Transport and release of insulin granules during biphasic insulin secretion

Sammanfattning: Glucose-evoked insulin secretion exhibits two phases: The first phase represents exocytosis of insulin granules situated at the plasma membrane, whereas the second phase reflects insulin release from granules continuously recruited to the plasma membrane. This thesis investigates different aspects of phasic insulin secretion, with the specific aims to elucidate the possible contribution of insulin granule transport, as well as to study the influence of the nucleotides cAMP and NADPH to this process. Time-lapse confocal imaging revealed that insulin granules move by an interplay of directed and random movements. The directed movements, representing kinesin-mediated transport of insulin granules along microtubules, constitute an efficient mechanism to transfer granules, whereas random movements are slow but essential for facilitating the interaction between granules and motorproteins. Glucose accelerated granule velocities by ~50%, but had only modest effects on the frequency of directed events. In contrast, cAMP elevated the number of translocations by ~90%, but had no effect on average granule velocity. This indicates that granule mobility is a regulated process and that different stimulators of insulin secretion influence granule mobility via distinct mechanisms. Near the plasma membrane granules are transported through the actin web via myosin 5a-mediated translocations. Silencing of myosin 5a expression using RNAinterference inhibited hormone secretion by 46% under stimulatory conditions. Myosin 5a-mediated transport was accelerated by glucose during late phase of insulin secretion, supporting the view that granules released at this stage are recruited from the inner side of the actin network. The possible role of NADPH and cAMP for granule transport and phasic insulin secretion was also investigated. Capacitance recordings of insulin exocytosis revealed that physiological concentrations of NADPH stimulated insulin release by 84% and 102% in mouse and rat beta-cells, respectively. NADPH was also found to increase in response to glucose, which suggests NADPH as a possible candidate metabolite for the amplifying pathway of glucose-evoked insulin secretion. Furthermore, infusion of NADPH in combination with the redox protein glutaredoxin stimulated insulin secretion by additional 54%, indicating that glutaredoxin acts as an NADPH effector. Finally, we demonstrated that the nature of cAMP-signals are transient (~40 s) but in spite of this are able to exert a time-dependent potentiation, or memory effect, on insulin secretion and granule mobility. We speculate that this potentiation might be due to elevated recruitment of new granules to the release sites at the plasma membrane. In conclusion: (1) Physical translocation of insulin granules is a regulated process that is stimulated during second phase insulin secretion. (2) Kinesin 1- and myosin 5a-mediated translocations in combination with diffusional movements are essential for efficient recruitment of new granules to the plasma membrane. (3) The potentiating effect of cAMP on insulin secretion might at least in part be a result of elevated granule transport. And finally (4) NADPH is a candidate for mediating the stimulatory effect of glucose on second phase insulin secretion.