Converging lines of behavioral and pharmacological evidence suggest that GABAergic synapses in the basolateral amygdala (BLA) may play an integral role in mediating the anxiolytic effects of ethanol (EtOH). Since anxiety is thought to play an important role in the development of, and relapse to, alcoholism, elucidating the mechanisms through which EtOH modulates GABAergic synaptic transmission in the BLA may be fundamental in understanding the etiology of this disease. A recent study in mice has shown that principal cells within the BLA receive inhibitory input from two distinct types of GABAergic interneurons: a loosely distributed population of local interneurons and a dense network of paracapsular (pcs) GABAergic cells clustered along the external capsule border. Here, we sought to confirm the presence of these two populations of GABAergic synapses in the rat BLA and evaluate their ethanol sensitivity. Our results suggest that rat BLA pyramidal cells receive distinct inhibitory input from local and pcs interneurons and that EtOH potentiates both populations of synapses, albeit via distinct mechanisms. EtOH enhancement of local inhibitory postsynaptic currents (IPSCs) was associated with a significant decrease in paired-pulse ratio (PPR) and was significantly potentiated by the GABAB receptor antagonist SCH 50911 [(+)-(S)-5,5-dimethylmorpholinyl-2-acetic acid], consistent with a facilitation of GABA release from presynaptic terminals. Conversely, EtOH enhancement of pcs IPSCs did not alter PPR and was not enhanced by SCH 50911 but was inhibited by blockade of noradrenergic receptors. Collectively, these data reveal that EtOH can potentiate GABAergic inhibitory synaptic transmission in the rat BLA through at least two distinct pathways.
Although the exact etiology of alcoholism has yet to be fully elucidated, it is known that alcohol abuse is often comorbid with anxiety disorders (Cornelius et al., 2003) and that self-medication of anxiety may be an important underlying factor in the development of (Carrigan and Randall, 2003), and relapse to (Breese et al., 2005), alcoholism. Previous studies have suggested that many of the pharmacological effects of alcohol (EtOH) in the mammalian central nervous system may be attributable to interactions with GABAergic synaptic transmission (Weiner and Valenzuela, 2006). In fact, behavioral studies have consistently shown that EtOH can mimic the actions of positive allosteric modulators of GABAA receptors (Criswell and Breese, 2005). Because drugs that enhance GABAergic inhibition in the central nervous system are often used to treat anxiety disorders (Allison and Pratt, 2003), it has been hypothesized that ethanol's anxiolytic effects may be mediated, in part, via interactions with the GABAergic system. Therefore, determining the mechanisms through which EtOH enhances GABAergic transmission in brain regions that mediate anxiety behaviors may provide insight into the pathophysiology of alcoholism and possibly reveal new targets for the development of pharmacotherapies to alleviate this disorder.
The amygdala is thought to be critically involved in the regulation of anxiety-like behaviors, with the basolateral nucleus of the amygdala (BLA) playing an essential role in determining the saliency of external stimuli and in the production of appropriate behavioral responses (LeDoux, 1996). For example, direct infusion of some GABAA receptor agonists into the BLA has been shown to reduce certain measures of anxiety-like behavior (Sanders and Shekhar, 1995a; Menard and Treit, 1999), whereas microinjection of GABAA receptor antagonists into this brain region increases some experimental anxiety-like behaviors (Sanders and Shekhar, 1995b; Sajdyk and Shekhar, 1997). Therefore, GABAergic synapses in the BLA may contribute to some of the acute anxiolytic effects of EtOH. Although many studies have demonstrated that EtOH can enhance GABAergic synaptic transmission via both pre- and postsynaptic sites in a brain region-specific manner (Siggins et al., 2005; Breese et al., 2006; Weiner and Valenzuela, 2006), the effects of EtOH at GABAergic synapses in the BLA are not fully known. Recent data suggest that EtOH may enhance GABAergic synaptic transmission in the rat BLA by increasing GABA release from presynaptic terminals (Zhu and Lovinger, 2006) while having no direct effect on GABAA receptor activity (McCool et al., 2003). It is noteworthy that Marowsky et al. (2005), using GAD-GFP mice, have recently reported that the primary glutamatergic output cells of the BLA receive inhibitory input from two distinct populations of GABAergic cells, local interneurons and paracapsular (pcs) cells. Although initial studies in mice have revealed important differences in the pharmacological properties of synapses arising from these two groups of GABAergic interneurons, these synapses have not been characterized in rat, and the ethanol sensitivity of these two populations of inhibitory synapses is not known. Therefore, in this study, we sought to characterize the acute effects of EtOH at local and pcs GABAergic synapses onto rat BLA projection neurons utilizing in vitro slice electrophysiology. Our findings suggest that EtOH enhances GABAergic transmission at synapses arising from both local and paracapsular GABAergic interneurons in the rat BLA, however via distinct mechanisms.
