Expression of macrophage inhibitory cytokine-1 in prostate cancer bone metastases induces osteoclast activation and weight loss.

BACKGROUND: Macrophage inhibitory cytokine-1 (MIC-1) belongs to the bone morphogenic protein/transforming growth factor-beta (BMP/TGF-beta) superfamily. Serum MIC-1 concentrations are elevated in patients with advanced prostate cancer. The effects of MIC-1 on prostate cancer bone metastases are unknown. METHODS: In vitro effects of MIC-1 on osteoblast differentiation and activity were analyzed with alkaline phosphatase and mineralization assays; osteoclast numbers were counted microscopically. MIC-1 effects on TLR9 expression were studied with Western blotting. Human Du-145 prostate cancer cells were stably transfected with a cDNA encoding for mature MIC-1 or with an empty vector. The in vivo growth characteristics of the characterized cells were studied with the intra-tibial model of bone metastasis. Tumor associated bone changes were viewed with X-rays, histology, and histomorphometry. Bone formation was assayed by measuring serum PINP. RESULTS: MIC-1 induced osteoblast differentiation and activity and osteoclast formation in vitro. These effects were independent of TLR9 expression, which was promoted by MIC-1. Both MIC-1 and control tumors induced mixed sclerotic/lytic bone lesions, but MIC-1 increased the osteolytic component of tumors. Osteoclast formation at the tumor-bone interface was significantly higher in the MIC-1 tumors, whereas bone formation was significantly higher in the control mice. At sacrifice, the mice bearing MIC-1 tumors were significantly lighter with significantly smaller tumors. CONCLUSIONS: MIC-1 up-regulates TLR9 expression in various cells. MIC-1 stimulates both osteoblast and osteoclast differentiation in vitro, independently of TLR9. MIC-1 over-expressing prostate cancer cells that grow in bone induce osteoclast formation and cachexia.

Toll-like receptor 9 mediates CpG oligonucleotide-induced cellular invasion.

Toll-like receptor 9 (TLR9) belongs to the innate immune system and recognizes microbial and vertebrate DNA. We showed previously that treatment with the TLR9-agonistic ODN M362 (a CpG sequence containing oligonucleotide) induces matrix metalloproteinase-13-mediated invasion in TLR9-expressing human cancer cell lines. Here, we further characterized the role of the TLR9 pathway in this process. We show that CpG oligonucleotides induce invasion in macrophages from wild-type C57/B6 and MyD88 knockout mice and in human MDA-MB-231 breast cancer cells lacking MyD88 expression. This effect was significantly inhibited in macrophages from TLR9 knockout mice and in human MDA-MB-231 breast cancer cells stably expressing TLR9 small interfering RNA or dominant-negative tumor necrosis factor receptor-associated factor 6 (TRAF6). Sequence modifications to the CpG oligonucleotides that targeted the stem loop and other secondary structures were shown to influence the invasion-inducing effect in MDA-MB-231 cells. In contrast, methylation of the cytosine residues of the parent CpG oligonucleotide did not affect the TLR9-mediated invasion compared with the unmethylated parent CpG oligonucleotide. Finally, expression of TLR9 was studied in clinical breast cancer samples and normal breast epithelium with immunohistochemistry. TLR9 staining localized in epithelial cells in both cancer and normal samples. The mean TLR9 staining intensity was significantly increased in the breast cancer cells compared with normal breast epithelial cells. In conclusion, our results suggest that TLR9 expression is increased in breast cancer and CpG oligonucleotide-induced cellular invasion is mediated via TLR9 and TRAF6, independent of MyD88. Further, our findings suggest that the structure and/or stability of DNA may influence the induction of TLR9-mediated invasion in breast cancer.

Inhibition of the mevalonate pathway and activation of p38 MAP kinase are independently regulated by nitrogen-containing bisphosphonates in breast cancer cells.

