The gastroprotective effect of DDF (3,6-dimethoxy-6'', 6''-dimethyl-[2'', 3'' : 7,8]-chromeneflavone) from Lonchocarpus araripensis Benth. (Leguminosae) on gastric damage induced by absolute ethanol (96%, 0.2 mL/mouse) and indometacin (30 mg kg(-1), p.o.) in mice was investigated. Intraperitoneally administered DDF at dose levels of 50, 100 and 200 mg kg(-1) markedly reduced the gastric lesions in the ethanol model by 62, 72 and 96%, and in the indometacin model by 34, 70 and 75%, respectively, as compared with misoprostol (50 microg kg(-1), p.o.), the reference compound that caused lesion suppression by 67% in ethanol model and by 72% against indometacin-induced ulceration. The ED50 of DDF in reducing gastric lesions induced by ethanol and indometacin (dose of the DDF that reduced the gastric lesion area by 50% in relation to the control value) was 50.87 and 61.56 mg kg(-1), respectively. Mechanistic studies were carried out at 100 mg kg(-1) DDF using the ethanol model. Compared with N-acetylcysteine (750 mg kg(-1), p.o.), a donor of sulfhydryls, DDF only partially replenished the ethanol-induced depletion of gastric mucosal NP-SH. Pretreatment with TRPV1 antagonist capsazepine (5 mg kg(-1), i.p.) or the non-selective cyclooxygenase inhibitor indometacin (10 mg kg(-1), p.o.) effectively blocked the gastroprotective effect of DDF (100 mg kg(-1)) against ethanol damage. Furthermore, the effect of DDF was significantly reduced in mice pretreated with L-NAME, or glibenclamide, the respective inhibitors of nitric oxide synthase and K+ ATP channel activation. These data provide evidence to show that DDF affords gastroprotection against gastric damage induced by ethanol and indometacin by different and complementary mechanisms, which include involvement of endogenous prostaglandins, nitric oxide release, the activation of TRPV1 receptor or K+ ATP channels, besides a sparing effect on NP-SH reserve.