The diversity of vacA and cagA genes of Helicobacter pyloriin East Asia

Wen Zhou a;b, Shiho Yamazaki a, Akiyo Yamakawa a, Masahiro Ohtani a,Yoshiyuki Ito a, Yoshihide Keida c, Hideaki Higashi d, Masanori Hatakeyama d,

Jianmin Si b, Takeshi Azuma a;�

a Second Department of Internal Medicine, Fukui Medical University, Matsuoka-cho, Yoshida-gun, Fukui 910-1193, Japanb Sir Run Run Shaw Hospital Clinical Research Institute, School of Medicine, Zhejiang University, Hangzhou, PR China

c Division of Internal Medicine, Okinawa Chubu Hospital, Okinawa, Japand Division of Molecular Oncology, Institute for Genetic Medicine and Graduate School of Science, Hokkaido University, Sapporo, Japan

Received 31 March 2003; received in revised form 4 July 2003; accepted 29 September 2003

First published online 1 November 2003

Abstract

It has been reported that Helicobacter pylori infection with the type I strain, which expresses the VacA and CagA antigens, is associatedwith duodenal ulcer. We examined the diversity of vacA and cagA genes in 143 isolates obtained from patients with duodenal ulcer orchronic gastritis in East Asia (two different areas of Japan, Fukui and Okinawa, and also in Hangzhou, China) by polymerase chainreaction (PCR) and sequence analysis. Diversities of cagA and vacA genes were detected in East Asia. The prevalence of cagA-positiveH. pylori was significantly different between Fukui and Okinawa (P=0.0032). The prevalence of Western type CagA was significantlyhigher in Okinawa than in Fukui (P6 0.0001). However, there was no significant association between the genotype of cagA and clinicaloutcome. In Japan, the predominant vacA genotype was s1c/m1b. In contrast, in Hangzhou, the predominant vacA genotype was s1c/m2,and they were all East Asian CagA-positive. These findings suggest that a distinct distribution of the vacA and cagA genotypes is presentin East Asia, regardless of clinical outcome.< 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: vacA ; cagA ; Duodenal ulcer ; Chronic gastritis ; Helicobacter pylori

1. Introduction

Helicobacter pylori infects about 50% of the world’spopulation and is thus a major cause of gastritis, pepticulcer, and an important risk factor for gastric cancer [1^4].Several putative H. pylori virulence factors have been iden-ti¢ed, including urease, vacuolating cytotoxin (VacA), andcytotoxin-associated gene A antigen (CagA) [5^9]. Mostclinical isolates of H. pylori from patients su¡ering frompeptic ulcer or malignant disease express these virulence-associated factors [10,11]. Consequently, H. pylori strainshave been divided into two broad families, type I and typeII, which are based on whether or not they express theVacA and CagA antigens. Type I strains can produceVacA and CagA, while type II strains lack this ability

[12]. VacA induces the formation of intracellular vacuolesin eukaryotic cells in vitro. The vacA gene is present in allH. pylori strains and contains at least two variable parts.The s-region (encoding the signal peptide) coexists as s1 ors2 allelic types. Among type s1 strains, subtypes s1a, s1b,and s1c have been identi¢ed. The m-region (middle) occursas the m1 or m2 allelic type [10^13]. Among type m1,three subtypes have been identi¢ed, designated m1a,m1b, and m1c [14,15]. Production of the vacuolating cy-totoxin is related to the mosaic structure of vacA. In gen-eral, type s1/m1 and s1/m2 strains produce high and mod-erate levels of toxin, respectively, whereas s2/m2 strainsproduce little or no toxin [11]. The CagA antigen is ahighly immunogenic protein of unknown function. ThecagA gene is part of a 40-kb DNA insertion that wasconsidered to have typical features of a bacterial pathoge-nicity island (PAI), and may have originated from a non-Helicobacter source [16,17]. The cag PAI contains about30 genes, and six of the cag genes are thought to encode a

0928-8244 / 03 / $22.00 < 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.doi :10.1016/S0928-8244(03)00299-2

* Corresponding author. Fax: +81 (776) 61-8110.E-mail address: azuma@fmsrsa.fukui-med.ac.jp (T. Azuma).

