Inhibition of G Protein-Activated Inwardly Rectifying K+ Channels by Different Classes of Antidepressants

Various antidepressants are commonly used for the treatment of depression and several other neuropsychiatric disorders. In addition to their primary effects on serotonergic or noradrenergic neurotransmitter systems, antidepressants have been shown to interact with several receptors and ion channels. However, the molecular mechanisms that underlie the effects of antidepressants have not yet been sufficiently clarified. G protein-activated inwardly rectifying K+ (GIRK, Kir3) channels play an important role in regulating neuronal excitability and heart rate, and GIRK channel modulation has been suggested to have therapeutic potential for several neuropsychiatric disorders and cardiac arrhythmias. In the present study, we investigated the effects of various classes of antidepressants on GIRK channels using the Xenopus oocyte expression assay. In oocytes injected with mRNA for GIRK1/GIRK2 or GIRK1/GIRK4 subunits, extracellular application of sertraline, duloxetine, and amoxapine effectively reduced GIRK currents, whereas nefazodone, venlafaxine, mianserin, and mirtazapine weakly inhibited GIRK currents even at toxic levels. The inhibitory effects were concentration-dependent, with various degrees of potency and effectiveness. Furthermore, the effects of sertraline were voltage-independent and time-independent during each voltage pulse, whereas the effects of duloxetine were voltage-dependent with weaker inhibition with negative membrane potentials and time-dependent with a gradual decrease in each voltage pulse. However, Kir2.1 channels were insensitive to all of the drugs. Moreover, the GIRK currents induced by ethanol were inhibited by sertraline but not by intracellularly applied sertraline. The present results suggest that GIRK channel inhibition may reveal a novel characteristic of the commonly used antidepressants, particularly sertraline, and contributes to some of the therapeutic effects and adverse effects.

G protein-activated inwardly rectifying K + (GIRK) channels (also known as Kir3 channels) are members of a major subfamily of inwardly rectifying K + (Kir) channels that includes seven subfamilies [24]. Four GIRK channel subunits have been identified in mammals [25][26][27]. Neuronal GIRK channels are predominantly heterotetramers composed of GIRK1 and GIRK2 subunits in most brain regions or homotetramers composed of GIRK2 subunits in the substantia nigra [27][28][29][30], whereas atrial GIRK channels are heterotetramers composed of GIRK1 and GIRK4 subunits [26]. The channels are activated by various G i/o -proteincoupled receptors, such as M 2 muscarinic, a 2 adrenergic, D 2 dopaminergic, opioid, nociceptin/orphanin FQ, CB 1 cannabinoid, and A 1 adenosine receptors, through the direct action of Gprotein bc subunits [31][32][33]. Additionally, ethanol activates GIRK channels independently of G-protein-coupled signaling pathways [34,35]. GIRK channels play an important role in regulating neuronal excitability, synaptic transmission, and heart rate [31,[36][37][38][39]. Furthermore, recent studies have suggested that GIRK channel modulation has the potential for treating several neuropsychiatric disorders and cardiac arrhythmias [33,40,41]. Therefore, GIRK channel modulators may affect various brain and cardiac functions. We have demonstrated the distinctive effects of several antidepressants on GIRK channels, even among the same class, particularly SSRIs [42,43]. To further clarify the interaction between various classes of commonly used antidepressants and GIRK channels may be useful for advancing our understanding of the pharmacological effects of antidepressants. In the present study, we examined the effects of various antidepressants on GIRK channels using the Xenopus oocyte expression assay.

Preparation of specific mRNAs
Plasmids that contain the entire coding sequences for the mouse GIRK1, GIRK2, and GIRK4 channel subunits were obtained previously [34,44,45]. cDNAs for mouse Kir2.1 in pcDNA1 [46] were generously provided by Dr. Lily Y. Jan (University of California, San Francisco). These plasmids were linearized by digestion with the appropriate enzymes as described previously [45,46]. The specific mRNAs were synthesized in vitro using the mMESSAGE mMACHINE TM In Vitro Transcription Kit (Ambion, Austin, TX, USA).

