Molecular Mechanism of the ATF6α/S1P/S2P Signaling Pathway in Hippocampal Neuronal Apoptosis in SPS Rats

Apoptosis of hippocampal neurons is one of the mechanisms of hippocampal atrophy in posttraumatic stress disorder (PTSD), and it is also an important cause of memory impairment in PTSD patients. Endoplasmic reticulum stress (ERS) mediated by activated transcription factor 6α (ATF6α)/site 1 protease (S1P)/S2P is involved in cell apoptosis, but it is not clear whether it is involved in hippocampal neuron apoptosis caused by PTSD. A PTSD rat model was constructed by the single prolonged stress (SPS) method. The study was divided into three parts. Experiment 1 included the control group, SPS 1 d group, SPS 7 d group, and SPS 14 d group. Experiment 2 included the control group, SPS 7 d group, SPS 7 d + AEBSF group, and control + AEBSF group. (4-(2-Aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF) is an ATF6α pathway inhibitor). Experiment 3 included the control group, SPS 4 d group, SPS 4 d + AEBSF group, and control + AEBSF group. The protein and mRNA expression levels of ATF6α, glucose-regulated protein (GRP78), S1P, S2P, C/EBP homologous protein (CHOP), and caspase-12 in the hippocampus of PTSD rats were detected by immunohistochemistry, Western blotting and qRT-PCR. Apoptosis of hippocampal neurons was detected by TUNEL staining. In experiment 1, the protein and mRNA expression of ATF6α and GRP78 increased gradually in the SPS 1 d group and the SPS 7 d group but decreased in the SPS 14 d group (P < 0.01). In experiment 2, compared with that in the control group, the protein and mRNA expression of ATF6α, GRP78, S1P, S2P, CHOP, and caspase-12 and the apoptosis rate were significantly increased in the SPS 7 d group (P < 0.01). However, the protein and mRNA expression of ATF6α, GRP78, S1P, S2P, CHOP, and caspase-12 and the apoptosis rate were significantly decreased after AEBSF pretreatment (P < 0.01). In experiment 3, compared with that in the control group, the protein and mRNA expression of ATF6α, GRP78, S1P, S2P, CHOP, and caspase-12 and the apoptosis rate were increased in the SPS 14 d group (P < 0.05). However, the protein and mRNA expression of ATF6α, GRP78, S1P, S2P, CHOP, and caspase-12 and the apoptosis rate were decreased after AEBSF pretreatment (P < 0.05). SPS induced apoptosis of hippocampal neurons by activating ERS mediated by ATF6α, suggesting that ERS-induced apoptosis is involved in the occurrence of PTSD.


Introduction
Posttraumatic stress disorder (PTSD) occurs in individuals who have experienced a sudden major disaster (such as an earthquake, tsunami, flood, avalanche, or war) or unusual threats (such as violence). The main manifestations of PTSD are (1) repeated intrusive memories of the traumatic experience and memory abnormalities; (2) persistent avoidance and emotional dysregulation; and (3) increased alertness. Recently, with the increasing number of sudden disasters at home and abroad, the incidence rate of PTSD has been increasing year by year. It has become a serious psychological obstacle and is the immediate focus of public attention after disasters (Qi et al. 2016;Sin et al. 2017). MRI examination of PTSD patients showed hippocampal shrinkage with abnormal memory function (Apfel et al. 2011;Hines et al. 2014). It has been found that apoptosis of hippocampal neurons may be one of the pathological mechanisms related to memory impairment in PTSD (Fan et al. 2011). However, the exact regulation mechanism needs to be further studied.
Excessive endoplasmic reticulum stress (ERS) can initiate the apoptotic pathway (Hetz and Saxena 2017;Lee et al. 2018;Sisinni et al. 2019). PTSD usually occurs after an individual experiences one of the various disasters mentioned above. These disaster stimuli are all supernormal stimuli that may cause ERS and initiate the unfolded protein response (UPR), positively regulate protein folding, promote unfolded protein degradation and restore the stability of the endoplasmic reticulum network (Wan and Jiang 2016). When excessive unfolded proteins are deposited, they exceed the processing capacity of the UPR, resulting in excessive ERS, and cell apoptosis will be initiated by the endoplasmic reticulum apoptosis pathway (Sharma et al. 2019).
The molecular weight of ATF6α, which belongs to a class of transcription factors containing a transmembrane domain, is 90 kDa (p90ATF6α). Under normal conditions, ATF6α mainly exists in the form of a zymogen (p90ATF6α) and is tightly bound to glucose-regulated protein (GRP78). When ERS occurs, ATF6α is separated from GRP78, released from the endoplasmic reticulum, and hydrolyzed to produce a fragment of approximately 50 kDa, namely, the P50 ATF6α fragment. The p50ATF6α fragment can be further translocated into the nucleus and affect the expression of ERS-related genes, such as C/ EBP-homologous protein (CHOP) (Stauffer et al. 2020). However, when endoplasmic reticulum homeostasis cannot be restored, the apoptosis process is initiated and executed by caspase 12, which resides in the endoplasmic reticulum and is the key signal by which ERS induces apoptosis (Fan et al. 2020). During ERS, the activation of ATF6α can specifically upregulate the expression of caspase 12, which leads to cell apoptosis (Martelli et al. 2020). Therefore, this study aimed to establish a PTSD rat model through single prolonged stress (SPS) to investigate the effect of the ATF6/S1P/S2P signaling pathway on the apoptosis of hippocampal neurons and to lay a theoretical foundation for further study of the PTSD mechanism.

