Although syntaxin 1 is generally thought to function as the primary target-N-ethylmaleimide-sensitive factor attachment protein receptor required for pancreatic  cell insulin secretion, we have observed that overexpression of a dominant-interfering syntaxin 4 mutant (syntaxin 4/⌬TM) attenuated glucose-stimulated insulin secretion in HC-9 cells. Furthermore, these cells express the selective syntaxin 4-binding protein Synip (syntaxin 4 interacting protein), and Synip was specifically co-immunoprecipitated with syntaxin 4 but not syntaxin 1. Overexpression of the full-length Synip protein (Synip/wild type) inhibited VAMP2 association with syntaxin 4 and decreased glucose-stimulated insulin secretion. This did not occur with a Synip mutant (Synip/ ⌬EF) that was incapable of binding syntaxin 4. Consistent with a functional role of syntaxin 4 in this process, expression of syntaxin 4/⌬TM also inhibited glucosestimulated insulin secretion. Furthermore, analysis of first and second phase insulin secretion demonstrated that syntaxin 4/⌬TM mainly suppressed the second phase of insulin secretion. In contrast, overexpression of Synip resulted in an inhibition of both the first and second phase of glucose-stimulated insulin secretion. These data demonstrate that syntaxin 4 plays a functional role on insulin release and granule fusion in  cells and that this process is regulated by the syntaxin 4-specific binding protein Synip.It is well established that glucose-stimulated insulin secretion occurs through a complex metabolic network in which the increased flux of glucose in pancreatic  cells elevates the intracellular ATP/ADP ratio (1-5). The increased ATP/ADP ratio results in the closure of the ATP-sensitive potassium channel (K ATP channels) and subsequent cellular depolarization (6, 7). In turn, depolarization causes the activation of the L-type voltage-dependent calcium channels that result in the influx of extracellular calcium. This increased intracellular calcium then serves as a trigger for the initial fusion of insulincontaining dense core granules with the plasma membrane (8, 9).Although the calcium trigger mechanism, trafficking, and fusion of insulin granules have not been fully elucidated, recent studies have suggested that the majority of granules that initially fuse with the plasma membrane are localized beneath the plasma membrane in a bound (docked) state (10 -12). This pool of insulin granules has been referred as the readily releasable pool, and it appears to account for the first phase of stimulusmediated insulin secretion (13). This initial release process requires the functional interactions of the t-SNARE 1 proteins, syntaxin 1 and SNAP25, with the v-SNARE protein, VAMP2. For example, the use of SNARE mutants or various SNARE protein-specific proteolytic toxins prevent regulated insulin secretion (14 -17). However, since these endotoxins do not proteolyze complexed SNARE proteins, it is thought that the plasma membrane-bound (docked) insulin granules first undergo a priming step (core SNARE complex...