FEMSIM 1639 15-12-03

FEMS Immunology and Medical Microbiology 40 (2004) 81^87

www.fems-microbiology.org

putative type IV secretion system specializing in the trans-fer of a variety of multimolecular complexes across thebacterial membrane to the extracellular space or into othercells [17,18]. Recent studies have provided a molecularbasis for the pathological actions of CagA on gastric ep-ithelial cells. After attachment of cagA-positive H. pylorito gastric epithelial cells, CagA is directly injected fromthe bacteria into the cells via the bacterial type IV secre-tion system and undergoes tyrosine phosphorylation in thehost cells [19^22]. Furthermore, we recently discoveredthat translocated CagA forms a physical complex withSRC homology 2 domain (SH2)-containing tyrosine phos-phatase SHP-2 in a phosphorylation-dependent manner,and stimulates phosphatase activity [23]. SHP-2 is knownto play an important positive role in mitogenic signaltransduction [24]. Deregulation of SHP-2 by CagA mayinduce abnormal proliferation of gastric epithelial cells.The molecular mass of CagA protein varies from 128 to140 kDa. Variation in size of the protein has been corre-lated with the presence of a variable number of repeatsequences located in the 3P region of the gene [25,26].The phosphorylation sites are located in the repeat regionof CagA. Recently, we also discovered that the predomi-nant CagA proteins isolated in East Asia have a distinctsequence at the region corresponding to the repeat se-quence of Western CagA. Upon tyrosine phosphorylation,this East Asian-speci¢c sequence confers stronger SHP-2binding and transforming activities to Western CagA [27].The potential of CagA to disturb host cell functions as avirulence factor could be determined by the degree ofSHP-2 binding activity. The diversity of the CagA phos-phorylation site, which collectively determines binding af-¢nity of CagA to SHP-2, may be an important variable indetermining the clinical outcome of infection by di¡erentH. pylori strains.In this study, we examined the diversity of vacA and

cagA genes in isolates obtained from patients with duo-denal ulcer or chronic gastritis in two di¡erent areas ofJapan, Fukui and Okinawa, and also in South EastChina (Hangzhou, Zhejiang Province) to investigate therelationship between genotype of H. pylori and clinicaloutcome.

2. Materials and methods

2.1. Patients

One hundred and forty three H. pylori clinical isolates(65 Fukui strains, 60 Okinawa strains, and 18 Hangzhoustrains) were obtained during upper gastroduodenal en-doscopy at the Second Department of Internal Medicine,Fukui Medical University, Okinawa Chubu Hospital, andSir Run Run Shaw Hospital, Zhejiang University, respec-tively. The 65 patients in Fukui consisted of 26 with duo-denal ulcer (17 male and nine female, mean age 48.8 years)

and 39 with chronic gastritis (19 male and 20 female, meanage 57.6 years). The 60 Okinawa patients consisted of 21with duodenal ulcer (18 male and three female, mean age55.2 years) and 39 with chronic gastritis (15 male and 24female, mean age 59.5 years). The 18 Hangzhou patientsconsisted of 11 with duodenal ulcer (four male and sevenfemale, mean age 42.9 years) and seven with chronic gas-tritis (three male and four female, mean age 36.1 years).Patients who had received non-steroidal anti-in£ammatorydrugs were excluded from this study, and none of thepatients had recently been prescribed antibiotics.

2.2. Isolation and culture of H. pylori

Gastric biopsy specimens from each patient were inoc-ulated onto a trypticase soy agar (TSA)-II/5% sheep bloodplate and cultured under microaerobic conditions (O2, 5%;CO2, 15%; N2, 80%) at 37‡C for 5 days. A single colonywas picked from each primary culture plate, inoculatedonto a fresh TSA-II plate, and cultured under the condi-tions described above. A few colonies were picked fromeach plate and transferred into 20 ml of brucella brothliquid culture medium containing 10% fetal calf serum,and cultured for 3 days under the same conditions asdescribed above. Some of the liquid culture samples werestored at 380‡C in 0.01 M phosphate-bu¡ered saline(PBS) containing 20% glycerol. DNA from each H. pyloriisolate was extracted from the pellet of the liquid culturesample by the protease/phenol-chloroform method, sus-pended in 300 Wl of a TE bu¡er (10 mM Tris^HCl,1 mM ethylenediamine tetraacetic acid (EDTA)) andstored at 4‡C until polymerase chain reaction (PCR) am-pli¢cation.

