Abstract
Chronic alcohol exposure leads to a neuroinflammatory response involving activation of the nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome and proinflammatory cytokine production. Acute ethanol (EtOH) exposure activates GABAergic synapses in the central and basolateral amygdala (BLA) ex vivo, but whether this rapid modulation of synaptic inhibition is because of an acute inflammatory response and alters anxiety-like behavior in male and female animals is not known. Here, we tested the hypotheses that acute EtOH facilitates inhibitory synaptic transmission in the BLA by activating the NLRP3 inflammasome-dependent acute inflammatory response, that the alcohol-induced increase in inhibition is cell type and sex dependent, and that acute EtOH in the BLA reduces anxiety-like behavior. Acute EtOH application at a binge-like concentration (22–44 mm) stimulated synaptic GABA release from putative parvalbumin (PV) interneurons onto BLA principal neurons in ex vivo brain slices from male, but not female, rats. The EtOH facilitation of synaptic inhibition was blocked by antagonists of the Toll-like receptor 4 (TLR4), the NLRP3 inflammasome, and interleukin-1 receptors, suggesting it was mediated by a rapid local neuroinflammatory response in the BLA. In vivo, bilateral injection of EtOH directly into the BLA produced an acute concentration-dependent reduction in anxiety-like behavior in male but not female rats. These findings demonstrate that acute EtOH in the BLA regulates anxiety-like behavior in a sex-dependent manner and suggest that this effect is associated with presynaptic facilitation of parvalbumin-expressing interneuron inputs to BLA principal neurons via a local NLRP3 inflammasome-dependent neuroimmune response.
SIGNIFICANCE STATEMENT Chronic alcohol exposure produces a neuroinflammatory response, which contributes to alcohol-associated pathologies. Acute alcohol administration increases inhibitory synaptic signaling in the brain, but the mechanism for the rapid alcohol facilitation of inhibitory circuits is unknown. We found that acute ethanol at binge-like concentrations in the basolateral amygdala (BLA) facilitates GABA release from parvalbumin-expressing (PV) interneuron synapses onto principal neurons in ex vivo brain slices from male rats and that intra-BLA ethanol reduces anxiety-like behavior in vivo in male rats, but not female rats. The ethanol (EtOH) facilitation of inhibition in the BLA is mediated by Toll-like receptor 4 (TLR4) and nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome activation and proinflammatory IL-1β signaling, which suggests a rapid NLRP3 inflammasome-dependent neuroimmune cascade that plays a critical role in acute alcohol intoxication.
Introduction
Acute alcohol intoxication has widespread effects in the central nervous system and is a risk factor for the development of alcohol use disorder (AUD; for review, see Vonghia et al., 2008; Jacob and Wang, 2020). Accumulating evidence suggests sex differences in rates and consequences of alcohol use in humans (Agabio et al., 2017; Randall et al., 2017; Peltier et al., 2019) and animals (Vetter-O'Hagen et al., 2009; Albrechet-Souza et al., 2020; Flores-Bonilla and Richardson, 2020). Acute ethanol (EtOH) is anxiolytic (LaBuda and Fuchs, 2001), whereas chronic EtOH exposure leads to heightened anxiety during withdrawal (Kliethermes, 2005). Several studies have shown a role for the central amygdala (CeA; see Gilpin et al., 2015; Roberto et al., 2021) and basolateral amygdala (BLA; see Silberman et al., 2009a,b; McCool et al., 2010; Silberman et al., 2012; Price and McCool, 2022) in alcohol's effects on anxiety-related behaviors.
Acute EtOH regulates inhibitory interneuron circuits in both the CeA (Roberto et al., 2003, 2004; for review, see Roberto et al., 2021) and BLA (Zhu and Lovinger, 2006) by facilitating presynaptic GABA release. EtOH-induced presynaptic modulation of inhibitory transmission in the CeA is specific to male rats (Kirson et al., 2021), but the molecular mechanisms underlying acute EtOH modulation of inhibitory circuits in the amygdala are not well characterized.
Chronic EtOH exposure causes neuroinflammation, which is implicated in the neuropathology of AUD (for review, see Crews et al., 2006; Cui et al., 2014; Erickson et al., 2019). The inflammasome is an intracellular oligomeric protein sensor complex that is part of the innate immune system and is activated by diverse cellular stimuli including toxins, pathogen-associated molecular patterns (PAMPs) and sterile danger signals like danger-associated molecular patterns (DAMPs). The nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome is comprised of the NLR sensor, procaspase-1, and the apoptosis-associated speck-like protein containing caspase activation and recruitment domain (ASC; see Kelley et al., 2019; Swanson et al., 2019). The Toll-like receptor 4 (TLR4) is another innate immune system signal and plays a critical role in EtOH-induced microglial activation and neuroinflammation (Fernandez-Lizarbe et al., 2009). Chronic EtOH treatment activates the NLRP3 inflammasome in seven-week-old female mouse brain via TLR4 signaling (Alfonso-Loeches et al., 2014, 2016). Also, adult male mice treated with chronic EtOH, postmortem brains from human subjects with AUD, and organotypic hippocampal-entorhinal cortex slice cultures treated with EtOH for 4d exhibit neuroimmune activation via high mobility group box 1 (HMGB1)/TLR (TLR2, TLR3, and TLR4) signaling (Crews et al., 2013). However, it is not known whether a single EtOH exposure recruits the TLR4–NLRP3 inflammasome cascade in the BLA to produce acute neuroinflammatory effects.
Polymorphisms of the IL-1 gene family are associated with increased susceptibility to alcohol misuse and dependence (Pastor et al., 2005; Saiz et al., 2009). Recent studies showed that chronic EtOH treatment dampened IL-1β–induced suppression of GABA release in the CeA of nonhuman primates during abstinence (Patel et al., 2022) and increased IL-1β expression in the mouse CeA (Patel et al., 2019), but did not report any interaction between acute IL-1β and EtOH modulation of GABA release in the CeA of either control or EtOH-dependent mice (Bajo et al., 2015a,b; Patel et al., 2019). Collectively, existing evidence suggests the involvement of inflammatory mechanisms in chronic alcohol-induced AUD, but not in the effect of acute alcohol exposure.
