Genetic and pharmacological inhibition of XBP1 protects against APAP hepatotoxicity through the activation of autophagy

Acetaminophen (APAP) hepatotoxicity induces endoplasmic reticulum (ER) stress which triggers the unfolded protein response (UPR) in hepatocytes. However, the mechanisms underlying ER stress remain poorly understood, thus reducing the options for exploring new pharmacological therapies for patients with hyperacute liver injury. Eight-to-twelve-week-old C57BL/6J Xbp1-floxed (Xbp1f/f) and hepatocyte-specific knockout Xbp1 mice (Xbp1∆hepa) were challenged with either high dose APAP [500 mg/kg] and sacrificed at early (1–2 h) and late (24 h) stages of hepatotoxicity. Histopathological examination of livers, immunofluorescence and immunohistochemistry, Western blot, real time (RT)-qPCR studies and transmission electron microscopy (TEM) were performed. Pharmacological inhibition of XBP1 using pre-treatment with STF-083010 [STF, 75 mg/kg] and autophagy induction with Rapamycin [RAPA, 8 mg/kg] or blockade with Chloroquine [CQ, 60 mg/kg] was also undertaken in vivo. Cytoplasmic expression of XBP1 coincided with severity of human and murine hyperacute liver injury. Transcriptional and translational activation of the UPR and sustained activation of JNK1/2 were major events in APAP hepatotoxicity, both in a human hepatocytic cell line and in a preclinical model. Xbp1∆hepa livers showed decreased UPR and JNK1/2 activation but enhanced autophagy in response to high dose APAP. Additionally, blockade of XBP1 splicing by STF, mitigated APAP-induced liver injury and without non-specific off-target effects (e.g., CYP2E1 activity). Furthermore, enhanced autophagy might be responsible for modulating CYP2E1 activity in Xbp1∆hepa animals. Genetic and pharmacological inhibition of Xbp1 specifically in hepatocytes ameliorated APAP-induced liver injury by enhancing autophagy and decreasing CYP2E1 expression. These findings provide the basis for the therapeutic restoration of ER stress and/or induction of autophagy in patients with hyperacute liver injury.


INTRODUCTION
Because of its extensive worldwide use and narrow therapeutic window, acetaminophen (APAP) hepatotoxicity is considered a significant public health concern. Despite half a century of research, the specific mechanisms of APAP hepatotoxicity are still not fully understood, and clinical treatment is limited.
The endoplasmic reticulum (ER) is a major intracellular organelle that performs multiple physiological functions including protein folding, post-translational modifications, biosynthesis of fatty acids and sterols, detoxification of xenobiotics, and storage of intracellular Ca 2+ [1]. Upon exposure to potential stressors such as drugs, the ER initiates the unfolded protein response (UPR) to restore homeostasis [2].
In mammalian cells, the UPR consists of three primary pathways that sense the accumulation of unfolded or misfolded proteins. These sensor protein-transcription factor pairs are: (i) Protein kinase RNA-like endoplasmic reticulum kinase (PERK) and eukaryotic Initiation Factor 2 alpha (eIF2α); (ii) the activating transcription factor 6 (ATF6); and (iii) Inositol requiring kinase 1α (IRE1α) and X-box binding protein 1 (XBP1) [3]. Upon activation of IRE1α, splicing of the Xbp1 mRNA occurs, thus enabling the translation of a spliced XBP1 protein (sXBP1). sXBP1 is a transcription factor that induces genes involved in chaperoning proteins through the ER and degrading proteins that cannot be properly folded in the ER [4]. The role of ER stress in APAP-induced hepatocellular injury is supported by a considerable amount of data. APAP hepatotoxicity results in increased expression of phosphorylated eIF2a and CCAAT-enhancer-binding protein homologous protein (CHOP) [5], whilst sublethal doses of APAP activate the transcription factor ATF6 [6]. More importantly, the IRE1α-XBP1 arm of the UPR response has been shown to play a critical role in APAP-induced liver injury via the regulation of cytochrome P450 activity [1].
