Inhibition of Calcium‑Sensing Receptor Alleviates Chronic Intermittent Hypoxia‑Induced Cognitive Dysfunction via CaSR‑PKC‑ERK1/2 Pathway

Obstructive sleep apnea–hypopnea syndrome (OSAHS) is typically characterized by chronic intermittent hypoxia (CIH), associated with cognitive dysfunction in children. Calcium-sensing receptor (CaSR) mediates the apoptosis of hippocampal neurons in various diseases. However, the effect of CaSR on OSAHS remains elusive. In the present study, we investigated the role of CaSR in CIH-induced memory dysfunction and underlying mechanisms on regulation of PKC-ERK1/2 signaling pathway in vivo and in vitro. CIH exposures for 4 weeks in mice, modeling OSAHS, contributed to cognitive dysfunction. CIH accelerated apoptosis of hippocampal neurons and resulted in the synaptic plasticity deficit via downregulated synaptophysin (Syn) protein level. The mice were intraperitoneally injected with CaSR inhibitor (NPS2143) 30 min before CIH exposure and the results demonstrated CaSR inhibitor alleviated the apoptosis and synaptic plasticity deficit in the hippocampus of CIH mice. We established intermittent hypoxia PC12 cell model and found that the activation of CaSR accelerated CIH-induced PC12 apoptosis and synaptic plasticity deficit by upregulated p-ERK1/2 and PKC. Overall, our findings revealed that CaSR held a critical function on CIH-induced cognitive dysfunction in mice by accelerating hippocampal neuronal apoptosis and reducing synaptic plasticity via augmenting CaSR-PKC-ERK1/2 pathway; otherwise, inhibition of CaSR alleviated CIH-induced cognitive dysfunction


Introduction
Obstructive sleep apnea-hypopnea syndrome (OSAHS) is mainly manifested by recurrent obstruction of the upper airway during sleep, contributing to hypopnea and sleep apnea. OSAHS can occur at any age, especially in preschool children [1]. Multiple systems were reported to be damaged by OSAHS in adults, including the cardiovascular system, respiratory system, and muscular system [2]. The others and our research have demonstrated the nervous system impairment, especially cognitive dysfunction, was observed more frequently in pediatric OSAHS [3,4]. Chronic intermittent hypoxia (CIH) is recognized as the main feature of OSAHS. Large amounts of researches indicated that CIH exposure impaired cognitive function by inducing apoptosis and the change of synaptic plasticity in hippocampus [5,6]. However, the underlying mechanisms remain unclear.
Calcium-sensing receptor (CaSR), a member of G-protein coupled receptors superfamily, senses extracellular levels of calcium ions and is essential for regulating systemic calcium homeostasis [7]. Apart from parathyroid glands, kidney, and bone, CaSR is also expressed in the hippocampus, and holds a critical function on neuronal proliferation and differentiation during early neural development [8]. Large amounts of researches demonstrated that CaSR expression was upregulated in rat Huiya Ying, Zilong Zhang, and Wei Wang contributed equally to this work. brains, especially in CA1 and CA3 of the hippocampus, after global or focal cerebral ischemia/reperfusion [9,10]. Moreover, CaSR, activated by extracellular calcium or CaSR agonists, increases the concentration of intracellular calcium ([Ca 2+ ]i), which then induces the apoptosis of hippocampal neurons [11]. However, the mechanisms of CaSR-induced apoptosis are not well understood.
Synaptic plasticity represents the information storage process of synapses and is considered to be related to learning and memory. Recent studies have shown that CaSR may exist widely in synapses, and CaSR plays a significant role in synaptic plasticity. Additionally, further research demonstrated that CaSR was likely to modify dendritic structure when synapses are involved in memory and learning [12]. The distribution of CaSR in nerve endings is related to the synthesis and release of neurotransmitters, participating in the regulation of synaptic response and synaptic plasticity. A recent study revealed that CaSR, which promotes dendritic development and axonal growth, is regulated via ERK signaling pathway [13]. Large amounts of evidence indicates that hippocampal synaptic plasticity is impaired after IH exposure [14,15], which may be the basis of memory dysfunction in OSAHS. Nevertheless, it is unknown whether CaSR has an impact on cognitive function by regulating synaptic plasticity under intermittent hypoxic conditions. CaSR can sense increased extracellular levels of calcium ions, which lead to calcium influx, resulting in increased [Ca 2+ ]i [16]. Intracellular calcium overload is a critical step for the apoptosis of hippocampal neurons undergoing H/R injury [17]. In addition, CaSR has been demonstrated to activate Ca 2+ -dependent protein kinase C (PKC) and induce the phosphorylation of extracellular signal-regulated kinases 1/2 (ERK1/2) [18]. PKC, a family of serine/threonine protein kinases, participates in many cellular responses such as cell proliferation [19]. To date, conventional PKC isoforms that have been identified require Ca 2+ influx for their activation [20]. ERK1/2, as a subfamily of mitogen-activated protein kinases (MAPKs), is generally thought to be involved in cell apoptosis, proliferation, and synaptic plasticity [21,22]. Further studies indicated that CaSR activation aggravates the apoptosis and the change of synaptic plasticity in hippocampal neurons during H/R [11,23] and the [Ca 2+ ]i overload induced by the activation of CaSR contributes to the apoptosis of PC12 cells in oxygen and glucose deprivation/reoxygenation [24]. However, it remains unclear whether CaSR holds a function on CIH-induced apoptosis and synaptic plasticity in hippocampal neurons by PKC-ERK1/2 signaling pathway. Therefore, our present study aimed to investigate the role of CaSR in neurocognitive deficits of childhood OSAHS and the underlying mechanisms.