Here, we tested the effects of acute “binge-like” EtOH, defined as ≥22 mm (i.e., ≥100 mg/dl), on BLA synaptic physiology and anxiety-like behavior. Our goals were (1) to test whether acute EtOH modulates cell type-specific inhibitory circuits in the BLA in a sex-dependent manner, (2) to determine whether the EtOH response in the BLA is mediated by the TLR4-NLRP3-IL-1β inflammatory cascade, and (3) to determine the effect of EtOH in the BLA on anxiety-like behavior in male and female rats.
Materials and Methods
Animals
For electrophysiology experiments, 8- to 16-week-old male and female Wistar rats (Charles River, stock #003) were group housed (two to three rats of same sex) in a climate-controlled (22°C) animal facility at Tulane University with ad libitum access to food and water and a 12/12 h reversed light/dark cycle (lights off at 7 A.M.). The animals were acclimated to the facility for at least one week before killing for brain slice preparation. All procedures were approved by the Institutional Animal Care and Use Committee of Tulane University and were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 8023, revised 1978). All efforts were made to minimize animal suffering and to reduce the number of animals used. The rats were killed in the morning and recordings were performed between 10 A.M. and 6 P.M., which was in the dark phase of the reversed light-dark cycle.
For behavioral experiments, eight-week-old male and female Wistar rats (Charles River; stock #003) were housed in same-sex pairs in a climate-controlled (22°C) facility in the Louisiana State University Health Sciences Center with ad libitum access to food and water and a 12/12 h reversed light/dark cycle (lights off at 7 A.M.). Animals were handled daily for 3 min for 7–10 d before the initiation of experimental protocols. All procedures were approved by the Institutional Animal Care and Use Committee of the Louisiana State University Health Sciences Center and were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 8023, revised 1978). Behavioral experiments were conducted between 9 A.M. and 12 P.M., which was during the dark phase of the reversed light-dark cycle.
Brain slice preparation
Unanesthetized rats were decapitated in a rodent guillotine using disposable rodent restrainers (DecapiCone; Model DC-200, Braintree Scientific Inc.). Brains were immediately removed and immersed for 1–2 min in ice-chilled cutting solution composed of the following: 252 mm sucrose, 2.5 mm KCl, 26 mm NaHCO3, 1.25 mm NaH2PO4.H2O, 10 mm glucose, 1 mm CaCl2 and 5 mm MgCl2, which was saturated with a continuous flow of carbogen gas (95% O2 and 5% CO2 gas mixture). The osmolarity of the cutting solution was adjusted to 295–305 mOsm and the pH was adjusted to 7.2–7.4 with KOH. Next, brain blocks containing the amygdala were glued to the stage of a vibrating slicer (Pelco 100 Vibratome, Ted Pella). Three to four brain slices of 350 µm thickness were sectioned rapidly on the vibrating slicer in a carbogenated (95% O2 and 5% CO2), ice-cold cutting solution. The slices were then immediately transferred to artificial CSF (aCSF; 290–310 mOsm; pH 7.2–7.4 adjusted with NaOH) containing 126 mm NaCl, 2.5 mm KCl, 26 mm NaHCO3, 1.25 mm NaH2PO4.H2O, 10 mm glucose, 2.5 mm CaCl2 and 1.3 mm MgCl2, which was continuously saturated with the carbogen gas and maintained at 32°C in a water bath (Intertek 110–031 PSC/EN, PolyScience). Each slice was gently bisected down the midline and the hemisections were put back into the aCSF, where they were maintained in the water bath for 60–90 min.
Patch-clamp recordings
Hemi-slices were transferred one at a time from the slice holding chamber at 32°C to a submersion recording chamber where they were continuously perfused with aCSF saturated with carbogen gas at room temperature at a rate of 1.5–2 ml/min. All electrophysiological recordings were performed at room temperature. The BLA principal cells were identified by their location and their teardrop-shaped or pyramidal-shaped soma visualized under infrared differential interference contrast illumination with an infrared-sensitive camera (Dage MTI IR-1000) attached to an upright, fixed-stage microscope (Olympus BX51 WI). Patch-clamp recording electrodes were pulled from borosilicate glass with filament (O.D.: 1.65 mm, I.D.: 1.20 mm, length: 10 cm; item #BF165-120-10, Sutter Instrument) using a horizontal micropipette puller (Model P-97, Sutter Instrument) to a tip resistance of 2–4 MΩ. Whole-cell voltage-clamp electrophysiological data were acquired using a Multiclamp 700A/700B amplifier, a Digidata 1440A/1440B digitizer, and pCLAMP 10 software (Molecular Devices LLC). A built-in bessel filter was used to low pass filter the analog data at 2 kHz, and the data were digitized at 10 kHz. Data were stored on a computer hard drive for offline analysis. IPSC and EPSC recordings were analyzed with Mini Analysis Program version 6.0.7 (Synaptosoft Inc.). Only the cells in which the series resistance (access resistance) did not change by >20% during recordings were included in the analyses.
Recorded cells were allowed to stabilize for ∼10 min before the start of experiments. Following this baseline stabilization, the baseline synaptic activity was recorded for 5 min before a 5-min EtOH application followed by a 5-min washout. For experiments in which calcium channels were blocked, slices were preincubated for 45–60 min in either the P/Q-type calcium channel inhibitor ω-Agatoxin IVA (250 nm; Alomone Labs) or the N-type calcium channel inhibitor ω-Conotoxin GVIA (1 μm; Alomone Labs) at 32°C. For experiments in which the NLRP3 inflammasome was targeted, slices were preincubated for 90–120 min at 32°C in the NLRP3 inhibitor MCC950 (10 μm; InvivoGen Inc.). For experiments where TLR4 was targeted, slices were preincubated for ≥45 min in the TLR4 inhibitor TAK-242 (50 nm; Apexbio Technology LLC). The slices were then transferred to the recording chamber where they were perfused with a continuously gassed (95% O2 and 5% CO2) aCSF containing the respective inhibitors at room temperature.