Materials and Methods
Slice Preparation. Transverse amygdala slices (400 μm) were prepared from 4 to 6-week-old male Sprague-Dawley rats. Slices were maintained at ambient temperature for at least 2 h in oxygenated artificial cerebrospinal fluid containing 124 mM NaCl, 3.3 mM KCl, 2.4 mM MgCl2, 2.5 mM CaCl2, 1.2 mM KH2PO4, 10 mM d-glucose, and 25 mM NaHCO3, saturated with 95% O2 and 5% CO2.
Electrophysiological Recordings. Slices were transferred to a recording chamber and superfused with aerated artificial cerebrospinal fluid at 2 ml/min using a calibrated flowmeter (Gilmont Instruments, Racine, WI). Experiments were performed at ambient temperature because our previous studies have found that this promotes the stability of patch-clamp recordings in brain slices and does not influence ethanol enhancement of GABAA IPSCs (Ariwodola and Weiner, 2004). Recording electrodes were prepared from filamented borosilicate glass capillary tubes (inner diameter, 0.86 mm) using a horizontal micropipette puller (P-97; Sutter Instruments, Novato, CA). Patch-clamp recordings were made using a filling solution containing 130 mM K-gluconate, 10 mM KCl, 1 mM EGTA, 100 μM CaCl2, 2 mM Mg-ATP, 200 μM Tris-guanosine 5′-triphosphate, and 10 mM HEPES, pH adjusted with KOH, 275 to 280 mOsm. In all experiments, 5 mM QX-314 was included in the recording solution to block voltage-gated sodium currents and GABAB IPSCs in the BLA neurons being recorded (Horn et al., 1980; Nathan et al., 1990). Whole-cell patch-clamp recordings were made from BLA pyramidal neurons voltage-clamped at –30 to –40 mV (not corrected for junction potential). Only cells with a stable access resistance of 5 to 20 MΩ were used in these experiments. Whole-cell currents were acquired using an Axoclamp 2B or Axopatch 200B amplifier, digitized (Digidata1200 or Digidata 1321A; Axon Instruments, Union City, CA), and analyzed on- and off-line using an IBM-compatible personal computer and pClamp 9.0 software (Axon Instruments).
Pharmacological Isolation of Synaptic Currents. GABAA IP-SCs were evoked every 20 s by electrical stimulation (0.2-ms duration) using a concentric bipolar stimulating electrode (FHC, Bowdoinham, ME) placed near (50–100 μm) the recording electrode (“local” stimulation) or along the external capsule (“distal” stimulation). In all experiments, stimulation intensity was adjusted to evoke responses that were 10 to 20% of maximal currents (typically 80–120 pA). GABAA IPSCs were pharmacologically isolated using a mixture of 50 μM dl-2-amino-5-phosphonovaleric acid and 20 μM 6,7-dinitroquinoxaline-2,3-dione to block N-methyl-d-aspartate and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate receptors, respectively. In some experiments, the effects of dual stimulation on evoked GABAA IPSCs were studied by delivering a conditioning pulse either locally or distally, 250 ms before a local “test” pulse. To conduct an initial assessment of the synaptic locus of EtOH effects on local or distal GABAA IPSCs, a paired-pulse protocol with an interstimulus interval (ISI) of 50 ms was also utilized in some experiments. In both of these experiments, GABAA IPSCs were evoked every 45 s. Unless otherwise stated, all drugs used were purchased from Sigma (St. Louis, MO). Drugs were made up as 100 to 400-fold concentrates and applied to slices via calibrated syringe pumps (Razel Scientific Instruments, Stamford, CT). A 4 M ethanol solution was prepared immediately before each experiment from a 95% stock solution (Aaper Alcohol and Chemical, Shelbyville, KY) kept in a glass storage bottle.
