C1q/TNF-Related Protein 3 Prevents Diabetic Retinopathy via AMPK-Dependent Stabilization of Blood-Retinal Barrier Tight Junctions

Abstract

Background The impairment of the inner blood-retinal barrier (iBRB) increases the pathological development of diabetic retinopathy (DR), a severe complication in diabetic patients. Identifying approaches to preserving iBRB integrity and function is a significant challenge in DR. C1q/tumor necrosis factor-related protein-3 (CTRP3) is a newly discovered adipokine and a vital biomarker, predicting DR severity. We sought to determine whether and how CTRP3 affects the pathological development of non-proliferative diabetic retinopathy (NPDR). Methods To clarify the pathophysiologic progress of the blood-retinal barrier in NPDR and explore its potential mechanism, a mouse Type 2 diabetic model of diabetic retinopathy was used. The capillary leakage was assessed by confocal microscope with fluorescent-labeled protein in vivo. Furthermore, the effect of CTRP3 on the inner blood-retinal barrier (iBRB) and its molecular mechanism was clarified. Results The results demonstrated that CTRP3 protects iBRB integrity and resists the vascular permeability induced by DR. Mechanistically, the administration of CTRP3 activates the AMPK signaling pathway and enhances the expression of Occludin and Claudin-5 (tight junction protein) in vivo and in vitro. Meanwhile, CTRP3 improves the injury of human retinal endothelial cells (HRMECs) induced by high glucose/high lipids (HG/HL), and its protective effects are AMPK-dependent. Conclusions In summary, we report, for the first time, that CTRP3 prevents diabetes-induced retinal vascular permeability via stabilizing the tight junctions of the iBRB and through the AMPK-dependent Occludin/Claudin-5 signaling pathway, thus critically affecting the development of NPDR.

Keywords: CTRP3; diabetic retinopathy; iBRB; permeability; tight junction proteins.

Figure 1

gCTRP3 ameliorated T2D-induced capillary leakage. (A), Representative images are showing that gCTRP3 (0.5 μg/g/d) reduced Evans blue dye leakage to dermal adjacent tissue in T2D mice. (B), Quantification with column graph analysis (n = 8–10). (C), gCTRP3 failed to affect body weight, blood glucose levels (D), or insulin resistance (E,F) in the diabetic retinopathy group when compared with Sham. All data are shown as means ± SD. n = 18 for each group. ** p < 0.01, *** p < 0.001, **** p < 0.0001; NS, no significance. Blue arrows indicate ear capillary leakage. HFD, high-fat diet. ND, normal diet.

Figure 2

The protective effect of CTRP3 on DR. (A), Representative images showing that CTRP3 reduced the retinal vascular leakage, which was increased in the T2D DR group. (B), Representative images showed no visible neovascularization was evaluated with TRITC-ConcanavalinA signal (intact retinal vessels, red). (C), Significant extravasation of FITC-dextran (vascular leakage, green) was observed in T2D-DR mice, but it was reduced by CTRP3. All data are shown as means ± SD. n = 8–10 for each group. ** p < 0.01, *** p < 0.001; NS, not significant. White arrows indicate retinal vascular leakage. T2D, Type 2 diabetes. DR, diabetic retinopathy.

Figure 3

gCTRP3 protected barrier function of iBRB against HGHL-induced damage. (A), Representative images of tube formation assay showing that gCTRP3 failed to promote HREMC tube formation. (B), Bar graph for tube formation analysis, n = 5. (C), Representative images for xCELLigence electrical conductivity assays showing that gCTRP3 increased cellular connection. (D), Bar graph analysis for xCELLigence conductivity assay. (E), Bar graph analysis for trans-well endothelial permeability assay. (F), Immunoflurencence images showing that HGHL disrupted cell-to-cell junctions in HRMECs staining with Occludin (red), Claudin-5 (red), Nuclei (blue). (G,H), Fluorescence intensity quantification of Occludin (red), Claudin-5. HRMECs were treated with vehicle or HGHL for 24 h followed by CTRP3 administration (3 μg/mL). All data are shown as means ± SD. n = 5 for each group. Bar graph represents analysis from at least 5 independent repeated experiments. * p < 0.05, ** p < 0.01, *** p < 0.001. HGHL, high glucose, and high lipids.

Figure 4

CTRP3 increased Occludin and Claudin-5 expression and protected the integrity of iBRB against HGHL-induced damage. (A), Representative immunoblots showing that Claudin-1, ZO-1, Claudin-5, and Occludin protein levels in HRMECs pre-treated with gCTRP3 (3 μg/mL) followed by HGHL administration. (B), Bar graph analysis for quantification of ZO-1 protein expression. (C), Bar graph analysis for the quantification of Occludin protein expression. (D), Bar graph analysis for quantification of Claudin-1 protein expression. (E), Bar graph analysis for quantification of Claudin-5 protein expression. All data are shown as means ± SD. n = 5–7. Bar graph represent analysis from at least 5 independent repeated experiments. * p < 0.05, ** p < 0.01. *** p < 0.001; NS, no significance.

Figure 5

Inhibition of AMPK blocked CTRP3′s role in the upregulation of Occludin and Claudin 5. (A), Representative immunoblots show AMPK and Akt activation in HREMCs treated with gCTRP3. (B), The bar graph shows the quantification of AMPK and Akt activation. HRMECs were challenged by gCTRP3 (3 µg/mL) for 15 min followed by HGHL administration (n = 5–7). (C), Western blot analysis confirmed successful knockdown of AMPK by siRNA and expression of pAMPK, AMPK, pACC, ACC, Occludin, and Claudin-5. (D), Bar graph for quantification of the level of key molecules. (E), xCELLigence electrical conductivity assays showed the role of AMPK in HREMCs pretreated with CTRP3 followed by the HGHL challenge. (F), Bar graph for analysis of xCELLigence electrical conductivity assay. n = 5–7. Bar graph represents analysis from at least 5 independent repeated experiments. All data are shown as means ± SD. * p < 0.05, ** p < 0.01; NS, no significance. HGHL, high glucose and high lipids.

Figure 6

AMPK deficiency blocked CTRP3 upregulated the protein level of Occludin, Claudin-5 in DR. (A): Representative Western blots of pAMPK, AMPK, pACC, ACC, Claudin-1, ZO-1, Claudin-5 and Occludin in the retina from Sham and T2D/STZ DR mice with/without gCTRP3 treatment and Compound (C) administration. (B), Bar graph for quantification of the expression of pAMPK, AMPK, pACC, and ACC. (C), Bar graph for quantification of the expression of Claudin-1, ZO-1, Claudin-5, and Occludin. (D), Diagram depicts mechanism responsible for the iBRB protective effects of CTRP3 on diabetic retinopathy. n = 6–8. Bar graph represents analysis from at least 6 independent repeated experiments. Arrow pointing up indicates upregulation; Arrow pointing down indicates downregulation. All data are shown as means ± SD. * p < 0.05, ** p < 0.01. *** p < 0.001; NS, no significance. DR, diabetic retinopathy. T2D, Type 2 diabetes. Comp. C, Compound C.