miR-204 Targets PERK and Regulates UPR Signaling and β-Cell Apoptosis

Abstract

Endoplasmic reticulum (ER) stress plays an important role in the pathogenesis of diabetes and the associated β-cell apoptosis. Although microRNAs (miRNAs) have been widely studied in various diseases including diabetes, the role of miRNAs in ER stress and β-cell apoptosis has only started to be elucidated. We recently showed that diabetes increases β-cell miR-204 and have now discovered that miR-204 directly targets the 3′untranslated region of protein kinase R-like ER kinase (PERK), 1 of the 3 ER transmembrane sensors and a key factor of the unfolded protein response (UPR). In addition, by using primary human islets, mouse islets, and INS-1 β-cells, we found that miR-204 decreased PERK expression as well as its downstream factors, activating transcription factor 4 and CCAAT enhancer-binding protein homologous protein, whereas it had no effect on the other 2 ER transmembrane sensors, activating transcription factor 6 and inositol-requiring enzyme-1α. Interestingly, we discovered that miR-204 also inhibited PERK signaling in the context of ER stress, and this exacerbated ER stress-induced β-cell apoptosis. This effect could be mimicked by PERK inhibitors supporting the notion that the miR-204-mediated inhibition of PERK and UPR signaling was conferring these detrimental effects on cell survival. Taken together, we have identified PERK as a novel target of miR-204 and show that miR-204 inhibits PERK signaling and increases ER stress-induced cell death, revealing for the first time a link between this miRNA and UPR.

Loss of functional pancreatic β-cells is a key factor of both type 1 and type 2 diabetes (1, 2), and endoplasmic reticulum (ER) stress has been suggested to be involved in this process (3). In pancreatic β-cells, the ER is the cellular organelle for insulin folding and maturation. Under certain conditions, such as high glucose, cytokine exposure, and insulin mutations, unfolded and misfolded insulin accumulates in the ER and causes ER stress (3). In response to this ER stress, an unfolded protein response (UPR) is activated through 3 ER transmembrane proteins, protein kinase R-like ER kinase (PERK), inositol-requiring enzyme-1 (IRE1α), and activating transcription factor 6 (ATF6), to relieve the stress by reducing the protein load and inducing molecular chaperon expression (4). All 3 ER transmembrane factors have been shown to play important roles in β-cell function and survival (3). In particular, PERK, also named eukaryotic translation initiation factor 2α kinase 3, is mutated in patients with Wolcott-Rallison syndrome, which is a rare, autosomal recessive disorder characterized by permanent neonatal or early infancy insulin-dependent diabetes (5, 6). In mice, deficiency of germ line PERK causes diabetes by increasing β-cell death (7, 8).

In the past decade, microRNAs (miRNAs) have been found to play important roles in various diseases including diabetes (9), but their function in pancreatic β-cells is still largely unknown. miRNAs are single-stranded approximately 22-nt small RNAs that bind to complementary sequences classically located in the 3′untranslated region (UTR) of target genes and cause mRNA degradation and/or translational repression (10). Because binding is primarily conferred by the first 7- to 8-nucleotide seed sequence of the miRNA, 1 miRNA can typically target multiple genes due to complementarity adding to the complexity of the signaling network (10). Although a variety of programs exist to help predict putative targets genes based on the presence of seed sequence binding sites (11), identifying bona fide target genes has remained a challenging yet important task especially in order to better understand the biological and pathophysiological role of any given miRNA.

miR-204 is highly expressed in human pancreatic islets and enriched in β-cells as opposed to α-cells (12, 13), suggesting that it might play an important role in β-cell biology. Indeed, we recently discovered that diabetes induces islet miR-204 expression and that miR-204 in turn targets the insulin transcription factor v-maf avian musculoaponeurotic fibrosarcoma oncogene homolog A (MAFA) and thereby inhibits insulin transcription (14). However, very little is known about other potential miR-204 targets and their role in β-cell biology and the aim of the present study was therefore to address this question. Using prediction software as well as extensive experimental testing we discovered that miR-204 also targets the UPR protein PERK and regulates PERK signaling under normal and ER stress conditions thereby modulating ER stress-induced β-cell apoptosis.

