Title

 

Authors

A. Toniolo

PhD School in Pharmacological Sciences – Pharmacology Toxicology and Therapy
Dept. of Pharmacology and Anesthesiology, University of Padova 

 

Abstract

We previously demonstrated that the production of prostacyclin (PGI₂) by endothelial cells undergoing different modalities of oxidative stress is mediated by COX-1 and not by COX-2 (Bolego et al; 2009). Epigallocatechin-3-gallate (EGCG), the main catechin of green tea, produces multiple protective effects in the cardiovascular system mainly attributed to its antioxidative properties. However, it has been also demonstrated that EGCG is susceptible to auto-oxidation and production of reactive oxygen species (ROS), which in turn contribute to cytoprotective effects (Elbling et al., 2010) and endothelium-dependent relaxation (Kim et al., 2007). The present study is aimed at evaluating the effects of EGCG on endothelial PGI₂ production using both in vitro and ex vivo models of endothelium. Human umbilical vein endothelial cells (HUVECs) from healthy subjects were used at second passage. Cells challenged with EGCG induced a concentration- (10-100µM) and time- (15’-60’) dependent production of ROS as measured by flow cytometry, using 6-carboxy-2’-7’-dichlorofluorescein diacetate (10 µM). Addition of catalase (300 U/ml) completely abolished the generation of ROS in EGCG treated cells (p<0.05 vs ECGC alone). Conversely, pretreating HUVECs with either the NADPH inhibitor diphenyleneiodonium (DPI, 1 µM) or SOD (0.2 U/ml) did not significantly affect the EGCG-mediated ROS production. The formation of 6-keto-PGF1α, the stable hydrolysis product of PGI₂, was measured in culture medium samples by specific EIA. In the presence of exogenous arachidonic acid and EGCG (100 µM, 30’), PGI₂ production by HUVECs was concentration-dependently inhibited by the selective COX-1 inhibitor SC560 (1-100nM, p<0.001 vs control), but not by the COX-2 inhibitor SC236. In addition, EGCG caused concentration-dependent vasorelaxation (0.1-100 µM) in rat aortic rings precontracted with 10 µM phenylephrine. Pre-incubation of aortic rings with L-NAME (10 µM) only partially abolished EGCG-mediated vasodilating response (p<0.05), while the curve returned to control values in the presence of SC560 (10 nM), pointing to a critical involvement of COX-1-derived PGI₂ in EGCG-mediated vasorelaxation. In agreement with the functional data, SC560 (10 nM) completely prevented the increase in PGI₂ released into the medium from EGCG-treated aortic rings. In conclusion, we here demonstrated that ECGC generate a large amount of ROS, mostly H₂O₂, likely resulting from auto-oxidation of the molecule. Moreover, we provide evidence that in the presence of EGCG-derived hydroperoxides COX-1, and not COX-2, is responsible for endothelial PGI₂ production. We also demonstrated that EGCG-mediated endothelium-dependent vasorelaxation requires COX-1-derived PGI₂. Overall, these data further support the notion that in conditions of altered oxidative tone COX-1 is the main activity responsible for the production of endothelial prostacyclin.

References
Bolego C et al. 2009 FASEB J. 23:605-12.
Elbling L et al. 2010 Free Radic Biol Med. 49:1444-52.
Kim JA et al. 2007 J Biol Chem. 282:13736-45.