Theses Doctoral

Complex Encoding of Olfactory Information by Primary Sensory Neurons

Xu, Lu

The encoding of olfactory information starts from the interaction between odorant molecules and olfactory sensory neurons (OSNs). In mouse, one mature olfactory sensory neuron (OSN) almost exclusively expresses one out of ~1,000 odorant receptors (ORs). The relationship between odorants and ORs is promiscuous: one odorant can activate multiple ORs and one OR can be activated by many odorants.

This combinatorial olfactory coding scheme is fundamental, but not sufficient to fully understand the peripheral encoding of odor mixtures. Almost all naturally-occurring smells consist of many different odorous compounds; for example, the perception of rose is composed of (-)-cis-rose oxide, beta-damascenone, bata-ionone and many other odorants. It is well appreciated in psychology and perfumery that different components in an odor blend can affect each other, producing modulation effects. However, these effects are often considered to be the results of higher center processing, while odor interactions at the peripheral level have not been comprehensively measured.

To evaluate peripheral neuronal responses to odor blends, it is necessary to profile the response patterns of a large population of OSNs while the responses of each individual OSN can be resolved. Conventionally, this has been achieved by imaging OSNs acutely dissociated from the olfactory epithelium with a regular epi-fluorescent microscope. In Chapter 2 of this thesis, such method was utilized to characterize the response patterns of three groups of bio-isosteres. This study reveals that OSNs discriminate odors primarily based on their topological properties rather than chemical properties.

Chapter 3 investigates the modulation effects of Hedione, a chemical that has been widely used in perfumery for 60 years. Hedione is psychophysically known as an enhancer that brings up the volume of floral and citrus odors, but the underlying mechanism remains largely unknown. Our study showed that Hedione could both enhance and inhibit odor responses in peripheral neurons, with inhibition being the dominant effect. Moreover, dose-dependent analyses have shown that odorant receptors with lower binding affinity are more prone to inhibition, leading to the hypothesis that Hedione may act as a weak antagonist, which highlights the scent of the leading compound through contrast enhancement.

However, the cell imaging method in Chapter 2 and 3 was limited by the low throughput (200 cells per field of view) and cell damage during digestion. Utilizing a new advance in microscopy, Swept Confocally Aligned Planar Excitation (SCAPE), I was able to perform 3D volumetric imaging on the intact olfactory epithelium of OMP-CRE+/-GCaMP6f-/- mice with a perfused half-head preparation. This method is capable of recording over 10,000 OSNs simultaneously with high spatial and temporal resolution. The process of establishing the imaging protocol and data analysis pipeline has been detailed in Chapter 4.

Chapter 5 characterizes OSN responses to odor blends using the SCAPE microscopy. A large number of responding cells showed inhibited or enhanced responses to odor mixtures compared with responses to each individual component. Eight structurally and perceptually distinct chemicals were tested, all shown to act as antagonists or enhancers to some extent. Compared with a monotonically additive coding scheme, the presence of widespread modulation effects could diversify the output, thereby increasing the capacity of the olfactory system to distinguish complex odor mixtures.

Taken together, these results show that olfactory information is subject to widespread modulation in the olfactory epithelium. This unusual complexity at the primary receptor level implies an information coding strategy different from those utilized by visual and acoustic systems, where complex interactions among stimuli only occur at higher levels of processing. Further experiments are needed to explain the mechanisms at the molecular level and to link peripheral neuronal responses to psychophysics and behavior.


  • thumnail for Xu_columbia_0054D_16004.pdf Xu_columbia_0054D_16004.pdf application/pdf 5.7 MB Download File

More About This Work

Academic Units
Biological Sciences
Thesis Advisors
Firestein, Stuart J.
Ph.D., Columbia University
Published Here
June 8, 2020