Experimental Animals and Grouping
All experimental protocols were approved by the Animal Research Committee of Wenzhou Medical University (wydw 2021-0628). Forty healthy male C57BL/6 mice (18-22 g; 4 weeks) were purchased from the Experimental Animal Center (License no. SCXK 2018-0017) of Wenzhou Medical University, Zhejiang Province, China. The mice were randomly divided into four groups: control group (C), chronic intermittent hypoxia model group (CIH), CIH with CaSR inhibitor (NPS2143) treatment group (CIH + NPS2143), and C with NPS2143 treatment group (C + NPS2143). The mice in C and CIH group were not subjected to NPS2143 injection. Mice in CIH + NPS2143 and C + NPS2143 groups were intraperitoneally injected with NPS2143 (3 mg/kg, GLPBIO, Cat. No: GC169431) 30 min before CIH exposure everyday [25]. All animals used in the experiment were cared for in accordance with the ethical guidelines on animal experimentation of the Laboratory Animals of China National Institutes of Health.

Animal Models
The CIH model in mice is established by using a computercontrolled chamber with a nitrogen/oxygen delivery system to establish hypoxic-reoxygenation conditions that we previously designed [5]. In simple terms, in an intermittent hypoxic (IH) chamber, the oxygen concentration fluctuated from 9.0% ± 1.5% to 21% ± 0.5% in 90 s per cycle, 40 cycles per h, 7 h a day (from 8:00 to 15:00) during the light cycle, for 4 weeks. The CIH and CIH + NPS2143 groups were placed in the IH chamber, and the C and C + NPS2143 groups were placed in a cabin filled with compressed air.

8-Arm Radial Maze Test
An eight-arm radial maze was used to evaluate spatial memory of mice after CIH. As reported previously [26], 2 days before the CIH exposure, the mice were restricted to eat until their weight was reduced to 80-85% of their baseline body weight. After acclimation experiment, mice were trained following CIH at 19:00 every night, once a day for 7 consecutive days. The 8-arm radial maze test was performed instantly after 7 h of IH exposure on the last day. During the test, mice covered with an opaque box were placed in the center of the maze for 15 s, and then released for free feeding when the box was removed. The hind paws of mice completely enter into one arm is considered that arm entry was complete. A trial was terminated immediately when all pellets were consumed or 5 min had elapsed. The performance of mice was evaluated according to the number of error choices as follows: working memory errors (WME: re-entry into arms where food pellet had been eaten), reference memory errors (RME: entry into a never-baited arm), and the total errors (TE: mice enter a baited arm and eat the pellet as the right choice, otherwise it was counted as an error).

Animal Sample Collection
After the 8-arm radial maze test, five mice randomly selected from each group were anesthetized with 2% sodium pentobarbital (45 mg/kg, Sigma, USA) via intraperitoneal injection. Their brain tissues were dissected into left and right hemispheres immediately on ice, and placed into 4% paraformaldehyde for Nissl staining, TUNEL assay, and immunohistochemistry. Subsequently, the remaining five mice in each group were dissected into left and right hippocampus, and were frozen at − 80 °C for Western blotting analysis.

Nissl Staining
Following dewaxing with dimethylbenzene, 4-μm-thick brain sections were treated with 100%, 95%, 85%, and 75% gradient alcohol by turn. After rinsing with water, slices were immersed in Nissl staining solution (Beyotime Biotechnology, China) for 20 min at 37 °C. Then, the tissue sections were dehydrated in 95% alcohol twice (5 min per time) and permeabilized with xylene twice (15 min each time). The images of hippocampal CA1 regions were captured by light microscope with 400 × magnification (OLYM-PUS 1X70-SIF2, Japan) and ImageJ software were applied to quantify the numbers of neurons. The size and number of Nissl body can reflect the degree of neuronal injury.

