Helicobacter pylori induced transactivation of SRE and AP-1 through the ERK signalling pathway in gastric cancer cells
- Dr Y Mitsuno, Department of Gastroenterology, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan.
- Accepted 29 January 2001
BACKGROUND AND AIMS Helicobacter pylori infection induces expression of proinflammatory cytokines such as interleukin (IL)-8 and tumour necrosis factor α (TNF-α) in gastric mucosa, and their genes have AP-1 binding sites in the promoter region. c-Fos is important for transactivation of AP-1 which has SRE in the promoter region. We conducted this study to confirm H pylori induced transactivation of these binding sites.
METHODS Transactivation of SRE and AP-1 was evaluated in human gastric cancer cells TMK1 and MKN45 by luciferase reporter assay in transient transfection. We compared the effects of coculture with four H pylori strains, a cag pathogenicity island (PAI) positive strain TN2, its isogenicvacA negative (TN2-ΔvacA) orcagE negative (TN2-ΔcagE) mutants, and acag PAI negative clinical isolate T68. Phosphorylation of ERK1/2, JNK, and c-Jun was measured by immunoblot, induction of IL-8 secretion by ELISA, and the effects of MEK by inhibitor U0126.
RESULTS Both SRE and AP-1 were transactivated by coculture with TN2. Although TN2-ΔvacA induced comparable transactivation, TN2-ΔcagE and T68 showed decreased transactivation of SRE (65% and 51%) and AP-1 (71% and 54%, respectively, of TN2). Heat killed TN2 or indirect contact using a permeable membrane inhibited transactivation. Levels of phosphorylated ERK1/2, JNK, and c-Jun were increased by coculture with TN2. MEK inhibitor U0126 reduced TN2 induced transactivation of SRE and AP1, as well as secretion of IL-8, by 83%, 87%, and 53%, respectively, of TN2.
CONCLUSIONS Transactivation of SRE and AP-1, through ERK/MAPK and JNK/SAPK cascades, respectively, was found in gastric cancer cells cocultured with H pylori. Direct contact with viable bacteria possessing intactcag PAI is a prerequisite for the onset of intracellular signalling leading to AP-1 transactivation.
Helicobacter pylori infects human gastric mucosa and induces chronic active gastritis.1 2 H pylori mediated inflammation is characterised by neutrophil infiltration mediated by interleukin (IL)-8 secretion from gastric epithelial cells.3-5 Activation of transcriptional factors nuclear factor κB (NFκB) and AP-1 plays an important role in IL-8 induction.6 Thecag pathogenicity island (cag PAI) of H pylori, a cluster of approximately 30 genes, is a prerequisite, at least for induction of NFκB activation.7 8
AP-1 is a multipotential transcriptional factor with various cytokines and chemokines such as IL-2, IL-3, IL-4, IL-6, IL-8, and tumour necrosis factor α (TNF-α), with its binding site in their promoter region.9-12 AP-1 is formed either as a homodimer of c-Jun or as a heterodimer of c-Jun and c-Fos, with the latter form known to be more potent than the former.13 14 WhileH pylori mediated activation of the JNK/SAPK cascade leading to AP-1 transactivation was recently reported,15 the status of c-Fos in relation toH pylori infection has not yet been elucidated. However, it is known that c-Fos expression is dependent on the ERK/MAPK cascade where activated Elk1 together with SRF16 17 binds the SRE binding site in the promoter region of c-fos.
Thus we hypothesised that the AP-1 binding site is activated inH pylori infection by the c-Jun/c-Fos heterodimer and sought to examine transactivation of SRE. In this study, we examined H pylori mediated transactivation of both SRE and AP-1 directly by using a luciferase reporter assay, and analysed the ERK/MAPK cascade, a pathway upstream of SRE transactivation.
