|Home | About | Journals | Submit | Contact Us | Français|
Helicobacter pylori infection disrupts the balance between gastric epithelial cell proliferation and apoptosis, which is likely to lower the threshold for the development of gastric adenocarcinoma. H. pylori infection is associated with EGF receptor (EGFR) activation through metalloproteinase-dependent release of EGFR ligands in gastric epithelial cells. Since EGFR signaling regulates cell survival, we investigated whether activation of EGFR following H. pylori infection promotes gastric epithelial survival.
Mouse conditionally immortalized stomach epithelial cells (ImSt) and AGS cells, as well as wild-type and kinase-defective EGFR (EGFRwa2) mice, were infected with the H. pylori cag+ strain 7.13. Apoptosis, caspase activity, EGFR activation (phosphorylation) and EGFR downstream targets were analyzed.
Inhibiting EGFR kinase activity or decreasing EGFR expression significantly increased H. pylori-induced apoptosis in ImSt. Blocking H. pylori-induced EGFR activation with a heparin-binding (HB)-EGF neutralizing antibody or abrogating a disintegrin and matrix metalloproteinase-17 (ADAM-17) expression increased apoptosis of H. pylori-infected AGS and ImSt, respectively. Conversely, pretreatment of ImSt with HB-EGF completely blocked H. pylori-induced apoptosis. H. pylori infection stimulated gastric epithelial cell apoptosis in EGFRwa2, but not in wild-type mice. Furthermore, H. pylori-induced EGFR phosphorylation stimulated PI3K-depnedent activation of the anti-apoptotic factor Akt, increased expression of the anti-apoptotic factor Bcl-2, and decreased expression of the pro-apoptotic factor Bax.
EGFR activation by H. pylori infection has an anti-apoptotic effect in gastric epithelial cells that appears to involve Akt signaling and Bcl family members. These findings provide important insights into the mechanisms of H. pylori-associated tumorigenesis.
Helicobacter pylori, a Gram-negative bacterium that selectively colonizes the gastric mucosa, increases the risk of developing gastric adenocarcinoma and gastric non-Hodgkin’s lymphoma in humans and animal models.1, 2 Host responses that mediate H. pylori-associated disease include disruption of the balance between gastric epithelial cell proliferation and apoptosis, which can manifest as increased proliferation and either increased or decreased apoptosis at different stages of infection.3, 4 Increased proliferation that is not balanced by compensatory increases in apoptosis in advanced stage of infection may heighten the risk for development of gastric neoplasia.5 Thus, understanding the mechanism through which H. pylori regulates cell fate is critical for establishing therapeutic approaches to potentially prevent H. pylori infection-induced gastric tumorigenesis.
Both in vivo and in vitro studies have demonstrated that several H. pylori virulence factors stimulate gastric epithelial apoptosis, such as products of the cag pathogenicity island,4, 6 a secreted vacuolating toxin (VacA),7 and lipopolysaccharide.8 The cellular mechanisms by which H. pylori induces apoptosis include activation of Fas9 and cytokine receptors, such as tumor necrosis factor (TNF) receptor10 and induction of spermine oxidase (SMO) activity,11 which lead to Caspase activation and apoptosis in gastric epithelial cells.
H. pylori has also been reported to be associated with attenuated apoptosis following infection,12 and chronically colonized humans and Mongolian gerbils harboring H. pyloricag+ strains develop increased gastric epithelial cell proliferation without an associated increase in apoptosis, suggesting that increased proliferation in the absence of a concomitant increase in apoptosis may contribute to increased risk for gastric cancer with cag+ strains.13–15 However, the mechanism through which H. pylori exerts anti-apoptotic effects remains unclear.