Terminal Transferase-Mediated dUTP Nick End Labeling (TUNEL) Staining
TUNEL assay was used to detect apoptosis according to the manufacturer's recommended procedure. The tissue sections were dewaxed, sealed, and treated with TUNEL reaction mixture for 30 min and POD for 60 min. The slices were then incubated with DAB, counter-stained with a hematoxylin dye, and stained on Eukitt (Sigma-Aldrich). Apoptotic cells are those with brown granules in the nucleus, namely TUNEL positive staining cells. Under a 400 × magnification light microscope, 5 sections in each group were randomly selected with 3 fields in each section. The positive staining cells in the hippocampus CA1 region were counted, and the apoptosis index was determined. Apoptosis index (AI) was calculated as number of TUNEL positive cells/total number of cells × 100 to evaluate apoptosis.

Immunohistochemistry
Brain tissue sections were dewaxed in xylene, rehydrated through a serial alcohol gradient, and washed in PBS. And then, the brain tissue sections were heated by microwave and cooled naturally to repair the antigen in 10 mmol/L citrate buffer with pH 6.0. The primary antibody rabbit anti-CaSR (1:200, Affinity), rabbit anti-p-ERK1/2 (1:200, Cell Signaling), and anti-synaptophysin (Syn) (1:100, Affinity) were used for immunostaining and brain tissue sections were incubated with above antibody at 4 °C overnight, and, after incubation with the appropriate biotin-conjugated secondary antibody, stained with hematoxylin and eosin. Brain tissue sections were visualized through microscopy. Under a 400 × magnification, 5 sections in each group were randomly selected with 3 fields in each section. The integrated density in the hippocampus CA1 region were calculated via ImageJ software (National Institutes of Health).

Tissue Protein Extraction
Quantitative expression level of CaSR, p-ERK1/2, ERK1/2, Syn, Bcl-2, and cleaved Caspase-3 proteins in the hippocampus were measured by western blots. Hippocampal tissues of each group were homogenized in ice-cold radio immunoprecipitation assay (RIPA) lysis buffer containing 1 mM phenylmethanesulfonyl fluoride (PMSF) (Beyotime Biotechnology, China), followed by centrifugation at 12,000 rpm and 4℃ for 15 min to obtain the supernatant. After, supernatant was measured by BCA Protein Assay Reagent Kit (Beyotime Biotechnology, China) to concentrate the protein for western blot assay.

Cell Counting Kit 8 (CCK-8)
Cell viabilities in every group were detected via a CCK-8 kit (DojinDo, Japan) after IH exposure for 9 h. In addition, cytotoxicity of CaSR inhibitors (NPS2143) and ERK inhibitors (U0126) was determined by CCK8. The cells treated with 2.0 × 10 5 cells/ml were planted on 96-well plates and cultured at 37 °C in a humidified atmosphere (95% air and 5% CO 2 ) for 24-48 h depended on cell growth rate. After treatment with reagents of different concentrations, the cells rinsed with PBS were added CCK8 reaction mixture (10 μl CCK-8 stock solution, 90 μl DMEM medium) to each well and incubated at 37℃ for 1 h. Cell viability was determined by measuring absorbance at 450 nm with a fluorometer.

Statistical Analysis
All data are normally distributed, expressed as means ± SD. The RME, WME, and TE in each group with seven-day training were compared via repeated measures two-way analysis of variance (ANOVA), and one-way ANOVA analysis was used to evaluate spatial memory (RME, WME, and TE) on the testing day, whereas all other comparisons between groups with different treatments were analyzed by analysis of one-way ANOVA, followed by LSD or Dunnett's T3 posthoc analysis. All statistical analyses were conducted using SPSS software (IBM Corp., Armonk, NY, USA). The statistical significance level was set at 0.05 for all comparisons.

Inhibition of CaSR Alleviated CIH-Induced Cognitive Dysfunction of Mice
Eight-arm radial maze test is a widely recognized method to detect spatial memory and learning function. In this study, we used eight-arm maze to evaluate the effect of CIH on learning and memory function. Repeated measures ANOVA showed that WME, RME, and TE were gradually reduced in each group with 7-day training ( Fig. 1a-c). Following the learning and memory processes, these data quantify the formation of spatial memory throughout training. In the testing day, the number of WME, RME, and TE increased significantly after intermittent hypoxia exposure(all p < 0.001), indicating that CIH induced the memory impairment, while inhibition of CaSR decreased significantly the above errors (WME, p = 0.007; RME, p = 0.017, TE, p = 0.008; CIH + NPS2143 vs CIH group, Fig. 1d). We found from the data in behavioral test that inhibition of CaSR significantly alleviated CIH-induced cognitive dysfunction.

