GABAA receptor agonists modulate intracellular Ca2+ levels in activated human CD4+ T cells Authors: Sergiy V. Korol, Zhe Jin, Amol K. Bhandage, Hayma Hammoud, Bryndis Birnir Published: 2026 ABSTRACT GABA binding to GABAA receptors induces transient elevation in intracellular Ca2+ concentrations in activated human CD4+ T cells. Here we examined the Ca2+ response evoked by the GABAA receptor agonists THIP, muscimol, isoguvacine, and TACA and further, effects of KV1.3 and KCa3.1 channel blockers on the GABA-evoked Ca2+ signal. The cells were activated by anti-CD3 antibody and cultured for 72 h before the effects of agonists and antagonists on intracellular Ca2+ levels were examined by Ca2+ imaging. mRNA sequencing, qPCR, and immunofluorescence imaging identified the GABAA subunits expressed. Both the half-maximal concentration for activation (the apparent affinity (EC50)) and the evoked, peak cellular Ca2+ response varied among the agonists. The agonists EC50 values varied more than 1,000-fold, but the compounds can, nevertheless, all be classified as high-affinity agonists (EC50 < 1 μM). The EC50 sequence was as follows: GABA (0.005 nM) ≈ THIP (0.005 nM) < muscimol (0.162 nM) < TACA (2.2 nM) < isoguvacine (54.4 nM). In contrast, the peak increase in the cellular Ca2+ signal followed the sequence: GABA >> THIP > muscimol > isoguvacine > TACA. For all agonists, the transient Ca2+ signal was attenuated at micromolar concentrations. ShK-Dap22 and TRAM-34 that block KV1.3 and KCa3.1 channels, respectively, inhibited the GABA-activated Ca2+ response. The CRAC channel inhibitor YM58483 reduced the Ca2+ current by 40%. The pharmacology is consistent with the ρ2 GABAA receptor subunit being a part of the high-affinity GABAA receptors expressed in human CD4+ T cells. The agonists at the ρ2-containing GABAA receptors induced intracellular Ca2+ signals in human CD4+ T cells that were related to the agonist concentration. All tested agonists have a high apparent affinity but differ in efficacy at the receptors. Ca2+ signaling enables high-throughput assays for fast screening of potential drugs aimed at GABAA receptors regulating the T cell functions. Introduction γ-Aminobutyric acid is a well-known inhibitory neurotransmitter and activates GABAA and GABAB receptors in the central nervous system (CNS) and the periphery (1–3). The highest concentration of GABA is in the brain (4), but physiologically meaningful levels of GABA are normally present in the blood (5–7) and various organs, including the pancreatic islets, lymph nodes, testes, and pituitary glands (2, 4). Immune cells do express a variety of receptors activated by their respective neurotransmitters (8). In the plasma membrane of T cells, high-apparent-affinity (EC50 < 1 μM) GABAA receptors are present (9–11). Recently, in human CD4+ T cells, we recorded currents with a pharmacological profile resembling ρ-containing GABAA receptors (12). In studies using expression systems, the pharmacology of homomeric ρ GABAA receptors is significantly different from the more common heteromeric αβγ or αβδ GABAA receptors (1, 13). We have shown in human CD4+ T cells that the activation of GABAA receptors by GABA leads to a graded, transient increase in intracellular Ca2+ signal (12). Interestingly, the GABA EC50 values of the Ca2+ signal and values for the GABA-activated GABAA receptor currents differed by a factor of approximately 10 (12). Here we examined if the GABAA receptor agonists—muscimol, TACA, isoguvacine, and THIP—similarly induce elevations in intracellular Ca2+ concentrations in the activated CD4+ T cells. We further examined whether the GABA-activated Ca2+ signal was affected by the activation of potassium channels. The pharmacological profile and expression of subunits confirm the presence of high-affinity ρ2-containing GABAA receptors in the cells that effectively induce Ca2+ transients when activated by the GABAA receptor agonists. Materials and methods Human CD4+ T lymphocyte isolation and activation Human blood samples and blood buffy coats were obtained from Uppsala University Hospital (Uppsala Akademiska sjukhuset), the protocol of which was approved by the Regional Research Ethics Committee in Uppsala (Dnr 2013/347). All individuals who donated blood were voluntarily recruited. Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples using density-gradient centrifugation with Ficoll-Paque™ PLUS (Sigma Aldrich, Sweden). Subsequently, CD4+ T lymphocytes were isolated from the PBMCs by negative selection, utilizing a human CD4+ T lymphocyte isolation kit (Miltenyi Biotec, Germany) as per the manufacturer’s instructions. The freshly isolated CD4+ T lymphocytes (1 million cells/mL) were suspended in human plasma-like medium (HPLM) (Gibco, Fisher Scientific, Sweden) containing 10% heat-inactivated dialyzed fetal bovine serum (Gibco, Fisher Scientific, Sweden), 100 μg/mL streptomycin, 100 U/mL penicillin (Gibco, Fisher Scientific, Sweden), and 5 μM β-mercaptoethanol (Gibco, Fisher Scientific, Sweden). The cells were activated in a 96-well plate (1 × 105–1.2 × 105 cells/well) with 3 μg/mL plate-bound anti-CD3 antibody (BD Biosciences, USA) for 72 h. Insulin (1 nM) was added at 48 h after the activation of the cells. The anti-CD3 concentration (5 μg/mL) used by Jacobs et al. (14) is similar to the one employed in this study. They showed that when anti-CD3 alone is used, the proliferation of the T cells was approximately one-third of the maximal stimulation obtained when co-stimulated with anti-CD28 at the maximal concentration of anti-CD3 (5 μg/mL). Akt-dependent signaling is activated by insulin at 48 h post-activation and is reminiscent of the anti-CD28-evoked signaling (14). Calcium imaging Calcium imaging was performed as previously described (12). In brief, cells were loaded with 3 μM Fluo-8 AM calcium indicator (AAT Bioquest, Pleasanton, CA, USA) for 10 min at 37 °C and then seeded onto coverslips coated with 5% (3-aminopropyl)triethoxysilane (Sigma Aldrich) for 20 min at 37 °C. Calcium imaging was performed using a Zeiss LSM700 confocal microscope with a ×20 objective, capturing images at 0.5-s intervals. The recording solution was prepared by mixing two types of RPMI 1640 media, one containing 11 mM glucose (Gibco, Fisher Scientific, Sweden) and the other is glucose-free (Gibco, Fisher Scientific, Sweden). They were mixed in a 1:1 ratio, giving a final glucose concentration of 5.5 mM. The drugs were diluted in the recording solution and perfused into the chamber at specific concentrations using a peristaltic pump. After each drug application, the cells were perfused with the recording solution only to remove the drug. When the cells were again in the recording solution only, then the next test drug concentration was perfused into the chamber. This was done so that the receptors/cells had always been in a drug-free solution before a drug application, thus avoiding potential cumulative drug effects such as run-down or desensitization. The area around the individual cells was marked, and absolute fluorescence intensity values (F) were extracted using ZENBlue software. The relative intensity (F/Fmedian) for each cell was calculated for every recorded image and plotted against time. The peak relative intensity during the drug application was considered the drug’s response. Data are presented for all cells as scatter plots and box and whisker plots. Electrophysiological patch-clamp recordings To study the effect of YM58483 and picrotoxin (PTX) on CRAC-channel-mediated currents, the electrophysiological patch-clamp recordings were done in the whole-cell configuration. The extracellular solution was composed of the following (in mM): 155 NaCl, 4.5 KCl, 2 CaCl2, 1 MgCl2, 5 HEPES, and 5.6 glucose (pH 7.4, adjusted with NaOH). The pipette solution consisted of the following (in mM): 140 Cs methanesulfonate, 10 NaCl, 3 MgCl2, 10 HEPES, and 10 EGTA (pH 7.3, with CsOH). We further examined the recovery of currents during repeated drug applications