Abstract

The protein kinase RNA-like endoplasmic reticulum kinase (PERK) pathway, which is a branch of the unfolded protein response, participates in a range of pathophysiological processes of neurological diseases. However, few studies have investigated the role of the PERK in intracerebral hemorrhage (ICH). The present study evaluated the role of the PERK pathway during the early phase of ICHinduced secondary brain injury (SBI) and its potential mechanisms. An autologous whole blood ICH model was established in rats, and cultured primary cortical neurons were treated with oxyhemoglobin to mimic ICH in vitro. We found that levels of phosphorylated alpha subunit of eukaryotic translation initiation factor 2 (p-eIF2“) and activating transcription factor 4 (ATF4) increased significantly and peaked at 12 h during the early phase of the ICH. To further elucidate the role of the PERK pathway, we assessed the effects of the PERK inhibitor, GSK2606414, and the eIF2“ dephosphorylation antagonist, salubrinal, at 12 h after ICH both in vivo and in vitro. Inhibition of PERK with GSK2606414 suppressed the protein levels of p-eIF2“ and ATF4, resulting in increase of transcriptional activator CCAAT/enhancer-binding protein homologous protein (CHOP) and caspase-12,which promoted apoptosis and reduced neuronal survival. Treatment with salubrinal yielded opposite results, which suggested that activation of the PERK pathway could promote neuronal survival and reduce apoptosis. In conclusion, the present study has demonstrated the neuroprotective effects of the PERK pathway during the early phase of ICH-induced SBI. These findings highlight the potential value of PERK pathway as a therapeutic target for ICH.

Keywords:Intracerebral hemorrhage · Endoplasmic reticulum stress · Unfolded protein response · PERK pathway · Neuroprotection

Introduction

Stroke, also known as a cerebrovascular accident, is a morbid state produced by insufficient blood flow to meet the metabolic demands of the brain. Intracerebral hemorrhage (ICH) is the deadliest type of stroke with a 30-day mortality up to 40% and severe disability in the majority of survivors [1]. The mechanisms of ICH are extremely complex, including primary brain injury and secondary brain injury (SBI). At present, it is generally accepted that SBI plays a more critical role in the poor prognosis of hemorrhagic stroke. Unfortunately, we currently have no effective solutions to SBI, which involves oxidation, inflammation, apoptosis, and hematotoxicity [2]. SBI results in disruption of cellular metabolism and activation of a series of stress responses such as the unfolded protein response (UPR) in endoplasmic reticulum (ER) stress [3].

The ER is an important subcellular organelle in eukaryotic cells. It plays a vital role in many cellular processes that include folding of newly synthesized secretory and membrane proteins, posttranslational modifications, and regulation of intracellular Ca2+ homeostasis [4]. Normally, only properly folded proteins are transported from the ER to the Golgi apparatus; unfolded or misfolded proteins are degraded. ER stress occurs when unfolded or misfolded proteins accumulate and the folding capacity of ER chaperones exceeds the capacity of the ER lumen to facilitate their disposal. As a consequence, a battery of adaptive processes, collectively known as the UPR, can be activated that transmit signals from the ER to the cytosol and nucleus to combat harmful effects of ER stress and restore normal cellular homeostasis [5]. The UPR can remove unfolded or misfolded proteins when ER stress occurs, and it might play a significant role in cell survival [6]. However, if stimuli are severe or prolonged, ER stress responses may be unable to compensate,and cell apoptosis may be induced [7].

The UPR is triggered by activation of three sensor proteins at the ER membrane: activating transcription factor-6 (ATF6), inositol-requiring enzyme-1 (IRE1), and protein kinase RNA-like ER kinase (PERK) [8]. Activated PERK phosphorylates the alpha subunit of eukaryotic translation initiation factor 2 (eIF2“), which can block the initiation stage of translation, thereby reducing protein synthesis and decreasing the ER load [9]. If ER stress is sustained, the ER-specific apoptosis pathway is activated by promoting expression of transcriptional activator CCAAT/enhancerbinding protein homologous protein (CHOP) and caspase-12 (CASP12) [10]. In recent years, several studies have reported that the UPR plays a vital role in the fate of neuronal cells following ischemic stroke. Although ICH only accounts for 10–20% of all cerebrovascular accidents worldwide [11], it is the most devastating type of stroke with a high morbidity and mortality; up to 50% of patients die within the first 24 h [12].

