SPARC promotes insulin secretion through down-regulation of RGS4 protein in pancreatic β cells

Abstract

SPARC-deficient mice have been shown to exhibit impaired glucose tolerance and insulin secretion, but the underlying mechanism remains unknown. Here, we showed that SPARC enhanced the promoting effect of Muscarinic receptor agonist oxotremorine-M on insulin secretion in cultured mouse islets. Overexpression of SPARC down-regulated RGS4, a negative regulator of β-cell M3 muscarinic receptors. Conversely, knockdown of SPARC up-regulated RGS4 in Min6 cells. RGS4 was up-regulated in islets from sparc -/- mice, which correlated with decreased glucose-stimulated insulin secretion (GSIS). Furthermore, inhibition of RGS4 restored GSIS in the islets from sparc -/- mice, and knockdown of RGS4 partially decreased the promoting effect of SPARC on oxotremorine-M-stimulated insulin secretion. Phosphoinositide 3-kinase (PI3K) inhibitor LY-294002 abolished SPARC-induced down-regulation of RGS4. Taken together, our data revealed that SPARC promoted GSIS by inhibiting RGS4 in pancreatic β cells.

Figure 1

SPARC expression levels were regulated by glucose and insulin. (A) Min6 cells were incubated with increasing concentrations of glucose (0–40 mM) for 16 h. Cell lysates were subjected to western blotting with anti-SPARC antibody or β-actin antibody. (B) Min6 cells were stimulated with increasing concentration of insulin (0 to 100 nM) for 16 h. Cell lysates were subjected to Western blotting with anti-SPARC antibody or β-actin antibody. (C,D) The SPARC levels were quantified relative to loading control β-actin each from three experiments. * denotes P < 0.05, n = 3. Two-way ANOVA followed by Tukey’s multiple comparisons test was used in statistical analysis.

Figure 2

SPARC enhanced muscarinic receptor agonist-stimulated insulin secretion. (A) Mouse islets were isolated and culture in RPMI1640 containing 11 mM glucose overnight. The islets were infected with AdV-EGFP or AdV-SPARC for 24 h. The islets were stimulated with 16.7 mM glucose, or 16.7 mM glucose and 100 µM Oxo-M in Krebs Ringer Bicarbonate buffer for 1 h. The supernatants were collected, and insulin was determined with insulin ELISA kit. Shown were the levels of insulin secretion normalized to protein concentration. The results were derived from three individual experiments. * denotes P < 0.05, n = 3. (B) Mouse islets were cultured in RMPI1640 medium with 11 mM glucose overnight. Groups of islets were then incubated with Krebs Ringer Bicarbonate buffer and with 16.7 mM glucose in the presence or absence of recombinant SPARC, 1 µg/ml, or 16.7 mM glucose and 100 µM Oxo-M in the presence or absence of SPARC for 1 h. The supernatants were collected, and insulin was determined with insulin ELISA kit. Shown were the levels of insulin secretion normalized to protein concentration. * denotes P < 0.05, n = 3. Statistical significance was assessed by unpaired two-tailed Student’s t test.

Figure 3

Overexpression of SPARC down-regulates RGS4 mRNA and protein in Min6 cells. Min6 cells were infected with AdV-EGFP or AdV-SPARC for 24 h. (A) mRNAs were prepared from AdV-EGFP and AdV-SPACR treated Min6 cells and were subjected to cDNA microarray. The microarray data measuring relative expression of selected genes were shown in the heat map. (B) RGS4 mRNA was downregulated by overexpression of SPARC in Min6 cells. The cells were harvested for RNA preparation. RT-qPCR was performed to determine RGS4 mRNA levels after SPARC overexpression. (C) The cell lysates were subjected to Western blotting with anti-SPARC antibody or anti-β-actin antibody, and SPARC protein was shown to be overexpressed. The cell lysates were also subjected to Western blotting with anti-RGS4 antibody, and the results showed down-regulation of RGS4 protein by overexpression of SPARC in Min6 cells. (D) Relative RGS4 protein levels were quantified and normalized with β-actin. Statistical significance was assessed by unpaired two-tailed Student’s t test. ** denotes P < 0.01, n = 3. *** denotes P < 0.001, n = 3.

