9.1%of the control after drug washout for 20 min (n=7, Fig. 2F).For some neurons (n=7), we added (2S)-2-amino-2-[(1S,2S)-2-carboxycycloprop-1-yl]-3-(xanth-9-yl) propanoicacid (LY341495)(10 nM), the mGluR II antagonist, to the LY379268 solution. Theinhibitory effect produced by LY379268 was reversed byLY341495 (Fig. 2B, bottom trace). LY341495 (100 nM) alone didnot have any significant effect on AMPA receptor-mediatedEPSCs of the ICC neurons (n=7, p>0.05). Three neurons whoseAMPA receptor-mediated responses were reduced by LY379268were labeled by Lucifer Yellow. One was a bipolar cell (data notshown) and the other two were multipolar cells (Fig. 2D).

A summary of the effects of the mGluR I, II, and III agonistsandmGluR II antagonist on the AMPA receptor-mediated EPSCsis shown in Fig. 2C. ACPD (50 μM) significantly decreased theamplitude of the EPSCs to 60.8±7.4% of the control level (n=12,p<0.001); the EPSCs returned to 99.1±5.7% of the control levelafter drug washout (p>0.05). LY379268 (20 nM) reduced theAMPA receptor-mediated EPSCs to 45.3±10.5% of the controllevel (n=7, p<0.001); the EPSCs returned to 97.8±2.4% of thecontrol levelwith the addition of LY341495 (10 nM) (p>0.05). TheEPSCswerenotnotablyaffectedbyDHPG (100 μM) (n=8) or L-SOP(100 μM) (n=7) (DHPG: 98.7±5.4%of the control, L-SOP: 96.5±4.7%of the control) (p>0.05). All of the above results suggest that theAMPA receptor-mediated responses can be modulated byactivation of mGluR II but not groups I and III.

To test whether the reduction of the AMPA receptor-mediated EPSCs was due to the activation of the mGluR IIthrough a presynaptic mechanism, we applied paired-pulsestimulation to the LL with a 200ms interpulse interval. Incontrol solution, the paired-pulse stimulation induced twoEPSCs, and in most cases, the second EPSC had a smalleramplitude than the first one. Adding LY379268 (20 nM) to thecontrol solution decreased both EPSCs, with more reduction ofthe first one (Fig. 3A1). In 9/10 cells tested, the ratio of the secondto the first EPSCswas increasedbyLY379268compared to that inthe control solution (Fig. 3B, left panel). The average ratio of thesecond EPSC amplitude to the first one was significantly largerduring application of LY379268 (0.950±0.065) than that in thecontrol solution (0.796±0.037) (n=10, p<0.01) (Fig. 3C, left twocolumns). The reduction of the first EPSC was highly correlatedwith the increase of the paired-pulse response ratio (n=10,p<0.01, Fig. 3D). We also examined the effects of LY379268 onthe AMPA receptor-mediated responses with the slices in asolution with a low extracellular Ca2+ concentration (1.0 mM).TheAMPAreceptor-mediatedEPSCswere smaller in the1.0-mMCa2+ solution. The effect of LY379268 on the AMPA receptor-mediated EPSCs in the low extracellular Ca2+ solution wassimilar to that with the normal extracellular Ca2+ solution.LY379268 (20 nM) reduced the AMPA receptor-mediated EPSCsbut produced a greater suppression of the first one (Fig. 3A2). Forboth 1.0-mM and 2.4-mM Ca2+ solutions, the first responses inLY379268 were normalized to the first control responses, andthe EPSCs in the control and the agonist solutions weresuperimposed. The smaller reduction of the secondEPSCduringLY379268 (red traces) can be seen clearly (Fig. 3, bottom traces inA1 andA2). In 1.0-mMCa2+ solution, the ratio of the second to thefirst EPSCswas increased in8/10 cells tested (Fig. 3B, rightpanel).The average ratio of the second to the first EPSCs wassignificantly increased from 0.924±0.023 to 1.019±0.044 byLY379268 (n=10, p<0.05) (Fig. 3C, right two columns). Further-more, low extracellular Ca2+ had similar effects to thoseproduced by LY379268 on the ratio of second/first EPSCs. Theratio with the 1.0-mM Ca2+ concentration was significantlylarger than that with the neurons in the 2.4-mM Ca2+ solution(unpaired t-test, p<0.05) (Fig. 3C, two white columns).

2.3. Effects of activation of mGluR on synaptic responsesmediated by GABAA receptors

The IPSCs mediated by GABAA receptors in ICC neurons werefirst pharmacologically isolated by suppressing glycine andionotropic glutamate (AMPAandNMDA) receptors (see details inSection 4). Administration of ACPD (50 μM) substantially

Fig. 2 – Effects of the mGluR agonists and antagonist on AMPA receptor-mediated EPSCs. (A1,2,3) AMPA receptor-mediatedEPSCs before (top traces), during (middle traces), and after (bottom traces) ACPD, DHPG, and L-SOP, in three neurons,respectively. The effect of ACPD was reversible. (B) AMPA receptor-mediated EPSCs of a neuron before (top trace) and during(middle trace) application of LY379268. The response returned to the control level when LY341495 was added to the LY379268solution (bottom trace). (C) The effects of ACPD, DHPG, L-SOP, and LY379268 on the averaged amplitude of the AMPA receptor-mediated EPSCs (***p<0.001). (D) A neuron was labeled with Lucifer Yellow and classified as a multipolar cell. AMPA receptor-mediated EPSCs of this neuron were reduced by LY379268. *Location of the neuron in the ICC. Scale=20 μm. (E) LY379268 at 5,10, 20, 50, and 100 nMdecreased the EPSCs to 76.6%, 61.7%, 49.5%, 41.0%, and 38.4% of the control response, respectively. (F) Thechanges of the average amplitude of the AMPA receptor-mediated EPSCs as a function of time before, during, and afterapplication of LY379268.

