Journal Facts

Journal
BoneKEy Knowledge Environment
ISSN
1533-4368
Volume
Year
2002

Article Facts

Title
How metastases home to bone: The attraction of chemokines
DOI
10.1138/2002052
Author

Online Date
07/09/2002

Article Links

BoneKEy-Osteovision | Commentary

How metastases home to bone: The attraction of chemokines


DOI:10.1138/2002052

Commentary on: Taichman RS, Cooper C, Keller ET, Pienta KJ, Taichman NS, McCauley LK. Use of the stromal cell-derived factor-1/CXCR4 pathway in prostate cancer metastasis to bone. Cancer Res. 2002 Mar 15;62(6):1832-7.

Metastasis is a highly organized, non-random multi-step process that results in tumor growth at distant sites. Certain solid tumors have a propensity to metastasize to specific organs; the pattern of which cannot be explained by any theory of embolism alone. Such is the case of the proclivity of breast and prostate cancer to metastasize to bone, as noted in the autopsy series of 1889 by Paget. He proposed the "seed and soil" hypothesis to explain the phenomenon (). That is, tumor cells or seeds are randomly distributed, but only grow in sites that produce the appropriate growth factors. More recently, two other theories to explain site-specific metastases have been put forth (). The first implicates "vascular zip codes": vascular beds from specific organs express adhesion molecules to which circulating tumor cells attach. Tumor cells express peptide zip codes that direct them to an organ- or tumor-specific vascular bed (). The last, or chemoattraction theory, proposes that organ-specific attractant molecules cause tumor cells to home to specific sites (). Indeed, the complexity of science dictates and evidence to date indicate that all three processes are involved. Most recently, Taichman et al. () and Müller et al. () provide tantalizing evidence for the chemoattraction theory. Stromal cell-derived factor-1 (SDF-1 or CXCL12) and its receptor, CXCR4, appear to be central mediators of site-specific metastases.

Of note, the above three theories of site-specific metastasis have also been implicated in the specific homing of hematopoietic stem cells to the bone marrow, a normal process that shares many similarities with solid tumor metastasis to bone (). Chemokines expressed in bone are responsible for stem cells’ setting up residence in the bone marrow. These small proteins induce cytoskeletal rearrangement, adhesion to endothelial cells, and directional migration in leukocytes via G-protein coupled receptors (). The responsible chemokine, SDF-1, is expressed by osteoblasts and bone marrow endothelial cells and binds to CXCR4 on human hematopoietic progenitor cells which then engraft in bone marrow. In SDF-1 or CXCR4 null mice, this engraftment does not occur (). CXCR4 is also a co-factor required for supporting T-lymphocyte tropic HIV infection into permissive cells. CXCR4 and its ligand SDF-1 play additional important roles in hematopoiesis, organogenesis, vascularization and embryogenesis ().

So how do SDF-1 and CXCR4 contribute to bone metastasis? Taichman et al. () present convincing in vitro data that implicate this ligand/receptor complex in prostate cancer metastasis to bone. First, prostate cancer cell lines express CXCR4 and human osteoblasts express SDF-1 at the RNA and protein levels. The CXCR4 on the prostate cancer cell lines is functional: SDF-1 rapidly induced phosphorylation of ERK-1/ERK-2. Furthermore, SDF-1 1) enhanced the binding of prostate cancer cells to both endothelial cells and osteosarcoma cells; and 2) stimulated prostate cancer cells to migrate across bone marrow endothelial cells and to invade extracellular matrix. Finally, serum-free osteoblast-conditioned media promoted invasion of prostate cancer cells through extracellular matrix, and this was blocked by a neutralizing antibody to CXCR4. These data demonstrate a role for bone-derived SDF-1, via CXCR4 expressed on prostate cancer cells, in the metastatic processes of directional migration and invasion. Despite the exciting nature of these findings, one must use caution when extrapolating the data to the clinical situation of prostate cancer metastases to bone. Interpretation of these studies is limited by the fact that only prostate cancer cell lines, and not primary or metastatic prostate cancer, were studied. There was no comparison made between CXCR4 expression in primary prostate cancer tissue and normal prostate. Finally, the lack of in vivo studies using established bone metastases models limit the extrapolation of these in vitro findings into the clinic.