Figure 4

Knocking down of SPARC with siRNA in Min6 cells up-regulated RGS4 mRNA, and loss of sparc up-regulated RGS4 protein in isolated islets. Min6 cells were transfected with SPARC siRNA and cultured for 48 h. (A,B) The cell lysates were subjected to Western blotting with anti-SPARC antibody or anti-β-actin antibody (A). (B) Relative RGS4 levels were quantified and normalized with β-actin. *** denotes P < 0.001, n = 3. Statistical significance was assessed by unpaired two-tailed Student’s t test. (C,D) Up-regulation of RGS4 protein in the islets of sparc −/− mice. Mouse islets were isolated from WT or sparc −/− mice and subject to Western blot analysis of RGS4 protein (C). Average RGS4 levels from two experiments were quantified and normalized to β-actin levels (D). ** denotes P < 0.01, n = 3.

Figure 5

SPARC-induced AKT phosphorylation and RGS4 down-regulation in Min6 cells is abolished by PI3 kinase inhibitor LY-294002. (A) SPARC-induced RGS4 downregulation was blocked by PI3K inhibitor. Cultured Min6 cells after serum starvation for 12 h were pretreated with the PI3K inhibitor LY-294002 (10 μM) for 1 h before treatment with or without recombinant SPARC (1 µg/ml) for 12 h followed by Western blot analysis with the indicated specific antibodies. Down-regulation of RGS4 by SPARC protein was abolished by LY-294002 (B). Lower panel was ratio of p-AKT S473 /total AKT quantified from three experiments (C). ** denotes P < 0.01, n = 3. *** denotes P < 0.001, n = 3. Statistical significance was assessed by unpaired two-tailed Student’s t test.

Figure 6

RGS4 inhibitor CCG4986 restored insulin secretion in islets of sparc −/− mice. (A) Mouse islets were isolated from WT or sparc −/− mice. The isolated islets were cultured for 16 h in RPMI1640 with 11 mM Glucose and 10% FBS. The WT or sparc −/− islets were incubated for 1 h in 2.8 mM glucose, then the medium was changed to 16.7 mM glucose, 16.7 mM glucose with Oxo-M, 100 µM in Krebs Ringer Bicarbonate buffer for 1 h. The supernatants were collected for insulin assay. Shown were the levels of insulin secretion normalized to total protein concentrations. * denotes P < 0.05, n = 3. Statistical significance was assessed by unpaired two-tailed Student’s t test. (B) sparc −/− mouse islets were incubated for 1 h in 2.8 mM glucose, and then the medium was changed to 16.7 mM glucose, 16.7 mM glucose with Oxo-M, 100 µM, or 16.7 mM glucose + CCG4986, 100 µM, or 16.7 mM glucose, Oxo-M, 100 µM, CCG4986, 100 µM. The supernatants were collected, and insulin was measured. Shown were the levels of insulin secretion normalized to protein concentration. * denotes P < 0.05, n = 3. Two-way ANOVA followed by Tukey's multiple comparisons test was used in statistical analysis. (C,D) Min6 cells were transfected with RGS4 siRNA and cultured for 48 h. The cell lysates were subjected to Western blotting with anti-RGS4 antibody or anti-β-actin antibody (C). (D) After being transfected with RGS4 siRNA for 48 h, groups of Min6 cells were then incubated with 16.7 mM glucose, or 16.7 mM glucose and 100 µM Oxo-M in the presence or absence of recombinant SPARC, 1 µg/ml for 1 h. The supernatants were collected, and insulin was measured. Shown were the levels of insulin secretion normalized to protein concentration. n.s. denotes P > 0.05, n = 3. * denotes P < 0.05, n = 3. Statistical significance was assessed by unpaired two-tailed Student’s t test. (E) The graph schematically outlined a potential novel mechanism for the regulation of insulin secretion by SPARC. Increased SPARC inhibited RGS4 expression, which, in turn, would result in increased M3 receptor / G-protein coupling, leading to increased insulin secretion in β cells.