Helicobacter pylori heat-shock protein 60
induces inflammatory responses through the
Toll-like receptor-triggered pathway in cultured
human gastric epithelial cells
Ryuta Takenaka,1Kenji Yokota,2Kiyoshi Ayada,2Motowo Mizuno,1
Ying Zhao,2Yoshihito Fujinami,2Song-Nan Lin,2Tatsuya Toyokawa,1
Hiroyuki Okada,1Yasushi Shiratori1and Keiji Oguma2
Department of Medicine and Medical Science (Medicine 1)1and Department of Bacteriology2,
Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama
Received 29 July 2004
Accepted 2 September 2004
Contact between Helicobacter pylori and gastric epithelial cells results in activation of NF-kB
followed by secretion of interleukin (IL)-8. However, host-cell receptor(s) and their ligands
involved in H. pylori-related IL-8 production have yet to be fully defined. In this study, the
interaction between Toll-like receptors (TLRs), which are host receptors for pathogens involved
in the innate immune response, and heat-shock protein (HSP) 60, an immune-potent antigen of
H. pylori, was examined during H. pylori-induced IL-8 secretion in vitro. Recombinant H. pylori
HSP60 (rHpHSP60) was prepared and added to cultured KATO III human gastric epithelial
cells with or without pre-incubation with mouse monoclonal anti-TLR2 or anti-TLR4 antibodies.
IL-8 mRNA expression and IL-8 protein release were analysed by Northern blotting and
immunoassay. Involvement of NF-kB activation was analysed immunocytochemically by
anti-NF-kB p65 antibody and ammonium pyrrolidinedithiocarbamate (PDTC), an inhibitor of
NF-kB-mediated transcriptional activation. rHpHSP60 induced IL-8 mRNA expression and
IL-8 secretion in a dose-dependent manner in KATO III cells. Anti-TLR2 antibody inhibited
rHpHSP60-induced IL-8 secretion by 75%, and anti-TLR4 antibody inhibited it by 30%.
rHpHSP60 induced nuclear translocation of NF-kB p65, which was inhibited by pretreatment
with anti-TLR2 antibody. Treatment with PDTC significantly decreased the secretion of IL-8
induced by rHpHSP60. These findings suggest that H. pylori HSP60 activates NF-kB and
induces IL-8 production through TLR-triggered pathways in gastric epithelial cells. Thus, it is
possible that H. pylori HSP60 and TLR interaction in host cells contributes to the development
of gastric inflammation caused by H. pylori infection.
Chronic infection with Helicobacter pylori is recognized as
thecause ofchronicactive gastritis,pepticulcer diseases and
mucosa-associated lymphoid tissue lymphoma, and is an
important risk factor in gastric carcinoma development
(Blaser, 1990; Parsonnet et al., 1991, 1994). Binding of
H. pylori to gastric epithelial cells, in particular through
the blood group antigen-binding adhesin, and the binding
of strains expressing the cag pathogenicity island results in
the production of proinflammatory chemokines such as
interleukin (IL)-8 (Crabtree et al., 1995a; Rad et al., 2002).
NF-kB and AP-1 function as intracellular messengers in this
process (Meyer-ter-Vehn et al., 2000; Sharma et al., 1998).
The 60 kDa heat-shock protein (HSP60), an immune-
potent antigen of H. pylori (Sharma et al., 1997), induces
(Yamaguchi et al., 1999). We have reported that humoral
immune responses to HSP60 are closely associated with
gastric inflammation and play a role in the pathogenesis of
mucosa-associated lymphoid tissue lymphoma (Hayashi
et al., 1998; Kawahara et al., 1999). With regard to host-cell
receptors for HSP60, sulfated glycolipids are reportedly
involved (Huesca et al., 1996), but the host-cell receptor(s)
involved in HSP60-induced IL-8 production have not been
fully elucidated. Some studies have shown that Toll-like
receptors (TLRs) act as receptors for bacterial structures
Abbreviations: FCS, fetal calf serum; G3PDH, glyceraldehyde-3-
phosphate dehydrogenase; HSP, heat-shock protein; IL, interleukin;
PDTC, ammonium pyrrolidinedithiocarbamate; PMB, polymyxin B;
rHpHSP60, recombinant H. pylori HSP60; TLR, toll-like receptor;
TNF-a, tumour necrosis factor a.
0002-7527 G 2004 SGMPrinted in Great Britain 3913
Microbiology (2004), 150, 3913–3922
such asLPS and HSP60 (Kawahara et al.,2001;Ohashi et al.,
2000; Sasu et al., 2001; Vabulas et al., 2001) in the develop-
ment of an innate immune response against bacterial
infection. Among the TLR family, TLR2 and/or TLR4
(Ohashi et al., 2000, Sasu et al., 2001; Vabulas et al., 2001)
are reportedly important in the recognition of HSP60 in
human cells and chlamydia. TLR2 and TLR5 are reported
to be required for H. pylori-induced NF-kB activation and
chemokine expression by epithelial cells (Smith et al., 2003).
In the present study, to investigate the involvement of
the interaction between bacterial HSP60 and host TLRs
in inflammatory responses against H. pylori infection, we
reacted cultured KATO III human gastric epithelial cells
with H. pylori HSP60 or recombinant H. pylori HSP60
(rHpHSP60) in the presence or absence of anti-TLR anti-
bodies. The effects on IL-8 production and NF-kB activa-
tion were measured.