In the present study we sought to investigate the molecular mechanisms associated with high dose, APAP-induced hepatic ER stress in vivo and in vitro. Our study reveals a novel mechanism of protection against APAP hepatotoxicity by which decreased sXBP-1 induces autophagy.

MATERIALS AND METHODS Human samples
Livers from patients undergoing urgent liver transplantation for ALF resulting from APAP overdose were evaluated for XBP1 expression in liver paraffin sections. The diagnosis of ALF was established in Newcastle Hospitals NHS Foundation Trust (Newcastle, UK). The samples were obtained by the Newcastle Biomedicine Biobank (12/NE/0395), and approved by Newcastle Research Ethics Committee North East Newcastle and North Tyneside. Patients´clinicopathological characteristics are shown in Table 1. All patients gave informed consent for all clinical investigations, according to the principles embodied in the Declaration of Helsinki. Samples are explants from patients undergoing urgent liver transplantation for ALF, with a timing between paracetamol overdose and explant of within 12 h.

Experimental models
All animals were randomly allocated to the experimental groups. Eight-totwelve-week-old C57/BL6J mice purchased from ENVIGO (Valencia, Spain) were fasted overnight and challenged with an intraperitoneal (i.p.) injection of APAP [500 mg/kg]. Mice harboring a conditional floxed allele of Xbp1 (Xbp1 f/f ), a gift from Richard S. Blumberg (Harvard Medical School) were crossed with Alb-Cre mice (Charles River España, Cerdanyola del Vallés, Barcelona) to obtain a liver specific knockout of Xbp1 (Xbp1 Δhepa ). Overnight-fasted 8-12 week-old male Xbp1 f/f and Xbp1 Δhepa mice were challenged with an i.p. injection of APAP [500 mg/kg] or vehicle. Since it is a high dose and APAP is insoluble in water, APAP was first dissolved in DMSO then diluted to 50 mg/ml with PBS (final concentration of DMSO < 1%

Other methodology
For immunoblot analysis membranes were incubated overnight with the primary antibodies shown in Supplementary Table 1. GAPDH was used as a loading control. For RT-PCRs, human and mouse primers were purchased from Merck (sequences are shown in Supplementary Tables 2 and 3, respectively). Enzymatic colorimetric test was performed using the Triglycerides kit (RAL, Barcelona, Spain) for quantitative determination of triglycerides in the liver.

Statistical analysis
Statistical analysis was performed using the software Prism v8.0 (GraphPad Software). HepaRG cells and C57BL/6J mice were compared using one-way analysis of variance (ANOVA) and statistical significance was assessed by two-way ANOVA adjusted for Tukey's multiple comparisons. Data are expressed as means ± standard error of the mean (SEM). A p value equal or less than 0.05 was considered to be statistically significant.

Increased XBP1 expression in human and murine APAP hepatoxicity
To investigate whether XBP1 could play a role in the UPR associated with APAP overdose, XBP1 expression was investigated in liver tissue sections obtained from patients undergoing emergency liver transplantation for APAP-induced ALF. The clinicopathological characteristics of these patients are summarized in Table 1.
Given that the antibody detects both the unspliced and the spliced forms of XBP1, most hepatocyte nuclei expressed XBP1. However, cytoplasmic expression of XBP1 became more evident as hepatocyte architecture was altered (Fig. 1A). These samples showed increased levels of serum markers of liver disease Fig. 1 The expression of XBP1 increases during APAP overdose both in mouse and human. A Representative XBP1 staining performed on paraffin liver sections of patients with acute liver failure (ALF) due to APAP overdose. Scale bars, 100 μm (top) and 50 μm (bottom) (N = 13). B mRNA levels of sXBP1/uXBP1 in HepaRG cells, were determined by RT-qPCR, 24 h after APAP challenge. C mRNA level of sXBP1/uXBP1 in C57BL/6J wild-type (WT) mice was determined by RT-qPCR, 24 h after APAP injection. D Representative XBP1 staining was performed on paraffin liver sections of C57BL/6 wild-type (WT) mice, 24 h post-APAP [500 mg/kg]. Scale bars, 100 μm and 50 μm. E Quantification of XBP1positive area was performed. Data are expressed as mean ± SEM and graphed, separately (*p < 0.05-****p < 0.0001; N = 3-5 animals per experimental group).