2.3. Nucleotide sequences of the 3P region of the cagA gene

The 3P region of cagA was ampli¢ed by PCR using thefollowing primers [28]. Forward primer: 5P-GAATTG-TCTGATAAACTTGAAA, reverse primer: 5P-GCGTA-TGTGGCTGTTAGTAGCG. Ampli¢cation conditionswere as follows: heating at 94‡C for 5 min, followed by25 cycles consisting of 94‡C for 30 s, 55‡C for 30 s, and72‡C for 30 s. The tubes were kept at 72‡C for 7 minbefore storage at 4‡C. PCR products were examined by2% agarose gel electrophoresis. The gels were stained withethidium bromide to detect the size of the 3P fragment ofcagA gene. PCR products were then puri¢ed with Centri-con-100 concentrator columns (Amicon, Beverly, MA,USA). DNA sequencing was performed by the dideoxynu-cleotide chain termination method with a BigDye termi-nator cycle sequencing ready reaction mix (Applied Bio-systems, Tokyo, Japan) in an ABI Prism 310 geneticanalyzer (Applied Biosystems, Tokyo, Japan) using thesame primers shown above and the following additionaltwo primers, 5P-GCCCTACCTTACTGAGATCAT and5P-AGCTTCTGATACCGCTTGACTG. According to the

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^8782

manufacturer’s protocol, reagent mixtures containing 5 Wlof puri¢ed PCR product, 3.2 pmol of primer, 8 Wl ofterminator cycle sequencing ready reaction mix, and 5 Wlof sterilized distilled water were prepared. Reaction tubeswere placed in the thermal cycler, and thermal cycling wasstarted under the following conditions: 96‡C for 10 s,50‡C for 5 s, and 60‡C for 4 min, repeated for 25 cycles.Cycle sequencing reactions were performed for both DNAstrands. Nucleotide sequences were aligned and analyzedwith GENTYX-Mac, version 10.0 (Software DevelopmentCo., Tokyo, Japan).

2.4. Detection of vacA gene diversity

We previously reported full-length vacA gene sequencesof 37 Fukui strains [29]. Fourteen Fukui strains isolatedfrom duodenal ulcer patients (GenBank accessionnumbers AF049620, AF049621, AF049623, AF049625,AF049627, AF049629, AF049630, AF049631, AF049637,AF049639, AF049641, AF049645, AF049647, andAF049652) and 17 Fukui strains isolated from chronicgastritis patients (GenBank accession numbers AF049619,AF049622, AF049624, AF049626, AF049628, AF049632,AF049633, AF049635, AF049636, AF049642, AF071095,AF049643, AF049644, AF049646, AF049648, AF049649,and AF071096) were included in this study. In this study,genotyping of vacA s- and m-region alleles was performedin the remaining 112 strains as described in detail previ-ously. Brie£y, parts of the vacA s- and m-regions wereampli¢ed, using SS1-F, SS3-F, S1c-F and VA1-R primers

for the s-region and VA3-F, VAm-F3, VA4-F, VA3-R,VAm-R3, and VA4-R primers for the m-region (Table 1)[14,29,30]. The conditions of PCR were described above.

2.5. Statistical analysis

Di¡erences in the distribution of genotypes of the cagAand/or vacA genes and the associations between the geno-types of the cagA and/or vacA gene and clinical outcomewere analyzed with the M-squared test and Fisher’s exactprobability test. Di¡erences with a P-value of less than0.05 were considered signi¢cant.

3. Results

3.1. Diversity of cagA

All isolates from Fukui and Hangzhou were cagA-pos-itive strains. In contrast, ¢ve of 60 (8.3%) isolates fromOkinawa were cagA-negative strains. All cagA-negativestrains were isolated from patients with chronic gastritis.The prevalence of cagA-positive H. pylori was signi¢cantlydi¡erent between Fukui and Okinawa (P=0.0032) (Ta-ble 2).We previously demonstrated that the Western type

CagA possesses WSS (Western CagA-speci¢c, SHP-2binding sequence), while the East Asian type CagA doesnot have the WSS, but instead has a distinct sequencedesignated ESS (East Asian CagA-speci¢c, SHP-2 bindingsequence) in the corresponding region (Fig. 1). All isolatesfrom Fukui and 94.4% (17/18) of the isolates from Hang-zhou had East Asian type CagA. In contrast, 21.7% (13/60) of the strains had Western type CagA in Okinawa. Theprevalence of Western type CagA was signi¢cantly higherin Okinawa than in Fukui (M2 = 15.7, P6 0.0001) (Table2). Almost all East Asian type CagA (122/124, 98.4%) hada single ESS region. Two East Asian type CagA-positivestrains had two ESS regions. Predominant Western typeCagA (11/14, 78.6%) had a single WSS region. One West-ern type CagA-positive strain had two WSS regions, andtwo Western type CagA-positive strains had three WSSregions.We examined the relationship between the genotype of

cagA and clinical outcome. There was no signi¢cant asso-ciation between the genotype of cagA and clinical outcome

