Original Article Possible Mechanisms of the PERK Pathway on Neuronal Apoptosis in a Rat Model of Surgical Brain Injury

Am J Transl Res 2021;13(2):732-742 www.ajtr.org /ISSN:1943-8141/AJTR0110812

Original Article Possible mechanisms of the PERK pathway on neuronal apoptosis in a rat model of surgical brain injury

Mu-Yao Wu1*, Fan Gao1*, Jia-Feng Tang1, Jin-Chao Shen2, Rong Gao3, Bao-Qi Dang1, Gang Chen4

1Department of Rehabilitation, Zhangjiagang TCM Hospital Affiliated to Nanjing University of Chinese Medicine, Suzhou, China; 2Anesthesiology Department, Zhangjiagang TCM Hospital Affiliated to Nanjing University of Chinese Medicine, Suzhou, China; 3Department of Neurosurgery, Zhangjiagang TCM Hospital Affiliated to Nanjing University of Chinese Medicine, Suzhou, China; 4Department of Neurosurgery and Brain and Nerve Research Laboratory, The First Affiliated Hospital of Soochow University, Suzhou, China. *Equal contributors. Received March 13, 2020; Accepted December 11, 2020; Epub February 15, 2021; Published February 28, 2021

Abstract: Protein kinase R-like endoplasmic reticulum kinase (PERK) is an important transmembrane protein in the endoplasmic reticulum (ER). PERK signaling has a critical function in neuronal apoptosis. This work aimed to assess PERK signaling for its function in surgical brain injury (SBI) and to explore the underlying mechanisms. Totally 120 male Sprague Dawley (SD) rats were assessed in an SBI model. The effects of the PERK inhibitor GSK2606414 were examined by Western-blot, immunofluorescent staining, TUNEL staining, fluoro-jade C (FJC) staining and neurological assays in rats with SBI. In this study, p-PERK and p-eIF2α protein amounts were increased upon SBI establishment, peaking at 24 h. Meanwhile, administration of GSK2606414 reversed these effects and prevented neuronal apoptosis. The PERK pathway has a significant function in neuronal apoptosis, and its suppression after SBI promotes the alleviation of brain injury. This suggests that targeting the PERK signaling pathway may represent an efficient therapeutic option for improving prognosis in SBI patients.