Experimental Animals
Three hundred healthy male Wistar rats (provided by China Medical University) weighing 150-180 g and aged 6-7 weeks were given standard food and water and were housed at 5 per cage with an indoor temperature of 22 ± 2 °C and a humidity of 55 ± 5%. The experimental study was carried out after 1 week of adaptive feeding. The bedding was changed every 2 days, and the rats were placed under a 12-h light/dark cycle (lighting time 8:00-20:00).

PTSD Model Preparation and Grouping
After the rats were adapted and processed in the laboratory, the study comprised three experiments. Experiment 1 included the control group, SPS 1 d group, SPS 7 d group, and SPS 14 d group, with 25 rats in each group. Experiment 2 included the control group, SPS 7 d group, SPS 7 d + AEBSF group, and control + AEBSF group, with 25 animals in each group. (4-(2-Aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) is an ATF6 pathway inhibitor). Experiment 3 included the control group, SPS 14 d group, SPS 14 d + AEBSF group, and control + AEBSF group, with 25 animals in each group. The specific steps of PTSD modeling were as follows: first, the rat was restrained for 2 h, then it was immediately forced to swim for 20 min, and then it was allowed to recover for 15 min. After that, diethyl ether was used to anesthetize the rat and induce consciousness. The rats were fed freely and had a regular diet. The administration method of AEBSF was stereotactic brain injection: after the rat was anesthetized and fixed on the stereotaxic instrument, the hippocampus was located on both sides, the cranium was drilled, and microinjection was performed at a speed of 0.2 μl/min and a dose of 10 μl (300 μM). The needle was left in place for 10 min after the injection; the skin was subsequently sutured, and the rat was kept warm until it recovered from anesthesia.

Behavioral Tests
To evaluate the success of PTSD modeling in the rats, Morris water maze experiments and open-field experiments (Beijing Dongle Technology Co., Ltd, China) were carried out. The rat behavior tests were performed 7 days after SPS. The general state of each group of rats (including diet and mental state) was carefully observed. The room for behavioral testing was illuminated by indirect white light. The rats were tested at 9:00-11:00 in the morning every day. The experimenter was at least 1 m away from the equipment during the test. The experimental data from each group were recorded, and then the rats were returned to their cages. Additionally, 75% alcohol was used to clean the equipment to avoid residual stimuli (such as feces, urine, and odor) from affecting the subsequent experimental rats.

Water Maze Test
To test the learning and spatial memory abilities of the rats, a Morris water maze experiment was performed on the rats 7 days after SPS. There were 10 rats in the control group and 10 rats in the SPS 7 d group. The water maze consisted of a circular pool, a video analysis system, and a computer. The diameter of the ferrous metal circular pool was approximately 150 cm. Tap water was injected into the pool, and the water temperature was controlled at 20-22 °C. The experimental process was divided into two parts, namely, the positioning and navigation experiment and the space exploration experiment. (1) Positioning and navigation experiment: Each rat was placed in the water in the four quadrants of the pool facing the pool wall. After the rat climbed onto the platform and remained on it for 2 s, the timer was stopped, and the time was recorded as the escape latency period. The video tracking system recorded the swimming path of the rat. After the rats boarded the platform, they were kept on the platform for 15 s, and then the rats were wiped dry and placed in a warm environment. On days 1 to 3, the platform was placed approximately 1 cm above the liquid surface; on days 4 to 5, the platform was placed approximately 1 cm below the liquid surface.
(2) Space exploration experiment: After 24 h, the platform was removed, and the space exploration experiment was carried out. Each mouse was placed in the water from the quadrant opposite to the quadrant where the original platform was located, and the swimming speed of the mouse within 60 s and the time spent in the original platform area were recorded.

Open-Field Experiment
To evaluate the autonomous exploration behavior of the rats and evaluate their anxiety and fear behaviors in unfamiliar environments, an open-field experiment was performed. There were 10 rats in the control group and 10 rats in the SPS 7 d group. The rat open-field reaction box was 30-40 cm high, and the bottom side was 100 cm long. The bottom was divided into 25 small squares of 4 cm × 4 cm. The inner wall was painted black. A digital video camera was installed 2 m above the box. The field of view included the entire open field. Before the experiment, the rats were placed in a small box for 20 min. In a quiet environment, the rats were placed in the middle of the open-field box. The behavior of the rats, the number of times the rats entered the central area, and the percentage of movement distance in the central area over 5 min were recorded.

