Odorant receptor (OR) genes were discovered by Buck and Axel (1991). Around 1,100 ORs are expressed in mouse olfactory sensory neurons (OSNs) in a monoallelic and monogenic manner (Chess et al., 1994). These OSNs are located in the main olfactory epithelium and project their axons to the olfactory bulb (OB) where they form homogeneous glomeruli. The glomeruli are formed at conserved positions in the OB. Several laboratories have studied the mechanisms that guide the OSN axons to their target where they form synapses with the mitral cells that convey the olfactory signal to other brain regions. Several theories have emerged to explain these mechanisms, but none has yet achieved full explanation for the precise axonal wiring in the mouse olfactory system.

This thesis describes three approaches for gaining further insight into the mechanisms that govern axonal wiring. In the first approach a gene-targeting vector was designed to target the M71 locus and knock in the constitutively active Gαs subunit (caGs) in M71-expressing OSNs. It has been reported that OSNs expressing caGs shift their glomerular position to a posterior position in the OB. However this shift was shown only with transgenic mice. A gene-targeted expression from the M71 locus would have provided a better understanding, because it is closer to the native mechanism. Unfortunately the assembly of the targeting vector was more difficult than anticipated and the project was suspended after a year.

The second approach was an empirical study of the fundamental components of the olfactory system. Green-fluorescent OSNs in 11 gene-targeted strains coexpressing GFP with a specific OR gene were evaluated quantitatively. It was shown that their OSN number is proportional to the respective total glomerular volume (TGV). These data enable new methods of quantifying and qualifying the properties of the OSN population and allow an estimate of the OSN numbers from the easier to measure TGV, which is much easier to measure than counting cells. With the first broad collection of complete datasets of the number of OSNs, several other results could be extrapolated. The full datasets allowed an in-depth analysis of the sampling error for the 11 strains and established a well-grounded threshold for sampling of the main olfactory epithelium from every fifth section. I also demonstrated that these 11 strains have different stabilities of expression in terms of numbers of labeled OSNs, measured by the coefficient of variation (CV), with the strain MOR23-IRES-tauGFP being the most stable. It was also shown that most of the strains tested have a unique expression pattern along the anterior-posterior axis, except for OR-expressing OSNs in the classical zone 1, which displayed a similar expression pattern. For this study several reporter genes were expressed from the M71 locus, in addition gene replacements and downregulation of M71 were measured as well. These strains showed differences in the OSN numbers compared to M71-IRS-taulacZ expressing OSNs. The latter being expressed three times more than in the strains expressing fluorescent marker proteins. However each strain maintained the characteristic expression pattern for M71.

The final approach discussed in this thesis is the influence of a conditional Neuropilin-1 (Nrp1) knockout on axonal wiring of the mouse olfactory system. It has been reported that the position of a glomerulus in the OB along the anterior-posterior axis is determined by a gradient of Nrp1. However this finding was in mice expressing a transgene. Here an approach was chosen to generate a better test of this hypothesis. Conditional knockouts of Nrp1 in M71-expressing OSNs were generated, exploiting the unusual feature of monoallelic expression of ORs to provide a control in the same individual mouse. The results show not a gradual shift of glomeruli to the anterior OB as reported previously (Imai et al., 2009), but the formation of ectopic glomeruli in the anterior bulb, resembling the effect an adenylyl cyclase 3 (AC3) knockout has on M71-expressing OSNs. This phenotype is established directly at the onset of glomerular formation and is maintained during adulthood. These finding leads to the conclusion that the signal transduction cascade of OSNs is indeed linked to the expression of Nrp1, but a simple Nrp1-gradient hypothesis cannot explain the phenotype M71-expressing OSNs.