Natural Product Isoliquiritigenin Activates GABAB Receptors to Decrease Voltage-Gate Ca2+ Channels and Glutamate Release in Rat Cerebrocortical Nerve Terminals

Reduction in glutamate release is a key mechanism for neuroprotection and we investigated the effect of isoliquiritigenin (ISL), an active ingredient of Glycyrrhiza with neuroprotective activities, on glutamate release in rat cerebrocortical nerve terminals (synaptosomes). ISL produced a concentration-dependent inhibition of glutamate release and reduced the intraterminal [Ca2+] increase. The inhibition of glutamate release by ISL was prevented after removing extracellular Ca2+ or blocking P/Q-type Ca2+ channels. This inhibition was mediated through the γ-aminobutyric acid type B (GABAB) receptors because ISL was unable to inhibit glutamate release in the presence of baclofen (an GABAB agonist) or CGP3548 (an GABAB antagonist) and docking data revealed that ISL interacted with GABAB receptors. Furthermore, the ISL inhibition of glutamate release was abolished through the inhibition of Gi/o-mediated responses or Gβγ subunits, but not by 8-bromoadenosine 3′,5′-cyclic monophosphate or adenylate cyclase inhibition. The ISL inhibition of glutamate release was also abolished through the inhibition of protein kinase C (PKC), and ISL decreased the phosphorylation of PKC. Thus, we inferred that ISL, through GABAB receptor activation and Gβγ-coupled inhibition of P/Q-type Ca2+ channels, suppressed the PKC phosphorylation to cause a decrease in evoked glutamate release at rat cerebrocortical nerve terminals.


Introduction
Isoliquiritigenin (ISL, Figure 1A), a flavonoid with a chalcone structure, is the main active ingredient in the Glycyrrhiza glabra L. root. ISL has received attention because of its various pharmacological benefits, including antibacterial, antiviral, antidiabetic, antioxidant, anti-inflammatory, anticarcinogenic, analgesic, and neuroprotective effects [1][2][3]. Regarding its neuroprotective activity, previous animal studies found that ISL can penetrate the blood-brain barrier [4] and protects against brain damage and cognitive impairment induced by kainic acid, lipopolysaccharide (LPS), ischemia, or traumatic brain injury [5][6][7]. In vitro, ISL attenuates glutamate-, H 2 O 2 -or amyloid beta-protein (Aβ) (25-35)-induced Glutamate is the predominant excitatory neurotransmitter in the central nervous system (CNS) that produces excitation through glutamate receptors, supporting normal synaptic transmission and sustaining learning and memory processes [13,14]. However, in various pathological conditions, including stroke and neurodegenerative diseases, excessive glutamate release may mediate neuronal injury or death [15,16]. Thus, reduced glutamatergic transmission through the inhibition of released glutamate has been proposed as a neuroprotective mechanism [17][18][19]. Studies have reported that several natural products with neuroprotective effects may reduce presynaptic glutamate release [20][21][22][23]. Likewise, ISL has a neuroprotective-like effect and whether or not ISL can influence presynaptic glutamate release should be evaluated. Therefore, this study investigated the effect and possible mechanisms of ISL on glutamate release in rat cerebrocortical nerve terminals (synaptosomes).

Animals
Male Sprague-Dawley rats (150-200 g) were maintained at 21-23 • C with free access to water and food, under a 12:12 h light/dark cycle (lights-on at 07:00 h). The procedures used in this study were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The experiments were developed after the protocol approved by the Institutional Ethics Committee (A10814). In this study, all efforts were made to minimize the number of animals used and their suffering.

Preparation of Synaptosomes
Rats were killed by cervical dislocation and decapitation. The cerebral cortex was rapidly removed and homogenized in ice-cold hepes-buffered medium containing 0.32 M sucrose (pH 7.4). The homogenate was centrifuged at 3000× g for 10 min at 4 • C. The supernatant was retained and centrifuged at 14,500× g for 12 min at 4 • C. The pellet was resuspended and layered on top of a discontinuous Percoll gradient and centrifuged at 32,500× g for 7 min at 4 • C. Protein concentration was determined using the Bradford assay. Synaptosomes were centrifuged in the final wash to obtain synaptosomal pellets with 0.5 mg protein [24,25].