Table 1Primer sets used for vacA typing

Region Primer code Nucleotide sequence

s1a SS1-F 5P-GTCAGCATCACACCGCAAC-3PVA1-RP 5P-CTGCTTGAATGCCGCCCAAC-3P

s1b SS3-F 5P-AGCGCCATACCGCAAGAG-3PVA1-R 5P-CTGCTTGAATGCCGCCCAAC-3P

s1c S1c-F 5P-CTCTCGCTTTAGTGGGGYT-3PVA1-R 5P-CTGCTTGAATGCCGCCCAAC-3P

m1a VA3-F 5P-GGTCAAAATGCGGTCATGG-3PVA3-R 5P-CCATTGGTACCTGTAGAAAC-3P

m1b VAm-F3 5P-GGCCCCAATGCAGTCATGGAT-3PVAm-R3 5P-GCTGTTAGTGCCTAAAGAAGCAT-3P

m2 VA4-F 5P-GGAGCCCCAGGAAACATTG-3PVA4-R 5P-CATAACTAGCGCCTTGCAC-3P

Table 2Distribution of the cagA allele types

Fukui Okinawa Hangzhou

cagA (3) East Asian Western cagA (3) East Asian Western cagA (3) East Asian Western

Duodenal ulcer 0 26 0 0 14 7 0 10 1Chronic gastritis 0 39 0 5 28 6 0 7 0Total 0 65 0 5a 42 13b 0 17 1

aThe prevalence of cagA-negative H. pylori was signi¢cantly higher in Okinawa than in Fukui (P=0.0032).bThe prevalence of Western CagA-positive strain was signi¢cantly higher in Okinawa than in Fukui (P6 0.0001).

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^87 83

(chronic gastritis and duodenal ulcer) in Fukui, Okinawa,and Hangzhou.

3.2. Diversity of vacA

In Fukui, all strains contained s1 in the signal sequenceregion and m1 in the middle region (Fig. 2). The predom-inant vacA genotype was s1c/m1b (55/65, 84.6%). In Oki-nawa, although the predominant vacA genotype was s1c/m1b (38/60, 63.3%), four strains (6.7%) were s1a/m1b, onestrain (1.7%) was s1b/m1b, seven strains (11.7%) were s1a/m2, ¢ve strains (8.3%) were s1c/m2, and ¢ve strains (8.3%)were s2/m2. In contrast, in Hangzhou the predominantvacA genotype was s1c/m2 (11/18, 61.1%). The prevalence

of s1c/m1b was signi¢cantly higher in Fukui than in Oki-nawa and Hangzhou (Fukui vs. Okinawa; M2 = 11.9,P=0.0006, Fukui vs. Hangzhou; M2 = 22.7, P6 0.0001).The prevalence of s1c/m2 was signi¢cantly higher in Hang-zhou than in Fukui and Okinawa (Hangzhou vs. Fukui;M2 = 45.8, P6 0.0001, Hangzhou vs. Okinawa; M2 = 23.7,P6 0.0001). There was no m1c genotype in East Asia.There was no signi¢cant association between the genotypeof vacA and clinical outcome (chronic gastritis and duo-denal ulcer) (Table 3).

3.3. Relationship between cagA and vacA alleles

The predominant East Asian CagA-positive strains had

Fig. 1. The alignment of the deduced amino acid sequence in the 3P region of the cagA gene among strains F36 (GenBank accession numberAB090080), F75 (AB090106), OK112 (AB090088), OK111 (AB090140), and OK107 (AB090086). Identical amino acids are indicated by hyphens. Dotsindicate gap. The sequences start at position 918 of a reference CagA amino acid (strain NCTC11637, AE202973). ESS region is indicated by underline,and WSS region is indicated by double underline. F36 and F75 had one and two ESS regions, respectively. OK112, OK111 and OK107 had one, two,and three WSS regions, respectively.