Materials and methods
The TN2 strain, generously donated by Dr Nakao (Takeda Chemical Industries, Ltd, Osaka, Japan), was positive for two known virulence factors, CagA and VacA (vacuolating cytotoxin), and possessed intactcag PAI.18 Infection with this strain induced gastric cancer in Mongolian gerbils.19 The T68 strain, isolated from a Japanese patient at our institution, was negative for the above two virulence factors and lackedcag PAI.18 An isogeniccagE negative mutant (TN2-ΔcagE) was constructed by inserting a kanamycin resistant gene cassette into thecagE locus of cagPAI of TN2, as described in one of our previous papers.20The isogenic vacA negative mutant (TN2-ΔvacA) was constructed by disrupting the vacA gene of TN2 which was confirmed by Southern hybridisation and vacuolation assay, as we described previously.21 These strains were cultured on Columbia agar with 5% (vol/vol) horse blood and Dent antibiotic supplement (Oxoid, Basingstoke, UK) at 37°C under microaerobic conditions (Campy-Pak Systems; BBL, Cockeysville, Maryland, USA). Heat killedH pylori was prepared by treating the bacteria at 65°C for 60 minutes.
The reporter plasmids (Stratagene, La Jolla, California, USA) contained a Photinus pyralis luciferase gene driven by a basic promoter element (TATA) and seven AP-1 binding sites in pAP1-Luc, or by the basic promoter element and five SRE binding sites in pSRE-Luc. The efficacy of transfection was verified by cotransfection of a control plasmid (pRL-TK; Promega, Madison, Wisconsin, USA) containing the Renilla reniformis luciferase gene driven by the herpes simplex virus thymidine kinase.
Recombinant human IL-1β was purchased from Upstate Biotechnology (Lake Placid, New York, USA); recombinant human epidermal growth factor (EGF) from Sigma (St Louis, Missouri, USA); polyclonal antibodies for phospho-p44/p42 ERK1/2 (Thr202/Tyr204), p44/p42 ERK1/2, phospho-JNK/SAPK (Thr183/Tyr185), JNK/SAPK, phospho-c-Jun (Ser63), and c-Jun from New England BioLabs, Inc. (Beverly, Massachusetts, USA); and MEK inhibitor U0126 from Promega.
HUMAN CELL LINES
Human gastric cancer cells TMK1 (a gift from Dr E Tahara, Hiroshima University School of Medicine, Hiroshima, Japan)22 and MKN45 were maintained in RPMI 1640 (Sigma) supplemented with 10% fetal bovine serum (FBS) (Gibco/BRL),l-glutamine, penicillin G, and streptomycin. In coculture experiments, cancer cells (4×105 cells/ml) andH pylori (107 colony forming units/ml) were cultured in RPMI 1640 without antibiotics and supplemented with 10% FBS. To assess the effects of direct contact, cancer cells and bacteria were separated by a membrane filter (Nunc Tissue Culture Inserts No 162138, Nunc, Roskilde, Denmark).
TRANSACTIVATION OF SRE AND AP-1 SITES
Transactivation of SRE and AP-1 was evaluated with the luciferase reporter assay. Approximately 4×105 cells were plated onto six well tissue culture plates (Iwaki Glass, Chiba, Japan) and transfected 24 hours later with the expression vector pSRE-Luc (0.6 μg) or AP1-Luc (0.6 μg) using Effectene transfection reagent (Quiagen, Hilden, Germany). Control vector pRL-TK (0.01 μg) was added to each sample for standardisation of transfection efficacy. When pSRE-Luc reporter plasmids were transfected, the concentration of FBS in RPMI 1640 was reduced to 0.5% to minimise the effects of various factors in serum. After 24 hours, H pylori(106 colony forming units/ml) was added. After another 20 hours of coculture, cells were harvested in phosphate buffered saline and lysed in a luciferase lysis buffer, and luciferase assays were carried out using the PicaGene dual seapansy system (Toyo Ink, Tokyo, Japan). Firefly luciferase activity and seapansy luciferase activity were measured as relative light units with a luminometer (Lumat LB9507; EG&G Berthold, Bad Wildbad, Germany). Firefly luciferase activity was normalised for transfection efficiency based on seapansy luciferase activity. All assays were performed in four independent experiments.