Materials and Methods

Tissue culture

INS-1 β-cells were grown in RPMI 1640 medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, 1mM sodium pyruvate, 2mM L-glutamine, 10mM HEPES, and 0.05mM 2-mercaptoethanol and were last authenticated by insulin expression in November 2015. To inhibit PERK activity, INS-1 cells were preincubated for 1 hour and treated for 5 hours with 2 different established PERK inhibitors (PERK inhibitor 1, GSK2606414 [100nM] or PERK inhibitor 2, GSK2656157 [100nM]; EMD Millipore) (15, 16). To induce ER stress, INS-1 cells were treated with 0.5μM thapsigargin (TG) for 5 hours. HEK293 cells were grown in DMEM (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Mouse islets were isolated from C57BL/6 mice by collagenase digestion as described previously (17). All mouse studies were approved by the University of Alabama at Birmingham Animal Care and Use Committee and complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Human islets were obtained from the Integrated Islet Distribution Program and always islets from the same donor were used as control and at least islets from 3 different donors were used per experiment.

Plasmid construction, transfection, and luciferase assays

The wild-type human PERK 3′UTR region containing the miR-204-binding site was amplified from human genomic DNA by using the primers listed in Supplemental Table 1. To generate the human PERK 3′UTR with mutated miR-204-binding site, mutations were introduced by PCR and primers listed in Supplemental Table 1. PCR products were subcloned into the SpeI and PmeI sites of the pMIR-REPORT Luciferase vector (Thermo Fisher Scientific) yielding the human PERK-WT and PERK-M 3′UTR luciferase reporter plasmids. All plasmids were confirmed by sequencing. For luciferase assays, HEK293 cells were plated in 12-well plates and grown overnight to approximately 60% confluence. Cells were transfected with hPERK-WT or hPERK-M together with hsa-miR-204 mirVana miRNA mimic (pre-miR-204) or pre-miR negative control 2 (Thermo Fisher Scientific) using DharmaFECT Duo transfection reagent (Thermo Fisher Scientific). To control for transfection efficiency, cells were cotransfected with pRL-TK (Promega) control plasmid expressing Renilla luciferase, and 24 hours after transfection, firefly as well as Renilla luciferase activity were determined using the Dual Luciferase Assay kit (Promega). For mRNA and protein analysis, INS-1 cells were plated in 6-well plates and grown overnight to approximately 60% confluence. Human islets (500 per tube) or mouse islets (100 per tube) were gently dispersed by incubation for 5 minutes in 200 μL of 0.05% Trypsin-EDTA (Thermo Fisher Scientific) at 37°C and washed and resuspended in culture medium. Cells were transfected with pre-miR-204 or pre-miR negative control 2 (Thermo Fisher Scientific), or with miRIDIAN hairpin inhibitor miR-204 (anti-miR-204) or miRIDIAN microRNA hairpin inhibitor negative control 1 (GE Dharmacon) at a final concentration of 25nM using DharmaFECT1 transfection reagent (GE Dharmacon) and harvested 72 hours after transfection.

Quantitative real-time RT-PCR

Total RNA was extracted using miRNeasy kit (QIAGEN) according to the manufacturer's instructions. RNA (1 μg) was reverse transcribed to cDNA using the first strand cDNA synthesis kit (Roche). Quantitative real-time PCR was performed on a LightCycler 480 System (Roche) using SYBR Green (Thermo Fisher Scientific). All primers used are listed in Supplemental Table 1. All samples were corrected for the 18S ribosomal subunit run as an internal standard.

Western blotting

Protein extracts from INS-1 were prepared using lysis buffer containing HEPES (50mM), Nonidet P-40 (10%), sodium fluoride (100mM), sodium pyrophosphate (10mM), EDTA (4mM), phenylmethanesulphonyl fluoride (1mM), leupeptin (2μM), activated sodium orthovanadate (2mM), and okadaic acid (100nM). Protein concentrations were measured by Pierce BCA protein assay (Thermo Fisher Scientific), and equal amounts of protein were loaded. Bands were visualized by enhanced chemiluminescence plus (GE Healthcare) and quantified by ImageQuant TL software (GE Healthcare). The following antibodies were used: rabbit anti-PERK IgG (3192; Cell Signaling), mouse anti-ATF6 IgG (ab11909; Abcam), rabbit anti-IRE1α IgG (3294; Cell Signaling), rabbit anti-phospho-PERK IgG (3179; Cell Signaling), rabbit anti-ATF4 IgG (11815; Cell Signaling), mouse anti-CCAAT enhancer-binding protein homologous protein (CHOP) IgG (2895; Cell Signaling), rabbit anti-Caspase-3 (14220; Cell Signaling), mouse antiactin IgG (ab3280; Abcam), goat antirabbit IgG (sc-2004; Santa Cruz Biotechnology, Inc), and goat antimouse IgG (sc-2005; Santa Cruz Biotechnology, Inc).