Bacterial cells and culture. H. pylori (ATCC 43504) was cultured
on Brucella agar supplemented with 7% horse blood at 37uC under
microaerophilic conditions. After 5 days incubation, bacteria were
harvested in RPMI 1640 with or without 10% fetal calf serum
(FCS) to an OD600 of 1?0, corresponding to approximately
KATO III human gastric cell lines (obtained from the Japanese
Collection at Research Bioresources, Tokyo, Japan) were grown in
RPMI 1640 supplemented with 10% FCS, 50 U penicillin G sodium
5% CO2and 99% humidity. KATO III cells were used at a final
concentration of 56105cells ml21.
Preparation of rHpHSP60. rHpHSP60 was prepared as described
by Ishii et al. (2001). In brief, the ORF of HSP60 taken from
H. pylori ATCC 43504 genomic DNA was amplified and cloned into
the vector pGEX-5X3 (Amersham Biosciences). The resultant
plasmid was transformed into Escherichia coli DH5a. Cells were
cultured in Luria–Bertani broth containing 2% glucose and
ampicillin (100 mg ml21), harvested by centrifugation and disrupted
by sonication on ice for 5 min with a probe sonicator (Astron).
Soluble rHpHSP60 was purified by glutathione-Sepharose 4B affinity
chromatography (Amersham Biosciences) according to the manufac-
Protein production in KATO III cells (ELISA). KATO III cells
were co-cultured with various amounts of intact H. pylori or
various concentrations of rHpHSP60. Cell-culture supernatants were
collected and IL-8 was measured by ELISA (IL-8 ELISA Develop-
ment kit; Techne). To evaluate the interaction between rHpHSP60
and TLRs in IL-8 production, cells were preincubated with
10 mg TL2.1 ml21(anti-human TLR2 mouse mAb; Sanbio), 10 mg
HTA125 ml21(anti-human TLR4 mouse mAb; Sanbio) or 10 mg
control irrelevant mouse mAb ml21(anti-botulinum toxin type C),
which we had prepared previously (Mahmut et al., 2002), for
30 min before stimulation under FCS-free conditions. In addition,
KATO III cells were pretreated with ammonium pyrrolidinedithio-
carbamate (PDTC) (Sigma–Aldrich), an inhibitor of NF-kB-mediated
transcriptional activation (Kawai et al., 1995; Ziegler-Heitbrock
et al., 1993), for 1 h before being reacted with intact H. pylori
(16107c.f.u. ml21) or rHpHSP60 (50 mg ml21). Cell-culture super-
natants were collected and IL-8 was measured as described above.
To exclude the possible effects of trace amounts of LPS on IL-8
secretion induced by rHpHSP60 preparations, 20 mg polymyxin B
ml21(PMB) (Wako Pure Chemical Industries), a potent LPS
antagonist (Stokes et al., 1989), was added when KATO III cells
were reacted with rHpHSP60, and E. coli LPS (Sigma–Aldrich) was
used as control. In addition, a PMB-agarose gel column (Pierce) was
used to remove LPS and LPS-associated molecules in rHpHSP60
preparations. The endotoxin activity of rHpHSP60 and PMB-agarose
column fractions was determined by using the LAL assay kit
(BioWhittaker) according to the manufacturer’s instructions. IL-8
secretion from KATO III cells that had been reacted with the original
rHpHSP60 preparation was compared with that from cells reacted
with the PMB-agarose column pass-through fractions.
NF-kB activation in KATO III cells that had been reacted with
rHpHSP60 was detected and quantified by the use of Trans AM NF-
kB (Active Motif) according to the manufacturer’s instructions.
Briefly, nuclear extracts were prepared and incubated for 1 h on a 96-
well plate with immobilized oligonucleotide containing the NF-kB
to p65 (RelA), a subunit of the activated NF-kB dimer, horseradish
peroxidase-conjugated secondary antibody was added, followed by
developing solution included in Trans AM NF-kB. The optical density
was read at 450 nm with a reference wavelength of 655 nm on a
spectrophotometer. Recombinant tumour necrosis factor a (TNF-a)
(Sigma–Aldrich) was used as a positive control.
Detection of rHpHSP60 in KATO III cells with bacterial or
rHSP stimulation. To measure the rHpHSP60 concentration when
live or heat-killed bacteria contacted and stimulated KATO III cells,
the culture supernatants and cells were analysed by Western blotting.
KATO III cells (56106cells ml21) were reacted with several con-
centrations of rHpHSP60 (100–1 mg ml21), or intact or heat-killed
H. pylori [from OD600=0?5 (approx. 56107c.f.u. or cells ml21) to
OD600=0?01 (approx. 16105c.f.u. or cells ml21)]. After incubation
for 8 h, supernatants were gently collected. Cells were resuspended
in 1 ml PBS. Aliquots of 10 ml of supernatants and cell suspensions
were taken for SDS-PAGE and transferred to polyvinylidene difluoride
membranes (Millipore). The membranes were reacted with 1:3000-
diluted anti-H. pylori HSP60 polyclonal antibody prepared as
described previously (Yunoki et al., 2000). After washing with PBS,
the membranes were reacted with peroxidase-conjugated anti-rabbit
immunoglobulins (Dako). The reactions were detected by ECL
Western blotting detection reagents (Amersham Pharmacia). The
luminol reaction was visualized by a LAS-1000mini Bio-Imaging
Analyser System (Fuji).