It is not clear whether ER stress and the UPR are involved in mechanisms that underlie ICH-induced SBI. The purpose of this study was to investigate the role of the PERK pathway during the early phase of ICH-induced SBI and its potential mechanisms. We monitored the time course of expression of the PERK pathway and utilized two experimental tools, PERK inhibitor GSK2606414 [13, 14] and eIF2“ dephosphorylation inhibitor salubrinal [15, 16], which exert opposite effects both in vivo and in vitro.

Materials and Methods

Animals

Adult male Sprague-Dawley rats (250–300 g, Animal Center of the Chinese Academy of Sciences, Shanghai, China) were raised with free access to water and food and housed in temperatureand humidity-controlled animal quarters with a 12-h light/dark cycle. All animal experiments were approved by the Ethics Committee of the First Affiliated Hospital of Soochow University and in accordance with the National Institutes of Health Guide.

ICH Model

The ICH model was established in rats using stereotaxic injection of autologous whole blood according to a previous report [17] with some modifications. In brief, rats were anesthetized and then mounted on a stereotaxic frame (ZH-Lanxing Btype, Anhui Zhenghua Biological Equipment Co. Ltd. Anhui, China). Then, a cranial burr hole was drilled 0.2 mm anterior to bregma and 3.5 mm lateral to the midline, which corresponded to the right basal ganglia. Autologous whole blood (100 μL) was adjunctive medication usage collected by cardiac puncture and injected slowly (5.5 mm ventral to the cortical surface, 20 μL/min) with a microinjector (Hamilton Company, NV, USA). To prevent reflux, the needle was kept in place for an additional 5 min. The bone hole was sealed with bone wax, and the scalp was then disinfected and sutured. During the entire surgery, rats were placed on a heating pad in a supine position, and the pad was maintained at ~27–35 °C. Vital signs were monitored continuously. After establishment of the ICH model, the rats were returned to their cages with food and water. A representative brain coronal section was shown in Fig. 1a.

Experimental Design

There were two types of in vivo experiments. In experiment 1, we analyzed the time course of changes in levels of p-eIF2“ and ATF4 after ICH. A total of 72 rats (80 rats were used, 72 rats survived after surgery) were randomly (used the randomization table) divided into six groups of 12 rats per group, which included a sham group and five experimental groups arranged by time after ICH: 4, 8, 12, 16, and 24 h. At the indicated time point after ICH, rats were killed, and the brain samples of six rats in each group were dissected and used for Western blot analysis. Double immunofluorescence staining of p-eIF2“ and ATF4 with neuronal nuclei (NeuN) was performed in the sham group and 12 h after ICH (Fig. 1b).
In experiment 2, 108 rats (129 rats were used, 108 rats survived) were randomly (used the randomization table) divided into six groups of 18 rats per group: sham, ICH, ICH + vehicle (for GSK2606414), ICH + GSK2606414 (90 μg in 5 μL sterile saline), ICH + vehicle (for salubrinal), and ICH + salubrinal (1 mg/kg body weight). Neurological scoring and brain edema were assessed at 12 h after ICH. Expression levels of p-eIF2“, ATF4, CHOP, and CASP12 were determined by Western blot analysis at 12 h after ICH. Finally, terminal deoxynucleotidyl transferasemediated dUTP nick end labeling (TUNEL) and fluoro-jade B (FJB) staining were also performed at 12 h after ICH in each group (Fig. 1c).

Effects of PERK pathway on neurons subjected to OxyHband the potential mechanisms

Fig. 1 Intracerebral hemorrhage model and experimental design. (a) Representative whole brains and brain slices from ICH model rats. (b) Experiment 1 was designed to evaluate expression of p-eIF2α and ATF4 at different time points. (c) Experiment 2 was designed to investigate effects of the PERK pathway on ICH-induced SBI and potential mechanisms. (d) Experiment 3 was designed to investigate the role of the PERK pathway in vitro.

In experiment 3, primary rat cortical neurons were treated with oxyhemoglobin (OxyHb) (10 μmol/L) to mimic effects of ICH in vitro. The experimental groups were similar to those of experiment 2 in vivo, and we assessed changes in protein levels of p-eIF2α, ATF4, CHOP, and cleaved CASP12. At 12 h after OxyHb treatment, a sulforhodamine B (SRB) assay was used to test cell viability, and the cell culture supernatants were collected for lactate dehydrogenase (LDH) activity detection. Double immunofluorescence staining of TUNEL and NeuN was performed in all groups (Fig. 1d).

For neurological scoring and brain edema evaluation, the observers did not know group of rats, either the component of infusion. For Western blot analysis, the bands were collected from one independent experiment using one rat, and the statistical data were from at least six rats. For all the immunofluorescence analysis, the representative images were from at least three independent experiments using six rats.

