A well-balanced bone structure involves a balanced role of the osteoclasts, osteoblasts, and osteocytes. The balance is maintained by different internal and external factors like calcium demands, the immune system and its cytokines, or mechanical loading.

To reduce inflammation and bone loss in arthritis, we subjected arthritic bones in male mice to mechanical loading and a specific COX-2 inhibitor and investigated their effects in paper I. COX-2 is one of the genes upregulated in response to loading. However, inhibiting it did not interfere with the loading response, shown by prevention of bone loss, as also seen in a similar study with female mice, in addition to decreased swelling in the knee. This study could support physicians in recommending exercise to arthritic patients with COX-2 specific inhibitors to relieve pain. However, contrary to the female study, the osteoclast number was unchanged in our study.

Owing to the sex differences seen in osteoclast number, we studied the sex-specific gene expression in osteoclast differentiation in paper II. Previous studies have shown discrepancies to the question if bone marrow cells from males or females differentiate faster into osteoclasts. Therefore, we selected a naïve source of unselected precursors from bone marrow cells. We observed that even though the female cells adhered better than the male cells, eventually the male cells differentiated faster into osteoclasts. Gene expression analysis indicated that female cells have a higher expression of adhesion genes, while male cells are inclining towards differentiation genes, and suggest involvement of the PI3K-AKT and the CSF1R-RANK signalling pathways.

Further, to understand the role of TRAP, a known osteoclast marker, we studied the sex- and site- specific differences in bone of TRAP knockout mice in paper III. We found that in contrast to females, male KO mice has a wider proximal tibia with more trabecular bone, and decreased number and density of bony bridges in the growth plate compared to the wild-type controls. Thus, the differences between the sexes are site-specific, which in part depends on TRAP in mice. Since the role of TRAP, specially the 5b isoform in osteoclasts, is important for the bone phenotype, it would be a suitable marker in prognosis or diagnosis of different bone-associated diseases. Therefore, in paper IV we developed a double sandwich TRAP ELISA with antibodies specific to TRAP 5a or the whole TRAP protein, to detect both isoforms in the same sample. TRAP 5b was associated to resorption activity, and the 5b/5a ratio correlated with CTX levels, indicating a relation between osteoclast number and activity with TRAP 5b, as well as a minimal 5a involvement in the remodelling process.

To summarise, this thesis contributes to the existing knowledge on mechanical loading, proving its positive effects in a diseased state, such as arthritis. Further, it deepens our understanding around osteoclast gene expression and the role of its crucial marker, TRAP, in bone development in the two sexes.