Oxidative Stress Activates the Human Histidine Decarboxylase Promoter in AGS Gastric Cancer Cells*

Oxidant stress is thought to play a role in the pathogenesis of many gastric disorders. We have recently reported that histidine decarboxylase (HDC) promoter activity is stimulated by gastrin through a protein kinase C- and extracellular signal-regulating kinase (ERK)-dependent pathway in gastric cancer (AGS-B) cells, and this transcriptional response is mediated by a downstream cis-acting element, the gastrin response element (GAS-RE). To study the mechanism through which oxidant stress affects gastric cells, we examined the effects of hydrogen peroxide (H2O2) on HDC promoter activity and intracellular signaling in AGS-B cells. H2O2(10 mm) specifically activated the HDC promoter 10–12-fold, and this activation was blocked by both mannitol andN-acetylcysteine. Hydrogen peroxide treatment of AGS-B cells increased the phosphorylation and kinase activity of ERK-1 and ERK-2, but did not affect Jun kinase tyrosine phosphorylation or kinase activity. In addition, treatment of AGS-B cells with H2O2 resulted in increasedc-fos/c-jun mRNA expression and AP-1 activity, and also led to increased phosphorylation of epidermal growth factor receptor (EGFR) and Shc. H2O2-dependent stimulation of HDC promoter activity was completely inhibited by kinase-deficient ERKs, dominant-negative (N17 and N15) Ras, and dominant-negative Raf, and partially blocked by a dominant-negative EGFR mutant. In contrast, protein kinase C blockade did not inhibit H2O2-dependent induction of the HDC promoter. Finally, deletion analysis demonstrated that the H2O2 response element could be mapped to the GAS-RE (nucleotides 2 to 24) of the basal HDC promoter. Overall, these studies suggest that oxidant stress activates the HDC promoter through the GAS-RE, and through an Ras-, Raf-, and ERK-dependent pathway at least partially involving the EGFR.

Reactive oxygen metabolites (ROMs), 1 such as superoxide anion O 2 . , hydroxyl radical OH, and H 2 O 2 , are generated in cells as physiological by-products of electron transfer reactions and arachidonic acid metabolism (1,2). An elevated level of ROMs reflecting oxidative stress is observed during a number of acute physiologic and pathological states, including sepsis, exposure to ionizing radiation, ischemia and reperfusion, and diverse inflammatory conditions (2). In addition, oxidative stress is thought to play a role in many chronic disease processes, included atherosclerosis, aging, and cancer (3)(4)(5).
The gastric mucosa is continuously exposed to luminal oxidants generated from ingested food, bacteria, and shed mucosal cells (6). The gastric epithelium, in conjunction with the surface mucous layer, represents the first line of defense against luminal oxidative stress (7); despite constant exposure to luminal oxidants, the gastric epithelium remains unaffected. However, enhanced production of ROMs due to acute and chronic inflammation of the gastrointestinal tract may contribute to the mucosal injury. Inflammatory and ulcerative diseases of the gastric mucosa, such as those associated with ethanol, nonsteroidal anti-inflammatory drugs, cold stress, burn stress, ischemia-reperfusion, and Helicobacter pylori, are associated with increased oxidative stress (8 -10). H. pylori-infected gastric mucosa, for example, is characterized by an accumulation of active neutrophils (polymorphonuclear leukocytes) (11,12), and these activated neutrophils may result in increased oxidative DNA damage (13). Thus, reactive oxygen metabolites may be involved in the pathogenesis of both peptic ulcer disease and gastric cancer (14).