Immunocytochemistry. Rats were anesthetized with an i.p. injection of pentobarbital (200 mg/kg) and perfused transcardially with saline followed by 4% paraformaldehyde in phosphate buffer. The brains were then dissected from the cranial vault and blocked by two coronal cuts. Tissue blocks were postfixed overnight, cryoprotected in a graded concentration series of sucrose (10, 20, and 30%), frozen in OCT on dry ice, and stored at –80°C. Brains then were sectioned coronally at 40 μm on a freezing microtome. Sections through the BLA were immunoreacted with either GAD67 or PV primary antibodies. Free-floating sections were rinsed in phosphate-buffered saline, blocked in Triton X and normal horse serum, and immunoreacted with mouse anti-GAD67 monoclonal antibody (1:3000; MAB5406; Chemicon International, Temecula, CA) or mouse anti-parvalbumin monoclonal antibody (1:6000; Sigma, P 3088) in Triton X and normal horse serum followed by biotinylated, rat-adsorbed horse anti-mouse IgG secondary antibody (1:1000; Vector Laboratories, Burlingame, CA), incubated in avidin-biotin complex (ABC Elite kit, Vector; 30 min), and visualized with diaminobenzidine (Elite Kit, Vector). Finally, sections were washed in phosphate-buffered saline, mounted, dehydrated in a graded series of ethanol, cleared in xylene, and coverslipped. As a control, one section was processed according to the same protocol with the omission of primary antibody (GAD67 or PV).
Statistics. Ethanol and other drug effects on evoked IPSCs were quantified as the percentage change in the area under the curve of synaptic currents relative to the mean of control and washout values. The effect of ethanol on paired-pulse ratio (PPR) was assessed by averaging five consecutive sweeps in the presence and absence of the drug (Kim and Alger, 2001). PPR was calculated as the amplitude of the second response divided by the amplitude of the first, relative to the same “prestimulation” baseline, using an ISI of 50 ms. Although fully separable IPSCs could not be evoked at short ISIs under our recording conditions, this protocol proved to be a reliable index of presynaptic changes in empirical experiments in our lab and has been used in another recent study where IPSCs did not decay fully to baseline (Liang et al., 2006). Statistical analyses of drug or stimulation effects were performed using the two-tailed Student's paired, unpaired Student's t tests, or a one-way analysis of variance followed by the Newman-Keuls post hoc test, where applicable, with a minimal level of significance of p < 0.05.
Physiological Evidence of Distinct Populations of GABAergic Interneurons in the Rat BLA. As noted in the introduction, a recent study using GAD-GFP mice provided morphological and electrophysiological evidence of two distinct populations of GABAergic interneurons that synapse onto glutamatergic pyramidal neurons in the BLA (Marowsky et al., 2005). One population is comprised of sparsely distributed “classical” local interneurons that mediate the majority of feedback inhibition onto BLA pyramidal cells. The other population is a densely clustered network of paracapsular interneurons located along the external capsule border that mediate most cortical feed-forward inhibition in the BLA. Immunohistochemical studies in these mice showed that the population of local interneurons was found to be immunopositive for GAD, parvalbumin (PV), and other typical interneuron markers (e.g., calbindin, cholecystokinin). Interestingly, paracapsular interneurons were immunopositive only for GAD. To determine whether similar populations of interneurons are present in the rat, we performed immunohistochemical studies to examine the distribution of GAD- and PV-positive interneurons in the rat BLA. As observed in the mouse, GAD staining revealed classical local interneurons, scattered loosely throughout the BLA, as well as dense clusters located along the medial and lateral borders of the BLA (Fig. 1A). It is noteworthy that although many local interneurons stained positive for PV, the dense clusters of cells located along the BLA borders were devoid of this marker, as also reported in mouse (Fig. 1B).
In our next experiment, we employed a paired-pulse depression (PPD) protocol to determine whether synapses arising from these two populations of GABAergic interneurons could be recorded in rat amygdala slices. Whole-cell patch recordings were made from BLA pyramidal neurons, and pairs of pharmacologically isolated GABAA IPSCs were evoked by one of two stimulating electrodes. One electrode was positioned distally, along the external capsule border of the BLA to target the lateral pcs clusters. The other electrode was placed proximal to the recording electrode to stimulate local interneurons (Fig. 2A). In many brain regions, including the BLA, dual stimulation of the same population of GABAergic synapses at interstimulus intervals of 100 to 500 ms results in a depression of the amplitude of the second response relative to the first, termed PPD (Davies et al., 1990; Pearce et al., 1995; Szinyei et al., 2000). We hypothesized that, if the local and distal stimulating electrodes were activating distinct, nonoverlapping populations of GABAergic interneurons, then PPD should be evident when the paired stimuli were delivered from the same electrode but not when a locally evoked IPSC was preceded by a distal response. Under our recording conditions, paired local stimulation at an ISI of 250 ms resulted in a significant depression of the second IPSC amplitude relative to the first (39.7 ± 2.9%, p < 0.05, n = 14; Fig. 2B). In contrast, pairing a distal and a local IPSC of comparable size, at the same 250-ms ISI, had no significant effect on the amplitude of the local response (7.4 ± 4.9%, p > 0.05, n = 14; Fig. 2B). These findings suggest that, as observed in the mouse BLA, pcs and local GABAergic interneurons in the rat BLA activate functionally distinct populations of GABAergic synapses onto individual pyramidal neurons in this brain region. Interestingly, paired distal stimulation at an ISI of 250 ms did not result in significant paired pulse depression (9.6 ± 7.3%, p > 0.05, n = 13, data not shown), suggesting that there may be important differences in the physiological processes that regulate short-term plasticity between local and pcs synapses.
