The Journal of Immunology

Dual Role of the Leukocyte aMb2 in

Dmitry A. Soloviev,* Stanley L. Hazen,† Dorota Szpak,* Kamila M. Bledzka,* Christie M. Ballantyne,‡ Edward F. Plow,* and Elzbieta Pluskota*

Polymorphonuclear neutrophils (PMNs) and macrophages are crucial contributors to neovascularization, serving as a source of

chemokines, growth factors, and proteases. aMb2(CD11b/CD18) and aLb2(CD11a/CD18) are expressed prominently and have been implicated in various responses of these cell types. Thus, we investigated the role of these b2 in angiogenesis. 2/2 2/2 Angiogenesis was analyzed in wild-type (WT), aM-knockout (aM ), and aL-deficient (aL ) mice using B16F10 melanoma, 2/2 RM1 prostate cancer, and Matrigel implants. In all models, vascular area was decreased by 50–70% in aM mice, resulting in stunted tumor growth as compared with WT mice. In contrast, aL deficiency did not impair angiogenesis and tumor growth. The 2/2 neovessels in aM mice were leaky and immature because they lacked smooth muscle cell and pericytes. Defective angiogenesis 2/2 2/2 in the aM mice was associated with attenuated PMN and macrophage recruitment into tumors. In contrast to WT or the aL 2/2 leukocytes, the aM myeloid cells showed impaired (Plm)-dependent extracellular matrix invasion, resulting from 50–75% decrease in plasminogen (Plg) binding and pericellular Plm activity. Surface plasmon resonance verified direct interaction of the

aMI-domain, the major ligand binding site in the b2 integrins, with Plg. However, the aLI-domain failed to bind Plg. In addition, endothelial cells failed to form tubes in the presence of conditioned medium collected from TNF-a–stimulated PMNs derived from 2/2 the aM mice because of severely impaired degranulation and secretion of VEGF. Thus, aMb2 plays a dual role in angiogenesis, supporting not only Plm-dependent recruitment of myeloid cells to angiogenic niches, but also secretion of VEGF by these cells. The Journal of Immunology, 2014, 193: 4712–4721.

one marrow–derived myeloid cells, particularly poly- degradation and remodeling and regulate the bioavailability of morphonuclear neutrophils (PMNs) and macrophages, various proangiogenic stimuli (5), all requisite events in angio- B are key regulators of tumor progression and metastasis. genesis (reviewed in Refs. 4, 6–8). In addition, PMNs and mac- One of the major tumor promoting functions of these cells is their rophages secrete urokinase-type plasminogen activator (uPA), facilitation of angiogenesis (reviewed in Refs. 1, 2). PMNs and which converts plasminogen (Plg) to plasmin (Plm). There are macrophages contribute to angiogenesis via a variety of well- diverse Plg receptors on leukocyte surface (reviewed in Ref. 9), established mechanisms. One example is their capacity to produce and bound Plm facilitates leukocyte migration/invasion by directly

by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. and secrete a variety of proangiogenic factors such as vascular degrading ECM, and encourages leukocyte recruitment in a vari- endothelial growth factor A (VEGF-A), fibroblast growth factor ety of in vivo models of inflammation (10–12). (FGF), IL-8, IL-10, CXCL1/GRO, and COX-2 (3, 4). In addition, aMb2 (CD11b/CD18) and aLb2 (CD11a/CD18), the two most PMNs and macrophages are a rich source of numerous proteases, broadly studied members of the b2 integrin subfamily, are par- including neutrophil , cathepsin G, and several metal- ticularly enriched in PMNs and macrophages, where they regulate loproteinases, which are crucial for extracellular matrix (ECM) diverse cell functions, including migration, adhesion, the respi- ratory burst, and cytokine production (13). In addition, we have

previously demonstrated that aMb2 enhances uPA-dependent Plg *Department of Molecular Cardiology, Joseph J. Jacobs Center for Thrombosis and Vascular Biology, Learner Research Institute, Cleveland Clinic, Cleveland, OH 44195; activation on the PMN surface (14, 15), which has the potential to http://classic.jimmunol.org †Department of Molecular and Cellular Medicine, Cleveland Clinic, Cleveland, OH influence their recruitment to inflammatory or angiogenic sites ‡ 44195; and Baylor College of Medicine and Methodist DeBakey Heart and Vascular in vivo. Based on these observations, we hypothesized that a b Center, Houston, TX 77030 M 2 and aLb2, as key regulators of leukocyte functions, might be im- Received for publication January 24, 2014. Accepted for publication August 22, 2/2 2/2 2014. plicated in angiogenesis. In this study, we used aM and aL This work was supported by National Institute of Allergy and Infectious Diseases mice and three distinct in vivo angiogenesis models to show that Grant AIO 80596 (to D.A.S.), American Heart Association Grant SDG 0335088N (to aMb2, but not aLb2, is a critical contributor to angiogenesis. This

Downloaded from E.P.), and National Institutes of Health–National Heart, Lung, and Blood Institute a b Grants R01 HL17964 and P01 HL07331 (to E.F.P.). function of M 2 is mediated by two distinct mechanisms: 1) support of Plm-dependent PMN and macrophage recruitment to Address correspondence and reprint requests to Dr. Elzbieta Pluskota, Department of Molecular Cardiology, NB50, Lerner Research Institute, Cleveland Clinic, 9500 angiogenic niches; and 2) enhancement of leukocyte production and Euclid Avenue, Cleveland, OH 44195. Email address: [email protected] secretion of the primary angiogenic growth factor, VEGF-A. The online version of this article contains supplemental material. Abbreviations used in this article: BM, bone marrow; BMT, BM transplantation; EC, Materials and Methods endothelial cell; ECM, extracellular matrix; FGF, fibroblast growth factor; GST, gluta- Materials thione S-transferase; KC, keratinocyte-derived cytokine; MAEC, mouse aortic endothe- lial cell; MOMA-2, macrophage–monocyte mAb; NG2, neural/glial Ag 2; Plg, Mouse VEGF165 and keratinocyte-derived cytokine (KC) were from plasminogen; Plm, plasmin; PMN, polymorphonuclear neutrophil; sc-uPA, single- Biosource International (Camarillo, CA); was from Sigma- chain uPA; sFLT-1, soluble VEGF receptor-1; SMA, smooth muscle actin; SPR, surface plasmon resonance; uPA, urokinase-type plasminogen activator; VEGF-A, Aldrich (St. Louis, MO); biotin-conjugated anti-mouse CD31 mAb was vascular endothelial growth factor A. from BD Pharmingen (San Jose, CA); rabbit anti-smooth muscle actin (SMA; Abcam, Cambridge, MA); rabbit anti–neural/glial antigen 2 (NG2; Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 Millipore, Temecula, CA); rabbit anti-mouse laminin (Serotec, Oxford,

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400202 The Journal of Immunology 4713

