AGA Abstracts study we examine the role of mPGES1 in the acid-induced increase in PGE2 production and cell proliferation, and studied the role of NADPH oxidase NOX5-S in acid-induced upregulation of mPGES1 in an EA cell line FLO. RT-PCR shows that mPGES1, mPGES2 and cytosolic PGES (cPGES) are present in FLO cells. Pulsed acid treatment increased mPGES1 mRNA by 110% and mPGES2 by 10%, but did not have any effect on cPGES mRNA. Acid treatment significantly increased mPGES1 protein expression. Knockdown of mPGES1 by mPGES1 siRNA blocked acid-induced increase in PGE2 production and thymid- ine incorporation (an indicator of cell proliferation rate). Knockdown of NOX5-S by NOX5 siRNA significantly inhibited acid-induced increase in mPGES1 expression, PGE2 production and thymidine incorporation. Overexpression of NOX5-S significantly increased the luciferase activity in FLO cells transfected with a NF-κB In Vivo activation reporter plasmid pNF-κB- Luc. Knockdown of NF-κB1 p50 by p50 siRNA almost abolished acid-induced increase in mPGES1 expression, PGE2 production and thymidine incorporation. In a chromatin immunoprecipitation assay, the promoter region of mPGES1 DNA was detectable in the immunoprecipitated chromatin sample of FLO cell lysate with a p50 antibody, indicating that p50 binds to mPGES1 promoter. We conclude that mPGES1 mediates acid-induced increase in PGE2 production and cell proliferation. Acid-induced mPGES1 expression depends on activation of NOX5-S and NF-κB1 p50. Microsomal PGES1 may be a potential target to prevent or treat EA. Supported by NIH NIDDK R01 DK080703. Sa1081 Signaling in H 2 O 2 -Induced Increase in Cell Proliferation in Barrett's Esophageal Adenocarcinoma Cells Xiaoxu Zhou, Dan Li, Murray Resnick, Jose Behar, Jack R. Wands, Li Juan Wang, Ronald A. DeLellis, Weibiao Cao Gastro-esophageal reflux disease complicated by Barrett's esophagus (BE) is a major risk factor for esophageal adenocarcinoma (EA). The mechanisms whereby acid reflux may accelerate the progression from BE to EA are not fully understood. Reactive oxygen species (ROS) have been reported to be increased in BE and EA. We have previously shown that Barrett's metaplastic cells may be a source of ROS and that NADPH oxidase NOX5-S is responsible for acid-induced H 2 O 2 production in Barrett's cells and in EA cells. Besides metaplastic cells, other H 2 O 2 -producing cells (e.g. inflammatory cells) present in BE mucosa may produce additional ROS which may also affect metaplastic cells contributing to eso- phageal tumorigenesis. In this study we investigated whether exogenous H 2 O 2 affects NOX5- S expression and cell proliferation in a Barrett's EA cell line FLO. Low doses (10 -11 & 10 - 13 M) of H 2 O 2 significantly increased thymidine incorporation, whereas high dose (10 -5 M) of H 2 O 2 decreased thymidine incorporation. 10 -13 MH 2 O 2 markedly increased NOX5-S mRNA and protein expression in FLO cells. H 2 O 2 -induced increase in NOX5-S expression was significantly inhibited by knockdown of NF-κB1 p50 with p50 siRNA. H 2 O 2 significantly increased the luciferase activity in FLO cells transfected with a NF-κB In Vivo activation reporter plasmid pNF-κB-Luc, indicating H 2 O 2 may activate NF-κB. Overexpression of NF- κB p65 and p50 remarkably increased the luciferase activity in FLO cells transfected with a NOX5-S reporter plasmid NOX5LP. In addition, H 2 O 2 -induced increase in NOX5-S expression and thymidine incorporation was significantly decreased by the MAP kinase inhibitor PD98059 and ERK2 siRNA, but not by ERK1 siRNA. We conclude that low doses of H 2 O 2 increase cell proliferation. H 2 O 2 -induced increase in cell proliferation may depend on activation of ERK2 MAP kinase, NF-κB1 p50 and NOX5-S. It is possible that in Barrett's esophagus ROS produced by inflammatory cells may activate ERK2 MAP kinase and