Fig 2: Melatonin prevents retinal pericyte loss in diabetes retinopathy. (A) Representative images of IB4 (green) and NG2 (gray) immunolabeling in normal control and diabetic rat retinas with or without melatonin treatment. (B) Quantification of pericyte coverage. (C) NG2 expression in normal control and diabetic rat retinas with or without melatonin treatment (n = 4 retinas per group). (D) Representative images of IB4 (green), IBA-1 (red), and NG2 (blue) immunolabeling in normal control and 8-week diabetic rat retinas with or without melatonin treatment. Endothelial loss is indicated by white arrowheads. Data are expressed as mean ± SEM (n = 3 retinas per group). N, normal control group; M, melatonin treatment group; D, diabetes group; D+M, diabetes with melatonin treatment group; **p < 0.01, ***p < 0.001 by one-way ANOVA. Scale bar, 50 µm.

Fig 3: Melatonin protects the barrier function of pericytes and endothelial cells in vitro. (A) HBVPs were cocultured with HRMECs for 72 h, and then the supernatant of CoCl2-treated BV2 cells was added into the upper chamber of transwell. (B) The transendothelial electrical resistance (TEER) was detected in the HBVP/HRMEC coculture system under different conditions (n = 3 per group). (C) The FITC-dextran leakage was detected in the HBVP/HRMEC coculture system under different conditions (n = 3 per group). (D) The expressions of pericyte markers, including a-SMA, NG2, and PDGFR-ß in CoCl2-treated HBVPs with or without melatonin for 12 h (n = 6 per group). (E) Cell viability of CoCl2-treated HBVPs with or without melatonin for 12 h (n = 6 per group). (F) Cell death of pericytes in CoCl2-treated HBVPs with or without melatonin for 12 h (n = 3 per group). (G, H) Representative images of Claudin-5 (red) and ZO-1 (green) immunostaining in CoCl2-treated HRMECs with or without melatonin. The nuclei were stained with DAPI (blue). Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA. Scale bar, 40 µm.

Fig 4: Hippo signaling regulates CSPG4 levels.(A) Depiction of the Hippo signaling pathway and the effect of inhibition by XMU-MP-1 and TRULI. (B) CSPG4 transcript levels in HeLa and HT-29 cells treated with 3 µM XMU-MP-1 for 7 h. CSPG4 transcript levels are presented as mean (n = 3) ± S.D. (C) Immunoblot detection of CSPG4 from HeLa and HT-29 cells exposed to 6 µM of XMU-MP-1 for 24 h. (D) CSPG4 transcript levels in HeLa and HT-29 cells treated with 30 µM TRULI for 24 h. CSPG4 transcript levels are presented as mean (n = 3) ± S.D. (E) Immunoblot detection of CSPG4 from HeLa and HT-29 cells exposed to 30 µM of TRULI for 24 h. (F) CSPG4 transcript levels in HeLa and YAP1 knock out cells (HeLa?YAP1). CSPG4 transcript levels are presented as mean (n = 3) ± S.D. (G) Immunoblot analyzing YAP1 and CSPG4 levels in HeLa cells and HeLa?YAP1. *p < 0.05 determined by Student’s t-test.

Fig 5: CSPG4 levels in HeLa and HeLa R5 cells.(A) Flow cytometry analysis of HeLa and HeLa R5 cells stained with antibodies against CSPG4. (B) Immunoblot analysis of CSPG4 and GAPDH levels using protein lysates from HeLa and HeLa R5 cells. (C) RNA was extracted from HeLa and HeLa R5 cells, and RT-qPCR was used to quantify CSPG4 transcripts. RT-qPCR data are presented as mean (n = 3) ± S.D. (D) Immunoblot analysis of CSPG4 in HeLa R5 and HeLa R5 cells transfected with pEF6-CSPG4-myc-his. (E) Comparison of TcdB2 CPE in HeLa R5 and HeLa R5 (pEF6-CSPG4-myc-his) after treatment with 4 pM of toxin for 24 h. * p < 0.05 determined by Student’s t-test.