To record IPSCs, whole-cell patch-clamp recordings were performed in putative BLA principal neurons at room temperature in aCSF containing 20 μm DNQX and 50 μm AP5 (Hello Bio Inc.) to block glutamate AMPA and NMDA receptor-mediated currents, respectively. Cells were voltage clamped at −70 mV to record the IPSCs with a cesium-based, high-Cl– patch solution (∼290 mOsm, pH 7.25–7.30 adjusted with CsOH) composed of 110 mm CsCl, 30 mm K-gluconate, 1.1 mm EGTA, 10 mm HEPES, 0.1 mm CaCl2, 4 mm Mg-ATP, and 0.3 mm Na-GTP. QX-314 (4 mm) was included in the patch solution to prevent presynaptic action potential generation by reversed IPSCs. Spontaneous IPSCs (sIPSCs) were recorded without tetrodotoxin (TTX) and miniature IPSCs (mIPSCs) were recorded with TTX (1 μm) in the bath solution to block presynaptic action potential generation.
For detection of EPSCs, whole-cell patch-clamp recordings were performed in putative BLA principal neurons at room temperature in aCSF containing 100 μm picrotoxin to block GABAA receptor-mediated currents. Cells were voltage clamped at −70 mV with a low-Cl− patch solution (∼290 mOsm, pH 7.25–7.30 adjusted with KOH) composed of 130 mm K-gluconate, 10 mm HEPES, 10 mm phosphocreatine disodium, 4 mm Mg-ATP, 0.4 mm Na-GTP, 5 mm KCl, 0.6 mm EGTA, and 4 mm QX-314. To record miniature EPSCs (mEPSCs), TTX (1 μm) was added to the perfusion medium to block presynaptic action potentials.
Drugs used in electrophysiological experiments
All drugs used in the study were purchased from Sigma-Aldrich unless otherwise specified. Pure grade EtOH (≥99.5%) was diluted to its final concentrations in aCSF. The P/Q-type calcium channel antagonist ω-Agatoxin IVA (Alomone Labs) was dissolved in aCSF to prepare a 10 μm stock solution, which was aliquoted and stored at −20°C until the day of the experiment, when a final concentration of 250 nm was prepared by diluting the stock solution in aCSF. The N-type calcium channel antagonist ω-Conotoxin GVIA (Alomone Labs) was dissolved in aCSF to prepare a 100 μm stock solution, which was aliquoted and stored at −20°C until the day of the experiment, when a final concentration of 1 μm was prepared by dilution in aCSF. The NLRP3 inflammasome inhibitor MCC950 (InvivoGen Inc.) was dissolved in DMSO to prepare a 20 mm stock solution, which was stored at −20°C until the day of the experiment, when a final concentration of 10 μm was prepared by dilution in aCSF. The recombinant human IL-1RA was dissolved in aCSF and stored at −20°C in 100 µL aliquots. The TLR4 antagonist TAK-242 (Apexbio Technology LLC) was dissolved to 10 mm in DMSO to prepare a stock solution, which was aliquoted and kept at −20°C until the day of the experiment when the aliquots were diluted in aCSF to a final concentration of 50 nm.
Stereotaxic surgery
Male and female rats were anesthetized with isoflurane and surgically implanted with 26-gauge cannulas (Plastics One) aimed at the BLA. The stereotaxic coordinates according to Paxinos and Watson (2007) were: 2.4 mm posterior to bregma, ± 4.9 mm lateral to the midline, and 7.7 mm ventral to the skull. Following surgeries, rats were monitored to ensure recovery from anesthesia and were given 5–7 d to recover before experiments were performed. Rats were treated with the analgesic flunixin (2.5 mg/kg, s.c.) and the antibiotic cefazolin (20 mg/kg, i.m.) before the start of surgeries and once the following day.
In vivo stereotaxic injection
Reagent-grade 95% EtOH was diluted in 0.9% saline solution immediately before administration. On the day before the test day, rats received one sham injection, consisting of insertion of the 30-gauge injection needles into the cannulas for 2 min. On the test day, rats received bilateral intra-BLA infusions of EtOH (22 or 44 mm) or saline using 10-μl Hamilton microsyringes connected to the cannulas by polyethylene tubing (PE-20) and controlled by an infusion syringe pump (KD Scientific Inc.). The injection needles protruded 1 mm beyond the tips of the cannulas, and the injection volume (0.5 μl) was infused over a period of 60 s. The injection needles were retained within the cannulas for an additional 60 s after drug infusion to maximize diffusion and minimize EtOH backflow into the cannulas.
Elevated plus maze test
The elevated plus maze (EPM) was a black Plexiglas apparatus consisting of two closed arms (50 × 10 × 40 cm) and two open arms (50 × 10 cm) attached to metal legs elevating the maze 50 cm above the ground. All testing was conducted under dim illumination (∼10 lux). Rats were placed individually in the center of the maze facing a closed arm and allowed 5 min of free exploration. Behavior was recorded with a video camera positioned above the maze. The EPM was cleaned thoroughly between subjects using Quatricide (Pharmacal Research Labs). The absolute and the percent time the rats spent in the open arms, the number of open arm entries, and the number of closed arm entries were scored by an observer blind to the conditions. Entry into an arm was defined as all four paws entering the arm.
Statistical analysis
IPSCs and EPSCs were analyzed using the Mini Analysis Program version 6.0.7 (Synaptosoft Inc.). For all datasets, comparison among three or more groups was done by one-way ANOVA. In cases of a significant ANOVA, post hoc comparisons were performed using Tukey's multiple comparisons test. For comparison of PSCs recorded during baseline and drug treatments, averages of the last 3 min of recordings of baseline and of drug treatments were compared. When data were normally distributed, two groups were compared by a paired (within-cell comparison) or unpaired (between-cell comparison) Student's t test. When data were not normally distributed, two groups were compared by a Wilcoxon matched-pairs signed rank test (within-cell comparison) or Mann–Whitney test (between-cell comparison). Statistical analyses were performed using Prism 9 (GraphPad). All values are expressed as mean ± SEM. Statistical significance was set at p < 0.05.
