*Article* **Activation of Neuronal Nicotinic Receptors Inhibits Acetylcholine Release in the Neuromuscular Junction by Increasing Ca 2+ Flux through Cav1 Channels**

**Nikita Zhilyakov 1, \* , Arsenii Arkhipov 1 , Artem Malomouzh <sup>1</sup> and Dmitry Samigullin 1,2, \***


**Abstract:** Cholinergic neurotransmission is a key signal pathway in the peripheral nervous system and in several branches of the central nervous system. Despite the fact that it has been studied extensively for a long period of time, some aspects of its regulation still have not yet been established. One is the relationship between the nicotine-induced autoregulation of acetylcholine (ACh) release with changes in the concentration of presynaptic calcium levels. The mouse neuromuscular junction of m. Levator Auris Longus was chosen as the model of the cholinergic synapse. ACh release was assessed by electrophysiological methods. Changes in calcium transients were recorded using a calcium-sensitive dye. Nicotine hydrogen tartrate salt application (10 µM) decreased the amount of evoked ACh release, while the calcium transient increased in the motor nerve terminal. Both of these effects of nicotine were abolished by the neuronal ACh receptor antagonist dihydro-beta-erythroidine and Cav1 blockers, verapamil, and nitrendipine. These data allow us to suggest that neuronal nicotinic ACh receptor activation decreases the number of ACh quanta released by boosting calcium influx through Cav1 channels.

**Keywords:** neuromuscular junction; neurotransmitter release; acetylcholine; nicotinic receptor; calcium channel; calcium transient

## **1. Introduction**

Acetylcholine (ACh) is the main neurotransmitter in the peripheral nervous system of vertebrates and humans. In particular, it is responsible for the transmission of signals from the motor nerve to the skeletal muscles [1,2]. Since the neuromuscular junction is a key linker in the initiation of any motor act (from voluntary movement of the limbs, to breathing, to contraction of the vocal cords), an investigation of the regulation of neuromuscular transmission is of great importance for both fundamental neurobiology and applied medicine.

Since the midst of the 20th century, the data began to accumulate indicating that ACh, released in the synaptic cleft from the nerve endings, activates presynaptic cholinergic receptors, thus exerting a modulatory effect on the neurotransmission process by changing the amount and/or dynamics of subsequent portions of neurotransmitter release [2–7]. Initially pharmacologically, and later by other methods it has been shown that both ionotropic nicotinic and metabotropic muscarinic cholinergic receptors are present in the motor nerve terminal, and the activation of these receptors can lead to autoregulation of ACh release [4,5,8–11]. According to the data of morphological and functional analysis, presynaptic cholinergic receptors can be located both near the active zones and relatively far from the synaptic cleft [4,12,13].

**Citation:** Zhilyakov, N.; Arkhipov, A.; Malomouzh, A.; Samigullin, D. Activation of Neuronal Nicotinic Receptors Inhibits Acetylcholine Release in the Neuromuscular Junction by Increasing Ca 2+ Flux through Cav1 Channels. *Int. J. Mol. Sci.* **2021**, *22*, 9031. https://doi.org/ 10.3390/ijms22169031

Academic Editors: Piotr D. Bregestovski and Carlo Matera

Received: 10 August 2021 Accepted: 19 August 2021 Published: 21 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

When studying autoregulation mediated by muscarinic cholinergic receptors, it was found that the activation of the M1-subtype receptors led to facilitation of the release. In contrast, activation of the M2-subtype caused inhibition of the ACh quanta release [14,15]. Both the M1- and M2-mediated mechanisms depend on calcium influx [14,16–18].

Studies of the mechanisms of the autoregulation of ACh release mediated by nicotinic cholinergic receptors are complicated by the fact that the predominant population of these proteins is located in the postsynaptic membrane. Their activation is accompanied by the depolarization of sarcolemma and the subsequent generation of action potential, which ultimately leads to muscle contraction. The data indicate that the activation of presynaptic nicotinic cholinergic receptors leads to inhibition of the process of ACh release [19–21].

Also, experimental evidence was obtained indicating the possible involvement of voltage-gated calcium channels (VGCCs) of the L-type (Cav1) in the modulation of neurotransmission [20,22]. Meanwhile, the results of a number of studies demonstrate that neither the Cav1 type nor the N-type (Cav2.2) VGCCs participate in the evoked release of ACh in mammalian neuromuscular contacts [23–26].

Thus, the question of the role of calcium channels in the mechanisms of the regulation of ACh release, as mediated by nicotinic cholinergic receptors, remains open as of now.

In the present study, using a pharmacological approach, electrophysiological techniques and the method of the optical registration of changes in calcium levels in the motor nerve endings, we made the following observations. An agonist of nicotinic receptors (at a concentration not significantly affecting the state of the postsynaptic membrane) leads to a decrease in the amount of released ACh quanta. This effect is accompanied not by a decrease, but by an increase of calcium ion entry into the motor nerve terminal. Our data suggest that the nicotinic cholinergic receptors responsible for the mechanism of ACh release autoregulation are the receptors of neuronal type. Activation of these receptors leads to upregulation of the Cav1 type of VGCCs, resulting in the enhancement of Ca2+ entry into the nerve ending.

#### **2. Results**

#### *2.1. Effects of Nicotine on the Electrophysiological Parameters of the Neuromuscular Junction*

Using the intracellular microelectrode technique, we recorded the resting membrane potential (RMP) of the muscle fiber, amplitude of miniature endplate potentials (mEPPs), frequency of occurrence of mEPPs, and the amplitude of the evoked potentials of the end plate (EPPs).

Alterations in RMP and mEPP amplitude indicate the postsynaptic action of the pharmacological agent. Changes in the frequency of occurrence of mEPPs suggest the presynaptic action of the drug. The EPP amplitude, in turn, can vary due to changes at both the pre- and post-synaptic levels. Therefore, it was necessary to assess the effect of nicotine on every parameter mentioned above to determine the optimal effective concentration of nicotine to study the autoregulation.

The control RMP value of muscle fibers was −71.48 ± 0.77 mV (*n* = 30). Application of nicotine at concentrations of 0.1 µM, 1 and 5 µM did not affect the RMP significantly, providing values of −69.72 ± 0.90 mV (*n* = 30), −70.91 ± 0.78 mV (*n* = 30), and −70.85 ± 0.87 mV (*n* = 30), respectively (Figure 1a). A slight significant depolarization was observed when nicotine concentration was increased to 10 µM (−67.04 ± 0.79 mV; *n* = 30). At a concentration of 50 µM, a more pronounced depolarization was observed, and the mean RMP value decreased to −56.94 ± 1.29 mV (*n* = 30; Figure 1a).

Another sign of the postsynaptic action of nicotine was the change in the amplitude of mEPP. The mean value of the amplitude of the spontaneous signal in the control was 0.88 ± 0.05 mV (*n* = 30). Application of nicotine in concentration up to 10 µM did not affect mEPP amplitude significantly, with the values being 0.83 ± 0.04 mV (*n* = 30) for 0.1 µM, 0.84 ± 0.05 mV (*n* = 30) for 1 µM, 0.96 ± 0.06 mV (*n* = 30) for 5 µM, and 0.83 ± 0.05 mV (*n* = 30) for 10 µM (Figure 1b). Asignificant decrease in mEPP amplitude to 0.55 ± 0.04 mV (*n* = 30) was observed only with 50 µM nicotine (Figure 1b).

application (the range from 0.1 to 50 μM). Results are expressed as mean ± SEM. Asterisk way ANOVA test with Dunnet's post **Figure 1.** Effects of nicotine on the electrophysiological parameters registered at the mouse neuromuscular junction. Changes in absolute values are shown (**a**) resting membrane potential (RMP) of muscle fibers, (**b**) amplitudes of miniature endplate potentials (mEPP), (**c**) frequency of the mEPPs, and (**d**) the amplitudes of evoked endplate potentials (EPP) in the control and 15 min after nicotine application (the range from 0.1 to 50 µM). Results are expressed as mean ± SEM. Asterisks (\*) indicate significant effects (*p* < 0.05, one-way ANOVA test with Dunnet's post-hoc comparison).

) for 0.1 μM,

). After application of 5 μM

) for 1 μM, 0.96 ± 0.06 mV ( ) for 5 μM, and 0.83 ± 0.05 mV (

of the frequency of mEPPs upon application of 0.1 and 1 μM nicotine

0.1 μM, 1 and 5 μM did not alter the average amplitude

starting at the concentration of 10 μM (28.36 ± 1.27 mV; ), while at 50 μM the ampli-

centration of 10 μM nicotine was chosen. When using nicotine at this concentration, a de-

for 10 μM and 0.57 ± 0.06 Hz

) for 10 μM (Figure

μM (Figure

) was observed only with 50 μM nicotine (Figure

In contrast to the amplitude of mEPPs, the effect of nicotine on the frequency of occurrence of spontaneous signals was detected at significantly lower concentrations. That is, the average values of the frequency of mEPPs upon application of 0.1 and 1 µM nicotine were 1.64 ± 0.15 Hz (*n* = 30) and 1.22 ± 0.12 Hz (*n* = 30), respectively, and did not differ from the control value of 1.57 ± 0.14 Hz (*n* = 30; Figure 1c). After application of 5 µM nicotine, the frequency significantly decreased to 1.07 ± 0.08 Hz (*n* = 30), and inhibition was further enhanced to 0.98 ± 0.07 Hz (*n* = 30) for 10 µM and 0.57 ± 0.06 Hz (*n* = 30) for 50 µM (Figure 1c).

The amplitude of the EPP, which reflects the level of evoked ACh release and depends on changes in the sensitivity of the postsynaptic membrane in the area of the neuromuscular contact, was 32.80 ± 1.02 mV (*n* = 30) in the control. Application of nicotine at concentrations of 0.1 µM, 1 and 5 µM did not alter the average amplitudes of EPP, which were equal to 31.93 ± 1.21 mV (*n* = 30), 32.43 ± 1.11 mV (*n* = 30), and 32.33 ± 1.18 mV (*n* = 30), respectively (Figure 1d). However, nicotine produced a decrease in EPP amplitude, starting at the concentration of 10 µM (28.36 ± 1.27 mV; *n* = 30), while at 50 µM the amplitude decreased almost twofold to 16.89 ± 1.09 mV (*n* = 30; Figure 1d).

Thus, for further investigations of the ACh release autoregulation mechanisms, concentration of 10 µM nicotine was chosen. When using nicotine at this concentration, a decrease in the EPP amplitude was observable, while there were no changes in the mEPP amplitude (with only a slight depolarization of the sarcolemma).

#### *2.2. Activation of Neuronal Nicotinic Receptors Leads to Downregulation of the EPP Quantal Content*

Under control conditions, the quantal content (QC) was 46.8 ± 4.5. The bath application of nicotine (10 µM) decreased the QC of EPP significantly by 12.0 ± 4.4% (*n* = 7; Figure 2).

centration of 1 μ

g the antagonist of neuronal nicotinic ACh receptors (nNAChRs) DHβE

after pretreatment with DHβE, the inhibitory effect of nicotine on the quantal release of

of DHβE alone at a con-

treatment with the neuronal cholinergic receptor antagonist DHβE (1 μM; ), DHβE ( and DHβE way repeated ANOVA test with Tukey's post **Figure 2.** Nicotine inhibits the evoked release of ACh quanta (quantal content, QC) by activating nNAChRs. Panels on the top are representative traces of EPP and mEPP (50 signals averaged) in separate experiments with nicotine application (Nic, 10 µM; (**a**)), and nicotine application after pretreatment with the neuronal cholinergic receptor antagonist DHβE (1 µM; (**b**)). (**c**) Results are expressed as mean ± SEM and SD of QC, as percentages with nicotine (*n* = 7), DHβE (*n* = 9), and DHβE plus nicotine (*n* = 9) applications versus control. Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test with Tukey's post-hoc comparison).

Δ The nicotine-induced decrease in the number of ACh quanta released in response to stimulation of the motor nerve suggests the involvement of presynaptic cholinergic receptors. Using the antagonist of neuronal nicotinic ACh receptors (nNAChRs) DHβE [27,28], we obtained the data supporting this suggestion. The application of DHβE alone at a concentration of 1 µM did not change the QC of EPP (101.5 ± 1.3%; *n* = 9, Figure 2), however, after pretreatment with DHβE, the inhibitory effect of nicotine on the quantal release of ACh was completely abolished (105.5 ± 6.6%; *n* = 9; Figure 2).

#### *2.3. Activation of Neuronal Nicotinic Receptors Induces an Increase of the Calcium Level in the Motor Nerve Terminal*

Since the process of the evoked release of a neurotransmitter is triggered by the entry of calcium ions into the nerve ending [29,30], it was suggested that the inhibitory effect of nicotine on ACh release could be related to a decrease in Ca 2+ influx.

