Intraocular gene transfer of pigment epithelium-derived factor rescues photoreceptors from light-induced cell death

In this study, we investigated whether intraocular gene transfer of pigment epithelium-derived factor (PEDF) ameliorates the extent of light-induced photoreceptor cell death. Lewis rats received intravitreous injection of 3 x 10(9) particles of adenoviral vector expressing PEDF (AdPEDF.11) in one eye and 3 x 10(9) particles of empty adenoviral vector (AdNull.11) in the contralateral eye. The rats were then dark-adapted for 3 days after which they were continuously exposed to fluorescent light (2,500 lux) for 0, 6, 24, 96, and 168 h. Both eyes were then enucleated and processed for morphometric analysis. Cell death in the retina was examined using TUNEL staining with a propidium iodide counterstain. The photoreceptor cell counts in each of the three groups were significantly different (P < 0.001). Eyes that received intravitreous injection of AdNull.11 or no injection showed a greater number of pyknotic photoreceptor cells and a reduced photoreceptor cell density as compared to eyes treated with intravitreous AdPEDF.11 injection. AdNull.11 treated eyes showed a lesser but still significant protection of photoreceptor cells when compared to untreated eyes. Fewer TUNEL-positive photoreceptor cells were present in AdPEDF.11 treated eyes than in AdNull.11 treated or untreated eyes (P = 0.004). The amplitudes of the ERG a-wave, b-wave, and oscillatory potentials (OPs) were increased significantly by treatment (P < 0.05). These data suggest that adenovirus vector-mediated intraocular expression of PEDF significantly increases photoreceptor cell survival following excessive light exposure. Neuroprotection may result from inhibition of light-induced apoptotic processes. This study provides proof of concept for a gene transfer approach to modulating retinal cell death resulting from photo-oxidative damage and supports the hypothesis that gene transfer of PEDF is broadly applicable to modulating apoptosis in the retina.     

Angiopoietin-2 enhances retinal vessel sensitivity to vascular endothelial growth factor

Increased expression of vascular endothelial growth factor (VEGF) in the retina starting after postnatal day (P)7 results in neovascularization originating from deep retinal capillaries, but not those in the superficial capillary bed. Doxycycline was administered starting P0 to double transgenic mice with inducible expression of VEGF in the retina. These mice showed proliferation and dilation of superficial retinal capillaries, indicating that at this stage of development, the superficial capillaries are sensitive to the effects of VEGF. Angiopoietin-2 (Ang2) is expressed along the surface of the retina for several days after birth, but by P7 and later, Ang2 is only expressed in the region of the deep capillary bed. In mice with ubiquitous doxycycline-inducible expression of Ang2, in the absence of doxycycline, intravitreous injection of a gutless adenoviral vector expressing VEGF (AGV.VEGF) resulted in neovascularization of the cornea and iris, but no retinal neovascularization. After treatment with doxycycline to induce Ang2 expression, intravitreous injection of AGV.VEGF caused retinal neovascularization in addition to corneal and iris neovascularization. The retinal neovascularization originated from both the superficial and deep capillary beds. These data suggest that Ang2 promotes sensitivity to the angiogenic effects of VEGF in retinal vessels.     

VEGF-TRAP(R1R2) suppresses choroidal neovascularization and VEGF-induced breakdown of the blood-retinal barrier

Vascular endothelial growth factor (VEGF) plays a central role in the development of retinal neovascularization and diabetic macular edema. There is also evidence suggesting that VEGF is an important stimulator for choroidal neovascularization. In this study, we investigated the effect of a specific inhibitor of VEGF, VEGF-TRAP(R1R2), in models for these disease processes. VEGF-TRAP(R1R2) is a fusion protein, which combines ligand binding elements taken from the extracellular domains of VEGF receptors 1 and 2 fused to the Fc portion of IgG1. Subcutaneous injections or a single intravitreous injection of VEGF-TRAP(R1R2) strongly suppressed choroidal neovascularization in mice with laser-induced rupture of Bruch's membrane. Subcutaneous injection of VEGF-TRAP(R1R2) also significantly inhibited subretinal neovascularization in transgenic mice that express VEGF in photoreceptors. In two models of VEGF-induced breakdown of the blood-retinal barrier (BRB), one in which recombinant VEGF is injected into the vitreous cavity and one in which VEGF expression is induced in the retina in transgenic mice, VEGF-TRAP(R1R2) significantly reduced breakdown of the BRB. These data confirm that VEGF is a critical stimulus for the development of choroidal neovascularization and indicate that VEGF-TRAP(R1R2) may provide a new agent for consideration for treatment of patients with choroidal neovascularization and diabetic macular edema.     

