Activation of α₆-containing GABA_A receptors induces antinociception under physiological and pathological conditions

Abstract

The loss of GABAergic inhibition is a mechanism that underlies neuropathic pain. Therefore, rescuing the GABAergic inhibitory tone through activation of GABA_A receptors is a strategy to reduce neuropathic pain. This study was designed to elucidate the function of the spinal α₆-containing GABA_A receptor in physiological conditions and neuropathic pain in female and male rats. Results show that α₆-containing GABA_A receptor blockade or transient α₆-containing GABA_A receptor knockdown induces evoked hypersensitivity and spontaneous pain in naïve female rats. The α₆ subunit is expressed in IB4⁺ and CGRP⁺ primary afferent neurons in the rat spinal dorsal horn and dorsal root ganglia (DRG), but not astrocytes. Nerve injury reduces α₆ subunit protein expression in the central terminals of the primary afferent neurons and DRG, whereas intrathecal administration of positive allosteric modulators (PAMs) of the α₆-containing GABA_A receptor reduces tactile allodynia and spontaneous nociceptive behaviors in female, but not male, neuropathic rats and mice. Overexpression of the spinal α₆ subunit reduces tactile allodynia and restores α₆ subunit expression in neuropathic rats. PAMs of the α₆-containing GABA_A receptor induces a greater antiallodynic effect in females compared to male rats and mice. Finally, α₆ subunit is expressed in humans. This receptor is found in CGRP⁺ and P2X3⁺ primary afferent fibers but not astrocytes in the human spinal dorsal horn. Our results suggest that the spinal α₆-containing GABA_A receptor has a sex-specific antinociceptive role in neuropathic pain, suggesting that this receptor may represent an interesting target to develop a novel treatment for neuropathic pain.

1. Introduction

Neuropathic pain is a debilitating chronic condition caused by a direct injury to peripheral nerves or the spinal cord. Neuropathic pain is a major health problem affecting around 17% of people around the world [8,12]. Unfortunately, the current pharmacological approaches for treating neuropathic pain are insufficient and are accompanied by many negative side-effects [8,20]. As such, it is critical to identify alternative analgesic strategies for neuropathic pain management. Part of the problem is that neuropathic pain is a complex disease with many mechanistic changes occurring at the level of the peripheral and central nervous system.

One such change is a loss of the GABAergic inhibition that contributes to the neuronal excitability in the peripheral and spinal nociceptive processing [21,29,31,53]. It has been hypothesized that restoring GABAergic tone by modulating the function of extra synaptic GABA_A receptors may be beneficial for pain management [14,68,69,81]. GABA_A receptors are ligand-dependent channels. There are 19 GABA_A subunits up to date, and each one confers classical pharmacological or biophysical properties [19,54]. Recent work form our group has shown that activation of spinal α₅-containing GABA_A receptors leads to chronic pain [14]. However, little is known about the role of other extra synaptic GABA_A receptors on neuropathic pain.

Extra synaptic α₆-containing GABA_A receptors are expressed in the trigeminal ganglion and the spinal cord, which are essential sites for pain processing [5,17,26,58,75]. Positive allosteric modulators (PAMs) of the α₆-containing GABA_A receptor reduce migraine-type and trigeminal neuropathic pain in mice and rats [17,75,77]. Moreover, sexual hormones can regulate the function and expression of extra synaptic GABA_A receptors, including α₆-containing GABA_A receptors [21,43,56,57,74]. Nevertheless, the role of spinal α₆-containing GABA_A receptors in neuropathic pain in both sexes has not been investigated.

Based on these considerations, we hypothesized that the spinal α₆-containing GABA_A receptor has an antinociceptive function under physiological conditions and neuropathic pain. We first evaluated the role of α₆-containing GABA_A receptor on nociceptive behaviors and its expression within the spinal cord and dorsal root ganglia (DRG) in naïve rats. We then evaluated the effect of nerve injury on α₆ subunit expression. We administered α₆-containing GABA_A receptor PAMs and induced overexpression of this receptor via a non-viral vector to assess its effects on nociceptive behaviors. We also determined the effect of α₆-containing GABA_A receptor PAMs in female and male neuropathic rats and mice. Finally, we investigated the presence of the α₆ subunit in the spinal cord of humans. Our findings support the view that α₆-containing GABA_A receptor activation present in the spinal cord and DRG has a relevant participation in the expression of neuropathic pain in rodents.

2. Material and methods

2.1. Animals

Six to seven-week-old age male and female Wistar rats (140–160 g) and 10 weeks-old female and male Swiss wesbter mice (about 25 g) were housed (4 per cage) in acrylic cages (44 cm width × 33 cm length × 20 cm height) under controlled temperature and humidity conditions (22 °C ± 1 °C, 50% humidity, and 12/12 light-dark cycle) and provided with tap water and food ad libitum. Rats were at least 8–9 weeks-old (200 g), while mice were 12 weeks-old (25–30 g) at the time of testing. The animals were obtained from our breeding facility (Cinvestav, Mexico City) or Charles River Laboratories. Rats from Charles River Laboratories were used one week after arrival at the animal facility at the University of Texas at Dallas. All animal procedures and protocols were performed with the approval of the Institutional Ethics Committee (Cinvestav, Mexico City) and The University of Texas at Dallas following the established regulations by the Mexican Official Norm for the Use and Welfare of Laboratory Animals (NOM-069-ZOO-1999), the National Institutes of Health Guide for Care and Use of Laboratory Animals (Publication No. 85–23, revised 1985), and the Guidelines on Ethical Standards for Investigation of Experimental Pain in Animals [84]. Lastly, some of the data sets (as mentioned in its respective figure legend) were reused according to the 3Rs principles for animal experiments to save the number of animals in this study.

2.2. Spinal nerve ligation model

All rats and mice surgeries were conducted according to a previous method described by Kim and Chung in 1992 [36]. In brief, animals were anesthetized with a mixture of ketamine (50 mg/kg, ip) and xylazine (10 mg/kg, ip). The left spinal nerves L5 and L6 (for rats) and L4 and L5 (for mice) were surgically exposed and ligated with 6–0 silk suture in a region distal to the DRG. In the sham-operated animals, the nerves were exposed but not ligated.

2.3. Stimulus-evoked nociceptive behavior

Animals were habituated in transparent acrylic cages on a mesh grid floor for 30–45 min before the experiment. Subsequently, von Frey filaments were used to determine the 50% paw withdrawal threshold using the up-down method as previously described [10,16]. The 50% withdrawal threshold was determined according to the following equation:

$$50\%\text{\hspace{0.17em}Threshold\hspace{0.17em}}\left(\text{g}\right)=(10^[\text{Xf}+\text{k}\delta ])/10,000$$

where Xf is the value of the last von Frey filament used (in logarithmic units); k is the correction factor based on the response patterns of a calibration table and the tabulated value based on the pattern of positive and negative responses, and δ indicates the average differences between stimuli in logarithmic units [10,16]. In non-lesioned animals, a value 11.98–15 g (for rats) or 1.2–1.6 g (for mice) are considered normal, while the presence of allodynia is considered when the 50% withdrawal threshold of the limb is < 4 (for rats) or 0.2 (for mice) g, respectively.