mRNA expression in KATO III cells (RT-PCR and Northern
blotting). Using Northern blotting analysis, we analysed IL-
8 mRNA expression in KATO III cells treated with H. pylori and
rHpHSP60. Total RNA was isolated from KATO III cells using RNA
zol-B (Tel-Test Inc.), and IL-8 cDNA was prepared by RT-PCR
using ReverTra Dash (Toyobo) with the following oligonucleotide
primer set for IL-8 (59-AAGGAACCATCTCACTG-39 and 59-GAT-
TCTTGGATACCACAGAG-39; Crabtree et al., 1994) and labelled
with horseradish peroxidase by an ECL direct nucleic acid labelling
and detection system (Amersham Biosciences). Human glyceraldehyde-
3-phosphate dehydrogenase (G3PDH) cDNA was also prepared, as a
control, with a primer set in an RT-PCR kit (ReverTra Dash)
according to the manufacturer’s protocol. After treatment with
rHpHSP60 (50 mg ml21), total RNA was isolated from KATO III
cells and separated by electrophoresis (15 mg per lane) through a
1% agarose gel containing formaldehyde and transferred to a nylon
membrane (Hybond-N+; Amersham Biosciences) in 206 SSC
(16 SSC is 0?15 M sodium chloride/15 mM sodium citrate). The
membrane was baked for 30 min at 80uC and cross-linked by a UV
cross-linker (Stratagene). After pre-hybridization, the membrane was
R. Takenaka and others
hybridized with labelled IL-8 or G3PDH cDNA probes for 12 h at
42uC. The membrane was washed twice with 0?56 SSC containing
0?4% SDS for 10 min at 55uC and then washed twice more with
26 SSC for 5 min at room temperature. RNA in the blots was detected
by using the ECL direct nucleic acid labelling and detection system
and analysed with the LAS-1000mini Bio-Imaging Analyser System.
TLR mRNA and protein expression in KATO III cells was examined by
RT-PCR. Total RNA was isolated from KATO III cells by using RNA
zol-B. After DNase I (Invitrogen) treatment, RT-PCR was performed
using ReverTra Dash. PCR amplification was performed for 35 cycles
at 95uC for 10 s, 53uC for 2 s and 74uC for 30 s using the following
oligonucleotide primer sets: for TLR2, 59-CTTATCCAGCACACGA-
ATACACAG-39 (sense 1455–1478) and 59-TGGAGAGTCACACAG-
GTAATTTGC-39 (antisense 1807–1830); and for TLR4, 59-CTCCA-
GGTTCTTGATTACAGTCTC-39 (sense 1800–1823) and 59-TAGCT-
CATTCCTTACCCAGTCCT-39 (antisense 2218–2240). These primer
sets were designed by using GENETYX-MAC (Genetyx Corporation)
based on the DNA sequences of TLR2 and TLR4 from the GenBank
Database. As a housekeeping gene for the control reaction, PCR was
performed simultaneously with G3PDH primers (ReverTra Dash).
Appropriate negative controls, including amplification of a non-
specific RT-PCR reaction product, were included in each run. PCR
products were resolved on 2% agarose gels and DNA fragments were
visualized under UV light by staining with ethidium bromide.
Real-time PCR for mRNA of TLR2 and 4. Total RNA was iso-
lated from KATO III cells using RNA zol-B (Tel-Test Inc.). After
DNase I (Invitrogen) treatment, cDNA was prepared using ReverTra
Dash. Quantitative PCR analysis was performed on a LightCycler
using the LightCycler Primer set of human TLR2, TLR4 and
G6PDH. Real-time PCR was done in a 24 ml volume consisting of
10 ml cDNA sample, 10 ml H2O, 2 ml LightCycler primer set and 2 ml
LightCycler FastStart DNA master SybrGreen I. The reaction
mixtures were loaded into capillary tubes and thermal cycling was
carried out as follows: activation of hotstart Taq DNA polymerase,
95uC for 10 min; cycling, 95uC for 10 s, 68uC for 10 s, 72uC for
16 s for 40 cycles. The DNA concentration was calculated by
LightCycler software and the ratio of TLR2 or TLR4/GAPDH was
Immunocytochemistry of TLR and NF-kB in KATO III cells.
KATO III cells were seeded on chamber slides and fixed in 70%
methanol. After incubation with TL2.1 (anti-human TLR2 antibody)
or HTA125 (anti-human TLR4 antibody) as primary antibodies,
the cells were incubated with tetraethylrhodamine isothiocyanate-
conjugated goat anti-mouse immunoglobulins (Sigma–Aldrich). As
negative controls, irrelevant mouse mAb (anti-botulinum toxin type
C) and PBS were used instead of the primary antibodies. The stained
cells were observed under a confocal laser-scanning microscope
(LSM510; Carl Zeiss).
For NF-kB activation analysis, KATO III cells were seeded in poly-L-
lysine (Wako Pure Chemicals Industries)-treated chamber slides
3 days before the experiment. After pretreatment with TL2.1 (anti-
human TLR2 antibody), HTA125 (anti-human TLR4 antibody) or
medium control for 30 min, cells were stimulated with rHpHSP60
or intact H. pylori for 1 h and fixed in methanol (70%) containing
H. pylori (10
IL-8 (ng ml )
IL-8 (ng ml )
0 0.11 10 50100 200
_7 c.f.u. ml
_1) rHpHSP60 (pg ml
IL-8 mRNAIL-8 mRNA
G3PDH mRNAG3PDH mRNA
Fig. 1. IL-8 protein release from and mRNA expression in KATO III cells by H. pylori and rHpHSP60. KATO III cells were
treated for 24 h with the indicated numbers of intact H. pylori (a) or concentrations of rHpHSP60 (b) and harvested. IL-8
protein levels in the culture supernatant were measured by immunoassay. Both H. pylori and rHpHSP60 induced IL-8 release
in a dose-dependent manner. Data are expressed as means±SD (n=3). *, P<0?01 vs control. After treatment of KATO III
cells with intact H. pylori (c) or rHpHSP60 (50 mg ml”1) (d) for the indicated time periods, Northern blotting analysis was
performed with IL-8 and G3PDH cDNA probes. IL-8 mRNA increased and reached a maximum 3 h after rHpHSP60
stimulation and returned to its original level 6 h after the stimulation.