Nonetheless, one can appreciate the implications of targeting the SDF-1/CXCR4 pathway to prevent bone or other metastases due to prostate cancer. In studies that complement and extend the prostate studies of Taichman et al., Müller et al. () proposed that metastatic cancer cells co-opt signals that normally control leukocyte transport. Upon screening human breast cancer lines for 17 chemokine receptor genes, they demonstrated that only CXCR4 and CXCR7 were overexpressed. CXCR4 was also overexpressed in primary breast cancer and metastases, but not normal mammary epithelia or stroma. Further, when they screened a panel of normal organs for the CXCR4 ligand, SDF-1, they found that it was only overexpressed in organs which represent the most common sites of breast cancer metastases: lymph node, lung, liver and bone marrow. Similar to the findings of Taichman et al., breast cancer cells expressed functional CXCR4. Specifically, SDF-1 induced a transient increase in intracellular filamentous actin (F-actin), intense F-actin staining in the periphery of cells and a redistribution of F-actin toward the leading edge. In adherent cells, this was associated with distinct pseudopodia formation. Likewise, in breast cancer cell lines as well as primary breast cancer cells, SDF-1 induced directional migration of breast cancer cells and directional invasion through extracellular matrix. Migration of the breast cancer cells in response to conditioned medium from human bone marrow or lymph node was inhibited by CXCR4 blockade. In contrast, protein extracts from organs that are rare targets of breast cancer metastasis, such as skin or muscle, showed weak chemoattractive properties for breast cancer cells, and this was not affected by CXCR4 blockade. Finally, in two different mouse models of breast cancer metastasis to lung, in which the CXCR4-positive MDA-MB-231 breast cancer cell line is inoculated either intravenously or into the mammary fat pad, a neutralizing antibody to CXCR4 significantly reduced lung metastases.

Taken together, these findings imply an important role for SDF-1 signaling through CXCR4 to promote metastasis by stimulating tumor cell adhesion, migration and invasion. However, the definitive study of SDF-1/CXCR4 in the development of bone metastases due to breast and prostate cancer has yet to be performed. Demonstration that CXCR4 blockade reduces the development of bone metastasis due to the MDA-MB-231 breast cancer or the PC-3 prostate cancer (both of which are CXCR4-positive) is a crucial and feasible experiment. These mouse models are established, and the Müller study confirms that CXCR4 blockade is effective in other mouse models of metastases.

The role of SDF-1/CXCR4 to promote metastasis is not unique to breast and prostate cancer, but may be integral in many tumor types which eventually house in bone. Renal cell carcinoma (), acute myeloid and lymphoblastic leukemia (), chronic lymphocytic leukemia () and non-Hodgkin B-cell lymphoma () all express functionally active chemokine receptors that mediate tumor cell migration in vitro. Furthermore, SDF-1 signaling via CXCR4 may induce other behavior besides adhesion, migration and invasion that promote tumor survival. For example, in chronic lymphocytic leukemia SDF-1 signaling through CXCR4 reduced tumor cell apoptosis (); this was not observed in the prostate cancer study of Taichman et al. Finally, although data from several studies implicate SDF-1/CXCR4 pathway in the pathogenesis of bone metastasis, it is clear that this pathway is not bone-specific. The Taichman data also imply that SDF-1/CXCR4 pathway may not be specific to bone metastases since the prostate cancer line DU145, derived from a brain metastases, also expressed CXCR4.

As such, these studies raise other questions about the role of SDF-1/CXCR4 in metastases. 1) Is SDF-1 the primary determinant of site-specific metastasis or just a facilitator? 2) In many cases of metastases due to breast and prostate cancer, bone is the only site of metastases. In this setting, if SDF-1/CXCR4 has a role in targeting tumor cells to bone, what prevents growth of metastasis at other target organs which produce large amounts of SDF-1, such as liver and lung? 3) Is SDF-1/CXCR4 only responsible for attraction of the tumor cells to bone or does it also play a role in the phenotype of respective metastases: osteolytic in breast cancer or osteoblastic in prostate cancer. Does SDF-1 signaling through CXCR4 differentially regulate the production of osteolytic and osteoblastic factors respectively by breast and prostate cancer? 4) With respect to generalized metastasis, which step in the metastatic cascade is affected by SDF-1/CXCR4? Does SDF-1/CXCR4 alter tumor expression of proteases that portend the invasive phenotype? Does it alter tumor cell expression of integrins, which mediate important interactions between the tumor cell and vasculature as well as the tumor cell and the bone microenvironment? 5) It has been proposed that patients with long-term nonprogression of AIDS have alterations in chemokines or chemokine receptors that prevent entry of HIV into T-cells. If such patients develop malignancy, are they less likely to develop metastases? 6) Will targeted therapy directed against CXCR4 be effective to prevent metastases? The short-term animal studies of Müller suggest this to be the case, but the lethality of the CXCR4 null state suggests that significant toxicity may be evident with long-term treatment.

The story of SDF-1/CXCR4 in metastases provides us with an example of a normal physiologic process, that of hematopoietic cells homing to bone marrow which, when dysregulated in cancer, is associated with skeletal complication of malignancy. Once again, the biology of skeletal development continues to provide us with the key to unlock the mysteries which underly the skeletal complications of malignancy. Although the evidence to date provides strong support for the chemoattraction theory of metastasis, the SDF-1/CXCR4 interaction alone is insufficient to explain all aspects of metastases. Thus, it seems likely that all three theories of metastases contribute to the complex, yet highly organized, process of cancer metastases. Future investigation is sure to delineate the specific contributions of each.

Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.