The Experimental Study Of Integrin-linked Kinase(ILK) And Retinal Neovascularization

Abnormal retinal neovascularization is a main cause of vision loss for inducing retinal edema, hemorrhage and traction retinal detachment. A various retinal diseases can lead to abnormal angiogenesis such as diabetic retinopathy, retinopathy of prematurity, retinal vein occlusion. In developed countries diabetic retinopathy is the major cause of severe vision loss. Over the past several years, a variety of proteins have been identified as potential targets of anti-angiogenesis therapy, but poor results in clinical trails of some of the antiangiogenic strategies.ILK is an important component of the PI-3K signal pathway. The protein is composed of distinct domains, including four N-terminal ankyrin repeats followed by a central PH-like sequence and a C-terminal region that is homologous to the catalytic domain of protein kinases. It can be activated by PI-3K and sequentially activate PKB and GSK3. Recent data from several lines of investigation suggest that ILK play a role in physiological and pathological neovascularization.In this study, we investigate the role of ILK in promoting the formation of retinal neuvasculerization in both RF/6A cell and mice model of OIR by using LY294002(a highly selective small molecule inhibitor of ILK activity) or a specific, effective RNA-interference.Materials and Methods1 cell culture and hypoxia RF/6A cells were cultured in DMEM (Gibco/BRL) with high glucose supplemented with 10%FBS(Gibco/BRL) and antibiotics at 37℃in 5% CO 2 and 95% air. All cells were incubated in DMEM supplemented with only 1%FBS over night before used in this study. After changed the medium with glucose-free DMEM(Sigma) containing 1% fatal bovine serum, the cells were incubated in 94%N2, 5%CO2 and 1%O2 environment for 0, 1, 2, 3, 4 hours and harvested for RT-PCR or western blot analysis. In the LY294002 treated group, cells were stimulated by 20μmol/L LY294002(Sigma) for 30 minutes before incubated in 94%N2, 5%CO2 and 1%O2 environment.2 siRNA transfection RF/6A cells were transfedted with small interfering RNA(siRNA)that specifically targets the ILK gene. RF/6A cells were seeded into 25 cm2 bottles one day before transfection. At the time of transfection with siRNA, the cells were about 60% confluent. The siRNA sequences used in this study to target ILK were 5'CCUGACGAAGCUCAACGAGAA(TT)3'and5'UUCUCGUUGAGCUUCGUCAGGd(TT)3'.The control sequences were 5'UUCUCCGAACGUGUCACGUd(TT)3'and5'ACGUGACACUUCGGAGAAd(TT)3'.SiRNA were prepared by serum-free DMEM medium and transfection reagent Lipofaectamine2000 Ineitrogen).45 hours after transfection, hypoxia was induced in the cells.3 RT-PCR Total RNA was isolated from cultured cells by TRIzol reagent(Gibco/BRL). The RNA concentration and purity was determined by spectrophotometer(A260/A280).Total cellular RNA was reversed- transcribed using oligo(dT) primers, reverse transcriptase and cDNA was made by the reverse-transcription reaction incubated at 37℃for 50 min.The amplification profile was 94℃for 120s, 72℃for 60s,and 72℃for 420 s for 35 cycles. The PCR products were separated by 1% agarose gel electrophoresis.4 Western blot Analyses Cells were homogenied in lysis buffer containing 8M urea, 10%SDS, 1M DTT and protease inhibitors. SDS-PAGE was performed with a 8% gel, according to a standard procedure, after boiling with sample buffer, and protein in the gel was transferred electrophoretically to a nitrocellulose membrane at 400mA for 3.5 hours.HIF-1α,ILK andα-actin were visualized by using an polyclonal antibody(Stressgen) and a horseradish-peroxidase-linked anti-mouse Ig-G secondary antibody. The complexes were detected by autoradiography, using a chemiluminescence detection system (ECL, Amersham)5 ELISA for VEGF The concentration of VEGF produced was measured with ELISA kit(Biosource). The culture supernatants were collected after stimulation for every time point, and the absorbency was measured at 492 nm. VEGF production was normalized to the volume of the medium and cell number.6 Cell-adhesion assay The adhesion assay was performed in 96-well microtiter plates covered with fibronectin(5μg/cm2). After coating, the wells were washed with PBS and blocked with 0.1% fatty-acid-free BSA in DMEM for 1 hour at 37°C. Trypsinized cells(5.0×104cells ) resuspended in 100μml serum-free DMEM containing 