UK), goat anti-Fibrin II (Accurate Chemical, Westbury, NY), purified, or transferred to 1.25% dextran T500 solution (1:9) to sediment erythrocytes FITC-labeled rat anti-Ly6G, clone 1A8, specific for mouse PMNs were for 30 min at room temperature (14, 18). Leukocyte-rich supernatants were from BD Pharmingen (San Jose, CA); anti-mouse macrophage–monocyte washed with PBS once, and PMNs were isolated by magnetic positive mAb (MOMA-2) was from Chemicon (Temecula, CA); and rat LEAF TM selection using mouse anti-Ly6G MicroBead Isolation Kit (Miltenyi Bio- purified anti-mouse aM integrin (clone M1/70) was from BioLegend (San tec, Auburn, CA) according to the manufacturer’s instructions. Eluted cells Diego, CA). Glu-Plg was isolated from normal human plasma by affinity were 98% granulocytes, of which more than 96% were neutrophils and chromatography on lysine-sepharose followed by gel filtration. Growth 1–2% were eosinophils. Contaminating lymphocytes and monocytes were factor–reduced Matrigel matrix was from BD Biosciences (San Diego, less than 2% as determined by Wright staining. PMN viability was .98% CA). Murine recombinant TNF-a was from R&D Systems. Cycloheximide as determined by trypan blue staining. The PMN yield was usually ∼0.5 3 6 and pentoxifylline were from Sigma-Aldrich. 10 per mouse, and blood pooled from 10–15 mice was used. For Plg activation assays, lymphocytes and monocytes were collected from buffy Mice coats and washed twice with the HBSS buffer. Leukocytes were obtained a 2/2 a 2/2 from blood pooled from 7–10 mice. The M mice were generated as described previously (16), and L For matrix invasion and Plg binding assays, macrophages and PMNs mice were purchased from the Jackson Laboratory. Both strains were were isolated from peritoneal lavages, PMNs at 6 h and macrophages at 72 h backcrossed for 12 generations into a C57BL/J6 background. The study was after i.p. thioglycolate injection, when their recruitment was at the highest conducted under protocols approved by the IACUC of the Cleveland Clinic. levels for each cell type (10). PMNs constitute 92% and macrophages Angiogenesis models in vivo constitute 90% of all cells in the 6- and 72-h peritoneal lavages, respec- tively. PMNs and macrophages harvested from lavages are referred to as Eight- to twelve-week-old mice were injected s.c. with 106 murine B16F10 “peritoneal PMN” or “peritoneal macrophages” in the manuscript to dis- melanoma or RM1 prostate cancer cells. Tumors were collected 8– tinguish them from blood cells. 14 d after injection and were weighed, photographed, and processed for immunohistochemical staining. Endothelial cells (ECs) were identified Matrix invasion assays ex vivo using a biotinylated mouse CD31 mAb, SMAs with anti-SMA Ab, peri- m cytes with anti-NG2 Ab, fibrin with anti-Fibrin II Ab, basement membrane Prechilled tissue culture inserts with porous (8- m pore size) polyester m with anti-laminin Ab, PMNs with rat anti-Ly6G (clone 1A8), and mono- membrane (Costar, Corning, NY) were coated with 50 l/insert Matrigel (0.5 mg/ml) overnight at 4˚C until dry. Next, matrices were rehydrated cytes–macrophages with MOMA-2 mAbs. Stained sections were analyzed m using fluorescent imaging microscopy (Leica, Wetzlar, Germany) and with 600 l DMEM F-12 for 1 h. Peritoneal PMNs or macrophages were suspended in serum-free DMEM F-12 medium (Life Technologies, ImagePro Plus Capture and Analysis software (Media Cybernetics). CD31, 3 5 fibrin, and Ly6G- or MOMA2-positive area was quantified in 5–10 inde- Carlsbad, CA), and added to matrix-coated inserts (1 10 /insert), which pendent fields. The average area per field was determined from duplicate were placed in a 24-well plate containing serum-free DMEM F-12 sup- plemented with or without 100 ng/ml KC. Plg (90 mg/ml) was added to measurements of each of the fields analyzed. Matrigel plug assay was appropriate inserts, and cells were incubated for 18 h at 37˚C. Assays were performed as described (17). Mice were injected with 500 ml growth stopped by removing the inserts and wiping the inside with a cotton swab factor-reduced Matrigel mixed at 4˚C with heparin (26 U/ml) alone or with to remove nonmigrated cells. The migrated cells were quantified using the KC (500 ng/ml) or VEGF 165 (100 ng/ml; R&D Systems). Matrigel plugs Cyquant Cell Proliferation Kit (Molecular Probes, Eugene, OR) according were harvested 8 d after injection and snap-frozen, and 8-mm sections were to the manufacturer’s instructions. processed for immunohistochemical analyses as described above. In aM- integrin blocking experiments, WT mice were injected intravenously with Plg activation on the leukocyte surface rat LEAF anti-mouse aM integrin (clone M1/70; BioLegend) or isotype- matched normal rat IgG2b (3.5 mg/kg), 4 h before and then 2, 4, and 6 d after Peripheral blood PMNs or lymphocyte–monocyte mixtures were incubated Matrigel-KC implantation. The Matrigel plugs were collected 8 d after in- with KC (100 ng/ml) for 1 h at 37˚C in the presence of 25 nM single-chain

by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. jection, sectioned, and stained with anti-CD31 to examine vascular formation. uPA (sc-uPA) in 10 mM Tris-Cl buffer (pH 7.4) containing 0.14 mM NaCl, 0.1% BSA, 1 mM CaCl2, and 1 mM MgCl2. The cells were washed three Bone marrow transplantation times, and 50 ml of the cell suspension (1 3 106 cells/well) were added to Two-month-old recipient mice were lethally X-irradiated with a total dose each well of the microtiter plates. Peritoneal macrophages, which had not been m of 9 Gy and reconstituted with i.v. injection of 107 bone marrow (BM) cells incubated with KC or sc-uPA, were also added to the plates. Next, 100 lof m isolated from the femurs of donor mice. Mice were used 6–8 wk after BM a Glu-Plg (1 M) and Plm-specific fluorescent substrate H-D-Val-Leu-Lys-7- transplantation (BMT). Engraftment efficiency was examined 6 wk after amido-4-methylcoumarin (2 mM) mixture was added, and Plm formation was 2 2 2 2 2 2 / / / monitored over 45 min at 37˚C at aem = 370 nm and aex = 470 nm using BMT in chimeric aM →WT and WT→aM mice using WT and aM mice, which did not undergo BMT as controls did. Single-cell suspensions a fluorescence plate reader (SpectraMax GeminiXS; Molecular Devices). from spleens and thymuses from these mice were prepared, and percentages of individual leukocyte subsets were measured with flow cytometry using Regulation of VEGF by PMNs http://classic.jimmunol.org FITC-labeled Abs to cell-specific markers (Ly-6G for PMN, F4/80 for 2/2 2/2 Peripheral blood WT, aM or aL PMNs were incubated in 24-well macrophages, CD19 for B lymphocytes, CD3 for T lymphocytes) and FITC- tissue culture plates (Costar; 3 3 106 cells/well) in 250 ml DMEM F-12 labeled isotype-matched Abs as controls. medium in the absence or presence of TNF-a (20 ng/ml) for 2 h at 37˚C. m Plg binding to a and a -I domains The inhibitors of synthesis (cycloheximide-10 g/ml) or PMN M L degranulation (pentoxifylline, 3,7-dimethyl-1-[5-oxohexyl]-xantine, 300 m a Glutathione S-transferase (GST)-fused aMI and aLI domains were purified M) were added to PMNs 60 min before the addition of TNF- . Super- with glutathione chromatography. Real-time protein-protein interactions natants were collected and centrifuged at 1500 rpm in a Beckman GS-6 Downloaded from were analyzed using a Biacore 3000 instrument (Biacore AB). Plg was centrifuge for 10 min, and VEGF was measured using mouse VEGF immobilized on CM5 biosensor chips by amine coupling. Experiments Quantikine ELISA Kit (R&D Systems). In parallel, concen- were performed at 22˚C in 10 mM HEPES buffer (pH 7.4) containing 150 tration was measured in PMN supernatants using mouse Lactoferrin mM NaCl and 0.005% surfactant P20 (flow rate, 25 ml/min). Surface LTF/LF Elisa Kit (Cusabio). plasmon resonance (SPR) sensorgrams were obtained by injecting various concentrations of GST-tagged aMIoraLI domain over immobilized Plg Quantitative real-time PCR and reference flow cells. Surfaces were regenerated with 30-s pulses of Total RNA was isolated from peripheral blood mouse PMNs, either resting 10 mM NaOH. Association–dissociation curves were determined after or stimulated with TNF-a (20 ng/ml) for 2 h at 37˚C, and from WT and the subtraction of the reference surface values and buffer binding at six 2 2 a / mouse CD117+ bone marrow progenitor cells with Trizol reagent selected concentrations. Sensorgrams were analyzed using BIAevaluation M according to the manufacturer’s protocol. Total RNA (∼4 mg) was reverse software (version 4.01; GE Healthcare). transcribed using the SuperScript III First-Strand Synthesis System for RT- PMN and macrophage isolation PCR (Invitrogen) and random hexamers, according to the manufacturer’s protocol. Real-time quantitative PCR of the cDNA template was performed Mouse PMNs for use in mouse aortic endothelial cell (MAEC) tube for- with an iCycler (Bio-Rad). VEGF and GAPDH primers were from mation and Plg activation assays (see below) were isolated from blood SABiosciences (Frederick, MD). The PCR contained 150 ng of cDNA, drawn from hearts of anesthetized animals into sterile acid-citrate-dextrose 10 mM forward and reverse primers, and 12.5 ml23 RT2 Real Time SYBR (1:7 volume 145 mM sodium citrate, pH 4.6, and 2% dextrose). Blood was Green PCR Master mix (SABiosciences) in a total volume of 25 ml. All 4714 INTEGRIN aMb2-DEPENDENT ANGIOGENESIS

