The gene for Jamaican sprinting success? No, not really.

Anyone who has walked past a TV set over the last few days will have seen footage of the remarkable Jamaican sprinter Usain Bolt, who comfortably cruised to victory (and a world record) in the Olympic 100 metre sprint, and as I write this has just done precisely the same thing in the 200 metre sprint. The interest in Bolt stems not from the fact that he wins his races, but rather from the contemptuous ease with which he does so.

And Bolt is not the only Jamaican to impress in short distance events in Beijing: the country's women's sprint team took all three medals in their 100 metre dash.

Naturally, these performances have provoked widespread speculation about the basis of Jamaica's sprinting success, and the short-distance prowess of other populations of West African ancestry. One controversial suggestion has drawn the most headlines: that sprinting is in their genes, or rather in one gene in particular - variously referred to as "Actinen A" or "ACTN3".

This gene has been the subject of a recent rash of news stories sparked by Bolt's victories, all of which refer to comments by Jamaican academic Errol Morrison in the Jamaica Gleaner over a month ago. The Gleaner article summarised the (unpublished) results of a collaboration between Morrison and a group at the University of Glasgow:

At the base of sprint speed are the fast-twitch muscle fibres stocked with the speed protein Actinen A. And early data indicate that 70 per cent of Jamaican athletes have the gene for Actinen A. Only 30 per cent of Australian athletes studied had the gene.

(The Gleaner reporter, Martin Henry, astonishingly went on to speculate that this gene may help to explain why Jamaicans are "also disproportionately aggressive and violent".)

The Daily Mail followed up on the story two weeks later with a marginally more coherent account:

What they have found - and Morrison emphasises the findings are preliminary - is that fast men have a special component called Actinen A in their fast-twitch muscles, which determine whether humans are sprinters or plodders. It is found in 70 per cent of Jamaicans. In a control study of Australians, only 30 per cent were found with it.

The "preliminary" nature of the findings didn't stop the Daily Mail reporter from following this paragraph with the conclusion that this result "would seem to explain why Jamaicans punch above their weight among sprinters". Similarly definitive statements were made by other reporters continuing the story after Bolt's 100 metre victory; one rare exception was a fairly well-balanced piece in Slate.

The stories take advantage of a widespread perception - by no means totally unjustified, but nonetheless controversial - that Jamaicans and other groups of West African ancestry have a genetic advantage when it comes to raw muscle power. Having apparent scientific evidence to support this perception is a reporter's dream; the headlines write themselves.

So, how good is this scientific evidence? Does the "Actinen A" gene (whatever that is) actually influence sprinting performance? And if so, does it explain the difference in explosive power between Jamaicans and the rest of the world? The answers, as it turns out, are "probably" and "not really".

The ACTN3 gene and muscle performance
At this point I probably should confess to having a more than casual interest in this story: I was one of the authors on the first study showing an association between this gene and elite athlete status back in 2003, and this gene has been the central focus of my research for a good part of the last six years. (The opinions I express here are purely my own, by the way, and in no way are meant to represent the views of my research institute.)

The ACTN3 gene encodes a protein called α-actinin-3 ("Actinen A" is a misnomer of uncertain origin propagated by lazy reporters), which is found within the fast fibres of muscle - the cells that are required for generating rapid, forceful contraction in activities such as sprinting and weightlifting. Interestingly, the human ACTN3 gene comes in two forms in the general population: there's a normal, functional version called 577R, and a "defective" version called 577X, which contains a single base change that prevents the production of α-actinin-3. People who have two copies of the 577X version (I'll refer to them as X/X) produce absolutely no α-actinin-3 in their fast muscle fibres.

These people don't suffer from muscle disease as a result of this deficiency - in fact, there's a pretty good chance that you're one of them. The frequency of the 577X variant differs around the world, but overall somewhere between one-sixth and one-quarter of the world's population (at least a billion people worldwide) are X/X, and therefore completely deficient in α-actinin-3.

So lack of α-actinin-3 clearly doesn't destroy your muscle; however, over the last five years we and other groups have assembled evidence suggesting that it does influence how good your muscle is at generating explosive power. We first showed in 2003 that X/X individuals are significantly under-represented among elite Australian sprint/power athletes, suggesting that the absence of α-actinin-3 in X/X individuals is detrimental to optimal muscle power generation. This association has since been replicated in four separate athlete studies by groups in Europe and the US; there is also weaker but reasonably consistent evidence that α-actinin-3 deficiency results in slightly higher endurance capacity, both in human athletes and in a mouse model generated by our group. In addition, several groups have reported that X/X individuals in the general population display lower muscle strength and reduced sprint performance.

Importantly, the latter two studies suggest that the proportion of the variance in strength and sprint performance in the general population explained by the ACTN3 variant is around 2-3%. So for most of us lazy slobs this gene has a pretty trivial effect - almost completely drowned out by noise from the effects of diet, exercise levels and other genes. (Certainly there are dozens or even hundreds of other genes influencing physical performance, some of which - like the ACE gene - have been fairly well-studied, but most of which are completely unknown and uncharacterised; and environmental factors play about as large a role as genes do in traits like muscle strength and cardiorespiratory performance.)