The amplitude changes of the optical signal (∆F/F0) from the calcium dye loaded into the nerve terminal in response to a single stimulus (with the same characteristics as during EPP registration) averaged about 30% (Figure 3). Nicotine application did not lead to a decrease, as expected, but instead caused a significant increase in the amplitude of the calcium transient by 13.7 ± 4.3% (*n* = 8; Figure 4). Thus, in the presence of a nicotinic receptor agonist, the presynaptic calcium level increases more strongly in response to nerve stimulation than in its absence. Is this increase indeed triggered by nNAChRs, whose activation leads to a decrease in the subsequent ACh release? The answer to this question was obtained in the experiments with an antagonist of nicotinic receptors, DHβE.

**Figure 3.** Pseudo-color calcium images of a motor nerve terminal loaded with Oregon Green 488 BAPTA-1 Hexapotassium Salt. The axon is imaged before and during single electrical stimulus (0.2 ms duration). Scale bar, 20 µm. , after pretreatment with DHβE, th

βE

βE

ntagonist DHβE (1 μM; ntagonist DHβE (1 μM; ), DHβE ( ) and DHβE test with Tukey's post **Figure 4.** Nicotine increases the calcium transient in the motor nerve ending by the activation of neuronal ACh receptors. Blockade of the receptors leads to a decrease in the amplitude of the calcium signal. Panels on the top are representative traces of calcium transient from separate experiments with nicotine application (Nic, 10 µM; (**a**)), and nicotine application after pretreatment with an nNAChR antagonist DHβE (1 µM; (**b**)). (**c**) Mean ± SEM and SD of the amplitude of the calcium signal, expressed as a percentage of control when applying nicotine (*n* = 8), DHβE (*n* = 15) and DHβE plus nicotine (*n* = 15). Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test with Tukey's post-hoc comparison).

), DHβE ( ) and DHβE test with Tukey's post Application of the antagonist alone led to a decrease in the calcium transient significantly by 12.9 ± 1.5% (*n* = 15; Figure 4). However, after pretreatment with DHβE, the calcium signal-enhancing effect of nicotine was completely abolished (100.0 ± 0.8%; *n* = 15; Figure 4).

#### *2.4. Neuronal Nicotinic Receptors Alter Calcium Level in Presynaptic Terminal by Gating L-Type (Cav1) Calcium Channels*

To identify the source of the increase in the calcium signal upon the activation of presynaptic nNAChRs, cadmium chloride, a nonselective blocker of calcium-permeable channels, was used at a concentration of 10 µM. After the application of cadmium chloride, a decrease in the amplitude of the calcium transient was observed by 54.5 ± 2.7% (*n* = 9). In the presence of cadmium, the effect of nicotine on the alterations in calcium levels was completely abolished (101.4 ± 3.4%; *n* = 9; Figure 5). Therefore, the observed increase in the presynaptic calcium level upon the activation of nNAChRs is mediated by proteins (channels) which are permeable for Ca 2+ . Further experiments were carried out to establish which type of VGCCs are involved in nicotine-induced increases in calcium transients.

**Figure 5.** The calcium transient-enhancing effect of nicotine is abolished a 2.1 VGCC blocker ω incubation with ω **Figure 5.** The calcium transient-enhancing effect of nicotine is abolished after a nonspecific calcium channel blockade, but not after inhibition of Cav2.1-type calcium channels. Panels on the top are representative traces of calcium transient from individual experiments: (**a**) the effect of the nonspecific calcium channel blocker CdCl<sup>2</sup> (Cd 2+ , 10 µM), (**b**) no effect of nicotine (Nic, 10 µM) after pretreatment with CdCl<sup>2</sup> , (**c**) the effect of the Cav2.1 VGCC blocker ω-agatoxin IVA (Aga, 40 nM), and (**d**) the effect of nicotine on the calcium transient after pre-incubation with ω-agatoxin IVA. (**e**) Mean ± SEM and SD of calcium signal amplitudes obtained in the above series and expressed as a percentage of control or value after CdCl<sup>2</sup> (*n* = 9), CdCl<sup>2</sup> + Nic (*n* = 9), Aga (*n* = 5;) and Aga + Nic (*n* = 7) application. Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test).

The application of the specific P/Q-type (Cav2.1) VGCCs blocker ω-agatoxin IVA at a concentration of 40 nM that blocks only a certain proportion of channels [31] led to a significant decrease in the calcium transient by 67.0 ± 4.4% (*n* = 5; Figure 5). In the case of a partial blockade of the main type of VGCCs Cav2.1, nicotine application (10 µM) led to an increase in the amplitude of the calcium transient significantly by 29.9 ± 3.8% (*n* = 10, Figure 5). Therefore, the effect of the activation of nNAChRs on the intracellular calcium level is not mediated by Cav2.1 channels.

2.1) VGCCs blocker ω

Cav1 calcium channel blockers such as verapamil (50 µM) and nitrendipine (25 µM), produced significant calcium transient decreases of 25.0 ± 4.4% (*n* = 9) and 18.8 ± 1.1% (*n* = 17), respectively (Figure 6). The application of nicotine after pretreatment by these blockers did not cause any changes in the calcium transient: the amplitudes were 101.8 ± 1.5% (*n* = 9) and 100.7 ± 1.2% (*n* = 17), respectively (Figure 6).

Lack of effect of nicotine (an increase in the amplitude of the calcium transient) and DHβE ) lack of DHβE (1 μM) effect after nitrendipine pre DHβE ( **Figure 6.** Lack of effect of nicotine (an increase in the amplitude of the calcium transient) and DHβE (a decrease in the amplitude of the calcium transient) after blockade of the Cav1 channels. Panels on the top are representative traces of calcium transient from individual experiments: (**a**) effect of Cav1 calcium channel blocker nitrendipine (Nitre, 25 µM), (**b**) lack of nicotine (Nic, 10 µM) effect after pre-application of nitrendipine, (**c**) lack of DHβE (1 µM) effect after nitrendipine pre-treatment, (**d**) effect of Cav1 type VGCCs blocker verapamil (50 µM), (**e**) no effect of nicotine after pre-application of verapamil. (**f**) Mean ± SEM and SD of calcium signal amplitudes obtained in the above series and expressed as a percentage of control or value after Nitre (*n* = 17), Nitre plus Nic (*n* = 17), Nitre plus DHβE (*n* = 7), verapamil (*n* = 9) and verapamil plus Nic (*n* = 9) application. Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test).

These data allow us to conclude that the observed increase in the calcium level in the nerve ending upon activation of nNAChRs by an exogenous agonist is due to mediation by Cav1 type VGCCs. Therefore, the phenomenon of the endogenous activation of presynaptic cholinergic receptors discovered by us should also be mediated by calcium channels of this type. Indeed, the calcium transient-reducing effect of DHβE, when Cav1 channels were blocked by nitrendipine, was completely abolished (100.0 ± 0.8%; *n* = 7; Figure 6).

Thus, the results obtained allow us to conclude that activation of nNAChRs leads to an additional increase in the entry of Ca 2+ into the nerve ending through VGCCs of the Cav1 type. Therefore, if this is the mechanism underlying the decrease in the quantal content upon activation of this type of cholinergic receptor, then it should be expected that the blockade of Cav1 type of calcium channels will eliminate the nicotine-induced decrease in the amount of released ACh quanta. Examining this assumption became the scope of the next step of the study.

reducing effect of DHβE, wh

#### *2.5. Nicotine-Induced Decrease in Acetylcholine Release Is Mediated by L-Type (Cav1) Calcium Channels and Not Coupled to Apamin-Sensitive KCa Channels*

To assess the possible role of Cav1 type of calcium channels in the nicotine-induced mechanism of ACh release autoinhibition, verapamil and nitrendipine were used at the same concentrations as in experiments with calcium transients.

Verapamil and nitrendipine application resulted in a significant decrease in the QC by 14.2 ± 3.2% (*n* = 6) and 11.2 ± 2.9% (*n* = 6), respectively (Figure 7). Nicotine application after pre-treatment with Cav1 channel blockers did not cause any changes in the evoked ACh release, and the QCs were 101.6 ± 1.9% (*n* = 6) and 103.8 ± 2.4% (*n* = 6), respectively (Figure 7).

**Figure 7.** Nicotine-induced decrease in the ACh release (quantal content, QC d ANOVA test with Tukey's post **Figure 7.** Nicotine-induced decrease in the ACh release (quantal content, QC) involves L-type Cav1 channels. Panels on the top are representative traces of EPP and mEPP (50 signals averaged) from individual experiments: (**a**) and (**b**) lack of nicotine (Nic, 10 µM) effect after pre-application with the Cav1 VGCC blockers nitrendipine (Nitre, 25 µM) and verapamil (50 µM); (**c**) mean ± SEM and SD of QC, expressed as a percentage of control or value after Nitre (*n* = 6), Nitre plus Nic (*n* = 6), Verapamil (*n* = 6) and Verapamil plus Nic (*n* = 6) application. Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test with Tukey's post-hoc comparison).

One of the possible calcium-activated targets that may be responsible for the nicotineinduced depression of ACh release is the apamin-sensitive KCa channel [21]. To examine the involvement of this channel into the nicotine-induced decrease of ACh release from the motor nerve terminal, experiments with KCa blocker apamin were conducted.

The application of this blocker to small-conductance Ca 2+ -activated K<sup>+</sup> channels alone at a concentration of 100 nM did not change the QC of EPP (95.9 ± 1.8%; *n* = 6, Figure 8).

In the presence of apamin the effect of nicotine on the QC of EPP remained unchanged (10.5 ± 1.4%; *n* = 6, Figure 8). Therefore, the effect of the activation of nNAChRs on the quantal release of ACh is mediated by Cav1 channels and not coupled to apamin-sensitive KCa channels.

nicotine (Nic, 10 μM) application after pretreatment repeated ANOVA test with Tukey's post **Figure 8.** Nicotine-induced decrease in ACh release (quantal content, QC) is observed in the presence of apamin (Apa), the blocker of small conductance Ca 2+ -activated K + channels. Panel on the top (**a**) is representative traces of EPP and mEPP (50 signals averaged) from individual experiments with apamin application (100 nM) and with nicotine (Nic, 10 µM) application after pretreatment with apamin; (**b**) Mean ± SEM and SD of QC, expressed as a percentage of control or value after apamin (*n* = 6), apamin + Nic (*n* = 6). Asterisks (\*) indicate significant effects (*p* < 0.05, one-way repeated ANOVA test with Tukey's post-hoc comparison).

#### **3. Discussion**

#### *3.1. Effects of Nicotine on ACh Release*

, nicotine at concentrations up to 1 μM has no effect aptic membrane. At the concentration of 5 μM, the presynap of presynaptic cholinergic receptors). An increase in concentration to 10 μM enhances the The results of our study demonstrate that in the mouse neuromuscular preparation of m.LAL, nicotine at concentrations up to 1 µM has no effect on either the processes of ACh release from the nerve terminal or on the processes of its interaction with the postsynaptic membrane. At the concentration of 5 µM, the presynaptic effect of nicotine appears to become detectable (inhibition of the spontaneous release of ACh due to the activation of presynaptic cholinergic receptors). An increase in concentration to 10 µM enhances the presynaptic effect of the alkaloid (suppression of not only spontaneous, but also of the evoked ACh release) and leads to a weak postsynaptic effect (decrease in RMP due to the activation of postsynaptic cholinergic receptors). With an increase in nicotine

concentration to 50 µM, dramatic changes become evident in all recorded parameters of neurotransmission.

Thus, nicotine at a concentration of 10 µM exerts both postsynaptic and presynaptic inhibitory effects on the neuromuscular synapse, causing a decrease in the number of ACh quanta released in response to action potential. A similar decrease in the QC during the activation of cholinergic receptors has been noted earlier [17,20,21,32], however, these results were obtained on other preparations, in conditions of initially reduced QC, or in cut fiber preparations.

#### *3.2. Presynaptic Cholinergic Receptors and the Role of Calcium Influx in the Mechanism of ACh Release Autoinhibition*

Since the inhibitory effect of nicotine on the QC was completely abolished by the application of the antagonist of neuronal cholinergic receptors DHβE, it was concluded that these receptors are involved in the cholinergic mechanism of regulation of ACh release. DHβE binds to the β2 subunits of neuronal receptors and is a selective antagonist for nonα7 nAChRs [33]. In the heteromeric receptors of ganglionic neurons the primary α subunit is α3, whereas in the rodent central nervous system the primary α subunit is α4 [34]. The α4β2 nNAChR is the most abundant subtype expressed in the brain, and studies have demonstrated that this receptor subtype is located presynaptically [35]. Therefore, we can assume that, in the neuromuscular synapse, neuronal cholinergic receptors have the α4β2 subunit composition. This assumption is supported by the data that at 1 µM DHβE, that we used in this study, the mouse α4β2 receptors are almost completely blocked, while the α3β4 subunit receptors remain essentially active [34].