Caspase activity assays

Apoptosis was assessed by using Caspase-Glo 3/7 Assay kit (Promega) according to the manufacturer's instructions. DNA content was measured for normalization by using Quant-iT PicoGreen dsDNA Assay kit (Thermo Fisher Scientific) according to the manufacturer's instructions.

Statistical analysis

Student's t tests were used to calculate the significance of a difference between 2 groups. For datasets of more than 2 groups and to analyze changes overtime, we performed one-way ANOVA calculations.

Results

miR-204 targets and inhibits β-cell PERK

miR-204 is the most highly enriched microRNA in pancreatic β-cells as compared with α-cells (12, 13), and its expression is elevated in diabetes (14), suggesting that it may have important functions in β-cell biology and pathology. However, aside from our recent discovery that miR-204 inhibited insulin transcription by targeting the insulin transcription factor MAFA, the targets and functions of miR-204 have remained largely elusive. Using miRWalk prediction, we now have found that PERK (and 1 of the 3 ER transmembrane sensors), is a putative target of miR-204. Indeed, we identified a conserved miR-204-binding site in the 3′UTR of human and rat PERK (Figure 1A). To confirm the prediction, we generated reporter constructs by inserting the wild-type human PERK 3′UTR or the human PERK 3′UTR with a mutated miR-204-binding site downstream of the luciferase gene, and assessed miR-204-directed inhibition of reporter gene expression. miR-204 significantly decreased luciferase activity through the wild-type PERK 3′UTR, whereas no reduction was found with the mutant PERK 3′UTR (Figure 1B), indicating that PERK is a direct target of miR-204. To further determine whether miR-204 can reduce PERK expression, we overexpressed miR-204 precursors in INS-1 β-cells and assessed the miR-204 effects on PERK mRNA and protein levels. We found that miR-204 overexpression significantly reduced both PERK mRNA (Figure 2A) and protein levels (Figure 2, B and C). To confirm the effect of miR-204 on PERK expression in primary islets, we overexpressed miR-204 precursors in primary human and mouse islets. The results showed that miR-204 overexpression significantly reduced PERK mRNA levels in both human islets (Figure 2D) and mouse islets (Figure 2E). In addition, inhibition of endogenous miR-204 by knockdown with anti-miR-204 significantly increased PERK mRNA levels in INS-1 cell (Figure 2F). These results suggest that miR-204 targets and inhibits PERK. On the other hand, miR-204 had no effect on the mRNA expression or protein levels of the other 2 ER stress sensors, IRE1α (Figure 3, A–C) and ATF6 (Figure 3, D–F), consistent with the lack of a miR-204-binding site in their 3′UTRs and indicating that the effect of miR-204 on PERK was specific.

Figure 1.

PERK as a target of miR-204. A, Alignment of miR-204 seed sequence (arrow) and its binding site in wild-type human (hPERK-WT) and rat PERK 3′UTR (rPERK-WT) (bold), as well as sequence of the mutated human PERK 3′UTR (hPERK-M) (gray). B, miR-204-directed repression of the luciferase reporter gene bearing hPERK-WT or hPERK-M 3′UTR segments as assessed 24 hours after cotransfecting HEK293 cells with miR-204 precursor (pre-miR-204) or negative control and the wild-type or mutant reporter plasmids. All data represent mean ± SEM of 3 independent experiments. N.S., not significant.

Figure 2.

miR-204 effects on PERK expression. PERK mRNA (A) and PERK protein levels (B and C) as assessed by qRT-PCR and immunoblotting in INS-1 β-cells transfected with pre-miR-204 or negative control. Human islets (D) and mouse islets (E) were transfected with pre-miR-204 or negative control, and PERK expression was assessed by qRT-PCR. F, PERK mRNA expression in INS-1 cells transfected with miR-204 inhibitor (anti-miR-204) or negative control. All data represent mean ± SEM of 3 independent experiments, and a representative immunoblot is shown. N.S., not significant.