Keywords: ER stress, UPR, PERK, eIF2α, apoptosis, SBI

Introduction tion and reduce perioperative costs. However, few studies have focused on the pathophysiol- Surgical brain injury (SBI) is caused by neuro- ogy of SBI or the development of specific treat- surgery and often occurs at the edge of the ments. Research into SBI pathology has made area upon which the surgery is performed [1]. It some progress, but the specific mechanisms may lead to multiple postoperative complica- underlying its occurrence and development are tions such as cerebral edema, ischemia, intra- very complex and remain unclear. Studies have cranial hematoma, and neurological impair- shown that SBI can lead to a series of patho- ment and behavioral degeneration [2]. SBI is logical processes such as brain edema, BBB thought to be inevitable, and is often associat- damage, inflammatory reactions, and apopto- ed with postoperative neurologic deficits [3]. sis around the surgical site [5]. Recent reports Studies have confirmed that brain edema revealed that endoplasmic reticulum (ER) stress occurs several hours after SBI, which may is substantially involved in the pathological pro- impair blood-brain barrier (BBB) integrity and cess of secondary brain damage following SBI. aggravate craniocerebral injury, including secondary inflammatory reactions [4, 5]. SBI- The ER is an important membranous organelle induced nerve injury and cerebral edema may in mammalian cells, mostly contributing to pro- prolong hospital stay. Therefore, it is very im- tein folding and modification and Ca2+ storage portant to reduce secondary brain injury ca- and release. As a site of protein maturation in used by SBI, maintain neurological function cells, the ER is sensitive to changes in the inter- and reduce perioperative morbidity related to nal and external cellular environments. When intracranial neurosurgery [6]. All these actions the internal environment of the ER changes, have significant impacts on patients’ rehabilita- stress response is activated to attempt to res- Inhibiting PERK pathway alleviates apoptosis tore homeostasis; these changes in the ER are that PERK is activated after brain injury wh- called ER stress (ERs) [7]. The unfolded protein en phosphorylated at ser51. Phosphorylated response (UPR) is one of the cell response eIF2α (p-eIF2α) can also increase apoptosis by mechanisms involved in defense activation of regulating AFT4 expression [15]. Yan et al. stud- adaptation. The UPR restores the ER’s capabil- ied a rat subarachnoid hemorrhage model and ity of controlling protein folding, processing and showed that inhibiting PERK may alleviate early secretion [8]. Through the expansion of the brain damage upon subarachnoid hemorrhage membrane, the ER reduces the rate of synthe- via an Akt-related anti-apoptotic signaling path- sis of new proteins, increases partner molecule way [16]. Our previous report confirmed that synthesis and promotes the degradation of inhibiting PERK signaling in intracerebral hem- misfolded proteins to maintain its own steady orrhage can reduce secondary brain injury fol- state, thereby restoring the normal function of lowing intracerebral hemorrhage and promote cells [9]. However, when the damage exceeds neuron survival by inhibiting apoptosis [17]. the system’s repair capacity, excessive accu- However, whether PERK signaling has a func- mulation of misfolded proteins leads to pro- tion in SBI remains undefined. grammed cell death or apoptosis [10]. In this way, the intensity and duration of stimulation Therefore, the present work aimed to assess determine whether UPR promotes cell survival PERK levels and PERK-eIF2α signaling in a rat or induces apoptosis. SBI model, and to explore the possible mechanism of the PERK signaling pathway in SBI. The UPR has three parallel signal branches, including protein kinase R-like endoplasmic Materials and methods reticulum kinase-eukaryotic translation initiation factor 2α (PERK-eIF2α), inositol requiring Experimental design protein 1-x box binding protein 1 (IRE1-XBP1), and activating transcription factor 6 (ATF6) In Experiment 1, animal weights, feed intake, pathways. Each branch can regulate a series and motor function were similar in all animals. of downstream transcription factors to partici- To examine PERK, p-PERK, eIF2α and p-eIF2α pate in the survival or death of cells [2]. Under amounts following SBI, 36 rats (36 survivors normal conditions, the ER’s chaperone glu- in totally 40) were randomized to six groups cose-regulated protein 78 (GRP78/BIP) inter- using computer-based randomization, includ- acts with the transmembrane proteins PERK, ing the Sham, SBI 6 h, SBI 12 h, SBI 24 h, SBI IRE1, and ATF6, respectively, preventing signal 48 h and SBI 72 h groups. Brain tissues sur- transduction. GRP78 can promote the folding rounding the damaged area were sampled for and assembly of nascent polypeptides, prevent Western blot (WB) and double immunofluores- their misfolding and aggregation, induce prote- cence analysis. asomal degradation of misfolded proteins, and control the initiation of each UPR pathway [11]. In Experiment 2, in order to assess the func- When unfolded and misfolded proteins in the tion of PERK signaling in SBI, 72 rats (72 survi- ER increase in aggregated amounts, GRP78 vors in totally 80 animals) were randomized to dissociates from these three proteins, which four groups using computer-based randomiza- activates them and increases folding capacity tion, including the Sham, SBI, SBI+DMSO, and [12]. Increased GRP78 level has a protective SBI+GSK groups. Euthanasia was performed effect on ischemic nerves [13]. at 24 h post-SBI (according to Experiment 1), and the damaged brain tissues were collected. More and more animal model evidence sug- Neurological tests were carried out in all ani- gests that the UPR is involved in many diffe- mals prior to euthanasia. Tissues inside of the rent diseases. The UPR is gradually becoming injury were evaluated by WB, those obtained an attractive new therapeutic drug target. Di- from the back were used to prepare paraffin seases in which UPR is involved include can- sections. TUNEL and Fluoro-jade C (FJC) stain- cer, metabolic diseases, and cerebral ischemia ing were performed for detecting neuronal [14]. Because it is one of the three main path- apoptosis and necrosis. All assays were strictly ways of the UPR, the PERK-eIF2α pathway has blinded, with specimens coded by an indepen- seen increasingly studied. It has been reported dent investigator.

733 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

Animals (1:500, Abcam), p-eIF2α (1:500, Abcam), ATF4 (1:1000, Abcam) and GAPDH (1:10000, Sigma, Here, totally 120 male Sprague Dawley (SD) USA). The membranes were then incubated rats (300-350 g) were obtained from JOINN with HRP-linked goat anti-rabbit IgG (Invitro- Laboratories, Suzhou, China. Of these, 108 gen, USA) secondary antibodies for 2 h at 4°C. were analyzed. Animal housing was carried out Detection used chemiluminescent substrate under a 12 h/12 h light/dark cycle at 25°C and (Millipore) and a LUMINESCENT IMAGE ANAL- controlled relative humidity, with rodent chow YZER (GE Healthcare Bio-Sciences AB, Swe- and water ad libitum. den). ImageJ (National Institutes of Health, USA) was used to analyze immunoreactive The SBI model bands.

