Reduction of Autophagosome Overload Attenuates Neuronal Cell Death After Traumatic Brain Injury

Xingyun Quan, ay Li Song, ay Xiaomei Zheng, b Shenjie Liu, a Huaqiang Ding, c Sijing Li, a Guanghui Xu, a Xin Li a and Liang Liu a,d,e,f*
a Department of Neurosurgery, The Affiliated Hospital of Southwest Medical University, China
b Department of Neurology, The Affiliated Hospital of Southwest Medical University, China
c Department of Neurosurgery, The People ’s Hospital of Chongqing Yubei, China
d Sichuan Clinical Research Center for Neurosurgery, China
e Neurological Diseases and Brain Functions Laboratory, Clinical Medical Research Center of Southwest Medical University, China
f Academician (Expert) Workstation of Sichuan Province, China

Abstract—
Previous studies have shown that alterations in autophagy-related proteins exist extensively after trau- matic brain injury (TBI). However, whether autophagy is enhanced or suppressed by TBI remains controversial. In our study, a controlled cortical impact was used to establish a model of moderate TBI in rats. We found that a sig- nificant increase in protein levels of LC3-II and SQSTM1 in the injured cortex group. However, there were no sig- nificant differences in protein levels of VPS34, Beclin-1, and phosphor-ULK1, which are the promoters of autophagy. Lysosome dysfunction after TBI might lead to autophagosome accumulation. In addition, the highly specific autophagy inhibitor SAR405 administration reduced TBI-induced apoptosis-related protein cleaved caspase-3 and cleaved caspase-9 levels in the ipsilateral cortex, as well as brain edema and neurological defects accessed by mNSS. Furthermore, chloroquine treatment reversed the beneficial effects of SAR405 by increasing the accumulation of autophagosomes. Finally, our data showed that autophagy inhibition by VPS34 gene knock- out method attenuated cell death after TBI. Our findings indicate that impaired autophagosome degradation is involved in the pathological reaction after TBI, and the inhibition of autophagy contributes to attenuate neuronal cell death and functional defects. © 2021 IBRO. Published by Elsevier Ltd. All rights reserved.

Key words: autophagy, autophagosome, VPS34, SAR405, neuronal cell death, traumatic brain injury.

INTRODUCTION
Traumatic brain injury (TBI) is a major medical problem worldwide. It leads to disability and death in adults, causing a public burden on modern society (Zeng et al., 2020a, 2020b). Reducing secondary brain damage is the key to saving patient lives and improving neurological function. Many pathological processes, including apopto- sis, inflammation, and oxidative stress, lead to aggra- vated secondary brain damage (Zhang and Wang, 2018). Although secondary brain damage may be reduced by various treatments, the curative effect and

*Correspondence to: L. Liu, Department of Neurosurgery, The Affiliated Hospital of Southwest Medical, University, NO.25 of Taiping Street, Luzhou, Sichuan 646000, China.
E-mail address: [email protected] (L. Liu).
y These authors contributed equally to this work.
Abbreviations: 3-MA, 3-methyladenine; CCI, controlled cortical impact; CQ, chloroquine; GFP, green fluorescent protein; PIK3C3, phosphatidylinositol 3-kinase catalytic subunit type 3; TBI, traumatic brain injury; TEM, transmission electron microscopy.
prognosis of TBI patients are poor (Anthony Jalin et al., 2019). Therefore, new treatment strategies are urgently needed to reduce secondary brain injury.Macroautophagy (hereafter referred to as autophagy) is an evolutionarily conservative process in which cells mediate the retention and turnover of organelles and cytoplasm through lysosome-dependent pathways. In the process of autophagy, autophagosomes separate components for autophagy by forming a closed bilayer membrane structure and transporting it to lysosomes for degradation (Grishchuk et al., 2011; Cui et al., 2017a, 2017b). Some studies have shown that Autophagy is associated with apoptosis, neuroinflammatory response, and neurological deficiency in various neurological condi- tions, including TBI (Grishchuk et al., 2011; Xing et al., 2012; Zhang et al., 2017a, 2017b). Liang et al. found that the autophagy marker protein, Beclin-1, interacts with Bcl- 2 to regulate apoptosis (Liang et al., 1998). Luo et al. fur- ther found that reducing the Beclin-1/Bcl-2 ratio might reduce TBI-induced neuronal cell apoptosis (Luo et al., 2011). A growing body of research has proved that some

https://doi.org/10.1016/j.neuroscience.2021.02.007

0306-4522/© 2021 IBRO. Published by Elsevier Ltd. All rights reserved.
drugs may reduce nerve cell damage by regulating the autophagy pathway (Smith et al., 2011; Zhang et al., 2017a, 2017b). However, there remains a lack of direct evidence on the process and role of autophagy after TBI.

Previous researches have shown that autophagy is activated after TBI (Sadasivan et al., 2008; Luo et al., 2011). TBI was observed to induce an increase in Beclin-1 and LC3-II (the phosphatidyl ethanolamine- conjugated form) levels. Conversely, Sarkar et al. found that autophagy might be inhibited after TBI (Sarkar et al., 2014). They found that LC3-II levels were increased during the early stage after TBI due to autophagosome accumulation, but Beclin-1 levels were not altered. More evidence is needed to detect the state of autophagy after TBI before exploring its effect on secondary brain injury.
The effect of autophagy on the apoptotic signalling pathway in TBI remains unclear probably because existing studies employ non-selective drugs to regulate autophagy. For example, the widely used autophagy inhibitor, 3-methyladenine (3-MA), is thought to reduce neuronal apoptosis in TBI mice by inhibiting autophagy activation (Luo et al., 2011). On the contrary, Jin et al. showed that 3-MA might weaken the protective effect of mild hypothermia in brain injury treatment (Jin et al., 2016). Additionally, 3-MA inhibits both VPS34/PIK3C3 and PIK3C1. Wu et al. reported that continuous treatment with 3-MA might significantly increase autophagy by inhibiting PIK3C1 (Wu et al., 2010). Therefore, the use of selective inhibitors that target VPS34 might provide more convincing evidence on the role of autophagy in TBI. SAR405 is a highly selective molecular mass kinase inhibitor that has been demonstrated to inhibit VPS34 without off-target activity on PIK3C1 kinases (Pasquier, 2015). Besides, the application of autophagy-related gene knockdown methods in TBI animal models has not been explored.

In the present study, we investigated the role and state of autophagy flux in rats after TBI and the precise regulation of autophagy using highly specific inhibitors and autophagy gene knockdown methods that target VPS34.