Collection of Rat Tissue
Five rats in each group were anesthetized with 2% pentobarbital sodium (Shanghai Sixin Biotechnology Co., Ltd., China) and perfused with 4% paraformaldehyde (PFA) until their limbs were stiff. Gradient alcohol dehydration, xylene clearing, and routine paraffin embedding were performed, and 5-μm-thick coronal sections were sliced.

Immunohistochemical Staining
The sections prepared by fixed paraffin embedding were deparaffinized and washed in PBS and then 3% H 2 O 2 at room temperature for 12 min, PBS washed 3 times, 3 min each; 3% goat serum (Beijing Zhongshan Jinqiao, China) for 30 min. Without washing, the sections were incubated overnight at 4 °C with the following antibodies: mouse anti-ATF6α monoclonal antibody (1:300), mouse anti-GRP78 monoclonal antibody (1:200), and rabbit anti-CHOP polyclonal antibody (1:200) (Santa Cruz Biotechnology, USA). The sections were washed with PBS 3 times for 3 min each; biotin secondary antibody (Beijing Zhongshan Jinqiao, China) was added and incubated at room temperature for 30 min; the sections were washed with PBS; and SABC reagent (Beijing Zhongshan Jinqiao, China) was added dropwise and incubated at room temperature for 30 min. The sections were washed with PBS 3 times for 3 min each, DAB color development (Beijing Zhongshan Jinqiao, China) was performed, and the sections were counterstained with hematoxylin, dehydrated with gradient alcohol, cleared with xylene, and mounted. The slides were observed and photographed under an optical microscope (Olympus, BX60, Japan), Image-Pro Plus software was used to calculate the average positive rate for semi-quantitative analysis.

TUNEL Staining
Sections were deparaffinized, washed with PBS, washed with 4% PFA for 30 min, washed with PBS, and washed with 0.1% Triton X-100 (Boster, China) for 2 min. TUNEL detection solution was prepared in an ice bath as follows. TdT Enzyme: TUNEL detection solution (Suzhou Kaiji Biotechnology Co., Ltd., China) was prepared via 1:24 dilution in fluorescent labeling solution. The cells were washed with PBS, 50 μl TUNEL reaction solution was added dropwise to the cells and incubated at 37 °C for 60 min, and the cells were washed with PBS. After mounting the slides, photos were taken with a fluorescence microscope, and the apoptosis rate was calculated according to the following equation: apoptosis rate = number of TUNEL-positive cells/total number of cells × 100%.

Western Blotting to Detect the Expression of ATF6α, GRP78, S1P, S2P, CHOP, and Caspase-12 in the Hippocampus
Hippocampal tissue was isolated following decapitation. After weighing, lysis buffer was added at a ratio of 1:3, and the tissue was homogenized, ultrasonically crushed, and centrifuged for 30 min (12,000 rpm/min). Determination of protein concentration was performed by a BCA Kit (SuoLaibao Bio, Beijing, China) and was 3 mg/ml; 50 µg of protein from each tissue sample was subjected to 10% SDS-PAGE denaturing gel electrophoresis (Bio-Rad, USA) at 90 v for 30 min (concentration gel) and then at 110 v for 120 min (separation gel). The proteins were transferred to PVDF membranes (Beijing Solebao, China) at a constant pressure overnight at 4 °C, and the membranes were blocked with 5% BSA (SuoLaibao Bio, Beijing, China) for 2 h. The following primary antibody solutions were prepared: mouse anti-ATF6α (1:1000), mouse anti-GRP78 (1:1000), rabbit anti-S1P and S2P polyclonal antibody (Thermo Fisher, USA; 1:500), rabbit anti-CHOP (1:5000), rabbit anti-Caspase-12 polyclonal antibody (Santa Cruz Biotechnology, USA; 1:5000), mouse anti-GAPDH, and mouse anti-β-actin monoclonal antibody (Proteintech, China; 1:5000). The PVDF membranes were incubated in the antibody incubation solutions, and after they were rinsed 3 times with TBST, the second antibody (1:5000) was added and incubated at room temperature for 2 h, and the membranes were rinsed 3 times with TBST. An ECL Luminometer (Bio-Rad, USA) was used to detect luminescence. ImageJ software was used to analyze the gray values of the bands, and the ratio of the target band and the internal reference GAPDH/β-actin × 100% was used as the relative protein expression.

qRT-PCR
Ten rats in each group were anesthetized and decapitated to isolate the hippocampus. Total RNA was extracted with TRIzol reagent (SuoLaibao Bio, Beijing, China) and quantified with a spectrophotometer (Thermo Fisher, USA). One microgram of RNA was reverse transcribed into cDNA under the following conditions: 37 °C, 15 min; 85 °C, 5 s. Quantitative PCR detection was performed under the following conditions: 95 °C, 2 min; 95 °C, 30 s; 60 °C, 30 s. The fluorescent signal (40 cycles) was collected to start the reaction. GAPDH was used as an internal reference. The sequence of each primer is shown in Table 1.