In the next step of the mechanism of autoinhibition triggered by nNAChRs, it was necessary to answer the following key question: how is the activation of the presynaptic nicotinic receptors coupled to changes in the intracellular calcium level? Previous data [20,32,36] were indicating such a coupling, but there was no direct evidence found for this prior to our study. Using standard electrophysiological methods, combined with the fluorescent method for registration of calcium transients, which reflect changes in the calcium level within the presynaptic terminal upon action potential arrival, enabled us to obtain data on changes in the ACh release. Our results demonstrate that the activation of nNAChRs (sensitive to DHβE), leading to a decrease in ACh release, is accompanied by an increase in the level of calcium in the nerve terminal. Another important observation made was the significant effect of DHβE on the amplitude of the calcium transient when applied alone. This may indicate that there exists a background tonic activation of nNAChRs which results in a tonic increase in calcium entry into the nerve terminal. We suggest that the release of endogenous ACh during motor nerve stimulation and the spontaneous quantal release under physiological conditions results in the modulation of presynaptic Ca2+ entry and provides a physiologically important feedback [17]. In the absence of impulse activity, the largest amount of ACh is released from the nerve terminal through non-quantal release [37]. This ACh is able to tonic the activation of muscarinic cholinergic receptors at the neuromuscular synapse [38]. Therefore, it can be assumed that DHβEsensitive nicotinic cholinergic receptors can also be activated by a non-quantal ACh.

The complete absence of the effect of nicotine on the calcium signal after pre-treatment with cadmium (10 µM), which is a nonselective blocker of all types of calcium Ca<sup>v</sup> channels [39] allows two conclusions to be put forward: (i) the increase in the calcium level in the nerve terminal is mediated by transmembrane proteins permeable for Ca2+ from the environment; (ii) the observed entry of Ca2+ is mediated by channels other than those of nNAChRs. The last suggestion is very important, because it has been shown earlier that nNAChRs are more permeable to Ca2+ as compared to permeability of their muscle-type counterparts [40,41]. At the same time, it was shown that cadmium up to a concentration of 200 µM does not block currents through nNAChRs [42], but significantly blocks currents through VGCCs [43,44].

After the inactivation of VGCCs of Cav2.1 type, which are key to triggering the process of evoked ACh release [31,45,46] the effect of nicotine on calcium entry into the terminal was preserved, while after the blockade of Cav1 type channels, it was completely abolished. It should be noted that the possibility of the involvement of these channels in the evoked release of ACh quanta remained under debate until recently, as in [20,47–49] versus [23–26]. We have obtained clear evidence of the involvement of the Cav1 type of calcium channels in the regulation of the bulk calcium level in the terminal, and of the process of neurotransmission in the mammalian neuromuscular junction.

#### *3.3. Critical Issues in Establishing the Coupling between nNAChRs and L-Type (Cav1) Calcium Channels while Using a Pharmacological Approach*

The phenylalkylamine (verapamil), dihydropyridine (nitrendipine) and benzothiazepine classes of Cav1 type calcium channel blockers are capable of blocking nNAChRs [42,50]. The absence of the effects of nicotine (both on the calcium transient and on the QC) after the application of verapamil and nitrendipine may well be related to a simple direct blockade of nNAChRs (sensitive to DHβE and permeable to Ca2+). Indeed, DHβE, verapamil, and nitrendipine all lead to a decrease in the transients. At the same time, all three pharmacological agents abolish the effect of nicotine.

If even the direct blockade of nNAChRs by verapamil and nitrendipine does take place, a number of additional facts still point to the involvement of the Cav1 calcium channels in the mechanism of modulation of calcium entry and the process of ACh release in the nerve terminal. So, after the application of all three agents, the calcium entry decreases, however, the effect of verapamil and nitrendipine is almost two times more pronounced than that of DHβE. Furthermore, the QC does not change after the addition of DHβE, whereas in the presence of verapamil and nitrendipine it is decreased by more than 10%. In addition, the activation of cholinergic receptors (by nicotine) and presumed blockade of cholinergic receptors (by verapamil and nitrendipine) have not an opposite, but a unidirectional effect: a decrease in the QC. And, last but not least, the absence of the effect of nicotine on the calcium transient after the application of cadmium, which does not affect the functioning of nNAChRs (including α4β2 nNAChRs) or even potentiate them [42,51], indicates that in our case a pharmacological effect on two different targets took place.

#### *3.4. How the Activation of nNAChRs Modulates L-Type (Cav1) Channel Functioning*

nNAChRs have high Ca2+ permeability [41], which does not have a direct effect on the calcium transient amplitude. However, calcium entry through these receptors can lead to two potential outcomes and one can observe an increase in the amplitude of the calcium transient as a result of two mechanisms: (i) the CDF process (calcium dependent facilitation) of the Cav1 type channel is triggered [52] (Figure 9a); (ii) the CDI (calcium dependent inactivation) process of Cav1 type calcium channels, mediated by the interaction between CaM and Ca2+ channel, is disrupted by increasing CaMKII activity [53] (Figure 9b). However, according to [20], a decrease in ACh release caused by activation of nNAChRs is not associated with CaM.

The existence of a functional interaction between nNAChRs and the channels of the Cav1 type was established in a primary culture of neurons in the mouse cerebral cortex [54]. It should be noted that this work revealed the interaction of calcium channels with α4β2 nNAChRs. Potentially, a similar interaction takes place in the muscle-nerve junction. The authors believe that activation of presynaptic receptors leads to depolarization sufficient for opening of Cav1 calcium channels and entry of calcium into the neuron [54] (Figure 9c).

The question that arises is how enhanced Ca2+ entry via Cav1 channels can result in the downregulation of the ACh release. Firstly, the results obtained on the neuromuscular preparations suggest that the calcium channels of Cav1 type are located far from the active zones [55], an therefore that Ca2+ entering through them cannot directly interact with the exocytosis machine. Secondly, it was shown repeatedly that the elevation of the intra-terminal Ca2+ level could activate calcium-sensitive proteins participating in the downregulation of neurotransmitter release. Among them, the most probable candidates are calmodulin, calcineurin, and Ca2+-activated K<sup>+</sup> (KCa) channels [21,56,57]. However, according to our data, apamin-sensitive KCa channels are not involved into the nicotineinduced effects on the evoked ACh release.

**Figure 9.** Schematic drawing showing possible mechanisms of the coupling between nAChRs and L-type (Cav1) calcium channels in the motor nerve terminal. (**a**) Calcium dependent facilitation of the Cav1 type channel, mediated by CaM and triggered by calcium [52], which enters through nNAChRs; (**b**) calcium dependent inactivation process of Cav1 type calcium channels, mediated by the interaction between CaM and the Ca 2+ channel, is disrupted by an increase in CaMKII activity [53]; (**c**) opening of Cav1 calcium channels and entry of calcium into terminal caused by depolarization due to activation of nAChRs [54].

#### **4. Materials and Methods**

#### *4.1. Animals*

Mice BALB/C (20–23 g, 2–3 months old) of either sex were used in this study. All animal care and experimental protocols met the requirements of the European Communities Council Directive 86/609/EEC and was approved by the Ethical Committee of Kazan Medical University (No. 10; 20 December 2016). Animals were housed in groups of 10 animals divided by gender inside plastic cages with plenty of food (standard mice chow) and water ad libitum. The temperature (22 ◦C) of the room was kept constant and a 12 h light/dark cycle was imposed. Animal studies are reported in compliance with the ARRIVE guidelines [58] and with the recommendations made by the British Journal of Pharmacology [59].

#### *4.2. Tissue Preparations and Solutions*

Animals were euthanized by cervical dislocation in accordance with the approved project protocol. The *Levator auris longus* muscle (m. LAL) was quickly removed [60], then put into a Sylgard® chamber with bubbled (95% O<sup>2</sup> and 5% CO2) Ringer solution (pH 7.4) containing (in mM): NaCl 137, KCl 5, CaCl<sup>2</sup> 2, MgCl<sup>2</sup> 1, NaH2PO<sup>4</sup> 1, NaHCO<sup>3</sup> 11.9, and glucose 11. Bath solution temperature was controlled by a Peltier semiconductor device. Experiments were performed at 20.0 ± 0.3 ◦C. Muscle contractions were prevented using µ-conotoxin GIIIB in a 2 µM concentration [61].

#### *4.3. Electrophysiology*

We used a standard intracellular recording technique [14,62]. Microelectrodes were prepared from borosilicate glass (World Precision Instruments, Sarasota, FL, USA) using a P97 micropipette puller (Sutter Instrument, Novato, CA, USA). Recording electrodes were 20–30 MΩ and filled with 3 M KCl. To record evoked and spontaneous (miniature) endplate potentials (EPPs and mEPPs, respectively) an Axoclamp 900 A amplifier and a DigiData 1440 A digitizer (Axon Instruments, San Jose, CA, USA) were used. Membrane potential was at −60 to −80 mV and recorded by a miniDigi 1B (Axon Instruments, San Jose, CA, USA). Experiments with membrane potential deviations over 7 mV were declined. The nerve was stimulated with rectangular suprathreshold stimuli (0.2 ms duration, 0.5 Hz frequency) via a suction electrode connected to a model 2100 isolated pulse stimulator (A-M Systems, Sequim, WA, USA). All electronic devices were driven by pClamp v10.4 software (Molecular Devices, San Jose, CA, USA). Bandwidth was from 1 Hz to 10 kHz. After collecting 35 EPPs, mEPPs during 2 min were recorded. Quantal content was estimated as ratio of averaged EPPs to mEPPs amplitude.

#### *4.4. Calcium Transient Recording*

Nerve motor endings were loaded with Oregon Green 488 BAPTA-1 Hexapotassium Salt 1 mM high-affinity calcium-sensitive dye (Molecular Probes, Eugene, OR, USA) through the nerve stump, as described previously [44]. The fluorescence signal was recorded using an imaging setup based on an Olympus BX-51 microscope with a ×40 water-immersion objective (Olympus, Tokyo, Japan). Calcium transient registration was performed via a high-sensitivity RedShirtImaging NeuroCCD-smq camera (RedShirtImaging, Decatur, GA, USA), at 500 fps (exposure time 2 ms) and 80 × 80 pixels, which was sufficient for calcium transient registration with good temporal resolution. As the source of light, a Polychrome V (Till Photonics, Munich, Germany) was used with a set light wavelength of 488 nm. The following filter set was used to isolate the fluorescent signal: 505DCXT dichroic mirror, E520LP emission (Chroma, Bellows Falls, VT, USA). We used Turbo-SM software (RedShirtImaging, Decatur, GA, USA) for data recording. In each experiment, 8 fluorescence responses were recorded, then averaged. This was an optimal amount to obtain data of sufficient quality and to reduce excitotoxicity and photobleaching of the fluorophore.

To analyze the recorded images, ImageJ software (NIH, Bethesda, MD, USA) was used. We picked regions of interest in the motor nerve ending image and background manually. Subsequent data processing was performed in Excel (Microsoft, Redmond, WA, USA). Background values were averaged and subtracted from signal values. Data were represented as a ratio: (∆F/F<sup>0</sup> − 1) × 100 %, where ∆F is the fluorescence intensity during stimulation, and F<sup>0</sup> is the fluorescence intensity at rest [63].

#### *4.5. Materials*

Nicotine hydrogen tartrate salt (nicotine), nitrendipine, verapamil, ω-agatoxin IVA, cadmium chloride, and dimethylsulfoxide (DMSO) were obtained from Sigma Aldrich (St. Louis, MO, USA). Dihydro-β-erythroidine hydrobromide (DHβE), apamin (TOCRIS, Bristol, UK), and µ-conotoxin GIIIB were obtained from Peptide Institute Inc. (Osaka, Japan). Drugs were dissolved in distilled water with the exceptions of nitrendipine and verapamil, which were dissolved in DMSO. Further dilutions for all drugs were done in Ringer solution. In experiments with drugs which were dissolved in DMSO, the same concentration of DMSO was added to the control solution as was present in the solution with the agent. Finally, the DMSO concentration in the solution did not exceed 0.01%.

#### *4.6. Data and Statistical Analysis*

Data collection and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology [64]. The number of experiments in each experimental group was selected on the basis of observing a statistically significant effect while using the minimum number of animals (3R principles) and on experience from previous studies. Animals were randomly assigned to the different experimental groups, with each group having the same number of animals by design. Blinding of the operator was not feasible, but data analysis was performed semi-blinded by an independent analyst.

Statistical analysis was performed using Statistica 6.1 Base (Tulsa, OK, USA). A Shapiro–Wilk normality test was used to analyze the data distribution. Null-hypothesis testing was performed by ANOVA. One-way ANOVA followed by Dunnett's or Tukey's test for multiple comparison post hoc was used. For related groups, a one-way repeated ANOVA test was performed. Data are presented as mean ± SEM. Values of *p* < 0.05 were considered significant.