32 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

decreased the IPSC (Fig. 4A1), but L-SOP (100 μM) hadno effect onthe IPSCs (Fig. 4A3). To examine further if the effect of ACPDwasdue to the activation of mGluR I or II, DHPG (100 μM) andLY379268 (20 nM) were applied, respectively. DHPG did not haveany effect on the GABAA receptor-mediated IPSCs (Fig. 4A2).However, applicationof LY379268greatly reduced theamplitudeof the IPSCs (Fig. 4B, middle trace). The effect of LY379268 on theGABAA receptor-mediated IPSCs was dose-dependent (n=8,Fig. 4E). The concentration of LY379268 that produced a 50%reduction in the IPSC was 10–20 nM. Application of LY379268(20 nM) for 10 min reduced the IPSCs to 51.8±5.7%; the IPSCreturned to 77.0±6.2% of the control after 20min of drug wash-out (n=6, Fig. 4F). For some neurons (n=7), we added LY341495(10 nM) to the LY379268 solution and the IPSCs returned to the

control level (Fig. 4B, bottom trace). LY341495 (100 nM) alone didnot have any significant effect on the GABAA receptor-mediatedIPSCs of the ICC neurons (n=6, p>0.05). Three neurons werelabeled with Lucifer Yellow. One of these cells was bipolar (datanot shown), and theother twoweremultipolarcells (Fig. 4D). TheIPSCs were reduced by LY379268 in each of these three cells.

A summary of the effects of the mGluR I, II, and III agonistsand mGluR II antagonist on the GABAA receptor-mediatedIPSCs is shown in Fig. 4C. ACPD (50 μM) significantly decreasedthe IPSC amplitude to 37.1±3.9% of the control level (n=10,p<0.001). The IPSC returned to 102.5±6.6% of the control afterdrug washout (p>0.05). LY379268 (20 nM) also significantlyreduced the IPSC amplitude to 41.0±6.3% of the control (n=7,p<0.001). The response returned to 97.2±2.3% of the control

Fig. 3 – Effects of LY379268 on AMPA receptor-mediated EPSCs elicited by paired-pulse stimulation of the LL. (A1) With 2.4 mMextracellular Ca2+, AMPA receptor-mediated EPSCs were evoked by paired-stimulus pulses before (top traces) and duringLY379268 (middle traces) in a neuron. Bottom traces: the EPSCs shown in the middle trace were normalized (red) andsuperimposedwith the control EPSCs (black). (A2)With 1.0 mMextracellular Ca2+, AMPA receptor-mediated EPSCswere evokedby paired-stimulus pulses before (top traces) and during LY379268 in a neuron (middle traces). Bottom traces: the EPSCs shownin the middle trace were normalized (red) and superimposed with the control EPSCs (black). (B) The ratios of the second to thefirst EPSCs in control and LY379268 solutions with 2.4 mM (left panel) and 1.0 mM extracellular Ca2+ (right panel). Two dotsconnecting a line represent one neuron. (C) Comparisons of the ratios of the second to the first EPSCs with extracellular Ca2+ of2.4 mM and 1.0 mM, and before and after application of LY379268 with either Ca2+ concentration (**p<0.01; *p<0.05).(D) Correlation of the reduction in the amplitude of the first EPSC to the increase in the ratio of the second to the first EPSCsduring application of LY379268 with 2.4 mM extracellular Ca2+ (p<0.01).

33B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

with the addition of LY341495 (10 nM) to the LY379268 solution(p>0.05). However, there was no significant difference be-tween the average IPSC amplitude before and after applica-tion of DHPG (100 μM) (101.2±2.8% of control) (n=9, p>0.05) orL-SOP (100 μM) (98.0±5.5% of control) (n=10, p>0.05).

Paired-pulse stimulation of the LL was used to investigatethe site of the mGluR II action on the GABAA receptor-mediated IPSCs (n=10). For all cells tested, the second IPSCwas always smaller than the first one. LY379268 (20 nM)reduced both IPSCs but had a greater effect on the first one(Fig. 5A1). In 8/10 cells tested, the ratio of the second to the firstresponses was increased in LY379268 compared to that in thecontrol solution (Fig. 5B, left panel). The average ratio of the

second IPSC amplitude to the first one was significantlyincreased from 0.686±0.022 in the control solution to 0.923±0.095 in the LY379268 solution (n=10, p<0.01) (Fig. 5C, left twocolumns). The amplitude change in the first IPSC correlatedwell with the change in the ratio of the second/first IPSCs(n=10, p<0.001, Fig. 5D). We also investigated effects of themGluR II agonist on the GABAA receptor-mediated IPSCs withthe slices in a solution with a low extracellular Ca2+ con-centration (1.0 mM). Similar to the effect of LY379268 on theGABAA receptor-mediated IPSCs in the 2.4-mM Ca2+ solution,LY379268 reduced both IPSCs but suppressed the first IPSCmore than the second one (Fig. 5A2). For both 2.4-mM and 1.0-mM Ca2+ solutions, the first IPSCs during LY379268 were

Fig. 4 – Effects of the mGluR agonists and antagonist on GABAA receptor-mediated IPSCs. (A1,2,3) GABAA receptor-mediatedIPSCs before (top traces), during (middle traces), and after (bottom traces) ACPD, DHPG, and L-SOP, in three neurons,respectively. The effect of ACPD was reversible. (B) GABAA receptor-mediated IPSCs of a neuron before (top trace) and during(middle trace) application of LY379268. The response returned to the control level when LY341495 was added to theLY379268 solution (bottom trace). (C) The effects of ACPD, DHPG, L-SOP and LY379268 on the average amplitude of the GABAA

receptor-mediated IPSCs (***p<0.001). (D) A neuron was labeled with Lucifer Yellow and classified as a multipolar cell. GABAA

receptor-mediated IPSCs of this neuron were reduced by LY379268. * Location of the neuron in the ICC. Scale=20 μm. (E) TheIPSCs were reduced to 86.7%, 54.9%, 37.6%, 30.1%, and 30.4% of the control level at 5, 10, 20, 50, and 100 nM LY379268,respectively (n=8). (F) The changes of the average amplitude of the GABAA receptor-mediated IPSCs as a function of time before,during, and after application of LY379268.