Influence of plasminogen activator inhibitor type 1 on choroidal neovascularization.

High levels of the plasminogen activators, but also their inhibitor, plasminogen activator inhibitor 1 (PAI-1), have been documented in neovascularization of severe ocular pathologies such as diabetic retinopathy or age-related macular degeneration (AMD). AMD is the primary cause of irreversible photoreceptors loss, and current therapies are limited. PAI-1 has recently been shown to be essential for tumoral angiogenesis. We report here that deficient PAI-1 expression in mice prevented the development of subretinal choroidal angiogenesis induced by laser photocoagulation. When systemic and local PAI-1 expression was achieved by intravenous injection of a replication-defective adenoviral vector expressing human PAI-1 cDNA, the wild-type pattern of choroidal angiogenesis was restored. These observations demonstrate the proangiogenic activity of PAI-1 not only in tumoral models, but also in choroidal experimental neovascularization sharing similarities with human AMD. They identify therefore PAI-1 as a potential target for therapeutic ocular anti-angiogenic strategies.     

Nitric oxide is proangiogenic in the retina and choroid

Nitric oxide (NO) has been shown to have proangiogenic or antiangiogenic effects depending upon the setting. In this study, we used mice with targeted deletion of one of the three isoforms of nitric oxide synthase (NOS) to investigate the effects of NO in ocular neovascularization. In transgenic mice with increased expression of vascular endothelial growth factor (VEGF) in photoreceptors, deficiency of any of the three isoforms caused a significant decrease in subretinal neovascularization, but no alteration of VEGF expression. In mice with laser-induced rupture of Bruch's membrane, deficiency of inducible NOS (iNOS) or neuronal NOS (nNOS), but not endothelial NOS (eNOS), caused a significant decrease in choroidal neovascularization. In mice with oxygen-induced ischemic retinopathy, deficiency of eNOS, but not iNOS or nNOS caused a significant decrease in retinal neovascularization and decreased expression of VEGF. These data suggest that NO contributes to both retinal and choroidal neovascularization and that different isoforms of NOS are involved in different settings and different disease processes. A broad spectrum NOS inhibitor may have therapeutic potential for treatment of both retinal and choroidal neovascularization.     

Helper-dependent adenovirus for the gene therapy of proliferative retinopathies: stable gene transfer, regulated gene expression and therapeutic efficacy

BACKGROUND: Ocular neovascular disorders, such as diabetic retinopathy and age-related macular degeneration, are the principal causes of blindness in developed countries. Current treatments are of limited efficacy, whereas a therapy based on intraocular gene transfer of angiostatic factors represents a promising alternative. For the first time we have explored the potential of helper-dependent adenovirus (HD-Ad), the last generation of Ad vectors, in the therapy of retinal neovascularization. METHODS: We first analyzed efficiency and stability of intraretinal gene transfer following intravitreous injection in mice. A HD-Ad vector expressing green fluorescent protein (GFP) under the control of the cytomegalovirus (CMV) promoter (HD-Ad/GFP) was compared with a first-generation (E1/E3-deleted) Ad vector carrying an identical GFP expression cassette (FG-Ad/GFP). We also constructed HD-Ad vectors expressing a soluble form of the VEGF receptor (sFlt-1) in a constitutive (HD-Ad/sFlt-1) or doxycycline (dox)-inducible (HD-Ad/S-M2/sFlt-1) manner and tested their therapeutic efficacy upon intravitreous delivery in a rat model of oxygen-induced retinopathy (OIR). RESULTS: HD-Ad/GFP promoted long-lasting (up to 1 year) transgene expression in retinal Müller cells, in marked contrast with the short-term expression observed with FG-Ad/GFP. Intravitreous injection of HD-Ad vectors expressing sFlt-1 resulted in detectable levels of sFlt-1 and inhibited retinal neovascularization by more than 60% in a rat model of OIR. Notably, the therapeutic efficacy of the inducible vector HD-Ad/S-M2/sFlt-1 was strictly dox-dependent. CONCLUSIONS: HD-Ad vectors enable stable gene transfer and regulated expression of angiostatic factors following intravitreous injection and thus are attractive vehicles for the gene therapy of neovascular diseases of the retina.     

