Zinc Inhibits the GABAAR/ATPase during Postnatal Rat Development: The Role of Cysteine Residue

Zinc ions (Zn2+) are concentrated in various brain regions and can act as a neuromodulator, targeting a wide spectrum of postsynaptic receptors and enzymes. Zn2+ inhibits the GABAARs, and its potency is profoundly affected by the subunit composition and neuronal developmental stage. Although the extracellular amino acid residues of the receptor’s hetero-oligomeric structure are preferred for Zn2+ binding, there are intracellular sites that, in principle, could coordinate its potency. However, their role in modulating the receptor function during postembryonic development remains unclear. The GABAAR possesses an intracellular ATPase that enables the energy-dependent anion transport via a pore. Here, we propose a mechanistic and molecular basis for the inhibition of intracellular GABAAR/ATPase function by Zn2+ in neonatal and adult rats. The enzymes within the scope of GABAAR performance as Cl−ATPase and then as Cl−, HCO3−ATPase form during the first week of postnatal rat development. In addition, we have shown that the Cl−ATPase form belongs to the β1 subunit, whereas the β3 subunit preferably possesses the Cl−, HCO3−ATPase activity. We demonstrated that a Zn2+ with variable efficacy inhibits the GABAAR as well as the ATPase activities of immature or mature neurons. Using fluorescence recording in the cortical synaptoneurosomes (SNs), we showed a competitive association between Zn2+ and NEM in parallel changes both in the ATPase activity and the GABAAR-mediated Cl− and HCO3− fluxes. Finally, by site-directed mutagenesis, we identified in the M3 domain of β subunits the cysteine residue (C313) that is essential for the manifestation of Zn2+ potency.


Introduction
Neuronal plasticity is directly associated with alterations in the expression, properties, and function of plasma membrane channels and transporters that dissipate or generate ion gradients, respectively [1][2][3]. The divalent transition metal cation zinc (Zn 2+ ) is concentrated in multiple brain regions, including the cerebral cortex, hippocampus, hypothalamus, and amygdala and seems to act as a neuromodulator, targeting a wide spectrum of postsynaptic receptors [4,5]. Specifically, Zn 2+ modulates the sensitivity of inhibitory and excitatory receptors to their respective neurotransmitters [6][7][8].
As members of the pentameric ligand-gated ion channel (pLGICs) family, γ-aminobutyric acid type A receptors (GABA A Rs) are inhibited allosterically by micromolar concentrations of Zn 2+ [9]. Recent structural studies have provided some insight into the mechanism of inhibition underlying GABA-induced responses to Zn 2+ [10]. Once in the channel, Zn 2+ may first bind to a histidine residue and then nullify the γ-aminobutyric acid (GABA) response by physically plugging the pore, thereby reducing the single Cl − current. This residue is located at the extracellular end of the ion channel lining the TM2 domain of the β3 subunit and may form part of a zinc-binding site [11]. The Zn 2+ inhibitory potency is profoundly affected by the subunit composition (αβ or αβγ) of GABA A Rs [12] and varies according to the developmental stage [13,14]. This is physiologically relevant because Zn 2+ ions could be Int. J. Mol. Sci. 2023, 24, 2764 2 of 18 critically involved in various pathological processes (e.g., seizures and temporal-lobe epilepsy) where subunit expression may be altered [15][16][17]. In addition to the extracellular sites, there are intracellular amino acid residues in the receptor's hetero-oligomeric structure that could, in principle, coordinate Zn 2+ binding; however, their role in modulating the receptor remains unclear. The GABA A R β3 subunit possesses a zinc-sensitive ATPase that enables the energy-dependent transport of Cl − via a receptor pore [18,19]. In addition, such a GABA A R-coupled ATPase is involved in seizures [20]. Given that Zn 2+ ions are released both extra-and intracellularly [4,5], and ATPase is localized on the intracellular side of the channel [21], there is considerable interest in determining the molecular mechanisms through which Zn 2+ modulates ATPase function. Furthermore, the molecular determinants that mediate the sensitivity of GABA A R to Zn 2+ ions at different developmental stages are not fully understood.
Zinc is a signaling molecule involved in the regulation of various enzymes by inhibiting their catalytic activity [22]. Zn 2+ at nano-or micromolar concentrations inhibits "metallobinding" proteins by interacting with an active center that contains catalytic dyads or triads of glutamate, histidine, and cysteine residues. Early studies have provided evidence of the sulfhydryl and disulfide groups' involvement in GABA A R responses [23], notably two cysteine residues located at the external N-terminus and most of the extra cysteines (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11) of the GABA A R subunits that can significantly contribute to redox modulation [24]. In addition, transmembrane domains can contain cysteines whose positions show a relatively high degree of conservation. Specifically, a single cysteine at the M3 domain is present in virtually all GABA A R subunit subtypes (including β subunits). Although the modulation of such cysteine residue (C313) by redox agents has been shown in homomeric GABA A R β3 subtypes [25], its sensitivity to zinc is unknown.
There is strong evidence showing that synaptic sulfhydryl groups of ionic channels and transporters are targets for electrophiles [26]. N-ethylmaleimide (NEM) is a membranepermeant alkylating agent that modifies the thiol groups of cysteine residues via the formation of a covalent thioether bond. Specifically, NEM can cause an increase in the frequency of GABA A R-mediated postsynaptic currents [27,28] or eliminate the depolarizationor post-burst-induced suppression of GABA A R-mediated inhibition in CA1 pyramidal cells [28,29]. Moreover, NEM modulates the GABA A R-mediated Cl − or HCO 3 − fluxes in cortical neurons via regulating the desensitization/resensitization in a bicarbonatedependent fashion [19]. However, the specific mechanism by which NEM functions to modulate GABA A Rs has yet to be established. Considering that NEM, in contrast to other thiol agents, causes a decrease in the neuronal concentration of ATP [19,30], we aimed to use NEM to test the hypothesis that cysteine residues contribute significantly to the Zn 2+ -mediated modulation of the ATPase activity.