( Table 1). We first investigated the activation of the UPR response after exposure to APAP in a mycoplasma-free human hepatic cell line, HepaRG, and in a preclinical in vivo model. Interestingly, we found that splicing of Xbp1 occurred under high concentrations of APAP in vitro (Fig. 1B) and in a murine model after high dose APAP (Fig. 1C).
Additionally, we evaluated the expression of XBP1 by IHC staining in liver tissues from mice treated with either vehicle or high-dose APAP [500 mg/kg] for 24 h (Fig. 1D). Positive staining for XBP1 was evident in hepatocyte nuclei in control-vehicle treated mice. However, 24 h after high-dose APAP triggered significantly increased XBP1 expression in the liver parenchyma, and specifically in hepatocyte cytosols (Fig. 1D, E), suggesting that transcriptional activation of XBP1 is an event directly related with the UPR in human and murine APAP-induced liver injury.
Xbp1 Δhepa mice show decreased hepatic UPR and JNK1/2 activation in response to APAP Since our results evidenced splicing of XBP1 in APAP intoxication for the first time in human explants and hepatocyte cell line, we generated mice with specific deletion of Xbp1 in hepatocytes (Xbp1 Δhepa ). Littermate Xbp1 f/f mice were used as controls. Interestingly, livers from untreated Xbp1 Δhepa mice displayed significant hyperactivation of IRE1α and loss of XBP1 protein ( Supplementary Fig. 1A), whilst loss of Xbp1 in hepatocytes caused hyperactivation of IRE1α, as previously shown [1]. Therefore, we next sought to investigate the UPR after high-dose APAP in Xbp1 Δhepa mice. Whilst APAP did not greatly increase IRE1α phosphorylation in Xbp1 f/f , phospho-IRE1α protein levels were overexpressed in mice with hepatocyte-specific deletion of Xbp1 (Xbp1 Δhepa ) ( Fig. 2A). In turn, 24 h post-APAP, an increase in CHOP and BiP/GRP78 protein levels in Xbp1 f/f compared with Xbp1 Δhepa livers was observed ( Fig. 2A). Additionally, PERK protein levels were decreased in both untreated, and to a greater extent, in APAP-treated Xbp1 Δhepa mice, compared with Xbp1 f/f livers. Interestingly, peIF2 was absent in untreated Xbp1 f/f but significantly overexpressed in Xbp1 Δhepa livers. APAP induced also the expression of peIF2 in both Xbp1 f/f and Xbp1 Δhepa mice ( Fig. 2A). As expected, the levels of uXBP1 were not altered by high-dose APAP; however, sXBP1 protein was overexpressed in Xbp1 f/f compared with Xbp1 Δhepa livers ( Fig. 2A).
JNK is activated by APAP and plays a crucial role in hepatotoxicity [9][10][11]. To determine the effect of XBP1 ablation on JNK activation, we examined the JNK signaling pathway in liver tissues by immunoblotting, 24 h after high-dose APAP. Increased levels of pJNK were found in Xbp1 f/f mice, whereas lack of JNK induction was characteristic of Xbp1 Δhepa livers at this time point (Fig. 2B). Moreover, JNK1 and JNK2 phosphorylation was slightly decreased in Xbp1 Δhepa compared with Xbp1 f/f (Fig. 2B).