Glutamate Release
Synaptosomal pellets were analyzed for glutamate release using enzyme-linked fluorescence method previously described [24,26]. As synaotosome is not amenable to electrical stimulation, a number of biochemical depolarization protocols have been developed, includ- ing the use of K + channel blocker 4-AP or high external [K + ] [24]. Therefore, synaptosomes were suspended in hepes-buffered medium containing 2 mM NADP + , 50 units of glutamate dehydrogenase, and 1.2 mM CaCl 2 , and the synaptosome suspension was stimulated with either 1 mM 4-AP or 15 mM KCl. Increases in fluorescence due to production of NADPH was determined using a PerkinElmer LS55 spectrofluorimeter with excitation and emission wavelengths set at 340 nm and 460 nm, respectively. Released glutamate was calibrated by a standard of exogenous glutamate (5 nmol) and expressed as nanomoles glutamate per milligram synaptosomal protein (nmol/mg protein). Values quoted in the text and depicted in bar graphs represent the levels of glutamate cumulatively released after 5 min of depolarization, and are expressed as nmol/mg protein/5 min.

Immunohistochemistry
Immunofluorescence analysis was performed on synaptosomes as described previously [29,30]. The synaptosomes were attached to the polylysine-coated coverslips for 40 min, fixed with 4% paraformaldehyde for 5 min, and permeabilized with 0.2% Triton X-100 in phosphate-buffered saline (PBS) for 1 h. Subsequently, the synaptosomes were incubated with primary antibody solutions containing anti-GABA B receptor antibody (1:100; Abcam, Cambridge, UK) and anti-vesicular glutamate transporter 1 (VGLUT1) antibody (1:100; Abcam, Cambridge, UK) overnight. Synaptosomes were then washed with PBS and incubated in a mixture of goat anti-mouse DyLight 549-and goat antirabbit fluorescein isothiocyanate (FITC)-conjugated secondary antibodies (1:200; Invitrogen, Carlsbad, CA, USA) for 2 h at room temperature. Immunoreactivity was visualized on a Image Xpress Micro confocal microscope (Molecular Devices, San Jose, CA, USA). The estimation of the percentage of glutamatergic terminals positive for GABA B receptor was counted three randomly selected areas (255 × 255 µm 2 ) from each coverslip and averaged using ImageJ (Bio-Rad, Hercules, CA, USA).

Molecular Docking Study
The tools of CDOCKER in software of Discovery Studio 4.1 client was used for the molecular docking experiment. Firstly, the molecular structure of GABA B protein (PDB ID 4MGF) was downloaded from the RCSB Protein Data Bank, and then prepared by the molecular modeling software. The structure of ISL was established following the standard protocols in software, and then docked into the binding site of the active site in GABA B protein.

Statistical Analysis
The data were checked for normal distribution and analyzed by Student t-tests or one-way analysis of variance (ANOVA), followed by Fisher's LSD multiple comparisons tests as appropriate using GraphPad Prism 7 (La Jolla, CA, USA). Data are expressed as mean values ± stander error of the mean (SEM) of at least five independent experiments. A p-value < 0.05 was considered statistically significant.

