The characterization of a novel monoclonal antibody against CD93 unveils a new antiangiogenic target

Abstract

The inhibition of tumor angiogenesis is one of the main challenges in cancer therapy. With the aim of developing monoclonal antibodies able to inhibit angiogenesis, we immunized mice with proliferating human umbilical vein endothelial cells. We generated a library of monoclonal antibodies able to recognize antigens expressed on endothelial cells and screened the antibodies for their ability to inhibit endothelial cell proliferation, migration, and sprouting in vitro. Here, we show that the antibody, designated as 4E1, is able to neutralize the formation of new vessels both in vitro and in vivo without affecting endothelial cell survival. By mass spectrometry we identified CD93 as the antigen bound by 4E1 and mapped the recognized epitope. CD93 is a transmembrane protein heavily glycosylated preferentially expressed in the vascular endothelium. CD93 silencing by lentiviral-mediated small hairpin RNA expression impairs human endothelial cell proliferation, migration, and sprouting. Altogether these findings reveal 4E1 as a novel antiangiogenic antibody and identify CD93 as a new target suitable for antiangiogenic therapy.

Figure 1. The mAb 4E1 affects cell proliferation, migration, and in vitro sprouting of human endothelial cells

A: Cell proliferation expressed as thymidine uptake in HUVEC. Cells were grown in 96-well-plates, serum starved, and induced to proliferate with complete medium (induct.) in the presence of different concentrations of 4E1 or unrelated purified antibodies (NC, 500 nM). B: Migration assay on HUVEC in the presence of 4E1 (500 nM) or unrelated purified antibodies (NC, 500 nM). Cells were grown in growth factor-depleted culture medium and plated in Boyden chambers. Chemotaxis was stimulated with 10 ng/ml VEGF (VEGF) or complete medium (M199). Migratory cells were stained and counted under a light microscope. C: Sprouting of HUVEC spheroids embedded into collagen gels in the absence (NT) or presence of 10 ng/ml VEGF (VEGF). NC, unrelated purified antibodies (500 nM), mAb 4E1 (500 nM). A representative experiment is shown (original magnification, x40). Data represent the ± SD of three-five independent experiments each in triplicate.

Fugure 2. The mAb 4E1 impairs angiogenesis in vivo

A: HUVEC were grown on Matrigel in the presence of 4E1 (500 nM) or unrelated antibodies (NC, 500 nM) and the formation of capillary-like structures was observed 20 h after seeding. A representative experiment is shown (original magnification, x100). B: Quantification of tube length was performed based on the results shown in panel A. Results were expressed as means ± SD of four different fields randomly chosen from each group from three independent experiments. C: Nude mice were injected subcutaneously with Matrigel containing HUVEC in the presence of 4E1 or unrelated antibodies (NC). Matrigel plugs were excised, sectioned and subjected to immunofluorescence analysis. Representative sections, stained with anti-von Willebrand factor antibodies, are shown. Scale bar represents 150 μm. D: Quantification of microvessel density (MVD) was performed based on the results shown in panel C as reported in Materials and methods. E: Quantification of vessel length was performed based on the results shown in panel C as reported in Materials and methods. At least ten sections/plug were examined. Data represent the ± SD of three independent experiments.

Figure 3. The mAb 4E1 recognizes the membrane protein CD93 and does not affect endothelial cell survival

A: Mouse BALB/c fibroblasts were transfected with a construct expressing the human chimeric protein CD93-YFP under the control of a constitutive promoter. Cells were plated on glass coverslips, fixed and subjected to immunofluorescence analysis in the absence of permeabilization by using the mAb 4E1. Scale bar represents 8 μm. B: Vein section through human umbilical cord stained by immunofluorescence using anti-von Willebrand factor (von Will) and 4E1 antibodies. The merge of the double staining and a DIC image of the same cross section are shown. Scale bar, 22 μm. C: The effect of 4E1 treatment on survival of growth-arrested endothelial cells. HUVEC were grown to confluence and treated with 500 nM of 4E1 or unrelated antibodies (NC) for 48 h. The apoptotic percentage was calculated by flow cytometric analysis of nuclei stained with propidium iodide. D: Cell viability assay on proliferating (prolif.) and quiescent (confl.) endothelial cells treated with 500 nM 4E1 or unrelated antibodies (NC) for 96 h. The optical density values were expressed as relative cell viability. All data represent the ± SD of three independent experiments.

Figure 4. 4E1 recognizes a conformational epitope localized on a region overlapping D1 and DX domains of CD93

A: Human Lenti-X 293T cells were transiently transfected with empty vector (mock) or a construct expressing human CD93 full-length. Cell extracts were immunoprecipitated with 4E1 and immunoprecipitates were analyzed by Western blotting with anti-CD93 antibodies (H190). Equal loading of cell lysates was confirmed by using anti-β-actin antibodies. B: The schematic diagram illustrates the deletion mutants of the CD93 extracellular domains fused with a 6X Myc tag. The deletion mutants were designated in accordance with Wu and colleagues [15]. LS, signal peptide; CTLD, C-type lectin-like domain; X, domain of unknown structural function; EGF-like, EGF-like repeats; Mucin mucin-like domain; Myc, 6X Myc tag. C: Lenti-X 293T cells were transiently transfected to generate recombinant CD93 domain proteins fused with a 6X Myc tag. For each different deletion mutant two different cDNA constructs were transfected (#1 and #2). Cell extracts were analyzed by immunoblotting using anti-Myc antibodies. Anti-β-actin antibodies were used to confirm equal loading. D: Cell lysates from cells transfected as in C, were subjected to immunoprecipitation analyses by using 4E1. Immunoprecipitates were analyzed by Western blotting with anti-Myc antibodies.

Figure 5. The mAb 4E1 impairs endothelial cell adhesion on different substrates

Gelatin (gel), laminin (lam) and collagen (coll) adhesive substrates were used for coating. HUVEC, grown in complete medium, were biochemically detached, incubated in the presence of different concentrations of 4E1 (50, 250, and 500 nM) or unrelated antibodies (500 nM, NC) for 15 min in ice and allowed to adhere on substrate-coated surfaces for 10 min. Fixed cells were stained with crystal violet solution, and the optical density values were expressed as relative cell adhesion. Data represent the ± SD of four independent experiments each in triplicate.

Figure 6. CD93 silencing affects cell proliferation, adhesion, migration, and in vitro sprouting of human endothelial cells

HUVEC were infected with a lentiviral vector expressing unrelated (unr) or CD93 shRNAs (clone 85 or 86). A: Cell extracts from shRNA expressing HUVEC were analyzed by Western blotting using anti-CD93 (H190) and anti-β-actin antibodies to confirm equal loading. B: Cell proliferation expressed as thymidine uptake in infected HUVEC. Cells were grown in 96-well-plates, serum starved (starv) and induced with complete medium (induct). Not infected endothelial cells were also analyzed (-). C: Cell adhesion assay of infected HUVEC. Cells were biochemically detached and allowed to adhere on gelatin-coated surfaces for 15 min. Fixed cells were stained with crystal violet solution and the optical density values were expressed as relative cell adhesion. D: Migration assay on infected HUVEC. Cells were grown in growth factor-depleted culture medium and plated in Boyden chambers. Chemotaxis was stimulated with 10 ng/ml VEGF. Migratory cells were stained and counted under a light microscope. E: Sprouting assay of infected endothelial cells embedded into collagen gels in the absence (NT) or presence of 10 ng/ml VEGF (VEGF). A representative experiment is shown (original magnification, x40). Data represent the ± SD of four-five independent experiments each in triplicate.