Group II metabotropic glutamate receptors inhibit glutamate release at thalamocortical synapses in the developing somatosensory cortex

Abstract

Thalamocortical synapses provide a strong glutamatergic excitation to cortical neurons that is critical for processing sensory information. Unit recordings in vivo indicate that metabotropic glutamate receptors (mGluRs) reduce the effect of thalamocortical input on cortical circuits. However, it is not known whether this reduction is due to a reduction in glutamate release from thalamocortical terminals or from a decrease in cortical neuron excitability. To directly determine whether mGluRs act as autoreceptors on thalamocortical terminals, we examined the effect of mGluR agonists on thalamocortical synapses in slices. Thalamocortical excitatory postsynaptic currents (EPSCs) were recorded in layer IV cortical neurons in developing mouse brain slices. The activation of group II mGluRs with DCG IV reduced thalamocortical EPSCs in both excitatory and inhibitory neurons, while the stimulation of group I or group III mGluRs had no effect on thalamocortical EPSCs. Consistent with a reduction in glutamate release, DCG IV increased the paired pulse ratio and the coefficient of variation of the EPSCs. The reduction induced by DCG IV was reversed by the group II mGluR antagonist, LY341495, and mimicked by another selective group II agonist, APDC. The mGluR2 subtype appears to mediate the reduction of thalamocortical EPSCs, since the selective mGluR3 agonist, NAAG, had no effect on the EPSCs. Consistent with this, we showed that mGluR2 is expressed in the barrels. Furthermore, blocking group II mGluRs with LY341495 reduced the synaptic depression induced by a short stimulus train, indicating that synaptically released glutamate activates these receptors. These results indicate that group II mGluRs modulate thalamocortical processing by inhibiting glutamate release from thalamocortical synapses. This inhibition provides a feedback mechanism for preventing excessive excitation of cortical neurons that could play a role in the plasticity and refinement of thalamocortical connections during this early developmental period.

Introduction

Thalamic stimulation of cortical circuits is the initial step in the cortical processing of sensory information. Thalamocortical terminals excite cortical neurons via the release of glutamate which activates AMPA and NMDA receptors on the cortical neurons (Castro-Alamancos and Connors, 1997). To monitor and regulate the amount of glutamate released, many glutamatergic terminals express presynaptic metabotropic glutamate receptors (mGluRs), which are divided into three groups based on sequence homology, pharmacological profiles, and biochemical signaling (Conn and Pin, 1997, Anwyl, 1999). Group II and III mGluRs decrease the release of glutamate (Pisani et al., 1997, Poisik et al., 2005), while group I mGluRs either enhance (Reid et al., 1999, Schwartz and Alford, 2000) or reduce the release of glutamate (White et al., 2003).

Previous experiments suggest that mGluRs could be modulating thalamocortical inputs within the rodent somatosensory system. In rodents, fibers arising from the ventrobasal nucleus of the thalamus project in a well organized somatotopic pattern to specialized structures in layer IV known as barrels (Woolsey and Van der Loos, 1970). Each barrel receives a dense thalamic input conveying sensory information from a single whisker on the rodent snout. Immunohistochemical studies have demonstrated that group I and group II mGluRs are densely expressed in the barrel hollows in layer IV of the somatosensory cortex where the thalamocortical terminals are located (Ohishi et al., 1998, Muñoz et al., 1999). These studies indicate that mGluRs are expressed in the vicinity of thalamocortical synapses, but they did not determine whether the mGluRs were expressed on the thalamocortical terminals or on the cortical neurons. In addition, the stimulation of mGluRs in vivo reduces cortical activation induced by both somatosensory and visual stimuli (Cahusac, 1994, Taylor and Cahusac, 1994, Beaver et al., 1999), indicating that mGluRs modulate the cortical processing of sensory information. These experiments demonstrate that mGluRs decrease thalamocortical excitation; however, they did not determine whether the reduction is mediated by a presynaptic decrease in the release of glutamate from thalamocortical terminals or by a postsynaptic hyperpolarization of the cortical neurons.

To directly determine whether mGluRs inhibit glutamate release from thalamocortical synapses, we used the thalamocortical slice preparation (Agmon and Connors, 1991, Porter and Nieves, 2004) which provides better access to thalamocortical circuits.

