Activation of pH-Sensing Receptor OGR1 (GPR68) Induces ER Stress Via the IRE1α/JNK Pathway in an Intestinal Epithelial Cell Model

Proton-sensing ovarian cancer G-protein coupled receptor (OGR1) plays an important role in pH homeostasis. Acidosis occurs at sites of intestinal inflammation and can induce endoplasmic reticulum (ER) stress and the unfolded protein response (UPR), an evolutionary mechanism that enables cells to cope with stressful conditions. ER stress activates autophagy, and both play important roles in gut homeostasis and contribute to the pathogenesis of inflammatory bowel disease (IBD). Using a human intestinal epithelial cell model, we investigated whether our previously observed protective effects of OGR1 deficiency in experimental colitis are associated with a differential regulation of ER stress, the UPR and autophagy. Caco-2 cells stably overexpressing OGR1 were subjected to an acidic pH shift. pH-dependent OGR1-mediated signalling led to a significant upregulation in the ER stress markers, binding immunoglobulin protein (BiP) and phospho-inositol required 1α (IRE1α), which was reversed by a novel OGR1 inhibitor and a c-Jun N-terminal kinase (JNK) inhibitor. Proton-activated OGR1-mediated signalling failed to induce apoptosis, but triggered accumulation of total microtubule-associated protein 1 A/1B-light chain 3, suggesting blockage of late stage autophagy. Our results show novel functions for OGR1 in the regulation of ER stress through the IRE1α-JNK signalling pathway, as well as blockage of autophagosomal degradation. OGR1 inhibition might represent a novel therapeutic approach in IBD.


Results
OGR1 induces ER stress under acidic conditions. In order to investigate the role of the pH-sensing OGR1 receptor in the induction of ER stress, OGR1-overexpressing Caco-2 and VC Caco-2 cells, were subjected to an acidic pH shift for 24 h. The stress inducer tunicamycin induced protein expression of the ER stress marker BiP in a dose dependent manner in VC Caco-2 cells and Caco-2 cells overexpressing OGR1 (  1C) and p-IRE1α to total IRE1α (Fig. 1D) is presented. BiP mRNA expression also significantly increased under acidic conditions in Caco-2 overexpressing OGR1 compared to VC cells (Fig. 1E). Interestingly, at acidic pH the expression of BiP and phosphorylation of IRE1α were markedly reduced in OGR1-overexpressing Caco-2 cells in the presence of the OGR1 inhibitor ( Fig. 1F and Supplementary  Figure 3), suggesting that ER stress is induced by proton-activated OGR1 signalling. In OGR1 overexpressing Caco-2 cells, pH-dependent OGR1 signalling triggered the splicing of XBP1, which was prevented in the presence of the OGR1 inhibitor (Fig. 1G,H, and Supplementary Figure 4), confirming the role of OGR1 in the induction of ER stress. OGR1 induces ER stress via IRE1α/JNK signalling. Next, we sought to identify the signalling factors involved in acidic pH-induced OGR1-mediated ER stress. Acidic pH induced BiP expression and JNK phosphorylation in OGR1-overexpressing Caco-2 cells compared to VC cells ( Fig. 2A and Supplementary Figure 5). Importantly, BiP expression and JNK phosphorylation were prevented in the presence of the OGR1 inhibitor ( Fig. 2A and Supplementary Figure 5). Strikingly, in OGR1-overexpressing cells the expression of JNK was increased in the presence of the OGR1 inhibitor. This result suggests a compensatory mechanism that would trigger JNK expression following blockade of JNK phosphorylation. Of note, acidic pH failed to induce cleavage of ATF6 ( Fig  . ER stress is induced by acidosis activated OGR1-mediated signalling. Caco-2 cells were subjected to different pH medium, following 4-6 h incubation in pH 7.6 serum free medium. (A) Vector control Caco-2 (VC) and OGR1 overexpressing Caco-2 cells where treated with tunicamycin at the indicated concentrations for 24 h. Total protein was isolated and Western blotting was performed. The results are representative of two independent experiments. (B) After 24 h pH shift, total protein was isolated and Western blotting was