H. pylori HSP60 induces IL-8 through TLR
glycerol (30%) at 220uC for 30 min. After washing, cells were treated
with 0?1% Triton X100 for 4 min, incubated with rabbit anti-p65
isothiocyanate-conjugated F(ab9)2 fragments of goat anti-rabbit
immunoglobulins (Sigma–Aldrich). The stained cells were observed
under a confocal laser-scanning microscope.
Statistical analysis. Results were expressed as means±SD. Data
were compared using Scheffe’s multiple comparison test and differ-
ences were considered significant at P values <0?05.
rHpHSP60 increased the protein release and
mRNA expression of IL-8 in KATO III cells
rHpHSP60, as well as intact H. pylori, significantly
increased IL-8 release from KATO III gastric epithelial
cells in a dose-dependent manner (Fig. 1a, b). Recently,
release of TNF-a from murine macrophages that had been
reacted with recombinant human HSP60 was found due
to LPS contamination of the human HSP60 preparations
(Gao & Tsan, 2003). To exclude the possibility that the
release of IL-8 induced by our rHpHSP60 similarly was
caused by contaminating LPS, we reacted KATO III cells
with LPS or rHpHSP60 in the presence or absence of PMB;
PMB did not inhibit IL-8 release, whereas LPS-induced IL-8
secretion was inhibited by PMB (Fig. 2a). We also searched
for endotoxin activities in our rHpHSP60 preparations
(Fig. 2b). Even though the endotoxin activity of the original
rHpHSP60 preparation was negligible, we further purified
rHpHSP60 by passing it through a PMB-agarose column.
No endotoxin activity was detected in fractions passed once
(PMB-1) or twice (PMB-2) through the column (Fig. 2b),
yet both these fractions induced as much IL-8 release from
KATO III cells as the original rHpHSP60 preparation
for IL-8 secretion induced by rHpHSP60.
rHpHSP60 (mg ml
IL-8 (ng ml )
IL-8 (ng ml )
500 105010 50 10
activity (EU ml
500 100 50
500 100 50
100 50 10510
_5 Intact H. pylori
_5 Heat-killed H. pylori
Fig. 2. Analysis of the possible involvement of contaminating LPS in IL-8 release induced by rHpHSP60. (a) Effect of PMB
treatment. KATO III cells were pretreated with PMB at a dose of 20 mg ml”1before being reacted with rHpHSP60
(50 mg ml”1) or LPS (1 mg ml”1). The treatment with PMB inhibited release of IL-8 induced by LPS, but not by rHpHSP60.
Data are expressed as means±SD (n=3). *, P<0?05 vs LPS without PMB pretreatment. (b) Endotoxin activity in rHpHSP60
and fractions passed through a PMB-agarose column. Endotoxin activity of the rHpHSP60 (50 mg ml”1) preparation is
negligible when compared with that of LPS (1 mg ml”1), and no endotoxin activity is detected in fractions passed once
(PMB-1, 50 mg ml”1) or twice (PMB-2, 50 mg ml”1) through a PMB-agarose column. Endotoxin activity was estimated by use
of an LAL assay kit. (c) IL-8 release induced by rHpHSP60 compared with IL-8 release induced by the rHpHSP60 fractions,
PMB-1 and PMB-2. KATO III cells were treated with PMB-1 or PMB-2 at the indicated concentrations and IL-8 release was
measured by immunoassay. The PMB-1 or PMB-2 fractions induced the same level of IL-8 release as that induced by the
original rHpHSP60 preparation. Data are expressed as means±SD (n=3). *, P<0?001 vs control without stimulation; 3,
P<0?01 vs control without stimulation. (d) The concentration of rHpHSP60 in KATO III during H. pylori stimulation of the cells
was measured by Western blotting.
R. Takenaka and others
The concentrations of rHpHSP60 in KATO III cells while
H. pylori stimulates the cells were measured by Western
blotting (Fig. 2d). The rHpHSP60 concentration of the
co-culture with intact or heat-killed H. pylori (16107cells
ml21) and KATO III cells was the same as that of culture
supernatant following stimulation with rHpHSP60 (50 mg
ml21). rHpHSP60 was detected in supernatants stimulated
at 100 and 50 mg rHpHSP60 ml21; however, it was not
detected in KATO III cells taken from co-culture with
rHpHSP60. In supernatant collected from intact H. pylori
stimulation, rHpHSP60 was detected at 16106bacterial
cells; however, it was not detected following stimulation
with heat-killed H. pylori.
Expressionof IL-8 mRNA in KATO III cells stimulated with
H. pylori (Fig. 1c) or rHpHSP60 (Fig. 1d) was examined by
Northern blotting analysis. IL-8 mRNA expression increa-
sed and reached a maximum level at 3 h after rHpHSP60
Involvement of TLRs in rHpHSP60-induced IL-8
release from KATO III cells
We next examined whether interaction of rHpHSP60 with
TLRs was involved in IL-8 secretion induced by rHpHSP60.
First, we confirmed the expression of TLR2 and TLR4
in KATO III cells. TLR2 and TLR4 mRNAs were constitu-
tively expressed in the cells, irrespective of the presence
of rHpHSP60 stimulation, as demonstrated by RT-PCR
(Fig. 3a). Also, TLR2 and TLR4 proteins were identified
immunocytochemically along the cell surface of KATO III
cells (Figs 4b and 3c). Real-time PCR was performed to
confirm the expression of TLR2 or TLR4 mRNA. The
number of PCR product (copies) are shown in Fig. 3(d)
and the TLR2 or TLR4/G6PDH ratio is shown in Fig. 3(e).
rHSP60 did not enhance TLR mRNA expression.