Muscle hyperalgesia was measured as previously described [43]. In brief, rats were immobilized, and their left hind paw was placed in the Randall-Selitto apparatus to receive an increasing pressure in the gastrocnemius muscle until the limb was withdrawn. All animals were handled for 3 days before the test day. The average of three consecutive tests with an inter-stimulus interval of 1 min was used to determine the muscle hyperalgesia threshold. In naïve rats, a value of 250 g was considered a cut-off of stimulation, and in neuropathic rats, muscle hyperalgesia was considered when the muscle threshold was ≤ 150 g.

2.4. Spontaneous nociceptive behavior

Spontaneous nociception was observed without experimenter intervention. Animals were placed in transparent acrylic cylinder without prior habituation. Rearing was measured during 5 min, and a positive response was considered every time the animal stood on its hindlimbs uniformly, raised its forelimbs from the ground, and extended its head upwards. Flinches were measured as the numbers of shakings in the injury hindlimb during 30 min [4,52].

2.5. Western blotting

The animals were anesthetized with isoflurane and killed by decapitation. The ipsilateral dorsal lumbar section of the spinal cord was dissected and immediately put on dry ice. Tissues were homogenized in ice-cold cOmplete lysis buffer containing Tris 50 mM, NaCl 150 mM, Triton 100-X 1%, sodium deoxycholate 5%, SDS 0.1%, and proteases inhibitors (cOmplete^™, EDTA-free protease inhibitor cocktail; Cat. No. 04693124001, Sigma-Aldrich, St. Louis, MO). Homogenized tissues were centrifuged at 14,000 rpm at 4°C for 10 min to remove cellular debris. Protein concentration was measured by the Bradford method according to the manufacturer instructions (Quick Start^™ Bradford Protein Assay; Cat. No. 5000203, Bio-Rad, Hercules, CA). A total of 50 μg of protein was separated by a 10% SDS-polyacrylamide gel electrophoresis for 90 min, 120 V at room temperature. Proteins were transferred to a polyvinylidene difluoride membrane (PVDF, Millipore Sigma, MA) for 1.5 h, 70 V at 16°C. Then the membranes were blocked with 3% non-fat milk in 1X Tris buffer solution containing Tween 20 0.1% (TBS-T) for 80 min at room temperature. Membranes were washed in 1X TBS-T two times for 5 min each before incubation with the primary antibody rabbit anti-GABA_A receptor α₆ subunit (1:300, Cat. No. G5544, Sigma Aldrich, St. Louis, MO, Table 1) overnight at 4°C. The next day, the membranes were washed three times in 1X TBS-T for 10 min each, and then incubated with the secondary antibody (anti-rabbit, 1:5000, Cat. No. 11-035-033, Jackson InmunoResearch Laboratories, West Grove, PA) for 2 h at room temperature. Protein signal was detected using Immobilon Western Chemiluminescent HRP substrate (Sigma-Aldrich, St. Louis, MO) and visualized with Bio-Rad ChemiDoc XRS+ Imaging System. Membranes were stripped using Restore Western Blot Stripping buffer (ThermoFisher Scientific, Waltham, MA) and re-tested with anti-beta actin (1:10000, Cat. No. GT5512, GeneTex, Irvine, CA), which was used as internal load control to normalize α₆ subunit protein level. Bands were quantified by densitometry and analyzed using Image lab 6.0.1 software (Bio-Rad, Hercules, CA). Anti-GABA_A receptor α₆ subunit specificity was tested by using siRNA knockdown (see in vivo Gabra6 knockdown section), adding a peptide to the antibody, or removing of the primary antibody (Supplementary Fig. 1). All the western blot analysis were performed in independent groups from those used to the behavioral assays.

Table 1.

List of antibodies used for the immunohistochemistry.

Antibody	Vendor	Cat. No.	Dilution/Conc.
Mouse-anti-CGRP	Novus Biologicals	NBP2–88945	1:100
Mouse-anti-GFAP	NeuroMab	75–240	2.5 μg/ml
Mouse-anti-IB4 conjugated Alexa 568	ThermoFisher	I21412	1:500
Mouse-anti-NeuN	Millipore	MAB377	10 μg/ml
Mouse-anti-P2X3	Santa Cruz	sc-390572	5 μg/ml
Rabbit-anti-α6-GABAAR	Sigma-Aldrich	G5544	5 μg/ml
Guinea Pig-anti-TRPV1	Neuromics	GP14100	1:500
Goat-anti-guinea pig H&L 568	ThermoFisher	A11075	1:2000
Goat-anti-mouse IgG1 488	ThermoFisher	A21121	1:2000
Goat-anti-rabbit H&L 555	ThermoFisher	A21428	1:2000
Goat-anti-rabbit H&L 647	ThermoFisher	A21245	1:2000

2.6. Immunohistochemistry

Animals were anesthetized with isoflurane and killed by decapitation. Spinal cord tissue was dissected and embedded in OCT in a cryomold and flash-frozen immediately on dry ice. All human tissue procurement procedures were approved by the Institutional Review Boards at the University of Texas at Dallas (UTD) and in collaboration with the Southwest Transplant Alliance. Human lumbar spinal cords were collected from organ donors within 4 hours of cross-clamp and from neurologic determination of death donors. Donor information is provided in Table 1. The human lumbar dorsal spinal cord was collected and frozen immediately on dry ice and then stored in a −80°C freezer. The human tissues were gradually embedded in OCT in a cryomold by adding small volumes of OCT over dry ice to avoid thawing. All the tissues were randomly sectioned at 20 μm in a cryostat (Leica CM1950, Nussloch, Germany), and the sections were mounted onto SuperFrost Plus charged slides (ThermoFisher Scientific, Waltham, MA). The slides remained in the −20°C cryostat chamber until completion of sectioning and were immediately used for immunohistochemistry.