Bisphosphonates are widely used inhibitors of bone resorption. They also inhibit the growth of various cancer cells in vitro, but the clinical significance of this effect is unclear. The cancer growth inhibitory effects of nitrogen-containing bisphosphonates, (i.e. zoledronate) have been attributed to their ability to inhibit the mevalonate pathway. We have shown that bisphosphonates also induce p38 activation, which signals resistance against the drug-induced growth inhibition through an unknown mechanism. We show here that zoledronate induces a G1/S cell cycle arrest in human MDA-MB-231 breast cancer cells. Furthermore, p38 inhibitor augments bisphosphonate-induced growth inhibition by inducing an additional G2-phase cell cycle arrest. We also show that the nitrogen-containing bisphosphonate-induced effects on p38 phosphorylation occur before accumulation of unprenylated Rap1A or Rac1 activation. Geranylgeranyl pyrophosphate, an end-product of the mevalonate pathway, reversed the accumulation of unprenylated Rap1A but not phosphorylation of p38. Geranylgeranyl pyrophosphate also reversed n-BP induced growth inhibition, but the completeness of this reversal was nitrogen-containing bisphosphonate concentration dependent. Also mevastatin induced the accumulation of unprenylated Rap1A, but it did not induce p38 phosphorylation. In conclusion, our results suggest that in addition to the previously reported effects on apoptosis, nitrogen-containing bisphosphonates also inhibit the growth of MDA-MB-231 breast cancer cells by inducing G1/S cell cycle arrest. The bisphosphonate-induced p38 activation signals for resistance against these drugs, by promoting progression through the G2/M-checkpoint. Of these pathways only growth inhibition is mediated via inhibition of the mevalonate pathway in MDA-MB-231 cells. Combining p38 inhibitors with bisphosphonates may result in increased anti-cancer efficacy.

Differential effects of Ca(2+) on bisphosphonate-induced growth inhibition in breast cancer and mesothelioma cells.

Bisphosphonates are widely clinically used inhibitors of bone resorption. Pre-clinical studies indicate that bisphosphonates also inhibit the growth of various cancer cells in vitro, but their in vivo anti-cancer activity varies greatly, depending on the tumor type. We compared the various cellular effects of bisphosphonates in breast cancer and mesothelioma cells, with differences in growth inhibition responses to bisphosphonate-treatment in vivo. We show that the growth inhibitory effects of nitrogen-containing bisphosphonates are significantly affected by excess Ca(2+) in a cell- and bisphosphonate-specific fashion. Furthermore, excess pyrophosphate-resembling bisphosphonates prevent nitrogen-containing-bisphosphonate-induced accumulation of unprenylated Rap1A, p38 phosphorylation and growth inhibition in human MDA-MB-231 breast cancer and mouse AB-12 mesothelioma cells. For some, but not all tested, pyrophosphate-resembling bisphosphonate: nitrogen-containing bisphosphonate combinations these results may be partially explained by the ability of the excess pyrophosphate-resembling bisphosphonates to chelate Ca(2+). In mice, subcutaneous AB-12 and MDA-MB-231 tumors exhibit positive staining for Ca(2+) minerals, as revealed with Von Kossa stainings. We further show that the AB-12 tumors accumulate significantly more of the bone scanning bisphosphonate, Tc99m-medronate, as compared with MDA-MB-231 tumors. In conclusion, our results suggest that Ca(2+) regulates the growth inhibitory effects of bisphosphonates in a target cell and drug-specific fashion. These findings may be of physiological relevance since many tumor types are calcified. They further suggest that bisphosphonates can accumulate in tumors that are growing at the visceral sites and that differences in tumor accumulation of bisphosphonates may regulate their in vivo sensitivity to these drugs.

Bisphosphonates inhibit the growth of mesothelioma cells in vitro and in vivo.

PURPOSE: Bisphosphonates (such as risedronate and zoledronate) are widely used inhibitors of bone resorption. Despite their in vitro antiproliferative effects in various cancer cells, bisphosphonates have not exhibited significant antitumor efficacy in animal models of visceral cancer, which may be due to their poor bioavailability. The diagnostic use of radioactive bisphosphonates has revealed the accumulation of bisphosphonates in mesothelioma, which prompted us to test the antitumor efficacy of bisphosphonates in this disease. EXPERIMENTAL DESIGN AND RESULTS: Treatment with either risedronate or zoledronate (2 x 10(-4) to 2 x 10(-6) mol/L) inhibited the growth of AB12 and AC29 mouse mesothelioma cells and induced the accumulation of unprenylated Rap1A in these cells. Both these in vitro effects were reversed by geranygeraniol, an end product of the mevalonate pathway that these bisphosphonates inhibit. Both bisphosphonates also induced the phosphorylation of the p38 mitogen-activated protein kinase in AB12 and AC29 cells. The inhibition of p38 augmented bisphosphonate-induced growth inhibition in these cells. Bisphosphonate-induced p38 phosphorylation was not reversible by geranylgeraniol. Risedronate (15 mg/kg) and zoledronate (0.5 mg/kg) inhibited the growth of s.c. tumors and increased the median survival of mice with i.p. mesothelioma tumors in vivo. DISCUSSION: In conclusion, risedronate and zoledronate inhibit the mevalonate pathway and induce p38 activation in mesothelioma cells in vitro. The effects on the mevalonate pathway dominate because the net result is growth inhibition. Both bisphosphonates also inhibit mesothelioma tumor growth in vivo and prolong the survival of mesothelioma-bearing mice. These results support further study of bisphosphonates in the management of mesothelioma.