Macrophages and tumor angiogenesis

Angiogenesis plays a crucial role in the cancerogenesis, growth and progression of human solid and hematological tumors.^{1, 2} The passage from the preangiogenic phenotype to the angiogenic phenotype, referred to as the ‘angiogenic switch’, allows the formation of a neovasculature that is indispensable for tumor growth and metastatic dissemination.³

In 1889, the English surgeon Stephen Paget published his ‘seed and soil’ explanation of nonrandom pattern of metastasis, and was the first to suggest that interactions between tumor cells and host cells in the microenvironment are critical in regulating tumorigenesis.⁴ Certain favored tumor cells (the ‘seed’), he said, had a specific affinity for the growth-enhancing milieu within specific organs (the ‘soil’), and hence metastasis only occurred when the ‘seed’ and ‘soil’ were compatible.⁵ The importance of several components of the ‘soil’ in regulating tumor growth has since been emphasized: (1) the extracellular matrix; (2) stromal cells and their growth factors and inhibitors; (3) microvessels and angiogenic factors; and (4) inflammatory cells.

The tumor microenvironment is a complex system of many cell type, including endothelial cells and their precursors, pericytes, smooth-muscle cells, fibroblasts, neutrophils, eosinophils, basophils, mast cells, T, B and natural killer lymphocytes, and antigen-presenting cells, such as macrophages and dendritic cells, which communicate through a complex network of intercellular signaling pathways that are mediated by surface adhesion adhesion molecules, cytokines and their receptors.

Tumor angiogenesis result not only from the interaction of cancer cells with endothelial cells, but surrounding inflammatory cells have also a crucial role in directing the neoformation of blood vessels.

Macrophages are derived from CD34-positive bone marrow progenitors that continually proliferate and shed their progeny in the bloodstream as promonocytes. They then develop into monocyte and extravasate into tissues where they differentiate into a specific type of ‘resident’ tissue macrophage.⁶ Metchnikoff was the first person in 1893 to use the term ‘macrophage’ to describea large cell able to take up microorganisms.⁷ The phenotype of these fully differentiated, resident macrophages can vary markedly within tissues, from that of microglial cells in the brain, Kupffer cells in the liver, alveolar macrophages in the lung and Langerhans cells in the skin. Resident macrophages share a set of common functions, including their ability to intervene against microbial infections, to regulate normal cell turnover and tissue remodeling, and to help repair sites of injury.⁶

Almost any local disturbance of tissue normality, be it infection, normal cell turnover or wounding, immune response or malignancy, caused rapid recruitment of macrophages. Recruited macrophages exhibit many phenotypic differences from resident tissue macrophages. The generic term, ‘macrophages activation’ is commonly used to describe this process, but the nature of an ‘activated macrophage’ population depends upon both the nature of the recruiting stimulus and the location.

It is now well established that the functional domain of the macrophage extends far beyond its originally recognized role as a scavenger cell. Its rich array of secretory products, anatomic diversity and functional heterogeneity is unmatched by any other cell type. As a result of this remarkable versatility, the macrophage is able to influence every facet of the immune response and inflammation as well as playing a central role in the etiology and/or pathogenesis of a number of disease processes.

Monocyte differentiated into polarized macrophage subset when exposed to different cytokine milieu.⁸ In the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon gamma (IFN-γ), lipopolysaccharide and other microbial products, monocyte differentiate into M1 macrophage. In the presence of macrophage colony-stimulating factor (M-CSF), interleukin (IL)-4, IL-13, IL-10, immunosuppressive agents (corticosteroids, vitamin D₃, prostaglandins) monocytes differentiate into M2 macrophages, involved in tumor angiogenesis.

