Elimination of a ligand gating site generates a supersensitive olfactory receptor

The interaction of odors with their cognate receptors constitutes one of the most complex ligand/receptor binding problems in biology due to the sheer quantity of potential odor molecules facing a limited albeit huge number of different olfactory receptors which in some species comprise close to 10% of all proteins1,2. The tuning width of these receptors is extremely variable, with odor spectra ranging from exceedingly broad3,4 to monospecific5. In some cases a single functional group of the ligand dominates the specificity of the ligand/receptor interaction, in other cases an ensemble of chemical features is recognized6,7,8. However, only in very few cases do we possess a molecular understanding of the binding interaction between an odorant and its receptor9. So far crystal structures are not available for any olfactory receptor, and thus prediction of olfactory receptor structures has relied on modeling studies using established templates such as the beta-adrenergic receptor (β-AR)10 and rhodopsin11, further supported by site-directed mutagenesis and subsequent functional analysis of mutant receptors12. We have used a similar approach to unravel the ligand interaction of a zebrafish olfactory receptor specific for aliphatic diamines, TAAR13c6. The trace amine associated receptor (TAAR) family is the only olfactory receptor family that is much larger in teleost fish compared to tetrapods, suggesting an essential role for TAARs in fish13. Zebrafish possess 112 taar genes, compared to only 15 in mouse and even less in the amphibian and avian lineages6. Since zebrafish serve as a model system for vertebrates, and their olfactory system is qualitatively similar to that of vertebrates including mammals14, zebrafish are well suited to gain deeper insight into vertebrate olfactory receptor properties. We have recently shown TAAR13c to be a highly sensitive and specific receptor for the death-associated odor cadaverine6, which emanates from carrion via bacterial decarboxylation of lysine. Cadaverine is strongly repulsive for humans and, interestingly, it also elicits strong innate aversive behavior in zebrafish6. At low concentrations of cadaverine mostly TAAR13c-expressing neurons get activated suggesting that a single olfactory receptor might suffice to generate a powerful odor-driven behavior6. Here we aimed to understand the molecular basis at the very beginning of this neural circuit, i.e. the interaction of TAAR13c with its native ligand cadaverine. We have performed thorough modeling of the TAAR13c receptor to identify potential binding site residues, and found all of them clustering in the upper third of the transmembrane domains of TAAR13c. We mutated several of these candidate residues and compared the activation of mutant to wildtype receptors in a heterologous cell expression system. Two aspartates, Asp1123.32 and Asp2025.42, buried in the plane of the membrane, were identified as essential components of an internal binding site for cadaverine. Another aspartate, Asp2796.58, was found to constitute an essential residue of a second binding site located at the extracellular surface of the receptor. Both conservative and non-conservative substitutions of Asp2796.58 generated supersensitive receptors. Based on our modeling data we suggest the external binding site to act as a gate, which cadaverine has to pass on its way to the internal binding site. As long as the external binding site is occupied, the gate is closed, and thus limits the free access of cadaverine to the internal binding site. This constitutes a novel molecular mechanism for regulating ligand access to the activating binding site. To the best of our knowledge, such a gating mechanism has not been suggested for any olfactory receptor of any species so far.