EXPERIMENTAL PROCEDURES
Animals preparation
All the rats were purchased from the Chengdu Dashuo Experimental Animal Co. Ltd. All surgical procedures and animal experiments were approved by the Institutional Animals Ethics Committees (IAEC) at Southwest Medical University, Luzhou, China (permit number: SYXK (Chuan) 2018-065), and complied with the Guidelines of the National Institutes of Health on the Care and Use of Laboratory Animals. A total of 228 adult (11–16 weeks) male Sprague-Dawley (SD) rats weighing 320 ± 10 g were used. They were maintained on a 12-h light/dark cycle at a constant temperature (22
±2 °C) and humidity (50 ± 10%) with food and water available ad libitum for a minimum of 3 d before surgery.

Controlled cortical impact (CCI)The animals were surgically prepared by anesthesia with sodium pentobarbital (50 mg/kg body weight) for Controlled Cortical Impact (Custom Design and Fabrication, VA, USA) induced injury or sham surgery. The CCI device was driven by compressed air and consisted of a pneumatic impactor with a 3.5 mm diameter tip. An incision was made along the midline of the scalp. A 4-mm circular craniotomy was performed on the left parietal bone. The skullcap was carefully removed without disrupting the underlying dura. A moderate injury was induced by the impact (the velocity of 4.5 m/s, the deformation depth of 1.6 mm, and the dwell time of 120 ms) as described previously (Zhao et al., 2017). Sham animals underwent the same proce- dures, including anesthesia and surgery, except for the impact.

Intracerebroventricular administration
Intracerebroventricular (i.c.v) drug administration was performed as previously described (Gao et al., 2018). The rats were briefly placed in a stereotaxic apparatus. The needle of a 10-ml Hamilton syringe (Gaoge Industry, Shanghai, China) was inserted into the right lateral ventri- cle through a burr hole following coordinates relative to bregma: 0.8 mm posterior and 1.5 mm lateral to the bregma, and 3.7 mm below the horizontal plane of the bregma. After injection, the needle was kept in place for an additional 10 min and retracted slowly. SAR405 (MCE, HY12481) was dissolved in a sterile saline solution containing 1% dimethyl sulfoxide (DMSO) with a total of 10 ll. The same volume of 1% DMSO in sterile saline was prepared as a negative control for SAR405. SAR405 (10 lL) or 1% DMSO (10 lL) was injected via i.
c.v (with the rate 1 ll/min) 20 min before TBI. An addi- tional 10 min was required before retracting the needle slowly. Then, bone was sealed on the burr hole immedi- ately, and the rats were put in a warm blanket waiting for sham or TBI operation. Adenovirus expressing the GFP-LC3B fusion protein (Ad-GFP-LC3B) (Beyotime Biotechnology, C3006) was transfected to rats via i.c.v as previously described (Zeng et al., 2018). Rats were anesthetized, and stereotaxic surgery was performed to deliver 4 lL of adenovirus. TBI or sham operation was performed 7 d after adenovirus transfection.

Chloroquine (CQ) injection
Chloroquine (MCE, HY-17589A) was dissolved in 0.9% sterile saline and injected via i.p immediately after TBI at a dose of 10 mg/kg as previously described (Ma et al., 2012; Cui et al., 2017a, 2017b). Rats in sham, naı¨ ve, and vehicle groups were injected by the same vol- ume of 0.9% sterile saline.

Lentiviral vector preparation
Three siRNA sequences targeting rat VPS34 and a negative control sequence were designed and synthesized by Hanbio Biotechnology (Shanghai,
China). The best performing VPS34 siRNA sequence was 50-GGGAAGAGAGAACAAAAGA-30. The negative control (NC) siRNA (scrambled siRNA) was 50-TTCTCCGAACG TGTCACGTAA-30. The preparation, packaging, and quality inspection of the shRNA lentivirus vector was completed by Hanbio Biotechnology (Shanghai, China). After the high purity endotoxin-free extraction of the three plasmid vectors, the vector plasmid of shRNA was co-transfected into 293 T cells with liposome transfection reagent Lipofiter (Hanbio Biotechnology) transfection reagent. The virus supernatant was collected at 48 h and 72 h after transfection, respectively. A total volume of 10 mL virus was added to Hela cells with a 96-well plate for verification, and after 24 h of culture, a microscopic examination was performed. The titer of lentivirus was detected by the dilution counting method for 1 109 transfection units (TU) per mL. Transfection was performed according to the manufacturer’s instructions. A total volume of 6 mL lentivirus solution was stereotactically administered into the lateral ventricle 7 d before sham or TBI operation.

Western blot
Cell lysates preparation and Western blot analysis were performed as previously described (Sarkar et al., 2020). To obtain whole tissue extract, the cortical tissue was subfractionated in homogenization buffer as described previously (Wang et al., 2015). Briefly, the lysate concen- tration was detected by using a BCA Protein Assay Kit (Beyotime Biotechnology, China). Equal amounts of pro- teins (50 lg) were resolved on an SDS-PAGE) gel and transferred onto a polyvinylidene difluoride (PVDF) mem- brane. The membrane was blocked with 5% non-fat dry milk in PBST (PBS + 0.05% Tween 20) for 2 h at room temperature and then incubated overnight at 4 °C, sepa- rately with the following primary antibodies:LC3 (1:600; Proteintech, 14600-1-AP), SQSTM1/P62 (1:1000; Proteintech, 66184-1-Ig), Beclin-1/BECN1 (1:1000; Proteintech, 11306-1-AP), VPS34/PIK3C3 (1:1000; Bey-
otime Biotechnology, AF1549), Phospho-ULK1 (Ser555) (D1H4) (1:1000; Cell Signaling Technology, 5869), ATG5 (1:1000; Sigma, A0731), anti-cathepsin D (CTSD) (1:2000, Abcam, ab75852), b-Actin (1:10000; Protein- tech, 60008-1-Ig), GAPDH (1:10000; Proteintech, 10494-1-AP), cleaved caspase 3/CASP3 (1:1000; Bey- otime Biotechnology, AF1150), caspase-3 (1:1000; Bey- otime Biotechnology, AC030), caspase-9 (1:1000; Beyotime Biotechnology, AF1264), Cytochrome C (1:200; Beyotime Biotechnology, AC909), BAX (1:5000; Proteintech, 50599-2-Ig), and Bcl2 (1:1000; Proteintech, 12789-1-AP). Afterward, the blot bands were washed three times with PBST for 15 min. The antibodies were then incubated with horseradish peroxidase conjugated anti-rabbit IgG and anti mouse IgG (1:2000; Abcam, ab205718 or 1:1000; Beyotime Biotechnology, A0192) for 2 h at room temperature. Then, the PVDF membranes were washed three times with PBST for 15 min and visu- alized with an ECL system. Using FIJI software to quantify the density of each band.
Subcellular fractionation Fractionation of subcellular was performed according to the methods introduced in the previous study (Sabirzhanov et al., 2014). The cell suspension was cen- trifuged at 500×g for 15 min at 4 °C and digitonin lysis buffer (pH 7.4, 250 mM sucrose, 1 mM EDTA, 200 lg/ml
×digitonin, 1 mM DTT, 20 mM HEPES, 80 mM KCl, 1 mM EGTA, and phosphatase inhibitor cocktails and protease inhibitor) was added for 10 min to re-suspend the cell pel- let. The lysate was centrifuged at 12,000 g for 15 min at 4 °C. The supernatant was recovered to represent the cytosolic fraction.