U0126, a potent and specific MEK inhibitor of MEK1 and MEK2,23 was added to the culture medium at a final concentration of 10 μM, two hours before addition ofH pylori. The final concentration of dimethyl sulphoxide, used as a solvent, was 0.1% (vol/vol), which did not affect the activity or activation of any protein kinase examined.
PHOSPHORYLATION OF ERK1/2, JNK, AND c-JUN
After coculturing with H pylori or stimulation by EGF or IL-1β, cells were lysed in lysis buffer (50 mM Tris HCl buffer (pH 7.4) containing 1 mM EGTA, 2 mM dithiothreitol, 25 mM sodium β-glycerophosphate, 0.1 mM phenylmethylsulfonyl fluoride, and 10 μg/ml aprotinin) and centrifuged at 15 000 g for 15 minutes. An aliquot of supernatant containing 10 μg protein was applied on sodium dodecyl sulphate-polyacrylamide gel and transferred to a PVDF membrane (Amersham Pharmacia Biotech., Buckinghamshire, UK). The membrane was probed with an antibody to phosphorylated or total ERK1/2, JNK, or c-Jun at a dilution ratio of 1:1000 in Tris buffered saline-Tween. After washing, the membrane was incubated with horseradish peroxidase conjugated goat antirabbit immunoglobulin G at a dilution ratio of 1:1000 in Tris buffered saline-Tween, and applied to the ECL detection assay (Amersham Pharmacia Biotech).
After TMK1 cells were cocultured with H pylori for 20 hours, IL-8 concentration in the culture supernatant was measured with an EIA (PerSeptive Diagnostics, Framingham, Massachusetts, USA) using a standard curve obtained from sequentially diluted recombinant IL-8 samples.
Differences in means were analysed using the unpaired Student'st test, and Dunnett's multiple comparison using SAS software version 6.12 (SAS Institute, Inc., Carey, North Carolina); p<0.05 was considered significant.
SRE TRANSACTIVATION BY H PYLORI
The luciferase assay using pSRE-Luc showed that coculture with TN2 strain enhanced SRE transactivation in TMK1 and MKN45 cells to 310% and 680% of responses in untreated cells, respectively (fig 1A, B). TN2-ΔvacA showed comparable enhancement. However, T68, lacking cag PAI, reduced transactivation of AP-1 to 52% and 19% of the response induced by TN2 in TMK1 (fig 1A) and MKN45 (fig 1B), respectively. SRE transactivation induced by TN2-ΔcagE was also reduced compared with intact TN2 to 65% in TMK1 and 18% in MKN45. No transactivation was observed with heat killed TN2 or with viable TN2 separated by a permeable membrane (fig 1A, B).
AP-1 TRANSACTIVATION BY H PYLORI
Coculture with TN2 increased AP-1 transactivation in TMK1 and MKN45 to 240% and 370% of responses in untreated cells, respectively (fig 2A, 2B). While TN2-ΔvacA induced AP-1 transactivation comparable with TN2, AP-1 transactivation by T68 was reduced to 54% and 38% of responses induced by TN2 in TMK1 (fig 2A) and MKN45 (fig 2B), respectively. TN2-ΔcagEinduced an intermediate response in TMK1 (71%) and a reduced response in MKN45 (38%). No transactivation was induced with heat killed TN2 or with viable TN2 separated by a membrane filter.