Effects of EtOH on Local and Distal BLA GABAergic Synapses. In our next series of experiments, we examined the acute effect of ethanol on GABAA IPSC area evoked by stimulation of either local interneurons (local) or the pcs interneuron cluster along the external capsule border (distal). In these and subsequent experiments, local and distal synapses were tested separately to minimize any possible influence of the dual stimulation protocol on ethanol effects. Bath application of EtOH (5–80 mM) significantly potentiated the area of local and distal GABAA IPSCs in a concentration-dependent manner (Fig. 3). This potentiation by EtOH was not associated with any changes in holding current or input resistance, even at the highest concentration tested (ΔIhold = 5.7 ± 4.8%; ΔRI = 5.0 ± 2.4%). Ethanol potentiation of both local and distal IPSCs was apparent within 5 to 8 min and reversed after a 10 to 15-min washout (Fig. 4, A1 and B1). At both types of synapses, significant effects of EtOH were observed at concentrations of 20 mM and higher and a statistical comparison of the data revealed a significant overall effect of EtOH (F = 4.365, p < 0.003) with no significant effect of stimulus location (F = 2.090; p > 0.05) and no significant stimulus-ethanol interaction effect (F = 0.3334, p > 0.05).
As observed in previous studies (e.g., Weiner et al., 1997), ethanol had somewhat variable effects on the amplitude and decay of local and distal evoked IPSCs. However, the IPSC area, which reflects the total charge transfer that occurs during each synaptic event, was consistently potentiated in a concentration-dependent manner at both types of synapses.
Mechanisms Underlying EtOH Potentiation of Local and Distal BLA GABAergic Synapses: Role of Presynaptic GABAB Receptors. The preceding experiment revealed that local and distal GABAergic synapses onto BLA pyramidal neurons were potentiated by EtOH. We next sought to determine the possible mechanisms through which EtOH enhances these two populations of inhibitory synapses. Previous research from our lab and others (Siggins et al., 2005; Breese et al., 2006; Weiner and Valenzuela, 2006) has shown that EtOH enhances GABAergic synaptic transmission primarily via a presynaptic facilitation of GABA release. In addition, it has previously been shown that agonist activation of presynaptic GABAB receptors in the BLA results in an inhibition of GABAergic transmission in this brain region (Yamada et al., 1999). Notably in both the hippocampus (Ariwodola and Weiner, 2004) and isolated BLA nerve bouton preparation (Zhu and Lovinger, 2006), the EtOH-mediated increase in GABA release activates presynaptic GABAB autoreceptors, which serves to limit the overall potentiating effect of EtOH at these synapses. This limiting effect can be unmasked by pretreating slices with a GABAB receptor antagonist, which significantly increases EtOH potentiation of GABAA IPSCs in these preparations. To determine whether a similar presynaptic mechanism modulates EtOH enhancement of local and distal synapses in BLA slices, we tested the effect of 80 mM ethanol on the area of local and distal IPSCs alone and in the presence of the GABAB receptor antagonist, SCH 50911 (20 μM). At local synapses, 80 mM EtOH significantly increased GABAA IPSC area by 48.0 ± 8.9% (p < 0.05, n = 10; Fig. 4A1), and bath application of SCH 50911 had no effect on its own. However, pretreatment with SCH 50911 significantly enhanced EtOH potentiation of local IP-SCs (97.3 ± 16.2%, p < 0.05, n = 9; Fig. 4A2). As observed in our first experiments, bath application of 80 mM EtOH significantly enhanced the area of distal IPSCs (64.3 ± 17.0%, p < 0.05, n = 10; Fig. 4B1). Interestingly, unlike the effect observed at local synapses, bath application of SCH 50911 alone significantly potentiated the area of distal IPSCs (47.0 ± 17.1%, p < 0.05, n = 10; Fig. 4B2), and, although pretreatment with the GABAB receptor antagonist did not unmask a greater EtOH potentiation of distal GABAA IPSCs, EtOH still potentiated distal IPSCs in the presence of SCH 50911 to a similar extent as that observed under control conditions (48.6 ± 13.5%, p > 0.05, n = 10; Fig. 5).