Ablation of Xbp1 in hepatocytes reduces hepatotoxicity both at early and late stages of high-dose APAP Concomitant to our results with APAP-derived hyperacute liver injury, marked splicing of XBP1 has been previously reported in mice, 24 h after high dose APAP [5]. Thus, we sought to thoroughly investigate the relevance of hepatocytic XBP1 in the pathophysiology of APAP-hepatotoxicity. Whilst APAP induced significant hyperacute liver injury in Xbp1 f/f animals, Xbp1 Δhepa mice displayed only a modest induction of serum ALT, AST and LDH levels, 24 h post-APAP ( Supplementary Fig. 1B-D). Macroscopic and histological analyses of the liver revealed markedly reduced liver injury in Xbp1 Δhepa mice compared with Xbp1 f/f animals which exhibited severe centrilobular necrosis ( Fig. 3A and Supplementary Fig. 1E), and significantly higher TUNEL-positive cells and cleavage of caspase-3 (CC3) (Fig. 3B and Supplementary  Fig. 1F). APAP-induced hyperacute liver injury is initiated by Nacetyl-p-benzoquinone imine (NAPQI), generated by several cytochrome P450 (CYP) isoenzymes, primarily, CYP2E1 [12]. CYP2E1 protein levels visibly increased after high-dose APAP in Xbp1 f/f mice, but with clearly reduced expression under basal conditions (DMSO, vehicle), and 24 h post-APAP in Xbp1 Δhepa compared with Xbp f/f mice ( Fig. 3C and Supplementary Fig. 1F).
Next, experiments for early time points of APAP hepatotoxicity were performed to examine the contribution of CYP2E1 at this stage. We thus challenged Xbp1 f/f and Xbp1 Δhepa mice with highdose APAP for 1 h and 2 h, and evaluated necrotic foci, cell death and CYP2E1 expression. Interestingly, 1 h post high-dose APAP challenge, triggered a tendency towards decreased hepatic injury and CYP2E1 expression, which was significant at the 2 h timepoint in Xbp1 Δhepa compared with Xbp1 f/f mice (Fig. 3D, E and Supplementary Fig. 2A-C).
CYP2E1 metabolism leads to the generation of reactive oxygen species (ROS) including lipid peroxidation, which can be measured as 4-HNE adducts using immunohistochemistry ( Supplementary  Fig. 3A), and triglyceride accumulation, a well-established consequence of severe ER stress [13], both in liver tissue and in serum ( Supplementary Fig. 3B-D). All these parameters were significantly reduced in Xbp1 Δhepa compared with Xbp1 f/f mice, 24 h after high-dose APAP. In addition, APAP treatment caused a significant upregulation of heme oxygenase-1 (Ho-1) in Xbp1 f/f compared with unaltered levels of this gene in Xbp1 Δhepa mice ( Supplementary Fig. 3E).
Increased ROS disrupts hepatic tight junctions (TJs) [14,15]. Both Xbp1 f/f and Xbp1 Δhepa mice treated with DMSO (vehicle) maintained normal cell polarity and adhesion as shown by ZO-1 western blots (Supplementary Fig. 1F). However, APAP caused a dramatic decrease of ZO-1 expression in Xbp1 f/f and to a lesser extent in Xbp1 Δhepa mice ( Supplementary Fig. 1F). Additionally, the number of CD45 and CD11b was significantly reduced in Xbp1 Δhepa mice, and a tendency towards reduced number of F4/80 cells was found ( Supplementary Fig. 4A-D). Concomitantly, reduced expression of mRNA transcripts for Tnfα in Xbp1 Δhepa indicated significantly decreased inflammation compared with Xbp1 f/f livers ( Supplementary Fig. 4E), as well as markers of inflammasome activation (eg IL-1β and Nrlp3) were significantly lower in Xbp1 Δhepa (Supplementary Fig. 4E).
Loss of Xbp1 in hepatocytes promotes autophagy after APAP treatment TEM analysis indicated the accumulation of abundant autophagosomes in livers of Xbp1 Δhepa mice following APAP treatment, whilst the hepatic parenchyma of Xbp1 f/f animals was predominantly necrotic ( Supplementary Fig. 5).
ER stress activation is frequently accompanied by Ca 2+ release and autophagy via the stimulation of AMPK, which leads to inhibition of the AKT/mTORC1 pathway, upregulation of the autophagy marker Atg5 and degradation of P62. This was characteristic of Xbp1 Δhepa compared with Xbp1 f/f livers, 24 h after high-dose APAP (Fig. 4A, B). Immunoblotting confirmed increased protein levels of pAMPK in livers of Xbp1 Δhepa mice, thus attenuating pAKT, and promoting the degradation of P62 and the conversion of microtubule-associated protein light chain 3 (LC3)-I to LC3-II associated with autophagosome formation and autophagy (Fig. 4C).
