INS-1 832/13 (Cell Culture Centre, CAMS, Beijing, China) cells were maintained in RPMI-1640 containing 11.1 mM D-glucose, 10% fetal calf serum, penicillin (100 U/ml) and streptomycin (100 mg/ml), 10 mM Hepes, 2 mM L-glutamine, 1 mM sodium pyruvate, and 0.05 mM 2-bmercaptoethanol. After the cells reached 70% confluence, the medium was replaced with RPMI-1640 containing BSA-conjugated sodium palmitate (PA; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) at concentrations of 0.5 mM, with or without EX-4 (50 nM; Sigma-Aldrich, Merck KGaA), or 0.5% BSA as control for 24 h.

To knock down IRE1α or RACK1 in INS-1 cells, cells were transfected with small interfering RNA (siRNA; Genepharma, Shanghai, China) targeting IRE1α (sequences: 5′-GGAATTACTGGCTTCTCATAG′) or RACK1 (target sequences: 5′-GCTAAAGACCAACCACATTGG-3′) at a concentration of 20 µM. Parallel cell cultures were transfected with control siRNA containing scrambled non-targeted sequence (5′-GTTCTCCGAACGTGTCACGTTT-3′) at the equal concentrations (Genepharma). INS-1 cells were transfected with the plasmid containing human XBP1 s (Addgene Inc., Cambridge, MA, USA) for XBP1s overexpression. At 48 h after transfection, the medium was replaced with regular medium containing PA (0.5 mM), with or without EX-4 (50 nM), Forskolin (10 µM; Sigma-Aldrich; Merck KGaA) or H89 (10 µM; Sigma-Aldrich; Merck KGaA) for 24 h, INS-1 cells precultured for 30 min with PKA inhibitor H89 were treated with EX-4.

Total RNA was isolated from INS-1 cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). First-strand cDNA was synthesized with moloney murine leukemia virus (M-MLV) reverse transcriptase (Invitrogen) and random hexamer primers (Invitrogen; Thermo Fisher Scientific, Inc.). RT-qPCR was conducted using the SYBR Green PCR system, following the manufacturer's recommendations (Applied Biosystems; Thermo Fisher Scientific, Inc.). GAPDH was used as an internal control for normalization. The oligonucleotide primers used are shown in Table I.

For coimmunoprecipitation analysis, INS-1 cells were lysed with the lysis buffer [20 mM tris-HCl (pH 7.5), 100 mM KCl, 0.1% Nonidet P-40, 1 mM EDTA, and 10% glycerol containing 1 mM phenyl-methyl-sulfonyl-fluoride (PMSF), 1% Protease Inhibitor Cocktail (Sigma-Aldrich; Merck KGaA), and 1% Phosphatase Inhibitor Cocktails I/II (Sigma-Aldrich; Merck KGaA) for 0.5 h at 4°C. After incubation with the desired primary antibody for 18 h at 4°C via gentle rocking, immune complexes were captured by mixing with a final concentration of 2.5% protein G Sepharosebeads (Amersham; GE Healthcare, Chicago, IL, USA) for 2 h at 4°C on a rotator. Anti-IRE1α antibody was used in the Co-IP assays. Beads were subsequently washed three times with the washing buffer [20 mM tris-HCl (pH 7.5), 150 mM KCl, 0.5% Nonidet P-40, 1 mM EDTA, and 10% glycerol supplemented with 1 mM PMSF, 1% Protease Inhibitor Cocktail, and 1% Phosphotase Inhibitor Cocktails I/II], followed by SDS-PAGE and immunoblotting analysis after elution by boiling in 2X SDS loading buffer.

Antibodies against phospho-Akt (Ser473, no. 9271), phosphor-eIF2α (Ser 51, no. 3597), phospho-FoxO1 (Thr24, no. 9464), IRE1α (no. 3294), XBP1s (no. 12782), phosphor-CREB (Ser133, no. 9196), eIF2α (no. 9722), Akt (no. 4691), FoxO1 (no. 2880), CREB (no. 9197), RACK1 (no. 5432), PKA (no. 4782) were purchased from Cell Signaling Technologies, Inc. (Danvers, MA, USA). α-tubulin antibody (T6199) from Sigma-Aldrich and antibody against phospho-IRE1α (Ser724, ab124945) from Abcam (Cambridge, MA, USA). Tubulin antibody was diluted 1:10,000 and all other antibodies were diluted 1:1,000. For immune-blotting, cellular lysates were prepared by RIPA buffer. Protein extracts were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane filter (Merck KGaA, Darmstadt, Germany). After incubation with desired antibodies, the blots were developed with Thermo Scientific's SuperSignal West Pico Chemiluminescent substrate or Millipore's Immunobilon Western Chemiluminescent HRP substrate.

INS-1 cell viability was measured using WST-8 assay using Cell Counting Kit-8 (CCK-8; Dojindo, Gaithersburg, MD, USA) according to manufacturer's instruction. INS-1 β cells were seeded in 96-well culture plates at a density of 10⁵ /ml. The next day, the culture medium was replaced with RPMI-1640 containing BSA-conjugated PA at concentrations of 0.5 mM, with or without EX-4, or 0.5% BSA as control for 24 h. The CCK-8 assay reagent was added to the culture medium for the final 3 h. A microplate reader was used to measure the absorbance at 450 nm.