Experimental Procedures

Thalamocortical slice preparation

Young mice from postnatal days 6 to 12 were obtained from our breeding facility at the Ponce School of Medicine. In compliance with the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals (Publication DHHS NIH 86-23), the Institutional Animal Care and Use Committee of the Ponce School of Medicine approved all procedures involving animals. The animals were anesthetized with halothane, decapitated, and the brains were removed and placed in ice-cold oxygenated artificial cerebrospinal fluid (ACSF) containing 126 mM NaCl₂, 3 mM KCl, 1.25 mM NaH₂PO₄, 2 mM CaCl₂, 1 mM MgSO₄, 26 mM NaHCO₃, and 20 mM glucose, bubbled with carbogen (95% O₂ and 5% CO₂). Brains slices of 300 to 400 μm thickness were prepared with a Vibratome 1000 Plus (Vibratome, St. Louis, MO) by standard procedures to preserve thalamocortical connectivity (Agmon and Connors, 1991, Porter et al., 2001). Slices containing both the ventrobasal thalamic nucleus and the barrel cortex were saved and incubated for one hour at room temperature in ACSF.

Electrophysiology

After the one hour incubation, individual slices were transferred to a submersion recording chamber mounted on a E600 upright microscope (Nikon Instruments, Melville, NY), and perfused with room temperature, carbogenated ACSF at a rate of 2-3 ml/min. To stimulate thalamocortical axons, we placed a bipolar tungsten microelectrode in the ventrobasal nucleus or internal capsule under brightfield illumination to view the anatomical landmarks of the thalamocortical pathway. Excitatory postsynaptic currents (EPSCs) were recorded with a patch-clamp amplifier (MultiClamp 700A, Axon Instruments, Union City, CA) using glass micropipettes that were pulled on a Flaming/Brown micropipette puller (Sutter Instruments, CA). The positions of the stimulating electrode and micropipettes were adjusted to maximize EPSC amplitude. Paired stimuli separated by 50 ms were given every 10-15 ms.

Single neurons were visually identified using a video camera (Dage MTI, Michigan City, IN) connected to the Nikon E600 microscope. All neurons included in this study were located in layer IV of the barrel cortex. Whole-cell current-clamp and voltage-clamp recordings were done with patch pipettes filled with an internal solution containing 12 mM KCl, 140 mM KGluconate, 0.2 mM EGTA, 10 mM HEPES, 0.3 mM GTP and 0.4 mM ATP (pH 7.3, 285 mOsm). Resting membrane potential was measured after achieving whole cell configuration and cells with a membrane potential more depolarized than -50 mV were discarded. To obtain an intrinsic firing pattern an initial hyperpolarizing square current step was given followed by depolarizing current steps in current-clamp mode. Recordings were filtered at 4 kHz, digitized at 10 kHz, and saved to a computer using pCLAMP9 software (Axon Instruments, Union City, CA). Membrane potentials were not corrected for the junction potential. Recordings were not compensated for series resistance, but changes in series resistance were continuously monitored and recordings were eliminated from analysis if the series resistance changed by more than 15%. As previously described (Agmon et al., 1996, Beierlein and Connors, 2002), evoked EPSCs which exhibited latencies that varied by less than one millisecond were considered monosynaptic thalamocortical EPSCs and were included in the analysis. The latencies of the evoked EPSCs showed an average latency from the beginning of the stimulation artifact of 6.2 ± 0.03 ms in slices from mice 6-12 postnatal days old (n = 44). These latencies are consistent with the latencies of thalamocortical EPSCs evoked by stimulation of either the ventrobasal nucleus or the internal capsule and measured in slices from similarly aged animals (Agmon and O’Dowd, 1992, Crair and Malenka, 1995, Agmon et al., 1996, Lu et al., 2001, Laurent et al., 2002). Since a previous study indicated that corticothalamic EPSCs exhibit paired pulse facilitation while thalamocortical EPSCs exhibit paired pulse depression (Beierlein and Connors, 2002), we only included inputs which displayed paired pulse depression in our analysis. To examine the effect of mGluR activation on EPSCs, we diluted the drugs in ACSF containing 100 μM DL-2-amino-5-phosphonopentanoic acid (AP5) and 10 μM bicuculline to block NMDA and GABA_A currents, respectively.

LY 341495, (2S,2′R,3′R)-2-(2′,3′-dicarboxycyclopropyl)glycine (DCG IV), 3,5-dihydroxyphenylglycine (DHPG), L-(+)-2-amino-4-phoshonobutyric acid (L-AP4), (2R, 4R)-4-aminopyrrolidine-2,4-dicarboxylic acid (APDC), and N-acetylaspartylglutamate (NAAG) were purchased from Tocris (Ellisville, MO). Bicuculline and AP5 were purchased from Sigma (St. Louis, MO).