Since decreased liver injury (eg necrosis) and impaired CYP2E1 activity was also observed in Xbp1 Δhepa animals, 1-2 h following high-dose APAP, we examined whether autophagy in these animals was also enhanced at early stages of hyperacute liver injury. Significantly increased Atg5 mRNA transcripts and decreased P62 mRNA expression was found at 2 h-but not at 1 h-after high-dose APAP in Xbp1 Δhepa livers (Fig. 4D, E). Moreover, we explored changes in enzymes of the glutathione (GSH) system, including glutathione peroxidase-1 (GPX1) and -4 (GPX4) which were restored at early stages (1-2 h) of APAP hepatoxicity in Xbp1 Δhepa but depleted in Xbp1 f/f mice (Fig. 4F, G).
Pretreatment with STF-083010 mitigated liver injury induced by high dose APAP STF-083010 (STF), an inhibitor of the endonuclease activity of IRE1α, blocks the splicing of XBP1 in vivo and thus prevents the initiation of the UPR [16]. Therefore, our next experimental approach was to pharmacologically mimic the protection from APAP-mediated liver injury observed in Xbp1 Δhepa mice. C57BL/6J mice were pretreated with STF-083010 and sacrificed 24 h after high-dose APAP. A significant amelioration of serum levels of ALT and AST (Fig. 5A-C), and a tendency towards decreased LDH levels (Fig. 5D) was observed in STF + APAP-treated mice. Blinded histopathological evaluation of liver tissue demonstrated that STF manifestly mitigated APAP-induced hepatic necrosis and inflammation (Fig. 5E). Moreover, STF markedly attenuated APAP- Fig. 2 Consequences of ablation of XBP1 in hepatocytes in the UPR and JNK activation, 24 h after high dose APAP. A Protein levels of pIRE1α, IRE1α, BiP, CHOP, peIF2α, PERK, sXBP1, and uXBP1, were determined. B Protein levels of pJNK, JNK, pJNK1, JNK1, pJNK2, JNK2 were determined. GAPDH was used as a loading control. Relative protein levels were quantified using ImageJ software. Data are expressed as mean ± SEM (<0.05-****p < 0.0001; # p < 0.05 ## p < 0.0001;*intragroup, # intergroup).   induced overexpression of BiP, CHOP, and sXBP1 but had no effect on peIF2α protein levels (Fig. 5F). Interestingly, STF promoted a slight activation of AMPK, and the degradation of P62 and the conversion of LC3-I to LC3-II (Fig. 5G).
STF-083010 post-treatment does not modify APAP-induced CYP2E1 activity In order to rule out possible off-target effects of STF, we injected STF at 1-2 h after high-dose APAP (Fig. 6A). APAP significantly increased the percentage of necrotic area, TUNEL-positive cells and CYP2E1 activity at early stages (1-2 h), whilst STF treatment had no effect on these parameters. In contrast, high-dose APAP and post-treatment with STF (APAP + STF) at 1 and 2 h, significantly decreased the percentage of necrotic areas and TUNEL-positive cells, but had no impact on CYP2E1 protein expression (Fig. 6B-G). These results were validated by immunoblot, where STF alone or APAP + STF had a greater impact on CYP2E1 expression after 2 h (Fig. 6H). Finally, while APAP + STF was capable of restoring Gpx1 depletion, STF post-treatment did not change the mRNA expression of Atg5 or P62 (Fig. 6I-K).