SBI modelling was based on a previous report Immunofluorescence [1]. The rats underwent fixation in a stereotac- tic instrument upon intraperitoneal injection of Double immunofluorescence (IF) was carried pentobarbital sodium (pentobarbital sodium for out according to a previous report [19]. Brie- intraperitoneal injection at 40 mg/kg). Midline fly, brain tissue specimens underwent fixation scalp incision was performed, followed by drill- with 4% formalin, paraffin embedding and lon- ing of a cranial window (5 mm×5 mm) in the gitudinal sectioning at 5 μm. After dewaxing right frontal lobe. The entire right frontal lobe and washing, the sections were blocked with was excised (2 mm lateral and 1 mm anterior to Immunostaining sealant (Beyotime) for 1 h at bregma). Pressure was applied to the surround- ambient, followed by incubation overnight at ing brain tissue to stop bleeding, followed by 4°C with rabbit primary antibodies targeting scalp suturing. The Sham group only underwent p-PERK (1:100, Abcam) and p-eIF2α (1:200, craniotomy and bone flap replacement without Abcam), and mouse anti-NeuN primary antibod- dural incision. Postoperatively, the rats were ies (1:1000, Abcam), respectively. Next, the euthanized at various times. specimens were incubated with Alexa Fluor 488-conjugated donkey anti-rabbit IgG (1:400, Drug injection Invitrogen) and Alexa Fluor 555-linked donkey anti-mouse IgG (1:400, Invitrogen) for 1 h at According to a previous study, GSK2606414 ambient. Finally, counterstaining was carried (MCE, USA) was dissolved at 90 μg/5 μl in 20% out with DAPI for 10 min, and a U-RFL-T fluores- dimethyl sulfoxide (DMSO), while animals treat- cence microscope (OLYMPUS, Japan) was uti- ed with the vehicle received 20% DMSO. All lized for analysis. treatments were administered intracerebroven- tricularly at 0.5 μl/min 1 h after operation [17]. TUNEL staining

Immunoblot TUNEL staining was carried out for apoptosis detection [17]. The sections were dewaxed be- WB was carried out according to a previous fore TUNEL staining, and successively incubat- report [18]. Briefly, cortex tissue specimens ed with proteinase K (37°C for 20 min) and were lysed with RIPA lysis buffer containing TUNEL detection liquid (Beyotime). TUNEL-posi- protease Inhibitors (CWBIO, China), and centri- tive neurons were assessed under a U-RFL-T fuged at 13,000×g and 4°C for 15 min. Protein fluorescence microscope (OLYMPUS, Japan). amounts in the supernatants were assessed with PierceTM BCA Protein Assay Kit (Thermo Fluoro-Jade C staining Fisher, USA). Equal amounts of protein underwent separation by 10% SDS-PAGE (Beyotime, Fluoro-Jade C (FJC) staining was performed for China) followed by transfer onto polyvinylidene detecting neuronal necrosis in accordance with difluoride membranes (Millipore, USA). Upon the standard procedure described in the kit blocking (5% skimmed milk for 2 h at ambient), (Biosensis, South Australia). Briefly, the sec- the membranes were submitted to overnight tions were dewaxed and incubated with potas- incubation at 4°C with rabbit primary antibod- sium permanganate (1:10 in distilled water) for ies targeting PERK (1:1000, Abcam, UK), 10 min. After rinsing with distilled water (2 min), p-PERK (1:500, Cell Signaling, USA), eIF2α the samples were incubated with FJC (1:10 in

