Protective effect of perindopril on diabetic retinopathy is associated with decreased vascular endothelial growth factor-to-pigment epithelium-derived factor ratio: involvement of a mitochondria-reactive oxygen species pathway

Abstract

Objective: This study aimed to verify whether the decreased vascular endothelial growth factor (VEGF)-to-pigment epithelium-derived factor (PEDF) ratio can serve as an indicator for the protective effect of angiotensin-converting enzyme inhibitors (ACEIs) on diabetic retinopathy (DR) and to investigate the role of mitochondrial reactive oxygen species (ROS) in the downregulated VEGF-to-PEDF ratio.

Research design and methods: Diabetic rats and control animals were randomly assigned to receive perindopril or vehicle for 24 weeks, and bovine retinal capillary endothelial cells (BRECs) were incubated with normal or high glucose with or without perindopril. VEGF, PEDF, PPARgamma, and uncoupling protein-2 (UCP-2) in the rat retinas or BREC extracts were examined by Western blotting and real-time RT-PCR. The levels of VEGF and PEDF in cell culture media were examined by ELISA. Mitochondrial membrane potential (Deltapsim) and ROS production were assayed using JC-1 or CM-H2DCFDA.

Results: The VEGF-to-PEDF ratio was increased in the retina of diabetic rats; perindopril lowered the increased VEGF-to-PEDF ratio in diabetic rats and ameliorated the retinal damage. In BRECs, perindopril lowered the hyperglycemia-induced elevation of VEGF-to-PEDF ratio by reducing mitochondrial ROS. We found the decreased ROS production was a result of perindopril-induced upregulation of PPARgamma and UCP-2 expression and the subsequent decrease of Deltapsim.

Conclusions: It is concluded that the protective effect of ACEI on DR is associated with a decreased VEGF-to-PEDF ratio, which involves the mitochondria-ROS pathway through PPARgamma-mediated changes of UCP-2. This study paves a way for future application of ACEI in treatment of DR.

FIG. 1.

Changes of retinal VEGF and pigment epithelium-derived factor (PEDF) levels in the control rats (control), diabetic rats (DM), and diabetic rats treated with perindopril (DM + P). A: Real-time RT-PCR determination of VEGF and PEDF mRNAs relative to the control values. B: Western blotting analysis of VEGF and PEDF protein expression in the three groups. Equal protein loading was confirmed with the β-actin antibody. C: VEGF-to-PEDF ratios (protein) in the three groups. Levels of VEGF and PEDF were first normalized by β-actin at each time point. The normalized VEGF level was divided by the corresponding PEDF levels for VEGF-to-PEDF ratio. Data are means ± SD from eight rats per group, and the experiments were repeated independently at least three times with similar results. **P < 0.01 versus control, #P < 0.05 versus DM.

FIG. 2.

Observation of early retinal histopathological lesions in the control, diabetic rats (DM), and diabetic rats treated with perindopril (DM + P) groups. A and B: Low- and high-magnification (×50 and ×400) photomicrographs of trypsin-digested retinal blood vessels obtained from the three groups. All preparations were stained with PAS and hematoxylin; arrowheads indicate acellular capillary. C: Transmission electron micrograph of a capillary from outer plexiform layer of rat retinas in the three groups. Arrows denote the segment of the outer capillary basement membrane between the endothelial cells and glia limitans, which was used to measure basement membrane width (original magnification ×9,000). D–F: Determination of pericyte/mm² of capillary area (D), acellular capillary segment/mm² in the retinal vessels (E), and retinal capillary basement membrane thickness (BMT, nm) (F) in the three groups. Data are means ± SD from eight rats per group, and the experiments were repeated independently at least three times with similar results. **P < 0.01 versus control, #P < 0.05 versus DM. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 3.

Retinal mitochondrial ROS production and membrane potential (the mitochondria suspensions were diluted to 1 mg protein/ml) and levels of UCP-2 and PPARγ in the control, diabetic rats (DM), and diabetic rats treated with perindopril (DM + P) groups. A: Mitochondrial ROS production in the retinas of the three groups was identified by the fluorescent probe CM-H2DCFDA. B: A correlation analysis between VEGF-to-PEDF ratio and mitochondrial ROS level in the perindopril-treated group. C: Mitochondrial membrane potential (Δψm) in the three groups was examined by the molecular probe JC-1. D: Western blotting analysis of UCP-2 protein and PPARγ protein expression in the three groups. Equal protein loading was confirmed with the β-actin antibody. Data are means ± SD from eight rats per group, and the experiments were repeated independently at least three times with similar results. *P < 0.05 versus control, **P < 0.01 versus control, #P < 0.05 versus DM, **P < 0.01 versus DM.