Antibody Characterization and Drugs

Anti-p-eIF2α antibody (ab32157), anti-eIF2α antibody (ab169528), anti-CHOP antibody (ab11419), anti-CASP12 antibody (ab62484), mouse anti-NeuN monoclonal antibody (ab104224), and anti-β-tubulin antibody (ab179513) were purchased from Abcam (Cambridge, MA, USA). Anti-ATF4 antibody (sc-200) was purchased from Santa Cruz (Santa Cruz, CA, USA). Salubrinal and GSK2606414 were purchased from TargetMol (Boston, MA, USA).

Drug Administration

One hour after surgery, the PERK pathway inhibitor, GSK2606414, was dissolved in dimethyl sulfoxide (DMSO) and further diluted in sterile saline to a final concentration of 0.5%. Five microliters of GSK2606414 (90 μg) was then administered intracerebroventricularly at a rate of 0.5 μL/ min [18]. The microsyringe was left in situ for another 10 min before being removed slowly. The eIF2α dephosphorylation inhibitor, salubrinal, was infused intraperitoneally at a dose of 1 mg/kg in saline with 1.5% DMSO [19]. Equal volumes of DMSO solutions were respectively administered to vehicle control animals.

Intracerebroventricular Injection

Intracerebroventricular injection was conducted as reported previously [20]. Briefly, rats were placed in a stereotaxic frame after anesthetization as described above. Then, a small burr hole was drilled into the skull 1.0 mm lateral to and 1.5 mm posterior to bregma over the left hemisphere. The needle of a 10 μL Hamilton syringe was slowly inserted through the burr hole into the left lateral ventricle 4.0 mm below the dural surface. A reagent was infused into the left lateral ventricle at a rate of 0.5 μL/min.

Establishment of the In Vitro ICH Model and Cell Treatment

Isolation and culture of primary cortical neurons has been described previously [21, 22]. Briefly, whole brains of 17-day rat embryos were used to prepare primary neuron-enriched cultures. Every effort was made to minimize the number of embryos used and their suffering. First, we removed the blood vessels and the meninges. Then, the brain tissues were digested with 0.25% trypsin for 5 min at 37 °C. After termination of digestion, the suspension was centrifuged at 1500 rpm for 5 min, and the pellet was resuspended in plates and cultured in Neurobasal Medium (GIBCO, Carlsbad, CA, USA). Cultures were maintained in an atmospheric incubator at SCH727965 37 °C with 5% CO2. Neurons were cultured for 2 weeks, and half of the media was replaced every 2 days. To mimic ICH, neurons were treated with 10 μM OxyHb [23]. The cultures were divided into four groups as follows: control; OxyHb treatment for 12 h; OxyHb + vehicle (for GSK2606414), pretreatment with GSK2606414 (1 μM) for 1 h, thorough rinsing, and OxyHb treatment for 12 h [24]; OxyHb + vehicle (for salubrinal); and pretreatment with salubrinal (50 μM) for 1 h,thorough rinsing, and OxyHb treatment for 12 h [25].

Neurological Scoring

At 12 hafterICH, rats in experiment 2 were assessed by neurological scoring before euthanasia. All rats were evaluated using a previously published scoring system that monitored appetite, activity, and neurological deficits [21] (Table 1).

Brain Edema

The indexof brain edema was determined using the wet/dry method as described previously [26]. Briefly, the brain tissue was removed and collected, and the samples were weighed immediately (wet weight), followed by drying at 100 °C for 72 h. And then the tissues were reweighted to obtain the dry weight. The percentage of water content was calculated as follows: [(wet weight − dry weight)/wet weight] × 100%.

Cell Viability

Neuronal viability was evaluated by SRB assay. Following treatment incubation, the culture medium was removed, and neurons were fixed with 10% trichloroacetic acid (TCA) followed by staining with 0.4% SRB. Absorbances were measured at 540 nm with a Bio-Rad Microplate reader. Cell viability was measured in triplicate and repeated at least three independent times.

LDHAssay

The concentrations of LDH in the culture medium were measuredusing a LDH detecting kit (A020-2; Jiancheng Biotech, Nanjing, China) according to the instructions. The data were presented relative to standard curves.

