Synthesis and activity of Ref-1 inhibitors. (A) Synthetic scheme for APX2009 (6a) and APX2014 (6b). Structure of APX3330 (7) included for reference. Reagents and conditions: a, 2-iodo-3-hydroxy-1,4-naphthoquinone (iodolawsone, 1), 2-propylacrylic acid (2), K₂CO₃, Pd(OAc)₂, argon, 100°C, 1 hour, 74%; b, (COCl)₂, DMF, DCM, RT overnight, 100%; c, DEA·HCl (APX2009) or CH₃ONH₂·HCl (APX2014), DIPA·HCl, RT, 45 minutes, 62% and 71% respectively; d, NaOCH₃/CH₃OH, argon, 30 minutes, RT, 96% and 86%, respectively. See Supplemental Methods for full synthetic details. (B) APX2009 and APX2014 are more effective inhibitors of Ref-1-induced AP-1 DNA binding than APX3330 in an EMSA. Two separate gels from the same experiment are shown. The IC₅₀ for redox EMSA inhibition was 25, 0.45, and 0.2 μM for APX3330, APX2009, and APX2014, respectively. These assays were performed multiple times with similar results.

Compounds APX2009 and APX2014 inhibit endothelial cell proliferation in HRECs and Rf/6a cells in vitro. Dose-dependent effects of APX2009 (A) and APX2014 (B) in HRECs, and dose-dependent effects of APX2009 (C) and APX2014 (D) in Rf/6a choroidal endothelial cells. In vitro proliferation was measured using an alamarBlue assay. GI₅₀ values are indicated. Mean ± S.E.M., n = 3 per dose.

Compounds APX2009 and APX2014 inhibit S phase in HRECs. After HRECs were treated with the indicated concentrations of APX2009 and APX2014, (A) EdU (red) and (C) Ki-67 (green) were detected and nuclei (blue) stained with DAPI; Scale bars, 100 μm. (B) Quantification of EdU and (D) quantification of Ki-67 in HRECs. Mean ± S.E.M., n = 3 fields per dose. **P < 0.01; ****P < 0.0001 compared with DMSO control (one-way ANOVA with Dunnett’s post-hoc test). Representative data from three independent experiments. See Supplemental Figs. 1A and 2.

Compounds APX2009 and APX2014 inhibit endothelial cell migration in HRECs and Rf/6a cells in vitro. (A) Effect of APX2009 and APX2014 on cell migration in HRECs. A confluent monolayer of HRECs with various treatments (highest doses shown) was wounded and wound closure was monitored for 8 hours. (B) Quantitative analysis of cell migration shows that APX compounds significantly block the migration of HRECs. (C) Effect of APX2009 and APX2014 on cell migration in Rf/6a cells. A confluent monolayer of Rf/6a with various treatments (highest doses shown) was wounded and wound closure was monitored for 16 hours. (D) Quantitative analysis of cell migration shows that APX compounds significantly block the migration of Rf/6a cells. Mean ± S.E.M., n = 3 per dose. **P < 0.01; ***P < 0.001 compared with DMSO control (one-way ANOVA with Dunnett’s post-hoc test). Representative data from three independent experiments. Scale bar, 500 μm. See Supplemental Figs. 4 and 5.

Compounds APX2009 and APX2014 inhibit endothelial tube formation in HRECs and Rf/6a cells in vitro. Representative images of tube formation on Matrigel by HRECs (A) and Rf/6a cells (C) in the presence of the indicated concentrations of APX compounds; Quantitative analysis of tube formation in HRECs (B) and Rf/6a cells (D) following APX2009 and APX2014 treatment. Tubular length was measured and represented as relative to DMSO control. Mean ± S.E.M., n = 3 wells. **P < 0.01; ***P < 0.001 compared with DMSO control (one-way ANOVA with Dunnett’s post-hoc test). Representative data from three independent experiments. Scale bar, 500 μm. See Supplemental Figs. 6 and 7.

Compounds APX2009 and APX2014 inhibit TNF-α-mediated NF-κB signaling and proangiogenic target gene mRNA expression. (A) After treating HRECs with the indicated concentrations of APX2009 and APX2014, p65 (red) was detected by immunofluorescence and nuclei (blue) stained with DAPI; compounds dose dependently reduced p65 nuclear translocation as evidenced by decreased overlap between red and blue signals. BAY 11-7082 is a positive control NF-κB inhibitor. Scale bars, 100 μm. (B) VEGFA, (C) VCAM1, and (D) CCL20 mRNA expression levels in HRECs. APX2009 and APX2014 dose dependently inhibited levels of each transcript. Mean ± S.E.M., n = 3 technical replicates. *P < 0.05; **P < 0.01; ***P < 0.001 compared with DMSO control (one-way ANOVA with Dunnett’s post-hoc test). Representative data from three independent experiments.

Compounds APX2009 and APX2014 inhibit choroidal sprouting in a concentration-dependent manner. Representative phase-contrast images of choroidal sprouts formed 48 hours after treatment with indicated concentrations of APX2009 (A) or APX2014 (C) or vehicle control (0.5% or 0.2% DMSO, respectively). Quantification of sprouting distance from the edge of the APX2009-treated (B) or APX2014-treated (D) choroidal tissue piece to the end of the sprouts averaged from four perpendicular directions using ImageJ software. Mean ± S.E.M., n = 4–6 choroids/per treatment; N = 3 to 4 eyes. ****P < 0.0001 (ANOVA with Dunnett’s post-hoc test). Scale bars, 500 μm.

Systemic Ref-1 inhibition with APX3330 blocks neovascularization in the laser-induced choroidal neovascularization mouse model. (A) Representative optical coherence tomography (OCT) images obtained 7 days post-laser, showing CNV lesions in eyes of vehicle (left) and 50 mg/kg i.p. APX3330-treated animals (right). (B) Representative images from confocal microscopy for agglutinin-stained CNV lesions 14 days post–laser treatment. (C) Quantification of CNV lesion vascular volumes from Z-stack images at day 14 using ImageJ software. Mean ± S.E.M., n = 7–9 eyes/treatment. *P < 0.05 (unpaired Student’s t test). Scale bars, 100 μm. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer.