Role of vascular endothelial growth factor-165b in the breakdown of the blood-retinal barrier after acute high intraocular pressure in rats

Background: The blood-retinal barrier (BRB) is essential in maintaining the retinal homeostasis of the microenvironment, previous studies have found that BRB breakdown occurs after acute high intraocular pressure (HIOP) in rats, elevated intraocular pressure can induce upregulation of vascular endothelial growth factor-165b (VEGF-165b) protein in the retina, but the role of VEGF-A165b in BRB breakdown after acute HIOP is still undetermined. Methods: In this study, the rat acute HIOP model was established before and after intravitreous injection of anti-VEGF-165b antibody. The expression of VEGF-165b and ZO-1 in rat retina was detected by immunohistochemistry or western blotting, and the breakdown of BRB was detected by Evans blue (EB) dye. Results: The normal retina of rats expressed VEGF-165b protein, which was mainly located in the retinal ganglion cell (RGC) layer and the inner nuclear layer and was coexpressed with tight junction protein ZO-1. After acute HIOP, the expression of VEGF-165b was upregulated (P < 0.01); The expression of ZO-1 was downregulated (P < 0.01) at 12 h and then recovered at 3 d; EB leakage increased, peaking at 12 h (P < 0.01). After intravitreous injection of anti-VEGF-165b antibody, the expression of VEGF-165b protein was significantly downregulated (P < 0.01); and the downregulation of the expression of ZO-1 was more obvious (P < 0.01); EB leakage became more serious, peaking at 3 d (P < 0.01). EB analysis also showed that EB leakage in the peripheral retina was greater than that in the central retina (P < 0.01). Conclusions: The endogenous VEGF-165b protein may protect the BRB from acute HIOP by regulating the expression of ZO-1. The differential destruction of BRB after

animal department, Hunan Agricultural University (Changsha, Hunan, China) and housed in a standard rat cage with unlimited access to water and food with a 12hour continuous light supply every 24 hours. Rats were randomly divided into a normal group and an experimental group. The experimental group consisted of a control group, a 12 h group and a 3 d group. Previous studies [8] found that BRB injury was more obvious in the early stage (within 24 hours) and gradually recovered in the late stage (1-7 days). Therefore, in this study, the 12

Establishment of acute HIOP models
The animal model of acute HIOP was established as previously described [8,9,16,17]. Briefly, animals were anesthetized by intraperitoneal injection of 2% pentobarbital solution (40 mg/kg). A sterile disposable intravenous infusion needle connected to an instillation instrument filled with normal saline was inserted into the anterior chamber. The intraocular pressure of the two eyes was gradually elevated to 120 mmHg and maintained for 60 min, then slowly lowered to the normal level. A drop of levofloxacin eye drops was administered to the conjunctival sac for infection prevention. For the sham surgery group as a control, the needle was inserted into the anterior chamber without elevating the pressure. Rats with cataract or eye inflammation were excluded as unsuccessful models. Each group was composed of 6 rats. 3.

VEGF-A165b antibody administration
Previous studies have reported that endogenous retinal VEGF can be antagonized by intravitreal injection of a VEGF inhibitor [18] or antibody [19]. Therefore, Park HY and his team's method of administration [19] were used in this study. Briefly, 3 μl of VEGF-A165b antibody (5 mg/ml) was intravitreally injected into the eyes of rats of all three groups. After the rats were anesthetized by intraperitoneal injection of 2%

Retinal tissue preparation
The retinal tissue preparation was performed as previously described [20]. Briefly, the rats were sacrificed with excessive 2% pentobarbital solution intravitreal injection (80 mg/kg). For immunofluorescence staining, rats were transcardially perfused with normal saline followed by paraformaldehyde (PF) solution. After perfusion, the eyeballs were dissected out. The eyecups were gradually dehydrated and subsequently embedded in Tissue-Tek optimal cutting temperature medium.
The preparation of retinal tissue for western blotting is summarized as follows: after anesthetizing rats with excessive 2% pentobarbital solution (80 mg/kg), the eyeballs were quickly dissected out over ice, the lens and vitreous body were removed, the retina was scraped and placed into the EP tube, and it was frozen with liquid nitrogen. Then the retina was kept at -80 °C.
The rats were anesthetized by intraperitoneal injection of 2% pentobarbital solution (40 mg/kg) 2 hours before execution. After that, 3% Evans blue (45 mg/kg, E2129, Sigma-Aldrich, CA, USA) was injected into the great saphenous vein within 2 minutes, the eyes and toes of the animals turned blue rapidly, indicating that the dye was evenly distributed. The rats were anesthetized by intraperitoneal injection of 2% pentobarbital solution (80 mg/kg) and transcardially perfused with normal saline followed by PF. After perfusion, the following operations were performed: (1) The eyeball was dissected out immediately, the retina was scraped with a glass curved needle, and the whole retina was mounted onto the slide in darkness. (2) The eyeball was dissected out immediately, and the retina was taken out. With the optic nerve papilla as the center, the retina was cut into four quadrants, laid flat on the glass plate, and cut at the halfway point from the optic nerve papilla; the outer part of the retina is the peripheral part retina, and the inner part is the central retina. Retinas were dried at room temperature and weighed accurately, then put into EP tubes for later use. 6.

