Love is in the air

A series of studies starting in the 1990s linked compatibility genes to human attractiveness. Claus Wedekind, a biologist at the University of Bern in Switzerland, asked 49 women to score how attractive they found the odour of tee-shirts worn for two nights by men who had used scent-free toiletries.² Wedekind reported that women preferred the smell of men with compatibility genes different to their own. He hypothesised that women preferred MHC-distinct partners because it would give their offspring a greater range of immune genes, protecting them against a wider variety of pathogens and toxins.

Other studies followed, including research by Carol Ober, a geneticist at the University of Chicago, US, looking at members of the Hutterite religious clan, an Amish-like closed society living in the rural Midwest of the US. Hutterites marry only other clan members, so the variety in their gene pool is relatively low. Ober found that within these limits, Hutterite women still managed to find partners who were MHC-distinct from them most of the time.

However, Philip Hedrick and Francis Black from Arizona State University, US, examined couples from South American Indian tribes and could not find any evidence that individuals had opted for partners with different MHC genes. They noted that their study reflected natural selective forces at play. Perhaps various ‘genetic, environmental and cultural factors’ found in real life were not present in controlled studies, allowing MHC genes to be a stronger cue for selecting a partner, they speculated.

The theory that MHC genes somehow influence human attractiveness remains controversial and has polarised scientific opinion, says Davis. For example, Hedrick has made it clear that he remains a sceptic. ‘Nobody knows whether or not a difference of a single point on a scale of one to 10 for the sexiness of tee-shirt would actually influence a person’s behaviour.’

Craig Roberts, who studies evolutionary human behaviour and biological psychology at Stirling University, Scotland, thinks that MHC genes undoubtedly influence the choice of partners in humans. ‘It does in all vertebrate taxa – mammals, reptiles, birds, fish – so why not in humans?

The lab experiments show that we have the capacity to detect it, in the same way as these other taxa. I believe the only question is not whether, but to what extent, MHC influences human mate choice and romantic relationships.’

A large number of studies on inbred mice and rats – who were alike in all genes but MHC – have found a tendency for mice to choose mates with different MHC genes, presumably based on what they could smell. This could be because it avoids inbreeding, allows mice to find relatives for communal nesting, or help keep immune system genes diverse.

Roberts points out, however, that it is not the case that MHC controls attractiveness in any animal species, even mice. ‘MHC is not the only aspect considered when weighing up attractiveness. I have showed how MHC compatibility in mice was overshadowed by effects of male dominance. Mouse mate choice is also known to be influenced by other genes, diet and nutrition, familiarity, body size, etc.’ Wedekind also studied mice and he found that tastes were reversed in pregnant females. Unlike non-breeding females, they chose the familiar odour of MHC-similar males. Perhaps, by nesting with relatives, the mothers get help looking after their offspring.

Wedekind’s tee-shirt studies also found that women who were on the contraceptive pill preferred the smell of MHC-similar males rather than MHC-distinct ones, perhaps because the pill raises oestrogen levels in the body just as in pregnancy. Could it be that when a female is pregnant, they are looking for safe, helpful partners rather than mates?

Research by Roberts also found women on the pill did tend to prefer the scent of men with similar MHC profiles.³ ‘If odour plays a role in human mate choice, our results suggest that contraceptive pill use could disrupt mate preferences,’ he concluded.

In another interesting finding, Roberts reported that single women preferred odours of MHC-similar men, while women in relationships preferred odours of MHC-dissimilar men. He suggested that volunteers in long-term relationships were able to ‘focus more successfully’ on what they wanted in a mate, or perhaps they could discriminate MHC dissimilarity more effectively because they had greater experience of male odour.

‘Women report that odour is very high on their priority list of wants during partner choice, much higher than in men,’ says Roberts. ‘Odour is probably not critical in deciding who to date, but becomes much more important when the relationship becomes intimate. It may ultimately be critical on the morning after the night before, in the decision about whether the relationship develops further.’

So does this mean that women who are on the pill tend to choose mates they would otherwise avoid? If so, there may be consequences for choosing a partner with immune systems that are not complementary. In Ober’s Hutterite study, for example, the few couples with a high degree of MHC similarity suffered increased rates of miscarriage and experienced longer intervals between pregnancies. One suggestion is that the body recognises a child who isn’t likely to have a strong immune system and miscarries it.