Figure 3.

miR-204 effects on IRE1α and ATF6 expression. INS-1 β-cells were transfected with pre-miR-204 or negative control and (A) IRE1α mRNA and (B and C) IRE1α protein levels as well as (D) ATF6 mRNA and (E and F) ATF6 protein levels were assessed by qRT-PCR and immunoblotting. All data represent mean ± SEM of 3 independent experiments, and a representative immunoblot is shown. N.S., not significant.

miR-204 inhibits PERK function and downstream signaling

Upon ER stress, PERK is activated by oligomerization and autophosphorylation (18). Activated PERK induces the expression of the transcription factor ATF4, which in turn activates the transcription factor and proapoptotic protein CHOP (4). To assess whether miR-204 might also regulate this downstream signaling pathway of PERK, we investigated the miR-204 effects on ATF4 and CHOP. miR-204 overexpression significantly decreased ATF4 expression in INS-1 cells, primary human islets and mouse islets (Figure 4, A–C). In contrast, inhibition of endogenous miR-204 significantly increased ATF4 expression in INS-1 cells (Figure 4D). Consistently, miR-204 overexpression also significantly decreased CHOP mRNA levels in INS-1 cells, primary human islets and mouse islets (Figure 4, E–G), whereas inhibition of endogenous miR-204 significantly increased CHOP mRNA levels in INS-1 cells (Figure 4H). These results suggest that miR-204 regulates not only PERK expression but also its function and downstream signaling factors.

Figure 4.

Effects of miR-204 on β-cell expression of the PERK downstream factors ATF4 and CHOP. ATF4 mRNA expression was measured by qRT-PCR in (A) INS-1 β-cells, (B) human islets, and (C) mouse islets transfected with pre-miR-204 or negative control, as well as in (D) INS-1 cells transfected with anti-miR-204 or negative control. CHOP mRNA expression in (E) INS-1 β-cells, (F) human islets, and (G) mouse islets transfected with pre-miR-204 or negative control, and (H) INS-1 cells transfected with anti-miR-204 or negative control. All data are shown as the mean ± SEM of 3 independent experiments.

miR-204 inhibits ER stress-induced PERK signaling

To further address the question whether miR-204 could also regulate PERK and its downstream factors under ER stress conditions, we overexpressed miR-204 in INS-1 cells treated with TG. TG treatment dramatically increased both PERK mRNA and phosphorylated protein levels, consistent with the expected induction of ER stress by TG and the resulting activation of the UPR, whereas miR-204 overexpression significantly inhibited these effects (Figure 5, A and B). Consistently, TG treatment also clearly induced ATF4 mRNA and protein levels and this effect was significantly blunted by miR-204 overexpression (Figure 5, C and D). Furthermore, CHOP mRNA and protein levels were significantly elevated in response to TG treatment and miR-204 overexpression significantly reversed or even abolished these effects (Figure 5, E and F). These results indicate that miR-204 not only regulates basal PERK expression but also inhibits ER stress-induced PERK signaling.

Figure 5.

miR-204 effects on ER stress-induced PERK signaling. INS-1 β-cells transfected with negative control (control) or with pre-miR-204 (miR-204) were treated with DMSO or TG 0.5μM for 5 hours to induce ER-stress and PERK mRNA (A) and protein levels (B), ATF4 mRNA (C), and protein levels (D), and CHOP mRNA (E) and protein levels (F) were assessed. All data are shown as the mean ± SEM of 3 independent experiments, and representative immunoblots are shown.

Because we previously have shown that miR-204 inhibits insulin expression by targeting MAFA (14), we also investigated whether miR-204 would still do so in the context of ER stress. Although TG treatment already led to a pronounced decrease in the expression of both, rat insulin1 and insulin2 genes, miR-204 overexpression resulted in an additional significant inhibition of insulin expression (Figure 6, A and B), and this was associated with a significant decrease in MAFA (Figure 6C). These findings are in alignment with the observation that ER stress leads to decreased insulin gene transcription (19) and demonstrate that miR-204 maintains its inhibitory effects on MAFA and insulin production even under conditions of ER stress and already reduced insulin transcription.