ISL Inhibits 4-AP-evoked Glutamate Release from Nerve Terminals in the Cerebral Cortex through Ca 2+ Influx Reduction
To examine the presynaptic action of ISL, isolated synaptosomes were depolarized with 4-APwhich has been shown to open voltage-gated Ca 2+ channels (VGCCs) and to induce the release of glutamate [31]. Under control conditions, 4-AP evoked a glutamate release of 7.2 ± 0.1 nmol/mg/5 min. Incubation with 10 µM ISL for 10 min prior to the addition of 1 mM 4-AP produced an inhibition of 4-AP-evoked glutamate release to 4.3 ± 0.1 nmol/mg/5 min [t(8) = 61.5, p < 0.001]. The inhibition of 4-AP-evoked glutamate release by ISL was concentration-dependent, with a maximum inhibition of 78.6 ± 0.5% produced at 50 µM; the half maximal effective concentration for this inhibition was 17.2 µM ( Figure 1B). Given the robust repression of evoked glutamate release that was seen with 10 µM ISL, this concentration of ISL was used in subsequent experiments to evaluate the mechanisms that underlie the ability of ISL to reduce glutamate release.
To investigatewhether the inhibition of 4-AP-evoked glutamate release by ISL was mediated by an effect on exocytotic vesicular release, we examine the action of ISL in the presence of bafilomycin A1 (0.1 µM), which causes the depletion of glutamate in synaptic vesicles. As shown in Figure 1C, bafilomycin A1 reduced control 4-AP-evoked glutamate release (31.8 ± 2.7% of control, p < 0.001). When bafilomycin A1 was present, 10 µM ISL failed to produce significant inhibition (31.2 ± 2.1% of control, p = 0.98 vs. bafilomycin A1-treated group). Similarly, the glutamate release evoked by 1 mM 4-AP was reduced in the presence of extracellular-Ca 2+ -free solution that contained 300 µM EGTA (26.4 ± 1.5% of control, p < 0.001 vs. control group). This Ca 2+ -independent release evoked by 4-AP was unaffected by 10 µM ISL (26.8 ± 1.5% of control, p = 0.99; Figure 1D). In addition, we examined the effect of ISL on the 15 mM KCl-evoked glutamate release, a process that involves Ca 2+ influx primarily through the opening of VGCCs [32]. As illustrated in Figure 1E, 10 µM ISL significantly inhibited the KCl-evoked glutamate release [60.4 ± 6.7% of control, t(8) = 5.9, p < 0.001 vs. control]. These results indicate that the observed inhibition of glutamate release by ISL is likely to be due to a decrease in Ca 2+ influx through VGCCs which are coupled to glutamate exocytosis in the nerve terminals. To verify this, [Ca 2+ ] C was determined in synaptosomes preloaded with Fura-2. Stimulation of synaptosomes with 1 mM of 4-AP elicited a rise in [Ca 2+ ] C . 10 µM ISL preincubation did not significantly affect basal [Ca 2+ ] C (control, 148.6 ± 1.1; ISL, 147.7 ± 2.1; p = 1 vs. control group; Figure 1F), but caused a significant decrease in the 4-AP-induced [Ca 2+ ] C elevation [64.1 ± 3.7% of control, t(8) = 50.9, p < 0.001 vs. control group].

Specifics of VGCCs Involved in ISL Inhibition of Glutamate Release
In the rat cerebrocortical nerve terminal preparation, the release of glutamate is primarily coupled to the entry of Ca 2+ through both N-and P/Q-type Ca 2+ channels (39,40). The results above described indicated that the ISL-induced inhibition of glutamate release involves a reduction in Ca 2+ influx through VGCCs. To verify this hypothesis, we used selective VGCC inhibitors to characterize the role of individual Ca 2+ channel subtypes in the observed ISL-mediated inhibition of glutamate release. As shown in Figure 2A, with the blockade of N-type VGCCs with ω-conotoxin GVIA (2 µM), glutamate release evoked by 4-AP under control conditions was reduced (70.6 ± 1.2% of control, p < 0.001). When ω-conotoxin GVIA was present, glutamate release was further inhibited by 10 µM ISL (39.6 ± 0.7% of control, p < 0.05 vs. ω-conotoxin GVIA -treated group, Figure 2A). The additive relation between ω-conotoxin GVIA and ISL indicates that N-type VGCCs appear not to mediate the action of ISL on glutamate release. In addition, selective blockade of P/Q-type VGCCs using ω-agatoxin IVA (0.5 µM) reduced control 4-AP-evoked glutamate release (44.7 ± 0.6% of control, p < 0.001). Notably, in the presence of ω-agatoxin IVA, 10 µM ISL inhibition of glutamate release was completely abolished (45.4 ± 0.4% of control, p = 0.92 vs. ω-agatoxin IVA-treated group, Figure 2B), indicating that P/Q-type VGCCs are involved in the observed modulation of glutamate release by ISL. In contrast, selective blockade of L-type VGCCs with nifedipine (1 µM) caused no significant changes on glutamate release evoked by 4-AP under control conditions (98.6 ± 1.5% of control, p = 0.83), indicating that L-type VGCCs are not directly support glutamate release from synaptosomes. When nifedipine was present, 10 µM ISL significantly reduced 4-AP-evoked glutamate release [42.4 ± 1.4% of control, p < 0.05 vs. nifedipine-treated group, Figure 2C), indicating that L-type VGCCs, similar to N-type VGCCs, are not involved. These results demonstrate that a reduction in Ca 2+ influx mediated by P/Q-type VGCCs is associated with the inhibition of glutamate release by ISL.

