Bone marrow–derived CMPs and GMPs represent highly functional proangiogenic cells: implications for ischemic cardiovascular disease

Therapeutic angiogenesis has emerged as an innovative strategy for the treatment of patients with cardiovascular ischemic disease states, such as myocardial ischemia or peripheral artery disease. Several clinical trials using autologous, bone marrow (BM)–derived mononuclear cells have demonstrated only modest improvements of function and/or perfusion of ischemic myocardium or limb in patients.^1-6  However, the identities of these BM-derived mononuclear cells and their functional ability to promote neovascularization remain poorly understood.

Accumulating evidence has indicated that proangiogenic cells (PACs) may originate from the BM and are capable of being recruited to sites of ischemic injury, where they contribute to neovascularization and tissue repair through paracrine/autocrine mechanisms or by direct incorporation into the vessel wall.^7-10  Indeed, PAC function and/or number have been found to be markedly impaired in patients with coronary or peripheral artery disease.^11-17  Herein, we provide evidence that a hierarchy exists among mouse hematopoietic stem cell (HSC) progenitors in their ability to differentiate into PACs and promote neovascularization. Complementary functional studies revealed that common myeloid progenitors (CMPs) and granulocyte-macrophage progenitors (GMPs) preferentially differentiate into PACs as opposed to megakaryocyte-erythrocyte progenitors (MEPs), HSCs, and common lymphoid progenitors. Moreover, only PACs derived from BM-derived CMPs and GMPs were able to enhance neovascularization in vivo. Taken together, these findings reveal a novel and unexpected hierarchy among BM-derived progenitors in their ability to promote neovascularization.

Progenitor cells were isolated from C57BL/6 mice (n = 8-10) BM by using multicolor FACSAria (BD Biosciences) as previously described.^18,19  Briefly, approximately 2 to 5 × 10⁵ BM-derived cells were prospectively isolated and purified as follows: HSCs (Lin⁻Sca1⁺c-kit⁺), CMPs (Lin⁻Sca1⁻c-kit⁺CD34⁺FcγRII/II^lo), GMPs (Lin⁻Sca1⁻c-kit⁺CD34⁺FcγRII/III^hi), and MEPs (Lin⁻Sca1⁻c-kit⁺CD34⁻FcγRII/III^lo; supplemental Figure 1A, available on the Blood Web site; see the Supplemental Materials link at the top of the online article). Subsequently, cells were incubated in endothelial growth media-2 (EGM-2; Lonza) for 8 to 10 days and subjected to flow cytometry to detect expression of vascular endothelial growth factor receptor-2 (VEGFR2), CD31, CD133, CD34, CD45, Tie2, von Willebrand factor, endothelial nitric oxide synthase, and VE-cadherin (eBioscience). Cells were also incubated in the presence of 10 μg/mL for 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl indocarbocyanine perchlorate–acetylated low density lipoprotein (DiI-AcLDL; Biomedical Technologies) or FITC-labeled Ulex lectin (Vector Laboratories).

Matrigel was added to 96-well culture plates, and EGM-2–cultured progenitors were admixed with human umbilical vein endothelial cells to each well at a 1:1 ratio. Six hours later, tube-like network formation was assessed.

Matrigel plugs admixed with 0.8 mL of Matrigel, basic fibroblast growth factor (250 ng/mL; R&D Systems), and heparin (60 units/mL), and BM-derived progenitors grown in either EGM-2 or IMDM for 7 days were implanted subcutaneously into 10-week-old C57BL/6 mice. Eight days later, explanted plugs were assessed for angiogenesis by probing paraffin sections with CD31 antibodies.

Ischemic injury was produced by unilateral right femoral artery ligation in C57BL/6 mice. The deep femoral artery was cauterized, and the complete severance of the main femoral artery was performed. Immediately after surgery, mice were imaged on a Laser Doppler Imager-2. PACs were also injected intramuscularly into the right quadriceps and extent of blood flow recovery evaluated over a 14 day period.

Values are expressed as mean ± SD. Differences between values were examined using 2-tailed Student t test and were considered significant at P < .05.

Given the fact that BM-derived mononuclear cells can promote neovascularization,^4,20,21  we hypothesized that specific subsets within this population may possess distinctly high PAC differentiation potential. To assess the PAC differentiation potentials among hematopoietic stem cells or progenitors, isolated HSCs, CMPs, GMPs, MEPs, and common lymphoid progenitors were exposed to EGM-2 medium for 8 to 10 days and evaluated by flow cytometry to determine induced expression of VEGFR2 as a hallmark PAC marker. As shown in Figure 1A, CMPs and GMPs expressed approximately 4- and 6-fold higher VEGFR2 levels, respectively, than MEPs. In contrast, common lymphoid progenitors did not survive (data not shown). A similar pattern was observed among human BM-derived progenitors and human umbilical cord-derived progenitors (supplemental Figure 1B). One of the characteristics of PACs is their ability to uptake AcLDL and Ulex lectin.^4,20,21  As shown in Figure 1B, PACs derived from CMPs and GMPs exhibited markedly higher uptake of DiI-AcLDL and Ulex lectin than the MEP- or HSC-derived PACs. Compared with MEP- and HSC-derived PACs, CMP- and GMP-derived PACs also expressed higher levels of the markers CD31, CD133, CD34, CD45, Tie2, von Willebrand factor, endothelial nitric oxide synthase, and VE-cadherin (Figure 1C; supplemental Figure 1C) and the chemokine receptors CXCR4, CXCR3, CCR2, and CCR7 (supplemental Figure 2A) by flow cytometry. Finally, cell growth curves revealed that CMP- and GMP-derived PACs grew up to 2-fold faster than MEP- and HSC-derived PACs over 7 days (supplemental Figure 2B-C). Collectively, these observations indicate that CMPs and GMPs have greater potential to differentiate into PACs.

