Abstract
Bone is a common site for cancer metastasis. To create space for their growth, cancer cells stimulate bone resorbing osteoclasts. Cytokine RANKL is a key osteoclast activator, while osteoprotegerin (OPG) is a RANKL decoy receptor and an inhibitor of osteoclastogenesis. Consistently, systemic application of OPG decreases metastatic tumor burden in bone. However, OPG produced locally by cancer cells was shown to enhance osteolysis and tumor growth. We propose that OPG produced by cancer cells causes a local reduction in RANKL levels, inducing a steeper RANKL gradient away from the tumor and towards the bone tissue, resulting in faster resorption and tumor expansion. We tested this hypothesis using a mathematical model of nonlinear partial differential equations describing the spatial dynamics of OPG, RANKL, PTHrP, osteoclasts, tumor and bone mass. We demonstrate that at lower expression rates, tumor-derived OPG enhances the chemotactic RANKL gradient and osteolysis, whereas at higher expression rates OPG broadly inhibits RANKL and decreases osteolysis and tumor burden. Moreover, tumor expression of a soluble mediator inducing RANKL in the host tissue, such as PTHrP, is important for correct orientation of the RANKL gradient. A meta-analysis of OPG, RANKL and PTHrP expression in normal prostate, carcinoma and metastatic tissues demonstrated an increase in expression of OPG, but not RANKL, in metastatic prostate cancer, and positive correlation between OPG and PTHrP in metastatic prostate cancer. The proposed mechanism highlights the importance of the spatial distribution of receptors, decoys and ligands, and can be applied to other systems involving regulation of spatially anisotropic processes.
Highlights
Primary cancers develop metastatic tumors in distant sites and tissues of the body, and frequently, fatal outcome is due to those secondary rather than the primary tumors [1]
The distinction of the two regimes for tumor-derived osteoprotegerin provides a potential explanation of the differential experimental outcomes in [9] and [12]. To test this hypothesis we developed a mathematical model of tumor-osteoclast interactions, including the cytokine fields of RANK ligand (RANKL), OPG and parathyroid hormone-related protein (PTHrP), and examined the model predictions by means of appropriate in silico experiments focusing on the following main questions: 1) How does the impact of systemic OPG compare to the impact of cancer–cell derived OPG production? 2) How is indirect stimulation of RANKL production via PTHrP different from direct production of RANKL by tumor cells?
We introduce the PTHrP concentration wP as a new state-variable: once produced by cancer cells, PTHrP diffuses across the tissue and is degraded by proteases, While diffusing across the tissue, PTHrP induces the expression of RANKL by osteoblastic cells in the bone tissue, and we describe this by adding a source term to the RANKL equation (3)
Summary
Primary cancers develop metastatic tumors in distant sites and tissues of the body, and frequently, fatal outcome is due to those secondary rather than the primary tumors [1]. Bone is remodeled through the process where old or damaged tissue is resorbed by cells specialized in bone destruction, osteoclasts, and new bone is produced by specialized bone-forming osteoblasts [3,4,5]. Once osteoclasts have removed the old tissue, they move forward and recruit osteblasts, which in turn fill the previously resorbed trench with osteoid. The latter eventually mineralizes, and the process of mass-neutral bone renewal is complete. The mostly membrane-bound RANKL binds to its receptor RANK, expressed on osteoclasts and their precursors, inducing osteoclast differentiation and stimulating resorptive activity. The resulting osteolysis provides in turn more space for the growing tumor – thereby closing the so-called ‘vicious cycle’ of bone resorption and tumor growth (Figure 1)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have