1 This study investigated several mechanisms involved in the vasorelaxant effects of (À)-epigallocatechin-3-gallate (EGCG). 2 EGCG (1 mM-1 mM) concentration dependently relaxed, after a transient increase in tension, contractions induced by noradrenaline (NA, 1 mM), high extracellular KCl (60 mM), or phorbol 12-myristate 13-acetate (PMA, 1 mM) in intact rat aortic rings. In a Ca 2 þ -free solution, EGCG (1 mM-1 mM) relaxed 1 mM PMA-induced contractions, without previous transient contraction. However, EGCG (1 mM-1 mM) did not affect the 1 mM okadaic acid-induced contractions. Removal of endothelium and/or pretreatment with glibenclamide (10 mM), tetraethylammonium (2 mM) or charybdotoxin (100 nM) plus apamin (500 nM) did not modify the vasorelaxant effects of EGCG. In addition, EGCG noncompetitively antagonized the contractions induced by NA (in 1.5 mM Ca 2 þ -containing solution) and Ca 2 þ (in depolarizing Ca 2 þ -free high KCl 60 mM solution). 3 In rat aortic smooth muscle cells (RASMC), EGCG (100 mM) reduced increases in cytosolic free Ca 2 þ concentration ([Ca 2 þ ] i ) induced by angiotensin II (ANG II, 100 nM) and KCl (60 mM) in 1.5 mM CaCl 2 -containing solution and by ANG II (100 nM) in the absence of extracellular Ca 2 þ . 4 In RASMC, EGCG (100 mM) did not modify basal generation of cAMP or cGMP, but significantly reversed the inhibitory effects of NA (1 mM) and high KCl (60 mM) on cAMP and cGMP production. 5 EGCG inhibited the enzymatic activity of all the cyclic nucleotide PDE isoenzymes present in vascular tissue, being more effective on PDE2 (IC 50 B17) and on PDE1 (IC 50 B25). 6 Our results suggest that the vasorelaxant effects of EGCG in rat aorta are mediated, at least in part, by an inhibition of PDE activity, and the subsequent increase in cyclic nucleotide levels in RASMC, which, in turn, can reduce agonist-or high KCl concentration-induced increases in [Ca 2 þ ] i .