Protein Kinase C Suppression Is Involved in the ISL-mediated Inhibition of Glutamate Release
It has been reported that increased Ca 2+ influx in nerve terminals enhances PKC activation and glutamate release [33]. We therefore examined whether the inhibition of Ca 2+ influx caused by ISL decreased PKC activity. We tested how GF109203X, a general inhibitor of PKC, changed the ISL effect. Specifically, in the absence of GF109203X, ISL significantly reduced the 4-AP-evoked glutamate release (p < 0.001). Application of 10 µM GF109203X significantly inhibited control 4-AP-evoked glutamate release (61.1 ± 0.9% of control, p < 0.001). In the presence of 10 µM GF109203X, 10 µM ISL failed to reduce 4-AP-evoked glutamate release (60.8 ± 0.7% of control, p = 0.98 vs. GF109203X-treated group, Figure 3A). Similar results were also observed in the presence of 10 µM of Go6976, an inhibitor of conventional Ca 2+ -dependent PKCα isozymes, which reduced the 4-APevoked glutamate release (69.4 ± 1.3% of control, p < 0.001). In addition, 10 µM ISL did not reduce the release of glutamate evoked by 4-AP when Go6976 was present (67.6 ± 0.9% of control, p = 0.49 vs. Go6976-treated group, Figure 3B). To verify that the PKC suppression was involved in ISL inhibition of glutamate release, we examined the effect of ISL on the phosphorylation of PKC and its substrate, myristoylated alanine-rich C kinase substrate (MARCKS), in cerebrocortical synaptosomes. Western blot analysis showed that 1 mM of 4-AP increased the phosphorylation of PKC (892.9 ± 42.5% of control, p < 0.001), PKCα (437.3 ± 81.5% of control, p < 0.01), and MARCKS (136.8 ± 91.6% of control, p < 0.001) in the presence of 1.2 mM CaCl 2 . These phosphorylation changes caused by 4-AP were markedly attenuated by the presence of 10 µM ISL (PKC, 109.4 ± 2.3% of control; PKCα, 99.9 ± 4.8% of control; MARCKS, 91.6 ± 20.8% of control; p < 0.05 vs. 4-AP-treated group; Figure 3C). These results suggest that the PKC signaling pathway was suppressed by ISL during its inhibition of 4-AP-evoked glutamate release.

Presynaptic GABA B Receptors Mediate ISL's Effect on Glutamate Release
ISL-mediated actions in the CNS have previously been associated with the modulation of γ-aminobutyric acid type B (GABA B ) receptors [34]. Since GABA B receptor activation inhibits Ca 2+ influx and glutamate release [35,36], we investigated whether ISL inhibition of glutamate release was dependent on GABA B receptor activity. In these experiments, we first identified the presence of the GABA B receptor protein in the cerebrocortical glutamatergic synaptosomes through immunocytochemical analysis. As detailed in Figure 4A, a significant percentage (71%) of vesicular transporter of glutamate type 1 (VGLUT1)-positive glutamatergic particles (red) was also immunopositive for the GABA B receptor protein (green), indicated by the merged staining (yellow). In addition, the effect of the GABA B receptor selective agonist baclofen on the ISL-mediated inhibition of glutamate release was evaluated. As detailed in Figure 4B, baclofen (50 µM) reduced the 4-AP-evoked release of glutamate from the cerebrocortical synaptosomes (41.2 ± 0.9% of control, p < 0.001). In the presence of baclofen, the addition of 10 µM ISL failed to reduce 4-AP-evoked glutamate release (40.4 ± 1.9% of control, p = 0.99 vs. baclofen-treated group). The lack of additivity in the inhibitory actions of ISL and baclofen on glutamate release cans be explained by the inhibition of the same action mechanism by both compounds. Similar to lack of effect of ISL in the presence of baclofen, ISL had no effect on the 4-AP-evoked glutamate release in the presence of the GABA B receptor antagonist CGP35348 (100 µM) (96.4 ± 1.7% of control, p = 0.98 vs. CGP35348-treated group). CGP35348 alone elicited no significant effect on the 4-AP-evoked glutamate release (96.8 ± 1.2% of control, p = 0.39; Figure 4C). Therefore, these  Figure 3C). These results suggest that the PKC signaling pathway was suppressed by ISL during its inhibition of 4-AP-evoked glutamate release.  be explained by the inhibition of the same action mechanism by both compounds. Similar to lack of effect of ISL in the presence of baclofen, ISL had no effect on the 4-AP-evoked glutamate release in the presence of the GABAB receptor antagonist CGP35348 (100 μM) (96.4 ± 1.7% of control, p = 0.98 vs. CGP35348-treated group). CGP35348 alone elicited no significant effect on the 4-AP-evoked glutamate release (96.8 ± 1.2% of control, p = 0.39; Figure 4C). Therefore, these results suggest that GABAB receptor activation is involved in the ISL-mediated inhibition of 4-AP-evoked release.

