Probing the Validity of a 21st Century Theory of Smell
Shubin Ling and James P. Snyder
Department of Chemistry, Emory University

Abstract

Luca Turin's theory of smell, based on the vibrational modes of odorants rather than stereochemical factors alone, is the current hypothesis for the underlying mechanism of olfaction. Turin proposed that the unique vibrational spectrum of an odor molecule, when properly transformed, can be correlated with its smell. Compounds with similar spectra are presumed to have comparable smells; those with different spectra, unrelated smells. Turin suggests that a molecule docks at a G-protein's receptor site and is vibrationally excited by the tunneling of a high energy electron donated by NADPH across the binding site. Cellular signals coupled to the modified receptor ultimately lead to a response in the brain. One test of this version of the vibration theory would encapsulate a small odorous molecule in a framework like a buckyball. If the brain perceives the small molecule signal, it would be consistent with the theory. The molecular modeling computer program Spartan was utilized in order to obtain infrared spectra of various molecules placed inside buckyballs and buckytubes. The spectra are virtually identical to the free molecules. However, based on an analysis of buckyball vapor pressure, it was concluded that the encapsulated molecule may not be volatile. Thus focus was shifted to another of Turin's claims. Turin states that cedramber, Jeger's ketal, karanal, and timberol, molecules which have different structures, smell alike. He also asserts that 2-undecanone, 4-undecanone, and 6-undecanone, molecules with similar structures, smell different. Using Macromodel, Gaussian, and Molden, which are molecular computer programs, IR spectra were calculated using a higher level of theory (density functional theory (DFT) and the 6-31G* basis set) and compared to Turin's PM3 applications. It was found that Turin's claims do not hold as well at a higher level of calculation. This, however, is only the first step. Our experiment to this point omits partial charges and displacements from equilibrium, factors Turin considered in his algorithm. A statement about the validity of Turin's theory cannot be made until these factors are included, work that is underway.

Introduction

Luca Turin's theory of smell, based on the vibrational modes of odorants rather than stereochemical factors alone, is the current hypothesis for the underlying mechanism of olfaction. Turin proposed that the unique vibrational spectrum for an odor molecule, when properly transformed, can be correlated with its smell. Compounds with similar spectra are presumed to have comparable smells; those with different vibrational spectra have unrelated smells. Turin suggests that a molecules docks at a G-protein coupled receptor site and is vibrationally excited by the tunneling of a high energy electron donated by NADPH across the binding site. Cellular signals coupled to the modified receptor and accompanying G-proteins ultimately lead to a response in the brain. We have devised two experiments to evaluate Turin's views.

Methods and Materials

Molecular modeling programs (Spartan, Macromodel, Gaussian, and Molden) were used in order to optimize geometry of molecules and obtain IR spectra. The methods of calculation were semi-empirical PM3 and density functional (Becke3YLP/6-31G*).

Results

Encapsulating small, odorous molecules within buckyballs does not significantly alter the vibrational modes of a molecule. This inhibits binding while maintaining the same vibrations. Turin's results using the semi-empirical PM3 method were not consistent with the experimentally found data using the density functional method.

Conclusions and Future Studies

The PM3 method predicts that buckyballs do not interfere with the vibrations of a caged small molecule. Analysis of buckyball vapor pressure suggests that the encapsulated molecules may not be volatile. Thus. Vibrational spectra were evaluated at different levels of theory. Turin's PM3 results do not appear to match up with our 6-31G* calculations. The ketones and the organic molecules showed similar differences among their IR spectra, even though molecules in the ketone series smell different, while those in the organic molecule series smell the same. To fully evaluate Turin's theory, the IR spectra need to be transformed to electron scattering spectra; work in progress.

Acknowledgements and Funding Attributions

Funding for this experiment was provided by the Howard Hughes Medical Institute and the Summer Undergraduate Research at Emory program.

In Plain English

First, I compared the vibrations of a molecule in an isolated environment to the vibrations of the same molecule inside a carbon cage. Second, I took Turin's experiment and applied a more advanced technique to see whether the same results would be obtained.

Techniques

Optimization and frequency calculation.