Cells are highly sensitive to conditions that disrupt the environment of the endoplasmic reticulum (ER) or that increase demand on its machinery for synthesis, maturation and transport of secretory cargo. Various physiological and pathophysiological conditions interfere with protein folding processes in the ER, leading to accumulation of misfolded or unfolded proteins, a cellular condition referred to as “ER stress”. Adapting to the nature, intensity, and duration of ER stress, cells launch the unfolded protein response (UPR), a tri-partite intracellular signaling mechanism that triggers translational control and an extensive transcriptional response to balance client protein load with the folding capacity of the ER [1]. The three distinct mammalian UPR signaling pathways are initiated by the ER transmembrane sensors protein kinase RNA-activated (PKR)-like ER kinase (PERK), activating transcription factor 6 (ATF6), and inositol-requiring enzyme 1 (IRE1). Activated PERK phosphorylates the α subunit of eukaryotic initiation factor 2 (eIF2α), effectively down-regulating protein synthesis [2,3,4,5]. Proteolytic processing of ATF6 yields an active transcription factor (ATF6α(N)) that up-regulates expression of ER resident quality control proteins, including chaperones and ER-associated degradation (ERAD) components [6,7,8,9,10]. Upon activation of IRE1, its endoribonucleolytic activity catalyzes an unconventional cytosolic splicing of X-box binding protein 1 (XBP1) mRNA, resulting in a translational frameshift that generates XBP1S, a pro-adaptive, basic-region leucine zipper (bZIP) transcription factor [11,12,13]. If these adaptive mechanisms do not sufficiently recover ER homeostasis, the UPR can commit the damaged cell to death by apoptosis [14].

XBP1 is subject to transcriptional, post-transcriptional and post-translational control mechanisms [15,16,17,18,19,20,21,22,23,24,25,26], indicating that the activity of this crucial UPR transcription factor is carefully balanced. Indeed, XBP1 has tissue-distinct roles during the UPR in different physiological processes and diseases. XBP1, a member of the CREB/ATF (cyclic AMP response element binding protein/activating transcription factor), family of transcription factors, was originally identified as a protein that binds to the HLA (human leukocyte antigen) DRα promoter [27]. Since then, studies have unveiled multiple roles for XBP1 in diverse processes including secretory function, the inflammatory response, lipid metabolism and glucose homeostasis [28]. XBP1 deficiency impairs normal development and function of antibody-producing plasma cells [29] as well as pancreatic, liver and salivary gland cells [30,31,32,33]. XBP1 also plays a role in the development and survival of dendritic cells [34] and in the response of macrophages to certain Toll-like receptor agonists [35]. Defects in the XBP1 gene and/or protein have also been linked to critically important problems in human health such as ulcerative colitis/Crohn’s disease [36], insulin resistance and type 2 diabetes [37].

On the whole, these observations highlight the critical role of XBP1 as a pro-adaptive transcription factor in various physiological processes. Though genetically modified mice have allowed key discoveries of XBP1 function, a better understanding is needed concerning the mechanisms employed by cells to specifically moderate the activity of XBP1.

IRE1 is a bifunctional, single-pass ER transmembrane Ser/Thr kinase and endoribonuclease. This UPR transducer was originally identified in yeast [38,39,40] and is very similar across eukaryotic species. Mammals express two isoforms of IRE1, α and β. IRE1α is ubiquitously expressed [41], whereas IRE1β is limited to the gut epithelium [42]. Here, our discussion will focus on IRE1α since it is expressed in all cell types. In cells experiencing ER stress, active IRE1α catalyzes the cytosolic removal of a 26 nucleotide (nt) intron from XBP1 pre-mRNA (XBP1u) (Figure 1b). This unconventional splicing event results in a translational frameshift that joins the two open reading frames (ORFs) encoding a DNA binding domain (DBD) and a transcriptional activation domain (AD) [11,12,13] (Figure 1a). Translation of spliced XBP1 mRNA (XBP1s) generates XBP1S, a bZIP transcription factor that binds to and activates two cis-acting promoter motifs, the ER stress response element (ERSE) and unfolded protein response element (UPRE) [13]. In addition to activating its own transcription, XBP1S induces genes encoding: factors involved in translocation of nascent polypeptides into the ER (SRP54, SEC61A, SEC61G and TRAM1); ER chaperones (ERdj4, ERdj5, HEDJ, GRP58 and PDI-P5), ERAD components (EDEM, OS9, HERP, and p58^IPK); factors involved in vesicular transport (SEC23B and SEC24C)) [43,44]. More recently, XBP1S was shown to up-regulate expression of a microribonucleic acid (miRNA), miR-346, that targets the human antigen peptide transporter 1 (TAP1) mRNA [45].