Although oxidative stress has cytotoxic effects, studies have also indicated that ROMs mediate a number of adaptive biologic responses (15). Oxidative stress secondary to H 2 O 2 exposure may function as a local trigger for programmed cell death (16), but it can also modulate the expression of a variety of genes that are involved in the immune and inflammatory response, such as the nuclear transcription factor kB (NF-B) (17). In addition, ROMs at low levels have been shown to induce the expression of growth factor-regulated genes, such as AP-1 (c-fos and c-jun), c-myc, and egr-1 (17)(18)(19). Oxidative stress appears to affect numerous signaling pathways through protein phosphorylation and induce selectively a number of genes, and thus chronic exposure to low levels of ROM may conceivably influence cell growth and differentiation. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  Histidine decarboxylase (HDC) is the major enzyme involved in the production of histamine from the amino acid L-histidine, a key step in the stimulation of gastric acid secretion (20). Both histamine and HDC enzyme are activated in the rat stomach in response to cold-induced stress (21,22). Several previous observations have suggested that the HDC gene may be activated in the stomach under conditions of oxidative stress. For example, HDC activity has been shown to be increased during ischemia-reperfusion injury to the gastrointestinal tract (23)(24)(25). Experiments from our laboratory have shown that the HDC promoter is transcriptionally regulated by gastrin and the phorbol ester (phorbol 12-myristate 13-acetate, PMA) through a protein kinase C (PKC)-and MAP kinase-dependent pathway, acting on a 23-nucleotide cis-acting element, the gastrin response element (GAS-RE) (20,26,27). More recently, we have shown that activation of AP-1 is essential for gastrinstimulated HDC transcription, although the mechanism is most likely indirect (28), indicating that HDC may be an important downstream target of AP-1. Thus, we decided to study the effects of oxidative stress on gastric epithelial cells, AP-1 activity, and HDC promoter activity. These studies suggest that H 2 O 2 can selectively stimulate HDC promoter activity, most likely through an EGFR-dependent signaling pathway that leads to activation of the MAP kinase/ERK pathway.
DNA Constructs and Reporter Plasmids-The human 1.8-kb HDCluciferase construct and the GAS-RE/thymidine kinase-Luc construct have been described elsewhere (26). The GAS-RE/thymidine kinase-Luc construct is derived from thymidine kinase-Luc but also contains the human HDC GAS-RE sequences ligated upstream of the herpes simplex virus 1 thymidine kinase promoter. The 4XTRE-CAT construct has been previously reported (28). The c-Fos-luciferase construct was a kind gift of Richard Treisman (29). The EGFR dominant-negative expression (HER653) construct was a gift of Murray Korc and contains the kinasedeleted mutant of the human EGFR cDNA (dominant-negative (DN)-EGFR) under the control of the cytomegalovirus (CMV) promoter. Transfection of this DN-EGFR construct has been shown to block EGFR signaling (30). The ornithine decarboxylase-luciferase construct was a gift of Juanita L. Merchant, and contained 450 base pairs of human ornithine dcarboxylase 5Ј-flanking DNA. The GAL4-c-Jun wild type construct contains the GAL4 DNA binding domain linked to amino acids 1-246 of human c-Jun, while the GAL4-c-Jun mutant (AA) has Ser-63 and Ser-73 of c-Jun mutated to alanines. These constructs and the 5XGAL-Luc system have been described previously (27).
The ERK cDNAs and mutants were subcloned into the expression vector pCMV5 (in which cDNA expression is driven by a CMV promoter) and have been described previously (27,31). Wild type ERK constructs contained full-length human cDNAs, whereas the mutants K71R-ERK-1 and K52R-ERK-2 encode kinase-deficient, interfering proteins with kinase activities Ͻ5% of the corresponding wild type kinases (31). The constructs Ha-ras-ASN17 (RasN17) and Ha-ras-Ala15 (RasN15) represent dominant-negative Ras expression constructs in which mutated Ras genes are under control of the murine mammary tumor virus promoter (32,33). The dominant-negative Raf construct has also been previously described (27,34).
