Nitric Oxide Synthase Blockade Impairs Spontaneous Calcium Activity in Mouse Primary Hippocampal Culture Cells

Oscillation of intracellular calcium concentration is a stable phenomenon that affects cellular function throughout the lifetime of both electrically excitable and non-excitable cells. Nitric oxide, a gaseous secondary messenger and the product of nitric oxide synthase (NOS), affects intracellular calcium dynamics. Using mouse hippocampal primary cultures, we recorded the effect of NOS blockade on neuronal spontaneous calcium activity. There was a correlation between the amplitude of spontaneous calcium events and the number of action potentials (APs) (Spearman R = 0.94). There was a linear rise of DAF-FM fluorescent emission showing an increase in NO concentration with time in neurons (11.9 ± 1.0%). There is correlation between the integral of the signal from DAF-FM and the integral of the spontaneous calcium event signal from Oregon Green 488 (Spearman R = 0.58). Blockade of NOS affected the parameters of the spontaneous calcium events studied (amplitude, frequency, integral, rise slope and decay slope). NOS blockade by Nw-Nitro-L-arginine suppressed the amplitude and frequency of spontaneous calcium events. The NOS blocker 3-Bromo-7-Nitroindazole reduced the frequency but not the amplitude of spontaneous calcium activity. Blockade of the well-known regulator of NOS, calcineurin with cyclosporine A reduced the integral of calcium activity in neurons. The differences and similarities in the effects on the parameters of spontaneous calcium effects caused by different blockades of NO production help to improve understanding of how NO synthesis affects calcium dynamics in neurons.


Introduction
Nitric oxide (NO) is a small molecule that is a secondary gaseous messenger [1]. NO is crucial for various biological systems, including the central and peripheral nervous, cardiovascular, immune and reproductive systems [2][3][4][5]. Under physiological conditions, it contributes to regulating proliferation, survival and neuronal differentiation [6]. Three nitric oxide synthase (NOS) genes whose products have distinct tissue localization and properties are known: neuronal, inducible and endothelial NOS (nNOS, iNOS, and eNOS, respectively) [7]. nNOS participates in the regulation of neurogenesis processes presumably preventing cell proliferation, participating in memory formation, learning processes, sexual behavior, excitotoxicity and a number of neurodegenerative conditions [7,8]. NO released from a small subpopulation of NOS-containing neurons that lie in apposition to cerebral arterioles appears to be involved in mediating the coupling between neuronal activity and cerebral blood flow and metabolism [9].

NO Synthesis in Hippocampal Primary Cultures
Dissociated primary neuronal cultures are widely used as a model system to investigate the cellular and molecular properties of diverse neuronal populations and the mechanisms of action potential generation and synaptic transmission. Hippocampal cultures show all the features of neuronal activity, such as formation of synchronous burst activity and spontaneous calcium oscillations, and they share many key properties with CA1 pyramidal cells during long term potentiation (LTP) [43,44].
We recorded the increase in the amount of nitric oxide in neuronal cell bodies ( Figure 1A). DAF-FM fluorescent emission simply increased linearly, showing the rise of NO concentration with time in soma ( Figure 1B, n = 30, 11.9 ± 1.0%, 5th min). NO is a vital second messenger in cultured neuronal networks and is necessary for network activity.

NO Synthesis in Hippocampal Primary Cultures
Dissociated primary neuronal cultures are widely used as a model system to investigate the cellular and molecular properties of diverse neuronal populations and the mechanisms of action potential generation and synaptic transmission. Hippocampal cultures show all the features of neuronal activity, such as formation of synchronous burst activity and spontaneous calcium oscillations, and they share many key properties with CA1 pyramidal cells during long term potentiation (LTP) [43,44].
We recorded the increase in the amount of nitric oxide in neuronal cell bodies ( Figure  1A). DAF-FM fluorescent emission simply increased linearly, showing the rise of NO concentration with time in soma ( Figure 1B, n = 30, 11.9 ± 1.0%, 5th min). NO is a vital second messenger in cultured neuronal networks and is necessary for network activity.