734 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

Figure 1. Post-SBI PERK/p-PERK and eIF2α/p-eIF2α protein amounts in the peri-injury cortex. Immunoblot was carried out for determining PERK/p-PERK and eIF2α/p-eIF2α amounts in the Sham and SBI groups at 6 h, 12 h, 24 h, 48 h and 72 h following SBI. The Sham group was used for normalization, and quantitation used ImageJ. One-way analysis of variance (ANOVA) with post hoc Dunnett’s test was carried out for group comparisons (n=6 for each group; *P<0.05, **P<0.01, ***P<0.001 vs. Sham; ns P>0.05 vs. Sham). distilled water) in the dark for 10 min. Following Tukey’s test was performed for analyzing TUNEL 3 rinses with distilled water, the sections were and FJC staining, as well as neurological behav- dried at 60°C for 5 min, soaked in xylene for 1 ioral scores and brain water content. P<0.05 min and coverslipped with distyrene plasticizer indicated statistical significance. xylene (DPX). FJC-positive cells were photo- graphed under a U-RFL-T fluorescence micro- Results scope (OLYMPUS, Japan) and averaged in three high-power fields. Post-SBI brain expression of PERK pathway proteins Brain edema assessment PERK, p-PERK, eIF2α, and p-eIF2α protein The wet-dry technique was carried out for amounts at 6, 12, 24, 48, and 72 h after SBI assessing brain edema in the injured brain were assessed by Western blotting (Figure 1). [20]. Briefly, brain specimens were separated The results showed that p-PERK and p-eIF2α into ipsilateral and contralateral hemispheres. amounts started to increase at 6 h upon SBI After obtaining wet weights, the specimens and peaked at 24 h. Meanwhile, total protein underwent drying at 100°C for 48 h, followed levels of PERK and eIF2α were unaltered. by dry weight measurements. The rate of brain water content (%) was derived as [(wet weight- Post-SBI localization of p-PERK and p-eIF2α in dry weight)/(wet weight)] ×100%. cortical cells surrounding the injury

Neurological scoring p-PERK and p-eIF2α localizations were evaluated by IF staining with NeuN (a neuronal mark- Neurological deficiency was assessed at 24 h er) (Figure 2). Consistent with WB analysis, IF post-SBI based on the modified Garcia score findings revealed that there were numerous [21, 22]. Scores in each subtest ranged be- p-PERK-positive neurons (Figure 2A, 2C) in the tween 0 and 3 (maximum total score of 21, indi- 24 h post-SBI group compared with the Sham cating no neurological defects). group. There were also more p-eIF2α-positive neurons (Figure 2B, 2D) in the 24 h post-SBI Statistical analysis group compared with the Sham group.

Data analysis was carried out with GraphPad GSK2606414 intervention and the PERK path- Prism 8.0 (San Diego, USA). Data are mean ± way after SBI SD. One-way analysis of variance (ANOVA) was performed to compare immunoblot data, with After GSK2606414 intervention following SBI, post hoc Dunnett’s test. Student’s t-test was p-PERK and p-eIF2α protein amounts were performed for analyzing immunofluorescent markedly reduced in comparison with the SBI staining data. One-way ANOVA with post hoc group. The SBI+DMSO group was similar to the

735 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

Figure 2. Post-SBI IF micrographs depicting p-PERK and p-eIF2α in the peri-injury cortex. Representative double-immunofluorescence micrographs showing green-labeled p-PERK (A, C)/p-eIF2α (B, D) and red-labeled NeuN neurons in the Sham and 24 h post-SBI groups. Counterstaining utilized DAPI (blue). Arrows point to p-PERK/p-eIF2α co- localization with neurons. Scale bar, 50 µm. Student’s t test was performed for analyses (n=6, *P<0.05 vs. Sham).

SBI group in protein amounts. Additionally, GSK2606414 intervention and post-SBI neuro- ATF-4 amounts were significantly elevated in nal apoptosis and necrosis the SBI and SBI+DMSO groups compared with the Sham group. After GSK2606414 interven- Caspase-3 and Bcl-2 protein levels were lower tion, ATF-4 levels were remarkably decreased after GSK2606414 intervention post-SBI com- compared with the SBI group (Figure 3A). pared with those of the SBI group (Figure 3B).

736 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

Figure 3. Effect of GSK2606414 intervention on PERK signaling pathway 24 h post-SBI. PERK pathway protein amounts were assessed after GSK2606414 intervention in SBI rats (A). Caspase-3 and Bcl-2 protein amounts after GSK2606414 intervention in SBI rats (B). One-way analysis of variance (ANOVA) with post hoc Tukey’s test was carried out for comparisons (n=6 for each group; *P<0.05 vs. Sham; #P<0.05 vs. SBI+DMSO).