Statistical Analysis
Statistical analysis was performed using SPSS 22.0 statistical software. Measurement data are expressed as the mean ± SEM. One-way analysis of variance was used for comparisons between groups, and the LSD t test was used for multiple comparisons among groups. P < 0.05 indicates that the difference is statistically significant.

Behavioral Verification of the Rat Model of PTSD
The results of the Morris water maze positioning and navigation experiment showed that the average escape latency of rats in the SPS group was significantly higher than that of rats in the control group on the 1st to 5th day after SPS (P < 0.01, Fig. 1a, b). Analysis of the results of the space exploration experiment showed that the percentage of time spent in the target quadrant in the SPS group (13.01 ± 0.88) was significantly lower than that in the control group (22.43 ± 0.76) (P < 0.01, Fig. 1c). In the open-field experiment, the number of times the rats in the SPS group entered the central area (3.73 ± 1.16) and the percentage of the central area movement distance (8.27 ± 0.66) were significantly lower than those in the control group (6.87 ± 0.99 and 13.29 ± 1.43, respectively) (P < 0.01, Fig. 1d-f).

Changes in Apoptosis of Hippocampal Neurons of SPS Rats
After TUNEL staining, the cell nuclei that were observed to be stained red under a fluorescence microscope were considered apoptosis-positive cells. In the control group, a relatively small number of apoptotic cells were observed. The number  of apoptosis-positive cells increased at 1 day after SPS. The number of apoptotic cells in the SPS 7 d group was significantly higher than that in the control group, and the number of apoptotic cells reached a peak at 14 d after SPS (Fig. 2a). The difference in the number of apoptotic positive cells between the SPS groups and the control group is shown in Fig. 2b.

Changes in ATF6α Expression in the Hippocampus of SPS Rats
Positive expression of ATF6 α was indicated by brown-yellow granules (Fig. 3a). Compared with that in the control group, the expression of ATF6α in the SPS 1 d and SPS 7 d groups increased significantly, and the expression of ATF6α in the SPS 14 d group decreased significantly (P < 0.01, Fig. 3c).
When endoplasmic reticulum stress occurs, ATF6α can be hydrolyzed twice and release a 50-kDa fragment, namely, p50 ATF6α. To determine whether SPS induced the activation of ATF6α, Western blotting was performed, and the results showed that the expression of P90 ATF6 α was consistent with the immunohistochemistry results; moreover, ATF6α appeared in the area of approximately 50 kDa. The active fragment p50 ATF6α had a band, and its change was consistent with that of p90 ATF6α (P < 0.01, Fig. 3b, e, f).  The mRNA levels of ATF6α in the SPS 1 d, 7 d, and 14 d groups were, respectively, 1.18 ± 0.03, 1.54 ± 0.04, and 1.07 ± 0.02 times that in the control group. After SPS, the level of ATF6α mRNA in the rat hippocampus gradually increased and decreased on the 14th day (P < 0.01, Fig. 3d).

Changes in GRP78 Expression in the Hippocampus of SPS Rats
Positive staining of GRP78 was mainly distributed in the periplasm of neurons in the hippocampus, and the positive signal was visible as brown-yellow particles (Fig. 4a).
Statistical analysis showed that the expression of GRP78 increased on the 1st day after SPS, increased significantly on the 7th day after SPS, and decreased significantly on the 14th day after SPS (P < 0.01, Fig. 4c).
Western blotting was used to detect the expression of GRP78 in the hippocampus of each group, and a single band was observed at 78 kDa. The expression of the GRP78 protein gradually increased 1 d after SPS, reached a peak at 7 d, and then decreased. (P < 0.01, Fig. 4b, e).
qRT-PCR results showed that the expression levels of GRP78 mRNA in the SPS 1 d, 7 d, and 14 d groups were, respectively, 1.21 ± 0.02, 1.64 ± 0.06, and 1.43 ± 0.05 times d mRNA expression levels of ATF6α as measured by RT-qPCR. *P < 0.01 vs the control group, # P < 0.01 vs the SPS 1 d group, and △ P < 0.01 vs the SPS 7 d group that in the control group. That is, after SPS, GRP78 mRNA levels gradually increased, reached a peak at 7 d, and then decreased. (P < 0.01, Fig. 4d).

