ObjectiveTo observe the effect of conditional knocking out (KO) vascular endothelial growth factor (VEGF) gene on the mouse model of oxygen induced retinopathy (OIR).MethodsThe conditional VEGF KO mice were generated using Cre-Loxp technology, resulting in the deletion of VEGF in a portion of Müller cells permanently in mouse retina. Cre positive was CKO mice, Cre negative was NKO mice. OIR was induced by keeping mice in 75% oxygen at postnatal 7 days (P7) to P12 and in room air from P12 to P17 (each 20 mice for CKO and NKO, respectively). The mice mortality was analyzed. At day P17, the percentage of retinal avascular area was calculated using retinal flat-mounting with fluorescence angiography, the number of vascular endothelial cell nucleus breaking through retinal inner limiting membrane was counted with hematoxylin eosin (HE) staining of retinal sections, and the expression of hypoxia-inducible factor-1α (HIF-1α) was detected by immunofluorescence analysis. ResultsDuring the development of OIR, the mortality rate of CKO mice (65.00%) was higher than that of NKO mice (30.00%) with the significant difference (x2=4.912, P=0.027). At day P17, all the mice retinas were harvested. The retinal fluorescence angiography displayed that the normal retinal vascularization of CKO mice was delayed, and large avascular areas were observed. Meanwhile, rare new vascular plexus was found in CKO mice and the thickness of whole retina decreased dramatically. In contrast, NKO mice developed larger area of normal retinal vascular network structure with higher blood vessel density and more new vascular plexus with obvious fluorescein leakage. The percentage of avascular area in CKO mice [(28.31±11.15)%] was higher than NKO mice [(16.82±7.23)%] with the significant difference (t=2.734, P=0.014). The HE staining of retinal sections indicated smaller counts of vascular endothelial cell nucleus breaking through retinal inner limiting membrane in CKO mice (26.10±6.37) when compared to NKO mice (28.80±7.59) , the difference was significant (t=2.437, P=0.016). The immunofluorescence analysis showed stronger expression of HIF-1α in CKO mice than NKO mice, which was mainly located in the retinal ganglion cell layer.ConclusionsThe local VEGF gene knockout partially inhibits retinal neovascularization in OIR mice. However, it also suppresses the normal retinal blood vascular development with a decrease of OIR mice survival ability.
Objective To observe the inhibitory effect of intravitreal injection of triamcinolone acetonide (TA) on oxygen-induced retinal neovascularization, and to investigate its mechanism. Methods A total of 48 C57BL/6 mice at the age of 7 days were divided into normal group (groupA,n=6), highoxygen group (group B, n=6), TA control group (group C,n=18) and TA highoxygen group (group D,n=18). The retinal neovascularization of group B and D were induced by oxygen. One eye of each mouse of group C and D received an intravitreal injection 2 mu;l (20 mu;g /mu;l) of TA, and the same volume of BSS was injected into the other eye of the mice as BSS control group (group E) and BSS highoxygen group (group F). At postnatal day 17, the retinas were collected and the number of the endothelium cell nuclei of new vessels beyond the inner limiting membrane (ILM) was counted on HE-stained paraffin retina sections. The expression level of vascular endothelial growth factor (VEGF), stromal cellderived factor 1 (SDF-1) and CD14 were measured by immunohistochemical staining. The mRNA expression of VEGF and SDF-1 were detected by real-time RT-PCR. Results The numbers of the endothelium cell nuclei of new vessels beyond the ILM in group A - F were 0, 675, 0, 0, 110 and 688 respectively. In group A and D, it decreased than that in group B and F respectively (t=30.62, 19.532; P＜0.05). There was no difference of VEGF, SDF-1 and CD14 expression between group C and E (t=0.161, 0.284, 0.223; P＞0.05), but the differences were statistically significant between group D and F(t=-2.264, -2.358, -4.897;P＜0.05). There was no difference on mRNA level of VEGF and SDF-1 between group C and E(t=-0.497,-0.709;P＜0.05), but the differences were statistically significant between group D and F(z=-5.137,-4.411;P＜0.05). Conclusion Intravitreal injection with TA can inhibit oxygen-induced retinal neovascularization, down-regulated expression of VEGF and SDF-1 may be the mechanism.