Quantitative detection of BRB breakdown with Evans blue
The EB quantitative method was performed as previously reported [8,21] with modification. Briefly, 1 μl of EB (2%) was diluted 1000-fold in 1 ml formamide (F9037, Sigma-Aldrich, CA, USA) to a concentration of 20 ng/ml, then semidiluted 7 times in turn, and a total of 8 standard tubes, including a formamide blank tube, were used to prepare a standard curve of EB in formamide.
Each EP tube with dry retina was incubated with 160 ml formamide in a constanttemperature chamber at 60 °C for 24 hours. Then the extract was centrifuged at 4°C and 12 000 rpm for 30 minutes. The supernatant was taken (150 ml) and divided into three sample tubes (50 ml/tube). The optical density of each standard tube and sample tube was measured using a biospectrophotometer (Eppendorf, Germany) at 620 nm. The concentration of dye was calculated according to the standard curve of EB in formamide. EB (ng) content was standardized to retinal dry weight (mg), expressed as ng/mg. The obtained values are expressed as means and standard errors.

Immunofluorescence staining
Retina sections were incubated with 5% donkey serum for 1 hour at room temperature before being incubated with the primary antibodies in the refrigerator overnight. Subsequently, in a dark chamber, the sections were incubated with the secondary antibodies for 2 hours at room temperature. Sections incubated with 2% donkey serum without a primary antibody were used as a negative control.
VEGF-165b was used for single labeling of retina sections, and dual labeling of VEGF-165b and ZO-1 was performed in sections from the control animal. The antibodies used in this study are shown in Table 1
The supernatant was removed and subjected to protein quantification with a Pierce BCA reagent kit (Thermo Fisher Scientific, 23227). Then, 50 μg of the protein lysate was loaded onto a 10% SDS-PAGE gel for electrophoresis and subsequently transferred to a nitrocellulose (NC, PALL 66485) membrane. After blocking with a 5% milk-PBS solution, the membrane was incubated with the primary antibody (Table   1). Thereafter, the membrane was washed three times and incubated with an HRPconjugated secondary antibody. Membranes developed by incubation with β-actin were used as controls. Protein was visualized using the Pierce ECL reagent kit (Thermo Fisher Scientific, 32132). Quantitative analysis of proteins was carried out on the protein bands with ImageJ (National Institutes of Health) and Microsoft Excel (Microsoft Corp.). The amounts of VEGF-165b and ZO-1 protein were normalized to β-actin. The obtained values are expressed as means and standard errors.

Statistical analysis
Statistical analysis was performed with SPSS18.0 software (Statistical Product and Service Solutions18.0, Al Monk, New York, USA). Paired t tests were used for comparisons between paired data. All the other data were analyzed using one-way ANOVA. P < 0.05 was considered statistically significant.

1.
Dynamic changes in VEGF-165b protein level in rat retina with acute HIOP before and after treatment detected by immunofluorescence staining and western blotting.
Immunofluorescence staining of VEGF-165b showed that it was mainly expressed in the ganglion cell layer and the inner nuclear layer of the retina. The expression of VEGF-165b in the normal retina was similar to that in the control group (image not shown). The relative fluorescence intensity of VEGF-165b in the 12 h group and the 3 d group was stronger than that in the control group (P < 0.01) ( Figure 1A). The result is consistent with that of western blotting ( Figure 2A). Before vitreous injection of anti-VEGF-165b antibody, the expression of VEGF-165b/β-actin in the control group, 12 h group and 3 d group was 0.743 ± 0.023, 1.039 ± 0.035, and 1.974 ± 0.039, respectively. Compared with the control group, the expression of VEGF-165b protein in the 12 h group and the 3 d group was upregulated (P < 0.01).