to evaluate the level of run-down and desensitization, if present. In these experiments, in order to keep the intracellular chloride concentration intact, gramicidin (2.6 μg/mL) was used to obtain the perforated configuration. The extracellular solution (in mM) contained the following: 157 NaCl, 4.5 KCl, 0.5 CaCl2, 1 MgCl2, 5 HEPES, and 5.6 glucose (pH 7.4, adjusted with NaOH), and the pipette solution consisted of the following (mM): 149 KCl, 2 CaCl2, 1 MgCl2, 1 NaCl, 10 HEPES, and 5.6 glucose (pH 7.3, adjusted with KOH). The patch-clamp pipettes were produced from borosilicate glass and had a resistance range of 8–11 MΩ when filled with the pipette solution and brought into contact with the extracellular solution. Recordings were performed with the help of an Axopatch 200B amplifier, filtered at 2 kHz, and digitized at 10 kHz with an analog-to-digital converter. For the recordings, the Clampex 10.5 (Molecular Devices, San Jose, CA, USA) program was used. For current recordings, the following procedure was used: 200-ms voltage ramps (from -120 to +60 mV) were applied every 17 s from a holding potential -40 mV. After getting stable ramp recordings in control conditions under continuous recording chamber perfusion, the drug (GABA, YM58483, or picrotoxin) dissolved in the extracellular solution was applied to the bath. After the ramp was stabilized in the presence of the drug, the latter was washed out before the next current recording. The current values were taken at -80 and -40 mV and normalized to the current in the absence of the drugs. Chemical compounds The chemicals, unless otherwise specified, were purchased from Sigma-Aldrich/Merck (Steinheim, Germany). GABAA receptor agonists muscimol, isoguvacine, trans-4-aminocrotonic acid (TACA), and THIP hydrochloride were purchased from Tocris (UK) and used to examine the pharmacological response of the GABAA receptors (15) expressed in T cells. The agonists’ chemical structures (1, 13, 16) are shown in Figure 1. TACA (trans-3-aminocrotonic acid) is a GABA analog with an unsaturated carbon–carbon bond, whereas muscimol and isoguvacine are heterocyclic GABA analogs. THIP (4,5,6,7-tetrahydroisoxazolo(5,4-c)pyridin-3-ol) is a relatively rigid analog of GABA. Chemical structures of GABAA receptor agonists. Trans-4-aminocrotonic acid (TACA) is a GABA analog with an unsaturated carbon–carbon bond, whereas muscimol and isoguvacine are heterocyclic GABA analogs. 4,5,6,7-tetrahydroisoxazolo(5,4-c)pyridin-3-ol (THIP) is a relatively rigid analog of GABA. Total RNA isolation and quantitative PCR Total RNA was extracted from CD4+ T cells using the NucleoSpin RNA Mini kit (Macherery-Nagel, Germany) with on-column DNase treatment to remove genomic DNA. The extracted RNA was quantified with the NanoDrop 2000 spectrophotometer (Thermo Scientific, USA). Reverse transcription was performed using SuperScript reverse transcriptase IV (ThermoFisher Scientific, USA), according to the manufacturer’s protocol. Quantitative PCR was performed using qPCRBIO SYBR Mix low-ROX (PCR Biosystems, UK) and gene-specific primers (6) in a 384-well plate using a QuantStudio™ 6 Flex Real-Time PCR system (ThermoFisher Scientific, USA). The PCR cycling protocol consists of an initial denaturation step of 2 min at 95 °C, followed by 45 cycles of 95 °C for 5 s and 60 °C for 30 s. The relative expression levels of the target genes were expressed as 2-ΔCt (Ct, cycle threshold) by normalizing to reference genes (IPO8). All samples were run in duplicate. Immunofluorescence staining CD4+ T cells were collected, washed with PBS, and deposited on cytoslides (ThermoFisher Scientific, Sweden) using a Shandon Cytospin centrifuge (400 rpm for 4 min at room temperature (RT)). The cells were fixed with 4% paraformaldehyde (PFA) for 15 min and washed three times with PBS. Triton X-100 (0.1%) was applied for 5 min to permeabilize the cell membrane. The cells were further blocked with 5% bovine serum albumin and 10% donkey serum (Jackson ImmunoResearch) diluted