Slides were transferred to cold 10% formalin (pH 7.4) for 15 min. Next, the tissue slides were dehydrated at room temperature in 50% ethanol, 70% ethanol for 5 min each, and then immersed in 100% ethanol for 10 min. The slides were air-dried, and a ring was drawn around each section using ImmEdge PAP pen (Vector Labs, Burlingame, CA). Once the hydrophobic ring had dried, the slides were blocked with blocking buffer (10% normal goat serum, 0.3% Triton-X 100 in 0.1 M phosphate buffer (PB)) for 2 h at room temperature. Then, slides were washed in 0.1 M PB, placed in a light-protected humidity-controlled tray, and incubated with the primary antibodies (Table 1) diluted in blocking buffer overnight at 4°C. The following day, slides were washed in 0.1 M PB and incubated with respective secondary antibodies (1:2000) with DAPI (1:5000, Cat. No. 14285, Cayman Chemical, Ann Arbor, MI) diluted in blocking buffer for 2 h at room temperature. Human tissue slides were covered with TrueBlack® (diluted in 70% ethanol), a lipofuscin autofluorescence quencher, for 1 min. All slides were washed in 0.1 M PB, air-dried, and coverslipped with ProLong Gold Antifade reagent (Invitrogen, Waltham, MA).

2.6.1. Images analysis

All images were captured on a FluoView 1200 confocal microscope with 10x and 20x magnifications (Olympus corporation, Tokyo, Japan). One XY image was acquired of the spinal dorsal horn from each section, and 3–4 sections were imaged per animal (~12 images per experiment for an n = 3). Sections with any folding/damage artifact were discarded. Acquisition parameters were based on the users’ guidelines for Olympus. The gain was kept at the default setting, 1 Hz; HV ≤ 600, offset = 4, laser power ≤ 10% and default lens aperture (1 airy disk diameter). All images were acquired using the same image settings. Image J software was used to measure the signal intensity as cellular total corrected fluorescence (CTCF) by using the formula: integrated density - (area of cell selected x mean fluorescence of background readings). The co-localization percentage of the α₆ subunit was quantified by Mander’s overlapping coefficient using cellSens Dimension software (Olympus corporation, Tokyo, Japan).

2.7. Intrathecal administration in vivo

Intrathecal administration was performed in rats and mice under isoflurane anesthesia (2.5% isoflurane) in order to study the role of the α₆-containing GABA_A receptor within the spinal cord, as previously described [9]. In brief, rats and mice were placed in the prone position. A syringe (Cat # 80100, Hamilton, NV) connected to a 30-gauge needle was placed between L4-L5 vertebrae. The access to the subarachnoid space was confirmed by the presence of a ‘tail flick’. All rats and mice received a total volume of 10 μl or 5 μl, respectively.

2.8. In vivo Gabra6 knockdown

Small interference RNA (siRNA) treatment was given as previously reported [64]. In brief, scrambled (Cat. No. 4390843) or Gabra6 (Cat. No. 4390815, ID assay s132065, ThermoFisher Scientific, Waltham, MA) siRNA were mixed in 5 μl of lipofectamine RNAiMAX (ThermoFisher Scientific, Waltham, MA) to reach a final concentration of 1 μg/μl. The mix was prepared daily under sterile conditions immediately before each administration. Animals received a daily intrathecal injection of Gabra6 or scrambled siRNA every 24 h for three days. The α₆ subunit protein expression was measured on day 3 and 6 after the first siRNA administration to correlate α₆ subunit protein expression with nociceptive behaviors.

2.9. In vivo Gabra6 transfection

For plasmid amplification, E. Coli bacteria were transfected with 10 ng of Gabra6 plasmid (Cat. No. RR207696, OriGene, Rockville, MD) or control cDNA3 plasmid (Cat. No. V79020, ThermoFisher Scientific, Waltham, MA) through thermal shock. Next, the bacteria were resuspended in SOC medium (Invitrogen, Waltham, MA) with constant shaking at 37°C for 1 h. Bacteria with Gabra6 or cDNA3 plasmid were cultivated in LB Broth medium (ThermoFisher Scientific, Waltham, MA) containing the corresponding antibiotic (Kanamycin for Gabra6, 25 μg/ml; or ampicillin for cDNA3, 100 μg/ml) for 24 h at 37°C. The following day, two transformed bacteria colonies were selected and growing in LB Broth medium for 16 h at 37°C. The plasmid was isolated and purified using a Maxiprep Kit (ThermoFisher Scientific, Waltham, MA) according to the manufacturer instructions. The purity and concentration of the plasmids were detected using a NanoDrop 2000 (ThermoFisher Scientific, Waltham, MA). Transient expression of the α₆ subunit in vivo was performed as previously described [22]. In brief, 5 μg of Gabra6 or CDNA3 plasmid were mixed 1:1 with lipofectamine 2000 (ThermoFisher Scientific, Waltham, MA) and delivered by an intrathecal injection for three days every 24 h starting at day 11 post-surgery. The effectiveness of the Gabra6 plasmid was assessed by measuring the α₆ subunit protein expression 24 h after the last administration.

2.10. Drugs

Ro 15–4513 (8-azido-5,6-dihydro-5-methyl-6-oxo-4H-imidazol[1,5-a] [1,4]benzo-diazepine-3-carboxylic acid ethyl ester) and PZ-II-029 (2,5-dihydro-7-methoxy-2-(4-methoxyphenyl)-3H-pyrazolo [4,3-c] quinoline-3-one), PAMs of the α₆-containing GABA_A receptor were purchased from Tocris Bioscience (Bristol, UK). Drugs were dissolved in 30% dimethyl sulfoxide (DMSO). Furosemide, α₆-containing GABA_A receptor antagonist, was obtained from Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO) and dissolved in 100% DMSO. Anesthetics ketamine and xylazine were purchased from PiSA Farmacéutica (Guadalajara, Jalisco). Isoflurane was obtained from Vet One (Boise, Idaho). Doses of the drugs used in this study were chosen from preliminary studies carried out in our laboratory.

2.11. Data analysis

All pharmacological experiments were carried out by a tester blinded to the treatments. An experimenter injected vehicles and treatments, meanwhile a different tester, unaware of the groups or treatment, performed the behavioral testing in animals. Results are shown as box-and-whiskers plots with the boxes showing the median and the 25^th and 75^th quartiles and whiskers representing minimum and maximal values, except for time courses and western blotting where data are presented as mean ± standard error of the mean.

The differences between treated and vehicle groups for tactile allodynia and muscle hyperalgesia were evaluated by two-way analysis of variance (ANOVA) followed by the Tukey test since independent groups of animals were used. Rearing and flinching behavior, western blotting and immunohistochemistry results were analyzed by one-way ANOVA followed by the Dunnett post hoc test. Sex differences were analyzed by using a three-way ANOVA followed by the Tukey test. All the statistical analysis were performed with SigmaPlot v12.0 (Systat Software Inc, Chicago, IL), and the plots were created with GraphPad Prism 8 (GraphPad Software Inc, La Jolla, CA). Statistical significance was accepted when p < 0.05.

