CommentaryThe Search for Odorant Receptors tide

nally, odorants were reported to depolarize other cell The Question types and to even alter the membrane potential of artifiThe first time I thought about olfaction was when I read cial liposomes (Kashiwayanagi and Kurihara, 1984; Noa 1985 paper from Sol Snyder’s group that discussed mura and Kurihara, 1987). Thus it was not at all clear the unsolved question of how odors are detected in the what kind of proteins the odorant receptors were or, for nose (Pevsner et al., 1985). This paper opened up a that matter, whether they even existed. fascinating new world for me. It was estimated that huI decided to take an unbiased approach with regard mans could perceive 10,000 or more chemicals as havto the structure of odorant receptors and to focus on two ing distinct odors. Even more remarkably, subtle assumptions: first, odorant receptors would be proteins changes in an odorous chemical could dramatically encoded by a family of related genes and, second, odorchange its perceived odor. How could the olfactory sysant receptors would be selectively expressed by olfactem detect such an enormous diversity of chemicals? tory sensory neurons. I first tried an unconventional apAnd how could the nervous system translate this comproach in which I replica screened an olfactory cDNA plexity of chemical structures into a multitude of differlibrary with large amounts of 32p-labeled genomic DNA ent odor perceptions? To me, this was a monumental or brain cDNA. The idea was that clones containing problem and a wonderful puzzle. I was hooked. repetitive sequences would be labeled by both probes As a molecular biologist, the logical first question to whereas clones containing members of an olfactory ask was how the recognition of diverse chemical strucmultigene family would be labeled only by the genomic tures is accomplished in the nose. With this knowledge DNA probe. I also tried a cDNA subtraction approach to in hand, one might then be able to explore how sensory identify genes selectively expressed in olfactory sensory information is organized in the nose and the brain to neurons and, in addition, tried to develop a way of clonultimately yield odor perceptions. It seemed obvious ing genes that were related, but not identical. These from a molecular standpoint that there must be a family efforts yielded some genes that appeared to be specifiof odorant receptors that varied in ligand specificity. It cally expressed in olfactory sensory neurons, but none also seemed that olfactory sensory neurons in the nose belonged to a family, so I set them aside. that detect odorants must express different receptors The Discovery in order for odorants to elicit different signals in the brain Advances in technology often underpin advances in sciand thereby generate distinct odor perceptions. ence, and this was indeed the case in our discovery of The Search odorant receptors. The development of the polymerase In March 1988, I embarked on a search for odorant chain reaction (Saiki et al., 1985), coupled with the disreceptors; this search would prove arduous, but imcovery of a thermostable DNA polymerase (Saiki et al., mensely rewarding. At the time, I had already completed 1988) and the development of programmable thermocya postdoctoral project in Richard Axel’s lab on Aplysia clers (Weier and Gray, 1988), revolutionized molecular neurons. My background was in immunology and I had biological techniques. also been trying to develop a method to identify reIn 1989, an olfactory neuron-specific G protein was arranged genes in the mammalian nervous system, the identified, strengthening the case for a G protein-couidea being that such genes might provide insight into pled mechanism of olfactory transduction (Jones and its cellular and connectional diversity. I was intrigued Reed, 1989). In addition, while the sequences of only by the possibility that gene rearrangement or gene contwo types of G protein-coupled receptors (GPCRs) were version might be involved in the generation of a varied known in 1986, the number had grown to almost 20 by set of odorant receptors or regulate their expression, as 1989, and it was evident that the GPCBs all shared with antigen receptors in the immune system. I became limited sequence motifs and a potential seven transobsessed with finding the odorant receptors and stayed membrane domain structure. That year, it was shown on in Richard Axel’s lab to look for them. for the first time that degenerate oligonucleotide primers I first looked for clues as to the molecular nature of could be used in PCR reactions to uncover new memthe receptors. Odorants were reported to induce GTPbers of protein families, including GPCRs (Libert et al., dependent increases in adenylyl cyclase activity in the 1989; Wilks, 1989). I tried using the published GPCR cilia of olfactory sensory neurons (the apparent site of primer pair, but found only a dopamine receptor. odorant recognition), suggesting a role for G proteins At that point, I decided to conduct an exhaustive and cAMP in olfactory transduction (Pace et al., 1985; search for GPCRs in the olfactory epithelium by using Sklar et al., 1986). Moreover, the cilia had cyclic nucleoa number of different degenerate primers in a combina-