PCR were performed in triplicate. Results were calculated as expression of with Matrigel alone or with Matrigel supplemented with VEGF or the target gene relative to expression of the reference gene (GAPDH). KC (keratinocyte-derived factor) to stimulate angiogenesis. CD31 staining of Matrigel plugs containing VEGF or KC revealed well- MAEC tube formation assay 2/2 formed vasculature in the implants from WT and aL mice, m 2/2 Twenty-four-well tissue culture plates were coated with 250 l Growth whereas the Matrigel plugs from aM mice showed no distinct Factor-Reduced Matrigel (BD Biosciences, San Diego, CA) and incubated vascular formations, although a few CD31-positive ECs were dis- at 37˚C for 30min. When the Matrigel solidified, 1.25 3 105 of WT MAECs were added to each well in 250 ml of chosen PMN-conditioned cerned within the plugs (Fig. 1E). Regardless of the angiogenic 2/2 DMEMF-12 medium obtained as described in Regulation of VEGF by factor used, blood vessel area in the Matrigel implants in the aM PMNs and supplemented with 10% FBS and 90 mg/ml heparin. Inhibitors mice was reduced by ∼75% compared with the implants from the of VEGF, neutralizing LEAF rat anti-mouse VEGF mAb (BioLegend), a 2/2 and WT animals (p , 0.01, n = 8 per group; Fig. 1F). In m L isotype matched rat IgG2a (100 g/ml; BioLegend), and recombinant control Matrigel plugs without proangiogenic cytokines, no blood mouse soluble VEGF receptor-1 (sFLT-1; 100 ng/ml; R&D Systems) were preincubated for 60 min with conditioned media of WT TNF-a–stimulated vessels were detected in any of the three mouse genotypes tested PMNs before its addition to MAECs. Live time-lapse photography was (data not shown). performed for 12 h, using 5-min intervals on a Leica DMIRB Inverted 2/2 Microscope equipped with a Roper Scientific CoolSNAP HQ Cooled CCD Neovasculature in aM mice is immature camera, a temperature controller, and a CO2 incubation chamber. Snap- shots were taken using MetaMorph Software. Tube formation was ana- The presence of smooth muscle cells and pericytes within vas- lyzed and quantified using ImageJ software version 1.34. culature is a key indicator of its maturity because they stabilize blood vessels. We double-stained melanoma and prostate tumor Statistical analysis sections with Abs to CD31 and to SMA (Fig. 2A, top panel)orto Data are expressed as mean 6 SEM. To determine significance, a one-way NG2 chondroitin sulfate proteoglycan, a marker of pericytes 2/2 ANOVA test was performed to compare angiogenic responses between the (Fig,2A bottom panel). In prostate tumors grown in WT and aL three mouse genotypes, and a two-tailed Student t test was performed for mice 30–37% of total CD31+ blood vessels stained for SMA, a 2/2 2 2 comparisons between WT and the M mice using the Sigma-Plot a / , whereas only 15% of blood vessels formed in M mice software program (SPSS); p 0.05 was considered to be statistically , significant. expressed this marker (p 0.02, n = 8; Fig. 2B). Costaining for CD31 and NG2 revealed reduced pericytes interacting with blood 2/2 vessels in prostate tumors in aM mice as compared with WT Results 2/2 2/2 and aL mice (25% NG2-positive blood vessels in aM ver- Integrin aMb2 is critical in angiogenesis in vivo 2/2 sus 50–58% in aL and WT mice; p , 0.05, n = 8; Fig. 2C). Leukocytes, particularly PMNs and macrophages, are important Measurement of blood vessel diameter revealed that they were 2/2 supporters of angiogenesis as a source of proteases and angiogenic substantially smaller in prostate tumors of aM mice as com- 2/2 factors (reviewed in Refs. 1, 2). The b2 integrins are crucial in the pared with WT and aL mice (25 6 8 mm versus 50–55 6 regulation of a variety of leukocyte responses, including adhesion, 10 mm; p , 0.05, n = 60; 4 mice per group; Fig. 2D, top panel, migration, and cytokine production (reviewed in Ref. 13). Ac- and 2E). Immunostaining for laminin showed a 50% reduction in 2/2 cordingly, we examined the role of two prominent members of the basement membrane thickness of blood vessels in aM mice 2/2 by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. b2 integrin subfamily, aMb2 and aLb2, in angiogenesis. These two (2.7 6 0.3mm) compared with WT and aL mice (5.8–6 6 1.3 integrins share the same b2-subunit, and their a-subunits are 49% mm; p , 0.03, n = 60; 4 mice per group; Fig. 2D, lower panel, and 2/2 2/2 identical (19). Angiogenesis was analyzed in the aM , aL , 2F). With decreased maturation and laminin deposition, we con- 2/2 and control WT mice using two tumor models, murine B16F10 sidered that vasculature in aM mice might be leaky. Plasma melanoma, and RM1 prostate cancer. These tumors are highly leakage measured as an area that is positive for plasma-derived 2/2 vascularized, and their growth is heavily dependent on an angio- fibrin was enhanced by 2.5-fold in tumors grown in aM mice 2/2 genic response (20, 21). Staining of tumor sections for endothelial compared with WT and aL mice (13.8 6 1.8% versus 5.2 6 cells with CD31 (green fluorescence) revealed normal, well- 1.1% and 3.8 6 0.7%, respectively; p , 0.03, n = 20; 4 mice per 2/2 developed, thick vasculature in tumors grown in the aL and group; Fig. 2G, 2H). The vasculature in melanoma tumors grown 2/2 http://classic.jimmunol.org WT mice. In contrast, angiogenesis was significantly impaired in in aM mice also showed reduced maturity and leakiness (data 2/2 aM mice as only a few short and thin vessel-like structures were not shown). In addition, the permeability of preexisting blood 2/2 observed in melanomas and prostate tumors from these mice vessels in dorsal skin of aM and WT mice was examined using (Fig. 1A). Blood vessel area of the melanoma sections was atten- Evans blue dye injected i.v. Baseline permeability upon injection 2/2 2/2 uated by ∼70% in the aM compared with tumors in the aL of control PBS and VEGF-A–induced vascular permeability were 2/2 and WT mice (p , 0.001, n = 8 per group; Fig.1B, upper panel). In similar in aM and WT mice, indicating that preexisting vas- 2/2 Downloaded from addition, significantly reduced vascular area was observed in RM1 culature in aM mice was normal (Supplemental Fig. 1). 2/2 2/2 prostate tumors grown in aM mice compared with aL and WT mice: 6200 mm2 6 1000 versus 12800 6 2000 in a 2/2 and Impaired PMN and macrophage recruitment to angiogenic L a 2/2 12000 6 2150 mm2 in WT mice (p , 0.01, n = 8 per group; sites in M mice Fig. 1B, lower panel). Consistent with the blunted angiogenic re- The CD11b+/Gr-1+ myeloid cells consisting primarily of PMNs 2/2 sponse in the aM mice, melanomas grown in these mice were are crucial enhancers of angiogenesis, and they contribute to an- 70–80% smaller (p , 0.01, n = 7 per group) than those derived in giogenic switch in many tumors (4, 5, 22). Tumor growth and 2/2 2/2 2/2 2/2 the aL and WT mice: 40 6 10 mg in aM versus 202 6 42 angiogenesis were impaired in the aM (CD11b ) mice; 2/2 mg in WT and 178 6 34 mg in aL mice. In addition, the average therefore, we considered the possibility that this integrin regulates 2/2 weight of RM1 prostate tumors recovered from the aM mice was recruitment of these cells to growing tumors. First, we examined 2/2 + + ∼40–50% lower than from the aL and WT mice (p , 0.05, n =8 infiltration of CD11b /Gr-1 cells in prostate and melanoma 2/2 2/2 per group): 295 6 30 mg in aM versus 497 6 80 mg in WT and tumors in WT and aL mice by double staining with anti- 2/2 570 6 100 mg in aL mice (Fig. 1C, 1D). CD11b (green) and anti-Ly6G (Gr-1) mAbs (red). As shown in 2/2 + Impaired angiogenesis in the aM mice was corroborated Fig. 3A, 70–80% of CD11b cells were also positive for Gr-1, and using Matrigel as a third angiogenic model. Mice were injected numerous CD11b+/Gr-1+ cells were detected in prostate tumors in The Journal of Immunology 4715

