Free Radical Biology & Medicine, Vol. 36, No. 9, pp. 1126 –1133, 2004Copyright D 2004 Elsevier Inc.

Printed in the USA. All rights reserved0891-5849/$-see front matter

doi:10.1016/j.freeradbiomed.2004.02.001

Original Contribution

OXYGEN TENSION REGULATES REACTIVE OXYGEN GENERATION

AND MUTATION OF Helicobacter pylori

AH-MEE PARK,* QUAN LI,* KUMIKO NAGATA,y TOSHIHIDE TAMURA,

y KUNIO SHIMONO,z

EISUKE F. SATO,* and MASAYASU INOUE*,b

*Department of Biochemistry and Molecular Pathology, Osaka City University Medical School, 1-4-3 Asahimachi,Abeno, Osaka 545-8585, Japan; yDepartment of Bacteriology, Hyogo College of Medicine, 1-1 Mukogawa,

Nishinomiya, Hyogo 663-8501, Japan; and zHinode-En Elder Health Center, Ohishi-Minamimachi,Nada, Kobe, Hyogo 657-0852, Japan

(Received 3 October 2003; Revised 21 January 2004; Accepted 2 February 2004)

Ad

Bioche

School

6645-3

Abstract—Although both bacillary and coccoid forms of Helicobacter pylori reside in human stomach, the

pathophysiological significance of the two forms remains obscure. The present work describes the effect of oxygen

tension on the transformation and reactive oxygen species (ROS) metabolism of this pathogen. Most H. pylori cultured

under an optimum O2 concentration (7%) were the bacillary form, whereas about 80% of cells cultured under aerobic or

anaerobic conditions were the coccoid form. The colony-forming unit of H. pylori decreased significantly under both

aerobic and anaerobic culture conditions. The bacillary form of H. pylori generated predominantly superoxide radical,

whereas the coccoid form generated preferentially hydroxyl radical. Specific activities of cellular respiration, urease, and

superoxide dismatase decreased markedly after transformation of the bacillary form to the coccoid form, with

concomitant generation of protein carbonyls and 8-hydroxyguanine. The frequency of mutation of cells increased

significantly during culture under nonoptimum O2 conditions. These results indicate that ROS generated by H. pylori

catalyze the oxidative modification of cellular DNA, thereby enhancing the transformation from the bacillary to the

coccoid form. The enhanced generation of mutagenic hydroxyl radicals in the coccoid form might accelerate mutation

and increase the genetic diversity of H. pylori. D 2004 Elsevier Inc. All rights reserved.

Keywords—Helicobacter pylori, Oxidative stress, Superoxide, Mutation, Gastric cancer, Free radicals

INTRODUCTION

Helicobacter pylori is a gram-negative and microaerobic

bacterium that plays important roles in the pathogenesis

of gastritis, peptic ulcer, and gastric cancer [1,2]. Two

types of H. pylori, bacillary and coccoid forms, are seen

in human gastric mucosa; the bacillary form is the

predominant form found in human stomach, and the

coccoid form is the major form found in extragastric

environments [3]. Both oral-to-oral and fecal-to-oral

routes have been postulated to be involved in the

dress correspondence to: Dr. Masayasu Inoue, Department of

mistry and Molecular Pathology, Osaka City University Medical

, 1-4-3 Asahimachi, Abeno, Osaka 545-8585, Japan; Fax: +81-6-

721; E-mail: inoue@med.osaka-cu.ac.jp.

1126

mechanism of infection of human subjects [4–6]. We

previously reported that the oxygen tension in gastric

juice changes significantly depending on the injected

solutions [7]. Gastric mucosal oxygen tension has been

reported to decrease significantly in animals exposed to

restraint stress [8]. When the bacillary form of H. pylori

is exposed to various stresses, such as unfavorable O2

tension and antibiotics, it undergoes transformation to the

coccoid form [9]. The coccoid form of H. pylori has been

postulated to be viable but difficult to grow under culture

conditions [10]. The mechanism and pathophysiological

significance of the transformation of H. pylori remain

unknown. It has been postulated that the transformation

of H. pylori from the bacillary to the coccoid form might

occur irreversibly, and, hence, the latter form might

represent a dormant stage in the life cycle [11,12].

However, the coccoid form of H. pylori has been known

Transformation of H. pylori 1127

to exhibit activities to synthesize DNA and proteins [13].