Localization of ZO-1 in rat retina by double immunofluorescence staining and detection of the dynamic changes in ZO-1 expression before and after treatment by western blotting
Double immunofluorescence staining of VEGF-165b and ZO-1 in rat retinal sections showed that ZO-1 protein was mainly expressed on the retinal ganglion cell layer and both sides of the inner nuclear layer and was coexpressed with VEGF-165b protein ( Figure 1B). Western blotting showed that before vitreous injection of anti-VEGF-165b antibody, the expression of ZO-1/β-actin in the control group, 12 h group and 3 d group was 0.708 ± 0.025, 0.414 ± 0.032, and 0.582 ± 0.034, respectively.
Compared with the control group, the expression of ZO-1 protein in the 12 h group decreased significantly (P < 0.01); the expression of ZO-1 protein in the 3 d group recovered gradually but was still lower than that in the control group (P < 0.01).
After vitreous injection of anti-VEGF-165b antibody, ZO-1/β-actin in the control group, 12 h group and 3 d group was 0.735 ± 0.021, 0.406 ± 0.019, and 0.145 ± 0.026, respectively. Compared with the control group, the expression of ZO-1 protein in the 12 h group and the 3 d group was significantly downregulated (P < 0.01). Compared with the same group before treatment, the expression of ZO-1 protein in the 3 d group was significantly downregulated (P < 0.01).

Detection of blood-retinal barrier leakage in rats by Evans blue staining
After EB was injected into the great saphenous vein of normal rats, the image taken   (Figure 4).

Discussion
The normal retina of rats expressed the protein of VEGF-165b, which was mainly h. After treatment, EB leakage did not decrease but became more serious in the 3 d group. EB tracing on the whole mounted retina and EB quantification also showed that EB leakage in the peripheral retina was greater than that in the peripheral retina.
The expression of VEGF-165b protein was upregulated in the retina of rats with acute HIOP, which was consistent with Ergorul C and his team's report [12]. . After intravitreal injection of anti-VEGF-165b antibody, the decrease in ZO-1 protein was more obvious, indicating that BRB integrity was damaged more seriously after endogenous VEGF-165b was antagonized, suggesting that endogenous VEGF-165b has a protective effect on retinal endothelial cells, which is consistent with the results of Magnussen AL and his team [15]. These results also suggest that VEGF-165b may protect the BRB from acute HIOP by regulating the expression of ZO-1.
Our EB method for quantitative detection of BRB breakdown is not exactly the same as that reported by Xu et al. [21]. After EB was injected into the caudal vein of normal rats, confocal images were taken from whole-retinal mounted slices. Retinal blood vessels and their branches were clearly visible. No red EB spots were found in the extravascular retinal space (Figure 3 A2). After perfusion with 0.9% saline and 4% PF, no red fluorescent spots were found in the retina (Figure 3 A3). If red EB spots or plaques appear in the retina after perfusion, it is the EB that leaks into the retinal tissue space, indicating that BRB has been damaged or broken down. The results observed by this method and the analysis of the relative fluorescence intensity of EB ( Figure 3D) are basically consistent with the results of EB quantitative detection (Figure 4). This method is simple and effective; the difference in relative fluorescence intensity can also reflect the degree of BRB shows that the response of the inner retina and iBRB to hypoxia-ischemia is isotropic. We observed that EB leakage in the peripheral retina was more severe than that in the central retina after acute HIOP before and after treatment, showing that the breakdown of BRB had a regional difference. Tong JB and his team [17] reported that there were site differences in retinal blood supply after acute HIOP, which was associated with selective retinal ganglion cell death, but this only partially explained the difference in retinal ganglion cell vulnerability. Therefore, we speculate that the characteristics of BRB injury may be related to the selective loss of ganglion cells under acute HIOP.

Conclusions
We found that the expression of retinal VEGF-165b was continuously upregulated after acute HIOP, and the damage to BRB was severe in the early stage and recovered in the later stage. After inhibiting the endogenous VEGF-165b protein, the expression of VEGF-165b and ZO-1 was downregulated more obviously, and the damage to BRB was more serious, suggesting that VEGF-165b has a protective role  There is no conflict of interest to be disclosed by the author. The manuscript has been read and approved for publication by all authors.  Table   Table 1. The antibodies used in this study Figure 1 The expression of VEGF-165b and ZO-1 in retina as detected by immunofluorescence staining The ARRIVE Guidelines Checklist 2019-11-06.pdf