Dissecting the molecular mechanisms of olfaction in a malaria-vector mosquito.

H malaria is widely endemic in tropical and subtropical regions of the world, where ca. 1.5 billion people are at risk, ca. 500 million clinical cases occur, and 1–3 million deaths, mostly of children, are due wholly or in part to the disease (see, e.g., http: www.malaria.org ). All of the species of Plasmodium that infect humans and cause malaria are transmitted by mosquitoes of the genus Anopheles. The African malaria mosquito, Anopheles gambiae, is especially dangerous owing to its dramatic tendency to feed on humans (anthropophily) and resulting extraordinary efficiency as a vector of the most deadly of the parasites, Plasmodium falciparum (1). In mosquitoes, host-seeking and selection are mediated by volatile chemicals emanating from the host (2). Thus, the likelihood that the anthropophily and high vectorial capacity of A. gambiae are based on olfactory cues has stimulated interest in mosquito olfaction. Two papers published in PNAS (including one in this issue) begin to dissect the molecular mechanisms that mediate olfactory sensory transduction in the antennae of A. gambiae. The recent paper by Fox et al. (3) describes the identification and characterization of a family of G-protein-coupled receptors (GPCRs) that are thought to be the first identified mosquito odorant receptors (ORs). The paper by Merrill et al. in this issue (4) documents the cloning and characterization of an arrestin involved in the regulation of the olfactory response in A. gambiae. Both of these classes of proteins represent key parts of the signaling machinery that results in the response of antennal olfactory receptor cells (ORCs). The ORs, deployed in the surface membrane of the apical dendrites of ORCs, are believed to bind odorants and begin the signaltransduction cascades. Thus, understanding the nature, distribution, and function of the ORs is essential for understanding odor discrimination and sensitivity. Since the cloning of putative ORs from rodents by Buck and Axel in 1991 (5), the identification and function of these receptors have been studied vigorously. On the order of a thousand different ORs may be expressed in mammals, and a smaller number expressed in related vertebrates (6). With few exceptions, these receptors fit into the superfamily family of heptahelical GPCRs (7). The basic transduction mechanism subserved by these receptors is widely conserved within the animal kingdom and is thought to be essentially the same in vertebrates, nematodes, and insects (8). Binding of odorants to the receptors, with or without mediation by soluble odorant-binding proteins (9), triggers the activation of a G-protein and the production of a second messenger, either IP3 or cAMP (10). These second messengers, in turn, typically cause the opening of membrane ion channels and the depolarization of the ORC (11). Despite this conservation of general mechanism, and the conservation of heptahelical structure, the amino acid sequences of ORs can be widely divergent. Although they cluster into related families (12), there can be as little as 8% identity between ORs in the same organism. As a consequence, approaches to cloning insect ORs based on sequence similarity have failed. The emergence of genome-level analyses has enabled the identification of ORs in insects, for example in Drosophila melanogaster, in which analysis of the genomic sequence yielded approximately 40 candidate OR genes (13–15). One of these putative receptors, Or43a, has been studied in two different functional-expression systems, and in both cases it mediated sensitivity to cyclohexanone, cyclohexanol, benzaldehyde, and benzyl alcohol (16, 17). The functional-expression data confirm that at least a subset of the candidate receptors identified through the Drosophila genome project are in fact ORs. Analyzing just 5% of the genome of A. gambiae, Fox et al. (3) identified four candidate OR genes that code for heptahelical proteins and have olfactory-specific expression patterns. One of these proteins shows significant (36%) sequence identity to the functionally expressed Drosophila OR, OR43a, suggesting that this family of sequences also represents functional ORs in A. gambiae. Notably, the mRNA level for one of the candidate receptors, AgOR1, is down-regulated after blood feeding, paralleling the down-regulation of olfactory sensitivity associated with blood feeding. In sum, these findings suggest that these four genes may encode functional ORs. If so, a simple extrapolation suggests that A. gambiae may express about 80 different ORs. Although much work yet must be done to isolate additional receptors and to characterize their expression and odorant sensitivity, the work of Fox et al. is an important step toward understanding the mechanisms by which A. gambiae detects its human hosts. It is also clear, however, that even a complete understanding of the expression patterns and chemical specificity of the ORs would not explain the functioning of the peripheral olfactory system. The receptors must function in the context of the sensory machinery present in the ORCs, including the downstream components of the signaltransduction pathway, the mechanisms of adaptation of the response, and the potential variability in these systems from cell to cell. The work reported by Merrill et al. (4) in this issue begins the characterization of that signal-transduction machinery, and a mechanism for adaptation, by cloning and characterizing an arrestin, AgArr1, from A. gambiae. Arrestins are involved in the desensitization and internalization of GPCRs. Phosphorylation of a GPCR by a GPCR-kinase triggers the binding of arrestin. The arrestin classically has two functions: first to interfere with G-protein binding, thus terminating the immediate response to the ligand, and second, to interact with dynamin and other proteins to trigger the endocytotic internalization of the receptor (18). Both mechanisms are impor-