Roles of X-box binding protein 1 in liver pathogenesis
Article information
Abstract
The prevalence of drug-induced liver injury (DILI) and viral liver infections presents significant challenges in modern healthcare and contributes to considerable morbidity and mortality worldwide. Concurrently, metabolic dysfunction-associated steatotic liver disease (MASLD) has emerged as a major public health concern, reflecting the increasing rates of obesity and leading to more severe complications such as fibrosis and hepatocellular carcinoma. X-box binding protein 1 (XBP1) is a distinct transcription factor with a basic-region leucine zipper structure, whose activity is regulated by alternative splicing in response to disruptions in endoplasmic reticulum (ER) homeostasis and the unfolded protein response (UPR) activation. XBP1 interacts with a key signaling component of the highly conserved UPR and is critical in determining cell fate when responding to ER stress in liver diseases. This review aims to elucidate the emerging roles and molecular mechanisms of XBP1 in liver pathogenesis, focusing on its involvement in DILI, viral liver infections, MASLD, fibrosis/cirrhosis, and liver cancer. Understanding the multifaceted functions of XBP1 in these liver diseases offers insights into potential therapeutic strategies to restore ER homeostasis and mitigate liver damage.
INTRODUCTION
Recent epidemiological findings indicate an estimated annual occurrence of around 20 new cases of drug-induced liver injury (DILI) per 100,000 individuals. Idiosyncratic DILI constitutes approximately 11% of acute liver failure cases in the United States. Factors contributing to DILI risk include medication dosage, drug lipophilicity, and the degree of hepatic metabolism. While the impact of host factors such as age, gender, and chronic liver conditions on DILI development remains uncertain, existing evidence presents a mixed perspective [1-3]. Similar to DILI, viral liver infections have exerted a profound influence on human health, contributing to substantial morbidity and mortality among those afflicted with acute and chronic infections [4,5]. Initially shrouded in mystery regarding their origin, the identification of viral agents encouraged the scientific community’s interest in unraveling their pathogenesis and developing diagnostic tools to identify affected individuals. Over the centuries, rapid advancements in scientific and technological expertise have facilitated the possibility of controlling and even curing such infections. Preventive medicine, particularly through vaccination, has emerged as a pivotal focus in mitigating the impact of these infections [6]. Thus, DILI and viral liver infections are expected to pose significant challenges in the current century.
The term, metabolic dysfunction-associated steatotic liver disease (MASLD), initially proposed as non-alcoholic fatty liver disease (NAFLD) by Schaffner in 1986 [7], has emerged as a significant form of chronic liver illness, largely due to the increasing rates of obesity observed over the decades [8,9]. Recognized as the hepatic expression of metabolic syndrome, it encompasses factors such as abdominal obesity, hypertriglyceridemia, low high-density lipoprotein (HDL) cholesterol, arterial hypertension, and hyperglycemia [9]. The histopathological spectrum of MASLD ranges from simple steatosis to steatohepatitis and fibrosis, with potential progression to more severe conditions such as cirrhosis and hepatocellular carcinoma [10]. Therefore, the development of treatment for patients with MASLD would likely have a considerable effect on morbidity and mortality. As our understanding of the disease etiology and mechanisms has expanded, it has become clear that MASLD does not adequately encompass the diverse disease processes. While obesity and metabolic syndrome are recognized as significant risk factors for MASLD [9,11], a considerable proportion of the patients have a normal body mass index (BMI), that is, non-obese (lean or non-obese) [12-14]. Consequently, recognizing the limitations of the NAFLD definition, a new term, MASLD, was proposed in 2020 [15,16]. Unlike NAFLD, which is diagnosed after excluding other causes (e.g., alcohol) [17,18], MASLD emphasizes metabolic abnormalities [15]. In other words, we can diagnose MASLD if hepatic steatosis coexists with metabolic risk factors. The global adoption of the new name will take time, but this effort highlights the prevalence of MASLD patients in our society and the growing interest in treating health issues.
The discovery of X-box binding protein 1 (XBP1) via cloning marked an advancement in 1990. It was identified as a distinctive basic-region leucine zipper transcription factor, involved in regulating human major histocompatibility complex class II gene expression [19]. Over the subsequent decades, numerous studies have revealed that XBP1 is downstream of inositol-requiring enzyme 1 alpha (IRE1α) which induces Xbp1 mRNA into its mature form via alternative splicing and its essential role as a transcription factor within the unfolded protein response (UPR) [20-22]. This response mechanism, found across invertebrates and vertebrates, is crucial for cell survival under stress conditions [23-26]. Moreover, the spliced form of XBP1, known as XBP1s, modulates the transcription of specific genes. This process varies depending on the cell type, to regulate a wide array of cellular functions, including lipid metabolism, glucose biosynthesis, and immune response [27-31]. Several reviews have covered the diverse physiological aspects of endoplasmic reticulum (ER) stress and delineated its involvement in the development of various diseases, including cancer and diabetes, as well as neurodegenerative, metabolic, and inflammatory disorders [26,32-35]. Hence, it is important to understand the role of XBP1s in the pathogenesis of liver diseases. This review aims to provide a comprehensive overview of the emerging roles and molecular mechanisms underlying XBP1s-mediated signaling pathways in liver disease pathogenesis, in conjunction with an exploration of current trends in potential therapeutic strategies.
XBP1s IN THE UPR
ER stands as one of the largest compartments within cells and facilitates numerous biological processes, encompassing the folding of cytoplasmic proteins during translation, maintenance of Ca2+ levels, and synthesis of steroids and lipids. Moreover, it acts as a core for vesicular transport, including the movement of endosomes. Integral to its functions are key regulatory elements such as protein chaperones involved in oxidation-reduction reactions, enzymes responsible for processes such as proteolysis, glycosylation, and sulfation, as well as Ca2+ transporters and channels [36,37]. ER stress arises from the accumulation of unfolded or misfolded proteins within the ER lumen, induced by factors such as viral infections, nutrient scarcity, drug exposure, and other disruptions to typical ER functions [38,39]. Three signaling pathways, known as the UPR, activate in response to restore ER homeostasis [31]. The UPR activates to eliminate unfolded or misfolded proteins when mild to moderate ER stress occurs. This form of the UPR has adaptive or cytoprotective roles and aims to restore normal ER function. Conversely, during severe or prolonged ER stress, the UPR becomes excessively hyperactivated, triggering the initiation of intrinsic apoptotic pathways. This variant of the UPR is termed maladaptive, unchecked, or terminal UPR [40,41]. Three transmembrane sensors in the ER, namely IRE1α, protein kinase RNA-like endoplasmic reticulum kinase (PERK), and activating transcription factor (ATF) 6, mediate the UPR (Fig. 1).

Three branches of the unfolded protein response. Various stimuli trigger ER stress, causing the accumulation of unfolded or misfolded proteins and activating the UPR. The accumulation of unfolded or misfolded proteins triggers the UPR, causing ER chaperone GRP78 to dissociate from the luminal domains of three ER stress sensors. Upon activation, IRE1α undergoes dimerization and trans-autophosphorylation by activating its endoribonuclease activity and initiating RIDD, which breaks down multiple types of mRNA molecules to reduce protein synthesis. PERK phosphorylates eIF2α, resulting in a decrease in protein translation and phosphorylated-eIF2α selectively increases the translation of ATF4. ATF6, an ER membrane-bound inactive precursor transcription factor, is transported to the Golgi apparatus in COPII-coated vesicles. There, it undergoes proteolytic cleavage by the S1P and S2P, resulting in activation. Activated XBP1s, ATF4, and ATF6 translocate to the nucleus to initiate the transcription of genes encoding ER chaperones or proteins, restoring ER homeostasis or inducing cell death. ATF4, activating transcription factor 4; ATF6, activating transcription factor 6; COPII, coat protein complex II; eIF2α, eukaryotic translation initiation factor 2α; ER, endoplasmic reticulum; GRP78, glucose-regulatory protein 78; IRE1α, inositol-requiring enzyme 1 alpha; PERK, protein kinase RNA-like endoplasmic reticulum kinase; RIDD, regulated IRE1α-dependent decay; S1P, site-1 protease; S2P, site-1 protease; UPR, unfolded protein response; XBP1, X-box binding protein 1.