2/2 FIGURE 1. Angiogenesis is impaired in the aM mice. (A) Representative images of melanoma (left panel) and prostate (right panel) tumor sections stained with an EC marker (CD31 Ab, green). Scale bars, 50 mm. (B) Decreased area of CD31-stained vascula- 2/2 ture in tumors grown in aM mice. Data are repre- sentative of two independent experiments with eight mice per group. (C and D) Average weight of mela- 2/2 noma and prostate tumors grown in aM is less than in WT mice. Representative tumors are shown in (C). Data are means 6 SEM (n = 8 mice per group). (E) Representative images of Matrigel implant sections stained with CD31 Ab. Scale bars, 50 mm. (F) Reduced area of CD31-positive vasculature in the VEGF- or KC- 2/2 containing Matrigel implants from the aM mice. The data are representative of three independent experiments with eight mice per group. *p , 0.05. by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. http://classic.jimmunol.org 2/2 2 2 2/2 WT and aL mice, with no differences in their recruitment 5610 6 985 mm in WT and 5240 6 450 mm in aL mice, 2 2/2 observed in the two mouse strains. Similar results were obtained whereas in prostate tumors it was 3600 6 750 mm in aM mice 2 2 2/2 with melanoma tumors (data not shown). An assessment of versus 7500 6 1100 mm in WT and 7280 6 1300 mm in aL + + 2/2 2/2 CD11b /Gr-1 cell recruitment to tumors in aM (CD11b ) littermates (p , 0.05, n = 8 per group; Fig. 3E). In addition, mice was not feasible because of the absence of the aM integrin staining of Matrigel implant sections with anti-PMN and MOMA-

Downloaded from subunit on these cells; therefore, we stained tumor sections only 2 Abs revealed robust VEGF- and KC-dependent leukocyte in- 2/2 with PMN-specific anti-Ly6G (Gr-1) mAb. Indeed, PMN infil- filtration into the centers of the implants in WT and aL mice, 2/2 tration into both melanoma and prostate tumors was significantly whereas it was inhibited in the aM mice (data not shown). 2/2 2/2 reduced in the aM mice compared with WT and aL litter- mates (Fig. 3B). Quantification of Ly6G-positive areas in tumor BMT and suppression with blocking Abs confirm the 2 2/2 importance of aMb2 in angiogenesis sections verified these observations: 900 6 210 mm in aM 2 2 2/2 versus 3600 6 1120 mm in WT and 3950 6 350 mm in aL Next, we sought to confirm that reduced tumor growth and an- 2 2/2 mice in melanomas (p , 0.02, n = 8) and 1200 6 265 mm in giogenesis in the aM mice are due to impaired functions of 2/2 2 2 aM versus 6020 6 840 mm in WT and 7135 6 1780 mm in bone marrow-derived cells, including leukocytes, and are not 2/2 aL mice in prostate tumors (p , 0.01, n = 8 per group; caused by defective vascular cells. Therefore, we performed BMT Fig.3C). Next, tumor sections were stained with MOMA-2. experiments and examined growth and angiogenesis in RM1 tumors. 2/2 2/2 Macrophage infiltration into both tumors was decreased by 50– Transplantation of aM BM into WT hosts (aM →WT) resulted 2/2 2/2 60% in the aM mice as compared with WT and aL litter- in reduced RM1 tumor growth and angiogenesis (tumor weight: 2/2 2/2 mates (Fig. 3D, 3E). In melanomas grown in the aM mice, p = 0.0169; vascular area: p =0.028foraM →WT versus macrophage-positive area was 1820 6 160 mm2 compared with WT→WT; n = 5; Fig. 4A, 4B). Alternatively, transplantation of WT 4716 INTEGRIN aMb2-DEPENDENT ANGIOGENESIS

2/2 FIGURE 2. Neovasculature in prostate tumors in aM mice is immature and leaky. Immunohistochemistry and image analyses of RM1 prostate tumors 2/2 2/2 implanted into WT, aM , and aL mice. (A) Costaining for SMA (green) and CD31 (red; top panel) and for NG2 (green) and CD31 (red; bottom panel)in RM1 prostate tumors. Nuclei are stained with DAPI. Scale bars, 50 mm. (B and C) Quantification of the data presented in (A), top and bottom panels, respectively. Data are expressed as mean 6 SEM and are representative of three independent experiments (n = 8 mice/group). (D) CD31-stained (top panel) 2/2 2/2 and laminin-stained (bottom panel) blood vessels in tumors grown in WT, aM , and aL mice. Scale bars, 50 mm(top panel) and 25 mm(lower panel). (E) Vessel diameter was measured in 60 tumor vessels cut perpendicular to their longitudinal axis in 8-mm-thick sections, stained for CD31 from each mouse (n=4 mice/group). Data are expressed as mean 6 SEM (n = 60). (F) Thickness of laminin-positive basement membrane in blood vessels formed in 2/2 2/2 by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. WT, aM , and aL mice. Data are mean 6 SEM (n = 60) and are representative of three independent experiments including four mice per group. (G) 2/2 2/2 Representative photograph of fibrin content (brown) in tumors grown in WT, aM , and aL mice. Tumor sections were stained with anti-Fibrin II beta Ab. Scale bars, 25 mm. (H) Quantification of fibrin-positive areas in tumor sections stained with anti-fibrin Ab. Data are means 6 SEM (n = 8) and are representative of three independent experiments including eight mice per group.

2/2 2/2 BM into the aM hosts (WT→aM ) restored growth and (Ly-6G, F4/80, CD19, CD3, respectively) and FITC-labeled angiogenesis of prostate tumors to these of control WT mice isotype-matched Abs as controls. We found that the content of receiving WT bone marrow (RM1 weight: p = 0.0183; vascular individual leukocyte subsets was similar in respective organs in 2/2 2/2 area: p =0.012forWT→aM versus aM →WT; n =5; both chimeric mouse lines as well as in control mice (no BMT), http://classic.jimmunol.org Fig. 4A, 4B). These results suggest that blunted tumor growth indicating that aM deficiency did not affect BMT engraftment 2/2 and angiogenesis in the aM mice is a consequence of altered efficiency (Supplemental Table I). In addition, flow cytometry of bone marrow-derived cell functions. In addition, image analysis circulating total leukocytes with anti-mouse aM mAb confirmed 2/2 of RM1 tumor sections stained with PMN-specific anti- its absence in the aM →WT chimeras and its presence in the 2/2 Ly6G and anti-monocyte–macrophage MOMA-2 Abs revealed WT→aM mice (data not shown). impaired recruitment of these cells to tumors grown in To confirm the importance of the aM integrin in angiogenesis, a 2/2 a