Results
We first recorded spontaneous IPSCs (sIPSCs) and miniature IPSCs (mIPSCs) in principal neurons of the BLA using whole-cell patch clamp recordings in ex vivo slices of amygdala from male and female rats to determine the acute effect of bath application of 44 mm EtOH on basal inhibitory synaptic input. The IPSCs were recorded as inward membrane currents because of the high [Cl−] in the patch pipette. There was no difference in the baseline frequency, amplitude, or decay time of sIPSCs between principal neurons recorded in slices from males (n = 9 cells from 9 rats) and females (n = 11 cells from 10 rats). The basal sIPSC frequency was 3.81 ± 0.78 Hz in males and 3.04 ± 0.61 Hz in females (U = 41, p = 0.55); the basal sIPSC amplitude was 59.99 ± 9.79 pA in males and 48.66 ± 6.83 pA in females (U = 31, p = 0.18); and the basal sIPSC decay time was 10.2 ± 1.4 ms in males and 9.9 ± 0.8 ms in females (t = 0.184, df = 18, p = 0.86).
EtOH increases inhibitory synaptic inputs to BLA principal neurons
A 5-min bath application of EtOH (Fig. 1A) caused a concentration-dependent increase in the frequency of sIPSCs but had no effect on the sIPSC amplitude or decay time (from the peak of the IPSC to 63% decay to baseline) in BLA principal neurons in slices from male rats (Fig. 1B). EtOH caused a similar increase in the frequency of miniature IPSCs (mIPSCs) recorded in the presence of TTX (1 μm) and had no effect on the mIPSC amplitude or decay time in principal neurons from male rats. There was no difference between the EtOH-induced changes in sIPSC and mIPSC frequencies (Fig. 1C,D). The EtOH effect in principal neurons from male rats was concentration-dependent, at 22 mm causing an ∼80% increase in mIPSC frequency that did not reach significance (t = 1.465, df = 6, p = 0.19) and at 44 mm causing a significant ∼twofold increase in mIPSC frequency (t = 2.492, df = 7, p = 0.04; Fig. 1E). These data suggest that EtOH stimulates GABA release from presynaptic inhibitory interneurons in a concentration-dependent, action potential-independent fashion, suggesting that EtOH acts on GABA neuron axon terminals to facilitate GABA release onto the principal neurons of the BLA in male rats.
Sex-dependent increase in GABAergic IPSCs by acute EtOH in BLA principal neurons. A, Experimental timeline for whole-cell recordings. After a ∼10-min stabilization period, baseline was recorded for 5 min followed by EtOH application for 5 min and washout. When antagonists were applied, slices were preincubated in antagonist before EtOH application. B, Representative sIPSC (left; blue) and mIPSC (right; red) traces during baseline recording (top), in the presence of EtOH (middle), and during EtOH washout (bottom). C, In slices from male rats, EtOH (44 mm) induced an increase in sIPSC frequency (blue symbols; t = 2.865, df = 8, p = 0.02, two-tailed paired t test; n = 9 cells from 9 rats). In TTX, EtOH (44 mm) caused a similar increase in the frequency of mIPSCs (red symbols; t = 2.492, df = 7, p = 0.04, two-tailed paired t test; n = 8 cells from 6 rats). D, Summary graphs showing no difference between the effects of EtOH (44 mm) on sIPSC and mIPSC frequencies (t = 0.209, df = 15, p = 0.84, two-tailed unpaired t test), amplitudes and decay times. EtOH (44 mm) had no effect on sIPSC or mIPSC amplitudes (sIPSCs: t = 0.374, df = 8, p = 0.72; mIPSCs: t = 0.108, df = 7, p = 0.92) and decay times (sIPSCs: t = 1.508, df = 8, p = 0.17; mIPSCs: t = 0.674, df = 7, p = 0.52). E, The EtOH-induced increase in mIPSC frequency in slices from male rats was concentration dependent; the increase was significant at 44 mm (t = 2.492, df = 7, p = 0.04; two-tailed paired t test), but not at 22 mm (t = 1.465, df = 6, p = 0.19; two-tailed paired t test; n = 7 cells from 5 rats). F, In slices from female rats, EtOH did not elicit a significant increase in sIPSC frequency compared with baseline at 44 mm (t = 0.291, df = 10, p = 0.78, two-tailed paired t test; n = 11 cells from 10 rats) or 88 mm (t = 1.611, df = 9, p = 0.14, two-tailed paired t test; n = 10 cells from 7 rats). *p < 0.05.
In BLA principal neurons from female rats, EtOH did not have a significant effect on sIPSC frequency (p = 0.78), amplitude (p = 0.72), or decay time (p = 0.92) relative to the baseline at 44 mm or when the concentration was doubled to 88 mm (p = 0.14, 0.33, 0.82, respectively; Fig. 1F). The sIPSC frequency in 88 mm EtOH was not significantly different from that in 44 mm EtOH (t = 1.412, df = 19, p = 0.17). Therefore, binge-like EtOH concentrations of 22–44 mm facilitated inhibitory synaptic transmission in BLA principal neurons from male rats, but not in female rats even at a high EtOH concentration (44–88 mm), suggesting a pronounced sex difference in the sensitivity to EtOH of inhibitory synaptic transmission in the BLA ex vivo.
EtOH facilitation of inhibitory inputs is P/Q-type Ca2+ channel-dependent
Parvalbumin-expressing (PV) and cholecystokinin-expressing (CCK) interneurons comprise the basket cells that form perisomatic inhibitory synapses on BLA principal neurons (Freund and Katona, 2007). The two subtypes of basket interneurons can be distinguished based on the expression of different types of axonal voltage-gated Ca2+ channels: P/Q-type Ca2+ channels in PV neurons (Hefft and Jonas, 2005; Zaitsev et al., 2007) and N-type Ca2+ channels in CCK neurons (Wilson et al., 2001; Hefft and Jonas, 2005; Blazon et al., 2021). We next tested whether the presynaptic EtOH facilitation of GABA release in the BLA from male rats was dependent on voltage-gated Ca2+ channels specific to PV or CCK basket cells using selective Ca2+ channel antagonists. Slices were preincubated in aCSF containing the P/Q-type Ca2+ channel blocker ω-Agatoxin IVA (AgTx, 250 nm) or the N-type Ca2+ channel blocker ω-Conotoxin GVIA (CgTx, 1 μm) for 45–60 min at 32°C before recordings. The increase in sIPSC frequency induced by EtOH (44 mm) was blocked by bath application of the P/Q-type Ca2+ channel blocker ω-Agatoxin IVA (250 nm), but not by the N-type Ca2+ channel blocker ω-Conotoxin GVIA (1 μm; Fig. 2). These data suggest that EtOH facilitation of inhibitory synaptic inputs to BLA principal neurons in male rats is P/Q-type Ca2+ channel dependent and implicate PV-expressing basket cells in the EtOH modulation of synaptic inhibition in the BLA.