Inhibition of CaSR Alleviated Neuronal Injury and Apoptosis in CA1 Region of the Hippocampus with CIH Exposure
Neural apoptosis in hippocampus is one of the important mechanisms associated with cognitive dysfunction. In our study, Nissl staining was analyzed to assess neuronal viability in the hippocampus. A decreasing neuronal viability was showed in the hippocampal CA1 region in CIH group (p < 0.001; Fig. 2a-b). In addition, TUNEL staining was conducted to detect the apoptosis of neurons. Compared with control mice, CIH resulted in an increase of TUNEL-positive neurons in the hippocampal CA1 region with nuclear concentration (p < 0.001; Fig. 2c-d). On the contrary, after treatment with CaSR inhibitor, the neuronal injury and apoptosis rate in hippocampal CA1 region were significantly decreased, indicating that inhibition of CaSR alleviated CIH-induced neuronal apoptosis.
In order to further clarify the severity degree of CIHinduced neuronal apoptosis, we examined the expression level of cleaved Caspase-3, cleaved Caspase-7, cleaved Caspase-9, Bax, and Bcl-2 via western blot assay. Pro-enzyme of caspase-9 is cleaved into the active dimer form by apoptotic signals-induced cytochrome C. As an initiator caspase of the family of caspases, caspase-9 activates downstream executioner caspases, including caspase-3 and caspase-7, to initiate cell apoptosis [28]. Therefore, the expression levels of cleaved Caspase-9, cleaved Caspase-7, and cleaved Caspase-3 can reflect the severity of apoptosis. Bax, a proapoptotic protein in the Bcl-2 family, induces the release of cytochrome C to mediate cell apoptosis under stress. The pro-apoptotic protein is inhibited by Bcl-2, and the relative ratio between the Bcl-2 and the Bax is a marker to indicate whether the cells will survive or undergo apoptosis [29]. We detected the level of cleaved Caspase-3, -7, and -9, and the ratio between Bcl-2 and Bax after 4 weeks of CIH exposure. As shown in Fig. 2e-i, compared with C group, the level of cleaved Caspase-3, cleaved Caspase-7, cleaved Caspase-9 was upregulated (all p < 0.001) and the ratio between Bcl-2 and Bax was decreased in CIH group (p < 0.001). In contrast, these changes were reversed following intervention with CaSR inhibitor. These data suggested that inhibition of CaSR could significantly decrease CIH-induced cell apoptosis.

Inhibition of CaSR Attenuated Synaptic Plasticity Deficit in the Hippocampus of Mice Following CIH
Synaptophysin (Syn), as an important synaptic protein, holds a critical function on the regulation of synapse plasticity.
To determine the effect of CIH on synaptic plasticity, Syn expression levels in hippocampus were measured via immunohistochemistry (IHC) and western blot (WB) assay. The IHC results revealed that CIH exposure decreased Syn staining in hippocampal CA1 region (p < 0.001, Fig. 3a-b). Meanwhile, WB analysis showed that the levels of the Syn protein in CIH group were significantly downregulated compared with that in the control group (p < 0.001, Fig. 3c-d). However, Syn staining was increased and Syn level was upregulated significantly in mice treated with NPS2143 following Fig. 1 Inhibition of CaSR alleviated CIH-induced cognitive dysfunction of mice. a-c The numbers of spatial memory errors (WME, RME, and TE) on different training days via 8-arm radial maze test, respectively. N = 10 mice per group. d The numbers of WME, RME, and TE in each group on the testing day. Graphs display normalized results of mean ± SD for n = 10. WME, working memory errors; RME, reference memory errors; TE, total errors. one asterisk (*) p ≤ 0.01 vs. C, two asterisks (**) p ≤ 0.05 vs. C; one number sign ( # ) p ≤ 0.01 vs. CIH, two number signs ( ## ) p ≤ 0.05 vs. CIH; one ampersand ( & ) p ≤ 0.01 vs. NPS2143, two ampersands (. && ) p ≤ 0.05 vs. NPS2143 CIH exposure (p = 0.049, p < 0.001; CIH + NPS2143 vs CIH group, Fig. 3b and d). These findings indicated that inhibition of CaSR might attenuate the synaptic plasticity deficit in the hippocampus after CIH exposure.