When we pretreated KATO III cells with TL2.1 (anti-
TLR2 mAb) or HTA125 (anti-TLR4 mAb) before reaction
10 PCR products
25 3530 4025 3530 40
Fig. 3. TLR2 and TLR4 expression in KATO III cells. (a) KATO III cells were incubated in the presence or absence of
rHpHSP60 and TLR mRNA was detected by RT-PCR. PCR bands of TLR2 and TLR4 mRNAs are present irrespective of
rHpHSP60 treatment. (b, c) Immunocytochemical staining of KATO III cells with TL2.1 anti-TLR2 antibody (b) or HTA125 anti-
TLR4 antibody (c). TLR2 (b) and TLR4 (c) are stained on the surface of KATO III cells. (d) Increased numbers of PCR
products of TLR2 (left) and TLR4 (right) were detected in real-time PCR. Filled diamonds, 0 h; open squares, 1 h; filled
triangles, 3 h; crosses, 6 h; stars, 9 h. (e) The ratios of TLR2 or TLR4/G6PDH after 35 cycles were calculated.
H. pylori HSP60 induces IL-8 through TLR
with intact H. pylori, IL-8 secretion from KATO III was
significantly inhibited by approximately 50 or 40%,
respectively (Fig. 4a). Also in rHpHSP60-induced IL-8
secretion, a significant decrease (75%) was observed after
treatment with TL2.1, but not after treatment with HTA125
NF-kB activation in KATO III cells reacted with
Next, we investigated NF-kB activation in KATO III cells
reacted with rHpHSP60. An increase of intranuclear NF-
kB was observed 10 min after rHpHSP60 stimulation and
reached a maximum level 1 h after treatment (Fig. 5a). This
reaction was significantly inhibited by anti-TLR2 antibody
(TL2.1) treatment (Fig. 5b). Immunocytochemically, p65
(RelA), a subunit of the activated NF-kB dimmer, was
localized exclusively in the cytoplasm before reaction with
intact H. pylori or rHpHSP60 (Fig. 6a). After reaction with
intact H. pylori (Fig. 6b) or rHpHSP60 (Fig. 6c), p65 was
translocated into the nucleus. The HSP60-induced trans-
location of p65 was inhibited by pretreatment with TL2.1
anti-TLR2 antibody (Fig. 6d).
When we pretreated KATO III cells with PDTC, an inhi-
bitor of NF-kB-mediated transcriptional activation, the
release of IL-8 induced by reaction with intact H. pylori
or rHpHSP60 stimulation was significantly inhibited
(Fig. 7a, b).
In this study, we focused on the possibility that H. pylori
HSP60 is a bacterial ligand in the host inflammatory
response to H. pylori infection. We demonstrated in KATO
III human gastric cells that rHpHSP60 induces the secretion
IL-8 (ng ml )
IL-8 (ng ml )
Fig. 4. The effect of pretreatment of KATO III cells with anti-
TLR antibodies on rHpHSP60-induced IL-8 release. KATO III
cells were pretreated with 10 mg TL2.1 anti-TLR2 antibody
ml”1or 10 mg HTA125 anti-TLR4 antibody ml”1before the
rHpHSP60 (50 mg ml”1) (b); IL-8 release was measured by
immunoassay. Treatment with anti-TLR2 or anti-TLR4 antibody
significantly inhibited H. pylori-induced IL-8 release, by approxi-
mately 50 and 40%, respectively (a). In the experiments with
rHpHSP60, treatment with anti-TLR2, but not anti-TLR4 anti-
body significantly decreased IL-8 release (75%) (b). Results
are shown as means±SD (n=3). *, P<0?01 vs intact H. pylori
stimulation (a) or vs rHpHSP60 stimulation (b).
(16107c.f.u. ml”1) (a)or
0 103060 120240
rHpHSP60. NF-kB in nuclear extracts was measured after
treatment of KATO III cells with rHpHSP60. The amounts of
intranuclear NF-kB increased and reached a maximum level 1 h
after treatment (a). Treatment with anti-TLR2, but not anti-TLR4
antibody significantly inhibited NF-kB activation (b). Data are
shown as means±SD (n=3). *, P<0?05 vs control without
NF-kB activation inKATO IIIcells treatedwith
R. Takenaka and others
of the inflammatory cytokine IL-8 and increases the
expression of IL-8 mRNA. Also, we established that TLRs,
especially TLR2, are host receptors for H. pylori HSP60
that are likely to play an important role in the secretion
of IL-8, and that the interaction between rHpHSP60 and
TLR2 induced IL-8 production via the signalling pathway
involving NF-kB activation.
HSPs are a family of proteins induced in prokaryotic and
eukaryotic cells by environmental stress. These proteins
function as chaperones to facilitate folding, unfolding and
translocation of intracellular polypeptides (Ellis, 1990;
Young, 1990). HSP60 has been shown to play a role in
the adherence and attachment of H. pylori to gastric
epithelium (Huesca et al., 1996; Yamaguchi et al., 1997) and
Fig. 6. Immunocytochemical analysis of NF-kB nuclear translocation in KATO III cells treated with rHpHSP60. Cells were
treated with medium alone (a), intact H. pylori (b), rHpHSP60 (c) or rHpHSP60 after pre-incubation with TL2.1 anti-TLR2
antibody (d). The cells were then reacted with anti-p65 (RelA) antibody. Before stimulation, p65 is present in the cytoplasm
only (a). After stimulation with intact H. pylori (b) or rHpHSP60 (c), p65 is translocated into the nucleus. This rHpHSP60-
induced translocation of p65 was inhibited by pretreatment with TL2.1 anti-TLR2 antibody (d).