IRE1α pathway
IRE1α is the most evolutionarily conserved branch within the UPR signaling pathway [42]. Activation of IRE1α leads to the RNase domain facilitating the non-conventional splicing of XBP1u mRNA, thereby producing the homeostatic transcription factor XBP1s (Fig. 1) [20-22]. XBP1s functions as a transcription factor that promotes cell survival by upregulating genes involved in chaperone functions and lipid metabolism [43,44]. Moreover, the active IRE1α RNase domain induces the endonucleolytic degradation of numerous mRNAs resembling XBP1 mRNA, a process termed regulated IRE1α-dependent decay (RIDD), initially discovered in Drosophila melanogaster in 2006 (Figs. 1 and 2) [45-47]. RIDD operates continuously under baseline conditions or mild ER stress and progressively intensifies as the severity or duration of ER stress increases (Fig. 2) [48-50]. If ER stress remains unresolved, prolonged RIDD activity eventually becomes cytotoxic, marked by decreased XBP1 mRNA splicing and increased RIDD activity (Fig. 1) [51].

Implications of the IRE1α/XBP1 pathway based on the different forms of IRE1α complexes. Cell fate is determined by the downstream consequences of ER stress-induced IRE1α activation, including proadaptive XBP1s-dependent transcriptional signaling, RIDD, and JNK pathway activation. IRE1α activation occurs through conformational changes prompted by the dimerization of monomers within the membrane plane, followed by trans-autophosphorylation. Upon additional stress stimuli, higher-order oligomers may form, enhancing IRE1α RNase activity. Activated IRE1α facilitates the splicing of XBP1 mRNA in normal cells under mild ER stress and cancer cells under chronic ER stress. The removal of an intron from the XBP1 transcript induces frame-shift, leading to the synthesis of a potent transcription factor, XBP1-spliced, which regulates numerous UPR target genes for survival. Otherwise, IRE1α activation by dimerization induces degradation and translational attenuation in the RIDD pathway to promote cell death or phosphorylated IRE1α with oligomer that is the conformational change from IRE1α dimer. IRE1α oligomer interacts with TRAF2 and thereby activates the JNK pathway via ASK1, ultimately leading to cell death in normal cell under chronic ER stress. ASK1, apoptosis signal-regulating kinase 1, ER, endoplasmic reticulum; IRE1α, inositol-requiring enzyme 1 alpha; JNK, c-Jun N-terminal kinase; RIDD, regulated IRE1α-dependent decay; TRAF2, tumor necrosis factor receptor-associated factor 2; UPR, unfolded protein response; XBP1, X-box binding protein 1.
PERK pathway
PERK is a type I transmembrane protein in the ER membrane, that functions as a sensor for unfolded or misfolded proteins in the ER lumen, triggering a signaling cascade [52]. After glucose-regulated protein 78 (GRP78) dissociates, PERK is activated through dimerization and subsequent autophosphorylation. A key downstream target of active PERK is eukaryotic initiation factor 2α (eIF2α) [53], whose phosphorylation at serine 51 inhibits general protein translation, thereby decreasing the accumulation of proteins in the ER lumen. PERK-deficient mouse embryonic fibroblasts lack the translational block and are thus hypersensitive to ER stress [54]. Similarly, knock-in cells with non-phosphorylatable eIF2α exhibit improved sensitivity to ER stress agents and cell death [55]. In addition, phosphorylation of eIF2α promotes the selective translation of specific mRNAs, such as ATF4 which upregulates UPR target genes related to amino acid synthesis and transport, reducing ER stress and restoring ER homeostasis [56]. However, sustained activation of PERK and eIF2α phosphorylation may lead to apoptosis. ATF4 also induces the transcription of CCAAT-enhancer-binding protein (C/EBP) homologous protein (CHOP), which is associated with ER stress-induced apoptosis. Notably, CHOP is additionally regulated at the translational level by phosphorylated eIF2α (Fig. 1) [57].
ATF6 pathway
ATF6, a type II transmembrane protein and basic leucine zipper transcription factor, dissociates from GRP78 upon ER stress. It is then transported via COPII-coated vesicles from the ER to the Golgi apparatus, where its C-terminal half is cleaved by site-1 protease [58] and the membrane-anchored N-terminus is cleaved by site-2 protease to produce an active N-terminal cytosolic fragment (ATF6N). ATF6N translocates to the nucleus to regulate the transcription of ER stress target genes, including XBP1, CHOP, and protein chaperones (Fig. 1) [59,60].
IRE1α transitions from dimeric to oligomeric forms in response to ER stress
Previously, it was assumed that the dimerization of IRE1α modified intracellular signaling according to the persistence of ER stress. However, recent studies have demonstrated that intracellular signaling varies depending on the form of the IRE1α complex [61-63]. Moreover, IRE1α has different oligomer functions in different conditions because IRE1α is a dual-activity enzyme, consisting of an endoribonuclease domain and a serine-threonine kinase domain [21]. Therefore, understanding these differences is considered highly important for understanding cellular physiology and pathology. It dimerizes upon activation of IRE1α under mild ER stress, which enables trans-autophosphorylation of its kinase domain (Fig. 2) [64,65]. Following this phosphorylation, IRE1α rearranges into continuous dimers or higher-order oligomeric structures, leading to the activation of its RNase domain to produce XBP1s in mild ER stress (Fig. 2) [40]. The activated RNase domain cleaves a 26-nucleotide intron from the XBP1u mRNA, which requires the formation of an oligomer consisting of at least an IRE1α tetramer (Fig. 2) [42,62,66]. Additionally, the active RNase domain partially contributes to cell survival during this period through RIDD [67]. In contrast, the kinase domain of IRE1α with UPR signaling triggers pro-apoptotic signaling pathways under chronic and severe stress conditions [68-70]. Moreover, IRE1α not only degrades other ER membrane-associated mRNAs through RIDD to regulate cell death [71], but also triggers proapoptotic signaling pathways to recruit the adaptor molecule, tumor necrosis factor (TNF) receptor-associated factor 2 (TRAF2), under chronic and severe ER stress conditions (Fig. 2) [40]. Subsequently, TRAF2 activates apoptosis signal-regulating kinase 1 (ASK1/MAP3K5) and c-Jun N-terminal kinase 1 (JNK/MAPK8/SAPK1), both known to induce apoptosis (Fig. 2) [72,73]. Furthermore, cancer cells exposed to persistent stress exhibit overexpression of IRE1α and XBP1s, which shifts the cellular response towards survival pathways rather than apoptotic ones (Fig. 2) [65,74].
XBP1s-MEDIATED REGULATION OF TARGET GENES
Studies have demonstrated that XBP1s modulates genes implicated in a variety of cellular processes, such as the ER stress response, lipid metabolism, and glucose homeostasis (Fig. 3) [43,44,75]. Moreover, XBP1s is vital for the development and maintenance of highly secretory cells, including hepatocytes, pancreatic progenitor and β cells, and intestinal Paneth cells [76-79]. The absence of XBP1s leads to insulin resistance and type 2 diabetes or promotes vulnerability to intestinal inflammation [79,80]. Also, XBP1 plays a role in preserving retinal function and synaptic integrity, likely by regulating synaptic scaffold proteins [81]. In retinal neurons, XBP1 knockout (KO) aggravates and hastens neurodegeneration under aging or diabetic conditions [82,83]. This leads to a more pronounced decline in retinal function, with significant morphologic defects such as a decrease in ERG β-wave strength and ribbon synapses loss observed after 5 months of diabetes or at 1 year of age in XBP1 KO mice [82,83]. Additionally, XBP1 is essential for maintaining retinal pigment epithelium (RPE) tight junctions by regulating actin cytoskeletal reorganization and calcium-dependent RhoA/Rho kinase signaling [84]. In RPE cells, XBP1 promotes cell survival by promoting antioxidant genes and attenuating ER stress [85-87]. Moreover, XBP1-mediated activation of the UPR and inhibition of NF-κB activation play a protective role against retinal endothelial inflammation, particularly through ER stress preconditioning [88]. These findings highlight the key role of XBP1s in diverse physiological and pathological processes. Moreover, XBP1s is linked to other diseases, such as obesity, diabetes, cancer, and inflammatory disorders, drawing attention to it as a critical molecular target for therapeutic development [26,32,34]. Thus, comprehensive information on disease-specific XBP1s targets and intracellular mechanisms may provide essential insights for developing new therapeutic strategies for these diseases.