CaSR Was Involved in Hippocampal Neuronal Apoptosis and Synaptic Plasticity Deficit Induced by CIH, Accompanied by Decreased p-ERK1/2
Large amounts of evidence demonstrated CaSR was associated with apoptosis [30]. In our present study, CaSR inhibitor was used to explore the role of CaSR in CIH-induced hippocampal neuronal apoptosis. In Fig. 1d, inhibition of CaSR improved spatial memory function with reduced RME, WME, and TE (p = 0.007, 0.017, 0.008; CIH + NPS2143 vs CIH group). Nissl staining was performed to determine the neuronal viability and showed that hippocampal neurons in the CIH + NPS2143 group maintained a better functional status than those in the CIH group ( Fig. 2a-b). In addition, it was demonstrated that cell apoptosis was significantly alleviated in CIH + NPS2143 group with decreased TUNEL-positive neurons (p < 0.001, Fig. 2c-d), downregulated cleaved Caspase-3, cleaved Caspase-7, cleaved Caspase-9 (p = 0.018, p = 0.012, p < 0.001) and upregulated ratio between Bcl-2 and Bax ( p < 0.001), compared with the CIH group (Fig. 2ei). At the same time, we also found that treatment with NPS2143 significantly alleviated synaptic plasticity deficit in the mice following CIH exposure, manifested by upregulated Syn protein level (Fig. 3a-d).
ERK1/2 was involved in cell proliferation and synaptic plasticity [31,32]. In our study, CaSR and p-ERK1/2 expression in hippocampal CA1 region was determined by immunohistochemistry. As shown in Fig. 4a-c, compared with C group, the level of CaSR and p-ERK1/2 was elevated in hippocampal CA1 region (all p < 0.001), while in the CIH + NPS2143 group, CaSR and p-ERK1/2 were downregulated (p = 0.002, p < 0.001, CIH + NPS2143 vs CIH group). To further investigate the changes of CaSR and ERK1/2 in hippocampus, CaSR and p-ERK1/2 protein expression via western blots was determined. These data in Fig. 4d-f revealed CIH increased CaSR and p-ERK1/2 expression in hippocampus (all p < 0.001). Meanwhile, we found NPS2143 pretreatment decreased the above protein levels (p = 0.004, 0.013), while ERK1/2 was no difference in different groups (p = 0.072). These data indicated CaSR was involved in hippocampal neuronal apoptosis and synaptic plasticity deficit induced by CIH, accompanied by increased p-ERK1/2, while inhibition of CaSR alleviated CIH-induced neuronal apoptosis and synaptic plasticity deficit, with decreased p-ERK1/2.

IH Induced Caspase-3-Dependent Apoptosis in PC12 Cells
To further elucidate the underlying mechanism of CIHinduced apoptosis in hippocampus, we established the IH (hypoxia for 1 h/reoxygenation for 0.5 h reoxygenation) model in PC12 cells according to our previous methods [27]. In the experiment, cell viability was determined by CCK8 assay and the level of cleaved Caspase-3 was detected via western blots. As shown in Fig. 5, IH induced cell injury in an IH cycle-dependent manner. Compared to the control group, cell viability was gradually decreased from 2 to 6 cycles (IH2-6; all p < 0.001; Fig. 3a) and the level of cleaved Caspase-3 was increased for up to 4 cycles (IH4; all p < 0.001; Fig. 3b). IH for 6 cycles provided maximal cell injury compared to less cycles of IH. Therefore, hypoxia/ reoxygenation for 6 cycles (a total of 9 h) was selected for subsequent experiments. In addition, the data suggested IH induced Caspase-3-dependent apoptosis in PC12 cells.