#### **5. Conclusions**

In the present study, we found that activation of presynaptic nNAChRs leads to a decrease in the quantal ACh release from the nerve ending. This negative feedback mechanism is mediated by the modulating of the function of VGCCs of the Cav1 type, which leads to an increase in the entry of Ca2+ into the nerve terminal (Figure 10). Understanding of the peculiarities of action of ACh (nicotine) on the nNAChR-containing nerve endings (not only in cholinergic synapses [51,65]) has a broad scientific and clinical significance, since cholinergic nicotinic signaling (in addition to neuromuscular transmission and synaptic transmission in ganglia) is involved in the setting of a variety of processes, including anxiety, depression, arousal, memory, and attention [66,67].

**Figure 10.** Working model of the mechanism of autoregulation of ACh release in the peripheral cholinergic synapse via nNAChRs. Activation of nNAChRs is accompanied by an increase in the entry of calcium ions into the motor nerve terminal through the L-type (Cav1) calcium channels. The latter are involved in both the process of evoked ACh release and its modulation.

**Author Contributions:** N.Z. and D.S. conceived and designed research; N.Z. and A.A. performed experiments; N.Z., A.A, A.M. and D.S. did the data analysis and interpretation. N.Z., D.S. and A.M. drafted the manuscript. All authors have read and agreed to the published version of the manuscript.

FRC Kazan Scientific Center of RAS АААА А18 **Funding:** This research was funded by RFBR #19-04-00490 and by the government assignment for FRC Kazan Scientific Center of RAS AAAA-A18-118022790083-9.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Local Ethics Committee of Kazan State Medical University (No. 10; 20 December 2016).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data that support the findings of this study are available from the corresponding author upon reasonable request.

**Acknowledgments:** The authors are thankful to Isabel Bermudez-Diaz, Ellya Bukharaeva and Alexey Petrov for consulting us on some issues. The authors would like to thank Victor I. Ilyin, for his critical reading of this manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

s DHβE β


#### **References**


## *Article* **Simultaneous Monitoring of pH and Chloride (Cl** <sup>−</sup>**) in Brain Slices of Transgenic Mice**

**Daria Ponomareva 1,2,3,† , Elena Petukhova 2,3,† and Piotr Bregestovski 1,2,3, \***

1


**Abstract:** Optosensorics is the direction of research possessing the possibility of non-invasive monitoring of the concentration of intracellular ions or activity of intracellular components using specific biosensors. In recent years, genetically encoded proteins have been used as effective optosensory means. These probes possess fluorophore groups capable of changing fluorescence when interacting with certain ions or molecules. For monitoring of intracellular concentrations of chloride ([Cl −]i ) and hydrogen ([H<sup>+</sup> ] i ) the construct, called ClopHensor, which consists of a H<sup>+</sup> - and Cl <sup>−</sup>-sensitive variant of the enhanced green fluorescent protein (E <sup>2</sup>GFP) fused with a monomeric red fluorescent protein (mDsRed) has been proposed. We recently developed a line of transgenic mice expressing ClopHensor in neurons and obtained the map of its expression in different areas of the brain. The purpose of this study was to examine the effectiveness of transgenic mice expressing ClopHensor for estimation of [H<sup>+</sup> ]<sup>i</sup> and [Cl <sup>−</sup>]<sup>i</sup> concentrations in neurons of brain slices. We performed simultaneous monitoring of [H<sup>+</sup> ]<sup>i</sup> and [Cl <sup>−</sup>]<sup>i</sup> under different experimental conditions including changing of external concentrations of ions (Ca 2+ , Cl <sup>−</sup>, K + , Na + ) and synaptic stimulation of Shaffer's collaterals of hippocampal slices. The results obtained illuminate different pathways of regulation of Cl <sup>−</sup> and pH equilibrium in neurons and demonstrate that transgenic mice expressing ClopHensor represent a reliable tool for non-invasive simultaneous monitoring of intracellular Cl <sup>−</sup> and pH.

**Keywords:** genetically encoded biosensors; optopharmacology; transgenic mice; intracellular pH; intracellular chloride; brain slices; pH and Cl <sup>−</sup> transporters

#### **1. Introduction**

Neuronal activity is accompanied by dynamical changes in the intra- and extracellular concentration of ions. A number of studies demonstrated that activation or inhibition of neurons can cause a rapid shift of hydrogen (H<sup>+</sup> ) and chloride (Cl <sup>−</sup>) [1–3]. The physiological intracellular pH range in different eukaryotic cells is 6.5–8.0 [4], which corresponds to a very low free H<sup>+</sup> concentration, from 10 nM to 300 nM. In the mammalian brain the intracellular pH is 7.0–7.4 [5,6] reflecting the importance of maintaining the intracellular H<sup>+</sup> concentration ([H<sup>+</sup> ]i ) in a narrow range.

Intracellular Cl <sup>−</sup> concentration ([Cl −]i ) in different cell types varies from 3 mM to 60 mM, being around 5–10 mM in the majority of mammalian neurons [7]. Deviations from this physiological range can alter the excitability of cells, modulate the function of a variety of proteins including ion channels [8–10]. Abnormal changes in concentrations of these ions are associated with the development of pathological processes and some disorders including neurodegeneration, epilepsy and brain ischemia [11–13].

To maintain H<sup>+</sup> and Cl <sup>−</sup> in physiological ranges, various transporters, cotransporters, and other ion regulating proteins are expressed in cells of biological organisms. The

**Citation:** Ponomareva, D.; Petukhova, E.; Bregestovski, P. Simultaneous Monitoring of pH and Chloride (Cl−) in Brain Slices of Transgenic Mice. *Int. J. Mol. Sci.* **2021**, *22*, 13601. https://doi.org/10.3390/ ijms222413601

Academic Editor: Carlo Matera

Received: 28 November 2021 Accepted: 14 December 2021 Published: 18 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

regulation of intracellular [H<sup>+</sup> ]i is primarily driven by Na+/H<sup>+</sup> exchange, Na<sup>+</sup> -driven Cl−/HCO<sup>3</sup> <sup>−</sup> exchange, Na+/HCO<sup>3</sup> <sup>−</sup> cotransport, and Cl−/HCO<sup>3</sup> <sup>−</sup> exchange [1,14–16].

The level of [Cl−]<sup>i</sup> is maintained predominantly by a K+/Cl<sup>−</sup> cotransporter (KCC2), which pumps out Cl<sup>−</sup> from the cells, a Na+/K+/Cl<sup>−</sup> cotransporter NKCC1, which loads Cl<sup>−</sup> into the cell, and a Cl−/HCO<sup>3</sup> <sup>−</sup> exchanger [3]. Activation of the Cl−/HCO<sup>3</sup> <sup>−</sup> exchanger leads to local simultaneous pH and Cl<sup>−</sup> changes. These interrelated relationships highlight the necessity of simultaneous monitoring of changes in Cl<sup>−</sup> and H<sup>+</sup> concentrations.

To measure pH and Cl<sup>−</sup> in cells of biological organisms, several methods have been proposed. The most used are ion-selective microelectrodes [17–19], fluorescent dyes [20–22] and genetically encoded probes [7,23–26]. Especially promising are genetically encoded fluorescent sensors, which allow non-invasive monitoring of intracellular ion concentrations in specific cell types.

For simultaneous registration of [H<sup>+</sup> ]<sup>i</sup> and [Cl−]<sup>i</sup> , the construct ClopHensor has been developed, which consists of a H<sup>+</sup> - and Cl<sup>−</sup> -sensitive variant of the enhanced green fluorescent protein (E2GFP) fused with a monomeric red fluorescent protein (mDsRed) [27]. The sensor was studied at the heterologous expression in different cells. It has been shown that the construct possesses a pKa = 6.8 for H<sup>+</sup> and K<sup>d</sup> for Cl<sup>−</sup> in the range 40–50 mM at physiological pH ~ 7.3 [24,27,28]. Improved variants of ClopHensor have been developed and tested in cultured brain slices using biolistical transfection [29]. Another biosensor optimized for the simultaneous [Cl−]<sup>i</sup> and pH<sup>i</sup> imaging, called LSSmClopHensor, has been developed [30]. It was expressed in the cortex of rats using in utero electroporation and used for analysis in brain slices and in vivo [30,31].

Recently, we presented a line of transgenic mice expressing ClopHensor in neurons due to the neuronal-specific promoter Thy1 [32]. The pattern of ClopHensor expression across the brain has been revealed using the CLARITY method in combination with confocal and light-sheet microscopy. It showed a robust expression of ClopHensor in the hippocampal formation, thalamus, ponds, medulla, cerebellum and other areas [32].

The present study is devoted to the analysis of H<sup>+</sup> - and Cl−-transients at synaptic stimulation of neurons in hippocampal slices of Thy1: ClopHensor mice. We analyzed changes in both ion concentrations at different approaches including inhibition of GABA-ergic synaptic transmission, extracellular Ca2+-free and low Cl<sup>−</sup> or Na<sup>+</sup> conditions. Our observations demonstrate the efficiency of transgenic mice expressing ChlopHensor for reliable non-invasive monitoring of intracellular Cl<sup>−</sup> and pH in normal and pathological conditions.

#### **2. Results**

## *2.1. Simultaneous Monitoring of pH and Cl*<sup>−</sup> *Using ClopHensor*

Experiments were performed on brain slices of transgenic mice expressing ClopHensor using the mouse neuronal promoter Thy1. ClopHensor consists of a modified enhanced green fluorescent protein E2GFP connected with a monomeric DsRed (mDsRed) via a 20-amino-acid flexible linker (Figure 1A). E2GFP carries a specific anion-binding site engineered by the single substitution T203Y and elevation of Cl<sup>−</sup> causes static quenching of E2GFP's fluorescence [33]. Like all green fluorescent proteins, E2GFP is also sensitive to pH: its emission decreases when it is acidified. However, E2GFP's emission intensity does not depend on pH when excited at its isosbestic point of 458 nm, allowing one to perform ratiometric analysis. In addition, at wavelengths above 540 nm, signals of both fluorescent proteins, E2GFP and mDsRed, are pH and Cl−-independent (Figure 1B). This allows the simultaneous ratiometric analysis of changes in [H<sup>+</sup> ]<sup>i</sup> and [Cl−]<sup>i</sup> at monitoring upon excitation at three wavelengths: 488 nm: pH- and Cl−-dependent E2GFP signal, 458 nm: pH-independent E2GFP signal, and above 543 nm: Cl−- and pH-independent mDsRed signal [27].

= ΔF / ΔF

= ΔF / ΔF

− <sup>−</sup> – **Figure 1.** Monitoring pH and Cl <sup>−</sup> in brain slices of mice expressing ClopHensor. (**A**) Schematic representation of ClopHensor. (**B**) Excitation spectra of ClopHensor collected at different pH values (5.9, 6.9, 7.4, and 8.4) in the absence of Cl <sup>−</sup> (left) and at increasing Cl <sup>−</sup> concentration (0–1 M) and constant pH = 6.9 (right) (modified from Arosio et al., 2010 [27]). (**C**) Micrographs of pyramidal cells of CA1 hippocampal region, expressing ClopHensor, under illumination with excitation light with wavelengths of 455 nm, 505 nm and 590 nm (age P7).

In the present study, we recorded fluorescence at slightly different excitation wavelengths (455 nm, 505 nm and 590 nm) and estimated changes in concentrations of ions using the following ratios:

$$\mathbf{R\_{pH}} = \Delta \mathbf{F\_{505nm}} / \Delta \mathbf{F\_{455nm}} \tag{1}$$

$$\mathbf{R\_{C1}} = \Delta \mathbf{F\_{590nm}} / \Delta \mathbf{F\_{455nm}} \tag{2}$$

Examples of fluorescence at different excitation wavelengths are shown in Figure 1C. Details of calibration are presented in Section 4.

## *2.2. Effect of Bicuculline on Synaptically Induced Cl*<sup>−</sup> *and pH-Transients in Hippocampal Neurons*

To examine the properties of intracellular Cl<sup>−</sup> and H<sup>+</sup> transients during synaptic activation, we performed experiments on hippocampal slices of transgenic mice expressing ClopHensor. The first task was to establish how the changes in intracellular concentrations of Cl<sup>−</sup> and H<sup>+</sup> , evoked by electrical stimulation, are amenable to pharmacological modulation of synaptic transmission. Fluorescent emission signals were recorded from the hippocampal CA1 area upon sustained tetanic stimulation of Schaffer's collaterals (100 Hz for 20 s). Stimulus intensities ranged from 20 µA to 320 µA. The experiments were carried out on adult transgenic mice aged 2–8 months.

High-frequency stimulation of Schaffer's collaterals caused strong changes of fluorescence signals excited at 455 nm and 505 nm without effect on excitation at 590 nm (not shown). Ratiometric analysis revealed that these changes correspond to an elevation of both [Cl−]<sup>i</sup> and [H<sup>+</sup> ]i . In this set of experiments, in hippocampal CA1 neurons, the base level of [Cl−]<sup>i</sup> was 8.0 ± 1.1 mM (*n* = 5) and the mean intracellular pH was 7.30 ± 0.02 (*n* = 11). High frequency synaptic stimulation increased the [Cl−]<sup>i</sup> to 0.8 ± 0.3 mM (*n* = 5) and acidification of neurons to 0.024 ± 0.006 units pH (*n* = 11).