Recent reports show that H. pylori induces up-regulation of EGF receptor (EGFR) expression16 and EGFR activation by heparin-binding (HB)-EGF release from gastric epithelial cells.17, 18 EGFR is a member of the ErbB family which consists of four tyrosine kinase receptors, EGFR (ErbB1), and ErbB2-4. These four receptors are essential for modulating cell survival, proliferation, and differentiation in many tissue types.19
However, there is no report that has identified the role of EGFR activation in the pathogenesis of H. pylori infection. Therefore, the purpose of this study was to define gastric epithelial cellular responses regulated by EGFR activation following H. pylori infection. Our studies using in vitro cell culture and in vivo animal models show that transactivation of EGFR by H. pylori protects gastric epithelial cells from apoptosis and that this anti-apoptotic response is mediated by PI3K-dependent activation of Akt and Bcl family members. Thus, our studies demonstrate that activation of EGFR may represent a previously unrecognized molecular regulatory step in determining the effects of H. pylori on regulation of gastric epithelial cell homeostasis. Since EGFR has been implicated in a number of epithelial cancers and serves as a target for cancer therapy,20, 21 results in this report provide a novel focus for further investigations into the mechanisms of H. pylori-associated tumorigenesis
Information regarding inhibitors, antibodies, reagents, cell treatment, gastric epithelial cell isolation, cellular lysate preparations, immunohistochemistry, and apoptosis assays are provided in the Supplemental Data.
The H. pylori cag+ rodent adapted strain 7.13 was grown in Brucella broth with 5% FBS for 24 h.22 Bacteria were identified as H. pylori by urease and oxidase activity and Gram’s stain morphology. H. pylori were washed with PBS and cell culture medium before addition to cells.
Mouse conditionally immortalized stomach epithelial cells (ImSt) were isolated from the gastric epithelium of transgenic mice with a temperature-sensitive mutation of the simian virus 40 (SV40) large tumor antigen gene (tsA58) fused to the promoter of the mouse H-2Kb class I gene (H-2Kb-tsA58 mice).23 A disintegrin and matrix metalloproteinase-17 (ADAM-17)−/−ImSt were isolated from the gastric epithelium of ADAM-17 null heterozygous mice crossed with the H-2Kb-tsA58 mice.24
ImSt were maintained in RPMI 1640 medium containing 10% FBS, 5 U/ml of murine IFN-γ and 100 U/ml penicillin and streptomycin on collagen-coated plates and grown under permissive conditions at 33°C with 5% CO2. Prior to all experiments, cells were transferred to 37°C (non-permissive) conditions and cultured in medium containing 0.5% FBS, without IFN-γ or antibiotics for 16–18 hours.
ADAM-17−/− ImSt reconstituted to express wild-type (wt) ADAM-17 or catalytically-inactive E406A mutant ADAM-1725 were prepared as described before.26 Briefly, C-terminally HA-tagged murine wt and E406A ADAM17 were cloned into the pBM-IRES-PURO retroviral vector. All cDNA constructs were verified by DNA sequencing. For retroviral transduction, ADAM17−/− ImSt were seeded into tissue culture flasks and cultured for 24 h before incubation with virus stock for 12 h in the presence of 4 μg/ml polybrene. Cells were then grown for 48 h in normal cell culture medium before passage into medium containing 5 μg/ml puromycin.
ImSt were transiently transfected with either non-targeting siRNA or mouse EGFR SMARTpool siRNA at 70% confluence using Lipofectamine 2000.27 After 24 h of transfection, cells were cultured in medium containing 0.5% FBS for 16–18 hours before treatment.
AGS cells were grown in RPMI 1640 medium containing 10% FBS and 100 U/ml penicillin and streptomycin at 37°C with 5% CO2. Prior to all experiments, cells were cultured in medium containing 0.5% FBS without antibiotics for 16–18 h.
All animal experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee at Vanderbilt University. C57BL/6wa2 mice with a naturally occurring EGFR kinase defective mutation (Val to Gly mutation at residue 743 in the kinase domain)28 were obtained from Dr. David Threadgill (University of North Carolina, Chapel Hill). PCR primers specific for the EGFR sequence containing the point mutation were used for genotyping (sequences available upon request). Mice (8–10 weeks old, 25–30g) were anesthetized and then infected orally with H. pylori (109 cfu) or broth only in a total volume of 500 μl for 24 h.29
Gastric tissue was fixed in 10% neutral-buffered formalin and paraffin-embedded before sectioning. The gastric mucosa was scraped into homogenization buffer and tissue was lysed.30
Apoptosis in cell lines was detected using ApopTag In Situ Apoptosis Detection Kits (TUNEL).31, 32 For Annexin V-FITC staining, attached cells were dissociated using Accutase and double stained with Annexin V-FITC and propidium iodide.33 The percentage of apoptotic cell populations was measured by flow cytometry.11
Apoptosis was detected in gastric tissue sections using ApopTag™ In Situ Oligo Ligation (ISOL) Kit, and observed by DIC microscopy.31, 32 Apoptotic cells were quantified by counting the absolute number of positive stained cells in at least 500 gastric glands.