Downloaded from WT recipients receiving M BM as compared with control WT mice were injected with a rat anti- M blocking Ab (clone WT→WT mice, indicating a crucial role of aMb2 in this process M1/70) that has been shown to inhibit neointima formation in (p = 0.0184 for PMNs [Ly6G] and p = 0.0167 for MOMA-2; rabbits specifically (23). These neutralizing Abs bind to the ligand- n = 5; Fig. 4C, 4D). In contrast, transplantation of WT BM to binding site within the aM subunit and inhibit its interactions with 2/2 aM recipients restored not only angiogenesis, but also PMN ligands thereby mimicking the gene ablation. Matrigel implants were and macrophage migration into tumors (PMNs: p = 0.021; harvested after 8 d, and the Ab was injected before Matrigel injec- 2/2 2/2 MOMA-2: p = 0.0454 for WT→aM versus aM →WT; n =5; tion and then every 2 d. Vasculature was analyzed in Matrigel Fig. 4C, 4D). To exclude the possibility that aM deficiency might sections stained with anti-CD31 mAb (Fig. 4E). The anti-aM impair recovery of the immune system upon BMT, we assessed Ab reduced KC-induced angiogenesis in Matrigel implants by 2/2 engraftment efficiency 6 wk after BMT in chimeric aM →WT ∼60–80% as compared with noninjected mice, whereas isotype- 2/2 2/2 and WT→aM mice using WT and aM mice not undergoing matched normal rat IgG2b had no effect indicating specificity (p , BMT as controls. We have collected spleens and thymuses from 0.05 mice injected with anti-aM versus not injected; n = 4 mice these mice, prepared single-cell suspensions, and measured per- per group; Fig. 4E, 4F). These experiments verify the importance centages of PMN, macrophage, B cells, and T cells by flow of aMb2 on myeloid cells in tumor-induced angiogenesis via cytometry using FITC-labeled Abs to cell specific markers regulation of leukocyte recruitment to sites of neovascularization. The Journal of Immunology 4717

FIGURE 3. Reduced leukocyte infiltration of tumors 2/2 in the aM mice. (A) Prostate tumor sections from 2/2 WT and aL mice were double stained with FITC- labeled rat anti-CD11b (M1/70) mAb (green) and rat anti- Ly6G (Gr-1; clone 1A8-red). Numerous CD11b+/Gr-1+ cells are present on merged images (yellow-orange) in both mouse strains. Representative images of the melanoma and prostate tumor sections stained with PMN-specific anti-Ly6G (B) and the monocyte–mac- rophage-specific MOMA-2 (D) Abs. Image analysis shows reduced Ly6G-positive (C) and MOMA-2– positive (E) area in melanoma and prostate tumors in 2/2 the aM mice. Data are means 6 SEM (n =8mice per group) and are representative of three independent experiments. Scale bars, 50 mm. *p , 0.05. by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd.

aMb2 facilitates leukocyte recruitment to angiogenic sites via leukocytes to invade tumors could be caused by impaired Plg

interaction with Plg and enhancement of Plg activation activation on the leukocyte surface. Because aMb2 and aLb2 are As Plm-dependent ECM proteolysis greatly facilitates leukocyte the most abundant b2-integrins on PMNs and macrophages, we 2/2 invasion (10–12), we hypothesized that the inability of aM used a modified Boyden chamber system to elucidate the role of

http://classic.jimmunol.org FIGURE 4. Defective hematopoietic cells contribute 2/2 to reduced tumor growth and angiogenesis in the aM mice. (A) Average weight and (B) CD31-positive vas- cular area in RM1 prostate tumors in mice undergoing 2/2 BMT with WT or aM donor marrow. Data represent mean 6 SEM (n = 5 mice per group). (C) Ly6G-positive and monocyte–macrophage-positive (D)areainRM1 Downloaded from 2/2 tumors in mice undergoing BMT with WT or aM donor marrow. Data represent mean 6 SEM (n = 5 mice per group). (E and F) Intravenous administration of

blocking mAb (M1/70) to aM inhibits KC-dependent angiogenesis in Matrigel plug model in WT mice. M1/70

and normal rat IgG2b (3.5 mg/kg) were injected before Matrigel injection and on days 2, 4, and 6. Matrigel implants were harvested on day 8, sectioned, and stained with anti-CD31 mAb. (E) Representative images of Matrigel sections stained with anti-CD31 (brown). Scale bars, 50 mm. (F) Quantification of the CD31-positive area in Matrigel implants. Data are means 6 SEM (n = 4) and are representative of two independent experi- ments. *p , 0.05. 4718 INTEGRIN aMb2-DEPENDENT ANGIOGENESIS by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. http://classic.jimmunol.org

FIGURE 5. aMb2 directly interacts with Plg, enhances its activation, and facilitates Plm-dependent leukocyte recruitment. KC-directed transmigration of 2/2 peritoneal PMNs (A) and macrophages (B) through Matrigel-coated inserts is Plg-dependent and is impaired in aM leukocytes. The data are means 6

Downloaded from SEM of triplicate samples and are representative of two independent experiments including three mice per group. (C) Reduced binding of soluble Alexa488- 2/2 labeled Plg to peritoneal aM PMNs and macrophages. The cells were incubated with increasing concentrations of Plg as indicated for 30 min at 37˚C. After two washings, Plg binding to cell surface was analyzed using a FACSCalibur flow cytometer and CellQuest software. The cells incubated without Plg were set as negative controls. The data are means 6 SEM of triplicate samples (n=3 mice per group) and are representative of two independent

experiments. (D) Plg was immobilized on the CM5 sensor chip surfaces (500 RU). Sensorgrams obtained for a concentration series of GST-aM-I (left panel) 2/2 and GST-aL–I domain (right panel). (E and F) Reduced Plg activation on the surface of peripheral blood aM PMNs and peritoneal macrophages. The results are means 6 SEM of triplicate samples (n = 5 mice per group) and are representative of two independent experiments. *p , 0.05.

2/2 aMb2 and aLb2 in mouse peritoneal PMN and macrophage mi- effect was overcome by addition of Plg to the aL and WT 2/2 gration through a Matrigel matrix barrier (ECM extracted from leukocytes (Fig. 5A, 5B). In contrast, addition of Plg to the aM Engelbreth-Holm-Swarm mouse sarcoma) in response to KC cells failed to improve their migration through the Matrigel (Fig. 5A, 5B). In the absence of KC, there was minimal migration barrier, suggesting that the capability of these cells to bind and 2/2 of leukocytes into the lower chambers in the presence or absence activate Plg was limited (p = 0.014 aM versus WT PMNs and 2/2 of added Plg. Although KC-induced migration of PMNs and p = 0.005 aM versus WT macrophages; n = 3; Fig. 5A, 5B). macrophages was significantly impeded by Matrigel, this barrier Indeed, flow cytometry showed a 50–60% reduction in binding The Journal of Immunology 4719

2/2 of Alexa488-conjugated soluble Plg to aM peritoneal PMNs 2/2 and macrophages as compared with WT and aL leukocytes (Fig. 5C). To determine whether Plg interacts directly with the aMI- and aLI- domain, the major site of ligand binding in b2 integrins (13), we performed SPR experiments. GST-tagged aMI-domain interacted with Plg immobilized on biosensor chips in a concentration-dependent manner, whereas the GST-tagged aLI-domain did not bind Plg (Fig. 5D). GST alone also did not interact with Plg. From the progress curves of the Plg:aMI-domain interaction, we estimated a Kd = 1.76 6 0.9 3 1027. This value was derived by fitting the kinetic data to a 1:1 global Langmuir model, and the stoichiom- etry observed at ligand saturation was 1:1. 2/2 Next, we compared Plg activation on the surface of WT, aL , 2/2 and aM leukocytes using a fluorogenic plasmin-specific pep- tide substrate. Peripheral blood PMNs and lymphocytes were stimulated with KC and pretreated with sc-uPA to activate the integrin and enable sc-uPA binding to leukocyte surface, respec- tively. No plasmin activity was detected in the absence of leuko- 2/2 cytes. The aM PMNs showed a 50% reduction in Plm 2/2 generation as compared with WT and aL PMNs (p , 0.03, n = 2/2 5; Fig. 5E). In contrast, Plg activation was similar on WT, aM , 2/2 and aL lymphocytes (Fig. 5E). As with PMNs, peritoneal aM- deficient macrophages also exhibited severely reduced (by 75%) 2/2 Plg activation compared with the aL and WT macrophages (p , 0.05, n = 5; Fig. 5F). In control experiments, we examined aL expression on aM-deficient PMN, aM levels on aL-null, and control WT PMNs (both peritoneal and circulating) by flow cytometry. Neither aL nor aM deficiency altered expression of its counterpart b2-integrin on PMNs (data not shown), confirming that Plg recognition and activation is aMb2-specific. In addition, aMb2 functions as a Plg receptor, and this function is critical in Plm-dependent leukocyte recruitment to angiogenic sites.

aMb2 regulates secretion of VEGF-A by PMNs by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd.