Data analysis and statistics

Experiments were accepted for analysis if the effect of the mGluR agonist on the EPSCs reversed upon removal of the drug and there was no confounding di- or polysynaptic activity in the EPSCs. Data were analyzed using Clampfit (Axon Instruments, Union City, CA). Average EPSCs were taken from 10 consecutive traces including failures. The paired pulse ratio (PPR) was calculated from EPSCs evoked by paired stimuli as the mean amplitude of the EPSC evoked by the second stimulus (EPSC2) divided by the mean amplitude of the EPSC evoked by the first stimulus (EPSC1). The coefficient of variation (CV) was calculated as the standard deviation of the EPSC amplitude divided by the mean EPSC amplitude from 10 consecutive traces including failures. Data were analyzed using Student’s t-test or repeated measures ANOVA (Statistica, Statsoft, Tulsa, OK). After a significant main effect, post-hoc comparisons were done using the Tukey HSD test. Statistical significance was set at p < 0.05. Values are reported as the mean ± the standard error of the mean (S.E.M.).

mGluR2 Immunohistochemistry

Mice (P8-9) were anesthetized with halothane, decapitated, and the brains were removed and fixed in a 4% paraformaldehyde and sucrose solution overnight. After fixing overnight, slices (80 μm in thickness) were prepared with a freezing microtome (Microm HM 400, Fisher Scientific) and rinsed with tris-buffered saline (TBS) for an hour. Endogenous peroxidases were inhibited by incubating slices with 0.3% H₂O₂ in TBS for 30 min. To enhance epitope exposure the slices were incubated for 30 min in TBS preheated to 80° C in a water bath (Jiao et al., 1999). Nonspecific binding was prevented by incubating the slices in a blocking solution containing 0.25% Triton X-100 and 5% Bovine Serum Albumin (BSA) in TBS for one hour. A mouse anti-mGluR2 antibody (Advance Targeting System Inc, San Diego, CA) was diluted (1 μg/ml) in TBS containing 1% BSA and 0.25% Triton X-100 and incubated overnight. The next day, the slices were washed with TBS and incubated overnight with a biotinylated anti-mouse antibody (Vector Laboratories, Burlingame, CA) diluted (10 μg/ml) in TBS containing 1% BSA and 0.25% Triton X-100. Next, slices were washed with PBS and revealed with a standard avidin-biotin peroxidase procedure (Vectastain ABC kit, Vector Laboratories, Burlingame, CA) as previously described (Porter et al., 2001). In control slices the same protocol was followed except the primary antibody against mGluR2 was omitted. In the absence of the primary antibody no staining was observed. Images of the immunoreactivity were taken with a cooled CCD camera (CoolSNAPcf, Roper Scientific, Trenton, NJ) using MetaMorph software version 6.3r7 (Molecular Devices Co., Downingtown, PA).

Results

Group II mGluRs decrease thalamocortical EPSCs

In the present study, we determined whether mGluRs modulate thalamocortical synapses onto layer IV neurons of the developing somatosensory cortex. Monosynaptic thalamocortical EPSCs were evoked by stimulating either the ventrobasal nucleus of the thalamus or the internal capsule in the presence of bicuculline and AP5 to block GABA_A and NMDA currents, respectively. Consistent with previous studies of thalamocortical synapses, all connections included in this study displayed short latencies (6.2 ± 0.03 ms, range 3.6 to 8.1 ms, n = 44; Agmon and O’Dowd, 1992, Crair and Malenka, 1995, Agmon et al., 1996, Lu et al., 2001, Laurent et al., 2002) and paired pulse depression (paired pulse ratio = 0.79 ± 0.04, n = 44; Gil et al., 1997, Gibson et al., 1999, Porter et al., 2001). In order to determine which mGluRs modulate thalamocortical synapses the amplitudes of the evoked EPSCs were measured in the presence of selective agonists for group I, II, and III mGluRs. As shown in Figure 1, bath application of the selective group II mGluR agonist, 10 μM DCG IV (Gereau and Conn, 1995), reduced thalamocortical EPSCs by half (53 ± 7% of control, n = 15; t = 6.4, p < 0.0001). In contrast, the selective group I mGluR agonist, 20 μM DHPG (Schoepp et al., 1994, Gereau and Conn, 1995), and the selective group III mGluR agonist, 50 μM L-AP4 (Conn and Pin, 1997) had no effect on thalamocortical EPSCs. In the presence of DHPG and L-AP4 the EPSCs were 101 ± 2% (n = 4, t = 0.16, p = 0.88) and 94 ± 6% (n = 7; t = 0.94, p = 0.38) of control EPSCs, respectively. These experiments demonstrate that group II mGluRs modulate thalamocortical synapses onto layer IV neurons of the developing somatosensory cortex.