Table 3Distribution of the vacA allele types

s1a/m1a s1a/m1b s1b/m1a s1b/m1b s1c/m1a s1c/m1b s1a/m2 s1b/m2 s1c/m2 s2/m2

Fukui duodenal ulcer 0 4 1 0 0 21 0 0 0 0chronic gastritis 1 4 0 0 0 34 0 0 0 0total 1 8 1 0 0 55a 0 0 0 0

Okinawa duodenal ulcer 0 1 0 0 0 13 4 0 3 0chronic gastritis 0 3 0 1 0 25 3 0 2 5total 0 4 0 1 0 38 7 0 5 5

Hangzhou duodenal ulcer 0 0 0 0 0 2 0 8 0chronic gastritis 0 1 0 0 0 3 0 0 3 0total 0 1 0 0 0 5 1 0 11b 0

aThe prevalence of s1c/m1b was signi¢cantly higher in Fukui than in Okinawa and Hangzhou (Fukui vs. Okinawa, P=0.0006; Fukui vs. Hangzhou,P6 0.0001).bThe prevalence of s1c/m2 was signi¢cantly higher in Hangzhou than in Fukui and Okinawa (Hangzhou vs. Fukui, P6 0.0001; Hangzhou vs, Okinawa,P6 0.0001).

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^8784

Fig. 2. Nucleotide sequence alignment of vacA gene. A: vacA s-region sequences showing di¡erences between vacA s1a, s1b, and s1c alleles. Each 76-bpsequence presented here starts at position 47 of a reference vacA open reading frame (strain NCTC11638, GenBank accession number U07145). Identi-cal nucleotides are indicated by hyphens. B: vacA m-region sequences showing di¡erences between vacA m1a and m1b alleles. Each 291-bp sequencepresented here stars at position 2914 of NCTC11638 vacA open reading frame. GenBank accession numbers for the sequences depicted here areAF071095 (F37), AF049630 (F51), AF049633 (F56), AF049652 (F73), AE001511 (J99), AF049619 (F20), AF049620 (F21), AF049621 (F28), AF049622(F29), and AF035610 (China R13).

Table 4The relationship between cagA and vacA allele types

cagA status vacA genotype

s1a/m1a s1a/m1b s1b/m1a s1b/m1b s1c/m1a s1c/m1b s1a/m2 s1b/m2 s1c/m2 s2/m2

Fukui East Asian type 1 8 1 0 0 55 0 0 0 0Okinawa cagA-negative 0 0 0 0 0 2 0 0 0 3

Western type 0 0 0 1 0 2 5a 0 3a 2a

East Asian type 0 4 0 0 0 34b 2 0 2 0Hangzhou Western type 0 0 0 0 0 0 1 0 0 0

East Asian type 0 1 0 0 0 5 0 0 11 0

aWestern type CagA was signi¢cantly associated with the vacA m2 genotype (P6 0.0001).bEast Asian type CagA was signi¢cantly associated with the vacA s1c/m1b genotype (P6 0.0001).

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^87 85

vacA s1c/m1b genotype (55/65, 84.6% in Fukui and 34/42,81.0% in Okinawa) in Japan. In contrast, in Hangzhou, 11of 17 (64.7%) East Asian CagA-positive strains had thevacA s1c/m2 genotype. The predominant Western typeCagA-positive strains had the vacA m2 genotype (10/13,76.9%) (Table 4).

4. Discussion

H. pylori is one of the most genetically diverse bacterialspecies, with any given isolate easily distinguished frommost others by DNA ¢ngerprinting or the sequencing ofrepresentative gene segments. This mutational diversityhas been enhanced by extensive interstrain gene transferand recombination [18]. Strain-speci¢c genetic diversityhas been postulated to be involved in the organism’s abil-ity to cause di¡erent diseases. There are also indications ofsigni¢cant geographic di¡erences among strains. Only one-half to two-thirds of Western isolates carry cagA. In con-trast, nearly all East Asian strains carry cagA [29,31]. Ithas also been reported that large sequence di¡erences dis-tinguish the cagA gene fragments from Asian strains andother strains [27,32]. The present data showed the diversityof cagA even in East Asia: Japan (Fukui and Okinawa)and China (Hangzhou). All isolates from Fukui andHangzhou were cagA-positive strains. In contrast, 8.3%of isolates from Okinawa were cagA-negative strains.The prevalence of cagA-positive H. pylori was signi¢cantlydi¡erent between Fukui and Okinawa. In addition, allisolates from Fukui and 94.4% of the isolates from Hang-zhou had East Asian type CagA. In contrast, 21.7% of thestrains had Western type CagA in Okinawa. The preva-lence of Western type CagA was signi¢cantly higher inOkinawa than in Fukui. Fukui is a typical rural prefecturelocated in the central Japanese mainland (Honshu), whileOkinawa consists of islands in the southwestern part ofJapan. The two areas are separated by more than 1300km. Okinawa has had active international communicationwith the West historically and had a relative large Amer-ican presence in the last half century. The Western typeH. pylori may therefore have been introduced into Okina-wa from the West. In this study, all cagA-negative strainswere isolated from patients with chronic gastritis, how-ever, there was no signi¢cant association between the ge-notype of CagA and clinical outcome (chronic gastritisand duodenal ulcer).In the present study, we also examined the diversity of