Firstly in 1863, Rudolf Virchow noticed the infiltration of leukocytes into malignant tissues and suggested that cancers arise at regions of chronic inflammation.⁹ Leukocytes that do reach the tumor often remain localized in the tumor periphery or stroma and are often not able to exert strong antitumor activity.¹⁰ Both mouse and human tumors produce tumor-derived chemotactic factors capable of stimulating monocyte migration.¹¹

Tumor-associated macrophages (TAMs) derived from circulating monocytes and are recruited at the tumor site by a tumor-derived chemotactic factor from monocytes, originally described by Bottazzi et al.,¹² and later identified as the chemokine CCL2/MCP-1.^{13, 14} When exposed to vascular endothelial growth factor (VEGF)¹⁵ or to brief ischemia,¹⁶ endothelial cells synthesize MCP-1 and the extent of MCP-1 expression in human cancers correlated with both TAM infiltration and tumor malignancy in human melanoma, in Kaposi sarcoma cell lines and in human tumor cell lines of epithelial origin such as breast, colon and ovary.¹⁷ Moreover, MCP-1 is angiogenic when implanted into the rabbit cornea, where it exerted a potency similar to VEGF.¹⁸ MCP-1 expression has been shown to correlate significantly with levels of VEGF, tumor necrosis factor-α (TNF-α) and IL-8.^{19, 20, 21}

The expression of CCL5/RANTES is elevated in breast tumor cells synergistically by IFN-γ and TNF-α, regulating monocyte migration into tumor sites and stimulate them to secrete matrix metalloproteinase (MMP)-9 and MMP-19.^{22, 23}

CSF and GM-CSF are commonly produced in a range of different tumor types and are chemotactic for macrophages in vitro.²⁴ Transplanted mouse tumors transfected with the GM-CSF gene exhibit increased TAM infiltration²⁵ and genetic deletion of CSF-1 in the PyMT mouse model of breast cancer significantly decreased TAM infiltration and attenuated tumor progression to metastasis.²⁶ Increased expression of GM-CSF has been found in human breast, endometrial and ovarian tumors²⁷ and high GM-CSF expression is associated with high TAM accumulation in breast carcinomas.²⁸ Also VEGF is chemotactic for monocytes via VEGFR-1.^{29, 30}

TAM expresses and releases epidermal growth factor (EGF), fibroblast growth factor-2 (FGF-2),^{31, 32} transforming growth factor-α and -β (TGF-α and -β),^{33, 34, 35} VEGF,³⁶ TNF-α³⁷ IL-1,³⁸ IL-6, IL-8,³⁹ platelet-activating factor⁴⁰ platelet-derived growth factor (PDGF)⁴¹ G-CSF and GM-CSF.⁴² thymidine phosphorylase⁴³ and chemokines, such as CCL2.⁴⁴ TAMs produce, besides, angiogenic factors, angiostatic molecules such as thrombospondin-1,⁴⁵ IL-12,⁴⁶ IL-18⁴⁷ and MMP-12.⁴⁸ TAMs can produce angiogenesis regulators and may also induce tissue remodeling by producing various proteinase activators and inhibitors that may destroy the integrity of the basement membrane and extracellular matrix, liberating matrix-bound factors, including MMP-2, MMP-9, MMP-12 and cyclooxygenase-2 (COX-2).^{49, 50, 51} TAM production of MMP-9 has been shown to be crucial for angiogenesis in a human papillomavirus-16-induced model of cervical carcinogenesis. In this model, inhibition of MMP-9 in macrophages blocked the release of VEGF and thereby inhibited angiogenesis and tumor growth.⁵² Inhibition of CSF-1 function in human tumors xenografted into immunocompromised mice reduced their growth and this was correlated with poor macrophage recruitment and reduced angiogenesis due to a depletion of VEGF.