in PBS for 60 min. After PBS washing, the cells were incubated with primary antibodies—rabbit anti-insulin receptor (ThermoFisher Scientific, 1:500), rabbit anti-GABRR2 (ThermoFisher Scientific, 1:600), mouse anti-GABRB3 (UC Davis, USA, 1:500)—diluted in blocking solution overnight at 4 °C, followed by incubation with a secondary antibody—AlexaFluor 488-conjugated AffiniPure Donkey Anti-Rabbit IgG (Jackson ImmunoResearch, 1:500) or Rhodamine Red X-conjugated AffiniPure Donkey anti-Mouse IgG (Jackson ImmunoResearch, 1:600)—for 1 h at RT. The cells were washed and counterstained with DAPI for 5 min. The cytoslides were mounted in ProLongTM Gold antifade reagent (ThermoFisher Scientific, USA). Images were acquired using a confocal microscope (LSM700, Zeiss) with ×63 objective. Data presentation Single-cell data presentation and agonist-Ca2+-response data were fitted in GraphPad Prism 10 (GraphPad Software Inc., La Jolla, CA, USA). The data were fitted using nonlinear regression with a standard dose–response model: where (F/Fmedian)max is the peak Ca2+ response to different concentrations of the agonist, [agonist] is the concentration of the agonist (GABA, muscimol, TACA, isoguvacine, or THIP), and (EC50)agonist is the half-maximal concentration of the agonist. Statistical analysis For the comparison of two groups, paired t-test or Wilcoxon test for paired data or Mann–Whitney test for unpaired data was used. The statistical significance level was set to 0.05. Results Agonist-evoked cytoplasmic Ca2+ response In human-activated CD4+ T cells, the GABA activation of GABAA receptors leads to a graded, transient increase in the intracellular Ca2+ signal (12). In contrast, GABA has no effect on the intracellular Ca2+ levels in human resting CD4+ T cells (Supplementary Figure 1). We examined if muscimol, isoguvacine, TACA, and THIP similarly induced Ca2+ signals in the human CD4+ T cells and then how they were related to the agonist concentration. Typical Ca2+ signal-intensity responses are exemplified by muscimol applications to the CD4+ T cells (Figure 2A). Increasing the agonist concentration evoked larger Ca2+ signals until a maximal response was attained, and then at supersaturating concentrations, the Ca2+ signal decreased in accordance with the desensitization of GABAA receptors at saturating agonist concentrations (17). Muscimol is a selective and potent agonist of most GABAA receptors (13). The applications of muscimol to the CD4+ cells revealed a dose-dependent Ca2+ response in the range of concentrations from 10 pM to 1 μM muscimol (Figures 2B, C). The relative maximum Ca2+ signal of approximately 1.17 was recorded at 1 μM muscimol (Figures 2B, C). At higher concentrations, the Ca2+ signal decreased. The data were fitted to a dose–response curve, and the estimated half-maximal activation concentration, EC50, was 0.162 nM. TACA is an agonist at heteromeric GABAA receptors but is more selective for homomeric ρ receptors (13). The relative maximum Ca2+ signal in 500 nM TACA was 1.10 (Figures 2D, E). At higher concentrations, the Ca2+ response sharply decreased. The data were fitted to a dose–response model, and the estimated half-maximal concentration was calculated, EC50 = 2.2 nM. Isoguvacine is effective at many GABAA receptors but has a significantly lower EC50 at homomeric ρ2 receptors compared with GABA (16). The relative maximum Ca2+ signal in 10 μM isoguvacine was 1.13 (Figures 2F, G). At higher concentrations, the Ca2+ response decreased. The data were fitted to a dose–response model, and the estimated half-maximal concentration was calculated, EC50 = 54.4 nM. THIP or gaboxadol is a potent GABAA receptor agonist with some preference for receptors containing the δ subunit in the receptor complex (18, 19). The relative maximum Ca2+ signal in 100 nM THIP was 1.20 (Figures 2H, I). At higher concentrations, the Ca2+ response decreased slightly. The data were fitted to a dose–response model, and the estimated half-maximal concentration