Calcium-Sensing Receptor Mediates β-Amyloid-Induced Synaptic Formation Impairment and Cognitive Deficits via Regulation of Cytosolic Phospholipase A2/Prostaglandin E2 Metabolic Pathway

Calcium-sensing receptor (CaSR) is a G protein-coupled receptor (GPCRs). Soluble β-amyloid peptide (Aβ) is one of the orthosteric modulators of CaSR, while, the role and underlying mechanism of CaSR in cognitive decline in Alzheimer’s disease (AD) is unclear. In this study, molecular technology such as live-cell imaging combined with behavioral tests were used to explore the role and the underlying mechanism of CaSR in the cognitive deficits in AD mice. The expression levels of CaSR were increased both in AD mice and Aβ1–42 (β-amyloid protein)-treated primary cultured neurons. Pharmacological inhibition of CaSR ameliorated recognitive and spatial memory deficits of Aβ1–42-oligomer-treated mice in a dose-dependent manner. Pharmacological inhibition of CaSR or down-regulation of the expression of CaSR by CaSR-shRNA-lentivirus prevented the impairment of filopodia, and the synapse induced by oligomeric Aβ1–42. The contents of cytosolic phospholipase A2 (cPLA2) and prostaglandin E2 (PGE2) in hippocampal neurons and tissue were increased after treatment with Aβ1–42 oligomers. Inhibition or down-regulation of CaSR mediates Aβ-induced synapse formation and cognitive deficits partially, through the activation of the cPLA2/PGE2 pathway. This study provides novel insights on CaSR, which is a promising therapeutic target for AD.


INTRODUCTION
Alzheimer's disease (AD) is one of the most common neurodegenerative diseases with a slow and gradual deterioration of the memory that affects language, personality, and cognitive control (Wyss-Coray and Rogers, 2012;Mhatre et al., 2014;Dos Santos Picanco et al., 2018). AD is a severe threat for human health, along with an aging population, however, the pathogenesis of it is still unclear. Mounting evidence shows that the accumulation of β-amyloid peptide (Aβ) is an important contributing factor in the pathology of AD (Baek et al., 2017;Choi et al., 2017). Aβ has different aggregated forms, including monomers, oligomers, protofibrils, and mature fibrils (Ahmed et al., 2010). It is well demonstrated that soluble oligomers of Aβ is the pertinent toxic form of Aβ (Wang H. C. et al., 2016;Wang T. et al., 2016). Both others, and our previous studies have shown that soluble Aβ oligomers could decrease the number of dendritic spines and inhibit the long-term potentiation (LTP), leading to the decline of cognitive function in AD mice (Price et al., 2014;Jiang et al., 2016). Disrupting synapse function plays an important role in the memory deficits of AD, which stands out in early AD pathological changes (Price et al., 2014;Teich et al., 2015;Wang X. et al., 2018). Therefore, it is a promising strategy to consider amelioration of synaptic impairment induced by Aβ oligomers as the target of prevention or treatment of AD.
Aβ is also one of the allosteric agonists of calcium-sensing receptor (CaSR). Aβ has been shown to interact directly with CaSR via a proximity ligation assay (Diez-Fraile et al., 2013;Leach et al., 2015). CaSR, a member of the G protein-coupled receptor (GPCRs) C family, is a seven-transmembrane protein (Brauner-Osborne et al., 2007;Conigrave and Ward, 2013;Summers, 2016). CaSR has been detected in the hippocampus which is an important brain structure that is essential for spatial memory, language learning, and episodic memory (Ferry et al., 2000). CaSR is mainly localized in nerve endings in the neurons and is involved in regulating brain excitability (Chen et al., 2010). CaSR activation depends on the persistent interaction with its agonists. At physiological conditions, CaSR is partially activated (Ruat and Traiffort, 2013;Díaz-Soto et al., 2016). In cultured cells, Aβ-activated CaSR could cause excessive release of Aβ (Armato et al., 2013;Dal Prà et al., 2014). The expression of CaSR was significantly increased in AD transgenic mice (Leach et al., 2015;. The role of CaSR in AD is unclear and the cellular mechanisms have not been well characterized. As a promising therapeutic target, we therefore evaluated the role of CaSR in cognitive deficits in the mouse model of AD and its underlying cellular mechanisms. The effects of CaSR on oligomeric Aβ-induced synaptic injury are unknown. In the current study, we have also evaluated the role and the underlying mechanisms of CaSR in Aβ-mediated synaptic impairment.

