Corneal neovascularization may be caused by several pathogenic conditions, including inflammatory, ischemic, degenerative or traumatic diseases of the cornea and loss of the limbal stem cell barrier. To date, numerous models of corneal angiogenesis have been developed. Thermal injury and alkali-burn models have been utilized in many studies, since these are easy to perform and corneal neovascularization may be conveniently observed . Inflammation is the major mechanism of cornea angiogenesis in these models and these models were predominantly used in studies focused on inflammation. Infectious keratitis and corneal ulcers, which brought unstable experimental results and difficulties to precise quantitative analysis, are also commonly seen in these models. Mouse tumor models induce corneal neovascularization by intrastromal implantation of tumor cells in the rat or mouse cornea, where vessel growth is stimulated by tumor-secreted angiogenic factors, inflammation or immunologic mediators . We previously reported the anti-angiogenic activity of H-RN in a choroid-retinal endothelial cell line (RF/6A), in the chick chorioallantoic membrane and in a mouse model of oxygen-induced retinopathy .
In the present study, we investigated the anti-angiogenic effect of H-RN in VEGF-induced corneal neovascularization in vitro using HUVECs, and in vivo using a mouse cornea micropocket assay, which is dependent on direct stimulation of neovascularization rather than indirect stimulation by inflammation or tumors. Our results obtained from both in vitro and in vivo models were highly reproducible and easily quantifiable . In addition, VEGF was used as a direct angiogenesis stimulator in the models, thus providing meaningful results for the evaluation of an anti-VEGF and anti-angiogenic reagent.
Previously, we reported a new peptide derived from HGF, H-RN, which exhibited anti-angiogenic activity in a choroid-retinal endothelial cell line (RF/6A) and in the chick chorioallantoic membrane, as well as in a mouse model of oxygen-induced retinopathy . In the present study, we investigated the anti-angiogenic activity of H-RN on corneal neovascularization. HUVECs were used for in vitro studies, and the effects of H-RN on VEGF-stimulated proliferation, cell migration and endothelial cell tube formation were investigated. Similar results were found as those obtained from our in vitro study of RF/6A cells. H-RN significantly inhibited HUVEC proliferation, migration and tube formation stimulated by VEGF. The inhibitory effects were particularly intense at high concentrations, though not dose-dependent. The scrambled peptide did not show any inhibitory effect at any concentration. In the mouse cornea micropocket assay, we found that VEGF significantly stimulated corneal angiogenesis. Neovascularization derived from the corneal limbus developed towards VEGF-containing pellets, with bushy and thick vessels migrating towards and over the surface of the white pellet. This growth was significantly inhibited following administration of H-RN. We infer that H-RN has the potential for treating pathological corneal neovascularization, and sustained release delivery may be an effective drug delivery option, although further investigation of H-RN pharmacokinetics is still required.
Li et al.  recently reported that topical administration of KH906, a recombinant human soluble VEGF receptor fusion protein, was capable of significantly inhibiting angiogenesis in an alkali burn corneal neovascularization rabbit model by topical administration. KH906 was administrated in three different concentrations; 5 mg/ml, 10 mg/ml and 20 mg/ml. Rabbits received topical administration (50 μl) of the different solutions four times daily for 14 days. Corneal neovascularization was analyzed 10 and 14 days after chemical cauterization. In this study, corneal neovascularization was significantly reduced in KH906-treated groups compared to control treated animals. Compared to the effective peptide quantity in our study, the total drug quantity applied in Li’s study is much higher (250 μg vs 5 μg), indicating that H-RN has a much lower effective concentration than KH-906. Furthermore, the treatment cycle of H-RN is significantly shorter than KH-906 (7 d vs 14 d), and the production cost of H-RN is also lower than KH-906. In this study, the level of VEGF in the cornea in KH906-treated groups was significantly decreased. In our study, we found that the ability of H-RN to inhibit the anti-apoptotic activity of VEGF and to induce G0/G1 phase cell cycle arrest is related to its anti-angiogenesis properties. Taken together, this demonstrates that both H-RN and KH-906 inhibit neovascularization through an anti-VEGF mechanism.
Previous reports have shown that VEGF may inhibit vascular endothelial cell apoptosis . We infer that H-RN may inhibit the anti-apoptosis activity of VEGF, as demonstrated by flow cytometric analysis of apoptosis. It is well established that VEGF inhibits endothelial cells apoptosis via activation of the PI3K signaling pathway  and by upregulating the expression of Bcl-2 and A1 , important anti-apoptosis genes. Additional studies are required to determine whether H-RN inhibits activation of the PI3K signaling pathway or the expression of Bcl-2 and A1. We further investigated the effect of H-RN on the cell cycle of HUVECs, which may relate to its inhibitory effect on endothelial cell proliferation. We observed a G0/G1 phase arrest in H-RN treated cells, indicating that H-RN inhibits DNA replication of HUVECs. Cyclin, cyclin dependent kinase (CDK)  and cyclin dependent kinase inhibitor (CDKI)  are critical factors regulating the cell cycle. Future studies will aim to investigate the relationship between H-RN and these factors.