Western Blot Analysis

After perihematomal tissues were collected, the brain samples of each animal were homogenized separately and then mechanically lysed in lysis buffer (Beyotime Institute of Biotechnology, Jiangsu, China). After centrifuging at 15000 × g for 10 min at 4 °C, the supernatants were collected immediately. Protein concentration was determined using an enhanced bicinchoninic acid (BCA) protein assay kit (Beyotime Institute of Biotechnology). Then, the protein (30 μg/lane) were loaded on a 10% SDS-PAGE gel, separated, and then electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore Corporation, Billerica, MA, USA). The membrane was blocked with 5% bovine serum albumin (Biosharp, Hefei, AH, China) for 1 hat room temperature and then probed with the primary antibody overnight at 4 °C. Next, the membrane was incubated with the corresponding HRP-conjugated secondary antibody for 2 hat 37 °C and then washed with phosphate buffer saline (PBS)Tween20 (PBST). Finally, bands were visualized by enhanced chemiluminescence (ECL) as reported previously [26] and analyzed using ImageJ software. Relative quantity of proteins was determined by normalizing to levels of loading controls.

Immunofluorescence Microscopy

Brain tissues were fixed in 4% paraformaldehyde and embedded in paraffin. The tissues were cut into 4 μmsections and dewaxed immediately before immunofluorescence staining. Double immunofluorescence was performed with primary antibodies for p-eIF2“ or ATF4 and NeuN. After washing three times with PBS, the samples were stained with appropriate secondary antibodies. All primary antibodies were applied at a dilution of 1:100, and all secondary antibodies were diluted 1:500. Normal rabbit IgG was used as a negative control (data not shown). Sections were observed with a fluorescence microscope (BX50/BX-FLA/DP70, Olympus Co., Japan), and relative fluorescence intensity was analyzed as described previously [27].

TUNEL Staining

Quantitation of apoptotic cells was performed using TUNEL staining according to the manufacturer’s protocol (DeadEnd FluorometricKit, Promega, WI, USA). Three sections per rat were examined and photographed in parallel for TUNELpositive cell counting.

FJB Staining

FJB staining was used to reveal the neuronal degradation, which was sensitive and highly specific [28]. The procedures were performed as previously described [29]. In brief, the brain sections were deparaffinized and then dried in an oven. Then, sections were rehydrated using xylene and a series of graded ethanol solutions followed by water. Brain sections were permeabilized in 0.04% Triton X-100 and incubated with FJB dye solution. Then they were observed and photographed in parallel by a fluorescence microscope (BX50/ BX-FLA/DP70, Olympus Co.). The FJB-positive cell numbers were counted after being observed and photographed in parallel for six microscopic fields in each tissue. Microscopy was performed by an observer blind to the experimental group.

Statistical Analysis

GraphPad Prism 7 was used for all statistical analysis. Neurobehavioral scoring is presented as the median with the interquartile range. All other data represent mean ± SEM. One-way ANOVA for multiple comparisons and the Student-Newman-Keuls post hoc test were used to assess differences among all groups. Differences were considered significant atp < 0.05.Post hoc power analysis was performed according to a power analysis (PRISM, t-test comparison of the mean). Based on a two-sample t-test with a specified mean difference between the sham and ICH group, an estimated standard deviation was calculated, and alpha = 0.05, power > 0.75 for a sample size of n = 6 per groups. We assigned six rats in each groups because this number was close to the prediction.

Results

Elevation of p-eIF2“ and ATF4 Levels in Brain Tissues After ICH

In experiment 1, the Western blot analysis showed that the ICH group expressed higher protein levels of p-eIF2“ and ATF4 compared with the sham group. After induction of ICH, protein levels of p-eIF2“ and ATF4 in brain tissues were significantly elevated at 4 hand peaked at 12 h, which were remarkably higher in the 12 h group compared with the 8 h and 16 h groups (Fig. 2a, b). Double immunofluorescence staining in shamand ICH groups further verified that p-eIF2“ and ATF4 were markedly expressed in neurons and increased at 12 hafter ICH (Fig. 2c, d). Hence, we focused on the PERK pathway in neurons at 12 h after ICH in the following studies.

Fig. 2 Protein levels of p-eIF2“ and ATF4 in brain tissues after ICH.Immunofluorescence in brain tissues. Double immunofluorescence was (a) Western blot analysis and quantification of p-eIF2“ and eIF2“ properformed with ATF4 antibodies (green) and a neuronal marker (NeuN, tein levels at different time points following ICH in brain tissues. (b) red). Nuclei were fluorescently labeled with DAPI (blue). Scale Western blot analysis and quantification of ATF4 protein levels atdifbar = 30 μm. In A and B, mean values for the sham group or control ferent time points following ICH in vivo. (c) Immunofluorescence in group were normalized to 1.0. One-way ANOVA followed by Studentbrain tissues. Double immunofluorescence was performed with p-eIF2“ Newman-Keuls post hoc tests were used. Data are mean ± SEM. antibodies (green) and a neuronal marker (NeuN, red). Nuclei were *p < 0.05, **p < 0.01 vs. sham group; ##p < 0.01 12 h group vs. 8 h group; fluorescently labeled with DAPI (blue). Scale bar = 30 μm. (d) &p < 0.05 12 h group vs. 16 h group, n = 12.