was calculated, EC50 = 0.005 nM. Relative calcium responses to different concentrations of the GABAA receptor agonists examined in human CD4+ T lymphocytes. Cells were activated and cultured for 72 h, insulin (1 nM) was included for the last 24 h, and glucose concentration was 5.5 mM. (A) Representative mean Ca2+ signal intensity trace (blue line) and SEM (light pink whiskers) recorded in cells (n = 35 from one donor) at 72 h post-activation. The bars show the perfusion time of the drugs. Between applications of muscimol, the recording chamber was perfused with the bath solution. Relative calcium signals evoked by (B) muscimol (n = 72 cells, N = 6 donors), (D) TACA (n = 66 cells, N = 3 donors), (F) isoguvacine (n = 77 cells, N = 3 donors), and (H) THIP (n = 92 cells, N = 4 donors) in cells from different donors (marked by different colors). Mean dose–response values (mean ± SEM), from the lowest agonist concentration to the concentrations giving the maximal response for muscimol (C), TACA (E), isoguvacine (G), or THIP (I) and fitted with a standard dose–response curve model. The results show that the apparent affinity (EC50) and the efficacy of the induced Ca2+ signals varied among the agonists (Figure 2). The EC50 values are related to the potency of the compounds and resulted in the following sequence: GABA (0.005 nM) (12) ≈ THIP (0.005 nM) < muscimol (0.162 nM) < TACA (2.2 nM) < isoguvacine (54.4 nM). In contrast, efficacy, which is related to the maximal cellular Ca2+ signal, resulted in the following sequence: GABA (12) >> THIP > muscimol > isoguvacine > TACA. Interestingly, GABA remained both most potent and with the highest efficacy of the five agonists tested in this study. Expression of GABAA receptor subunits in the CD4+ T cells We have previously shown that human immune cells can express a variety of GABA receptor subunits (6, 20) and that, in the CD4+ T cells, insulin shifts the receptors GABA EC50 to lower concentrations. This is, in part, related to the increased expression of the insulin receptor with time after the activation of the T cells (Supplementary Figure 2) as insulin enhances the expression of the GABAA receptors (12). We examined the expression level of the GABAA receptor subunits in the CD4+ T cells 72 h after activation (Figures 3A–C). We studied which GABAA receptor subunits could be detected in the cells. Interestingly, the ρ2 subunit was most highly expressed in the activated CD4+ T cells (Figures 3A–C), but a few other subunits were expressed, albeit at low levels, including β3 that could also be detected at the protein level in the cells (Figure 3B). Clearly, both homomeric and heteromeric ρ-containing GABAA receptors may be present in the CD4+ T cells. Expression of GABAAR subunits in activated human CD4+ T cells. (A, B) GABAAR subunit mRNA expression detected by RNAseq (A) (N = 6) and qPCR (B) (N = 5) in activated CD4+ T cells at 72 h post-activation. (C) Immunofluorescent staining of GABRR2 (ρ2) and GABRB3 (β3) in activated CD4+ T cells at 72 h post-activation. The nuclei were counterstained with DAPI. RNAseq data were extracted from the Gene Expression Omnibus (GEO) database (GSE230041). GABA-activated Ca2+ signals are reduced by CRAC and K+ channel blockers We have previously shown the involvement of the CRAC channels in the GABA-activated cascade (12) where the acutely applied 1 μM YM58483 specifically blocked the channels (21). A higher concentration (10 μM YM58483) acutely applied also reversibly inhibited the Ca2+ current and the Ca2+-evoked signal (Figures 4A, B; Supplementary Figures 3A, B) in accordance with other studies (21–23). Picrotoxin, an inhibitor of GABAA receptors (24), did not reduce the Ca2+ current (Figures 4C, D). YM58483 reduced the Ca2+ current by approximately 40%, indicating that even other plasma membrane Ca2+ channels may be activated, may be open, and may allow Ca2+ entry in the activated T cells. Previous studies have established that potassium ion channels not only maintain the