While in vitro studies have shown that mechanical loading can result in changes in the composition of intervertebral disc matrix, the effects of cyclic loading in vivo have not been considered. The objective of this study was to assess the effect of static and cyclic compression of different frequencies on the nuclear composition of the intervertebral disc. Thirty-six Sprague-Dawley rats were randomly divided into a control group (no pin insertion, no loading), a sham group (pins inserted in sixth and seventh caudal vertebrae, no loading), a static loading group (compression applied via pins) and cyclic loading groups (loading at 0.5, 1.5 or 2.5 Hz). Loading was applied for 1 h each day from the third to 17th day following pin insertion, and the caudal 5-6, 6-7 and 7-8 discs harvested to quantify proteoglycan content, collagen content and chondrocyte density in the nucleus pulposus. Static compression resulted in a significant reduction in total proteoglycan content as compared with the adjacent control disc, but this effect was not seen in any of the cyclic loading groups. However, comparison with the sham group appears to indicate an overall decrease in total proteoglycan content at the targeted and adjacent levels following cyclic loading. The 0.5 Hz loading group showed a significantly greater total proteoglycan content than all other compression groups, and also showed a lower total collagen content than the sham group. Results suggest that frequency dependent changes in composition occur in response to cyclic loading, but are not limited to the directly loaded disc alone. Further studies are required to verify this, but the choice of control appears to need careful consideration in all studies of this nature.