Figure 6.

Effects of miR-204 on insulin and MAFA expression under ER stress. INS-1 β-cells transfected with negative control (control) or with pre-miR-204 (miR-204) were treated with DMSO or TG 0.5μM for 5 hours to induce ER stress and insulin1 (Ins1) (A), insulin2 (Ins2) (B) and MAFA (C) gene expression was assessed by qRT-PCR. All data are shown as the mean ± SEM of 3 independent experiments. N.S., not significant.

miR-204-mediated PERK inhibition exacerbates ER stress-induced β-cell apoptosis

ER stress is a major stimulus for β-cell apoptosis (3, 20, 21), raising the possibility that, by regulating ER stress signaling, miR-204 might also regulate ER stress-induced β-cell death. To test this possibility, INS-1 β-cells were exposed to TG-induced ER stress in the presence or absence of miR-204 overexpression. As expected, we observed a significant increase in cleaved-caspase-3 in response to TG treatment. Interestingly, miR-204 overexpression resulted in an even further increase in TG-induced apoptosis (Figure 7A). In contrast, under basal conditions and in the absence of ER stress, miR-204 overexpression did not alter β-cell apoptosis (Figure 7B), consistent with our previous observations (14).

Figure 7.

Effects of miR-204 and PERK inhibition on ER stress-induced β-cell apoptosis. A, Apoptosis as assessed by cleaved-Caspase-3 levels in INS-1 β-cells transfected with pre-miR-204 or negative control and treated with DMSO control or TG (0.5μM for 5 h) to induce ER-stress. B, Apoptosis in INS-1 cells transfected with pre-miR-204 or negative control in the absences of ER-stress induction. C, INS-1 β-cells were treated with TG (0.5μM) or TG together with the PERK inhibitors (PI1) or (PI2) (100nM) for 5 hours, and apoptosis was assessed by Caspase activity. D, Apoptosis in INS-1 cells treated with the PERK inhibitors PI1 or PI2 in the absence of ER-stress. All data are shown as the mean ± SEM of 3 independent experiments and a representative immunoblot is shown. N.S., not significant.

As part of the UPR, activated PERK not only induces ATF4 and CHOP, but also phosphorylates and inactivates the translation initiation factor eIF2α, leading to reduced protein translation and protein loading into the ER and thereby helps alleviate ER stress. Moreover, deficiency of PERK has been shown to cause β-cell death and lead to diabetes in humans and mice (5, 7), suggesting that the net effects of PERK inhibition are detrimental to β-cell survival. We therefore hypothesized that the observed exacerbation of β-cell apoptosis might have been caused by miR-204-mediated PERK inhibition. If so, one would expect that PERK inhibitors would mimic the miR-204 effects and this is exactly what we observed when we treated INS-1 β-cells with 2 established PERK inhibitors (15, 16). Both PERK inhibitors led to a significant additional increase in TG-induced apoptosis (Figure 7C) very similar to the results with miR-204 overexpression. However, neither of the PERK inhibitors had again any effect on apoptosis under normal conditions (Figure 7D). These results suggested that targeting and inhibition of PERK by miR-204 was in fact conferring the observed miR-204-mediated exacerbation of ER stress-induced β-cell apoptosis.

Discussion

The results of this study reveal for the first time that the ER transmembrane sensor protein PERK is posttranscriptionally regulated by a microRNA and demonstrate that miR-204 directly targets and inhibits PERK and its downstream signaling pathway and thereby exacerbates ER stress-induced β-cell apoptosis.

So far, PERK has been known to be primarily regulated by autophosphorylation upon ER stress (4). However, how the expression of PERK mRNA and protein is regulated has remained largely elusive. We now report for the first time that PERK expression is regulated by a microRNA, miR-204, which is highly enriched in pancreatic β-cells and directly targets the PERK 3′UTR. Moreover, results of the current studies show that miR-204-mediated inhibition of PERK occurs in rat INS-1 β-cells as well as in primary mouse and human islets and is specific to PERK, as other ER stress sensors such as ATF6 and IRE1α are not affected by miR-204. Importantly, miR-204 was able to inhibit PERK expression and UPR signaling even in the context of ER stress.