Objective To observe the expression of C9 in laser-induced choroidal neovascularization (CNV) and the inhibitory effect of cobra venom factor (CVF) on CNV. Methods Thirty-six C57BL/6J mice were randomly divided into control group (n=12), laser photocoagulation group (n=12) and CVF group(n=12). The mice in control group had no interference treatment. CNV of the latter 2 groups were induced by photocoagulation. The mice of CVF group received intraperitoneal injection of CVF (0.1 mu;g/g) 2 days before and once a day after laser treatment. Six days after photocoagulation, the examination of fundus fluorescein angiography was used to confirm that CNV existed in the mice of laser the photocoagulation group. Then at 1, 3, 5, 7 days, the C9 expression was observed using SABCFITC histochemical stain and HE stain. The absorbance was measured by Image-Pro Plus 6.0 image software. Results Only in the laser group had formed CNV. At different time points, C9 expression was observed in the laser photocoagulation group but not in the CVF group. The A value showed a significant difference among 3 groups at different time points (F=146.045, P=0.000). The significant difference of A value exited in each group at different time points (F=9725, P=0.000). Conclusions There is C9 expression in laser-induced choroidal neovascularization. CVF can inhibit CNV formation by inducing complement consumption.
Objective To observe the inhibition effect of selective cyclooxygenase2 inhibitor（celecoxib）on the experimental choroidal neovascularization（CNV）. Methods Thirty 8-10 weeks old healthy male Brown-Norway（BN）rats were randomly divided into the control, laser and celecoxib group,with 10 rats in each group. At the dosage of 50 mg/kg, celecoxib was gavaged twice per day. After 7 days, experimental CNV was induced by Krypon laser on laser group and celecoxib group. Fundus fluorescein angiography (FFA) was performed on days 3, 7,14,21,30 after laser photocoagulation.On days 21 after photocoagulation, 5 rats in each group were sacrificed and the relative thickness of CNV membranes, the expression of COX-2, vascular endothelial growth factor(VEGF) and matrix metalloproteinase-2(MMP-2) were studied by histopathologic or immunohistochemistry examination.Results On days 21 after photocoagulation, the incidence of CNV in the celecoxib group is significantly lower than that in the laser group (chi;2=7.1068,P=0.0077); the relative thickness of the CNV membranes in the celecoxib group is reduced 41.38% compared to the laser group, the difference is statistically significant (t=16.760 0，P=0.0000).COX-2,VEGF and MMP-2 expression in the CNV membrane of celecoxib group were significantly lower than in control group (t=5.710 0,5.840 0, 8.020 0; P=0.000 0); the COX-2, VEGF and MMP-2 expressions in choroid and retina of control group were weak. Conclusion Prophylactic celecoxib can reduce the expression of VEGF and MMP-2 by inhibiting COX-2, and prevent the CNV induced by laser photocoagulation.