EtOH modulation of synaptic inhibition is P/Q-type Ca2+ channel dependent. A, B, Representative traces of sIPSCs in slices from a male rat before (top), during (middle), and after washout (bottom) of 44 mm EtOH in the presence of (A) the N-type Ca2+ channel blocker CgTx (1 μm) and (B) the P/Q-type Ca2+ channel blocker AgTx (250 nm). C, Time course of the effect of 44 mm EtOH on sIPSC frequency in BLA principal neurons from male rats normalized to the baseline. The EtOH-induced increase in sIPSC frequency was blocked in the presence of the P/Q-channel blocker AgTx (t = 0.384, df = 5, p = 0.72; two-tailed paired t test; n = 6 cells from 5 rats), but not in the N-channel blocker CgTx (t = 3.677, df = 6, p = 0.01; two-tailed paired t test; n = 7 cells from 7 rats). D, Summary of the changes in mean sIPSC frequency. Left, Changes in sIPSC frequencies in CgTx and AgTx. Right, Changes in mean sIPSC frequencies relative to baseline. *p < 0.05, **p < 0.01.
Intra-BLA EtOH produces anxiolytic effects in male but not female rats
We next tested the effect of acute EtOH application in the BLA on anxiety-like behavior in male and female rats. EtOH (22 mm or 44 mm) was injected bilaterally directly into the BLA, and anxiety-related behavior was tested in the EPM. Cannula placements were confirmed through histologic examination, and only animals with accurate bilateral placement were included in the final analysis. Intra-BLA injection of 22 mm EtOH, but not 44 mm EtOH, caused a significant increase in the time spent in the open arms (F(2,22) = 9.446, p = 0.001), in the percentage of time spent in the open arms (F(2,22) = 9.291, p = 0.001) and in the number of open arm entries (F(2,22) = 8.039, p = 0.002) in male rats (Fig. 3). There was no effect of intra-BLA EtOH infusion on the number of closed-arm entries (F(2,22) = 0.433, p = 0.65) in male rats, suggesting this treatment did not have a nonspecific effect on locomotor activity in this assay. In contrast, there was no significant effect in female rats of either 22 mm or 44 mm EtOH on the time spent in the open arms (F(2,25) = 0.352, p = 0.70), percentage of time spent in the open arms (F(2,25) = 0.205, p = 0.81), number of open-arm entries (F(2,25) = 0.669, p = 0.52), or closed-arm entries (F(2,25) = 0.265, p = 0.77; Fig. 3). In order to determine whether there was a sex difference in the EtOH effect on anxiety-like behavior, we performed a 2-way ANOVA with sex and treatment as factors. Although there was no significant sex-by-treatment interaction (time in the open arms: F(2,47) = 3.101, p = 0.054; percent time in the open arms: F(2,47) = 2.003, p = 0.15; number of open-arm entries: F(2,47) = 2.105, p = 0.13), there were significant effects of sex (time in the open arms: F(1,47) = 12.16, p = 0.001; percent time in the open arms: F(1,47) = 12.84, p = 0.0008; number of open-arm entries: F(1,47) = 12.40, p = 0.001) and treatment (time in the open arms: F(2,47) = 4.366, p = 0.01; percent time in the open arms: F(2,47) = 3.945, p = 0.02; number of open-arm entries: F(2,47) = 4.033, p = 0.02). These data suggest that intra-BLA EtOH reduces anxiety-like behavior in a concentration-dependent manner in male but not female rats.
Intra-BLA EtOH infusion promotes a dose-dependent anxiolytic-like effect in male, but not in female rats. Male and female rats were treated with bilateral intra-BLA infusions (0.5 µl per side) of EtOH (22 mm or 44 mm) or saline and tested in the elevated plus maze for anxiety-like behavior. Male rats treated with 22 mm EtOH (n = 9), but not 44 mm EtOH (n = 8), showed a significant increase in (A) the time spent in the open arms (F(2,22) = 9.446, p = 0.001), (B) the % time spent in the open arms (F(2,22) = 9.291, p = 0.001) and (C) the number of open-arm entries (F(2,22) = 8.039, p = 0.002), but not in (D) the number of entries in the closed arms (F(2,22) = 0.433, p = 0.65) compared with controls (n = 8). No effect of 22 mm (n = 10) or 44 mm EtOH (n = 9) was seen on any of these measures in female rats compared with controls (n = 9). *p < 0.05, **p < 0.01; one-way ANOVA followed by Tukey's post hoc test. E, Correct placements of intra-BLA cannulas in male and female rats. The number of points in the figure is less than the total number of animals because of overlapping injection sites.
EtOH has no effect on excitatory synaptic input to BLA principal neurons
We next asked whether the lack of effect of the intra-BLA EtOH on anxiety-like behavior at the 44 mm concentration in male rats, despite its increased effect on synaptic inhibition, might be attributable to an emergent activation by EtOH of excitatory synaptic circuits in the BLA at higher concentrations, which could oppose the effect of the EtOH-induced increase in inhibitory synaptic transmission on anxiety-like behavior. To test this, we again turned to ex vivo slices of amygdala to perform whole-cell recordings of miniature EPSCs and spontaneous EPSCs (mEPSCs and sEPSCs, respectively) in BLA principal neurons. Bath application of 44 mm EtOH had no effect on the frequency (t = 0.405, df = 5, p = 0.70), amplitude (t = 0.785, df = 5, p = 0.47), or decay time (t = 0.174, df = 5, p = 0.87) of mEPSCs (Fig. 4), or on the frequency (t = 0.585, df = 5, p = 0.58), amplitude (t = 0.640, df = 5, p = 0.55), or decay time (t = 0.998, df = 5, p = 0.36) of sEPSCs (n = 6 cells from 4 rats) in BLA principal neurons in slices from male rats. These data demonstrate a lack of effect of 44 mm EtOH on basal glutamate release and sensitivity.