Moreover, both forms of H. pylori are known to adhere

to epithelial cells and activated cellular metabolism,

including tyrosine phosphorylation of specific proteins

[14]. Thus, not only the bacillary form but also the

coccoid form of H. pylori has been postulated to con-

tribute to infection [15]. We previously showed that the

bacillary form of H. pylori generates the superoxide

radical, and the coccoid form generates preferentially

the hydroxyl radical [16]. These results suggested that

the transformation of H. pylori might affect its biological

properties, including infectivity and toxicity to gastric

mucosal cells. To understand the roles of the transfor-

mation of H. pylori in the mechanism of infection and in

the pathogenesis of gastric mucosal injury, we studied the

effects of O2 tension on its activities urease, SOD, and

cytochrome c oxidase activities; generation of reactive

oxygen species, oxidative modification of proteins and

DNA, and growth.

EXPERIMENTAL PROCEDURES

Materials

The reagents used in the present experiments, such as

2-methyl-6-[ p-methoxyphenyl]-3,7-dihydroimidazol

[1,2- a]pyrazin-3-one (MCLA), 8-amino-5-chloro-7-phe-

nyl-pyridopyridazine (L012), Brucella broth and agar,

and ascorbic acid, were purchased from Tokyo Kasei

(Tokyo, Japan), Wako Company (Osaka, Japan), Becton

Dickinson (Cockeysville, MD, USA) and Sigma Chem-

ical Company. (St Louis, MO, USA). All other reagents

used were of the highest grade commercially available

and were obtained from Wako Company.

H. pylori NCTC-11637 was cultured in Brucella broth

containing 5% horse serum under optimal microaerobic

conditions (7% O2) with gentle shaking at 37jC [13]. H.

pylori was also cultured under anaerobic (<1% O2) and

aerobic (21% O2) conditions without using CO2. Anae-

roPack Campylo and AnaeroPack Keep (Mitsubishi Gas

Chemical, Tokyo, Japan) were used to obtain micro-

aerobic and anaerobic atmospheres, respectively. The

number of suspended cells was determined optically;

an optical density of 0.2 at 550 nm corresponds to

approximately 108 cells/ml. The number of living cells

was determined by a colony-forming unit (CFU). Cul-

tured cells were washed with 10 mM Hepes buffer (pH

7.4) containing 0.9% NaCl and used for experiments.

Assay of cellular respiration and enzyme activities

Respiration and cytochrome c oxidase activity of H.

pylori were polarographically monitored using a Clark-

type oxygen electrode (Rank Brothers, Ltd., Cambridge,

UK) in a closed cell containing 10 mM Hepes buffer (pH

7.4), 0.9% NaCl, and 108 to 109 cells/ml at 37jC as

described previously [17]. Pyruvate 5 mM and ascorbic

acid 5 mM with N,N,NV,NV-tetramethyl-p-phenylenedi-

amine (TMPD) 0.5 mM were used as substrates for

respiration and cytochrome c oxidase, respectively. Urease

and SOD activities were assayed essentially by the me-

thod’s of Ferraro et al. [18] and Crapo et al. [19],

respectively, after solubilization with 1% Triton X-100

containing 10 mM potassium phosphate (pH 7.4) at 4jCfor 10 min, followed by centrifugation at 15,000g for 10

min. Enzyme activities in the soluble fractions were

determined.

Analysis of reactive oxygen species

Superoxide and hydroxyl radicals generated by H.

pylori (5 � 108 cells/ml) were assayed in 10 mM Hepes

buffer (pH 7.4) containing 0.9% NaCl and 0.5% Triton

X-100 in the presence of either 1 AM MCLA [20] or 0.4

mM L012 [21] using a luminescence reader BLR-201

(Aloka Co., Tokyo, Japan). To determine the specificity

of MCLA and L012 for reactive oxygen species (ROS),

chemiluminescence intensity was analyzed in the pres-

ence and absence of either deferoxamine (50 AM) or Cu/

Zn-SOD (200 U/ml).

