Nanoparticle-mediated expression of a Wnt pathway inhibitor ameliorates ocular neovascularization

Abstract

Objective: The deficiency of very low-density lipoprotein receptor resulted in Wnt signaling activation and neovascularization in the retina. The present study sought to determine whether the very low-density lipoprotein receptor extracellular domain (VLN) is responsible for the inhibition of Wnt signaling in ocular tissues.

Approach and results: A plasmid expressing the soluble VLN was encapsulated with poly(lactide-co-glycolide acid) to form VLN nanoparticles (VLN-NP). Nanoparticles containing a plasmid expressing the low-density lipoprotein receptor extracellular domain nanoparticle were used as negative control. MTT, modified Boyden chamber, and Matrigel (™) assays were used to evaluate the inhibitory effect of VLN-NP on Wnt3a-stimulated endothelial cell proliferation, migration, and tube formation. Vldlr(-/-) mice, oxygen-induced retinopathy, and alkali burn-induced corneal neovascularization models were used to evaluate the effect of VLN-NP on ocular neovascularization. Wnt reporter mice (BAT-gal), Western blotting, and luciferase assay were used to evaluate Wnt pathway activity. Our results showed that VLN-NP specifically inhibited Wnt3a-induced endothelial cell proliferation, migration, and tube formation. Intravitreal injection of VLN-NP inhibited abnormal neovascularization in Vldlr(-/-), oxygen-induced retinopathy, and alkali burn-induced corneal neovascularization models, compared with low-density lipoprotein receptor extracellular domain nanoparticle. VLN-NP significantly inhibited the phosphorylation of low-density lipoprotein receptor-related protein 6, the accumulation of β-catenin, and the expression of vascular endothelial growth factor in vivo and in vitro.

Conclusions: Taken together, these results suggest that the soluble VLN is a negative regulator of the Wnt pathway and has antiangiogenic activities. Nanoparticle-mediated expression of VLN may thus represent a novel therapeutic approach to treat pathological ocular angiogenesis and potentially other vascular diseases affected by Wnt signaling.

Keywords: VLDLR; Wnt; eye; nanoparticle; neovascularization.

Fig. 1. VLN-NP mediated expression and activity of VLN in HRMEC

A: Conditioned media of VLN-NP or LN-NP were concentrated 3-fold and used for Western blotting. B: HRMEC were quantified with MTT assay at indicated durations of VLN-NP or LN-NP treatment. **p<0.01, n=6. C: HRMEC were quantified after treatment with VLN-NP or LN-NP for 72 hr in the presence of LCM or WCM. **p < 0.01, n=6. D, F: HRMEC were seeded on one side of the membrane in the insert and treated with VLN-NP or LN-NP in the presence of WCM or LCM for 12 hr. The cells migrated to the other side of the membrane were stained, and micrographs captured (D), and cells counted in five fields per insert (F). **p<0.01, n=3. E, G: HRMEC were treated with VLN-NP and LN-NP for 72 hr, seeded on Matrigel-coated plates and incubated with WCM or LCM at 37°C for 6 hr. E: Representative micrographs of tube formation (×200). G: Branching points were counted in five fields per dish. **p<0.01, n=3. H: HRMEC were transfected with a VLDLR siRNA or a control siRNA after incubation with VLN-NP or LN-NP for 48 hr for additional 24 hr. Cells were then quantified with MTT assay. **p<0.01, n=6.

Fig. 2. VLP-NP suppresses Wnt signaling activated by Wnt ligand in cultured cells

A: HRMEC were transfected with VLN-NP or LN-NP for 48 hr and treated with LCM or WCM for 4 hr (p-LRP6, t-LRP6, n-p-β-catenin, t-β-catenin) or 24 hr (VEGF). Levels of p-LRP6, t-LRP6, n-p-β-catenin, t-β-catenin and VEGF in the cell lysates were measured by Western blot analysis. B–D: Semi-quantification of p-LRP6 and t-LRP6 (B), n-p-β-catenin and t-β-catenin (C), and VEGF (D) levels by densitometry and normalized by β-actin levels. **p<0.01, * p<0.05, n=3. E: HRMEC were transfected with VLN-NP or LN-NP for 48 hr and treated with Mab2F1 (50 µg/ml) 4 hr before adding LCM/WCM for additional 24 hr. VEGF mRNA levels were measured by q-PCR and normalized by 18S. **p<0.01, * p<0.05, NS: not significant, n=3; F, G: HRMEC were transfected with VLN-NP or LN-NP for 48 hr and infected with Ad-GFP/Ad-S37A for additional 24 hr. T-β-catenin and VEGF protein levels were measured by Western blot analysis (F), and VEGF mRNA levels were measured by q-PCR and normalized by 18S (G). H: hTERT-RPE cells were transfected with TOPFLASH vectors and then treated as indicated for 24 hr. TOPFLASH activity was measured using Luciferase assay and expressed as the firefly/renilla ratio relative to that of LCM-LN group. **p<0.01, n=3.