Cell Line and Transfections-AGS-B, the cell line used in these studies, was prepared by stable transfection of the human gastric adenocarcinoma cell line AGS (ATCC CRL 1739) with the CCK-B/ gastrin receptor and has been reported (20). The AGS-B/1.8-kb hHDC-Luc cells, derived from the AGS-B cells through transfection of the 1.8-kb hHDC-Luc construct, have also been described (26). AGS-B and AGS-B/1.8-kb hHDC-Luc cells were grown in Dulbecco's modified Eagle's medium plus 10% bovine calf serum, 100 IU/ml penicillin, and 100 IU/ml streptomycin in a humidified atmosphere (5% CO 2 ). Transient transfections of cultured cells were carried out using the calcium phos-phate precipitation technique (DNA transfection kit, 5 Prime 3 3 Prime, Inc.) as described previously (20). H 2 O 2 stimulation was performed for 10 min, and cells were harvested 6 h thereafter. In time course studies, cells were harvested 0, 2, 4, and 6 h after a 10-min H 2 O 2 pulse. Gastrin and PMA were added at stimulatory concentrations (12-24 h after transient transfection), cells were harvested, and luciferase assays were performed at 48 h. Luciferase assays were carried out using luciferin, ATP, coenzyme A (Promega system) with a Monolight Luminometer (Analytical Luminescence Laboratory). CAT assays were performed as described previously (28). Experiments were performed in triplicate or quadruplicate, and results were calculated as mean Ϯ S.D. Values for HDC luciferase were expressed as a fold increase in luciferase compared with untreated controls. Activities in all transfection experiments represented the mean Ϯ S.D. of at least 4 -6 independent transfections. Activities varied Ͻ15% among transfection experiments. Statistical differences were calculated using the Student's t test (Microsoft Excel).
RNA Isolations and Northern Blot Analysis-Total RNA from AGS-B cells was prepared using TRIzol R reagent (Life Technologies, Inc.) according to the manufacturer's instructions. 20 g of total RNA were separated on 2.2 M formaldehyde, 1.2% agarose gels. Ethidium bromide staining confirmed that equal amounts of RNA were loaded per lane (data not shown). Probes for c-jun and c-fos probes were excised from expression constructs which have previously been reported (28) and were labeled with the Ready-to-go kit (Amersham Pharmacia Biotech) according to the manufacturer's instructions. Membranes were prehybridized for 30 min and hybridized for 60 min using Rapid-Hyb buffer (Amersham Pharmacia Biotech) at 65°C. Following hybridization the membranes were washed twice for 20 min each in 2ϫ SSC, 0.1% SDS at 25°C, then once in 2ϫ SSC, 0.1% SDS at 55°C for 20 min. Membranes were exposed to x-ray film for 16 h at Ϫ80°C.
Immunoprecipitation and Western Blotting-AGS-B cells were grown to subconfluence and then starved for 36 h in Dulbecco's modified Eagle's medium without serum. Cells were then stimulated with H 2 O 2 (10 mM final concentration) for 10 min at 37°C. After stimulation, cells were washed in ice-cold phosphate-buffered saline and lysed in lysis buffer (20 mM Tris, pH 7.8, 150 mM NaCl, 2 mM EDTA, 0.5% Nonidet P-40, 10 mM NaF, 10 mM sodium orthovanadate, 2 mM phenylmethylsulfonyl fluoride, and 10 g/ml each of leupeptin, pepstatin, and aprotinin). The lysates were precleared for 1 h by incubating them with normal rabbit serum and protein A-Sepharose beads (Pharmacia). Antibodies to ERK-1, JNK-1, EGF-R, and Grb2 (Santa Cruz Biochemicals), and Shc (Transduction Laboratories) were used to immunoprecipitate these proteins from cells lysates. Immune complexes were collected on protein A-Sepharose beads and washed three times with ice-cold lysis buffer. SDS-sample buffer was added, and the beads were boiled for 5 min. Proteins were separated by SDS-polyacrylamide gel electrophoresis on 7.5% polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Millipore). The membrane was then probed with anti-phosphotyrosine antibody (4G10) (1 g/ml) (Upstate Biotechnology Inc.). Bands were visualized with the enhanced chemiluminescence system (Amersham). Membranes that were probed with anti-phosphotyrosine antibody were reprobed after stripping the membrane. Stripping was performed by washing extensively with TTBS (20 mM Tris pH 7.6, 150 mM NaCl plus 0.05% Tween 20) for 4 h, then incubating the membrane in a solution of 62.5 mM Tris-HCl, pH 6.8, 2% SDS, and 100 mM 2-mercaptoethanol for 30 min at 55°C followed by washing in TTBS for 2 h changing the buffer every 30 min.
