Inhibition of eIF2α Dephosphorylation Protects Hepatocytes from Apoptosis by Alleviating ER Stress in Acute Liver Injury

Objectives Protein kinase R-like ER kinase (PERK)/eukaryotic initiation factor 2 alpha (eIF2α) is an important factor along the main pathways for endoplasmic reticulum (ER) stress-mediated apoptosis. In this study, we investigated the effects of eIF2α phosphorylation on hepatocyte apoptosis and the ER stress mechanisms in acute liver injury. Methods eIF2α phosphorylation and apoptosis under ER stress were monitored and measured in male BALB/c mice with acute liver injury and human hepatocyte line LO2 cells. Results Carbon tetrachloride (CCl4) administration triggered ER stress and hepatocyte apoptosis, as well as eIF2α phosphorylation in mice. Inhibition of eIF2α dephosphorylation, as the pretreatment with 4-phenylbutyric acid (chemical chaperone, ER stress inhibitor), mitigated CCl4-induced intrahepatic ER stress, apoptosis, and liver injury. In an ER stress model of LO2 cells induced by thapsigargin (disrupting ER calcium balance), inhibition of eIF2α dephosphorylation reduced ER stress and apoptosis, while PERK knockdown reduced eIF2α phosphorylation and exacerbated ER stress and apoptosis. Conclusions eIF2α phosphorylation is one of the mechanisms employed by ER stress for restoring cellular homeostasis. Inhibition of eIF2α dephosphorylation mitigates hepatocyte apoptosis by alleviating ER stress in acute liver injuries.


Introduction
Liver injury can be initiated by a variety of causes, including infection with hepatitis viruses, alcohol, drugs, metabolic abnormalities, autoimmunity, ischemia, and hypoxia [1]. However, hepatocyte injury remains the most common pathophysiological basis of various liver diseases and the main cause of liver dysfunction [2]. Apoptosis, as it relates to a form of hepatocyte injury, can be triggered by intra-or extracellular signaling. Endoplasmic reticulum (ER) stress is one of the intracellular signaling pathways for mediation of apoptosis. ER stress is initiated when unfolded/misfolded proteins accumulate in the ER and bind to glucoseregulated protein 78 (GRP78) [3]. This particular binding event leads to phosphorylation of protein kinase R-like ER kinase (PERK) and inositol-requiring enzyme 1 alpha (IRE1α) and cleavage of the activating transcription factor 6 (ATF6) [4,5]. Phosphorylated PERK can thereby also phos-phorylate the alpha subunit of eukaryotic initiation factor 2 (eIF2α). Current theory suggests that phosphorylated eIF2α represses protein synthesis and reduces protein load in the ER [6]. On the other hand, the phosphorylated eIF2α selectively induces the response of activating transcription factor 4 (ATF4) [7,8], which regulates the expression of GRP78, growth arrest and DNA damage 34 (GADD34), and C/EBP homologous protein (CHOP). Research further suggests that GADD34 can interact with protein phosphatase 1 (PP1), thereby dephosphorylating eIF2α and effectively forming a negative feedback loop to restore protein synthesis [9]. ER stress results in proteolytic cleavage of ATF6, generating a 50 kD active fragment [10], whereby ATF6 activation leads to an increased transcription of a network of genes, including GRP78 and X-box binding protein 1 (XBP1). Koh et al. discovered that spliced XBP1 (XBP1s) is converted from a nonspliced isoform by IRE1α endonuclease, facilitating the expression of a number of unfolded protein response (UPR) responsive genes [11,12], similar to the types of UPRs found in ER stress environments.
While research suggests a multitude of naturally occurring ER stress regulators, studies continue to demonstrate the efficacy of ER stress regulation via chemical treatment. 4-Phenylbutyric acid (PBA, a chemical chaperone) alleviates ER stress in a variety of cell types [13,14]. Salubrinal, a treatment alternative method, selectively suppresses eIF2α dephosphorylation by inhibiting PP1 activity, sustaining the phosphorylated eIF2α status, while ISRIB inhibits the eIF2α phosphorylation [15][16][17]. In addition, DnaJC3 is an ER stress-regulated chaperone and can inhibit eIF2α kinases including PERK, protein kinase R (PKR), general control nonderepressible 2 (GCN2), and heme-regulated inhibitor (HRI) [18,19]. Taken together, PERK, ATF6, and IRE1α can impede protein synthesis, upregulate an ER response protein, activate ER-related degradation, and promote cell survival [20]. If ER homeostasis is disturbed, ER stress will trigger proapoptotic signaling, such as CHOP, c-Jun N-terminal kinase (JNK), and caspase-12 [21,22]. Caspase-3 responds to both intra-and extracellular signals and is subject to cleavage in an effort to initiate apoptosis [23,24]. The impact of ER stress on apoptosis is shown in Figure 1.
ER stress inevitably occurs in the pathogenesis of various liver diseases [25,26]. The PERK/eIF2α relationship provides a key component for the resulting ER stress-mediated apoptosis [27]. This study utilized a carbon tetrachloride-(CCl 4 , through conversion into reactive trichloromethyl to injure the liver) induced acute liver injury mouse model and a thapsigargin-(TG, through disruption of the ER calcium balance) induced ER stress model in cultured hepatocytes to determine the effect of inhibited eIF2α dephosphorylation on hepatocyte apoptosis and investigated in detail the molecular mechanism.

