Differential effects of arachidonic acid and docosahexaenoic acid on cell biology and osteoprotegerin synthesis in osteoblast-like cells

Abstract

The purpose of the study was to elucidate the mechanisms by which polyunsaturated fatty acids (PUFAs) prevent bone loss. MG-63 human osteoblasts and MC3T3-E1 murine osteoblasts were exposed to the n-6 PUFA arachidonic acid (AA) and the n-3 PUFA docosahexaenoic acid (DHA) as well as oestrogen (E2) and parathyroid hormone (PTH) and the effects thereof tested on a variety of biological parameters characteristic of osteoblasts. These parameters included prostaglandin E2 (PGE2) synthesis, proliferation, differentiation to mature mineralising osteoblasts as well as osteoprotegerin (OPG) and receptor activator of nuclear factor êB ligand (RANKL) secretion. Results showed that AA stimulates PGE2 production significantly in both cell lines. Stimulated PGE2 production by MC3T3-E1 cells however, was significantly higher, which might be attributed to auto-amplification by PGE2 itself in this cell line. Pre-incubation of the MG-63 cells with cyclo-oxygenase (COX)-blockers inhibited PGE2 production significantly, suggesting that both COX enzymes were involved in PGE2 synthesis. The number of functional osteoblasts is important for bone formation therefore in vitro osteoblastic cell proliferation was investigated. In contrast to the hormones E2 and PTH, both AA and DHA inhibited proliferation significantly. The AA-mediated anti-proliferative effect is possibly independent of PGE2 production, as PGE2 per se had little effect on proliferation. DHA inhibited proliferation of MG-63 cells more severely, which might be attributed to the osteosarcoma nature of the MG-63 cells. The anti-proliferative effect of these PUFAs might be attributed to modulation of cell cycle progression or anti-mitotic effects of PUFA peroxidation products. Morphological studies showed apoptotic cells after DHA exposure in MG-63 cells. There is a reciprocal relationship between reduced proliferation and the subsequent induction of cell differentiation in vitro. High basal levels of alkaline phosphatase (ALP) activity, a marker of the mature mineralising osteoblastic phenotype, were detected in MC3T3-E1 cells. Long-term exposure to AA inhibited ALP activity in these cells. This process might be PGE2-mediated. Exposure to PUFAs, however, did not compromise the ability of the MC3T3-E1 cells to differentiate to mature mineralising osteoblasts. In contrast with MC3T3-E1 cells, MG-63 cells demonstrated low basal ALP activity and were unable to differentiate to mature mineralising osteoblasts. In the absence of osteogenic-inducing supplements, PUFAs induced adipocyte-like features that might be due to the expression of high levels of PPARã in this cell line. Lipid-filled vacuoles were absent in the MC3T3-E1 cells suggesting that the MC3T3-E1 cell line may not express PPARã mRNA. The study furthermore demonstrated that PUFAs are able to modulate OPG and RANKL secretion in osteoblasts. AA inhibited OPG secretion dose-dependently in both cell lines, this could be PGE2-mediated. AA dose-dependently stimulated soluble RANKL (sRANKL) secretion in MC3T3-E1 cells thereby affecting the OPG/RANKL ratio in a negative way, supporting various reports that AA and PGE2 do cause bone resorption. No sRANKL could be detected after exposing the MC3T3-E1 cells to DHA suggesting that DHA could be protective to bone. In conclusion, contrary to in vivo evidence, this in vitro study could not indisputably demonstrate protective effects of PUFAs on the osteoblastic cell lines tested.