Immunofluorescence and TUNEL staining
Immunofluorescence staining was performed as described previously (Geronimo-Olvera et al., 2017). Briefly, rats were euthanized and perfused with 4% paraformaldehyde in 0.1 mM phosphate-buffered saline (PBS, pH 7.4). Brains were removed carefully and fixed with 4% paraformaldehyde (PFA) 24 h before being fro- zen. Sucrose (30%) was used to immerse brain samples at 4 °C until the brain sunk to the bottom. A cryostat (CM3050S; Leica Microsystems, Germany) was used to section 8 mm-thick coronal brain slices. Sections were permeabilized with 0.1% Triton X-100 in PBS for 10 min, then, blocked with 10% goat serum for 2 h, and incubated overnight at 4 °C with following primary antibodies: LC3B (1:400; Cell Signaling Technology, 83506), RBFOX3/ NeuN (1:100; Proteintech, 26975-1-AP), cleaved caspase 3/CASP3 (1:200; Beyotime Biotechnology, AF1150), SQSTM1/P62 (1:100; Proteintech, 66184-1-Ig), RBFOX3/NeuN (1:200; Proteintech, 26975-1-AP), GFAP (1:200; Proteintech, 16825-1-AP), and AIFM1/AIF (1:100; Proteintech, 17984-1-AP). Afterward, slices were washed with PBS and incubated with secondary antibod- ies as follows: CoraLite488 – conjugated Affinipure Goat Anti-Mouse IgG (1:500; Proteintech, SA00013-1) and CoraLite594 – conjugated Goat Anti-Rabbit IgG(H + L) (1:500; Proteintech, SA00013-4). Finally, the nuclei were stained with DAPI (Beyotime Biotechnology, C1002). A terminal deoxynucleotidyl transferase dUTP nick end- labeling (TUNEL) detection kit (Beyotime Biotechnology, C1086) was performed for detecting apoptotic cells according to the manufacturer’s instruction. Exposure times were kept constant for all sections in each experi- ment. The numbers of positive cells were identified and counted in the ipsilateral cortex from three random coro- nal slices per brain. The numbers of double-labeled cells in a defined region were analyzed by investigators who were blinded to groups.

Transmission electron microscopy (TEM)
TEM analysis was performed as described previously (Wang et al., 2018). Briefly, brains were fixed with 3% glu- taraldehyde and postfixed with 1% OsO4 in 0.1 M cacody- late buffer for 2 h. The samples were then stained with 1% Millipore-filtered uranyl acetate, dehydrated in increasing concentrations of ethanol, and infiltrated and embedded in epoxy resin (Epon812; NEOCERA, USA). Finally, elec- tron photomicrographs were taken of ultrastructures of neuronal cells with a transmission electron microscope (JEM-1400PLUS; JEOL, Tokyo, Japan).

Brain water content
Brain edema was evaluated by brain water content analysis, as previously described (Peng et al., 2019). The whole rat brains were harvested at 24 h after TBI and separated. The brain specimens were immediately weighed to obtain the wet weight (WW) and dried at 100 °C for 72 h before determining the dry weight (DW). The percentage of brain water content = [(WW — DW)/ WW] × 100%.

Evaluation of the neurological function
Neurological function evaluation was performed as previously described (Cui et al., 2017a, 2017b). Briefly, blinded and trained observers evaluated neurological function with the modified Neurological Severity Score (mNSS) test. The composite neuroscore indicated to quantify the for alterations of motor/sensory functions, reflexes and behaviours. Each function is graded on a scale of 0 to 18 (normal score, 0; maximal deficit score, 18). High scores indicate severe behavioural disfunction.
(1) the rats were raised with their tails and walked on the floor (6 scores), (2) sensation test (2 scores), (3) beam balance test (6 scores), (4) loss of reflex and abnormal movement (4 scores), those who were unable to perform tasks or lack tested reflexes (1 scores).

Statistical analysis
Statistical analysis was performed with Graph Pad Prism (Graph Pad Software, San Diego, CA). All data were expressed as the mean and standard deviation (mean ± SD). One-way analysis of variance (ANOVA) was used, followed by multiple comparisons between groups using Tukey’s HSD post hoc test (for more than two groups) and Holm–Bonferroni correction. A P value of
<0.05 was regarded as statistically significant.

RESULTS
Autophagosomes accumulated in the ipsilateral cerebral cortical neurons after TBI
To examine the state of autophagy after injury, we determined the time course of autophagy marker protein levels in the ipsilateral cortex. VPS34/PIK3C3 (phosphatidylinositol 3-kinase, catalytic subunit type 3)- Beclin-1 complex and ULK1 (unc-51 like autophagy activating kinase 1) complex are regarded as the markers of autophagy initiation, ATG12 (autophagy- related 12)–ATG5 (autophagy-related 5) conjugation initiates autophagy independently of Beclin-1. Western blot analysis showed that there were no significant differences in VPS34, Beclin-1, ATG12–ATG5 conjugate or phosphor-ULK1 levels in the injured cortex group, compared with in the sham group, at all the time points (Fig. 1A–D). LC3 (micro-tubule-associated protein 1 lightchain 3) is cleaved to membrane-bound LC3-II during autophagosome formation. Thus, the LC3-II protein level indirectly reflects the number of autophagosomes. In order to examine the number of autophagosomes after TBI, we determined levels of the autophagy marker protein MAP1LC3B. There was a significant increase in the LC3-II protein level, which peaked on Day 1 and decreased from Days 3 to 7 (Fig. 1A ,B). The change in SQSTM1/p62 (Sequestosome 1) protein levels is widely used to measure the state of autophagy flux. We also assessed SQSTM1levels in the injured cortex by western blotting. Western blotting showed a time- dependent increase in the SQSTM1 level 4 h after injury and a decrease in the level by Day 3 (Fig. 1A, C).
The GFP-tagged LC3 aggregated on the autophagosome membrane, appearing as spots on the fluorescence microscopy images (Li et al., 2019). Autophagosomes can be recognized as double- membrane vacuoles at the ultrastructural level. We per- formed image analysis of cortical slices obtained directly from transgenic rats pre-treated with adenovirus- expressing green fluorescent protein (GFP)-LC3 fusion protein. A significant increase in the number of GFP- LC3B-positive cells was observed in the injured cortex rats, compared with in the sham controls, 4 h and 1 d after injury (Fig. 2A–C). Furthermore, we performed double immunofluorescence staining with antibodies against LC3, RBFOX3/NeuN (RNA binding protein, fox-1 homolog [C. elegans] 3), and GFAP (glial fibrillary acidic protein) antigens to investigate the cell-type specificity for autophagosomes in the injury cortex. The results showed that LC3 was mainly expressed in RBFOX3-positive neu- ronal cells, but not GFAP-positive astrocytes, in the injured cortex (Fig. 2F–G). We also examined the autophago- some level in the ipsilateral cortex by transmission elec- tron microscopy (TEM). We observed an increased number of double-membrane vacuoles 1 d after controlled cortical impact (CCI). In contrast, double-membrane vac- uoles were not visible in the sham cortices (Fig. 2D, E). Our data demonstrated that autophagosomes were accu- mulated in ipsilateral cortical neuronal cells after TBI.