H. pylori HSP60 induces IL-8 through TLR
to induce IL-8 secretion from KATO III human gastric
epithelial cells (Yamaguchi et al., 1999). The results of
the present study, which confirmed that rHpHSP60 can
induce IL-8 release from KATO III cells and further found
that IL-8 secretion was accompanied by an increase of
IL-8 mRNA, indicate that H. pylori HSP60 is a bacterial
virulence factor that can induce host inflammatory
It has been reported that commercially available recombi-
nant human HSP60 induced TNF-a release from murine
macrophages in a manner similar to that of LPS, but Gao
& Tsan (2003) reported that the TNF-a-inducing activity
of this HSP60 was due to the presence of contaminating
LPS and LPS-associated molecules. In the present study,
three experiments excluded the possibility that LPS con-
tamination of our rHpHSP60 preparation was responsible
for the release of IL-8: (1) the addition of PMB did not
inhibit rHpHSP60-induced IL-8 release; (2) endotoxin
activities in the rHpHSP60 preparation were negligible;
and (3) fractions of rHpHSP60 further purified by passage
through a PMB-agarose column induced as much IL-8
release from KATO III as did the original rHpHSP60
Sulfated glycolipid is reportedly a host-cell receptor for H.
pylori HSP60 (Huesca et al., 1996), and HSP60s of human
cells and chlamydia bind to TLR2 or TLR4 (Ohashi et al.,
2000; Sasu et al., 2001; Vabulas et al., 2001). In H. pylori
infection, TLR4 involvement is reported in the activation
of mitogen oxidase 1 by LPS of H. pylori in gastric pit cells
(Kawahara et al., 2001). Other investigators reported that
human gastric epithelial cells recognize and respond to
H. pylori at least in part via TLR2 and TLR5 (Smith et al.,
2003). Recently, H. pylori HSP60 was reported to mediate
IL-6 production by macrophage via TLRs independent
mechanism (Gobert et al., 2004). However, the interaction
of the TLRs with H. pylori HSP60 on IL-8 production by
human gastric epithelial cells had not been studied. In the
present study, we demonstrated that treatment of KATO III
cells with anti-TLR antibodies, especially anti-TLR2, inhi-
bited HSP60-induced IL-8 secretion. This result is persua-
sive evidence that TLRs are host-cell receptors for H. pylori
HSP60 in the induction of initial inflammatory responses.
In vertebrates, TLRs play a key role in the recognition of
infectious pathogens and initiate innate immunity as a first
lineof defence(Aderem&Ulevitch, 2000).Forthis purpose,
it is reasonable that TLRs should be directed to specific
motifs on pathogens. Because HSPs are highly conserved
proteins with significant molecular mimicry among prokar-
yotic and eukaryotic cells, the recognition of H. pylori HSP
by TLRs is a likely mechanism for the initiation of innate
immunity and the concurrent host inflammatory response
against this bacterium. In support of this possibility, we
found that anti-TLR2 antibody significantly inhibited IL-8
secretion. However, the inhibition was incomplete, even
when anti-TLR4 antibody was added. Thus, receptors
other than TLR2 and TLR4 may also recognize H. pylori
HSP60. At present it remains unclear whether the observed
induction of IL-8 is specific to HSP60 from H. pylori.
However, preliminary data suggest that production of
this cytokine from monocytes involves the less conserved
regions of the protein.
Activation of the TLR signalling pathway leads to NF-kB
translocation to the nucleus and transactivation (Aderem
& Ulevitch, 2000; Akira & Hemmi, 2003). In the present
study, we demonstrated that nuclear translocation of RelA
(p65) was induced by the addition of rHpHSP60 and
inhibited by pretreatment with the anti-TLR antibody.
Involvement of NF-kB activation was further confirmed
by the observation that IL-8 secretion was inhibited after
treatment with PDTC, an inhibitor of NF-kB-mediated
transcriptional activation. However, the possible involve-
ment of other signalling pathways, such as the mitogen-
activated protein kinase pathway, which is also triggered
after TLR ligation (Aderem & Ulevitch, 2000; Akira &
Hemmi, 2003), deserves consideration.
IL-8 (ng ml )
IL-8 (ng ml )
Fig. 7. The effect of PDTC on rHpHSP60-induced IL-8 release
from KATO III cells. KATO III cells were pretreated with the
indicated amounts of PDTC, an inhibitor of NF-kB-mediated
transcriptional activation, 1 h before reaction with intact H.
pylori or rHpHSP60; IL-8 release was measured by immuno-
assay. PDTC treatmentsignificantly
induced by either intact H. pylori (a) or rHpHSP60 (b). Data
are shown as means±SD (n=3). *, P<0?001 vs H. pylori (a)
and rHpHSP60 (b) stimulation without PDTC pretreatment. 3,
P<0?01 vs rHpHSP60 stimulation without PDTC pretreatment.
R. Takenaka and others
Adherence and attachment of H. pylori to gastric epithelial
cells cause host responses, including the synthesis of various
inflammatory mediators and cytokines that lead to the
development of gastric injury. The cag pathogenicity island
of H. pylori is reportedly important in the production of
these cytokines through NF-kB activation (Covacci et al.,
1999; Crabtree et al., 1995b; Sharma et al., 1998). Other
reports have indicated that not only cag pathogenicity
island, but also flagella, LPS, outer-membrane protein
and HSP60 of H. pylori are associated with IL-8 induction
from epithelial cells or monocytes (Bhattacharyya et al.,
2002; Cunningham et al., 2000; Lee et al., 2003; Yamaguchi
et al., 1999; Yamaoka et al., 2002). These non-cag antigens
are located on the bacterial surface and make direct contact
with host cells. The cag-independent pathway of cytokine
induction may also involve induction of mucosal inflam-
mation. The results of the present study, which indicate that
H. pylori HSP60 acts as a bacterial virulence factor for the
induction of host inflammatory responses, acting through
binding to TLRs, further characterize the mechanisms by
which H. pylori can cause disease.