Here, we propose a mechanistic and molecular basis for the inhibition of intracellular GABA A R/ATPase function by zinc in neonatal and adult rats. The enzymes within the scope of GABA A R performance as Cl − ATPase and then as Cl − , HCO 3 − ATPase form during the first week of postnatal rat development. The GABA A R β1 and β3 isoforms, in contrast to the β2 subunit, possess the ATPase activity that facilitates re-establishing the anion gradients into neurons. In addition, we have shown that the Cl − ATPase form belongs to the β1 subunit, whereas the β3 subunit preferably possesses the Cl − , HCO 3 − ATPase activity. We demonstrated that a Zn 2+ with variable efficacy inhibits the GABA A R as well as the ATPase activities of immature or mature neurons. Using fluorescence recording in the cortical synaptoneurosomes (SNs), we showed a competitive association between Zn 2+ and NEM in parallel changes both in the ATPase activity and the GABA A R-mediated Cl − and HCO 3 − fluxes. Finally, by site-directed mutagenesis, we identified in the M3 domain of β subunits the cysteine residue (C313) that is essential for the manifestation of Zn 2+ potency.

Zn 2+ Inhibits the ATPase Activity in Neonatal and Adult Rats
To study the importance of ATPase in the early "ontogenetic switch" of GABAergic transmission [1,31,32], we assessed the effects of Cl − and HCO 3 − in ATPase activation during postnatal development (days P1-P30) in rats. The results revealed the essential role of Cl − (30-60 mM) in the ATPase activity (537.8 ± 26.0 nmol P i ·min −1 ·mg −1 ) at days P1 to P2, and to a lesser extent at days P5 to P7 ( Figure 1A). While HCO 3 − does not affect the ATPase activity during days P1-P9, it caused a significant increase in enzyme activity (569.3 ± 66.8 nmol P i ·min −1 ·mg −1 ) at days P10 to P14 with a maximal effect in the concentration range of 20 to 30 mM ( Figure 1B). Both the Cl − ATPase as well as the Cl − , HCO 3 − ATPase activity, in contrast to HCO 3 − ATPase (not shown), were inhibited by 40 µM bicuculline ( Figure 1C). To determine whether there are two enzyme forms (Cl − ATPase and Cl − , HCO 3 − ATPase), we tested these enzyme activities at days P1 to P35. The Cl − ATPase activity appeared for the duration of time between days P1 to P7 with maximal effectiveness at days P1 to P3. While the Cl − , HCO 3 − ATPase was observed after day P8, the maximal activity (673.4 ± 35.5 nmol P i ·min −1 ·mg −1 ) occurred at day P15 of postnatal development and older ( Figure 1D). These results are consistent with observations showing that [HCO 3 − ] i homeostasis is provided by the plasma membrane transporters (primary, NCBE, and carbohydrase) around postnatal day 12 [2,32,33]. By way of comparison, we tested the effect of Zn 2+ on the Cl − ATPase or Cl − , HCO 3 − ATPase activities of the plasma membranes from the brains of neonatal (P1) and adult (P35) rats. The Zn 2+ at concentrations ranging from 0 to 100 µM completely blocked these enzyme activities, with a maximum effect at 1 µM (I 50 = 0.2 µM) and 10 µM (I 50 = 2.0 µM), respectively ( Figure 1E). As shown in Figure 1F, the inhibiting effect of Zn 2+ on the studied ATPase activity was dependent on the concentration of Mg 2+ -ATP (0.3-3 mM) in the incubation medium.
NEM (200-400 µM) can both activate as well as inhibit GABA A R efficacy [28,29]. To clarify the role of cysteine residues, we explored the effect of NEM on the Cl − ATPase and the Cl − , HCO 3 − ATPase. As illustrated in Figure 1G, NEM at concentrations ranging from 1 to 100 µM completely blocked the Cl − ATPase activity in neonate (P1) rats. Contrariwise, the Cl − , HCO 3 − ATPase activity was elevated approximately two-fold in the presence of NEM (200-400 µM). The inhibition of Cl − ATPase activity did not significantly recover in the presence of 20 µM Zn 2+ , whereas the NEM activation of the Cl − , HCO 3 − ATPase was eliminated by Zn 2+ , suggesting a similar binding site ( Figure 1H). In addition, the activating effect of 300 µM NEM was eliminated by dithiothreitol (DTT) at 2 mM, indicating the involvement of the thiol groups of cysteine residues ( Figure 1I). Meanwhile, DTT activated the Cl − , HCO 3 − ATPase activity in the control animals (without NEM) this did not appear in the presence of zinc. These data concur with the literature. In particular, DTT has enhanced the responses of functional recombinant GABA A Rs by a mechanism in addition to Zn 2+ chelation [25].