34 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

normalized to the first control responses and then the IPSCsin the control and agonist solutions were superimposed. Asmaller reduction in the second EPSC by LY379268 (red traces)is clearly displayed (Fig. 5, bottom traces in A1 and A2). With1.0 mM extracellular Ca2+, the ratio of the second to the firstIPSCs was increased by LY379268 in 8/10 cells (Fig. 5B, rightpanel). The average ratio of the second to the first IPSCs wasincreased from 0.891±0.022 to 0.981±0.035 (n=10, p<0.05)(Fig. 5C, right two columns). Low extracellular Ca2+ also hadsimilar effects on the ratio of the second/first IPSCs to thoseobtained with LY379268. The ratio was greatly increased withthe neurons in the 1.0-mM Ca2+ solution compared to that inthe 2.4-mM Ca2+ solution (unpaired t-test, p<0.01) (Fig. 5C, twowhite columns).

3. Discussion

We found that the mGluR agonists, ACPD (for mGluR I and II)andLY379268 (formGluR II), significantly reduced theEPSCsandIPSCs evoked by stimulation of the LL. ThemGluR II antagonist,LY341495, reversed the depressant action of LY379268. Theseresults indicate that the fast excitatory glutamatergic andinhibitory GABAergic transmission in the ICC can bemodulatedby thegroup IImGluRs.ACPDproducedcomparable reduction inthe amplitude of EPSCs and IPSCs to that found with LY379268.Furthermore, the group I agonist, DHPG, had no effect on theEPSCs and IPSCs. Thus, the depression produced by ACPDappeared to be a result of activation of the group II not group I

Fig. 5 – Effects of LY379268 on GABAA receptor-mediated IPSCs elicited by paired-pulse stimulation of the LL. (A1) With 2.4 mMextracellular Ca2+, GABAergic IPSCs were evoked by paired-pulse stimulation before (top traces) and during LY379268 (middletraces) in aneuron. Bottom traces: the IPSCs shown in themiddle tracewerenormalized (red) and superimposedwith the controlIPSCs (black). (A2) With 1.0 mM extracellular Ca2+, GABAergic IPSCs were evoked by paired-pulse stimulation before (top traces)and during LY379268 (middle traces) in a neuron. Bottom traces: the IPSCs shown in themiddle tracewere normalized (red) andsuperimposedwith the control IPSCs (black). (B) The ratios of the second to the first IPSCs in control and LY379268 solutionswith2.4 mM and 1.0 mM extracellular Ca2+. Two dots connecting a line represent one neuron. (C) Comparisons of the ratios of thesecond to the first EPSCs with extracellular Ca2+ of 2.4 mM and 1.0 mM, and before and after application of LY379268 in eitherCa2+ concentration (**p<0.01; * p<0.05). (D) Correlationof the amplitude reduction in the first IPSC to the increase in the ratio of thesecond to the first IPSCs during application of LY379268 with 2.4 mM extracellular Ca2+ (p<0.001).

35B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

mGluRs.We chose the concentration of 100 μMforDHPG,whichis at least 2–5 times its EC50 (the concentration of an agonist thatprovokes a response halfway between the baseline andmaximumresponse) for group ImGluRs (Cartmell and Schoepp,2000). The concentration of 100 μM for L-SOP is high enough forthe activation of most receptor types in the group III mGluRs(Cartmell and Schoepp, 2000). At these concentrations, neitherof these drugs had an effect on the EPSCs and IPSCs of the ICCneurons. Therefore, we conclude that group I and III mGluRsplay little or no role in themodulation of synaptic transmissionin the ICC. In contrast, LY379268 reduced the EPSCs and IPSCs ofthe ICC neurons in a dose-dependent manner. The suppressiveeffect of LY379268 was reversed by LY341495 at 10 nM, aconcentration at which only the mGluR II, not I and III, recep-

tors are blocked (Johnson et al., 1999). These results clearlydemonstrate the modulatory role of the group II mGluRs insynaptic transmission in the ICC. Similar suppressive effects onthe glutamatergic and/or GABAergic responses by activation ofthegroup IImGluRshavebeenobserved invarious regionsof theCNS (striatum: Pisani et al., 2002; hippocampus: Price et al., 2005;thalamus: Alexander and Godwin, 2005, 2006; superior collicu-lus: Neal and Salt, 2006; amygdala: Muly et al., 2007).

Immunoreactivity and mRNA for all three groups of themGluRs have been detected in the IC. However, the expressionof the different groups and different types of mGluRs withineach group varies in the IC. There is also a differentialdistribution of each type of mGluRs in different regions ofthe IC. A strong expression of mGluR5 was observed only in

36 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

the shell region of the IC, perhaps the dorsal and externalnuclei of the IC (Shigemoto et al., 1993; Romano et al., 1995;Ferraguti and Shigemoto, 2006). There was scattered distribu-tion of themRNA for mGluR1 in the IC (Shigemoto et al., 1992).Our results did not show any effect of the agonist of the group ImGluRs (mGluR 1 and 5) on membrane properties andsynaptic responses of ICC neurons, which is consistent withthe low labeling of the group I mGluR in the ICC.

