Olfactory design: smell and spectroscopy
Our sense of smell is actually a complex system designed to detect thousands of chemicals. It helps warn of us of danger, e.g. rotting food—we can sense one component of rotten meat, ethyl mercaptan, at a concentration of 1/400,000,000th of a milligram per litre of air.1 Smell also helps us distinguish types of foods and flowers. The sense of smell is actually responsible for most of the different ‘tastes’ of foods. In many animals, this sense is even more important than in humans—it helps bees find nectar, for example.
The nose contains millions of receptors, of 500–1000 different types. They are in the yellow olfactory epithelium, that covers about 2.5 cm2 on each side of the inner nose. The different types of receptors are proteins folded so a particularly shaped odour molecule can dock. Each receptor is coupled to a g-protein. When the odour molecule docks, the g-protein is released. This sets off a second messenger to stimulate a neuron to send a signal. This is transmitted by olfactory nerve fibres which enter either of two specialized structures (olfactory bulbs), stemlike projections under the front part of the brain. They sort the signals, and transmit them to the brain for processing.1,2
Recently, Luca Turin, a biophysicist at University College, London, proposed a mechanism where an electron tunnels from a donor site to an acceptor site on the receptor molecule, causing it to release the g-protein. Tunnelling requires both the starting and finishing points to have the same energy, but Turin believes that the donor site has a higher energy than the receptor. The energy difference is precisely that needed to excite the odour molecule into a higher vibrational quantum state. Therefore when the odour molecule lands, it can absorb the right amount of the electron’s energy, enabling tunnelling through its orbitals.3
This means the smell receptors actually detect the energy of vibrational quantum transitions in the odour molecules, as first proposed by G.M. Dyson in 1937.4 This energy decreases with increasing mass of the atoms, and increases with increasing bond strength. It also depends on the symmetry of the molecule. For a diatomic molecule,5 the fundamental transition energy is:
E = h⁄2π(k/µ)½
Where h is Planck’s constant; k is the force constant of the bond; and µ is the reduced mass, which is related to the masses of the two atoms by:
µ = m1m2⁄(m1 + m2)
A transition can sometimes be caused by incident electromagnetic radiation of the right frequency (ν, the Greek letter nu). This frequency is related to the energy by:
E = hν
The vibrational spectrum is normally measured in wavenumbers , the reciprocal of the wavelength, so its units are cm-1 (reciprocal centimeters). Wavenumber is related to energy by:
As this energy is in the infrared region, infrared absorption spectroscopy is a common tool for measuring vibrational energies and bond strengths (together with the complementary technique of Raman spectroscopy).
This means certain groups of atoms have similar energies, so have similar vibrational spectra. For example, chemicals with sulfur-hydrogen bonds tend to vibrate at about 2500 cm-1 and this is often perceived as a ‘rotten’ smell—rotten eggs produce chemicals like hydrogen sulfide (H2S), and ethyl mercaptan produced by rotting meat is C2H5SH.
Turin supports his theory by noting that decaborane (B10H14) smells very similar to S–H compounds, and it has nothing in common with them apart from similar vibrational energies. Although boron has a much lower atomic mass than sulfur, B–H bonds are much weaker than S–H bonds, and these effects happen to cancel out.
Further support was provided by the analogous compounds ferrocene and nickelocene. These have a divalent metal ion (iron and nickel respectively) sandwiched between two cyclopentadienyl anions (C5H5–). The main vibrational difference between them is that the metal ring bond in ferrocene vibrates at 478 cm-1, while in nickelocene it is 355 cm-1. Ferrocene smells rather spicy, while nickelocene smells like the aromatic hydrocarbon rings. Turin proposes that below a threshold of 400 cm-1, the vibrational signal is swamped by ‘background noise’, so is not detected by the nose.
As different isotopes have different masses but similar chemical properties, they affect the vibrational energy. It can be seen from the formula for reduced mass that the biggest difference results from replacing hydrogen (Ar = 1) with deuterium (Ar = 2)—the numerator is doubled. Indeed, deuterated acetophenone smells fruitier than ordinary acetophenone (C6H5COCH3). It also smells slightly of bitter almonds, just like many compounds containing the cyanide or nitrile group (C≡N)—both C–D and C≡N bonds vibrate at about 2200 cm-1.
One challenge to Turin’s theory is the different smells of some enantiomers (optical isomers), as they have identical vibrational spectra. For example, R-carvone smells like spearmint, and S-carvone like caraway. The answer is: the spectra are identical only in an achiral medium, as in solution or gas phase. But the smell receptors are chiral and orient the two enantiomers differently. This means that different vibrating groups lie in the tunnelling direction in each enantiomer. Turin thinks that the caraway S-carvone is oriented so a carbonyl (C=O) group lies in that direction, so is detected; in the minty R-carvone, it lies at right angles, so is ignored. Turin supported this by manufacturing a caraway scent by mixing the minty carvone with the carbonyl-containing butanone (C2H5COCH3).
If Turin’s theory were true, then infrared and Raman spectroscopy would be essential tools for the perfume industry! Turin is also using inelastic tunnelling spectroscopy—‘inelastic’ refers to the energy loss before tunnelling, as with the proposed sensory mechanism.
The precise chemistry of olfaction is still little understood. But Turin believes he has found a sequence of amino acid residues that could function as the electron donor together with NADPH. He has also found five residues coordinated to a zinc atom that could be the acceptor site. One warning sign of zinc deficiency is loss of the sense of smell, and zinc is often involved in biological electron-transfer reactions.
Whether or not Turin’s idea is correct, the olfactory system exhibits what the biochemist Michael Behe calls irreducible complexity, and is therefore evidence of design.6 This means the system requires many parts for it to work, and would not function if any were missing. The chemical sensing machinery needs proteins with just the right shape to accommodate the odour molecules. And under Turin’s model, the right energy levels as well. And even if the sensors were fully operational, the chemical information gathered by the nose would be useless without nerve connections to transmit it and the brain to process it.
- Sensory reception: smell (olfactory) sense. Britannica CD, Version 97. Encyclopædia Britannica, Inc. 1997. Return to text
- Hill, S., 1998. Sniff’n’shake. New Scientist 157(2115):34–37. Return to text
- Turin, L., 1996. A spectroscopic mechanism for primary olfactory reception. Chemical Senses 21:773. Cited in Hill, Ref. 2 and Sell, Ref. 4. [Note added subsequently: Dr Turin himself wrote (email 9 February 2000): ‘Dear Dr Sarfati, I write to congratulate you on your lucid and accurate description of my spectroscopic theory of smell ….’ However, he said he didn’t necessarily agree with my conclusion that a Creator was responsible. But he continued, ‘I entirely agree, however that if true, my theory is one more example of the wonderful design of living things,’ but he left the question of the cause of this design open.] Return to text
- Sell, C., 1997. On the right scent. Chemistry in Britain, 33(3):39–42. Return to text
- For more complicated molecules, see Wilson, E.B., Decius, J.C. and Cross, P.C., 1955. Molecular Vibrations: the Theory of Infrared and Raman vibrational spectra, McGraw-Hill, New York. Return to text
- Behe, M. J., 1996. Darwin’s Black Box: The Biochemical Challenge to Evolution, The Free Press, New York. See Product information, above right. Return to text