The results show that the chloride-sensitive fluorescent quenching dye MQAE (n-(ethoxycarbonylmethyl)-6-methoxyquinolinium bromide) is a useful tool for determining intracellular chloride activity, and for the quantitative determination of chloride fluxes in neurons [36]. Using MQAE we measured the [Cl − ] i in the SNs (P1), which was approxi-mately 28 mM. To explore the GABA A R activity, the SNs were initially loaded with MQAE and then exposed to a mediator. Previously, it was shown that GABA (1-100 µM) increased the Cl − , HCO 3 − ATPase activity and GABA A R-mediated Cl − transport in the HEK 293FT and neuronal cells with a maximum effect at 100 µM [7,18]. As shown in Figure 2A, SNs (P1) demonstrated a fast (10 s) Cl − efflux in response to the application of 100 µM GABA in HCO 3 − free experimental medium with a maximum peak in fluorescence changes of 14.4 ± 1.6% followed by a reduction in MQAE fluorescence. However, in the presence of 25 mM HCO 3 − in the experimental medium, the GABA-mediated Cl − efflux did not appear. The addition of 40 µM of bicuculline or 25 µM of VO 4 3− to the medium led to a suppression of GABA-mediated fluorescence changes ( Figure 2B), indicating that Cl − efflux from SNs (P1) is an ATP-dependent process via the GABA A R pathway [19]. Early research suggests that the inhibition of the GABA A R-mediated Cl − current in the hippocampus by 50 µM Zn 2+ was stronger in neonate rats (29.3%) than in adults (13%) [13]. In our study, Zn 2+ in the range of 1 to 300 µM inhibited peak fluorescence changes at IC 50 = 80 µM, with the maximum effect (100%) at a concentration of 200 µM ( Figure 2C). The GABA concentration response curve (EC 50 = 9.0 µM) revealed that the inhibition of GABA A R-mediated fluorescence changes by Zn 2+ (100 µM) is non-competitive with little dependence on the mediator concentration (EC 50 = 12.0 µM) ( Figure 2D). In the presence of 25 µM VO 4 3− , the GABA-mediated response was slightly restored (not statistically significant) ( Figure 2E). To clarify the role of cysteine residues in the Zn 2+ modulating effect and considering the data in Figure 1G, we used NEM at 300 µM. As shown in Figure 2F, NEM completely eliminated the GABA-mediated Cl − efflux from the neurons. In the presence of 100 µM Zn 2+ , the NEM effect was noticeably restored. In order to test whether a high concentration of GABA affects ligand modulation [37,38], we applied a low concentration of mediator. Specifically, 10 µM GABA induced a Cl − efflux from the neurons and Zn 2+ (100 µM) or NEM (300 µM) completely eliminated the effect of the mediator, confirming the absence of changes in the pharmacological properties of the receptor.
In mature neurons isolated from adult rats, the [Cl − ] i is approximately 6 mM [2], E Cl − is negative, and GABA A R activation triggers a Cl − influx and subsequent hyperpolarization [32]. We measured the [Cl − ] i in the SNs (P35) to be approximately 5 mM. To clarify the role of the GABA A R activity, the SN S were initially loaded with chloride-sensitive dye and then exposed to the mediator. As shown in Figure 2G, SNs (P35) demonstrated a fast (10 s) Cl − efflux in response to the application of 100 µM GABA in HCO 3 − free experimental medium with a maximum peak in fluorescence changes of 14.0 ± 1.0%. The application of mediator induced a non-significant Cl − influx into the SNs (P35) in the HCO 3 -free medium with a maximum peak in fluorescence changes of 6.0 ± 0.5% ( Figure 2G). The added 40 µM bicuculline or 25 µM VO 4 3− in the medium led to the suppression of the GABA-mediated fluorescence changes, confirming that Cl − influx is an ATP-dependent process via a receptor pore ( Figure 2H). Zn 2+ in the range of 1 to 1000 µM inhibited the peak in fluorescence changes at IC 50 = 100 µM, with maximum effect at a concentration of 500 µM ( Figure 2I). The GABA concentration response curve (EC 50 = 7.0 µM) revealed that inhibition of GABA A R-mediated fluorescence changes by 200 µM Zn 2+ is non-competitive (EC 50 = 7.3 µM) ( Figure 2J). In the presence of 25 µM VO 4 3− , the GABA-mediated response was partially recovered, indicating close binding sites in the ATP-hydrolyzing center ( Figure 2K). To establish the role of cysteine residues in the Zn 2+ modulation effect, we used NEM as a drug-specific modulator of the ATPase activity. As shown in Figure 2L, NEM (300 µM) eliminated the GABA A R-mediated Cl − influx from the neurons. In the presence of 200 µM Zn 2+ , the NEM effect was significantly restored.  In mature neurons isolated from adult rats, the [Cl − ]i is approximately 6 mM [2], ECl − is negative, and GABAAR activation triggers a Cl − influx and subsequent hyperpolarization [32]. We measured the [Cl − ]i in the SNs (P35) to be approximately 5 mM. To clarify the role of the GABAAR activity, the SNS were initially loaded with chloride-sensitive dye and then exposed to the mediator. As shown in Figure 2G, SNs (P35) demonstrated a fast (10 s) Cl − efflux in response to the application of 100 µM GABA in HCO3 − free experimental medium with a maximum peak in fluorescence changes of 14.0 ± 1.0%. The application of mediator induced a non-significant Cl − influx into the SNs (P35) in the HCO3-free medium with a maximum peak in fluorescence changes of 6.0 ± 0.5% ( Figure 2G). The added 40 µM bicuculline or 25 µM VO4 3− in the medium led to the suppression of the GABA-mediated fluorescence changes, confirming that Cl − influx is an ATP-dependent process via a receptor pore ( Figure 2H). Zn 2+ in the range of 1 to 1000 µM inhibited the peak in fluorescence changes at IC50 = 100 µM, with maximum effect at a concentration of 500 µM ( Figure  2I). The GABA concentration response curve (EC50 = 7.0 µM) revealed that inhibition of

Zn 2+ Inhibits the GABA A R-Mediated HCO 3 − Flux in Mature Neurons
The CO 2 production and conversion in the intracellular HCO 3 − has an outwardly directed flow after prolonged GABA exposure, despite a lower permeability via the receptor pore, with a consequent dramatic recovery in [HCO 3 − ] i [39,40]. pH-sensitive fluorescent dyes have been widely applied to monitor changes in intracellular pH. Among them, 2 ,7bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF AM) is cell-permeable. With an increase in the pH of the cells, an increase in fluorescence is observed, and with a decrease in pH, the fluorescence is quenched [41]. To clarify the role of the GABA A Rs in the bicarbonate transport, the SNs were initially loaded with pH-sensitive dye (BCECF) and then exposed to GABA. Since the extracellular concentration of bicarbonate is about 25 mM [42], we studied the GABA A R-mediated HCO 3 − efflux in the absence or presence of 25 mM HCO 3 − in the experimental medium. The SNs (P1) in the absence or presence of 25 mM HCO 3 − showed a rapid decrease in intracellular pH (pH i ) in response to the GABA (100 µM) addition, with a maximum peak in fluorescence changes of approximately 10.6 ± 1.9% followed by a plateau ( Figure 3A). Neither bicuculline at 40 µM or 25 µM VO 4 3− affected the GABA-mediated HCO 3 − efflux in the absence or presence of 25 mM HCO 3 − ( Figure 3B), indicating that it was not via a receptor-dependent pathway and was without the involvement of the ATPase system. Zn 2+ in the range of 1 to 1000 µM did not affect the peak in fluorescence changes ( Figure 3C). As shown in Figure 3D, in the presence of 25 µM VO 4 3 , the GABA-mediated response was unchanged. Similarly, NEM (300 µM) with or without 300 µM Zn 2 in the experimental medium did not change the fluorescence peak ( Figure 3E).