Low to moderate immunolabeling and [3H] LY341495binding for mGluR2/3 were found in the IC (Petralia et al.,1996; Wright et al., 2001). The moderate labeling of themGluR2/3 may reflect presynaptic group II mGluRs onglutamatergic and GABAergic terminals that we found inthis study. However, several immunohistochemical and in situhybridization studies detected only mGluR3 and found nomGluR2 in the IC (Ohishi et al., 1993a,b, 1998; Tamaru et al.,2001). In the present study, we could not distinguish whichtype of the mGluRs in group II influenced the synapticresponses. More specific agonist or antagonists are requiredto determine the type of group II mGluRs that might beinvolved in synaptic regulation in the IC. Among the group IIImGluRs, distribution of mGluR7 is the most extensive in thecentral nervous system (Ferraguti and Shigemoto, 2006).Kinzie et al. (1995) detected low levels of mGluR7 mRNA inthe IC. However, immunoreactivity for mGluR7a andmGluR7bin the IC was negative (Kinoshita et al., 1998). Our results didnot detect any notable effect of the agonist of the group IIImGluRs on the membrane properties or the synapticresponses of the ICC neurons, suggesting little or no physio-logical role of the mGluRs III in the ICC.

The present study supports the presynaptic origin of thegroup II mGluR mediated response inhibition. In our data,there was no indication of direct effects of the group II mGluRson the postsynaptic membrane; viz., ACPD did not affect anyintrinsic membrane properties of the ICC neurons. Weobserved paired-pulse depression (PPD) for the AMPA recep-tor-mediated EPSCs and GABAA receptor-mediated IPSCs inthe ICC neurons. Lowering extracellular Ca2+ concentration,which decreases Ca2+ influx to presynaptic terminals andreduces transmitter release, caused reduction of the ampli-tudes of the EPSCs and IPSCs, and significantly reduced theextent of the PPD measured as the ratio of the second to thefirst response amplitudes. The results of the high correlationbetween the degree of the PPD reduction and the decreasein the amplitude of the first response provide additionalevidence to support the presynaptic Ca2+ regulation in thePPD. Similar to lowering Ca2+ concentration, the mGluR IIagonist, LY379268, reduced the degree of the PPD in both2.4 mM and 1.0 mM extracellular Ca2+, indicating that activa-tion of the mGluR II may decrease Ca2+ influx and then alterthe process that transiently inhibits synaptic vesicle release inresponse to the second presynaptic action potential (Belling-ham and Walmsley, 1999).

Postsynaptic desensitization may also be a possiblemechanism for the PPD (Jones and Westbrook, 1996). AMPAreceptor desensitization has been suggested to play a role insynaptic depression in auditory neurons (Otis et al., 1996).However, this type of postsynaptic modification unlikelyaccounted for the PPD in this study. Trussell et al. (1993)reported that the AMPA receptor desensitization in auditory

neurons of chick nucleus magnocellularis was almost fullyrecovered 60 ms after exposure to glutamate of 3 mM for 5 ms.We used a 200-ms interpulse interval to induce PPD, and ourstimulus pulse was very brief (0.1 ms). With these stimulusparameters, the possibility that smaller EPSCs to the secondpulse was due to desensitization of AMPA receptors is verylow. Recovery from desensitization of GABAA receptorsappeared slower than AMPA receptors. In IC neurons afterexposure to GABA at 1 mM for 1 ms, the response took a fewseconds to recover completely from desensitization of theGABAA receptors (Kraushaar and Backus, 2002). However,Pugh and Raman (2005) suggested that if GABA concentrationis at 100 μM or less, trains of synaptic currents may not showovert evidence of a desensitization-based reduction in re-sponse amplitude. Indeed, in our previous study, we applied10 electrical pulses (0.1 ms for each pulse) to elicit GABAA

receptor-mediated IPSCs in ICC neurons and found only aslight reduction in the IPSCs at 5 Hz and temporal summationand facilitation at higher frequencies (Wu et al., 2004). Thetype of modification of the GABAergic IPSCs in our data wasvery similar to that of the changes in the responses induced byrepetitive administration of GABA of 30 μM at the samefrequency range in Pugh and Raman's study (2005). Thus,each of our electrical pulses probably induced a release ofGABA at a concentration of about 30 μM at the synaptic cleft.Therefore, it is unlikely that the PPD of the GABAA receptor-mediated IPSCs in this study was attributed to desensitizationof the GABAA receptors. Comparable decreases in the PPD ofthe AMPA and GABAA receptor-mediated responses by themGluR II agonist further support the idea that activation of thegroup II mGluRs affects the AMPA and GABAA receptor-mediated responses through the samemechanism, i.e., actingpresynaptically and suppressing glutamate and GABA release.

Our results also indicate the presence of functional group IImGluRs in glutamatergic as well as in GABAergic terminals inboth bipolar and multipolar ICC neurons. There was nocorrespondence between the intrinsic membrane properties ormGluR-mediated changes and the morphology of the cellfrom which recordings were made. We did not analyze furtherthe cell morphology in three-dimensional axes because ourrecordings were made from neurons near the surface of a thinslice and part of their dendritic tree in the Z-axis (rostral tocaudal) might have been sectioned. Therefore, it is difficult toconclude that our bipolar and multipolar cells correspondto “flat” or “nonflat” cells that are classified according to theorientation and thickness of the dendritic tree (Malmierca et al.,1993; Oliver andMorest, 1984; Peruzzi et al., 2000). Nevertheless,it is clear that our patch clamp electrodes sampled the mostprevalent cell types and that the modulation of the EPSCs andIPSCs by group II mGluRs was present in all of them.

It appears that, in the ICC, the physiological role of themGluRs is very similar to that of themetabotropic type of GABAreceptors, i.e., GABAB receptors (Ma et al., 2002; Sun et al., 2006)(Fig. 6). Our previous studies suggest that GABAB receptors exerttheir action presynaptically in the ICC. Activation of the GABAB

receptors in the ICC neurons suppresses the EPSCsmediated byAMPA and NMDA receptors, and the IPSCs mediated by GABAA

receptors, but does not affect the neurons’ intrinsic membraneproperties. The fact that both excitatory and inhibitory synaptictransmission can be modulated by either GABAB receptors or

Fig. 6 – Schematic view of the sites and action of presynapticmGluRs II and GABAB receptors in the ICC. ●: GABAmolecules. o: glutamate molecules. The location of GABAB

receptors on glutamatergic and GABAergic terminals wassuggested by our previous studies (Ma et al., 2002; Sun et al.,2006).