Effect of AEBSF on the ATF6α Pathway in the Hippocampus of SPS Rats
Western blotting was used to detect the protein expression of ATF6α in the hippocampus. The results showed that after pretreatment of SPS 7 d rats with AEBSF, the expression of p90 ATF6α and p50 ATF6α was significantly reduced (P < 0.01, Fig. 5a-c), and the expression of S1P and S2P, the two proteases that cleave ATF6α, also decreased (P < 0.01, Fig. 5e-g). However, pretreatment of normal rats with AEBSF had no significant effect on the ATF6α pathway (P > 0.05).
qRT-PCR: The mRNA expression levels of ATF6α in the SPS 7 d, SPS 7 d + AEBSF, and control + AEBSF groups were, respectively, 1.55 ± 0.02, 1.21 ± 0.03, and 1.08 ± 0.02 times that in the control group. After pretreatment with AEBSF, the mRNA expression of ATF6α was significantly reduced in the rats (P < 0.01, Fig. 5d). The fold changes in S1P mRNA expression in the SPS 7 d, SPS 7 d + AEBSF, and control + AEBSF groups relative to that in the control group were 4.22 ± 0.08, 1.54 ± 0.04, and 0.99 ± 0.02, respectively. That is, after pretreatment of SPS 7 d rats with AEBSF, the mRNA expression of S1P was significantly reduced (P < 0.01, Fig. 5h). The fold changes in S2P mRNA expression in the SPS 7 d, SPS 7 d + AEBSF, and control + AEBSF groups relative to that in the control group were 5.15 ± 0.08, 2.57 ± 0.05, and 1.19 ± 0.03 times, respectively. That is, after pretreatment of SPS 7 d rats with AEBSF, S2P mRNA expression was significantly reduced (P < 0.01, Fig. 5i). After pretreatment of normal rats with AEBSF, there was no significant effect on the mRNA expression levels of ATF6α, S1P, or S2P (P > 0.05).

Effects of AEBSF on the Expression of CHOP and Caspase-12 in the Hippocampus of SPS Rats
The expression of CHOP was visible as brown-yellow particles (Fig. 6a). Statistical analysis showed that the expression of CHOP was significantly reduced after pretreatment of SPS 7 d rats with AEBSF (P < 0.01, Fig. 6a, c). Pretreatment with AEBSF in normal rats had no significant effect on CHOP expression (P > 0.05).
Western blotting: The expression of CHOP and caspase-12 was significantly reduced after pretreatment with AEBSF in SPS 7 d rats. After pretreatment of normal rats with AEBSF, there was no significant effect on the expression of CHOP or caspase-12 (P > 0.05).
qRT-PCR: The mRNA expression levels of CHOP in the SPS 7 d, SPS 7 d + AEBSF, and control + AEBSF groups were, respectively, 6.17 ± 0.05, 4.43 ± 0.06, and 1.05 ± 0.03 times that of the SPS 7 d group. After pretreatment of rats with AEBSF, the mRNA expression of CHOP was significantly reduced (P < 0.01, Fig. 6d). The mRNA expression levels of caspase-12 in the SPS 7 d, SPS 7 d + AEBSF, and control + AEBSF groups were, respectively, 18.57 ± 0.87, 12.19 ± 0.27, and 1.17 ± 0.02 times that in the control group. That is, after pretreatment of SPS 7 d rats with AEBSF, the mRNA expression of caspase-12 was significantly reduced (P < 0.01, Fig. 6e). Pretreatment with AEBSF in normal rats had no significant effect on the mRNA expression of CHOP or caspase-12 (P > 0.05).

Fig. 5
Effect of AEBSF on the ATF6α pathway in the hippocampus of rats after SPS. a, b, c, e, f, g Expression levels of ATF6α and S1P and S2P in the rat hippocampus as measured by Western blotting. d, h, i mRNA expression levels of ATF6α, S1P, and S2P as measured by qRT-PCR. *P < 0.01 vs the control group, # P < 0.01 vs the SPS 7 d group, and △ P < 0.01 vs the SPS 7 d + AEBSF group

Effects of AEBSF on the Apoptosis of Hippocampal Neurons in SPS 7-d Rats
TUNEL staining showed that there were fewer apoptosispositive cells in the control group. The number of apoptosispositive cells was significantly increased at 7 days after SPS. After pretreatment of SPS 7 d rats with AEBSF, apoptosis was significantly reduced. Statistical analysis showed that after SPS 7 d rats were pretreated with AEBSF, the apoptosis rate was significantly reduced (P < 0.01, Fig. 7a, b), while after AEBSF pretreatment in normal rats, there was no significant effect on the apoptosis rate (P > 0.05).

