In skeletal muscle, α-actinin proteins (α-actinin-2 and α-actinin-3) are an important structural component of the Z disc, where they anchor actin thin filaments, helping to maintain the myofibrillar array.1, 2 In humans, α-actinin-2 is expressed in all skeletal muscle fibers, while α-actinin-3 is expressed only in type 2 fibers.3 The α-actinin-3 gene (ACTN3) has attracted considerable attention recently due to a frequent nonsense polymorphism (R577X) that may influence muscular performance.4 The R577X polymorphism is not associated with overt muscle disease despite the complete absence of α-actinin-3 protein in individuals carrying the X/X genotype.3 Subsequent studies addressed the question of whether the X/X genotype may somehow affect muscle function in otherwise healthy individuals, especially for movements involving high force, given the protein's localization to only type 2 muscle fibers.5, 6

Yang et al7 first reported in 2003 an under-representation of the X/X genotype group in elite sprint athletes; those findings have been replicated in Finnish and Greek populations8, 9 and extended to show lower sprint performance in X/X genotype carriers in a Greek population.10 These results indicate an advantage for sprint-related athletes expressing the α-actinin-3 protein compared to X/X athletes deficient in the protein.

In the present case–control study, we sought to determine if the previous findings of under-representation of the ACTN3 X/X genotype in sprint-related athletes would be extended to other primarily anaerobic athletes such as bodybuilders and powerlifters.



Subjects included 75 elite-level bodybuilders (n=18 ranked in the top 100 worldwide), powerlifters, and college level strength athletes from across the United States, and 876 men and women from the general population of the Maryland region participating in the Baltimore Longitudinal Study of Aging. The recruitment strategies and characteristics of the strength athletes have been described previously in detail;11 no direct physiological data are available on the athletes. The Baltimore Longitudinal Study of Aging is comprised of healthy men and women across the adult age-span (19–90 years); very few of the Baltimore Longitudinal Study of Aging subjects participate in strength training.12 All subjects were either White or Black by self-report (Table 1). Written informed consent was obtained from all subjects under protocols approved by the Institutional Review Boards of the University of Maryland, Johns Hopkins Bayview, and Harbor Hospital.

Table 1 Sex and race sample sizes of the strength athletes and the general population control subjects

DNA and genotyping

All subjects were genotyped for the ACTN3 R577X polymorphism using restriction digest methods described previously6 on DNA collected from whole blood. Sequence verified control samples of all three genotypes were used for all assays.


One-tailed χ2tests were performed to compare genotype frequencies between athletes and controls. Because X-allele frequencies are known to differ between Whites and Blacks,6 analyses were performed both in the total group and within each race group separately. Statistical significance was accepted at P<0.05.


The sex and race information of the 75 strength athletes and 876 control subjects is listed in Table 1. No significant genotype frequency differences were observed between men and women within the control group. Hardy–Weinberg equilibrium (HWE) calculations showed no deviation from expected frequencies in controls, but deviation was observed in the total athlete group (P=0.001), White athletes (P=0.01), and Black athletes (P=0.06). A significantly lower frequency of the X/X genotype was observed in the athletes (6.7%) compared to controls (16.3%; P=0.005; Table 2). Within race groups, the X/X genotype was significantly lower in White athletes vs controls (P=0.018). No Black athletes were observed with the X/X genotype, though this finding only approached statistical significance (P=0.10; Table 2).

Table 2 ACTN3 R577X genotype frequencies (%) for controls and athletes


The present report is the first to demonstrate a lower proportion of the ACTN3 R577X X/X genotype in elite-level strength athletes. The findings extend previous reports showing lower frequency of the X/X genotype in sprint-related athletes of various events. The athletes in the present study were predominantly male, so whether the results extend to female elite strength athletes, who differ from similarly trained males,13 is uncertain.

The present study relied on bodybuilders, powerlifters, and other strength athletes, many of whom were highly competitive within their discipline (eg, 18 of them were ranked among the top 100 bodybuilders worldwide), while previous cross-sectional studies of ACTN3 genotype have examined sprinters, short-track cyclists, etc.7, 8, 9 The strength athletes exhibit extreme muscle hypertrophy,14, 15 with performance measured either as physical appearance or single-repetition maximal force production (strength), rather than the multiple contraction events typical of sprint-related athletes.7, 8, 9, 10 Because the α-actinin protein is important for maintaining the myofibrillar array within the muscle fiber,1, 2 deficiency of the protein may be important for maintaining the contractile protein arrangement across multiple contractions involving type 2 muscle fibers, as can be inferred from previous studies;7, 8, 9, 10 the importance of single-contraction performance (eg, one-repetition maximum) is unknown. In this regard, strength athletes may demonstrate a lower proportion of X/X genotype compared to controls based more on their training regimens of multiple repetitions of heavy resistance16 compared to other athletes, rather than on their performance outcomes. Alternatively, recently published data by Vincent et al17 indicate altered morphology of type 2 fibers in X/X carriers. Preferential hypertrophy of type 2 fibers has been observed in bodybuilders and strength athletes,14, 18 so the findings of Vincent et al17 may represent another mechanism explaining our findings. The mechanisms underlying our findings will require further testing in future studies. No differences were noted in allele frequency between bodybuilders and powerlifters, though sample size limitations prevented a conclusive analysis in this regard.

The finding of low frequencies of the X/X genotype in Black individuals in the present study is consistent with the previous finding of low X-allele frequencies in African-descent populations.6 While no Black athletes carried the X/X genotype in the present study, this finding only approached statistical significance compared to controls, due in part to sample size limitations in the Black athlete group. Thus, verification of this result is warranted. The finding of significant deviations from HWE in the athletes but not in controls is consistent with a true genotype association.19

That higher R/R genotype frequencies were observed in controls compared to athletes does not have a clear explanation, nor do we know if the data are meaningful. The known biology of the R577X polymorphism would suggest a recessive model, with α-actinin-3 deficiency present in X/X carriers only.3 A recently developed Actn3-knockout mouse20 and the data from humans showing altered muscle fiber morphology in X/X carriers17 are beginning to address the mechanisms underlying the functional differences observed in the present and previous studies.

While cross-sectional studies to date have consistently shown that the X/X genotype is under-represented in sprint-related athletes7, 8, 9 or associated with inferior performance,10 longitudinal studies have produced more mixed results.21, 22 Additional studies are required to more completely address the influence of α-actinin-3 deficiency on muscular performance, especially focused on identifying the specific muscle phenotypes that are most affected by the deficiency (eg, multiple vs single-contraction measures). The ACTN3 findings reported here extend previous reports showing lower X/X genotype frequency in elite sprint-related athletes and provide further support for the hypothesis that α-actinin-3 protein deficiency (X/X genotype) impairs muscle performance.