an increase in the pH of the cells, an increase in fluorescence is observed, and with a decrease in pH, the fluorescence is quenched [41]. To clarify the role of the GABAARs in the bicarbonate transport, the SNs were initially loaded with pH-sensitive dye (BCECF) and then exposed to GABA. Since the extracellular concentration of bicarbonate is about 25 mM [42], we studied the GABAAR-mediated HCO3 − efflux in the absence or presence of 25 mM HCO3 − in the experimental medium. The SNs (P1) in the absence or presence of 25 mM HCO3 − showed a rapid decrease in intracellular pH (pHi) in response to the GABA (100 µM) addition, with a maximum peak in fluorescence changes of approximately 10.6 ± 1.9% followed by a plateau ( Figure 3A). Neither bicuculline at 40 µM or 25 µM VO4 3− affected the GABA-mediated HCO3 − efflux in the absence or presence of 25 mM HCO3 − ( Figure 3B), indicating that it was not via a receptor-dependent pathway and was without the involvement of the ATPase system. Zn 2+ in the range of 1 to 1000 µM did not affect the peak in fluorescence changes ( Figure 3C). As shown in Figure 3D, in the presence of 25 µM VO4 3 , the GABA-mediated response was unchanged. Similarly, NEM (300 µM) with or without 300 µM Zn 2 in the experimental medium did not change the fluorescence peak ( Figure 3E). application: Student's unpaired t-test, p < 0.0001 (n = 9) and p < 0.0165 (n = 5). (J) Bar chart showing average maximum GABA response before and after 300 µM NEM or 300 µM NEM + 300 µM Zn 2+ application: Student's unpaired t-test, p < 0.0011 (n = 6) and p < 0.0004 (n = 6). Data are presented as mean ± SEM. * p < 0.05, ns, not significant.
To explore the GABA A R activity in adult rats, the SN S (P35) were initially loaded with BCECF and then exposed to GABA. The SNs (P35) in the HCO 3 − -free medium showed a rapid pH i decrease in response to the GABA (100 µM) addition with a maximum peak in fluorescence changes of 7.3 ± 1.0%, while in the presence of 25 mM HCO 3 − , the GABAmediated HCO 3 − efflux was elevated twice with a maximum peak in fluorescence change of 15.7 ± 1.0% and a subsequent rapid (30 s) recovery of [HCO 3 − ] i ( Figure 3F). As illustrated in Figure 3G, 40 µM bicuculline, as well as 25 µM VO 4 3− , eliminated the recovery of GABAmediated pH i changes in the presence of 25 mM HCO 3 − , confirming that it is an energydependent process within the channel [19]. Zn 2+ in the range of 1-1000 µM inhibits the recovery peak in fluorescence changes at IC 50 = 90 µM and is completely eliminated at 400 µM ( Figure 3H). As shown in Figure 3I, in the presence of 25 µM VO 4 3− , Zn 2+ -induced inhibition at 300 µM was partially restored, indicating close site binding in the ATP-hydrolyzing center. NEM (300 µM) completely eliminated the GABA-mediated [HCO 3 − ] i changes, but in the presence of 300 µM Zn 2+ , the NEM effect was noticeably restored ( Figure 3J).

β3 Subunit Is Responsible for Zn 2+ -Sensitive [HCO 3 − ] i Recovery
Although, the structures of synaptic GABA A Rs are composed of 2α, 2β, and γ subunits, the functional and pharmacological properties are largely ensured by β subunits, of which the β3 subunit has an important role to play [43]. Previous studies of recombinant GABA A Rs report that in these three subunits (α2, β3, and γ2) only the β3 subunit possesses the Cl − , HCO 3 − ATPase activity [18]. To determine whether such ATPase characteristics are unique to β3 or whether β1 and β2 subunits could also manifest an enzyme activity that will be inhibited by Zn 2+ [18], we expressed single rat β1, β2, or β3 subunits. The homomeric GABA A R β1 and β3 isoforms demonstrated Cl − ATPase or Cl − , HCO 3 − ATPase activities (263.6 ± 22.0 and 357.6 ± 32.7 nmol P i ·min −1 ·mg −1 , respectively), but did not appear in cells expressing the β2 subunit ( Figure 4A). HEK 293FT cells expressing the GABA A R β1, β2, or β3 isoforms showed one band in the VLPs, with a molecular weight of approximately~54 kDa, that bound to the antibodies against the GABA A R β1, β2, or β3 subunits, respectively ( Figure 4B). As shown in Figure 4C, Zn 2+ (1-100 µM) completely suppressed the enzyme activity of the GABA A R β1 and β3 isoforms. The addition of NEM (300 µM) in the experimental medium resulted in an increase in ATPase activities by approximately one and eight-tenths. In the presence of 20 µM Zn 2+ , the activating effect of NEM did not appear ( Figure 4D).
Several recombinant receptor studies demonstrated that the potency of Zn 2+ is higher on the αβ subtypes than the αβγ receptors [7,10,11]. Furthermore, the homomeric channels formed by only α2 or β2 subunits were non-competitively blocked by 10 µM Zn 2+ to approximately the same extent (>80%) as the α1β2 isoform [44]. To ensure that the homomeric GABA A R subtypes were functional and would be inhibited by Zn 2+ , we examined the whole-cell flow evoked by GABA (100 µM) in HEK 293FT cells transfected with plasmid vectors containing GABA A R subunit cDNAs to produce the β1 or β3 subtypes. Considering that the homomeric β3 GABA A R isoforms demonstrated a marginal effect on the GABA-mediated Cl − inflow in cells [18], in the present study, we used the more effective benzamidine (5 mM) instead of GABA [45]. By way of comparison, we tested the effect of Zn 2+ on the homomeric β3 and the heteromeric α2β3γ2 GABA A R isoforms.