Objective To observe the effects of CXCR4 inhibitor (AMD3100) combined with anti-vascular endothelial growth factor (VEGF) antibody on experimental choroidal neovascularization. Methods Choroidal neovascularization (CNV) was induced in 48 BrownNorway (BN) rats by Krypton red laser photocoagulation, and those rats were randomly divided into AF564 group (group A), AMD3100 group (group B), combined treatment group (group C) and PBS group (group D), 12 rats in each group. Left eyes were the experimental eyes. The rats of group A-D received intravitreal injection of 5mu;l of AF564, AMD3100, AF564/AMD3100 and PBS after laser photocoagulation respectively. Fourteen days after photocoagulation, fundus fluorescein angiography (FFA), pathological section analysis and choroidal vascular wholemount were used to observe the degree of fluorescein leakage, the relative thickness and areas of CNV. Results Fourteen days after photocoagulation, the scores of fluorescein leakage in group A - D were 2.16plusmn;0.91, 2.16plusmn;0.91, 1.92plusmn;1.03, 1.39plusmn;0.93 respectively. Fluorescein leakage in group A - C was obviously reduced compared to group D (F=12.91,P＜0.001), while fluorescein leakage in group C was reduced compared to group A and B (F=9.21,P＜0.05). The CNV relative thicknesses in group A-D were 1.82plusmn;0.11, 1.90plusmn;0.22, 1.12plusmn;0.12, 2.82plusmn;0.29 respectively. Group A -C had thinner CNV compared to group D (F=5.92，P＜0.001), while group C had thinner CNV compared to group A and B (F=5.16, P＜0.05). The CNV areas in group A -D were (8204plusmn;122), (9332plusmn;211), (6533plusmn;101), and (13644plusmn;255) mu;m2 respectively. Group A -C had smaller CNV area compared to group D (F=147.50，P＜0.001), while group C had smaller CNV area compared to group A and B (F=112.60, P＜0.05). Conclusion Combined treatment with CXCR4 inhibitor and anti-VEGF antibody can inhibit laser-induced CNV significantly.
Objective To establish and evaluate a rat model of nonarteritic anterior ischemic optic neuropathy (NAION). Methods The rats were randomly divided into control group (n=13), sham laser group (n=11) and NAION group (n=23). The right eye was set as the experimental eye. NAION model was induced by directly illuminating the optic nerve (ON) of the right eye with 532 nm green laser, after intravenous infusion with the photosensitizing agent Rose Bengal. Sham laser treatment consisted of illuminating the ON region with 532 nm laser without Rose Bengal injection. Rats in control group underwent no intervention. The appearance of optic disc was observed with funduscope at 12 hours, 1, 3, 7, 28 days post-illumination. The histologic changes in the retina and ON of the NAION model were evaluated qualitatively with hematoxylin and eosin (HE) staining and transmission electron microscopy. The retrograde-labeled retinal ganglion cells (RGC) were counted on photographs taken from retinal flat mounts in a masked fashion. Results The optic disc in NAION eyes were swollen 3 days after photodynamic treatment. HE-stained longitudinal ON sections of NAION revealed vacuolar degeneration on day 3 after induction. Besides, ultrastructural study showed axonal edema and collapsed sheaths in the ischemic optic nerve at the same time point after modeling. ON edema resolved 7 days after induction. The final results revealed optic disc atrophy, extensive axonal loss, severe glial scar, and RGC death in large numbers 4 weeks after modeling. There were no aforementioned manifestations in control and sham laser group. The RGC density of the right eyes was statistically significantly lower in NAION group than that in control group and in sham laser group (t=−14.142, −14.088; P=0.000, 0.000). The survival rate of RGC was statistically significantly lower in NAION group than in control group and in sham laser group (t=−17.048, −16.667; P=0.000, 0.000). There was no difference of RGC density and survival rate of RGC between control and sham laser group (t=0.050, 0.348; P=0.961, 0.731). Conclusion A rat model of NAION was established successfully by photodynamic treatments with Rose Bengal, which induce optic nerve damage and RGC death.
Leber hereditary optic neuropathy (LHON) is a blinding disease caused by mutations in mitochondrial DNA. It is a classic disease model for studying mitochondrial abnormalities. Its main mutation sites are m11778G.A, m.3460G.A and m.14484T.C. LHON cell models are mainly produced by lymphoblasts, fibroblasts, cell hybrids and induced pluripotent stem cells, while LHON animal models are mainly mice, which are produced by rotenone and ND4 mutants. Although the research on the LHON model has achieved good results, there are still many difficulties in constructing an ideal experimental model, which severely limit the exploring to the pathogenesis and therapeutic drugs of LHON. A detailed understanding of the application and characteristics of existing models in LHON will help improve experimental design and construct new models.