Regulation of target gene transcription by XBP1. XBP1 modulates various cellular processes by regulating the transcription of numerous target genes. XBP1 regulates cell proliferation, affects glucose and lipid metabolism, and induces tumor formation and growth. XBP1, X-box binding protein 1.
Proliferation
Several researches have shown that XBP1 regulates the expression of various genes that maintain ER homeostasis, including those encoding ER-resident chaperones and elements of the ER-associated degradation (ERAD) machinery [44,89]. Also, IRE1α, XBP1s upstream regulator, is a substrate of the SEL1L-HRD1 protein (ERAD complex), and ER stress attenuates ERAD-mediated IRE1α degradation [90]. Recently reported new targets for XBP1s are peptidyl-prolyl cis-trans isomerase NIMA-interacting 1 (PIN1) and hepatocyte nuclear factor 4α (HNF4α). XBP1 binds to the WW domain of PIN1, with the Ser288-Pro motif of XBP1s being crucial for this interaction. This binding enhances the stability of XBP1s; however, PIN1 deficiency impairs XBP1s-induced cell proliferation and size increase (Fig. 3) [91]. Also, HNF4α controls the transcription of Xbp1 in pancreatic β-cells. Thus, silence or mutation of HNF4α reduces Xbp1 mRNA levels in INS-1 and MIN6 cells. In mice, the deletion of HNF4α causes a loss of XBP1 expression in pancreatic islets, resulting in altered ER structures and reduced Ca2+ levels in the ER (Fig. 3) [92].
Other XBP1 targets have been studied; XBP1 deletion in mice suppressed cyclin-dependent kinase 4 (CDK4) expression and affected BECLIN-1 transcriptional activation which is a major player in the initiation of autophagy (Fig. 3) [93,94]. Moreover, hepatocellular carcinoma downregulated 1 (HEPN1), a gene associated with cell growth arrest and apoptosis. The level of HEPN1 was upregulated by reactive oxygen species (ROS) and transcriptionally regulated by XBP1s (Fig. 3) [95].
Lipid metabolism
The IRE1α/XBP1 pathway is a key regulator of hepatic lipid metabolism (Fig. 3). Deleting IRE1α specifically in hepatocytes increased hepatic lipid levels and decreased plasma lipids by controlling lipid metabolism genes, including C/ebpβ, C/ebpδ, Pparγ, and enzymes involved in triglyceride biosynthesis [96]. Also, XBP1s directly regulates hepatic lipogenic genes, such as Dgat2, Scd1, and Acc2 [28]. Additionally, liver-specific deletion of Xbp1 impaired de novo hepatic lipogenesis, resulting in lower serum triglycerides, cholesterol, and free fatty acids [28]. CHOP, which may also be regulated by XBP1s, appears to influence lipid metabolism by inhibiting genes encoding C/EBPα and other proteins involved in lipid metabolism [97]. Hence, liver-specific CHOP-deficient mice exhibited less suppression of transcriptional regulators, such as C/EBPα, Pparα, Pgc1α, and Srebp1 under ER stress conditions compared to wild-type mice (Fig. 3). However, CHOP KO mice showed reduced expression of lipogenic genes and less lipid accumulation than wild-type mice when exposed to human immunodeficiency virus protease inhibitors (Fig. 3) [98]. Thus, the precise molecular and cellular mechanisms behind CHOP-mediated dysregulation of hepatic lipid metabolism are yet to be fully understood so further mechanism studies of CHOP regulated by XBP1s are needed.
Glucose metabolism
Recent studies have suggested that XBP1 may control glucose homeostasis by modulating intracellular signaling in pancreatic cells and hepatocytes and influencing adipocyte function (Fig. 3) [99-101]. Also, the IRE1/XBP1 pathway is associated with insulin resistance, type 2 diabetes, and obesity via its activation of c-Jun N-terminal kinase (JNK). In addition, ER stress triggers IRE1/JNK signaling, leading to a reduction in tyrosine phosphorylation (pY896) of insulin receptor substrate 1 (IRS1) and Akt phosphorylation in hepatocytes. At the same time, it increases serine phosphorylation of IRS1, ultimately impairing insulin signaling [27,101].
Moreover, XBP1 engages with insulin signaling mediators p38 mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K) to regulate the UPR in the liver [100]. Interestingly, XBP1 silencing led to inhibited viability of glioma cells which depend on glycolysis to make energy and sustain their survival, and decreased ATP/lactate production under insufficient oxygen, which is possibly regulated by repression of hexokinase II (HK2) expression (Fig. 3) [102]. A recent study demonstrated that XBP1 directly interrelates with the transcription factor forkhead box O1 (FoxO1) in the nucleus, which is involved in hepatic gluconeogenesis (Fig. 3). This interaction between XBP1 and FoxO1 is specific to certain cell types, such as hepatocytes, pancreatic cells, and adipocytes, and occurs independently of XBP1’s effects on ER folding capacity [99]. Another finding has suggested that XBP1s enhances the transcriptional activity of the Fgf21 promoter and its expression as a peroxisome proliferator-activated receptor (PPAR) α target gene, suggesting a physiological role for the IRE1α-XBP1 pathway in the response to starvation (Fig. 3) [103].
Cancer tumorigenesis/metastasis
As cancer cells spread from the primary tumor site and enter the bloodstream, only tumor-initiating cells can establish secondary tumors in distant organs. XBP1 plays a role in driving epithelial-to-mesenchymal transition (EMT), cell invasion, and metastasis in breast cancer (Fig. 3). Thus, XBP1 is significantly associated with the invasion and metastasis of both oral squamous cell carcinoma and breast cancer cells, and its expression may contribute to the proliferation of tumor cells [104,105]. Data from the cancer genome atlas reveal a link between elevated XBP1 levels and ER-positive tumors. The formation of the XBP1-HIF1α complex on target promoters is crucial for their transcriptional activation by RNA polymerase II in breast cancer (Fig. 3) [23]. Also, the IRE1/XBP1 pathway is implicated in cancers driven by N-Myc and c-Myc which influenced this pathway through various molecular mechanisms (Fig. 3) [106].
In addition, the IRE1-XBP1 pathway activates nuclear receptor coactivator 3 (NCOA3), which is also a transcriptional target during the UPR, forming a feedback loop between XBP1 and NCOA3 that regulates cell fate in ER-positive breast cancer cells (Fig. 3). Moreover, the upregulation of NCOA3 is essential for XBP1-mediated resistance to antihormonal treatments [107]. This study also found that XBP1 inhibition significantly suppressed cell proliferation in breast cancer and regulated cell differentiation [108]. Interestingly, lysyl oxidase-like 2 (LOXL2) facilitates EMT by accumulating in the ER, where it binds to GRP78, activating the IRE1-XBP1 pathway (Fig. 3). This activation results in the upregulation of EMT transcription factors, such as SNAI1, SNAI2, ZEB2, and TCF3, which are all direct targets of XBP1 [109]. Another research report on the XBP1 target is that upregulated XBP1 levels have been detected in esophageal squamous cell carcinoma and clinical samples and this overexpression highly correlates with advanced tumor stage, lymph node metastasis, and unfavorable prognosis. Furthermore, XBP1 promotes both cell proliferation and invasion, as demonstrated in vitro and in vivo, through its active regulation of matrix metalloproteinase-9 (Fig. 3) [110].