In order to test the assumption that during synaptic stimulation, the accumulation of Cl<sup>−</sup> in the hippocampal neurons is, at least partially, due to Cl<sup>−</sup> influx via activated GABAA receptors, we applied bicuculline, an antagonist of these receptors [34]. The traces shown in Figure 2A illustrate that upon the synaptic stimulation, bicuculline strongly decreased the amplitude of synaptically induced changes of [Cl−]<sup>i</sup> . On average, the mean amplitude of Cl<sup>−</sup> influx decreased to 41.2 ± 12.2% (Figure 2B, *n* = 7, *p* < 0.01). In contrast, upon application of bicuculline the amplitude of synaptically induced pH transients increased by 90.7 ± 17.4% (Figure 2C, *n* = 8, *p* < 0.01). Picrotoxin, a blocker of Cl−-selective GABA receptor channels, caused similar changes in the amplitudes of Cl<sup>−</sup> and pH transients (data not shown).

The augmentation of pH transients can be, at least partially, explained by the fact that bicuculline blocks inhibitory transmission, which leads to an increase in the general excitability [35]. To check the involvement of this mechanism in the potentiation of synaptically induced H<sup>+</sup> responses upon inhibition of GABA receptors, we analyzed the effect of bicuculline on the amplitude of evoked local field potentials (eLFP). After stimulation of Schaffer's collaterals, a population of CA1 neurons are activated simultaneously and fire an action potential in synchrony, giving rise to a single eLFP. As illustrated in Figure 2D, the addition of bicuculline (40 µM) led to an elevation of eLFPs. The mean potentiation was 24 ± 3.8% (Figure 2E). Consequently, in the presence of bicuculline, the effectiveness of the stimulation increased, causing stronger depolarization and augmentation of pH transients. A competitive AMPA/kainate receptor antagonist (CNQX, 40 µm) completely abolished stimulation-induced changes of ratiometric emission signals, confirming that they have emerged from synaptic activation (not shown).

Altogether, these data demonstrate that: (i) bicuculline modulates in opposite ways the synaptically evoked Cl−- and pH-transients in neurons expressing ClopHensor; (ii) bicuculline only partially inhibits synaptically induced elevation of [Cl−]<sup>i</sup> ; (iii) inhibition of GABA receptor by bicuculline causes potentiation of eLFPs.

**Figure 2.** Effect of bicuculline on Cl−- and pH-transients induced by synaptic stimulation of hippocampal CA1 neurons in brain slices of transgenic ClopHensor mice. (**A**) Traces of Cl−- (top) and pH-specific emission signals (bottom) in control (black) and the presence of 40 µM bicuculline (green). Arrows indicate stimulation (100 µA, 100 Hz, 20 s). (**B**) Mean inhibition by bicuculline (40 µM) of Cl<sup>−</sup> transients induced by synaptic stimulation of CA1 hippocampal neurons (mean percentage ± SEM, *n* = 7, 2–8 months). \*\* Significant difference with *p* < 0.01 (Paired Sample Wilcoxon Signed Rank Test). (**C**) Mean potentiation by bicuculline (40 µM) of stimulation-induced pH-transients. Summary data from 8 slices (mean percentage ± SEM, *n* = 8, 2–8 months). \*\* Significant difference with *p* < 0.01 (Paired Sample Wilcoxon Signed Rank Test). Age of mice 2–8 months. (**D**) Superimposed traces of evoked local field potentials (eLFP) induced by single stimulation of Schaffer's collaterals in control (black) and the presence of 40 µM bicuculline (green). Presented averaged traces of 10 individual eLFPs (stimulation: 20 µA, single pulse width 200 µs; age 7 months). (**E**) Summary of the eLFP amplitude potentiation by 40 µM bicuculline. 100% is the amplitude of eLFPs in control condition. Values are mean ± SEM (*n* = 7). \*\* Significant difference with *p* < 0.01 (Paired Sample Wilcoxon Signed Rank Test). Age of mice 2–8 months.

#### *2.3. Effect of Bicuculline on Intracellular pH Changes after Tetanic Stimulation, Assessed Using BCECF-AM*

To ensure the specificity of the pH changes when using ClopHsensor, we performed a series of experiments with the pH-sensitive dye BCECF-AM [36]. This compound is well tolerated by cells and offers ratiometric estimation of pH values.

We determined the pH changes by calculating the ratio of two emission signals obtained after illumination by light at 455 nm and 505 nm. The intracellular pH changes after tetanic stimulation and its modulation by bicuculline were analyzed. Experiments were conducted on wild type juvenile mice (P10–P12). Registration conditions were similar to those performed on slices from transgenic ClopHensor mice. Examples of

4

CA1 hippocampal neurons loaded with BCECF and excited at different wavelengths are presented in Figure 3A. Evoked pH-specific ratiometric emission signals were obtained by giving the same pattern of electrical stimulation (100 Hz for 20s); the stimulus intensities ranged from 200 µA to 250 µA.

3 **Figure 3.** The effect of bicuculline on the evoked pH signals recorded using BCECF-AM. (**A**) Micro-graphs of pyramidal cells of the CA1 hippocampal region, loaded by BCECF-AM, under illumination with excitation wavelengths of 505 nm and 455 nm (P12). (**B**) Representative traces of pH-specific ratiometric emission transients evoked by high-frequency electrical stimulation (200 µA, 100 Hz for 20 s) in control (left trace, black) and after addition of 10 µM bicuculline (right trace, green). The arrows indicate the moments of stimulation (P11). (**C**) Summary of bicuculline action on the amplitude of pH changes induced by synaptic stimulation. Normalization to the control values (mean ± SEM, P10–12, *n =* 5). \* Significant difference with *p* < 0.05 (Paired Sample Wilcoxon Signed Rank Test).

We then performed a comparative analysis of the kinetics of synaptically induced Cl<sup>−</sup> and pH transients. It was evaluated by determining the decay time constants, i.e., the time during which the peak amplitude of Cl<sup>−</sup> and H<sup>+</sup> components decreases by e-times (τdecay).

2

3

that of pH signals.

4 **Figure 4.** Comparison of the decay kinetics of stimulation-induced Cl<sup>−</sup> and pH transients in hippocampal CA1 neurons. (**A**) Representative traces of normalized Cl−-(red trace) and pH-(black trace) transients registered using the transgenic ClopHensor mouse (stimulation 100 Hz for 20 s; age 6 months), and pH-specific signals using BCECF-AM (blue trace; stimulation: 100 Hz for 20 s; age P11). (**B**) The mean time constants of decay of the Cl<sup>−</sup> and pH stimulation-induced responses were recorded with ClopHensor (age 2–7 months) and BCECF (age P10-12). Data from 9–16 slices. Values are mean ± SEM. \*\*\* Significant difference with *p* < 0.001 (Mann Whitney U test).

2 These results suggest that the mechanisms involved in the withdrawing of ions from neurons operate much faster for Cl<sup>−</sup> than for H<sup>+</sup> and different mechanisms are involved in the maintenance of normal physiological transmembrane concentrations of these ions.

Next, we analyzed how changes of physiologically important ions, involved in regulation of Cl<sup>−</sup> and H<sup>+</sup> transporters, modulate the amplitude of Cl<sup>−</sup> and pH transients induced by depolarization.

#### *2.5. Changes of [H<sup>+</sup> ]i - and [Cl*−*]<sup>i</sup> -Induced by High [K<sup>+</sup> ]o-Depolarization*

Several studies used [K<sup>+</sup> ]o-induced depolarization as a tool for stimulation of pH changes in different cell types [37–40]. We used this approach to analyze the properties of pH<sup>i</sup> and [Cl−]<sup>i</sup> transients in neurons of hippocampal slices under different experimental conditions.

In all experiments, the high [K<sup>+</sup> ]<sup>o</sup> solution contained 20 mM, i.e., 17.5 mM of potassium gluconate was added to normal ASCF containing 2.5 mM K<sup>+</sup> . Depolarization induced by application of high [K<sup>+</sup> ]<sup>o</sup> ASCF caused a remarkable elevation of [Cl−]<sup>i</sup> in all slices, with an average of 11.4 ± 1.1 mM (*n* = 11). Concerning pH, two types of responses were observed. In the majority of slices, [K<sup>+</sup> ]o-induced depolarization caused acidification of neurons by, on average, 0.15 ± 0.02 pH units (*n* = 23). In 8 cases, biphasic pH<sup>i</sup> -shifts composed of an early short-lasting alkalinisation that turned into a longer-lasting acidification were observed. We have not analyzed yet the reasons for this difference in the pH responses.

#### *2.6. [K<sup>+</sup> ]o-Induced Changes of pH and Cl*<sup>−</sup> *in a Ca2+-Free ACSF*

Observations of cells in culture conditions demonstrated that elevation of external K<sup>+</sup> causes strong depolarization of neurons, accompanied, in addition to acidification, by a rise in intracellular Ca2+ [38,39]. In these studies, acidic [K<sup>+</sup> ]o-induced pH responses were either completely prevented by the use of a Ca2+-free medium [38], or even inversed

and became alkalizing [39], suggesting a key role of Ca2+ ions in depolarization-induced pH changes.

We performed an analysis of the removal of external Ca2+ on depolarization-induced transients of [Cl<sup>−</sup> ]<sup>i</sup> and pH<sup>i</sup> in brain slices of ClopHensor mice. Under control conditions, when high K<sup>+</sup> was applied to slices for 3–6 min, an elevation in [Cl−]<sup>i</sup> and decrease in pH were observed (Figure 5A). In this set of experiments, [K<sup>+</sup> ]o-induced intracellular acidification occurred, on average by 0.12 ± 0.11 pH (*n* = 4, Figure 5C), and an increase in [Cl−]<sup>i</sup> occurred by 10.12 ± 1.21 mM (*n* = 4, Figure 5B). After washing, [Cl−]<sup>i</sup> recovered toward the baseline, while the base level of pH<sup>i</sup> sometimes followed to more alkaline values (Figure 5A).

Applying a Ca2+-free ACSF containing 1 mM EGTA did not change the base level of [Cl−]<sup>i</sup> , while a decrease in pH, i.e., acidification of the cell cytoplasm was observed (Figure 5A). On average, the base level of pH<sup>i</sup> decreased by 0.09 ± 0.02 (Figure 5D, *n* = 8, *p* < 0.01). [K<sup>+</sup> ]o-induced depolarization caused an elevation of [Cl−]<sup>i</sup> by 12.75 ± 0.83 mM (*n* = 4, Figure 5B). Surprisingly, unlike reports on preparations in culture [38,39], in our conditions, the pH responses to high K<sup>+</sup> were acidifying, only with reduced amplitudes compared to the control. On average, the [K<sup>+</sup> ]o-induced pH decrease was by 0.10 ± 0.08 pH (*n* = 4, Figure 5C).

## *2.7. The Effect of Extracellular Cl*<sup>−</sup> *on Synaptically-Induced Changes of [Cl*−*]<sup>i</sup> and pH<sup>i</sup>*

Our next task was to determine the contribution of Cl<sup>−</sup> ions to the [Cl−]<sup>i</sup> and pH<sup>i</sup> transients. For this, we analyzed the effects of decreasing Cl<sup>−</sup> concentration in ACSF from 134.7 mM to 7.2 mM ("low Cl−" conditions). The changes in base levels of pH<sup>i</sup> and [Cl−]<sup>i</sup> , as well as the amplitudes of Cl<sup>−</sup> and H<sup>+</sup> transients evoked by a high-frequency stimulation, were determined. On average, the mean amplitude of Cl<sup>−</sup> influx on highfrequency stimulation was 0.8 ± 0.3 mM (Figure 6A,B). At the same time, stimulation caused acidification by 0.012 ± 0.005 pH (Figure 6A,C). After adding the "low-Cl−" ACSF, a diminishing base level of [Cl−]<sup>i</sup> , synchronously with alkalinisation of neurons, was observed. New levels of [Cl−]<sup>i</sup> and pH<sup>i</sup> were completely stabilized in about 15 min (Figure 6A). On average, [Cl−]<sup>i</sup> diminished by 7.1 ± 1.1 mM (*p* < 0.05, *n* = 5) and <sup>i</sup> increased by 0.19 ± 0.01 units (*p* < 0.05, *n* = 5; Figure 6D,E).