As shown in Figure 4E, Zn 2+ at concentrations ranging from 0 to 500 µM completely blocked the MQAE fluorescence changes with maximum effects at 30 µM and 300 µM, respectively. While in the homomeric GABA A R ensembles containing the homomeric β1 or β3 isoforms, GABA at 100 µM induced a similar decrease in the pH i of cells previously resuspended in a medium with or without 25 mM HCO 3 − by approximately 8.0 ± 0.7% ( Figure 4H). The HEK 293FT cells expressing the GABA A R β3 subtype in the HCO 3 − -free medium also manifested only a GABA-mediated decrease in pH i (7.9 ± 0.9%) while in the presence of 25 mM HCO 3 − , the cells expressing the GABA A R β3 subtype in contrast to the β1 isoform, showed the GABA-mediated decrease in pH i with a maximum peak fluorescence change of 14.0 ± 1.3%, which was recovered over 30 s (9.9 ± 0.5%) ( Figure 4I). As shown in Figure 4J, NEM (300 µM) had no significant effect on the GABA-mediated HCO 3 − flow in the cells expressing the homomeric GABA A R β1 isoform in the absence or presence of 20 µM Zn 2+ . While in the cells expressing the GABA A R β3 isoform, the NEM inhibiting effect was partially recovered by 20 µM Zn 2+ ( Figure 4K). meric GABAAR subtypes were functional and would be inhibited by Zn 2+ , we examined the whole-cell flow evoked by GABA (100 µM) in HEK 293FT cells transfected with plas mid vectors containing GABAAR subunit cDNAs to produce the β1 or β3 subtypes. Con sidering that the homomeric β3 GABAAR isoforms demonstrated a marginal effect on the GABA-mediated Cl − inflow in cells [18], in the present study, we used the more effective benzamidine (5 mM) instead of GABA [45]. By way of comparison, we tested the effect o Zn 2+ on the homomeric β3 and the heteromeric α2β3γ2 GABAAR isoforms.  A single cysteine residue (C313) in the M3 domain is conserved in all GABA A R α, β, and γ subunits ( Figure 4F). To test whether this cysteine residue formed, at least partially, the molecular basis for NEM modulation, this residue was mutated to alanine in the β1 and β3 subunits (C313A). The HEK 293FT cells expressing the mutant GABA A R β1 or β3 isoforms showed one band in the VLPs with a molecular weight of approxi-mately~54 kDa, that bound to the antibodies against the GABA A R β1 or the β3 subunit, respectively ( Figure 4N). As shown in Figure 4O, the mutant (C313A) GABA A R β1 and β3 isoforms demonstrated Cl − and Cl − , HCO 3 − ATPase activities of 218.8 ± 24.8 and 279.2 ± 29.0 nmol −1 ·P i ·min −1 ·mg −1 , respectively. The addition of 300 µM NEM in the experimental medium did not change the enzyme activities. The homomeric mutant GABA A R β1 or β3 (C313A) isoform still displayed the GABA-mediated pH i drop in either the absence or presence of HCO 3 − by approximately 8.0 ± 0.9% ( Figure 4L,M), and this did not change after the application of NEM at 300 µM in the absence or presence of 20 µM Zn 2+ (Figure 4P,R).

Discussion
Currently, the properties of GABA A Rs are studied using electrophysiological (e.g., patch-clamp) and non-electrophysiological (e.g., fluorescence-based) methods. Although the fluorescence-based, in contrast to the patch-clamp method, does not directly measure ionic current and ion-concentration-dependent changes of fluorescence signals as a result of ionic flux, using environmentally sensitive dyes may allow to detect the conformational changes [46]. We used fluorescence recordings to directly assess the contribution of anions on GABA A R functional activity during ontogenesis. Our results shed light on the substantial role of HCO 3 − in the GABA efficacy in SNs isolated from adult rats (P35). In addition, we focused on the role of ATPase in the rapid re-establishment of anionic gradients after GABA A R responses. During the first days (P1-P4), Cl − , but not HCO 3 − , played a dominant role in the ATPase activity, while after postnatal day 10, a dominant role for HCO 3 − and a minor role for Cl − became apparent (Figure 1). The dramatic transient switching in the effectiveness and performance of the enzyme forms from Cl − ATPase to Cl − , HCO 3 − ATPase during early postnatal development demonstrates the involvement of primary-active transport in the recovery not only of [Cl − ] i , but also [HCO 3 − ] i after day P10. In addition, it was established that Cl − ATPase is related to the GABA A R β1 isoform, whereas the Cl − , HCO 3 − ATPase belonged to the GABA A R β3 isoform. Most likely, this is associated with both the distinct properties of the subunits and their traffic changes to the cell surface [47]. Data from the literature shows the variable expression of mRNA GABA A R β1, β2, or β3 subunits in hippocampal granule cells during postnatal development. Specifically, in neurons isolated in samples between days P5 and P7, there is a significant expression of the GABA A R β1 subunit mRNA in contrast to the β2 or β3 subunits [14], and neurons isolated in samples from postnatal days 17-21 had an increased expression of the GABA A R β2 and β3 subunits' mRNA. In this study, we show for the first time that the homomeric β3 isoform, in contrast to the β1 isoform, manifests an essential GABA-mediated HCO 3 − outflow and its recovery. These data suggest the high probability of the presence of a distinctive mechanism for HCO 3 − transport via the GABA A Rs. Notably, previous work has gained attention in the concluding observations that, out of the three β subunits, only the expression of β3 in the β1-β2 subunit knockout can fully maintain or restore inhibitory responses to control levels in the hippocampus [48].