CaSR Mediated PC12 Cell Apoptosis Induced by IH
To investigate the role of CaSR in IH-induced cell apoptosis, NPS2143, an inhibitor of CaSR, was selected to pretreat PC12 cells. Firstly, according to previous studies [33], several different concentrations of inhibitors in our experiment were selected. Cell toxicity of the inhibitors was detected via CCK8 assay, and then the expression level of CaSR was determined by western blots. In Fig. 6a, CCK8 assay showed no apparent cytotoxicity on PC12 cells for up to 10 μM of NPS2143 (NPS2143(10)), but cell viability in C + NPS2143(15) group was significantly decreased (p = 0.006), compared the other groups, which suggested that 15 μM of NPS2143 was injury to PC12 cells. In Fig. 6b-c, western blot assay showed that various concentrations of NPS2143 decreased the level of CaSR in PC12 cells induced by IH exposure, and pretreatment with 5 μM NPS2143 (IH + NPS2143(5)) provided maximal inhibition compared to lower concentrations (2.5 μM; IH + NPS2143(5) vs IH + NPS2143(2.5) group, p = 0.011). Moreover, there was no statistically significant difference in inhibitory effect compared with higher concentration of NPS2143 (10 μM; IH + N5 vs IH + N10 group, p = 0.257). Therefore, pretreatment of NPS2143 at 5 μM was selected for subsequent experiments in IH.
We investigated whether CaSR held a critical function on PC12 cell apoptosis with IH exposure. Immunofluorescence assay and western blot (Fig. 6) demonstrated that CaSR and p-ERK1/2 expression was increased in PC12 cells after IH (all p < 0.001), followed by increasing cleaved Caspase-3, cleaved Caspase-7, and cleaved Caspase-9 protein level and decreasing ratio between Bcl-2 and Bax (all p < 0.001), which was consistent with the results in vitro. The above makers of PC12 cells pretreated with NPS2143 were significantly reversed (IH + NPS2143 vs IH group, all p < 0.05). As shown in Fig. 6g and o, the level of PKC was upregulated (p < 0.001) in CIH-exposed PC12 cells, while NPS2143 pretreatment downregulated the level of PKC (p = 0.003). Collectively, these data suggested that CaSR played an important role in IH-induced cell apoptosis by increasing p-ERK1/2 and PKC, while inhibition of CaSR alleviated the IH-induced cell apoptosis via decreased p-ERK1/2 and PKC.

CaSR Modulated the Synaptic Plasticity in PC12 Cells with IH Exposure
Synapse-associated proteins Syn and GAP-43 play an important part in the regulation of synapse plasticity. Syn and GAP-43 expression levels in PC12 cells were measured via immunofluorescence (IF) and WB assay. The IF results revealed that IH exposure decreased Syn level in hippocampal CA1 region (p < 0.001, Fig. 7a-b). Meanwhile, WB analysis showed that the levels of the Syn and GAP-43 protein in IH group was significantly downregulated compared with that in the control group (p = 0.007; p = 0.003, Fig. 7c-e). However, Syn and GAP-43 levels were upregulated significantly in mice treated with NPS2143 following IH exposure (p = 0.033, p = 0.019; CIH + NPS2143 vs CIH group). These findings indicated that inhibition of CaSR might attenuate synaptic plasticity deficit in the hippocampus after IH exposure.

Phosphorylated ERK1/2 and PKC Were Involved in the IH-Induced Cell Apoptosis and Synaptic Plasticity Deficit
To identify whether p-ERK1/2 held a critical function on IHinduced cell apoptosis and synaptic plasticity deficit, U0126, a potent and selective MEK inhibitor, which can inhibit ERK through MEK/ERK pathway [34], was selected to pretreat PC12 cells. Firstly, according to previous research [35], several different concentrations of inhibitors in our experiment were selected. The toxicity of inhibitors to cells was detected via CCK8 assay, and then the protein level of p-ERK1/2 was determined by western blots. In Fig. 8a, CCK8 assay showed no apparent cytotoxicity on PC12 cells with the concentrations from 5 to 15 μM of U0126 under normal cell viability (all p > 0.05). In Fig. 8b, western blot assay showed that various concentrations of U0126 decreased the level of p-ERK1/2 in PC12 cells induced by IH exposure, and pretreatment with 10 μM U0126 (IH + U0126(10)) provided maximal inhibition compared to higher concentrations (IH + U0126(10) vs IH + U0126(5) group, p < 0.001; IH + U0126(10) vs IH + U0126(15) group, p = 0.635). Therefore, pretreatment of U0126 at 10 μM was selected for subsequent experiments in IH. As Fig. 8c-d shown, cleaved Caspase-3 expression was decreased in PC12 cells with treatment of U0126 after IH, compared with IH group (p = 0.002), but there was no statistical difference in the level of CaSR (p = 0.489). Chelerythrine chloride (CHE) is a potent inhibitor of PKC. In our previous study, CHE (10 µM) was selected in IH-induced PC12 cell injury [36]. In order to identify PKC was involved in the IH-induced cell apoptosis, the cells in C + CHE and IH + CHE group were pretreated with CHE (10µ mol/L) for 0.5 h and respectively cultured in an intermittent air or intermittent hypoxia incubator. As Fig. 8c-e and h shown, cleaved Caspase-3 expression was decreased in PC12 cells with treatment of CHE after IH, while the level of synaptic associated protein Syn was increased compared with IH group (p = 0.013; p < 0.001), but there was no statistical difference in the level of CaSR (p = 0.316). In addition, compared with IH group, p-ERK1/2 and PKC expression was downregulated (p = 0.02, p < 0.001) in IH + CHE group, while the level of PKC was no change (p = 0.549) and p-ERK1/2 was downregulated in IH + U0126 group (p < 0.001, Fig. 8f-g). All the data indicated that p-ERK1/2 and PKC were involved in the IH-induced cell apoptosis and synaptic plasticity deficit.