Figure 1.

Perfusion of a group II, but not a group I or III, mGluR agonist reduces thalamocortical EPSCs. A) Average of ten consecutive EPSCs taken before and after perfusion of the group II mGluR agonist DCG IV (10 μM). B) Time course showing the depression of evoked thalamocortical EPSCs by DCG IV in a cortical neuron. C) Average of ten consecutive thalamocortical EPSCs recorded before and after the application of the group I mGluR agonist DHPG (20 μM). D) Average of ten consecutive EPSCs taken before and after application of the group III mGluR agonist L-AP4 (50 μM).

DCG IV reduces thalamocortical EPSCs in both inhibitory and excitatory neurons

To determine whether group II mGluRs modulate thalamocortical synapses onto both inhibitory and excitatory neurons we measured the effect of DCG IV on evoked thalamocortical EPSCs recorded in both types of neurons. To record from inhibitory and excitatory neurons we selected cells located in the layer IV barrels with large elongated cell bodies and small round cell bodies respectively (Simons and Woolsey, 1984, Lin et al., 1985). Neurons were further classified using electrophysiological criteria as previously described (McCormick et al., 1985, Connors and Gutnick, 1990, Porter et al., 2001, Fontanez and Porter, 2006). In brief, inhibitory neurons fired action potentials at higher maximal frequencies and exhibited spikes with large rapidly repolarizing afterhyperpolarizations when depolarized with injected current pulses (Fig. 2A), while excitatory neurons discharged spikes exhibiting slowly repolarizing afterhyperpolarizations at lower maximal frequencies (Fig. 2B). The maximum firing frequency of the inhibitory neurons (47 ± 4.0 Hz; n = 13) was higher than the maximum frequency of the excitatory neurons (19 ± 1.7 Hz, n = 11, t = 5.97, p < 0.0001). In addition, the amplitudes of the afterhyperpolarization of the inhibitory neurons (10 ± 0.5 mV; n = 13) were greater than those of the excitatory neurons (5 ± 0.4 mV; n = 11; t = 7.55, p < 0.0001). As shown in Figure 2C, inhibitory and excitatory neurons were effectively separated into two populations by their maximal firing frequency and the amplitude of their afterhyperpolarizations. Bath application of 10 μM DCG IV caused a significant reduction of the thalamocortical EPSCs in both types of neurons. As shown in Figure 2D, the amount of depression of the thalamocortical EPSCs was equal in the inhibitory (46 ± 8% reduction, n = 9) and the excitatory neurons (47 ± 14% reduction, n = 6, t = 0.07, p = 0.94). These results indicate that group II mGluRs modulate thalamocortical synapses onto both inhibitory and excitatory neurons.

Figure 2.

DCG IV equally depresses thalamocortical synapses onto excitatory and inhibitory neurons. A and B represent firing discharges of an inhibitory and an excitatory neuron, respectively, in response to injection of a depolarizing current pulse. C) Graph showing the separation of inhibitory and excitatory neurons by afterhyperpolarization (AHP) amplitude and maximum firing frequency. D) Average amplitude of thalamocortical EPSCs recorded in 9 inhibitory neurons and 6 excitatory neurons during (DCG IV) and after (Wash) application of 10 μM of DCG IV. *p < 0.05.