In the final stage of autophagy, hydrolases in lysosomes degrade autophagosomes and their cargo (Zeng et al., 2020a, 2020b). Therefore, the impairment of lysosome function may lead to the accumulation of autophagosomes. In order to explore the function of lyso- somes, we tested both precursor CTSD (cathepsin D) and mature CTSD in the cortices of injured and sham rats by western blot. We found slightly lower levels of precursor CTSD and mature CTSD in the ipsilateral injured cortex as compared to sham groups 1 d after TBI (Fig. 1E–G). Our data shows that lysosomal dysfunction might con- tribute to autophagosomes accumulation after TBI.

SAR405 attenuated TBI-induced apoptosis
Cleaved caspase-3, an active form of caspase-3, is the major executioner caspase. Increased cleaved caspase-
3 was observed in the injured cortex rats, compared with in the sham controls, by SAR405 western blot analysis 4 h to 3 d after injury (Fig. 3A, D). In the present study, significantly reduced LC3-II levels in rats treated with a

Time course of autophagy marker protein and cathepsin D expression after TBI. (A–F) Representative western blots band (A, D, E) of time course and densitometric quantification of endogenous LC3-II (B), Beclin-1, VPS34/PIK3C3, P-ULK1 (phosphor-ULK1), SQSTM1 (C), ATG12- 1TG5 (D), precursor CTSD (F) and mature CTSD (G) after TBI. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Sham group. Data were represented as mean ± SD, n = 6 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction.

single intracerebroventricular (i.c.v.) injection of SAR405 at dose 0.04 mg/kg and 0.4 mg/kg, compared with in rats treated with the vehicle (solution of 1% DMSO), were observed at 1 d after TBI (Fig. 3B, E). The dose of
0.04 mg/kg was chosen for the remaining experiments based on the protein gel blot results. To confirm further that SAR405 inhibited autophagosomes, we performed image analysis of GFP-LC3 fluorescence in the cortex. A significant decrease in GFP-LC3-positive cortical cells was observed in the SAR405-treated group, compared with in the vehicle-treated group at 1 d after TBI (Fig. 3G, H). Since LC3 was predominantly co-localized in neuronal cells, SAR405 could successfully inhibit autophagosome formation in neuronal cells.
Next, we performed a protein gel blot analysis to determine the effect of SAR405 on neuronal apoptosis. LC3-II, cleaved caspase-3, and cleaved caspase-9 protein levels were significantly decreased, but SQSTM1 protein levels were increased, in the SAR405- treated TBI group, compared with in the vehicle-treated TBI group (Fig. 4A–F). These findings indicated that SAR405 single treatment might attenuate caspase- dependent apoptosis by inhibiting VPS34.
We also tested if the effect of SAR405 was dependent of an increase of lysosomal activity. There was no significant difference in lysosomal enzyme CTSD ((Fig. 5A–C) in injured cortex 1 d after TBI, suggesting that the effect of SAR405 was independent of lysosomal activity.

SAR405 ameliorated brain oedema, and mNSS test was performed after TBI Since SAR405 attenuated apoptosis, we hypothesized that it might reduce TBI-induced cerebral oedema. As a result, the water content in the ipsilateral cerebral Autophagosomes accumulated in cortical neurons after TBI. (A–C) Representative images are shown of GFP signal (A) in the cortex of sham and TBI transgenic rats expressing GFP-LC3. Nuclei were stained with DAPI (blue). Scale bar = 50 mm. (D, E) Electron micrographs revealing increased autophagic vacuoles (green square) in the ipsilateral cortex at 1d after TBI. The blue arrow represents the myelin sheath. Scale bar = 1 mm. (F, G) Immunofluorescent staining against LC3, RBFOX3/NeuN (F) and GFAP (G) of the ipsilateral cortex at 1 d after TBI. Scale bar = 50 mm. Data were represented as mean ± SD, n = 4 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test (for more than two groups) and Holm–Bonferroni correction. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Sham group. *The figure (B) was created and exported with BioRender.com under a paid subscription. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)hemisphere was significantly reduced in SAR405-treated rats, compared with in 1% DMSO-treated rats, 24 h after TBI. No significant changes in oedema were observed in
the contralateral cerebral hemisphere of each group (Fig. 6A). Furthermore, to check the effects of SAR405 on neurological functions, we performed the mNSS test Time course of apoptotic protein as well as effects of SAR405 and chloroquine (CQ) on regulation of autophagosomes accumulation after TBI. (A, D) Western blots of apoptotic marker pro caspase-3 and cleaved caspase-3 in cortical tissue lysates from sham and TBI animals at the indicated time points. ***P < 0.001 vs. Sham group. (B, E) Treatment with SAR405 reduced LC3-II protein levels at different concentrations.
*P < 0.05 vs. Sham group, ##P < 0.01 vs. TBI + SAR405 at a dose of 0 mg/kg group. (C, F) CQ treatment increased LC3-II protein level both in naı¨ ve and TBI rats. *P < 0.05 vs. Sham group, #P < 0.05 vs. naı¨ ve + vehicle group, @P < 0.05 vs. TBI + vehicle group. Data were represented as mean ± SD, n = 6 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction. (G, H) Representative images of GFP-LC3 positive rats treated with 1% DMSO and SAR405. Scale bar = 50 mm. *P < 0.05, **P < 0.01 vs. Sham group, ###P < 0.001 vs TBI + DMSO group. Data were represented as mean ± SD, n = 4 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction. DMSO, dimethyl sulfoxide; Vehicle, 0.9% saline.

