Glucose-induced [Ca2+]i oscillations in β cells are composed of trains of spikes within a subplasmalemmal microdomain

Simultaneous TIRF imaging of subplasmalemmal Ca2+ dynamics and granule fusions in insulin-secreting INS-1 cells reveals coexistent synchronized and asynchronous release

The basal and glucose-induced insulin secretion from pancreatic beta cells is a tightly regulated process that is triggered in a Ca2+-dependent fashion and further positively modulated by substances that raise intracellular levels of adenosine 3',5'-cyclic monophosphate (cAMP) or by certain antidiabetic drugs. In a previous study, we have temporally resolved the subplasmalemmal [Ca2+]i dynamics in beta cells that are characterized by trains of sharply delimited spikes, reaching peak values up to 5 µM. Applying total internal reflection fluorescence (TIRF) microscopy and synaptopHluorin to visualize fusion events of individual granules, we found that several fusion events can coincide within 50 to 150 ms. To test whether subplasmalemmal [Ca2+]i microdomains around single or clustered Ca2+ channels may cause a synchronized release of insulin-containing vesicles, we applied simultaneous dual-color TIRF microscopy and monitored Ca2+ fluctuations and exocytotic events in INS-1 cells at high frame rates. The results indicate that fusions can be triggered by subplasmalemmal Ca2+ spiking. This, however, does account for a minority of fusion events. About 90 %-95 % of fusion events either happen between Ca2+ spikes or incidentally overlap with subplasmalemmal Ca2+ spikes. We conclude that only a fraction of exocytotic events in glucose-induced and tolbutamide- or forskolin-enhanced insulin release from INS-1 cells is tightly coupled to Ca2+ microdomains around voltage-gated Ca2+ channels.

STIM1 regulates pancreatic β-cell behaviour: A modelling study

Pancreatic β-cells are equipped with the molecular machinery allowing them to respond to high glucose levels in the form of electrical activity and Ca2+ oscillations. These oscillations drive insulin secretion. Two key ionic mechanisms involved in this response are the Store-Operated Current and the current through ATP-dependent K+ channels. Both currents have been shown to be regulated by the protein STIM1, but this dual regulation by STIM1 has not been studied before. In this paper, we use mathematical modelling to gain insight into the role of STIM1 in the β-cell response. We extended a previous β-cell model to include the dynamics of STIM1 and described the dependence of the ATP-dependent K+ current on STIM1. Our simulations suggest that the total concentration of STIM1 modifies the bursting frequency, the burst duration and the intracellular Ca2+ levels. These results are in good agreement with experimental reports, and the contribution of the studied currents to electrical activity and Ca2+ dynamics is discussed. The model predicts that in the absence of STIM1 the excitability of the plasma membrane increases and that the glucose threshold for electrical activity is shifted to lower concentrations. These computational predictions may be related to impaired insulin secretion under conditions of reduced STIM1 in the diabetic state.

Indicator-dependent differences in detection of local intracellular Ca2+ release events evoked by voltage-gated Ca2+ entry in pancreatic β-cells