DCG IV acts presynaptically

Group II mGluRs are located both on presynaptic axon terminals and on postsynaptic membranes (Cartmell and Schoepp, 2000). Therefore the reduction in thalamocortical EPSCs could be caused by a presynaptic decrease in glutamate release or by a postsynaptic modification of ionotropic glutamate receptors. To distinguish between these two possible mechanisms, we measured the effect of 10 μM DCG IV on the paired-pulse ratio (PPR), the coefficient of variance (CV) of the EPSCs, the kinetics of the EPSCs, and the membrane resistance. A decrease in neurotransmitter release is associated with an augmentation in the PPR and CV of the EPSCs (Zucker, 1989, Ohana and Sakmann, 1998). As shown in Figure 3A and B, the thalamocortical fibers exhibited paired pulse depression. Application of 10 μM DCG IV increased the PPR of thalamocortical synapses from 0.77 ± 0.05 to 1.17 ± 0.17 (n = 9) in inhibitory neurons. Repeated measures ANOVA indicated a main effect of condition (F (2, 16) = 6.31, p =0.01) and post-hoc comparisons indicated that the PPR was greater in the presence of DCG IV than under control conditions (p = 0.009) and washout conditions (p = 0.04). In excitatory neurons DCG IV also increased the PPR from 0.83 ± 0.07 to 1.44 ± 0.09 (n = 6). Repeated measures ANOVA indicated a main effect of condition (F (2, 10) = 18.20, p < 0.001) and post-hoc comparisons indicated that the PPR was greater in the presence of DCG IV than control (p < 0.001) and washout conditions (p = 0.004). Therefore the stimulation of group II mGluRs converted the short-term plasticity of the synapses from paired-pulse depression to paired-pulse facilitation. To further verify the presynaptic nature of the DCG IV-mediated reduction of thalamocortical EPSCs, the CV was calculated in these same neurons (Fig. 3C and 3D). In inhibitory neurons (n = 9), DCG IV increased the CV of the EPSCs from 0.19 ± 0.04 to 0.38 ± 0.08 which gave a main effect of condition (F _{(2, 16)} = 11.97, p = 0.001). Post-hoc comparisons indicated that the CV was greater in the presence of DCG IV than either control (p = 0.001) or washout conditions (p = 0.004). In excitatory neurons (n = 6), DCG IV also increased the CV from 0.21 ± 0.04 to 0.36 ± 0.03 with a main effect of condition (F (2, 10) = 6.29, p = 0.017). Post-hoc analysis showed that the CV was greater during DCG IV application than during control (p = 0.02) or washout conditions (p = 0.04). These results suggest that DCG IV inhibits the thalamocortical EPSCs by reducing glutamate release via presynaptic mGluRs.

Figure 3.

DCG IV increases the PPR and CV of the thalamocortical EPSCs. A) Average of 10 consecutive EPSCs in a P9 inhibitory neuron in response to paired thalamocortical stimuli during control conditions and during application of 10 μM DCG IV. B) Average PPR in 9 inhibitory neurons and 6 excitatory neurons taken before (control), during (DCG IV), and after DCG IV application (wash). C) Overlays of 10 consecutive EPSCs under control conditions and during DCG IV application showing that DCG IV increased the CV. D) Average CV of the EPSCs in the same 9 inhibitory and 6 excitatory neurons before, during, and after DCG IV application. E) Average of 10 consecutive thalamocortical EPSCs in an excitatory neuron dialyzed internally with 1 mM GDPβS during control conditions and during application of 10 μM DCG IV. F) Average effect of DCG IV on membrane resistance (Rm) and thalamocortical EPSCs in 4 excitatory neurons dialyzed internally with GDPβS. *p < 0.05.

To determine whether DCG IV has postsynaptic effects, we measured the kinetics of the EPSCs and the membrane resistance before and during the application of DCG IV. Application of 10 μM DCG IV did not significantly affect the EPSC decay time constant in either the excitatory (control = 3.8 ± 1.2 ms; DCG IV = 3.3 ± 1.0 ms; n = 6, t = 0.90, p = 0.4) or the inhibitory neurons (control = 3.0 ± 0.2 ms; DCG IV = 3.1 ± 0.1 ms; n = 9, t = 0.93, p = 0.38). DCG IV also did not affect the membrane resistance of the inhibitory neurons (96 ± 8% of control, n = 9, t = 0.53, p = 0.61). In contrast, DCG IV reduced the membrane resistance of the excitatory neurons (69 ± 6% of control, n = 6, t = 5.22, p = 0.003) consistent with the postsynaptic activation of K⁺ channels (Luscher et al., 1997, Dutar et al., 2000). These results indicate that DCG IV has presynaptic effects in inhibitory neurons and both presynaptic and postsynaptic effects in excitatory neurons. To determine whether the postsynaptic effects are responsible for the decrease in thalamocortical EPSCs recorded in the excitatory neurons, we examined the effect of DCG IV on EPSCs after blocking postsynaptic G protein activation with 1 mM GDPβS in the patch pipette (Dutar et al., 2000). In excitatory neurons dialyzed with GDPβS, DCG IV still reduced the thalamocortical EPSCs (43 ± 7% of control, n = 4, t = 4.3, p = 0.024) while not affecting membrane resistance (101 ± 4% of control, n = 4, t = 0.32, p = 0.77; Figure 3E, F). Thus, postsynaptic effects of DCG IV cannot explain the reduction of thalamocortical EPSCs. In summary, these analyses demonstrate that DCG IV stimulates presynaptic mGluRs on thalamocortical synapses on both inhibitory and excitatory neurons to reduce glutamate release.