The authors wish to thank Dr William R. Brown (Denver Health
Medical Center, Denver, CO, USA) for his assistance in preparation
of this manuscript.
Aderem, A. & Ulevitch, R. J. (2000). Toll-like receptors in the
induction of the innate immune response. Nature 406, 782–787.
Akira, S. & Hemmi, H. (2003). Recognition of pathogen-associated
molecular patterns by TLR family. Immunol Lett 85, 85–95.
Bhattacharyya, A., Pathak, S., Datta, S., Chattopadhyay, S., Basu, J.
& Kundu, M. (2002). Mitogen-activated protein kinases and nuclear
factor-kappaB regulate Helicobacter pylori-mediated interleukin-8
release from macrophages. Biochem J 368, 121–129.
Blaser, M. J. (1990). Helicobacter pylori and the pathogenesis of
gastroduodenal inflammation. J Infect Dis 161, 626–633.
Covacci, A., Telford, J. L., Del Giudice, G., Parsonnet, J. &
Rappuoli, R. (1999). Helicobacter pylori virulence and genetic
geography. Science 284, 1328–1333.
Crabtree, J. E., Wyatt, J. I., Trejdosiewicz, L. K., Peichl, P., Nichols,
P. H., Ramsay, N., Primrose, J. N. & Lindley, I. J. (1994).
Interleukin-8 expression in Helicobacter pylori infected, normal,
and neoplastic gastroduodenal mucosa. J Clin Pathol 47, 61–66.
Crabtree, J. E., Covacci, A., Farmery, S. M., Xiang, Z., Tompkins,
D. S., Perry, S., Lindley, I. J. & Rappuoli, R. (1995a). Helicobacter
pylori induced interleukin-8 expression in gastric epithelial cells is
associated with CagA positive phenotype. J Clin Pathol 48, 41–45.
Crabtree, J. E., Xiang, Z., Lindley, I. J., Tompkins, D. S., Rappuoli, R.
& Covacci, A. (1995b). Induction of interleukin-8 secretion from
gastric epithelial cells by a cagA negative isogenic mutant of
Helicobacter pylori. J Clin Pathol 48, 967–969.
Cunningham, M. D., Shapiro, R. A., Seachord, C., Ratcliffe, K.,
Cassiano, L. & Darveau, R. P. (2000). CD14 employs hydrophilic
regions to ‘capture’ lipopolysaccharides. J Immunol 164, 3255–3263.
Ellis, R. J. (1990). The molecular chaperone concept. Semin Cell Biol
Gao, B. & Tsan, M. F. (2003). Recombinant human heat shock
protein 60 does not induce the release of tumor necrosis factor alpha
from murine macrophages. J Biol Chem 278, 22523–22529.
Gobert, A. P., Bambou, J. C., Werts, C., Balloy, V., Chignard, M.,
Moran, A. P. & Ferrero, R. L. (2004). Helicobacter pylori heat shock
protein 60 mediates interleukin-6 production by macrophages via a
toll-like receptor (TLR)-2-, TLR-4-, and myeloid differentiation
factor 88-independent mechanism. J Biol Chem 279, 245–250.
Hayashi, S., Sugiyama, T., Yokota, K., Isogai, H., Isogai, E.,
Oguma, K., Asaka, M., Fujii, N. & Hirai, Y. (1998). Analysis of
immunoglobulin A antibodies to Helicobacter pylori in serum and
gastric juice in relation to mucosal inflammation. Clin Diagn Lab
Immunol 5, 617–621.
Huesca, M., Borgia, S., Hoffman, P. & Lingwood, C. A. (1996).
Acidic pH changes receptor binding specificity of Helicobacter pylori:
a binary adhesion model in which surface heat shock (stress)
proteins mediate sulfatide recognition in gastric colonization. Infect
Immun 64, 2643–2648.
Ishii, E., Yokota, K., Sugiyama, T. & 7 other authors (2001).
Immunoglobulin G1 antibody response to Helicobacter pylori heat
shock protein 60 is closely associated with low-grade gastric mucosa-
associated lymphoid tissue lymphoma. Clin Diagn Lab Immunol 8,
Kawahara, Y., Yokota, K., Mizuno, M. & 7 other authors (1999).
Antibodies to human gastric epithelial cells and heat shock protein
60 in Helicobacter pylori positive mucosa associated lymphoid tissue
lymphoma. Gut 45, 20–23.
Kawahara, T., Teshima, S., Oka, A., Sugiyama, T., Kishi, K. &
Rokutan, K. (2001). Type I Helicobacter pylori lipopolysaccharide
stimulates toll-like receptor 4 and activates mitogen oxidase 1 in
gastric pit cells. Infect Immun 69, 4382–4389.
Kawai, M., Nishikomori, R., Jung, E. Y., Tai, G., Yamanaka, C.,
Mayumi, M. & Heike, T. (1995). Pyrrolidine dithiocarbamate inhibits
intercellular adhesion molecule-1 biosynthesis induced by cytokines
in human fibroblasts. J Immunol 154, 2333–2341.
