To examine the interactions between acute exercise and GH on markers of bone and collagen turnover and to assess the potential for detecting GH abuse in athletes using these markers, we studied 17 aerobically trained males (age, 26.9+/-1.5 yr). Sequential studies of exercise, GH administration, and GH withdrawal were undertaken. A randomized, controlled study of rest vs. exercise showed that exercise did not change serum osteocalcin; other markers of formation increased transiently (each P<0.001): bone-specific alkaline phosphatase (+16.1%), carboxyterminal propeptide of type I procollagen (+14.1%), and procollagen III N-terminal extension peptide (+5.0%). The carboxyterminal cross-linked telopeptide of type I collagen, a bone resorption marker, increased 9.7% (P = 0.018) in response to exercise. A randomized, double blind, placebo-controlled, parallel study of recombinant human GH treatment (0.15 IU/kg x day) for 1 week increased serum osteocalcin (net increase preexercise, +/-10.0%; P = 0.017), carboxyterminal propeptide of type I procollagen (+17.6%; P = 0.002), procollagen III N-terminal extension peptide (+48.4%; P = 0.001), and carboxyterminal cross-linked telopeptide of type I collagen (53.3%; P = 0.009). Disappearance half-times after cessation of recombinant human GH for pre- and postexercise markers ranged from 248-770 h. We conclude 1) endurance exercise transiently activates bone and collagen turnover; 2) brief GH administration results in similar but quantitatively greater augmentation; and 3) these data will assist in designing a GH detection strategy.