Our data demonstrate a critical role of aMb2 in leukocyte re- cruitment to angiogenic sites. However, we considered a possibil- ity that aMb2 might also regulate other proangiogenic leukocyte functions such as production and secretion of angiogenic stim- ulators. The CD11b+/Gr-1+ cells, which primarily constitute PMNs, are of particular interest because they secrete high levels of MMP-9 and VEGF, leading to “angiogenic switch” in many FIGURE 6. aMb2 supports angiogenesis via regulation of VEGF se- tumors and to a failure of anti-VEGF therapies (5, 22). Therefore, 2/2 2/2 cretion by PMNs.(A) Peripheral blood WT, aM ,oraL PMNs (3 3 http://classic.jimmunol.org we compared the VEGF-A content of supernatants released from 6 10 cells/well) were incubated in 24-well TC plates in the absence or a 2/2 a 2/2 peripheral blood PMN of WT and M and L mice. In the presence of TNF-a (20 ng/ml) for 2 h at 37˚C. Cycloheximide (10 mg/ml) a absence of TNF- stimulation, VEGF-A levels were low and or pentoxifylline (300 mM) were added 60 min before addition of TNF-a. similar in PMNs of all three genotypes. Notably, upon stimulation VEGF concentration was measured in supernatants using mouse VEGF with TNF-a, VEGF-A content was substantially lower (by ∼60%) Quantikine ELISA Kit. Data are means 6 SEM of triplicate samples and 2/2 B in the aM PMN-conditioned medium compared with medium are representative of three independent experiments. ( ) Bright field mi- 2/2 ΑΕ Downloaded from harvested from WT or aL PMNs (p , 0.05, n = 6; Fig. 6A). croscopy of tube formation by WT M Cs in the presence of conditioned a 2/2 a 2/2 VEGF is stored in specific granules in human PMNs (24). To media collected from WT, M ,or L peripheral blood PMNs a determine whether a b regulates de novo synthesis of VEGF its stimulated with TNF (upper panels). Inhibitors of VEGF: neutralizing M 2 m secretion, or both, we used cycloheximide, an inhibitor of protein anti-VEGF mAb, isotype matched rat IgG2a (100 g/ml), and recombinant mouse sFLT-1 (100 ng/ml) were preincubated for 60 min with conditioned synthesis, and pentoxifylline, an inhibitor of PMN degranulation. media of WT TNF-a–stimulated PMNs before its addition to MAECs Cycloheximide did not affect VEGF content in supernatants of (lower panels). The images were taken after 6 h incubation in 37˚C, 5% TNF-a–stimulated PMNs of any mouse strains tested, suggesting CO2. Scale bars, 75 mm. (C) Quantification of tube formation. The number that TNF-a did not stimulate de novo VEGF synthesis (Fig. 6A). of closed tubes was counted in 20 different fields of each treatment This interpretation was corroborated by quantitative real-time and plotted as mean 6 SEM and are representative of two independent polymerase chain reaction assays, which revealed that VEGF experiments. mRNA levels were similar not only in resting and TNFa-treated 2/2 2/2 PMNs, but also in WT, aM , and aL PMNs (Table I). In (Fig. 6A). Taken together, these data suggest that aMb2 integrin contrast, pentoxifylline inhibited secretion of VEGF from PMNs does not regulate VEGF synthesis, but rather its secretion via and reduced its concentration by 80–85% in supernatants from control of PMN degranulation, which is consistent with prior data 2/2 2/2 WT and aL PMNs and by additional 20% from aM PMNs implicating aMb2 in degranulation of human PMNs ex vivo (25). 4720 INTEGRIN aMb2-DEPENDENT ANGIOGENESIS

a 2/2 2/2 Table I. Comparison of VEGF-A mRNA content in WT, M , and ated as compared with the aL and WT cells. These data were a 2/2 L peripheral blood PMNs consistent with the SPR sensorgrams showing that recombinant aMI-domain directly interacts with Plg, but the aLI-domain does a 2/2 a 2/2 Treatment WT M L not. These findings are in agreement with prior studies showing Untreated 0.22 0.19 0.2 that aMb2 recognizes urokinase (uPA) and Plg enhancing their TNF-a (20 ng/ml) 0.24 0.20 0.24 reciprocal activation on PMN surface (14, 15, 28). To our a b Values are expressed relative to GAPDH mRNA levels. PMNs were incubated in knowledge, this is the first report implicating M 2 in angiogen- the presence or absence of TNF-a (20 ng/ml) for 2h at 37˚C. Total RNA was isolated esis and demonstrating its intimate interplay with Plg in vivo. using Trizol reagent, and RT-PCR has been performed as described in Materials and b Methods. Although, the 2-deficient mice showed slowed angiogenesis in healing wounds (29), none of the individual b2 integrin family members was shown to contribute to this process. The implication of aMb2 in angiogenesis and Plg binding and activation is highly To corroborate this conclusion, we measured the concentration of specific as aLb2 did not show any impairment in these responses. lactoferrin, a marker of PMN-specific granules, in PMN super- This distinction may be explained, at least in part, by the relatively natants. The relative changes of lactoferrin and VEGF in the low sequence identity between the ligand binding aMI- and aLI- PMN-conditioned media were highly similar. Importantly, the a b 2/2 2/2 domains resulting in the M 2 promiscuity for many structurally aM PMNs, but not the aL PMNs, showed severely impaired unrelated ligands, whereas aLb2 shows a limited ligand repertoire release of lactoferrin into the supernatants of TNF-a–stimulated a b 2/2 with little overlap in ligand recognition with M 2 (19). Regarding aM PMNs, which were almost as low as in supernatants of the ligand repertoire, the two b2 integrins aXb2 and aDb2 are more resting PMNs from each mouse strain (Supplemental Fig. 2). We a b a b + similar to M 2 than to L 2, and it would be interesting to examine also sought to examine VEGF production and secretion in CD117 their roles in angiogenesis. The critical role of a b -dependent Plg a 2/2 M 2 progenitor cells isolated from bone marrow of WT, M , and activation in PMN and macrophage recruitment to angiogenic niches a 2/2 L mice. However, we failed to detect mRNA for VEGF in is in accord with previous studies demonstrating the importance of these immature cells. cell-bound plasmin in leukocyte recruitment in a variety of in vivo Next, we analyzed the capacity of WT MAECs to form tubes in models of inflammation (10–12) and with a crucial role of the Plg a 2/2 the presence of the conditioned media derived from WT, M ,or system in angiogenesis (30–36). Pericellular proteolysis is critical for a 2/2 a L TNF- –stimulated PMNs. MAECs in the presence of me- initiation of angiogenesis as evidenced by suppression of neo- a 2 /2 dia collected from WT or L PMNs formed well-organized vascularization in mice deficient in various proteases (31, 37–39) and tubelike networks. In contrast, tubes formed by MAECs in the by administration of a variety of protease inhibitors (40, 41). Among a 2/2 presence of the M PMN-conditioned media were incomplete the proteases implicated in angiogenesis, in addition to plasmin, are , a (p 0.05, n = 20; Fig. 6B, 6C). In control samples, TNF- alone its activators and (reviewed in Ref. 4, 8). Plasmin did not support tube formation by ECs. To confirm that tube is one of pro-MMP-9 activators (42) and PMN-derived MMP-9 is a 2/2 formation by MAECs in the presence of WT or L PMN responsible for angiogenic switch in some tumors (5). Consistent supernatants is VEGF-dependent, we added neutralizing rat anti- with the finding of decreased Plm activity and the capacity of aMb2 mouse VEGF mAb (clone 2G11-2A05), its isotype control rat by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. to bind and activate MMP-9 (43), we observed reduced MMP-9 IgG , or sFLT-1. The effectiveness of these VEGF inhibitors has 2/2 2a activity in tumor extracts from the aM mice (data not shown). been previously established (26, 27). These inhibitors of VEGF As a key source of proteases and proangiogenic factors, PMNs almost completely inhibited (by 75–80%) tube formation by and macrophages are essential for angiogenesis. Angiogenesis is MAEC (p , 0.05, n = 20). In contrast, the isotype control Ab did severely blunted in neutropenic mice (5, 29, 44, 45) or in mice in not have any effect (Fig. 6B, bottom panel, and 6C). Although other which macrophages have been eliminated (2). Robust PMN and proangiogenic factors are likely to be present in PMN supernatants, macrophage recruitment is observed into ischemic tissues in- VEGF appears to be the key stimulator of this process in our ex- cluding tumors (46), and in many tumors recruitment of these cells a 2/2 perimental system. Finally, supplementation of M PMN con- tumors correlates with poor host survival (reviewed in Ref. 2). http://classic.jimmunol.org ditioned medium with recombinant mouse VEGF to the same With such evidence, the virtual absence of PMNs and macro- 2/2 concentration as in medium from WT PMNs (520 pg/ml) enhanced phages in the angiogenic tissue of the aM mice provides the numbers of closed tubes by 2.5-fold and almost completely a mechanism to account for profound reduction in neovas- a b restored MAEC tube formation (Fig. 6B, top panel). M 2 does not cularization and tumor growth in these animals. In addition to 2/2 directly regulate VEGF-A de novo synthesis; however, it enhances defective recruitment, the aM PMNs also exhibited severely VEGF-A secretion from PMN intracellular stores via its enhance- attenuated secretion of VEGF-A because of impaired degranula-