Electrophysiological analysis
Adult female Xenopus laevis frogs (Copacetic, Soma, Aomori, Japan) were anesthetized by immersion in water that contained 0.15% tricaine (Sigma-Aldrich, St. Louis, MO, USA). A small incision was made on the abdomen to remove several ovarian lobes from the frogs, which were humanely killed after the final collection. All procedures for the care and treatment of animals were performed in accordance with National Institutes of Health guidelines and were approved by the Institutional Animal Care and Use Committee of Niigata University (Permit Number: 172-2). Xenopus oocytes (Stages V and VI) were manually isolated from the ovary and maintained in Barth's solution [47]. Oocytes were injected with mRNA for GIRK1/ GIRK2 or GIRK1/GIRK4 combinations (0.15 ng each) or Kir2.1 (0.3 ng). The oocytes were incubated at 19uC in Barth's solution and manually defolliculated after treatment with 0.8 mg/ml collagenase (Wako Pure Chemical Industries, Osaka, Japan) for 1 h. The wholecell currents of the oocytes were recorded from 3 to 9 days after injection with a conventional two-electrode voltage clamp [34,48]. The membrane potential was held at 270 mV unless otherwise specified. Microelectrodes were filled with 3 M KCl. The oocytes were placed in a 0.05 ml narrow chamber and continuously superfused with a high-potassium (hK) solution (96 mM KCl, 2 mM NaCl, 1 mM MgCl 2 , 1.5 mM CaCl 2 and 5 mM HEPES, pH 7.4 with KOH) or a K + -free high-sodium (ND98) solution (98 mM NaCl, 1 mM MgCl 2 , 1.5 mM CaCl 2 and 5 mM HEPES, pH 7.4 with NaOH) at a flow rate of 2.5 ml/min. In the hK solution, the K + equilibrium potential was close to 0 mV, and the inward K + current flow through the Kir channels was observed at negative holding potentials as previously shown [25,27,43]. Additionally, to examine the effects of antidepressants on outward K + currents, a perfusion solution that contained 4 mM K + (K4 solution) was made by substituting NaCl with KCl in the ND98 solution. To examine the effects of an antidepressant on GIRK channels activated by Gprotein activation, 13.8 nl of 100 mM Li 4 -guanosine-59-O-(3-thiotriphosphate) (GTPcS; Sigma-Aldrich), a nonhydrolyzable G-protein activator, dissolved in distilled water was injected into an oocyte using a nanoliter injector (World Precision Instruments, Sarasota, FL, USA) as described previously [49]. Furthermore, to examine the effects of intracellular sertraline, 23 nl of 10 mM sertraline dissolved in distilled water was injected into an oocyte using a Nanoliter injector as described previously [50], and the oocyte currents were then continuously recorded for approximately 30-40 min. Because the volume of the Xenopus oocytes used was approximately 1 ml, the intracellular concentration of sertraline was presumed to be approximately 225 mM. For the analysis of concentration-response relationships, the data were fitted to a standard logistic equation [51] using KaleidaGraph (Synergy Software, Reading, PA, USA). The concentration of a drug that produces 50% of the maximal current response for that drug (EC 50 ), the concentrations required to reduce control currents by 25% and 50% (IC 25 and IC 50 , respectively), and the Hill coefficient (n H ) were obtained from the concentrationresponse relationships.

Data analyses
The data are expressed as mean 6 SEM, and n is the number of oocytes tested. The statistical analysis of differences between groups was performed using paired t-test, one-way analysis of variance (ANOVA), or two-way ANOVA followed by the Tukey-Kramer post hoc test. Values of P,0.05 were considered statistically significant.

Compounds
All of the antidepressants tested were commercially purchased. Amoxapine and nefazodone hydrochloride were obtained from Sigma-Aldrich. Mirtazapine and mianserin hydrochloride were obtained from Tocris Bioscience (Bristol, UK). Sertraline hydrochloride and duloxetine hydrochloride were obtained from Tronto Research Chemicals (North York, Canada). Venlafaxine hydrochloride was obtained from LKT Laboratories (St. Paul, MN, USA). Sertraline was dissolved in dimethyl sulfoxide (DMSO) or distilled water, and venlafaxine was dissolved in distilled water. The other antidepressants were dissolved in DMSO. The stock solution of each compound was stored at 230uC until use. Ethanol was purchased from Wako Pure Chemical Industries. Each compound was added to the perfusion solution in appropriate amounts immediately before the experiments.

