Identification of new targets for therapy of osteolytic bone disease in multiple myeloma.

One of the most characteristic features of multiple myeloma is the development of osteolytic bone lesions. Myeloma-associated bone disease is caused by an increase in osteoclastic bone resorption and a decrease in osteoblastic new bone formation. Insight into the molecular mechanisms of osteoclastogenesis has been provided by the detection of receptor activator of NF-kappaB ligand (RANKL), its specific receptor (RANK) and its decoy receptor antagonist osteoprotegerin (OPG). The RANK signaling system is abnormally regulated in multiple myeloma and targeting this system may ameliorate myeloma bone disease. Less is known about the development of osteoblastic dysfunction, and further knowledge about the interaction between myeloma cells and osteoblasts is required. The aim of this review is to focus on the principles of bone biology for a better understanding of the development of myeloma bone disease and to identify possible therapeutic targets.

Salvage bortezomib-dexamethasone and high-dose melphalan (HDM) and autologous stem cell support (ASCT) in myeloma patients at first relapse after HDM with ASCT. A phase-2 trial.

Until recently, only retrospective studies had been published on salvage high-dose melphalan (HDM) with autologous stem cell 'transplantation' (ASCT). In a prospective, nonrandomized phase-2 study, we treated 53 bortezomib-naïve patients with bortezomib-dexamethasone as induction and bortezomib included in the conditioning regimen along with the HDM. Median progression-free survival (PFS), time to next treatment (TNT) and overall survival (OS) after start of reinduction therapy were 21.6, 22.8 and 46.6 months, respectively. For 49 patients who completed salvage bortezomib-HDM(II) with ASCT, there was no significant difference of PFS and TNT after HDM (II) compared with after the initial HDM(I), and thus patients were their own controls (PFS (I: 20.1 vs II: 19.3 months (P=0.8)) or TNT (I: 24.4 vs II: 20.7 months (P=0.8)). No significant differences in the response rates after salvage ASCT compared with the initial ASCT. Bortezomib-HDM conditioning combo was feasible, and toxicity was as expected for patients treated with bortezomib and ASCT. In conclusion, in bortezomib-naïve patients treated at first relapse with salvage ASCT including bortezomib, PSF and TNT did not differ significantly from initial ASCT and median OS was almost 5.5 years with acceptable toxicity. A recent prospective randomized study confirms salvage ASCT to be an effective treatment.

Leptin is expressed in and secreted from primary cultures of human osteoblasts and promotes bone mineralization.

The adipose hormone leptin and its receptor are important for regulation of food intake and energy metabolism. Leptin also is involved in the growth of different tissues. In this study, we show the expression of leptin in primary cultures of normal human osteoblasts (hOBs) as evidenced by reverse transcriptase-polymerase chain reaction (RT-PCR) and immunocytochemistry. Release of leptin into the medium also was found. Leptin was not detected in commercially available hOBs (NHOst) or in three different human monoclonal osteosarcoma cell lines. Leptin expression was observed in OBs in the mineralization and/or the osteocyte transition period but not during the matrix maturation period. Furthermore, hOBs and osteosarcoma cell lines expressed the long signal-transducing form of the leptin receptor (OB-Rb) as shown by RT-PCR. We observed no significant changes in leptin or OB-Rb genes in hOBs after incubation with recombinant leptin, indicating no autoregulation of the leptin expression. Incubation of both hOBs entering the mineralization phase and osteosarcoma cell lines with recombinant leptin markedly increased the number of mineralized nodules as shown by alizarin S staining. These findings indicate that leptin may be of importance for osteoblastic cell growth and bone mineralization.

Comparison of the effects of 2-chlorodeoxyadenosine and melphalan myeloma cell lines.

We have studied the effects of 2-chlorodeoxyadenosine (2-CdA) and melphalan on DNA synthesis and cell proliferation of five myeloma cell lines and one primary culture of highly purified myeloma cells. The DNA synthesis was estimated by thymidine incorporation and cell death was estimated by a coluorimetric assay sensitive to mitochondrial activity. The concentrations of 2-CdA giving 50% inhibition of DNA synthesis (ID50) in four cell lines were 8, 100, 500 and 2500 nM, whereas the corresponding concentrations of melphalan were 600, 600, 1000 and 7500 nM, respectively. In one cell line, the ID50 for melphalan was 400 nM, whereas 2-CdA apparently was without inhibitory effect and the ID50 was not reached. The ID50 for 2-CdA in the primary culture was 250 nM. When compared on a molar basis, 2-CdA had a more potent inhibitory effect than melphalan on four out of five myeloma cell lines. This is in contrast to clinical experiments which have shown lack of effect of 2-CdA in myeloma patients. Our study shows that 2-CdA has a marked heterogeneous effect on myeloma cell lines and opens up the possibility that a similar variation in sensitivity may also exist among myeloma cell clones in vivo.