Downloaded from ment of PMN degranulation. tion, and supernatants collected from these cells did not sup- port EC tube formation in in vitro assays. As enhanced VEGF is Discussion the hallmark and a key contributor to CD11b+/Gr-1+-dependent The goal of this study was to examine involvement of the two major (mostly PMNs) resistance of many tumors to VEGF targeting anti- leukocyte b2 integrins in angiogenesis: aMb2 and aLb2. Using the cancer therapies, we have focused our investigations on the reg- aM- and aL-deficient mice, we demonstrate that aMb2 promotes ulation of this pivotal cytokine. We cannot exclude that other angiogenesis in model melanoma and prostate tumors, as well as proangiogenic factors, particularly those stored in PMN granules, a b in Matrigel implants, whereas the L 2 integrin does not. Blood might also be regulated by aMb2 via its influence on degranula- 2/2 vessel formation and tumor growth were impaired in aM mice tion. This important issue is open for further investigation. Even if a 2/2 2/2 as compared with the L or WT mice. Impaired angiogenesis aM PMNs had been able to migrate, they would likely to have 2/2 in aM mice was due to dramatic reduction in recruitment of failed to promote angiogenesis because of an inability to supply PMNs and macrophages into the tumors and Matrigel implants. the requisite amounts of the major proangiogenic stimulus, VEGF. Furthermore, we showed that Plg binding and activation on the The neovasculature in growing tumors, but not preexisting blood 2/2 2/2 surface of aM PMNs and macrophages and their Plm- vessels, in aM mice showed immature and leaky phenotype. dependent invasion through Matrigel were significantly attenu- This might be caused by insufficient infiltration of CD11b+/Ly6G+ The Journal of Immunology 4721