0.1% fatty-acid-free BSA and plated on 96-well culture plants. In the rhVEGF treated group, cells were stimulated by 30ng?ml-1 rhVEGF for 10 minutes before seeded to the plant. In the rhVEGF+ LY294002 treated group, 20μmol?L-1 LY294002 were added 30 minutes before rhVEGF. After incubated for 8 hours, non-adherent cells were removed by extensively washing with PBS and remaining cells were indicated with DMEM containing 1mg?ml-1MTT for an additional 4 hours. The number of cells were determined by spectrophotometer(OD=490nm) after indicated with DMSO(150μl?well-1)for another 10 minutes(n=10/per group).7 Cell proliferation assay Trypsinized cells(1.0×104cells ) resuspended in 100μml DMEM medium supplemented with 10% FBS were seeded on 96-well culture plates which were covered with fibronectin(5μg/cm2,Roche). After incubated over night with DMEM medium containing 10% FBS, non-adherent cells were removed by extensively washing with PBS. The remaining cells were indicated with DMEM medium containing 0.1%FBS for an additional 48 hours after stimulated by 30ng?ml-1 rhVEGF and/or 20μmol?L-1 LY294002. The number of cells were detected by MTT as described before.8 Migration assay Migration was analyzed with a modified Boyden chamber assay(cell culture inserts with a polycarbonete-filter [PVP,8um pore size, corning incorporated]) covered with growth factor reduced Matrigel(120μg?ml-1).Trypsinized cells(2.0×104) resuspended in 100μml DMEM containing 0 or 20μmol?L-1 LY294002 were added to the upper filter .The lower compartment was filled with 600ul DMEM medium containing 0.1%FBS and 0 or 20ng?ml-1 rhVEGF. Chambers were incubated for 16 hours at 37℃in 5% CO2. Cells on the under side of the filter were quantified by staining the cells with hematoxylin. Four different fields of each filter were photographed and cells were counted.9 Endothelial tube formation assay Endothelial tube formation assay was analyzed with a 24-well culture plates covered with 2mg?ml-1Matrigel(100ul?well-1). Trypsinized cells(5×104cells) resuspended in 500μml DMEM supplemented with 1%FBS were seeded on the matrix. After incubated for 8 hours at 37℃in 5% CO2, Cells were fixed in 4% PFA and stained with Hematoxylin-Eosinde(H&E). Morphological changes were microscopically monitored.10 Oxygen induced retinopathy (OIR) model C57BL/6 mice were obtained and housed in Institute of Laboratory Animals School of Medicine of Beijin University. OIR was induced according to the protocol described by Smith et al. Briefly, postnatal day(P)7 pups and their mothers were transferred from room air to 75%oxygen for 5 days and then returned to room air. From P12, animals were dosed daily with LY294002(at a concentration of 10mg/ml in 5% Tween80 in saline) by intraperitoneal injection. Control mice received equivalent volumes of vehicle(5% Tween80 in saline). Mice were killed at P17 and the eyeballs were enucleated for retinal wholemount and retinal cross-section.11 Retinal Wholemount Preparation and quantifying non-perfusion areasMice were killed after collected the whole blood and the eyeballs were enucleating. After fixed in 4% paraformaldehyde(PFA) for 1 hour at 4℃, cornea and lens, vitreous were excised and retinas were isolated. Retinal vasculature were then stained for adenosine diphosphatase (ADPase) activity. Images of the retinal vasculature were taken using microscope and non-perfusion areas were analyzed on computer with the program NIH Image(n=6/per group) .12 Retinal Cross-Sections and Pre-ILM Nuclei CountingEnucleated eyes were fixed (4%paraformaldehyde in 0.1M PBS)for 48 hours, and 6μm paraffin-embedded sections were cut. Standard hematoxylin and eosin (H&E) staining was performed. six sections were selected randomly and all nuclei present on the vitreous side of the ILM were counted. The average number of nuclei per section was computed.13 Statistical analysis All results were expressed as mean±SD and statistical comparisons were made on computer by SPSS 12.0 software. Result1 RT-PCR analysis of ILK,HIF-1αand VEGF mRNA expressionThe expression of ILK,HIF-1αand VEGFmRNA were increased gradually as the time of hypoxia prolonged. Compared with the expression of mRNA at 4h after hypoxia, the expression of ILK in LY294002 group was not inhibited, however, the expression of HIF and VEGFmRNA were strongly inhibited .Expression of HIF-1αand VEGFmRNA were strongly inhibited