Although ethanol consistently potentiated the area of local and distal IPSCs in these experiments, ethanol effects on IPSC amplitude and decay were more variable, as observed under control conditions. However, ethanol modulation of local and distal IPSC amplitude and decay were not selectively influenced by SCH 50911 pretreatment (data not shown).
Mechanisms of EtOH Action: Paired-Pulse Ratio. The results from the preceding experiment suggest that, although EtOH significantly enhances both local and distal GABAergic synapses in the BLA, these synapses may be potentiated via distinct mechanisms. To further examine the mechanisms of EtOH action at these two populations of synapses, we next examined the effect of EtOH on IPSC PPRs using an ISI of 50 ms. Many studies have demonstrated that experimental manipulations that result in a presynaptic increase in neurotransmitter release probability are typically associated with a decrease in PPR (Siggins et al., 2005; Liang et al., 2006), whereas postsynaptic manipulations generally have no effect on PPR. Under our recording conditions, PPRs were similar between local and distal synapses (1.2 ± 0.1 versus 1.2 ± 0.2, p > 0.05, n = 7 and 6, respectively; Fig. 6), suggesting that were no significant differences in basal release probability at both populations of synapses. Bath application of 80 mM EtOH significantly decreased PPR at local synapses (to 0.9 ± 0.1, p < 0.05, n = 7), consistent with a presynaptic locus of action. In contrast, although 80 mM EtOH potentiated distal pairs of IPSCs (see representative traces; Fig. 6, top), it had no significant effect on PPR at these synapses (1.1 ± 0.1, p > 0.05, n = 6; Fig. 6, bottom).
Mechanisms of EtOH Action: Role of Norepinephrine Receptors. The results of our first mechanistic studies are consistent with a presynaptic mechanism of EtOH action at local GABAergic synapses onto BLA pyramidal neurons. However, the findings that EtOH potentiation of distal IPSCs is not modulated by presynaptic GABAB autoreceptors and does not decrease PPR are not consistent with a facilitatory effect of EtOH on terminal GABA release at lateral pcs synapses. Rather, these findings raise the intriguing possibility that GABAergic synapses arising from lateral pcs interneurons may be facilitated by EtOH through a postsynaptic mechanism. Given the novel nature of EtOH effects on distal IPSCs in the BLA, we reviewed the relevant literature to identify possible candidate mechanisms that could potentially account for a postsynaptic mechanism of EtOH action. Interestingly, the BLA receives substantial noradrenergic innervation from the locus coeruleus (Roder and Ciriello, 1993), and a recent study has demonstrated a diffuse pattern of norepinephrine (NE)-positive terminals on pcs interneurons in the BLA (Fuxe et al., 2003). In addition, NE can potentiate GABAA receptor function (Cheun and Yeh, 1996), and NE significantly increases EtOH potentiation of GABA-mediated inhibition of Purkinje cell firing in the cerebellum (Lin et al., 1991). Based on these findings, we hypothesized that EtOH enhancement of pcs IPSCs may be influenced by NE receptor activity. To that end, we tested the effect of 80 mM EtOH on local and pcs IPSCs in the presence of a noradrenergic antagonist cocktail consisting of 1 μM prazosin, 20 μM yohimbine, and 10 μM propranolol to block α1, α2, and β adrenergic receptors, respectively. Bath application of the NE blocker cocktail alone had no significant effect on GABAA IPSCs regardless of stimulation locale (Fig. 7). In addition, pretreatment with the NE blocker cocktail did not significantly modulate EtOH potentiation of local IPSCs (48.0 ± 14.3% n = 6 versus 39.7 ± 10.4% n = 7, p > 0.05; Fig. 7). In contrast, the NE blockers completely abolished EtOH potentiation of pcs IPSCs (64.4 ± 17.0% n = 10 versus –2.6 ± 10.5% n = 16, p < 0.05). These findings suggest that NE may preferentially modulate GABAergic synaptic transmission from lateral pcs neurons. To that end, we tested the effect of NE on local and distal IPSCs. Bath application of 20 μMNE had no significant effect on local IPSC area (6.0 ± 8.4% n = 5, p > 0.05 Fig. 8) but did significantly potentiate the area of distal IPSCs (62.0 ± 17.2% n = 8, p < 0.05). In addition, pretreatment with the same NE antagonist cocktail described above completely blocked the effect of 20 μM NE on distal IPSCs (to –9.7 ± 7.7% n = 5, p > 0.05).