Analysis of protein carbonyls

Oxidatively modified proteins in H. pylori were

analyzed by an immunoblotting method as described

previously [22]. Briefly, cells were disrupted in 10 mM

sodium phosphate buffer (pH 7.4) containing 0.5 mM

phenylmethylsulfonyl fluoride and 0.5 mM deferox-

amine in the presence of either 0.5% Triton X-100 or

2% SDS at 4jC for 10 min and centrifuged at 15,000g

for 10 min. The supernatant fractions were used for the

assay of protein carbonyls. Carbonyl groups in cellular

proteins were reacted with 8 mM 2,4-dinitrophenyl-

hydrazine at 15jC for 60 min. Dinitrophenol (DNP)-

conjugated proteins were determined by SDS-polyacryl-

amide gel electrophoresis followed by Western blotting

analysis using specific antibodies to DNP (Intergen Co.,

Manhattanville, NY, USA) and an ECL kit (Amersham,

Buckinghamshire, England).

Analysis of oxidative modification and degradation of

DNA

DNA was purified from H. pylori cultured under the

three different conditions with the use of a DNA extrac-

tion WB kit (Wako Co.). The amount of 8-OHdG was

determined by the HPLC method as previously described

[16]. The integrity of DNA from the H. pylori specimens

was evaluated by electrophoresis on 0.7% agarose gels

containing SYBR Green I (FMC BioProducts, Rockland,

ME, USA).

Fig. 2. Effect of oxygen tension on respiration and cytochrome coxidase. Activities of respiration and cytochrome c oxidase in H. pyloriwere determined polarographically in 10 mM Hepes buffer (pH 7.4)containing 0.9% NaCl using a Clark-type oxygen electrode at 37jC.Open columns: H. pylori respiration with 5 mM succinate; closedcolumns: cytochrome c oxidase activity measured in the presence of 5mM ascorbate plus 0.5 mM TMPD. Data are means F SD derived fromfour separate experiments. Statistical analysis was performed usingStudent’s t test. Respiration: *p < .001 versus cells cultured under 7%O2; oxidase activity:

#p < .05 versus cells cultured under 7% O2.

A.-M. PARK et al.1128

Detection of mutation frequency

The frequency of H. pylori mutation was determined

20 h after culture under different O2 tensions as described

previously [23]. Briefly, the cells cultured under three

different O2 tensions were collected by centrifugation at

15,000g for 5 min at 25jC and suspended in 500 Al ofBrucella broth medium. H. pylori thus obtained were

subsequently cultured under microaerobic conditions on

Brucella agar plates containing 5% horse serum in the

presence and absence of either 1 Ag/ml ciprofloxacin, 10

Ag/ml rifampicin, or 5 Ag/ml metronidazole. After 3 days

of culture, the colonies on the plate were counted. The

frequency of mutation was expressed as the ratio of the

number of colonies grown in the presence of antibiotics

to that in the absence of the agents.

Statistical analysis

Results are expressed as means F SD, and Student’s

t test was used for statistical analysis. A level of p < .05

was considerd significant.

RESULTS

Effect of oxygen concentrations on the growth and

structure of H. pylori

When cultured under optimal microaerobic conditions

(7% O2) for 22 h, the CFU of H. pylori increased about

10-fold. However, when cultured under either anaerobic

(<1%) or aerobic (21%) conditions, CFU decreased

significantly (Fig. 1). Although the optical density at

550 nm of H. pylori cultured for 22 h under 7% O2

Fig. 1. Effect of oxygen tension on the growth of H. pylori. H. pyloriwas cultured for 22 h under oxygen concentrations of < 1, 7, and 21%.The number of living cells was determined by a colony-forming unit(CFU). Data are means F SD derived from four separate experiments.Statistical analysis was performed using Student’s t test. *p < .001versus cells cultured under 7% O2.

increased by about 10-fold, it remained unchanged

under both anaerobic and aerobic conditions. Although

most H. pylori cultured under 7% O2 were of the

bacillary form as observed under light microscopy, about

80% of cells cultured under either anaerobic or aerobic

conditions were found to be of the coccoid form. These

results indicate that during culture under either anaerobic

or aerobic conditions, the bacillary form of H. pylori

undergoes transformation to the coccoid form, which is

slow-growing.

Respiration and cytochrome c oxidase in H. pylori

The activity of cytochrome c oxidase, a terminal

oxidase of the respiratory chain, in H. pylori cultured

under either anaerobic or aerobic conditions was lower

than that in cells cultured under 7% O2 by about 55 and

30%, respectively (Fig. 2). Although the bacillary form

of H. pylori showed strong respiration in the presence of

pyruvate, respiratory activity of the coccoid form cul-

tured under either anaerobic or aerobic conditions was

negligible.