ImSt were fixed with 0.1% paraformaldehyde and incubated with a rabbit anti-active caspase-3 antibody and a mouse anti-Bcl-2 antibody, and were further stained using FITC-conjugated anti-rabbit and APC-conjugated anti-mouse antibodies, respectively.33 The percentage of active caspase-3 and/or Bcl-2 positive staining cell populations was measured by flow cytometry.
Statistical significance in each study was determined by one-way ANOVA followed by Newman-Keuls analysis for multiple comparisons using Prism 5.0 (GraphPad Software, Inc.). A p-value < 0.05 was defined as statistically significant. All data presented are representative of at least three repeat experiments and are presented as mean±S.E.M.
H. pylori induces up-regulation of EGF receptor (EGFR) expression16 and EGFR transactivation18 in gastric epithelial cells. Phosphorylation of Y1068 on EGFR mediates activation of the ERK/MAP and results in recruitment of the p85 subunit of PI3K to the EGF receptor, leading to activation of Akt.34 One significant role of EGFR activation is to promote cell survival. Therefore, we studied the effect of inhibiting EGFR tyrosine kinase activity on H. pylori-stimulated apoptosis in ImSt. H. pylori were co-cultured with ImSt in the presence or absence of the EGFR tyrosine kinase inhibitor, AG1478, and EGFR Y1068 phosphorylation and apoptosis were analyzed. As expected, H. pylori-induced EGFR activation was inhibited by the EGFR tyrosine kinase inhibitor (Figure 1A). However, the expected increase in apoptosis initiated by H. pylori was further increased by blocking EGFR tyrosine kinase activity, as detected by Annexin V staining and flow cytometry (Figure 1B and C) and TUNEL assay (Figure 1D and E).
To investigate a potential role for EGFR in suppressing H. pylori-induced apoptosis, we used siRNA directed against EGFR to reduce EGFR expression in ImSt (Figure 2A). Consistent with the results of experiments in which EGFR kinase activity was inhibited, loss of EGFR expression significantly enhanced H. pylori-stimulated apoptosis, compared to non-targeting control siRNA transfection (Figure 2B). Taken together, these results demonstrate that EGFR expression and EGFR tyrosine kinase activity enhance survival of gastric epithelial cells infected by H. pylori.
Our in vitro data indicate that EGFR activation plays an anti-apoptotic role in H. pylori-infected cells. To determine whether these effects could be extended in vivo, wild type (wt) or EGFR kinase defective (EGFRwa2) mice on a C57BL/6 background were infected with H. pylori. Mice were sacrificed 5 h or 24 h after infection and tissue was fixed and prepared for immunohistochemistry or mucosal scrapings were prepared for Western blot analysis. The presence of H. pylori in both wt and EGFRwa2 mice was confirmed by immunostaining (Figure 3A). H. pylori-induced EGFR Y1068 phosphorylation was detected in gastric mucosal lysates isolated from wt, but not in EGFRwa2 mice, despite similar levels of bacterial infection (Figure 3B). Importantly, immunostaining showed that H. pylori-stimulated EGFR Y1068 phosphorylation was present in wt, but was absent in EGFRwa2 mice (Figure 3C), which is consistent with our in vitro findings using ImSt (Figure 1A).
To determine if EGFR mediates H. pylori regulation of cell survival in vivo, we performed ISOL and anti-active caspase-3 staining to detect apoptosis. H. pylori infection for 24 h did not stimulate apoptosis in the gastric epithelium of wt mice. However, apoptosis was induced by H. pylori infection in EGFRwa2 mice (Figure 3D). Thus, these data indicate that survival of gastric epithelial cells in mucosa infected by H. pylori requires functional EGFR in vivo.