Effect of AEBSF on the Expression of ATF6α, S1P, S2P, CHOP, and Caspase-12 in the Hippocampus of SPS Rats
Western blotting: The results showed that at 14 d after SPS, the expression levels of ATF6α, S1P, S2P, CHOP, and Caspase-12 were increased, and after pretreatment of SPS 14 d rats with AEBSF, the expression of ATF6α, S1P, S2P, CHOP, and Caspase-12 was reduced (P < 0.05, Fig. 8a, b, c, e, f, h, j, k, m). However, pretreatment of normal rats with AEBSF had no significant effect on the expression of ATF6α, S1P, S2P, CHOP, or Caspase-12 (P > 0.05).  (× 400). b, f, g Relative protein expression of CHOP and caspase-12 in the hippocampus of rats as measured by Western blotting. d, e mRNA expression of CHOP and caspase-12 in the hippocampus of rats as measured by RT-qPCR. *P < 0.01 vs the control group, # P < 0.01 vs the SPS 7 d group, and △ P < 0.01 vs the SPS 7 d + AEBSF group Fig. 7 Apoptosis detected by TUNEL assays. Representative positive cells are shown (a, red, × 400). b Quantification of apoptotic cells. *P < 0.01 vs the control group, # P < 0.01 vs the SPS 7 d group, and △ P < 0.01 vs the SPS 7 d + AEBSF group qRT-PCR: The mRNA expression of ATF6α in the SPS 14 d, SPS 14 d + AEBSF, and control + AEBSF groups was, respectively, 1.35 ± 0.05, 1.15 ± 0.01, and 1.03 ± 0.15 times that in the control group. After pretreatment with AEBSF, the mRNA expression of ATF6α was reduced (P < 0.01, Fig. 8d). The fold changes in S1P mRNA expression levels in the SPS 14 d, SPS 14 d + AEBSF and control + AEBSF groups relative to that in the control group were 1.83 ± 0.08, 1.60 ± 0.06, and 1.00 ± 0.02, respectively. That is, after pretreatment of SPS 14 d rats with AEBSF, the mRNA expression of S1P was reduced (P < 0.05, Fig. 8g). The fold changes in S2P mRNA expression levels in the SPS 14 d, SPS 14 d + AEBSF, and control + AEBSF groups relative to that in the control group were 1.86 ± 0.07, 1.54 ± 0.06, and 1.11 ± 0.04, respectively. That is, after pretreatment of SPS 14 d rats with AEBSF, the mRNA expression of S2P was reduced (P < 0.05, Fig. 8i). The fold changes in CHOP mRNA expression in the SPS 14 d, SPS 14 d + AEBSF and control + AEBSF groups relative to that in the control group were 2.10 ± 0.06, 1.81 ± 0.05, and 1.05 ± 0.02, respectively. That is, after pretreatment of SPS 14 d rats with AEBSF, the expression of CHOP mRNA was reduced (P < 0.05, Fig. 8l). The fold changes in caspase12 mRNA expression in the SPS 14 d, SPS 14 d + AEBSF, and control + AEBSF groups relative to that in the control group was 2.21 ± 0.03, 1.60 ± 0.03, and 1.18 ± 0.04, respectively. That is, after pretreatment of SPS 14 d rats with AEBSF, the expression of caspase12 mRNA was reduced (P < 0.05, Fig. 8n). After pretreatment of normal rats with AEBSF, there was no significant effect on the mRNA expression levels of ATF6α, S1P, S2P, CHOP or Caspase-12 (P > 0.05).

Effects of AEBSF on Apoptosis of the Hippocampal Neurons of SPS 14-d Rats
TUNEL staining showed that there were fewer apoptosis-positive cells in the control group. The number of apoptosis-positive cells was increased at 14 days after SPS. After pretreatment of 14 d SPS rats with AEBSF, apoptosis was reduced. Statistical analysis showed that after pretreatment of the 14 d SPS rats with AEBSF, the apoptosis rate was reduced (P < 0.05, Fig. 9a, b), while after pretreatment of normal rats with AEBSF, there was no significant effect on the apoptosis rate (P > 0.05).

Fig. 8
Effect of AEBSF on the expression of ATF6α, S1P, S2P, CHOP, and Caspase-12 in the hippocampus of rats after SPS. a, b, c, e, f, h, j, k, m Expression levels of ATF6α, S1P, S2P, CHOP, and Caspase-12 in the rat hippocampus as measured by Western blotting. d, g, i, l, n mRNA expression levels of ATF6α, S1P, S2P, CHOP, and Caspase-12 as measured by qRT-PCR. *P < 0.05 vs the control group, # P < 0.05 vs the SPS 14 d group, and △ P < 0.05 vs the SPS 14 d + AEBSF group Discussion PTSD refers to abnormal psychological and physiological responses to severe stressors such as war. Its core clinical symptoms include recurrent memories of traumatic experiences, increased alertness, continuous avoidance, and negative cognitive and emotional changes. Although the understanding of the molecular mechanism of PTSD has made great progress over the years, the exact pathogenesis is still unclear. MRI examination of PTSD patients showed that the hippocampal volume decreased (Milani et al. 2017;Moyer 2016;Nisar et al. 2020), and animal experiments have also found that the hippocampus in PTSD patients was closely related to memory loss after a long delay (Joshi et al. 2020;Shan et al. 2020). The above results show that abnormalities in the hippocampus are important in the pathogenesis of PTSD. Therefore, in this study, we focused on testing the hippocampus of rats to better understand the pathogenesis of PTSD. SPS is a widely adopted model of PTSD (Lisieski et al. 2018;Yamamoto et al. 2009). In this study, we observed that the SPS group showed PTSD-like symptoms, such as loss of appetite and irritability. This may be due to the changes in the behavior and cognitive function of the rats caused by the stress stimulation, which affects the feeding behavior of the rats and causes cognitive and behavioral abnormalities. The Morris water maze showed that the average escape latency of the SPS group was significantly higher than that of the control group, and the percentage of time spent in the target quadrant was significantly lower in the SPS group than in the control group, suggesting that SPS caused memory impairment in the rats. The open-field experiment showed that the number of times the rats entered the central area and the percentage of movement distance in the central area decreased significantly in the SPS group. The above results showed that rats after SPS showed a strong similarity to PTSD patients, which laid a foundation for the further development of this study.