Lee, S. K., Stack, A., Katzowitsch, E., Aizawa, S. I., Suerbaum, S. &
Josenhans, C. (2003). Helicobacter pylori flagellins have very low
intrinsic activity to stimulate human gastric epithelial cells via TLR5.
Microbes Infect 5, 1345–1356.
Mahmut, N., Inoue, K., Fujinaga, Y. & 7 other authors (2002).
Characterisation of monoclonal antibodies against haemagglutinin
associated with Clostridium botulinum type C neurotoxin. J Med
Microbiol 51, 286–294.
Meyer-ter-Vehn, T., Covacci, A., Kist, M. & Pahl, H. L. (2000).
cascades and induces expression of the proto-oncogenes c-fos and
c-jun. J Biol Chem 275, 16064–16072.
Ohashi, K., Burkart, V., Flohe, S. & Kolb, H. (2000). Cutting edge:
heat shock protein 60 is a putative endogenous ligand of the toll-like
receptor-4 complex. J Immunol 164, 558–561.
Parsonnet, J., Friedman, G. D., Vandersteen, D. P., Chang, Y.,
Vogelman, J. H., Orentreich, N. & Sibley, R. K. (1991). Helicobacter
pylori infection and the risk of gastric carcinoma. N Engl J Med 325,
Parsonnet, J., Hansen, S., Rodriguez, L., Gelb, A. B., Warnke, R. A.,
Jellum, E., Orentreich, N., Vogelman, J. H. & Friedman, G. D. (1994).
Helicobacter pylori infection and gastric lymphoma. N Engl J Med
Rad, R., Gerhard, M., Lang, R., Schoniger, M., Rosch, T., Schepp, W.,
Becker, I., Wagner, H. & Prinz, C. (2002). The Helicobacter pylori
H. pylori HSP60 induces IL-8 through TLR
blood group antigen-binding adhesin facilitates bacterial coloniza- Download full-text
tion and augments a nonspecific immune response. J Immunol 168,
Sasu, S., LaVerda, D., Qureshi, N., Golenbock, D. T. & Beasley, D.
(2001). Chlamydia pneumoniae and chlamydial heat shock protein 60
stimulate proliferation of human vascular smooth muscle cells via
toll-like receptor 4 and p44/p42 mitogen-activated protein kinase
activation. Circ Res 89, 244–250.
Sharma, S. A., Miller, G. G., Peek, R. J., Jr, Perez-Perez, G. & Blaser,
M. J. (1997). T-cell, antibody, and cytokine responses to homologs of
the 60-kilodalton heat shock protein in Helicobacter pylori infection.
Clin Diagn Lab Immunol 4, 440–446.
Sharma, S. A., Tummuru, M. K., Blaser, M. J. & Kerr, L. D. (1998).
Activation of IL-8 gene expression by Helicobacter pylori is regulated
by transcription factor nuclear factor-kappa B in gastric epithelial
cells. J Immunol 160, 2401–2407.
Smith, M. F., Jr, Mitchell, A., Li, G., Ding, S., Fitzmaurice, A. M.,
Ryan, K., Crowe, S. & Goldberg, J. B. (2003). Toll-like receptor
(TLR) 2 and TLR5, but not TLR4, are required for Helicobacter
pylori-induced NF-kappa B activation and chemokine expression by
epithelial cells. J Biol Chem 278, 32552–32560.
Stokes, D. C., Shenep, J. L., Fishman, M., Hildner, W. K., Bysani,
G. K. & Rufus, K. (1989). Polymyxin B prevents lipopolysaccharide-
induced release of tumor necrosis factor-alpha from alveolar
macrophages. J Infect Dis 160, 52–57.
Vabulas, R. M., Ahmad-Nejad, P., da Costa, C., Miethke, T.,
Kirschning, C. J., Hacker, H. & Wagner, H. (2001). Endocytosed
HSP60s use toll-like receptor 2 (TLR2) and TLR4 to activate the toll/
interleukin-1 receptor signaling pathway in innate immune cells.
J Biol Chem 276, 31332–31339.
Yamaguchi, H., Osaki, T., Kurihara, N., Taguchi, H., Hanawa, T.,
Yamamoto, T. & Kamiya, S. (1997). Heat-shock protein 60
homologue of Helicobacter pylori is associated with adhesion of
H. pylori to human gastric epithelial cells. J Med Microbiol 46,
Yamaguchi, H., Osaki, T., Kurihara, N. Kitajima M., Kai, M.,
Takahashi, M., Taguchi, H. & Kamiya, S. (1999). Induction of
secretion of interleukin-8 from human gastric epithelial cells by heat-
shock protein 60 homologue of Helicobacter pylori. J Med Microbiol
Yamaoka, Y., Kikuchi, S., el-Zimaity, H. M., Gutierrez, O., Osato,
M. S. & Graham, D. Y. (2002). Importance of Helicobacter pylori oipA
in clinical presentation, gastric inflammation, and mucosal inter-
leukin 8 production. Gastroenterology 123, 414–424.
Young, D. B. (1990). Chaperonins and the immune response. Semin
Cell Biol 1, 27–35.
Yunoki, N., Yokota, K., Mizuno, M. & 7 other authors (2000).
Antibody to heat shock protein can be used for early serological
monitoring of Helicobacter pylori eradication treatment. Clin Diagn
Lab Immunol 7, 574–577.
Ziegler-Heitbrock, H. W., Sternsdorf, T., Liese, J., Belohradsky, B.,
Weber, C., Wedel, A., Schreck, R., Bauerle, P. & Strobel, M. (1993).
Pyrrolidine dithiocarbamate inhibits NF-kappa B mobilization and
TNF production in human monocytes. J Immunol 151, 6986–6993.
R. Takenaka and others