Aβ 1-42 Oligomers Preparation
Preparation of soluble Aβ 1-42 oligomers was done according to the protocol previously described (Jiang et al., 2016;Ding et al., 2019). One milligram of Aβ 1-42 (Bachem, Cat# H-1368.1000) powder was dissolved in 400 µl ice-cold 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP; Aladdin, Cat# K1625063), and incubated at room temperature for 20 min. Hundred microliter of this complete solution was diluted into 900 µl of deionized water to a final concentration of 0.25 g/l. After centrifugation at 14,000 g for 15 min, the supernatant was collected and the HFIP was completely evaporated. Then the collected supernatant was kept stirring for 48 h at room temperature. A 50 µM Aβ 1-42 solution was obtained and stored at 4 • C. The preparation is the combination of low molecular weight forms of soluble Aβ (Chunhui et al., 2018).

Animals
ICR mice (RRID:IMSR_CRL:22) or B6C3-Tg (APPswe/PSEN1dE9) mice were used in our experiments. Breeding pairs of APPswe/PSEN1dE9 transgenic mice were originally purchased from Jackson Laboratories, USA. A breeding colony of APPswe/PSEN1dE9 mice was established at the Medical School of Ningbo University. All experimental animals were housed in a temperature and humidity-controlled animal facility (22 ± 3 • C, 60% ± 5%) with a 12 h light and dark cycle and free access to standard chow and water. Experiments were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23, revised 1996) and approved by the Institutional Animal Care and Use Committee of the Ningbo University. The approval number for the animal experiments is SYXK (ZHE) 2013-0191. Genotypes of APPswe/PSEN1dE9 mice were analyzed as follows: DNA was isolated from the tail tip of each mouse and PCR was performed using the following primer pairs: APP, forward primer 5 -GACTGCCACTCGACCAGGTTCTG-3 , reverse primer 5 -CTTGTAAGTTGGATTCTCATATCCG-3 ; PS1, forward primer 5 -GTGGATAACCCCTCCCCCAGCCTAGACC-3 , reverse primer 5 -AATAGAGAACGGCAGGAGCA-3 . APPswe/PSEN1dE9 transgenic mice and wild type mice were identified by agarose gel electrophoresis. In each identification experiment, both positive and negative controls were designed.

Animal Surgery
Two-month-old healthy male ICR mice (25-30 g) were pseudorandomly assigned to the experimental groups using a random number generated from Excel, and all mice were marked by staining in different parts of the back. A pre-test open field experiment was first performed on all mice to determine the locomotor activity before the formal experiments and to exclude those with an obvious movement disorder. No obvious movement disorder was found among the mice. Thus, no mice were excluded. ICR mice were anesthetized by intraperitoneal administration of sodium pentobarbital (50 mg/kg) before they were placed in a stereotaxic apparatus (Stoelting, Wood Dale, IL, USA). Cannulas (RWD Life Science, Shenzhen, China) were surgically implanted into bilateral hippocampal regions of the mice using the following coordinates: AP −1.7 mm from Bregma; ML ± 1.0 mm from the midline; DV −1.5 mm from pia mater. After 7 days of post-operative recovery, the minipump needle tips were inserted into the ventricle through cannulas to inject pharmaceuticals. Experimental mice were given three consecutive infusions of Aβ 1-42 (4 µmol/kg) and/or NPS 2143 (Sigma, Cat# SML0360, 0.08 or 0.16 µmol/kg), while the control mice received saline injections instead of the drugs.

