Date of Degree


Document Type


Degree Name





Kevin Ryan

Committee Members

Paul Feinstein

Khosrow Kashfi

Bin Tian

Hoau-Yan Wang

Subject Categories

Biochemistry | Molecular and Cellular Neuroscience | Molecular Biology


Olfactory code, odorant receptor antagonism, vibrational theory, cleavage and polyadenylation, pre-mRNA processing, pre-mRNA, 3’ end formation


Mammals can detect and discriminate uncountable odors through their odorant receptors. To accommodate the countless and diverse odors, exceptionally large numbers of odorant receptor (OR) genes are expressed in mammals. In addition, the mammals utilize a combinatorial code, where an odorant molecule can activate multiple ORs; an OR also responds to a set of multiple odorants. In nature, an odor is often a complex mixture of multiple odorant molecules. The combination of the ORs activated by each constituent generates the unique olfactory code for the particular odor.

Some odorants can antagonize select ORs, as discussed in Chapter 1. An antagonist within an odor mixture can conceivably affect the olfactory code of the odor mixture by inhibiting other constituents that alone would function as agonists. While it is clear that some odorants can be the antagonists of select odorant receptors, the degree to which odorant antagonism contributes to the olfactory code of naturally occurring odor mixtures is unknown. As described in the following chapters, my studies aimed to obtain better understanding of the odorant antagonism at the molecular level, using populations of dissociated primary mouse olfactory sensory neurons and calcium imaging.

Firstly, we probed for the prevalence of odorant antagonism in a naturally occurring odor, using charred wood odor mixture as an example. This mixture was chosen because its constituents have structural similarities between each other, as described in Chapter 2; the odorant antagonists that have been identified thus far show structural similarities to their cognate agonists. The results of this study suggested that the role of odorant antagonism is insignificant in encoding of the charred wood mixture; each constituent contributed additively to make up the olfactory code of the mixture.

The structure and conformation of the odorants are important for OR activation and antagonism. Conversely, the vibrational theory of olfaction states that the activation of the OR depends on the intramolecular vibration of the odorant.1, 2 Previously, numerous behavioral and molecular studies have suggested that the insects could differentiate the deuterated odorants from the non-deuterated odorants, in support of the vibrational theory, although the structure of the insect ORs are unrelated to that of the mammalian ORs. We posited that if the deuterated odorants could bind but fail to activate the ORs that are activated by the non-deuterated odorants, the deuterated odorants could serve as antagonists. However, the psychophysical studies on humans have shown mixed results. A recent study using a set of modified mouse and human ORs expressed in a heterologous system showed that the deuterated and non-deuterated odorants activated the cognate recombinant ORs with equal potency, arguing against the vibrational theory. Herein, we tested this controversy on the primary mouse olfactory sensory neurons using calcium imaging, where the un-modified, endogenous ORs are expressed along with the endogenous downstream signaling molecules. As described in Chapter 3, at the concentrations above the detection threshold, the individual mouse ORs could not differentiate the isotopologues. At the molecular level, there was no evidence of vibrational mechanism of OR activation. Our findings confirmed the implausibility of the vibrational theory of olfaction in the primary mouse OSNs. Consequently, the deuterated odorants were not suitable as odorant antagonists. In this chapter we also point out that replacing H with D in small molecules decreases the hydrophobicity of the molecule. The hydrophobic effect contributes to ligand-receptor binding, and may explain differences observed in insect studies.

Some odorants are known for their synergistic ability to enhance the percept of other odorants when used in a mixture.3, 4 One such example is methyl dihydrojasmonate (MDHJ), which itself has a weak floral odor.4-6 While its odor is weak on its own, MDHJ is known for its ability to “boost” the floral character of other odorants.4 Published molecular level studies on this widely known observation are lacking. To address this deficiency, and as described in Chapter 4, we examined how MDHJ affects the olfactory code of a floral odor mixture, using a rose oil odor mixture as an example. Our results revealed that when added to the mixture, MDHJ can activate additional OSNs, aside from the OSNs that are activated by the rose oil odor constituents. MDHJ also inhibited some ORs that are otherwise activated by the rose oil odor constituents. Our results suggested that MDHJ can “fine-tune” the olfactory code of the rose oil odor mixture by not only by activating ORs but also through odorant antagonism. However, a control experiment indicated that this property is likely not unique to MDHJ.

Also described in this thesis is some preliminary work aimed at profiling the 3’ untranslated regions of the odorant receptor genes. The mRNA 3’ untranslated regions (3’ UTR) often contain the gene regulatory elements such as AU-rich elements and binding sites for miRNA and RNA binding proteins.7 Thus the 3’ UTR can provide dynamic regulation of gene expression. While the importance of 3’ UTR in gene regulation is in general well-established, the 3’ ends of the odorant receptor genes have been incompletely annotated. In relation to this side project, the methods that can be used to target 3’ UTR of the odorant receptor genes for profiling were reviewed, and a technical study of in vitro 3’ processing completed. (Chapters 5 - 7)