XBP1s IN DIFFERENT TYPES OF CELL DEATH
Among the UPR transducers, IRE1α is recognized as the most evolutionarily conserved and contains two functional domains for serine/threonine-protein kinase and endoribonuclease (Fig. 4) [111,112]. When IRE1α is activated, it facilitates the unconventional splicing of Xbp1 mRNA into its mature form [22,113]. This spliced Xbp1 mRNA causes the potent transcription factor XBP1s, which in turn promotes the transcription of genes related to UPR, ERAD, protein folding, and degradation to recover ER homeostasis and prevent the liver from further damage [114,115]. However, the cytosolic kinase domain of IRE1α engages TRAF2, stimulating ASK1 and its downstream effector JNK1. Consequently, this activation triggers the inflammatory response and oxidative stress, which play a role in regulating apoptosis [115]. Furthermore, the CHOP pathway is highly involved in increasing apoptosis activation through the IRE1α-XBP1 pathway [116,117]. In line with published reports, our group revealed that toxicant challenge upregulated Xbp1s and pleckstrin homology-like domain, family A, member-3 (Phlda3) mRNA levels in the liver, indicating a positive association between the mRNA levels of Xbp1s and Phlda3 [118]. Notably, XBP1s, downstream of IRE1α transcriptionally induce Phlda3 upon ER stress (Fig. 4), suggesting that XBP1 regulates various targets as a transcriptional factor in toxicant situations. Further, the induction of PHLDA3 through the IRE1α–XBP1 pathway caused hepatic apoptosis by suppressing AKT (Fig. 4).

Activation of apoptosis, necroptosis, pyroptosis, and ferroptosis regulated by XBP1s. The pathway of ER stress-induced apoptosis is activated not only by IRE1α, which initiates the TRAF2-ASK1-JNK inflammatory/oxidative stress pathway, but also by the CHOP pathway, facilitated by XBP1s. Similarly, PHLDA3 induction by XBP1s exacerbates apoptosis by suppressing p-AKT. Overexpression of XBP1s promotes necrotic cell death; XBP1s activates the CHOP and downstream protein NLRP3, resulting in caspase-1 activation and pyroptosis. XBP1s mediates ferroptosis by inducing Gα12 overexpression through enhanced ROCK1 via dysregulated ALOX12, miR-15a, and GPX4, facilitating lipid peroxidation. AKT, serine/threonine protein kinase B; ALOX12, arachidonate 12-lipoxygenase; ASK1, apoptosis signal-regulating kinase 1; CHOP, CCAAT-enhancer-binding protein homologous protein; ER, endoplasmic reticulum; Gα12, G protein subunit alpha 12; GPX4, glutathione peroxidase 4; IRE1α, inositol-requiring enzyme 1 alpha; JNK, c-Jun N-terminal kinase; miR-15a, microRNA-15a; NLRP3, nucleotide-binding domain, leucine-rich–containing family, pyrin domain–containing-3; PHLDA3, pleckstrin homology-like domain, family A, member-3; RIDD, regulated IRE1α-dependent decay; ROCK1, rho associated coiled-coil containing protein kinase 1; TRAF2, tumor necrosis factor receptor-associated factor 2; UPR, unfolded protein response; XBP1, X-box binding protein 1.
Necroptosis: Necroptosis represents a programmed cell death mechanism, characterized as a regulated form of necrosis, which is distinguished by cellular swelling and the subsequent rupture of the plasma membrane [119]. It is orchestrated by the enzymatic actions of receptor-interacting serine/threonine-protein kinase (RIPK) 1, RIPK3, and mixed-lineage kinase domain-like protein (MLKL) [120-122]. A recent study has demonstrated that the inhibition of XBP1s protects toxicants from increasing necroptotic cell death in the liver (Fig. 4) [123].
Pyroptosis: Pyroptosis, a form of programmed cell death, is initiated by inflammasomes, protein complexes formed in response to different stimuli. These inflammasomes serve as intracellular sensors, detecting danger signals such as pathogen-associated molecular patterns or danger-associated molecular patterns [124,125]. Once activated, inflammasomes recruit and activate pro-caspase-1/4/5/11, inactive precursors of a family of cysteine proteases crucial for pyroptosis [126]. Current studies have revealed that ER stress inducers stimulate IRE1α, potentially promoting hepatocyte pyroptosis by upregulating CHOP and the NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome. Conversely, tauroursodeoxycholic acid reduces CHOP expression and suppresses the activities of caspase-1/3/11, thereby decreasing interleukin-1β (IL-1β) secretion in response to toxicant exposure [127]. Similar findings suggested that the IRE1α-XBP1 pathway influences the activation of the hepatocellular NLRP3 inflammasome. The absence of XBP1 in mice resulted in the reversal of NLRP inflammasome activation (Fig. 4) [128].
Ferroptosis: Ferroptosis is a new mode of programmed cell death that differs from apoptosis and necrosis. It is defined by the accumulation of lipid peroxidation products in the cell membrane, causing oxidative injury and the reduction of glutathione peroxidase 4 (GPX4), an antioxidant enzyme crucial for protecting cells from membrane lipid peroxidation. This cascade ultimately triggers ferroptotic cell death [129]. In recent study, Gα12 overexpression by IRE1α-XBP1s either inhibits GPX4 or promotes arachidonate 12-lipoxygenase (ALOX12) during ferroptosis (Fig. 4) [130]. This finding is in line with the observation that dihydroartemisin-induced ferroptosis is significantly attenuated in the knockdown of XBP1s [131].
ROLE OF XBP1s IN DILI
Acetaminophen (APAP)
APAP is a commonly used pain-relieving medication. While generally safe within recommended doses, excessive intake of APAP may lead to liver injury marked by centrilobular hepatic necrosis [132]. The primary mechanism behind APAP-induced liver damage involves the formation of N-acetyl-p-benzoquinone imine (NAPQI), a toxic metabolite generated through the oxidation of APAP by CYP1A2, CYP2E1, and CYP3A4, resulting in glutathione (GSH) depletion [133-137]. A reduction in GSH levels enables NAPQI to interact with cysteine and lysine residues within hepatocytes, forming adducts with proteins [138]. Additionally, NAPQI triggers the generation of ROS, initiates ER stress, and disturbs organelle function, culminating in the activation of JNK as a key mediator of APAP hepatotoxicity and subsequent hepatocyte death [139-144].
In a recent investigation, APAP-induced liver injury was identified as involving ferroptosis mediated by GPX4 due to XBP1s-induced Gα12 overexpression (Fig. 4). Also, accumulative data demonstrated that both pharmacological intervention and genetic suppression of XBP1 in hepatocytes prevent cell death from APAP-induced injury by inhibiting p-JNK1/2 (Fig. 5A) [145]. Similar results have shown that the absence of XBP1 reduces necrotic cell death in mice treated with APAP. This effect is attributed to the heightened activity of IRE1α and the subsequent degradation of Cyp1a2 and Cyp2e1 mRNA [123]. Moreover, APAP-induced XBP1s/NLRP3 activation is inhibited in Notch-activated mesenchymal stromal/stem cells through not only an AMPK/SIRT1-dependent manner but also XBP1s deacetylation [146]. These findings emphasize the role of ferroptosis and hepatotoxicity in the development of APAP-induced injury and highlight the necessity for additional research into the interplay between APAP toxicity accompanying XBP1 and ferroptosis [130].

Overview of the roles of XBP1s in drug-induced liver injury and viral hepatitis. (A–D) The induction of XBP1s by APAP causes hepatotoxicity via phosphorylation of JNK1/2, which is recovered by STF-083010 (A). Efavirenz, an anti-HIV drug, exacerbates hepatotoxicity by increasing XBP1s (B). Anti-depression drugs, sertraline, and nefazodone aggravate cytotoxicity in the liver through XBP1s overexpression, which is restored by 4-phenylbutyrate (4-PBA) (C). Anti-type 2 diabetes drugs including rosiglitazone and fenofibrate increase XBP1s levels and regulate lipid accumulation (D). (E, F) HBV augments EDEM through the IRE1α-XBP1 pathway (E). HCV increases XBP1s and PHLDA3 expression (F). HCV also can inhibit XBP1s trans-activating activity, suppressing EDEM (F). 4-PBA, 4-phenylbutyrate; EDEM, ER degradation-enhancing-mannosidase-like protein; HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus; IRE1α, inositol-requiring enzyme 1 alpha; JNK, c-Jun N-terminal kinase; PHLDA3, pleckstrin homology-like domain, family A, member-3; XBP1, X-box binding protein 1.