FIG. 4.

ROS production and levels of VEGF and PEDF in BRECs exposed to normal glucose (NG) or high glucose (HG). A: Intracellular ROS production was determined by the fluorescent probe CM-H2DCFDA in BRECs exposed to NG, HG, HG + perindopril (HG + P), HG + UCP-2 sense, HG + UCP-2 antisense, or HG + P + UCP-2 antisense for 24 h. B: VEGF and PEDF mRNAs in BRECs were determined by real-time RT-PCR in NG, HG, HG + P, or HG with NAC (HG + NAC) for 72 h. Results are expressed relative to the NG values. C and E: Western blotting analysis of VEGF and PEDF protein expression in BREC extracts (C) and ELISA analysis of VEGF and PEDF protein expression in BREC media (E) in the four groups. D and F: VEGF-to-PEDF ratios (protein in BRECs extracts [D] and in BREC media [F]) in the four groups. Levels of VEGF and PEDF were first normalized by β-actin at each time point in Western blotting; the normalized VEGF level was divided by the corresponding PEDF levels for calculation of VEGF-to-PEDF ratio. Data are means ± SD from nine cells per group, and the experiments were repeated independently at least three times (real-time RT-PCR and Western blotting) or two times (ELISA) with similar results. **P < 0.01 versus NG, ##P < 0.01 versus HG.

FIG. 5.

Levels of VEGF and PEDF in BRECs exposed to normal glucose (NG) or NG + H₂O_2. A: VEGF and PEDF mRNAs in BRECs were determined by real-time RT-PCR in the control group and the H₂O₂-treated group. Results are expressed as relative to the NG values. B and D: Western blotting analysis of VEGF and PEDF protein expression in BREC extracts (B) and ELISA analysis of VEGF and PEDF protein expression in BREC media (D) in the two groups. C and E: The VEGF-to-PEDF ratios (protein in BRECs extracts [C] and in BREC media [E]) in the two groups. Data are means ± SD from nine cells per group, and the experiments were repeated independently at least three times (real-time RT-PCR and Western blotting) or two times (ELISA) with similar results. **P < 0.01 versus NG.

FIG. 6.

Mitochondrial membrane potential (Δψm) and UCP-2 level in BRECs. A: Δψm in BRECs exposed to normal glucose (NG), high glucose (HG), HG + perindopril (P), HG + UCP-2 sense, HG + UCP-2 antisense, and HG + P + UCP-2 antisense for 24 h was examined by the molecular probe JC-1. B: A correlation analysis between intracellular ROS production and Δψm level in BRECs exposed to HG. C: UCP-2 mRNAs in BRECs exposed to NG, NG plus perindopril (NG + P), HG, HG plus perindopril (HG + P), HG plus NAC (HG + NAC), and HG plus NAC and perindopril (HG + NAC+ P) were determined by real-time RT-PCR. Results are expressed relative to the NG values. D: Western blotting analysis of UCP-2 protein expression in BRECs in NG, NG + P, HG, HG + P, HG + NAC, and HG + NAC + P groups. Equal protein loading was confirmed with the β-actin antibody. Data are means ± SD from nine cells per group, and the experiments were repeated independently at least three times with similar results. *P < 0.05 versus NG, **P < 0.01 versus NG, #P < 0.05 versus HG, ##P < 0.01 versus HG, $P < 0.05 versus HG + NAC.

FIG. 7.

Levels of PPARγ and UCP-2 in BRECs. A and B: PPARγ1 (A) and PPARγ2 (B) mRNAs in BRECs were determined by real-time RT-PCR in normal glucose (NG), NG + perindopril (P), high glucose (HG), and HG + P for 24 h. Results are expressed relative to the NG values. C: Western blotting analysis of PPARγ protein expression in the four groups. D: Western blotting analysis of UCP-2 protein expression in NG, HG, HG with GW9662 (HG + GW9662), HG + P, and HG plus P and GW9662 (HG + P + GW9662) groups. Equal protein loading was confirmed with the β-actin antibody. Data are means ± SD from nine cells per group, and the experiments were repeated independently at least three times with similar results. *P < 0.05 versus NG, **P < 0.01 versus NG, #P < 0.05 versus HG, $$P < 0.01 versus HG + P.