Interestingly, we recently reported that miR-204 also targets MAFA and thereby inhibits β-cell insulin transcription and production (14), and we have now found that miR-204 continues to do so in the context of ER stress. This in turn would also be expected to reduce the ER work load and help prevent activation of the UPR. Together with the current results in terms of PERK, this suggests that the physiological role of miR-204 is to prevent excessive insulin production and prolonged UPR activation. This idea is also in alignment with the notion that the evolutionary importance of microRNAs is based on their ability to buffer gene expression (22), eg, in this case targeting MAFA and PERK, respectively. Under normal conditions, this would be beneficial and indeed even transient overexpression of miR-204 did not induce β-cell apoptosis in the absence of ER stress. However, prolonged miR-204 overexpression as observed in diabetes (14) or in the context of ER stress would be expected to be harmful to β-cells, because it would cause insufficient insulin production and inadequate PERK activation resulting in β-cell dysfunction and increased susceptibility to ER stress. In fact, this is exactly what we observed earlier in regard to β-cell dysfunction (14) and in the current studies in terms of ER stress-induced β-cell apoptosis.

PERK is highly expressed in pancreatic β-cells and plays an important role in β-cell biology (7, 23, 24). As a key ER transmembrane factor responsible for induction of the UPR, PERK activation is critical for the compensatory cell response to ER stress, even though it is also associated with induction of proapoptotic CHOP (4). On the other hand, CHOP acts as a dominant negative inhibitor of CCAAT enhancer-binging protein (C/EBP) and C/EBP in turn has been shown to decrease β-cell mass, whereas ablation of C/EBP increased β-cell survival (25) raising the possibility that by inhibiting C/EBP, CHOP may also have some beneficial effects. In any case, the miR-204-mediated detrimental effects associated with the overall inhibition of the compensatory PERK signaling pathway and its more protective members seem to outweigh any potential positive effects. Of note, germ line PERK mutations are associated with Wolcott-Rallison syndrome, which is characterized by severe diabetes and β-cell loss in humans (5, 6) and mice (7, 8). Pancreas-specific PERK knockout mouse models were also found to exhibit hyperglycemia and low β-cell mass and PERK was shown to be required for fetal and neonatal β-cell development and differentiation (23). Interestingly, more recent studies demonstrated that inducible, global as well as β-cells-specific deletion of PERK in adult mice also resulted in diabetes, β-cell ER distension and decreased β-cell number by apoptosis (24). These results suggest that PERK expression is essential for β-cell development, integrity and survival and our current findings that by targeting and inhibiting PERK miR-204 increased ER stress-induced β-cell apoptosis are very much in alignment with this notion.

PERK has also been shown to be involved in inflammation by playing a role in nuclear factor κB-mediated transcription (26) and in ER stress-induced expression of thioredoxin-interacting protein (27, 28), which in turn contributes to inflammasome activation (29). By targeting PERK, miR-204 would therefore also be expected to at least indirectly be involved in inflammation. In fact, inhibition of PERK and UPR has been shown to potentiate nuclear factor κB-regulated gene expression and cytokine-induced β-cell death (30), very similar to the exacerbation of β-cell apoptosis we observed in response to miR-204 overexpression. In addition, we have previously shown that miR-204, itself is induced by thioredoxin-interacting protein (14). Although future studies will have to elucidate the exact role miR-204 plays in these different pathways, it seems that this microRNA is well positioned to modulate the important cross talk between inflammation and ER stress.

β-Cell dysfunction and apoptosis are key features of diabetes and we previously found that miR-204 expression is increased in diabetic islets (14). Our previous results further showed that miR-204 has detrimental effects on β-cell function by targeting the insulin transcription factor MAFA and inhibiting insulin production (14), and we now show that these effects are maintained under ER stress. Interestingly, our present study further demonstrates that miR-204 also has a detrimental effect on β-cell survival by targeting PERK and increasing ER stress-induced apoptosis. Taken together, this suggests that lowering miR-204 levels may have beneficial effects on both β-cell survival and β-cell function in the context of diabetes.

In summary, we have found that miR-204 targets and inhibits PERK and exacerbates ER stress-induced β-cell apoptosis, for the first time revealing a link between miR-204 and UPR signaling. The results thereby also shed new light on the role of this microRNA in β-cell biology and ER stress signaling.