performed. The results are representative of three independent experiments. (C) Densitometry after normalization of BiP to β-actin and (D) p-IRE1α to total IRE1α. Statistical analysis was performed using oneway ANOVA followed by Tukey's post-test. Data are presented as means ± SE of three independent experiments (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). (E) After 24 h pH shift, total RNA was isolated and mRNA expression was investigated by qPCR. Statistical analysis was performed using one-way ANOVA followed by Tukey's post-test. Data are presented as means ± SE of three independent experiments (*p < 0.05; **p < 0.01). (F) A specific small molecule OGR1 inhibitor (10 µM) was tested and the cells were subjected to low pH for 24 h, following 4-6 h incubation in pH 7.6 serum free medium. After 24 h pH shift, total protein was isolated and Western blot performed. Results are representative of two independent experiments. (G) Cells were treated as described in (F), then total RNA was extracted and analysed for expression of XBP1 (XBP1u) and spliced XBP1 (XBP1s) by conventional PCR. Results are representative of three independent experiments. Acidosis activated OGR1-mediated signalling does not induce apoptosis. Since IRE1α/JNK signalling has been shown to trigger apoptosis by inhibiting Bcl-2, we investigated the impact of OGR1 activation on the induction of apoptosis. VC and OGR1-overexpressing cells where subjected to an acidic pH shift in the presence or absence of the OGR1 inhibitor for 24 h. Annexin V and PI staining followed by FACS analysis revealed that the population of apoptotic cells was not affected by the acidic pH shift in OGR1-overexpressing cells ( Fig. 3A-C). Furthermore, cleavage of caspase 3 and poly (ADP-ribose) polymerase (PARP) were investigated by immunoblotting. Under the condition that BiP was upregulated on activation of OGR1, neither cleaved caspase 3 nor cleaved PARP was observed ( Fig. 3D and Supplementary Figure 9), confirming that apoptosis was not induced in OGR1-overexpressing cells.
Acidosis activated OGR1-mediated signalling blocks autophagy. ER stress has been linked to the blockage of autophagy. Therefore, we sought to investigate the role of OGR1 in autophagy. VC and OGR1-overexpressing Caco-2 cells were subjected to an acidic pH shift for 24 h and protein levels of LC3-I and LC3-II were investigated by immunoblotting. Acidic pH reduced the conversion of LC3-I into LC3-II, and blocked autophagosome degradation, evidenced by the accumulation of total LC3 in OGR1-overexpressing Caco-2 cells compared to VC cells ( Fig. 4A and Supplementary Figure 10). We confirmed these results using immunofluorescence microscopy. OGR1-overexpressing cells subjected to an acidic pH shift showed increased LC3 staining, which was reversed in the presence of the OGR1 inhibitor. On the other hand, no changes were observed in the VC under different pH conditions with or without OGR1 inhibitor ( Fig. 4B-D). These results suggested that autophagy is blocked by proton-activated OGR1 signalling.

Discussion
Our results show that proton-activated OGR1-mediated signalling triggers the expression of the ER stress marker BiP together with the phosphorylation of IRE1α and splicing of XBP1 in a human intestinal epithelial cell line stably overexpressing OGR1. Furthermore, we found that activation of OGR1 triggers the IRE1α-JNK signalling pathway, but not the other branches involved in the UPR, namely PERK or ATF6. Acidosis and activation of the UPR in intestinal epithelial cells are closely linked to the development of intestinal inflammation (Fig. 5). Our results provide confirmatory evidence of a crucial role for OGR1-mediated IRE1α/JNK activation in the induction of ER stress under low pH conditions, which might underlie the reported impact of OGR1 in the development of IBD 18,19 . In our previous studies, we observed significant and pH-dependent OGR1-mediated signalling, including IP3/Ca 2+ /ERK signalling and enhanced SRF transcription under acidic pH conditions (pH = 6.8) 17 .