EtOH has no effect on mEPSCs in BLA principal neurons of male rats. A, Time series of mean mEPSC frequency normalized to baseline. EtOH (44 mm) was applied at 5 min for a duration of 5 min. B, Mean (±SEM) mEPSC frequency (Freq), amplitude (Amp), and decay time (Decay) during baseline and EtOH application. EtOH had no effect on mEPSC frequency (t = 0.405, df = 5, p = 0.70), amplitude (t = 0.785, df = 5, p = 0.47), or decay (t = 0.174, df = 5, p = 0.87; two-tailed paired t test; n = 6 cells from 3 rats). C, Representative traces of mEPSCs at baseline and during the EtOH application at compressed (top) and expanded (bottom) time scales.
The acute EtOH effect on BLA inhibitory transmission requires activation of the NLRP3 inflammasome
Several lines of evidence suggest the involvement of the innate immune system in the pathophysiological effects of chronic alcohol (for review, see Gao et al., 2011; Szabo and Saha, 2015; Coleman and Crews, 2018). Here, we tested the hypothesis that EtOH increases GABAergic transmission in the BLA by rapidly recruiting an intracellular inflammation signal, the NLRP3 inflammasome. Preincubation of slices in the NLRP3 inflammasome inhibitor MCC950 (10 μm) had no significant effect on baseline sIPSC frequency (2.948 ± 0.887 Hz, p = 0.48), amplitude (55.367 ± 7.008 pA, p = 0.71) or decay time (9.877 ± 1.465 ms, p = 0.88) compared with control aCSF, suggesting that MCC950 alone does not change basal GABA release or sensitivity. However, the increase in sIPSC frequency in BLA principal neurons from male rats induced by 44 mm EtOH was blocked in the presence of the NLRP3 inhibitor MCC950 (10 μm; Fig. 5). This suggests that the EtOH-induced increase in presynaptic GABA release is dependent on the rapid activation of an NLRP3 inflammasome-dependent cellular inflammatory response.
The EtOH-induced increase in sIPSC frequency is dependent on activation of the NLRP3 inflammasome. A, sIPSCs recorded in a BLA principal neuron from a male rat during baseline recording in the NLRP3 inflammasome inhibitor MCC950 (10 μm) and after co-application of EtOH (44 mm). B, Time series of mean sIPSC frequency recorded in MCC950, normalized to baseline. EtOH (44 mm) was applied at 5 min for a duration of 5 min. C, Summary histograms of the EtOH effects on the mean (±SEM) sIPSC frequency (Hz), amplitude (pA) and decay time (ms; top) and on the normalized mean (±SEM) sIPSC frequency, amplitude, and decay time relative to baseline (bottom). EtOH had no effect on the sIPSC frequency (p = 0.09; Wilcoxon matched-pairs signed rank test), amplitude (p = 0.84; Wilcoxon matched-pairs signed rank test), or decay time (p > 0.99; paired t test; n = 6 cells from 4 rats).
The acute EtOH facilitation of sIPSCs in BLA principal neurons is IL-1 receptor dependent
Stimulation of the innate immune response via NLRP3 inflammasome activation leads to the production of the cytokines IL-1β and IL-18 and downstream activation of IL-1 receptors as part of the cellular inflammatory response. Therefore, we next tested for the proinflammatory cytokine dependence of acute EtOH facilitation of synaptic inhibition in the BLA using the IL-1 receptor antagonist (IL-1RA). Bath application of IL-1RA (200 ng/ml, 11.63 nm) alone caused a significant decrease in the sIPSC frequency (from 3.511 ± 0.998 to 2.415 ± 0.829 Hz; F(1.3,6.6) = 16.36, p = 0.004; post hoc p = 0.003) and amplitude (from 61.942 ± 13.074 to 51.446 ± 9.313 pA; F(1.6,8) = 9.866, p = 0.009; post hoc p = 0.04), without affecting the sIPSC decay time (from 16.940 ± 1.216 to 18.117 ± 6.119 ms; F(1.1,5.6) = 0.130, p = 0.76; post hoc p = 0.83). Subsequent addition of EtOH (44 mm) failed to further change the sIPSC frequency (p > 0.999), amplitude (p = 0.60), or decay time (p = 0.88) compared with the IL-1RA alone (Fig. 6).
The acute EtOH effect requires activation of the IL-1 receptor. A, Representative continuous sIPSC recording from a BLA principal neuron during IL-1RA and EtOH application and washout. B, Segments of a representative sIPSC recording from a different BLA principal neuron at baseline, during IL-1RA, IL-1RA + EtOH applications, and with washout at compressed (left) and expanded (right) time scales. C, Summary of the normalized mean changes in sIPSC frequency, amplitude, and decay time. IL-1RA alone reduced the sIPSC frequency (from 3.511 ± 0.998 Hz to 2.415 ± 0.829 Hz; F(1.3,6.6) = 16.36, p = 0.003) and amplitude (from 61.942 ± 13.074 pA to 51.446 ± 9.313 pA; F(1.6,8) = 9.866, p = 0.04; n = 6 cells from 5 rats). EtOH in IL-1RA failed to change the sIPSC frequency (p > 0.999) or amplitude (p = 0.60). *p < 0.05, **p < 0.01, repeated measures ANOVA followed by Tukey's multiple comparison test.