37B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

mGluRs in the ICC indicates that both excitatory and inhibitorysynaptic transmissions in the ICC may require multiplemechanisms to balance the level of excitation and inhibition(Fig. 6). Interestingly, in avian cochlear nucleus magnocellularneurons, glutamatergic transmission is not subject to modula-tion by mGluRs but is regulated by GABAB receptors, whereasGABAergic transmission is regulated bymGluRs (Brenowitz andTrussell, 2001; Brenowitz et al., 1998; Lu, 2007). The significanceof differential distribution of functional mGluRs and GABAB

receptors indifferentneurons (or indifferent regionsordifferentspecies) is not well understood. Nevertheless, in the ICC, twotypes of metabotropic receptors play similar roles in synapticmodulation.

Complex modulatory processes by metabotropic receptorsin the ICC may take place in the ICC in vivo. PresynapticmGluRs and GABAB receptors may work synergically on eitherglutamatergic or GABAergic terminals independently. Alter-natively, glutamatergic excitation and GABAergic inhibitionmay be modulated at the same time in different proportionsby mGluRs and GABAB receptors in the ICC. Further in vivoelectrophysiological investigation with application of mGluRand GABAB agonists and antagonists is required to elucidatehowmGluRs and GABAB receptors contribute to the balance ofglutamatergic excitation and GABAergic inhibition in proces-sing acoustical signals in the ICC.

4. Experimental procedures

4.1. Preparation of brain slices

The methods used to prepare brain slices were in agreementwith the guideline of Canadian Council on Animal Care, andwere approved by the Carleton University Animal Care Com-mittee. Long Evans rats (Charles River, St. Constant, Quebec,Canada) aging 8–17 days old were anaesthetized with isoflur-ane. The animals were then euthanized by decapitation, and

their brains were rapidly taken out from their skulls. The brainswere dissected in artificial cerebrospinal fluid (ACSF) beingbubbled with 95% O2 and 5% CO2. The ASCF consisted of CaCl2(2.4 mM), glucose (10 mM), HEPES (3mM), KCl (3 mM), KH2PO4

(1.2 mM), MgSO4 (1.3 mM), NaCl (129mM), and NaHCO3 (20 mM)in deionized water. The pH of the ACSF was adjusted to 7.4.When theCa2+ concentrationwas lowered from2.4 to 1.0 mM inACSF, 1.4 mM MgCl2 was added to the ACSF. Coronal sectionsthrough the IC, tilted 15°–20° toward the caudal side, weremadeat 200-μmthickness using a tissue slicer (Campden, UK) in roomtemperature. Slices havingboth the ICC and LLwere chosen andtransferred to a recording chamber that was continu-ously perfused by a warm ACSF of 30±1 °C at a flow rate of 10–12 ml/min.

4.2. Whole-cell patch clamp recording

The ICC neurons were visualized with an upright lightmicroscope equipped with a 40× water immersion objectivelens and differential interference contrast (Zeiss, Axioskop 2, FSplus,Germany). Patchelectrodesweremade fromglasscapillarytubes (1.1 mmODand0.8 mmID;Kimble,Mexico)withaverticalpuller (PP830,Narishige,Tokyo, Japan). Thepatchelectrodeshada resistance of 4–6 MΩ. The electrodes were filled with internalsolution containing Kgluconate (130mM), MgCl2 (2 mM), KCl(5 mM), GTP (0.3 mM), ATP (2mM), EGTA (1.6 mM), and HEPES(10 mM). The solutionwas filteredwith a 0.2-µm filter andhad apHof7.2. For recording synaptic responses,QX-314 (0.5 mM)wasadded to the internal solution.

Whole-cell patch clamp recordings were made with anAxopatch 200A or 200B amplifier (Axon Instruments, USA)from the ICC neurons. Signals were filtered at 5 kHz by theamplifier and digitized at 10 kHz by a Digidata 1320A interface(Axon Instruments, USA). The series resistance (10–25 MΩ) wascompensated by 75% or more. Intrinsic membrane propertiesof neurons were examined by injecting current pulses of −200to 200 pA in 10-pA steps with a duration of 200 ms.

AMPA receptor-mediated EPSCs and GABAA receptor-me-diated IPSCs were recorded at a holding potential of −60 and−40 mV, respectively. An insulated tungsten bipolar electrodewas placed in the LL just below the ICC. The LL was stimulatedby an electrical square pulse of 0.1 ms with an intensitybetween 5 and 30 V. The intensity was set at a level that couldevoke two-thirds of a maximal response. The synaptic re-sponseswere recorded every 20–100 s, and the average of threeconsecutive responses was used for further analysis.

4.3. Drugs

To isolate AMPA receptor-mediated responses, strychnine(0.5 μM), bicuculline (10 μM), and L-2-amino-5-phosphonovalericacid (L-AP5, 100 μM)were added to theACSF to suppress glycine,GABAA, and NMDA receptors, respectively. GABAA receptor-mediated responses were pharmacologically isolated by addingstrychnine (0.5 μM) and kynurenic acid (5 mM) or a combinationof L-AP5 (100 μM) and 6-cyano-7-nitroquinoxaline-2,3-dione(CNQX, 10 μM) to the ACSF to suppress glycine and glutamatereceptors. All the mGluR agonists and antagonists includingACPD,DHPG, LY379268, L-SOP, and LY341495were administeredby bath application. Strychnine, bicuculline, L-AP5, CNQX, and

38 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

kynurenic acid were purchased from Ascent Scientific, UnitedKingdom. ACPD, L-SOP, LY379268, and LY341495 were obtainedfrom Tocris, USA. DHPGwere purchased fromWako Chemicals,USA.

