WNK1 Antagonist Decreased Sevourane-induced Neurotoxicity in HT22 Hippocampal Neurons via the WNK1/NKCC1 Signalling Pathway

Background: . The GABA A R plays an important role in the neurotoxicity caused by sevourane, but the mechanism of its regulation in this context is unknown. The present study aimed to reveal the WNK1/NKCC1 pathway as a regulator of the neurotoxicity caused by sevourane on the HT22 hippocampal neuron cell line. Methods: HT22 cultured hippocampal neurons were divided into three groups: control group, 4.1% sevourane treatment group for 6h, WNK-463 (specic antagonist of WNK1, 1µmol) + sevourane treatment group. Cell viability and the optimum concentration of WNK-463 were measured by MTS method. Cell apoptosis was detected by Tunel method, and the content of cleaved caspase-3 protein apoptosis factor was detected by Western blot. Pathway protein detection including the expression of WNK1, NKCC1 were detected. Calcium imaging measures intracellular calcium ion concentration and veries downstream targets. Results: The neurotoxic effects of sevourane on hippocampal neurons were observed. As setting the results of control group is “1”, compared with sevourane group, WNK-463 downregulated the protein expression of cleaved caspase-3 (1.96±0.49 vs 1.29±0.38), increased cell viability (0.68 vs 0.96) and decreased apoptosis (13.31±4.67/vision vs 7.05±3.82/vision, ***P < 0.001). Compared with the control group, sevourane treatment increased the expression of WNK1 kinase and NKCC1 protein, whereas WNK-463 reversed this increase without affecting the control HT22 cells (****P < 0.0001). Compared with the control group, sevourane exposure in HT22 cells increased intracellular Ca 2+ concentrations, while WNK-463 reversed this change (1.0 vs 2.08±1.36 vs 1.26±0.77, **P < 0.01). Conclusion: This study demonstrated a neuroprotective role of the WNK1 antagonist WNK-463 in sevourane-induced neurotoxicity. WNK-463 promoted hippocampal neuron viability and reduced the apoptosis and intracellular calcium overload caused by sevourane on HT22 hippocampal neurons, possibly via the modulation of the WNK1/NKCC1 pathway.


Background
The neurotoxic effect caused by general anaesthesia in the developing brain has caused continual concern. Numerous animal studies have found that an exposure to most general anaesthetics in mammals at their peak of development leads to a range of morphological changes [1][2][3] . These changes include neurodegenerative changes [3,4] , neural apoptosis or cell death, and impaired neurogenesis [5] . In addition, anaesthesia exposure in infancy has also been associated with memory impairment, behavioural disorders, and poor intellectual development persisting to early adulthood [6][7][8][9][10] . Repeated and prolonged exposure to general anaesthetics in the developing brain is still a high-risk factor for neuronal injury [6,11,12] . Therefore, it is of great signi cance to explore the mechanism of neurotoxicity caused by general anaesthetics in developing neurons.
The γ-aminobutyric acid type A receptor (GABA A R) plays an important role in the mechanism of general anaesthesia [13,14] . Sevo urane, a representative inhaled general anaesthetic, stimulates GABA A R to induce the ow of chloride ions (Cl -) [14] . In the adult brain, GABA A R activation triggers Clowing into cells, membrane hyperpolarization and synaptic inhibition. Conversely, in the developing brain, GABA A R activation triggers Clowing out of cells, membrane depolarization and synaptic excitation [15,16] . This developmental "switch" in GABA A R function from excitability to inhibition has been attributed to the difference in the [Cl -]i of immature versus mature neurons [16,17] . This difference in [Cl -]i at different stages results from the conversion of receptors with a dominant function from Na + -K + -Clcotransporter 1 (NKCC1) to K + -Clcotransporter 2 (KCC2) [17] .
WNK (with no lysine) kinases are Cl --sensing kinases, including WNK1, that regulate the activities of NKCC1 and KCC2 [17] . In our previous study, we found that sevo urane's neurotoxicity on the developing brain was related to the changes in the transition of GABA A R a1 and a2 [18] . This neuronal damage was decreased by the NKCC1 inhibitor bumetanide [18] . Stevens [19] also found that an NKCC1 inhibitor prevented ketamine neurotoxicity. Therefore, we hypothesized that when a WNK1 inhibitor (WNK-463) is administered, NKCC1 activity can be downregulated, the concentration of chloride ions in cells can be reduced, and the excitatory neurotoxic effects caused by sevo urane can be decreased in the developing brain.
As the research object, hippocampal neuronal HT22 cell line from mouse [20] were used to verify the above hypothesis and explore the neurotoxicity mechanisms of sevo urane on neurons, with the aim to provide new ideas for ameliorating the neurotoxicity of inhalational anaesthetics.