1, 3, and 7 d after TBI. We noticed an increase in neurological damage in all the injured cortex groups, compared with in the sham rats. The neurological scores were significantly decreased in the SAR405- treated groups, compared with in the sham rats (Fig. 6B).Chloroquine (CQ) reversed the neuroprotective effects of SAR405 by suppressing autophagosome degradation We selected CQ to increase the number of autophagosomes to observe whether it increased apoptosis. CQ is a late-stage inhibitor of autophagy, which leads to the accumulation of LC3-II-labeled autophagosomes by destroying the vesicle fusion between lysosomes and other vesicles. Administration of CQ injection at 10 mg/kg immediately after CCI significantly increased LC3-II levels in both naı¨ ve and TBI rats (Fig. 3C, F). LC3-II and SQSTM1 expression was significantly upregulated in the CQ-treated TBI group, compared with in the vehicle-treated TBI group. However, cleaved caspase-3 and cleaved caspase-9 levels were increased in the CQ+SAR405-treated TBI group, compared with in the SAR405-treated TBI group (Fig. 4A–F). Additionally, SAR405 and CQ did not affect the Beclin-1 protein levels. These findings suggested that CQ treatment could reverse the neuroprotective effects of SAR405.VPS34 knockdown reduced neuronal damage in the ipsilateral cortex after TBI
We also performed gene knockdown methods to confirm that early autophagosome inhibition might regulate cell death after TBI. An i.c.v. injection of the lentivirus (6 mL) with VPS34 siRNA was administered to rats 7 d before Effects of SAR405 and CQ on apoptosis and autophagy markers after TBI. (A–F) Representative protein gel blot band (A, D) and quantification of LC3-II (B), SQSTM1 (C), cleaved caspase-9 (E), and cleaved caspase-9 (F). **P < 0.01, ***P < 0.001 vs. Sham group; #P < 0.05, ###P < 0.001 vs. TBI group; @P < 0.05, @@P < 0.01, @@@P < 0.001 vs. TBI + CQ + SAR405 group. Data were represented as mean ± SD, n = 6 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction.Effects of SAR405 on cathepsin D expression after TBI. Representative western blots band (A) of SAR405 on cathepsin D expression in injured and control cortex after TBI. *P < 0.05, **P < 0.01 vs. Sham group. Data were represented as mean ± SD, n = 6 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction.CCI. VPS34 expression was decreased in both the VPS34-KD (VPS34 knockdown) naı¨ ve and TBI groups, compared with in the VPS34-NC (VPS34 negative control) rats, 1 d after TBI, confirming the knockdown efficacy of VPS34 siRNA in this model (Fig. 7A, B).

TheBcl-2/Bax ratio is considered the key for neuronal survival (Mao et al., 2013). TBI induces direct and indirect activation of Bax, resulting in Bax translocation to the mitochondria. Changes in the outer mitochondrial mem- brane permeability release the mitochondrial apoptotic Effects of SAR405 on brain water content and neurological functions after TBI. SAR405 treatment reduced brain water content (A) at 1 d after TBI. ***P < 0.001 vs. Sham group; ##P < 0.01 vs. DMSO group. Treatment with SAR405 diminished modified neurological severity scores (B) at 1 d, 3 d, and 7 d after TBI. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Sham group; #P < 0.05, ##P < 0.01, ###P < 0.001 vs. DMSO group. Data were represented as mean ± SD, n = 6 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction.proteins, cytochrome c and AIFM1, into the cytoplasm to
upstream autophagy regulators, such as VPS34-Beclin-1 complex, ULK1 complex and ATG12–ATG5 conjugate remained unchanged in the cortex at all examined time points after TBI; (2) a significant increase in LC3-II and SQSTM1 protein levels, which peaked 1 d after TBI, was observed, and GFP-LC3 signal was mainly located in RBFOX3-positive neuronal cells; (3) the increase of apoptotic protein was observed in the first 3 days after TBI; (4) i.c.v. SAR405 injection 20 min before TBI significantly decreased apoptosis and brain water content 1 d after TBI and improved
neurobehavioral defects 1, 3, and 7 d after TBI; (5) LC3-II protein levels and apoptosis were significantly increased 1 d after TBI in the intraperitoneal CQ- induce caspase-dependent and caspase-independent neuronal apoptosis, respectively. We observed significant decreases in the LC3-II and cleaved caspase-3 levels and increases in the SQSTM1 level and the Bcl-2/Bax ratio in the VPS34-KD TBI group, compared with in the negative control-treated group in the whole-tissue lysates from cortex (Fig. 7C–F, H). We also observed significant decreases in the cytochrome C level in the VPS34-KD TBI group, compared with in the negative control-treated group in the cytosolic fractions from cortex tissues (Fig. 7G, I). Consistent with the SAR405 treatment data, there was no difference in Beclin-1 expression in the VPS34-KD group, compared with in the VPS34-NC group (Fig. 7D).
To confirm our findings, we performed double immunofluorescence staining with antibodies against autophagy and apoptosis markers 24 h after TBI. The total number of LC3-, cleaved caspase-3-, and SQSTM1-positive cells were markedly increased in the injured cortex, compared with in the sham cortex, after TBI (Fig. 8A–D). LC3 and cleaved caspase-3 levels were significantly decreased, but SQSTM1 expression was increased, in the VPS34-KD groups, compared with in the negative control-treated group (Fig. 8A–C). Besides, VPS34-KD significantly decreased the number of TBI-induced AIFM1 (recombinant human apoptosis- inducing factor 1)-positive cells, indicating that autophagy inhibition might also reduce caspase- independent cell death (Fig. 8A, D). Similarly, TUNEL staining showed that the number of apoptotic cells was significantly decreased in the VPS34 siRNA-treated injured cortex, compared with in the negative control cortex (Fig. 8A, E).