4.4. Histology

The slices with cells injected with Lucifer Yellow (2 mM) werefixed in 4% paraformaldehyde overnight and subsequentlywashed in 0.1 M PBS. The cells were then visualized with aPerkin Elmer Ultraview Vox spinning disk confocal system(PerkinElmer Inc., Waltham, MA) using 20× and 40× objectives.A Z-stack of confocal images at 0.3- to 0.5-μm intervals wascollected, and a two-dimensional reconstruction of the neuronwas generated with Volocity software 5.2.0 (Improvision Ltd.,Waltham, MA), and the image was then photographed.

4.5. Data analysis

Changes in membrane potential elicited by current injectionwere obtained by measuring voltages between the baselinemembrane potential and the peak hyperpolarization ordepolarization. A plot of current–voltage (I–V) relationshipwas made using values of the membrane potentials and thecorresponding injected current levels. The slope of the I–Vcurve in the range of current injections from 0 to −100 pA wasused to determine the input resistance of a neuron. The am-plitude of a synaptic current was measured from the peak ofthe response to the baseline established before the stimulusartifact. Numerical averages are presented as means±SEMs.Paired and unpaired-t tests were performed to comparedifference between means. In every case where comparisonswere made, the data were normally distributed. A minimumcriterion of p<0.05 was set for statistical significance.

Acknowledgments

We thank Dr. J.B. Kelly for a critical reading of the manuscriptand many helpful suggestions. We thank Dr. H. Sun foranalysis of Lucifer-Yellow labeled neurons. This work wassupported by research grant from Natural Sciences andEngineering Council of Canada (NSERC) to S.H. Wu.

R E F E R E N C E S

Alexander, G.M., Godwin, D.W., 2005. Presynaptic inhibition ofcorticothalamic feedback by metabotropic glutamatereceptors. J. Neurophysiol. 94, 163–175.

Alexander, G.M., Godwin, D.W., 2006. Unique presynaptic andpostsynaptic roles of group II metabotropic glutamatereceptors in the modulation of thalamic network activity.Neruoscience 141, 501–513.

Anwyl, R., 1999. Metabotropic glutamate receptors:electrophysiological properties and role in plasticity. Brain Res.Rev. 29, 83–120.

Bandrowski, A.E., Aramakis, V.B., Moore, S.L., Ashe, J.H., 2001.Metabotropic glutamate receptors modify ionotropicglutamate responses in neocortical pyramidal cells andinterneurons. Exp. Brain Res. 136, 25–40.

Bandrowski, A.E., Moore, S.L., Ashe, J.H., 2002. Activation ofmetabotropic glutamate receptors by repetitive stimulation inauditory cortex. Synapse 44, 146–157.

Bellingham, M.C., Walmsley, B., 1999. A novel presynapticinhibitory mechanim underlies paired pulse depression at afast central synapse. Neuron 23, 159–170.

Barnes-Davies, M., Forsythe, I.D., 1995. Pre- and postsynapticglutamate receptors at a giant excitatory synapse in ratauditory brainstem slices. J. Physiol. 488, 387–406.

Brenowitz, S., Trussell, L.O., 2001. Minimizing synaptic depressionby control of release probability. J. Neurosci. 21, 1857–1867.

Brenowitz, S., David, J., Trussell, L., 1998. Enhancement of synapticefficacy by presynaptic GABA(B) receptors. Neuron 20, 135–141.

Cartmell, J., Schoepp, D.D., 2000. Regulation of neurotransmitterrelease bymetabotropic glutamate receptors. J. Neurochem. 75,889–907.

Conn, P.J., Pin, J.P., 1997. Pharmacology and functions ofmetabotropic glutamate receptors. Annu. Rev. Pharmacol.Toxicol. 37, 205–237.

D'Angelo, W.R., Sterbing, S.J., Ostapoff, E.M., Kuwada, S., 2005. Therole of GABA-ergic inhibition in the coding of interaural timedifference of low-frequency sounds in the inferior colliculus.J. Neurophysiol. 93, 3390–3400.

Faingold, C.L., Gehlbach, G., Caspary, D.M., 1991. Functionalpharmacology of inferior colliculus neurons. In: Altschuler, R.A.,Bobbin, R.P., Clopton, B.M., Hoffman, D.W. (Eds.), Neurobiology ofHearing: The Central Auditory System. Raven Press, New York,pp. 223–251.

Ferraguti, F., Shigemoto, R., 2006. Metabotropic glutamatereceptors. Cell Tissue Res. 326, 483–504.

Fuzessery, Z.M., Hall, J.C., 1996. Role of GABA in shaping frequencytuning and creating FM sweep selectivity in the inferiorcolliculus. J. Neurophysiol. 76, 1059–1073.

Ingham, N.J., McAlpine, D., 2005. GABAergic inhibition controlsneural gain in inferior colliculus neurons sensitive tointeraural time differences. J. Neurosci. 25, 6187–6198.

Johnson, B.G., Wright, R.A., Arnold, M.B., Wheeler, W.J., Ornstein,P.L., Schoepp, D.D., 1999. [3H]-LY341495 as a novel antagonistradioligand for group II metabotropic glutamate (mGlu)receptors: characterization of binding to membranes of mGlureceptor subtype expressing cells. Neuropharmacology 38,1519–1529.

Jones, M.V., Westbrook, G.L., 1996. The impact of receptordesensitization on fast synaptic transmission.Trends Neurosci. 19, 96–101.

Kelly, J.B., Caspary, D.M., 2005. Pharmacology of the inferiorcolliculus. In: Winer, J.A., Schreiner, C.E. (Eds.), The InferiorColliculus. Springer, New York, pp. 248–281.

Kew, J.N., Kemp, J.A., 2005. Ionotropic andmetabotropic glutamatereceptor structure and pharmacology. Psychopharmacology179, 4–29.

Kinoshita, A., Shigemoto, R., Ohishi, H., van der Putten, H., Mizuno,N., 1998. Immunohistochemical localization of metabotropicglutamate receptors, mGluR7a and mGluR7b, in the centralnervous system of the adult rat andmouse: a light and electronmicroscopic study. J. Comp. Neurol. 393, 332–352.