Cell culture
The hippocampal neuronal HT22 cell line of mouse were purchased from the American Type Culture Collection (ATCC). HT22 cells were maintained in Dulbecco's modi ed Eagle's medium (Gibco, Grand Island, USA) containing 10% foetal bovine serum (Gibco, Grand Island, USA) and 1% penicillinstreptomycin solution (Life Technologies, Carlsbad, ON, Canada) at 37°C in a humidi ed incubator with 5% CO 2 . The medium was completely changed every 72 h.
All HT22 cell lines were divided into three groups. The control group, sevo urane treatment group and sevo urane-WNK-463 group. The cell culture was exposed in an airtight plastic chamber with inlet and outlet connectors. The inlet port of the chamber was used to adjust the concentration of sevo urane (AbbVie Inc., North Chicago, IL, USA), which was connected to a sevo urane vaporizer. The outlet of the chamber was used to monitor sevo urane concentration through a gas monitor (PM 8060, Drager, Lübeck, Germany) until the target concentration was reached.
In the control group, the cell culture was grown in the chamber with humidi ed atmosphere with 5% CO 2 (95% air/5% CO 2 ) at 37°C for 6 h.
In the sevo urane group, the chamber was gassed with a concentration of sevo urane (4.1%) in the carrier gas (95% air/5% CO 2 ) for 30 min as described previously [21] . And the cell culture was put in the chamber and then kept tightly sealed for 6 h at 37°C.

MTS assay
The CellTiter 96® AQueous One Solution Cell Proliferation Assay Kit (Promega Corporation, Madison, WI, USA) is a colourimetric method for determining the number of viable cells and screening of optimum drug concentration (WNK-463) in proliferation in cytotoxicity assays according to a previous study [22] . We added WNK-463 at a certain concentration gradient of 0, 0.5, 1.0, 2.5, 5.0 µmol, in order to nd the optimum concentration of WNK-463 through the results of MTS. Move 20µl of CellTiter 96® AQueous One Solution Reagent into each well of the 96-well assay plate containing the samples in 100µl of culture medium. The culture wells with MTS reagent was placed in an incubator at 37℃for 2 hours and then recording the absorbance at 490 nm with a 96-well plate reader. The results of the control group were set at 100% and the survival rates of the other two groups were calculated. From the MTS results (absorption was detected at 490 nm using a microplate reader), we found that with the increase in WNK-463 concentration, the protective effect was gradually enhanced and reached a peak at 1.0 µmol before it decreased which was used in our subsequent experiments (P < 0.01, see supplementary le, S- Figure 1).

TUNEL assay
The terminal deoxynucleotidyl transferase (TdT) dUTP nick-end labelling (TUNEL) method was performed to label the 3'-terminus of the fragmented DNA of apoptotic cells according to the manufacturer's protocol [20] (One Step TUNEL Apoptosis Assay Kit, Beyotime, China). Phosphate-buffered saline (PBS) containing 4% paraformaldehyde was added to the three groups of cells and incubated for 1 hour and washed with 1x PBS 3 times for 5 min. After permeabilization with 1x PBS containing 0.1% Triton X-100 for 2 min on ice, the cell slices were incubated with a TUNEL working solution for 1h at 37°C in darkness. The TUNEL-positive cells were labelled by Cy3 and were examined with uorescence microscopy (Olympus IX710 Camera, Japan). The cells with green uorescence were considered apoptotic cells. To count the number of apoptotic cells, we selected the same-sized areas of each cell slice for analysis with Image Pro Plus software.

Measurement of intracellular calcium levels ([Ca 2+ ]i)
According to before the treatment method of cells after collecting, learned that, the supernatant uid to join without phenol red culture medium, Hank's balanced salt solution (HBBS) wash three times, to join the Fluo-4, AM concentration of 5 umol/L working liquid Beyotime biological co, LTD(Shanghai), incubation in 37 ℃ for 30 min, using HBSS ushing cells 3 times, and continue in HBSS incubation for 30 min, under laser confocal microscope test, excitation wavelength of 494 nm, emission wavelength of 516 nm.

Western blot analysis
The proteins including caspase-3, WNK1, NKCC1 were detected by western blot analysis. The proteins were extracted using RIPA lysis buffer (Beyotime Biotechnology, Shanghai, China). The protein concentrations were quanti ed using the BCA™ Protein Assay Kit (Pierce, Appleton, WI, USA). Equal amounts of protein were resolved over 8-12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene uoride (PVDF) membranes. The membranes were incubated in 5% non-fat milk for 1 h at room temperature and then incubated with primary antibodies overnight at 4°C. The primary antibodies were caspase-3 antibody (#9662, 1:1000 dilution), WNK1 antibody (NB600-225, 1:500 dilution) and NKCC1 antibody (#14581,1:500 dilution). The membranes were then incubated with the secondary antibodies (Abcam) for 1 h at room temperature. Positive signals were visualized by enhanced chemiluminescence (ECL) reagent (GE Healthcare, Little Chalfont, UK). The intensity of the bands was quanti ed using Image Lab TM Software (Bio-Rad, CA, USA).

