Soluble vascular endothelial growth factor decoy receptor FP3 exerts potent antiangiogenic effects

Abstract

The binding of vascular endothelial growth factor (VEGF) to its receptors stimulates tumor growth; therefore, modulation of VEGF would be a viable approach for antiangiogenic therapy. We constructed a series of soluble decoy receptors containing different VEGF receptor 1 (FLT1) and VEGF receptor 2 (KDR) extracellular domains fused with the Fc region of human immunoglobulin (Ig) and evaluated their antiangiogenic effects and antitumor effects. Results of in vitro binding and cell proliferation assays revealed that decoy receptor FP3 had the highest affinity to VEGF-A and -B. Compared with bevacizumab, FP3 more effectively inhibited human umbilical vein endothelial cell (HUVEC) migration and vessel sprouting from rat aortic rings. FP3 significantly reduced phosphorylation of AKT and ERK1/2, critical proteins in the VEGF-mediated survival pathway in endothelial cells. Moreover, FP3 inhibited tumor growth in human hepatocellular carcinoma (HepG2), breast cancer (MCF-7), and colorectal cancer (LoVo) tumor models, and reduced microvessel density in tumor tissues. The FP3-mediated inhibition of tumor growth was significantly higher than that of bevacizumab at the same dose. FP3 also demonstrated synergistic antitumor effects when combined with 5-fluorouracil (5-FU). Taken together, FP3 shows a high affinity for VEGF and produced antiangiogenic effects, suggesting its potential for treating angiogenesis-related diseases such as cancer.

Figure 1

Schematic representation and binding affinities of chimeric decoy receptors FP1 through FP6. (a) The fusion proteins were constructed by fusing extracellular domains of vascular endothelial growth factor (VEGF) receptors VEGF receptor 1 (FLT1) and VEGF receptor 2 (KDR) with the Fc region of human immunoglobulin G1 (IgG1). R1 represents FLT1; R2, KDR; D1–5, Ig-like extracellular domains 1–5; Fc, human IgG1 constant region. (b) The binding affinities of the chimeric decoy receptors to VEGF₁₆₅ were determined. Proteins (FP1–FP6; 0.05–1,500 pmol/l) were incubated with immobilized VEGF₁₆₅ (10 pmol/l), and binding was determined by optical density at 450 nm (OD₄₅₀). (c) The inhibitory effects of chimeric proteins or bevacizumab on human umbilical vein endothelial cell (HUVEC) proliferation. Inhibition was determined by incubating the FP1, FP3, or bevacizumab (35–700 nmol/l) with VEGF₁₆₅ (0.3 nmol/l) for 3 days in endothelial basal growth medium plus 1% fetal bovine serum. Cell viability was determined using a colorimetric assay and presented as OD₄₅₀ value. Assays were performed in triplicate and three independent experiments were conducted. **P < 0.01, ***P < 0.001 compared with FP1.

Figure 2

Effects of FP3 on VEGF₁₆₅-induced human umbilical vein endothelial cell (HUVEC). (a) Cell migration was studied using a modified transwell migration chamber. HUVECs were treated with control immunoglobulin G (IgG) (35 nmol/l) or various concentrations of bevacizumab, FP1, or FP3 in the presence of VEGF₁₆₅ (0.3 nmol/l). After 4-hour incubation at 37 °C, cells that had migrated to the lower filter surface were counted (original magnification ×200). Migration was evaluated relative to untreated cells (100%). Results are reported as the mean ± SEM of 10 independent high-power fields/well. Assays were performed in triplicate and three independent experiments were conducted. *P < 0.05, **P < 0.01, ***P < 0.001 compared with FP1. (b) The effects of bevacizumab, FP1, or FP3 on capillary tube formation (in vitro differentiation). HUVEC seeded on Matrigel were stimulated with VEGF₁₆₅ (0.3 nmol/l) and treated with FP1, FP3, or bevacizumab (350 nmol/l). After 20-hour incubation, cell morphology was evaluated. Representative images show organized lumen-containing structures in cells stimulated with VEGF₁₆₅ alone or human IgG plus VEGF₁₆₅ (controls) (original magnification ×100). (c) Area covered by the endothelial cell tube network. Results are expressed as mean ± SEM (n = 3). **P < 0.01, FP3 compared with FP1 or bevacizumab. All assays were conducted six times. (d) Aortic rings were embedded in Matrigel in 24-well plates and incubated for 6 days with VEGF₁₆₅ (1.2 nmol/l), and PBS, bevacizumab (700 or 1,400 nmol/l), or FP3 (700 or 1,400 nmol/l). Tissues were then fixed, stained, and photographed (original magnification ×40). (e) Quantitative analysis of microvessel sprouting. Microvessels were counted in high-power fields (×200), and aortic rings were assigned a rating of 0 (no sprouts) to 5 (profuse sprouting). Results are expressed as mean ± SEM. **P < 0.01 compared with VEGF₁₆₅; *P < 0.05, FP3 (700 nmol/l) compared with bevacizumab (700 nmol/l). VEGF, vascular endothelial growth factor.