the vacA gene and the relationship between vacA andcagA. In Fukui, all strains contained s1 in the signal se-quence region and m1 in the middle region. The predom-inant vacA genotype in Japan was s1c/m1b (55/65, 84.6%in Fukui, and 38/60, 63.3% in Okinawa). Van Doorn et al.also reported that subtype s1c was prevalent in East Asia,but appears to be rare in other parts of the world [13]. Wealso demonstrated in this study that the predominant East

Asian type CagA-positive strains had vacA s1c/m1b geno-type in Japan. In contrast, in Hangzhou the predominantvacA genotype was s1c/m2 (11/18, 61.1%). Initially, Pan etal. reported that more than 80% of Chinese isolates(Shanghai and Guangzhou, Guangzhou is s 1000 kmfrom Shanghai) carried s1a/m2 [14]. In their study, theydid not discriminate between subtypes s1a and s1c. Hang-zhou is close to Shanghai; thus, the predominant vacAgenotype in China seems to be s1c/m2. All s1c/m2 strainswere East Asian CagA-positive strains in this study. Ourresults indicate a distinct distribution of the vacA geno-type, even in East Asian CagA-positive strains. It has beenreported that carriage of certain genotypes (cagA-positive,vacA type s1) is signi¢cantly associated with more severeclinical outcome, such as duodenal ulcer and gastric can-cer, in the West [31]. However, there was no signi¢cantassociation between the genotype of vacA and duodenalulcer in this study, although all vacA s2/m2 strains wereisolated from patients with chronic gastritis.In East Asia, the prevalence of gastric cancer is much

higher than in Western countries. However, the gastriccancer risk is di¡erent even in East Asia. The death ratefrom gastric cancer is 43.7/100 000 in Fukui, 18.2/100 000in Okinawa, and 23.1/100 000 in Hangzhou, respectively.Although there was no signi¢cant association between thegenotype of cagA and/or vacA and clinical outcome(chronic gastritis and duodenal ulcer) in this study, furtherprecise analysis of gastric mucosal atrophy as a gastriccancer risk is required.

Acknowledgements

This work was supported by Special Coordination Fundfor Promoting Science and Technology, Ministry of Edu-cation, Culture, Sports, Science and Technology, Japan,and a grant-in-aid for Scienti¢c Research (B) and (C)from Japan Society for the Promotion of Science.

References

[1] Warren, J.R. and Marshall, B.J. (1983) Unidenti¢ed curved bacilli ongastric epithelium in active chronic gastritis. Lancet 1, 1273^1275.

[2] NIH Consensus Conference (1994) Helicobacter pylori in peptic ulcerdisease. NIH consensus development panel on Helicobacter pylori inpeptic ulcer disease. J. Am. Med. Assoc. 272, 65^69.

[3] Marshall, B.J., Goodwin, C.S., Warren, J.R., Murray, R., Blincow,E.D., Blackbourn, S.J., Phillips, M., Waters, T.E. and Sanderson,C.R. (1988) Prospective double-blind trial of duodenal ulcer relapseafter eradication of Campylobacter pylori. Lancet 2, 1437^1442.

[4] Parsonnet, J., Friedman, G.D., Vandersteen, D.P., Chang, Y., Vogel-man, J.H., Orentreich, N. and Sibley, R.K. (1991) Helicobacter pyloriinfection and the risk of gastric carcinoma. N. Engl. J. Med. 325,1127^1131.

[5] Covacci, A., Censini, S., Bugnoli, M., Petracca, R., Burroni, D.,Macchia, G., Massone, A., Papini, E., Xiang, Z., Figura, N. andRappuoli, R. (1993) Molecular characterization of the 128-kDa im-

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^8786

munodominant antigen of Helicobacter pylori associated with cyto-toxicity and duodenal ulcer. Proc. Natl. Acad. Sci. USA 90, 5791^5795.