The link between acidic activation of GPCRs and MAPKs has been long established. Several reports have demonstrated that GPCRs can induce intracellular signal transduction through ERK1/2 and MAPK pathways 53,54 . Acidic OGR1 stimulation has been shown to trigger IL-6 expression through ERK1/2 and p38 activation in human airway smooth muscle cells 30 . Proton-dependent Ca 2+ release from intracellular stores has been shown to trigger the MEK/ERK1/2 pathway, thereby linking acidification with cell proliferation 2 . Recent reports have shown that ER stress triggers apoptosis via the activation of the IRE1α-JNK signalling pathway 55,56 . Surprisingly, we did not detect apoptotic processes following acidic activation of the IRE1α-JNK pathway, suggesting that OGR1-mediated IRE1α-JNK signalling may therefore promote cell survival together with OGR1 inflammatory signalling in intestinal epithelial cells. Interestingly, the pro-apoptotic role of JNK has been suggested to be strongly influenced by the parallel activation of cell survival pathways and the strength of the apoptotic response. Several reports indicate that while the sustained activation of JNK is associated with apoptosis, the acute and transient activation of JNK is crucial for cell proliferation and survival [57][58][59] . In this regard, several studies have also suggested that two functionally distinct phases of JNK signalling are involved in the ER stress response, an early phase that promotes survival and a late phase associated with cell death. Brown et al. showed that early JNK activation in ER-stressed cells triggers the expression of several apoptosis inhibitors early in the ER stress response. Using MEFs from IREα-and TRAF2-deficient mice, these authors showed that the early JNK activation requires both IRE1α and TRAF2 60 .
Additionally, acidic activation of OGR1 has been suggested to enhance survival in osteoclasts through the induction of PKC activation, which may affect the phosphorylation of pro-or anti-apoptotic proteins, or stimulate ERK1/2 signalling 61,62 . Although the role of PKC in autophagy regulation is still controversial, several studies have suggested that PKC is involved in the suppression of autophagy 63 . In HEK293 cells stably expressing LC3, activation of PKC significantly attenuated autophagy induced by starvation or rapamycin through the phosphorylation of LC3, while inhibition of PKC with pharmacological inhibitors increased autophagy 64 . PKC has also been shown to mediate cisplatin nephrotoxicity in vivo by suppressing autophagy 49 . Moreover, PKC has also been suggested to block autophagy in pancreatic ductal carcinoma cells through the activation of tissue transglutaminase 2 65,66 .
Acidic activation of OGR1 triggers the activation of JNK-mediated ER stress, which suggests a role of IRE1-JNK signalling in controlling autophagy 71 . Strikingly, our results show an increase in total LC3 Caco-2 cells were subjected to different pH medium, following 4-6 h incubation in pH 7.6 serum free medium. After 24 h pH shift, total protein was isolated and Western blotting was performed. Autophagy was measured by variations in the ratio of LC3-II/ LC3-I and the total amount of LC3 (LC3-I plus LC3-II) relative to GAPDH. Results are representative of two independent experiments. (B,C) Caco-2 cells were subjected to different pH medium, with or without OGR1 inhibitor (10 µM, following 4-6 h incubation in pH 7.6 serum free medium.) After 24 h pH shift, cells were fixed in 4% paraformaldehyde and stained with an anti-LC3 antibody. Cells were analysed by immunofluorescence microscopy and images were acquired under a confocal laser microscope. Results are representative of three independent experiments. Scale bars indicate 50 µm. (D) Quantification of the ratio of LC3/DAPI is presented. Changes in LC3 accumulation were calculated relative to DAPI staining from at least 4 areas. Statistical analysis was performed using one-way ANOVA followed by Tukey's post-test. Data are presented as means ± SE of three independent experiments (*p < 0.05; **p < 0.01). pH conditions: High pH 7.5-7.8; Normal pH 7.2-7.4; Low pH 6.6-6.8.

