Epigallocatechin-3-gallate (EGCG), the most abundant polyphenolic compound of green tea, exerts anti- or pro-oxidant activities in endothelial cells depending on pre-existing redox conditions.

Anti-atherosclerotic and anti-hypertensive effects of EGCG have been related to its anti-oxidant activity on endothelial function (1). Interestingly, EGCG may increase- rather than decrease - production of reactive oxygen species (ROS) under certain conditions (2). We hypothesized that anti- or pro-oxidant activities of EGCG may depend on previous cellular redox state, and evaluated whether EGCG-dependent ability to modulate cellular redox balance in bovine aortic endothelial cells (BAEC) might differ under healthy or pathophysiological conditions. These latter were obtained by mimicking endothelial oxidative stress produced by hyperglycemia (HG; 55 mM/48 h) and insulin resistance state (insulin 100 nM plus wortmannin IW; 100 µM/24 h). Time- (30min – 48h) and dose-response (1-100 µM) effects to EGCG were investigated in BAEC under normal or oxidative stress conditions. For all conditions, total and phosphorylated levels of eNOS, AMPK and LKB1 were analyzed by immunoblotting and compared in the absence and in the presence of the AMPK- inhibitor Compound C (CC, 20 µM/2h). For ROS detection, intracellular H2O2 and superoxide O2- levels were measured by CM-H2DCF-DA and DHE fluorescent probes, respectively. In each experimental set, quantitative analysis of red or green intensity fluorescence compared to respective basal levels were performed and results expressed as means ± SE of n independent experiments. Student’s t test was used and values of P < 0.05 were considered statistically significant. In BAEC under healthy conditions, acute (1 h) stimulation with EGCG elicited beneficial effects by increasing eNOS, AMPK and LKB1 phosphorylation levels, whereas prolonged (3 h) EGCG-stimulation enhanced production of intracellular ROS and impaired cell survival. In HG- or IW-treated BAEC, nuclear O2- and cytoplasmic H2O2 concentrations were significantly increased over basal levels. Interestingly, EGCG stimulation significantly decreased ROS production to levels comparable with those observed in BAEC pre-treated with conventional anti-oxidants such as SOD, apocynin, or catalase. Importantly, CC-mediated inhibition of AMPK activity significantly blunted the anti-oxidant effects of EGCG in BAEC under IW conditions, but had no effect in EGCG-treated BAEC under healthy conditions. Figure 1 shows the intensity fluorescence levels measured in EGCG-treated BAEC in the absence and in the presence of CC, under healthy and IW-conditions (n = 4; *P < 0.05 vs. basal; °P < 0.05 vs. EGCG+CC). Our findings suggest that EGCG behaves mainly as anti-oxidant under pre-existing oxidative stress conditions, but it may become pro-oxidant when cellular ROS production is low. Moreover, AMPK activity plays a crucial role on antioxidant effects of EGCG. Overall, potential detrimental effects of EGCG obtained here may warn on inappropriate consumption of natural products based on alleged, but incompletely characterized, anti-oxidant properties. 1. Stangl V, Dreger H, Stangl K, et al. (2007) Cardiovasc Res 73, 348-358 2. Elbling L, Weiss RM, Teufelhofer O, et al (2005) FASEB J 19, 807-809