The mammalian olfactory system can recognize and discriminate a large number of different odorant molecules. The detection of odorants results from the interaction with G-protein coupled receptors on the surface of olfactory neurons. In 1991 Buck and Axel discovered a large multigene family encoding seven-transmembrane proteins, that represent these olfactory receptors. An olfactory sensory neuron expresses a small number of receptors, probably only a single one from one allele. Neurons expressing a given receptor are localized in one of four broad zones in the olfactory epithelium, but within a zone their distribution is scattered and appears to be random.
My current work focuses on axon guidance in the olfactory system. My approach is based on gene targeting technology and exploits tau-lacZ as a novel axonal marker. As a first experiment, together with Fan Wang and Richard Axel (Columbia University), I modified the olfactory receptor gene P2 in the germline of mice by inserting immediately 3D of the stop codon the sequences encoding the reporter tau-lacZ. Tau is a microtubule binding protein that, when fused to the histochemical marker lacZ, enables it to be transported down the axon. The translation of the reporter from the bicistronic message is controlled by an internal ribosome entry site (IRES) of viral origin. Olfactory neurons that have chosen - by unknown mechanisms - to express P2, will generate a bicistronic message, from which both the P2 receptor and tau-lacZ are translated. In these mutant mice, neurons expressing P2, including their axons, can be stained blue upon exposure to X-gal. We were able to demonstrate that these neurons converge onto three glomeruli in the olfactory bulb, out of a total of 2,500. From early on in development, the axonal projections of P2 expressing neurons take a path that points towards the future location of the P2 glomeruli. Importantly, the location of these glomeruli displays left/right and medial/lateral symmetry, and is invariant from mouse to mouse, implying the existence of a topographic map in the olfactory bulb.
In another mouse strain, the tau-lacZ reporter sequences were introduced into the P2 locus by targeted mutagenesis in such a way that the P2 coding sequence is deleted. lacZ positive neurons in these mice project their axon randomly over the surface of the olfactory bulb, in striking contrast to the axonal convergence observed in the mice carrying the mutation that leaves the coding region intact. Taken together, these results implicate the olfactory receptor itself in the establishment of axonal convergence. To distinguish between a permissive (indirect) or an instructive (direct) role of the receptor in this process, we are designing additional mouse strains in which the coding sequence of P2 is replaced by that of another receptor.
Use of the tau-lacZ marker requires paraformaldehyde fixation, precluding visualization of live neurons. We are in the process of creating a mouse strain with a P2-IRES-tau-GFP mutation. In these mice, P2 expressing neurons should be identifiable while they are alive by virtue of their green fluorescence. Observing how axons of P2 expressing neurons approach their targets in the bulb will be informative in the formulation of models of axon guidance: Do these axons fasciculate prior to synapse formation ? Do they form synapses in a restricted area, or is there a process of trial and error ? Do the axons contact non-neuronal cells in the bulb ? In addition, the ability to identify and isolate single P2 expressing cells by fluorescence should permit electrophysiological analysis of their response to odorants. This mouse strain could constitute an in vivo assay to find physiological ligands for P2, which remains - like all other olfactory receptors - an orphan receptor.
Finally, FDG staining of live lacZ expressing neurons or fluorescent visualization of GFP expressing neurons may allow us to purify by flow cytometric methods P2 expressing olfactory neurons. A pool of sorted P2 expressing neurons will be used to construct a cDNA library, aided by the Polymerase Chain Reaction. Similarly, a cDNA library will be made from neurons expressing a different receptor, isolated from another mouse strain carrying an IRES-tau-lacZ or IRES-tau-GFP mutation in a different receptor gene. Differential screening of these cDNA libraries should result in the isolation of genes expressed selectively in P2 expressing neurons but not in neurons expressing the other olfactory receptor. Such genes may be involved in axon guidance, gene regulation or signal transduction. Their function can be tested by conventional "knock-out" targeted mutations.