The DCG IV-mediated depression is reversed by LY 341495 and mimicked by APDC

To confirm that DCG IV reduces thalamocortical EPSCs by stimulating group II mGluRs, we tested whether the effect of DCG IV could be reversed by the selective group II mGluR antagonist, LY 341495 (Bandrowski et al., 2003), and mimicked by another selective group II mGluR agonist, APDC (Schoepp et al., 1995). As shown in Figure 4A and B, 200 nM LY 341495 reversed the DCG IV-mediated inhibition of thalamocortical EPSCs in inhibitory neurons to 94 ± 6% of control (n = 4, t = 4.15, p = 0.025) and in excitatory neurons to 72 ± 9% of control (n = 4, t = 4.26, p = 0.024). In addition, as shown in Fig 4C and D, bath perfusion of 10 μM APDC reduced thalamocortical EPSCs in inhibitory neurons (45 ± 10% of control, n = 4, t = 5.4, p = 0.012) and in excitatory neurons (60 ± 10% of control, n = 5, t = 4.11, p = 0.015). Therefore, DCG IV and APDC equally reduced the thalamocortical EPSCs in inhibitory (t = 0.52, p = 0.61) and excitatory neurons (t = 0.44, p = 0.67). Together these results demonstrate that DCG IV depresses thalamocortical EPSCs by activating group II mGluRs.

Figure 4.

DCG IV’s reduction of thalamocortical EPSCs is reversed by a group II mGluR antagonist and mimicked by APDC. A) Average of 10 consecutive EPSCs recorded under control conditions, during perfusion of 10 μM DCG IV, and after subsequent perfusion of 10 μM DCG IV plus 200 nM LY 341495 illustrating the reversal of DCG IV-mediated depression by LY 341495. B) Average effect of DCG IV and DCG IV plus LY 341495 on thalamocortical EPSCs in 4 inhibitory and 4 excitatory neurons. C) Average of 10 consecutive EPSCs recorded before and during application of 10 μM APDC illustrating the reduction of thalamocortical synapses by APDC. D) Average effect of 10 μM APDC on thalamocortical synapses in 4 inhibitory and 5 excitatory neurons. * p < 0.05.

The mGluR3 subtype does not depress thalamocortical synapses

Group II mGluRs consist of two subtypes, mGluR2 and mGluR3. Since DCG IV and APDC equally stimulate both subtypes and LY 341495 equally blocks both subtypes, we could not determine which subtype mediates the reduction of thalamocortical EPSCs. The endogenous peptide, N-acetylaspartylglutamate (NAAG), selectively activates mGluR3 (Wroblewska et al., 1997) and the application of 30 μM NAAG reduces evoked EPSPs in cerebellar Purkinje neurons (Sekiguchi et al., 1989). Therefore to determine whether the mGluR3 subtype is responsible for the reduction of thalamocortical EPSCs, we bath applied NAAG. To prevent the breakdown of NAAG by peptidases like glutamate carboxypeptidase II (GCP II) which are found in brain slices (Riveros and Orrego, 1984, Robinson et al., 1987, Sanabria et al., 2004), we applied NAAG in the presence of phosphonomethyl-pentanedioic acid (PMPA) which inhibits GCP II (Bzdega et al., 2004, Alexander and Godwin, 2006). As shown in Figure 5, application of 10 μM of PMPA alone did not reduce the thalamocortical EPSCs (93 ± 4% of control, n = 6, t = 1.9, p = 0.11). Furthermore, application of 100 μM NAAG in the presence of PMPA did not reduce the thalamocortical EPSCs (88 ± 5% of control, n = 6). There was no difference between the EPSCs in the presence of PMPA alone or PMPA plus NAAG (n = 6, t = 1.1, p = 0.33), indicating that the mGluR3 subtype does not decrease glutamate release from thalamocortical terminals.