Spontaneous Electric Activity in Cultured Neurons Is Closely Related to Calcium Oscillations
Spontaneous activity in cultures varies from single APs to AP bursts of different sizes [45]. Concurrent recording of neuronal firing via patch-clamp and calcium fluorescent imaging showed action potential (AP)-associated calcium activity ( Figure 1C). The amplitude of SCEs was related to the number of APs in the burst. Spearman correlation analysis showed high codependency (Figure 2A, n = 5, R = 0.94). The integral of SCEs can be interpreted as the total change in [Ca 2+ ]in in the cytosol over time, or calcium flow throughout the cytosol. In addition to SCE amplitude, we evaluated the correlation of the amount of APs and the SCE integral ( Figure 2A, n = 5, R = 0.91). Thus, it can be concluded that the indicators of the amplitude and OG integral of SCEs depend on the number of APs generated by the cells.

Spontaneous Electric Activity in Cultured Neurons Is Closely Related to Calcium Oscillations
Spontaneous activity in cultures varies from single APs to AP bursts of different sizes [45]. Concurrent recording of neuronal firing via patch-clamp and calcium fluorescent imaging showed action potential (AP)-associated calcium activity ( Figure 1C). The amplitude of SCEs was related to the number of APs in the burst. Spearman correlation analysis showed high codependency (Figure 2A, n = 5, R = 0.94). The integral of SCEs can be interpreted as the total change in [Ca 2+ ] in in the cytosol over time, or calcium flow throughout the cytosol. In addition to SCE amplitude, we evaluated the correlation of the amount of APs and the SCE integral ( Figure 2A, n = 5, R = 0.91). Thus, it can be concluded that the indicators of the amplitude and OG integral of SCEs depend on the number of APs generated by the cells. To assess the relationship between NO synthesis and calcium activity in neurons, we performed a correlation analysis of NO production and [Ca 2+ ]in by measuring DAF-FM and OG fluorescence. We used recordings from different neuronal cultures, since the fluorescence emission spectra of OG and DAF-FM overlap. Fluorescence measurements were performed for 5 min. Minute-by-minute evaluation using Spearman correlation analysis showed a rather high degree of correlation between NO increase and calcium integrals ( Figure 2B, n = 6, R = 0.58). The cumulative calculations of OG integral and NO production from the 1st to the 5th minute also indicated a high correlation over a longer period ( Figure 2B, n = 6, R = 0.82). These correlations clearly indicate a link between intracellular calcium and NO, and thus involvement of NO in the regulation of spontaneous neuronal activity.

Effect of NOS-Blockade on Calcium Activity Is Associated with Spontaneous Neuronal Firing
NO participates in many intracellular processes linked to calcium signaling and neurotransmission. Considering there is NO synthesis in hippocampal primary cell cultures, we tested how NO synthesis blockade contributes to the calcium activity associated with spontaneous neuronal firing. Two different NOS blockers were used in the study. Nw-Nitro-L-arginine (L-NNA 100 μM; Sigma Aldrich, St. Louis, MI, USA) is a blocker of all NOS isoforms, and 3-Bromo-7-Nitroindazole (3-Br-7Ni 10 μM; Tocris, Bristol, UK) is a selective blocker of the nNOS isoform. For 3-Br-7Ni, the average amplitude did not change ( Figure 3B, n = 6, 22.00 ± 4.55 vs. 22.67 ± 5.16%, n.s.). The average frequency decreased significantly ( Figure 3B, n = 6, 0.21 ± 0.01 vs. 0.08 ± 0.02 Hz, p = 0.001). The calcium integral also significantly decreased ( Figure 3B, n = 6, 15.9 ± 3.0 vs. 5.2 ± 1.9 a.u.*ms, p = 0.0009). Two-fold reduction in the [Ca 2+ ]in integral indicates powerful suppression of calcium-conducting systems. At the same time, the rise slope (20-80%) showed no significant increase ( Figure 3B, n = 6, 5.7 ± 1.5 vs. 4.8 ± 1.3 a.u./s, n.s.), as well as the decay slope (0-50%) ( Figure  3B  To assess the relationship between NO synthesis and calcium activity in neurons, we performed a correlation analysis of NO production and [Ca 2+ ] in by measuring DAF-FM and OG fluorescence. We used recordings from different neuronal cultures, since the fluorescence emission spectra of OG and DAF-FM overlap. Fluorescence measurements were performed for 5 min. Minute-by-minute evaluation using Spearman correlation analysis showed a rather high degree of correlation between NO increase and calcium integrals ( Figure 2B, n = 6, R = 0.58). The cumulative calculations of OG integral and NO production from the 1st to the 5th minute also indicated a high correlation over a longer period ( Figure 2B, n = 6, R = 0.82). These correlations clearly indicate a link between intracellular calcium and NO, and thus involvement of NO in the regulation of spontaneous neuronal activity.