3. Results

3.1. α₆-containing GABA_A receptor exerts an antinociceptive role under physiological conditions

First, we sought to determine the role of the α₆-containing GABA_A receptor in pain signaling under physiological conditions. For that purpose, female rats received the intrathecal administration of furosemide (10–300 nmol), an α₆-containing GABA_A receptor antagonist [39,73]. Then, we measured stimulus-evoked mechanical thresholds and spontaneous nociceptive behaviors for the next 6 hours (Fig. 1A). Furosemide, but not vehicle, reduced paw withdrawal and muscle pressure threshold in a dose-dependent manner (Fig. 1B–C). Furosemide also increased flinching in the hind paw, but it did not modify the number of rearing behaviors (Figs. 1D–E). Interestingly, furosemide induced a lower effect in paw withdrawal response in male rats (Figs. 1F–G). To further confirm the participation of the α₆-containing GABA_A receptor, we carried out an in vivo transient knockdown of the spinal α₆-containing GABA_A receptor in female rats (Fig. 2A). Despite this effect seem to be short-lasting, it should be pointed out that the intrathecal administration of the siRNA anti-α₆ subunit reduced protein expression of the α₆ subunit at the spinal cord on the day 3, but not day 6, after the first siRNA administration (Fig. 2B–C). Repeated spinal injection of the anti-α₆ subunit siRNA diminished paw withdrawal and muscle pressure thresholds for about 4 days (Fig. 2D–E). In addition, intrathecal α₆ subunit knockdown induced spontaneous pain in female naïve rats (Fig. 2F–G). These results strongly suggest that the spinal α₆-containing GABA_A receptor exerts an antinociceptive role in physiological conditions in female and, at a lesser extent, in male rats.

Figure 1.

Furosemide induces nociceptive behaviors in naïve female rats. A) Experimental protocol for nociceptive behavior tests in naïve female rats. Effect of i.t. administration of furosemide on B) withdrawal threshold, C) muscle pressure, D) number of flinches, and E) the number of rearing in naïve female rats. F) Effect of i.t. administration of furosemide on withdrawal threshold in naïve female and male rats. G) Area under the curve (AUC) of the time course of the effect induced by furosemide. Data are presented as the mean ± SEM (n = 6 animals per group for behavioral assessment). In B and C: *P < 0.05, **P < 0.01 ***P < 0.001 Furosemide (Furo) vs vehicle in female rats, as determined by two-way ANOVA, followed by the Tukey test. In D, E and G: **P < 0.01 ***P < 0.001, as determined by one-way ANOVA, followed by the Tukey test. In F: * P < 0.05, ***P < 0.001 Furosemide (Furo) vs vehicle in female rats; and ^&&P < 0.01 Furo in female vs male, as determined by three-way ANOVA, followed by the Tukey test. ns: not significant; Furo: Furosemide; Veh: Vehicle. Data for vehicle group in female rats depict the same data set from Fig. 1B. Black dotted line represents the tactile allodynia threshold.

Figure 2.

Knockdown of the α₆-containing GABA_A receptor induces nociceptive behaviors in naïve female rats. A) Timeline of the experimental design used to test the effect of the α₆-containing GABA_A siRNA in naïve female rats. Effect of the transient knockdown of α₆-containing GABA_A receptor on the protein expression of the α₆ subunit in the spinal dorsal horn on day 3 (B) and day 6 (C) after 1^st siRNA administration. D) withdrawal threshold, E) muscle pressure, F) number of flinches, and G) number of rearing in naïve female rats. Data are presented as the mean ± SEM (n = 6 animals per group for behavioral assessment, n = 4 for western blotting analysis). In B and C: ***P < 0.001, as determined by one-way ANOVA, followed by the Tukey test. In D, E, F and G: *P < 0.05, **P < 0.01 ***P < 0.001 Gabra6 siRNA vs Scrambled siRNA in female rats, as determined by two-way ANOVA, followed by the Tukey test. In G: ^#P < 0.05 naïve vs scrambled siRNA by two-way ANOVA, followed by the Tukey test. ns: not significant; Scr: Scramble. Black dotted line represent the tactile allodynia threshold.

As our pharmacological results in naïve female rats suggest the participation of the α₆-containing GABA_A receptor in physiological conditions, we then determined the cellular localization of the α₆ subunit. Spinal cord sections from naïve female rats were labeled with NeuN (a neuronal marker) as well as GFAP (an astrocytes marker). α₆ subunit co-localized with NeuN (3.22 ± 0.05 % of overlapping) (Fig. 3A), but not with GFAP (no overlapping found) (Fig. 3). We then investigated the distribution of this subunit in the spinal dorsal horn. As CGRP (16.50 ± 1.27 % of overlapping with the α₆ subunit) and IB4 (33.51 ± 3.58 % of overlapping with the α₆ subunit) are good markers for laminae I and II labeling, immunohistochemistry assay showed α₆ subunit expression at the dorsal horn mainly from laminae II to V of the spinal cord (Fig. 3B–C). Besides, α₆ subunit did not co-localize with GFAP (no overlapping found) (Fig. 3D). Since pre- (nociceptive primary afferents terminals) and post-synaptic (spinal dorsal horn neurons) terminals are very adjacent to each other, it is hard to distinguish between them by using the immunohistochemistry assay [47]. Therefore, we assessed the expression of the α₆ subunit within the DRG neurons from naïve female rats (Fig. 4). We found the expression of the α₆ subunit in 74.11% of the DRG neurons where the overlapping of the α₆ subunit was greater with IB4⁺ (about 15%) than TRPV1⁺ (about 11%) neurons (Fig. 4A–B). The expression of the α₆ subunit was mainly found in medium-diameter neurons of the DRG, which may suggest that the α₆-containing GABA_A receptor is located in neurons related to pain signaling (Fig. 4C). Furthermore, we found that the α₆ subunit is also expressed in the satellite glia cells of DRG (Fig. 4D). Previous bulk sequencing analysis in female mice DRG showed the expression of α₆ subunit mRNA in this tissue [59]. The specificity of primary antibodies was assessed through the immunostaining in the absence of the primary antibody (Supplementary Fig. 2). Also, positive control immunostaining was performed using cerebellum, since this receptor is widely expressed in this brain area (Supplementary Fig. 2). Our data demonstrate that α₆-containing GABA_A receptor is expressed in the central terminals of the primary afferents within the spinal cord, as well as in DRG neurons.

Figure 3.

Expression of the α₆-containing GABA_A receptor in the spinal dorsal horn of female naïve rats. Determination of cellular localization of the α₆ subunit in different types of neurons population, A) neuronal marker (NeuN), B) peptidergic neurons (CGRP), C) non-peptidergic neurons (IB4), and D) astrocytes (GFAP). Scale, 50 μm and 100 μm.

Figure 4.

