The relationship between GABA
A and analgesia has been becoming a hot pursuit in pain research in recent years. GABA
A receptors are the most important of the three subtypes of receptors and are currently the most intensively studied. GABA
A receptors belong to pentameric ligand-gated ion channels mediating most of the rapid inhibitory synaptic transmission in central nervous system, and GABA
A receptor dysfunction is associated with many psychiatric and neurological disorders, such as anxiety, epilepsy and substance use disorders [
9]. The GABA
A receptor is comprised of five distinct subunits whose expression varies according to brain region, cell type and subcellular structural domain, as well as their function. To date, at least 19 receptor subunits are found, which are divided into subclasses according to homologous amino acid sequences: α1, α2, α3, α4, α5, α6, β1, β2, β3, and γ1, γ2, γ3, δ, ε, π, θ, and ρ1, ρ2, ρ3 [
17]. Although the receptor subunits vary in combination and function, not all subunits assemble efficiently into functional receptor subtypes. Most GABA
A receptors include two α-, two β- and one γ- subunits[
18,
19], which form an integrated negative ion channel that is permeable to chloride and bicarbonate ions. GABA
A receptors are primarily involved in antinociceptive effects at the supraspinal level, modulating GABA-induced inhibition of short duration monosynapses. When GABA combines with GABA
A receptors, chloride ion permeability on the nerve membrane increases, causing an anion-selective channel to open. This channel mainly gates chloride ions which allows them to flow inside neurons via ion concentration gradients, leading to hyperpolarization of the cell membrane, thereby inhibiting the neural excitability of GABAergic interneurons. As a result, this reduces pain signal transduction and further produces analgesic effect (see
Figure 1) [
20].
GABA
A receptors are the most prevalent receptors of inhibitory neurotransmitter in the CNS, and primary afferent neurons (PAD) are principally depolarized by GABA through the action of the cationic chloride transporter protein (NKCC1) [
21]. NKCC1 helps maintain a high intracellular concentration of chloride ion (Cl
-) [
22], and the binding of GABA to GABA
A receptors leads to anion efflux to depolarize PAD. The subsequent shunting of afferent action potentials leads to a decrease in excitatory transmitters released from injurious terminals. This process is known as presynaptic inhibition and plays a crucial role in controlling the hyperexcitability of neurons in the dorsal horn of the spinal cord [
23]. However, there are exceptions. For example, certain types of injury reportedly can enhance PAD such that spiking potentials can be sequentially evoked at primary afferent terminals, thus leading to the change in the PAD from a normal inhibitory process to an excitatory one [
24].
2.2.1. Mechanism of analgesic action of alpha(α) subtype receptors
Studies have shown that the cerebral cortex, hippocampus, and caudate nucleus exhibit complex expression patterns of multiple GABA
A receptor subtype combinations. Subunits are expressed in the hypothalamus, striatal-nigral fibers and cerebellar stellate/basket cells, and are also extensively existed in the spinal cord. Alpha (α) subunits are important determining factors for localization and function of receptors. The two main isoforms of α, α1 and α2, are widely distributed in the brain. Especially, the most abundant subunit, α1, is almost ubiquitously present in the brain [
25]. The α1 subunit in the spinal cord is concentrated around the central canal. Although the α2 subunit is widely expressed, the regional expression patterns of the α1 and α2 subunits in the brain are negatively correlated. The α2 subunit is highly present in the cerebellum and forebrain, including the amygdala, hippocampus, olfactory bulb, lateral septum and granule cells of the medial preoptic area of the hypothalamus, and is most abundant in the superficial dorsal horn of the spinal cord. The α3 subunit expression is localized in the olfactory bulb, cerebral cortex and brainstem nuclei, and is also distributed throughout the dorsal horn of the spinal cord and around the central canal. The α4 subunit expression is limited to the hippocampus, thalamus and striatum. The α5 subunit is abundant in the hippocampus and is also expressed in the granule cells of the olfactory bulb, neocortex and hypothalamus, but is less abundant [
26]. It is also weakly expressed in the spinal cord. The expression pattern of the α6 subunit is the most restricted of the α subunits, existing in cochlear nucleus granule cells and cerebellar granule cells [
27]. Taken together, α1, α2 and α3 are mainly concentrated at synaptic sites, whereas α4, α5 and α6 are usually located extrasynaptically [
28].
Table 1.
Main distribution of the alpha subunit of the GABAA receptor.
Table 1.