Syndecan-1 is targeted to the uropods of polarized myeloma cells where it promotes adhesion and sequesters heparin-binding proteins.

Syndecan-1 (CD138) is a heparan sulfate-bearing proteoglycan present on the surface of myeloma cells where it mediates myeloma cell-cell and cell-extracellular matrix adhesion. In this study, we examined myeloma cell lines for cell membrane localization of syndecan-1. On some cells we note a striking localization of syndecan-1 to a single small membrane protrusion, with the remainder of the cell surface being mostly negative for syndecan-1. Examination of cell morphology reveals that a proportion of cells from myeloma cell lines, as well as primary myeloma cells, are polarized, with a uropod on one end and lamellipodia on the other end. On these polarized cells, syndecan-1 is specifically targeted to the uropod, but in contrast, on nonpolarized cells syndecan-1 is evenly distributed over the entire cell surface. In addition to syndecan-1, several other cell surface molecules localize specifically to the uropod, including CD44 and CD54. Functional assays reveal that myeloma cell lines with a high proportion of polarized cells have a much higher migratory potential than cell lines with few polarized cells. Moreover, the uropod is the cell pole preferentially involved in aggregation of myeloma cells and in adhesion of myeloma cells to osteoblast-like cells. When polarized myeloma cells are incubated with heparin-binding proteins, like hepatocyte growth factor or osteoprotegerin, they concentrate in the uropod. These data indicate that syndecan-1 is targeted to the uropod of polarized myeloma cells and that this targeting plays a role in promoting cell-cell adhesion and may also regulate the biological activity of heparin-binding cytokines.

Bone morphogenetic protein-4 inhibits proliferation and induces apoptosis of multiple myeloma cells.

Bone morphogenetic proteins (BMPs) can be isolated from organic bone matrix and are able to initiate de novo cartilage and bone formation. Here it is shown that BMP-4 inhibited DNA synthesis in a dose-dependent manner in 3 IL-6-dependent multiple myeloma (MM) cell lines (OH-2, IH-1, and ANBL-6). In contrast, no effect on DNA synthesis was observed in 3 IL-6-independent MM cell lines (JJN-3, U266, and RPMI 8226). BMP-4 induced cell cycle growth arrest in the G(0)/G(1) phase in OH-2 and ANBL-6 cells but not in IH-1 cells. BMP-4 induced apoptosis in OH-2 and IH-1 cells, but not significantly in ANBL-6 cells. Furthermore, BMP-4 induced apoptosis in freshly isolated MM cells from 4 of 13 patients. In the OH-2 and ANBL-6 cell lines and in a patient sample, immunoblotting showed that BMP-4 down-regulated IL-6-induced tyrosine phosphorylation of Stat3, suggesting a mechanism for the apparent antagonism between IL-6 and BMP-4. BMP-4 or analogues may be attractive therapeutic agents in MM because of possible beneficial effects on both tumor burden and bone disease.

Hepatocyte growth factor (HGF) induces interleukin-11 secretion from osteoblasts: a possible role for HGF in myeloma-associated osteolytic bone disease.

Multiple myeloma is associated with unbalanced bone remodeling causing lytic bone lesions. Interleukin-11 (IL-11) promotes osteoclast formation and inhibits osteoblast activity and may, thus, be one factor involved in cancer-induced bone destruction. We have previously shown that myeloma cells produce hepatocyte growth factor (HGF). We now report that HGF induces IL-11 secretion from human osteoblast-like cells and from the osteosarcoma cell lines Saos-2 and HOS. In coculture experiments, both the myeloma cell line JJN-3 and primary myeloma cells from 3 patients induced IL-11 secretion from osteoblasts, whereas no induction was observed with the non-HGF producing myeloma cell line OH-2. Enhanced IL-11 induction was observed with physical contact between osteoblasts and myeloma cells as compared with experiments in which contact was prohibited by tissue inserts. Anti-HGF serum strongly reduced the myeloma cell-induced IL-11 secretion. Furthermore, we show that JJN-3 cells express HGF on the cell-surface. Removal of surface-bound HGF on JJN-3 cells reduced IL-11 production induced in cocultures. Transforming growth factor beta1 and IL-1 potentiated the effect of HGF on IL-11 secretion, whereas an additive effect was observed with tumor necrosis factor. Thus, myeloma-derived HGF can influence the bone marrow environment both as a soluble and a surface-bound factor. Furthermore, HGF emerges as a possible factor involved in myeloma bone disease by its ability to induce IL-11.