Expression of the α₆-containing GABA_A receptor in the DRG neurons of female naïve rats. A) Determination of cellular localization of the α₆ subunit in different types of nociceptive neurons, B) analysis of the percentage of overlapping of the α₆ subunit with IB4 and TRPV1, and C) histogram with Gaussian distribution displaying the size profile of all α₆-containing GABA_A positive neurons. Data are presented as the mean ± SEM (n = 9 sections from 3 animals). Scale, 50 μm and 100 μm.

3.2. Stimulation of α₆-containing GABA_A receptor reduces nociceptive behavior in neuropathic rats and mice

Next, we assessed protein expression of the α₆ subunit in the spinal dorsal horn and DRG of female rats after nerve injury using the spinal nerve ligation (SNL) model. SNL induced tactile allodynia, muscle hyperalgesia and spontaneous nociceptive behaviors in female rats from 1 to 21 days post-SNL (Supplementary Fig. 3). Besides its nociceptive effects, SNL decreased α₆ subunit protein expression at the spinal cord (Fig. 5A) and DRG (Fig. 5B) from 3 to 21 days after injury. These results demonstrate that the α₆ subunit is downregulated after nerve injury, suggesting that this process contributes to the loss of GABAergic inhibition and neuropathic pain.

Figure 5.

Nerve injury decreases protein expression of the α₆-containing GABA_A receptor in the spinal dorsal horn and DRG in female rats. Time course of the western blotting analysis of the α₆ subunit in (A) the spinal dorsal horn and (B) DRG. Data are presented as the mean ± SEM (n = 4–5). *P < 0.05, **P < 0.01, ***P < 0.001, as determined by one-way ANOVA, followed by the Tukey test.

Indeed, the loss of GABAergic inhibition has been proposed as a mechanism that contributes to neuropathic pain [13,21,29,31,53]. Considering these results, we hypothesized that down-regulation of the spinal α₆-containing GABA_A receptor could be one mechanism contributing to the loss of the GABAergic inhibition. In order to counter act this phenomenon, we tested the effect of the intrathecal administration of PAMs of the α₆-containing GABA_A receptor in neuropathic female rats (Fig. 6A). Intrathecal administration of Ro 15–4513 (1–10 nmol), but not vehicle, reduced tactile allodynia for 8 h in a dose-dependent manner (Fig. 6B–C). Besides allodynia, it is well known that neuropathic pain induces muscle hyperalgesia [35,66]. Intrathecal administration of Ro 15–4513 (10 nmol) partially reduced muscle hyperalgesia for 2 h (Fig. 6D). Spontaneous pain represents an important symptom experienced by several patients with neuropathic pain [33]. Moreover, non-stimulus evoked pain assessment in animal models is relevant to the translation into the clinic. Based on this, we explored the effect of Ro 15–4513 on flinching and rearing behaviors 2 h after intrathecal administration. This schedule was chosen as R015–4513 reached its maximal effect at this time. Interestingly, intrathecal administration of Ro 15–4513 reduced the number of spontaneous flinches and increased rearing behavior in neuropathic rats (Fig. 6E–F). Admittedly, Ro 15–4513 is a selective PAM of α₄- and α₆-containing GABA_A receptors [27,78,80]. Thus, our data with this drug could result from activation of both receptors. To avoid this concern, we also used PZ-II-029 (a PAM with higher selectivity for the α₆-containing GABA_A receptor) [76]. PZ-II-029 (1–30 nmol) increased in a dose-dependent fashion the paw withdrawal threshold for 6 h (Fig. 6G–H). In contrast, intrathecal injection of PZ-II-029 (30 nmol) did not affect nerve injury-induced muscle hyperalgesia (Fig. 5I). Moreover, PZ-II-0–29 reduced the number of spontaneous flinches and increased rearing behavior at 2 h post-administration (Fig. 6J–K). Taken together, our results suggest that activation of spinal α₆-containing GABA_A receptor reduces evoked and spontaneous manifestations of neuropathic pain. In order to support our pharmacological data, we investigated whether PZ-II-029 repetitive administration every 6 h could induce an antiallodynic effect and restore α₆ subunit protein expression in neuropathic rats. For that purpose, we measured the withdrawal threshold 2 h and 6 h after each PZ-II-029 injection during 24 h. (Fig. 6L). Interestingly, the repetitive PZ-II-029 administration induced a sustained antiallodynic effect during the 24 h of evaluation, except at 6 h after the first administration (Fig. 6M). Furthermore, repeated administration of PZ-II-029 also restored protein expression of the α₆ subunit in the spinal cord (Fig. 6N). These results suggest that sustained activation of the α₆-containing GABA_A receptor diminishes tactile allodynia by increasing protein expression of the α₆ subunit, as previously reported for the α₄ subunit [2,3]. Finally, to further confirm the participation of α₆-containing GABA_A receptor in this effect, we tested the effect of PZ-II-029 in presence of furosemide (Fig. 7A). Of note, the α₆-containing GABA_A receptor antagonist furosemide (10–30 nmol, it) dose-dependently reduced the antiallodynic effect of PZ-II-029 in neuropathic female rats (Fig. 7B–C). Since the furosemide is a blood brain barrier-impermeable drug [65], we gave furosemide by systemic administration to elucidate the role of peripheral α₆-containing GABA_A receptor in the antinociceptive effect induced by PZ-II-029 (Fig. 7D). Furosemide (20 mg/kg, i.p.) partially prevented PZ-II-029-induced antiallodynia (Fig 7E–F). We also assessed the administration of furosemide in absence of PZ-II-029 to exclude its potential per se effect on the withdrawal threshold. Administration of furosemide (10–30 nmol, i.t. or 20 mg/kg, i.p.) did not modify the withdrawal threshold of the paw compared to vehicle (Fig. 7). Taken together, these results suggest that the antinociceptive effect induced by PZ-II-029 is mainly mediated by local mechanisms in the spinal cord and DRG.

Figure 6.

