Is irisin a human exercise gene?

Journal name:
Date published:
Published online

Arising from P. Boström et al. Nature 481, 463468 (2012)

Boström et al. report that exercise training induces the expression of the FNDC5 gene in human muscle, producing irisin, which can convert white fat into brown fat, so enhancing metabolic uncoupling and hence caloric expenditure, and propose that this is a new health promoting hormone1. This assertion is based on experimental evidence that exogenous FNDC5 induces uncoupling protein 1 (UCP1) expression in white subcutaneous adipocytes; overexpression of FNDC5 in liver (elevating systemic irisin) prevents diet-induced weight gain and metabolic dysfunction and stimulates oxygen consumption in mice; and FNDC5 mRNA expression levels double after exercise training in eight human skeletal muscle samples. However, the UCP1 induction was lower than observed during Brite2 formation or the level associated with an improved diabetes profile in humans3. Here we demonstrate that muscle FNDC5 induction occurs only in a minority of subjects—whereas all types of exercise training programmes4, 5, 6, 7, in the vast majority of people, yield some gain in cardiovascular or metabolic health, in our analysis of ~200 subjects muscle FNDC5 was increased only in highly active elderly subjects, whereas FNDC5 expression was unrelated to metabolic status, which casts doubt over the general relevance of skeletal muscle FNDC5 to human health.

At a glance


  1. Irisin is not routinely activated by exercise in humans.
    Figure 1: Irisin is not routinely activated by exercise in humans.

    a, mRNA expression in young sedentary males before and 24 h after 6 weeks intense endurance cycling (n = 24); first published in 2005 (ref. 8) (FDR using SAMR >10%). b, Correlation between FNDC5 and the change in aerobic capacity after 6 weeks supervised training (n = 24, R2 = 0.05, P = 0.37)8. c, FNDC5 mRNA expression in subjects (n = 43) before and after supervised resistance training12 (n = 43, P = 0.6). d, Change in fasting insulin (P = 0.03) and glucose (P = 0.04) levels after 20 weeks of training. e, Correlation of the change in fasting insulin versus the change in FNDC5 mRNA levels after 20 weeks of training (R2 = 0.01, P = 0.7). f, FNDC5 mRNA expression in young and older sedentary and age-matched endurance trained subjects (n = 10 per group, P = 0.52 and P = 0.02, respectively, derived from gene-chip study (GEO accession GSE9103). Data are MAS5 normalized, quality controlled for outliers. Significance analysis of microarrays paired or unpaired t-tests were used as appropriate. Sed., sedentary; Ex., exercise trained. y axis units are arbitrary units. Error bars indicate standard error.

  2. Irisin expression is not related to diabetes status in humans.
    Figure 2: Irisin expression is not related to diabetes status in humans.

    Plots correlating body mass index (BMI; a), 2-h glucose (b), fasting insulin (c) and fasting glucose (d) with skeletal muscle FNDC5 mRNA expression in 118 diabetes medication-free subjects aged ~55 years demonstrates that FNDC5 mRNA expression does not relate to peripheral insulin sensitivity (for example, fasting glucose (R2 = 0.01, not significant), fasting insulin (R2 = 0.042, not significant)) or glucose response during a glucose tolerance test (R2 = 0.016, not significant)13. Furthermore, rather than being expressed at higher levels in the leaner subjects, there was a statistically significant positive association between BMI and FNDC5 mRNA abundance (R2 = 0.1, P = 0.004), albeit with a very modest degree of shared variance (~10%). Data are MAS5 normalized, quality controlled for outliers and log-transformed. Linear regression analysis was carried out and adjusted for four dependent tests.

Accession codes

Primary accessions

Gene Expression Omnibus


  1. Boström, P. et al. A PGC1-α-dependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature 481, 463468 (2012)
  2. Petrovic, N. et al. Chronic peroxisome proliferator-activated receptor γ (PPARγ) activation of epididymally derived white adipocyte cultures reveals a population of thermogenically competent, UCP1-containing adipocytes molecularly distinct from classic brown adipocytes. J. Biol. Chem. 285, 71537164 (2010)
  3. Timmons, J. A. & Pedersen, B. K. The importance of brown adipose tissue. N. Engl. J. Med. 361, 415416 (2009)
  4. Boule, N. G. et al. Effects of exercise training on glucose homeostasis: the HERITAGE family study. Diabetes Care 28, 108114 (2005)
  5. Holten, M. K. et al. Strength training increases insulin-mediated glucose uptake, GLUT4 content, and insulin signaling in skeletal muscle in patients with type 2 diabetes. Diabetes 53, 294305 (2004)
  6. Keller, P. et al. Using systems biology to define the essential biological networks responsible for adaptation to endurance exercise training. Biochem. Soc. Trans. 35, 13061309 (2007)
  7. Babraj, J. A. et al. Extremely short duration high intensity interval training substantially improves insulin action in young healthy males. BMC Endocr. Disord. 9, 3 (2009)
  8. Timmons, J. A. et al. Human muscle gene expression responses to endurance training provide a novel perspective on Duchenne muscular dystrophy. FASEB J. 19, 750760 (2005)
  9. Keller, P. et al. A transcriptional map of the impact of endurance exercise training on skeletal muscle phenotype. J. Appl. Physiol. 110, 4659 (2011)
  10. Melov, S., Tarnopolsky, M. A., Beckman, K., Felkey, K. & Hubbard, A. Resistance exercise reverses aging in human skeletal muscle. PLoS ONE 2, e465 (2007)
  11. Vollaard, N. B. J. et al. Systematic analysis of adaptations in aerobic capacity and submaximal energy metabolism provides a unique insight into determinants of human aerobic performance. J. Appl. Physiol. 106, 14791486 (2009)
  12. Phillips, B. et al. Resistance exercise training improves age-related declines in leg vascular conductance and rejuvenates acute leg blood flow responses to feeding and exercise. J. Appl. Physiol. 112, 347353 (2011)
  13. Gallagher, I. J. et al. Integration of microRNA changes in vivo identifies novel molecular features of muscle insulin resistance in type 2 diabetes. Genome Med. 2, 9 (2010)

Download references

Author information


  1. University of London, Royal College Street, London NW1 0TU, UK

    • James A. Timmons &
    • Peter K. Davidsen
  2. University of California Davis, Davis, California 95616, USA

    • Keith Baar
  3. University of Nottingham, Derby Royal Infirmary, Derby DE22 3DT, UK

    • Philip J. Atherton


J.A.T. and P.K.D. carried out the data analysis while J.A.T., K.B. and P.J.A. drafted the article. J.A.T., P.K.D., K.B. and P.J.A. edited the final article.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Additional data