Jesper Olesen:
Role of PGC-1α in acute and low-grade inflammation

Date: 13-08-2013    Supervisor: Henriette Pilegaard

The aim of the present thesis was to examine the role of the exercise-induced transcriptional co-activator, PGC-1α, in acute and low-grade inflammation. To investigate this, the following three hypotheses were tested: 1) Skeletal muscle PGC-1α plays an important role in acute LPS-induced systemic inflammation as well as in the inflammatory response in mouse skeletal muscle. 2) Long-term exercise training and/or resveratrol supplementation prevents age-associated low-grade- and skeletal muscle inflammation in mice with PGC-1α being required for these improvements. 3) Exercise training and/or resveratrol supplementation reduces systemic- as well as skeletal muscle inflammation in aged human subjects.

Study I demonstrated an impaired LPS-induced plasma TNFα and skeletal muscle TNFα response in PGC- 1α muscle specific knockout mice compared with WT mice. Conversely, mice with transgenic overexpression of PGC-1α in skeletal muscle showed a greater fold increase in plasma TNFα than WT mice, when stimulated with LPS. Taken together, these results suggest that skeletal muscle PGC-1α is required for a robust LPS-induced TNFα response.

Study II demonstrated that plasma TNFα and IL-6 as well as liver TNFα mRNA and protein, visceral adipose tissue TNFα mRNA and skeletal muscle TNFα protein were all increased in old mice (15 month old) compared with young mice (3 month old), confirming that aging is associated with systemic low-grade inflammation and tissue inflammation in mice.

Study II and III demonstrated that exercise training reduced skeletal muscle TNFα protein content and systemic IL-6 levels in mice and skeletal muscle TNFα mRNA content in aged human subjects. This importance of physical activity in reducing inflammation is supported by results from our inactivity study (7 days of bed-rest) in young men, showing increased inflammation as evidenced by enhanced skeletal muscle IL-6 mRNA and adipose tissue iNOS mRNA content. In conjunction, these results may indicate that skeletal muscle inflammation is inversely related to the level of physical activity. However, no clear association between the physical activity level and the level of systemic inflammation existed in aged exercise trained human subjects or in young inactive human subjects.

Study II further demonstrated that PGC-1α was required for the exercise training-induced prevention of an age-associated increase in skeletal muscle TNFα protein content. However, PGC-1α was not mandatory for the exercise training-induced reductions in systemic IL-6 in mice, suggesting that additional factors contribute to the systemic anti-inflammatory effects of exercise training.

Study II and III demonstrated that while resveratrol increased the protein content of the anti-oxidant enzyme GPX1 and appeared to reduce oxidative stress in mouse skeletal muscle, resveratrol did not elicit any antiinflammatory effects neither in mice nor in human subjects. In contrast, resveratrol even impaired the exercise training-induced reduction in protein carbonylation and TNFα mRNA in human subjects.

Study II demonstrated that the minor effects on GPX1 and oxidative stress observed with resveratrol in mice were independent of PGC-1α and these findings were further supported by preliminary in vitro data showing that the resveratrol-induced up-regulation of cyt c mRNA in primary myotubes was independent of PGC-1α.

In conclusion, skeletal muscle PGC-1α was required for a robust LPS-induced TNFα response. No antiinflammatory effects of resveratrol were observed in mice or human subjects. Skeletal muscle inflammation was increased by physical inactivity in humans and reduced by exercise training in mice and human subjects. The observed anti-inflammatory effect of exercise training in mice was partly mediated through PGC-1α.