Enzyme activity in H. pylori

Urease is an important enzyme for H. pylori to survive

in acidic gastric juice. The specific activity of urease in

the coccoid forms of H. pylori cultured under either

anaerobic or aerobic conditions was also lower than that

of the bacillary form cultured under 7% O2 (Fig. 3). The

specific activity of SOD in the coccoid forms was also

lower than that of the bacillary form.

Fig. 3. Effect of oxygen tension on enzyme activities. Activities ofSOD (open columns) and urease (closed columns) in H. pyloricultured under oxygen concentrations of < 1, 7, and 21% weremeasured as described in the text. Data are means F SD derivedfrom four separate experiments. Statistical analysis was performedusing Student’s t test. SOD activity: *p < .001 versus cells culturedunder 7% O2; urease activity: #p < .001 versus cells cultured under7% O2.

Transformation of H. pylori 1129

Generation of reactive oxygen species by H. pylori

Both MCLA and L012 have been used as chemilu-

minescence probes highly sensitive for superoxide and

hydroxyl radicals, respectively [20,21]. The specificity of

the two probes can be analyzed by the combined use of

SOD and deferoxamine, a potent chelator of free iron.

The bacillary form of H. pylori cultured under 7% O2

Fig. 4. Effect of culture conditions on the generation of reactive oxyg(5 � 108 cells/ml) cultured under oxygen concentrations of < 1%monitored by the chemiluminescence method using 1 AM MCLA ((closed squares) or 200 U/ml Cu/Zn-SOD (closed circles) was adderesults represent one experiment typical of four similar runs.

generated the superoxide radical as revealed by MCLA-

dependent chemiluminescence (Fig. 4). The presence of

SOD, but not deferoxamine, in the assay mixture com-

pletely suppressed the intensity of MCLA chemilumi-

nescence, suggesting that the bacillary form generated

predominantly the superoxide radical. The chemilumi-

nescence intensity of the bacillary form was significantly

stronger than that of the coccoid forms cultured under

either anaerobic or aerobic conditions. In contrast, the

chemiluminescence intensity of L012 was significantly

higher for the coccoid form of H. pylori cultured under

either anaerobic or aerobic conditions than for the

bacillary form cultured under 7% O2. The presence of

either SOD or deferoxamine completely inhibited the

L012-dependent chemiluminescence. Thus, the transfor-

mation of H. pylori from the bacillary to the coccoid

form appears to be accompanied by an increased gener-

ation of hydroxyl radicals through the Fenton-like reac-

tion enhanced by redox cycling of a transition metal(s)

and superoxide radical.

Oxidative modification of H. pylori proteins

Reactive oxygen species interact with a wide vari-

ety of cellular constituents including lipids, proteins,

and DNA [24–26]. Interaction of ROS with certain

amino acid residues in proteins results in the formation

of reactive carbonyl groups by the Michael addition-

type reaction [27]. Thus, we analyzed the amount of

protein carbonyl groups in H. pylori cultured under

three different sets of conditions (Fig. 5). Western

blotting analysis revealed that the amount of carbonyl

en species. Generation of reactive oxygen species by H. pylori(open squares), 7% (circles), and 21% (open triangles) was

a) or 0.4 mM L012 (b) at 37jC. Either 50 AM deferoxamined to the assay mixtures, under 7% oxygen concentration. The

Fig. 5. Effect of oxygen tension on protein carbonyls in H. pylori. Reactive carbonyl groups in H. pylori proteins were determined bySDS–PAGE followed by staining with anti-dinitrophenol antibody. Proteins were extracted with either 0.5% Triton X-100 or 2% SDSfrom H. pylori cultured under different oxygen concentrations as in Fig. 1. Proteins were incubated with 8 mM 2,4-dinitrophenylhydrazine at 15jC for 1 h. Proteins were extracted, electrophoresed, and subjected to immunoblot analysis with the antibodies (left).Coomassie blue staining (CBB) of proteins was carried out (right).