The results of this study provide anatomical and electrophysiological evidence that, as recently described in GAD-GFP mice (Marowsky et al., 2005), glutamatergic pyramidal cells in the rat BLA receive distinct inhibitory input from local and lateral pcs interneurons. In addition, our data further suggest that pharmacologically relevant concentrations of ethanol significantly enhance GABAergic synaptic inhibition arising from both local and pcs interneurons in the rat BLA, albeit via distinct mechanisms. At local GABAergic synapses, ethanol potentiation is significantly enhanced by pretreatment with a GABAB receptor antagonist and is associated with a significant decrease in PPR, consistent with a presynaptic facilitation of GABA release. In contrast, the present findings suggest that ethanol enhancement of GABAergic synapses arising from pcs interneurons may occur through a mechanism that does not involve a presynaptic alteration in GABA release because this enhancement is not influenced by antagonism of GABAB receptors and is not accompanied by a change in PPR. Moreover, blockade of NE receptors completely antagonized ethanol potentiation of pcs, but not local, GABAergic synapses, providing further evidence that ethanol enhances these two populations of GABAergic synapses via distinct mechanisms.
As noted above, a recent study using GAD-GFP mice found that BLA pyramidal cells, the predominant output neurons of this nucleus, receive inhibitory innervation from two distinct classes of GABAergic interneurons (Marowsky et al., 2005). One population of classical interneurons was scattered throughout the BLA, was immunopositive for typical interneuron markers like PV, could be activated by electrical stimulation proximal to BLA pyramidal cells, and likely mediates the majority of feedback inhibition onto BLA pyramidal cells. A second group of GABAergic interneurons (pcs) was detected along the border between the external capsule and the BLA, but these cells were not immunopositive for typical interneuron markers like PV. These densely packed clusters of cells exhibited properties very similar to the more extensively characterized population of intercalated interneurons located along the medial border of the BLA that receive excitatory input from BLA pyramidal neurons (Pare and Smith, 1993a,b) and project to cells within the central nucleus of the amygdala (CeA) (Royer et al., 1999). The pcs interneurons along the external capsule border could be readily evoked by stimulation within the external capsule and were shown to mediate the majority of cortical feed-forward inhibition onto BLA pyramidal neurons.
Our immunohistochemical studies suggest that similar populations of local and pcs interneurons may also be present in the rat BLA. GAD immunostaining revealed sparsely distributed local interneurons throughout the BLA as well as dense clusters localized to the lateral and medial borders of this brain region. It is noteworthy that PV staining was observed among the local interneurons but was not detected in the GAD-positive cells located within the dense clusters along the border of the BLA. Moreover, in the presence of glutamate receptor antagonists, GABAA IPSCs onto BLA pyramidal neurons could be evoked by stimulating electrodes placed distally, within the external capsule or locally, proximal to the cell being recorded. In addition, using a dual stimulation protocol, we demonstrated that a locally evoked conditioning IPSC significantly depressed the amplitude of a subsequent local test IPSC, evoked 250 ms later. In contrast, a distal conditioning IPSC, evoked by stimulating within the external capsule, had no significant effect on the amplitude of a subsequent local test IPSC. Taken together, the immunohistochemical and electrophysiological data provide empirical evidence that local and lateral pcs interneurons provide distinct, largely nonoverlapping inhibitory input onto rat BLA pyramidal neurons.
Although many studies have now demonstrated that ethanol enhances GABAergic inhibition in a number of different brain regions, the ethanol sensitivity of GABAergic responses can vary markedly between (Criswell et al., 1995) and even within a given brain area (Weiner et al., 1997). Multiple pre- and postsynaptic mechanisms have been proposed to account for these effects (Siggins et al., 2005; Weiner and Valenzuela, 2006). Our results suggest that, within the BLA, ethanol significantly enhances both local and pcs IPSCs at pharmacologically relevant concentrations. However, despite the fact that both populations of GABAergic synapses were potentiated at ethanol concentrations as low as 20 mM, these inhibitory pathways appear to be enhanced via distinct mechanisms.
