Things will be a lot different when we are older. Perhaps when your grandchild is born, right after they snip the umbilical cord, they will sequence their genome. Immediately, some computer will pick out genetic markers for specific disease states, and maybe a doctor (possibly some gene-ologist of sorts) will do a little bit of gene therapy to prevent the more serious diseases from developing. Perhaps for those really wealthy parents, a few more options may open for tweaking. If you don’t like the hair colour your child’s genes code for, just check the appropriate box and they can take care of that. What about sports? Would you prefer a power athlete, maybe a sprinter, or would you rather your child be tailored for endurance? For that option, perhaps they would look at tinkering with the ACTN3 gene, which encodes the muscle fiber protein alpha-actinin-3.

As it turns out, you could probably already predict genetic predisposition of muscles for either power or endurance. There is evidence for certain alleles of the ACTN3 gene (alleles being different variants of a gene) are more prominent in different types of athletes, as shown in Yang et al. (2003).

If you looked closely at your bicep, you would find that it is made up of myocytes (literally muscle cells). Even closer, and you would see that each myocyte is made up of myofibres, and each fibre is packed full of these long parallel running filaments. Most of these filaments are either thin actin filaments, or thick myosin filaments. Look carefully and you would find that these filaments interdigitate with each other, as you contract and relax your muscle, the amount of overlap in among these fibres change. Of course this sort of action uses a lot of energy in the form of ATP, and there is something about calcium gradients and action potentials, but in essence, it is about actin and myosin fibres sliding parallel to each other, changing the length of a muscle fibre. Where alpha-actinin comes into play is at a part of the myofibre called the Z-line. This line runs perpendicular to the myofibre, anchoring the thin actin filaments to each other. The alpha-actinin protein is perfect for this job because it is kind of long, with an actin binding domain at one end, and alpha-actinin binding domains running along its length. These proteins form dimers, with one sticky actin binding domain at each end. Currently, scientists have found several functions of alpha-actinin (MacArthur and North, 2004). In addition to actin, it has been shown to interact with other components of the contractile machinery including titin and nebulin. alpha-actinin has also been linked to signaling and channel proteins, as well as metabolic enzymes. These results demonstrate the broad functions of alpha-actinin proteins, not only as structural proteins, but also involved in contraction and muscle metabolic activities.

There are four genes that encode the alpha-actinin proteins in humans, these being ACTN1-4. These alpha-actinins have slightly differing functions, some bind actin located in the cytoskeleton, others in muscle fibres at the Z-line. In addition, the type of alpha-actinin used also depends on the type of muscle, whether it is the cardiac muscle of your heart or skeletal muscle in your arms and legs (the third type of muscle, smooth muscle, is a little different and has no Z-lines). alpha-actinin-2 is a sort of base model of alpha-actinin, and is found equally in both cardiac and skeletal muscle. alpha-actinin-3, on the other hand, is expressed specifically in fast glycolytic skeletal muscle.

The impetus for the research into ACTN3 and its relation to different athletes and populations actually began as something completely different. Scientists were focused on a specific null allele of the ACTN3 gene. This 577X allele introduces a premature stop codon in the gene, causing deficient alpha-actinin-3 expression. In people homozygous for this allele (577XX) no functional alpha-actinin-3 is produced. Heterozygous (577RX) and homozygous wild type (577RR) produce normal alpha-actinin-3. Initially, Dr. Kathryn North and her lab were interested in the study of congenital muscular dystrophy. During this work, they came upon the ACTN3 577X null allele and thought that maybe they had stumbled upon a genetic link to the disease. However, the excitement was short lived after it was soon discovered that this null allele was found in healthy populations as well. So they asked the question: Why is a significant proportion of the population homozygous for the 577X null allele and, consequently, alpha-actinin-3 deficient? The initial speculation was that other alpha-actinin proteins may be functionally redundant to alpha-actinin-3. alpha-actinin-2 indeed seems an ideal candidate, showing the same subcellular localization in muscle fibres when alpha-actinin-3 is not present. Still, ACTN3 has been evolutionarily conserved since its divergence from the ACTN2 gene some 300 million years ago, suggesting it has some unique functional phenotype that has made it worth keeping unchanged all this time.

