Defining the chemical space for ligands of odorant - binding proteins

The last Nobel Prize for Medicine was awarded to Linda Buck and Richard Axel for their work on the sense of smell, called olfaction, in humans. Some aspects of human olfaction are difficult to study, but in insects, where olfaction is even more important, the systems involved are more accessible to investigation. Although there are some differences between olfaction in mammals and insects, in many important ways they are similar and for this study, we propose to work on the olfactory system of the fruit fly, Drosophila melanogaster. This insect is very amenable to our proposed work because all the genetic information for this insect has been assembled and a great deal is known about the influence of olfaction on its behaviour. Also, we know almost all of the individual smells (about 90), each of which comprises a specific small molecular weight chemical, which it recognises. These small molecular weight chemicals can be brought together into libraries for the investigations planned here. Because of the complete genetic information that we have for this insect, we also know all of the genes involved in the olfactory processes and we have selected one set of these genes which code for proteins involved in the first stage of the olfactory process. These proteins are called odorant- binding proteins (OBPs) and these interact with particular members of the chemical libraries of individual smells that we will create. We plan to search these libraries with individual OBPs to find out which smells interact with which particular proteins. Although we have almost all of the smell compounds used, we will also search the much larger collectionss of other natural compounds that the fly may interact with for any active small molecule chemicals that have been missed. For these collections, each containing hundreds of compounds, we will take smells from situations in which the fly is likely to have an interest, for example, from flies of the same species or from their food, which is rotting fruit. Having linked individual smells to particular OBPs by allowing the protein to absorb individual compounds from our libraries and natural product collections, we will design ranges of chemicals (about 30 for each of the smells that we have matched to an OBP) that are similar to the natural smell compounds but with some structural differences. If these compounds still interact with the protein, and whether this interaction is stronger or weaker, will allow us to work out exactly what shape and electronic charge characteristics the smell molecule must have to interact successfully with the protein. This will allow us to make alterations to the natural smells, which could be used to fool these insects and other pest insect species and offer new ways to control insect pests. In addition, we will investigate whether we can use OBPs to detect small molecular weight compounds that act as key indicators of commercial or other interest. For example, when food begins to deteriorate, some small molecules are produced which are extremely difficult to detect amongst the large amounts of small molecules normally produced by this material, but a device based on OBPs could do this and in other examples, such as the detection of illegal drugs and explosives.

This proposal seeks to identify the chemical space necessary for an olfactory stimulus to interact specifically with the olfactory recognition system at the molecular level. The proposal will study the olfactory system of the fruit fly, Drosophila melanogaster, which, because of the availability of the full genome sequence and considerable knowledge of the nature and role of olfactory signals, will present a highly amenable target that is representative of animal olfaction and possibly of biological molecular recognition more generally. Although the genes involved in the olfactory processes of D. melanogaster are all annotated as such, there is as yet no link for most of these genes with their olfactory ligands. This work will take libraries of all known olfactory ligands, comprising semiochemicals and including pheromones (ca. 90), which by acting as signals influence the behaviour and development of the insect, and sequentially present these to individual odorant-binding proteins (OBPs) which form the first stage in the process of olfactory recognition in insects and are analogous to related proteins in vertebrates. It is expected that these libraries will contain almost all of the natural semiochemical ligands for D. melanogaster. However, more extensive collections of natural products containing hundreds of compounds (extracted from the main materials with which this insect interacts, i.e. its own kind, and food and oviposition site material being deteriorating fruit) will also be searched for any other compounds which interact with the OBPs. Each time the ligand, or in some cases ligands, for specific OBPs are identified, a new range of homologues and other analogues will be constructed (ca. 30per binding ligand) so that the chemical space for this particular protein-ligand interaction can be determined. The first identification of ligand protein interaction will be done by the absorption or subtraction from a library or collection of natural metabolites by interaction with the protein as measured by high resolution gas chromatography. Confirmation of ligand interaction will be by electrospray mass spectrometry and high resolution nuclear magnetic resonance spectroscopy. Ligands will be labelled with NMR active isotopes (13C, 2H, 19F) and NMR spectroscopy will be used to investigate binding between labelled ligands and proteins, using detection of the isotopes. This will allow us to determine whether slow or fast exchange binding occurs in solution. Producing recombinant OBPs in the presence of a 15N source will allow detection of the OBP by 15N/1H heteronuclear single quantum coherence 2-dimensional experiments. Chemical shift perturbations of amino acids in the binding site by addition of the ligand will be detected and the amino acids identified by characterisation of part of the OBP. This will identify the amino acids that define the chemical space for small molecule chemical binding. Detailed knowledge of which ligands fulfil the chemical space requirements of the natural ligand will allow new semiochemical-based strategies for pest control to be developed, including against agricultural crop pests and insects causing nuisance and transmitting pathogens to human and farm animals, and to the exploitation of beneficial insects such as the parasites and predators of pests. Knowledge of gene products recognising specific novel and anthropogenic small molecules will allow attempts to devise new biosensors based on immobilised natural or synthetic OBPs for the detection of key compounds associated with food safety and quality, illicit materials and other aspects of community and state security.