Abstract
Purpose :
Glycoproteins and glycan-binding factors are critical regulators of angiogenesis. We have shown that depletion of the glycoprotein endomucin (EMCN) prevents VEGF-induced endothelial cell (EC) migration and proliferation. While the mechanism is unknown, interactions between EMCN and glycan-binding factors such as galectin-3 (Gal3) may be involved. We hypothesize that Gal3 and EMCN work in tandem to promote VEGFR2 signaling.
Methods :
Knockdown was achieved in human retinal microvascular ECs using siRNA (Dharmacon) against EMCN, Gal3, or control. Gal3 was conjugated to cyanogen beads for immunoprecipitation. EMCN (Abcam), Gal3 (Abcam) and VEGFR2 (R&D) antisera were used for immunocytochemistry (ICC). Following VEGF (10 ng/ml) or Gal3 (50 ug/ml) addition, migration was measured after 15 h; proliferation after 24, 48 and 72 h. Endogenous VEGF was neutralized with ranibizumab (10 ug/ml). Surface proteins were labeled with NHS-SS-Biotin (Life Technology) and visualized by Western blot using VEGFR2 and CD31 (Abcam) antisera.
Results :
EMCN and Gal3 knockdown showed robust depletion: siEMCN (1 ± 0.1 vs 0.01 ± 0.001; p<0.01) and siGal3 (1 ± 0.3 vs. 0.01 ± 0.008; p<0.05). EMCN interacted with Gal3 in a carbohydrate-dependent manner. ICC confirmed co-localization of EMCN and Gal3 with VEGFR2. Gal3 activated HRECs and increased V-CAM (1 ± 0.03 vs 3 ± 0.2; p<0.01) and E-selectin (1 ± 0.07 vs 15 ± 1.2; p<0.01) gene expression. Migration was seen following VEGF (1 ± 0.1 vs 1.7 ± 0.1; p<0.01) and Gal3 addition (1 ± 0.1 vs 1.6 ± 0.03; p<0.05). EMCN depletion prevented VEGF (1 ± 0.1 vs 1.2 ± 0.05) and Gal3-induced migration (1 ± 0.1 vs 1 ± 0.02). In the absence of exogenous VEGF, blocking endogenously produced VEGF by ranibizumab prevented Gal3 migration (37 ± 3 vs 21 ± 0.7; p<0.01), but not proliferation. While surface VEGFR2 was reduced after VEGF stimulation in control ECs (1 ± 0.1 vs 0.3 ± 0.08; p<0.01), EMCN depletion retained surface VEGFR2 (0.8 ± 0.1 vs 0.7 ± 0.1). Loss of Gal3 did not impact VEGFR2 endocytosis (1 ± 0.1 vs 0.3 ± 0.08; p<0.01).
Conclusions :
Results show that EMCN interacts with Gal3 and potentially VEGFR2. VEGF- and Gal3-induced migration are EMCN-dependent. Endogenous VEGF is required for Gal3-mediated migration but not proliferation. EMCN, but not Gal3, regulate VEGF-induced VEGFR2 endocytosis. These results highlight the significance of EMCN and Gal3 in VEGFR2 signaling with applications for angiogenic therapy.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.