We next examined the relationship between EGFR-regulated apoptosis and proliferation in H. pylori-infected mice. Interestingly, apoptotic cells were limited to the epithelial cell layer of the lower one third of the gastric glands (Figure 3D). However, proliferating gastric epithelial cells were localized to the upper one third of the gastric glands, and no difference in gastric epithelial cell proliferation rates was detected in wt mice after 24 h of H. pylori infection compared to non-infected mice (SD Figure 1). Therefore, no overlap was present between proliferative and apoptotic cell populations. Given the evidence that both gastric epithelial cells with H. pylori-activated EGFR and apoptotic gastric epithelial cells localized to the lower one-third of gastric glands (Figure 3C and D) and no proliferation was stimulated by H. pylori in this zone, these data suggest that H. pylori-activated EGFR after 24 h infection exerts an anti-apoptotic, but not a proliferative role, in gastric epithelial cells.
Activation of Src family protein kinases, leading to metalloproteinase-dependent ligand release, serves as one of the mechanisms by which EGFR is transactivated.35 Previous studies have reported that H. pylori activated EGFR is mediated by stimulation of HB-EGF release from gastric epithelial cells,17, 18 since ADAM-17 can regulate HB-EGF cleavage and ligand binding,36, 37 we determined whether ADAM-17 is involved in H. pylori-regulated EGFR transactivation and apoptosis. In ADAM-17−/− ImSt, H. pylori failed to activate EGFR (Figure 4A) and stimulated increased apoptosis (Figure 4B–C). Infection of wt ADAM-17, but not the catalytically-inactive E406A mutant ADAM-17 restored EGFR activation (Figure 4A) and reduced apoptosis induced by H. pylori (Figure 4B–C), indicating a requirement for ADAM-17 in the H. pylori-induced EGFR activation-dependent cell survival phenotype.
Since HB-EGF is a ligand generated by ADAM-17 to activate EGFR38 and H. pylori stimulates production of HB-EGF, we next determined whether inhibition of HB-EGF altered H. pylori-stimulated EGFR activation and apoptosis. Pretreatment of ImSt with HB-EGF neutralizing antibody blocked H. pylori activation of EGFR (Figure 5A) and increased apoptosis (Figure 5B). Importantly, pretreatment of AGS cells with HB-EGF completely blocked H. pylori-stimulated apoptosis (Figure 5B), suggesting that HB-EGF is not only required for EGFR activation, but also for gastric epithelial cell survival in response to H. pylori infection.
One of the major downstream signaling targets of EGFR is PI3K. Akt is the principal mediator of PI3K-regulated signaling pathways for promoting cell survival. Akt is also an anti-apoptotic target for EGFR transactivation by TNF.27 Therefore, we tested whether EGFR is required for activation of Akt by H. pylori. H. pylori-induced AKT and EGFR activation was blocked in ADAM-17−/−ImSt, and these responses were reconstituted in cells expressing wt, but not catalytically-inactive E406A ADAM-17−/− (Figure 4A). Furthermore, HB-EGF neutralizing antibody inhibited Akt and EGFR activation by H. pylori (Figure 5A). H. pylori infection-induced Akt activation in ImSt was inhibited by blocking EGFR or PI3K kinase activities using an EGFR kinase inhibitor AG1478 or a PI3K kinase inhibitor wortmannin, respectively (Figure 6A). Importantly, pretreatment of ImSt with wortmannin enhanced H. pylori-induced apoptosis (Figure 6B).