Behavior Tests
The novel object recognition (NOR) task, consisting of a familiarization phase and a test phase, was carried out in an open-field arena (60 × 60 × 15 cm) on the 11th to 12th day after the first injection. On the first day, they were familiarized with two identical objects for 5 min. On the second day, one of the objects was replaced by a novel one with a different shape and color, and the mice were allowed to explore the arena for 5 min. To ensure the absence of olfactory cues, the open-field arena and the objects were cleaned thoroughly. Exploration was defined as sniffing or touching the objects. The distance between the nose and object was no more than 2 cm. If the mice traveled around or sat on the objects, this was not defined as object recognition. The discrimination index, the ratio of the amount of time spent exploring any one of the two objects (training session) or the novel object (retention session), over the total time spent exploring both objects, was used to measure the cognitive function of animals.
A Morris water maze (MWM) was performed as described (Jiang et al., 2016). Briefly, the equipment included a pool with a diameter of 110 cm that was filled with opaque water at approximately 22 ± 1 • C. Spatial memory was assessed by recording the latency time for the animal to escape from the water onto an escape platform during the place navigation phase. At the learning phase, mice were given 90 s to find the hidden platform that was 1 cm below the water surface. The place navigation test of the MWM, which consisted of four trials (interval 20-30 min) each day, took place during the 14th day to the 17th day and the latency time was recorded. On the 18th day, the platform was removed from the maze. A probe trial was conducted to measure the trajectories and entries of mice to the target quadrant with a video tracking system (Ethovision XT). The assessor was blinded to the experimental conditions for analysis of the behavioral tests.

Lentivirus Construction
The Lentivirus was purchased from Shanghai Genechem Company, Limited (China). The company used the following procedures: The coding sequence of the CaSR was from GenBank: NM_013803.3. For small hairpin RNA (shRNA) against mouse CaSR, vectors were constructed from the original plasmid GV lentiviral vector, and the GV118 serotype was selected; the reaction element sequence of which is U6-MCS-Ubi-EGFP. The target gene was inserted into the MCS element by HpaI and XhoI, two restriction endonucleases. The optimal target sequence (TCTTCATCAAGTTCCGAAA) was selected for small hairpin RNA (shRNA) against mouse CaSR, and a scrambled shRNA (TTCTCCGAACGTGTCACGT) served as a control. For viral packaging, the respective recombinant plasmids were cotransfected into 293T cells (ATCC, RRID:CVCL_0063). The GV stocks were titered by quantitative PCR, stored at −80 • C of a titer of 10 9 particles/ml and shipped with 20 µl in every tube. We then used a total of 5 * 10 5 TU/ml of the virus to transfect the hippocampal neurons. All procedures were performed under a biosafety cabinet in a biosafety level 2 facility.

Confocal Imaging and Analysis
At DIV 7 or DIV 14 of culture, living neurons were captured by a Fluoview-1000 confocal microscope. After drug treatments, the neurons were maintained in a recording chamber with extracellular solution (148.00 mM NaCl, 3.00 mM KCl, 3.00 mM CaCl 2 , 10.00 mM HEPES, and 8.00 mM glucose, pH 7.3) at room temperature. Digital images of GFP were collected on a Fluoview-1000 confocal microscope (Olympus) using a 60× oil objective lens without optical zoom at an excitation wavelength of 488 nm. They were analyzed using Fluoview-1000 software. All lengths of the secondary dendritic branches were measured by tracing their extension, and the filopodia and spines were counted. For all analyses, images were analyzed blind to treatments and data were collected from at least three independent experiments.

Cytosolic Phospholipase A2 (cPLA2) and Prostaglandin E2 (PGE2) Assay
Cytosolic phospholipase A2 (cPLA2) or prostaglandin E2 (PGE2) levels in cultured hippocampal neurons and tissues were assayed using a commercial mouse cPLA2 (Qiaodu-Bio, Cat# CK-E92479M) or PGE2 ELISA Kit (Qiaodu-Bio, Cat# CK-E90213M), respectively. The hippocampal neurons and tissues were homogenized in ice-cold 70 µl RIPA (Solarbio, Cat# R0010) buffer. The samples were centrifuged at 13,000 g (4 • C for 10 min). Ten micro-liter of the supernatant from the hippocampal homogenate was used to assay cPLA2 or PGE2 level, according to the manufacturer's protocols. The content of cPLA2 was based on measures of absorbance at 450 nm/well in a 96 well plate reader (Thermo Scientific, USA).