ISL Interacts with the GABA B Receptors
In the last decade, machine learning applications is appropriate to use in pharmacological research. Among them, molecular docking is appropriate to use pharmacology to predict the target of natural products [37].To estimate the interaction between ISL and GABA B receptors, a molecular docking experiment was performed using the CDOCKER tools in the Discovery Studio 4.1 client. The molecular structure of a GABA B receptor (PDB ID 4MGF) was downloaded from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (PDB ID 4MGF). The results revealed that ISL had hydrogen-bonding interactions with the amino acid residue Glu349. Furthermore, the residues His170 and Cys129 in LB1 domain make a van der Waal and a lipophilic interaction with isoliquiritigenin, respectively ( Figure 5). Both His170 and Glu 349 are major residues of LB1 domain in GABA B receptor contact to agonists [38][39][40]. tools in the Discovery Studio 4.1 client. The molecular structure of a GABAB receptor (PDB ID 4MGF) was downloaded from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (PDB ID 4MGF). The results revealed that ISL had hydrogen-bonding interactions with the amino acid residue Glu349. Furthermore, the residues His170 and Cys129 in LB1 domain make a van der Waal and a lipophilic interaction with isoliquiritigenin, respectively ( Figure 5). Both His170 and Glu 349 are major residues of LB1 domain in GABAB receptor contact to agonists [38][39][40].

Gβγ-coupling Mechanism Is Involved in the ISL-mediated Inhibition of Glutamate Release
GABAB receptors are Gi/o-protein-coupled receptors that activate different downstream effectors, such as the well-characterized Gi/o-cyclicmonophosphate (cAMP)-protein kinase A pathway [41]. To determine whether the Gi/o proteins were involved in the ISL-mediated inhibition of glutamate release, synaptosomes were incubated for 4 h with the Gi/o protein inhibitor pertussis toxin (PTX; 2 µ g/mL) [42]. As presented in Figure 6A, PTX did not change the glutamate release evoked by 4-AP under control conditions (101.2 ± 0.8% of control, p = 0.98). In the presence of PTX, the addition of 10 µ M ISL failed to reduce 4-AP-evoked glutamate release (99.8 ± 0.4% of control, p = 0.87 vs. PTX-treated group). To further elucidate the transduction pathways lying downstream of the ISL-mediated activation of Gi/o, we tested the influence of exogenous cAMP on the effect of ISL. In the presence of the membrane-permeable cAMP analogue 8-bromo-adenosine-3′,5′-cyclic monophosphate (8-bromo-cAMP; 250 µ M), control glutamate release evoked by 4-AP was increased (143.2 ± 2,5% of control, p < 0.001), but 10 µ M ISL significantly inhibited glutamate release (87.6 ± 0.6% of control, p < 0.05 vs. 8-bromo-cAMP-treated group, Figure  6B). In addition, the adenylate cyclase inhibitor MDL12330A (10 µ M) reduced the control