Discussion
This study is devoted to the dependence of spontaneous AP-associated calcium activity on NO/NOS activity in hippocampal neuron-glial cultures. We evaluated AP-associated calcium activity and found that different blockades of NOS have similar and different effects on calcium dynamics, allowing us to make hypotheses as to why.
Concurrent patch-clamp and imaging revealed a direct correlation between the number of APs in bursts and fluorescence intensity. The correlation between the number of APs and calcium increase was established for the amplitude and integral parameters of SCEs. Changes in SCE amplitude may indirectly indicate changes in the activity of APrelated molecular targets. This link is mostly related to synaptic targets and voltage-dependent channels [46].
The linear increase in DAF-FM fluorescence indicates stable somatic NO production. We evaluated the correlation of OG and DAF-FM integrals. The coefficient obtained from the analysis is evidence for this connection. In soma, voltage-dependent calcium channels (VDCCs) and ryanodine receptors (RyRs) play an important role in increasing intracellular calcium concentration in response to external and internal stimuli. NO participates in the regulation of these systems, and the NOS blockade led to a reduction in AP-associated calcium activity. Due to the dual role of NO and its participation in physiological and

Discussion
This study is devoted to the dependence of spontaneous AP-associated calcium activity on NO/NOS activity in hippocampal neuron-glial cultures. We evaluated AP-associated calcium activity and found that different blockades of NOS have similar and different effects on calcium dynamics, allowing us to make hypotheses as to why.
Concurrent patch-clamp and imaging revealed a direct correlation between the number of APs in bursts and fluorescence intensity. The correlation between the number of APs and calcium increase was established for the amplitude and integral parameters of SCEs. Changes in SCE amplitude may indirectly indicate changes in the activity of AP-related molecular targets. This link is mostly related to synaptic targets and voltage-dependent channels [46].
The linear increase in DAF-FM fluorescence indicates stable somatic NO production. We evaluated the correlation of OG and DAF-FM integrals. The coefficient obtained from the analysis is evidence for this connection. In soma, voltage-dependent calcium channels (VDCCs) and ryanodine receptors (RyRs) play an important role in increasing intracellular calcium concentration in response to external and internal stimuli. NO participates in the regulation of these systems, and the NOS blockade led to a reduction in AP-associated calcium activity. Due to the dual role of NO and its participation in physiological and pathological effects, the issue of increased neuronal activity due to increased NO production needs to be resolved in the future. Based on the results, an increase in the rates of spontaneous calcium events and action potentials are expected. nNOS is mostly located in the post-synapse, which can explain the difference in the effect of 3-Br-7Ni and L-NNA on the amplitude and frequency of SCEs from cellular somas. The nNOS blockade affects the frequency of SCEs to a greater extent than amplitude, since nNOS is mostly localized in synapses [10,11]. Neuronal network activity depended on nNOS blockade, while not affecting the amplitude of SCEs on neuronal somas. A decrease in the frequency of SCEs when blocking NOS is associated with the need for the presence of NO to maintain calcium homeostasis in synapses [47]. For example, PMCA4b is one of the major players in calcium balancing; it interacts with nNOS at the PDZ site reducing NO production [21]. The dose-dependent manner of L-NNA action may also be an answer to the question about the difference between the effects of L-NNA and 3-Br-7Ni on the frequency of SCEs [48].
NOS blockade decreased the frequency, amplitude and integral of SCEs in neurons. This indirectly indicates reduced neuronal excitability. At the postsynapse, ionotropic and metabotropic glutamate receptors (NMDARs, AMPARs, mGluRs) are affected by NO [49].
As a consequence, postsynaptic activation may suffer in the absence of NO. An additional explanation of the SCE frequency decrease might be connected to NO-dependent trafficking and expression of AMPARs on the postsynapse [20]. The absence of NO affects the activity of mGluRs and G-proteins, reducing the excitability of the postsynaptic membrane [50]. Insufficient calcium activity of neuronal soma in the presence of L-NNA due to reduced postsynapse excitability may be the cause of insufficient presynaptic vesicular release. Normally, APs from the soma arrive at the presynapse leading to the activation of VDCCs, RyRs and, as a consequence, calcium-dependent presynaptic processes: vesicular transport and neurotransmitter release [51]. NO also enhances calcium release from intracellular stores [52,53]. S-nitrosylation plays a role in the activation of L-type VDCCs, and increases the release of glutamate. Blocking NO synthesis affected the presynaptic targets and the processes of presynaptic function associated with them. This may explain the decrease in the frequency of SCEs, which reflect network activity, when blocking the synthesis of NO. Interestingly, both 3-Br-7Ni and L-NNA reduced the integral to similar values that might be a threshold calcium flow level, which neurons can maintain without NO activity. The rise and decay slopes mostly reflect the speed of calcium propagation with calcium intake channels and calcium removal systems (PMCA and SERCA). The absence of rise and decay slope changes with the nNOS blocker 3-Br-7Ni reveals no changes in calcium intake and calcium removal. This might be connected to the strong balancing mechanisms of the calcium amount for intracellular calcium-dependent processes. With L-NNA, we see the influence on the rise and decay slope. It changes, meaning a longer front of calcium intake and slower calcium removal from the soma. This may be due to overfilled calcium stores. RyRs cannot act normally without nitrosylation, leading to a reduction in calcium leaving the endoplasmic reticulum (ER) via RyRs, lowering SERCA activity because the calcium concentration in the ER remains high [51].
There is no NO-dependent decrease in SCE frequency when blocking calcineurin synaptic activity [24]. CysA did not affect SCE amplitude and frequency, showing its activity is mostly related to postsynaptic rather than somatic calcineurin targets. Calcineurin also showed lower calcium flow control than NOS, but the rise slope changed in the same direction as in the presence of L-NNA. This might indicate some downregulation of calcium intake but not removal when calcineurin is blocked [54]; however, calcineurin has a number of other effects on synaptic physiology.
These results reveal the calcium dynamics in relation to NO production in neurons, and lead to hypotheses that must be further investigated.

