Prostate cancer cells metastasize to the hematopoietic stem cell niche in bone.

T he majority of men with advanced prostate cancer develop bone metastases as opposed to metastases at other sites. It has been unclear why prostate cancer selectively metastasizes to and proliferates in bone. Recently, Shiozawa et al. delineated a mechanism that may account for the establishment of prostate cancer in bone. Specifically, they identified that prostate cancer cells compete with hematopoietic stem cells (HSC) for the osteoblast in the HSC niche of the bone. Defining the mechanisms through which prostate cancer cells establish themselves in bone is critical towards developing effective therapeutic strategies to prevent or target bone metastases. HSCs localize to the bone marrow microenvironment where multiple cells may contribute to the HSC niche. However, the osteoblast has risen as a key cellular component of the HSC niche. Several studies have provided strong support to the importance of the osteoblast in maintaining the HSC population including the observations that decreasing osteoblasts results in contraction of the HSC population, whereas increasing osteoblasts results in expansion of the HSC population. Osteoblasts regulate the HSC through several mechanisms. These include promoting chemotaxis to the HSC niche through production of CXCL12 (also known as stromal-derived factor-1), serving as a docking site for HSC in the endocortical bone microenvironment through production of vascular adhesion molecule-1 and annexin II, and signaling the HSC to become quiescent through CXCL12. The concept that cancer cells can co-opt normal cellular activities has been previously demonstrated in the context of bone metastases. For example, the importance of CXCL12-mediated chemotaxis to drive cancer cells to bone, similar to the function on HSC to enter the bone marrow microenvironment, has been identified in multiple tumor types including breast cancer and prostate cancer. However, once in the bone marrow microenvironment, it was unknown where the tumor cells localize. Through several elegant experiments, Shiozawa et al. have now convincingly demonstrated that prostate cancer cells and HSCs occupy the same endosteal niche and furthermore that they compete with each other for niche occupancy on the osteoblast. Initially, they demonstrated that subcutaneous pre-implantation of mice with prostate cancer cells, but not non-metastatic transformed prostate epithelial cells, diminished the ability of HSC to engraft in an in vivo model of bone marrow transplant. This initial experiment provided the first indication that the prostate cancer cells compete with HSC for the HSC niche. They followed these experiments with direct evaluation of the prostate cancer cells’ ability to compete with HSC for binding to osteoblasts both in vitro and in vivo. To perform this, they injected prostate cancer cells, which were irradiated to inhibit their proliferation, directly into the blood stream, and determined that they were able to inhibit HSC engraftment and marrow recovery in irradiated mice. Furthermore, using confocal and multiphoton microscopy, they demonstrated that HSC and prostate cancer cells colocalized at the osteoblasts in identical regions of the bone marrow. To provide further evidence of the importance of the osteoblast niche on development of bone metastases, they manipulated the niche volume. In these studies, they found that increasing the HSC niche through administration of a bone anabolic regimen of parathyroid hormone increased the metastatic burden, whereas decreasing the HSC niche through ablation of osteoblasts resulted in decreased metastatic burden. Additionally to ascertain the similarity in the niche occupation between HSC and prostate cancer cells, they explored if prostate cancer cells can be mobilized from the niche similar to HSC. They found two different factors that mobilize HSC from marrow: AMD3100, which blocks binding of CXCL12 to its receptor and granulocyte-colony-stimulating factor, both inducing an increase of circulating prostate cancer cells. Finally, the authors wanted to provide some clinical evidence for the competition between prostate cancer cells and HSC for the niche. Consistent with their concept of competition for the HSC niche, they found that men with metastatic disease had more circulating hematopoietic progenitor cells (derived from HSC) than men with only local prostate cancer. These findings provide a new mechanism to account for the ability of prostate cancer cells to target and become established in bone. Importantly, they also suggest several significant clinical ramifications. A frustrating aspect of prostate cancer is that it is possible to subject a patient to prostatectomy with an apparent cure, but they develop bone metastases many years later. In context of the results presented by Shiozawa et al., it is plausible to consider that prostate cancer cells are maintained in a quiescent state, similar to HSC, when they are in the HSC niche. This dormant state may allow prostate cancer cells to remain in the bone microenvironment for years until some event activates them or releases them from the niche. Another challenging aspect of prostate cancer bone metastases is that they appear protected from chemotherapeutics due to being sequestered in the bone microenvironment. The osteoblast-induced dormancy that occurs in the bone microenvironment could contribute to chemoresistance as most chemotherapeutic Departments of Urology and Pathology, University of Michigan, Ann Arbor, MI 48109, USA Correspondence: Dr ET Keller ([email protected]) Asian Journal of Andrology (2011) 13, 622–623 2011 AJA, SIMM & SJTU. All rights reserved 1008-682X/11 $32.00