Susceptibility of pancreatic beta cells to fatty acids is regulated by LXR/PPARalpha-dependent stearoyl-coenzyme A desaturase

Abstract

Chronically elevated levels of fatty acids-FA can cause beta cell death in vitro. Beta cells vary in their individual susceptibility to FA-toxicity. Rat beta cells were previously shown to better resist FA-toxicity in conditions that increased triglyceride formation or mitochondrial and peroxisomal FA-oxidation, possibly reducing cytoplasmic levels of toxic FA-moieties. We now show that stearoyl-CoA desaturase-SCD is involved in this cytoprotective mechanism through its ability to transfer saturated FA into monounsaturated FA that are incorporated in lipids. In purified beta cells, SCD expression was induced by LXR- and PPARalpha-agonists, which were found to protect rat, mouse and human beta cells against palmitate toxicity. When their SCD was inhibited or silenced, the agonist-induced protection was also suppressed. A correlation between beta cell-SCD expression and susceptibility to palmitate was also found in beta cell preparations isolated from different rodent models. In mice with LXR-deletion (LXRbeta(-/-) and LXRalphabeta(-/-)), beta cells presented a reduced SCD-expression as well as an increased susceptibility to palmitate-toxicity, which could not be counteracted by LXR or PPARalpha agonists. In Zucker fatty rats and in rats treated with the LXR-agonist TO1317, beta cells show an increased SCD-expression and lower palmitate-toxicity. In the normal rat beta cell population, the subpopulation with lower metabolic responsiveness to glucose exhibits a lower SCD1 expression and a higher susceptibility to palmitate toxicity. These data demonstrate that the beta cell susceptibility to saturated fatty acids can be reduced by stearoyl-coA desaturase, which upon stimulation by LXR and PPARalpha agonists favors their desaturation and subsequent incorporation in neutral lipids.

Figure 1. Agonists for LXR protect primary beta cells against palmitate toxicity.

a) Rat and human beta cells were exposed to 500 µM-P, or 250 µM-P in the absence or presence of agonists for LXR (10 µM TO1317 or 1 µM GW3965), FXR (5 µM GW4064), or PXR (30 µM PCN). Data shown as mean+sd, n = 4-10, * p<0.05, ** p<0.01, *** p<0.001, versus palmitate b) Time course over 16 days for the toxicity of 250 µM-P±TO1317 (1 µM) in rat beta cells. Data shown as mean+sd, n = 7-10.

Figure 2. Effect of LXR agonist on intracellular lipid accumulation and palmitate metabolism.

a) Representative figure showing electron microscopy and vital nile red staining on control cells, and beta cells exposed to 250 µM-P for 2 days±TO1317 (10 µM). Magnification: EM 2550x, Nile red (green) + nuclear Hoechst (red) staining: 60x. b) Effect of 250 µM-P±TO1317 on ¹⁴C-palmitate metabolism. Beta-cells were cultured for 24 h in the absence (white bars) or presence (gray bars) of 250 µM-P±TO1317, and subsequently incubated for 2 hrs with ¹⁴C-palmitic acid. Metabolism was measured as recovery of ¹⁴C-label in CO₂, in lipids and lipid intermediates (pmol/2 h/10³ cells) and expressed as percentages of the control condition. Data shown as mean±se (n = 4), Student's t-test, ^$$ P<0.01 and ^$$$ P<0.001, versus control; * P<0.05 and ** P<0.01 versus palmitate.

Figure 3. Enzymes involved in conversion of palmitate in response TO1317.

a) Overview of enzymes involved in conversion of palmitate (C16:0) to palmitoylate (C16:1), vaccinate (C18:1 n-7), stearate (C18:0) and oleate (C18:1 n-9): stearoyl CoA desaturase (SCD1, SCD2), elongases (ELOvl5, ELOvl6), diacylglycerol transferase (DGAT1, DGAT2), sterol-O-acyl transferase (SOAT1, SOAT2). TG, triglycerides; CE, cholesteryl esters. b) qPCR result examining the mRNA levels of the above enzymes in freshly isolated rat beta cells, alpha-cells and liver, white adipose and brain tissue. Gene expression levels were normalized to actin, n = 3. c) qPCR analysis for SCD1 and 2, showing the effect of 8 days 250 µM-P±TO1317. Data shown as mean±sd, n = 4, *** p<0.001, compared to control; ^$ p<0.05, ^$$$ p<0.001, versus TO1317. d) Representative immunoblot showing SCD protein levels in beta cells cultured for 8 days with or without 250 µM-P±TO1317 (10 µM). Equal amounts of total protein were added on a 10% polyacrylamide gel. Representative for 4 independent experiments.