resting membrane potential in T cells but also enhance calcium entry through the CRAC channels (25). The KV1.3 and/or KCa3.1 channels are expressed at a sufficiently high density so that their activation prevents the cell from depolarizing even when CRAC channels are open (25). Thus, the K+ channels contribute to maintaining a high driving force for Ca2+ influx. As the Cl- equilibrium potential is more depolarized than the resting membrane potential (12, 25), it is possible that opening the GABAA receptor channels will promote depolarization and increase the openings of the KV1.3 channel, which will enhance Ca2+ influx through the CRAC channels and, in turn, enhance further the activation of KCa3.1 channels (25). We examined if this coordinated channel activity took place by activating the GABAA receptors in the absence or presence of blockers for these K+ channels, and the results are shown in Figures 4E, F. Both ShK-Dap22 (Figure 4E), a potent blocker of KV1.3 channels, and TRAM-34 (Figure 4F), a selective blocker of KCa3.1 channels, significantly reduced the GABA-activated Ca2+ signal in T cells. The results are consistent with the coordinated effects of the GABAA receptors—KV1.3, KCa3.1, and CRAC channels—on the Ca2+ signal in T cells upon application of GABAA receptor agonist to the cells. Inhibition of ion channels in human CD4+ T cells. Human CD4+ T cells derived from healthy donors were activated for 72 h in a medium with 5.5 mM glucose, and 1 nM insulin was added in the last 24 h of activation. (A) Inhibition of CRAC channel current by YM58483. (B) Statistical analysis of the YM58483 experiments (P = 0.0312, n = 5 cells from N = 2 donors, Wilcoxon matched-pairs signed rank test). (C) Picrotoxin (PTX) does not block the CRAC channel current. (D) Statistical analysis of the PTX experiments (P = 0.5, n = 3 cells from N = 2 donors, Wilcoxon matched-pairs signed rank test). (A–D) Voltage ramps of 200 ms (from -120 to +60 mV) were applied every 17 s from a holding potential of -40 mV in a whole-cell configuration. After getting stable ramp recordings in control conditions under continuous recording chamber perfusion, the drug (YM58483 or PTX) was applied to the bath. After the ramp was stabilized in the presence of the drug, the latter was washed out. The current values were taken at -80 mV at every condition and were normalized to the control current. (E) The GABA (500 nM)-induced increase in the intracellular Ca2+ levels is inhibited by ShK-Dap22 (500 pM) by 47% on average (n = 151 cells, N = 3 donors). (F) TRAM-34 (1 µM) inhibited the increase in the intracellular Ca2+ levels by almost 100% on average (n = 169 cells, N = 3 donors). The box-and-whiskers plot represents the maximum relative intracellular Ca2+ levels (F/Fmedian)max achieved during the drug application period from individual cells. The blue, green, and red dots indicate cells from a specific human donor (p < 0.0001). Discussion In activated human CD4+ T cells, GABA binding to GABAA receptors leads to a transient increase in intracellular Ca2+ concentrations that modulate the metabolism and secretion of inflammatory molecules from the cells (12). Here we examined the ability of several GABAA receptor agonists to induce Ca2+ signaling in human CD4+ T cells. THIP, muscimol, isoguvacine, and TACA all induced graded, concentration-dependent Ca2+ signals in the CD4+ T cells, but they varied in both the apparent affinity and efficacy, and none surpassed GABA (Figure 5A). The GABA-activated Ca2+ influx was dependent on active K+ channels (Figure 5B). The effects of GABAA receptor agonists on CD4+ T cell Ca2+ homeostasis enable a “tap-like” on/off effect on intracellular Ca2+, with significant potential to modulate the cells. Furthermore, the enabled Ca2+ signal may be a very effective tool when screening for new, efficacious GABAA receptor agonists regulating T cell functions. Dose–response curves of the Ca2+ signals for GABAA receptor agonists and a model for the GABA activation of