TAMs accumulate in hypoxic regions of tumors and hypoxia triggers a proangiogenic program in these cells. TAM adaptation to hypoxia, which is achieved by the increased expression of hypoxia-inducible and proangiogenic genes, such as VEGF, FGF-2 and CXCL8, as well as glycolytic enzymes, whose transcriptions are controlled by the transcription factors hypoxia-inducible factors 1 and 2 (HIF-1 and -2).⁵³ Increased number of TAMs in hypoxic regions may promote tumor progression in part by stimulating levels of angiogenesis, such as occurring in breast carcinoma.⁵⁴

White et al.⁵⁵ used adenoviral infection to overexpress HIF-2 in human macrophages and found it to be the primary inducer of genes encoding angiogenic cytokines in these genes. Macrophages also upregulates VEGF and other proangiogenic factors in response to hypoxia. TAMs express VEGF almost exclusively in avascular and perinecrotic areas of human breast carcinoma.⁵⁶

Macrophages also synthesize increased levels of MMP-7 when exposed to hypoxia in vitro and in an avascular area of human tumors.⁵⁷ MMP-7 is known to stimulate endothelial cell proliferation and migration.⁵⁸

A complementary DNA array study has identified upregulation of messenger encoding >30 proangiogenic genes in primary macrophages exposed to hypoxia, including CXCL8, angiopoietin, COX-2 and other factors.⁵⁵

Bingle et al.⁵⁹ found that when macrophages are co-cultured in vitro with human tumor spheroids, they infiltrate deep into the central, hypoxic areas of these structures. The release of VEGF by macrophage-infiltrated spheroids was significantly higher than that seen for noninfiltrated spheroids. This increase translated into a significant stimulation of angiogenesis in vivo when implanted into microcirculation window chambers on the flags of nude mice for 3 days.⁵⁹

Evans^{60, 61} has shown that mice depleted of macrophages by whole-body X-irradiation or azathioprine administered before or after implantation of a syngeneic fibrosarcoma showed a delay in the appearance of tumors, and a marked reduction in tumor vascularization. Mostafa et al.^{62, 63} and Stenzinger et al.⁶⁴ showed that vascularization of several human tumor cell lines grown on the chorioallantoic membrane of the chick embryo or subcutaneously in nude mice occurred coincidentally with mononuclear cell infiltration at the tumor site.

Polverini and Leibovich⁶⁵ isolated macrophages from a transplantable rat fibrosarcoma and examined them and their serum-free conditioned media for angiogenic activity in rat corneas. Results showed that TAM and their conditioned media were potently angiogenic in vivo and stimulated proliferation of bovine aortic endothelial cells in culture. Moreover, when TAMs were combined with tumor cells at a concentration equivalent to the number of macrophages originally present in the tumor, there was a marked enhancement of tumor neovascularization and growth. Polverini and Leibovich⁶⁶ reported that hamsters bearing chemical carcinogen-induced squamous-cell carcinomas showed a marked reduction in the thymidine incorporation by endothelial cells and neovascularization of tumors when treated with low doses of steroids and anti-macrophage serum.

In the mouse model of breast cancer caused by the mammary epithelial cell restricted expression of the Polyoma middle T oncoprotein (PyMT mice) infiltration of TAM in primary tumors is positively associated with tumor progression to malignancy.²⁶ Depletion of macrophages in this model severely delayed tumor progression and reduced metastasis, whereas an increase in macrophage infiltration remarkably accelerated these processes. By using the PyMT-induced mouse mammary tumors, Lin et al.⁶⁷ have characterized the development of the vasculature in mammary tumors during their progression to malignancy. They have shown that both angiogenic switch and the progression to malignancy are regulated by infiltrated macrophages in the primary mammary tumors. Moreover, inhibition of the macrophage infiltration into the tumor delayed the angiogenic switch and malignant transition, whereas genetic reduction of the macrophage population specifically in these tumors rescued the vessel phenotype. Finally, premature induction of macrophage infiltration into premalignant lesions promoted an early onset of the angiogenic switch independent of tumor progression.⁶⁷

De Palma et al.⁶⁸ have shown that a subset of monocytes that express the angiopoietin receptor Tie-2 are inducers of angiogenesis in both spontaneous and orthotopic tumor models. Knockout of these Tie-2-expressing cells in vivo markedly reduced angiogenesis in human glioma xenografts and prompted tumor regression.