Main distribution of the alpha subunit of the GABAA receptor.
Subunit |
Main distribution areas |
α1
|
Hippocampus, cerebral cortex, pericentral canal of spinal cord |
α2
|
Cerebellum, forebrain, superficial dorsal horn of spinal cord |
α3
|
Cortical, dorsal horn of the spinal cord and pericentral canal |
α4
|
Striatum, thalamus |
α5
|
Olfactory bulb, hippocampus |
α6
|
Cochlear nucleus granule cells, Cerebellar granule cells |
The involvement of the GABA
A receptor subunit in analgesia has been well documented. Among the GABA
A receptor binding sites, we focus on benzodiazepine receptors, which are often used in chronic pain patients to facilitate their sleeping. However, benzodiazepines do not have clear analgesic efficacy, particularly when they are administered systemically. The high affinity of the benzodiazepine receptor binding site needs the existence of a histidine residue at a conserved site in the
N-terminal structural domain of the α subunit, which is present in the α1, α2, α3, α5 subunits but not in the α4 and α6 subunits. Mutation of histidine residues to arginine effectively reduces the affinity of GABA
A receptors for benzodiazepine receptor binding sites without altering their response to GABA. McKernan
et al. [
29] reported that the sedative effect of benzodiazepines is primarily mediated through the α1-GABA
A receptor as a significant reduction in the sedative effect of benzodiazepines was observed in mice carrying the α1 subunit H→R point mutation. In contrast, one study by Crestani
et al. [
30] found that point mutations in the α2 subunit could lead to a loss of anxiolytic effects of benzodiazepines. Through the study of benzodiazepine-mediated effects in mice with various α-subunit point mutations [
31], it is generally believed that GABA
A receptors containing α1 subtype mediate sedative effects, whereas receptors containing α2/α3 are associated with anxiety and those containing α5 are associated with memory function.
Knockout mice proves to be a powerful tool to dissect the specific involvement of different α subtypes in pain sensation. In a series of studies, specific point mutant mice were created, i.e. knocking out one of four alpha subtype: α1H→R(RHHH), α2H→R(HRHH), α3H→R(HHRH), α5H→R(HHHR) [
32]. Mouse models of inflammatory pain and neuropathic pain were then established by injecting formalin into the hind paw or performing chronic constriction injury surgery (CCI). After intrathecal injection of diazepam into these mice, respectively, the antinociceptive/hyperalgesic effect was found to be significantly reduced in α2 (HRHH) and α3 (HHRH) point-mutant mice, suggesting that GABA
A receptors containing α2 and α3 subtypes in the spinal cord mediated the analgesic effect of intrathecal diazepam administration and were the main contributors [
32]. In addition, a minimal effect of spinal GABA
A receptors containing α5 was found in the α5 (HHHR) point-mutant mice in a model of inflammatory pain. Compound NS11394 is a partial agonist with functional selectivity for α5>α3>α2>α1 GABA
A receptors [
33]. In rodent central sensitization-related pain models, NS11394 can increase mechanical pain and thermal pain thresholds and effectively alleviate arousal behavior in rats. After treated with flumazenil, the effect of NS11394 was reversed, thus demonstrating that NS11394 can produce antinociceptive effects by binding to GABA
A receptors
in vivo [
34]. NS11394 was then made a comparison with other known positive allosteric modulators. Knabl
et al. [
32] selected an anxiolytic and selective GABA
A receptor partial agonist, L-838417, which positively modulates α2, α3 and α5 GABA
A receptors (43%, 43%, and 39%), but lacks α1 GABA
A receptor activity (1.5%) [
35]. L-838417 not only is an α2, α3 and α5 partial agonist, but also is an α1 antagonist. Since α2, α3 and α5 subunits are the major α subunits in the spinal cord, and α1 subunits are mainly expressed in the periaqueductal gray (PAG). L-838417 may positively modulate spinal antinociceptive alloreceptors while blocking central prenociceptive GABA
A receptors. Both actions contribute to its antinociceptive effects, suggesting that L-838417 has analgesic effects in inflammatory and neuropathic pain models in wild-type rats. These findings suggest that α2/α3 selective or α2/α3/α5 selective agonists may be a new class of analgesics that can be used alone or in combination with existing analgesics for disorders associated with inflammatory pain or neuropathic pain. This demonstrates that, in addition to selectivity
per se, a minimal level of positive regulation of individual α subunits (α2, α3 and possibly α5) may be a key determinant of antinociceptive hyperalgesia. On the other hand, TPA023 also activates α2 and α3 GABA
A receptors. Although TPA023 is not as potent as L-838417, it also lacks activity at α1 GABA
A receptors and is significantly less potent at α5 GABA
A receptors [
36]. It is an α2/α3 selective partial agonist, but the analgesic effect is not significant, probably due to insufficient α2/α3 efficacy [
37]. Similarly, compound HZ166, an imidazolobenzodiazepine with selective partial agonism for receptors containing α2 and α3 subunits, showed reversible hyperalgesic effects of flumazenil in mice with yeast polysaccharide A-induced inflammation and CCI-induced mechanical pain hypersensitivity [
38], which demonstrated its analgesic effects in mouse models of neuropathic and inflammatory pain. Taken together, NS11394, L-838417, and HZ166 are all potent agonists (positive allosteric modulators) at α2/α3-GABA
A receptors with
in vivo analgesic efficacies in preclinical studies.