For apoptosis analysis, cells were washed twice with 1X binding buffer then labelled with Annexin V and propidium iodide (PI) following the manufacturer's instructions. The Apoptosis Analysis kit was ordered from Beyotime Institute of Biotechnology (Haimen, China). Cell apoptosis was analyzed by fluorescence-activated cell sorting (FACS) using a FACScan flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA).

All data are expressed as the mean ± standard error mean. Student's t-test was used to compare mean values between two groups using the Graph-Pad Prism 5 program (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

Exposure to free fatty acids such as PA has been shown to cause ER stress (17). In our study, 0.5 mM PA for 24 h did cause a significant increase in the protein level of eukaryotic initiation factor 2α (eIF2α), a ER stress marker, as well as the decrease in the phosphorylation of Akt (p-Akt) and its physiological target FoxO1 (p-FoxO1) (Fig. 1A), which leads to decreased cell viability (Fig. 1B) and increased Bim mRNA expression in INS-1 cells (Fig. 1C). The activation of Bim has been supposed to trigger cleavage of caspases and is critical for the apoptosis in β cells (18). In accord with this, PA induced significant cell apoptosis in INS-1 cells (Fig. 1D). However, EX-4 treatment induced a significant reverse in the protein level of p-Akt and p-FoxO1, displaying a protective role on the INS-1 cells, as demonstrated by the increased cell viability and decreased expression of Bim. IRE1α-Xbp1 signaling pathway has been implicated in the homeostatic regulation of pancreatic islet β cells, we then tested whether EX-4-induced protective roles on INS-1 cells was associated with this signaling pathway. We found that PA triggers the mild increase in the phosphorylation at the Ser 724 activation site in IRE1α as detected by a phosphorylation site-specific antibody, which triggered the increase of XBP1 s protein levels (Fig. 1A). Interestingly, the protein level of p-IRE1α was further increased in response to EX-4 treatment (Fig. 1A), accompanied by the enhancement in XBP1s, suggesting that EX-4 could stimulate IRE1α-Xbp1 signaling pathway.

To determine whether IRE1α-Xbp1 signaling pathway acts as a critical component in mediating EX-4's beneficial effects in PA-induced INS-1 cells, we used siRNA to silence the IRE1α gene. INS-1 cells were transfected with siRNA specific for IRE1α and the efficiency of IRE1α knockdown was validated (Fig. 2A). In nontransfected cells treated with EX-4, a significant increase in p-Akt and a drastic decrease in p-eIF2α protein levels was observed (Fig. 1A), in contrast, transfection of INS-1 cells with siRNA targeting IRE1α, the expression of p-Akt was only partially improved by EX-4 treatment under PA-treated conditions (Fig. 2A). The decreased cell viability and increased mRNA level of Bim and cell apoptosis were also not ameliorated (Fig. 2B-D). As IRE1α associates with various signaling molecules, we then tested whether EX-4-induced upregulation of p-Akt is through the splicing of Xbp1 mRNA by IRE1α phosphorylation. In INS-1 cells with RNAi-mediated knockdown of IRE1α, the decreased protein levels of p-Akt and cell viability under EX-4 treatment was rescued by XBP1s overexpression (Fig. 2A and B), while the mRNA level of Bim and cell apoptosis markedly decreased (Fig. 2C and D) This data indicates that the beneficial roles of EX-4 treatment on PA-induced INS-1 cells might be associated with the IRE1α-Xbp1 signaling pathway.

Previous research has supported the idea that under ER stress conditions, IRE1α is activated through dimerization and transautophosphorylation (8,9). To test the idea that a protein kinase other than IRE1α itself may link EX-4 with the observed stimulation of IRE1α phosphorylation, we examined whether protein kinase A (PKA) is involved in the stimulation of IRE1α phosphorylation by EX-4. We found that EX-4 could promote the activation of PKA, as evidenced by the increased expression of cAMP response element-binding protein (CREB) phosphorylation (p-CREB), as well as the upregulation of IRE1α phosphorylation (Fig. 3). In accordance with this, forskolin, a chemical activator of PKA, also triggered the increase of IRE1α phosphorylation, while the increase in the IRE1α phosphorylation by EX-4 was suppressed by the inhibition of PKA using H89 (Fig. 3), a pharmacological PKA inhibitor, implying that EX-4-regulated IRE1α-phosphorylation is mediated through a PKA-dependent manner.

RACK1, binding to membrane receptors and protein kinases, coordinates the interactions between signaling components in multiple cellular processes (19). We found that EX-4 treatment induced the association of IRE1α with PKA and RACK1 by Co-IP analysis (Fig. 4A). More importantly, INS-1 cells with knockdown of RACK1, although presenting increased p-CREB by EX-4, showed the absent increased IRE1α phosphorylation compared to cells transfected with a scrambled negative control siRNA (Fig. 4B). This confirms a possible function of RACK1 in the recruitment of PKA to phosphorylated IRE1α in response to EX-4 treatment.