Statistical Analyses
Statistical analyses were performed with GraphPad PRISM software (GraphPad Software Inc., La Jolla, CA, USA, RRID:SCR_00298). Data were presented as mean ± SEM. Two-group comparisons were analyzed by a two-tailed student's t-test. Multiple group data were analyzed using a one-way analysis of variance (ANOVA) followed by a Tukey post hoc test, with the exception of the data of the place navigation test of the MWM tests, which were analyzed by two-way repeated-measures ANOVA with Tukey post hoc comparisons. P < 0.05 was considered statistically significant. A test for outliers was not performed on the data.

The Expression Levels of CaSR Are Increased in AD Mice and Aβ 1-42 -Treated Primary Hippocampal Neurons
The expression levels of CaSR in APPswe/PSEN1dE9 transgenic mice and wild type mice were detected (Figures 1A,B). The expression levels of CaSR were significantly increased in the 9month-old AD mice (P < 0.01, Figure 1G). The expression levels of CaSR in the control and Aβ 1-42 -treated primary hippocampal neurons were also detected (Figures 1C-F). Both the mRNA and protein levels were increased by Aβ 1-42 treatment (P < 0.05, Figure 1H; P < 0.01, Figure 1I). The results demonstrated that Aβ 1-42 treatment upregulates the expression level of CaSR.

Pharmacological Inhibition of CaSR Prevents Dendritic Filopodium Loss Caused by Oligomeric Aβ 1-42 in Hippocampal Neurons
To find out whether CaSR mediates Aβ 1-42 -induced early synapse formation impairment, we measured the density of dendritic filopodium of hippocampal neurons at DIV 7 treated with Aβ 1-42 oligomers, in the presence or the absence of CaSR antagonist NPS 2143 the density of dendrite filopodium of the Aβ 1-42 -treated group was significantly reduced compared to the Frontiers in Aging Neuroscience | www.frontiersin.org control group (P < 0.01, Figure 2). Treatment with NPS 2143 (0.1 µM) significantly ameliorated the reduction of filopodium density induced by Aβ 1-42 oligomers (P < 0.01), while, NPS 2143 treatment alone did not alter the density of dendrite filopodium (P > 0.05, Figure 2).

NPS 2143 Prevents Spine Loss and Synaptic Impairment Caused by Oligomeric Aβ 1-42 in Hippocampal Neurons
To further analyze the role of CaSR in synapse formation, the dendritic spine densities at DIV 14 were quantified in the control group, oligomeric Aβ 1-42 -treated group, oligomeric Aβ 1-42 + NPS 2143 group and the NPS 2143 alone group. The oligomeric Aβ 1-42 -treated group prominently decreased the spine density (P < 0.01, Figure 3). Treatment with NPS 2143 (0.1 µM) significantly prevented the decreased spine density induced by Aβ 1-42 oligomers (P < 0.01, Figure 3), while, NPS 2143 treatment alone did not alter the density of the spine (P > 0.05, Figure 3). These results suggest that CaSR is involved in dendritic spine loss caused by Aβ 1-42 oligomers in hippocampal neurons.

Down-Regulation of CaSR Expression Prevents Synaptic Impairment Induced by Oligomeric Aβ 1-42 in Hippocampal Neurons
To further verify that CaSR mediates the synaptic impairment induced by Aβ 1-42 , the synaptic densities of neurons treated by the CaSR-shRNA-lentivirus or NC shRNA were analyzed by immunocytochemistry (Figures 5A-P). The expression level of CaSR was down regulated by the CaSR-shRNAlentivirus (P < 0.01, Figure 5Q). Compared with the control group, the numbers of synaptotagmine-1-positive puncta and PSD 95-positive puncta were significantly decreased (P < 0.01, Figures 5R,S) in the oligomeric Aβ 1-42 -treated cells.
Down-regulation of the CaSR expression significantly prevented the decreased puncta numbers of synaptotagmine-1 and PSD 95 (P < 0.01, Figures 5R,S). The synapse number was also measured, and down-regulation of CaSR significantly protected hippocampal neurons from synapse loss induced by oligomeric Aβ 1-42 (P < 0.01, Figure 5T).

