Lack of evidence for an increase in interleukin-6 expression in adult murine bone, bone marrow, and marrow stromal cell cultures after ovariectomy†

Interleukin-6 (IL-6) has been implicated as a mediator of postmenopausal bone loss. In vitro studies of bone and bone marrow cells have suggested that estrogen regulates bone turnover by controlling the production of IL-6, a potent stimulator of osteoclastogenesis and bone resorption. To investigate this hypothesis in an in vivo model, we examined the effect of ovariectomy or estrogen replacement on IL-6 mRNA and protein expression in adult mouse bone and bone marrow in vivo and in marrow stromal cell cultures. Eight-week-old CD-1 mice were sham-operated (SHAM), ovariectomized (OVX), or ovariectomized and subcutaneously implanted with 21-day slow-release pellets containing 10 μg of 17β-estradiol (O+E). Placebo pellets were implanted in the SHAM and OVX mice. Uterine weights at 1, 2, or 3 weeks after surgery were significantly decreased (68-76%) in OVX animals compared with SHAM or O+E. In mice sacrificed at 1 or 3 weeks after surgery, we found by nonquantitative reverse transcribed polymerase chain reaction (RT-PCR), that SHAM, OVX, and O+E calvariae (CALV) constitutively expressed IL-6 mRNA. In contrast, IL-6 mRNA was either barely detectable or absent in the tibia (TIB) and bone marrow (BM). In the mice sacrificed 3 weeks after surgery, we determined by quantitative RT-PCR that IL-6 mRNA in the CALV from the OVX and O+E groups were decreased by 81 and 92%, respectively, compared with SHAM. IL-6 protein levels in the flushed bone marrow (BMSups) were detectable and were not significantly different among the groups. In bone marrow cells that were cultured for 1 week, basal levels of IL-6 protein were low and did not differ significantly among the SHAM, OVX, or O+E groups sacrificed 1, 2, or 3 weeks after surgery. After the addition of hrIL-1α, IL-6 protein levels increased 100- to 1300-fold over control. IL-6 levels in cells from animals sacrificed 2 weeks after surgery were significantly lower in the hrIL-1α-stimulated OVX and O+E groups than in hrIL-1α-stimulated SHAM cell cultures. In conclusion, in this model we could find no increase in IL-6 production with in vivo estrogen withdrawal in calvaria, long bones, bone marrow, or marrow stromal cell cultures. If increases in IL-6 expression are involved in the effects of estrogen withdrawal on bone, the magnitude of these changes are relatively small and below the limits of detection of the assays that we employed.