Increased serum levels of inflammatory mediators have been associated with numerous disease states including atherosclerosis, Type II diabetes, hypertension, depression, and overall mortality. We hypothesized that a long-term exercise intervention among older adults would reduce serum inflammatory cytokines, and this reduction would be mediated, in part, by improvements in psychosocial factors and/or by beta-adrenergic receptor mechanisms. Adults age 64 were randomly assigned to either an aerobic exercise treatment (CARDIO) or a flexibility/strength exercise treatment (FLEX) 3 days/week, 45 min/day for 10 months. A subgroup of subjects treated with non-selective beta(1)beta(2) adrenergic antagonists were included to evaluate the potential role of beta-adrenergic receptor adaptations as mediators of an exercise-induced change in inflammation. The inflammatory mediators [C-reactive protein (CRP), IL-6, tumor necrosis factor (TNF)-alpha, and IL-18] and the psychosocial factors (depression, perceived stress, optimism, sense of coherence, and social support) were measured pre- and post-intervention. The CARDIO treatment resulted in significant reductions in serum CRP, IL-6, and IL-18 compared to the FLEX treatment (significant treatment x time interaction, p<.05), whereas TNFalpha declined in both groups (main effect of time, p=.001). However, several psychosocial factors (depression, optimism, and sense of coherence) improved in both groups suggesting that the reduction of CRP, IL-6, and IL-18 in the CARDIO group was not mediated by improvements in psychosocial scores. With respect to the potential role of beta-adrenergic receptors, both CARDIO subjects treated with beta-adrenergic antagonists and those who were not treated with those medications demonstrated similar reductions in serum CRP, IL-6, IL-18, and TNFalpha. In summary, we have observed that an aerobic exercise intervention can significantly reduce serum inflammatory mediators, but beta-adrenergic receptors and psychosocial factors do not appear to be involved.