Neuron necrosis in the SBI group was remark- Discussion ably elevated in comparison with that of the Sham group (Figure 4A, 4B). Necrosis levels After primary injury, SBI causes postoperative were similar in the SBI and SBI+DMSO groups. nerve injury and brain edema, aggravating cra- Meanwhile, necrosis level was starkly reduced niocerebral injury. Identifying a tool to reduce in the SBI+GSK group compared with the SBI nerve injury after SBI is an important factor group. in prognosis. We here described the neuroprotective effects of the PERK signaling pathway GSK2606414 intervention improves neurologi- on preserving neurons and attenuating brain cal behavioral scores in SBI rats edema after SBI. First, we measured p-PERK and p-eIF2α protein amounts in the peri-surgi- Neurological behavioral scores were starkly cal brain tissue after SBI. The results show- decreased in the SBI group in comparison with ed that p-PERK and p-eIF2α amounts were the Sham group, while the SBI and SBI+DMSO markedly elevated and peaked at 24 h after groups had comparable values. The SBI+GSK SBI. Next, we determined whether suppressing group had overtly ameliorated neurological be- PERK signaling would be beneficial in SBI, havioral scores in comparison with the SBI especially in reducing apoptosis and brain group (Figure 4C). In addition, brain edema was edema surrounding the surgical injury. As markedly reduced in the injured hemispheres shown above, the PERK signaling pathway following GSK2606414 intervention post-SBI. inhibitor GSK2606414 markedly ameliorated However, brain edema had no significant alte- neurological function and reduced brain ede- ration in the contralateral brain (Figure 4D). ma in SBI rats. The present study firstly described the possible neuroprotective effects of Taken together, ER stress was activated in rats the PERK signaling pathway in a rat model of after SBI, and the PERK signaling pathway was SBI. activated after SBI with increased p-PERK and p-eIF2α amounts. This increased ATF4 pro- The ER lumen constitutes an oxidative environ- tein levels and led to neuronal apoptosis (Figure ment, and is essential for disulfide bond forma- 5). tion and the proper folding of proteins. ER

737 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

Figure 4. Effect of GSK2606414 intervention on PERK signaling pathway 24 hours after SBI. TUNEL (A) and FJC (B) staining in SBI rats after GSK2606414 intervention. Neurological behavioral scores in SBI rats following GSK2606414 administration (C). Brain water amounts of the bilateral hemispheres were assessed by the wet-dry technique (D). One-way analysis of variance (ANOVA) with post hoc Tukey’s test was carried out for comparisons (n=6, *P<0.05, **P<0.01 vs. Sham; #P<0.05 vs. SBI+DMSO). stress is due to the destruction of ER function curs when ERs is too severe to digest unfold- [23]. Several studies have shown that oxidative ed proteins and compensate for cell dama- stress [24], ionic cell alteration [25], high mito- ge [28]. Cerebral ischemic preconditioning can chondrial calcium amounts [26], and toxic glu- protect the brain from ischemia and reperfu- tamate release [27] can induce ER stress in sion injury via the inhibition of ERs-associated various diseases. Dual effects of ERs on cells apoptosis [29]. Under ERs, the IRE1 pathway, have been demonstrated. Mild ERs can acti- one of the three parallel signaling branches, vate autophagy, facilitate the digestion of un- mediates both adaptive and pro-apoptotic folded proteins and damaged organelles, and pathways in CNS diseases [30], inhibits the improve cell survival. However, apoptosis oc- activation of Caspase-3 and reduces free cal-

738 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

ered to mediate ER-stress- induced autophagy, causing prolonged inhibition of global protein biosynthesis, promot- ing neuronal loss and clinic- al disease [35, 36]. Inhibiting the PERK signaling pathway can improve motor function and promote neurological function recovery by attenuating neuronal damage [37]. The PERK signaling pathway is induced by high amounts of unfolded proteins in the ER and leads to neuronal death [38]. Inhibiting PERK signaling reduces early brain injury upon subarachnoid hemorrhage [39]. Consistent with the above observations, our previous studies demonstrated in cell and animal models of cerebral hemorrhage that p- elF2α and ATF4 amounts in neurons are remarkably increased, reaching the highest levels at 48 h [17]. In accordance with our previous findings, p-PERK and p-eIF2α amounts peaked at 24 h after Figure 5. Probable mechanisms underpinning PERK signaling pathway’s SBI, and GSK2606414 inhib- effects on SBI. After SBI, PERK signaling was activated and p-PERK and ited cell apoptosis and pro- p-eIF2α amounts were increased. In addition, the protein levels of ATF4 moted neuronal survival by were increased and led to neuronal apoptosis. GSK2606414 intervention suppressing p-PERK, p-eIF2α reversed these effects, reducing neuronal apoptosis and necrosis, which and ATF4 after SBI. PERK in- played a protective role in brain injury after SBI. hibition decreased Caspase-3 and Bcl-2 levels, thereby de- cium accumulation in the cytosol [31]. The inhi- creasing the rate of nerve apoptosis. The de- bition of ERs can increase nerve cell survival gree of cerebral edema was also reduced by and improve nerve function [32]. GSK2606414. Therefore, inhibiting PERK signaling showed a neuroprotective effect after PERK signaling, one of the three classical UPR SBI. branches, has a critical function in neuronal death. When the central nervous system is This study had several limitations. Other stud- damaged, its constituent proteins PERK and ies have shown that ERs has dual effects on eIF2α are phosphorylated and upregulated in the cell fate, with mild ERs activating auto- neurons. In cerebral ischemia, the PERK-eIF2α phagy to provide nutritional support for cells signaling pathway is activated [33]. Inhibition [28]. In addition, the present work only investi- of the PERK and IRE1 signaling pathways can gated the effect of severe ER stress on nerve suppress ERs-dependent autophagy and allevi- apoptosis after SBI. The mechanism by which ate acute nerve injury caused by ischemic the UPR pathway protects cells and promot- stroke [22]. The levels of PERK and eIF2α are es apoptosis between survival and death is increased in traumatic brain injury [34]. PERK- unclear. Furthermore, the GRP78 protein is eIF2α pathway induction after UPR is consid- thought to bind to the three transmembrane