[6] Cover, T.L. and Blaser, M.J. (1992) Puri¢cation and characterizationof the vacuolating toxin from Helicobacter pylori. J. Biol. Chem. 267,10570^10575.

[7] Cover, T.L. (1996) The vacuolating cytotoxin of Helicobacter pylori.Mol. Microbiol. 20, 241^246.

[8] Tee, W., Lambert, J.R. and Dwyer, B. (1995) Cytotoxin productionby Helicobacter pylori from patients with upper gastrointestinal tractdiseases. J. Clin. Microbiol. 33, 1203^1205.

[9] Hirai, M., Azuma, T., Ito, S., Kato, T., Kohli, Y. and Fujiki, N.(1994) High prevalence of neutralizing activity to Helicobacter pyloricytotoxin in serum of gastric carcinoma patients. Int. J. Cancer 56,56^60.

[10] van Doorn, L.J., Figueiredo, C., Sanna, R., Plaisier, A., Schneeberg-er, P., de Boer, W. and Quint, W. (1998) Clinical relevance of cagA,vacA, and iceA status of Helicobacter pylori. Gastroenterology 115,58^66.

[11] Atherton, J.C., Cao, P., Peek, R.M., Tummuru, M.K., Blaser, M.J.and Cover, T.L. (1995) Mosaicism in vacuolating cytotoxin alleles ofHelicobacter pylori : Association of speci¢c vacA types with cytotoxinproduction and peptic ulceration. J. Biol. Chem. 270, 17771^17777.

[12] Xiang, Z., Censini, S., Bsyeli, P.F., Telford, J.L., Figura, N., Rap-puoli, R. and Covacci, A. (1995) Analysis of expression of cagA andvacA virulence factors in 43 strains of Helicobacter pylori reveals thatclinical isolates can be divided into two major types and that cagA isnot necessary for expression of the vacuolating cytotoxin. Infect.Immun. 63, 94^98.

[13] van Doorn, L.J., Figueiredo, C., Sanna, R., Pena, S., Midolo, P., Ng,E.K., Atherton, J.C., Blaser, M.J. and Quint, W.G. (1998) Expandingallelic diversity of Helicobacter pylori vacA. J. Clin. Microbiol. 36,2597^2603.

[14] Pan, Z., Berg, D.E., van der Hulst, R.W.M., Su, W., Raudonikiene,A., Xiao, S., Dankert, J., Tytgat, G.N.J. and van der Ende, A. (1998)Prevalence of vacuolating cytotoxin production and distribution ofdistinct vacA alleles in Helicobacter pylori from China. J. Infect. Dis.178, 220^226.

[15] Mukhopadhyay, A.K., Kersulyte, D., Jeong, J., Datta, S., Ito, Y.,Chowdhury, A., Chowdhury, S., Santra, A., Bhattacharya, S.K.,Azuma, T., Nair, B. and Berg, D.E. (2000) Distinctiveness of geno-types of Helicobacter pylori in Calcutta, India. J. Bacteriol. 182,3219^3227.

[16] Covacci, A., Censini, S., Bugnoli, M., Petracca, R., Burroni, D.,Macchia, G., Massone, A., Papini, E., Xiang, Z., Figura, N. andRappuoli, R. (1993) Molecular characterization of the 128-kDa im-munodominant antigen of Helicobacter pylori associated with cyto-toxicity and duodenal ulcer. Proc. Natl. Acad. Sci. USA 90, 5791^5795.

[17] Censini, S., Lange, C., Xiang, Z., Crabtree, J.E., Ghiara, P., Boro-dovsky, M., Rappuoli, R. and Covacci, A. (1996) Cag, a pathoge-nicity island of Helicobacter pylori, encodes type I-speci¢c and dis-ease-associated virulence factors. Proc. Natl. Acad. Sci. USA 93,14648^14653.

[18] Covacci, A., Telford, J.L., Del Giudice, G., Parsonnet, J. and Rap-puoli, R. (1999) Helicobacter pylori virulence and genetic geography.Science 284, 1328^1333.

[19] Asahi, M., Azuma, T., Ito, S., Suto, H., Nagai, Y., Tsubokawa, M.,

Tohyama, Y., Maeda, S., Omata, M., Suzuki, T. and Sasakawa, C.(2000) Helicobacter pylori CagA protein can be tyrosine phosphory-lated in gastric epithelial cells. J. Exp. Med. 191, 593^602.

