The secreted proMMP-9 and its activation determine angiogenesis-inducing capacity of human monocytes and macrophages. (A) Comparative analysis of angiogenesis-inducing capacity of purified proMMP-9 produced by different cell types. ProMMP-9 released by neutrophils or secreted into SF medium by monocytes or macrophages was purified by affinity chromatography and incorporated into collagen mixture to provide 3 ng proMMP-9 per onplant. Control onplants contained PBS only. (Upper) Zymographic images above the corresponding bars in the graph below indicate equal supplementation of proMMP-9 into the onplants. Triangles indicate positions of the monomer, heterodimer and homodimer of proMMP-9 released by neutrophils (left) and the proMMP-9 monomer secreted by monocytes and macrophages (right). Bar graph, levels of angiogenesis were determined as described in Figure 2B. From 4 to 6 embryos, each grafted with 6 collagen onplants, were analyzed per variable in 3 independent experiments. The pooled data are shown as fold changes (means ± SEM) in the levels of angiogenesis induced by proMMP-9 produced by different cell types compared with PBS control (1.0). Note that, although all tested purified proMMP-9 induced angiogenesis at levels significantly higher than the levels observed in PBS control, equal amounts of proMMP-9 produced by distinct cell types induce different levels of angiogenesis (***P < .0001). (B) The angiogenesis-inducing capacity of the secretate produced by monocytes and macrophages is contained in proMMP-9. (Upper) The flow-through fractions depleted of proMMP-9 by affinity chromatography (compare zymographic images in B and A), but containing >95% of initial protein content (insets from silver-stained gels positioned below zymographic images), were incorporated into collagen onplants at amounts equalized as to being produced by the same number of cells (1.5 × 10⁴/onplant). Open triangle in the zymograph indicates the position where proMMP-9 band would be localized before depletion. Bar graph, levels of angiogenesis were determined 3 days later as described in Figure 2B. The data are from a representative experiment using from 4 to 6 embryos per variant. Bars are means ± SEM of fold-changes in the levels of angiogenesis induced by cell secretates depleted of proMMP-9 compared with PBS control (1.0). (C) Activation of proMMP-9 produced by intact cells is required for induction of angiogenesis. Intact neutrophils and M1 or M2 macrophages were incorporated into collagen onplants (1.5 × 10⁴ cells/onplant), along with 2 µg/mL of either mAb 8-3H, which binds to proMMP-9, but does not block the activation of the proenzyme, or mAb 7-11C, which effectively blocks activation of proMMP-9 into the proteolytically capable MMP-9 enzyme. Control onplants contained PBS only. The levels of angiogenesis were determined as described in Figure 2B. For each variant, from 4 to 6 embryos grafted with 6 collagen onplants were analyzed in 2 independent experiments. The pooled data are shown as fold changes in the levels of angiogenesis induced by different cell types compared with PBS control (1.0). Bars are means ± SEM of fold changes in angiogenic indices. ***P < .0001. (D) Activation of proMMP-9 secreted by macrophages is required for induction of angiogenesis. Neutrophil releasate and SF CM from M1 or M2 macrophages was incorporated into collagen onplants at 1.5 ng of proMMP-9 per onplant along with 2 µg/mL of MMP-9–specific mAbs, namely control mAb 8-3H or activation-blocking mAb 7-11C. Control onplants contained PBS only. The levels of angiogenesis were determined as described in Figure 2B. The data are from a representative experiment using from 4 to 6 embryos per variant. Bars are means ± SEM of fold changes in the levels of angiogenesis induced by CM from different cell types compared with PBS control (1.0). **P < .005; ***P < .0001.