PERK Pathway Activation Ameliorated Neurological Behavior Impairment and Brain Edema in the Early Phase of ICH

The PERK inhibitor, GSK2606414, was injected intracerebroventricularly at 1 hafterICH, and the eIF2“ dephosphorylation inhibitor, salubrinal, was injected intraperitoneally at 30 min before ICH. Then the protein levels of p-eIF2“ and ATF4 were detected by Western blot. It was shown that administering GSK2606414 significantly suppressed the increases in protein levels of p-eIF2“ and ATF4 after ICH. On the contrary, the inhibitor of eIF-2“ dephosphorylation,salubrinal, could significantly increase protein levels of p-eIF2“ and ATF4 (Fig. 3a–c). To assess the effects of manipulating the PERK pathway on neurological behavioral impairment after ICH,all rats were subjected to behavioral testing before being killed. Remarkable neurological behavioral impairment was observed in the ICH, ICH + vehicle (for GSK2606414), and ICH + vehicle (for salubrinal) groups compared with the sham group at 12 hafterICH. After intracerebroventricular GSK2606414 injection, neurological behavioral impairment was exacerbated. In contrast, after intraperitoneal injection of salubrinal, neurobehavioral deficits were significantly ameliorated (Fig. 3d). Furthermore, we evaluated brain water content in each group. In the ICH group, brain water content was significantly increased when compared with the sham group. Salubrinal treatment significantly reduced ICH-induced brain water content, whereas GSK2606414 treatment significantly increased brain edema (Fig. 3e).

PERK Pathway Activation Inhibited Neuronal Apoptosis and Necrosis Induced by ICH at 12 hIn Vivo

It has been reported that PERK signaling pathway was involved in ER stress-induced apoptosis. The ER-specific apoptosis pathway is activated by promoting expression of CHOP and CASP12 [10]. Western blot was used to measure the protein levels of CHOP and the cleavage of CASP12.The protein levels of CHOP and cleaved CASP12 significantly increased at 12 h in the ICH, ICH + vehicle (for GSK2606414), and ICH + vehicle (for salubrinal) groups (Fig. 4a). With the administering of GSK2606414,the levels of CHOP and cleaved CASP12 were significantly increased. Otherwise, with the treatment of salubrinal, it showed an opposite effect, exhibiting that significantly suppressed the increase of CHOP and cleaved CASP12 induced by ICH (Fig. 4a). In addition, histological examination showed that the number of TUNEL-positive neurons and FJB-positive cells significantly increased at 12 h in the ICH, ICH + vehicle (for GSK2606414), and ICH + vehicle (for salubrinal) groups (Fig. 4b, c). Treatment with GSK2606414 significantly increased the total number of TUNEL and NeuN double-stained cells at 12 h after ICH, as well as the FJBpositive cells, compared with the ICH + vehicle (for GSK2606414) group at 12 h after ICH (Fig. 4b, c). Compared with the inhibition experiments, activation of the PERK pathway yielded opposite results. Treatment with salubrinal significantly lowered the total number of TUNEL and NeuN double-stained cells at 12 h after ICH (Fig. 4b). Similarly, the ICH + salubrinal group showed a significant reduction in the number of FJB-positive cells compared with the ICH + vehicle (for salubrinal) group at 12 h after ICH (Fig. 4c).

PERK Pathway Activation Promoted Neuronal Survival at 12 hAfterICH In Vitro

In vitro, primary cortical neurons were subjected to OxyHb to mimic the ICH model. Similar trends were observed in expression of p-eIF2“ and ATF4 with the in vivo experiment after the treatment of GSK2606414 and salubrinal. The results demonstrated that GSK2606414 treatment could significantly suppress the increases in protein levels of p-eIF2“ and ATF4 induced by the OxyHb treatment. On the other hand, salubrinal significantly increased protein levels of p-eIF2“ and ATF4 (Fig. 5a). Compared with the control group, a significant decrease in neuronal viability was observed in the OxyHb group, and this was exacerbated by inhibition of the PERK pathway (Fig. 5b). On the contrary, after the treatment of salubrinal, the cell viability was rescued in neurons under OxyHb stimulus (Fig. 5b). Similar to the results of cell viability, the LDH release was elevated after OxyHb stimulus. And with the treatment of GSK2606414 and salubrinal, the release of LDH was exacerbated rescued respectively compared to vehicle group (Fig. 5c).