Anti-human immunodeficiency virus (HIV) drugs
Anti-HIV drugs are medicines used to treat HIV infection by targeting different stages of the virus life cycle and are associated with dyslipidemia and idiosyncratic hepatotoxicity, which are prevalent complications of HIV treatments. Several categories of anti-HIV medications, including nonnucleoside reverse transcriptase inhibitors (NNRTIs), NRTIs, and HIV protease inhibitors, have been associated with adverse effects [147].
Efavirenz is an NNRTI that inhibits HIV-1 reverse transcriptase by binding to the enzyme reversibly and in a noncompetitive manner. While typically regarded as safe, there is evidence suggesting that efavirenz disturbs lipid metabolism and triggers liver fibrosis [147-149]. Recent studies have demonstrated the ER stress markers, including CHOP, GRP78, phosphorylated eIF2α, and XBP1s are increased when primary hepatocytes are treated with efavirenz (Fig. 5B) [150]. One of the mechanisms involved in efavirenz-induced liver injury is mitochondrial dysfunction with ER stress, as evidenced by observations of morphological changes, including ER expansion [151-153]. Similarly, human hepatocytes exhibit efavirenz-induced XBP1 splicing and apoptotic phenotypes, with an inhibitor of XBP1 splicing attenuating the apoptotic observation [154], indicating the involvement of ER stress, particularly XBP1s, in efavirenz-induced hepatotoxicity (Fig. 5B).
Several HIV protease inhibitors may induce ER stress in both human and animal hepatocytes [155-157]. The ER stressinducing potentials of ritonavir, lopinavir, saquinavir, nelfinavir, and atazanavir have been investigated in experimental cell models at concentrations relevant to the clinical setting [156]. The presence of CHOP, ATF4, ATF3, and specific ER chaperones has shown that these protease inhibitors cause ER stress in HepG2 cells. The ER stress response is associated with its ability to inhibit the proteasome.
Recent studies have suggested that lopinavir and ritonavir induce ER stress in mice, primarily human and mouse hepatocytes, as well as the combined effect of alcohol-induced hepatic lipid accumulation and ER stress response. Both ritonavir and lopinavir induce mRNA levels of spliced Xbp1 in mice but do not increase serum alanine aminotransferase (ALT) activity. However, co-administration of these HIV protease inhibitors with alcohol increases ALT activity and synergistically enhances lipid accumulation and ER stress responses. Moreover, the combination of alcohol with either ritonavir or lopinavir reduces intracellular calcium levels and exacerbates cell death. These findings suggest that HIV protease inhibitors exacerbate alcohol-induced ER stress and liver dysfunction [155]. Similar results have shown that nelfinavir, atazanavir, lopinavir, and ritonavir increase GRP78, CHOP, ATF3, ATF4, and XBP1 expression [157-160].
Anti-depression drugs
Sertraline is prescribed to relieve depression, panic disorder, obsessive-compulsive disorder, and post-traumatic stress disorder. Despite being generally recognized as safe, cases of acute liver failure have been reported associated with sertraline use [161-165]. Sertraline has been found to exhibit toxicity in primary hepatocytes and HepG2 cells, with mitochondrial dysfunction and apoptosis identified as its underlying mechanisms [166,167]. Sertraline-treated hepatocytes show an increase in the expression of ER stress markers, including PERK, IRE1α, CHOP, and XBP1s (Fig. 5C). In addition, sertraline treatment elevates the expression of TNF and phosphorylated JNK, extracellular signalregulated kinase 1/2, and p38, which are components of the MAPK pathway [168,169]. Similar findings have been observed that nefazodone promotes CHOP, ATF4, p-eIF2α, and XBP1s levels in HepG2 cells (Fig. 5C). Moreover, 4-phenylbutyrate (4-PBA), an ER stress inhibitor, reduces sertraline and nefazodone-induced ER stress and cytotoxicity (Fig. 5C). The findings suggest that the mechanism underlying sertraline-induced liver toxicity is complicated, involving the interplay between ER stress, mitochondrial dysfunction, and apoptosis, along with the participation of the MAPK signaling pathway.
Anti-type 2 diabetes drugs
Peroxisome PPARγ agonists are oral medications developed to treat type 2 diabetes, characterized by insulin resistance. Ciglitazone, a PPARγ agonist undergoing early clinical trials, has not been released due to concerns about hepatotoxicity. Troglitazone, the first PPARγ agonist approved by the US FDA, was withdrawn from the market because of severe liver failure [170,171]. The mechanisms underlying this liver toxicity include mitochondrial dysfunction, apoptosis, impaired calcium balance, oxidative stress, and ER stress [172-179]. Treatment with fenofibrate and rosiglitazone increases in hepatic triglyceride (TG) compared to the control, enhancing both the amount and size of lipid droplets. These drugs activate both the fatty and anti-fatty liver metabolic pathways, worsening hepatic lipid accumulation (Fig. 5D). Rosiglitazone treatment enhances lipid and glucose transport in the liver through Fabp4 and Slc2a2 and modifies the immune and oxidative responses by downregulating Emr1 and Pon1. Of note, rosiglitazone treatment significantly induces the Xbp1 gene, which tends to be slightly upregulated by fenofibrate (Fig. 5D) [180].
Among several types of anti-type 2 diabetes drugs, pioglitazone improves insulin sensitivity in humans but does not affect indicators of ER stress. Changes in pre- and posttreatment ER stress levels are not associated with alterations in insulin sensitivity or BMI. In an in vitro assay, palmitate, thapsigargin, and tunicamycin induced ER stress in HepG2 cells, resulting in elevated levels of transcripts such as CHOP, ERN1, GADD34, and PERK. Additionally, they increased XBP1 splicing and phosphorylation of eIF2, JNK1, and c-JUN. However, pre-treatment with pioglitazone did not affect ER stress, regardless of the inducing agents such as palmitate, tunicamycin, and thapsigargin [181]. These findings suggest that a decrease in ER stress does not mediate the improvement of insulin sensitivity using pioglitazone.
ER STRESS IN VIRAL HEPATITIS
Viral hepatitis represents a considerable public health issue, affecting millions worldwide due to infection by hepatotropic viruses such as hepatitis A-E. In 2015, ~approximately 257 million people worldwide were living with hepatitis B virus (HBV), and 71 million were living with hepatitis C virus (HCV) [182]. Both HBV and HCV have a long-term infection, causing chronic infections that are associated with complications such as liver fibrosis and hepatocellular carcinoma (HCC) [183-185]. Despite advances in antiviral treatments, resistance remains an obstacle, especially in the treatment of HBV, highlighting the need to explore new therapeutic targets [186]. The pathogenesis and treatment of HBV and HCV infections, as well as their associated diseases, increasingly recognize the role of ER stress.
ER plays a role throughout the replication cycle of HBV and HCV [187,188]. As viral replication progresses, the accumulation of viral proteins and replication intermediates occurs, triggering ER stress [189]. Moreover, when there is an imbalance between the rapid processing demand of viral proteins and the normal protein processing capacity, ER undergoes stress [190]. In response, infected cells activate the UPR to suppress viral activity. However, viruses have developed mechanisms to evade this inhibition of viral replication [191]. A recent study has shown that HBV can enhance viral replication by utilizing the autophagosome/multivesicular body axis following ER stress [192]. In this article, we discuss how viruses interact with pathways associated with the ER stress response.