Materials and Methods
2.1. Animals and Induction of Liver Injury. Male BALB/c mice (18 ± 2 g) were supplied by the Animal Center of Zunyi Medical University (Guizhou, China) and housed in a specific pathogen-free facility where room temperatures varied between 20 and 24°C. Mice were acclimated for one week prior to the start of experimental procedures, where they were then monitored for health and behavior every 12 h. Before the experimental procedure was initiated, investigators and technicians were educated by ethics experts on experimental animal welfare and animal use ethics. All mouse studies were carried out in accordance with the guidelines of China Animal Care and Research. The animal study protocol was approved by the Animal Care and Use Committee of the Affiliated Hospital of Zunyi Medical University (Guizhou Province, China).
To induce acute liver injury, mice were randomly grouped and injected intraperitoneally with 10 mL/kg body weight of olive oil alone (control) or a mixture of CCl 4 (25%) and olive oil (75%) (acute liver injury model group, n = 10). In regulating ER stress chemically, mice were pretreated with salubrinal (1 mg/kg body weight, vehicle: dimethyl sulfoxide (DMSO); Sigma), ISRIB ( Figure 1: The impact of ER stress on apoptosis. PERK/eIF2α is an important factor in the main pathways for ER stress-mediated apoptosis. eIF2α integrates multiple signals and involves both prosurvival and proapoptotic pathways of ER stress.

Humane Endpoints.
At the conclusion of each experiment, mice were sacrificed unless they died prematurely. One to three mice were placed in a 4000 mL euthanasia box. The box chamber air was then replaced with 100% carbon dioxide at a turnover rate of 30% of the chamber volume per minute. Mice were anesthetized under these conditions for up to 3 minutes. After losing consciousness and presenting an inability to respond to stimuli, blood circulation was maintained while tissue and blood were harvested. The entire anesthetization procedure was performed within 10 minutes. . The bound antibodies were detected with horseradish peroxidase-(HRP-) conjugated anti-mouse or anti-rabbit IgG and visualized by enhanced chemiluminescent reaction. The relative levels of target protein compared to the control were determined via densitometry, using Quantity One software (Bio-Rad, Hercules, CA, USA). In addition, a single membrane can be probed multiple times after striping earlier bound antibodies. Briefly, the visualized membranes were immersed in stripping buffer for 10-60 minutes to remove the bound antibodies and then primed for incubation with a new set of antibodies against a different protein.
2.6. Histology and Immunohistochemistry. Liver tissues were fixed in 10% formalin and embedded in paraffin. Sections (5 μm thickness) were stained with hematoxylin and eosin (HE). In addition, the tissue sections were subjected to immunohistochemistry using monoclonal antibodies against cleaved caspase-3 (9664, Cell Signaling Technology) and CHOP (ab11419). The stained sections were viewed under a light microscope (Olympus CX31). The liver histology was independently scored by two experienced pathologists using the Histology Activity Index-(HAI-) Knodell score. Image-Pro Plus 6.0 was used for quantitative analysis of histology and immunohistochemistry [28].    BioMed Research International cytometry. The apoptotic index was calculated as the percentage of annexin V-positive cells divided by the total number of cells in the gated region as previously outlined [31].