Complete deletion of XBP1 in mice results in embryonic lethality at ~13 weeks gestation [33]. Rescue of embryonic lethality by targeting an XBP1 transgene selectively to hepatocytes led to early post-natal lethality via activation of ER stress-mediated proapoptotic pathways [30]. Specifically, the phenotype consisted of weak expression of ER chaperone genes and poorly developed ER in pancreatic and salivary gland acinar cells, and this correlated with impaired production of pancreatic digestive enzymes [30]. Similarly, XBP1S is vital for ER expansion and induction of high-rate immunoglobulin synthesis during plasma cell differentiation [44,46].

Recent reports indicate that post-transcriptional mechanisms influence the fate of XBP1 mRNA. Regulatory mechanisms implicated include unique localization of XBP1 mRNA at the ER membrane and translational pausing that facilitates IRE1α-dependent splicing. In addition, XBP1 mRNA is targeted by miRNA.

IRE1α-mediated splicing of XBP1 mRNA occurs in the cytosol [47,48], in contrast to conventional mRNA splicing that takes place in the nucleus. Only recently have discoveries shed light on underlying mechanisms that orchestrate the localization of XBP1 mRNA within proximity of IRE1α at the ER membrane (Figure 1b). A novel observation of cellular localization of total XBP1 mRNA was reported in a study examining mRNA partitioning and translation in the ER and cytosolic compartments during the UPR [49]. Surprisingly, total XBP1 mRNA was found to be predominantly membrane associated, although its protein products, XBP1U and XBP1S, are soluble [49]. A subsequent study confirmed XBP1u mRNA association with the ER membrane, but reported XBP1s mRNA re-distribution to cytosolic compartments for translation [24]. Yanagitani and colleagues [24] further implicated a conserved hydrophobic region (HR2) near the carboxyl-terminus of XBP1U as an ER membrane association domain (Figure 1a, b). This group speculated that the HR2 of nascent XBP1U polypeptide chains might cotranslationally recruit XBP1u mRNA to the ER membrane as part of a mRNA-ribosome-nascent chain complex (R-RNC) [24] (Figure 1b). In addition, they recently reported that translation of the XBP1u mRNA transiently pauses to stabilize the R-RNC complex [25]. This entire process is dependent on XBP1U sequences that are highly similar across multiple species, specifically the HR2 and an additional region near the carboxyl-terminus [25] (Figure 1a).

While the Stephens [49] and Yanagitani [24,25] studies agree that XBP1u mRNA localizes at the ER membrane, ambiguity remains as to whether XBP1 mRNA shifts from the ER membrane to the cytosol after IRE1α-mediated splicing has occurred. Notably, the two studies were conducted in different cell lines under different strengths of ER stress inducers. Importantly, the HR2 is located within the 3’ segment of the XBP1u coding region where the translational frame is altered by IRE1α−mediated splicing, resulting in XBP1S which lacks the HR2 [24]. Finally, studies of XBP1-deficient mice have revealed hyperactivation of IRE1α associated with splicing of a truncated XBP1 mRNA in liver and intestinal tissue [32,36], indicating that expression of XBP1U is not required for splicing. Perhaps, the sub-cellular distribution of total XBP1 mRNA is determined in a tissue- and/or stress-specific fashion. Further studies are required to delineate a full understanding of these mechanisms and their relevance in vivo.

More recently, we and others have reported miRNA-mediated regulation of XBP1 expression via the 3’ untranslated region (UTR) of XBP1 mRNA [15,50] (Figure 1b). miRNA are a class of endogenous, non-coding, single-stranded RNAs ~22 nts long that typically function as post-transcriptional repressors of gene expression [51]. Although the specific biological functions of miRNA in ER stress and the UPR remain largely unknown, a few ER stress-inducible miRNAs have been identified [15,45,52].