Kinzie, J.M., Saugstad, J.A., Westbrook, G.L., Segerson, T.P., 1995.Distribution of metabotropic glutamate receptor 7 messengerRNA in the developing and adult rat brain. Neuroscience 69,167–176.

Klug, A., Park, T.J., Pollak, G.D., 1995. Glycine and GABA influencebinaural processing in the inferior colliculus of the mustachebat. J. Neurophysiol. 74, 1701–1713.

Koch, U., Grothe, B., 1998. GABAergic and glycinergic inhibitionsharpens tuning for frequency modulations in the inferiorcolliculus of the big brown bat. J. Neurophysiol. 80, 71–82.

Kotak, V.C., Sanes, D.H., 1995. Synaptically evoked prolongeddepolarizations in the developing auditory system.J. Neurophysiol. 74, 1611–1620.

39B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

Kraushaar, U., Backus, K.H., 2002. Characterization of GABAA andglycine receptors in neurons of the developing rat inferiorcolliculus. Eur. J. Physiol. 445, 279–288.

Kudoh, M., Sakai, M., Shibuki, K., 2002. Differential dependence ofLTD on glutamate receptors in the auditory cortical synapsesof cortical and thalamic inputs. J. Neurophysiol. 88,3167–3174.

Le Beau, F.E.N., Rees, A., Malmierca, M.S., 1996. Contribution ofGABA- and glycine-mediated inhibition to the monauraltemporal response properties of neurons in the inferiorcolliculus. J. Neurophysiol. 75, 902–919.

Le Beau, F.E.N., Malmierca, M.S., Rees, A., 2001. Iontophoresis invivo demonstrates a key role for GABA(A) and glycinergicinhibition in shaping frequency response areas in the inferiorcolliculus of guinea pig. J. Neurosci. 21, 7303–7312.

Lu, Y., 2007. Endogenous mGluR activity suppresses GABAergictransmission in avian cochlear nucleus magnocellularisneurons. J. Neurophysiol. 97, 1018–1029.

Lu, Y., Rubel, E.W., 2005. Activation of metabotropic glutamatereceptors inhibits high-voltage-gated calcium channelcurrents of chicken nucleus magnocellularis neurons.J. Neurophysiol. 93, 1418–1428.

Ma, C.L., Kelly, J.B., Wu, S.H., 2002. Presynaptic modulation ofGABAergic inhibition by GABAB receptors in the rat's inferiorcolliculus. Neuroscience 114, 207–215.

Malmierca, M.S., Blackstad, T.W., Osen, K.K., Karagülle, T.,Molowny, R.L., 1993. The central nucleus of the inferiorcolliculus in rat: a Golgi and computer reconstruction study ofneuronal and laminar stricture. J. Comp. Neurol. 333, 1–27.

Muly, E.C., Mania, I., Guo, J.D., Rainnie, D.G., 2007. Group IImetabotropic glutamate receptors in anxiety circuitry:correspondence of physiological response and subcellulardistribution. J. Comp. Neurol. 505, 682–700.

Nataraj, K., Wenstrup, J.J., 2005. Roles of inhibition in creatingcomplex auditory responses in the inferior colliculus:facilitated combination-sensitive neurons. J. Neurophysiol. 93,3294–3312.

Nataraj, K., Wenstrup, J.J., 2006. Roles of inhibition in complexauditory responses in the inferior colliculus: inhibitedcombination-sensitive neurons. J. Neurophysiol. 95,2179–2192.

Neal, S.A., Salt, T.E., 2006. Modulation of GABAergic inhibition inthe rat superior colliculus by a presynaptic group IImetabotropic glutamate receptor. J. Phsyiol. 577, 659–669.

Nishimaki, T., Jang, I.-S., Ishibashi, H., Yamaguchi, J., Nabekura, J.,2007. Reduction of metabotropic glutamate receptor-mediatedheterosynaptic inhibition of developing MNTB-LSO inhibitorysynapses. Eur. J. Neurosci. 26, 323–330.

Ohishi, H., Shigemoto, R., Nakanishi, S., Mizuno, N., 1993a.Distribution of the messenger RNA for a metabotropicglutamate receptor, mGluR2, in the central nervous system ofthe rat. Neuroscience 53, 1009–1018.

Ohishi, H., Shigemoto, R., Shigetada, N., Mizuno, N., 1993b.Distribution of the mRNA for a metabotropic glutamatereceptor (mGluR3) in the rat brain: an in situ hybridizationstudy. J. Comp. Neurol. 335, 252–266.

Ohishi, H., Neki, A., Mizuno, N., 1998. Distribution of ametabotropic glutamate receptor, mGluR2, in the centralnervous system of the rat and mouse: animmunohistochemical study with a monoclonal antibody.Neurosci. Lett. 30, 65–82.

Oliver, D.L., Morest, D.K., 1984. The central nucleus of the inferiorcollculus in the rat. J. Comp. Neurol. 222, 237–264.

Otis, T., Zhang, S., Trussell, L.O., 1996. Direct measurement ofAMPA receptor desensitization induced by glutamatergicsynaptic transmission. J. Neurosci. 16, 7496–7504.

Palombi, P.S., Caspary, D.M., 1996. GABA inputs control dischargerate primarily within frequency receptive fields of inferiorcolliculus neurons. J. Neurophysiol. 75, 2211–2219.

Peruzzi, D., Sivaramakrishnan, S., Oliver, D.L., 2000. Identificationof cell types in brain slices of the inferior colliculus.Neuroscience 101, 403–416.

Petralia, R.S., Wang, Y.-X., Niedzielski, A.S., Wenthold, R.J., 1996.The metabotropic glutamate receptors, mGluR2 and mGluR3,show unique postsynaptic, presynaptic and glial localizations.Neuroscience 71, 949–976.