A relationship between the macrophage content of tumors, the rate of tumor growth and the extent of their vascularization has been shown in several tumors, including breast carcinoma where TAM presence focally in large numbers correlates with a high level of angiogenesis and with poor prognosis, decreased relapse-free and overall survival of the patients,^{69, 70} malignant uveal melanoma,⁷¹ glioma,⁷² squamous-cell carcinoma of the esophagus,⁷³ bladder carcinoma⁷⁴ and prostate carcinoma.⁷⁵ In lung cancer, TAM may favor tumor progression by contributing to stroma formation and angiogenesis through their release of PDGF in conjunction with TGF-β-1 production by cancer cells.⁷⁶

Monocytes/macrophages display a high degree of plasticity, as shown by their ability to transdifferentiate into endothelial cells in vitro and in vivo.^{77, 78, 79, 80, 81, 82, 83, 84, 85} CD14⁺ mononuclear cells have been used as the starting population for cultivation of endothelial progenitor cells (EPCs).⁷⁷ Cultivated EPC grown from different starting populations, including peripheral blood mononuclear cells, have been shown to express endothelial markers such as von Willebrand factor, VEGFR-2, VE-cadherin, CD156 and CD31.⁸⁶ Monocytes coexpress endothelial lineage markers such as VEGFR-2 and AC133 and have the capacity to differentiate into adherent endothelial cells and to form cord-like structures in Matrigel.^{80, 87}

Kamihata et al.⁸⁸ and Shintani et al.⁸⁹ have shown that bone marrow mononuclear cells not only contain EPC but also angiogenic factors and cytokines and that implantation of bone marrow mononuclear cells into ischemic tissues augments collateral vessel formation.

Elsheikh et al.⁸¹ have shown that peripheral blood monocytes CD14⁺ and VEGFR-2⁺ exhibited the potential to differentiate in vitro into cells with endothelial characteristics. Moreover, these cells transduced by a lentiviral vector driving expression of green fluorescent protein (GFP) and transplantation of these cells into ballon-injured femoral arteries of nude mice significantly contributed to efficient re-endothelialization. Rehman et al.⁷⁸ reported that peripheral-blood endothelial-like cells are derived from monocytes/macrophages and secrete angiogenic factors.

Maniotis et al.⁹⁰ described a new model of formation of vascular channels by human melanoma cells and called it ‘vasculogenic mimicry’ to emphasize the de novo generation of blood vessels without the participation of endothelial cells and independent of angiogenesis. The word ‘vasculogenic’ was selected to indicate the generation of the pathway de novo and ‘mimicry’ was used because tumor cell pathways for transporting fluid in tissues were clearly not blood vessels.

Recently, we have demonstrated that multiple myeloma bone marrow TAM exposed to VEGF and FGF-2 develop a number of phenotypic properties similar to those of paired bone marrow endothelial cells, and form capillary-like structures overlapping morphologically those produced by endothelial cells.⁹¹ At ultrastructural level, multiple myeloma TAMs exhibit numerous cytoplasmic extroversions arranged in tube-like structures and these data suggest that TAMs contribute to build neovessels in multiple myeloma through a vasculogenic mimicry.

TAMs can influence angiogenesis by releasing angiogenic cytokines directly, or indirectly by secreting extracellular matrix-degrading enzymes that release angiogenic factors that have been sequestered by the matrix. TAMs are found in abundance in many tumor types and once the monocytes have entered the tumor environment they are activated by factors peculiar to the tumor microenvironment and migrate toward the areas of hypoxia perhaps following an oxygen gradient or chemokines released by surrounding tumor cells or other inflammatory cells. TAMs then release factors that initiate angiogenesis into the area and their production of extracellular matrix-degrading enzymes facilitates the growth of endothelial cells and loosens the fibrous network of the extracellular matrix thus allowing potentially metastatic tumor cells increased mobility.

With the seemingly central role that TAM could play in tumor angiogenesis, the macrophage itself becomes an appealing target for future antiangiogenic therapeutic strategies through two approaches: (1) compounds that suppress secretion of angiogenic substances by macrophages; and (2) compound that inhibit macrophage infiltration into the tumor mass.