MAP Kinase Activity Assays-To determine ERK activity in response to H 2 O 2, cell-free extracts were prepared after treatment of subconfluent AGS-B cells with H 2 O 2 (10 mM). Extracts were boiled with Laemmli sample buffer, and aliquots (10 g protein) were separated on a 10% SDS-polyacrylamide gel containing 0.5 mg/ml ERK substrate myelin basic protein (35). Gels were subsequently submitted to denaturationrenaturation and kinase activity was measured by incubating the gels in kinase buffer containing 25 M ATP and 100 Ci [␥-32 P]ATP for 90 min at room temperature. Kinase reaction was terminated by washing the gels in 5% trichloroacetic acid, 10 mM sodium pyrophosphate. Dried gels were exposed to x-ray films, and signal intensity was determined by scanning densitometry using a Molecular Dynamics personal densitometer with Imagequant version 3.22 software.

RESULTS
Hydrogen Peroxide Activates the HDC Promoter-To determine whether oxidative stress stimulates HDC promoter activity, AGS-B cells stably transfected with the 1.8-kb hHDC-Luc construct were exposed to varying concentrations of hydrogen peroxide for 10 min, and cell extracts were assayed for luciferase activity 6 h later (Fig. 1A). Hydrogen peroxide stimulated HDC promoter activity in a dose-dependent fashion, with a 10 -12-fold increase seen when using 10 mM H 2 O 2 . Stimulation by 10 mM H 2 O 2 resulted in a measurable increase in HDC promoter activity as early as 3 h post-stimulation, with a maximal increase seen at 5-7 h after treatment (Fig. 1B). The effect of H 2 O 2 appeared to be selective, since no activation of the ornithine decarboxylase promoter nor of the basal thymidine kinase promoter was observed under similar conditions (data not shown).
To confirm that induction of HDC promoter activity was due to increased oxidant stress and free radical generation, we examined the effect of pretreatment with N-acetylcysteine and mannitol on stimulation by H 2 O 2 . N-Acetylcysteine is a glutathione precursor known to enhance cellular antioxidant potential, while mannitol is a free radical scavenger with relative specificity for the hydroxyl radical (3). Both N-acetylcysteine and mannitol completely abolished HDC promoter activation by H 2 O 2 , consistent with a mechanism involving oxidant stress and a hydroxyl radical-like species (Fig. 2).

Hydrogen Peroxide Activates AP-1 and c-Myc in AGS-B Gas-
tric Cells-Previous studies from our laboratory showed that HDC promoter activity was stimulated by gastrin and PMA through an AP-1-dependent mechanism (28). Therefore, we examined the effect of H 2 O 2 on the expression of c-fos and c-jun mRNA in AGS-B gastric cells using Northern blot analysis. Treatment of AGS-B cells with 10 mM H 2 O 2 led to a marked increase in mRNA expression of both c-fos and c-jun, with initial increases seen at 30 min and the largest increases observed at 120 min (Fig. 3A). We also examined the effect of H 2 O 2 on c-Jun-dependent transactivation, using the GAL4-c-Jun (wild type and mutant) expression vectors and the 5XGAL-Luc reporter gene (see "Experimental Procedures"). Both H 2 O 2 and PMA stimulated transactivation by GAL4-c-Jun wild type but had no effect on the mutant construct, GAL4-c-Jun mutant (AA) (Fig. 3B). Hydrogen peroxide also stimulated AP-1-dependent transactivation, as measured using AGS-B cells transiently transfected with the 4XTRE-CAT construct (Fig. 3C). The effect of H 2 O 2 on c-fos gene expression was most likely due to transcriptional induction, as evidenced by a 10-fold stimulation effect of H 2 O 2 on c-fos promoter activity in transiently transfected AGS-B cells (Fig. 3D). Finally, H 2 O 2 also stimulated c-Myc-dependent transactivation, as shown in studies using wild type and mutant GAL4-c-Myc expression vectors and the 5XGAL-Luc reporter gene (Fig. 3E). These studies indicate that H 2 O 2 transactivates c-Myc, and that the AP-1 pathway is activated both transcriptionally and post-translationally by oxidative stress in gastric cells.
