Calcium signaling and secretory granule pool dynamics underlie biphasic insulin secretion and its amplification by glucose: experiments and modeling

Glucose-stimulated insulin secretion from pancreatic β-cells is controlled by a triggering pathway that culminates in calcium influx and regulated exocytosis of secretory granules, and by a less understood amplifying pathway that augments calcium-induced exocytosis. In response to an abrupt increase in glucose concentration, insulin secretion exhibits a first peak followed by a lower sustained second phase. This biphasic secretion pattern is disturbed in diabetes. It has been attributed to depletion and subsequent refilling of a readily releasable pool of granules or to the phasic cytosolic calcium dynamics induced by glucose. Here, we apply mathematical modeling to experimental data from mouse islets to investigate how calcium and granule pool dynamics interact to control dynamic insulin secretion. Experimental calcium traces are used as inputs in three increasingly complex models of pool dynamics, which are fitted to insulin secretory patterns obtained using a set of protocols of glucose and tolbutamide stimulation. New calcium and secretion data for so-called staircase protocols, in which the glucose concentration is progressively increased, are presented. These data can be reproduced without assuming any heterogeneity in the model, in contrast to previous modeling, because of nontrivial calcium dynamics. We find that amplification by glucose can be explained by increased mobilization and priming of granules. Overall, our results indicate that calcium dynamics contribute substantially to shaping insulin secretion kinetics, which implies that better insight into the events creating phasic calcium changes in human β-cells is needed to understand the cellular mechanisms that disturb biphasic insulin secretion in diabetes.

Dynamics of glucose-induced insulin secretion in normal human islets

The biphasic pattern of glucose-induced insulin secretion is altered in type 2 diabetes. Impairment of the first phase is an early sign of β-cell dysfunction, but the underlying mechanisms are still unknown. Their identification through in vitro comparisons of islets from diabetic and control subjects requires characterization and quantification of the dynamics of insulin secretion by normal islets. When perifused normal human islets were stimulated with 15 mmol/l glucose (G15), the proinsulin/insulin ratio in secretory products rapidly and reversibly decreased (∼50%) and did not reaugment with time. Switching from prestimulatory G3 to G6–G30 induced biphasic insulin secretion with flat but sustained (2 h) second phases. Stimulation index reached 6.7- and 3.6-fold for the first and second phases induced by G10. Concentration dependency was similar for both phases, with half-maximal and maximal responses at G6.5 and G15, respectively. First-phase response to G15–G30 was diminished by short (30–60 min) prestimulation in G6 (vs. G3) and abolished by prestimulation in G8, whereas the second phase was unaffected. After 1–2 days of culture in G8 (instead of G5), islets were virtually unresponsive to G15. In both settings, a brief return to G3–G5 or transient omission of CaCl2 restored biphasic insulin secretion. Strikingly, tolbutamide and arginine evoked immediate insulin secretion in islets refractory to glucose. In conclusion, we quantitatively characterized the dynamics of glucose-induced insulin secretion in normal human islets and showed that slight elevation of prestimulatory glucose reversibly impairs the first phase, which supports the view that the similar impairment in type 2 diabetic patients might partially be a secondary phenomenon.

l-cysteine reversibly inhibits glucose-induced biphasic insulin secretion and ATP production by inactivating PKM2

Increase in the concentration of plasma l-cysteine is closely associated with defective insulin secretion from pancreatic β-cells, which results in type 2 diabetes (T2D). In this study, we investigated the effects of prolonged l-cysteine treatment on glucose-stimulated insulin secretion (GSIS) from mouse insulinoma 6 (MIN6) cells and from mouse pancreatic islets, and found that the treatment reversibly inhibited glucose-induced ATP production and resulting GSIS without affecting proinsulin and insulin synthesis. Comprehensive metabolic analyses using capillary electrophoresis time-of-flight mass spectrometry showed that prolonged l-cysteine treatment decreased the levels of pyruvate and its downstream metabolites. In addition, methyl pyruvate, a membrane-permeable form of pyruvate, rescued l-cysteine–induced inhibition of GSIS. Based on these results, we found that both in vitro and in MIN6 cells, l-cysteine specifically inhibited the activity of pyruvate kinase muscle isoform 2 (PKM2), an isoform of pyruvate kinases that catalyze the conversion of phosphoenolpyruvate to pyruvate. l-cysteine also induced PKM2 subunit dissociation (tetramers to dimers/monomers) in cells, which resulted in impaired glucose-induced ATP production for GSIS. DASA-10 (NCGC00181061, a substituted N,N′-diarylsulfonamide), a specific activator for PKM2, restored the tetramer formation and the activity of PKM2, glucose-induced ATP production, and biphasic insulin secretion in l-cysteine–treated cells. Collectively, our results demonstrate that impaired insulin secretion due to exposure to l-cysteine resulted from its direct binding and inactivation of PKM2 and suggest that PKM2 is a potential therapeutic target for T2D.

Assessment of the Metabolic Pathways Associated With Glucose-Stimulated Biphasic Insulin Secretion

Biphasic glucose-stimulated insulin secretion involves a rapid first phase followed by a prolonged second phase of insulin secretion. The biochemical pathways that control these 2 phases of insulin secretion are poorly defined. In this study, we used a gas chromatography mass spectroscopy-based metabolomics approach to perform a global analysis of cellular metabolism during biphasic insulin secretion. A time course metabolomic analysis of the clonal β-cell line 832/13 cells showed that glycolytic, tricarboxylic acid, pentose phosphate pathway, and several amino acids were strongly correlated to biphasic insulin secretion. Interestingly, first-phase insulin secretion was negatively associated with l-valine, trans-4-hydroxy-l-proline, trans-3-hydroxy-l-proline, dl-3-aminoisobutyric acid, l-glutamine, sarcosine, l-lysine, and thymine and positively with l-glutamic acid, flavin adenine dinucleotide, caprylic acid, uridine 5′-monophosphate, phosphoglycerate, myristic acid, capric acid, oleic acid, linoleic acid, and palmitoleic acid. Tricarboxylic acid cycle intermediates pyruvate, α-ketoglutarate, and succinate were positively associated with second-phase insulin secretion. Other metabolites such as myo-inositol, cholesterol, dl-3-aminobutyric acid, and l-norleucine were negatively associated metabolites with the second-phase of insulin secretion. These studies provide a detailed analysis of key metabolites that are either negatively or positively associated with biphasic insulin secretion. The insights provided by these data set create a framework for planning future studies in the assessment of the metabolic regulation of biphasic insulin secretion.

A subcellular model of glucose-stimulated pancreatic insulin secretion

When glucose is raised from a basal to stimulating level, the pancreatic islets respond with a typical biphasic insulin secretion pattern. Moreover, the pancreas is able to recognize the rate of change of the glucose concentration. We present a relatively simple model of insulin secretion from pancreatic β-cells, yet founded on solid physiological grounds and capable of reproducing a series of secretion patterns from perfused pancreases as well as from stimulated islets. The model includes the notion of distinct pools of granules as well as mechanisms such as mobilization, priming, exocytosis and kiss-and-run. Based on experimental data, we suggest that the individual β-cells activate at different glucose concentrations. The model reproduces most of the data it was tested against very well, and can therefore serve as a general model of glucose-stimulated insulin secretion. Simulations predict that the effect of an increased frequency of kiss-and-run exocytotic events is a reduction in insulin secretion without modification of the qualitative pattern. Our model also appears to be the first physiology-based one to reproduce the staircase experiment, which underlies ‘derivative control’, i.e. the pancreatic capacity of measuring the rate of change of the glucose concentration.