Expression of urokinase plasminogen activator and the urokinase plasminogen activator receptor in myeloma cells.

Binding of urokinase (uPA) to its receptor (uPAR; CD87) focuses proteolytic activity on the cell surface and this system is of importance in malignant matrix degradation and tumour invasion. By immunocytochemistry and flow cytometry, we found that primary myeloma cells and myeloma cell lines expressed uPA and uPAR. Soluble uPA was present in cell line supernatants and lysates in low concentrations. In cell lines, uPA and uPAR were located both on the cell surface and intracellularly, but the expression of both proteins was low. Higher levels of uPAR was detected on the cell surface of primary myeloma cells. When primary myeloma cells were gated by CD45 expression, stronger expression was found on immature CD45+ cells than on mature CD45-/dim cells. Finally, both myeloma cell lines and primary cells were able to cleave a uPA-specific substrate showing that the uPA system is functionally active. We conclude that myeloma cells are able to produce uPA and uPAR. This opens up a possible role of the uPA system in myeloma cell invasion and in the proteolytic digestion of bone matrix.

High levels of soluble syndecan-1 in myeloma-derived bone marrow: modulation of hepatocyte growth factor activity.

Syndecan-1 is a heparan sulfate proteoglycan expressed on the surface of, and actively shed by, myeloma cells. Hepatocyte growth factor (HGF) is a cytokine produced by myeloma cells. Previous studies have demonstrated elevated levels of syndecan-1 and HGF in the serum of patients with myeloma, both of negative prognostic value for the disease. Here we show that the median concentrations of syndecan-1 (900 ng/mL) and HGF (6 ng/mL) in the marrow compartment of patients with myeloma are highly elevated compared with healthy controls and controls with other diseases. We show that syndecan-1 isolated from the marrow of patients with myeloma seems to exist in an intact form, with glucosaminoglycan chains. Because HGF is a heparan-sulfate binding cytokine, we examined whether it interacted with soluble syndecan-1. In supernatants from myeloma cells in culture as well as in pleural effusions from patients with myeloma, HGF existed in a complex with soluble syndecan-1. Washing myeloma cells with purified soluble syndecan-1 could effectively displace HGF from the cell surface, suggesting that soluble syndecan-1 can act as a carrier for HGF in vivo. Finally, using a sensitive HGF bioassay (interleukin-11 production from the osteosarcoma cell line Saos-2) and intact syndecan-1 isolated from the U-266 myeloma cell line, we found that the presence of high concentrations of syndecan-1 (more than 3 microg/mL) inhibited the HGF effect, whereas lower concentrations potentiated it. HGF is only one of several heparin-binding cytokines associated with myeloma. These data indicate that soluble syndecan-1 may participate in the pathology of myeloma by modulating cytokine activity within the bone marrow.

Serum osteoprotegerin levels are reduced in patients with multiple myeloma with lytic bone disease.

Osteoprotegerin (OPG), the neutralizing decoy receptor for the osteoclast activator RANK ligand, was measured in serum taken from patients with multiple myeloma at the time of diagnosis. Median OPG was lower in the patients with myeloma (7.4 ng/mL; range, 2.6-80; n = 225) than in healthy age- and sex-matched controls (9.0 ng/mL; range 5.1-130; n = 40; P =.02). Importantly, OPG levels were associated with degree of radiographically assessed skeletal destruction (P =.01). The median OPG level in patients lacking osteolytic lesions was 9.1 ng/mL, as compared with 7.6 ng/mL and 7.0 ng/mL, respectively, in patients with minor or advanced osteolytic disease. Furthermore, OPG levels were associated with World Health Organization performance status (P =.003) and correlated to serum levels of carboxy-terminal propeptide of type I procollagen (PICP; P <.001) but not with clinical stage or survival. These findings suggest impaired OPG function in myeloma and give a rationale for OPG as a therapeutic agent against myeloma bone disease.