The protein families that mediate vesicle trafficking are conserved through phylogeny from yeast to humans, as well as throughout the cell from the endoplasmic reticulum to the plasma membrane (4). The membrane fusion reaction requires integral membrane proteins termed soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs). SNAREs function through the association of their cytoplasmic domains to form coiled coils. The Rab family of low-molecular-weight GTPases is also believed to regulate the processes of membrane targeting and fusion (30). Rab proteins are molecular switches that cycle between a GTP- and membrane-bound, active state and a GDP-bound, cytosolic, inactive state. These two forms of Rab are converted from one to the other by the action of GDP/GTP exchange factor and GTPase- activating protein. Active GTP-bound Rab is thought to function by binding to effector proteins associated with specific membrane components. In several systems, Rab proteins and their effectors have been shown to mediate a SNARE-independent initial association of membranes, termed tethering, which is the first step of the docking stage. Furthermore, subsequent trans-SNARE pairing may also be regulated by Rab proteins and their effectors. For example, a complex of Rab5 and the endosome-tethering protein EEA1 regulates endosome fusion by oligomerizing with syntaxin 13, a component of the SNARE machinery (24). Each vesicle-trafficking step involves a set of different members of the Rab and SNARE protein families localized to distinct membrane compartments. Among these trafficking routes, multicellular organisms have developed a highly regulated exocytosis mechanism that occurs only in response to an intracellular Ca2+ signal generated by secretagogues. These secretory cells contain the special vesicles to store and release secretory products. Recent progress in biochemical studies of synaptic vesicles at nerve termini has revealed a number of molecules possibly involved in these processes. To date, Rab3 and its putative effector proteins, rabphilin3 and RIM, are the best candidates for exocytotic Rab and Rab effectors in neurons (20). Although large dense-core granules, another type of secretory vesicles in endocrine and exocrine cells, have been less well characterized biochemically due to their low abundance, recent evidence suggests that Rab3 also functions in endocrine cells such as adrenal chromaffin cells (18) and pancreatic beta cells (32). rabphilin3, however, is not expressed in pancreatic beta cells (32), whereas it is specifically expressed in brain and adrenal gland tissues and is associated with synaptic vesicles at nerve termini. Thus, it is conceivable that any Rab and/or effector proteins may act as regulators for insulin secretion. Recently, we identified a novel rabphilin3-like protein, granuphilin, in pancreatic beta cells (37). Similar to rabphilin3, this protein contains an amino-terminal zinc-finger motif and carboxyl-terminal C2 domains. Granuphilin, however, differs from rabphilin3 in a number of ways, including tissue distribution and subcellular localization (37). Granuphilin is expressed in pancreatic beta cells and pituitary tissue but is not significantly expressed in other major organs such as the brain. It is associated with dense-core granules but not with synaptic-like microvesicles in pancreatic beta cells. The unique distribution pattern of granuphilin suggests that it may be involved in the regulated exocytosis of dense-core granules in these endocrine tissues. In the present study, we report that Rab27a interacts with granuphilin in beta cells. In addition, Rab27a and granuphilin show remarkably similar tissue and subcellular distributions. Moreover, overexpression of Rab27a in beta cell lines significantly enhances high K+-induced insulin secretion. Identification of a novel complex, Rab27a/granuphilin, should help elucidate the mechanism of regulated exocytosis in endocrine cells, which thus far has been poorly characterized.