Helicobacter pylori colonizes the human gastric mucosa, causing inflammation that leads to atrophic gastritis, and it can cause peptic ulcer and gastric cancer. We show that polyphenol administration to mice experimentally infected by H. pylori or treated with VacA toxin can limit gastric epithelium damage, an effect that may be linked to VacA inhibition.Helicobacter pylori chronically infects the gastric mucosa of Ͼ50% of the human population, causing gastritis. The infection can lead to the development of peptic ulcer (27) and gastric mucosa-associated lymphoid tissue lymphoma (6) and increases the risk of gastric cancer in humans (8,11,31). It has been proven that H. pylori infection can cause gastric cancer in animals (9, 33). In both humans and animals, the gastric pathology depends on the virulence of the H. pylori infecting strain and on the genetic background of the host (4, 10, 22). In humans, lifestyle is also relevant (13, 26). The current antibiotic-based therapies are generally effective but can fail due to antibiotic resistance or lack of patient compliance. Thus, there is a continuous effort to develop new tools to fight against this pathogen. In particular, plant extracts or plant-derived substances have been investigated for anti-H. pylori activity in vitro (1,2,14,15,18,20,24) and in vivo (12,19,28).H. pylori expresses several factors that allow host stomach colonization and can play a role in pathogenesis. One of the most important factors is the vacuolating cytotoxin VacA (3, 29). We have already reported that polyphenols inhibit VacAinduced ion/urea conduction and cell vacuolation (30). Also, we have observed the ability of dealcoholized wine and green tea to reduce gastritis in mice experimentally infected by H. pylori (unpublished data). Hop bract extract has been recently reported to exert anti-VacA activity (35). Here, we investigated whether pure polyphenols could influence gastric colonization or gastritis in mice and/or counteract the effects of VacA in vivo.Animal experiments were done in compliance with current law. Data were evaluated by one-tailed Mann-Whitney U test, with P values of Ͻ0.05 considered significant.H. pylori strain SPM 326 type I, expressing the s1/m1 VacA isoform, was used to intragastrically infect specific-pathogenfree CD1 mice (Charles River, Calco, Italy), a well-established H. pylori infection model (17, 32). Noninfected controls received saline only. Starting 24 h before the infection, the animals had free access to drinking water containing 1% glucose and a mixture of 1.5 mg/ml each of tannic acid (average molecular weight, 1,701.20) and n-propyl gallate (Sigma-Aldrich, Milan, Italy) (TAϩNPG). TA and NPG were selected because of their previously shown anti-VacA activity in vitro (30), their water solubility, and the relative abundance of tannins and gallates in common beverages, such as wine and tea. Also, TA is a GRAS (generally regarded as safe) food additive. Based on previous results obtained with wine (unpublished data), the polyphenol concentration cho...