as the ILK mRNA was strongly inhibited after siRNA transferred at 4h after hypoxia. However, there were no changes in the negative and blank control groups(figure 1).2 Western blot analysis of ILK and HIF-1αThe expression of ILK,HIF-1αwere markedly increased from the 1st hour to the 4th hour in hypoxic RF/6A cells, and the expression of HIF-1αat the 4th hour in hypoxic RF/6A cells were strongly inhibited following the treatment of 20μmol/L LY294002. HIF-1αexpression were strongly inhibited as the suppression of ILK after siRNA transferred at the 4h after hypoxia. However, there were no obviously change between the negative and blank control groups(figure 2).3 ELISA for VEGF The expression of VEGF was 66.649±7.010pg/ml under the nomoxia medium and elevated to 132.636±17.139pg/ml in hypoxia environment (t=7.539,P<0.01). VEGF reduced to 66.358±8.066pg/ml following the treatment of 20μmol/L LY294002(t=5.798,P<0.01) and to 91.605±9.048pg/ml after siRNA transfer(t=4.688,P<0.01)(F=20.803,P<0.01).4 Cell-adhesion assayThe OD values(490nm) were 0.0726±0.01961 in control group , 0.1137±0.02631 in VEGF treatment group , 0.0837±0.01503 in Ly294002+VEGF group and 0.0853±0.02454 in siRNA+VEGF group(F=6.44,P<0.01). Compared to control, the OD values was significantly increased in VEGF treatment group(t=4.211,P<0.01). However, the OD values was significantly decreased in LY294002+VEGF group and siRNA+VEGF group compared to VEGF treatment group(t=3.074 ,2.91,P<0.01).5 Cell proliferation assayThe OD values(490nm) were 0.4162±0.1392 in control group ,0.6412±0.2420 in VEGF treatment group,0.4476±0.1834 in Ly294002+VEGF group(F=3.99,P<0.05).Compared to control, the OD values was significantly increased in VEGF treatment group(t=2.608,P<0.05). However, the OD values was significantly decreased in Ly294002+VEGF group compared to VEGF treatment group(t=2.244,P<0.05).6 Migration assayThe numbers of cells were 83.66±30.283 in control group and 248±74.748 in VEGF treatment group, 138.5±38.167 in Ly294002+VEGF group(F=15.393,P<0.01).Compared to control, the number of cells was significantly increased in VEGF treatment group(t=5.436,P<0.01). There was also statistically significant reduction in LY294002+VEGF group compared to VEGF treatment group(t=3.682,P<0.01)(figure 3).7 Endothelial tube formation assayAs shown in figure 4, endothelial cells formed hollow tube-like structures in VEGF treatment group. But the endothelium tube-formation were completely inhibited in both groups either treated by LY294002 or treated by siRNA.8 The areas of avasculature in retinal whole mountAs shown in figure 5A, there are normal vasculature from disc to the ora serrata in air-raised mice. However, in hyperoxia-exposured mice, the large radial vessels degenerated and the central retina showed almost no perfusion. At the junction between perfusion and non-perfusion, new vessels grew from remained superficial plexus and extended to central avascular areas. Simultaneously, in mid- peripheral retinas, a great quantity of neovascular tufts could be observed(figure 5B). In this study, we evaluated the neovascularizaion by measuring the areas of non-perfusions in central retinas(NIH Image J version 1.31 software) and neovascular tufts(Pre-ILM Nuclei Counting, as shown in result 9). The areas of non-perfusions in central retinas were significantly increased from 62798±16995.62μm2 in no treatment group to 84722.65±10435.01μm2 in Ly294002 treatment group(F=12.086,t=3.476, P<0.01).9 Pre-ILM Nuclei CountingCompared to the control group, the number of pre-ILM nuclei significantly increased from 2.617±1.549 to 46.43±6.415 in hypoxia group. However, the number of pre-ILM nuclei were significantly decreased in the group of LY294002 from 46.43±6.415 to 35.65±6.697个(F=33.347,t=14.26,3.509, P<0.01)(figure 6).DiscussionVEGF, also called vascular permeability factor(VPF) or vasculotropin, is originally identified as an endothelial-specific mitogen and plays a critical role in neovascularization. The major pathysiological stimulus for VEGF expression in retinal neovascularization is hypoxia, resulting in the transcriptional induction of the VEGF gene by HIF-1α. Because of this, recent investigations of retinal neovascular treatment have focused on inhibiting VEGF expression and activity.In this study, we found that the expression of ILK mRNA and protein were increased gradually from