Fig. 3 Effects of PERK pathway on brain injury in vivo at 12 h after ICH. (a–c) Western blot analysis showing phosphorylation levels of eIF2“ and expression of ATF4 in the sham, ICH, ICH + vehicle (for GSK2606414), ICH + GSK2606414, ICH + vehicle (for salubrinal), and ICH + salubrinal groups at 12 hafter ICH onset. *p < 0.05 vs. sham group, **p < 0.01 vs. sham group; #p < 0.05, &p < 0.05, &&p < 0.01, n = 12. (d) Neurological scoring. ***p < 0.001 vs. sham group; #p < 0.05, &p < 0.05, n = 18. (e) Brain water content at 12 h post-ICH. ***p < 0.001 vs. sham group; #p < 0.05, &&p < 0.01, n = 6.

Fig. 4 PERK pathway was involved in ICH-induced neuronalapoptosis and necrosis in vivo at 12 h. (a) Western blot analysis showing expression of CHOP and cleaved CASP12 in the sham, ICH, ICH + vehicle (for GSK2606414), ICH + GSK2606414, ICH + vehicle (for salubrinal), and ICH + salubrinal groups at 12 h after ICH onset. **p < 0.01 vs. sham group; ##p < 0.01, &&p < 0.01, n = 6. (b) TUNEL staining showing apoptotic cells in the sham, ICH, ICH + vehicle (for GSK2606414), ICH + GSK2606414, ICH + vehicle (for salubrinal), and ICH + salubrinal groups at 12 hafter ICH onset. Double immunofluorescence was performed with TUNEL (green) and a neuronal marker (NeuN, red), and nuclei were fluorescently labeled with DAPI (blue). Scale bar = 30 μm. The percentage of TUNEL-positive neurons in each group. **p < 0.01 vs. sham group; #p < 0.05, &p < 0.05, n = 12. (c) FJB staining (green) shows neuronal degradation in the cerebral cortex. Scale bar = 26 μm. Arrows indicate FJB-positive cells. FJB-positive cells/mm2 was determined in the brain cortex at 12 h. *p < 0.05 vs. sham group; #p < 0.05, &p < 0.05, n = 12.

Fig. 5 Effects of PERK-eIF2-ATF4 pathway on in OxyHb-induced neuronal damage. (a) Western blot analysis showing phosphorylation levels of eIF2“ and expression of ATF4 in the control, OxyHb, OxyHb + vehicle (for GSK2606414), OxyHb + GSK2606414, OxyHb + vehicle (for salubrinal), and OxyHb + salubrinal groups at 12 h. **p < 0.01 vs. control group; #p < 0.05, &p < 0.05, &&p < 0.01, n = 6. (b) Cell viability in neurons was measured by SRB assay. ***p < 0.001 vs. control group; ###p < 0.001; &&p < 0.01, n = 6. (c) LDH analysis. ***p < 0.001 vs. control group; ###p < 0.001; &&&p < 0.001, n = 6.

PERK Pathway Activation Reduced Neuronal Apoptosis at 12 hAfterICH In Vitro

Experiments performed in vitro yielded similar results. With the treatment of GSK2606414, the protein levels of CHOP and cleaved CASP12 were significantly increased at 12 h in neurons after treatment with OxyHb (Fig. 6a), when compared with the OxyHb + vehicle (for GSK2606414) group. In addition, neurons subjected to OxyHb + GSK2606414 showed a significant increase in neuronal apoptosis compared with the OxyHb + vehicle (for GSK2606414) group measured by TUNEL staining (Fig. 6b). However, treatment with salubrinal could significantly suppress the expression of CHOP and cleaved CASP12 at 12 h in neurons after treatment with OxyHb (Fig. 6a) when compared with the OxyHb + vehicle (for salubrinal) group. Also, neurons subjected to OxyHb + salubrinal showed significant inhibition of neuronal apoptosis compared with the OxyHb + vehicle (for salubrinal) group (Fig. 6b).

