Bisphosphonates have recently been introduced in the therapeutic armamentarium for the long-term treatment of patients with multiple myeloma (MM). These pyrophosphate analogs not only reduce the occurrence of skeletalrelated events but also provide patients with a clinical benefit and improve the survival of some of them. We investigated the effects of two bisphosphonates, pamidronate and zoledronate, on both myeloma cells and bone marrow stromal cells (BMSCs). We show here that both bisphosphonates induce both myeloma cell and BMSC apoptosis. Furthermore, at lower concentrations, they induce a significant inhibition (40% and 60%, respectively) of the constitutive production of interleukin-6 (IL-6) by BMSCs. We have recently shown that BMSCs produce MMP-1, the major metalloproteinase involved in the initiation of bone resorption, production up-regulated by IL-1␤. Here, we demonstrate that zoledronate significantly inhibits MMP-1 production by BMSCs stimulated with IL-1␤ more efficiently than pamidronate. However, zoledronate and to a lesser extent pamidronate are responsible for an up-regulation of MMP-2 secretion by BMSCs. MMP-2 is involved both in bone resorption and in the metastatic process. In conclusion, the apoptosis of myeloma cells and BMSCs and the inhibition of both IL-6 and MMP-1 production induced by bisphosphonates, mainly zoledronate, could have antitumoral effects in patients with MM. However, the up-regulation of MMP-2 secretion observed in vitro suggests a putative risk of tumor cell dissemination in vivo when using these new potent bisphosphonates. This potentially deleterious effect could be abolished by combining bisphosphonates with metalloproteinase

Multiple myeloma is a very devastating cancer with a high capacity to destroy bone matrix. Matrix metalloproteinases (MMPs) play a critical role in bone remodeling and tumor invasion. In this study, we have investigated the involvement of interstitial collagenase (MMP-1) and gelatinases (MMP-2 and MMP-9) in the biology of multiple myeloma. We show (1) that myeloma cells express MMP-9 and (2) that this expression is not subjected to regulation either by interleukin-6 (IL-6), the major myeloma cell growth factor, or by other cytokines involved in the multiple myeloma cytokine network. In the tumoral environment, we show that bone marrow stromal cells express MMP-1 and MMP-2. Whereas MMP-1 is positively regulated by IL-1β, tumor necrosis factor-α, and Oncostatin M, MMP-2 is not modulated by any of these cytokines. To evaluate whether myeloma cells can modify the bone marrow stromal environment, we have examined these MMP activities in coculture. Interestingly, we have observed an upregulation of MMP-1 and a partial conversion of the proMMP-2 into its activated form. We conclude that the increase of MMP activity produced or induced by myeloma cells in these cocultures could favor bone resorption and tumor invasion. Inhibition of such activities could represent a new therapeutical approach in multiple myeloma.

Insulin-like growth factor 1 (IGF-1) is a well-known growth factor for myeloma cells. Thus, therapeutic strategies targeting IGF-1R have been proposed for multiple myeloma treatment. In this study, we investigated the effect of the antagonistic anti-IGF-1R murineAVE1642 Ab (mAVE1642). We show that mAVE1642 selectively inhibits IGF-1R but not insulin signaling in human myeloma cell lines. Since we have previously shown the functional relevance of CD45 expression in the growth of myeloma cells and the association of CD45-negative (CD45neg) status with a less favorable clinical outcome, both CD45-positive (CD45pos) and CD45neg myeloma cell lines were selected for our study. We found that mAVE1642 strongly inhibits the growth of CD45neg myeloma cell lines, leading to a G1 growth arrest, whereas it has almost no effect on the growth of CD45pos myeloma cell lines. Furthermore, mAVE1642 binding induced a significant reduction of IGF-1R expression. We next demonstrated that the overexpression of IGF-1R in the CD45pos myeloma cell line increased Akt phosphorylation but was not sufficient to sensitize these cells to mAVE1642. In contrast, we generated a stable CD45-silencing XG-1 cell line and showed that it became sensitive to mAVE1642. Thus, for the first time, we provided direct evidence that the expression of CD45 renders cells resistant to mAVE1642. Taken together, these results support that therapy directed against IGF-1R can be beneficial in treating CD45neg patients.

Previous studies have shown that bone marrow, especially the bone microenvironment, may play an important role in the pathogenesis of multiple myeloma (MM). To elucidate the relationship between myeloma cells and bone cells, mainly osteoblasts, we have established a coculture system between two interleukin-6 (IL-6)-dependent myeloma cell lines, XG1 and XG6, and the osteosarcoma cell lines Saos-2 and MG63. Both osteosarcoma cell lines have retained major functions of normal osteoblasts; principally, the capacity to produce hematopoietic growth factors (including IL-6) and osteocalcin, a noncollagenic protein essential in the bone formation process. Because IL-6 is a critical growth factor in MM, we have examined the IL-6 osteoblastic cell production in our coculture system. XG1 cells strongly upregulate IL-6 production by MG63 and Saos-2 cells. Of major interest, the triggering of IL-6 is totally dependent on the physical contact between myeloma cells and osteoblastic cells, contact that is partly mediated by CD44, CD56, and fibronectin interactions. Osteocalcin production by MG63 and Saos-2 cells has previously been shown to be dependent on 1,25-(OH)2D3. We demonstrate that XG1 and XG6 cells reduced the amount of osteocalcin in MG63 coculture cell supernatants, a reduction that is partly mediated by a soluble factor and by cell-to-cell contact. Notably, whereas one of the myeloma cell lines, XG6, has lost its capacity to stimulate IL-6 production by osteoblastic cell lines, both XG1 and XG6 cell lines remain able to reduce the osteocalcin amount, indicating that IL-6 and osteocalcin levels are regulated by two different pathways. In conclusion, these data strongly support the concept that the bone microenvironment is directly modified by contact with myeloma cells and are consistent with the characteristics observed in vivo in patients with MM patients, ie, abnormally high IL-6 and low osteocalcin levels, respectively.