XBP1 in HBV
HBV, which belongs to the Hepadnaviridae family, is characterized as a compactly enveloped DNA virus [193]. The HBV genome consists of partially double-stranded circular DNA containing four overlapping open reading frames, encoding the core protein, surface protein, DNA polymerase, and HBV X protein (HBx) [194]. It has been suggested that HBV activates the IRE1α-XBP1 pathway (Fig. 5E) [195]. HBx acts as either an adaptor or a kinase activator and promotes the ATF6 and IRE1α-XBP1 pathways of UPR, suggesting that it may enhance HBV replication and expression in l iver cel ls, thus playing a role in l iver pathogenesis [196]. Similarly, the mRNA level of Xbp1 increased in the livers of 4-month-old HBV mutant-1 mice compared to wild-type HBV transgenic and non-transgenic mice [197]. Upon activation of the IRE1α-XBP1 pathway, the expression of ER degradation-enhancing-α-mannosidase-like protein (EDEM) increases (Fig. 5E), aiming to regulate overall protein levels and alleviate stress, thereby defending HBV-infected cells [185]. Through this mechanism, HBV establishes persistence and induces chronic inflammation in hepatocytes.
XBP1 in HCV
HCV, categorized in the Flaviviridae family, is a small, enveloped RNA virus. The RNA genome with positive polarity consists of 5’ and 3’ untranslated regions and a long open reading frame encoding a polyprotein precursor. This polyprotein precursor undergoes processing to generate at least 10 structural and non-structural viral proteins (i.e., core, E1, E2, p7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B) [198]. Previous studies failed to identify a clear induction of UPR-responsive genes in the liver of patients with untreated chronic hepatitis C, but they did observe evidence of ER stress and activation of the three UPR pathways [199]. The researchers discovered that cells carrying HCV subgenomic replicons increase the expression of XBP1 but suppress its transactivating function, preventing the transcriptional activation of EDEM (Fig. 5F). Consequently, misfolded proteins remain stable in these cells, inhibiting the initiation of the UPR pathways. Thus, HCV might suppress the IRE1α-XBP1 pathway to enhance the production of its viral proteins [200]. Similarly, IRE1α-dependent splicing of XBP1 is seen in cytomegalovirus-infected cells, with no accumulation of the target gene EDEM [201]. Surprisingly, our recent findings have reported that PHLDA3 mRNA levels are significantly higher in HCV patients with hepatitis compared to patients without hepatitis. Consistently, PHLDA3 protein levels are increased in HCV patients with hepatitis and elevated XBP1s (Figs. 4, 5F) [118]. HCV regulates the bidirectional IRE1α-XBP1 pathway in infected hepatocytes during this process, necessitating further in-depth studies [200].
XBP1 IN MASLD
With shifts in dietary habits and lifestyles, MASLD has emerged as a global health concern [202,203]. MASLD is a spectrum of liver diseases encompassing simple hepatic steatosis to progressive metabolic dysfunction-associated steatohepatitis (MASH), which can lead to advancing fibrosis, cirrhosis, and HCC [204-206]. The escalation in the prevalence of obesity and metabolic syndrome has contributed to an estimated global MASLD prevalence of ~around 24% [203]. About 20% of individuals with MASLD progress to MASH, and up to 5% develop cirrhosis [203]. Given the unclear mechanisms underlying MASLD, identifying the molecular target responsible for MASH is imperative for developing effective treatments. Some studies suggest that ER stress in UPR proteins plays a role in human MASLD and MASH [207,208]. Specifically, MASH is linked to elevated p-eIF2α levels and impaired XBP1 expression, although conflicting data exist regarding increased XBP1 expression and its target genes in MASH liver biopsies [207,209,210]. Findings from animal models of MASLD support the significance of the three UPR pathways, including XBP1, in the progression of MASLD [211].
ER serves as the primary site for lipid metabolism within hepatocytes, and one of the key roles of the UPR is to uphold lipid balance in the liver [212-214]. TGs are synthesized from fatty acids (FAs) and glycerol predominantly within the ER [215-217]. Additionally, ER is where the assembly of very low-density lipoprotein (VLDL) occurs before its transport to the Golgi apparatus [218-221]. Hence, ER function is closely linked to lipid metabolism. Previous research has demonstrated that ER stress can induce hepatic steatosis in mice [222,223], while excess lipid accumulation may also lead to ER stress. Saturated FAs such as palmitate have been found to activate the UPR, including XBP1, both in liver tissue and in hepatocytes. Therefore, the hepatic IRE1α/XBP1 pathway plays a role in lipid metabolism by regulating hepatic lipid synthesis and the assembly and secretion of VLDL [224]. More specifically, deficiency of hepatic IRE1α, resulting in XBP1 inactivation, leads to an increase in basal liver lipid content and exacerbates liver steatosis elicited by ER stress inducers (e.g., tunicamycin), while reducing VLDL secretion [96]. Moreover, loss of hepatic IRE1α, leading to XBP1 inactivation, worsens high-fat diet-induced steatosis, partly by upregulating miR-200 and miR-34 and downregulating their targets PPARα and SIRT1 [225]. In addition, XBP1 transcriptionally regulates FGF21 to protect against ER stress-induced liver steatosis [103]. Consistently, overexpression of XBP1 reduces diacylglycerol and hepatic TG levels in diet-induced and genetically obese mice, associated with decreased hepatic fatty acid synthesis and augmented lipolysis [226]. In contrast, hepatic loss of XBP1 lowers serum cholesterol and TG levels by reducing de novo lipogenesis [28]. Thus, XBP1 plays a role in controlling lipogenesis in the liver, indicating that impaired lipid storage may contribute to hepatic insulin resistance in models of ER stress. Studies have shown that eliminating XBP1s in hepatocytes protected mice from liver fat accumulation and insulin resistance [227]. While XBP1 regulates lipid metabolism-related genes, lipogenic genes (e.g., diacylglycerol O-acyltransferase 2 and angiopoietin-like 3) are controlled by RIDD [27,75]. The activation of the RIDD pathway in XBP1-KO mice leads to the breakdown of mRNA of the lipogenic genes, thereby playing a role in the hypolipidemic phenotype in the mice [228]; XBP1 KO mice exhibit reduced hepatic steatosis when fed a high-calorie diet; however, their susceptibility to diet-induced liver injury depends on diet composition and strain backgrounds [228,229]. Given the existence of contradictory reports regarding the function of XBP1, it is imperative to carefully examine the physiological response.
XBP1 alleviates ER stress and prevents hepatic steato-sis, while in XBP1-KO mice, long-term unrelieved ER stress promotes steatosis [230]. Moreover, another study showed that SIRT6 mitigates hepatic ER stress by deacetylating XBP1s, thereby improving hepatic lipid accumulation, apoptosis, and insulin resistance [231]. SIRT6 ablation in hepatocytes enhances the XBP1 pathway of the UPR, potentially altering the protein stability of XBP1 through its deacetylation. Consistently, another research suggests relationships between prolonged and uncontrolled ER stress and MASLD [80], and that SIRT6 deficiency exacerbates hepatic insulin resistance [232,233].
Humans exhibit several prominent 12-hour oscillations affecting daily food intake, immune function, body temperature, cognitive performance, and various hormone levels in circulation [234-236]. These cycles, many of which are connected to ER stress and UPR pathways, play a role in maintaining metabolic balance [237-239]. In particular, as is well known, hepatic disturbances in glucose and lipid regulation are key features of MASLD [240-242]. Several studies have suggested that genetic deletion of XBP1 in the liver disrupts the 12-hour rhythms, adversely influencing metabolic parameters and the development of an MASLD phenotype [243-245] Collectively, XBP1 is involved in various aspects of the pathogenesis. It regulates hepatic lipid metabolism, modulates insulin resistance, and affects 12-hour rhythms in MASLD (Table 1). Diverse investigations into the intricate interactions of XBP1 with different molecular pathways will provide insights into the complexity of MASLD and guide the development of targeted treatments.