Hydrogen Peroxide Activates the ERKs but Not JNK-Since both HDC and c-fos (and to a lesser extent, c-jun) promoter activity are known downstream targets of ERK activation, we investigated the effect of H 2 O 2 on the phosphorylation and kinase activity of the ERKs. As shown in Fig. 4A, H 2 O 2 stimulated both the phosphorylation and kinase activity of the ERKs. H 2 O 2 stimulated ERK kinase activity to a greater degree (Ͼ15-fold) than either gastrin or PMA (5-10 fold) or dibutyryl cAMP (3-5-fold) in AGS-B cells (Fig. 4B). In contrast, the JNKs were not activated by H 2 O 2 , although they were stimulated (and phosphorylated) by UV light exposure (Fig. 4C). Taken together, these findings suggest selective activation of ERKs by H 2 O 2 in the AGS-B system.

H 2 O 2 Treatment Leads to Shc and EGFR Phosphorylation-
Previous studies have suggested that in some cell systems H 2 O 2 activates the ERKs through a Ras pathway that involves tyrosine phosphorylation of EGFR, followed by phosphorylation of Shc and the induction of Shc-Grb2-SOS complexes (3,36). To investigate the pathways utilized by gastric cells in response to oxidant stress, we analyzed protein extracts from H 2 O 2 -treated AGS-B cells for phosphotyrosine proteins by Western blotting. H 2 O 2 induced a marked increase in tyrosine phosphorylation of multiple proteins (Fig. 5A, lane 4) with molecular masses that ranged from 60 to 200 kDa. In order to examine the effect of H 2 O 2 on the phosphorylation of Shc and its association with the adaptor protein Grb2, Grb2 and Shc proteins were first immunoprecipitated with specific antibodies, and phosphorylation levels were assessed using an antiphosphotyrosine antibody. As shown in Fig. 5A (lanes 5-12), treatment of AGS-B cells with H 2 O 2 resulted in an increase in tyrosine phosphorylation of both the 52-and 46-kDa isoforms of Shc. The increase was greater than that seen with PMA or gastrin stimulation. In addition, multiple bands representing tyrosine phosphorylated proteins, not seen with PMA or gastrin stimulation, were co-immunoprecipitated when using antibodies to Grb2 or Shc (Fig. 5A).
Several of the proteins that were phosphorylated in response to H 2 O 2 ranged in size from 100 to 200 kDa. Since the EGFR protein is 170 kDa, we examined whether EGFR associates with Shc and undergoes phosphorylation in response to H 2 O 2 . As shown in Fig. 5B, treatment of AGS-B cells resulted in tyrosine phosphorylation of EGFR and association with Shc, while PMA treatment did not. Gastrin stimulation also did not lead to EGFR phosphorylation (data not shown). Activity-To define the functional role of PKCs for H 2 O 2 -dependent hHDC transactivation, we down-regulated PKCs using pretreatment with 10 Ϫ8 M phorbol ester PMA as described previously (20,26). These studies revealed that blockade of PKCs using PMA pretreatment had no effect on the transactivating effect of H 2 O 2 , whereas PMA-stimulated activity of the 1.8-kb hHDC-Luc construct was completely abolished (Fig. 6). hHDC promoter is preferentially mediated through activation of pathways independent of PKCs.

Effect of PKC Blockade on H 2 O 2 -stimulated hHDC Promoter
The Effect of H 2 O 2 on the hHDC Promoter Is Mediated through MAP Kinase/ERK Signaling Pathways-To evaluate the importance of ERK-dependent signaling pathways for H 2 O 2 -dependent regulation of the hHDC promoter, AGS-B cells were co-transfected with 1.8-kb hHDC-Luc, and expression constructs encoding dominant-negative ERK mutants and subsequently stimulated with a maximal concentration of H 2 O 2 . Expression of dominant-negative ERK proteins abolished the effect of H 2 O 2 on hHDC promoter activity (Fig. 7), demonstrating that H 2 O 2 -dependent transactivation of the hHDC promoter is mediated through activation of MAP kinase/ERKs.
