The Hematopoietic Stem Cell Is a Source of Endothelial Cells in Cancer Vasculogenesis and Depends on SDF-1.

Abstract

Cancer growth and metastasis depend upon a rich supply of oxygen and nutrients from blood vessels. Previously we found that the hematopoietic stem cell (HSC) is capable of contributing to vasculogenesis in settings of physiologic repair, and given the relative hypoxia within tumor microenvironments, we subsequently hypothesized that the HSC also acts as a pathologic hemangioblast. First, we injected cancers (lung, pancreatic, melanoma, lymphoma) into cohorts of C57BL/6 mice which were previously transplanted green fluorescent protein (GFP)-tagged whole bone marrow. All cancer specimens demonstrated blood vessels with marrow-derived endothelial cells. Approximately 25% of tumor vessels contained marrow-derived endothelial cells as demonstrated by GFP, CD31 and vWF expression. We further questioned whether the tumor neovasculogenesis is from a clonal, self-renewing HSC. Single HSC transplanted mice were used as donors for secondary transplant mice in order to select for the true HSC. Lung cancers grown in recipients of single cell and serially transplanted hematopoietic stem cells (n=9) demonstrate clonal, donor-derived endothelial cells in 5% of tumor vasculature, matching hematopoietic engraftment. Thus, our results indicate that the self-renewing, clonal adult HSC exhibits pathologic hemangioblast activity. We further hypothesized that factors that affect leukocyte trafficking likely affect the pathologic hemangioblast activity of HSC. To test this hypothesis, we made slight modifications to our transplant and tumor inoculation model by administering GCSF and SCF to mobilize marrow derived EPC. Over the ensuing 14 days, the tumors in the cytokine treated group grew at a much faster rate and to a much larger size than tumors in the control mice. After 14 days of cytokine treatment and tumor growth, microvessel density was not different between cytokine treated mice (n=4) and control mice (n=4); however, marrow-derived tumor vasculogenesis was markedly elevated in the cytokine treated compared to controls (63% vs. 26%). Given that the SDF1/CXCR4 axis is pivotal for marrow cell homing and migration, we hypothesized that blocking this axis would block marrow-derived blood vessels in cancer. To test this hypothesis, we transplanted green fluorescent protein (GFP) marrow into wild-type C57BL/6 mice and then inoculated these mice with lung cancer. An experimental cohort of mice (n=4) received intra-tumoral anti-SDF1. A control cohort included mice receiving intra-tumoral PBS (n=4). Over the ensuing 14 days, tumors in the anti-SDF1 treated group grew at a much slower rate and to a much smaller size, if at all. After 14 days of injections and tumor growth, microvessel density was markedly decreased in the anti-SDF1 cohort compared to the control cohort. Moreover, marrow-derived tumor vasculogenesis was decreased in the anti-SDF1 treated tumors compared to controls (18% of vessels with marrow-derived endothelial cells vs. 26%, respectively). Lung cancer cells grew normally in vitro in the presence of anti-SDF-1. In conclusion, the results of our studies indicate that the HSC contributes to blood vessels within a variety of cancers and that strategies targeting HSC and EPC mobilization and homing potentially represent excellent ant-neoplastic opportunities. Indeed, perturbing the SDF1/CXCR4 axis inhibits marrow-derived tumor vasculogenesis. These studies serve as the preclinical basis for anti-SDF-1 antibody therapy as an adjunct to anti-cancer therapy.