G βγ -coupling Mechanism Is Involved in the ISL-mediated Inhibition of Glutamate Release
GABA B receptors are G i/o -protein-coupled receptors that activate different downstream effectors, such as the well-characterized G i/o -cyclicmonophosphate (cAMP)-protein kinase A pathway [41]. To determine whether the G i/o proteins were involved in the ISLmediated inhibition of glutamate release, synaptosomes were incubated for 4 h with the Gi/o protein inhibitor pertussis toxin (PTX; 2 µg/mL) [42]. As presented in Figure 6A, PTX did not change the glutamate release evoked by 4-AP under control conditions (101.2 ± 0.8% of control, p = 0.98). In the presence of PTX, the addition of 10 µM ISL failed to reduce 4-AP-evoked glutamate release (99.8 ± 0.4% of control, p = 0.87 vs. PTXtreated group). To further elucidate the transduction pathways lying downstream of the ISL-mediated activation of G i/o , we tested the influence of exogenous cAMP on the effect of ISL. In the presence of the membrane-permeable cAMP analogue 8-bromo-adenosine-3 ,5 -cyclic monophosphate (8-bromo-cAMP; 250 µM), control glutamate release evoked by 4-AP was increased (143.2 ± 2,5% of control, p < 0.001), but 10 µM ISL significantly inhibited glutamate release (87.6 ± 0.6% of control, p < 0.05 vs. 8-bromo-cAMP-treated group, Figure 6B). In addition, the adenylate cyclase inhibitor MDL12330A (10 µM) reduced the control glutamate release evoked by 4-AP (84.6 ± 0.9% of control, p < 0.001). The inhibition caused by 10 µM ISL on glutamate release was unchanged by the presence of MDL12330A (60.8 ± 2.4% of control, p < 0.05 vs. MDL12330A-treated group, Figure 6C). Notably, the G protein βγ (G βγ ) subunit inhibitor gallein (10 µM) had no significant effect on the control 4-Ap-evoked glutamate release (99.4 ± 1.9% of control, p = 0.98); in the presence of gallein, 10 µM ISL exhibited no further effect on 4-AP-evoked glutamate release (98.6 ± 2.1% of control, p = 0.94 vs. gallein-treated group, Figure 6D). These results suggest that ISL-mediated action on glutamate release is dependent on G βγ subunits, but not on changes in cAMP.

Discussion
ISL is an active ingredient of Glycyrrhiza, with pharmacological functions such as antioxidation, anti-inflammation, antinociception, neuroprotection, and antitumor effects [2]. To date, no data documenting the effect of ISL on glutamate release was available. Several reports have indicated that natural products may effectively reduce the release of glutamate [20][21][22][23]. We examined whether ISL exhibited an effect on glutamate release in rat cerebrocortical nerve terminals. The main study findings were as follows: (i) ISL exerts an inhibitory effect on VGCC-dependent glutamate release from cerebrocortical nerve terminals and (ii) this inhibitory action is dependent on the presynaptic GABAB receptor activation and PKC suppression.