Statistical analysis
All experiments were repeated three times. The results of multiple experiments are presented as the means ± standard deviations (SD). Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software Inc., San Diego, CA, USA). We used one-way variance analysis (ANOVA) to compare the overall difference, and posthoc test analysis to compare the differences between two groups. The Pvalues were calculated using LSD, posthoc test analysis. A P-value of <0.05 was considered to indicate statistical signi cance.

WNK-463 alleviated the sevo urane-induced reduction in cell activity
The MTS method was used to detect hippocampal neuronal HT22 cell activity. The l control group was 100% and that of the sevo urane group was 75.40% (P < 0.0001, S- Figure 2). According to the results of different concentration of WNK-463 on MTS (S- Figure 1), 1.0 µmol WNK-463 was the optimum concentration which was used in the next experiment. Compared to control group, sevo urane obviously reduced HT22 cell activity, and 1 µmol WNK-463 showed a signi cant protective effect (P = 0.006, Figure  1).

WNK-463 decreased the HT22 cell apoptosis caused by sevo urane
The results of TUNEL, in the control group, the positive cells were 4.44±2.53/mm 2 through the observation of the eld of uorescence microscope, whereas in the sevo urane group the positive cells were 13.31±4.67/vision. WNK-463 decreased the positive apoptosis cells compared with sevo urane group (7.05±3.82/vision vs 13.31±4.67/vision) (P<0.001, Figure 2A, B, C, D).
Through the detection of caspase-3, the results indicated that cleaved caspase-3 protein was signi cantly upregulated in cells after sevo urane treatment compared to the control cells. Moreover, cleaved caspase-3 activity was signi cantly reduced after WNK-463 treatment compared to the sevo urane group (P = 0.034, Figure 2 F).

WNK-463 inhibited the WNK1 kinase concentration increase caused by sevo urane
To explore the mechanism of apoptosis and the protective mechanism of the WNK1 inhibitor, the expression levels of WNK1 kinase proteins were investigated in HT22 cells after treatment with sevo urane. The results indicated, that the WNK1 kinase proteins in cells in the sevo urane group were signi cantly upregulated (P < 0.001; Figure 3). There was no signi cant difference between the WNK-463 group and the control group. However, after exposure to sevo urane, the WNK1 concentration was signi cantly increased, and this high expression could be inhibited by WNK-463.

WNK-463 inhibited the NKCC1 concentration increase caused by sevo urane
To verify the role of WNK1/NKCC1 signalling pathway, the expression levels of NKCC1 proteins were investigated in HT22 cells after treatment with sevo urane. The results indicated, compared with those in the control group, that the NKCC1 proteins in cells in the sevo urane group were signi cantly upregulated after exposure to sevo urane (P = 0.000), and this high expression could be inhibited by WNK-463. (P = 0.000; Figure 4).

WNK-463 reduced the calcium overload in HT22 cells after exposure to sevo urane
Calcium ion imaging technology (Fluo-4) was used to detect the basic values of calcium in the cells. The Fluo-4 images showed that the uorescence intensity of the sevo urane treatment group was signi cantly enhanced compared to that of the control group. However, compared with the sevo urane group, the WNK-463 treatment group showed a weak uorescence intensity differ from the control group ( Figure 5A, B, C), indicating that the WNK1 inhibitor reduced the calcium overload caused by sevo urane. The results indicated, compared with those in the control group, that the calcium ion in cells in the sevo urane group were signi cantly upregulated (P < 0.001; Figure 5D).