Serum
2.10. Statistical Analysis. The normally distributed data were expressed as mean ± standard deviation (SD). Differences between groups were statistically analyzed using the oneway analysis of variance (ANOVA) with Tukey's post hoc analysis. The one-sample Kolmogorov-Smirnov test was used to analyze the normality of distribution for continuous variables. A P value of <0.05 was considered statistically significant [32].

Inhibition of eIF2α Dephosphorylation
Mitigates the CCl 4 -Induced Apoptosis and Liver Injury in Mice. As shown in Figure 3(a), salubrinal, ISRIB, or PBA pretreatment alone did not significantly change intrahepatic p-eIF2α, ATF4, and cleaved caspase-3 protein expression in mice without liver injury, suggesting that the phosphorylation of eIF2α is contingent upon stimulation as identified in liver injury.
Once injury was induced in our mouse model, the salubrinal pretreatment significantly increased the intrahepatic p-eIF2α and ATF4 protein expression (P < 0:05; Figure 3(      BioMed Research International Figure 3(c)). Similar patterns were obtained in liver sections by immunohistochemistry (P < 0:05; Figure 3(d)). Serum ALT levels were significantly reduced with the salubrinal or PBA pretreatment and significantly elevated by the pretreatment with ISRIB or DnaJC3 overexpression post-CCl 4 injection (P < 0:05; Figure 3(e)). Histologically, though the necroinflammation in mice pretreated with salubrinal or PBA developed, the extent and severity of hepatic necrosis were significantly reduced (P < 0:05; Figure 3(f)). In contrast, the hepatic necrosis was significantly increased   XBP1s CHOP Cleaved caspase-3 , the CCl 4 -induced ATF6, XBP1s, and CHOP expression in the liver was significantly reduced by the salubrinal or PBA pretreatment and significantly increased by ISRIB or DnaJC3 overexpression pretreatment (P < 0:05; Figure 4(c)). The immunohistochemistry staining resulted in similar patterns as seen with the intrahepatic CHOP expression, as detected by Western blot (P < 0:05; Figure 4(d)).

PERK Knockdown Reduces TG-Induced eIF2α
Phosphorylation and Aggravates ER Stress and Apoptosis in LO2 Cells. Compared to control-treated cells, the expression of PERK was significantly lowered following PERK shRNA pretreatment in TG-untreated LO2 cells (P < 0:05; Figure 6(a)). The decreased PERK expression significantly suppressed TG-induced p-PERK, p-eIF2α, and ATF4 protein expression and increased ATF6, XBP1s, CHOP, and cleaved caspase-3 protein levels. Furthermore, compared to the TG group, the cell viability was significantly reduced (P < 0:05; Figure 6(b)), and the apoptotic index was significantly increased in the PERK shRNA+TG group (P < 0:05; Figure 6(c)).