PAMs of the α₆-containing GABA_A receptor decrease nociceptive behaviors induced by nerve injury in female rats. A) Timeline of experimental approaches used to determine the antinociceptive effect of PAMs of the α₆-containing GABA_A receptor in female neuropathic rats. B) Effect of the intrathecal administration of Ro 15–4513 on tactile allodynia in female neuropathic rats. C) Area under the curve (AUC) of the antiallodynic effect induced by Ro 15–4513 in female neuropathic rats. Effect of the intrathecal administration of Ro 15–4513 on D) muscle hyperalgesia, E) number of flinches, and F) number of rearing in female neuropathic rats. G) Effect of the intrathecal administration of PZ-II-029 on tactile allodynia in female neuropathic rats. H) Area under the curve (AUC) of the antiallodynic effect induced by PZ-II-029 in female neuropathic rats. Effect of the intrathecal administration of PZ-II-029 on I) muscle hyperalgesia, J) number of flinches, and K) number of rearing in female neuropathic rats. L) Experimental design of repetitive administration of PZ-II-029 in female neuropathic rats. Effect of i.t. repetitive administration of PZ-II-029 on M) tactile allodynia and N) protein expression of the α₆ subunit at the spinal dorsal horn in female neuropathic rats. The withdrawal threshold was measured 2 and 6 hours after each PZ-II-029 injection. The continuous blue line represents the time course for 24 h; orange dotted line represents the time course at 2 h after PZ-II-029 injection; purple dotted line represents the time course at 6 h after the PZ-II-029 administration. Data are presented as the mean ± SEM (n = 6 animals per group for behavioral assessment, n = 3 for western blotting analysis). In B, D, G and I: *P < 0.05, **P < 0.01 ***P < 0.001, as determined by two-way ANOVA, followed by the Tukey test. In C, E, F, H, J, K and N: *P < 0.05, **P < 0.01 ***P < 0.001, as determined by one-way ANOVA, followed by the Tukey test. ns: not significant.; PZ: PZ-II-029; Ro15: Ro 15–4513; Veh: Vehicle. Black dotted line represent the tactile allodynia threshold.

Figure 7.

Blockade of the α₆-containing GABA_A receptor decreases PZ-II-029-induced antiallodynic effect. A) Timeline of the experimental approach used to determine the antagonistic effect of furosemide on the antiallodynic effect induced by PZ-II-029 in female neuropathic rats. B) Effect of the intrathecal administration of furosemide on the antinociceptive effect induced by PZ-II-029 in female neuropathic rats. C) Area under the curve (AUC) of antagonistic effect of intrathecal administration of furosemide in female neuropathic rats. D) Timeline of the experimental approach used to determine the antagonistic effect of intraperitoneal administration of furosemide on the antiallodynic effect induced by PZ-II-029 in female neuropathic rats. E) Effect of the intraperitoneal administration of furosemide on the antiallodynic effect induced by PZ-II-029 in female neuropathic rats. F) Area under the curve (AUC) of antagonistic effect of intraperitoneal administration of furosemide in female neuropathic rats. In B and E: *P < 0.05, **P < 0.01, ***P < 0.001 vs Veh Furo + PZ (30 nmol) by two-way ANOVA followed by the Tukey test. In C and F: *P < 0.05, ***P < 0.001 by one-way ANOVA followed by the Tukey test. ns: not significant; PZ: PZ-II-029; Furo: Furosemide; Veh: Vehicle. Black dotted line represent the tactile allodynia threshold.

In order to determine whether the effects of Ro 15–4513 and PZ-II-029 are specific for rats, we tested these drugs in spinal nerve injured mice. SNL induced tactile allodynia in mice which was attenuated after intrathecal injection of Ro 15–4513 (Fig. 8A–B) or PZ-II-029 (Fig. 8C–D) in female neuropathic mice in a dose-dependent manner for both drugs. Our data indicate that activation of spinal α₆-containing GABA_A receptor decreases mechanical hypersensitivity in both rats and mice.

Figure 8.

PAMs of the α₆-containing GABA_A receptor decrease nociceptive behaviors induced by nerve injury in female mice. A) Effect of the intrathecal administration of Ro 15–4513 on tactile allodynia in female neuropathic mice. B) Area under the curve (AUC) of the antiallodynic effect induced by Ro 15–4513 in female neuropathic mice. C) Effect of the intrathecal administration of PZ-II-029 on tactile allodynia in female neuropathic mice. D) Area under the curve (AUC) of the antiallodynic effect induced by PZ-II-029 in female neuropathic mice. In A and C: **P < 0.01 ***P < 0.001, as determined by two-way ANOVA, followed by the Tukey test. In B and D: **P < 0.01 ***P < 0.001, as determined by one-way ANOVA, followed by the Tukey test. ns: not significant. PZ: PZ-II-029; Ro15: Ro 15–4513; Veh: Vehicle. Black dotted line represent the tactile allodynia threshold

3.3. Restitution of spinal α₆-containing GABA_A receptor reduces nociceptive behaviors in neuropathic rats

Our data suggest that nerve injury-induced down-regulation of spinal α₆ subunit contributes to neuropathic pain. Thus, we sought to determine the effect of the α₆ subunit overexpression through the injection of a non-viral plasmid in neuropathic rats. For that purpose, neuropathic rats received an intrathecal injection of a purified Gabra6 plasmid each day for 3 days (Fig. 9A). Gabra6 plasmid injection restored the α₆ subunit protein expression in the dorsal portion of the spinal cord (Fig. 9B). In addition, plasmid administration reduced tactile allodynia (Fig. 9C), muscle hyperalgesia (Fig. 9D) and spontaneous pain (Fig. 9E–F). Altogether, these results demonstrate that nerve injury-induced down-regulation of the spinal α₆-containing GABA_A receptor plays a role in tactile allodynia, muscle hyperalgesia and spontaneous pain.

Figure 9.

Administration of the Gabra6 plasmid restores the α₆-containing GABA_A receptor and decreases nociceptive behaviors induced by nerve injury female neuropathic rats. A) Timeline of experimental approaches used to determine the antinociceptive effect of the Gabra6 plasmid in female neuropathic rats. Effect of the intrathecal repeated administration of the Gabra6 plasmid on B) protein expression of the α₆ subunit at the spinal dorsal horn of female neuropathic rats 24 h after the last administration, C) tactile allodynia, D) muscle hyperalgesia, E) number of flinches, and F) number of rearing in female neuropathic rats. Data are presented as the mean ± SEM (n = 6 animals per group for behavioral assessment, n = 3 for western blotting analysis). In B, C, D and E: *P < 0.05, **P < 0.01 ***P < 0.001, as determined by two-way ANOVA, followed by the Tukey test. In F: *P < 0.05, as determined by one-way ANOVA, followed by the Tukey test. ns: not significant. Black dotted line represent the tactile allodynia threshold

3.4. Activation of spinal α₆-containing GABA_A does not affect nociceptive behavior in neuropathic male rats

We have previously shown that the spinal α₅-containing GABA_A receptor shows a sex-dependent antiallodynic effect in neuropathic rats and mice [21,43]. Here, we observed that α₆-containing GABA_A receptor antagonist furosemide induces mechanical hypersensitivity in a sex-dependent manner. Moreover, as α₆-containing GABA_A receptor is also affected by sex hormones [42], we evaluated the effect of PAMs in male and female rats and mice. Interestingly, we found that pharmacological activation of the spinal α₆-containing GABA_A receptor with PAMs, at the same dose that reduced tactile allodynia in female neuropathic rats and mice, failed to induce an antinociceptive effect in male rats (Fig. 10A–B) and mice (Fig. 10C–D). Interestingly, immunohistochemistry (Fig. 10E–F) and western blot analyses (Fig. 10G) showed that the α₆ subunit is less expressed in the dorsal horn of naïve male rats, compared to naïve female rats.