739 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis protein receptors PERK, IRE1, and ATF6 under [2] Huang L, Sherchan P, Wang Y, Reis C, Applegate physiological conditions. Moreover, under se- RL, Tang J and Zhang JH. Phosphoinositide vere ERs conditions, the GRP78 protein is se- 3-kinase gamma contributes to neuroinflam- parated from PERK, ATF6, and IRE1 when mis- mation in a rat model of surgical brain injury. J folded proteins accumulate to a certain extent Neurosci 2015; 35: 10390-10401. in the ER lumen [40]. Here, only PERK path- [3] Fugate JE. Complications of neurosurgery. way’s effects were evaluated, and the two other Continuum 2015; 21: 1425-1444. [4] Vikram J, Gerald M, Hsu FP and Zhang JH. transmembrane protein receptors IRE1 and Inhibition of Src tyrosine kinase and effect on ATF6 require further investigation. Finally, PERK outcomes in a new in vivo model of surgically signaling initiation following SBI deserves fur- induced brain injury. J Neurosurg 2007; 106: ther assessment. These issues are being add- 680-686. ressed in ongoing experiments in our group. [5] Matchett G, Hahn J, Obenaus A and Zhang J. Surgically induced brain injury in rats: the Overall, these findings indicate that under effect of erythropoietin. J Neurosci Methods severe ERs, PERK signaling pathway inhibition 2006; 158: 234-241. can suppress apoptosis and ultimately reduce [6] Benggon M, Chen H, Applegate RL and Zhang secondary brain injury after SBI. Therefore, J. Thrombin preconditioning in surgical brain PERK signaling may represent an important injury in rats. Acta Neurochir Suppl 2016; 121: endogenous physiological regulatory signaling 299-304. pathway in neurons, and could be targeted for [7] Torres M, Matamala JM, Duran-Aniotz C, the alleviation of brain damage after SBI. Cornejo VH, Foley A and Hetz C. ER stress signaling and neurodegeneration: at the intersec- Acknowledgements tion between Alzheimer’s disease and Prion- related disorders. Virus Res 2015; 207: 69-75. We are grateful to LetPub (www.letpub.com) for [8] Bocai NI, Marcora MS, Belfiori-Carrasco LF, help in language editing of the present manu- Morelli L and Castano EM. Endoplasmic reticu- script. This work was funded by the Suzhou lum stress in tauopathies: contrasting human people’s livelihood science and technology brain pathology with cellular and animal mod- project (SYS2019002), Gusu health personnel els. J Alzheimers Dis 2019; 68: 439-458. training project (GSWS2019076), Zhangjiagang [9] Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and science and technology project (ZKS1712, beyond. Nat Rev Mol Cell Biol 2012; 13: 89- ZKS1942), and Zhangjiagang youth science 102. and technology project (ZJGQNKJ201913). [10] Uddin MS, Tewari D, Sharma G, Kabir MT, Barreto GE, Bin-Jumah MN, Perveen A, Abdel- Disclosure of conflict of interest Daim MM and Ashraf GM. Molecular mechanisms of ER stress and UPR in the pathogene- None. sis of Alzheimer’s disease. Mol Neurobiol 2020; 57: 2902-2919. Address correspondence to: Rong Gao, Department [11] Park KW, Eun Kim G, Morales R, Moda F, of Neurosurgery, Zhangjiagang TCM Hospital Affiliat- Moreno-Gonzalez I, Concha-Marambio L, Lee ed to Nanjing University of Chinese Medicine, No.77 AS, Hetz C and Soto C. The endoplasmic reticu- Chang’an Southern Road, Suzhou 215600, China. lum chaperone GRP78/BiP modulates prion E-mail: [email protected]; Bao-Qi Dang, Depart- propagation in vitro and in vivo. Sci Rep 2017; ment of Rehabilitation, Zhangjiagang TCM Hospital 7: 44723. Affiliated to Nanjing University of Chinese Medicine, [12] Bertolotti A, Zhang Y, Hendershot LM, Harding No.77 Changan Southern Road, Suzhou 215600, HP and Ron D. Dynamic interaction of BiP and China. E-mail: [email protected] ER stress transducers in the unfolded-protein response. Nat Cell Biol 2000; 2: 326-332. References [13] Ouyang YB, Lu Y, Yue S, Xu LJ, Xiong XX, White RE, Sun X and Giffard RG. miR-181 regulates [1] Huang L, Woo W, Sherchan P, Khatibi NH, GRP78 and influences outcome from cerebral Krafft P, Rolland W, Applegate RL, Martin RD ischemia in vitro and in vivo. Neurobiol Dis and Zhang J. Valproic acid pretreatment reduc- 2012; 45: 555-563. es brain edema in a rat model of surgical brain [14] Cnop M, Foufelle F and Velloso LA. Endoplasmic injury. Acta Neurochir Suppl 2016; 121: 305- reticulum stress, obesity and diabetes. Trends 310. Mol Med 2012; 18: 59-68.