Our data suggest that ethanol potentiation of local GABAergic synapses in the BLA is mediated predominantly via a facilitation of GABA release from presynaptic terminals. This mechanism is supported by the observation that ethanol enhancement of local IPSCs is accompanied by a significant decrease in PPR, similar to its effect at GABAergic synapses in the central nucleus of the amygdala (Roberto et al., 2003, 2004) and hippocampus (Sanna et al., 2004). Indeed, several studies in the hippocampus have shown that ethanol significantly increases GABA release onto CA1 pyramidal neurons (Ariwodola and Weiner, 2004; Sanna et al., 2004; Li et al., 2006), and we have demonstrated that this effect serves to limit ethanol's overall potentiating effect at these synapses by activating presynaptic GABAB autoreceptors to reduce subsequent evoked GABA release (Ariwodola and Weiner, 2004). The finding that blockade of GABAB receptors significantly increased ethanol enhancement of local IPSCs in the BLA suggests that a similar presynaptic mechanism contributes to ethanol actions in this brain region as well.
These findings are also consistent with a recent report that ethanol significantly increases the frequency of spontaneous IPSCs in BLA neurons in brain slices and in an acutely isolated BLA neuron/bouton preparation (Zhu and Lovinger, 2006). This study also demonstrated, in the isolated neuron preparation, that ethanol had no effect on GABA-evoked currents or on the amplitude of tetrodotoxin-resistant miniature IPSCs, suggesting that postsynaptic mechanisms do not contribute to ethanol potentiation of GABAergic synapses in this preparation. Although the location of local and pcs synapses onto BLA neurons is not known, kinetic differences between local and pcs IPSCs observed during paired recordings suggest that pcs synapses may be located at more distal dendritic loci (Marowsky et al., 2005) or that possible differences in GABAA receptor subunit composition may exist at these two synapses. These findings, coupled with our data suggesting a presynaptic mechanism of ethanol action at local, but not pcs, synapses, may indicate that spontaneous IPSCs recorded from BLA pyramidal neurons in the bouton preparation may primarily reflect the activity of local interneuronal GABAergic synapses.
Although the ethanol sensitivity of pcs IPSCs was very similar to that observed at local synapses, our findings suggest that ethanol may not act via a facilitation of terminal GABA release at these synapses. First of all, ethanol potentiation of pcs IPSCs was not associated with a significant change in PPR. Moreover, this potentiation was not modulated by blockade of GABAB receptors, suggesting that acute ethanol application may not release sufficient GABA at these synapses to engage presynaptic GABAB receptors. Interestingly, ethanol potentiation of GABAergic inhibition in the CeA is also not influenced by blockade of presynaptic GABAB receptors (Roberto et al., 2003). It is noteworthy that application of the GABAB receptor antagonist SCH 50911 alone, which had no effect at local synapses, significantly potentiated evoked IPSCs at pcs synapses. This finding suggests that pcs GABAergic synapses can be modulated by presynaptic GABAB receptors and that these receptors are tonically active under our standard recording conditions.
Together, these finding further suggest that either the GABA sensitivity of presynaptic GABAB receptors may be significantly higher at pcs synapses than local synapses or that ambient GABA levels are elevated at pcs synapses, further underscoring the differential properties of these two populations of BLA GABAergic synapses. In fact, our finding that there is minimal PPD at pcs synapses may indicate that pcs GABAB receptors are maximally active under basal conditions. Such maximal activity may have resulted in a “ceiling effect” that could have occluded any interaction between EtOH and SCH 50911, even if ethanol was acting via a presynaptic mechanism at these distal synapses. However, this interpretation is not supported by the observation that EtOH potentiation of pcs IPSCs was not associated with a change in PPR. Clearly, additional experiments will be needed to fully resolve the synaptic locus of ethanol action at distal synapses and to more extensively characterize the differences in GABAB receptor modulation of local and pcs IPSCs.
Collectively, our data suggest that ethanol may potentially enhance pcs synapses via a postsynaptic mechanism or at the very least, via a mechanism that may not involve a presynaptic facilitation of GABA release. Although there is now compelling evidence that ethanol potentiates GABAergic synapses in many brain regions, relatively few studies have demonstrated direct postsynaptic actions of ethanol at concentrations associated with the pharmacological effects of this drug (Siggins et al., 2005; Weiner and Valenzuela, 2006). It is noteworthy that a postsynaptic mechanism was shown to contribute to ethanol enhancement of GABAA IPSCs in the CeA, although this effect was reported to be more variable than presynaptic measures of ethanol facilitation (Roberto et al., 2003, 2004). Interestingly, cells within the CeA also receive inhibitory input from local and intercalated GABAergic neurons, with the latter originating from a population of medial intercalated cells residing along the CeA-BLA border (Royer et al., 1999). It will be interesting in future studies to determine whether the medial intercalated synapses contributed to the apparent postsynaptic effects of ethanol observed in the CeA.