During the first minutes after establishing new steady-state levels in "low-Cl−" external solution, high-frequency stimulation of Schaffer's collaterals caused the acidification directed pH responses, in which the amplitude was much higher than in control ACSF (Figure 6A). On average, ∆pH<sup>i</sup> increased by 5 times in comparison with that recorded in the normal ACSF and became 0.063 ± 0.013 units pH (*p* < 0.05, *n* = 5, Figure 6C). The Cl<sup>−</sup> transients on stimulation were nearly completely suppressed (Figure 6A,B). With an increase in the duration of incubation of slices in the "low-Cl−" ACSF, the amplitude of the evoked pH responses decreased until they completely disappeared (data not shown). Our preliminary observations suggest that the reason for this elimination of pH responses is a continuous strong decrease in the amplitude of evoked local field potentials upon transition to "low-Cl−" ACSF conditions.

–

−

0.02 рН (Figure 5D,

−

−

−

− − о − **Figure 5.** Effect of external Ca 2+ removal on depolarization-induced changes in [Cl <sup>−</sup>]<sup>i</sup> and pH<sup>i</sup> in neurons of hippocampal slices. (A) Typical changes in [Cl −]i (top) and pH<sup>i</sup> (bottom) evoked by the application of 17.5 mM K + in the presence of 2.3 mM external Ca 2+ (Control) and in Ca 2+ -free ASCF. The periods of K <sup>+</sup> application and changes in external Ca 2+ are indicated by the bars above the traces. Note the decrease of pH upon elimination of external Ca 2+ . (**B,C**) The mean [K<sup>+</sup> ]o-induced changes of [Cl − i (**B**) and pH<sup>i</sup> (**C**) in control and in Ca 2+ -free ASCF. Summary of data from 4 slices (mean percentage ± SEM, age 4 months). (**D**) Mean values of base level of pH<sup>i</sup> in the control condition (2.3 mM of external Ca 2+ ) and in Ca2+-free ASCF. \* Significant difference with *p* < 0.05.

−

Figure 6. Effect of reduction of extracellular Cl and high-frequency stimulation on [Cl ]i and pHi in the hippocampal CA1 neurons of ClopHensor mice. (A) Typical traces of continuous recording of [Cl ]i (top) and pHi (bottom) illustrating the effects of the transition to low Cl ASCF and responses on high-frequency stimulation (age 6 months). The low Cl ACSF application is highlighted by yellow. Arrows indicate moments of stimulation. (B,C) Mean changes of synaptically induced [Cl ]i (B) and pHi (C) in conditions of high and low external Cl (mean ± SEM, n = 5, age 6 months). (D,E) Summary of the effect of reduction of external Cl concentration on base levels of [Cl ]i (D) and pHi (E) (mean percentage ± SEM, n = 5, 6 months). \* Significant difference with p < 0.05 (Paired Sample Wilcoxon Signed Rank Test). **Figure 6.** Effect of reduction of extracellular Cl<sup>−</sup> and high-frequency stimulation on [Cl−]<sup>i</sup> and pH<sup>i</sup> in the hippocampal CA1 neurons of ClopHensor mice. (**A**) Typical traces of continuous recording of [Cl−]<sup>i</sup> (top) and pH<sup>i</sup> (bottom) illustrating the effects of the transition to "low Cl−" ASCF and responses on high-frequency stimulation (age 6 months). The "low Cl−" ACSF application is highlighted by yellow. Arrows indicate moments of stimulation. (**B**,**C**) Mean changes of synaptically induced [Cl−]<sup>i</sup> (**B**) and pH<sup>i</sup> (**C**) in conditions of high and low external Cl<sup>−</sup> (mean ± SEM, *n* = 5, age 6 months). (**D**,**E**) Summary of the effect of reduction of external Cl<sup>−</sup> concentration on base levels of [Cl−]<sup>i</sup> (**D**) and pH<sup>i</sup> (**E**) (mean percentage ± SEM, *n* = 5, 6 months). \* Significant difference with *p* < 0.05 (Paired Sample Wilcoxon Signed Rank Test).

#### 2.8. Changes of [Cl ]i and pHi during [K<sup>+</sup> ] -Induced Depolarization in the Low-Cl ACSF *2.8. Changes of [Cl*−*]<sup>i</sup> and pH<sup>i</sup> during [K<sup>+</sup> ]o-Induced Depolarization in the Low-Cl*<sup>−</sup> *ACSF*

We then tested how a decrease in extracellular Cl affects the amplitude of [Cl ]i and pHi changes caused by the [K<sup>+</sup> ] -induced depolarization. Similarly to previously described observations (Section 2.5), the addition of 17.5 mM K<sup>+</sup> in normal ACSF induced reversible transients of acidification and synchronous elevation of [Cl ]i in neurons (Figure 7A). In this set of experiments, the mean K<sup>+</sup> -induced decrease in pHi was 0.15 ± 0.02 pH and the We then tested how a decrease in extracellular Cl<sup>−</sup> affects the amplitude of [Cl−]<sup>i</sup> and pH<sup>i</sup> changes caused by the [K<sup>+</sup> ]o-induced depolarization. Similarly to previously described observations (Section 2.5), the addition of 17.5 mM K<sup>+</sup> in normal ACSF induced reversible transients of acidification and synchronous elevation of [Cl−]<sup>i</sup> in neurons (Figure 7A). In this set of experiments, the mean K<sup>+</sup> -induced decrease in pH<sup>i</sup> was 0.15 ± 0.02 pH and the elevation of [Cl−]<sup>i</sup> was 4.36 ± 0.77 mM (*n* = 5).

elevation of [Cl ]i was 4.36 ± 0.77 mM (n = 5). As previously described, upon perfusion of slices with "low Cl−" ACSF, the base level of pH increased and [Cl−]<sup>i</sup> synchronously decreased (Figure 7A). In "low Cl−" ACSF conditions, depolarization caused by the addition of 17.5 mM K<sup>+</sup> , resulted in a significant increase in the amplitude of pH acidification responses, while [Cl−]<sup>i</sup> responses were nearly completely suppressed or even oppositely directed, i.e., showed a small efflux of intracellular Cl<sup>−</sup> (Figure 7A). On average, the mean amplitude of "high K+"-induced pH<sup>i</sup> responses was 0.33 ± 0.06 (*p* < 0.05, *n* = 5), i.e., about 2 times higher than in the control ACSF, while the mean change of [Cl−]<sup>i</sup> was −0.14 ± 0.03 (*p* < 0.05, *n* = 5), i.e., about 30 times smaller in comparison with control conditions. After washing with normal ACSF, the base levels of pH<sup>i</sup> and [Cl <sup>−</sup>]<sup>i</sup> returned to initial values; in addition, the amplitudes of K + -induced transients were fully recovered (Figure 7B,C).

These results are in accord with the above-described effect of lowering external Cl − on synaptically induced changes of pH<sup>i</sup> and [Cl −]i .

− − − ) illustrating the effects of the transition to "low Cl " ASCF and reto hippocampal slices (age 7 months). The "low Cl <sup>−</sup>" ACSF application is highlighted − **Figure 7.** Effect of extracellular Cl <sup>−</sup> on changes of [Cl <sup>−</sup>]<sup>i</sup> and pH<sup>i</sup> caused by K + -induced depolarization. (**A**) Typical traces of continuous recording of [Cl −]i (top) and pH<sup>i</sup> (bottom) illustrating the effects of the transition to "low Cl <sup>−</sup>" ASCF and responses to the addition of 17.5 mM K<sup>+</sup> to hippocampal slices (age 7 months). The "low Cl <sup>−</sup>" ACSF application is highlighted by the green bar. (**B**,**C**) Summary of the effect of K+-induced depolarization on changes of [Cl −]i (**B**) and pH<sup>i</sup> (**C**). Data from 5 slices (mean ± SEM, *n* = 5, 7 months). \* Significant difference with *p* < 0.05 (Paired Sample Wilcoxon Signed Rank Test).

#### <sup>−</sup> " −" *2.9. Changes in [Cl* <sup>−</sup>*]<sup>i</sup> and [H<sup>+</sup> ]<sup>i</sup> during [K + ]o-Induced Depolarization in the Low-Na <sup>+</sup> ACSF*

"

−"

− −

−

−

− <sup>−</sup> " " <sup>−</sup> − Extracellular Na + is one of the important participants and regulators of transporters of intracellular proton concentration [1]. For instance, the electrogenic sodium bicarbonate cotransporter NBCe1 (SLC4A4) contributes to intracellular as well as extracellular acid/base homeostasis in the brain and its dysfunctions are associated with pathophysiological states [16,41,42]. In addition, the Na + -dependent 2HCO<sup>3</sup> <sup>−</sup>/Cl <sup>−</sup>-exchanger (NDCBE) represents an important pH- Cl <sup>−</sup>-coupled physiological controller of ions [15].

− − Lastly, we analyzed how a decrease in external Na <sup>+</sup> affects changes in pH<sup>i</sup> and [Cl <sup>−</sup>]<sup>i</sup> caused by high K<sup>+</sup> depolarization. Similar to the above-presented results, in ASCF containing normal Na + concentration, K<sup>+</sup> -induced depolarization caused intracellular elevation of both H<sup>+</sup> and Cl <sup>−</sup> ions (Figure 8A). On average, the increase in [Cl <sup>−</sup>]<sup>i</sup> was 12.06 ± 1.19 mM (Figure 8B) and acidification was 0.18 ± 0.04 pH (Figure 8C).

*− о* Changing the solution to a low-Na <sup>+</sup> ASCF (replacement of NaCl with choline Cl) led to a decrease in the base level of intracellular Cl <sup>−</sup> and H<sup>+</sup> (Figure 8A). The mean changes were for [Cl <sup>−</sup>]<sup>i</sup> by 1.98 ± 0.15 mM (*n* = 5, *p* < 0.05, 6 months) and for H+ by 0.08 ± 0.01 pH units (*n* = 5, *p* < 0.05, 6 months). In low-Na <sup>+</sup> ASCF, responses to K<sup>+</sup> -induced depolarization were potentiated by more than 2 times for both ions, resulting in the mean changes of 25.77 ± 2.74 mM and 0.40 ± 0.05 pH units (*n* = 5, Figure 8B,C).

− −

−

−

−

− ) illustrating the effects of the transition to "low Na " ASCF and **Figure 8.** Effect of extracellular Na <sup>+</sup> on [Cl <sup>−</sup>]<sup>i</sup> and pH<sup>i</sup> responses induces by the high K <sup>+</sup> depolarization. (**A**) Typical traces of continuous recording of [Cl −]i (top) and pH<sup>i</sup> (bottom) illustrating the effects of the transition to "low Na <sup>+</sup>" ASCF and responses to the addition of 17.5 mM K<sup>+</sup> to hippocampal slices (age 6 months). (**B**,**C**) Summary of the high K<sup>+</sup> -induced changes in concentrations of Cl <sup>−</sup> (**B**) and H<sup>+</sup> (**C**) in control (grey columns) and in ACSF containing low Na + (26.5 mM) (red columns). Summary from 5 slices (mean ± SEM, 6 months).

#### **3. Discussion**

− − Our study provides evidence that the optosensoric reporter, ClopHensor, expressed in transgenic mice, represents an efficient tool for non-invasive monitoring of intracellular Cl − and H<sup>+</sup> ions. Applying different conditions for modulation of neuronal activity in brain slices from transgenic mice, we demonstrated the ability of ClopHensor to simultaneously monitor and analyze changes in [Cl <sup>−</sup>]<sup>i</sup> and [H<sup>+</sup> ]<sup>i</sup> during artificial changes of its equilibrium.

− – – Maintaining physiologically relevant concentrations of H<sup>+</sup> and Cl <sup>−</sup> has a pivotal role in controlling neuronal excitability in the adult brain and during development, and is likely to be crucial in pathophysiological conditions. Both ions play an important role in many cellular processes, including neurotransmission, regulation of membrane potential, cell volume and water–salt balance, modulation of the functions of different proteins, including voltage-gated and receptor-operated channels [1,12,43–46]. Because of high metabolic activity, accompanied by depolarization, neurons may be susceptible to acidification-induced injury, as can happen in excessive synaptic activation or during conditions of anoxia or ischemia.

− The steady-state physiologically relevant dynamic range of pH<sup>i</sup> and [Cl −]i is determined by the balance of both passive transport, through ion channels and by active mechanisms via exchangers or cotransporters [1,46,47].

− – − – − − The intracellular Cl <sup>−</sup> in neurons of the mammalian brain is primary regulated by two cation–Cl <sup>−</sup> cotransporters, the Na <sup>+</sup>/K <sup>+</sup>/2Cl <sup>−</sup> cotransporter 1 (NKCC-1) and the K <sup>+</sup>/Cl − co-transporter 2 (KCC-2) [46–49]. NKCC1 pumps Cl <sup>−</sup> into neurons, while KCC2 uses the energy of the K <sup>+</sup> gradient to extrude Cl <sup>−</sup> from neurons and maintain relatively low [Cl −]i (Figure 9). Dysfunction or changes in the expression of these cotransporters is critically linked to the etiology of several neurologic disorders including epilepsy, acute trauma, ischemia, autoimmune disorders, and neuropathic pain [48,50,51].