XBP1 IN FIBROSIS/CIRRHOSIS
Liver fibrosis is a serious outcome resulting from a range of chronic or prolonged insults, culminating in the progression of liver cirrhosis and HCC [246-250]. The activation of hepatic stellate cells (HSCs) and the accumulation of extracellular matrix (ECM) are widely regarded as key factors driving the onset and progression of fibrosis [251-254]. This pathological process is linked to the imbalance between ECM production and degradation [255]. Transforming growth factor-β (TGF-β) is a pro-fibrotic cytokine that drives the initiation and progression of liver fibrosis. TGF-β facilitates the accumulation of ECM while preventing its breakdown [256-258]. Additionally, TGF-β activates HSCs, further contributing to the fibrotic process [259]. A study found that TGF-β activates IRE1α-XBP1 during HSC activation, promoting the development of fibrosis [260]. In addition, XBP1-induced UPR stimulates HSCs via autophagy independently of TGF-β [209]. Through bioinformatics analysis, this finding is associated with the development of fibrosis in animal models of liver damage and MASH patients. Conversely, in XBP1 KO mice, continuous UPR activation is associated with progressive apoptosis, liver damage, and fibrosis [230]. Ablation of hepatic XBP1 and the preferential induction of JNK by activated IRE1α may contribute to increased apoptosis. XBP1 also plays a role in the progression of fibrosis by affecting macrophages. Further, mice deficient in hepatic XBP1 experience enhanced liver injury and fibrosis when consuming a high-fat sugar diet [229,261]. Depletion of XBP1 in macrophages hinders hepatic fibrosis in models of MASH and suppresses the activation of HSCs by reducing the expression of TGF-β1 [262]. Additionally, pharmacological inhibition of XBP1 prevents steatohepatitis and fibrosis in mice. Moreover, the activation of XBP1 inhibits BNIP3-mediated mitophagy, worsening liver fibrosis. XBP1 amplifies the activation of proinflammatory macrophages through a stimulator of interferon gene-dependent mechanisms, which play a role in advancing liver fibrosis [263].
Another study demonstrated that FXR KO mice deficient in hepatic XBP1 exhibit greater liver damage and increased fibrosis [264], indicating that the activation of the hepatic XBP1 pathway protects against depleting FXR mice, in line with the known adaptive and defensive function of XBP1 in resolving or mitigating ER stress. Furthermore, XBP1 LKO mice display exacerbated liver injury and fibrosis when subjected to ER stress [230]. On the other hand, during the progression of liver fibrosis, there is a notable increase in the expression of angiotensin II type-2 receptor (AT2R), which acts to mitigate liver fibrosis by inhibiting the IRE1α-XBP1 pathway [265]. In addition, the loss of AT2R during liver fibrosis leads to increased dimerization activation of IRE1α and increased XBP1 splicing. Consequently, spliced XBP1 facilitates target gene transcription by binding to the AT2R gene promoter.
The harmful effects of hepatocyte-specific XBP1 ablation primarily result from the excessive activation of IRE1α [266]. Considering that XBP1 activation depends on IRE1α, a comprehensive understanding of the interplay between XBP1 and IRE1α seems essential in liver fibrosis. Xbp1 mRNA splicing, JNK activation, and IRE1α-dependent RNA decay consistently correlate with increased activity of IRE1α in fructose-induced liver injury in hepatocyte-specific XBP1 KO mice. Additionally, there is evident activation of eIF2α, accompanied by the upregulation of pro-apoptotic molecules such as CHOP, BIM, and PUMA. Moreover, XBP1 is also involved in the ERAD pathway [267]. The XBP1-mediated ERAD machinery in the ER stress pathway shows elevated activity, while the NRF2-driven antioxidant response pathway is diminished in the cirrhotic human liver (Table 1).
XBP1 IN HCC
Cancer remains a significant global health concern, particularly HCC, which has poor survival rates attributed to the absence of targeted therapies [250,268-272]. HCC ranked as the seventh most prevalent cancer in 2020 and ranked second in cancer-related mortality [273]. Hence, there is a pressing need to explore more efficacious treatment modalities to enhance the outlook for patients with HCC.
ER stress generally occurs in cancer cells and has been involved in various cancer types, including HCC. Globally, both the occurrence and mortality rates of HCC are increasing, especially in individuals with pre-existing liver cirrhosis. Intriguingly, HBV infection may lead to HCC even without cirrhosis and is a prevalent cause of liver cancer mortality worldwide. Furthermore, recent research suggests that HCC may develop at the stage of MASH before liver cirrhosis occurs [274]. The activation of ATF6, XBP1, and BIP has been observed in human HCC [275], and the UPR pathways are promoted at stages of tumorigenesis in mouse models [276]. The IRE1α pathway, which triggers XBP1 activation, may be particularly important for the initiation of HCC. IRE1α LKO mice, leading to XBP1 inactivation, show a decrease in HCC incidence when treated with diethylnitrosamine [277]. This alleviation in HCC is associated with the activation of STAT3, attenuated proliferation, increased apoptosis, and reduced production of inflammatory cytokines (e.g., TNFα, IL-6) [277]. Moreover, increased levels of IL-6, induced by XBP1, promote the proliferation of HCC cells through STAT3. This impact of IRE1α-XBP1 is nullified upon blocking IL-6-STAT3 signaling [278]. XBP1 is increased in human HCC tissues, and there is a positive correlation between hepatic IL-6 levels and XBP1. XBP1 binds to the IL6 promoter and enhances its transcription in human HCC cells, indicating a consistent role of XBP1 in regulating IL-6 expression. During the UPR, activation of the IRE1α/XBP1 pathway is crucial for tumor cell survival in adverse conditions.
Previous research has shown elevated UPR activity and increased expression of RACK1, an intracellular scaffold protein, in HCC cell lines and human tissue samples [279]. The activation of the IRE1α/XBP1 signaling pathway protects HCC cells when they encounter ER stress elicited by treatment with sorafenib, an apoptosis-inducing agent. Moreover, the overexpression of RACK1 augments IRE1α phosphorylation and enhances XBP1 mRNA splicing, protecting HCC cells from sorafenib-induced apoptosis. Furthermore, XBP1 plays a role in promoting the survival of HCC cells both in vitro and in vivo [280]. The mechanism involves XBP1-activating LEF1, a predominant co-factor of β-catenin, by binding to its gene promoter. Additionally, XBP1 physically interacts with LEF1, forming a complex that facilitates WNT signaling. Human HCC samples consistently show this relationship between XBP1 and LEF1, which is associated with disease prognosis. Particularly, inhibiting XBP1 selectively using an IRE1α inhibitor suppressed the viability of HCC. Further, XBP1 modulates the processes of EMT, cancer cell invasion, and metastasis via the Twist/Snail pathway, and its expression is linked with a poor prognosis [281]. Its impact on cell invasion has also been observed in other cancers, including oral and lung cancers [105].
The functions of XBP1 extend beyond its involvement in the UPR. XBP1 plays a protective role in endothelial cells by facilitating the phosphorylation of AKT, thereby mitigating oxidative stress [282]. Additionally, it contributes to tumor cell autophagy by facilitating FoxO1 degradation through recruitment to the proteasome [283]. Also, XBP1 promotes tumorigenesis by accelerating the ubiquitination and degradation of the tumor suppressor p53, leading to enhanced tumor cell proliferation [93]. Another study has also highlighted the role of XBP1 in promoting tumorigenesis by increasing cholesterol biosynthesis [284]. By inhibiting SREBP2 ubiquitination, XBP1 enhances de novo cholesterol biosynthesis, fostering HCC cell proliferation and tumorigenic potential. These findings emphasize the importance of metabolism changes in the XBP1-mediated tumorigenic effects, highlighting its involvement in tumor cell metabolic reprogramming. On the other hand, ROS stimulates the generation of XBP1, which in turn enhances the expression of HEPN1 by binding to its gene promoter and functioning as a transcriptional activator [95]. Consequently, ROS-induced activation of XBP1 mediates the elevation of HEPN1, inducing cell growth arrest and apoptosis in HCC.