Pisani, A., Bonsi, P., Catania, M.V., Giuffrida, R., Morari, M., Marti,M., Centonze, D., Bernardi, G., Kingston, A.E., Calabresi, P., 2002.Metabotropic glutamate 2 receptors modulate synaptic inputsand calcium signals in striatal cholinergic interneurons.J. Neurosci. 22, 6176–6185.

Pollak, G.D., Burger, R.M., Park, T.J., Klug, A., Bauer, E.E., 2002. Rolesof inhibition for transforming binaural properties in thebrainstem auditory system. Hear. Res. 168, 60–78.

Portfors, C.V., Sinex, D.G., 2005. Coding of communication soundsin the inferior colliculus. In: Winer, J.A., Schreiner, C.E. (Eds.),The Inferior Colliculus. Springer, New York, pp. 411–425.

Portfors, C.V., Roberts, P.D., Jonson, K., 2009. Over-representationof species-specific vocalizations in the awake mouse inferiorcolliculus. Neuroscience 162, 486–500.

Price, C.J., Karayannis, T., Pál, B.Z., Capogna, M., 2005. Group II andIII mGluRs-mediated presynaptic inhibition of EPSCs recordedfrom hippocampal interneurons of CA1 stratum lacunosummoleculare. Neuropharmacology 49, 45–56.

Pugh, J.R., Raman, I.M., 2005. GABAA receptor kinetics in thecerebellar nuclei: evidence for detection of transmitter fromdistant release sites. Biophys. J. 88, 1740–1754.

Romano, C., Sesma, M.A., McDonald, C.T., O'Malley, K., van denPol, A.N., Olney, J.W., 1995. Distribution of metabotropicglutamate receptor mGluR5 immunoreactivity in rat brain.J. Comp. Neurol. 355, 455–469.

Sanes, D.H., McGee, J., Walsh, E.J., 1998. Metabotropic glutamatereceptor activation modulates sound level processing in thecochlear nucleus. J. Neurophysiol. 80, 209–217.

Shigemoto, R., Shigetada, N., Mizuno, N., 1992. Distribution ofthe mRNA for a metabotropic glutamate receptor (mGluR1)in the central nervous system: an in situ hybridizationstudy in adult and developing rat. J. Comp. Neurol. 322,121–135.

Shigemoto, R., Nomura, S., Ohishi, H., Sugihara, H., Nakanishi, S.,Mizuno, N., 1993. Immunohistochemical localization of ametabotropic glutamate receptor, mGluR5, in the rat brain.Neurosci. Lett. 163, 53–57.

Sun, H., Ma, C.L., Kelly, J.B., Wu, S.H., 2006. GABAB

receptor-mediated presynaptic inhibition of glutamatergictransmission in the inferior colliculus. Neurosci. Lett. 399,151–156.

Schwarz, D.W., Tennigkeit, F., Puil, E., 2000. Metabotropictransmitter action in auditory thalamus. Acta Otolaryngol. 120,251–254.

Takahashi, T., Forsythe, I.D., Tsujimoto, T., Barnes-Davies, M.,Onodera, K., 1996. Presynaptic calcium currentmodulation by ametabotropic glutamate receptor. Science 274, 594–597.

Tamaru, Y., Nomura, N., Mizuno, N., Shigemoto, R., 2001.Distribution ofmetabotropic glutamate receptormGluR3 in themouse CNS: differential location relative to pre- andpostsynaptic sites. Neuroscience 106, 481–503.

Trussell, L.O., Zhang, S., Raman, I.M., 1993. Desensitization ofAMPA receptors upon multiquantal neurotransmitter release.Neuron 10, 1185–1196.

Vater, M., Habbicht, H., Kössl, M., Grothe, B., 1992. The functionalrole of GABA and glycine in monaural and binaural processingin the inferior colliculus of horseshoe bats. J. Comp. Physiol.171, 541–553.

Voytenko, S., Galazyuk, A., 2008. Sound level processing in theinferior colliculus of awake bats is modulated by metabotropicglutamate receptors. Abstract of the 31st Annual MidwinterResearch Meeting. Phoenix, AZ, p. 116.

40 B R A I N R E S E A R C H 1 3 2 5 ( 2 0 1 0 ) 2 8 – 4 0

Wang, X., Jen, P.H., Wu, F.J., Chen, Q.C., 2007. Preceding weak noisesharpens the frequency tuning and elevates the responsethreshold of the mouse inferior collicular neurons throughGABAergic inhibition. Brain Res. 1167, 80–91.

Wright, R.A., Arnold, M.B., Wheeler, W.J., Ornstein, P.L., Schoepp,D.D., 2001. [3H]LY341495 binding to group II metabotropicglutamate receptors in rat brain. J. Pharmacol. Exp. Ther. 298,453–460.

Wu, S.H., Fu, X.W., 1998. Glutamate receptors underlyingexcitatory synaptic transmission in the rat's lateral superiorolive studied in vitro. Hear. Res. 122, 47–59.

Wu, C.H., Jen, P.H., 2006. The role of GABAergic inhibition inshaping duration selectivity of bat inferior collicular neuronsdetermined with temporally patterned sound trains. Hear. Res.215, 56–66.

Wu, S.H., Ma, C.L., Kelly, J.B., 2004. Contribution of AMPA, NMDA,and GABA(A) receptors to temporal pattern of postsynapticresponses in the inferior colliculus of the rat. J. Neurosci. 24,4625–4634.

Yin, S., Chen, Z., Feng, Y., Wang, J., 2008. The roles of localinhibition mediated by gamma-aminobutyric acid (GABA)-Areceptor in duration tuning in the inferior colliculus of guineapigs. Acta Otolaryngol. 31, 1–9.

Zhang, H., Kelly, J.B., 2001. AMPA and NMDA receptors regulateresponses of neurons in the rat's inferior colliculus.J. Neurophysiol. 86, 871–880.

Zhang, H., Kelly, J.B., 2003. Glutamatergic and GABAergicregulation of neural responses in inferiorcolliculus to amplitude-modulated sounds. J. Neurophysiol. 90,477–490.