NPS 2143 Prevents the Decreased Expression Levels of Synaptotagmine-1 and PSD 95 in AD Model Mice
Collective evidence shows that oligomeric Aβ 1-42 is regarded as the pertinent toxic form of Aβ (Baek et al., 2017;Choi et al., 2017). In order to study the single factor of Aβ and the underlying mechanism of CaSR in Aβ-mediated synaptic and cognitive impairment, the AD mouse model, made by microinjection with Aβ 1-42 oligomers, were used in the rest of our study.

Pharmacological Inhibition of CaSR Prevents Cognitive Deficits of Aβ 1-42 Oligomer-Treated Mice
Normal synapse formation is considered to be the structure basis of cognitive function. To find out whether the impairment of synapse formation mediated by CaSR was also involved in the cognitive deficits of the AD mouse model, NOR tests and MWM tests were used to estimate the role of CaSR in soluble Aβ 1-42 oligomer induced recognitive and spatial memory deficits (Figure 7), respectively.
The discrimination indexes were used to evaluate the recognitive memory of animals in NOR tests. There was no significant difference in discrimination indexes among these groups in the training session (P > 0.05, Figure 7B). In the retention session, discrimination indexes were decreased in the mice injected with soluble Aβ 1-42 oligomers (P < 0.05, Figure 7C). Inhibition of CaSR with NPS 2143 (0.16 µmol/kg) had no effect on discrimination indexes, but it significantly attenuated the Aβ 1-42 oligomer-induced reduction of discrimination indexes (P < 0.05, Figure 7C). The NPS 2143 (0.08 µmol/kg) treatment had no effect on Aβ 1-42 -induced change in the discrimination indexes (P > 0.05, Figure 7C). These results indicated that NPS 2143 rescues Aβ 1-42 -induced recognitive deficits in a dose dependent manner.
To further investigate whether CaSR is involved in Aβ 1-42 oligomer-mediated spatial memory impairment, we examined memory performance with MWM tests in mice treated with Aβ 1-42 oligomers in presence or absence of CaSR antagonist NPS 2143. Two-way ANOVA for repeated-measures revealed significant changes in drug effects (P < 0.01, Figure 7D) and time effects (P < 0.01, Figure 7D), but no interaction was found (P > 0.05, Figure 7D). The escape latency of Aβ 1-42 -treated mice on day 3 and 4 was significantly longer compared with the control mice (P < 0.05 and P < 0.01 for day 3 and 4 respectively, Figure 7D). The escape latency of NPS 2143-treated mice (0.08 or 0.16 µmol/kg) was stable compared to that of the control mice. The NPS 2143 (0.08 µmol/kg) did not prevent the increase escape latency of Aβ 1-42 -treated mice, however, the NPS 2143 (0.16 µmol/kg) treatment prevented the prolongation of latency induced by Aβ 1-42 on day 3 and 4 (P < 0.05, Figure 7D). In the probe test, after the hidden platform was removed from the target quadrant, the Aβ 1-42 -treated mice spent a shorter time in the target quadrant compared with the control mice (P < 0.01, Figure 7E). A 0.16 µmol/kg NPS 2143 treatment reversed the decrease of time in the target quadrant of the Aβ 1-42 -treated mice (P < 0.01, Figure 7E). However, treatment of NPS 2143 alone, at the dose of 0.08 or 0.16 µmol/kg, had no effect on the time spent in the target quadrant ( Figure 7E). Moreover, we measured the numbers of times the mice swam cross the place where the original platform was. The numbers of the target platform crosses were decreased in the Aβ 1-42 -treated mice (P < 0.01, Figure 7F). Injection of NPS 2143 (0.16 µmol/kg) reversed the decreased crosses over the target platform of Aβ 1-42 -treated mice (P < 0.01, Figure 7F). However, treatment of NPS 2143 alone had no effect on the numbers of the target platform crosses (Figure 7F). We did not observe a significant difference in velocity among the six groups in the MWM tests (Figure 7G), indicating that the differences in latency, time and number of crosses over the target platform among the groups were not caused by the differences in velocity. Altogether, these data suggested that pharmacological inhibition of CaSR prevents mice from Aβ 1-42 oligomer-induced spatial learning and memory deficits in MWM tests.