Scientific RepoRtS |
(2020) 10:1438 | https://doi.org/10.1038/s41598-020-57657-9 www.nature.com/scientificreports www.nature.com/scientificreports/ accumulation, but not in LC3-I to LC3-II conversion in OGR1 overexpressing cells following acidic pH shift, indicating an OGR1/IRE1/JNK-mediated blockage of the final stages of autophagy. The role of IRE1α-JNK in regulating autophagy remains a matter of controversy. Notably, JNK has been shown to play a role in autophagy suppression in neurons 72 . Conversely, the activation of ER stress triggered both apoptosis and autophagy through the IRE1/JNK/beclin-1 axis in breast cancer cells 73 . Another study showed that IRE1α upregulated autophagy under ER stress independently of XBP1 signalling 71 . Recently, phosphorylation of the anti-apoptotic protein BCL-2 by IRE1α was linked to the initiation of autophagy through the modulation of the activity of Beclin-1 74 , an essential component of the autophagy machinery 72,75,76 . JNK has been shown to participate in the expression of MAP1LC3 following TNF stimulation in vascular smooth muscle cells 77 . Inhibition of the JNK pathway blocked ceramide-induced autophagy and up-regulation of LC3 expression 78 . Xie et al. reported that JNK plays a crucial role in bufalin-induced autophagy in HT-29 and Caco-2 cells 79 .
In our hands, acidic activation of OGR1 in an OGR1-overexpressing cell model increased accumulation of LC3, but not the conversion of LC3-I into LC3-II, pointing to a blockage of late stage autophagy. Of note, our results suggest that partial activation of OGR1 under normal pH conditions is able to block late-stage autophagy in OGR1 overexpressing cells, and this effect is enhanced when OGR1 is fully activated at low pH. Interestingly, ROS-induced JNK activation induces both autophagy and apoptosis in cancer cells 80 . Taken together, our results suggest that acidic activation of OGR1 triggers opposite pathways leading to cell survival as well as the blockage of the late stages of autophagy. It is plausible that acidic activation of OGR1 initiates autophagy through IRE1α-JNK signalling together with parallel signals that block autophagosomal degradation, thereby contributing to the pro-survival and pro-inflammatory effects of OGR1.
Further investigations are required to elucidate the exact mechanisms of OGR1/IRE1/JNK-mediated blockage of the late stages of autophagy. Taken together, our results indicate that OGR1 may have novel functions in the regulation of ER stress and autophagy and could represent a novel therapeutic target of IBD. www.nature.com/scientificreports www.nature.com/scientificreports/ Methods Reagents. All chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA), including Tunicamycin (T7765) and Staurosporine (S6942), unless otherwise stated. A specific c-Jun N-terminal kinase (JNK) inhibitor (SP600125) was purchased from Calbiochem (La Jolla, CA). The OGR1 inhibitor was kindly provided by Takeda Pharmaceuticals San Diego, USA. All cell culture reagents were obtained from Thermo Fisher (Allschwil, Switzerland), unless otherwise specified.