Figure 10.

Activation of the spinal α₆-containing GABA_A receptor induces a greater antiallodynic effect in female compared with male rats and mice. Effect of the intrathecal administration of Ro 15–4513 and PZ-II-029 (α₆-containing GABA_A receptor PAMs) on the tactile allodynia in neuropathic female and male rats (A and B) and mice (C and D). E) Representative immunohistochemistry images of the α₆ subunit expression at the spinal dorsal horn of naïve female and male rats. F) CFCT analysis of immunohistochemistry staining. G) Western blotting of the spinal dorsal horn of naïve female and male rats. Data are presented as the mean ± SEM (n = 9 sections from 3 animals for immunohistochemistry, n = 3 for western blotting analysis, n = 6 animals per group for behavioral tests). In F and G: *P < 0.05, by the Studentś t-test. In A, B, C and D: ***P < 0.001 Ro 15–4513 or PZ-II-029 vs vehicle in female rats; ^##P < 0.01, ^###P < 0.001 Ro 15–4513 or PZ-II-029 vs vehicle in male rats; ^&&&P < 0.001 Ro 15–4513 or PZ-II-029 in female vs male, as determined by three-way ANOVA, followed by the Tukey test. Data for female rats depict the same data set from Fig. 6B (Ro 15–4513) and 6G (PZ-II-029). Black dotted line represent the tactile allodynia threshold.

3.5. α₆-containing GABA_A receptors are expressed in the human spinal dorsal horn

Since it has been recently shown species-specific differences of protein expression in tissues relevant for pain signaling [67], we examined the expression of α₆ subunit in the human lumbar spinal cord (Table 2). We found that the α₆ subunit is expressed in neurons of the spinal dorsal horn (Fig. 11), mainly in CGRP⁺ (2.65 ± 0.23 % of overlapping with the α₆ subunit) and P2X3⁺ neurons (1.93 ± 0.24 % of overlapping with the α₆ subunit), but not astrocytes (Fig. 12). The specificity of antibodies was assessed through the immunostaining in the absence of the primary antibody (Supplementary Fig. 4). It should be pointed out that DRG from human donors was not considered since a negligible amount of α₆-containing GABA_A receptor, compared to the spinal cord, has been reported (0.01 TPM for DRG versus 4.097 TPM for the spinal cord) [59].

Table 2.

Human donor information.

Donor	Spinal Cord	Sex	Age	Cause of death	Patient history
1	Lumbar	Male	19	Anoxia/Drug overdose	Anxiety and depression
2	Lumbar	Male	27	CNS tumor/ICH stroke	Headaches and seizures; glioblastoma
3	Lumbar	Female	29	Anoxia/Drug overdose	Seizures, bipolar disorder, depression

CNS, central nervous system; ICH, intracerebral hemorrhage

Figure 11.

The α₆-containing GABA_A receptor is expressed in the human spinal dorsal horn. Representative immunohistochemistry images of the α₆-containing GABA_A receptor expression at the spinal dorsal horn of human donors and its co-localization with neurons (NeuN) and nucleus (DAPI). Scale, 100 μm.

Figure 12.

The α₆-containing GABA_A receptor is expressed in peptidergic and non-peptidergic fibers of the human spinal dorsal horn. Cellular localization of the α₆-containing GABA_A receptor in peptidergic (CGRP⁺) and non-peptidergic (P2X3⁺) neurons, but not in GFAP⁺ cells at the human spinal dorsal horn. Scale, 100 μm.

4. Discussion

In this study, we investigated the role of the α₆-containing GABA_A receptor in physiological conditions and neuropathic pain. We first established that α₆-containing GABA_A receptor is mechanistically instrumental in normal pain processing by inhibiting it using intraperitoneal and spinal administration of furosemide. While furosemide has been shown to have other off-target effects such as inhibiting NKCC1 and KCC2 [5,15,17,24,30,75,83], these effects were only found when much higher μmol concentrations were used [5]. Therein, we used a lower nmol dose to prevent off-target drug effects and found that spinal α₆-containing GABA_A receptor blockade by furosemide induces evoked and spontaneous nociceptive behaviors in naïve female rats. Similarly, transient knockdown of α₆-containing GABA_A receptor with a siRNA also caused evoked and spontaneous nociceptive behaviors in naïve rats. In support of this, siRNA treatment reduced α₆ subunit expression at 3, but not 6, days after the first administration of the siRNA which correlate with the presence of nociceptive behaviors. Together, these data suggest that α₆-containing GABA_A receptors exert an antinociceptive role under physiological conditions. Previous work has also shown that in vitro α₆-containing GABA_A receptor knockdown increases cell excitability in trigeminal ganglion neurons [58]. Our data advances the idea that reducing the activity of GABA_A receptors with pharmacological or molecular tools leads to nociception in naïve animals [6,13,21,29,43,45,61].

It has been widely reported that GABA_A receptors expressed in the spinal cord regulate sensory information [14,21,29,31,82]. Previous work has shown that the α₆ subunit is expressed in lamina II of the spinal dorsal horn ¹⁴. However, we detected it throughout all laminae, ranging from lamina II to lamina V. Other studies have reported the expression of the α₆ subunit in the trigeminal ganglion of rats [17,26,41,58]. As such, we found that the α₆-containing GABA_A receptor is expressed in 74.11% of the DRG neurons which is similar to previous findings in TG from adult Wistar male rats and CD1 mice [18,75] but higher than those reported in TG from neonates male Sprague-Dawley rats [28]. Moreover, α₆-containing GABA_A receptor co-localized mainly with IB4⁺ (about 15%) and, at a lesser extent, TRPV1⁺ (about 11%) neurons at the DRG. To the best of our knowledge, this is the first study demonstrating that this receptor is expressed on sensory neurons of the DRG. Thus, our results confirm that α₆-containing GABA_A receptor is expressed in critical sites for nociceptive signaling.

Neuropathic pain has been associated with changes in the inhibition-excitation balance at the dorsal horn of the spinal cord. Indeed, nerve injury decreases the function and expression of KCC2 in neurons of lamina I of the superficial dorsal horn, events that are associated with the development of pain behaviors [13,40]. We observed that nerve injury decreased α₆ subunit protein expression in DRG and spinal cord for at least 21 days. These results suggest that neuropathic pain could result not only from reduction of KCC2, but also from down-regulation of the α₆-containing GABA_A receptor at the DRG and deep dorsal horn. In support of this idea, there is evidence that transient silencing of the α₆-containing GABA_A receptor induces nociception in a model of inflammatory pain in rats [41,42]. Previous studies from our laboratory found that nerve injury increases expression of spinal α₅-containing GABA_A receptors at the spinal dorsal horn, which then promotes neuropathic pain [21,29,43]. Data suggest that α₆-containing GABA_A and α₅-containing GABA_A receptors are regulated differentially under pathological conditions and have different functions in neuropathic pain.