Discussion
ISL is an active ingredient of Glycyrrhiza, with pharmacological functions such as antioxidation, anti-inflammation, antinociception, neuroprotection, and antitumor effects [2]. To date, no data documenting the effect of ISL on glutamate release was available. Several reports have indicated that natural products may effectively reduce the release of glutamate [20][21][22][23]. We examined whether ISL exhibited an effect on glutamate release in rat cerebrocortical nerve terminals. The main study findings were as follows: (i) ISL exerts an inhibitory effect on VGCC-dependent glutamate release from cerebrocortical nerve terminals and (ii) this inhibitory action is dependent on the presynaptic GABA B receptor activation and PKC suppression.
The results suggested that the ILS inhibition of glutamate release from cerebrocortical nerve terminals was associated with a decrease in Ca 2+ influx through the VGCCs, because ISL was unable to inhibit 4-AP-evoked glutamate release in the absence of extracellular Ca 2+ or blocking vesicular glutamate transporters and ISL reduced intraterminal [Ca 2+ ]. In addition, ISL also inhibited the release of glutamate induced by 15 mM of external KCl, which is a VGCC-dependent process [31,32]. Moreover, this research demonstrated that after the blockage of P/Q-type Ca 2+ channels, the remaining 4-AP-evoked glutamate release was unresponsive to ISL. Thus, the suppression of Ca 2+ influx through P/Q-type Ca 2+ channels is involved in the inhibition of glutamate release caused by ISL. P/Q-type Ca 2+ channels have been reported to participate in triggering glutamate release from nerve terminals [43][44][45]. In the nerve terminals, VGCCs can be mediated by the membrane potential; for example, the inhibition of Na + channels and Na + influx or activation of K + channels and K + efflux leads to cell membrane hyperpolarization. This changing membrane potential closes the VGCCs, which in turn reduces [Ca 2+ ] C and neurotransmitter release [31,46]. Although ISL has been shown to modulate Na + and K + channels [3,47,48], the inhibitory effect of ISL on the VGCCs observed in our study is not caused by a change in synaptosomal membrane potential because ISL inhibited the release of glutamate evoked by 4-AP and KCl. Both of these depolarizing treatments are thought to activate VDCCs coupled to glutamate release similarly, and this should be indicated by qualitatively similar modulation if it occurs at the level of the voltage-dependent Ca 2+ channel. The two depolarizing paradigms differ in that 4-AP-evoked glutamate release involves the action of Na + andCa 2+ channels, whereas 15 mM external KCl-evoked glutamate release involves only Ca 2+ channels [30,31]. This indicates that Na + channels are not involved in the effect of ISL on glutamate release. Furthermore, ISL did not affect the 4-AP-evokedCa 2+ -independent glutamate release, a component of glutamate release that is exclusively dependent on membrane potential [49,50]. Therefore, our findings indicate that the inhibition of releasecoupled VGCCs by ISL reflects a direct effect on VGCC function.
Some ISL-mediated action in the nervous system has been associated with the modulation of GABA B receptors [34]. The GABA B receptor, which is a G-protein-coupled receptor, regulates a variety of intracellular signaling systems, acting on both the pre-and postsynaptic membrane [51]. Presynaptic GABA B receptors can be coupled with several intracellular effector systems to inhibit glutamate release, including the inhibition of adenylate cyclase activity, membrane-delimited G i/o -coupled inhibition of VGCCs, and activation of G i/ocoupled K + channels [52][53][54]. In this study, we determined that the GABA B receptor protein was co-expressed with the glutamatergic terminal marker protein VGLUT1 within the same nerve terminals, indicating the presence of GABA B receptors on cerebrocortical glutamatergic terminals. The activation of GABA B receptors at cerebrocortical nerve terminals was involved in the ISL-mediated inhibition of glutamate release, because ISL was unable to inhibit glutamate release in the presence of the GABA B receptor agonist baclofen and antagonist CGP35348. Our docking data also revealed that ISL can interact with the amino acid residues (Glu349 and His170) of the GABA B receptor; both His170 and Glu 349 are major residues of the LB1 domain in GABA B receptor agonist contact [38][39][40]. In addition, the ISL-mediated action through the GABA B receptor may be mediated by the sequential activation of G i/o proteins because the ISL inhibition of glutamate release was abolished through the inhibition of G i/o -mediated responses with PTX. Moreover, ISL-mediated action on glutamate release is dependent on G βγ subunits, but not on changes in cAMP, as the inhibition of glutamate release caused by ISL was antagonized through galleon-induced G βγ subunit inhibition; no such changes were observed with cAMP analogue 8-bromo-cAMP or with the inhibition of adenylate cyclase with MDL12330A. These data, together with the blockade of the ISL-mediated inhibition of glutamate release in the presence of the selective inhibitor of P/Q-type Ca 2+ channels, indicated that the ISL-mediated inhibition of glutamate release was mediated by GABA B receptor activation through the G βγ -coupled inhibition of VGCCs. G βγ has been demonstrated to regulate neurotransmitter release through a direct inhibition of VGCCs [55,56], supporting our results.
The ISL inhibition of glutamate release observed in this study was associated with a suppression of the PKC pathway, because ISL failed to inhibit glutamate release in the presence of the selective PKC inhibitors GF109203X and Go6976; ISL inhibited the phosphorylation of PKC and its substrate MARCKS. PKC is a modulator of the exocytotic pathway, where it enhances both the priming and fusion steps of vesicle exocytosis through the phosphorylation of several proteins (including MARCKS) of the exocytotic machinery [57][58][59]. Increased [Ca 2+ ] in nerve terminals can activate PKC and subsequently phosphorylate MARCKS, which increases synaptic vesicle availability and glutamate release [33]. Therefore, the inhibitory effect of ISL on Ca 2+ entry observed in this work may cause a decrease in PKC-induced MARCKS phosphorylation, in turn resulting in decreased glutamate release. Overall, we inferred that ISL, through GABA B receptor activation and the G βγ -coupled inhibition of P/Q-type Ca 2+ channels, suppresses PKC and MARCKS phosphorylation, causing a decrease in evoked glutamate release from rat cerebrocortical nerve terminals (Figure 7). presence of the selective PKC inhibitors GF109203X and Go6976; ISL inhibited the phosphorylation of PKC and its substrate MARCKS. PKC is a modulator of the exocytotic pathway, where it enhances both the priming and fusion steps of vesicle exocytosis through the phosphorylation of several proteins (including MARCKS) of the exocytotic machinery [57][58][59]. Increased [Ca 2+ ] in nerve terminals can activate PKC and subsequently phosphorylate MARCKS, which increases synaptic vesicle availability and glutamate release [33]. Therefore, the inhibitory effect of ISL on Ca 2+ entry observed in this work may cause a decrease in PKC-induced MARCKS phosphorylation, in turn resulting in decreased glutamate release. Overall, we inferred that ISL, through GABAB receptor activation and the Gβγ-coupled inhibition of P/Q-type Ca 2+ channels, suppresses PKC and MARCKS phosphorylation, causing a decrease in evoked glutamate release from rat cerebrocortical nerve terminals (Figure 7).