Inhibition of GIRK channels by antidepressants
In Xenopus oocytes injected with GIRK1 and GIRK2 mRNAs, basal GIRK currents, which depend on free G-protein bc subunits present in the oocytes because of the inherent activity of Gproteins [32], were observed at a holding potential of 270 mV in an hK solution that contained 96 mM K + (Fig. 1A). The 3 mM Ba 2+ -sensitive current components (1042.8690.1 nA, n = 30) correspond to the magnitude of GIRK currents in oocytes that express GIRK channels [34]. Extracellular application of 30 mM sertraline, an SSRI, reversibly reduced the inward currents through the expressed GIRK channels (Fig. 1A). The current responses to an additional 100 mM sertraline during the application of 3 mM Ba 2+ , which blocks Kir channels, were not significant (reduction of inward currents by 4.563.5 nA; less than 1% inhibition of the Ba 2+ -sensitive current components, n = 4). Sertraline at 100 mM produced no significant response in a K +free ND98 perfusion solution that contained 98 mM Na + instead of the hK solution (3.061.8 nA, n = 4), suggesting that the SSRIsensitive current components show K + selectivity. Additionally, the application of DMSO or distilled water, the solvent vehicles, at the highest concentration (0.3%) induced no significant current response in the hK or ND98 solutions (n = 5; data not shown). In contrast, in oocytes injected with mRNA for Kir2.1, a constitutively active Kir channel [46], extracellular application of 300 mM sertraline had no significant effect on the inward currents through the channels in the hK solution (less than 2% change of the Ba 2+ -sensitive current components; 848.36322.0 nA, n = 4; Fig. 1B). In uninjected oocytes, 300 mM sertraline and 3 mM Ba 2+ caused no significant response (2.062.0 nA, n = 4, and 3.161.7 nA, n = 4, respectively; Fig. 1C) compared with oocytes injected with GIRK mRNA, suggesting no significant effect of sertraline or Ba 2+ on intrinsic oocyte channels. Furthermore, in oocytes injected with GIRK1 and GIRK4 mRNAs, 30 mM sertraline similarly inhibited basal GIRK currents under the same conditions (51.664.3% inhibition of 3 mM Ba 2+ -sensitive current components, 561.7658.2 nA, n = 11). Additionally, the Ba 2+sensitive current components in oocytes injected with mRNA for GIRK1/GIRK2 or GIRK1/GIRK4 combinations were very significantly larger than those in oocytes injected with the same small amount of a single GIRK mRNA (less than 20 nA, n = 7, respectively). The results indicate that sertraline predominantly inhibited GIRK1/2 and GIRK1/4 heteromultimeric channels, but not Kir2.1 channels. Moreover, the effects of different classes of antidepressants on GIRK channels were examined using the same expression assay. Amoxapine, a second generation TCA, and duloxetine, an SNRI, significantly inhibited basal GIRK currents at 100 mM (45.664.4 and 65.661.0% inhibition for GIRK1/2, n = 5 and 7, respectively; 27.664.0 and 49.762.2% inhibition for GIRK1/4, n = 4 and 6, respectively). However, the 5-HT 2 receptor antagonist nefazodone, NaSSAs mianserin and mirtazapine, and SNRI venlafaxine weakly inhibited the currents at 100 mM (35.963.5, 24.165.5, 17.663.5, and 19.464.6% inhibition for GIRK1/2, n = 11, 4, 4, and 4, respectively; 30.363.4, 18.861.3, 12.162.4, and 22.863.2% inhibition for GIRK1/4, n = 13, 5, 5, and 5, respectively). Additionally, the inhibitions were reversible with washout, similar to sertraline (data not shown). In contrast, Kir2.1 channels were insensitive to these drugs at 100 mM (less than 4% change of the Ba 2+ -sensitive current components; 912.56182.8 nA, n = 4). In uninjected oocytes, 300 mM of the drugs caused no significant response (less than 6 nA; n = 4 for each of the drugs). Altogether, the results suggest significant inhibition of GIRK channels by sertraline, duloxetine, and amoxapine, weak inhibition of the channels by nefazodone, mianserin, mirtazapine, and venlafaxine, and no significant effects of the drugs on Kir2.1 channels.

