Systemic Rho-kinase inhibition using fasudil in mice with oxygen-induced retinopathy.

PURPOSE: To investigate the influence of the selective Rho-kinase (ROCK) inhibitor, fasudil, on the mRNA level of proinflammatory factors and the retinal vascular development in mice with oxygen-induced retinopathy (OIR). METHODS: C57BL/6J mice underwent standard protocol for OIR induction from postnatal days 7 to 12. Subsequently, they received a daily intraperitoneal injection of fasudil or sodium chloride from P12 to P16. Analyses were performed using vascular staining on retinal flat mounts, RNA expression by qPCR, and immunohistochemistry on paraffin sections. RESULTS: On retinal flat mounts, the proportion of avascular area and tuft formation did not differ between the fasudil and NaCl group. Immunohistochemical staining revealed a less intense staining with inflammatory markers after fasudil. Nevertheless, there were no differences on RNA level between the two groups. CONCLUSIONS: In conclusion, our findings support that daily systemic application of fasudil does not decrease retinal neovascularization in rodents with oxygen-induced retinopathy. The results of our study together with the controversial results on the effects of different ROCK inhibitors from the literature makes it apparent that effects of ROCK inhibition are more complex, and further studies are necessary to analyze its potential therapeutic effects.

Local partial depletion of CD11b+ cells and their influence on choroidal neovascularization using the CD11b-HSVTK mouse model.

PURPOSE: To assess the influence of retinal macrophages and microglia on the formation of choroidal neovascularization (CNV). Therefore, we used a transgenic mouse (CD11b-HSVTK) in which the application of ganciclovir (GCV) results in a depletion of CD11b+ cells. METHODS: We first investigated if a local depletion of CD11b+ macrophages and microglia in the retina is feasible. In a second step, the influence of CD11b+ cell depletion on CNV formation was analysed. One eye of each CD11b-HSVTK mouse was injected with GCV, and the fellow eye received sodium chloride solution (NaCl). Cell counting was performed at day 3 and 7 (one injection) or at day 14 and 21 (two injections). Choroidal neovascularization (CNV) was induced by argon laser and analysed at day 14. RESULTS: The most effective CD11b+ cell depletion was achieved 7 days after a single injection and 14 days after two injections of GCV. After two injections of GCV, we found a significant reduction of CD11b+ cells in central (52 ± 23.9 cells/mm2 ) and peripheral retina (53 ± 20.6 cells/mm2 ); compared to eyes received NaCl (216 ± 49.0 and 210 ± 50.5 cells/mm2 , p < 0.001, respectively). Regarding CNV areas, no statistical significance was found between the groups. CONCLUSION: The CD11b-HSVTK mouse is a feasible model for a local depletion of CD11b+ cells in the retina. Nevertheless, only a partial depletion of CD11b+ cells could be achieved compared to baseline data without any intravitreal injections. Our results did not reveal a significant reduction in CNV areas. In the light of previous knowledge, the potential influence of systemic immune cells on CNV formation might be more relevant than expected.

Spatial distribution of CD115+ and CD11b+ cells and their temporal activation during oxygen-induced retinopathy in mice.

PURPOSE: The model of oxygen-induced retinopathy (OIR) is widely used to analyze pathomechanisms in retinal neovascularization. Previous studies have shown that macrophages (MP) play a key role in vessel formation in OIR, the influence of microglia (MG) having been discussed. The aim of our study was to analyze the spatial and temporal distribution and activation of MP/MG expressing CD115 and CD11b during the process of neovascularization in OIR. METHODS: We used MacGreen mice expressing the green fluorescence protein (GFP) under the promoter for CD115. CD115+ cells were investigated in vivo by scanning laser ophthalmoscopy at postnatal days (P) 17 and 21 in MacGreen mice with OIR (75% oxygen from P7 to P12), and were compared to MacGreen room-air controls. In addition MP/MG were examined ex vivo using immunohistochemistry for CD11b+ detection on retinal flatmounts at P14, P17, and P21 of wild type mice with OIR. RESULTS: In-vivo imaging revealed the highest density of activated MP/MG in tuft areas at P17 of MacGreen mice with OIR. Tufts and regions with a high density of CD115+ cells were detected close to veins, rather to arteries. In peripheral, fully vascularized areas, the distribution of CD115+ cells in MacGreen mice with OIR was similar to MacGreen room-air controls. Correspondingly, immunohistochemical analyses of retinal flatmounts from wild type mice with OIR induction revealed that the number of CD11b+ cells significantly varies between vascular, avascular, and tuft areas as well as between the retinal layers. Activated CD11b+ cells were almost exclusively found in avascular areas and tufts of wild type mice with OIR induction; here, the proportion of activated cells related to the total number of CD11b+ cells remained stable over the course of time. CONCLUSIONS: Using two different approaches to monitor MP/MG cells, our findings demonstrated that MP/MG concentrate within pathologically vascularized areas during OIR. We were able to clarify that reactive changes of CD11b+ cell distribution to OIR primarily occur in the deep retinal layers. Furthermore, we found the highest proportion of activated CD11b+ cells in regions with pathologic neovascularization processes. Our findings support previous reports about activated MP/MG guiding revascularization in avascular areas and playing a key role in the formation and regression of neovascular tufts.

Intravitreal inhibition of complement C5a reduces choroidal neovascularization in mice.

PURPOSE: To investigate the influence of complement component C5a inhibition on laser-induced choroidal neovascularization (CNV) in mice using a C5a specific L-aptamer. METHODS: In C57BL/6 J mice CNV was induced by argon-laser, C5a-inhibitor (NOX-D20) was intravitreally injected in three concentrations: 0.3, 3.0, and 30 mg/ml. The unPEGylated derivate (NOX-D20001) was applied at 3.0 mg/ml; the vehicle (5 % glucose) was injected in controls. Vascular leakage was evaluated using fluorescence angiography, CNV area was examined immunohistochemically. Activated immune cells surrounding the CNV lesion and potential cytotoxicity were analyzed. RESULTS: Compared to controls, CNV areas were significantly reduced after NOX-D20 injection at a concentration of 0.3 and 3.0 mg/ml (p = 0.042; p = 0.016). NOX-D20001 significantly decreased CNV leakage but not the area (p = 0.007; p = 0.276). At a concentration of 30 mg/ml, NOX-D20 did not reveal significant effects on vascular leakage or CNV area (p = 0.624; p = 0.121). The amount of CD11b positive cells was significantly reduced after treatment with 0.3 and 3.0 mg/ml NOX-D20 (p = 0.027; p = 0.002). No adverse glial cell proliferation or increased apoptosis were observed at effective dosages. CONCLUSIONS: Our findings demonstrate that the targeted inhibition of complement component C5a reduces vascular leakage and neovascular area in laser-induced CNV in mice. NOX-D20 was proven to be an effective and safe agent that might be considered as a therapeutic candidate for CNV treatment. The deficiency of activated immune cells highlights promising new aspects in the pathology of choroidal neovascularization, and warrants further investigations.