PACs facilitate angiogenesis, in part, by their abilities to migrate, adhere, and incorporate into vascular networks, where they promote neovascularization by secreting angiogenic cytokines.^4,20,21  As shown in Figure 1D-E, CMP- and GMP-derived PACs adhered to fibronectin-coated plates more efficiently and migrated faster than the MEP- or HSC-derived PACs or the CMPs and GMPs that had been grown in hematopoietic growth medium. Using the Matrigel network formation assay in vitro, BM-derived PACs were added to human umbilical vein endothelial cells; CMP- and GMP-derived PACs produced approximately 2-fold more network projections per microscopic field and incorporated into network structures at approximately 2-fold higher rate than the MEP- or HSC-derived PACs (Figure 1F-G). Finally, CMP- and GMP-derived PACs elaborated higher levels of the angiogenic growth factor VEGF compared with MEP- or HSC-derived PACs, whereas there were no differences detected in IL-1β levels (Figure 1H). Collectively, these in vitro data indicate that CMP- and GMP-derived PACs possess more favorable angiogenic properties than the MEP- and HSC-derived PACs, and even outperform CMPs and GMPs grown in hematopoietic medium alone.

One of the central functional properties of both BM-derived mononuclear cells and PACs is their ability to promote neovascularization within complex living systems. To assess this in mice, we performed subcutaneous implantation of Matrigel with BM-derived PAC subsets grown in EGM-2 or hematopoietic progenitors grown in IMDM.²²  Explanted plugs containing CMP- and GMP-derived PACs displayed up to approximately 6-fold higher CD31 staining and up to approximately 17-fold higher isolectin staining compared with plugs with MEP- and HSC-derived PACs (Figure 2A). Importantly, CMPs and GMPs grown in IMDM only minimally promoted neovascularization. To verify these findings in a pathophysiologic model of ischemic injury, PACs derived from CMPs, GMPs, MEPs, or HSCs were injected intramuscularly, immediately after femoral artery ligation in mice. As shown in Figure 2B, CMP- and GMP-derived PACs increased blood flow recovery to a greater extent than did the MEP- or HSC-derived PACs. Taken together, these findings indicate that the markedly enhanced angiogenic properties of CMP- and GMP-derived PACs are maintained in the in vivo setting.

Clinical trials applying various PAC subsets for therapeutic angiogenesis of myocardial or peripheral artery disease have thus far achieved modest functional effects,¹  or in some cases, the absence of any dose-response effect.^5,6  Because heterogeneity of the PACs used in this therapeutic strategy may be obscuring the identities and function of more potent subsets, identifying those cells with significantly robust angiogenic activity is necessary to the efficacy of PAC-based protocols in treating ischemic cardiovascular disease. This study demonstrated that CMP- and GMP-derived PACs possess these robust angiogenic properties and may provide an opportunity to improve cell-based therapies for ischemic cardiovascular disease.

This work was supported by the National Institutes of Health (HL080174 and HL091076, M.W.F.; and F32HL088819, A.K.W.), the American Cancer Society (Research Scholar Grant RSG0719501-LIB, M.W.F.), the Boston Area Diabetes Endocrinology Research Center (Pilot Grant P30DK057521, M.W.F.), and a Watkins Cardiovascular Medicine Discovery Award (M.W.F.). A.R. is a principal faculty member of the Harvard Stem Cell Institute. S.Y.F. was supported by the Clinical Investigators Training Program (Beth Israel Deaconess Medical Center–Harvard/MIT Health Sciences and Technology) in collaboration with Pfizer Inc and Merck & Co. This work was also supported in part by the Leducq Foundation Network of Research Excellence (A.R. and S.Y.F).

National Institutes of Health

Contribution: A.K.W. designed experiments, performed research, analyzed and interpreted data, and wrote the manuscript; K.C., S.Y.F., X.S., B.I., and F.E. performed research and analyzed the data; Y.T. performed research; A.R. interpreted the data and edited the manuscript; and M.W.F. designed experiments, performed research, interpreted the data, and wrote the manuscript.

Conflict-of-interest disclosure: M.W.F. and the Brigham and Women's Hospital have a patent pending related to the work that is described in the present study. The remaining authors declare no competing financial interests.

Correspondence: Mark W. Feinberg, Department of Medicine, Cardiovascular Division, Brigham and Women's Hospital, Harvard Medical School, 77 Ave Louis Pasteur, NRB-742F, Boston, MA 02115; e-mail: mfeinberg@rics.bwh.harvard.edu.