Theories of smell

Theories of olfaction

Vanillin

Smelly molecules have specific features like side groups, shape, chain length, charge. For example vanillin (left) has a carbon ring, a hydroxyl group, an aldehyde group and an ether side group. These components of the molecule bind to specific receptors on an olfactory receptor neuron (ORN) and activate it. The axon of the ORN projects to areas of dense neuronal tangles in the olfactory bulbs called "glomeruli" . These are the units of olfactory feature detection that will respond to these side-groups and features. At each glomerulus the converging axons of one particular type of olfactory receptor neuron (ORN) will synapse onto a mitral cell, so glomeruli are responsive to one and only one particular molecular feature detected by their ORNs. The two olfactory bulbs have their glomeruli distributed over them that are in more or less the same location in each bulb. These glomeruli detect the features of the odorant (smelly molecule). One odorant may activate one or more glomeruli and the result is an electrical activation of certain areas of the bulb - a pattern of activation will be specific to a particular odorant. So, an odorant will set in motion a pattern of electrical excitation over the olfactory bulbs that will be specific for that odorant. This pattern is relayed to higher brain centres via the lateral olfactory tract and where it will be processed and interpreted. The big question is, how do the smelly molecules interact with their receptors. There are a number of theories:

1. Molecular shape 
2. Diffusion pore 
3. Piezo-effect 
4. Molecular vibration 
5. Nose as a spectroscope - Luca Turin's vibrational theory of smell  
6. Walter Freeman and chaos theory applied to olfaction

1). Molecular shape

Chemists noted that C4-C8 chains of certain aldehydes/alcohols had strong odours and the odour changed as the chain length increased. 6-C benzene ring altered its smell greatly according to where the side chains were situated, whereas larger rings (14-19C atoms) could be rearranged considerably without altering their odour. The "lock and key" hypothesis (Moncrieff, The Chemical Senses, 2nd ed., 1951) was borrowed from enzyme kinetics and applied to smell. He proposed that distinct primary odours had receptor sites. In the early '50s an American, John Amoore, was on a fellowship to Oxford. He proposed a stereochemical theory of smell. Amoore was very complimentary about Moncrieff's (a dour Scotsman) theory. In Amoore's 1952 paper he refers to Moncrieff's book in the opening paragraph. He says,
"In chapter 12......Moncrieff presents his own theory, which has much to recommend it..in order to be odorous, a substance must possess two properties: (1) volatility; (2) a molecular configuration that is complementary to certain sites on the receptor system".
Amoore went on to propose 7 primary odours because of their high frequency of occurrence amongst 600 organic compounds; camphor, musk, floral, peppermint, ether, pungent and putrid.

Amoore's primary odours

These 7 primary odours were proposed to have different shaped receptors corresponding to the shape of the molecules. Moncrieff, in the next edition of his Chemical Senses (3rd. ed., 1967), was not so kind about Amoore and it maybe instructive to quote from the book:

"In 1952, Amoore suggested that there were seven primary classes of odours......The basic idea was that there were seven kinds of olfactory receptor site, places whereon odorant molecules could lodge when adsorbed on the olfactory sensitive area, and that the odorous molecules had shapes and sizes that were complementary to the shape and size of the seven olfactory receptors.....Molecular shape and size certainly play a part, perhaps a main part, in determining odour. But little confidence can be felt in the restriction of odours to seven classes, and particularly to the seven suggested classes. The "proofs" and "predictions" that have been quoted by Amoore are not satisfying...The suggestion that there are only seven kinds of receptor and seven classes of odour is apparently untenable. Unexpectedly, at the Corvallis, Oregon symposium on Food Flavours in September 1965, much of Amoore's theory was withdrawn. Difficulties had arisen with which it could not cope. So ended apparently one of the many theories of olfaction. Looking back there was never much solid evidence to support it, and there were difficulties all along the line, but it did stimulate a lot of useful thought".