Figure 7.
Schematic representation of the main mechanism involved in ISL-mediated inhibition of glutamate release from cerebrocortical nerve terminals. Giβγ, Gi protein βγ subunits; PKC, protein kinase C; MARCKS, myristoylated alanine-rich C kinase substrate.
The concentration (10 μM) of ISL used to depress glutamate release in the present work is consistent with that used in other studies. For example, ISL, at 10 μM, attenuated glutamate-induced increases in intracellular Ca 2+ and neuronal death in cultured cortical neurons and HT22 hippocampal neuronal cells [9,60]. The inhibition of glutamate release may be a vital mechanism in pathological conditions that cause elevated extracellular glutamate concentrations, such as hypoxia or ischaemia, epileptic seizures, or neurodegeneration [15,61]; several neuroprotective compounds can inhibit glutamate release to ensure extracellular glutamate concentrations remain below neurotoxic levels [17][18][19]. Thus, the inhibition of glutamate release caused by ISL may contribute to explain its anti-excitotoxic The concentration (10 µM) of ISL used to depress glutamate release in the present work is consistent with that used in other studies. For example, ISL, at 10 µM, attenuated glutamate-induced increases in intracellular Ca 2+ and neuronal death in cultured cortical neurons and HT22 hippocampal neuronal cells [9,60]. The inhibition of glutamate release may be a vital mechanism in pathological conditions that cause elevated extracellular glutamate concentrations, such as hypoxia or ischaemia, epileptic seizures, or neurodegeneration [15,61]; several neuroprotective compounds can inhibit glutamate release to ensure extracellular glutamate concentrations remain below neurotoxic levels [17][18][19]. Thus, the inhibition of glutamate release caused by ISL may contribute to explain its antiexcitotoxic actions in vitro [9,10,62] and in vivo [4][5][6][7]. In fact, the beneficial effects of ISL on brain function have been reported following oral and intraperitoneal administration in animals, suggesting that it can cross the blood-brain barrier to reach the brain (Jia et al., 2008; Zhu et al., 2019). However, the brain distribution of ISL needs to be investigated in further studies.

Conclusions
In conclusion, the inhibitory effect of ISL on glutamate release from cerebrocortical nerve terminals is linked to a decrease in [Ca 2+ ] C associated with GABA B receptor activation and the G βγ -coupled inhibition of VGCCs as well as the subsequent suppression of PKC and MARCKS phosphorylation. Our finding provided valuable new insights into the mechanism by which ISL operates in the brain and offered a potentially therapeutic treatment for neuronal diseases that involve excessive glutamate release.