A.-M. PARK et al.1130

proteins extracted with SDS was higher than that

extracted with Triton X-100 under all culture condi-

tions. The amount of carbonyl proteins in H. pylori

Fig. 6. Effect of oxygen tension on the generation of 8-hydroxyguanineextracted from H. pylori cultured under different oxygen tensions wereGreen I. (b) Amounts of 8-hydroxyguanine in DNA specimens werdescribed in the text. Data are means F SD derived from four indeStudent’s t test. *p < .05 versus cells cultured under 7% O2.

cultured under nonoptimum O2 tension was significantly

higher than that of microaerobically cultured cells

(microaerobic < anaerobic < aerobic).

and degradation of genomic DNA. (a) DNA specimens (0.2 Ag)subjected to 0.7% agarose gel electrophoresis containing SYBRe determined by the HPLC method. Other conditions were aspendent experiments. Statistical analysis was performed using

Transformation of H. pylori 1131

Changes in DNA after transformation of cells

We analyzed the amounts of 8-OHdG in DNA of H.

pylori cultured under three different O2 tensions (Fig.

6). The DNA sample from the coccoid form was

significantly larger than that from the bacillary form.

Interestingly, the amount of 8-OHdG in the coccoid

form cultured under anaerobic conditions was signifi-

cantly larger than that in the coccoid form cultured

under aerobic conditions. Agarose gel electrophoresis

revealed that, although most DNA from the bacillary

form cultured under 7% O2 was intact, significant

fractions of DNA from the coccoid forms cultured

under either anaerobic or aerobic conditions had under-

gone degradation.

Frequency of H. pylori mutations

Because ROS are potential mutagens, we analyzed the

frequency of mutation of H. pylori cultured under dif-

ferent O2 tensions in the presence and absence of various

antibiotics. The resistance of H. pylori to antibiotics,

such as rifampicin, metronidazole, and ciprofloxacin, can

be achieved through mutation of the responsive genes:

rpoB, rdxA, and gyrA respectively [23]. Although the

frequency of H. pylori mutation cultured under 7% O2

was fairly low, it differed significantly (~50-fold)

depending on the antibiotic used for the analysis (met-

ronidazole < rifampicin < ciprofloxacin). The frequency

of H. pylori mutation was significantly higher under

anaerobic culture conditions than under microaerobic

conditions irrespective of the antibiotic used (Fig. 7).

Fig. 7. Effect of oxygen tension on the mutation of H. pylori. H. pyloriwas cultured under < 1% (closed columns), 7% (open columns), and21% (hatched columns) O2 conentrations for 20 h. Then, cells werecollected by centrifugation and cultured again on Brucella agar platesunder 7% O2 in the presence or absence of either 1 Ag/ml ciprofloxacin,10 Ag/ml rifampicin, and 5 Ag/ml metronidazole. After 3 days, thecolonies were counted. The frequency of mutation was expressed as theratio of drug-resistant colonies to viable cells cultured in the absence ofantibiotics. Data are means F SD derived from four independentexperiments. Statistical analysis was performed using Student’s t test.Ciprofloxacin: *p < .05 versus cells cultured under 7% O2.

The frequency of mutation was also higher with H. pylori

cultured under anaerobic conditions than that of H. pylori

cultured under aerobic conditions.

DISCUSSION

The present work reports that the bacillary form of H.

pylori cultured under optimum O2 tension generates

preferentially substantial amounts of the superoxide

radical, whereas cells cultured under nonoptimum O2

tension generate predominantly the hydroxyl radical.

Cytochrome c oxidase activity in the coccoid forms of

H. pylori cultured under nonoptimum O2 tension was

lower than that in cells cultured under optimum O2

tension. The specific activities of SOD and urease are

also lower in the former than in the latter. Oxidative

modification of H. pylori proteins and DNA increased in

cells cultured under nonoptimum O2 tension.

These results suggest that growth of the coccoid forms

of H. pylori cultured under nonoptimal O2 tension

decreased presumably due to oxidative injury of the

proteins and DNA by the increased generation of hy-

droxyl radicals as compared with the bacillary form.

Because the hydroxyl radical is highly toxic [28,29],

oxidative injury of H. pylori constituents including

various enzymes required for cell growth might occur

nonspecifically. When H. pylori was cultured under

nonoptimum O2 tension, its pyruvate-dependent respira-

tion decreased more markedly than did the ascorbate/

TMPD-dependent respiration. This observation suggests

that some components of the electron transport system,

such as cytochromes bc1 and c553, were impaired more

strongly than other components, such as cytochrome c

oxidase. Thus, the terminal oxidase might seem to be

more resistant to oxidative stress than other constituents

in the respiratory chain.