the 1st to the 4th hour in hypoxic RF/6A cells , which was similar to the studies by Scandurro AB et al in human hepatocellular carcinoma cells.As the same, the expression of HIF-1αand VEGF were also increased in hypoxic RF/6A cells. To determine whether ILK is required for HIF-1αand VEGF expression in hypoxic RF/6A cells, we detected the expression of HIF-1αand VEGF at the 4th hour in hypoxic cells, in which time the expressions of ILK,HIF-1αand VEGF were strongly increased. After inhibited the expression of ILK by siRNA transfer, the expression of HIF-1αand VEGF were markedly reduced. Similar to this, the expression of HIF-1αand VEGF were also significantly decreased after the inhibition of ILK activity by LY294002, although the expression of ILK mRNA and protein had no changes. LY294002 is a highly selective small molecule inhibitor of PI-3K activity and ILK is a PI-3 kinase-dependent kinase, so LY294002 can inhibit ILK activity indirectly through the property of inhibiting PI-3K activity. These data suggest that ILK plays a role in elevation of VEGF in hypoxic RF/6A cells.On the other hand, neovascularization is a complex processes including ECM degradation , endothelial cells adhesion, proliferation, migration and new connection of these elements to form tube structure. Since VEGF play a key role in neovasculerization by participating these processes, it is necessary to determine the role of ILK in VEGF- mediated RF/6A cell adhesion, proliferation, migration and tube formation.We have shown that VEGF stimulates the adhesion of RF/6A cells and inhibition of ILK expression or activity results in inhibition of VEGF-mediated RF/6A cell adhesion. Similarly, the same results can be seen in cell proliferation and migration. These data demonstrated that the inhibition of ILK expression/activity has a dramatic effect on RF/6A cell function in response to VEGF, suggesting an essential role of ILK in blood vessel formation.The endothelial tube formation assay is frequently used for the demonstration of angiogenesis in vitro, in which endothelial cells placed on matrigel in the presence of angiogenic factors results in the endothelial cells forming tube-like structures morphologically similar to capillaries. This tube formation represents the contribution of cell adhesion, migration and proliferation. In this study, we used this assay to determine whether ILK is required for VEGF-mediated blood vessel formation. As shown in the result ,RF/6A cells cultured on matrigel formed tube structures in the presence of VEGF, however, it was completely inhibited after the treatment of LY294002 or siRNA transfer. This inhibition appears to be due primarily to inhibition of cell migration and proliferation.These data in RF/6A cell demonstrated that ILK not only plays a role in elevation of VEGF in hypoxic RF/6A cells, but also plays roles in VEGF–stimulated adhesion, proliferation, migration and tube-formation in RF/6A cells. Consequently, ILK plays important roles in two key aspects of retinal neovascularization: VEGF expression and VEGF- stimulated blood vessel formation. These findings suggest that ILK is a promising therapeutic target for the inhibition of retinal neovascularization. So we further detected whether ILK is required for the retinal neovascularization in vivo by using oxygen-induced retinopathy mice model which ILK activity were inhibited by LY294002Our results have shown that the inhibition of ILK activity by LY294002 has a dramatic effect on RF/6A cell function and the effect just similar for the inhibition of ILK expression by siRNA transfer. So we determined the role of ILK in vivo by using OIR mice model in which the ILK activity were inhibited by LY294002. As expected, the data of avascular area and pre-ILM nuclei number suggested that retinal neovascularization were decreased after treatment by LY294002.Our in vitro and in vivo studies demonstrated that ILK plays a role in elevation of VEGF in hypoxic RF/6A cells and VEGF–induced adhesion, proliferation, migration and tube-formation in RF/6A cells. Furthermore, ILK is required for retinal neovascularization in OIR mice models. These data suggest that ILK is a promising therapeutic target for the inhibition of retinal neovascularization in clinic. However, it still a long way to investigate the method of application, the dose of therapy and the toxicity of Ly294002.