DISCUSSION
In the present study, we established rat models of moderate brain injury and found that: (1) levels of
treated and CQ + SAR405-treated groups, compared with in the sham and SAR405-treated groups, respectively; (6) VPS34 siRNA significantly attenuated caspase-dependent and caspase-independent neuronal cell death 1 d after TBI. As far as we know, this is the first study to demonstrate the effect of autophagy after TBI using specific inhibitors or gene interference. The results of this study open up the prospect of TBI treatment via autophagy regulation. Previous studies have shown that autophagy initiation markers are increased in both experimental animals and clinical patients with TBI. However, this finding remains controversial (Lai et al., 2008; Au et al., 2017; Hui and Tanaka, 2019). Due to the lack of available evidence, we must first determine the state of autophagy after TBI. In the current study, the LC3-II protein level in the cortex was increased. We found that the number of fluo- rescent spots in the cerebral cortex was increased after TBI, and these results were also confirmed by TEM anal- ysis. The increased number of autophagosomes sug- gested autophagosome upregulation or accumulation in the injured cortex. However, in the present study, VPS34, Beclin-1, phosphor-ULK1, and ATG12-ATG5 conjugate remained unchanged during the first 7 d after TBI. The results showed no activation of autophagy after TBI.
Since autophagy was not activated, the increase in autophagosomes might be due to obstacles in the degradation process. Ubiquitinated cargo, including injured protein, is delivered to autophagosomes by SQSTM1, a receptor protein, and degraded by lysosomes. In the current study, SQSTM1 protein levels, which peaked on day 1, were significantly increased over 3 d after TBI, indicating that autophagy degradation was impaired in the early stage after TBI. Although the level of LC3-II protein still increased at 7 d after TBI, the accumulation of autophagy substrate SQSTM1 seem to resolve. This suggests that Inhibition of autophagy by VPS34 knockdown and apoptosis in the cortex after TBI. (A, B) VPS34-knockdown efficiency evaluation. VPS34 siRNA significantly decreased VPS34 protein expression both in normal and TBI rats. Data were represented as mean ± SD, n = 3 per group. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction. *P < 0.05 vs. VPS34NC Naı¨ ve group; ##P < 0.01 vs. VPS34KD TBI group. (C–F, H) Representative protein gel blot bands and quantification of LC3-II, cleaved caspase-3, Bcl-2/Bax, Beclin-1 and SQSTM1 in whole-tissue lysates from cortex tissues. (G, I) Representative protein gel blot bands and quantification of cytochrome c in cytosolic fractions from cortex tissues. Data were represented as mean ± SD (n = 6 per group). One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Sham group; #P < 0.05, ##P < 0.01 vs. VPS34-NC TBI group. VPS34-NC, VPS34 negative control; VPS34-KD, VPS34 knockdown groupautophagy flux may recover at the later stage after TBI. In addition, the activation of other autophagy pathways, such as chaperone-mediated autophagy, may also contribute to the eventual clearance of SQSTM1 (Sarkar et al., 2014). Lysosomal dysfunction has been previously reported in neurodegenerative diseases, such as Hunt-ington’s disease, Alzheimer’s disease, and Parkinson’s disease (Settembre et al., 2008; Nixon, 2013; Xu et al., 2016). We found that lysosomal enzyme CTSD decreased significantly 1 day after TBI and recovered later. This suggested that lysosomal dysfunction might contribute to impaired autophagy degradation after TBI.

Effects of VPS34 knockdown on neuronal cell death after TBI. (A–E) Representative images of immunofluorescent staining against LC3 (A– E), cleaved casp-3 (B), SQSTM1 (C), AIFM1 (D), and TUNEL (E). Data were represented as mean ± SD. One-way ANOVA was used, followed by Tukey’s HSD post hoc test and Holm–Bonferroni correction. ***P < 0.001 vs. Sham group; #P < 0.05, ##P < 0.01 vs. VPS34-NC TBI group. Scale bar = 50 mm, n = 4 per group. VPS34-NC, VPS34 negative control; VPS34-KD, VPS34 knockdown group; casp-3, caspase-3. Autophagy maintains cell survival by phagocytosis of damaged organelles and ubiquitin proteins. Excessive autophagy may lead to autophagic death. In the current study, we found that apoptosis-related protein, cleaved caspase-3, was significantly elevated over 3 d after TBI. This coincided with the time point at which autophagosomes were accumulated, indicating that cortical apoptosis occurred at the time point at which autophagy was most damaged. We also observed that LC3 was strongly co-localized with AIFM1 and cleaved caspase-3 in the cortex by immunofluorescence analysis. AIFM1 is transported from the inner mitochondrial membrane to the cytoplasm to participate in caspase-independent cell death (Norberg et al., 2010). Since LC3 is mainly located in neurons and LC3- positive cells showed the morphology of neurons, these results suggested that accumulated autophagosomes were related to nerve cell death in the cortex after TBI. The role of autophagy after TBI remains controversial. Previous research has shown that therapy with the non- specific autophagy inhibitor, 3-MA, can reduce nerve cell death by inhibiting autophagosome formation after TBI (Luo, et al., 2011). However, Jin et al. reported that 3- MA might aggravate brain damage after TBI (Jin et al., 2016). Another study showed that the autophagy agonist, rapamycin, reversed the protective effect of the neu- rotrophic factor, fibroblast growth factor-2 (Tang et al., 2017). Those available studies employ non-selective inhi- bitors to modulate autophagy after TBI (Smith, et al., 2011; Zhang and Wang, 2018). Since rapamycin and 3- MA show off-target effects, we must consider that they also regulate apoptosis through other pathways. Highly selective autophagy inhibitors or genetic intervention may help in clarifying the role of autophagy in TBI (Zhang and Wang, 2018). Our study provided first insight into the relationship between autophagy and neuronal apoptosis after TBI via selective inhibition of VPS34.
Our data showed that SAR405 significantly reduced LC3-II protein levels 1 d after TBI. Fluorescence analysis also showed that SAR405 treatment reduced the number of GFP-LC3-positive cells in the cortex.

SAR405 decreased the cleaved caspase-3 and cleaved caspase-9 protein levels in the cortex, demonstrating the neuroprotective effect of reducing autophagosome synthesis. Our data showed CQ treatment further increased the LC3-II and SQSTM1 protein levels 1 d after TBI. Since CQ further increased the accumulation of autophagosomes, these data indicated a partially impaired autophagy flux during the early stage after TBI. In addition, we observed a significant increase in the LC3-II protein levels in the CQ + SAR405-treated TBI group, compared with in the SAR405-treated TBI group. Our study suggested that SAR405 at our experimental concentration (0.04 mg/kg) did not inhibit autophagosome formation completely. SAR405 might reduce the burden of autophagy degradation by reducing part of autophagosome production. This relieved the pressure on lysosomes to degrade autophagosomes, resulting in complete removal of autophagosomes. Furthermore, Cui et al. found that CQ improved prognosis after TBI (Cui et al., 2015). We found that the neuroprotective effect of SAR405 could be reversed by CQ treatment. Cleaved caspase-3, and cleaved caspase-9 protein levels were increased in the CQ + SAR405-treated group, compared with in the SAR405-treated group. These results sug-gested that aggravation of autophagosome accumulation promoted neuronal apoptosis. In the present study, SAR405 treatment also showed protective effects on cerebral oedema and neurological function.
Emerging data support that knockdown of autophagy- associated genes, such as Beclin-1 or ATG7, may be a treatment strategy for focal cerebral infarction and haemorrhage (Chen et al., 2012; Xing et al., 2012).