Discussion
Our present study was the first to investigate that CaSR held a critical function on CIH-induced cognitive dysfunction of mice. Herein, we provided the evidence that CaSR exerted neuronal apoptosis and synaptic plasticity deficit by elevating PKC and ERK1/2 phosphorylation levels in hippocampus. Therefore, these data not only revealed the underlying mechanisms of CIH-induced neuronal apoptosis and synaptic plasticity in hippocampus but also further demonstrated the important role of CaSR in CIH. This study may provide a new target for drug selection in the treatment of OSAHS.
Spatial memory, responsible for recording a course to a location and recalling the location of an object, is a manifestation of cognitive function [37]. A variety of studies have demonstrated that mice showed deficits in spatial memory after IH exposure [5,38]. In our present study, we simulated memory impairment caused by OSAHS via a model of spatial memory dysfunction following CIH exposure in mice. The spatial memory was examined via an 8-arm radial maze on growing mice subjected to CIH. Consistent with previous studies [39], our results suggested that CIH impaired the mice spatial memory function, evidenced by more working and reference memory errors in CIH mice (Fig. 1a). Hippocampus CA1 region and superior frontal gyrus were identified as the main brain area of learning and memory [40]. Our recent study and others have demonstrated that the hippocampus of rats was significantly damaged by CIH exposure, especially the hippocampus CA1 region [41,42], and the present study also supports the result. These findings were in agreement with previous studies [14]. Our previous study found that CIH-induced cognitive dysfunction was associated with hippocampal apoptosis and we detected the neuronal viability in hippocampus CA1 region via Nissl staining and the apoptosis cells by TUNEL assay for further confirmation, and the present results were consistent with other studies [43,44]. Furthermore, we found CIH exposure increased cleaved Caspase-3, -7, and -9 protein expression and decreased the ratio between Bcl-2 and Bax protein expression of the mice. Thus, we suggested that CIH induced memory dysfunction of mice by promoting neuronal apoptosis in hippocampal CA1 region.
Large amounts of evidence demonstrated that CaSR played a critical role in neuronal proliferation and differentiation during early neural development [22,45]. In our present study, CaSR upregulated in CIH-induced neuronal apoptosis and synaptic plasticity deficit in hippocampus, contributing to the cognitive dysfunction in mice. To gain further insights into the association between the activation of CaSR and neuronal apoptosis and synaptic plasticity during IH exposure, NPS-2143, a specific CaSR inhibitor, was used to inhibit the expression of CaSR in mice subjected to IH. We found that CaSR inhibition attenuated neuronal apoptosis via Nissl and TUNEL staining, following downregulated cleaved Caspase-3, -7, and -9 level and upregulated the ratio between Bcl-2 and Bax. Meanwhile, CaSR inhibition also improved synaptic plasticity in the hippocampus, as evidenced by increased Syn expression levels. The present study demonstrated that CaSR inhibition contributed to the protective effect on CIH-induced neuronal injury. Moreover, CaSR was upregulated in CIH-induced apoptosis and synaptic plasticity deficit of hippocampus, accompanying the increasing p-ERK1/2 and PKC, while inhibition of CaSR downregulated p-ERK1/2 levels and PKC, so we suggested that phosphorylation of ERK1/2 and PKC may act as downstream products of CaSR and be involved in neuronal injury and synaptic plasticity deficit induced by CIH, which was consistent with our results in vitro.
To explore the mechanism of CaSR in CIH-induced cognitive impairment of the mice, we selected highly differentiated PC12 cell lines expressing nerve growth factor (NGF) receptors that can induce neurophenotypes. Studies of NGF-stimulated differentiation in the PC12 cells were conducted as in the previous reports [46]. PC12 line, a useful model system for the study of numerous problems in neurobiology and neurochemistry, is widely used for neuronal development. So in others and our researches, NGF-stimulated differentiation of PC12 cells was used to replace hippocampal primary cultures [27,36]. We successfully established the intermittent hypoxia PC12 model The ERK1/2 cascade, as the primary pathway of the MAPK family, was involved in cell proliferation and cell survival [47]. Previous research indicated that the phosphorylation of ERK1/2 plays an important role in cell apoptosis, but had a bidirectional regulation to apoptosis at different disease models [48,49]. Recently, some studies revealed a