Increased expression of VEGF in retinal pigmented epithelial cells is not sufficient to cause choroidal neovascularization

Increased expression of vascular endothelial cell growth factor (VEGF) in the retina is sufficient to stimulate sprouting of neovascularization from the deep capillary bed of the retina, but not the superficial retinal capillaries or the choriocapillaris. Coexpression of VEGF and angiopoietin 2 (Ang2) results in sprouting of neovascularization from superficial and deep retinal capillaries, but not the choriocapillaris. However, retina-derived VEGF and Ang2 may not reach the choriocapillaris, because of tight junctions between retinal pigmented epithelial (RPE) cells. To eliminate this possible confounding factor, we used the human vitelliform macular dystrophy 2 (VMD2) promoter, an RPE-specific promoter, combined with the tetracycline-inducible promoter system, to generate double transgenic mice with inducible expression of VEGF in RPE cells. Adult mice with increased expression of VEGF in RPE cells had normal retinas and choroids with no choroidal neovascularization (CNV), but when increased expression of VEGF in RPE cells was combined with subretinal injection of a gutless adenoviral vector containing an expression construct for Ang2 (AGVAng2), CNV consistently occurred. In contrast, triple transgenic mice with induced expression of Ang2 and VEGF in RPE cells, did not develop CNV. These data suggest that increased expression of VEGF and/or Ang2 in RPE cells is not sufficient to cause CNV unless it is combined with a subretinal injection of a gutless adenoviral vector, which is likely to perturb RPE cells. These data also suggest that the effects of angiogenic proteins may vary among vascular beds, even those that are closely related, and, therefore, generalizations should be avoided.     

Intraocular gutless adenoviral-vectored VEGF stimulates anterior segment but not retinal neovascularization

Vascular endothelial growth factor (VEGF) and insulin-like growth factor-1 (IGF-1) have been implicated as important stimulatory factors for retinal neovascularization. In this study, we used intraocular gene transfer with gutless adenoviral (AGV) vectors to determine the effect of increased intraocular expression of VEGF, IGF-1, or sphingosine kinase (SPK), which produces sphingosine-1-phosphate, another angiogenic factor. Retinal neovascularization did not occur from intravitreous AGV-vectored VEGF, IGF-1, SPK, or combined VEGF and IGF-1, except occasionally adjacent to the retinal penetration site from the injection. However, corneal and iris neovascularization occurred after 2 weeks in all eyes injected with AGV.VEGF, but not those injected with only AGV.IGF-1 or AGV.SPK. These data suggest that the superficial capillary bed of the retina is relatively insensitive to VEGF, IGF-1, or SPK in adult mice, except when combined with retinal trauma. However, AGV-vectored VEGF is sufficient to consistently cause severe corneal and iris neovascularization. This provides a model for anterior segment neovascularization, which unlike previous models is relatively inexpensive and is not plagued by spontaneous regression, and therefore, may be useful for identification of new treatments.     

Vascular endothelial growth factor upregulates pigment epithelium-derived factor expression via VEGFR-1 in human retinal pigment epithelial cells

We previously demonstrated that differentiated retinal pigment epithelial (RPE) cells express high levels of vascular endothelial growth factor (VEGF) and pigment epithelium-derived factor (PEDF), and a critical balance between VEGF and PEDF is important to prevent the development of choroidal neovascularization. We report here that VEGF secreted by RPE cells upregulates PEDF expression via VEGFR-1 in an autocrine manner. PEDF mRNA and protein expression was downregulated by neutralizing antibody against VEGF in differentiated human RPE cells. VEGFR-1 neutralization decreased PEDF mRNA and protein expression whereas anti-VEGFR-2 antibody had no effect. Addition of placenta growth factor (PlGF) restored PEDF expression in the presence of anti-VEGF antibody. These results demonstrate a regulatory interaction between angiogenesis stimulators and inhibitors to maintain homeostasis in normal human retina.     