Concentration-dependent inhibition of GIRK channels by various antidepressants
The concentration-response relationships for the inhibitory effects of different classes of antidepressants on GIRK1/2 and GIRK1/4 channels were investigated. Figure 2 shows that the inhibitions of both types of GIRK channels by various antidepressants were concentration-dependent with distinctive potency and effectiveness at micromolar concentrations. The rank order of the inhibition of GIRK channels by 100 mM of these drugs was the following: duloxetine$sertraline.amoxapine.nefazodone.mianserin$venlafaxine<mirtazapine for GIRK1/2 channels and sertraline.duloxetine&nefazodone, amoxapine. venlafaxine, mianserin.mirtazapine for GIRK1/4 channels. Table 1 shows the EC 50 and n H values obtained from the concentration-response relationships for sertraline, duloxetine and amoxapine, and the percentage inhibition of the GIRK currents by the drugs at the highest concentrations tested. Additionally, because the drugs could not completely block these types of GIRK channels even at the highest concentrations tested, the IC 25 and IC 50 values were also calculated to further compare the effects of the drugs ( Table 1). The inhibition of GIRK1/2 channels by sertraline was similar to that by duloxetine (Fig. 2). Furthermore, the inhibition of GIRK1/2 channels by sertraline was statistically similar to the inhibition of GIRK1/4 channels (P.0.05 at each concentration, Tukey-Kramer post hoc test; Fig. 2, Table 1). In contrast, the inhibition of GIRK1/2 channels by duloxetine and amoxapine was more effective than the inhibition of GIRK1/4 channels (P,0.05 at 30, 100, and 300 mM for duloxetine and P,0.05 at 300, 500, and 1000 mM for amoxapine, Tukey-Kramer post hoc test; Fig. 2, Table 1).

Characteristics of inhibition of GIRK channels by the SSRI sertraline and SNRI duloxetine
Sertraline and duloxetine, which belong to commonly used classes of antidepressants, effectively inhibited GIRK channels, and we further investigated the effects of these drugs in more detail. Instantaneous GIRK1/2 currents elicited by the voltage step to 2100 mV from a holding potential of 0 mV were diminished in the presence of 30 mM sertraline applied for 3 min (Fig. 3A). The percentage inhibition of the steady-state GIRK current at the end of the voltage step by sertraline was not significantly different from that of the instantaneous current (P.0.05, paired t-test; n = 9 at 240, 260, 280, 2100, and 2120 mV, respectively). For duloxetine, the instantaneous currents were primarily diminished in the presence of 30 mM duloxetine, and the currents gradually increased in the voltage step (Fig. 3A). The percentage inhibition of the steady-state GIRK current at the end of the voltage step by duloxetine significantly decreased compared with that of the instantaneous current (P,0.05 at 280, 2100 and 2120 mV, paired t-test, n = 6). Figure 3B shows that 30 mM sertraline-and duloxetine-sensitive currents in oocytes that expressed GIRK1/2 channels increased with negative membrane potentials, and the current-voltage relationships showed strong inward rectification (n = 9 and 6, respectively), similar to 3 mM Ba 2+ -sensitive currents that corresponded to basal GIRK currents, indicating a characteristic of GIRK currents. The percentage inhibition of GIRK1/2 currents by 30 mM sertraline at the end of the voltage pulses showed no significant difference across voltages between 2120 and 240 mV (no significant sertraline effect6membrane potential effect interaction, P.0.1, one-way ANOVA; P.0.1 across voltages, Tukey-Kramer post hoc test; Fig. 3C), suggesting voltage-independent inhibition of GIRK channels by sertraline. In contrast, the GIRK current inhibition by duloxetine at the end of the voltage pulses was voltage-dependent, with weaker inhibition at more negative membrane potentials (significant duloxetine effect6membrane potential effect interaction, P,0.05, one-way ANOVA; significant differences between 2120 and 260 mV, between 2120 and 240 mV, between 2100 and 260 mV, and between 2100 and 240 mV, P,0.05, Tukey-Kramer post hoc test, n = 6, Fig. 3C). The voltagedependency was associated with a time-dependent decrease in the inhibition by duloxetine in the voltage pulses at more negative membrane potentials. Furthermore, similar results were obtained in oocytes that expressed GIRK1/4 channels (n = 4 for each of the drugs; data not shown). Altogether, sertraline and duloxetine primarily inhibited GIRK channels at the holding potential of 0 mV before the voltage pulses. The inhibitory effects of sertraline were voltage-independent and time-independent during each voltage pulse, whereas those of duloxetine decreased voltagedependently with negative membrane potentials and timedependently up to a steady state current level in each voltage pulse.
