Vesicular nucleotide transporter-mediated ATP release regulates insulin secretion

Abstract

Extracellular ATP plays a critical role in regulating insulin secretion in pancreatic β cells. The ATP released from insulin secretory vesicles has been proposed to be a major source of extracellular ATP. Currently, the mechanism by which ATP accumulates into insulin secretory granules remains elusive. In this study, the authors identified the expression of a vesicular nucleotide transporter (VNUT) in mouse pancreas, isolated mouse islets, and MIN6 cells, a mouse β cell line. Immunohistochemistry and immunofluorescence revealed that VNUT colocalized extensively with insulin secretory granules. Functional studies showed that suppressing endogenous VNUT expression in β cells by small hairpin RNA knockdown greatly reduced basal- and glucose-induced ATP release. Importantly, knocking down VNUT expression by VNUT small hairpin RNA in MIN6 cells and isolated mouse islets dramatically suppressed basal insulin release and glucose-stimulated insulin secretion (GSIS). Moreover, acute pharmacologic blockade of VNUT with Evans blue, a VNUT antagonist, greatly attenuated GSIS in a dose-dependent manner. Exogenous ATP treatment effectively reversed the insulin secretion defect induced by both VNUT knockdown and functional inhibition, indicating that VNUT-mediated ATP release is essential for maintaining normal insulin secretion. In contrast to VNUT knockdown, overexpression of VNUT in β cells resulted in excessive ATP release and enhanced basal insulin secretion and GSIS. Elevated insulin secretion induced by VNUT overexpression was reversed by pharmacologic inhibition of P2X but not P2Y purinergic receptors. This study reveals VNUT is expressed in pancreatic β cells and plays an essential and novel role in regulating insulin secretion through vesicular ATP release and extracellular purinergic signaling.

Figure 1.

Expression of VNUT in the pancreas. A, Representative agarose gel electrophoresis showing the expression of VNUT and β-actin mRNA in total RNA samples isolated from mouse pancreas, isolated mouse islets, and MIN6 cells. B, Representative photomicrographs showing VNUT immunostaining in mouse pancreatic sections with (left) or without (right) preabsorption of the anti-VNUT serum with 30 μM mouse VNUT peptide. VNUT-positive cells (dark brown) were observed predominantly in the islets. Mouse islets in the left panel are indicated by a dashed line. Scale bar, 100 μm. C, Upper panels: Stacked serial confocal images of VNUT (upper left, green) and insulin (upper middle, red) staining in the islet. Upper right: Visualization of both VNUT and insulin simultaneously showing most VNUT and insulin colocalize in the same cells. Bottom panels: High magnification of the confocal images in the boxed area showing VNUT (lower left, green) and insulin (lower middle, red) in a single β cell. Bottom right: Merged image visualizing VNUT and insulin simultaneously. Representative VNUT/insulin double-labeled granules are indicated by white arrows. Scale bar: Top panel, 10 μm; Bottom panel, 2.5 μm.

Figure 2.

Endogenous VNUT is critical for ATP release from pancreatic β cells. A, Measurement of ATP release from MIN6 cells stimulated with various concentrations of glucose (G) or KCl (50 mM). B, Representative confocal images showing MIN6 cells stained with quinacrine and FM4–64 under basal conditions (2.8 mM) and after a 10-minute challenge with high glucose (25 mM). C, Summary of quinacrine staining in MIN6 cells treated with 2.8 mM or 25 mM glucose. D, VNUT mRNA expression measured by real-time PCR from MIN6 cells infected with either control or shRNA VNUT lentiviral vectors for 72 hours. E, MIN6 cells were infected with control or shRNA VNUT lentiviral vectors 72 hours before being stained with quinacrine. Representative monolayers were imaged using confocal microscopy under basal conditions (2.8 mM). F, Summary of quinacrine staining in MIN6 cells infected with control or shRNA VNUT lentiviral vectors. G, ATP release from control or VNUT shRNA-infected MIN6 cells under basal conditions (2.8 mM glucose) or after 20 minutes of glucose stimulation (25 mM glucose). Scale bar: 20 μm. * P < .05, ** P < .01.

Figure 3.

VNUT-mediated ATP release is essential for insulin secretion. A, Insulin secretion from MIN6 cells infected with either control or VNUT shRNA lentiviral vectors under basal conditions (2.8 mM G), stimulated with glucose (16.8 mM G) or tolbutamide (200 μM) for 60 minutes. B, Insulin secretion from MIN6 cells pretreated with various concentrations of Evans blue for 30 minutes before stimulation with high glucose (16.8 mM G) for 60 minutes. C, VNUT mRNA expression measured by real-time PCR from isolated mouse islets infected with either control or shRNA VNUT lentiviral vector for 72 hours. D, Insulin secretion from isolated mouse islets infected with either control or VNUT shRNA lentiviral vectors under basal conditions (2.8 mM G), stimulated with glucose (16.8 mM G) or tolbutamide (200 μM) for 60 minutes. E, Insulin secretion from control and VNUT shRNA-infected islets stimulated with glucose (16.8 mM) in the presence or absence of exogenous ATP (200 μM) for 60 minutes. F, Insulin secretion from islets pretreated with various concentrations of Evans blue for 30 minutes before stimulation with glucose (16.8 mM) with or without exogenous ATP (200 μM) for 60 minutes. * P < .05, ** P < .01.

Figure 4.

Overexpression of VNUT enhances ATP release from β cells. A, VNUT mRNA expression measured by real-time PCR from MIN6 cells infected with either control or o/e VNUT lentiviral vectors for 72 hours. B, ATP release was measured from control and o/e VNUT MIN6 cells under basal conditions (2.8 mM glucose) or after 20 minutes of glucose stimulation (25 mM glucose). C, Representative confocal images showing quinacrine (green) and FM4–64 (red) staining in MIN6 cells infected with either control (upper panel) or o/e VNUT lentiviral vectors (lower panel) in basal media (2.8 mM glucose). D, Summary of quinacrine staining in control and o/e VNUT MIN6 cells. E, Quantification of the average number of quinacrine positive vesicles per 100 μm². Scale bar: 20 μm. * P < .05, ** P < .01, *** P < .001.

Figure 5.

Overexpressing VNUT potentiates insulin secretion. A, VNUT mRNA expression measured by real-time PCR from isolated mouse islets infected with either control or o/e VNUT lentiviral vectors for 72 hours. B, Insulin secretion from isolated mouse islets infected with either control or o/e VNUT lentiviral vectors under basal conditions (2.8 mM G) or glucose stimulation (16.8 mM G) for 60 minutes. C and D, Insulin secretion from control and o/e VNUT-infected islets under basal conditions or stimulated with glucose (16.8 mM) for 60 minutes in the presence or absence of iso-pyridoxalphosphate-6-azophenyl-2′,4′-disulfonic acid (iso-PPADS; 50 μM) (C) or MRS 2179 (0.5 μM) (D). * P < .05, *** P < .001.