Recent studies have confirmed the involvement of XBP1 in the progression of HCC through microRNA regulation. Exosomal miR-21-5p, which HCC cells release, specifically influences HCC cell development by modulating XBP1’s activity [285]. This modulation promotes the M2 polarization of tumor-associated macrophages, consequently leading to an unfavorable prognostic outcome in HCC patients. Moreover, reduced levels of miR-199 observed in HCC are associated with elevated expression of XBP1 and cyclin D [286]. Both bioinformatics analysis and in vitro experiments suggested that XBP1 is a potential target of miR-199. Furthermore, a negative correlation is found between XBP1 and miR-199 levels. In addition to this miR-199/XBP1/cyclin D axis, XBP1 plays a role in promoting HCC cell proliferation (Table 1). In summary, as facilitated UPR serves a protective role in cancer cell survival and resistance to chemotherapy, targeting the UPR, particularly XBP1, may be a potential therapeutic strategy for cancer treatment (Table 1).
THERAPEUTIC STRATEGIES FOR XBP1 IN LIVER DISEASES
As mentioned above, XBP1 is widely recognized as a pivotal regulator in the incidence and exacerbation of liver diseases, indicating its potential as a significant therapeutic target for liver injury. Therefore, modulating XBP1 activity may contribute to the diagnosis and treatment of liver diseases. A variety of investigational compounds have developed with the potential to selectively modulate the IRE1α–XBP1 pathway, rather than directly targeting XBP1 itself, representing a promising avenue for pharmacological intervention (Table 2) [287]. Among these compounds are inhibitors that target the RNase domain of IRE1α, including toyocamycin, salicylaldehyde, 4μ8C, hydroxyl-aryl-aldehydes, STF-083010, and MKC-3946 [288-293]. The inhibitors have demonstrated efficacy in blocking XBP1 splicing, thereby impeding its downstream transcriptional activity [294-300]. In addition, compounds such as kinase competitor compound 3 have been identified to inhibit XBP1 splicing by selectively targeting the serine/threonine kinase domain of IRE1α, disrupting its function at the ATP-binding site [295,301]. Furthermore, there is an increasing emphasis on investigating the inhibition of IRE1α kinase activity as a potential strategy for managing metabolic defects associated with ER stress. This approach aims to modulate XBP1 activity without inducing ER stress, potentially mitigating adverse effects associated with prolonged ER responses [302]. responses. Conversely, alternative strategies involve the activation of the IRE1α RNase domain through compounds like quercetin, which binds to its Q site, triggering XBP1 splicing [303]. However, it is essential to acknowledge that these compounds are still in the early stages of preclinical testing, and their safety profiles and potential side effects remain to be fully elucidated. Furthermore, while the significance of ER stress, particularly the IRE1α-XBP1 pathway, in liver disease is undeniable, there remains a notable insufficiency of studies confirming the efficacy of these compounds specifically in hepatic disorders. Consequently, there is a need for research across various liver diseases to ascertain the potential therapeutic effects of the compounds. In addition, the complex interplay between ER stress and cellular homeostasis necessitates thorough evaluation to confirm the safety and efficacy of these chemicals before clinical translation. Despite these challenges, the ability to selectively modulate XBP1 activity holds significant therapeutic potential, particularly in diseases where dysregulated ER stress contributes to pathogenesis.
Therefore, with the ongoing advancement of research in this area, it becomes imperative to investigate personalized treatment strategies customized to the individual disease circumstances. This underscores the importance of understanding the intricate mechanisms underlying ER stress and the IRE1α–XBP1 pathway to develop targeted interventions that effectively manage diverse liver diseases while minimizing adverse effects.
CONCLUSIONS AND PERSPECTIVES
The presented data indicates that XBP1 plays critical roles in mediating various target genes related to lipid metabolism, cell survival/death, proliferation, autophagy, cell toxicity, cholesterol metabolism, inflammation, and cancer cell invasion/metastasis, beyond its traditional role as a regulator of ER stress. Additionally, it influences various signaling pathways independent of ER stress, providing partial insights into the pathogenesis of liver diseases. Accumulative data demonstrate that targeting XBP1s could be beneficial for the prevention and treatment of liver diseases. Notably, XBP1s exhibits contrasting roles across different cell types and liver diseases. For instance, while some studies indicate that XBP1s exacerbates liver injury by inducing cell death in DILI, it promotes cancer cell survival, thus exacerbating HCC progression. Furthermore, XBP1s plays a protective role in hepatocytes, while it exacerbates liver disease in macrophages and HSCs. In conclusion, studies on XBP1s present new prospects for understanding the pathogenesis of various liver diseases and formulating effective therapeutic strategies. Although modulating XBP1s activity using specific agonists or antagonists may offer attractive pharmacological strategies for manipulating liver injury, attention should be paid to some contradictory results that are yet to be clarified. Therefore, additional preclinical evidence to validate the efficacy of therapies targeting XBP1s for liver disease is needed.
Notes
Authors’ contribution
JT and YSK: Conceptualization, writing, and editing the manuscript. SGK: Conceptualization, critical revision, and finalization of the manuscript.
Conflicts of Interest
The authors have no conflict of interest to declare.
Acknowledgements
This research was supported by the National Research Foundation (NRF) grant funded by the Korea government (MSIT) (NRF-2022R1A6A3A01086695 to JT; RS-2023-00209701 to YSK; RS-2024-00441114 and NRF-2018R1A5A2023127 to SGK).
Abbreviations
ALOX12
arachidonate 12-lipoxygenase
ALT
alanine aminotransferase
APAP
acetaminophen
ASK1
apoptosis signal-regulating kinase 1
AT2R
angiotensin II type-2 receptor
ATF
activating transcription factor
ATF6N
active N-terminal cytosolic fragment
BMI
body mass index
CDK4
cyclin-dependent kinase 4
C/EBP
CCAAT-enhancer-binding protein
CHOP
C/EBP homologous protein
DILI
drug-induced liver injury
ECM
extracellular matrix
EDEM
ER degradation-enhancing-α-mannosidase-like protein
EMT
epithelial-to-mesenchymal transition
ER
endoplasmic reticulum
ERAD
ER-associated degradation
eIF2α
eukaryotic initiation factor 2α
FoxO1
forkhead box O1
GPX4
glutathione peroxidase 4
GRP78
glucose-regulated protein 78
GSH
glutathione
HBV
hepatitis B virus
HBx
HBV X protein
HCC
hepatocellular carcinoma
HCV
hepatitis C virus
HDL
high-density lipoprotein
HEPN1
hepatocellular carcinoma downregulated 1
HK2
hexokinase II
HNF4α
hepatocyte nuclear factor 4α
HSCs
hepatic stellate cells
IL-1β
interleukin-1β
IRE1α
inositol-requiring enzyme 1 alpha
IRS1
insulin receptor substrate 1
LOXL2
lysyl oxidase-like 2
MASLD
metabolic dysfunction-associated steatotic liver disease
MAPK
mitogen-activated protein kinase
MLKL
mixed-lineage kinase domain-like protein
JNK
c-Jun N-terminal kinase 1
KO
knockout
NAFLD
non-alcoholic fatty liver disease
NAPQI
N-acetyl-p-benzoquinone imine
NCOA3
nuclear receptor coactivator 3
NLRP3
pyrin domain-containing protein 3
NNRTIs
nonnucleoside reverse transcriptase inhibitors
UPR
unfolded protein response
4-PBA
4-phenylbutyrate
PERK
protein kinase RNA-like endoplasmic reticulum kinase
PHLDA3
pleckstrin homology-like domain
PIN1
peptidyl-prolyl cis-trans isomerase NIMA-interacting 1
PI3K
phosphoinositide 3-kinase
PPAR
peroxisome proliferator-activated receptor
RIDD
regulated IRE1α-dependent decay
RIPK
receptor-interacting serine/threonine-protein kinase
ROS
reactive oxygen species
RPE
retinal pigment epithelium
TG
triglyceride
TGF-β
transforming growth factor-β
TNF
tumor necrosis factor
TRAF2
TNF receptor-associated factor 2
VLDL
very low-density lipoprotein
XBP1
X-box binding protein 1