Furthermore, the effects of the two antidepressants on GIRK channels under a physiological K + condition were examined. In oocytes injected with GIRK1 and GIRK2 mRNAs, outward currents observed at a holding potential of 210 mV in a K4 solution that contained 4 mM K + were reversibly reduced by 30 mM sertraline (n = 4), 30 mM duloxetine (n = 4), and 3 mM Ba 2+ (the Ba 2+ -sensitive current components, 49.062.8 nA, n = 8; Fig.  S1), whereas in uninjected oocytes, the drugs at 100 mM and 3 mM Ba 2+ caused no significant response (3.060.9 nA for sertraline, 060 nA for duloxetine, and 7.661.3 nA for Ba 2+ ; n = 4, 4, and 8, respectively). The results suggest that the antidepressants also inhibited outward GIRK currents at a physiologically extracellular K + concentration.
Sertraline and duloxetine possess a secondary amine group with pK a values of 8.9 and 9.34, respectively (Data Sheets of Pfizer and  Eli Lilly and Company). At physiological pH or below, sertraline and duloxetine exist mainly in a protonated form, approximately 96.9% and 98.9% at pH 7.4, respectively, and the proportion of the uncharged form increases with an increase in pH. We examined whether changes in extracellular pH would affect GIRK channel inhibition by sertraline or duloxetine. However, in oocytes that expressed GIRK1/2 channels, the percentage inhibition of GIRK channels by sertraline or duloxetine at the same concentrations was not significantly affected by extracellular pH 7.4 and 9.0 (no significant pH6drug interaction, P.0.05, two-way ANOVA; P.0.05 at each concentration, Tukey-Kramer post hoc test; Fig. 4). The results indicate that a marked increase in the proportion of the uncharged form of sertraline and duloxetine may not significantly affect all of the inhibitory effects on GIRK channels, suggesting that GIRK channel inhibition may be mediated by both forms of the drugs with similar effectiveness. Additionally, the inhibition by the antidepressants was unlikely mediated by nonspecific membrane perturbation induced by the uncharged form.
Effects of sertraline on GIRK channels activated by GTPcS, a nonhydrolyzable GTP GIRK channels are activated by various G i/o -protein-coupled receptors through the direct action of G-protein bc subunits released from the heterotrimeric G-protein complex [32,33]. The effects of sertraline on GIRK channels activated by G-proteincoupled signaling mechanisms were further examined using GTPcS, a nonhydrolyzable GTP analog that maintains Gproteins in an activated state. Injection of GTPcS into Xenopus oocytes injected with GIRK1 and GIRK2 mRNAs increased inward currents with time and reached a steady-state level (516.06123.7 nA, n = 5) as reported previously [49,51]. The increased inward currents were completely blocked by 3 mM Ba 2+ , whereas GTPcS injection into uninjected oocytes had no significant effect on current responses to 3 mM Ba 2+ (3.962.1 nA, n = 5). Increased GIRK currents composed of basal GIRK currents and GTPcS-induced GIRK currents were inhibited by sertraline (IC 25 = 5.560.7 mM; IC 50 = 18.163.0 mM; n H = 1.2460.09; n = 5; Fig. 5). The concentration response curve for the inhibition of total GIRK currents by sertraline was partially different from that for the inhibition of basal GIRK currents in GTPcS-untreated oocytes injected with GIRK1 and GIRK2 mRNAs (P,0.05 at 30 mM, Tukey-Kramer post hoc test, Fig. 5). The results suggest that the potency of the inhibition of GIRK channels activated by GTPcSinduced G-protein activation may be slightly higher than that of basally active GIRK channels, although the maximal efficacy was similar.

