In this study, possible mechanisms underlying soleus muscle atrophy after spinal cord transection and attenuation of atrophy with cycling exercise were studied. Adult female Sprague-Dawley rats were divided into three groups; in two groups the spinal cord was transected by a lesion at T10. One group was transected and killed 10 days later, and another group was transected and exercised for 5 days starting 5 days after transection. The third group served as an uninjured control. All animals received a continuous-release 5'-bromo-2'-deoxyuridine pellet 10 days before they were killed. Transection alone and transection with exercise lead to activation of satellite cells, but only the exercise group showed a trend toward an increase in the number of proliferating satellite cells. In all cases the number of activated satellite cells was significantly higher than the number that divided. Although the number of cells undergoing proliferation increased with exercise, no increase in fusion of satellite cells into muscle fibers was apparent. Spinal cord transection resulted in a 25% decrease in myonuclear number, and exercise was not associated with a restoration of myonuclear number. The number of apoptotic nuclei was increased after transection, and exercise attenuated this increase. However, the decrease in apoptotic nuclei with exercise did not significantly affect myonuclear number. We conclude that apoptotic nuclear loss likely contributes to loss of nuclei during muscle atrophy associated with spinal cord transection and that exercise can maintain muscle mass, at least in the short term, without restoration of myonuclear number.