Cell Culture Preparation
A mixed C57Bl/6 mouse primary embryonic neuronal cell culture was prepared as described in [46], with the protocol number 357 for animal handling and cell culture preparation approved by the Ethics Committee of the Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry. Mice husbandry included breeding pairs in the cage ad libitum. For neuron-glial culture preparation and cultivation, Thermo Scientific mediums and supplements (Gibco, New York, NY, USA) were used. Briefly, pregnant mice at stage E18 were euthanized with commonly used isoflurane and then decapitated; the uterus was dissected out and the embryos were placed in a sterile Petri dish with Hanks' solution. Dissected hippocampi were collected in a tube with pre-heated Versen solution. Cut hippocampi were trypsinized with 0.31% solution for 15 min in an incubator. After incubation, trypsin was removed and cells on the bottom were rinsed with Plating medium containing DMEM, 10% FBS and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). Next, 500 µL Plating medium with 40 µL DNAse was added for 10-15 s. Washed cells were resuspended and counted in the Goryaev chamber. Polyethylene-imine-coated covers were placed in plate wells and 4 × 104 cells were plated on each. After 1 h incubation, the wells were filled with Full NBM I, containing Neurobasal media, 0.05% β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA), 1% Glutamax, 2% B27 supplement and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). After 7 days, the medium was replaced with Full NBM II, containing Neurobasal media, 1% Glutamax, 2% B27 supplement and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). Cultures were fed every 2-3 days. All experiments were conducted during the daylight ( Figure 6). taining DMEM, 10% FBS and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). Next, 500 uL Plating medium with 40 ul DNAse was added for 10-15 s. Washed cells were resuspended and counted in the Goryaev chamber. Polyethylene-imine-coated covers were placed in plate wells and 4 × 104 cells were plated on each. After 1 h incubation, the wells were filled with Full NBM I, containing Neurobasal media, 0.05% β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA), 1% Glutamax, 2% В27 supplement and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). After 7 days, the medium was replaced with Full NBM II, containing Neurobasal media, 1% Glutamax, 2% В27 supplement and 1% gentamicin (Sigma-Aldrich, St. Louis, MO, USA). Cultures were fed every 2-3 days. All experiments were conducted during the daylight ( Figure 6).