Discussion

ER stress-induced cell death is one of the most significant causes of brain injury [30]. When ER stress occurs, cells restore ER function by initiating a series of adaptive processes through the UPR [31]. Previous reports have suggested that the PERK pathway, which is part of the UPR, may participate in a range of pathophysiological processes of neurological diseases [32, 33]. However, it is unclear whether the PERK pathway is involved in the occurrence and development of post-ICH brain injury. Here, for the first time, we explored a possible role of the PERK pathway during the early phase of ICH-induced SBI both in vivo and in vitro. As previously reported, when unfolded or misfolded protein accumulates, PERK is activated by oligomerization and trans-autophosphorylation [34]. Phosphorylated PERK specifically induces phosphorylation of eIF2“ at ser51 (p-eIF2“), which then upregulates transcription factor ATF4 (Fig. 7) [35]. In experiment 1, we first investigated spatial-temporal expression of p-elF2“ and ATF4 protein after ICH. As shown in Fig. 2a, b, under ICH condition, p-eIF2“ consolidated bioprocessing and downstream ATF4 showed the same trend, such that the ratio of p-elF2“/ elF2“ and protein levels of ATF4 were remarkably elevated at 4 hand peaked at 12 h. Furthermore, to investigate spatial expression of the PERK pathway in brain tissue, as shown in Fig. 2c, d, double immunofluorescence staining indicated that p-eIF2“ and ATF4 were markedly expressed in neurons after ICH. This suggests that the PERK pathway was activated and that this pathway may play a vital role in ICHinduced SBI. Based on these findings, subsequent experiments focused on the PERK pathway in neurons at 12 hafter ICH.

It is well known that levels of phosphorylated protein can be regulated by inhibiting both phosphorylation and dephosphorylation. Consistent with previous studies [36, 37], as shown in Fig. 3a, the potent p-eIF2“ dephosphorylation inhibitor, salubrinal, significantly increased expression of p-eIF2“ and ATF4 at 12 hafter ICH, whereas levels of both proteins were decreased by the selective PERK inhibitor, GSK2606414. Previous research has demonstrated that p-eIF2“ suppresses initiating translation of global protein synthesis, which promotes cell survival by preventing further accumulation of unfolded or misfolded proteins in the ER [38, 39]. In addition, it is important for recovery from various stresses that ATF4 triggers expression of genes involved in amino acid metabolism, antioxidant stress, protein folding, and autophagy [40]. To define the effects of the PERK pathway on ICH-induced neurological behavioral impairment and brain edema, we performed neurological scoring and measured brain water content.

As shown in Fig. 3d, e, rats showed severe neurological behavioral impairment and brain edema compared with the sham group at 12 hafter ICH induction. ICH-induced neurological deficits and edema were ameliorated after activating the PERK pathway with salubrinal and aggravated by blocking the PERK pathway with GSK2606414. In addition, TUNEL and FJB staining were utilized to explore effects of the PERK pathway on apoptosis in brain tissues at 12 h post-ICH induction. As shown in Fig. 4b, c, the numbers of TUNEL-positive cells and FJB-positive cells in brain tissue around the hematomas were significantly increased in the ICH group compared with the sham group, and these cell numbers were augmented in the ICH + GSK2606414 group and reduced in the ICH + salubrinal group. These data clearly suggest that increasing expression of p-eIF2“ and ATF4 promotes neuronal survival and suppresses apoptosis, and this process can be reversed by reducing these protein levels after ICH.

At least three branches participate in ER stress-induced apoptotic events. These include the CHOP pathway [41], the ER-associated CASP12 pathway [42], and the cJUN NH2terminal kinase (JNK) pathway [43, 44]. Interestingly, as shown in Fig. 4a, GSK2606414 administration significantly increased CHOP expression at 12 h after ICH, while treatment with salubrinal remarkably suppressed expression of Representative images from control, OxyHb, OxyHb + vehicle (for GSK2606414), OxyHb + GSK2606414,OxyHb + vehicle (for salubrinal), and OxyHb + salubrinal groups. Each group was subjected to OxyHb except for the control group. Scale bar = 20 μm. The percentage of TUNELpositive cells. **p < 0.01 vs. control group; #p < 0.05, &p < 0.05, n = 6.

Fig. 6 PERK-eIF2-ATF4 pathway participated in OxyHb-induced neuronal apoptosis in vitro. (a) Western blot analysis showing expression of CHOP and CASP12 in the control, OxyHb, OxyHb + vehicle (for GSK2606414),OxyHb + GSK2606414,OxyHb + vehicle (for salubrinal), and OxyHb + salubrinal groups at 12 h. **p < 0.01 vs. control group; ##p < 0.01; &&p < 0.01, n = 6. (b) TUNEL staining to elucidate the role of PERK in OxyHb-treated neurons in vitro at 12 h.

