Despite significant improvements in patient survival and local disease control, 25% of patients with osteosarcoma develop metastases and the surgery for many extremity lesions still includes amputation. Thus, modalities that inhibit tumour growth and the metastatic cascade will have a significant impact on patient survival and potential for limb sparing surgery.Tumours induce bone destruction by cellular processes, such as osteoclast-mediated bone lysis. The importance of osteoclastmediated lysis has been demonstrated in other malignancies, where the development of osseous metastases has been shown to be mediated by soluble tumour-related osteoclast activating factors (Galasko, 1976). More recently, the induction of specific osteoclast recruiting factors in osteoblasts/stromal cells following contact with myeloma and breast cancer cells has been demonstrated (Chikatsu et al, 2000). Bisphosphonates have recognized efficacy in reducing bone destruction, pain and pathological fracture in a variety of lytic primary and metastatic diseases of the skeleton (Thiébaud et al, 1991;Coleman and Purohit, 1993;Shipman et al, 1997;Bloomfield, 1998;Diel et al, 1998). They have more recently been shown to inhibit establishment and growth of prostate cancer metastases, which are generally considered to be osteoblastic in their growth pattern (Adami, 1997). We proposed that, as with the intraosseous growth of prostate cancer, the local spread of osteosarcoma involves proteolytic and osteoclast-mediated bone destruction. The bisphosphonates may reduce bone destruction by uncoupling close regulation of osteoclast activity by osteoblast-like cells.The aims of our study were to examine the effects on the regulation of proliferation and apoptosis by bisphosphonate treatment of a clonal osteosarcoma cell line. Further, we assessed the expression of an osteoclast differentiating factor, receptor activator of NF-κβ (RANKL) (Yasuda et al, 1998b) and an osteoclastogenesis inhibitory factor, osteoprotegerin (OPG) (Simonet et al, 1997;Yasuda et al, 1998a). Finally, we investigated the effects on osteoblast-related gene expression by an aminobisphosphonate and a non-aminobisphosphonate. The use of a bisphosphonate from each group allows comparison of their relative potencies with the known effects on osteoclasts and bone resorption studies (Fleisch, 1993).
MATERIALS AND METHODS
Cell cultureUMR 106-01 cells derived from a 32 P-induced osteosarcoma in rats (Martin et al, 1976) were used. Cells were cultured in 75 cm 2 tissue culture flasks (Greiner Cellstar) in α-modified Minimal Essential Medium (α-MEM) containing hepes 4 g l Ϫ1 , sodium bicarbonate 1.95 g 6 Ϫ1 , gentamicin 80 mg 6 Ϫ1 , pH 7.4 and 10% fetal bovine serum (FBS), incubated at 37˚C and equilibrated in 5% CO 2 in air. Subcultures were performed using 0.0125% trypsin in 0.5 mM Na 2 EDTA in calcium and magnesium-free phosphate buffer (1 ϫ versene) to harvest the cells. Summary Local growth of osteosarcoma involves destruction of host bone by proteolytic mechanisms and/or host osteocl...