the Ca2+ influx in the CD4+ T cells. (A) Family of dose–response curves for the GABAA receptor agonists muscimol, TACA, isoguvacine, and THIP that induced Ca2+ signals in the CD4+ T cells, activated for 72 h and incubated with insulin (1 nM) for the last 24 h (glucose concentration 5.5 mM). For comparison, the dose–response curve for GABA is also shown (12). (B) In activated human CD4+ T cells, at 72 h post-activation, the voltage-gated K+ channel, KV1.3, and the Ca2+-activated K+ channel, KCa3.1, are active and maintain the resting membrane potential (ΨRMP) of the cells. When agonists of GABAA receptors (GABAAR) are applied (red circle), the GABAA receptor channels open. Since the Cl– reversal potential in T cells is more positive than the resting membrane potential, opening GABAA receptor channels depolarizes the cell membrane potential (Ψdepol) that, in turn, will result in opening many more KV1.3 channels that then hyperpolarize the membrane potential (Ψhyperpol). This hyperpolarization of the membrane potential increases the driving force for Ca2+ ion influx through CRAC channels. Increased intracellular Ca2+ concentration then increases KCa3.1 channel activity that, together with KV1.3 channels, contributes to the further increase in Ca2+ influx through the CRAC channels. Among the plethora of potential heteromeric GABAA receptors expressed in neurons and other cell types in the periphery (2, 26–28), ρ-containing receptors have a more restricted distribution and subunit composition. The ρ2 subunit can be found in the hippocampus, cerebellum, and spinal cord (29–31) but is mostly associated with the bipolar and horizontal cells in the retina (29, 32, 33). It is mostly studied as a ρ homomeric receptor in Xenopus oocytes (13, 16) and has also been reported to form heteromeric receptors, with at least the γ2 and potentially the α1 subunit (34–36). In human peripheral blood mononuclear cells, the ρ2 subunit is the most highly expressed GABAA receptor subunit (20), similar to what we report here for CD4+ T cells. Furthermore, insulin enhances the expression of ρ2 after the activation of CD4+ T cells (12). However, as other subunits can also be expressed (11), not only homomeric but also heteromeric ρ2-containing GABAA receptors may potentially be formed in CD4+ T cells. How GABAA receptors’ activation evokes a Ca2+ signal in T cells is not well understood. However, we have previously shown that the signal can be inhibited by a CRAC channel blocker at 1 μM concentration (12). The CRAC channel blocker reduced the Ca2+ current by 40%, indicating that even other Ca2+ channels may participate in the cascade. In this study, blockers of potassium channels KV1.3 and KCa3.1 reduced the intracellular Ca2+ increase. This is in line with the established, coordinated channel activation observed in T cells after T cell activation, which has been shown to regulate Ca2+ entry into the cells (25). Our findings demonstrate that GABA and other GABAA receptor agonists can transiently activate this cascade. The Cl– reversal potential is normally less negative than the resting membrane potential in human CD4+ T cells (12, 25). Thus, the activation of GABAA receptors leads to membrane potential depolarization in the cells. In T cells, the resting membrane potential is maintained by K+ channels, where it depends on the state of the cells if it is predominantly regulated by KV1.3 or KCa3.1 (25). These channels are expressed at sufficiently high density (25) so that their activation prevents the cell from depolarizing even when the CRAC channels are open. These K+ channels thus maintain a high driving force for Ca2+ influx as long as they are open. When the GABAA receptor channels are opened, the membrane potential will depolarize and a channel activation cascade starts (Figure 5B). KV1.3 channels that require depolarization for opening will open and shift the membrane potential to hyperpolarizing values, which increases Ca2+ influx through the CRAC channels, KCa3.1 will