Since Akt has direct effects on the apoptotic machinery through regulating expression of several factors, including a pro-apoptotic protein, Bax, and an anti-apoptotic protein, Bcl-2,39 we studied expression of the Bcl family member Bcl-2 in gastric epithelial cells. Bcl-2 and active caspase-3, markers for apoptosis, were assessed in the same cells by flow cytometry. Blocking EGFR kinase activity by AG1478 decreased the population of ImSt with H. pylori-stimulated Bcl-2 expression. However, the total population of cells with active caspase-3 was enhanced by AG1478 (Figure 7A–B), due to a large increase in Bcl-2−/active caspase-3+ cells (left upper quadrant, Figure 7A)
We also investigated Bcl-2 and Bax expression in the gastric mucosa of mice infected with H. pylori. Infection with H. pylori decreased Bax expression and increased Bcl-2 expression in wt mice, but these effects were not observed in the EGFRwa2 mice (Figure 7C). In fact, the ratio of Bax:Bcl-2 was reversed in EGFRwa2 mice, compared to wt mice (Figure 7D), which represents an increased apoptotic response in the EGFRwa2 mice upon H. pylori infection. This regulatory effect of H. pylori on Bcl-2 and Bax was confirmed in gastric epithelial cells isolated from the stomachs of mice infected with H. pylori (SD Figure 2). Thus, these results suggest that Bax and Bcl-2 are downstream targets of EGFR transactivation that promote an anti-apoptotic response in H. pylori-infected gastric epithelial cells.
In this study, we have demonstrated that activation of EGFR is a key regulatory step in promoting an anti-apoptotic response in gastric epithelial cells infected with H. pylori in vitro and in vivo. Furthermore, our data provide evidence that H. pylori-induced EGFR activation results in activation of Akt/PI3K kinase activity and regulation of Bcl family members which promote gastric epithelial cell survival. These novel findings indicate that activation of EGFR may be a previously unrecognized “molecular switch” determining the survival response of H. pylori-infected gastric epithelial cells. We speculate that EGFR activation may function to inhibit apoptosis caused by activation of Fas9 and cytokine receptors10 and SMO activation that leads to generation of H2O2, which results in cytochrome c release and caspase-3 activation11, 33 (Figure 8).
EGFR is a member of the ErbB family which consists of four tyrosine kinase receptors, EGFR (ErbB1), and ErbB2-4. These receptors promote cell survival, proliferation, migration, and differentiation in many tissue types.19 EGFR and other ErbB family members can be activated by direct interaction with EGF-like ligands or transactivated by various extracellular stimuli, such as TNF,27 which initiate formation of homo- and/or hetero-dimers and increased kinase activity.40, 41 Therefore, we recognize that other ErbBs may also be potential targets of H. pylori infection. In fact, H. pylori stimulates ErbB3 tyrosine phosphorylation, but not ErbB2 or 4 in ImSt (data not shown), thus, H. pylori may increase EGFR/ErbB3 heterodimer formation. Since ErbBs have been shown to couple EGFR to the PI3K/Akt pathway in gefitinib-sensitive non-small cell lung cancer cell lines,42, EGFR activation by H. pylori may be amplified by heterodimer formation with ErbB3. These studies are currently underway in our laboratory.
Given a requirement for metalloproteinase activity in H. pylori-initiated HB-EGF release for EGFR transactivation, ADAM-17 is an ideal candidate enzyme for regulation of this pathway.18, 43 Our data show that ADAM-17 is required for preventing H. pylori-induced apoptosis (Figure 4). ADAM-17 is the first ADAM to have a defined physiological substrate, which is the precursor transmembrane form of TNF. ADAM-17 is required for the release of soluble TNF and at least 3 EGFR ligands, HB-EGF, amphiregulin (AR), transforming growth factor (TGF)-α.36, 44 Previous studies have shown that H. pylori stimulates AR release from gastric mucosal cells.45 Thus AR and TNF release may be another ligand for H. pylori activation of EGFR. TNF production is increased in a number of gastrointestinal disorders, including H. pylori-induced gastritis,46 and TNF transactivates EGFR to promote cell survival in colon epithelial cells. 27 Although ADAM-17 is ubiquitously present in the human gastrointestinal tract, its expression and catalytic activity are increased in ulcerative colitis,47 and it regulates TNF levels in rodent models of colitis.48 However, while ADAM-17 is a target of drug development for inflammatory conditions,49 the disorganized and inflamed nature of the gastrointestinal tract seen in ADAM-17-deficient mice suggests this metalloproteinase may also play important roles in gut epithelial homeostasis, perhaps through regulation of EGFR ligands.36, 44, 50, 51 Therefore, a better understanding of the function of ADAM-17 during H. pylori-induced gastric epithelial injury may provide insights into its potential role in gastric adenocarcinoma.