[20] Stein, M., Rappuoli, R. and Covacci, A. (2000) Tyrosine phosphor-ylation of the Helicobacter pylori CagA antigen after cag-driven hostcell translocation. Proc. Natl. Acad. Sci. USA 97, 1263^1268.

[21] Segal, E.D., Cha, J., Lo, J., Falkow, S. and Tompkins, L.S. (1999)Altered states: involvement of phosphorylated CagA in the inductionof host cellular growth changes by Helicobacter pylori. Proc. Natl.Acad. Sci. USA 96, 14559^14564.

[22] Odenbreit, S., Puls, J., Sedlmaier, B., Gerland, E., Fischer, W. andHaas, R. (2000) Translocation of Helicobacter pylori CagA into gas-tric epithelial cells by type IV secretion. Science 287, 1497^1500.

[23] Higashi, H., Tsutsumi, R., Muto, S., Sugiyama, T., Azuma, T., Asa-ka, M. and Hatakeyama, M. (2002) SHP-2 tyrosine phosphatase asan intracellular target of Helicobacter pylori CagA protein. Science295, 683^686.

[24] Feng, G., Hui, C. and Pawson, T. (1993) SH2-containing phospho-tyrosine phosphatase as a target of protein-tyrosine kinases. Science259, 1607^1611.

[25] Yamaoka, Y., Kodama, T., Kashima, K., Graham, D.Y. and Sepul-veda, A.R. (1998) Variants of the 3P region of the cagA gene inHelicobacter pylori isolates from patients with di¡erent H. pylori-as-sociated diseases. J. Clin. Microbiol. 36, 2258^2263.

[26] Yamaoka, Y., El-Zimaity, H.M., Gutierrez, O., Figura, N., Kim,J.G., Kodama, T., Kashima, K., Graham, D.Y. and Kim, J.K.(1999) Relationship between the cagA 3P repeat region of Helico-bacter pylori, gastric histology, and susceptibility to low pH. Gastro-enterology 117, 342^349.

[27] Higashi, H., Tsutsumi, R., Fujita, A., Yamazaki, S., Asaka, M.,Azuma, T. and Hatakeyama, M. (2002) Biological activity of theHelicobacter pylori virulence factor CagA is determined by variationin the tyrosine phosphorylation sites. Proc. Natl. Acad. Sci. USA 99,14428^14433.

[28] Azuma, T., Yamakawa, A., Yamazaki, S., Fukuta, K., Ohtani, M.,Ito, Y., Dojo, M., Yamazaki, Y. and Kuriyama, M. (2002) Correla-tion between variation of the 3P region of the cagA gene in Helico-bacter pylori and disease outcome in Japan. J. Infect. Dis. 186, 1621^1630.

[29] Ito, Y., Azuma, T., Ito, S., Miyaji, H., Hirai, M., Yamazaki, Y.,Sato, F., Kato, T., Kohli, Y. and Kuriyama, M. (1997) Analysisand typing of the vacA gene from cagA-positive strains of Helico-bacter pylori isolated in Japan. J. Clin. Microbiol. 35, 1710^1714.

[30] Ito, Y., Azuma, T., Ito, S., Suto, H., Miyaji, H., Yamazaki, Y.,Kohli, Y. and Kuriyama, M. (1998) Full-length sequence analysisof the vacA gene from cytotoxic and noncytotoxic Helicobacter pylo-ri. J. Infect. Dis. 178, 1391^1398.

[31] Van Doorn, L.J., Figueiredo, C., Megraud, F., Pena, S., Midolo, P.,Queiroz, D.M., Carneiro, F., Vanderborght, B., Pegado, M.D., San-na, R., De Boer, W., Schneeberger, P.M., Correa, P., Ng, E.K.,Atherton, J., Blaser, M.J. and Quint, W.G. (1999) Geographic dis-tribution of vacA allelic types of Helicobacter pylori. Gastroenterol-ogy 116, 823^830.

[32] Achtman, M., Azuma, T., Berg, D.E., Ito, Y., Morelli, G., Pan, Z.,Suerbaum, S., Thompson, S.A., van der Ende, A. and van Doorn, L.(1999) Recombination and clonal groupings within Helicobacterpylori from di¡erent geographical regions. Mol. Microbiol. 32, 459^470.

FEMSIM 1639 15-12-03

W. Zhou et al. / FEMS Immunology and Medical Microbiology 40 (2004) 81^87 87