Inhibition of corneal neovascularization with endostatin delivered by adeno-associated viral (AAV) vector in a mouse corneal injury model

The use of a recombinant adeno-associated viral (rAAV) vector carrying endostatin gene as an anti-angiogenesis strategy to treat corneal neovascularization in a mouse model was evaluated. Subconjunctival injection of recombinant endostatin-AAV was used to examine the inhibition of corneal neovascularization induced by silver nitrate cauterization in mice. The results showed that gene expression in corneal tissue was observed as early as 4 days after gene transfer and stably lasted for over 8 months with minimal immune reaction. Subconjunctival injection of a high-titer rAAV-endostatin successfully inhibited neovascularization. Immunohistchemistry staining of CD 31 and endostatin showed that the treatment significantly inhibits angiogenesis in cornea. We concluded that the rAAV was capable of directly delivering genes to the ocular surface epithelium by way of subconjunctival injection and was able to deliver sustained, high levels of gene expression in vivo to inhibit angiogenesis.     

Prolonged blockade of VEGF family members does not cause identifiable damage to retinal neurons or vessels

Several ocular diseases complicated by neovascularization are being treated by repeated intraocular injections of vascular endothelial growth factor (VEGF) antagonists. While substantial benefits have been documented, there is concern that unrecognized damage may be occurring, because blockade of VEGF may damage the fenestrated vessels of the choroicapillaris and deprive retinal neurons of input from a survival factor. One report has suggested that even temporary blockade of all isoforms of VEGF-A results in significant loss of retinal ganglion cells. In this study, we utilized double transgenic mice with doxycycline-inducible expression of soluble VEGF receptor 1 coupled to an Fc fragment (sVEGFR1Fc), a potent antagonist of several VEGF family members, including VEGF-A, to test the effects of VEGF blockade in the retina. Expression of sVEGFR1Fc completely blocked VEGF-induced retinal vascular permeability and significantly suppressed the development of choroidal neovascularization at rupture sites in Bruch's membrane, but did not cause regression of established choroidal neovascularization. Mice with constant expression of sVEGFR1Fc in the retina for 7 months had normal electroretinograms and normal retinal and choroidal ultrastructure including normal fenestrations in the choroicapillaris. They also showed no significant difference from control mice in the number of ganglion cell axons in optic nerve cross sections and the retinal level of mRNA for 3 ganglion cell-specific genes. These data indicate that constant blockade of VEGF for up to 7 months has no identifiable deleterious effects on the retina or choroid and support the use of VEGF antagonists in the treatment of retinal diseases.     

Blockade of sphingosine-1-phosphate reduces macrophage influx and retinal and choroidal neovascularization

Sphingosine-1-phosphate (S1P) is a bioactive lipid molecule that stimulates endothelial cell migration, proliferation, and survival in vitro, and tumor angiogenesis in vivo. In this study, we used a humanized monoclonal antibody (sonepcizumab) that selectively binds S1P to investigate its role in retinal and choroidal neovascularization (NV). Intraocular injection of sonepcizumab significantly reduced macrophage influx into ischemic retina and strongly suppressed retinal NV in mice with oxygen-induced ischemic retinopathy. In mice with laser-induced rupture sites in Bruch's membrane, intraocular injection of sonepcizumab significantly reduced the area of choroidal NV and concomitantly reduced fluorescein leakage from the remaining choroidal NV. Four weeks after intraocular injection of up to 1.8 mg of the sonepcizumab in non-human primates, electroretinograms and fluorescein angiograms were normal, and light microscopy of ocular sections showed no evidence of structural damage. These data show for the first time that S1P stimulates both choroidal and retinal NV and suggest that sonepcizumab could be considered for evaluation in patients with choroidal or retinal NV.     

A novel bispecific molecule delivered by recombinant AAV2 suppresses ocular inflammation and choroidal neovascularization