cause upregulation of NOX5-S in Barrett's metaplastic cells, which further enhances production of ROS (a positive feedback) and increases cell proliferation and DNA damage, thereby contributing to the esophageal tumorigenesis. Supported by NIH NIDDK R01 DK080703. Sa1082 NADPH Oxidase NOX5-S Mediates Acid-Induced DNA Damage in Barrett's Esophageal Adenocarcinoma Cells Dan Li, Xiaoxu Zhou, Murray Resnick, Jose Behar, Jack R. Wands, Li Juan Wang, Ronald A. DeLellis, Weibiao Cao Gastro-esophageal reflux disease complicated by Barrett's esophagus (BE) is a major risk factor for esophageal adenocarcinoma (EA). The mechanisms whereby acid reflux may accelerate the progression from BE to EA are not fully understood. Acid and reactive oxygen species (ROS) have been reported to cause DNA damage in Barrett's cells. We have previously shown that NADPH oxidase NOX5-S is responsible for acid-induced H 2 O 2 production in Barrett's cells and in EA cells. In this study we examined whether NADPH oxidase NOX5- S mediates acid-induced DNA damage in a Barrett's EA cell line FLO. DNA damage was measured by using a Comet assay which is based on the ability of denatured cleaved DNA fragments to migrate out of the cell under the influence of an electric potential. It is quantitated by measuring the tail length, tail area and tail moment. Pulsed acid treatment significantly increased tail length from 1.6±0.5 to 6.2±1.2 pixels, tail area from 51.7±15.6 to 164.5±47.3 pixels, and tail moment from 0.9±0.3 to 2.2±0.6, suggesting that pulsed acid treatment increases DNA damage. To further confirm this result, the phosphorylation of histone H2AX, an indicator of double strand DNA break, was examined. Pulsed acid treatment markedly increased histone H2AX phosphorylation. Acid-induced increase in tail length, tail area, tail moment and histone H2AX phosphorylation was significantly decreased by knockdown of NOX5-S with NOX5 siRNA (figure 1). Conversely, overexpression of NOX5-S significantly increased tail length, tail area, tail moment and histone H2AX phosphorylation. We conclude that pulsed acid treatment causes DNA damage via activation of NOX5-S. It is possible that in Barrett's esophagus acid activates NOX5-S and increases ROS production, which causes DNA damage, thereby contributing to the progression from BE to EA. Supported by NIH NIDDK R01 DK080703. S-220 AGA Abstracts Sa1083 A Possible Role for Toll-Like Receptor 4 Activation in Barrett's Esophagus in the Development of Esophageal Adenocarcinoma Romy E. Verbeek, Peter D. Siersema, Fiebo J. ten Kate, Kees Fluiter, Frank P. Vleggaar, Jantine W. van Baal BACKGROUND: Barrett's esophagus (BE) is thought to develop and progress to esophageal adenocarcinoma (EAC) in the setting of chronic inflammation. However, the underlying signaling pathways are incompletely understood. We hypothesize that Toll-Like Receptor (TLR) 4 activation in BE may be involved in the development of EAC through induction of COX2 and resistance to apoptosis. AIM: To determine the expression and functional activity of TLR4 in the esophagus and changes in COX2 expression and susceptibility to apoptosis following TLR4 activation in BE. METHODS: Esophageal biopsies were obtained during endoscopy from normal squamous epithelium (SQ), BE and EAC of 14 patients with BE and 8 with BE associated EAC. Expression of TLR4 mRNA was assessed by in situ hybridization using antisense oligonucleotides containing locked nucleic acid and 2'-O- methyl-RNA moieties, and TLR4 protein by immunohistochemistry. The former was quanti- fied in frozen esophageal biopsies using Q-RT-PCR. To assess functional activity of TLR4, BE biopsies from 5 patients were stimulated with lipopolysaccharide (LPS), the natural TLR4 agonist. After 3 hours, IL8 production was determined in the supernatant by ELISA. RNA was isolated for determination of relative