Article Text

Download PDFPDF
4 Associations of bone mineral density-related genes and marathon performance in elite european caucasian marathon runners
  1. AJ Herbert1,
  2. AG Williams1,
  3. SJ Lockey2,
  4. RM Erskine3,
  5. PJ Hennis1,
  6. C Sale4,
  7. SH Day1,
  8. GK Stebbings1
  1. 1MMU Sports Genomics Laboratory, Department of Exercise and Sport Science, Manchester Metropolitan University, Crewe, UK
  2. 2School of Medical Education, Newcastle University, Newcastle-Upon-Tyne, UK
  3. 3School of Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK
  4. 4Musculoskeletal Physiology Research Group, Sport, Health and Performance Enhancement Research Centre, Nottingham Trent University, Nottingham, UK


Bone mineral density (BMD) is a multi-factorial phenotype determined by factors such as physical activity, diet and a sizeable genetic component. Athletic populations tend to possess higher BMD than non-athletes due to a larger volume of exercise completed. Despite this, some endurance runners can possess low BMD and/or suffer stress fractures, which can have negative impacts on their health and performance. Therefore, we hypothesised that elite endurance runners would possess a genotype associated with enhanced BMD and a reduced risk of injury, resulting in less training interruption and greater potential success. The study compared the genotype and allele frequencies of 5 genetic variants associated with BMD (LRP5 rs3736228, TNFRSF11B rs4355801, VDR rs2228570, WNT16 rs3801387, AXIN1 rs9921222) in elite (men <2 hour 30 min, n=110; women <3 hour 00 min, n=98) and sub-elite (men 2 hour 30 min – 2 hour 45 min, n=181; women 3 hour 00 min – 3 hour 15 min, n=67) marathon runners with those of a non-athlete control population (n=474). We also investigated whether marathon personal best time was associated with a more ‘advantageous’ BMD genotype. Congruent with our hypothesis, the ‘risk’ T allele for the AXIN1 rs9921222 polymorphism was 5% more frequent in the control group than in sub-elites (p=0.030, χ2=4.69) but no further differences were observed for this variant (p≥0.083, χ2 ≤4.98). WNT16 rs3801387 genotype frequency differed between athletes and controls (p=0.002, χ2=12.02) and elites vs controls (p=0.008, χ2=9.72), as did allele frequency. However, contrary to our hypothesis, it was the ‘risk’ A allele that was ~5% more frequent in athletes than controls. Similarly, when combining data from all 5 variants, the athletes had a lower Total Genotype Score than controls (53.6 vs 65.7; p≤0.001), again suggesting greater genetic susceptibility to bone injury in athletes. Personal best times were not associated with genotype in any comparison. These results suggest that high-level endurance runners do not benefit from genetic resistance to bone injury and a resulting ability to sustain large training volumes, contradicting our hypothesis. High-level endurance runners appear to be at a higher risk of bone injury from a genetic perspective, for as yet unexplained reasons, although large inter-individual differences in genetic risk exist.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.