Intracellular Calcium Recordings
Oregon Green-488 BAPTA-1 AM (OG; Invitrogen, Carlsbad, CA, USA) dissolved in dimethylsulfoxide (1 mM stock) was added to each culture before the experiment to a final concentration of 1 μM. Cultures were incubated for 40 min at 37 °C, CO2 5%. Then, coverslips were submerged in a recording chamber. The recording chamber was continuously perfused at 1.5 mL/min with a solution containing (mM): 130-NaCl, 2.5-KCl, 1.5-MgCl2, 1.5-CaCl2 10-glucose, 10-HEPES, 300 ± 5 mOsm, pH-7.33. An Evolve512 EMCCD camera (Photometrics UK Ltd., Marlow, UK), mounted on the SliceScope Pro 2000 (Scientifica, Uckfield, UK) microscope, was used ( Figure 6). Data acquisition was made with 3.6 fps. ROI positioning and dF/F0 analysis were performed with AxioVision Imaging System (Carl Zeiss, Jena, Germany). Calcium events were analyzed using Mini Analysis Program (Synaptosoft, Decatur, GA, USA), and pClamp (Molecular Devices, San Jose, CA, USA). The baseline of Ca 2+ -sensitive fluorescence for each cell at rest (between two neighboring calcium events) was taken as a 100% reference point. Thus, the amplitude of each SCE is represented as n% from the baseline. We evaluated the following SCE parameters: amplitude, frequency, individual integral, total integral (sum of integrals), rise slope and decay

Intracellular Calcium Recordings
Oregon Green-488 BAPTA-1 AM (OG; Invitrogen, Carlsbad, CA, USA) dissolved in dimethylsulfoxide (1 mM stock) was added to each culture before the experiment to a final concentration of 1 µM. Cultures were incubated for 40 min at 37 • C, CO 2 5%. Then, coverslips were submerged in a recording chamber. The recording chamber was continuously perfused at 1.5 mL/min with a solution containing (mM): 130-NaCl, 2.5-KCl, 1.5-MgCl 2 , 1.5-CaCl 2 10-glucose, 10-HEPES, 300 ± 5 mOsm, pH-7.33. An Evolve512 EMCCD camera (Photometrics UK Ltd., Marlow, UK), mounted on the SliceScope Pro 2000 (Scientifica, Uckfield, UK) microscope, was used ( Figure 6). Data acquisition was made with 3.6 fps. ROI positioning and dF/F0 analysis were performed with AxioVision Imaging System (Carl Zeiss, Jena, Germany). Calcium events were analyzed using Mini Analysis Program (Synaptosoft, Decatur, GA, USA), and pClamp (Molecular Devices, San Jose, CA, USA). The baseline of Ca 2+ -sensitive fluorescence for each cell at rest (between two neighboring calcium events) was taken as a 100% reference point. Thus, the amplitude of each SCE is represented as n% from the baseline. We evaluated the following SCE parameters: amplitude, frequency, individual integral, total integral (sum of integrals), rise slope and decay slope. The SCE calculations were carried out on a 5 min recording segment for control and 5 min recording segment for tested blockers, with the 5 min gap after the blockers were added to the recording solution. Arbitrary units for integrals and slopes were presented in tenths. Names of used chemicals and blockers with the catalogue number are showed in Table 1.

Intracellular NO Recordings
DAF FM is almost nonfluorescent until it reacts with NO; DAF FM is a dye that irreversibly binds to NO. The fluorescence quantum yield of DAF-FM is ∼0.005, but increases about 160-fold, to ∼0.81, after reacting with NO. DAF-FM Diacetate (Invitrogen, Carlsbad, CA, USA) was dissolved in dimethylsulfoxide (1 mM stock) and added to each culture before the experiment to a final concentration of 1 µM. Cultures were incubated for 40 min at 37 • C, CO 2 5%. Then, coverslips with cultures were submerged in a recording chamber. The recording chamber was continuously perfused as described above, and the microscopy setup was the same as the above (Figure 6). Data acquisition and ROI positioning were performed with AxioVision Imaging System (Carl Zeiss, Jena, Germany). The NO signal was analyzed using pClamp (Molecular Devices, San Jose, CA, USA). The calculation of DAF-FM fluorescence was carried out on a 5 min recording segment with 1 frame per 5 s. Arbitrary units for integrals and slopes were presented in tenths.

Statistical Analysis
All values are given as mean ± SEM from n-cells calculated for at least 4 separate cultures from 3 independent cell passages. For analysis, random spontaneously active cells from the field of view were taken. Data passed the Kolmogorov-Smirnov normality test; statistical analysis was performed using paired parametric t-test in Graphpad (Prism, San Diego, CA, USA). The difference between groups was considered significant at p < 0.05. No blinding was used.  Institutional Review Board Statement: The animal study protocol was approved by the Ethics Committee of Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry (protocol code 357, 04.08.2022).

Informed Consent Statement: Not applicable.
Data Availability Statement: Data available upon reasonable request.

Conflicts of Interest:
Authors declare no conflict of interest.