Discussion
Our study showed that the hippocampal neuronal HT22 cell had lower cell activity and higher apoptosis rate after exposure to sevo urane. These changes related to the overload of intracellular calcium and high expression of WNK1 kinase. However, the WNK1 inhibitor WNK-463 could improve cell activity, decrease the apoptosis caused by sevo urane, downregulate the expression of WNK1/NKCC1 and alleviate calcium overload. Therefore, we propose that the WNK1/NKCC1 signalling pathway may play a major role through intracellular calcium overload in sevo urane-induced neurotoxicity, which may be a potential protective mechanism against sevo urane-related neurotoxicity.
Most general anaesthetics are excitants of GABA A R and inhibitors of the N-methyl-D-aspartate (NMDA) receptor. Currently, many studies demonstrated that neurotrophic apoptosis [2,3] , mitochondrial damage and accumulation of ROS [23] , in ammatory factors [16,24] , the excitatory neurotoxicity [3] , the effects on the regeneration of neurons, changes in receptor subtypes [18] were the possible mechanisms of general anaesthetics' neurotoxicity [19][20][21][22][23][24][25] . In the present study, we found that exposure of sevo urane caused neuron apoptosis and decreased cell vitality, which was consistent with the above conclusions.
In the present study, we found that WNK-463, an inhibitor of WNK1 kinase, may play a protective role in developing neurons. WNK1 kinase is a Cl --sensitive serine threonine protein kinase expressed in the central nervous system. Affected by intracellular chloride ions, this kinase can directly inhibit KCC2 activity and enhance NKCC1 activity [17] . According to the level of intracellular chloride ion concentration, WNK1 kinase negatively regulates intracellular Clconcentration by inhibiting its downstream KCC2 and raise NKCC1 channel protein, maintaining intracellular osmotic pressure. WNK1 kinase directly on KCC2/NKCC1. It also activates the SPS1-related proline/alanine rich kinase (SPAK) and the related oxidative stress response protein OSR-1 (OSR-1), namely, those ghting-regulating the endogenous neural cells [Cl -] through the WNK1-SPAK-OSR-KCC2 /NKCC1 pathway, then the oxidative stress affected by the oxidative stress of SPS1-related proline/alanine rich kinase (SPAK) and those related to it. Furthermore, it affects the depolarization of GABA A R during excitation of developing neurons, and transforms to inhibitory function [25][26][27] . Friedel [28] found that in the process of neuron development, GABA A R is open, which transforms from the excitatory effect of immature neurons to the inhibitory effect of mature neurons, and an important guaranteed factor is that as KCC2 transports Clsynergistically out of cells, [Cl-]i decreases. As an inhibitor of NKCC1, bumetanide can attenuate the increase in apoptosis induced by sevo urane [18,29] . Stevens [19] also found that the NKCC1 inhibitor alleviated the decline in the learning and memory function of newborn rats caused by ketamine. Therefore, we propose that as a cotransporter of NKCC1 and KCC2, WNK1 can affect the transformation of NKCC1-mediated Clow and affect the excitability of GABA A R in the immature brain. In the present study, we found that sevo urane caused a signi cant decrease in cell activity and an increase in apoptosis and cleaved caspase-3 expression in the hippocampal neuronal HT22 cells of neonatal rats. After the administration of the WNK1 inhibitor WNK-463, cell activity was improved, and apoptosis and the expression of cleaved caspase-3 were decreased.
These results demonstrated that the WNK1 inhibitor was a protective factor against the neuroexcitatory toxicity of sevo urane in developing neurons, which is consistent with our hypothesis.
We also found that sevo urane exposure caused calcium overload in HT22 cells. In addition, the calcium current generated by the NMDA receptor directly activates voltage-dependent Ca 2+ channels (VDCCs), leading to increased intracellular Ca 2+ and calcium overload. Calcium is an important regulator, and intracellular Ca 2+ overload can lead to apoptosis through the activation of apoptosis-related enzymes, such as phospholipase, protease, and nucleic acid enzymes, as well as mitochondrial and cell membrane damage, and the generation of a large number of free radicals [30] . In the present study, the results of the Fluo-4 imaging of intracellular calcium showed a signi cant accumulation of Ca 2+ in HT22 cells after treatment with sevo urane. WNK1 inhibitor administration signi cantly decreased the calcium density, which demonstrated that the WNK1 inhibitor may in uence NKCC1 and then GABA A R excitability and intracellular Ca 2+ overload in HT22 cells after sevo urane exposure.
There were some limitations in this study. Firstly, it was only in vitro. However, this study is just the rst step to test our hypothesis. On the basis of this study, we will perform the further experiments to test the signaling pathway in vivo. Secondly, WNK463 is a speci c compound. It is still far from being used as a protective agent in vivo. However, the results of this study will help us to understand the possible mechanism of this pathway in sevo urane neurotoxicity. So as to provide new ideas for us to nd more practical and transformable protective agents on the WNK1/NKCC1/GABA A R pathway.

Conclusions
The results of this study indicate that sevo urane can inhibit the proliferation of HT22 cells and induce cell apoptosis, while WNK1 kinase antagonist can reverses these changes and produces protective effects. The WNK1/NKCC1 signalling pathway plays an important role. Reducing intracellular calcium in ux may be a change in the downstream of this signalling pathway.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
Not applicable.

Competing interests
The authors declare that they have no competing interests.