Fig. 3. Anti-angiogenic activity of VLN-NP in the retina

A: Expression of VLN in the mouse retinas was measured using Western blot analysis at 1, 2, 3 and 4 weeks following an intravitreal injection of VLN-NP. B: Representative retinal angiographs of Vldlr^−/− (VKO) mice(upper panels, ×40; lower panels, higher magnification of the boxed areas). VLN-NP was injected into the VKO eyes on P10, with LN-NP as control. Retinal angiography was performed on P30. C: Quantification of intra-retinal neovascular blebs (IRN blebs) in VKO mice treated with VLN-NP or LN-NP. **p<0.01, n=6. D: Immunostaining of BS-1 lectin (green) in VKO retinal sections. The nuclei were counterstained with DAPI (blue) (×200). Arrows indicate abnormal vessels in the retinas. GCL: ganglion cell layer; INL: inner nuclear layer; ONL: outer nuclear layer; RPE: retinal pigment epithelium. E–H: VLN-NP or LN-NP was intravitreally injected into OIR mice on P12 and retinal NV analyzed at P17. E: Representative retinal angiographs (upper panels, ×40; lower panels, higher magnification of the boxed areas). F: Quantification of non-perfusion area in the retina. **p<0.01, n=6. G: Representative retinal sections with H&E staining. Arrows indicate pre-retinal vascular cells (×200). H: Quantification of pre-retinal vascular cells. **p<0.01, n=6.

Fig. 4. The inhibition of Wnt signaling by VLN-NP in Vldlr^−/− and OIR mice

A–D: Eyecups were isolated at P30 from LN-NP-treated WT (WT+LN), LN-NP-treated Vldlr^−/− (VKO+LN) and VLN-NP-treated Vldlr^−/− (VKO+VLN) group. Levels of proteins were measured by Western blot analysis (A). Each lane represents an individual animal. Densitometry was performed to semi-quantify p-LRP6 and t-LRP6 (B), n-p-β-catenin and t-β-catenin (C), and VEGF (D) which were normalized by β-actin levels. **p<0.01, *p<0.05, n=4. E–H: The retinas were dissected on P17 from LN-NP-treated Normoxia (N+LN), LN-NP-treated OIR (OIR+LN) and VLN-NP-treated OIR (OIR+VLN) groups. Levels of protein were measured by Western blot analysis (E). Densitometry was performed to semi-quantify p-LRP6 and t-LRP6 (F), n-p-β-catenin, t-β-catenin (G), and VEGF (H) levels, which were normalized by β-actin levels. **p<0.01, *p<0.05, n=3. I: X-gal staining of retinal sections: VLN-NP or LN-NP was injected intravitreally into BAT-gal mice (50 µg/eye) with OIR at P12. The retinas were dissected at P17 and sectioned, and β-galactosidase activities were evaluated by X-gal staining (blue). GCL: ganglion cell layer; INL: inner nuclear layer; ONL: outer nuclear layer; RPE: retinal pigment epithelium.

Fig. 5. The effect of VLN-NP in rat alkali burn-induced corneal NV model

VLN-NP or LN-NP was administered topically onto the cornea 1 hr after alkali burn, and photographs were taken on day 7 following the administration. A: Representative photographs of rat corneas with indicated treatments. Arrows indicate corneal NV areas. B, C: Quantification of corneal NV by NV clock hour (B) and area (C) in each group. *p<0.05, **p<0.01, n=6. D: Representative rat cornea sections with H&E staining and immunostaining of an anti-CD31 antibody (green), and the nuclei counterstained with DAPI (blue). Magnification ×100. Arrows indicate NV areas in the cornea. Epi: epithelium; S: stroma; Endo: endothelium.

Fig. 6. VLN-NP inhibits Wnt signaling in neovascularized cornea after alkali burn

A: Corneal levels of proteins were measured by Western blot analysis at day 7 following the alkali burn. Each lane represents an individual rat. B–D: Semi-quantification of p-LRP6 and t-LRP6 (B), n-p-β-catenin and t-β-catenin (C), and VEGF (D) levels by densitometry and normalized by β-actin levels. **p<0.01, *p<0.05, n=3. E: X-gal staining of the whole corneas and corneal sections. Corneas from BAT-gal mice were isolated at Day 7 after the alkali burn and fixed. Corneal flat-mount and frozen sections were stained with X-gal to evaluate the expression of β-galactosidase reporter (blue). Epi: epithelium; S: stroma; Endo: endothelium.