With this in mind, Dr. North and her lab sought to identify certain populations with higher frequencies of functional alpha-actinin-3. This would in turn provide clues as to what, exactly, was so special about ACTN3. Athletes seemed a logical group to investigate, especially given alpha-actinin-3 function in fast-type skeletal muscle. To do this, North and her lab turned to the Australian Institute of Sport, from which they obtained over 400 DNA samples from elite national athletes from a variety of disciplines including swimming and track. Athletes were classified as either power or endurance based on their different events, for example sprinters versus long distance runners. ACTN3 genotype was compared between the different athlete and to a random cohort of 436 unrelated samples. The data showed that power athletes had a statistically significant lower frequency of the 577XX genotype relative to controls. It also showed a higher frequency of the null allele in endurance athletes, though this was only statistically significant in females. In addition, these genotypic differences were more pronounced in the more elite Olympic athletes in the study. A possible explanation for the difference between males and females may be due to lower levels of the hormone testosterone in females, possibly resulting in a greater need for genetic variation to determine muscle performance. In any case, initial data from this study suggests that having functional alpha-actinin-3 improves power performance in muscle.

So if the 577R allele and functional alpha-actinin-3 is best for power and speed, then why would there be such a significant frequency of the null allele in the human population at all? Given the higher suggestive frequencies of the null allele in endurance athletes, part of the answer may lie in efficiency of muscle performance. It is possible that endurance activities may positively select for the null allele. In this case, the need for extended muscle activity would have selected for the 577X allele. Another possibility to explain the presence of the null allele is not selection for a particular allele, but the selection against alpha-actinin-3, where 577X allele provides greater metabolic efficiency because of an absence of alpha-actinin-3 in the muscle. In this case, the null allele may confer greater fitness during times of famine or environmental stress. Dr. North believes the answer to this question may reside with the specific actions of alpha-actinin-3 compared to other alpha-actinins. Her lab is currently working on phenotypic studies and assays to investigate the effects of the null ACTN3 allele. A third possibility to consider is the presence of linked alleles of other genes in close proximity to the 577X allele. It is possible this linked gene may be the one being positively selected for, and that the null ACTN3 allele is merely hitching a ride on the road of genetic selection. The North lab is currently mapping regions close to the 577X allele to look into this possibility.

Beyond athletes, preliminary ACTN3 genotype frequency studies between different ethnic populations has revealed some interesting numbers. In people of European descent, 18% have the alpha-actinin-3 deficient genotype. In Asians, the frequency is 25%. At the other end of the spectrum are the Bantu tribes people of Africa, who show a less than 1% frequency of the null genotype. These different frequencies suggest that, following the divergence of the ACTN3 gene from other alpha-actinins, different forces of selection, possibly the need for power and speed versus endurance or efficiency, shaped the prevalence of the various ACTN3 alleles in populations. As of yet, the exact environmental conditions or physical demands responsible for the varying levels of the ACTN3 alleles is unknown, however, population comparisons may provide yet another clue to the evolutionary history of ACTN3. In the mean time, the puzzle of the power and speed allele for muscle performance remains an intriguing avenue into the study of recent genetic evolution in humans.


MacArthur, D.G. and North, K.N. A gene for speed? The evolution and function of alpha-actinin-3. BioEssays 26, 786-95 (2004).

Mills, M.A., Yang, N., Weinberger, R.P., Vander Woude, D.L., Beggs, A.H., Easteal, S., North, K.N. Differential expression of the actin-binding proteins, alpha-actinin-2 and -3 in different species: implications for the evolution of functional reduncancy. Human Molecular Genetics 10, 1335-46 (2001).

Yang, N., MacArthur, D.G., Gulbin, J.P., Hahn, A.G., Beggs, A.H., Easeal, S., North, K.N. ACTN3 genotype is associated with human elite athletic performance. American Journal of Human Genetics 73, 627-31 (2003).