Multiple myeloma–derived MMP-13 mediates osteoclast fusogenesis and osteolytic disease
Jing Fu, … , Stephen J. Weiss, Suzanne Lentzsch
Jing Fu, … , Stephen J. Weiss, Suzanne Lentzsch
Published April 4, 2016
Citation Information: J Clin Invest. 2016;126(5):1759-1772. https://doi.org/10.1172/JCI80276.
View: Text | PDF
Research Article Oncology

Multiple myeloma–derived MMP-13 mediates osteoclast fusogenesis and osteolytic disease

Abstract

Multiple myeloma (MM) cells secrete osteoclastogenic factors that promote osteolytic lesions; however, the identity of these factors is largely unknown. Here, we performed a screen of human myeloma cells to identify pro-osteoclastogenic agents that could potentially serve as therapeutic targets for ameliorating MM-associated bone disease. We found that myeloma cells express high levels of the matrix metalloproteinase MMP-13 and determined that MMP-13 directly enhances osteoclast multinucleation and bone-resorptive activity by triggering upregulation of the cell fusogen DC-STAMP. Moreover, this effect was independent of the proteolytic activity of the enzyme. Further, in mouse xenograft models, silencing MMP-13 expression in myeloma cells inhibited the development of osteolytic lesions. In patient cohorts, MMP-13 expression was localized to BM-associated myeloma cells, while elevated MMP-13 serum levels were able to correctly predict the presence of active bone disease. Together, these data demonstrate that MMP-13 is critical for the development of osteolytic lesions in MM and that targeting the MMP-13 protein — rather than its catalytic activity — constitutes a potential approach to mitigating bone disease in affected patients.

Authors

Jing Fu, Shirong Li, Rentian Feng, Huihui Ma, Farideh Sabeh, G. David Roodman, Ji Wang, Samuel Robinson, X. Edward Guo, Thomas Lund, Daniel Normolle, Markus Y. Mapara, Stephen J. Weiss, Suzanne Lentzsch

×

Figure 4

MMP-13 proteolytic activity is not required for OCL induction.

Options: View larger image (or click on image) Download as PowerPoint
MMP-13 proteolytic activity is not required for OCL induction. (A) SDS-P...
(A) SDS-PAGE analysis of purified WT pro–MMP-13 and E223A pro–MMP-13. M, molecular weight marker. (B) Proteolytic activity of purified MMP-13 WT and E223A proteins was analyzed by fluorometric assay and compared with standard MMP-13 (mean ± SD; n = 3). Data are representative of 3 independent experiments. (C) For OCL induction, mouse nonadherent BM cells were cultured with WT pro–MMP-13 or the pro–MMP-13 E223A mutant in the presence of M-CSF (10 ng/ml) and RANKL (50 ng/ml) for 4 days. OCLs were stained for TRAP. Scale bar: 100 μm. Data are representative of 5 independent experiments. (D) OCL size, number of nuclei per OCL, and number of OCLs per field in C were determined as described above. Data represent the mean ± SEM (n = 3). **P ≤ 0.01, by ANOVA (left panel) and *P ≤ 0.05, by Poisson ANOVA (middle panel). (E and F) Mouse nonadherent BM cells were seeded atop dentin slices and cultured as described in C for 9 days. Bone resorption lacunae were visualized by hematoxylin staining. Bone resorption area was calculated by ImageJ software. Data represent the mean ± SEM (n = 3) and are representative of 3 independent experiments. **P ≤ 0.01, by ANOVA (E). Scale bars: 100 μm (F).(G and H) Mouse nonadherent BM cells were cultured as described in C, with or without pro–MMP-13 WT or the pro–MMP-13 E223A mutant, in the presence of U0126 (5 μM) or vehicle control for 4 days. (G) DC-STAMP, NFATc1, p-ERK1/2, and ERK2 expression levels were determined in OCL lysates by Western blot analysis. Band intensities were quantified by ImageJ software, normalized to β-actin, and compared with control. Relative intensities are shown. (H) OCL differentiation analyzed by TRAP staining. Data are representative of 2 independent experiments. Scale bars: 100 μm.