Although early studies reported that Zn 2+ inhibits both primary active transporters and secondary active cation-chloride cotransporters (CCCs), there are essential differences in their sensitivity to its inhibiting action. Specifically, Zn 2+ inhibits the erythrocyte Ca 2+ ATPase with a K i of 80 pM [49], whereas CCCs are inhibited at high concentrations (I 50 = 50 µM) [50]. In the present study, we propose a mechanistic and molecular basis for the inhibition of GABA A R-coupled ATPase by Zn 2+ and its dependence on the stage of postembryonal development. The ATPase activity was inhibited by Zn 2+ , but its sensitivity to cations differed in neonatal (IC 50 = 0.2 µM) and adult rats (IC 50 = 2.0 µM), which is simi-lar to the results of electrophysiological studies. Specifically, the GABA A R-mediated Cl − current recorded in hippocampal slices isolated from postnatal rats (P1-P5) was inhibited by 80% at 100 µM Zn 2+ , in contrast to adult rats (30% inhibition at 400 µM Zn 2+ ) [13,14]. In confirmation of these data, the GABA A R-mediated Cl − flow in the SNs (P1) was inhibited by~90% at 100 µM Zn 2 + (I 50 = 80 µM), in contrast to adult rats (80% inhibition at 250 µM Zn 2+ ) (I 50 = 100 µM) (Figure 2). Based on the received data, the physiological role of zinc in the modulation of GABA A R/ATPase functional activity is in question because zinc ions are released both extra-and intracellularly [16]. Cellular "free" zinc concentrations are between 10 and 1000 pM and these concentrations are similar to the affinity of Zn 2+ for cytosolic "metallobinding" enzymes [22]. Effective micromolar zinc-inhibition can have physiological significance only at an extracellular binding with the receptor complex. However, some studies have shown that Zn 2+ can penetrate via the channel as a complex ion with permeating anions resulting in the intracellular or membrane inhibition effect [51,52]. Specifically, on cultured hippocampal neurons it was showed that a continuous background release of GABA induced a standing-sensitive inward Cl − -current that was inhibited by bicuculline [51]. This leakage current is initially reduced in amplitude by 300 µM Zn 2+ and eventually converted, in the continued presence of zinc, into discrete discontinuous transients appearing. It can be assumed that in our study we have also observed both the extra-and intracellular effects of Zn 2+ on the GABA A R-mediated anion transport and AT-Pase activity ( Figure 5). In addition, in our study, the substrate (Mg 2+ -ATP) and Zn 2+ were likely to have competed for the same ATP-hydrolysis site, indicating the binding of Zn 2+ to the active site. This finding aligns with the data showing that Zn 2+ can competitively bind to the catalytic center of various enzymes [22,53]. For example, Zn 2+ at nanomolar concentrations inhibits the receptor protein tyrosine phosphatase β activity, which contains a catalytic cysteine residue [53].
Recently, three distinct Zn 2+ -binding sites on the GABA A R were identified: one site within the ion pore of the β3 subunit is for His267 and Glu270 residues, and the other two occur on the external amino (N)-terminal face between the β (Glu 182) and α (Glu137 and His141) subunits [10,11]. Early research questioned the possible role of a cysteine residue in the structure of pLGICs, which theoretically could interact with Zn 2+ [8]. However, the involvement of cysteine in the Zn 2+ -inhibition potency of GABA A Rs was not demonstrated. Here, NEM completely inhibits the Cl − ATPase activity and Zn 2+ eliminates the NEM effect on Cl − , HCO 3 − ATPase form (Figure 1), implicating it as the catalytic cysteine (Cys313) and nearby residues in the coordination of Zn 2+ in the M3 domain of these β subunits ( Figure 4F). This line of reasoning confirms that mutant isoforms do not show the activation of ATPase by NEM. In addition, the NEM effect on the GABA A Rmediated Cl − or HCO 3 − fluxes was eliminated by vanadate in the presence of HCO 3 -, which denoted a close site of localization of the ATP-hydrolyzing center and cysteine residue (C313) in the M3 domain of the β3 subunit [19]. However, it should be noted that a side chain of Cys313 faces inside the β-subunit and is buried between the TM domains of the surrounding residues making it difficult to access. Moreover, this amino residue is not a part of the channel and does not ensure its formation. Therefore, we can assume that NEM may cause not a direct, but an allosteric effect on the enzyme activity that does not consider cysteine residue as absolutely catalytic. Based on the data obtained, two possible molecular mechanisms for Zn 2+ potency can be suggested: (1) the reaction of sulfhydryl bonds in the receptor-channel protein with Zn 2+ , and (2) the formation of inactive complexes between Zn 2+ and the ATPase ( Figure 5). Given that the ATP-hydrolyzing site is localized intracellularly and Cys313 is located in the transmembrane domain approximately in the middle of the membrane ( Figure 4G), it is more likely that not only cysteine, but also other amino acid residues that form at the active center are also involved in the zinc-induced inhibition of enzyme activity. residue as absolutely catalytic. Based on the data obtained, two possible molecular mechanisms for Zn 2+ potency can be suggested: (1) the reaction of sulfhydryl bonds in the receptor-channel protein with Zn 2+ , and (2) the formation of inactive complexes between Zn 2+ and the ATPase ( Figure 5). Given that the ATP-hydrolyzing site is localized intracellularly and Cys313 is located in the transmembrane domain approximately in the middle of the membrane (Figure 4G), it is more likely that not only cysteine, but also other amino acid residues that form at the active center are also involved in the zinc-induced inhibition of enzyme activity.  Intracellular neuronal zinc modulation is associated with a variety of physiological signaling pathways (including protein kinases and protein phosphatases) [54][55][56]. Here, we expand on these data and demonstrate that, during brain maturation, Zn 2+ with various efficacies inhibited not only the passive GABA A R-mediated responses, but also the ATPase compartment determined by β1 or β3 subunits. In addition, we established that the Cl − ATPase form belongs to the β1 subunit, whereas the β3 subunit preferably possesses the Cl − , HCO 3 − ATPase activity. Overall, we describe a new a role for Zn 2+ in the inhibition of GABA A R-coupled ATPase activity and present evidence of the existence of a new intracellular site responsible for its potency via binding with cysteine. In this context, given the current structural and kinetic data, identifying the molecular determinants underlying the extracellular regulation of GABA A R function, intracellular Zn 2+ regulation can have physiological and pathophysiological implications [57]. Indeed, GABAergic signaling is unique in that its polarity of action depends on [Cl − ] i and [HCO 3 − ] i [2,3], which are highly labile, leading not only to inhibitory, but also depolarizing/excitatory actions under certain conditions (for example, massive activation and spinal cord lesions, etc.) [35,57]. Moreover, altering [Cl − ] i on the second scale through changing GABA A R desensitization/resensitization may cause the collapse of the anion gradients and contribute to the induction of pathological conditions (e.g., seizures or epilepsy) modulated by Zn 2+ [15,58]. A mechanistic understanding of the interactions between Zn 2+ and the ATP-hydrolysis center within the receptor molecule may have clinical implications for the therapy of brain disorders by regulating the formation of an unstable, high-energy, phosphorylated intermediate and could pave the way for novel drug design.