Figure 3

Effect of FP3 on ERK and AKT phosphorylation in human umbilical vein endothelial cells (HUVECs). (a) HUVECs were pretreated with bevacizumab, FP1, or FP3 (3 nmol/l) for 30 minutes, followed by treatment with vascular endothelial growth factor (VEGF) (10 ng/ml) for 10 minutes. Cells were harvested and analyzed by Western blot using antibodies specific to p-Akt, Akt, p-ERK1/2, or ERK1/2. β-Actin was used as a loading control. Results are representative of at least three independent experiments. (b) The bar graph is the ratio of phosphorylated AKT (pAKT) to total AKT. (c) The bar graph is the ratio of phosphorylated ERK (pERK) to total ERK.

Figure 4

Effect of systemic administration of FP3 in xenograft tumor models. (a) Hepatocellular carcinoma HepG2 xenograft tumors after systemic administration of FP3. On day 2 following tumor cell implantation, mice were treated with phosphate-buffered saline (PBS, control), FP3 (4, 10, or 25 mg/kg), or bevacizumab (10 mg/kg) twice weekly for 7 weeks. Results are expressed as mean ± SEM. *P < 0.05, **P < 0.01 versus PBS. (b) Breast cancer MCF-7 xenograft tumors following systemic administration of FP3. When tumors reached the mean volume of 80–100 mm³, mice were treated with PBS, FP3 (2, 6, or 18 mg/kg), or bevacizumab (6 mg/kg) twice weekly for 5 weeks. Tumor sizes were determined weekly. Results are expressed as mean ± SEM (each group, n = 8). *P < 0.05, **P < 0.01 versus PBS. (c). Antitumor efficacy of FP1 or FP3 in LoVo xenografts. Tumors were treated with PBS, FP1 (10 mg/kg), or FP3 (10 mg/kg) twice weekly for 7 weeks. Tumor sizes were monitored once every week. Data shown are means ± SEM (each group, n = 8). *P < 0.05 versus FP1.

Figure 5

Effect of FP3 combined with 5-fluorouracil (5-FU) in colorectal cancer LoVo xenograft tumor model. (a) Mice were randomized to receive PBS, FP3, 5-FU, FP3 combined with 5-FU. FP3 (6 mg/kg) was administered twice weekly and 5-FU (10 mg/kg) once per week by intravenous injection for 7 weeks. Results are expressed as mean ± SEM. *P < 0.05 compared with FP3 alone, **P < 0.01 compared with 5-FU alone. (b) Immunohistochemical analysis. Intratumoral vascularization was assessed by CD31 immunostaining 5 weeks after treatment. Tissues were counterstained with H&E (original magnification ×40, ×100). (c) Microvessel density. Mean microvessel density was determined as CD31⁺ cells/field. Results are expressed as mean ± SEM (n = 3). *P < 0.05 versus PBS or 5-FU; n.s., not significant. (d) Representative tumor sections. Tumors were stained with H&E (original magnification ×100), and apoptosis was evaluated by terminal uridine deoxynucleotidyl transferase dUTP nick end labeling staining (original magnification ×200).