Sertraline inhibits ethanol-induced GIRK currents.
GIRK channels are also activated by ethanol independent of Gprotein signaling pathways [34]. Sertraline was shown to reduce ethanol consumption in mice [52] and was effective in alcoholics [53]. Therefore, we also examined the effects of sertraline on GIRK channel activation induced by ethanol. The effects of sertraline were evaluated by measuring the amplitude of the ethanol-induced current response during the extracellular application of sertraline at different concentrations. In oocytes injected with GIRK1 and GIRK2 mRNAs, the GIRK currents induced by 100 mM ethanol (344.2640.3 nA, n = 6) were reversibly attenuated in the presence of sertraline (IC 25 = 6.261.4 mM; IC 50 = 29.665.5 mM; n H = 0.8760.17; n = 6; Fig. 6A, 6B). However, the 100 mM ethanol-induced GIRK currents were not significantly affected by intracellularly applied sertraline (104.969.1% of untreated control current, paired t-test, P.0.1, n = 6; Fig. 6C). Moreover, in oocytes that expressed GIRK channels, the basal currents were not substantially affected by intracellularly applied sertraline (92.561.6% of untreated control current, n = 6). The results indicate that intracellular sertraline could not inhibit GIRK channels. In contrast, GIRK channel inhibition induced by extracellularly applied sertraline, which is mainly protonated at pH 7.4, was reversible with washout ( Figs. 1A and 6A). Because the protonated form may not readily permeate the cell membrane, extracellularly applied sertraline may exist mainly on the extracellular side. Altogether, extracellular sertraline may inhibit GIRK channels activated by ethanol. Additionally, the extent of inhibition by sertraline of GIRK1/2 channels activated by ethanol was higher at 100 and 300 mM than that of basally active GIRK1/2 channels by G-proteins (P,0.05, Tukey-Kramer post hoc test), indicating a significant difference in the maximal efficacy of sertraline between ethanol activation of GIRK channels and G-protein activation of the channels.

Discussion
The present study demonstrated that the SSRI sertraline, SNRI duloxetine, and second-generation TCA amoxapine effectively inhibited brain-type GIRK1/2 channels and cardiac-type GIRK1/4 channels expressed in Xenopus oocytes. However, the 5-HT 2 receptor antagonist nefazodone, SNRI venlafaxine, and NaSSAs mianserin and mirtazapine weakly inhibited both types of GIRK channels even at high concentrations. The inhibitions by different classes of antidepressants were concentration-dependent with various degrees of potency and effectiveness. In contrast, Kir2.1 channels in other Kir channel subfamilies were insensitive to all of the drugs. Furthermore, the present results suggest that sertraline and duloxetine primarily inhibited GIRK channels at the holding potential of 0 mV before the voltage pulses. The effects of sertraline on GIRK channels were voltage-independent and time-independent during each voltage pulse, similar to the effects of various TCAs [42]. The effects of duloxetine decreased voltage-dependently with negative membrane potentials and timedependently up to a steady current level in each voltage pulse, and the voltage-dependency was associated with a time-dependent decrease in the inhibition by duloxetine at more negative membrane potentials. The present results also suggest that the site of action on the channels may be extracellular. In contrast, blockade of GIRK channels by extracellular Ba 2+ and Cs + , which occlude the pore of the open channel, increases concentrationdependently, voltage-dependently with negative membrane potentials, and time-dependently with a comparatively small effect on the instantaneous current but marked inhibition on the steady-state current at the end of the voltage pulses [27]. These observations suggest that sertraline and duloxetine may cause an allosteric conformational change in GIRK channels, rather than simple occlusion of the open channel. Additionally, sertraline may stably bind to the channels during the voltage pulses, whereas duloxetine may partially dissociate from the channels in the voltage pulses. The n H values obtained from the concentration-response relationships for sertraline and duloxetine were almost 1 ( Table 1), suggesting an one-to-one interaction between the drug and the binding site. Interestingly, GIRK channels were significantly inhibited by the SSRI sertraline and SNRI duloxetine, despite a great difference in the