Our results provided first insight into the use of precise gene knockdown methods to elucidate the role of autophagy in TBI. In the present study, VPS34 siRNA inhibited VPS34 expression and decreased LC3-II protein level in the rat cortex after TBI. Western blot analysis showed that VPS34 knockdown significantly increased the Bcl-2/Bax ratio in the whole-tissue lysates obtained from the rat cor- tex, decreased the cytochrome c protein level in the cytosolic fractions, and reduced the cleaved caspase-3 protein level. This suggested that autophagy inhibition might help in reducing caspase-dependent apoptosis. We also performed double immunofluorescence staining to examine whether autophagy regulated caspase- independent apoptosis. Co-localization of LC3 with AIFM1- and cleaved caspase-3-positive cells in the rat cortex was significantly increased after TBI. VPS34-KD significantly reduced AIFM1- and cleaved caspase-3- positive cells in the ipsilateral cortex. Fluorescence analy- sis also showed that VPS34 significantly decreased TUNEL-positive cells in the cortex. Since LC3 is mainly located in nerve cells, our findings indicated that VPS34 gene knockdown might prevent both caspase- dependent and caspase-independent neuronal cell death in the cortex after TBI.

However, the present study has a few limitations. The default of lysosomal and autophagosome fusion may also lead to autophagosome accumulation. Further research is still needed to investigate whether there was a default of fusion between autophagosome and lysosome after TBI. A study by Zeng et al. suggested that autophagy activation might be related to the severity of brain injury (Zeng et al., 2018). Our study was based on previous data on the induction of only moderate TBI. The state and role of autophagy after mild and severe brain injury must be studied further in the future. This study aimed to explore the role of autophagy during the early stage after TBI. Normal autophagy is necessary to keep cells alive. Autop- hagy degradation may resume during the later stage after TBI, and methods designed to promote autophagy may be used to alleviate secondary brain injury. In this study, we failed to find the protein that played a role in the link between autophagosomes and apoptosis. Although Beclin-1 was identified as a Bcl-2-binding protein, both SAR405 and VPS34 showed no effects on Beclin-1 pro- tein levels in this study. Moreover, the SAR405 treatment did not improve CTSD levels to attenuate lysosomal dys- function. Further studies are required to explain the inter- action between autophagosome overload and neuronal cell death.

In conclusion, these data indicated that impaired
autophagy degradation contributed to autophagosome overload, and was involved in the process of nerve cell death after TBI. We confirmed the autophagy degradation damage was not complete in the early stage after moderate TBI. Because the use of chloroquine to further inhibit autophagy degradation increased autophagosomes. We also confirmed that inhibition of autophagosome formation reduced nerve cell death, brain edema, and neurological defects induced by TBI in rats. The potential connection between autophagosome overload and cell death still needs to be confirmed in future studies. Taken together, our data provided novel insights into strategies in the treatment of traumatic brain injury.

AUTHOR CONTRIBUTIONS
Xingyun Quan, Li Song, and Xiaomei Zheng designed this research; Huaqiang Ding, Sijing Li, Guanghui Xu. and Xin Li. performed experiment; Xingyun Quan and Liang Liu. analyzed the data; Xingyun Quan and Li Song wrote the paper.

ACKNOWLEDGEMENTS
This work was supported by grants from the Project of Sichuan Provincial Health Department (110371); the Project of Sichuan Medical Association (S17074); the Key projects of Education Department of Sichuan (12ZA075); the Luzhou Science and Technology Bureau Project (2017-S-40).

CONFLICT OF INTEREST
The authors declare no conflict of interest.