Ralvenius
et al. [
39] also bred four point mutant mice with only one GABA
A receptor subtype remaining benzodiazepine-sensitive, HRRR, RHRR, RRHR, RRRH (mice only with α1 or α2 or α3 or α5-GABAA receptors are sensitive to benzodiazepines), and tested them with diazepam. Although diazepam is a classic non-selective benzodiazepine, its activity was only restricted to a single GABA
A subtype in these triple point mutant mice. The results showed that the order of action of the subtypes was α2>α5>α3 for anti-mechanical hyperalgesia, and α2>α3>α5 for anti-thermal hyperalgesia and chemical injury. These results further support the critical role of α2-GABA
A receptor as an antihyperalgesic target. The lack of consistent analgesic effect of benzodiazepines such as diazepam is likely due to their broad pharmacological actions which overshadow the
bona fide analgesic efficacy. These studies facilitate the enthusiasm of developing subtype-selective (i.e., α2 and α3-specific) benzodiazepines for pain management.
2.2.2. Mechanism of analgesic action of delta(δ) subtype receptors
GABA
A receptors can be divided into synaptic receptors and extrasynaptic receptors based on their distribution within neurons. The two receptor groups also have distinct molecular structural features: synaptic receptors typically contain the γ subtype, while most extrasynaptic receptors are composed of the delta (δ) subtype. Unlike the α subtype GABA
A receptors which are regulated by benzodiazepines through the benzodiazepine receptor binding site, the extrasynaptic delta-subtype-containing GABA
A receptors are insensitive to positive allosteric modulation by benzodiazepines [
40]. Instead, they are highly sensitive to the modulation of neurosteroids.
Peng
et al. [
41] demonstrated the protein expression of δ-GABA
A receptors in the isolated spinal cord by Western blotting. In another study [
42], the presence of δ-GABA
A receptor mRNA was found by reverse transcription-polymerase chain reaction analysis of lamina tissue. These results suggest that δ-GABA
A receptors are present in the spinal cord, By electrophysiological recording, neurons in the spinal cord laminae were found to display a δ-GABA
A receptor-mediated tonic inhibitory current [
43]. One known δ-GABA
A receptor-preferring agonist with analgesic properties is THIP. Studies showd that THIP's enhancement of δ-GABA
A receptor activity reduced the excitability of spinal neurons in wild-type mice, but not in Gabrd
-/- mice (δ-GABA
A receptor-deficient mice). Compared to GABA, THIP excites greater currents from GABA
A receptors containing the δ subunit than from GABA
A receptors lacking the δ subunit [
44]. The δ-GABA
A receptor produced low-amplitude tonic inhibitory currents on spinal lamina II neurons, suggesting its potential as a pharmacological target for reducing acute injurious sensations by modulating central sensitization [
45]. Another compound with well-characterized analgesic efficacy was flupirtine, whose analgesic activity has been demonstrated in various animal models and in humans [
46]. Flupirtine has long been thought of as a selective neuronal potassium (K
+) channel opener [
47]. However, Klinger
et al. [
48] found that flupirtine has different effects on synaptic and extrasynaptic GABA
A receptors and preferentially acts on extrasynaptic GABA
A receptors. This suggests that δ-containing GABA
A receptors may be an important target of flupirtine and its analgesic activities may be at least partially mediated through δ-containing GABA
A receptors. In summary, there are some evidence in the literature suggesting that δ-GABA
A receptors may also be involved in pain modulation and warrants further examination.