Long chain polyunsaturated fatty acids and their possible interaction with phytoestrogens : impact on bone and bone cell function in vivo and in vitro : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Biochemistry at Massey University, Palmerston North, New Zealand

Abstract

Inflammation is a major contributor to postmenopausal bone loss. Various long chain polyunsaturated fatty acids (LCPUFAs), particularly those of the n-3 family, are known to have anti-inflammatory activity and may have a role in minimising postmenopausal bone loss. The objectives of this thesis were to determine whether some LCPUFAs have greater bone-protective effects than others; to identify some of the mechanisms of action of LCPUFAs in bone and to explore the possibility that combined treatment with LCPUFAs and phytoestrogens offers greater bone-protective effects than either treatment alone. Using the ovariectomised rat model for postmenopausal bone loss, the relative effectiveness of eicosapentaenoic acid (EPA, 20:5n-3), docosahexaenoic acid (DHA, 22:6n-3) and gamma-linolenic acid (GLA, 18:3n-6) in minimising bone loss post-ovariectomy was investigated. GLA exacerbated bone loss post ovariectomy in rats. In vitro, treatment of MC3T3-E1/4 osteoblast-like cells with GLA resulted in greater membrane-bound RANKL expression suggesting a possible stimulatory effect of GLA on osteoclastogenesis and osteoclast activity. EPA had no effect on overall bone mass in vivo. DHA significantly ameliorated ovariectomy-induced bone loss possibly by increasing plasma IGF-1 concentration, modulating vitamin D metabolism and, as observed in a second study, by increasing the concentration of gamma-carboxylated osteocalcin. In vitro both EPA and DHA reduced the prostaglandin E2 (PGE2)-induced increase in membrane-bound RANKL expression in MC3T3-E1/4 osteoblast-like cells. However as RANKL-independent pathways are believed to be largely responsible for the ovariectomy-induced increase in osteoclastogenesis in vivo, inhibition of RANKL expression may not significantly contribute to the prevention of ovariectomy-induced bone loss. In a second study in ovariectomised rats, combined treatment with DHA and 17β-oestradiol was associated with significantly higher femur bone mineral content than either treatment alone. However, no beneficial effects of combined treatment with DHA and either of the phytoestrogens genistein or daidzein, on bone mass were apparent. In vitro, co-treatment of TNF-α - exposed MC3T3-E1/4 cells with DHA and 17β-oestradiol was associated with a higher cell number compared to either treatment alone indicating a protective effect of combined treatment against the cytotoxic and/or anti-proliferative effects of TNF-α. In contrast, combined treatment of MC3T3-E1/4 cells with DHA and genistein, but not daidzein, was associated with significantly lower cell number than either treatment alone. As genistein, but not daidzein, is a tyrosine kinase inhibitor, this may indicate that DHA requires tyrosine kinase activity for its protective effect on cell number in TNF-α - exposed osteoblasts. Whether DHA itself is bioactive in bone cells or whether lipid mediators formed from DHA are responsible for the observed bone-protective effects is unknown. Using lipid mediator lipidomic analysis, the presence of DHA-derived lipid mediators in bone marrow in quantities known to be physiologically significant in other tissues was confirmed. Further research into the effects of these lipid mediators in bone and confirmation of the mechanisms of action of DHA in bone cells is required. This thesis demonstrates that consumption of DHA provides some protection against ovariectomy-induced bone loss in vivo and mitigates the effects of inflammation on RANKL signalling and osteoblast cell number in vitro. The bone-protective effects of DHA are complemented by co-treatment with 17β-oestradiol but may be inhibited by co-treatment with the phytoestrogens daidzein or genistein.