Oestrogen inhibits bone resorption, at least in part, by regulating the production of several cytokines, including interleukin-6 (IL-6), IL-1, receptor activator of nuclear factor kappaB ligand (RANKL) and osteoprotegerin (OPG) by cells of the osteoblastic lineage. The selective oestrogen receptor modulator raloxifene (RAL) acts on bone in a similar manner to oestrogen, although the mechanisms of action of RAL on osteoblasts still remain unclear. We investigated and compared the effects of 17-beta oestradiol (E(2)) and RAL on the regulation of IL-6, IL-1, RANKL and OPG in vitro in primary human osteoblastic (HOB) cells and in an immortalised clonal human bone marrow stromal cell line (HCC1) with osteoblastic characteristics. We tested E(2) and RAL at concentrations ranging from 10(-12) to 10(-6) M. IL-6, IL-1alpha and IL-1beta, OPG and RANKL were measured by ELISA. RANKL and OPG mRNA steady state level was assessed by quantitative PCR analysis. Both E(2) and RAL led to a significant reduction in IL-6 production in the HOB cells, although the effect was more marked with E(2) (P<0.05). IL-1alpha and IL-1beta also decreased significantly following treatment with E(2) and RAL in the HCC1 cells (E(2) (10(-8), 10(-7) and 10(-6) M), % reduction (means+/-S.E.M.) compared with vehicle-treated cells - IL-1alpha: 84+/-7.4, 70.8+/-2.9*, 78.2+/-4.8*; IL-1beta: 79+/-10, 72.8+/-8.2*, 66.6+/-2.8*; RAL (10(-8), 10(-7) and 10(-6) M) - IL-1alpha: 72.4+/-5*, 79+/- 5.2*, 102+/-7.7; IL-1beta: 67.9+/-3.2*, 69+/-2.5*, 73.8+/- 6.2*; *P<0.05). OPG protein concentration decreased significantly in a dose-dependent manner following treatment with E(2) and RAL (% reduction E(2) (10(-8), 10(-7) and 10(-6) M) - HOB: 72.5+/-8.4*, 80+/-6.7*, 62.8+/-8.9*; HCC1: 109+/-4, 98.8+/-6, 54.5+/-3.4*; RAL (10(-8), 10(-7) and 10(-6) M) - HOB: 81.5+/-5.5*, 62.7+/-7.4*, 55.2+/-10.9*; HCC1: 92.7+/-7.4, 67+/-12.2*, 39+/-4.5*; *P<0.05). In the HCC1 cells, RANKL protein did not change significantly following E(2). In contrast, a significant reduction in RANKL was seen with RAL at 10(-7) and 10(-6) M (66+/-6.4% and 74+/-3% respectively). There was no change in OPG mRNA expression following E(2) or RAL in the HCC1 cells, although in the HOB cells we observed a significant reduction in OPG mRNA. RANKL mRNA decreased significantly in the HCC1 cells following RAL (10(-8), 10(-7)and 10(-6) M) treatment (% change from controls: 52+/-2*, 62+/-1*, 53+/-5.8*; *P<0.05). Similar results were seen in the HOB cells with RAL at 10(-6) M (RANKL mRNA: 72+/-5.5, P<0.05). In addition, there was a significant decrease in the RANKL/OPG ratio after RAL at 10(-6) M (HOB: 65.6+/-5*, HCC1: 56.9+/-20*; *P<0.05). RANKL/OPG ratio did not change significantly in the HCC1 cells following E(2). However, in contrast to RAL, we observed an increase in the RANKL/OPG ratio in the HOB cells following treatment with E(2). In conclusion, the study shows that RAL and E(2) have divergent cell-specific effects on the regulation of cytokines. The data also suggest that, in contrast to E(2), RAL may exert its anti-resorptive actions, at least in part, via the RANKL/OPG pathway. Further in vivo studies are required to confirm this.
Interleukin-6 (IL-6), IL-1, receptor activator of nuclear factor kappaB ligand (RANKL) and osteoprotegerin production by human osteoblastic cells: comparison of the effects of 17-beta oestradiol and raloxifene