Figure 5.

The mGluR2 subtype appears to mediate inhibition of thalamocortical EPSCs. A) Average of 10 consecutive thalamocortical EPSCs during control conditions, during application of 10 μM PMPA, and after perfusion of 100 μM NAAG plus PMPA. B) Average effect of PMPA and PMPA plus NAAG on thalamocortical EPSCs in 6 neurons. C) Image of immunohistochemical staining for mGluR2 in the barrel cortex of a P9 mouse. Note heavy staining in barrel hollows (arrows). Scale bar = 100 μm.

The mGluR2 subtype is expressed where thalamocortical synapses terminate

Currently there are no selective mGluR2 agonists or antagonists available to determine if the mGluR2 subtype functionally suppresses thalamocortical EPSCs. However, a previous study indicates that mGluR2 is expressed in the adult rat and mouse barrel cortex (Ohishi et al., 1998). To confirm that mGluR2 is expressed in the barrels of developing mice, we used the previously characterized monoclonal anti-mGluR2 antibody (Ohishi et al., 1998). As shown in Figure 5C, mGluR2 is prominently expressed in the mouse barrel cortex during development and concentrated in the barrel hollows where the majority of thalamocortical axons terminate (Agmon et al., 1993). These results suggest that the reduction observed in the thalamocortical EPSCs during the activation of group II mGluRs is mediated by the mGluR2 subtype.

Synaptic activation of group II mGluRs reduces thalamocortical EPSCs

Although the above experiments indicate that group II mGluRs reduce glutamate release from thalamocortical synapses, they do not indicate whether these receptors are synaptically activated. To measure the endogenous activation of group II mGluRs, we gave a short conditioning train of 9 thalamocortical stimuli at 16 Hz followed 1.0 sec later by a single test stimulus similar to previously used protocols (Yamada et al., 1999, Billups et al., 2005). As indicated in Figure 6A, B, the thalamocortical EPSCs exhibited depression during the train. Blockade of group II mGluRs with 200 nM LY 341495 did not significantly alter the amplitudes of the EPSCs during the train suggesting that group II mGluRs do not mediated the train-induced depression (p > 0.06 for all EPSCs). However, LY 341495 did increase the size of the EPSC evoked by the test pulse (Figure 6C, D; n = 7, t =3.75, p = 0.009) indicating that the stimulation of group II mGluRs contributes to the synaptic depression of thalamocortical fibers.

Figure 6.

Synaptic stimulation reduces thalamocortical EPSCs via activation of group II mGluRs. A) Average of 10 consecutive traces showing thalamocortical EPSCs evoked by a train of 9 stimuli at 16 Hz during control conditions and during perfusion of 200 nM LY 341495. B) Average of 10 consecutive traces showing thalamocortical EPSCs evoked by a test stimulus given 1.0 sec after the train of 9 stimuli at 16 Hz during control conditions and during perfusion of 200 nM LY 341495 in the same neuron as in A. C) Average effect of LY 341495 on the 9 thalamocortical EPSCs evoked by the train in 7 neurons. D) Average effect of LY 341495 on thalamocortical EPSCs evoked by the test stimulus in the same 7 neurons. * p < 0.05.

Discussion

We directly examined whether mGluRs modulate the thalamic stimulation of cortical neurons. Our main finding is that the activation of group II mGluRs with exogenous agonists or synaptic stimulation depresses thalamic inputs onto excitatory and inhibitory cortical neurons of the developing somatosensory cortex. This depression is mediated by a reduction in glutamate release through the apparent stimulation of presynaptic mGluR2s. In contrast, thalamocortical EPSCs were unaffected by the activation of group I or III mGluRs. Therefore, group II mGluRs provide an important feedback mechanism for regulating thalamocortical excitation.

Previous studies demonstrated that application of a nonselective group I and group II mGluR agonist to the somatosensory cortex in vivo depresses the activation of cortical neurons by whisker stimulation (Cahusac, 1994, Taylor and Cahusac, 1994). Our results extend these earlier observations by showing that presynaptic mGluRs inhibit thalamocortical synapses and that group II mGluRs mediate the inhibition in developing mice. Previous reports indicate that mGluRs can be either maintained into adulthood (Macek et al., 1996, Yokoi et al., 1996, Flavin et al., 2000) or lost during development (Baskys and Malenka, 1991, Ross et al., 2000, Doherty et al., 2004). Therefore, it is also possible that the group II mGluR-mediated modulation of thalamocortical EPSCs is important only during development and does not occur in the adult brain despite the expression of group II mGluRs in the barrels of adult rats and mice (Ohishi et al., 1998, Muñoz et al., 1999).