Autophagy modulation and CYP2E1 expression in Xbp1 Δhepa animals Given that hepatocyte-specific deletion or pharmacological blocking of XBP1 splicing might trigger autophagy activation, we further evaluated the consequences of inducing or blocking autophagy with drugs in our experimental setting. Because suppression of mTOR is a central molecular signaling pathway leading to autophagy induction [17], we used rapamycin (RAPA). In contrast, chloroquine (CQ), a well-known lysosomal inhibitor was utilized as an autophagy blocking agent. Both RAPA and CQ were administered i.p. 1 h prior to high-dose APAP to both Xbp1 f/f and Xbp1 Δhepa mice and sacrifice took place at early (1 h) and late stages (24 h) of hyperacute liver injury (Fig. 7A). APAP challenge induced LC3-I and mildly LC3-II in Xbp1 f/f livers, which was accentuated by RAPA, which strongly caused overexpression of LC3-I/II (Fig. 7B). In sharp contrast, APAP clearly triggered overexpression of LC3-I/II in Xbp1 Δhepa livers with no effect observed in RAPA-pretreated Xbp1 Δhepa mice. Interestingly CQ pre-treatment ameliorated autophagy both in Xbp1 f/f and more clearly in Xbp1 f/f animals (Fig. 7B).
We previously showed that Xbp1 Δhepa mice have decreased protein levels of CYP2E1 in basal conditions ( Fig. 3C and Supplementary Fig. 1F). Thus, we speculated that autophagy could prevent from CYP2E1-mediated hepatotoxicity to Xbp1 Δhepa mice. APAP challenged decreased CYP2E1 levels in Xbp1 f/f mice, a phenomenon reverted by pretreatment with RAPA or CQ. Expectedly, Xbp1 Δhepa livers displayed decreased basal CYP2E1 levels, which were barely induced by RAPA nor by CQ both at early (1 h) and late (24 h) stages of APAP hepatotoxicity (Fig. 7C), indicating that autophagy might be responsible for suppressed CYP2E1 activity in hepatocyte-specific Xbp1 knockout mice.

DISCUSSION
Although APAP hepatotoxicity remains the leading cause of acute liver failure (ALF), temporal and quantitative effects of direct APAP-mediated liver injury on the induction of endoplasmic reticulum (ER) stress remain poorly understood. Activated IRE1 splices the mRNA of XBP1 in a non-canonical fashion, yielding the potent transcription factor sXBP1 [18,19].
Interestingly, we found that whilst most hepatocyte nuclei expressed XBP1, cytoplasmic expression of XBP1 became more evident as hepatocyte architecture was altered and serum transaminases increased in patients with APAP overdose.
Mechanisms of APAP hepatotoxicity have been reported to be similar in both humans and mice [14]. Indeed, mRNA expression of spliced/unspliced (sXBP1/uXBP1) in human hepatic HepaRG cells was consistent with our findings in murine APAP hepatotoxicity.
Our data also showed that hepatic deletion of Xbp1 leads to marked hyperactivation of IRE1α as previously reported [1,20]. Moreover, we found that reduced CHOP, BiP/GRP78 and JNK activation were involved in protection against APAP-induced hepatotoxicity. Additionally, protection from APAP through activation of regulated IRE1-dependent decay (RIDD) might play a further role in the protection against APAP, since enhanced RIDD leads to pro-death signaling by reducing the levels of select microRNAs that normally repress pro-apoptotic targets [21]. Furthermore, decreased sXBP1 was linked to the suppression of cytochrome P450 activity [3].
While our data obtained in wildtype mice mirrors previous reports indicating that CYP2E1 activity increases at early time points (1-4 h), and then increases only at 24 and 96 h after APAP in rodents [22,23], Xbp1 Δhepa livers were characterized by decreased CYP2E1 both at early and late phases of APAP hepatoxicity. This was linked to restoration of the GSH enzymes, preventing the severe oxidative stress occurring in normal mice. These data also correlated with the cytoprotective effectreduced number of necrotic areas and decreased cell death-of Xbp1-deficient mice, at early and late stages after high dose APAP.
In recent years, accumulating evidence has suggested that autophagy is activated against APAP-induced liver injury [24]. In our mice, TEM showed hepatocytes had normal ER ultrastructure in Xbp1 Δhepa mice, but abundant autophagosome formation and less mitochondrial damage upon APAP toxicity. Additionally, hepatocytic Xbp1 deficiency significantly promoted the phosphorylation of AMPK, augmented the effect of high dose APAP on the induction of LC3-II. Moreover, while Atg5 levels were increased, the levels of P62 were decreased both at early and late stages of APAP hepatoxicity, indicating that APAP-induced autophagy in the liver was a consequence of Xbp1 deletion.