MHC genes do interact with another set of genes in pregnancy, says Davis. They are involved in constructing the placenta, allowing it to function properly. ‘Our variable immune system genes influence whether or not pregnancy is successful. Couples having certain combinations of these genes are more likely to miscarry or have other problems in pregnancy such as pre-eclampsia.’

So how do we detect if our date has a suitably different set of compatibility genes? Mice are known to sniff out scents in urine, says Davis. But there is no known and proven mechanism for how people might emit and detect the odour of MHC genes.

One theory is that somehow MHC genes could set in motion the formation and release of scent molecules through a person’s glands on their skin. Again this is an area of controversy. Davis explains: ‘Some scientists say that there is no way it could ever work.’

The experiments done so far use tee-shirts, says Roberts, suggesting the upper body provides a suitable odour source, presumably the armpits. Bacteria are the key. Odour compounds are metabolites from bacteria feeding on the skin surface. ‘There are two ideas as to how this can be linked to MHC genes,’ explains Roberts, ‘either a common bacterial microflora metabolise MHC peptides – the exact nature of which are related to MHC genotype – to produce a characteristic odour, or the MHC selects for a characteristic microflora associated with particular MHC type.’ The essential mechanisms are likely the same in both mice and humans, regardless of odour source, he adds.

Meanwhile, German researchers from the Max Planck Institutes of Immunobiology and Epigenetics in Freiburg, and for Evolutionary Biology in Plön, believe they may have discovered how animals, and possibly humans, detect the smell of MHC genes and alter behaviour as a result. In mice and fish, peptide ligands – bound and displayed by MHC proteins – activate nerve cells in the olfactory (smell-detecting) system, says Thomas Boehm of the Max Planck Institute in Freiburg. The usual role of these ligands is to help the immune system recognise pathogens, such as bacteria, viruses and parasites, but here they seem to be helping the animal’s olfactory system detect certain MHC proteins in the scent from prospective partners.

In recent research, the German team now claims to show that MHC peptide ligands may play a similar role in humans.⁴ In psychometric tests, volunteers recognised when their body odour had been supplemented by MHC peptides, and preferred the odour that was more like their own.

‘These results show that the protein ligands transported by the immune molecules actually determine the natural body odour not only in animals, but also in humans,’ says team leader Manfred Milinski, director at the Max Planck Institute in Plön, who was involved in Wedekind’s original tee-shirt experiments. ‘We are unconsciously aware of how our own immune system is constituted and can recognise that of a potential partner from their smell.’

Working with researchers at the University of Dresden Medical School in Germany, the scientists then examined whether this response to MHC molecules in odours produced any effect in the brain. Using functional magnetic resonance imaging (MRI), the team was astonished to find that a small area located in the right middle frontal cortex always responded when the test subjects smelled one of their own peptides. This suggests, Boehm says, the presence of an ‘internal reference’ for MHC molecules that helps interpret the signal coming from the olfactory system, and could influence behaviour.

Taking this one step further, Milinski says they have previously found a correlation between a person’s MHC genes and their preference for certain natural perfume ingredients.

‘This suggests that the function of selecting a perfume for yourself may be to amplify your own MHC odour signal. [Our recent findings] suggest synthetic MHC ligand peptides may be used in perfumes instead of expensive natural ingredients.’ In time, Milinski believes perfumes could be developed specifically to intensify the signal of an individual’s immune gene repertoire to a potential partner.

It is clear that compatibility genes influence mate selection in some animals, concludes Davis. There is evidence – contentious evidence – that this can also happen in humans, although this influence may get swamped by social and cultural conventions. The biggest problem, he thinks, is that the link between compatibility genes and smell is not yet understood.

But if Milinski is right, and our brains react to favourable scents of complementary MHC genes in our body odour, is there any way we can ensure we are choosing the best partner?

MHC genes differ so much from one person to the next that there’s no obvious way to reliably predict who’s MHC-compatible with whom, says Davis. ‘Overall, nobody has a better or worse set of compatibility genes: there’s no hierarchy in the system. The fact that we differ is what’s important.’

It seems we just have to trust our animal instincts after all.

1 Michael B. Lewis, Perception, 2010, 39(1), 136.
2 Claus Wedekind et al, Proc. R. Soc. B, 1995, 260, 245.
3 Craig Roberts et al, Proc. R. Soc. B, 2008, 275(1652), 2715.
4 Manfred Milinski et al, Proc. R. Soc. B, 2013, 280(1755,) 2012.