Elucidating the down-stream targets of EGFR activation by H. pylori is important for understanding the cellular responses induced by H. pylori. We show that H. pylori-initiated EGFR activation is required for activation of an anti-apoptotic signal, Akt/PI3K. In fact, since the submission of this current manuscript another report has now demonstrated specific reduction of Akt via siRNA increases apoptosis in response to H. pylori infection of human gastric epithelial cells (Nagy et al., Journal of Infectious Diseases, In press). However, inhibition of EGFR does not block H. pylori-induced pro-apoptotic signals, such as p38 (Figures 4 and and5),5), or JNK activation.18 This evidence supports a model whereby pro-apoptotic signals become physiologically predominant in cells that do not express EGFR or lose EGFR kinase activity. We show that H. pylori-initiated EGFR activation is required for Akt/PI3K and ERK1/2 MAPK activation in gastric epithelial cells (Figures 4, ,5).5). Interestingly, although H. pylori-stimulated ERK1/2 MAPK activation in ImSt requires EGFR (Figures 4A and and5A5A and18), inhibition of ERK1/2 MAPK activation does not enhance H. pylori-induced apoptosis (Figure 6B). These data suggest that Akt is a potential target of EGFR activation for preventing H. pylori-stimulated apoptosis in gastric epithelial cells. However, another group reported a role of H. pylori-induced ERK1/2 MAPK activation in prevention of apoptosis.52 This difference in the role of H. pylori-activated ERK1/2 MAPK may depend on host and/or bacterial strain differences.
In addition to PI3K-dependent Akt activation up-regulation of Bcl-2 and down-regulation of Bax expression (Figure 7), cyclooxygenase 2 (COX-2) may serve as another downstream target of EGFR activation by H. pylori that can induce anti-apoptotic responses in gastric epithelial cells. The COX-2 level in gastric epithelial cells is up-regulated by H. pylori infection,53, 54 and COX-2 inhibitors increase H. pylori-induced apoptosis.55 We have also found that transactivation of EGFR is required for H. pylori up-regulation of COX-2 expression (unpublished observation).
In summary, our study demonstrates that H. pylori-stimulated EGFR activation exerts an anti-apoptotic role in gastric epithelial cells in vitro and in vivo through its downstream targets, Akt and Bcl2 family members. In vivo chronically infected humans and Mongolian gerbils harboring cag+ strains have increased gastric epithelial cell proliferation without an associated increase in apoptosis, suggesting that increased proliferation in the absence of a concomitant increase in apoptosis may contribute to increased risk for gastric cancer.14, 15 Additionally, the lack of gastric epithelial cell apoptosis with acute H. pylori infection in vivo that we have attributed to EGFR transactivation may contribute to the ability of H. pylori to colonize, as there may be less neutrophil infiltration that would otherwise be recruited to engulf cellular debris and may serve to defend against H. pylori infection.56 Important questions regarding the role of EGFR transactivation in the progression from gastric infection to neoplasia and the interaction of this pathway with the recently reported links between EGFR activation and macrophage recruitment57 and activation58 need to be addressed to fully understand the preventive and therapeutic implications of the findings on gastric carcinoma development.
Grant Support: This work was supported by NIH grants DK56008 for DBP, DK053620 and Office of Medical Research, Department of Veterans Affairs for KTW, DK58587, DK73902, and CA77955 for RMP, AI 068009, AI39657, and Department of Veterans Affairs for TLC, DK63363 and a Crohn’s and Colitis Foundation of America grant for PJD, DK58404 for Vanderbilt University Digestive Disease Research Center, and a NCI program project grant CA116087.
No conflicts of interest exist.
Author contribution: FY and RC contributed to the collection, analysis and interpretation of date and manuscript preparation. HC, UK, SSH, PJD, and MKW contributed to the collection, analysis, and interpretation of date. RMP, TLC, KTW, and DBP contributed to the analysis and interpretation of date and manuscript preparation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.