Animals
Animal experiments were carried out using adult male Wistar rats purchased from the Institute of General Pathology and Pathophysiology vivarium and weighing 130-160 g at the time of arrival unless otherwise stated. Rats were always group-housed (5 per cage) and maintained in a temperature-controlled environment (23 ± 1) on a 12:12 h light-dark cycle and had access to food and water ad libitum. We performed all manipulations on animals in accordance with EU directive
PMs were prepared from control HEK 293FT cells and various GABA A R variants were detached using Hanks' balanced salt solution (Gibco, Waltham, MA, USA) without divalent cations (i.e., trypsin was not used), and the cells were centrifuged at 300× g for 3 min. The HEK 293FT cells or brain (mostly cortex) were homogenized in an ice-cold buffer containing 0.3 M sucrose, 0.5 mM EDTA-Tris, HEPES-Tris, 10 mM (pH 7.3), and protease inhibitor cocktail tablets (A32955, Thermo Fisher Scientific, Waltham, MA, USA), and centrifuged at 10,000× g for 15 min at 4 • C, after which the pellet was discarded. The supernatant was centrifuged for 1 h at 150,000× g and the resulting pellets were resuspended in 20 mM HEPES-Tris pH 7.3. This plasma membrane-enriched preparation was used for further measurements of the enzyme activity. Ethylenediaminetetraacetic acid (60-00-4), 4-(2hydroxyethyl)-1-piperazineethane-sulfonic acid (HEPES), and Tris(hydroxymethylaminomethane (77-86-1) were obtained from Merck (Kenilworth, NJ, USA). For transfection procedures and virus-like particle (VLP) production, the same growth medium with decreased FBS content up to 4% was used according to the manufacturer's recommendations. Geneticin G418 sulphate (11811031, Invitrogen, Waltham, MA, USA) was present in the growth medium at a concentration of 500 mg/mL constantly except during the transfection. The cells were subcultured at confluence by treatment with 0.05% trypsin and 0.02% EDTA in PBS. For selection purposes and improving the yield of VLPs, the transfection medium was removed after 24 h and a fresh growth medium with 10 µg/mL blasticidin (R21001 Gibco, Waltham, MA, USA) was added. Transfected cells and VLPs were collected and analyzed 24-48 h after transfection.

Molecular Biology
The genes encoding the full-length rat GABA A R β1, β2 or β3 subunits were amplified by PCR from the cDNA library (Evrogen, Moscow, Russia) using gene-specific primers with Kozak sequence at the 5 end of the forward primer based on "GenBank:NM_012956.1", "GenBank:NM_012957.2" "GenBank:NM_017065.1" sequences. The PCR products were cloned into the pEF6/V5-His TOPO TA vector (K961020, Invitrogen, Waltham, MA, USA) separately and verified by DNA sequencing. Each vector was amplified using E. coli TOP10 strain in LB medium supplemented with 20 µg/mL ampicillin. Isolation and purification of plasmids were performed with PureYieldTM Plasmid Miniprep System (Promega, Madison, WI, USA) and Plasmid Midiprep 2.0 (Evrogen, Moscow, Russia). The sterilization of plasmids was implemented via 0.22-µm filtration. The concentration of plasmids was evaluated on spectrophotometer NanoDrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA). The quality validation of cloning and growth was performed additionally through enzymatic restriction by XbaI and BamHI in BamHI buffer (Thermo Fisher Scientific, Waltham, MA, USA), and the following electrophoresis in 1% agarose gel.
The typical transfection procedure of GABA A R subunit-containing constructs for the subsequent biochemical, spectrofluorometric, and Western blot analyses was as follows. Approximately 5 × 105 HEK293FT cells were suspended in 8 mL DMEM, plated into a 90-mm culture dish, and maintained 24 h approximately until 50-90% confluence. Then, 5 µg of plasmid DNA (β3 alone) was added combined with Lipofectamine ® 3000 Reagent (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) in Opti-MEM ® I (1×) + GlutaMAXTM-I medium (51985-026, Gibco, Inchinnan, UK) accordingly the manufacturers' recommendations. For microscopy, the cells were plated in 35 mm dishes and were incubated with a proportional amount of reagents and vectors.