The acute EtOH facilitation of sIPSCs in BLA principal neurons is TLR4 dependent
Previous studies have shown that chronic EtOH activates TLR4 to recruit the NLRP3 inflammasome in the adult mouse brain (Alfonso-Loeches et al., 2014, 2016). To test the hypothesis that the acute facilitatory effect of EtOH on GABA release in the BLA is mediated by TLR4 activation, we blocked TLR4 during EtOH application with TAK-242. Brain slices were transferred after cutting into aCSF containing TAK-242 (50 nm) and were incubated for ≥45 min before transfer to the recording chamber and start of the experiment following a ≥15-min incubation. The sIPSC frequency (Fig. 7C) was not affected by the presence of the antagonist (F(1,15) = 2.782, p = 0.12), however, there was a significant effect of EtOH on sIPSC frequency (F(1,15) = 7.373, p = 0.02), with significant interaction between factors (F(1,15) = 5.236, p = 0.04). Post hoc analysis of sIPSC frequency revealed that EtOH significantly increased frequency in the absence of antagonist (p = 0.001) but not in the presence of antagonist (p = 0.78). The sIPSC amplitude (Fig. 7C) was not affected by the antagonist or EtOH, with no significant interaction between factors. The sIPSC decay time (Fig. 7C) was significantly increased by the antagonist (F(1,15) = 7.622, p = 0.01), but there was no significant effect of EtOH (F(1,15) = 2.408, p = 0.14) or interaction between factors (F(1,15) = 0.814, p = 0.38). These results indicate that the TLR4 is necessary for the acute effects of EtOH on GABA release onto BLA principal neurons and suggest that EtOH activates TLR4 to recruit the NLRP3 inflammatory cascade.
TLR4 antagonist prevents the acute EtOH-induced increase in sIPSC frequency in BLA principal neurons. A, Representative traces of sIPSCs in BLA principal neurons recorded in slices from male rats showing sIPSCs at baseline (left; blue) and following EtOH (44 mm; right; green) application with (top, compressed and expanded time scales) and without (bottom, compressed and expanded time scales) the TLR4 antagonist TAK-242. B, Time series of the effect of EtOH on the mean normalized sIPSC frequency recorded in BLA principal neurons in the presence and absence of the TLR4 antagonist. The TLR4 antagonist blocked the EtOH-induced increase in sIPSC frequency (F(1,15) = 5.236, p = 0.04, repeated measures two-way ANOVA). C, Summary graphs showing the EtOH-induced increase in sIPSC frequency was abolished by blocking TLR4 with TAK-242; there was no effect of TAK-242 or EtOH on the sIPSC amplitude; and the sIPSC decay time increased in the presence of the TLR4 antagonist, but there was no effect of EtOH on the sIPSC decay time. ns = not significant, *p < 0.05, **p < 0.01.
Discussion
Here, we report an acute, sex-dependent modulation of inhibitory synaptic circuits and anxiety-like behavior by EtOH in the BLA. Acute EtOH at a “binge-like” concentration induced an increase in the frequency of sIPSCs in BLA principal neurons in slices from male, but not female, rats. The rapid EtOH facilitation of inhibitory synaptic transmission in the BLA was blocked by inhibiting the cellular inflammatory response. Acute EtOH applied directly into the BLA in vivo reduced anxiety-like behavior in male, but not female, rats. Collectively, our data suggest that acute EtOH in the BLA triggers a cellular inflammatory response that increases inhibitory synaptic signaling and anxiolytic-like behavior in male, but not female, rats.
The site of EtOH modulation of inhibitory synapses was likely presynaptic, at GABAergic axon terminals, since it altered release probability (i.e., IPSC frequency) but not postsynaptic parameters (i.e., IPSC amplitude and decay kinetics) and was not spike dependent. This suggests, therefore, that EtOH modulates GABA release, as it does in the CeA (Varodayan et al., 2017), but not postsynaptic GABAA receptor channel opening or signaling in BLA principal neurons, unlike the postsynaptic EtOH effects reported in the CeA (Varodayan et al., 2017).
In male rats, the EtOH effect on synaptic GABA release and anxiety-like behavior was concentration dependent. Increasing the concentration of EtOH from 22 to 44 mm caused an increase in the inhibitory synaptic response in BLA slices from male rats, but the acute intra-BLA application of EtOH in vivo caused an anxiolytic response in male rats only at the lower concentration. This suggests that EtOH may elicit an effect on anxiety-like behavior at the higher concentration that opposes the effect of the increase in synaptic inhibition. The lack of anxiolytic effect of EtOH at 44 mm was not caused by the emergence of EtOH facilitation of excitatory synaptic signaling since EtOH failed to increase mEPSCs or sEPSCs (although we did not test for effects on NMDA receptor currents and cannot, therefore, rule this out; Lovinger and Roberto, 2013). EtOH had no effect on the waveform of m/sIPSCs or m/sEPSCs, which suggests that postsynaptic modulation of GABAA and AMPA receptors is also not responsible for the lack of behavioral effect of EtOH at 44 mm. The lack of anxiolytic-like effect of intra-BLA EtOH at the higher concentration, therefore, is likely because of a postsynaptic mechanism distinct from modulation of baseline excitatory and inhibitory synaptic signaling (Martin et al., 2004; Federici et al., 2009; Velázquez-Marrero et al., 2014). EtOH produced an inverted U-shaped concentration effect on inhibitory synaptic signaling in the CeA (Roberto et al., 2003), hippocampus (Wan et al., 1996), and nucleus accumbens (Nie et al., 2000), which we did not observe in the BLA at the concentrations we tested.
We did not observe a significant increase in GABA release or anxiety-like behavior in response to EtOH in female rats. We pushed the EtOH concentration to 88 mm, but even this high concentration failed to produce a statistically significant increase over baseline in GABAergic signaling in slices from female rats. Considering that a blood alcohol concentration of 88 mm (equivalent to >400 mg/dl) is usually toxic, this suggests that female rats exhibit a relative resistance to acute EtOH effects at BLA inhibitory synapses compared with males. This was supported by the sex differences in acute EtOH sensitivity observed in our behavioral experiments. A similar sexually dimorphic sensitivity to the synaptic effects of acute EtOH was reported in a recent study in the CeA (Kirson et al., 2021). Collectively, our findings suggest that female rats exhibit a relative resistance to intra-BLA EtOH compared with males, which could contribute to differences in the sensitivity to acute alcohol exposure between sexes.