Figure 4. Δ9-desaturase activity plays a key role in protection against palmitate toxicity.

a) Effect of inhibition of SCD. Rat and human beta cells were exposed to 250 µM-P or 500 µM-P±10 µM TO1317 or 250 µM clofibrate/2 µM 9RA, and/or c9,t11 CLA (40 µM), or t10,c12 CLA (40 µM). Percentage cytotoxicity shown as mean+sd, n = 3-4, *** p<0.001, versus palmitate. b) Effect of treatment on fatty acid composition as measured by GC-MS. Beta cells were exposed to 250 µM-P±10 µM TO1317 and±40 nM t10,c12 CLA for 8 days. The levels for C16:0, C18:0, C16:1, C18:1 and C20:4 measured as nmol/nmol Pi were expressed relative to their levels in control cells (ratios of saturated over monounsaturated FA are shown in Figure S1). Mean±sd, n = 4 - 6, * p<0.05 versus control, $ p<0.05 versus palmitate. c) RNA interference. Beta cells were transfected with a pool of siRNA's directed against SCD1 and 2 (Si-SCD1/SCD2), and exposed to 250 µM-P and/or TO1317 for 3 days. Si-control cells were treated with lipid transfection carrier (jetsi endo) only, or a pool of siGlo risk-free siRNA's (siGlo). Percentage cytotoxicity shown as mean+sd, n = 4, * p<0.05, *** p<0.001, versus Si-control cells.

Figure 5. Effect of LXR and PPARα agonists on palmitate toxicity in LXR KO beta cells.

Primary mouse beta cells were exposed to a) 250 µM-P or 250 µM oleate for 8 days in presence or absence of 1 µM TO1317 or 250 µM clofibrate/2 µM 9-cis RA. Percentages of cytotoxicity are shown as mean+sd, n = 3, *** p<0.001, TO1317 and clofibrate/9RA versus palmitate; ^$$$ p<0.001, palmitate toxicity in LXR^-/- as versus wild type. b) SCD1, SCD2 and SREBP1c mRNA levels in LXR^-/- islets as compared to wild type. Q-PCR signals shown as mean+sd, n = 3, * p<0.001.

Figure 6. SCD levels regulate beta cell susceptibility for palmitate in vivo.

Beta cells and alpha-enriched cell preparations were obtained from: a-b) Zucker fatty (fa/fa) and lean control rats (fa/-) and exposed for 8 days to 250 µM and 500 µM palmitate or oleate in the presence or absence of 10 µM TO1317±40 nM t10,c12 CLA. Percentages of cytotoxicity are shown as mean±sd, n = 3, * p<0.001 fa/fa versus lean control, $ p<0.001 versus palmitate; or isolated from c-d) Wistar rats treated by oral gavages for 5 days with 40mg/kg BW TO1317 or vehicle (DMSO/PBS, 1/3), and exposed for 2 days to 500 µM palmitate or oleate in the presence or absence of 1 µM TO1317 or 250 µM clofibrate +2 µM 9-cis RA. Results are shown as mean±sd, n = 3, * p<0.001 TO1317-cells versus vehicle, $ p<0.001 versus control; or e-f) Beta cells were FACS sorted based on their glucose-induced increase in [NAD(P)H]-autofluorescence into two subpopulations characterized by a high or low glucose-responsive phenotype , and exposed for 8 days to 250 µM palmitate or oleate in the presence or absence of 1 µM TO1317 or 250 µM clofibrate plus 2 µM 9-cis RA. Percentages of cytotoxicity are shown as mean±sd, n = 3, * p<0.01 low versus high NAD(P)H-cells; $ p<0.01 agonist treatment versus control. a,c,e) Q-PCR analysis showing the differences in expression levels of glucokinase, SCD1, SCD2, LXRα, LXRβ, PPARα and SREBP1c. Results are shown as mean±sd, n = 3-7, * p<0.05, ** p<0.01 versus the respective control preparations.

Figure 7. Hypothetical lipoprotection model using LXR or PPAR ligands.

Long chain saturated FA, such as palmitate, induce cell death through reactive intermediates, ceramide generation or via ER stress. Activation of LXR downstream of PPARα activates SREBP1c and induces transcription of genes encoding SCD1 and SCD2. This results in an increased formation of monounsaturated FA and directs the flow of palmitate towards intracellular storage as neutral lipids. Next to inducing SCD levels, activation of PPARα results in an increased turn-over of FA by stimulation of β-oxidation, with detoxifying effects on both saturated and monounsaturated FA.