The Effect of H 2 O 2 on the hHDC Promoter Is Mediated through Ras-and Raf-mediated Pathways-To define the initial steps in activation of intracellular signaling cascades transmitting the effect of H 2 O 2 on the hHDC promoter, we analyzed the role of Ras and Raf for H 2 O 2 -stimulated hHDC promoter activity. Expression of the interfering Ras mutants Ha-ras-ASN17 (RasN17) or Ha-ras-Ala15 (RasN15) abolished the effect of H 2 O 2 on the hHDC construct (Fig. 8). In contrast, no effect on PMA-stimulated hHDC promoter transactivation could be observed (27) (data not shown). Similarly, inhibition of Raf by expression of a dominant-negative Raf-1 mutant abolished the H 2 O 2 effect (Fig. 8). Raf blockade also inhibited hHDC promoter transactivation in response to PMA (27) (data not shown). Taken together, our data demonstrate transmission of H 2 O 2 -dependent hHDC transactivation through a Rasand Raf-dependent proximal pathway.
H 2 O 2 Induction of the HDC Promoter Is Partially Blocked by Dominant-Negative EGFR-The role of EGFR in mediating stimulation of the HDC promoter was further studied using a kinase-deficient EGFR mutant lacking the cytoplasmic domain, which inhibits EGFR downstream signaling by formation of signaling-defective heterodimers with the wild-type receptor. This DN-EGFR construct was co-transfected into AGS-B cells along with the 1.8-kb hHDC-Luc construct, and the response to H 2 O 2 and PMA was assessed. The DN-EGFR partially blocked H 2 O 2 -mediated induction of HDC promoter activity, but had no effect on the PMA response (Fig. 9). In addition, the DN-EGFR construct had no effect on the response to gastrin (not shown).
The GAS-RE Functions as an H 2 O 2 -responsive Element-To investigate further the downstream mechanism by which H 2 O 2 activates the HDC promoter, we analyzed a number of HDC deletion constructs for activation by H 2 O 2 treatment. These deletion studies showed that the H 2 O 2 responsive element was located within the basal (Ϫ59 to ϩ125) human HDC promoter (data not shown). In previous reports, we mapped the HDC GAS-RE to a 23-bp element located just downstream of the transcriptional start site (26) (Fig. 10A). Ligation of the GAS-RE enhancer upstream of the gastrin -insensitive thymidine kinase promoter in the thymidine kinase-Luc construct resulted in a 4-fold response to H 2 O 2 , slightly less than that seen with PMA (Fig. 10B). DISCUSSION Recent studies have suggested that H 2 O 2 can selectively activate a number of genes involved in a variety of biologic responses to oxidant stress, including immunologic, proliferative, and apoptotic responses (15-17, 37, 38). In previous reports, we demonstrated that transcription of the HDC gene was activated by gastrin and PMA through a PKC/ERK/AP-1 pathway (20,27,28). We now show that HDC promoter activity can be stimulated by H 2 O 2 . The responses were seen at a slightly higher H 2 O 2 concentration (10 mM) than that used in previous studies (3), but may reflect the greater resistance of gastric cells to oxidative stress. The responses were specific for oxidative stress pathways and reactive oxygen intermediates (such as the hydroxyl radical) as shown by the inhibition seen with N-acetylcysteine and mannitol. In addition, the activation of HDC was selective, since no effect of H 2 O 2 was observed on the ODC promoter under similar conditions. Further, we have shown that the downstream target of oxidant stress is the GAS-RE element, which we have previously shown to mediate the HDC transcription response to gastrin and PMA.