However, even 2-3% can make a striking difference at the very elite level: of the 51 Olympic-level sprint/power athletes analysed in our original study and a follow-up analysis in Greek athletes not a single individual was X/X (compared to about 10 expected). In fact, X/X Olympian sprint athletes are unusual enough that identifying a single Spanish Olympic short-distance hurdler with α-actinin-3 deficiency was enough to warrant its own publication.

So the absence of α-actinin-3 means very little to most of us, but to a young athlete craving 100 metre Olympic superstardom it could make all the difference in the world. The same could be said of many other genetic variants, of course; Olympic sprinters, essentially, are those unlikely individuals at the vanishing edge of the probability distribution for whom nearly every genetic coin has come up heads.

Does the ACTN3 gene explain Jamaican sprinting prowess?
The underlying argument here is intuitively simple: (1) variation in the ACTN3 gene is strongly associated with elite sprint athlete status; (2) the "sprint" version of ACTN3 is more common in Jamaicans than in individuals of European ancestry; therefore (3) this variant may well play a role in the increased sprinting prowess of Jamaicans relative to Europeans. At first blush this sounds pretty convincing; however, while ACTN3 may play some role in the disproportionate success of Jamaican sprinters, I'd argue that it's likely to be a pretty small one. Here's why:

1. The difference in frequency between Jamaicans and Europeans is not as great as it would appear. The articles quoted above describe the proportion of individuals who have two copies of the 577R ("sprint") version of the gene; a more appropriate comparison is the proportion of individuals who have at least one copy of 577R (that is, including both R/R and R/X individuals), since it's only the complete absence of α-actinin-3 that is reliably associated with reduced sprint performance. This starts to look less impressive: it's 98% in Jamaicans compared to about 82% in Europeans. In other words, in both populations a sizeable majority of individuals have an ACTN3 status compatible with elite sprint performance.


2. The ACTN3 frequency reported for the Jamaicans by Morrison is not unique to Jamaicans, nor is it particularly surprising - our group has previously reported virtually identical frequencies in individuals from both West Africa (the ancestral source of the bulk of the Jamaican gene pool) and East Africa, in a collaboration with the same group at the University of Glasgow that Morrison has been working with on the Jamaican study. In fact, that study showed that an even higher frequency of α-actinin-3 expression (99%) is found in Kenya - in members of tribes whose members dominate international long-distance events, but have a notable dearth of representatives in track sprinting; we have more recently found similarly low frequencies in populations across sub-Saharan Africa. There's simply no clear relationship between the frequency of this variant in a population and its capacity to produce sprinting superstars.
   


3. Finally, when Usain Bolt was pacing restlessly at the starting line of the 100 metre sprint - even in the very first round of Olympic heats - the very low frequency of X/X individuals among Olympic sprinters means he was lined up against a group of athletes who almost certainly all express α-actinin-3! In other words, while the ACTN3 variant may have played a small role in getting Bolt to the Olympics, it can't possibly explain the astonishing advantage he has over his competitors.
   

I'll concede that the small difference in the frequency of α-actinin-3 expression between Jamaicans and Europeans may result in a slightly larger fraction of the Jamaican population being suitable for elite-level sprinting (all else being equal), but it's a tiny piece of the overall explanation at best - and it can't possibly explain why Bolt is so much better than his fellow West Africans and other Olympic-level sprinters. Clearly there are other factors at work.

Beyond "the gene for speed"
I'm certainly not arguing here that genetics doesn't play any role in Bolt's success - or in the remarkable over-representation of West African descendents in Olympic short-distance track events, or the similarly impressive skew towards East Africans among marathon runners. In fact I think most geneticists would be staggered if this was the case, even though direct evidence for underlying genes is currently very thin on the ground.

Rather, my point is that an excessive emphasis on ACTN3 as a major explanation for Jamaican success does a grave disservice to the complex interplay of genetic and environmental factors required for top-level athletic performance. This suggestion goes against everything we've learnt about the genetics of complex traits from recent genome-wide association studies, which have revealed that quantitative traits (like height and body weight) are frequently influenced by dozens to hundreds of genes, each of small effect; if anything, it's likely that athletic performance will be even more genetically complex than these traits. The ACTN3-centred argument also dismisses the importance of Jamaica's impressive investment in the infrastructure and training system required to identify and nurture elite track athletes, the effects of a culture that idolises local track heroes, and the powerful desire of young Jamaicans to use athletic success to lift themselves and their families out of poverty.

It is almost certainly true that Usain Bolt carries at least one of the "sprint" variants of the ACTN3 gene, but then so do I (along with around five billion other humans worldwide). Indeed, I'm fortunate enough to be lugging around two "sprint" copies - but that doesn't mean you'll see me in the 100 metre final in London in 2012. Unfortunately for me, it takes a lot more than one lucky gene to create an Olympian.


(Image: Phil McElhinney.)

 Subscribe to Genetic Future.