**Figure 9.** Scheme of key transmembrane transport elements controlling intracellular Cl <sup>−</sup> and pH in brain cells.

−

–

'

− − − − − − – – − − − Similarly, a modest shift of intra- and extracellular H<sup>+</sup> ion concentration can have significant effects on brain functions, such as neuronal excitability, synaptic transmission and metabolism [1,52]. This is the result of the sensitivity of a large number of processes to protons, such as ion channel gating and conductance, synaptic transmission, cell-to-cell communication via gap junctions, and enzymatic activity in brain energy metabolism [52–55]. The major pH regulating transporters identified in the mammalian brain so far comprise the Na <sup>+</sup>/H<sup>+</sup> exchanger (NHE1), electrogenic Na <sup>+</sup>/HCO<sup>3</sup> <sup>−</sup> cotransporter 1 (NBCe1), Na + -dependent Cl <sup>−</sup>/HCO<sup>3</sup> <sup>−</sup> exchange (NDCBE), and Na + -independent anion exchanger (AE3) (Figure 9; [1,15,56]. NBce-1, NDCBE and AE3 transporters perform translocation of HCO<sup>3</sup> <sup>−</sup>, thus controlling the physiological range of pH<sup>i</sup> . The electrogenic Na <sup>+</sup>/HCO<sup>3</sup> <sup>−</sup> cotransporter NBCe1 is one of the major regulators of [H<sup>+</sup> ]<sup>i</sup> and is expressed in most brain cell types, with the most prominent expression being astrocytes [1]. It has been shown that NBCe1 is not only an acid extruder/base loader, as suggested in early studies [57,58], but also an acid loader/base extruder [16]. Anion Cl <sup>−</sup>/HCO<sup>3</sup> <sup>−</sup> exchange (AE3) is known as an important acid-loading performer in brain cells [16,59]. The finding that hippocampal neurons of knockdown AE3 mice (Ae3 –/– ) lack detectable Cl <sup>−</sup>/HCO<sup>3</sup> − exchange activity [60] suggests that AE3 plays a critical role in the maintenance of the Cl − equilibrium potential and/or pH<sup>i</sup> in neurons.

The role of monocarboxylate transporters (MCTs) in the regulation of the functional integrity of synaptic transmission has been demonstrated. In excitatory synapses, MCT constitutively supports synaptic transmission, even under conditions when there is a sufficient amount of glucose and intracellular ATP [61]. Monocarboxylates cause a decrease in the pH of neurons, which is associated with changes in bioelectric activity [62]. MCT's involvement in Cl modulation requires future analysis.

These features of transporters emphasize the need for simultaneous monitoring of [Cl <sup>−</sup>]<sup>i</sup> and [H<sup>+</sup> ]<sup>i</sup> when analyzing the mechanisms of ion homeostasis and neuronal activ-

ity, particularly using neuropathological models. A genetically encoded sensor, named ClopHensor, was proposed for simultaneous measurement of Cl<sup>−</sup> and H<sup>+</sup> ion concentrations [27]. This construct has shown its effectiveness at the heterologous expression in cell lines, demonstrating good sensitivity and high stability to bleaching during long fluorescence measurements [28]. We recently presented a line of transgenic mice expressing ClopHensor in neurons and obtained a detailed map of its distribution in the mouse brain [32]. Expression of this probe in transgenic mice was found to be highly specific and reproducible in different animals, suggesting that this experimental model represents a promising tool for analysis of dynamic changes in [Cl−]<sup>i</sup> and pH<sup>i</sup> .

In the present study, we analyzed the effectiveness of ClopHensor expressed in transgenic mice for monitoring [Cl−]<sup>i</sup> and [H<sup>+</sup> ]i in neurons of hippocampal slices upon changing the ionic equilibrium by pharmacological modulation of neuronal activity, changes in extracellular concentrations of Ca2+, Cl<sup>−</sup> or Na<sup>+</sup> , and by depolarization caused by high-frequency synaptic stimulation or the use of high K<sup>+</sup> .

#### *3.1. Kinetics of Synaptically Induced Transients of [Cl*−*]<sup>i</sup> and [H<sup>+</sup> ]i*

High frequency induced stimulation of Schaffer's collaterals resulted in the elevation of [Cl−]<sup>i</sup> and [H<sup>+</sup> ]<sup>i</sup> and slow recovery. Importantly, the decay kinetics of Cl<sup>−</sup> transients was nearly 5-fold faster than pH. The experiments on wild type mice using the chemical pH sensor, BCECF, confirmed remarkably slow kinetics of pH<sup>i</sup> recovery. These differences in the kinetics of recovery of Cl<sup>−</sup> and H<sup>+</sup> responses may reflect the involvement of different cellular participants in control of these ions' homeostasis.

In the study using the ClopHensorN indicator heterologously expressed in cultured hippocampal slices, about 1.5 times faster decay of intracellular Cl<sup>−</sup> concentrations were indicated [29]. A profound analysis is necessary to reveal relationships and efficacy of different transporters and other mechanisms involved in the recovery of [Cl−]<sup>i</sup> and [H<sup>+</sup> ]i from synaptically induced disequilibrium.

#### *3.2. Effect of GABA Receptor Inhibition on Synaptically Induced Transients of [Cl*−*]<sup>i</sup> and [H<sup>+</sup> ]i*

Results of our study demonstrate that in hippocampal slices of transgenic mice, high frequency stimulation of Shaffer's collaterals causes an intracellular increase of both ions, Cl<sup>−</sup> and H<sup>+</sup> . Synaptic stimulation causes the release of neurotransmitters, primary glutamate and GABA, which activates corresponding receptors, leading to the opening of cation and anion-selective ion channels and resulting depolarization of neurons.

As predicted, inhibition of GABA-mediated chloride conductance by bicuculline resulted in a decrease of the synaptically induced influx of Cl−. On the other side, this decrease was accompanied by potentiation of acidific [H<sup>+</sup> ]i transients. This could be a consequent of two main things: (i) increased synaptic stimulation due to blockade of inhibitory GABA receptors; (ii) changes in the activity of transporters.

In support of the first possibility, the electrophysiological analysis showed that bicuculline causes an elevation in the amplitude of evoked local field potentials. On the other side, experiments that lowered the extracellular Cl showed that in spite of the strong reduction of evoked local field potentials, the amplitude of synaptically induced pH responses was potentiated nearly 10 times in comparison with control.

We suggest that this may be a consequence of the weakening of the acidifying activity of the Cl−HCO<sup>3</sup> <sup>−</sup> exchange. This assumption was tested in experiments with high K<sup>+</sup> induced depolarization and changes of external concentrations of ion participating in the functioning of transporters.

#### *3.3. Analysis of High K<sup>+</sup> -Induced Depolarization on Changes of [Cl*−*]<sup>i</sup> and [H<sup>+</sup> ]i in External Ca2+-Free Conditions*

Depolarization induced by application of high K<sup>+</sup> or by other means has been used in a number of studies for analysis of pH<sup>i</sup> changes in cultured and freshly dissociated neurons, and brain slices [37–40,63]. In most studies, depolarization induced by application

of high external K<sup>+</sup> was accompanied by elevation of intracellular Ca2+ and decrease of pH, i.e., acidification [37,39]. Early observations performed on cultured cells showed that under external Ca2+-free conditions, the acidification responses were either completely prevented [38] or even inversed, i.e., exhibited an increase in pH<sup>i</sup> [39]. Moreover, a blocker of voltage-gated Ca2+ channels inhibited K<sup>+</sup> -induced responses, suggesting that an increase in [Ca2+]<sup>i</sup> is a key factor for acidosis to occur [38].

We tested this suggestion in experiments on brain slices of ClopHensor-expressing mice and obtained distinct results. In our experiments, transition to external Ca2+-free conditions alone caused a decrease in base pH, i.e., acidification. Surprisingly, K<sup>+</sup> -induced depolarization produced additional acidification. The responses were only partially attenuated in comparison with control, suggesting an only partial contribution of Ca2+-dependent processes in pH<sup>i</sup> responses induced by membrane depolarization.

The reasons for these contrast observations have to be clarified. Among them may be the use of HEPES solution in cell culture experiments and different energy states of neurons.

#### *3.4. Effect of Decreasing Extracellular Cl*<sup>−</sup> *and Na<sup>+</sup> on K<sup>+</sup> -Induced Changes in [Cl*−*]<sup>i</sup> and [H<sup>+</sup> ]i* 3.4.1. Decreasing Extracellular Cl<sup>−</sup>

In our experiments, decreasing Cl<sup>−</sup> in ASCF resulted in a slow-developing (10–15 min) strong decrease in the base level of cytoplasmic Cl<sup>−</sup> and elevation of pH, i.e., extruding of H+ from neurons. However, in "low-Cl−" conditions, high K<sup>+</sup> -induced depolarization caused remarkable potentiation of acidification pH responses, while Cl<sup>−</sup> responses were nearly completely suppressed. These observations are in accordance with results obtained at high-frequency synaptic stimulation and indicate that different mechanisms are involved in the control of [Cl−]<sup>i</sup> and pH<sup>i</sup> at changes of extracellular ion concentrations and during depolarization of neurons. A decrease in extracellular Cl<sup>−</sup> weakens the ability of the Cl−/HCO<sup>3</sup> <sup>−</sup> transporter to pump Cl<sup>−</sup> into neurons in exchange for HCO<sup>3</sup> <sup>−</sup>, which leads to slow alkalinization. At high K<sup>+</sup> application, depolarization induces elevation of intracellular Ca2+, resulting in stimulation of PMCA transporter activity and consequent acidification.

## 3.4.2. Decreasing Extracellular Na<sup>+</sup>

Since the NHE, NBCe1 and NDCBE transporters operate on a Na<sup>+</sup> gradient, a decrease in the extracellular Na<sup>+</sup> concentration should lead to a decline in the operation of these pumps and, as a result, to a weakening of protons extruding from the cytoplasm, and a decrease in intracellular HCO<sup>3</sup> <sup>−</sup> We assume that for these reasons, in our experiments, there was a decrease in the pH<sup>i</sup> of neurons during the transition to low Na<sup>+</sup> and a potentiation of acidic responses to depolarization caused by increased K<sup>+</sup> . In the future, these processes will be analyzed in detail.

In conclusion, the baseline level and fluctuations of intracellular Cl<sup>−</sup> and pH play a crucial role for intercellular and intracellular signaling, as well as for cellular and synaptic plasticity. Our study demonstrates that transgenic mice expressing ClopHensor provide ample opportunities for studying the homeostasis of chloride and hydrogen not only under normal conditions, but also in pathology.

In particular, Cl<sup>−</sup> gradients are disrupted in epilepsy, especially in early childhood, which significantly complicates treatment. Hydrogen gradients in the central nervous system are disrupted by neuroinflammatory processes of various origins and ischemia. Research in these areas is relevant today, and ClopHensor mice can be a reliable tool for in-depth analysis of the mechanisms controlling the physiological ranges of concentrations of these ions.

#### **4. Materials and Methods**

#### *4.1. Animals*

Experiments were conducted on laboratory ICR CD-1 outbreed mice of both sexes at postnatal days 10–12 and adult transgenic mice, aged 2–6 months, strain C57BL/6N

expressing ClopHensor (Diuba et al., 2020). Use of animals was carried out in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85– 23, revised 1996) and European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes (Council of Europe No. 123; 1985). All animal protocols and experimental procedures were approved by the Local Ethics Committee of Kazan State Medical University. Mice had free access to food and water and were kept under natural day length fluctuations. Animals were not involved in any previous procedures.

#### *4.2. Solutions and Drugs*

Brain slices were prepared in ice-cold high potassium solution containing (in mM): Kgluconate 120, HEPES acid 10, Na-gluconate 15, EGTA 0.2, NaCl 4 (pH 7.2, 290–300 mOsm). After cutting, slices were incubated in a high magnesium artificial cerebrospinal fluid (ACSF) containing (in mM): NaCl 125, KCl 2.5, CaCl<sup>2</sup> 0.8, MgCl<sup>2</sup> 8, NaHPO<sup>4</sup> 1.25, glucose 14, NaHCO<sup>3</sup> 24 (pH 7.3–7.4, 290–300 mOsm). Storage of slices and performing of experiments were conducted in ACSF, containing (in mM): NaCl 125, KCl 2.5, CaCl<sup>2</sup> 2.3, MgCl2, 1.3, NaHPO<sup>4</sup> 1.25, glucose 14, NaHCO<sup>3</sup> 24 (pH 7.3–7.4, 290–300 mOsm). The ACSF was continuously oxygenated with 95% O<sup>2</sup> and 5% CO<sup>2</sup> to maintain the physiological pH. The following drugs were used: APV (40 µM, Hello Bio, Bristol, UK, Cat# HB0225), CNQX (10–40 µM, Sigma Aldrich, St. Quentin Fallavier, France, CAS: 115066-14-3), (-)- Bicuculline methochloride (10 µM or 40 µM, Tocris, Science Park Abingdon, UK, Cat#0131), BCECF/AM (10 µM, Sigma Aldrich, St. Quentin Fallavier, France, CAS: 117464-70-7). All drugs were diluted on the day of the experiment from the 1000× to 4000× stocks kept at −20 ◦C.

