On this page
You can tell the smell of fresh coffee from burnt toast in a fraction of a second, yet for most of the twentieth century nobody knew how the nose did it. Linda Buck and Richard Axel answered the first part of that question by finding the genes for smell receptors. This post explains what they found, the counting idea that makes smell work, and what students can learn from Buck's approach.
Who are Linda Buck and Richard Axel?
The Nobel facts page for Linda Buck says she was born on 29 January 1947 in Seattle. A course in immunology at the University of Washington drew her into biology. She earned a bachelor's degree in microbiology in 1975 and a PhD in immunology in 1980 at the University of Texas Southwestern Medical Center in Dallas, then moved to Columbia University in New York. Richard Axel was born on 2 July 1946 in New York and was at Columbia. They shared the 2004 Nobel Prize in Physiology or Medicine "for their discoveries of odorant receptors and the organization of the olfactory system".
What did the 1991 paper show?
Their 1991 Cell paper begins with the problem: the mammalian nose can recognise a huge number of different odour molecules. They cloned and characterised 18 members of what they called "an extremely large multigene family". The genes code for proteins that cross the cell membrane seven times, and they are switched on only in the olfactory epithelium, the smell-sensing lining of the nose. That pattern made them strong candidates for odorant receptors.
How large is the family?
The 2004 Nobel press release says the family has some 1,000 genes, about three per cent of our genes. Mice, the species they studied, have about 1,000 odorant receptors; fish have about 100; humans have somewhat fewer than mice because some genes were lost in evolution.
The receptors belong to the G protein-coupled receptor family. When an odour molecule binds, the receptor changes shape and activates a G protein, which raises cyclic AMP, which opens ion channels. The cell fires and sends a signal towards the brain. If you have studied hormone action, you will recognise the same second-messenger idea.
One cell, one receptor
The press release describes a surprise: each olfactory receptor cell uses one and only one receptor gene. Cells with the same receptor send their nerve fibres to the same small region, a glomerulus, in the olfactory bulb. There are about 2,000 glomeruli, roughly twice the number of receptor types (2,000 ÷ 1,000 = 2).
How can 1,000 receptors tell apart thousands of smells?
Buck's later work answered this. In a 1999 Cell study, Combinatorial Receptor Codes for Odors, her group found that one receptor recognises several odour molecules, one odour molecule is recognised by several receptors, and different odour molecules activate different combinations. They also found that a small change in a molecule, or in its concentration, can change its code.
Here is why combinations are so powerful. The numbers below are a classroom illustration, not data from the study.
With 3 receptor types, each either on or off, there are 2 × 2 × 2 = 8 patterns. Leave out "all off", which is no smell, and you have 7 possible codes.
With 10 receptor types, there are 2^10 − 1 = 1,023 codes.
With about 1,000 receptor types, the number of possible on-off patterns is 2^1000, a number with over 300 digits.
Real receptors respond with different strengths, not just on or off, so the true capacity is even larger. The press release says this is the basis for recognising and remembering about 10,000 different odours.
What can a student take from Buck's way of working?
Use an outside skill on a new problem. Buck trained in immunology, where large gene families are familiar, then applied that thinking to smell.
Narrow with a clear filter. The 1991 team looked for genes with three features at once: a large family, seven membrane crossings, and expression only in the nose. Lists of conditions like this are how you eliminate options in multiple-choice questions.
Ask the second question. Finding receptors was step one. Asking how they combine was step two, and that is where the explanation lay.
Respect counting. 2^n grows quickly. Permutation and combination skills from maths turn up in biology more than you might expect.
To place receptors and signals in the wider nervous system, read our guide to neural control and coordination. If G proteins and second messengers still blur together, the Fermi Concept Library has every concept in physics, chemistry, biology, mathematics and computing, so you can look up that one link in the chain.
Next step: work out how many on-off codes 5 receptor types can make, then check your answer is 31.