Apoptosis of Hippocampal Neurons in SPS Rats
The apoptosis process involves the activation, expression, and regulation of multiple genes (Xu et al. 2019). PTSD and apoptosis of hippocampal neurons influence each other (Chen et al. 2020). In this study, TUNEL staining showed that the apoptosis rate in the hippocampus of PTSD rats was significantly increased, and the number of apoptotic cells gradually increased with time and reached a peak at 14 days after SPS, which may be one of the causes of the decreased size and function of the hippocampus. In previous research by our group, we proved that SPS induced ATF6α-dependent ERS and mPFC neuronal apoptosis (Yu et al. 2014). Since ERS induces cell apoptosis, in this article, we hypothesized that ERS and endoplasmic reticulum-related apoptosis participate in SPS-induced apoptosis in PTSD-like rat hippocampal neurons.

Abnormal Expression of ATF6α in the Hippocampal Neurons of SPS Rats
ATF6α is a typical ERS sensor (Kim et al. 2019;Oka et al. 2019). This study found that the expression of ATF6α increased on the 1st, 7th, and 14th days after SPS and was thereafter significantly reduced. When ERS occurs, ATF6α can be hydrolyzed twice and release a 50-kDa fragment, namely, p50 ATF6α. To determine whether SPS induced ATF6α activation, we then used Western blotting to detect the protein expression of ATF6α in hippocampal neurons. p90 ATF6α was expressed abundantly in normal rats. The expression increased significantly at 7 d after SPS and decreased significantly at 14 d after SPS. Moreover, at 14 d after SPS, a band of the active fragment of ATF6α, p50 ATF6α, appeared in the region of the Western blot corresponding to approximately 50 kDa, and the change was consistent with that of p90 ATF6α. The change in ATF6α mRNA expression was consistent with the Western blotting results. Therefore, SPS induced the activation of ATF6α. According to reports (Yu et al. 2020), activation of ATF6αrelated ERS can cause hepatocyte apoptosis. Our group also found that SPS-induced ATF6α-dependent ERS is involved in the apoptosis of mPFC neurons induced by PTSD (Yu et al. 2014). In this study, after ATF6α was activated, hippocampal neuronal apoptosis increased significantly. However, at 14 days after SPS, the expression of ATF6α was reduced, but apoptosis was not significantly reduced. This result suggests that other mechanisms may be involved in the apoptosis of hippocampal neurons induced by SPS.

Abnormal Expression of GRP78 in the Hippocampal Neurons of SPS Rats
GRP78 is closely related to ERS-UPR (Feng et al. 2019;Ibrahim et al. 2019). Under nonstress conditions, GRP78 binds with three endoplasmic reticulum transmembrane proteins, including ATF6α, to assist in the proper folding and transport of proteins. During ERS, GRP78 dissociates and initiates UPR signaling pathways to relieve stress and restore the homeostasis of the endoplasmic reticulum (Cassimeris et al. 2019). However, if the stimulation intensity is too high or the duration is too long and the UPR cannot effectively alleviate ERS, the UPR signaling pathway will eventually upregulate CHOP and other ERS-specific transcription factors and initiate cell apoptosis to eliminate damaged cells (Ardic et al. 2019;Lu et al. 2017). In this study, GRP78 expression gradually increased after SPS and decreased on day 14 after SPS. This indicates that the accumulation of GRP78 in hippocampal neurons is a response to SPS. These experimental data further indicate that SPS induces the neuronal chaperone GRP78 in the hippocampus to respond to the UPR. In the early stage of the UPR, cells tried to enhance the protein-processing ability of the endoplasmic reticulum to restore cellular homeostasis by increasing the level of GRP78. However, as time went on, the UPR weakened, so there was a relative decrease in GRP78 expression on the 14th day after SPS.