Genetically encoded Ca²⁺ indicators have become widely used in cell signalling studies as they offer advantages over cell-loaded dye indicators in enabling specific cellular or subcellular targeting. Comparing responses from dye and protein-based indicators may provide information about indicator properties and cell physiology, but side-by-side recordings in cells are scarce. In this study, we compared cytoplasmic Ca²⁺ concentration ([Ca²⁺]_i) changes in insulin-secreting β-cells recorded with commonly used dyes and indicators based on circularly permuted fluorescent proteins. Total internal reflection fluorescence (TIRF) imaging of K⁺ depolarization-triggered submembrane [Ca²⁺]_i increases showed that the dyes Fluo-4 and Fluo-5F mainly reported stable [Ca²⁺]_i elevations, whereas the proteins R-GECO1 and GCaMP5G more often reported distinct [Ca²⁺]_i spikes from an elevated level. [Ca²⁺]_i spiking occurred also in glucose-stimulated cells. The spikes reflected Ca²⁺ release from the endoplasmic reticulum, triggered by autocrine activation of purinergic receptors after exocytotic release of ATP and/or ADP, and the spikes were consequently prevented by SERCA inhibition or P2Y₁-receptor antagonism. Widefield imaging, which monitors the entire cytoplasm, increased the spike detection by the Ca²⁺ dyes. The indicator-dependent response patterns were unrelated to Ca²⁺ binding affinity, buffering and mobility, and probably reflects the much slower dissociation kinetics of protein compared to dye indicators. Ca²⁺ dyes thus report signalling within the submembrane space excited by TIRF illumination, whereas the protein indicators also catch Ca²⁺ events originating outside this volume. The study highlights that voltage-dependent Ca²⁺ entry in β-cells is tightly linked to local intracellular Ca²⁺ release mediated via an autocrine route that may be more important than previously reported direct Ca²⁺ effects on phospholipase C or on intracellular channels mediating calcium-induced calcium release.

Complex dynamics in an unexplored simple model of the peroxidase-oxidase reaction

A previously overlooked version of the so-called Olsen model of the peroxidase-oxidase reaction has been studied numerically using 2D isospike stability and maximum Lyapunov exponent diagrams and reveals a rich variety of dynamic behaviors not observed before. The model has a complex bifurcation structure involving mixed-mode and bursting oscillations as well as quasiperiodic and chaotic dynamics. In addition, multiple periodic and non-periodic attractors coexist for the same parameters. For some parameter values, the model also reveals formation of mosaic patterns of complex dynamic states. The complex dynamic behaviors exhibited by this model are compared to those of another version of the same model, which has been studied in more detail. The two models show similarities, but also notable differences between them, e.g., the organization of mixed-mode oscillations in parameter space and the relative abundance of quasiperiodic and chaotic oscillations. In both models, domains with chaotic dynamics contain apparently disorganized subdomains of periodic attractors with dinoflagellate-like structures, while the domains with mainly quasiperiodic behavior contain subdomains with periodic attractors organized as regular filamentous structures. These periodic attractors seem to be organized according to Stern-Brocot arithmetics. Finally, it appears that toroidal (quasiperiodic) attractors develop into first wrinkled and then fractal tori before they break down to chaotic attractors.

β-cells retain a pool of insulin-containing secretory vesicles regulated by adherens junctions and the cadherin binding protein p120 catenin

The β-cells of the islets of Langerhans are the sole producers of insulin in the human body. In response to rising glucose levels, insulin-containing vesicles inside β-cells fuse with the plasma membrane and release their cargo. However, the mechanisms regulating this process are only partly understood. Previous evidence indicated reductions in α-catenin elevate insulin release, while reductions in β-catenin decrease insulin release. α- and β-catenin contribute to cellular regulation in a range of ways but one is as members of the adherens junction complex and these contribute to the development of cell polarity in b-cells. Therefore, we investigated the effects of adherens junctions on insulin release. We show in INS-1E β-cells knockdown of either E- or N-cadherin had only small effects on insulin secretion, but simultaneous knockout of both cadherins resulted in a significant increase in basal insulin release to the same level as glucose-stimulated release. This double knockdown also significantly attenuated levels of p120 catenin, a cadherin binding partner involved in regulating cadherin turnover. Conversely, reducing p120 catenin levels with siRNA destabilized both E- and N-cadherin, and this was also associated with an increase in levels of insulin secreted from INS-1E cells. Furthermore, there were also changes in these cells consistent with higher insulin release, namely reductions in levels of F-actin and increased intracellular free Ca²⁺ levels in response to KCl-induced membrane depolarization. Taken together, these data provide evidence that adherens junctions play important roles in retaining a pool of insulin secretory vesicles within the cell and establish a role for p120 catenin in regulating this process.