#### *4.3. Preparation of Brain Slices*

Sagittal 350 µm thick sections of the cerebral hemispheres containing the hippocampus were obtained with the use of a vibratome of Model NVSLM1, World Precision Instruments. Mice were euthanized by decapitation. The brain was quickly removed and placed in a Petri dish filled with ice-cold high-K<sup>+</sup> solution. The cerebellum and olfactory bulbs were cut off using a scalpel; the cerebral hemispheres were separated by cutting along the longitudinal fissure and mounted onto the vibratome specimen disc using superglue, orienting them downward with the sagittal cut surface and the cortex facing the razor blade. Sagittal 350 µm thick sections were prepared in ice-cold high K<sup>+</sup> solution. After being cut, slices were incubated for 15 min at room temperature in a resting high magnesium oxygenated ACSF. Then, slices were placed in a chamber filled with oxygenated ACSF. Before use, slices were allowed to recover for at least 1 h at room temperature.

Experiments were conducted during the period of 1–6 h after slicing. For the recordings, the brain slices were placed in a chamber perfused with an oxygenated ACSF. Recordings were carried out at 30–31 ◦C with the speed of perfusion of 25 mL/min.

#### *4.4. Fluorescence Imaging in Brain Slices*

## 4.4.1. Monitoring of [Cl−]<sup>i</sup> and pH on Brain Slices ClopHensor Mice

Fluorescence images were obtained using an upright microscope Olympus BX51WI equipped with the iXon Life 897 EMCCD camera (Andor, Oxford Instruments, Abingdon, UK), a 4-Wavelength LED Source (LED4D001, Thorlabs, Newton, NJ, USA) accompanied with a Four-Channel LED Driver (DC4100, Thorlabs), a quad-band filter set (Cat# 9403, Chroma, Foothill Ranch, CA, USA), and a water-immersion objective (60× magnification, 1 numerical aperture; LumPlanFL N, Olympus, Tokyo, Japan) (Figure 10).

**Figure 10.** Setup for simultaneous monitoring of pH<sup>i</sup> and [Cl <sup>−</sup>]<sup>i</sup> using ClopHensor.

The 4-Wavelength LED Source was supplied by light-emitting diodes (LEDs) at 365 nm, 455 nm, 505 nm and 590 nm, the last three of which were used in our study. The capacity of lighting and the LED switching order were adjusted via the DC4100 driver. Cyan channel fluorescence (455 nm excitation) was detected from 460 nm to 485 nm, green channel (505 nm excitation) from 527 nm to 551 nm, and red channel (590 nm excitation) from 600 nm to 680 nm.

All peripheral hardware control, image acquisition and the average fluorescence intensity measurement were achieved using DriverLed software (KSMU, Kazan). Regions of interest (ROI) were set around pyramidal cell bodies of the CA1 hippocampal region. The average fluorescence intensity of the ROI was tracked online at the time of the real-time fluorescence imaging. Numerical data were output by the DriverLed software on the computer in the form of text documents, which were subsequently parsed using Excel 2016 (Microsoft).

Excitation of E2GFP was provided by light-emitting diodes (LEDs) at 455 nm and 505 nm, and a LED at 590 nm provided excitation of DsRed. The duration of excitation at each wavelength was usually 70–110 ms. Examples of the images taken are shown in Figure 1C. Fluorescent emission was recorded continuously from the hippocampal CA1 pyramidal cells with a sampling interval of 10 s. Evoked emission signals were obtained

by stimulation of Schaffer's collaterals by glass bipolar electrode, filled with ACSF, placed in the stratum radiatum of the CA2 hippocampal area (100 Hz for 20 s, 20–320 µA, single pulse width 200 µs). –320 μ 200 μs

#### 4.4.2. Monitoring of pH Using BCECF

'

–

As an additional control of proper pH monitoring of intracellular pH in hippocampal neurons, we used the pH-sensitive dye, 2 ′ ,7 ′ -Bis (2-carboxyethyl)- 5 (6)-carboxyfluoresceinacetoxymethyl ester (BCECF-AM) (Figure 11A). The fluorescent dye, BCECF, was introduced for measuring cytoplasmic pH by Roger Tsien and co-workers [36]. Presently used BCECF-AM is a mixture of three types of cell-permeable non-fluorescent molecules, which are converted to fluorescent non-membrane-penetrating form by intracellular esterases [21]. This dye provides dual-excitation ratiometric monitoring of intracellular pH. sensitive dye, 2′,7′

) The chemical structure of one of the molecules of 2′,7′ **Figure 11.** (**A**) The chemical structure of one of the molecules of 2 ′ ,7 ′ -Bis (2-carboxyethyl)- 5 (6)-carboxyfluoresceinacetoxymethyl ester (BCECF-AM). (**B**) Excitation spectra of BCECF collected at different pH values. The BCECF pH measurements were made by determining the pH-dependent ratio of emission intensity when it was excited at 505 nm versus the emission intensity when excited at 455 nm.

We determined the pH changes by calculating the ratio of two emission signals obtained after illumination by light at 455 nm and 505 nm (Figure 11B). The intracellular pH changes after tetanic stimulation and its modulation by bicuculline were analyzed. Experiments were conducted on wild type juvenile mice (P10-P12). Registration conditions were similar to those performed on slices from transgenic ClopHensor mice. Evoked pHspecific ratiometric emission signals were obtained by giving the same pattern of electrical stimulation (100 Hz for 20s); the stimulus intensities ranged from 200 µA to 250 µA.

μA to 250 μA. Hippocampal slices were stained with 10 µM BCECF-AM for 40 min in the oxygenated ACSF. Then the dye was washed out and slices were superfused with normal ACSF for at least 40 min to allow esterases to cleave AM and stabilise pHi. For fluorescent monitoring, slices were transferred to the optical recording chamber, which was mounted on the stage of an upright microscope (Olympus BX51WI) (Figure 10).

The setup consisted of a microscope (Olympus BX51WI equipped with the iXon Life 897 EMCCD camera (Andor camera)), a 4-Wavelength Thorlabs LED Source (Light source), and a quad-band filter set (excitation filter). Excitation light with wavelengths of 505 nm, 455 nm and 590 nm, each for 70–100 ms, was sequentially applied to slices and the emission signals from CA1 pyramidal cells were registered by an EMCCD camera. The stimulation electrode was placed in the stratum radiatum of the CA2 hippocampal area. The evoked local field potentials were recorded using a glass micropipette electrode, placed in the stratum radiatum of the CA1 hippocampal area and an HEKA EPC 10 Patch Clamp Amplifier (not shown).

Slices were alternately illuminated at 455 nm and 505 nm with duration of 100 ms. Light from both wavelengths was passed through appropriate excitation and emission filters. Fluorescence image pairs were captured every 10 s by an intensified camera. Evoked emission signals were obtained in the same way as for ClopHensor: by sustained highfrequency electrical stimulation of Schaffer's collaterals (100 Hz for 20 s, 20–320 µA, single pulse width 200 µs).

#### *4.5. Electrophysiological Recording*

Evoked local field potentials (eLFP) were recorded in the stratum radiatum of the CA1 hippocampal region using glass micropipette electrodes filled with ACSF (resistance 1–2 MΩ) and an HEKA EPC 10 Patch Clamp Amplifier (HEKA Elektronik, Lambrecht, Germany). The DS3 Constant Current Isolated Stimulator (Digitimer, Welwyn Garden, UK) and a bipolar stimulating electrode placed on the Schaffer's collaterals at the hippocampal area CA2 were used for the induction of eLFPs. Current pulses 20–300 µA in amplitude and 200 µs in duration were applied to obtain reliable eLFPs. Single eLFPs were recorded continuously every 20 s. PatchMaster software (HEKA Electronik, Lambrecht, Germany) was used to record eLFPs, control the HEKA EPC 10 Patch Clamp Amplifier and the DS3 Constant Current Isolated Stimulator.

## *4.6. Intracellular pH and Cl*<sup>−</sup> *Calibration of ClopHensor on Hippocampal Slices*

To perform pH and Cl<sup>−</sup> calibration, we used a double ionophore technique [64]. The brain slices were exposed to the antibiotic nigericin, which acts as H+/K<sup>+</sup> antiporter and the Cl−/OH<sup>−</sup> antiporter tributyltin, which forms pores in the cell membrane and allows external Cl<sup>−</sup> to equilibrate with intracellular Cl<sup>−</sup> [65]. For the action of the compounds, before fluorescent monitoring, the hippocampal slices were kept for 1–28 h at +4 ◦C in the following solution: 150 mM K-Gluconate, 20 mM HEPES, and 10 mM D-glucose) with addition of nigericin (20 mkM, pH = 7.22–7.25) and tributyltin (20 mkM).

For pH calibration, the solutions with different pH values (from 6.23 pH to 8.04 pH) were prepared by adding KOH for alkalization. For calibration of Cl−, the solutions were prepared at pH = 7.25 and contained a different concentration of Cl<sup>−</sup> (0, 3, 10, 30, 100 mM). To keep the osmolality of the solutions, KCl was correspondently substituted by the Kgluconate (150, 147, 140, 120, 50 mM). All of these solutions contained 20 mM HEPES, and 10 mM D-glucose.

Fluorescent signals were recorded from the CA1 zone of the hippocampus with the following lighting parameters in the Drive LED program: Exposition 70 ms, LED on 100 ms, binning 2, Voltage (590 nm) 70 mV, Voltage (505 nm) 10 mV, Voltage (455 nm) 15 mV. The slices were incubated in each solution until the fluorescence stabilized, usually for 20–30 min.

We obtained a linear dependence for pH (from 6.7 pH to 7.6 pH) on the fluorescence ratio (F505nm/F455nm):

$$\text{pH} = \text{F}\_{505\text{nm}} / \text{F}\_{455\text{nm}} \times \text{K}\_1 + \text{K}\_{2\prime} \tag{3}$$

where K<sup>1</sup> = 0.44, K<sup>2</sup> = 6.06.

Coefficients were obtained by fitting data for 6 slices.

To estimate the [Cl−]<sup>i</sup> from calibration data obtained on 4 slices, the following equation was used:

$$\left[\text{Cl}^{-}\right]\_{\text{i}} = \text{K}\_{0} + \frac{\text{K}\_{1} - \text{K}\_{0}}{1 + \left(\frac{\text{K}\_{3}}{\text{F}\_{990 \text{ nm}}/\text{F}\_{455 \text{ nm}}}\right)^{\text{K}\_{2}}} \tag{4}$$

where K<sup>0</sup> = −0.33, K<sup>1</sup> = 33.62, K<sup>2</sup> = 10.1, K<sup>3</sup> = 0.87.


All coefficients were obtained by fitting data using the Igor Pro 6.02 software.

#### *4.7. Data Analysis and Statistics*

Amplitudes of evoked local field potentials were measured using the Online Analysis function of PatchMaster software (HEKA Electronik, Lambrecht, Germany). Superimposed average traces of evoked local field potentials were generated in PatchMaster software and processed for further presentation in Igor Pro 6.02 software (WaveMetrics, Tigard, OR, USA).

The average emission intensity of the region of interest (ROI) was measured online at the time of the real-time fluorescence imaging or, if necessary, recalculated using DriverLed software. Excel 2016 (Microsoft) software was used to compute and plot Cl<sup>−</sup> and pHdependent ratios of emission intensities and to measure the amplitudes evoked by tetanic stimulation pH- and Cl−-specific ratiometric fluorescence emission signals.

Origin 15 software was used to perform a statistical analysis of the data, to plot the graphs and compute the decay times of the evoked pH- and Cl<sup>−</sup> -specific fluorescence emission signals.

Data are represented as means ± SEM. Statistical significance was determined using Paired Sample Wilcoxon Signed Rank Test and Mann Whitney U test. Differences were considered significant at *p* < 0.05.

**Author Contributions:** For Conceptualization, P.B.; methodology and investigation E.P., D.P. and P.B.; formal analysis, E.P., D.P.; writing—original draft preparation, E.P., D.P. and P.B. supervision, P.B.; project administration, P.B.; funding acquisition, P.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by the Russian Science Foundation (Grant: 18-15-00313).

**Institutional Review Board Statement:** The study was conducted according to the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85–23, revised 1996) and European Convention for the Protection of Vertebrate Ani-mals used for Experimental and other Scientific Purposes (Council of Europe No. 123; 1985). All animal protocols and experimental procedures were approved by the Local Ethics Committee of Kazan State Medical University.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We cordially thank Zakharov, A. for developing the program for the analysis of fluorescent signals. We are also grateful to Yu, N. Davidyuk and A. Yusupova for PCR analysis of transgenic mice, and Kiani, N. for useful text comments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