Elevated vascular endothelial growth factor (VEGF) and complement activation are implicated in the pathogenesis of different ocular diseases. The objective of this study was to investigate the hypothesis that dual inhibition of both VEGF and complement activation would confer better protection against ocular inflammation and neovascularization. In this study, we engineered a secreted chimeric VEGF inhibitor domain (VID), a complement inhibitor domain (CID) and a dual inhibitor (ACVP1). Vectors expressing these three inhibitors were constructed and packaged into AAV2 (sextY‐F) particles. The expression and secretion of the proteins were validated by Western blot. The effects of these inhibitors expressed from AAV2 vectors were examined in endotoxin‐induced uveitis (EIU), experimental autoimmune uveoretinitis (EAU) and choroidal neovascularization (CNV) mouse models. The AAV2 vectors expressing the CID‐ and ACVP1‐attenuated inflammation in EIU and EAU model, whereas the vector expressing VID showed improved retinal structure damaged by EAU, but not affect the infiltration of inflammatory cells in EAU or EIU eyes. Both VID and CID vectors improved laser‐induced retinal and choroid/RPE injuries and CNV, whereas ACVP1 vector provided significantly better protection. Our results suggest that gene therapy targeting VEGF and complement components could provide an innovative and long‐term strategy for ocular inflammatory and neovascular diseases.     

Retinal and choroidal neovascularization

The unique vascular supply of the retina, the ability to visualize the vasculature in vivo, and the ability to selectively express genes in the retina make the retina an ideal model system to study molecular mechanisms of angiogenesis. In addition, this area of investigation has great clinical significance, because retinal and choroidal neovascularization are the most common causes of severe visual loss in developed countries and new treatments are needed. As a result, interest in ocular neovascularization is rapidly growing and there has been considerable recent progress. Use of genetically engineered mice in recently developed murine models provides a means to investigate the role of individual gene products in neovascularization in two distinct vascular beds, the retinal vasculature and the choroidal vasculature. It appears that angiogenesis in different vascular beds has common themes, but also has tissue-specific aspects. This review summarizes recent progress in the field of ocular neovascularization and the prospects that it provides for the development of new treatments.     

Long-Term Retinal PEDF Overexpression Prevents Neovascularization in a Murine Adult Model of Retinopathy

Neovascularization associated with diabetic retinopathy (DR) and other ocular disorders is a leading cause of visual impairment and adult-onset blindness. Currently available treatments are merely palliative and offer temporary solutions. Here, we tested the efficacy of antiangiogenic gene transfer in an animal model that mimics the chronic progression of human DR. Adeno-associated viral (AAV) vectors of serotype 2 coding for antiangiogenic Pigment Epithelium Derived Factor (PEDF) were injected in the vitreous of a 1.5 month-old transgenic model of retinopathy that develops progressive neovascularization. A single intravitreal injection led to long-term production of PEDF and to a striking inhibition of intravitreal neovascularization, normalization of retinal capillary density, and prevention of retinal detachment. This was parallel to a reduction in the intraocular levels of Vascular Endothelial Growth Factor (VEGF). Normalization of VEGF was consistent with a downregulation of downstream effectors of angiogenesis, such as the activity of Matrix Metalloproteinases (MMP) 2 and 9 and the content of Connective Tissue Growth Factor (CTGF). These results demonstrate long-term efficacy of AAV-mediated PEDF overexpression in counteracting retinal neovascularization in a relevant animal model, and provides evidence towards the use of this strategy to treat angiogenesis in DR and other chronic proliferative retinal disorders.     

Unbalanced expression of VEGF and PEDF in ischemia-induced retinal neovascularization

Retinal levels of vascular endothelial growth factor (VEGF) and pigment epithelium-derived factor (PEDF), an angiogenic inhibitor, were measured and correlated with the ischemia-induced retinal neovascularization in rats. The retinas with neovascularization showed a 5-fold increase in VEGF while 2-fold decrease in PEDF, compared to the age-matched controls, resulting in an increased VEGF/PEDF ratio. The time course of the VEGF/PEDF ratio change correlated with the progression of retinal neovascularization. Changes in the VEGF and PEDF mRNAs preceded their protein level changes. These results suggest that an unbalance between angiogenic stimulators and inhibitors may contribute to retinal neovascularization.     