For VLP production, GABA A R subunit-containing constructs were transfected together with Membrane Pro TM Reagent (Invitrogen, Waltham, MA, USA) amenably. Transfected HEK293FT started to bud off VLPs from the cell membrane approximately 24 h after transfection. The harvesting procedure was executed in conformity with manufacturer's recommendations. Briefly, the VLP-containing medium was mixed with Membrane Pro TM Precipitation Mix in the ratio of 5 to 1, where 5 refers to the medium. Then, the mix was incubated at 4 • C overnight. After incubation, VLPs were pelleted by centrifugation at 5500× g for 30 min and resuspended in HEPES buffer for subsequent analysis or stored at −80 • C. SNs were loaded dye (MQAE or BCECF) in BSS for 1 h at 37 • C and stored in an opaque test tube at RT or 4 • C. For that, the control HEK 293FT cells and various GABA A R β3 isoforms cells were trypsinized by adding 0.05% trypsin-EDTA solution (25200056, Gibco BRL, Waltham, MA, USA), washed PBS twice, resuspended in the BSB, and then loaded with MQAE for 1 h at 37 • C. After loading, the suspension was centrifuged at 200× g for 5 min at RT and kept in the aforementioned medium at RT in the opaque test tubes. For analysis, the pellet was resuspended in the BSB. Monitoring was performed with cells continuously superfused with incubation medium composed of (mM):135 NaCl (or 135 NaCl and 25 NaHCO 3 ), 0.5 KH 2 PO 4 , 0.8 MgCl 2 , and 5 mM Hepes (pH 7.4). Dye-loaded cells (SNS or HEK 293FT) were equilibrated in the test tube in the incubation medium V = 200 µL) in the absence or presence of compounds (ZnCl 2 , Na 3 VO 4 , NEM, bicuculline, or ouabain) for about 10 min at 37 • C before initial fluorescence measurements, and then 150 µL of the suspension was added into quartz microcuvette (non-flow cell) and stirred. The GABA-mediated Cl − or HCO 3 − transport was initiated directly by addition of GABA in final concentration of 1-100 µM in the cuvette by an in-house solution supply system. GABA A R-mediated Clor HCO 3 transport was assessed by the dynamic measurements of the variations in the fluorescence intensity of Clsensitive fluorescent dye MQAE-loaded or BCECF-loaded HEK 293FT cells or SNs using a FluoroMax ® -4 spectrofluorometer (HORIBA Scientific Edison, Piscataway, NJ, USA), respectively. The excitation and emission wavelengths were 350 nm and 480 nm for the measurement of Cl − -transport or 490 nm and 535 nm for measurement of HCO 3 −transport, respectively. The ∆F/F of each trial was calculated as (F − F 0 )/F 0 , where F 0 is the baseline fluorescence signal averaged over a 10 s period (this was the control measurement) immediately before the start of the application of GABA and supplement compounds. The value of 100% was obtained as the fluorescence intensity before the application of GABA, in the absence or presence of test compounds. The maximum amplitude of GABA-mediated fluorescence responses was calculated as the maximal difference in fluorescence intensity in the absence or presence of an agonist.
After 15-20 min of incubation, the ATPase activity was stopped by the addition of reagents for measuring of inorganic phosphorus (P i ). The Cl − -and Cl − , HCO 3 − ATPase activities were determined as a difference in formation of P i in the absence and in the presence of NaCl (2-60 mM) or NaHCO 3 (2.5-25 mM) in the incubation medium, respectively. Adenosine 5 -triphosphate (ATP) disodium salt hydrate (34369-07-8) and adenosine 5 -triphosphate disodium salt hydrate (A26209) were obtained from Merck (Kenilworth, NJ, USA). The concentration of P i in the incubation medium was measured by a modified method of Chen et al., (1956) [19] using a Cary 60 UV-vis spectrophotometer (Agilent, Santa Clara, CA, USA) at wavelength of 650 nm. The γ-phosphate analog, orthovanadate (VO 4 3− ) (Sigma-Aldrich, St. Louis, MI, USA), was obtained by boiling the vanadate solution (pH 10; 10 min), and freshly boiled stock was diluted to the final concentration (pH 7.3) prior to use.

Western Blot
VLPs of transfected HEK 293FT cells were subjected to SDS-PAGE using the SDS-PAGE reagent starter kit (1615100 Bio-Rad, Hercules, CA, USA) and to Western blot analysis using the Pierce™ fast Western blot kit (35055 Thermo Scientific, Waltham, MA, USA), ECL Plus Western Blotting Detection System Substrate (GE Healthcare, Chicago, IL, USA). Samples were SDS-treated by boiling for 5 min in a buffer consisting of 62.5 mM Tris, 10% glycerol, 5% 2-mercaptoethanol, 4% SDS, and 0.001% bromophenol blue, and then~20 µg of total protein was loaded into SDS-PAGE. Electrophoresis parameters were: 70 V for 10 min on 4% SDS-PAGE stacking gel and 120 V for 50 min on 12.5% SDS-PAGE resolving gel. Proteins were transferred on PVDF membrane by the semi-dry method using 0.09 A/cm 2 for 1 h. After that, membranes were incubated for 1 h in blocking solution containing 5% milk, and then incubated at 4 • C overnight with primary Anti-GABRB1 Monoclonal Antibody (S96-55) (Catalog # MA5-27699, TermoFisher Scientific, Waltham, MA, USA), Anti-GABRB2 Recombinant Rabbit Monoclonal Antibody (ARC0631) or GABRB3 antibody [N87/25] (ab98968, Abcam, Cambridge, UK), diluted 1:1000 with the blocking solution. After incubation, the membranes were washed with TBS-T 4 times for 15 min each, and then incubated at RT for 1 h with secondary HRP-conjugated antibodies (62-6520 Thermo Fisher Scientific, Waltham, MA, USA) diluted 1:5000 with the blocking solution. Then, the membrane was washed with TBS-T four times and the GE Healthcare ECL Plus Western Blotting Detection System (Amersham TM , GE Healthcare, UK) was applied according to manufacturer's instructions. The visualization of the bands was performed using a Kodak Image Station 440 (Rochester, NY, USA).

Statistical Analysis
The data are expressed as the mean ± SEM, and differences were considered significant for p < 0.05. Statistical differences were determined by two-tailed Student's unpaired t-test for data with equal variances and which were assessed as normally distributed with the Shapiro-Wilk test. Graphs and statistical analysis were obtained by using GraphPad Prism 9.3 (GraphPad Software, San Diego, CA, USA).
Representative images of MQAE and BCECF fluorescence changes in synaptoneurosomes VLPs were analyzed and assessed using Origin Pro version 9.1 for Windows (OriginLab, Northampton, MA, USA). The mean fluorescence intensity from each treatment group was separately compared to the mean fluorescence intensity of the untreated control group. Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.