PMNs known to support vascular maturation via elevated levels 21. van der Schaft, D. W., R. P. Dings, Q. G. de Lussanet, L. I. van Eijk, A. W. Nap, R. G. Beets-Tan, J. C. Bouma-Ter Steege, J. Wagstaff, K. H. Mayo, and of proangiogenic factors VEGF and MMP-9 (22). A. W. Griffioen. 2002. The designer anti-angiogenic peptide anginex targets Taken together, our studies demonstrate that integrin aMb2 tumor endothelial cells and inhibits tumor growth in animal models. FASEB J. promotes angiogenesis in vivo via a dual mechanism: first, as a Plg 16: 1991–1993. a b 22. Jablonska, J., S. Leschner, K. Westphal, S. Lienenklaus, and S. Weiss. 2010. receptor, M 2 supports Plm-dependent recruitment of myeloid Neutrophils responsive to endogenous IFN-beta regulate tumor angiogenesis and cells to angiogenic niches; second, aMb2 enhances VEGF-A se- growth in a mouse tumor model. J. Clin. Invest. 120: 1151–1164. b cretion by PMN degranulation. Based on our findings, selective 23. Rogers, C., E. R. Edelman, and D. I. Simon. 1998. A mAb to the 2-leukocyte a b integrin Mac-1 (CD11b/CD18) reduces intimal thickening after angioplasty or antagonists of M 2 can be considered as a new target to inhibit stent implantation in rabbits. Proc. Natl. Acad. Sci. USA 95: 10134–10139. tumor angiogenesis. 24. Gaudry, M., O. Bre´gerie, V. Andrieu, J. El Benna, M. A. Pocidalo, and J. Hakim. 1997. Intracellular pool of vascular endothelial growth factor in human neu- Acknowledgments trophils. Blood 90: 4153–4161. 25. Richter, J., J. Ng-Sikorski, I. Olsson, and T. Andersson. 1990. Tumor necrosis We thank Dr. Tatiana Byzova for expertise and advice regarding the in vivo factor-induced degranulation in adherent human neutrophils is dependent on angiogenesis models. CD11b/CD18-integrin-triggered oscillations of cytosolic free Ca2+. Proc. Natl. Acad. Sci. USA 87: 9472–9476. Disclosures 26. Basu, A., A. G. Contreras, D. Datta, E. Flynn, L. Zeng, H. T. Cohen, D. M. Briscoe, and S. Pal. 2008. Overexpression of vascular endothelial growth factor and the The authors have no financial conflicts of interest. development of post-transplantation cancer. Cancer Res. 68: 5689–5698. 27. Sossey-Alaoui, K., E. Pluskota, G. Davuluri, K. Bialkowska, M. Das, D. Szpak, D. J. Lindner, E. Downs-Kelly, C. L. Thompson, and E. F. Plow. 2014. Kindlin-3 References enhances breast cancer progression and metastasis by activating Twist-mediated 1. Tazzyman, S., H. Niaz, and C. Murdoch. 2013. Neutrophil-mediated tumour angiogenesis. FASEB J. 28: 2260–2271. angiogenesis: subversion of immune responses to promote tumour growth. 28. Lishko, V. K., V. V. Novokhatny, V. P. Yakubenko, H. V. Skomorovska-Prokvolit, Semin. Cancer Biol. 23: 149–158. and T. P. Ugarova. 2004. Characterization of plasminogen as an adhesive ligand a b 2. Qian, B.-Z., and J. W. Pollard. 2010. Macrophage diversity enhances tumor for integrins alphaMbeta2 (Mac-1) and 5 1 (VLA-5). Blood 104: 719–726. progression and metastasis. Cell 141: 39–51. 29. Schruefer, R., S. Sulyok, J. Schymeinsky, T. Peters, K. Scharffetter-Kochanek, 3. Kusumanto, Y. H., W. A. Dam, G. A. Hospers, C. Meijer, and N. H. Mulder. and B. Walzog. 2006. The proangiogenic capacity of polymorphonuclear neu- 2003. Platelets and granulocytes, in particular the neutrophils, form important trophils delineated by microarray technique and by measurement of neo- compartments for circulating vascular endothelial growth factor. Angiogenesis 6: vascularization in wounded skin of CD18-deficient mice. J. Vasc. Res. 43: 1–11. 283–287. 30. Oh, C.-W., J. Hoover-Plow, and E. F. Plow. 2003. The role of plasminogen in 4. Noonan, D. M., A. De Lerma Barbaro, N. Vannini, L. Mortara, and A. Albini. angiogenesis in vivo. J. Thromb. Haemost. 1: 1683–1687. 2008. Inflammation, inflammatory cells and angiogenesis: decisions and inde- 31. Heymans, S., A. Luttun, D. Nuyens, G. Theilmeier, E. Creemers, L. Moons, cisions. Cancer Metastasis Rev. 27: 31–40. G. D. Dyspersin, J. P. M. Cleutjens, M. Shipley, A. Angellilo, et al. 1999. In- 5. Nozawa, H., C. Chiu, and D. Hanahan. 2006. Infiltrating neutrophils mediate the hibition of plasminogen activators or matrix metalloproteinases prevents cardiac initial angiogenic switch in a mouse model of multistage carcinogenesis. Proc. rupture but impairs therapeutic angiogenesis and causes cardiac failure. Nat. Natl. Acad. Sci. USA 103: 12493–12498. Med. 5: 1135–1142. 6. van Hinsbergh, V. W. M., M. A. Engelse, and P. H. A. Quax. 2006. Pericellular 32. Vogten, J. M., A. Reijerkerk, J. C. Meijers, E. E. Voest, I. H. Borel Rinkes, and proteases in angiogenesis and vasculogenesis. Arterioscler. Thromb. Vasc. Biol. M. F. Gebbink. 2003. The role of the fibrinolytic system in corneal angiogenesis. 26: 716–728. Angiogenesis 6: 311–316. 7. Lee, S., M. Zheng, B. Kim, and B. T. Rouse. 2002. Role of matrix 33. Drixler, T. A., J. M. Vogten, M. F. Gebbink, P. Carmeliet, E. E. Voest, and -9 in angiogenesis caused by ocular infection with herpes I. H. Borel Rinkes. 2003. Plasminogen mediates liver regeneration and angio- simplex virus. J. Clin. Invest. 110: 1105–1111. genesis after experimental partial hepatectomy. Br. J. Surg. 90: 1384–1390. 8. Shamamian, P., J. D. Schwartz, B. J. Pocock, S. Monea, D. Whiting, 34. Creemers, E., J. Cleutjens, J. Smits, S. Heymans, L. Moons, D. Collen, M. Daemen, S. G. Marcus, and P. Mignatti. 2001. Activation of progelatinase A (MMP-2) by and P. Carmeliet. 2000. Disruption of the plasminogen gene in mice abolishes wound by guest on October 1, 2021. Copyright 2014 Pageant Media Ltd. neutrophil elastase, cathepsin G, and proteinase-3: a role for inflammatory cells healing after myocardial infarction. Am. J. Pathol. 156: 1865–1873. in tumor invasion and angiogenesis. J. Cell. Physiol. 189: 197–206. 35. Ling, Q., A. T. Jacovina, A. Deora, M. Febbraio, R. Simantov, R. L. Silverstein, 9. Das, R., E. Pluskota, and E. F. Plow. 2010. Plasminogen and its receptors as B. Hempstead, W. H. Mark, and K. A. Hajjar. 2004. II regulates fibrin regulators of cardiovascular inflammatory responses. Trends Cardiovasc. Med. homeostasis and neoangiogenesis in vivo. J. Clin. Invest. 113: 38–48. 20: 120–124. 36. Chen, P. K., B. I. Chang, C. H. Kuo, P. S. Chen, C. F. Cho, C. F. Chang, 10. Ploplis, V. A., E. L. French, P. Carmeliet, D. Collen, and E. F. Plow. 1998. G. Y. Shi, and H. L. Wu. 2013. Thrombomodulin functions as a plasminogen Plasminogen deficiency differentially affects recruitment of inflammatory cell receptor to modulate angiogenesis. FASEB J. 27: 4520–4531. populations in mice. Blood 91: 2005–2009. 37. Johnson, C., H. J. Sung, S. M. Lessner, M. E. Fini, and Z. S. Galis. 2004. Matrix 11. Busuttil, S. J., V. A. Ploplis, F. J. Castellino, L. Tang, J. W. Eaton, and E. F. Plow. metalloproteinase-9 is required for adequate angiogenic revascularization of 2004. A central role for plasminogen in the inflammatory response to bio- ischemic tissues: potential role in capillary branching. Circ. Res. 94: 262–268. materials. J. Thromb. Haemost. 2: 1798–1805. 38. Masson, V., L. R. de la Ballina, C. Munaut, B. Wielockx, M. Jost, C. Maillard, 12. Swaisgood, C. M., M. A. Aronica, S. Swaidani, and E. F. Plow. 2007. Plas- S. Blacher, K. Bajou, T. Itoh, S. Itohara, et al. 2005. Contribution of host MMP-2 http://classic.jimmunol.org minogen is an important regulator in the pathogenesis of a murine model of and MMP-9 to promote tumor vascularization and invasion of malignant kera- asthma. Am. J. Respir. Crit. Care Med. 176: 333–342. tinocytes. FASEB J. 19: 234–236. 13. Tan, S. M. 2012. The leucocyte b2 (CD18) integrins: the structure, functional 39. Ahn, G. O., and J. M. Brown. 2008. Matrix metalloproteinase-9 is required for regulation and signalling properties. Biosci. Rep. 32: 241–269. tumor vasculogenesis but not for angiogenesis: role of bone marrow-derived 14. Pluskota, E., D. A. Soloviev, and E. F. Plow. 2003. Convergence of the adhesive myelomonocytic cells. Cancer Cell 13: 193–205. and fibrinolytic systems: recognition of urokinase by integrin a Mbeta 2 as well 40. Pepper, M. S. 2001. Role of the matrix metalloproteinase and plasminogen as by the urokinase receptor regulates cell adhesion and migration. Blood 101: activator-plasmin systems in angiogenesis. Arterioscler. Thromb. Vasc. Biol. 21: 1582–1590. 1104–1117. 41. Stefanidakis, M., and E. Koivunen. 2006. Cell-surface association between Downloaded from 15. Pluskota, E., D. A. Soloviev, K. Bdeir, D. B. Cines, and E. F. Plow. 2004. Integrin alphaMbeta2 orchestrates and accelerates plasminogen activation and matrix metalloproteinases and integrins: role of the complexes in leukocyte fibrinolysis by neutrophils. J. Biol. Chem. 279: 18063–18072. migration and cancer progression. Blood 108: 1441–1450. 16. Lu, H., C. W. Smith, J. Perrard, D. Bullard, L. Tang, S. B. Shappell, 42. Gong, Y., E. Hart, A. Shchurin, and J. Hoover-Plow. 2008. Inflammatory mac- M. L. Entman, A. L. Beaudet, and C. M. Ballantyne. 1997. LFA-1 is sufficient in rophage migration requires MMP-9 activation by plasminogen in mice. J. Clin. mediating neutrophil emigration in Mac-1-deficient mice. J. Clin. Invest. 99: Invest. 118: 3012–3024. 1340–1350. 43. Stefanidakis, M., T. Ruohtula, N. Borregaard, C. G. Gahmberg, and E. Koivunen. 17. Passaniti, A., R. M. Taylor, R. Pili, Y. Guo, P. V. Long, J. A. Haney, R. R. Pauly, 2004. Intracellular and cell surface localization of a complex between D. S. Grant, and G. R. Martin. 1992. A simple, quantitative method for assessing alphaMbeta2 integrin and promatrix metalloproteinase-9 progelatinase in neu- angiogenesis and antiangiogenic agents using reconstituted basement membrane, trophils. J. Immunol. 172: 7060–7068. heparin, and fibroblast growth factor. Lab. Invest. 67: 519–528. 44. Benelli, R., M. Morini, F. Carrozzino, N. Ferrari, S. Minghelli, L. Santi, 18. Cotter, M. J., K. E. Norman, P. G. Hellewell, and V. C. Ridger. 2001. A novel M. Cassatella, D. M. Noonan, and A. Albini. 2002. Neutrophils as a key cellular method for isolation of neutrophils from murine blood using negative immu- target for angiostatin: implications for regulation of angiogenesis and inflam- nomagnetic separation. Am. J. Pathol. 159: 473–481. mation. FASEB J. 16: 267–269. 19. Ustinov, V. A., and E. F. Plow. 2005. Identity of the amino acid residues involved 45. Shaw, J. P., N. Chuang, H. Yee, and P. Shamamian. 2003. Polymorphonuclear neu- in C3bi binding to the I-domain supports a mosaic model to explain the broad trophils promote rFGF-2-induced angiogenesis in vivo. J. Surg. Res. 109: 37–42. ligand repertoire of integrin alpha M beta 2. Biochemistry 44: 4357–4364. 46. Cursiefen, C., L. Chen, L. P. Borges, D. Jackson, J. Cao, C. Radziejewski, 20. Huang, X., T. Raskovalova, A. Lokshin, A. Krasinskas, J. Devlin, S. Watkins, P. A. D’Amore, M. R. Dana, S. J. Wiegand, and J. W. Streilein. 2004. VEGF-A S. F. Wolf, and E. Gorelik. 2005. Combined antiangiogenic and immune therapy stimulates lymphangiogenesis and hemangiogenesis in inflammatory neo- of prostate cancer. Angiogenesis 8: 13–23. vascularization via macrophage recruitment. J. Clin. Invest. 113: 1040–1050.