significantly increased p-ERK1/2 expression in the hippocampus of rats undergoing cerebral ischemia, as underlying apoptotic machinery of hippocampus injury [50]. In intermittent hypoxia condition, we further investigated the role of ERK1/2 in PC12 cells. We found that CIH elevated ERK1/2 phosphorylation levels in vitro and vivo experiments, while p-ERK1/2 inhibition via a selective MEK inhibitor attenuated cell injury, following decreased cleaved Caspase-3 level and increased cell viability. Therefore we suggested that ERK1/2 phosphorylation held a critical function on cell apoptosis following CIH. Of note, we found that the expression of CaSR showed no significant difference via treatment with ERK inhibitor following IH exposure, while PC12 cells, exposed to IH and treated with CaSR inhibitor, showed significantly decreased phosphorylation of ERK1/2. These results suggested that p-ERK1/2 played a key role in IH-induced apoptosis and maybe a downstream product of CaSR. ERK1/2 signaling pathways play a significant part in synaptic plasticity [51]. We also found that CIH induced synaptic plasticity deficit, manifested by a decrease in Syn, following upregulated p-ERK1/2, while p-ERK1/2 was downregulated and Syn expression was increased after ERK inhibitor intervention. The data revealed that p-ERK1/2 held on a critical function on synaptic plasticity deficit induced by CIH.
Various researches demonstrated PKC was an important regulator in the process of ERK1/2 phosphorylation, and ERK1/2 activation has been reported to be mediated by PKC in hippocampal neural stem cells [52]. Recently, several studies have shown that ERK phosphorylation regulates different disease processes under the activation of PKC [53]. We examined whether the CaSR/PKC/ERK1/2 pathway is involved in apoptosis and synaptic plasticity deficit of PC12 cells via the three corresponding inhibitors pretreated with PC12 cells after CIH. We further found p-ERK1/2 and PKC expression were downregulated with PKC inhibitor precondition, while the level of PKC was no change and p-ERK1/2 was downregulated treated with ERK inhibitor, along with CaSR as upstream, which regulated p-ERK1/2 and PKC expression. Therefore, our results demonstrated that p-ERK1/2 was regulated via PKC while both are regulated by CaSR in cell apoptosis and synaptic plasticity deficit following CIH, which was consistent with the previous researches [18]. Strictly speaking, we did not demonstrate whether there was a direct or indirect relationship among CaSR, PKC, and ERK1/2. In structure, Fig. 9 This diagram outlined the mechanism of CaSR to regulate the cognitive impairment caused by chronic intermittent hypoxia (CIH). Under intermittent hypoxic exposure, CaSR, activated by extracellular calcium, increased the concentration of intracellular calcium, and then activated PKC. The PKC activation mediated the phosphorylation of ERK1/2 (p-ERK1/2). On the one hand, upregulated p-ERK1/2 increased the cleaved Caspase-3 expression, promoting the apoptosis of neurons in hippocampus. On the other hand, pERK1/2 modulated the production of synaptic-associated proteins (GAP-43, and Syn) to impair the hippocampal synaptic plasticity. Both neuronal apoptosis and synaptic plasticity deficit in hippocampus resulted in CIHinduced cognitive dysfunction. By contrary, inhibition of CaSR could alleviate CIH-induced neuronal apoptosis and synaptic plasticity CaSR contains PKC consensus sequences and the phosphorylation site of PKC protein. Therefore, there can be a direct interaction between CaSR and PKC Co-immunoprecipitation studies which showed that ERK interacted indirectly with PKC via SGLT1 or Raf-1 [54,55]. No evidence of direct interaction between ERK1/2 and PKC has been reported so far. The results of this study are only applicable to intermittent hypoxia mouse models, and there is no clinical study on children with OSAHS. Although the current study focuses on molecular biology, it provides a theoretical basis for exploring the mechanism and developing new therapeutic drugs of cognitive dysfunction in children with OSAHS, which needs further research.

Conclusion
In summary, the present study demonstrated that CaSR hold a critical function on CIH-induced cognitive dysfunction. The activation of CaSR induced upregulated PKC and phosphorylation of ERK1/2 and finally resulted neuronal apoptosis and the synaptic plasticity deficit in hippocampus. In addition, we speculated that the increased p-ERK1/2 and PKC expression may be the underlying downstream of CaSR in CIH-induced cognitive dysfunction of mice (Fig. 9). Our study provides an alternative signaling pathway in terms of OSAHS-induced cognitive impairments and may contribute to a theoretical basis for the treatment of OSHAS.