CaSR Participates in Aβ 1-42 -Induced Increase in the Levels of cPLA2 and PGE2
To explore the downstream pathways mediated by CaSR and soluble Aβ 1-42 oligomers, the contents of cPLA2 and PGE2 were measured in hippocampal neurons. Aβ 1-42 treatment increased the content of cPLA2 (P < 0.01, Figure 8A). NPS 2143 (0.1 µM) treatment significantly prevented the increased level of cPLA2 induced by Aβ 1-42 oligomers (P < 0.01), while the level of cPLA2 remained stable under NPS 2143 (0.1 µM) alone treated neurons, compared with that of the control neurons. The expression level of CaSR was reduced by the CaSR-shRNA-lentivirus, and knocking down CaSR also prevented the increased cPLA2 content induced by Aβ 1-42 oligomers (P < 0.01, Figure 8C). Aβ 1-42 treatment also increased the content of PGE2 (P < 0.01, Figure 8B). The NPS 2143 (0.1 µM) treatment significantly prevented Aβ 1-42 oligomer-induced increase of the level of PGE2 (P < 0.01), while, the level of PGE2 remained stable under NPS 2143 (0.1 µM) alone treated neurons, compared with that of the control neurons (P > 0.05 Figure 8B). Down-regulation the expression level of CaSR also prevented the increased PGE2 content induced by Aβ 1-42 oligomers (P < 0.01, Figure 8D).

DISCUSSION
To the best of our knowledge, our results, for the first, time demonstrate that the expression level of CaSR is increased by Aβ 1-42 . The increase expression or activity of CaSR mediates the AD-like phenotypes/pathology induced by Aβ 1-42 oligomers partially through activation of the cPLA2/PGE2 pathway. CaSR is located in nerve terminals which are related to synaptic plasticity and neuronal transmission (Bandyopadhyay et al., 2010). In this study, we provided novel insights to the possible role and underlying mechanisms of CaSR in AD.
It is well accepted that the hippocampus is a region of the adult brain where neurogenesis occurs. Perturbations in synapse formation by forms of oligomeric Aβ are tightly correlated with memory deficits in AD (Bandyopadhyay et al., 2010;Marchetti and Marie, 2011;Ardiles et al., 2012;Ma and Klann, 2012;Sanchez et al., 2012;Xu et al., 2014). Growing evidence indicates that Aβ-induced synaptic loss in the hippocampus occurs at the early stage of AD (Teich et al., 2015;Wang X. et al., 2018). We demonstrated that exposure of Aβ 1-42 oligomers FIGURE 7 | Protective function of inhibiting CaSR on oligomeric Aβ 1-42 -induced recognitive and spatial learning deficits. (A) Experimental schedule of behavioral tests. Cannulas were implanted into bilateral hippocampal regions. Vehicle, Aβ 1-42 (4 µmol/kg) and/or CaSR inhibitor NPS 2143 (0.08 or 0.16 µmol/kg body weight per day) were microinjected for 3 days from day 7 to 9. From day 11 to day 12, the novel object recognition (NOR) tests were performed. Morris water maze (MWM) tests were conducted from day 14 to 18 day. Place navigation tests (PNT) of the MWM were conducted four times a day for four consecutive days, followed by a probe trial test (PT) 24 h after the last PNT. The mice were sacrificed after behavioral experiments. (B) Discrimination indexes displayed no significant difference among these groups in the training session. (C) Quantitative comparison of the discrimination indexes in the retention session, n = 8 animals for each group. (D) The escape latency of the control, the oligomeric Aβ 1-42 -treated group (4 µmol/kg), and/or the NPS 2143-treated group (0.08 or 0.16 µmol/kg) during the four training days. (E) Quantitative comparison of the time in the target quadrant of the six groups. (F) Quantitative comparison of the number of target platform crosses of the six groups. (G) Swimming speed of the six groups did not show significant different. *P < 0.05 vs. control group, **P < 0.01 vs. control group, # P < 0.05 vs. Aβ 1-42 oligomer-treated group, ## P < 0.01 vs. Aβ 1-42 oligomer-treated group. Values represent mean ± SEM, n = 8 animals for each group.
potently decreased filopodium density in hippocampal neurons, while CaSR inhibitor NPS 2143 significantly prevented Aβ 1-42 oligomer-induced filopodium loss. These results suggest that CaSR is involved in the negative role of Aβ 1-42 during initial synapse formation. Dendritic spines, small membranous protrusions from neuronal dendrites, are developed from filopodia. Consistent with the role of CaSR in the Aβ 1-42induced decrease in filopodium density, CaSR was also involved in the impairment of spine and synapse formation mediated by Aβ 1-42 oligomers. Both pharmacological inhibition of CaSR  and knockdown of CaSR prevented Aβ 1-42 -induced synapse developmental deficits. Moreover, we found that CaSR also mediated Aβ 1-42 -induced cognitive deficits. Using behavioral tasks, including MWM and NOR tests, we also showed that Aβ 1-42 oligomers cause recognitive and spatial memory impairment. Inhibition of CaSR with NPS 2143 prevented Aβ 1-42 -induced cognitive deficits in a dose-depended manner. Thus, CaSR mediates AD-like synaptic and cognitive impairment induced by Aβ 1-42 oligomers.
cPLA2/PGE2 signaling pathways might be involved in the effects of CaSR on Aβ-induced cognitive impairment. PGE2, a lipid molecule derived from arachidonic acid (AA; Brummett et al., 2014). cPLA2 affects the synthesis of PGE2 by promoting the release of AA (Bate and Williams, 2015a). It has been found that the activation of cPLA2/PGE2 is associated with multiple signaling pathways such as neuronal excitation, synaptic secretion, lipid metabolism, and neuroinflammation (Murakami and Kudo, 2004;Igarashi et al., 2011;Sun et al., 2014). Activation of cPLA2 enzymes plays an important role in age-associated neuronal and memory impairment (Hermann et al., 2014). Neurons isolated from mice deficient in cPLA2 −/− showed resistance to the toxic effects of Aβ (Desbàne et al., 2012). PGE2 also regulates synaptic function and plasticity (Koch et al., 2010). Addition of PGE2 reduced the content of synaptic proteins in cortical neurons, and impaired hippocampal presynaptic long-term plasticity in a mouse model of AD (Bate et al., 2010;Maingret et al., 2017). Our results showed that the contents of cPLA2 and PGE2 in hippocampal neurons and hippocampal encephalic region were significantly increased due to oligomeric Aβ 1-42 treatment. These results are consistent with previous studies showing that Aβ oligomers could activate cPLA2 and increase the level of PGE2, resulting in a reduction of synaptic markers and a decline in cognitive function (Desbàne et al., 2012;Bate and Williams, 2015b). Both pharmacological inhibition of CaSR and down-regulation of the expression of CaSR prevented Aβ-induced increase of cPLA2/PGE2 and synaptic damage. Thus, CaSR might mediate Aβ-induced synaptic and cognitive damage through increasing cPLA2 and PGE2 contents.
Besides activation of the cPLA2/PGE2 pathway, CaSR has also been reported to be involved in the Aβ-induced increase of Aβ and phospho-tau (Dal Prà et al., 2014;Chiarini et al., 2017); this might also contribute to the decrease in cognitive decline in AD mice. At physiological conditions, CaSR was only partially activated (Ruat and Traiffort, 2013;Díaz-Soto et al., 2016). Increased Aβ made CaSR full or over activated, which further induced the early impairment of synapse formation and cognitive function. Together with our observations, the current data showed that CaSR is an important factor mediating the progress of AD.
In conclusion, we demonstrated that CaSR is involved in oligomeric Aβ 1-42 -induced cognitive dysfunction as well as synapse formation and developmental impairment in the pathogenesis of AD, in addition, our results indicated that cPLA2 and PGE2 are downstream targets of CaSR, which mediate cognitive decline in AD. Thus, we provided support for the efficacy of specific antagonists of CaSR in the treatment of AD.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The animal study was reviewed and approved by Institutional Animal Care and Use Committee of the Medical School of Ningbo University [permission: SYXK(ZHE)2013-0191].

AUTHOR CONTRIBUTIONS
SX and XB were responsible for the design of the study. CF was in charge of molecular and cellular experiments. YL, YD, and YC were mainly involved in animal experiments and relative analysis. QW and JW provided valuable advice for the research. WC and LS provided language modification and data analysis. All authors contributed to the article and approved the submitted version.