Hyposmotic shock stimulates insulin secretion by two distinct mechanisms. Studies with the betaHC9 cell

Exposure of betaHC9 cells to a Krebs-Ringer bicarbonate-HEPES buffer (KRBH) made hypotonic by a reduction of 25 mM NaCl resulted in a prompt stimulation of insulin release. The stimulation was transient, and release rates returned to basal levels after 10 min. The response resembles that of the first phase of glucose-stimulated insulin release. The response did not occur if the reduction in NaCl was compensated for by the addition of an equivalent osmolar amount of sorbitol, so the stimulation of release was due to the osmolarity change and not the reduction in NaCl. The hyposmotic shock released insulin in KRBH with or without Ca(2+). The L-type Ca(2+) channel blocker nitrendipine inhibited the response in normal KRBH but had no effect in KRBH without Ca(2+) despite the latter response being larger than in the presence of extracellular Ca(2+). Similar data were obtained with calciseptine, which also blocks L-type channels. The T-type Ca(2+) channel blocker flunarizine was without effect, as was the chloride channel blocker DIDS. In parallel studies, the readily releasable pool of insulin-containing granules was monitored. Immunoprecipitation of the target-SNARE protein syntaxin and co-immunoprecipitation of the vesicle-SNARE VAMP-2 was used as an indicator of the readily releasable granule pool. After hypotonic shock in the presence of extracellular Ca(2+), the amount of VAMP-2 coimmunoprecipitated by antibodies against syntaxin was much reduced compared with controls. Therefore, under these conditions, hypotonic shock stimulates exocytosis of the readily releasable pool of insulin-containing granules. No such reduction was seen in the absence of extracellular Ca(2+). In conclusion, after reexamination of the effect of hyposmotic shock on insulin secretion in the presence and absence of Ca(2+) (with EGTA in the medium), it is clear that two different mechanisms are operative under these conditions. Moreover, these two mechanisms may be associated with the release of two distinct pools of insulin-containing granules.