740 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

[15] Majsterek I, Diehl JA, Leszczynska H, Mucha B, [26] Zhou Y, Peng S, Ting W, Kaixian C, Weiliang Z, Pytel D and Rozpedek W. The role of the PERK/ Heyao W and Tohru M. Inhibition of calcium in- eIF2α/ATF4/CHOP signaling pathway in tumor flux reduces dysfunction and apoptosis in lipo- progression during endoplasmic reticulum toxic pancreatic β-cells via regulation of endo- stress. Curr Mol Med 2016; 16: 533-544. plasmic reticulum stress. PLoS One 2015; 10: [16] Yan F, Cao S, Li J, Dixon B, Yu X, Chen J, Gu C, e0132411. Lin W and Chen G. Pharmacological inhibition [27] Li Y, Li J, Li S, Li Y, Wang X, Liu B, Fu Q and of PERK attenuates early brain injury after sub- Ma S. Curcumin attenuates glutamate neuro- arachnoid hemorrhage in rats through the acti- toxicity in the hippocampus by suppression of vation of akt. Mol Neurobiol 2017; 54: 1808- ER stress-associated TXNIP/NLRP3 inflamma- 1817. some activation in a manner dependent on [17] Meng C, Zhang J, Dang B, Li H, Shen H, Li X AMPK. Toxicol Appl Pharmacol 2015; 286: 53- and Wang Z. PERK pathway activation pro- 63. motes intracerebral hemorrhage induced sec- [28] Shao Z, Pei W, Xuefeng W, Meishan J, Shuang ondary brain injury by inducing neuronal apop- L, Xudong M and Huaizhang S. Tetramethylpy- tosis both in vivo and in vitro. Front Neurosci razine protects against early brain injury and 2018; 12: 111. inhibits the PERK/Akt pathway in a rat model [18] You W, Wang Z, Li H, Shen H, Xu X, Jia G and of subarachnoid hemorrhage. Neurochem Res Chen G. Inhibition of mammalian target of ra- 2018; 43: 1650-1659. pamycin attenuates early brain injury through [29] Hu YQ, Chen W, Yan MH, Lai JJ and Wu L. modulating microglial polarization after experi- Ischemic preconditioning protects brain from mental subarachnoid hemorrhage in rats. J ischemia/reperfusion injury by attenuating en- Neurol Sci 2016; 367: 224-31. doplasmic reticulum stress-induced apoptosis [19] Wang Z, Wang Y, Tian X, Shen H, Dou Y, Li H, through PERK pathway. Eur Rev Med Pharma- Chen G. Transient receptor potential channel col Sci 2017; 21: 5736-5744. 1/4 reduces subarachnoid hemorrhage-in- [30] Ni H, Rui Q, Li D, Gao R and Chen G. The role duced early brain injury in rats via calcineurin- of IRE1 signaling in the central nervous sys- mediated NMDAR and NFAT dephosphoryla- tem diseases. Curr Neuropharmacol 2018; 16: tion. Sci Rep 2016; 367: 224-31. 1340-1347. [20] Rosenberg GA, Estrada EY and Dencoff JE. [31] Ni H, Rui Q, Xu Y, Zhu J, Gao F, Dang B, Li D, Matrix metalloproteinases and TIMPs are as- Gao R and Chen G. RACK1 upregulation induc- sociated with blood-brain barrier opening after es neuroprotection by activating the IRE1- reperfusion in rat brain. Stroke 1998; 29: XBP1 signaling pathway following traumatic 2189-2195. brain injury in rats. Exp Neurol 2018; 304: [21] Wu M, Gao F, Yang X, Qin X, Chen G, Li D, Dang 102-113. B and Chen G. Matrix metalloproteinase-9 reg- [32] Tsujii S, Ishisaka M, Shimazawa M, Hashizume ulates the blood brain barrier via the hedgehog T and Hara H. Zonisamide suppresses endo- pathway in a rat model of traumatic brain inju- plasmic reticulum stress-induced neuronal cell ry. Brain Res 2020; 1727: 146553. damage in vitro and in vivo. Eur J Pharmacol [22] Feng D, Wang B, Wang L, Abraham N, Tao K, 2015; 746: 301-307. Huang L, Shi W, Dong Y and Qu Y. Pre-ischemia [33] Gharibani P, Modi J, Menzie J, Alexandrescu A melatonin treatment alleviated acute neuronal and Wu JY. Comparison between single and injury after ischemic stroke by inhibiting endo- combined post-treatment with S-Methyl-N, N- plasmic reticulum stress-dependent autopha- diethylthiocarbamate sulfoxide (DETC-MeSO) gy via PERK and IRE1 signalings. J Pineal Res and taurine following transient focal cerebral 2017; 62. ischemia in rat brain. Neuroscience 2015; [23] Roussel BD, Kruppa AJ, Miranda E, Crowther 300: 460-473. DC, Lomas DA and Marciniak SJ. Endoplasmic [34] Rubovitch V, Barak S, Rachmany L, Goldstein reticulum dysfunction in neurological disease. RB, Zilberstein Y and Pick CG. The neuropro- Lancet Neurol 2013; 12: 105-118. tective effect of salubrinal in a mouse model of [24] Goswami P, Gupta S, Biswas J, Sharma S and traumatic brain injury. Neuromolecular Med Singh S. Endoplasmic reticulum stress insti- 2015; 17: 58-70. gates the rotenone induced oxidative apopto- [35] Radford H, Moreno JA, Verity N, Halliday M and tic neuronal death: a study in rat brain. Mol Mallucci GR. PERK inhibition prevents tau-me- Neurobiol 2016; 53: 5384-5400. diated neurodegeneration in a mouse model [25] Varadarajan S, Tanaka K, Smalley JL, Bampton of frontotemporal dementia. Acta Neuropathol ET, Pellecchia M, Dinsdale D, Willars GB and 2015; 130: 633-642. Cohen GM. Endoplasmic reticulum membrane [36] Kouroku Y, Fujita E, Tanida I, Ueno T, Isoai A, reorganization is regulated by ionic homeosta- Kumagai H, Ogawa S, Kaufman RJ, Kominami sis. PLoS One 2013; 8: e56603. E and Momoi T. ER stress (PERK/eIF2a phos-