Cell culture and pH shift. Caco-2 cells (LGC Promochem, Molsheim, Switzerland) and derived clones
stably overexpressing OGR1 were cultured in a humidified atmosphere with 5% CO 2 at 37 °C in Dulbecco's Modified Eagle's Medium (DMEM) with GlutaMAX (Invitrogen, Carlsbad, CA USA) supplemented with 400 µg/ ml geneticin (G418)-selective antibiotic (Invitrogen) and 10% fetal bovine serum (Invitrogen). Construction of the hu-OGR1-pcDNA3.1 + plasmid, clone generation, selection and characterization has been previously described 17 . pH treatment. pH shift experiments were carried out in serum-free RPMI-1640 medium supplemented with 2 mM GlutaMAX and 20 mM HEPES (all from Invitrogen). For pH adjustment of the RPMI medium, the appropriate quantities of NaOH or HCl were added, and the medium was allowed to equilibrate in the 5% CO 2 incubator at 37 °C for at least 36 h before it was used. Caco-2 cells were seeded and cultured for 24-48 hours before the pH shift was performed. Cells were starved for 4-6 h in serum free RPMI medium, pH 7.6, and then subjected to an acidic pH shift for 24 h. No. A11032, Invitrogen) for 1 h and DAPI (Sigma-Aldrich) for 5 min before mounting with anti-fade medium (Dako, Glostrup, Denmark). Cells were analysed by a Leica SP5 laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany). Fluorescence images were processed using Leica confocal software (LAS-AF Lite, Leica Microsystems). Quantification of LC3/DAPI was performed using ImageJ software [National Institutes of Health] 81 using the software's colour threshold tool, which calculates the area of positive staining. The resulting value was normalised to quantification of nucleus staining and represents the positively stained area normalised to cell numbers present in the given area.

Annexin V staining. Externalization of phosphatidylserine in apoptotic cells was detected with Annexin
V and dead cells were stained with propidium iodide (PI), using the Dead Cell Apoptosis Kit (Annexin V FITC and PI, Cat. No. V13242, Thermo Fischer Scientific), according to the manufacturer's instructions. After 10 min incubation at room temperature in the dark, cells were washed in PBS and resuspended in the binding buffer. Single-cell suspensions were analysed by FACS-Canto II flow cytometry (BD Biosciences, Allschwil, Switzerland) using FlowJo software.

RNA extraction and real-time quantitative PCR (qPCR). Total RNA was isolated using the RNeasy
Mini Kit (Qiagen, Hombrechtikon, Switzerland) according to the manufacturers' instructions. For removal of residual DNA, a DNase treatment was performed, according to the manufacturer's instructions, for 15 min at room temperature. For reverse transcription, the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) was used following the manufacturer's instructions. Determination of mRNA expression was performed by qPCR on a 7900HT real-time PCR system (Applied Biosystems) under the following cycling conditions: 20 s at 95 °C, then 45 cycles of 95 °C for 1 s, and 60 °C for 20 s with the TaqMan Fast Universal Master Mix. Samples were analysed as triplicates. Relative mRNA expression was determined the by the ΔΔCt method, which calculates the quantity of the target sequences relative to the endogenous control β-actin and a reference sample. TaqMan Gene Expression Assays (all from Applied Biosystems), used in this study were human BiP (Hs 00268858-S1) and human β-actin Vic TAMRA (4310881E). (2020) 10:1438 | https://doi.org/10.1038/s41598-020-57657-9 www.nature.com/scientificreports www.nature.com/scientificreports/ XBP1 splicing assay. XBP1 splicing was measured by specific primers flanking the splicing site yielding PCR product sizes of 152 and 126 bp for unspliced XBP1 and spliced XBP1 mRNA, respectively. Primers (forward 5′-CCTGGTTGCTGAAGAGGAGG-3′, reverse 5′-CCATGGGGAGATGTTCTGGAG-3′) were used. PCR was carried out at 95 °C for 15 min, then 40 cycles at 94 °C for 30 sec, 56.5 °C for 30 sec, and 72 °C for 1 min. The size difference between the spliced and the unspliced XBP1 is 26 nucleotides. These products were resolved on 3.5% agarose gels. Band intensity of XBP1s and XBP1u was determined using ImageJ and the ratio of XBP1s/XBP1u was quantified.
Statistical analysis. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA). Data are presented as means ± SE and statistical significance was determined using the Kruskal-Wallis test. p < 0.05 was considered significant. Where indicated, one-way ANOVA was performed, followed by Tukey's post hoc test.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.