Several recent studies have also reported potent and direct potentiating effects of ethanol on native and recombinant GABAA receptors containing the δ subunit (Wallner et al., 2003; Hanchar et al., 2005; Glykys et al., 2007). Therefore, it is possible that the subunit composition of GABAA receptors at synapses originating from pcs cells contain high levels of the δ subunit. However, δ-containing receptors are typically localized to extrasynaptic loci, and this subunit is not abundantly expressed in the amygdala (Wisden et al., 1992; Pirker et al., 2000). In addition, we did not observe a significant effect of ethanol on the holding current during our recordings, suggesting that, even if there is an appreciable extrasynaptic GABAergic conductance on BLA neurons, these extrasynaptic GABAA receptors are not directly potentiated by low concentrations of ethanol.
Interestingly, in some brain regions where postsynaptic mechanisms have been implicated in ethanol potentiation of GABAergic inhibition, these effects have often been attributed to indirect mechanisms. For example, a neurosteroid-dependent mechanism mediates a postsynaptic effect of ethanol on GABAergic synapses in the hippocampus (Sanna et al., 2004). In addition, in cerebellar Purkinje cells, NE has been shown to directly potentiate GABAA receptor function (Cheun and Yeh, 1996), and pretreatment with NE significantly enhances ethanol-mediated potentiation of GABAergic inhibition of Purkinje cell firing (Lin et al., 1991). Our findings that 20 μM NE potentiated distal, but not local, IPSCs and that pretreatment with a NE antagonist cocktail completely blocked this NE effect as well as ethanol potentiation of distal IPSCs provide initial evidence that NE receptor activation plays an integral role in mediating the effects of ethanol at pcs synapses onto BLA pyramidal neurons. Further experiments will be needed to determine the synaptic locus through which ethanol potentiates pcs IPSCs as well as the specific role that NE plays in this enhancement. Our current data suggest that ethanol may possibly stimulate NE release at pcs synapses, which indirectly mediate ethanol potentiation of these synapses or, alternatively, that NE may be a necessary cofactor that is required for ethanol potentiation of these synapses, as observed in the cerebellum (Lin et al., 1991).
In summary, our findings suggest that ethanol significantly enhances GABAergic synaptic inhibition arising from both local and pcs interneurons in the BLA. Therefore, acute ethanol exposure increases both cortical feed-forward inhibition as well as local feedback inhibition onto the primary excitatory output cells of the BLA. Studies with benzodiazepines, which are potent allosteric modulators of synaptic GABAA receptors, suggest an integral role of BLA inhibition in the regulation of anxiety-like behavior. Microinjection studies have demonstrated that local infusion of benzodiazepines directly into the BLA decreases a wide range of anxiety-like behaviors (Menard and Treit, 1999), and focal injection of a benzodiazepine antagonist into the amygdala can block some of the anxiolytic effects of systemically administered benzodiazepines (Hodges and Green, 1987). Moreover, elegant studies with transgenic knock-in mice with deleted benzodiazepine binding sites on specific GABAA receptor α subunits have shown that α2-containing GABAA receptors, which are heavily expressed at GABAergic synapses in the BLA (Marowsky et al., 2004), mediate many of the anxiolytic effects of benzodiazepines (Suzdak et al., 1986). Collectively, these data suggest that the relatively potent, facilitatory effects of ethanol on local and pcs inhibition in the BLA probably play a significant role in mediating the acute anxiolytic effects of this drug. Given the important role that anxiety is thought to play in the etiology of alcoholism, it will be important in future studies to further resolve the specific mechanisms through which ethanol enhances GABAergic inhibition at both populations of synapses and to examine how these pathways may be influenced by chronic ethanol exposure and withdrawal.
This work was supported by National Institutes of Health Grants AA 13960, AA 11997, and AA10422.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: EtOH, ethanol; BLA, basolateral nucleus of the amygdala; GAD-GFP, glutamate decarboxylase tagged with green fluorescent protein; pcs, paracapsular; IPSC, inhibitory postsynaptic current; QX-314, N-(2,6-dimethyl-phenylcarbamoylmethyl)-triethylammonium chloride; ISI, interstimulus interval; PPR, paired-pulse ratio; PV, parvalbumin; PPD, paired pulse depression; SCH 50911, (+)-(S)-5,5-dimethylmorpholinyl-2-acetic acid; NE, norepinephrine; CeA, central nucleus of the amygdala.
- The American Society for Pharmacology and Experimental Therapeutics