STF-083010 (STF), a specific inhibitor of IRE1α RNase, inhibits both endogenous and chemically-induced Xbp1 splicing without affecting IRE1α phosphorylation. Pharmacological inhibition of STF not only markedly reduced the levels of serum transaminases, but also mitigated liver injury and prevented oxidative burst induced by high-dose APAP in mice. This is in agreement with previous reports demonstrating anti-oxidant and anti-inflammatory effects of STF [21,25]. Moreover, our study demonstrated that STF has a protective effect against APAP hepatotoxicity, and promoted the activation of autophagy, without affecting CYP2E1 activity. Our data are also in agreement with previous studies showing that pretreatment with STF facilitated hepatocyte autophagy in response to toxin stimulation [26].
We next evaluated whether induction or pharmacological blocking of autophagy using Rapamycin (RAPA) and Chloroquine (CQ), respectively, might affect CYP2E1 levels. Our results suggest that reduced CYP2E1-associated drug metabolism is associated with enhanced autophagy in the absence of hepatocytic XBP1. These data were similar at early and late stages of APAP hepatotoxicity.
In summary, this study provides new mechanistic information on the role of ER stress, UPR and autophagy in APAP-mediated Fig. 6 Post-treatment with STF does not modify CYP2E1 expression during APAP hepatoxicity. A C57BL/6J mice were injected APAP [500 mg/kg], and 1 h later STF [75 mg/kg] or an equal volume of PBS with 10% DMSO. Mice were sacrificed 1 and 2 h later. A control group was included with only STF-injected mice. B Representative H&E staining was performed in liver paraffin sections. Scale bars, 100 μm. C Representative TUNEL staining was performed on frozen liver sections Arrows denote positive cells. Scale bars, 100 μm. D Representative IHC staining for CYP2E1. Scale bars, 100 μm. E Necrotic foci were quantified and graphed. F TUNEL-positive cells were quantified and graphed. G Positive area for CYP2E1 staining was calculated using ImageJ and graphed. H Protein levels of CYP2E1 were determined using western blot. GAPDH was used as a loading control. Relative protein levels were quantified using ImageJ software. I-K mRNA expression was analyzed for Gpx1 (I), Atg5 (J), p62 (K), and were evaluated using qRT-PCR. Data are expressed as mean ± SEM (*p < 0.05-****p < 0.0001;* intragroup; N = 3-6 per experimental group). Fig. 7 Autophagy intervention using rapamycin (RAPA) or chloroquine (CQ) during early and late APAP-derived hyperacute liver injury. A Xbp1 f/f and Xbp1 Δhepa mice were intraperitoneally injected with either rapamycin (RAPA) [8 mg/kg] or chloroquine (CQ) [60 mg/kg], 1 h prior to APAP [500 mg/kg] challenge, and sacrificed at early (1 h) or late times (24 h) after APAP toxicity. B Protein levels of LC3I/II were determined in Xbp1 f/f and Xbp1 Δhepa mice, pretreated with RAPA or CQ and sacrificed 24 h APAP challenge. C Protein levels of CYP2E1 were determined in Xbp1 f/f and Xbp1 Δhepa mice, pretreated with RAPA or CQ and sacrificed 24 h (top panel) -same membrane used as in (B) and 1 h (bottom panel) post-APAP. GAPDH was used as a loading control. Relative protein levels were quantified using ImageJ software. D Schematic representation of the effects of genetic deletion of XBP1 in hepatocytes and pharmacological intervention, during APAP hepatotoxicity. hepatotoxicity (Fig. 7D). Altogether, our study confirms previous observations [27] suggesting that pharmacological modulation of autophagy, including decreasing expression of the transcription factor sXBP1, a key regulator of the UPR, may be a novel therapeutic approach for treating APAP-induced liver injury.

DATA AVAILABILITY
Datasets used or analyzed in the current study are available from the corresponding author upon reasonable request.