Back to top

The shape theory has waxed and waned over the years and the notion of primary odours is no longer tenable. (Although without such a concept it is difficult to explain specific anosmias). However, following recent experiments involving the transfection of a cloned olfactory receptor gene (receptor 17) into the rat (Zhao et al., 1998), it was been demonstrated that this receptor only recognised a limited number of odours (aldehydes). Activation was caused by aldehydes 7-10 carbon atoms long, and not to C6 or C11. Since all the aldehydes have the same side group (-C=O) it can be inferred that this particular receptor was responding to chain length (i.e. shape) and not side group. This does not rule out the possibility that other receptors respond to side groups. Nevertheless it is support for Amoore's Shape Theory.

Amoore, J.E. (1952) The stereochemical specificities of human olfactory receptors. Perfumery & Essential Oil Record 43, 321-330

Amoore, J.E. (1963a) The stereochemical theory of olfaction. Nature, 198, 271-272.

Amoore, J.E. (1963b) The stereochemical theory of olfaction. Nature, 199, 912-913.

Moncrieff, J.W. (1967) The Chemical Senses, pp 381-2.

Zhao, H., Ivic, L., Otaki, J.M., Hashimoto, M., Mikoshiba, K. and Firestein, S. (1998) Functional expression of a mammalian odorant receptor. Science 279, 237-241.

Back to top

2). Diffusion pore

This theory of Davies and Taylor (1959) suggests that the olfactory molecule diffuses across the membrane of the receptor cell forming an ion pore in its wake. The diffusion time and affinity for the membrane receptor determine thresholds. But, it is difficult to explain the different qualities of smell. The problem of frequency coding and stimulus intensity is difficult to resolve. The different odour would cause a different size pore and therefore a different receptor potential, giving rise to a particular firing rate - but in olfaction, stimulus intensity is frequency coded and not the different quality of the odour. However, many odorants are organic molecules that will dissolve in membranes and alter their properties.

Davies, J.T. and Taylor, F.H. (1959) The role of adsorption and molecular morphology in olfaction: the calculation of olfactory thresholds. Biol. Bull. Marine Lab, Woods Hole, 117, 222-238.

Back to top

3). Piezo effect

This slightly "off the wall" theory was proposed by Rosenberg et al (1968). They believed that the carotenoids (vitamin A, in the pigment of the olfactory cells) combine with the odorous gases giving rise to a semiconductor current. They argued that this curent could activate the olfactory neurons. This theory followed from an earlier paper by Briggs and Duncan who proposed "..it seems reasonable to assume that proetin bound carotenoids of the olfactory epithelium are the receptors of energy from olfactant molecules entering the nasal cavity". Rosenberg an colleagues tested the idea and found a reversible concentration-dependent increase in current of up to 10,000,000 times and proposed a weak-bond complex formation which increased the number of charge carriers. However, there were problems with this theory; (1) receptor cells do not contain the pigment and (2) weakly odorous short chain alcohols gave a greater increase in semiconductor current than smellier long-chain alcohols. Neverthless, several interesting facts remain to be explained;
vitamin A deficiency leads to anosmia
the greater the pigmentation (more vitA) of the olfactory epithelium the more sensitive it is to smell
Briggs and Duncan (1961) treated 56 anosmics with intramuscular injections of vitamin A. 50 patients recovered in part or completely.

As long ago as 1870, Ogle (Med.-chir. Trans. 53, 263-290) proposed that the pigment found in the nasal passages might act to absorb the vibrations of odorous substances and convert them into heat vibration that would activate the neighbouring nerve cells. We still do not know what the pigment is doing in the olfactory epithelium.

Briggs, M.H. and Duncan, R.B. (1961). Odour receptors Nature 191, 1310-1

Rosenberg, B., Misra, T.N. and Switzer R. (1968) Mechanisms of olfactory transduction. Nature 217, 423-427.