changes in COX2 expression by Q-RT-PCR. Western blot for the anti-apoptotic factors XIAP and BCL2 was performed in esophageal biopsies to study the effect on apoptosis. RESULTS: TLR4 mRNA as well as protein were for a large part expressed on the epithelial surface of BE and EAC but confined to the epithelial base in SQ. Q-RT-PCR showed a 9.1-fold (p<0.01) increase in TLR 4 expression in EAC (1.5± 0.6) and a 7.3-fold (p<0.01) increase in BE (1.2± 0.2) compared to SQ (0.2± 0.1). Incubation of BE biopsies with LPS resulted in increased IL8 production compared to negative controls (without LPS) (mean increase 397± 412 pg/ml, p=0.04). This finding was confirmed by Q- RT-PCR showing a 6.1 (IQR 2.1-12) fold increased IL8 mRNA expression in BE biopsies following LPS incubation compared to controls. In addition, a 6.7 (IQR 2.3-13) fold increased COX2 mRNA expression was observed in BE biopsies incubated with LPS compared to controls. Finally, protein expression of the anti-apoptotic markers BCL2 and XIAP was increased after LPS stimulation of BE biopsies. CONCLUSION: The innate immune receptor TLR4 is expressed on the epithelial surface of BE and EAC and was found to be functionally active, as suggested by IL8 production after LPS stimulation. As activation of TLR4 in BE biopsies also resulted in increased expression of COX2 and the anti-apoptotic factors BCL2 and XIAP, it may well be that TLR4 activation in BE contributes to malignant transformation. Sa1084 Chemopreventive Potential of Metformin in Barrett's Esophagus Alireza Faridar, Sonia Chowdhury, Cathrine J. DeMars, Abhayjit Singh, Gwen A. Lomberk, Ganapathy A. Prasad, Paul J. Limburg, Amitabh Chak, Navtej Buttar BACKGROUND & AIM: We recently published that carcinogenic bile salts activate oncogenic cascade mTOR and its downstream targets like pS6K1. Our In Vitro as well as In Vivo studies in animals and patients show that mTOR pathway is up-regulated during neoplastic transformation in Barrett's esophagus. We also show that targeting mTOR pathway with rapamycin decreases cell growth. However, it is increasingly being recognized that rapamycin resistance could develop due to up-regulation of AKT phosphorylation by rapamycin (Cancer Research 2008). In search of a combinatorial chemoprevention partner, we noted that anti- diabetic drug metformin, by inhibiting AKT phosphorylation, may prevent resistance to rapamycin. However, it remains to be determined if metformin could alter phospho-AKT levels and/or cell growth in Barrett's esophagus. METHODS & RESULTS: Using Western blot and IHC, we found that the expression of well known metformin target protein phospho- AMPK is markedly decreased in malignant SKGT-4 cells compared to premalignant BAR cells. Although a 16 hour exposure to 0.5mM metformin did not change the phospho-AMPK levels, it did decrease expression of phospho-AKT (Figure below-top panel). Furthermore, adenoviral transduction of SKGT-4 cells with constitutively active AKT-1 increased their proliferation, while the pharmacological inhibitors of EGFR-PI3K-AKT pathway decreased cell growth. Finally, treatment of SKGT-4 cells by increasing doses of metformin (0.1 to 0.5mM) showed a progressive decrease in SKGT-4 cell growth by up to 50% (Figure below, Lower panel BrdU incorporation, p<0.05). We are currently investigating if constitutively active AKT can abrogate the growth inhibitory effects of metformin and if metformin when used together with rapamycin exerts better growth inhibition compared to when these agents are used independently. CONCLUSION: Our results, for the first time show that metformin can inhibit oncogenic phospho-AKT and has chemopreventive potential in Barrett's eso- phagus. Metformin's ability to inhibit phospho-AKT could compliment growth inhibitory effects of rapamycin and provide a novel combinatorial chemoprevention approach.