DCZ19931, a novel multi-targeting kinase inhibitor, inhibits ocular neovascularization

Abstract

Neovascularization is a prominent cause of irreversible blindness in a variety of ocular diseases. Current therapies for pathological neovascularization are concentrated on the suppression of vascular endothelial growth factors (VEGF). Despite the remarkable efficacy of anti-VEGF drugs, several problems still exist, including ocular complications and drug resistance. Thus, it is still required to design novel drugs for anti-angiogenic treatment. This study aimed to investigate the anti-angiogenic effects of a small molecule multi-target tyrosine kinase inhibitor, DCZ19931, on ocular neovascularization. The results showed that administration of DCZ19931 at the tested concentrations did not cause obvious cytotoxicity and tissue toxicity. DCZ19931 could reduce the size of choroidal neovascularization (CNV) lesions in laser-induced CNV model and suppress ocular neovascularization in oxygen-induced retinopathy (OIR) model. DCZ19931 could suppress VEGF-induced proliferation, migration, and tube formation ability of endothelial cells, exhibiting similar anti-angiogenic effects as Ranibizumab. DCZ19931 could reduce the levels of intercellular cell adhesion molecule-1 (ICAM-1) expression in vivo and in vitro. Network pharmacology prediction and western blots revealed that DCZ19931 exerted its anti-angiogenic effects through the inactivation of ERK1/2-MAPK signaling and p38-MAPK signaling. In conclusion, this study indicates that DCZ19931 is a promising drug for anti-angiogenic therapy for ocular diseases.

Figure 1

Synthesis of DCZ19931.

Figure 2

DCZ19931 has no obvious cytotoxicity and tissue toxicity. (A–C) HUVECs were incubated with the test concentrations of DCZ19931 (1 nM to 10 μM) or left untreated (Ctrl) for 24 h. Cell viability was measured by MTT assays (A, n = 4). Flow cytometry using Annexin V-FITC/PI double staining (B, n = 4) and Calcein-AM/PI staining (C, n = 4, scale bar, 20 μm) was performed to detect the apoptosis of HUVECs. (D,E) C57BL/6J mice received intravitreal injections of PBS (2 μL, Ctrl), 10% DMSO (2 μL), or DCZ19931 (2 μL, 1 μg/μL) for seven days. H&E staining (D, n = 5 animals per group, scale bar, 50 μm) and TUNEL staining assays were performed to detect retinal histological changes and retinal cell apoptosis (E, n = 5 animals per group, scale bar, 50 μm). DNase I was detected as a positive control in TUNEL staining assays. Statistical significance was evaluated by one-way ANOVA followed by Bonferroni post hoc test. *P < 0.05 indicated significant difference between the marked groups.

Figure 3

DCZ19931 inhibits ocular neovascularization in vivo. (A,B) C57BL/6J mice received intravitreal injections of 10% DMSO (2 μL, Ctrl), DCZ19931 (2 μL, 1 μg/μL), Ranibizumab (2 μL, 10 mg/mL), Ranibizumab (1 μL, 10 mg/mL) plus DCZ19931 (1 μL, 1 μg/μL) immediately after laser injury. The thickness and areas of neovascular were determined by H & E staining (A, n = 5, scale bar, 50 μm). CNV lesions were visualized by IB4 staining (B, n = 5, scale bar, 100 μm). (C) OIR mouse pups at P12 received intravitreal injections of 10% DMSO (1 μL, Ctrl), DCZ19931 (1 μL, 1 μg/μL), Ranibizumab (1 μL, 10 mg/mL), 1 μL of Ranibizumab (10 mg/mL) plus DCZ19931 (1 μg/μL), respectively. The retinas were harvested on P17 and then stained with GS-IB4 staining to observe retinal vessels. White dashed lines highlighted avascular areas; Yellow area indicated angiogenic regions (n = 5, scale bar, 200 μm). Statistical significance was evaluated by one-way ANOVA followed by Bonferroni post hoc test. *P < 0.05 indicated significant difference between the marked groups.