Compared with males, female rats displayed lower baseline anxiety-like behavior in the EPM. Indeed, the effect of intra-BLA injection of 22 mm EtOH in males brought them to the same level of performance on the EPM as that displayed by saline-injected females. This suggests that female rats display less basal anxiety-like behavior in the EPM than males, and that intra-BLA EtOH reduces anxiety-like behavior in males to the baseline level of females. This difference should be interpreted with caution, however, since males and females use distinct behavioral strategies to navigate threatening environments (Gruene et al., 2015; Orsini et al., 2016; see Shansky, 2015, 2018; Orsini and Setlow, 2017). Nevertheless, this is consistent with the sex-specific effect of EtOH on BLA inhibitory circuits and raises the possibility of tonic inhibitory control of BLA principal neurons in females that is not present in males. We found no difference in basal sIPSC frequency, amplitude or decay kinetics between males and females, which suggests similar baseline stochastic inhibitory inputs across sexes. Further investigation of tonic GABAA receptor and GABAB receptor activation in BLA principal neurons is required to determine whether sex differences in GABAergic synaptic tone may underlie differences in baseline inhibition.
We found that the facilitatory effect of EtOH on GABA release in the BLA is dependent on P/Q-type Ca2+ channels. Roberto and colleagues found a similar P/Q channel dependence of EtOH facilitation of GABA release in the CeA (Varodayan et al., 2017). This suggests that EtOH modulates voltage-gated P/Q channel opening in the absence of action potentials, indicative of a spike-independent depolarization of the presynaptic terminal, a depolarizing shift in the voltage dependence of the Ca2+ channels, or a voltage-independent modulation of the channels. Also, it suggests that EtOH modulation occurs at parvalbumin-expressing GABAergic interneuron synapses onto BLA principal neurons, since release from parvalbumin cells is P/Q Ca2+ channel dependent (Freund and Katona, 2007; Fu et al., 2022).
Acute EtOH modulation of GABAergic synaptic inputs to BLA principal cells was abolished by blocking activation of the NLRP3 inflammasome and by inhibiting TLR4 and IL-1 receptors, suggesting that the EtOH effect depends on an acute inflammatory response and the release of proinflammatory cytokine in the BLA. This effect of EtOH was initiated within 2–3 min and peaked in ∼5 min. While chronic EtOH exposure has been shown to activate the NLRP3 inflammasome in neurons and neural progenitor cells (Lippai et al., 2013; Alfonso-Loeches et al., 2014; De Filippis et al., 2016; Priyanka et al., 2020), this is the first evidence, to our knowledge, of acute NLRP3 inflammasome activation that modulates synaptic transmission.
Activation of the NLRP3 inflammasome involves a two-step process, a transcriptional priming step followed by an activation step triggered by PAMPs or DAMPs, which leads to IL-1β and IL-18 synthesis. Chronic EtOH treatment in mice leads to TLR4-dependent activation of NLRP3 in cortical and cultured microglial cells and activates caspase-1 to cause proinflammatory cytokine release (Vallés et al., 2004; Alfonso-Loeches et al., 2014, 2016), but this effect was not tested with acute EtOH exposure. EtOH can cause a TLR4-dependent phosphorylation of MAP kinases in cortical astrocytes and microglia within 10 min, and this results in a rapid inflammatory response and proinflammatory cytokine release (Blanco et al., 2005; Fernandez-Lizarbe et al., 2009). A rapid, nontranscriptional mechanism involving TLR4 activation has been shown to prime the NLRP3 inflammasome in macrophages (Juliana et al., 2012). Our findings suggest that acute TLR4 activation by EtOH leads to a rapid NLRP3 assembly and production of IL-1β. It remains to be determined whether this rapid inflammatory response occurs in microglia, astrocytes and/or neurons in the BLA.
Blocking IL-1R with the IL-1RA alone (i.e., in the absence of exogenous agonist) significantly decreased both the sIPSC frequency and amplitude, indicating either a constitutive activation of IL-1R or tonic IL-1β release in BLA brain slices. This suggests both presynaptic and postsynaptic modulation of GABA release and GABAA receptors by IL-1β, which is consistent with presynaptic and postsynaptic effects of IL-1R activation reported in the CeA (Bajo et al., 2015b; Patel et al., 2019). If EtOH acts by activating the NLRP3 inflammasome and increasing IL-1β release, then the lack of postsynaptic effect of the EtOH-induced IL-1β actions may be explained either by saturation of the postsynaptic IL-1 receptors (i.e., because of tonic basal IL-1β release) or by an IL-1β release site near the presynaptic IL-1 receptors and distant from the postsynaptic receptors. Modulation of GABA release by the IL-1RA alone suggests near-saturating activation of the IL-1R by tonic IL-1β release at baseline, which raises the interesting possibility that a difference in tonic release of IL-1β may contribute to the different EtOH sensitivities in the BLA of males and females.
Previous studies reported that IL-1β and IL-1RA modulate spontaneous GABA release in the CeA, but they do not block or occlude the acute EtOH facilitation of GABA release (Patel et al., 2019). Our finding of IL-1RA blockade of acute EtOH facilitation of GABA release in the BLA suggests that there are divergent mechanisms of EtOH actions in the BLA and CeA. Whether IL-1β released in the BLA in response to EtOH acts directly at IL-1 receptors on GABA axons or via an intermediate intercellular signal to modulate GABA release will require further study.
Footnotes
This study was supported by the National Institutes of Health multi-PI Award R01 AA026531 (to N.W.G. and J.G.T.). This work was also supported in part by Merit Review Awards I01 BX003451 (to N.W.G.) and I01 BX005118 (to J.G.T.) from the United States Department of Veterans Affairs, Biomedical Laboratory Research and Development Service, and by the Catherine and Hunter Pierson Chair in Neuroscience (J.G.T.). We thank Dr. Laura Harrison for her expert scientific consultation on this project.
N.W.G. owns shares in Glauser Life Sciences, a company with interest in developing therapeutics for mental health disorders. There is no direct link between those interests and the work contained herein. All other authors declare no competing financial interests.
- Correspondence should be addressed to Jeffrey G. Tasker at tasker{at}tulane.edu