741 Am J Transl Res 2021;13(2):732-742 Inhibiting PERK pathway alleviates apoptosis

phorylation) mediates the polyglutamine-in- [39] Yan F, Cao S, Li J, Dixon B, Yu X, Chen J, Gu C, duced LC3 conversion, an essential step for Lin W and Chen G. Pharmacological inhibition autophagy formation. Cell Death Differ 2007; of PERK attenuates early brain injury after sub- 14: 230-239. arachnoid hemorrhage in rats through the acti- [37] Sun D, Wang J, Liu X, Fan Y, Yang M and Zhang vation of akt. Mol Neurobiol 2017; 54: 1808- J. Dexmedetomidine attenuates endoplasmic 1817. reticulum stress-induced apoptosis and im- [40] Nakka VP, Prakash-babu P and Vemuganti proves neuronal function after traumatic brain R. Crosstalk between endoplasmic reticulum injury in mice. Brain Res 2020; 1732: 146682. stress, oxidative stress, and autophagy: poten- [38] Li Y, Li J, Li S, Li Y, Wang X, Liu B, Fu Q and tial therapeutic targets for acute CNS injuries. Ma S. Curcumin attenuates glutamate neuro- Mol Neurobiol 2014; 53: 532-544. toxicity in the hippocampus by suppression of ER stress-associated TXNIP/NLRP3 inflamma- some activation in a manner dependent on AMPK. Toxicol Appl Pharmacol 2015; 286: 53- 63.

742 Am J Transl Res 2021;13(2):732-742