Group II mGluRs reduce synaptic transmission by inhibiting glutamate release (Cartmell and Schoepp, 2000, Alexander and Godwin, 2005) and by hyperpolarizing neuronal dendrites and cell bodies (Sodickson and Bean, 1998, Dutar et al., 2000, Watanabe and Nakanishi, 2003). Our results demonstrate that group II mGluRs decrease thalamocortical excitability by reducing neurotransmitter release, since the paired-pulse ratio and the coefficient of variance of the thalamocortical EPSCs increased in response to group II mGluR stimulation. Furthermore, mGluR activation reduced the thalamocortical EPSCs in the absence of any change in either the membrane resistance or the EPSC decay kinetics. Together, these two analyses indicate that the reduction induced by group II mGluR agonists is mediated by a presynaptic decrease in glutamate release.

Group II mGluRs include mGluR2 and mGluR3 both of which can presynaptically reduce glutamate release (Sanabria et al., 2004, Poisik et al., 2005). The lack of effect of the selective mGluR3 agonist NAAG suggests that mGluR2 is responsible for the inhibition of thalamocortical synapses. Consistent with the possibility that mGluR2 mediates the reduction produced by group II mGluR agonists, mGluR2 is expressed in the hollows of the barrel cortex where the thalamocortical axons terminate (Ohishi et al., 1998) Figure 5). Taking together these findings, we suggest that presynaptic mGluR2s modulate thalamocortical synapses. Recent experiments have shown that the mGluR2 subtype also decreases glutamate release from corticothalamic terminals (Alexander and Godwin, 2005, Alexander and Godwin, 2006) indicating that mGluR2s are intricately involved in negative feedback within the thalamocortical circuitry.

The mGluR2-mediated depression of thalamocortical synapses may play a number of physiological roles during development. First, it is likely to be important for preventing excessive cortical excitation, since group II mGluR agonists effectively reduce epileptic activity (Burke and Hablitz, 1995, Miyamoto et al., 1997, Folbergrova et al., 2001) and can prevent excitotoxicity (Bruno et al., 1994, Buisson et al., 1996). Second, since we found that the group II mGluRs are functional during the time of elaboration of thalamocortical connections in the barrel cortex (Senft and Woolsey, 1991, Agmon et al., 1995, White et al., 1997), the mGluRs may play a role in the formation of thalamocortical connectivity. Consistent with this idea, barrel formation is disrupted by the excessive stimulation of presynaptic 5-HT_1B receptors on thalamocortical terminals (Salichon et al., 2001, Rebsam et al., 2002), which like group II mGluRs also inhibit adenylate cyclase and inhibit glutamate release (Laurent et al., 2002). Third, the stimulation of group II mGluRs may regulate the plasticity of thalamocortical synapses, since the activation of these receptors induces long-term depression at hippocampal and cortical synapses (Yokoi et al., 1996, Li et al., 2002, Renger et al., 2002).

In conclusion, our results demonstrate that presynaptic group II mGluRs modulate thalamocortical excitation. This modulation is likely to provide an important feedback mechanism for preventing excessive thalamocortical excitation that may play a role in the plasticity and refinement of thalamocortical connections during development.

Abbreviations

mGluRs

metabotropic glutamate receptors

EPSCs

excitatory postsynaptic currents

DCG IV

(2R, 2′R, 3′R)-2-(2′, 3′ dicarboxycyclopropyl) glycine

L-AP4

L-(+)-2-amino-4-phosphonobutyric acid

APDC

(2R, 4R)-4-aminopyrrolidine-2, 4-dicarboxylic acid

AMPA

DL-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

NMDA

N-methyl-D-aspartate

NIH

National Institutes of Health

ACSF

artificial cerebrospinal fluid

AP5

DL-2-amino-5-phosphonopentanoic acid

GABA

gamma aminobutyric acid

NAAG

N-acetylaspartylglutamate

PPR

paired pulse ratio

CV

coefficient of variation

TBS

tris-buffered saline

BSA

bovine serum albumin

DHPG

3, 5 dihydroxyphenylglycine

GCP II

glutamate carboxypeptidase II

PMPA

phosphonomethyl-pentanedioic acid