Discussion
In the present study, we investigated the impacts of eIF2α phosphorylation on hepatocyte apoptosis and evaluated the therapeutic implication in acute liver injury. We investigated hepatocyte apoptosis that presented an increased phosphorylation of eIF2α as a direct result of CCl 4

12
BioMed Research International mice and TG incubation in LO2 cells. Inhibition of eIF2α dephosphorylation mitigated ER stress and hepatocyte apoptosis in mice and LO2 cells, as well as alleviating liver injury in mice. These results confirm that eIF2α phosphorylation is one of the cytoprotective mechanisms for ER stress. To our knowledge, this is the first study suggesting that inhibition of eIF2α dephosphorylation mitigates hepatocyte apoptosis in acute liver injury. CCl 4 can be converted into reactive trichloromethyl by cytochrome P450 2E1, which induces hepatic lipid peroxidation, oxidative stress, ER stress, and inflammation, leading to hepatocyte degeneration, apoptosis, and other injuries [33,34]. ER stress is a defensive response to a host of stimulants, but it can evolve to encompass cell injury [35]. PERK/eIF2α is one of the main pathways for ER stress-mediated apoptosis [36]. However, recent studies have shown that increasing eIF2α phosphorylation protected injured cells [37]. In this study, ER stress, as well as eIF2α phosphorylation, and hepa-tocyte apoptosis were induced through CCl 4 administration in mice. Inhibition of eIF2α dephosphorylation, as a direct result of a decrease in ER stress by PBA, significantly mitigated hepatocyte apoptosis and liver injury. Conversely, the pretreatment with ISRIB or DnaJC3 overexpression inhibited CCl 4 -induced eIF2α phosphorylation and increased hepatocyte apoptosis and liver injuries. Our data further supports the previous claims that the elevation of eIF2α phosphorylation moderates hepatocyte apoptosis [38,39]. However, PBA pretreatment, which reduced eIF2α phosphorylation, also lessened the degree of CCl 4 -induced hepatocyte apoptosis. PBA, a chemical chaperone, can relieve ER stress through enhancing a prosurvival response and is shown to protect various cells [40,41]. Previous studies confirm that the effective rescuer of ER stress can mitigate liver injury [42,43]. In addition, CCl 4 administration induced eIF2α dephosphorylation as well as apoptosis in mice. Therefore, eIF2α phosphorylation does not directly regulate hepatocyte

13
BioMed Research International apoptosis in acute liver injury. Instead, the inhibition of eIF2α dephosphorylation may mitigate CCl 4 -induced hepatocyte apoptosis through alleviating the cellular ER stress.
PERK, ATF6, and IRE1α pathways mediate ER stress and have been shown to trigger both prosurvival and proapoptosis pathways [44,45]. Although eIF2α phosphorylation may mediate ER stress-related apoptosis, as suggested by some studies, more recent studies demonstrated that increased eIF2α phosphorylation protected injured cells. In this study, we found that the PBA pretreatment decreased the CCl 4 -induced eIF2α phosphorylation and the expression of ATF6, XBP1s, and CHOP and exerted a form of hepatic protection, suggesting that PBA rescued the ER stress and reduced the need for the eIF2α phosphorylation-mediated protection. Salubrinal pretreatment inhibited the eIF2α dephosphorylation and mitigated ER stress in vivo. Adversely, ISRIB or DnaJC3 overexpression pretreatment reduced the intrahepatic eIF2α phosphorylation but aggravated ER stress. By inhibiting its dephosphorylation, eIF2α has elevated phosphorylation levels and may therefore have enhanced abilities to regulate ER stress, ultimately restoring homeostasis. Furthermore, salubrinal pretreatment inhibited TG-induced eIF2α dephosphorylation and reduced ER stress and apoptosis, while a low level of PERK phosphorylation reduced TG-induced eIF2α phosphorylation and aggravated ER stress and apoptosis in LO2 cells. These results support that eIF2α phosphorylation coupled with ATF6 and IRE1/XBP1 signals to regulate ER stress, and these three pathways together determine the initiation of ER stress-related apoptosis. Inhibition of eIF2α dephosphorylation mitigates apoptosis through decreasing ER stress.

Conclusion
Inhibition of eIF2α dephosphorylation mitigated intrahepatic ER stress and hepatocyte apoptosis in acute liver injury. Our findings provide new insights into the impacts of hepatic eIF2α phosphorylation on liver injury.

Data Availability
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.