increase in activity with increased intracellular Ca2+, thus maintaining the hyperpolarized membrane potential and the inward driving force for the Ca2+ flux. The result of activating the channel cascade, which can be started by the GABAA receptor agonists, is a transient increase in the intracellular Ca2+ in CD4+ T cells (Figure 5B). We have previously shown that the GABAA-receptor-specific agonists (15), GABA, muscimol, and TACA, activate GABAA receptor Cl- currents in human CD4+ T cells that are blocked by picrotoxin, a specific GABAA receptor open channel blocker (6, 12, 15, 37, 38). Muscimol (500 nM) or TACA (500 nM) also evoked Ca2+ signals that were blocked by picrotoxin (12). The GABA EC50 value (0.04 nM) for the GABA-activated Cl- current was larger than the EC50 value (0.005 nM) of the GABA-evoked Ca2+ signal in human CD4+ T cells (12). This difference in the apparent affinity (EC50) for GABA is probably related to the high sensitivity and amplification of the Ca2+ signal due to the activation of the potassium channels that then increase the driving force for Ca2+ entry, thus enabling the sensitization of GABA signaling. The GABA apparent affinity is approximately 105-fold higher in human CD4+ T cells than what is reported for the ρ-containing GABAA receptors expressed in Xenopus oocytes (13) and retinal receptors (34) but similar to the extrasynaptic GABAA receptors in rat CA1 pyramidal neurons (39). Whether the higher apparent affinity (lower EC50) in CD4+ T cells is related to the incorporation of other subunits than ρ2 in GABAA receptors (34), different intracellular modulations or tethering (28, 40–42) or modulation by other native molecules (27, 28, 39, 41) is not known to date. The agonists’ apparent affinity sequence for the homomeric GABAA ρ1–3 receptors expressed in Xenopus oocytes is somewhat different from the one we reported here for the GABA-evoked Ca2+ signals (13, 16). The EC50 value for the Ca2+ signal evoked by GABA was shifted by approximately 106-fold toward higher affinity in CD4+ T cells compared with the Xenopus oocytes’ EC50 for the GABA-activated Cl- currents (13), whereas this difference was only approximately 10 in the CD4+ T cells (12). TACA, on the other hand, had lower apparent affinity and approximately 500-fold higher EC50 value for the Ca2+-evoked signal than GABA. For comparison, in heterologously expressed receptors, the TACA’s EC50 value based on Cl- currents was approximately 10 times lower than that of GABA (13). Isoguvacine had the lowest apparent affinity of all the agonists in both systems. Interestingly, THIP, which is an antagonist at least at GABAA ρ1 and ρ3 receptors (13), is, in contrast, a high-apparent-affinity agonist in human CD4+ T cells, similar to GABA. However, THIP has lower efficacy in inducing the Ca2+ response than GABA and, in this aspect, is more similar to muscimol that had intermediate EC50 for the Ca2+ response. GABA did not only have the highest apparent affinity but was also the most efficacious agonist for evoking the Ca2+ signal in human CD4+ T cells, followed by THIP and muscimol and then by isoguvacine and TACA (Figure 5A). At supersaturating concentrations of the agonists, the Ca2+ signal decreased as expected since GABAA receptors desensitize faster and more completely at high agonist concentrations (17, 43). In CD4+ T cells, GABA not only induces transient Ca2+ signals in the cells but also decreases the activity and protein level of hexokinase, the gatekeeper enzyme of the glycolytic pathway (12). Whether the different agonists used in this study are similarly effective at modulating hexokinase remains to be determined. There are a number of other GABA analogues (see, e.g., (13, 16)) that may be interesting to explore and to examine whether they evoke or inhibit GABA-evoked Ca2+ signals in human CD4+ T cells. These and other compounds can potentially serve as the so-called lead compounds for drug development in drug discovery programs targeting T cell function.