Back to top

4). Molecular vibration

The frequency of many odours is in the infrared (IR). Is this resonance associated with their smell? This idea was suggested by Dyson (1938). Male moths are drawn to candles because the flickery IR emission is identical to that of the female moth's pheromone. Different frequencies of IR could give rise to different smells. If the whole vibrational range was used, up to 4000cm-1, the detection of functional groups would be explained since many compounds with distinctive odours vibrate at around 1000cm-1. There is an immediate problem - that of the body's natural IR heat. Perhaps the pigment acts to absorb this IR radiation. Another problem is that frequency coding is proportional to stimulus intensity in olfaction, so different frquencies of IR could not be converted into different nerve firing frequency.

Dyson, G.M. (1938) The scientific basis of odour. Chem. Ind., 57, 647-651.

Back to top

The nose as a spectroscope

This theory, proposed by Luca Turin (1996), originates from the work of Dyson (see above) who suggested that the olfactory organs might detect molecular vibrations.

Turin has proposed that when the olfactory receptor protein binds an odorant, electron tunneling can occur across the binding site if the vibrational mode equals the energy gap between filled and empty electron levels. The electron tunnelling then activates a G-protein cascade. Receptors are therefore "tuned" to the vibrational frequency of particular odorants, rather like cones are "tuned" to particular wavelengths of light.

Luca Turin's theory of olfactory transduction

Nose as a spectroscope. Turin, L. (1996) A spectroscopic mechanism for primary olfactory reception. Chem. Senses 21, 773-791.

• NADPH supplies electrons
  
• e- can only tunnel if; (1) an odorant fills receptor site, (2) vibrational energy equals energy gap
  
• e- s flow through protein and reduce disulphide bridge via Zn++.
  

Back to top

Support for theory

After receiving a heavily prejudiced critique from the editor of Nature Neuroscience (vol. 7(4), 2004, pp 315-6) and a test published in the same issue by Keller & Vosshall that mis-used statistics to find no evidence that subjects can sniff the difference between two isotopes, support for Luca Turin's theory has come from two sources recently. First, Roberts at the 2006 ECRO meeting presented data demonstrating that subjects can distinguish isotopes on the basis of smell and second, physicists from University College London have proved that humans could theoretically recognise odour by photon assisted tunneling (http://arxiv.org/abs/physics/0611205, pub.22 Nov 2006). There has been some new debate about the vibrational theory with some evidence, e.g.  Implausibility of the vibrational theory of olfaction, against (although this is contested) and some in support; New evidence for the vibration theory of smell. This latter was in February 2016 and shows that honey bees can distinguish between isoptomers of simple odorants in which some of the hydrogen atoms are replaced by deuterium. Same shape, different vibration.

More discussion on the VTO (Vibrational Theory of Olfaction) following a short article by Derek Lowe, with some sensible comments and some less so.

For further details of debate on this theory:

• see Wikepedia.
• More discussion of smell theories can be found here: Exploring Everyday Chemistry, University of York, UK
• Quantum ideas on smelling
• "Chemistry and the Sense of Smell", Charles Sell's book on smell theory and function. He is an advocate of the Shape Theory.

Walter Freeman and chaos theory applied to olfaction

Walter Freeman produced some inspirational ideas about olfaction and how the olfactory system detects and identifies odour as well as adapting to olfactory experience.

Isotonic EEG maps of the olfactory bulb following different stimuli

Freeman demonstrated that the pattern of activation of the olfactory bulb was random - chaotic - when not experiencing a smell. An odorant acts as an attractor and the pattern of activation in the olfactory bulb becomes ordered and regular with a different pattern for each olfactory object. The pattern can evolve as the olfactory system experiences the same odorant following an intervening experience, as shown in the figure above.
The Physiology of Perception, by Walter J. Freeman: The brain transforms sensory messages into conscious perceptions almost instantly Chaotic, collective activity involving millions of neurons seems essential for such rapid recognition.

Freeman generated "phase portraits" of the olfactory system as it emerged from chaotic activity following a smell stimulus

Phase portrait of the olfactory system is response to a smell stimulus