Our group identified a miRNA, miR-30c-2* (since designated miR-30c-2-3p), that targets a single site in the 3′-UTR of XBP1 mRNA (Figure 1b). Over-expressing miR-30c-2* reduced the levels of XBP1 and its target genes in stressed cells, whereas inhibiting miR-30c-2* activity had the opposite effect, boosting XBP1 levels and promoting cell survival [15]. Induction of ER stress by subjecting human and mouse cell lines to treatment with tunicamycin (Tm), an inhibitor of N-linked glycosylation, triggers the UPR and activates IRE1α, thereby generating XBP1S. Interestingly, miR-30c-2* is up-regulated during the Tm-induced UPR, concomitant with XBP1, suggesting that this miRNA might affect XBP1S expression levels as the UPR proceeds [15]. Our study further showed that up-regulation of miR-30c-2* is dependent on another key signaling component of the UPR, the protein kinase PERK. The mechanism involves the transcription factor nuclear factor-κB (NF-κB) and its interaction with a NF-κB enhancer motif upstream of the miR-30c-2* genes [15]. In the UPR, NF-κB is activated downstream of PERK [53,54]. Importantly, a separate study revealed that phosphorylation of eIF2α is required for optimal induction of XBP1S by a mechanism involving both XBP1 mRNA stabilization and translation inhibition [18]. Therefore, emerging evidence indicates that regulatory cross-talk between the IRE1/XBP1 and PERK pathways influences the strength of XBP1S induction.

Another miRNA, miR-214, was recently implicated as a negative regulator of XBP1 expression in hepatocellular carcinomas (HCC) [50] (Figure 1b). In HCC, miR-214 expression was down-regulated in response to pharmacologic ER stress-inducing agents, hypoxia and lipopolysaccharide stimulation, suggesting that miR-214 represses XBP1 expression until the UPR is activated. Down-regulation of miR-214 correlated with NF-κB activation, whereby NF-κB exerted a negative regulatory affect on expression of the miR-199a/214 gene cluster [50]. However, a direct binding site through which NF-κB mediates this effect was not demonstrated, raising the possibility of an indirect regulatory mechanism.

Together, these findings break the ice for exploration into potential roles for miRNA-mediated regulation of XBP1. However, the in vivo relevance of these mechanisms remains unexplored. It will be interesting to investigate the degree to which miRNA regulate XBP1 translation and affect gene expression, cell function and cell fate in various physiological and pathophysiological conditions involving XBP1, such as plasma cell differentiation [46], macrophage activation [35] and tumor cell survival [55].

A series of recent discoveries have implicated XBP1S as a target for a variety of post-translational modifications, including phosphorylation, SUMOylation, acetylation and deacetylation (Figure 1c). These modifications affect XBP1S turnover, sub-cellular localization and transcriptional activity.

A number of binding partners have recently been identified for the XBP1 proteins, including transcription factors and other types of regulatory proteins. Such protein-protein interactions have been implicated in regulating the turnover, nucleocytoplasmic shuttling and transcriptional activity of XBP1S (Figure 2).

Whether via transcriptional, post-transcriptional, co-translational, post-translational modifications and/or protein-protein interactions (Table 1), examples of mechanisms that can finely adjust XBP1S activity are increasingly evident. In the complexity of UPR signal transduction, the emergence of an IRE1/XBP1 signaling pathway that is subject to such diverse controls leads one to postulate that we have just begun to understand the breadth and importance of XBP1 regulatory activities. The combinatorial effects of these molecular mechanisms may vary widely depending on the tissue-type undergoing ER stress. For example, in the stressful conditions of the tumor microenvironment, different types of ER stressors and/or varying degrees of the same ER stress can be present, such as hypoxia and nutrient deprivation.

An obvious, albeit challenging, direction for this exciting field is to establish the degree to which XBP1 regulatory mechanisms occur and function in vivo. Moving forward, considerable attention should be given to interpreting how these findings translate to our understanding of XBP1 related pathologies and the therapeutic potential of exploiting these mechanisms in certain disease states such as inflammatory disorders [63] and cancer [64]. Further defining the dynamic interplay of XBP1 and the molecular mechanisms that modulate its activity will provide a deeper understanding of the fundamental roles of the UPR, specifically IRE1/XBP1 signaling, in cell physiology.