Glucose-stimulated insulin secretion: the hierarchy of its multiple cellular and subcellular mechanisms
- 1.7k Downloads
Glucose-stimulated insulin secretion is ensured by multiple molecular, cellular and tissue events. In this issue of Diabetologia, Low et al (DOI: 10.1007/s00125-013-3019-5) have taken an important new step towards understanding the hierarchical organisation of these events, by monitoring in vitro the individual exocytosis of multiple beta cells within intact mouse islets. The authors show that glucose stimulation markedly increases the number of exocytotic events per cell and, to a lesser extent, the number of beta cells contributing to this event. In this commentary we discuss these novel observations and propose that metabolic and electrical coupling of islet beta cells is responsible for a more homogeneous glucose-induced secretory response of cells in an intact islet as compared with isolated beta cells.
KeywordsBeta cells Confocal microscopy Connexin Cx36 Exocytosis Glucose Heterogeneity Insulin release Pancreatic islets Secretory granules Stimulus–secretion coupling
Glucose-stimulated insulin secretion is the hallmark of a differentiated beta cell, and the early loss of this feature is a prominent event in the pathogenesis of type 2 diabetes [1, 2]. A wealth of data has identified a number of proteins, organelles and cells that interact during this event. Still, the relative contribution of each of these factors, and their hierarchical organisation for the proper control of insulin secretion, remain open questions.
In this issue of Diabetologia, Low et al  have taken an in vitro step towards understanding this process. By combining a high spatial and temporal resolution two-photon microscope and an extracellular membrane-impermeant tracer that labels secretory granules upon their fusion to the cell membrane , they simultaneously monitored the individual exocytosis activity of multiple beta cells within mouse islets exposed to various glucose concentrations . Having documented that their approach provides estimates consistent with the actual glucose-induced release of insulin (yet not measured from exactly the same islets which were monitored microscopically), they show that glucose stimulation markedly increases the number of exocytotic events per cell and, to a lesser extent, the number of beta cells contributing to this event. Strikingly, both changes were most conspicuous in islets acutely exposed to 8 mmol/l glucose, a concentration close to the physiological postprandial levels of the sugar . First, these findings confirm the usefulness of this technology for evaluating the functioning of individual beta cells within the native environment of pancreatic islets . Second, they provide further support for the previous observations that insulin secretion does not take place simultaneously and at the same rate in all beta cells [6, 9], and that glucose stimulation promotes the recruitment of actively secreting beta cells [3, 5, 6, 7, 8, 9]. The findings strengthen the view that beta cells display intrinsic functional differences that decrease with glucose stimulation [3, 4]. The mechanism of this recruitment remains to be understood. Given that Cx36 channels ensure the electrical and metabolic coupling of stimulated beta cells and the intercellular synchronisation of their glucose-induced [Ca2+] transients [4, 10, 11], it is likely that Cx36 signalling is implicated (Fig. 1).
The noteworthy novel finding of the study by Low et al  is that the incidence of exocytosis of secretory granules is markedly increased by raising the glucose levels from a resting to a postprandial level, and shows a more modest increment with a further increase in environmental glucose . The data identify the frequency of exocytosis as a primary physiological glucose effect, even though they do not unravel the underlying molecular mechanism and they do not identify the type of granule or the region of the cell membrane involved in their release. Further investigations may indicate how the graded and exquisite sensitivity to glucose is selectively lost in the beta cells of type 2 diabetic patients [1, 2]. In turn, this understanding would be expected to stimulate the development of novel targeted drug and cellular therapies.
Still, there are several aspects of the study by Low et al that should be taken into consideration when interpreting these findings. First, the study is on isolated islets , leaving the question of whether a similar pattern of exocytosis takes place in vivo to be answered. Further technological developments in conjunction with methods to stimulate individual cells within living animals may help to address this critical issue. Second, and in contrast to previous reports , the authors exclude significant effects of glucose on the dynamics of granule opening and recapture. This conclusion may be premature as the experiments were conducted at 34°C, on islets exposed to only one glucose concentration, and using a monitoring method which, the authors concede, had a time resolution that may not consistently allow the detection of subtle dynamic differences . Third, Low et al propose that the glucose effect on the incidence of exocytosis largely predominates over that on beta cell recruitment. This conclusion should be re-evaluated, taking into account the fact that recordings were made at the periphery of the islets, a region where beta cells are stimulated by glucose at a slower pace than those in the centre of the islets , and that the two-photon microscope method imaged only 10% of the total membrane area of the few dozen beta cells monitored in each recording . Thus, the actual recruitment of the actively secreting beta cells may have been undervalued. This possibility could be tested by combining the novel exocytosis recordings with markers that differentiate between the highly and poorly secreting beta cells, including under conditions such as pregnancy, in which plasticity of the functional beta cell mass is required to accommodate variations in insulin needs . At the end of their paper Low et al comment that the observations made in wild-type murine islets may not apply to human islets , because of the several cytological and regulatory differences that have been documented between the two species [16, 17, 18]. This is a possibility, even though several similarities have been shown between the islets of the two species, notably with regard to several steps of the glucose-secretion coupling mechanism [1, 2], the secretory heterogeneity of individual beta cells , their expression of Cx36 channels  and their synchronisation within isolated islets . As noted in the conclusion of the paper by Low et al , repeating the studies using human islets could help to clarify whether the glucose-induced effects differ in rodents and humans, which may suggest that the differences are dispensable evolutionary modifications, or are similar in the two species, which would argue for some critical, if not obligatory, mechanism.
Supported by grants from the Swiss National Science Foundation (310000-141162, IZ73Z0_127935, CR32I3_129987), the Juvenile Diabetes Research Foundation (1-2010-393, 40-2011-11, 5-2012-281), the European Union (BETAIMAGE 222980; IMIDIA C2008-T7, BETATRAIN 289932), the Flemish Fund For Scientific Research (G.0672.12N) and the KU Leuven (GOA/14/010).
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
The two authors were responsible for the conception and design of the manuscript, drafting the article and revising it critically for important intellectual content. Both authors approved the version to be published.