Our data suggest that spinal α₆-containing GABA_A receptor also plays an antinociceptive role in rodents in pathological conditions. This suggestion is based on the following findings: i) nerve injury decreases α₆ subunit expression; ii) intrathecal injection of selective PAMs of α₆-containing GABA_A receptor Ro 15–4513 and PZ-II-029 reduces nerve injury-induced tactile allodynia, muscle hyperalgesia and spontaneous nociceptive behaviors in rats; iii) intrathecal administration of Ro 15–4513 and PZ-II-029 reduces tactile allodynia in mice; iv) furosemide blocks the antiallodynic effect induced by PZ-II-029 in rats; and v) repeated intrathecal administration of PZ-II-029 reduces tactile allodynia and restores α₆ subunit expression. Our results agree with previous observations showing that systemic administration of selective PAMs of the α₆-containing GABA_A receptor reduces trigeminal nerve injury and migraine [17,75,77]. Reinforcing this, acute administration of PZ-II-029 abates cFos expression, a marker of neuronal excitability, in a migraine-type pain model [17], whereas intrathecal treatment with a siRNA against the α₆-containing GABA_A receptor decreases α₆ subunit protein expression, increases p-ERK expression and promotes nociception in ligated rats [41,42]. Although we do not have a clear-cut explanation for the effect of repeated intrathecal administration of PZ-II-029 on the α₆ subunit protein expression, it should be pointed out that the treatment with THDOC, a neurosteroid modulator of the α₄ subunit, increases protein expression of the α₄ subunit by PKC-mediated phosphorylation [2,3]. Besides, α₆ subunit could be phosphorylated in the Ser385 [32,68]. Therefore, it is likely that PZ-II-029 could regulate the α₆ subunit expression by a phosphorylation-dependent mechanism. However, this possibility remains unknown and falls beyond the scope of our study.

We observed that intrathecal administration of Ro 15–4513 and PZ-II-029 induced a delayed (about 2 h) antiallodynic effect in rats and mice. This delay suggests that both drugs could be acting in a site different from the spinal cord, as DRG. The fact that treatment with systemic furosemide (a blood brain barrier-impermeable drug) can partially reduce the antiallodynic effect of intrathecal PZ-II-029 suggest that effects of the α₆-containing GABA_A receptor PAMs are mainly due to their effects at the DRG.

As we determined that nerve injury leads to the reduction of α₆ subunit protein expression at the spinal cord, we hypothesized that restoring α₆ subunit protein expression would reduce neuropathic pain in rats. Accordingly, in vivo repeated intrathecal administration of a purified Gabra6 plasmid reduced tactile allodynia, muscle hyperalgesia and spontaneous nociceptive behaviors induced by nerve injury. In addition, spinal injection of the Gabra6 plasmid fully restored α₆ subunit protein expression at the spinal cord. To the best of our knowledge, this is the first study about the effect of overexpression of the α₆-containing GABA_A receptor in neuropathic pain. We acknowledge that non-viral vectors induce transient expression and low gene delivery to cells [23]. However, it has been shown that intrathecal administration of plasmid DNA represents a helpful tool for neuropathic pain [22,51,79]. However, the fact that this plasmid, but not its control, induces antinociceptive effects suggest that it works properly. It should be mentioned that the loss of the GABAergic inhibition under neuropathic pain conditions could be mediated by the metabotropic GABA_B receptor [44], as well as synaptic GABA_A receptors [37,46,47]. However, since the α₆-containing GABA_A receptor is an extra-synaptic receptor that can be activated by low GABA concentration due to its high affinity for this neurotransmitter [19], our results imply that restoring the GABA-mediated tonic inhibition by increasing α₆ subunit protein expression at the spinal cord induces an antinociceptive effect in neuropathic pain, as previously suggested with other experimental approaches [34,48].

Differences in pain signaling regulation between males and females under physiological and pathological conditions have been widely studied [1,21,43,49,50,55,60,62,63,70–72]. In this study we observed that both PAMs of the α₆-containing GABA_A receptor (Ro 15–4513 and PZ-II-029) induced a greater antiallodynic effect in female compared with male neuropathic rats and mice. Interestingly, we also observed that the α₆-containing GABA_A receptor antagonist furosemide induced greater mechanical hypersensitivity in female than in male naïve rats. Furthermore, α₆ subunit protein expression at the spinal cord is greater in female compared to male naïve rats. Moreover, immunohistochemistry analysis in our study revealed the expression of the α₆ subunit in DRG of female rats whereas a whole transcriptome expression study has shown the absence of Gabra6 gene expression in thoracic DRG from male rats [38]. All these findings indicate that the activity and expression of the α₆-containing GABA_A receptor depends on sex hormones present in female rodents. Our results agree with a previous study showing that high estradiol levels upregulate the Gabra6 gene in the trigeminal ganglion under physiological and inflammatory pain conditions [57]. The mechanisms underlying the dimorphic sexual role of the α₆-containing GABA_A receptor in neuropathic pain fall beyond the scope of this study but may involve hormonal, epigenetic, or immune regulation [11,21].

The translational failure of some GABAergic drugs to the clinic makes it necessary to explore target proteins in human tissues to confirm their expression and suggest a potential clinical application [25,82]. Our results show that the α₆ subunit is expressed in the spinal dorsal horn of human donors in CGRP⁺ and P2X3⁺ nociceptive primary afferent terminals, and NeuN⁺ but not in GFAP⁺ cells (astrocytes). This implies that α₆ subunit is expressed within the spinal cord in nociceptive primary afferent terminals. Whether the α₆-containing GABA_A receptor plays a crucial role in pain signaling in humans remains to be determined. Therefore, the role of α₆-containing GABA_A receptor under physiological and chronic pain conditions in humans requires further exploration.

5. Conclusion

Results indicate that the α₆-containing GABA_A receptor, which is present in the central terminals of the primary afferent neurons and DRG, exerts an antinociceptive role under physiological and pathological conditions. Pharmacological activation or protein overexpression of the α₆-containing GABA_A receptor reduced nociceptive behaviors in neuropathic rats, while blockade of this receptor induces tactile allodynia in naïve rats. Moreover, the antiallodynic effect of the α₆-containing GABA_A receptor is greater in female than in male rodents. Finally, this receptor is expressed in the human spinal cord. Our findings suggest that the α₆-containing GABA_A receptor could represent a target for the treatment of neuropathic pain.