The chromosomal high mobility group box-1 (HMGB1) protein acts as a proinflammatory cytokine when released in the extracellular environment by necrotic and inflammatory cells. In the present study, we show that HMGB1 exerts proangiogenic effects by inducing MAPK ERK1/2 activation, cell proliferation, and chemotaxis in endothelial cells of different origin. Accordingly, HMGB1 stimulates membrane ruffling and repair of a mechanically wounded endothelial cell monolayer and causes endothelial cell sprouting in a three-dimensional fibrin gel. In keeping with its in vitro properties, HMGB1 stimulates neovascularization when applied in vivo on the top of the chicken embryo chorioallantoic membrane whose blood vessels express the HMGB1 receptor for advanced glycation end products (RAGE). Accordingly, RAGE blockade by neutralizing Abs inhibits HMGB1-induced neovascularization in vivo and endothelial cell proliferation and membrane ruffling in vitro. Taken together, the data identify HMGB1/RAGE interaction as a potent proangiogenic stimulus.

Angiogenesis plays a key role in various physiologic and pathologic conditions, including tumor growth. Drm/gremlin, a member the Dan family of bone morphogenic protein (BMP) antagonists, is commonly thought to affect different processes during growth, differentiation, and development by heterodimerizing various BMPs. Here, we identify Drm/gremlin as a novel proangiogenic factor expressed by endothelium. Indeed, Drm/gremlin was purified to homogeneity from the conditioned medium of transformed endothelial cells using an endothelial-cell sprouting assay to follow protein isolation. Accordingly, recombinant Drm/gremlin stimulates endothelial-cell migration and invasion in fibrin and collagen gels, binds with high affinity to various endothelial cell types, and triggers tyrosine phosphorylation of intracellular signaling proteins. Also, Drm/gremlin induces neovascularization in the chick embryo chorioallantoic membrane. BMP4 does not affect Drm/gremlin interaction with endothelium, and both molecules exert a proangiogenic activity in vitro and in vivo when administered alone or in combination.

The contribution of polymorphonuclear neutrophils (PMN) to host defense and natural immunity extends well beyond their traditional role as professional phagocytes. In this study, we demonstrate that upon stimulation with proinflammatory stimuli, human PMN release enzymatic activities that, in vitro, generate bioactive angiostatin fragments from purified plasminogen. We also provide evidence that these angiostatin-like fragments, comprising kringle domain 1 to kringle domain 3 (kringle 1–3) of plasminogen, are generated as a byproduct of the selective proteolytic activity of neutrophil-secreted elastase. Remarkably, affinity-purified angiostatin kringle 1–3 fragments generated by neutrophils inhibited basic fibroblast growth factor plus vascular endothelial growth factor-induced endothelial cell proliferation in vitro, and both vascular endothelial growth factor-induced angiogenesis in the matrigel plug assay and fibroblast growth factor-induced angiogenesis in the chick embryo chorioallantoic membrane assay, in vivo. These results represent the first demonstration that biologically active angiostatin-like fragments can be generated by inflammatory human neutrophils. Because angiostatin is a potent inhibitor of angiogenesis, tumor growth, and metastasis, the data suggest that activated PMN not only act as potent effectors of inflammation, but might also play a critical role in the inhibition of angiogenesis in inflammatory diseases and tumors, by generation of a potent anti-angiogenic molecule.

Tumor neovascularization (angiogenesis) is regarded as a promising target for anticancer drugs. Heparin binds to fibroblast growth factor-2 (FGF2) and promotes the formation of ternary complexes with endothelial cell surface receptors, inducing an angiogenic response. As a novel strategy to generate antiangiogenic substances exploiting binding to FGF2 while preventing FGF receptor (FGFR) activation, sulfation gaps were generated along the heparin chains by controlled alkali-catalyzed removal of sulfate groups of iduronic acid 2-O-sulfate residues, giving rise to the corresponding epoxide derivatives. A new class of heparin derivatives was then obtained by opening the epoxide rings followed by oxidative glycol-splitting of the newly formed (and the preexisting) nonsulfated uronic acid residues. In vitro these heparin derivatives prevent the formation of FGFR/FGF2/heparan sulfate proteoglycan ternary complexes and inhibit FGF2-stimulated endothelial cell proliferation. They exert an antiangiogenic activity in the chick embryo chorioallantoic membrane assay, where the parent heparin is inactive. Low and very low molecular weight derivatives of a prototype compound, as well as its glycine and taurine derivatives obtained by reductive amination of glycol-split residues, retained the angiostatic activity. A significant relationship was found between the extent of glycol-splitting and the FGF2-antagonist/angiostatic activities of these heparin derivatives. Molecular dynamics calculations support the assumption that glycol-split residues act as flexible joints that, while favoring 1:1 binding to FGF2, disrupt the linearity of heparin chains necessary for formation of active complexes with FGFRs.