Fig. 7 Schematic representation of potential mechanisms of the PERK pathway in neuroprotection under ICH conditions. Following ICH, the activation of the ER stress response induces neuronal apoptosis. Consequently, activated PERK pathway increases the protein levels of p-eIF2“ and ATF4. At the early phase of ICH-induced brain injury, the PERK pathway is triggered to block the initiation process of translation, thereby reducing protein synthesis and decreasing the ER load, which might play a significant role in neuroprotection these pro-apoptotic proteins. CHOP, also known as growth arrestand DNA damage-inducible gene 153 (GADD153), is an ER stress-specific transcription factor that consist of an N-terminal transcriptional activation domain and a C-terminal basic-leucine zipper [45]. Normally, CHOP is expressed at extremely low levels. However, once ER stress occurs, its expression significantly increases. CHOP can be activated via the PERK-eIF2“ phosphorylation pathway, which triggers an increase in expression of ATF4. ATF4 then binds to the site of an amino acid reaction element of the CHOP promoter. In this study, we found that although the PERK-eIF2“-ATF4 axis is indispensable for CHOP expression, after treatment, there is an inverse relationship between protein levels of CHOP and expression of p-eIF2“ and ATF4. This suggests that the PERK pathway primarily plays a neuroprotective rather than pro-apoptotic role at 12 hafter ICH.

Another important pro-apoptosis protein is CASP12, which can induce apoptosis alone through ER stress rather than other apoptotic pathways. Under normal physiological conditions, CASP12 exists as an inactive zymogen similar to other cysteine proteases. Abnormal calcium can trigger specific activation of CASP12 in the ER, which coordinates with other ER stress molecules to activate CASP9 that transmits information to CASP3, causing cells to eventually undergo apoptosis [46]. In CASP12-deficient cells, apoptosis can be evoked by certain stimuli other than ER stress, which suggests that CASP12 is a specific apoptosis factor associated withER stress [47]. Therefore, we selected CHOP and CASP12 as markers of ER stress and apoptosis. As shown in Fig. 3b, results support anti-apoptotic effects of the PERK pathway during the early phase of ICH-induced SBI. However, although CASP12 has been recognized as a marker of ER stress-induced apoptosis in rats and mice, humans lack a functional CASP12 homologue due to multiple stop codons [45]. This represents a major impediment in translation from basic experiments to clinical practice.

In many previous in vitro studies, ER stress could be induced by tunicamycin, which is a glycosylation inhibitor, or thapsigargin, which is a highly selective inhibitor of the ER Ca2+-dependent ATPase [48, 49]. To define further the role of the PERK pathway, we established an in vitro model of ICH by treating primary cortical neurons with OxyHb. As shown in Figs. 5 and 6, neurons subjected to OxyHb yielded similar results regarding TUNEL staining and expression of p-eIF2“, ATF4, CHOP, and CASP12. Compared with the control group, a significant decrease in neuronal viability was observed in the OxyHb group, and viability was reduced by GSK2606414 treatment and enhanced by salubrinal administration. Consistent with these findings, as shown in Fig. 6b, the necrosis index showed a trend similar to that of the apoptotic index. Taken together, these data further confirm that the PERK pathway plays a neuroprotective role during the early phase of SBI induced by ICH. In addition, studies of ICH have increasingly recognized the significance of particular blood components in brain injury [50]. Thus, different responses may be induced by different blood components. Although that OxyHb mimics ICH induction has been well-accepted, which specific blood component is predominantly responsible for activation of the PERK pathway is unknown.

The current study has several limitations. The generally accepted view is that in a physiological state, the three transmembrane protein receptors, ATF6, IRE1, and PERK, are bound by glucose-regulated protein 78 (GRP78), which dissociates from these receptors and allows their activation under stress conditions [51]. This point of view has been challenged because it has been reported that unfolded or misfolded proteins can bind directly to ER stress sensor proteins to activate the UPR [52]. In this study, we only focused on effects of the PERK pathway, and initiation of the PERK pathway after ICH requires further study. In addition, current knowledge indicates that the UPR protects cells from ER stress by reducing synthesis of new proteins and enhancing degradation of unfolded or misfolded proteins. However, failure of the UPR due to severe or prolonged ER stress eventually promotes apoptotic cell death, which is an effective measure of protecting an organism from rogue cells expressing dysfunctional signal molecules [53]. Unfortunately, it has not been clear how the UPR globally coordinates cytoprotective and pro-apoptotic outcomes between a survival or death fate [54].Based on this, we have proposed a hypothesis that the PERK pathway predominantly plays a neuroprotective role in the early phase and a pro-apoptotic role in the late phase of ICH-induced SBI. In our recent study, it has been reported that PERK pathway activation promoted ICH-induced SBI by inducing neuronal apoptosis in the late phase [55]. And in this study, the neuroprotective role of the PERK pathway in the early phase has been confirmed.