PHOSPHORYLATION OF ERK1/2, JNK, AND c-JUN BY H PYLORI
The intracellular level of phosphorylated ERK1/2 was increased at 30 minutes in TMK1 cells cocultured with H pylori (TN2) (fig 3). This phosphorylation was more prominent than that induced by EGF (10 ng/ml) and lasted for another 60 minutes. The intracellular level of total ERK1/2 remained constant. The level of phosphorylated JNK was increased at 30 minutes in TMK1 cells cocultured with TN2 (fig 4), and the level of phosphorylated c-Jun was increased in TMK1 cells after 60 minutes of co-culture with H pylori (fig5).
EFFECTS OF MEK INHIBITION
When TMK1 cells were pretreated with the MEK inhibitor U0126,H pylori induced transactivation of SRE and AP-1 was significantly inhibited by 83% and 83%, respectively (fig6). The value of IL-8 secretion from unstimulated cells was 1626 (435) pg/ml. The MEK inhibitor also inhibited H pylori induced secretion of IL-8 from TMK1 cells by 53% (from 3095 (666) to 1443 (221) pg/ml) (fig7).
We have confirmed in this study that SRE and AP-1 are transactivated in gastric cancer cells when cocultured withH pylori. We also demonstrated not only phosphorylation of ERK1/2 in the ERK/MAPK cascade upstream of SRE but also phosphorylation of JNK and c-Jun in the JNK/SAPK cascade upstream of AP-1. Furthermore, the MEK inhibitor U0126 inhibited transactivation of not only SRE but also AP-1. Although there are no SRE sites in the promoter region of IL-8, induction of IL-8 secretion was inhibited by U0126. These results indicate that H pyloriinfection transactivates AP-1 through activation of both the JNK/SAPK cascade, resulting in c-Jun activation, and the ERK/MAPK cascade, resulting in c-Fos expression. Although AP-1 can be transactivated by c-Jun homodimer, inhibition of AP-1 transactivation by MEK inhibitor indicates that the c-Jun/c-Fos heterodimer is the main form of AP-1 inH pylori infection.
H pylori infection in gastric mucosa induces not only IL-8 but also IL-6 and TNF-α whose promoter regions contain AP-1 binding sites.24-26 The TNF-α gene also has a NFκB binding site in the promoter region, and H pylori infection is known to activate NFκB. ThusH pylori infection may induce TNF-α through transactivation of both AP-1 and NFκB binding sites, and TNF-α binding to its receptors on adjacent cells will switch on intracellular signals leading to inflammatory and immunological responses.
Although the pathway upstream of the ERK/MAPK cascade activated inH pylori infection has not yet been identified, our study indicated that direct contact with viableH pylori having intactcag PAI is required for the onset of signalling, resulting in induction of IL-8 secretion. Clinical isolates lacking cag PAI did not transactivate SRE or AP1 while those with intact cag PAI did (unpublished data).
During submission of this paper it was reported thatH pylori activates MAP kinase cascades and induces expression of c-fos and c-jun.27 However, enhancement of SRE, a key cis element ofc-fos, was not shown directly. Furthermore, the importance of direct contact with viable H pylori having intact cag PAI was not assessed.
These phenomena are difficult to interpret if molecules secreted byH pylori or attached on the surface of bacterial body triggers the signals by binding unknown receptors. It is supposed that (some of) cag PAI genes construct a type IV secretion system that transports bacterial derived molecules into host cells.28 Thus the cascade may be triggered by some molecules of bacterial origin transported into host cells. Recently, one of the cag PAI products, CagA, was shown to be inserted into host cells29although its relation to the cascades is not known.
In conclusion, we have confirmed transactivation of SRE and AP-1 through the ERK/MAPK and JNK/SAPK cascades, respectively, in gastric cancer cells cocultured with H pylori. Direct contact with viable bacteria possessing intactcag PAI is a prerequisite for the onset of intracellular signalling leading to AP-1 transactivation.
We thank Ms Mitsuko Tsubouchi for excellent technical assistance.
- Abbreviations used in this paper:
- pathogenicity island
- nuclear factor κB
- epidermal growth factor
- fetal bovine serum