Characterization of a Novel Caveolin Modulator That Reduces Vascular Permeability and Ocular Inflammation

Abstract

Purpose: Caveolin (Cav) regulates various aspect of endothelial cell signaling and cell-permeable peptides (CPPs) fused to domains of Cav can reduce retinal damage and inflammation in vivo. Thus, the goal of the present study was to identify a novel CPP that improves delivery of a truncated Cav modulator in vitro and in vivo.

Methods: Phage display technology was used to identify a small peptide (RRPPR) that was internalized into endothelial cells. Fusions of Cav with the peptide were compared to existing molecules in three distinct assays, vascular endothelial growth factor-A (VEGF) induced nitric oxide (NO) release, VEGF induced vascular leakage, and in a model of immune mediated uveitis.

Results: RRPPR was internalized efficiently and was potent in blocking NO release. Fusing RRPPR with a minimal Cav inhibitory domain (CVX51401) dose-dependently blocked NO release, VEGF induced permeability, and retinal damage in a model of uveitis.

Conclusions: CVX51401 is a novel Cav modulator that reduces VEGF and immune mediated inflammation.

Translational relevance: CVX51401 is an optimized Cav modulator that reduces vascular permeability and ocular inflammation that is poised for clinical development.

Figure 1.

RRPPR intern a lizes at a faster rate and more completely than AP in cultured EC. (A) Comparative fluorescence readouts for similar concentrations of rhodamine-AP (rhod-AP) and carboxyfluorescein-Endo5 (cFluo-RRPPR) dissolved in the cell lysis buffer to assess the linearity between peptide concentration and fluorescence values. Peptides were assayed separately to prevent interference of absorption/emission spectra. (B) Rate of cFluo-RRPPR internalization is greater than that of rhod-AP. Cultured ECs were incubated for 1, 2, 4, or 6 hours with individual peptides, acid washed, rinsed, trypsinized, lysed, and fluorescence was determined and converted to moles of peptides per 10⁶ cells by using a standard curve. Cells incubated with peptides for 5 minutes and treated as described, were used as background values. Data are mean ± standard error of the mean (SEM). N = 4 wells per condition, in triplicate, representative data are shown.

Figure 2.

RRPPR-Cav is more potent than AP-Cav at blocking VEGF-induced NO release. (A) Cultured ECs were pretreated for 6 hours with the indicated peptides (10 µM) and stimulated with VEGF (1 nM) for 30 minutes as indicated. *P < 0.05 compared with vehicle, and **P < 0.05 compared with AP-Cav + VEGF. N = 4 in triplicate. (B) The inhibitory actions of AP-Cav and RRPPR-Cav on NO release are dose-dependent. Cultured ECs were pretreated with peptides (1–50 µM) for 6 hours and stimulated with VEGF as described in A. Data are mean+/− SEM with *P < 0.05 compared with vehicle, and **P < 0.05 compared with AP-Cav + VEGF. N = 4 experiments performed in duplicate. (C) Optimization of both the cell-penetrating sequence and Cav domain (CVX51401) results in more effective NO release inhibition. ECs were treated with AP-Cav (10 µM) or CVX51401 (2 and 10 µM) and VEGF induced NO release assessed. Data are mean +/− SEM with *P < 0.05 compared with vehicle, and **P < 0.05 compared with AP-Cav (10 µM). N = 4 experiments were performed in triplicate, representative data are shown.

Figure 3.

CVX51401 suppresses VEGF-A induced permeability in vivo. (A) C57/B6J mice were administrated of either vehicle or CVX 51401 (0.1–5 mg/kg, IP) for 1 hour; followed by VEGF-A (300 ng)or PBS injected intradermally into either ear for 30 minutes and the extravasation of Evans Blue measured spectrophotometrically (n = 4–6 mice/group). Data are mean +/− standard error of measurement with *P < 0.05 compared with vehicle. (B) Mice were injected on days 14, 7, 3, and 1 and after 1 hour, with a single dose of CVX 51401 (2.5 mg/kg; IP) or vehicle. All 6 groups of mice were then injected intradermally into the ear with either PBS or VEGF-A and Evans blue leakage quantified. Data are mean +/− standard error of measurement with data are mean ± SEM. *P < 0.05 compared with vehicle and PBS injected in the ears (n = 5 mice/group).

Figure 4.

CVX51401 reduces inflammation and retinal degeneration in a model of uveitis in mice. B10.RIII mice were immunized with IRBP in complete adjuvant and treated with vehicle, AP-Cav, or CVX51401 (at 5 mg/kg daily, IP) from days 7 to 14 post-immunization. At day 14, eyes were processed for histology followed by grading. (A) Representative retinal histologically of treated groups showing less inflammation and degeneration in CVX51401 treated mice and quantitative scoring of these results in (B). Data are mean +/− standard error of measurement with *P < 0.05 compared with vehicle (n = 7 mice), AP-Cav (n = 8 mice), and CVX51401 (n = 8 mice). P values for other groups are shown for comparison.