Inhibitory Effect of AEBSF on the ATF6α/S1P/S2P Pathway in the Hippocampus of SPS Rats
Based on the above results, we speculated that SPS may induce ATF6α activation and UPR enhancement, which in turn induces apoptosis of hippocampal neurons. However, at 14 days after SPS, the UPR weakened, and the number of apoptotic cells still increased, indicating that other mechanisms may mediate neuronal apoptosis. To determine whether ATF6α activation induces SPS to stimulate hippocampal neuron apoptosis, we added ATF6α signaling pathway inhibitors to the SPS 7 d, SPS 14 d, and control groups. It has been reported in the literature that AEBSF inhibits the nuclear transcription of ATF6 and prevents ERSinduced ATF6α cleavage, thus inhibiting the transcription of downstream genes of ATF6 (Okada et al. 2003). After pretreatment of SPS 7 d and SPS 14 d rats with AEBSF, the expression of p90 ATF6α and p50 ATF6α was significantly reduced. At the same time, the expression of the two proteases that cleave ATF6α, namely, S1P and S2P, also decreased. However, pretreatment of normal rats with AEBSF had no significant effect on the ATF6α pathway. This suggests that AEBSF can significantly inhibit the activation of the ATF6α pathway.

Inhibition of the ATF6α/S1P/S2P Pathway Downregulates the Expression of CHOP and Caspase-12
Under normal conditions, the binding of ATF6 to GRP78 is stable. When ERS occurs, ATF6 is separated from GRP78, released from the endoplasmic reticulum, and hydrolyzed into the active fragment p50ATF6α, which then exerts its own activity (La et al. 2017). After entering the nucleus, p50ATF6α participates in the transcription and expression of ERS-related genes, such as the apoptosis-related cellular molecule CHOP (Xu et al. 2018b). CHOP is an important intermediate molecule linking ERS and apoptosis. Under normal circumstances, CHOP is mainly present in the cytoplasm, and its content is very low (Xu et al. 2018a). During severe or prolonged ERS, the activation of three transmembrane proteins on the endoplasmic reticulum membrane can induce the upregulation of CHOP transcription. Continuous upregulation of CHOP is not conducive to cell survival in conditions of disrupted protein homeostasis and is the main way to induce apoptosis in the UPR Lei et al. 2017). CHOP overexpression can promote apoptosis by activating apoptosis-responsive proteins such as ERO1, GADD34, and the death receptor DR5. Moreover, CHOP may participate in cell death by regulating genes that control apoptosis, such as CHOP overexpression. It can significantly inhibit the transcriptional expression of the procell survival gene BCL2 (Moriguchi et al. 2019;Poone et al. 2015). In this study, CHOP expression in hippocampal neurons continued to increase after SPS, and CHOP mRNA expression continued to increase. These results indicate that CHOP in hippocampal neurons of PTSD rats was upregulated after SPS and that neuronal apoptosis changed. After pretreatment with AEBSF, the expression of CHOP was significantly reduced, and the rate of cell apoptosis was significantly reduced. The endoplasmic reticulum apoptosis pathway mainly induces cell apoptosis by activating caspase-12, CHOP, and other apoptotic signaling molecules (Chung et al. 2018). Studies have found that caspase-12-deficient mice can resist apoptosis caused by ERS, while other death stimuli can still induce apoptosis, indicating that caspase-12 is closely related to endoplasmic reticulum pathway apoptosis. Caspase-12 is considered to be a key molecule that mediates apoptosis in the endoplasmic reticulum pathway (Song et al. 2018). Caspase-12 upregulates caspase-9 and then activates Caspase-3, leading to apoptosis (Gao et al. 2017). We showed that at 7 d after SPS, the expression of caspase-12 in the hippocampus of rats was significantly increased. After pretreatment with AEBSF, caspase-12 expression and the apoptosis rate were significantly reduced. This is consistent with the results of TUNEL staining, suggesting that CHOP and caspase-12 are involved in the regulation of hippocampal neuronal apoptosis in PTSD rats. Furthermore, we found that after pretreatment with AEBSF, the apoptosis rate did not decrease significantly. In addition to the above results, ATF6α was significantly downregulated on the 14th day after SPS, but the apoptosis rate did not decrease significantly, indicating that the ATF6α signaling pathway was not the main mechanism of apoptosis on the 14th day of SPS, but other mechanisms were involved; these results lay a foundation for further study of the mechanism of apoptosis in hippocampal cells during PTSD.
In summary, SPS induced apoptosis of hippocampal neurons by activating the ATF6α-mediated ERS response, which suggests that ERS-induced apoptosis is involved in the occurrence of PTSD. However, since the activation of caspase-12 can also occur through the mitochondrial apoptosis pathway, it is necessary to determine whether the mitochondrial apoptosis pathway plays a key role in the induction of hippocampal neuron apoptosis after SPS, which is also one of the limitations of this experiment. In addition, the exact mechanism of SPS-induced ERS and ERS-related apoptosis remains to be further elucidated.