To evaluate the effectiveness of reactive oxygen species (ROS)-responsive nanomedicine in suppressing corneal neovascularization (CNV) in vivo.

ROS-responsive nanomedicine (ROS-TK-5/siVEGF), which consists of vascular endothelial growth factor (VEGF) small interfering RNA (siRNA) and thioketal linkage was synthesized by the Michael addition.The cumulative release of siVEGF from nanomedicine under oxidant conditions was assessed by agarose gel electrophoresis.Thirty-nine VEGFR2-luc-KI transgenic mice were used in this study, of which 30 mice were randomly divided into a normal control group, a PBS control group, an ROS-TK-5/NC group, an ROS-TK-5/siVEGF group, and a ranibizumab group, with 6 mice in each group.The ROS levels in the corneal tissue after alkali burning were tested by dihydroethidium (DHE) staining in the other 9 mice.In each group, alkali-burned mice were subconjunctivally injected with 10 μl of a different formula every two days.The effectiveness of nanomedicine in attenuating CNV was evaluated by slit-lamp microscopy and an in vivo imaging system (IVIS) at 7, 14, and 21 days after alkali burning. The use and care of the animals complied with the Statement of the Association for Research in Vision and Ophthalmology (ARVO) and the Guidelines of the Animal Experimental Committee of Liberation Army General Hospital.The study protocol was approved by the Ethics Committee of Liberation Army General Hospital (No.2018-X14-82).

After treathrent with an aqueous solution without ROS, only 5%-10% of the siVEGF was released from the nanoparticles within 10 hours.In contrast, about 70% of the siVEGF was released from the nanoparticles after treatment with 10 mmol/L H₂O₂ within 10 hours.The relative fluorescent intensities in the corneal stromal layer at 7 days and 14 days after alkali burning were 5.403±0.306 and 2.930±0.255, respectively, which was significantly greater than those in the normal control group (1.003±0.015) (both at P<0.05). The CNV areas were statistically different among the four groups at various time points (F_group=49.855, P<0.01; F_time=65.556, P<0.01). The CNV area was significantly reduced in the ROS-TK-5/siVEGF and ranibizumab groups compared with the PBS control and ROS-TK-5/NC groups at 7 days and 14 days after modeling, and the CNV area was more effectively reduced in the ROS-TK-5/siVEGF group than the ranibizumab group at 7 days and 14 days after modeling (all at P<0.05). At day 21 after modeling, the CNV area was significantly reduced in the ROS-TK-5/siVEGF and ranibizumab groups compared to the PBS control and ROS-TK-5/NC groups (all at P<0.05). IVIS showed that the corneal fluorescent intensity was statistically different among the four groups at various times (F_group=27.193, P=0.003; F_time=51.062, P<0.01). The corneal fluorescent intensities were significantly reduced in the ROS-TK-5/siVEGF and ranibizumab groups compared to the PBS control and ROS-TK-5/NC groups at 7 days and 14 days after modeling; in addition, the corneal fluorescent intensity was more effectively reduced in the ROS-TK-5/siVEGF group in comparison with the ranibizumab group at 7 days and 14 days after modeling (all at P<0.05). At 21 days after modeling, the corneal fluorescent intensity was significantly reduced in the ROS-TK-5/siVEGF and ranibizumab groups compared to the PBS control and ROS-TK-5/NC groups (all at P<0.05).

ROS-TK-5/siVEGF nanomedicine effectively attenuates alkali burn-induced CNV formation and appears to have a better effect in comparison with ranibizumab at an early stage.

In vivo imaging system; Alkali burn; Corneal neovascularization; Reactive oxygen species; Nanomedicine