Heterotrimeric G proteins relay information from cell surface receptors to effecters. Each receptor specifically detects an extracellular stimulus (a photon, odorant, pheromone, hormone, or ncurotransmitter) while each effector generates one or more intracellular second messengers. The family of G protein-coupled receptors comprises several hundred proteins, while the less numerous effecters include enzymes and ion channels in several protein families. G proteins transmit signals with high fidelity: each G protein directs the flow of information from a distinctive subset of potential receptors to a limited array of possible effecters. Here we review recent efforts to understand receptor-G protein (RG) and G protein-effector (GE) signaling, with a special emphasis on identifying the structural features of G protein a (Ga) subunits that interact with receptors and effecters. Duplication and divergence of genes during the course of evolution have created structural diversity in each component of the G protein heterotrimer. In mammals, G protein a, p, and y polypeptides are encoded by at least 16, 4, and 7 genes, respectively (Cali et al., 1992; Gautam et al., 1990; McLaughlin et al., 1992; Simon et al., 1991; von Weizsacker et al., 1992). a subunits have received the most attention, because they bind and hydrolyze GTP and because the known structural determinants of RG or GE specificity reside in a subunits. Recent investigations have begun to document contributions of the p and y polypeptides to RG specificity (Kleuss et al., 1992, 1993) and to identify cases in which fly subunit complexes, alone or in cooperation with a subunits, regulate effecters such as adenylyl cyclase (Federman et al., 1992; Tang and Gilman, 1992), phospholipase C (PLC)(Blank et al., 1992; Boyer et al., 1992; Camps et al., 1992; Katz et al., 1992; Smrcka and Sternweis, 1993), and ion channels (Kim et al., 1989; Logothetis et al., 1987) and the pheromone response pathway of Saccharomyces cerevisiae (Whiteway et al., 1989). The large number of possible afir heterotrimers furnishes plentiful opportunities for fine-tuning the specificity of transmitted signals (Birnbaumer, 1992; Simon et al., 1991).

As targets for analysis of structure and function, a sub-units have the advantage of belonging to a protein family with conserved primary structure (500/o-90% identity) and a shared molecular mechanism, but diverse and easily documented signaling functions (Table 1). For this reason chimeric a subunits have made ideal tools for pinpointing regions of primary structure involved in RG and GE interactions. Moreover, a plausible and testable structural model of Ga can be defined on the basis of biochemical, immunologic, and molecular genetic observations, plus