Treatment of AGS-B cells with H 2 O 2 led to significant increases in ERK-1 and AP-1 activity. Since earlier studies by our group demonstrated that activation of HDC transcription by gastrin and PMA occurred through ERK-and AP-1-dependent pathways (20,27,28) (3). However, in contrast to this earlier study, the induction of c-jun and c-fos gene expression in AGS-B gastric cells was more sus-tained. Indeed, the greatest levels of c-fos and c-jun gene expression in H 2 O 2 -treated AGS-B cells occurred at 120 min, suggesting a more prolonged growth factor response in gastric cells under conditions of oxidant stress.
Gastrin-dependent activation of the HDC promoter in AGS-B cells has previously been shown by our group to occur through a signaling pathway that involves both Raf-1 and MAP kinase (27). However, this gastrin-dependent pathway appears not to utilize Ras but is dependent on activation of PKCs (20,27). In contrast, the stimulation of ERK activity by H 2 O 2 in AGS-B cells occurs at least in part through activation of the Ras/EGFR pathway (Fig. 11). The increase in EGFR phosphorylation was observed only with H 2 O 2 stimulation and was not seen with gastrin or PMA stimulation. In addition, expression of the dominant-negative Ras and dominant-negative EGFR construct significantly inhibited H 2 O 2 stimulation of the HDC promoter, but did not affect gastrin-or PMA-stimulated HDC expression. In contrast, PKC blockade or down-regulation effectively blocked both PMA and gastrin stimulation of the HDC promoter, but did not inhibit stimulation of HDC transcription by H 2 O 2 . Thus, while both gastrin/PMA and H 2 O 2 pathways leading to induction of HDC appear to converge on Raf-1 kinase and the MAP kinases, the H 2 O 2 pathway is distinguished by its utilization of EGFR and Ras, and the lack of dependence on PKCs (Fig. 11). The mechanism by which H 2 O 2 stimulates EGFR phosphorylation is unclear, although some have speculated that it may involve inhibition of a protein tyrosine phosphatase or the activation of intracellular protein tyrosine kinases such as Src (36). In addition, recent studies suggest that H 2 O 2 may play a role in signaling downstream of EGFR (39). Oxidant-induced EGFR tyrosine phosphorylation may be involved in protective responses to oxidant-induced cell injury in the gastric mucosa, possibly through activation of MAP kinases (40).
Deletion analysis of the human HDC promoter showed that the major cis-acting region mediating responsiveness to H 2 O 2 could be mapped to the GAS-RE element (2 to 24) which mediates response to gastrin and PMA (26). The GAS-RE element appears to bind a novel nuclear transcription factor(s) (26). A few H 2 O 2 -responsive cis-acting elements have been reported in the literature, including the AP-1 binding site (41), the NF-B binding site (36), an enhancer (Ϫ976 to Ϫ890) in the ADF/Trx (Adult T cell leukemia-derived factor/human thioredoxin) promoter (42), and the metal-responsive transcription factor-1 binding site in the metallothionein-I gene promoter (43). Thus, the GAS-RE appears to belong to the growing family of H 2 O 2responsive elements.
The activation of HDC transcription by H 2 O 2 may have some human biologic significance, given the importance of oxidant stress in the pathogenesis of diverse stomach disorders including gastritis, peptic ulcer disease, and gastric cancer. In particular, H. pylori-associated gastritis has been associated with activated neutrophils and oxidant stress (11,12). Although reactive oxidative metabolites may contribute to the killing of H. pylori microorganisms, they may also injure the surrounding tissue (1). The role of HDC activation and histamine generation in the response to oxidant stress of the gastrointestinal tract remains unclear, although several studies have suggested that histamine synthesized by HDC may facilitate healing of the gut mucosa (23-25) and inhibit the further generation of ROMs by neutrophils (44,45). In addition, a role in acid secretion has been suggested by recent studies showing that H 2 O 2 at low concentrations is capable of stimulating acid secretion by sevenfold in isolated gastric mucosa (46). Analysis of HDC promoter activation by H 2 O 2 should provide insight into the molecular events characterizing the adaptive response of gastric cells to oxidant stress.