Prolonged blockade of VEGF receptors does not damage retinal photoreceptors or ganglion cells

It has recently been reported that relatively short‐term inhibition of vascular endothelial growth factor (VEGF) signaling can cause photoreceptor cell death, a potentially clinically important finding since VEGF blockade has become an important modality of treatment of ocular neovascularization and macular edema. However, in a set of studies in which we achieved extended and complete blockage of VEGF‐induced vascular leakage through retinal expression of a VEGF binding protein, we did not observe any toxicity to retinal neurons. To follow‐up on these apparently discrepant findings, we designed a set of experiments with the kinase inhibitor SU4312, which blocks phosphorylation of VEGF receptors, to look directly for evidence of VEGF inhibition‐related retinal toxicity. Using transgenic mice with sustained expression of VEGF in photoreceptors, we determined that periocular injection of 3 µg of SU4312 every 5 days markedly suppressed subretinal neovascularization, indicating effective blockade of VEGF signaling. Wild‐type mice given periocular injections of 5 µg of SU4312 every 5 days for up to 12 weeks showed normal scotopic and photopic electroretinograms (ERGs), no TUNEL stained cells in the retina, and no reduction in outer nuclear layer thickness. Incubation of cultured ganglion cells or retinal cultures containing photoreceptors with high doses of SU4312 did not reduce cell viability. These data suggest that blocking VEGF signaling in the retina for up to 12 weeks does not damage photoreceptors nor alter ERG function and should reassure patients who are receiving frequent injections of VEGF antagonists for choroidal and retinal vascular diseases. J. Cell. Physiol. 224:262–272, 2010 © 2010 Wiley‐Liss, Inc.     

Marked inhibition of retinal neovascularization in rats following soluble-flt-1 gene transfer

BACKGROUND: In mouse models of retinopathy of prematurity (ROP) inhibitors of vascular endothelial growth factor (VEGF) functions administered systemically completely block retinal neovascularization. In contrast, selective ocular VEGF depletion has achieved an approx. 50% inhibition of retinal neovascular growth. It is unclear whether a more complete inhibition of new blood vessel development can be obtained with an anti-VEGF therapy localized to the eye. Therefore, the objective of the present study was to determine the effect of local anti-VEGF therapy in a different animal model which closely mimics human ROP. METHODS: Rats were exposed to alternating cycles of high and low levels of oxygen for 14 days immediately after birth; thereafter, they were intravitreally injected with an adenoviral vector expressing a secreted form of the VEGF receptor flt-1 (Ad.sflt), which acts by sequestering VEGF. Contralateral eyes were injected with the control vector carrying the reporter gene expressing beta-galactosidase (Ad.betaGal). RESULTS: At the peak of retinal neovascular growth, i.e. post-natal day 21 (P21), we observed up to 97.5% decrease in retinal neovascularization in animals injected with Ad.sflt. At the end of observation (P28), no significant difference in retinal vessel number was detected in both oxygen-injured and normoxic Ad.sflt-treated retinas compared with untreated or Ad.betaGal-treated retinas. CONCLUSION: Adenoviral-mediated sflt-1 gene transfer induces a near-complete inhibition of ischemia-induced retinal neovascularization in rats without affecting pre-existing retinal vessels.     

Overexpressing kringle 1 domain of hepatocyte growth factor with adeno-associated virus inhibits the pathological retinal neovascularization in an oxygen-induced retinopathy mouse model

Current clinical treatments for ocular neovascularization are characterized by high possibility of damaging healthy tissues and high recurrence rates. It is necessary to develop new treatment methods to control neovascularization with a stable and effective effect. Kringle1 domain of hepatocyte growth factor (HGFK1) has anti-angiogenesis activity. Here, we established oxygen-induced retinopathy (OIR) model to study if using adeno-associated virus (AAV) as a delivery system to overexpression HGFK1 in retinal cells could benefit retinal neovascularization. We show that, overexpressed exogenous gene was mainly expressed in the inner and outer nuclear layer of the retina. Compared with control mice, the mice pretreated with rAAV-HGFK1 at P3 showed relatively normal vascular branches examined by fluorescence fundus angiography. Subsequent H&E staining and immunohistochemical staining of CD31 of the eye tissue sections showed that the mice received rAAV-HGFK1 had a relatively normal distribution of vascular endothelial cells. Additionally, immunohistochemical staining indicated a lower expression of VEGF in the eye tissues of rAAV-HGFK1 treated OIR mice. Further in vitro studies showed that HGFK1 could inhibit the proliferation but promote the apoptosis of bovine retinal microvascular endothelial cells (BRECs) under the presence of VEGF. Moreover, HGFK1 could inhibit VEGF induced ERK activation but promote p38 activation in BRECs. Therefore, we propose that intravitreal injection of rAAV-HGFK1 might be used to improve the retinal neovascularization and HGFK1 may function through regulating VEGF signaling pathway to inhibit neovascularization.