pharmacological profiles for monoamine transporters. The chemical structure of sertraline is distinct from that of duloxetine [2,54]. These antidepressants may act at different binding sites on the channels, and agents with similar structures may interact with GIRK channels. However, the SNRIs venlafaxine and milnacipran [43] had weak or little effects on GIRK channels, respectively. The distinctive effects of the SNRIs on GIRK channels may be attribute to their diverse chemical structures [54]. The Xenopus oocyte expression system is useful to determine drug actions on membrane proteins, such as voltage-gated Na + and Ca 2+ channels, glutamate receptor channels, 5HT 1C receptor [55]. Since neuronal and cardiac GIRK channels are considered to consist predominantly of GIRK1/2 channels and GIRK1/4 channels, respectively [26,29,36], the effects of antidepressants on GIRK1/2 and GIRK1/4 channels expressed in Xenopus oocytes were investigated in the present study. However, GIRK subunits have been suggested to form functional GIRK channels composed of several types of tetrameric stoichimetries in various cell populations, particularly neurons [56]. GIRK1 subunits are posttranslationally modified by glycosylation [26,56,57]. Furthermore, GIRK channels are regulated by not only G proteins but also phosphatidylinositol 4,5bisphosphate in the cell membrane, polyamines and protein kinases [33]. The effects of antidepressants on GIRK channels might be influenced by differences in composition of the channel subunits, levels of glycosylation of GIRK1 subunits, and interaction with membrane and intracellular factors between the Xennopus oocyte expression system and neurons. Further studies using neurons and cardiac myocytes may be useful for advancing our understanding of the effects of antidepressants on GIRK channels.
The therapeutic serum concentrations range from approximately 0.16 to 0.82 mM for sertraline, 0.07 to 0.27 mM for duloxetine, 0.57 to 1.9 mM for amoxapine, 0.02 to 0.64 mM for nefazodone, 0.06 to 0.26 mM for mianserin, 0.08 to 0.37 mM for mirtazapine, and 0.72 to 1.44 mM for venlafaxine [2,[58][59][60]. Additionally, increases in antidepressant doses are associated with increases in blood concentrations [59]. The concentrations in cases of overdose were reported to reach up to 13.7 mM for sertraline [61], 8.4 mM for duloxetine [62], 57.4 mM for amoxapine [63], 11.7 mM for nefazodone, 18.9 mM for mianserin [59], 8.7 mM for mirtazapine [64], and 302.8 mM for venlafaxine [65]. Most of the doses of antidepressants are distributed in various tissues from the blood, and antidepressants generally accumulate in the brain [2,58,66]. Indeed, brain levels of antidepressants were 40-fold higher for sertraline [67], 15-fold higher for duloxetine [68], 8.7-to 35.5-fold higher for amoxapine [69], 1.3-to 1.8-fold higher for nefazodone [70], 12.1-fold higher for mianserin [71], 3.2-fold higher for mirtazapine, and 4.9-fold higher for venlafaxine [66] compared with blood levels. Altogether, due to the high brain-to-blood partition ratios, presumed brain concentrations during treatment with therapeutic doses would range from approximately 6.4 to 32.8 mM for sertraline and 5.0 to 67.5 mM for amoxapine, and those after overdose would reach up to 548 mM for sertraline, 126 mM for duloxetine, 499 or 2038 mM for amoxapine, 229 mM for mianserin, and 1484 mM for venlafaxine. In addition, it has been shown that the therapeutic concentrations of some SSRIs in the brain were much higher than binding affinities of the antidepressants to monoamine transportors [72][73][74][75]. Brain concentrations at therapeutic doses of sertraline and amoxapine and after overdose of sertraline, duloxetine, amoxapine, mianserin and venlafaxine overlap with their effective concentrations in inhibiting predominant brain-type GIRK1/2 channels (Fig. 2). Therefore, the present results suggest that some inhibition of GIRK channels in the brain might occur with the antidepressant medication, particularly sertraline. However, mirtazapine and nefazodone may have small or little effects on GIRK channels even at toxic levels. Inhibition of GIRK channels causes a depolarization of membrane potential, resulting in an increase in cell excitability [38]. GIRK channels play an important role in regulating neuronal excitability and synaptic transmission [36,41]. Therefore, even partial inhibition of GIRK channels by the antidepressants may affect various brain functions.