REFERENCES

Anthony Jalin AMA, Jin R, Wang M, Li G (2019) EPPS treatment attenuates traumatic brain injury in mice by reducing Abeta burden and ameliorating neuronal autophagic flux. Exp Neurol 314:20–33.
Au AK, Aneja RK, Bayir H, Bell MJ, Janesko-Feldman K, Kochanek PM, Clark RSB (2017) Autophagy biomarkers beclin 1 and p62 are increased in cerebrospinal fluid after traumatic brain injury. Neurocrit Care 26:348–355.
Chen CW, Chen TY, Tsai KL, Lin CL, Yokoyama KK, Lee WS, Chiueh CC (2012) Inhibition of autophagy as a therapeutic strategy of iron-induced brain injury after hemorrhage. Autophagy 8:1510–1520.
Cui C, Cui J, Jin F, Cui Y, Li R, Jiang X, Tian Y (2017a) Induction of the vitamin D receptor attenuates autophagy dysfunction- mediated cell death following traumatic brain injury. Cell Physiol Biochem 42:1888–1896.
Cui CM, Gao JL, Cui Y, Sun LQ, Wang YC, Wang KJ, Li R (2015) Chloroquine exerts neuroprotection following traumatic brain injury via suppression of inflammation and neuronal autophagic death. Mol Med Rep 12:2323–2328.
Cui D, Sun D, Wang X, Yi L, Kulikowicz E, Reyes M, Zhu J (2017b) Impaired autophagosome clearance contributes to neuronal death in a piglet model of neonatal hypoxic-ischemic encephalopathy. Cell Death Dis 8 e2919.
Gao YY, Zhang ZH, Zhuang Z, Lu Y, Wu LY, Ye ZN, Zhang XS (2018) Recombinant milk fat globule-EGF factor-8 reduces apoptosis via integrin beta3/FAK/PI3K/AKT signaling pathway in rats after traumatic brain injury. Cell Death Dis 9:845.
Geronimo-Olvera C, Montiel T, Rincon-Heredia R, Castro-Obregon S, Massieu L (2017) Autophagy fails to prevent glucose deprivation/glucose reintroduction-induced neuronal death due to calpain-mediated lysosomal dysfunction in cortical neurons. Cell Death Dis 8 e2911.
Grishchuk Y, Ginet V, Truttmann AC, Clarke PG, Puyal J (2011) Beclin 1-independent autophagy contributes to apoptosis in cortical neurons. Autophagy 7:1115–1131.
Hui KK, Tanaka M (2019) Autophagy links MTOR and GABA signaling in the brain. Autophagy 15:1848–1849.
Jin Y, Lei J, Lin Y, Gao GY, Jiang JY (2016) Autophagy inhibitor 3-MA weakens neuroprotective effects of posttraumatic brain injury moderate hypothermia. World Neurosurg 88:433–446.
Lai Y, Hickey RW, Chen Y, Bayir H, Sullivan ML, Chu CT, Kochanek PM (2008) Autophagy is increased after traumatic brain injury in mice and is partially inhibited by the antioxidant gamma- glutamylcysteinyl ethyl ester. J Cereb Blood Flow Metab 28:540–550.
Li K, Chen H, Li D, Li H, Wang J, Jia L, Wu P (2019), SAR405, a
highly specific VPS34 inhibitor, disrupts auditory fear memory consolidation of mice via facilitation of inhibitory neurotransmission in basolateral amygdala. 85:214-225.
Liang X, Kleeman L, Jiang H, Gordon G, Goldman J, Berry G, Herman B (1998), Protection against fatal Sindbis virus encephalitis by beclin, a novel Bcl-2-interacting protein. 72:8586-8596.
Luo CL, Li BX, Li QQ, Chen XP, Sun YX, Bao HJ, Dai DK (2011)
Autophagy is involved in traumatic brain injury-induced cell death and contributes to functional outcome deficits in mice. Neuroscience 184:54–63.
Ma X, Liu H, Foyil S, Godar R, Weinheimer C, Hill J, Diwan AJC (2012), Impaired autophagosome clearance contributes to cardiomyocyte death in ischemia/reperfusion injury. 125:3170- 3181.
Mao W, Yi X, Qin J, Tian M, Jin G (2013) CXCL12 inhibits cortical neuron apoptosis by increasing the ratio of Bcl-2/Bax after traumatic brain injury. Int J Neurosci 124:281–290.
Nixon RA (2013) The role of autophagy in neurodegenerative disease. Nat Med 19:983–997.
Norberg E, Orrenius S, Zhivotovsky B (2010) Mitochondrial regulation of cell death: processing of apoptosis-inducing factor (AIF). Biochem Biophys Res Commun 396:95–100.
Pasquier B (2015) SAR405, a PIK3C3/Vps34 inhibitor that prevents autophagy and synergizes with MTOR inhibition in tumor cells. Autophagy 11:725–726.
Peng J, Pang J, Huang L, Enkhjargal B, Zhang T, Mo J, Wu P (2019) LRP1 activation attenuates white matter injury by modulating microglial polarization through Shc1/PI3K/Akt pathway after subarachnoid hemorrhage in rats. Redox Biol 21 101121.
Sabirzhanov B, Zhao Z, Stoica BA, Loane DJ, Wu J, Borroto C, Dorsey SG (2014) Downregulation of miR-23a and miR-27a following experimental traumatic brain injury induces neuronal cell death through activation of proapoptotic Bcl-2 proteins. J Neurosci 34:10055–10071.
Sadasivan S, Dunn Jr WA, Hayes RL, Wang KK (2008) Changes in autophagy proteins in a rat model of controlled cortical impact induced brain injury. Biochem Biophys Res Commun 373:478–481.
Sarkar C, Jones JW, Hegdekar N, Thayer JA, Kumar A, Faden AI, Kane MA (2020) PLA2G4A/cPLA2-mediated lysosomal
membrane damage leads to inhibition of autophagy and neurodegeneration after brain trauma. Autophagy 16:466–485.
Sarkar C, Zhao Z, Aungst S, Sabirzhanov B, Faden AI, Lipinski MM (2014) Impaired autophagy flux is associated with neuronal cell death after traumatic brain injury. Autophagy 10:2208–2222.
Settembre C, Fraldi A, Jahreiss L, Spampanato C, Venturi C, Medina D, de Pablo R (2008) A block of autophagy in lysosomal storage disorders. Hum Mol Genet 17:119–129.
Smith CM, Chen Y, Sullivan ML, Kochanek PM, Clark RS (2011) Autophagy in acute brain injury: feast, famine, or folly? Neurobiol Dis 43:52–59.
Tang C, Shan Y, Hu Y, Fang Z, Tong Y, Chen M, Wei X (2017) FGF2 attenuates neural cell death via suppressing autophagy after rat mild traumatic brain. Injury 2017 2923182.
Wang C, Xie N, Wang Y, Li Y, Ge X, Wang M (2015) Role of the mitochondrial calcium uniporter in rat hippocampal neuronal death after pilocarpine-induced status epilepticus. Neurochem Res 40:1739–1746.
Wang J, Qi Q, Zhou W, Feng Z, Huang B, Chen A, Zhang D (2018) Inhibition of glioma growth by flavokawain B is mediated through endoplasmic reticulum stress induced autophagy. Autophagy 14:2007–2022.
Wu Y, Tan H, Shui G, Bauvy C, Huang Q, Wenk M, Ong C (2010), Dual role of 3-methyladenine in SAR405 modulation of autophagy via different temporal patterns of inhibition on class I and III phosphoinositide 3-kinase. 285:10850-10861.
Xing S, Zhang Y, Li J, Zhang J, Li Y, Dang C, Li C (2012) Beclin 1 knockdown inhibits autophagic activation and prevents the secondary neurodegenerative damage in the ipsilateral thalamus following focal cerebral infarction. Autophagy 8:63–76.
Xu Y, Tian Y, Tian Y, Li X, Zhao P (2016) Autophagy activation involved in hypoxic-ischemic brain injury induces cognitive and memory impairment in neonatal rats. J Neurochem 139:795–805. Zeng XJ, Li P, Ning YL, Zhao Y, Peng Y, Yang N, Xu YW (2020a) A2A R inhibition in alleviating spatial recognition memory impairment after TBI is associated with improvement in
autophagic flux in RSC. J Cell Mol Med 24:7000–7014.
Zeng XJ, Li P, Ning YL, Zhao Y, Peng Y, Yang N, Zhao ZA (2018) Impaired autophagic flux is associated with the severity of trauma and the role of A2AR in brain cells after traumatic brain injury. Cell Death Dis 9:252.
Zeng Z, Zhang Y, Jiang W, He L, Qu H (2020b) Modulation of autophagy in traumatic brain injury. J Cell Physiol 235:1973–1985.
Zhang L, Wang H (2018) Autophagy in traumatic brain injury: a new target for therapeutic intervention. Front Mol Neurosci 11:190.
Zhang L, Wang H, Fan Y, Gao Y, Li X, Hu Z, Ding K (2017), Fucoxanthin provides neuroprotection in models of traumatic brain injury via the Nrf2-ARE and Nrf2-autophagy pathways. 7:46763.
Zhang P, Ye Y, Qian Y, Yin B, Zhao J, Zhu S, Zhang L (2017b) The effect of pyrroloquinoline quinone on apoptosis and autophagy in traumatic brain injury. CNS Neurol Disord Drug Targets 16:724–736.
Zhao WY, Chen SB, Wang JJ, Xu C, Zhao ML, Dong HJ, Liang HQ (2017) Establishment of an ideal time window model in hypothermic-targeted temperature management after traumatic brain injury in rats. Brain Res 1669:141–149.
(Received 8 December 2020, Accepted 7 February 2021)
(Available online 16 February 2021)