Figure 4

DCZ19931 exhibits anti-angiogenic effects in vitro. (A) HUVECs were pre-treated with VEGF (10 ng/mL) for 12 h and then treated with DCZ19931 at the concentration of 10 nM to 1 μM for 24 h. Cell viability was determined by MTT assay (n = 4). (B–D) HUVECs were pre-treated with VEGF (10 ng/mL) for 12 h and then treated with DCZ19931 (500 nM), Ranibizumab (250 μg/mL), DCZ19931 (500 nM) plus Ranibizumab (250 μg/mL), or left untreated (VEGF) for 24 h. The group without VEGF treatment was taken as Ctrl group. Cell proliferation was detected by EdU assays (B, n = 4, scale bar, 20 μm). The cells were allowed to migrate for 12 h at 37 °C and 5% CO₂ and cell migration ability was assessed by transwell assays (C, n = 4, scale bar, 20 μm). HUVECs were seeded on matrigel matrix and cultured. Tube formation ability was observed under a light microscope at 8 h after seeding (D, n = 4, scale bar, 100 μm). (E) The mouse choroidal sprouting assays were conducted to measure the angiogenic potency of choroidal explants at day 4, day 5, and day 6 after culture. Representative images of the choroidal sprouting areas at indicated time points were shown (n = 4, scale bar, 200 µm). Statistical significance was evaluated by one-way ANOVA followed by Bonferroni post hoc test. For (A–D): *P < 0.05 indicated significant difference between the marked groups. For (E): *P < 0.05 vs. VEGF group, ^#P < 0.05 vs. DCZ19931 + Ranibizumab group.

Figure 5

DCZ19931 inhibits vascular permeability via downregulation of ICAM-1 expression. (A) HUVECs were cultured with VEGF (Ctrl, 10 ng/mL), VEGF (10 ng/mL) plus DCZ19931 (500 nM), VEGF (10 ng/mL) plus Ranibizumab (250 μg/mL), VEGF (10 ng/mL) plus DCZ19931 (500 nM) and Ranibizumab (250 μg/mL) for 24 h. qRT-PCR assays were performed to detect the expression of ICAM-1 mRNA in HUVECs (n = 4). (B) HUVECs were cultured with VEGF (Ctrl, 10 ng/mL), VEGF (10 ng/mL) plus DCZ19931 (500 nM), VEGF (10 ng/mL) plus Ranibizumab (250 μg/mL), VEGF (10 ng/mL) plus DCZ19931 (500 nM) and Ranibizumab (250 μg/mL), or left untreated (WT) for 24 h. WT indicated normal HUVECs without VEGF induction. EB-transwell assays were conducted to detect cell permeability (n = 4). (C,D) Laser-induced CNV mice received intravitreal injections of 10% DMSO (2 μL, Ctrl), DCZ19931 (2 μL, 1 μg/μL), Ranibizumab (2 μL, 10 mg/mL), Ranibizumab (1 μL, 10 mg/mL) plus DCZ19931 (1 μL, 1 μg/μL) after laser injury. WT indicated normal mice without CNV induction. qRT-PCR assays were performed to detect the expression of ICAM-1 mRNA in choroidal tissues at day 7 after treatment (C, n = 5). Western blots were performed to determine ICAM-1 expression in choroidal tissues. GAPDH was used as the internal control. Representative immunoblots along with the quantitative results were shown (D, n = 5). (E) The pups of OIR mice at P12 received intravitreal injections of 10% DMSO (1 μL, Ctrl), DCZ19931 (1 μL, 1 μg/μL), Ranibizumab (1 μL, 10 mg/mL), 1 μL of Ranibizumab (10 mg/mL) plus DCZ19931 (1 μg/μL) mixture. ICAM-1 immunofluorescence assays were conducted to determine ICAM-1 expression in OIR model. The representative images with the quantitative results were shown (n = 5, scale bar, 50 μm; nuclei, blue; ICAM-1-positive cells, green). Statistical significance was determined by one-way ANOVA followed by Bonferroni post hoc test. *P < 0.05 indicated significant difference between the marked groups.

Figure 6

DCZ19931 inhibits ocular neovascularization through p38-MAPK and ERK1/2-MAPK signaling. (A) The drug-target-disease network diagram of DCZ19931 involved in ocular neovascularization. (B) PPI network analysis. (C) GO enrichment analysis. (D) KEGG pathways enrichment analysis. (E) HUVECs were treated with DCZ19931 (500 nM), Ranibizumab (250 μg/mL), DCZ19931 (500 nM) plus Ranibizumab (250 μg/mL), or left untreated (Ctrl). Then, HUVECs were stimulated with or without VEGF (50 ng/mL) for 30 min. The expression levels of total ERK1/2, p38, JNK, p-ERK1/2, p-p38, and p-JNK were detected by western blots (n = 4). (F) HUVECs were treated with DCZ19931 (500 nM) or SB203580 (5 μmol/mL) plus U0126 (5 μmol/mL) for 24 h, and then stimulated with VEGF (50 ng/mL) for 30 min. The group without any treatment was taken as the control (Ctrl). The phosphorylated levels of ERK1/2 and p38 proteins were detected by western blots (n = 4). GAPDH was detected as the internal control. Representative immunoblots along with the densitometric quantitative results were shown. Statistical significance was determined by one-way ANOVA followed by Bonferroni post hoc test. *P < 0.05 indicated significant difference between the marked groups.