Interestingly, GIRK2 knockout mice exhibit reduced anxietyrelated behavior [76]. Animal studies have shown that sertraline has anxiolytic properties [77,78]. Indeed sertraline is clinically effective in the treatment of panic disorder and posttraumatic stress disorder [79]. Although the therapeutic effects are generally thought to be primarily attributable to inhibition of serotonin reuptake in the brain [2], some inhibition of GIRK channels might also contribute to improvement of anxiety symptoms.
Although the risk of seizures with antidepressants is generally very low, the association with overdose is well established [80]. However, the molecular mechanisms by which antidepressants cause seizures have not been clarified. GIRK2 knockout mice exhibit spontaneous seizures and are more susceptible to seizures induced by pentylenetetrazol than wild-type mice [37]. The risk of seizures in overdoses with sertraline, duloxetine, mianserin, and venlafaxine significantly increases [80][81][82], and amoxapine overdose is more likely to cause seizures [83]. Brain levels of the drugs in overdose cases may be considerably higher than levels during treatment at therapeutic doses, suggesting significant inhibition of neuronal GIRK channels by the drugs. Additionally, other types of K + channels are inhibited by antidepressants at micromolar concentrations, that is, the two-pore-domain K + channel, TREK-1 for sertraline and voltage-gated K + channels for amoxapine and mianserin [16,17,21]. Therefore, the inhibition of GIRK channels by the drugs after overdose together with the different types of K + channels may contribute to increased seizure activity and the occurrence of other neurological side effects by increasing neuronal excitability.
In the heart, GIRK channels cause a slowing of heart rate in response to activation of M 2 muscarinic receptors through acetylcholine release from the stimulated vagus nerve [25,26]. GIRK1 and GIRK4 knockout mice exhibit slightly elevated resting heart rates [39]. The present results indicate that sertraline, duloxetine, amoxapine, and venlafaxine can partially inhibit cardiac-type GIRK1/4 channels at blood levels after overdose, although the corresponding heart concentrations were not determined. These antidepressants are associated with sinus tachycardia in cases of toxicity after overdose [81,82,84,85]. In addition, the drugs exhibit low micromolar binding affinities for the muscarinic receptor, with the exception of venlafaxine [2,86], and nanomolar to low micromolar binding affinities for norepinephrine transporters [2,68]. Altogether, sinus tachycardia associated with drug overdose may be related to partial inhibition of atrial GIRK channels as well as antagonism of the muscarinic receptor and enhancement of sympathetic nerve activity.
Sertraline was shown to be effective in the treatment of alcoholics [53]. Interestingly, GIRK2 knockout mice show reduced ethanol-induced conditioned taste aversion and conditioned place preference and are less sensitive than wild-types to some of the acute effects of ethanol, including anxiolysis, habituated locomotor stimulation, and acute handling-induced convulsions [76,87]. In the present study, sertraline inhibited ethanol-induced GIRK1/2 currents. Sertraline may suppress some of the GIRK-related effects of ethanol. Furthermore, GIRK knockout mice show an attenuation of the morphine withdrawal syndrome [88]. Sertraline reduced the severity of the naloxoneprecipitated opioid withdrawal syndrome in rats [89]. GIRK knockout mice also show reduced cocaine self-administration [90]. Inhibition of GIRK channels by sertraline may play a role in the treatment of addiction to these drugs. Figure S1 Effect of sertraline on outward GIRK currents. In a Xenopus oocyte injected with GIRK1 and GIRK2 mRNAs, current responses to 30 mM sertraline and 3 mM Ba 2+ at a membrane potential of 210 mV in a K4 solution that contained 4 mM K + are shown. Asterisk indicates the zero current level. (DOC)