Because both proteins and DNA were oxidatively

impaired more markedly in the coccoid form of H. pylori

that generated preferentially hydroxyl radicals than in the

bacillary form, transition metals in some proteins of the

former might have become unable to catalyze the Fenton-

like caged reaction [30]. In fact, the amount of protein

carbonyl groups in the coccoid form of H. pylori was

significantly larger than that in the bacillary form. Thus,

substantial amounts of cellular proteins including metal-

loproteins seem to undergo denaturation during and/or

after the transformation of H. pylori.

The present work also shows that oxidation and

degradation of DNA were apparent in H. pylori cultured

under nonoptimum O2 conditions. The rate of mutation

of H. pylori also increased markedly when cultured under

nonoptimum O2 conditions presumably due to the in-

creased generation of hydroxyl radicals. It should be

noted that the amount of 8-OHdG in H. pylori cultured

A.-M. PARK et al.1132

under anaerobic conditions was higher than that of H.

pylori cultured under 21 and 7% O2 by about 3.7- and 9-

fold, respectively. Furthermore, the mutation rate was

also higher in H. pylori cultured under anaerobic con-

ditions than in H. pylori cultured under aerobic condi-

tions by about 1.5- to 5-fold. It is known that DNA-

repairing enzymes are often induced in various cells

exposed to oxidative stress. In fact, Escherichia coli

shows adaptive response to hydrogen peroxide [31] and

superoxide [32] by upregulating OxyR, SoxR, and SoxS

[33–36]. Because H. pylori lacks these stress-induced

genes [37,38], degradation of oxidized DNA would have

been enhanced more markedly under nonoptimum O2

tension. Furthermore, ATP generation would not occur

under anaerobic conditions so that DNA-repairing

enzymes functioned minimally. Although the amounts

of hydroxyl radicals generated were similarly high in H.

pylori cultured under anaerobic and aerobic conditions,

gene mutation caused by DNA oxidation occurred more

markedly under anaerobic conditions than under aerobic

conditions. H. pylori is known to have an unusually large

genetic diversity; hence, the genomic DNAs of cells

obtained from one patient differ significantly from those

of another [39,40]. The types of mutation found in H.

pylori are point mutation, recombination, insertion, sub-

stitution, and deletion. It is well documented that ROS

accelerate the mutation of DNA [41,42]. Hence, the

property of H. pylori to generate ROS seems to be an

important factor that enhances the genetic diversity of

this bacterium.

Because H. pylori upregulates the activities of CO2-

fixation enzymes, particularly under aerobic conditions,

it grows preferentially in the presence of 5–10% CO2

[43]. However, the transformation of H. pylori from

bacillary to coccoid form occurs preferentially in the

absence of CO2. Thus, we cultured H. pylori without

exogenous use of CO2. Preliminary experiments in this

laboratory showed that H. pylori cultured under aerobic

or microaerobic conditions with 5% CO2 generated

predominantly superoxide radicals (data not shown).

Thus, the presence of CO2 seems to affect the ability

of H. pylori to generate either superoxide or hydroxyl

radicals irrespective of the morphological transformation.

Although the coccoid form of H. pylori does not

grow under culture conditions, this form has been

postulated to retain virulence and induce gastric inflam-

mation in BALB/cA mice [44]. In fact, the coccoid

form of H. pylori is frequently found in human stomach

and extragastric environments [4–6,45,46]. Chan et al.

showed that 94 and 50% of H. pylori-positive patients

with gastric adenocarcinoma and peptic ulcer have the

coccoid form of cells, respectively [45]. Thus, the ROS

generated by the bacillary and/or coccoid forms of H.

pylori may have cytotoxic effects not only on H. pylori

itself but also on gastric mucosal cells in its hosts. The

possible involvement of ROS generated by the two forms

of H. pylori in the mechanism of gastric carcinogenesis

should be studied further.

Acknowledgments—This work was supported by Special CoordinationFunds for Promoting Science and Technology from the Ministry ofEducation, Culture, Sports, Science and Technology, the JapaneseGovernment (11877031 to M.I.).

REFERENCES

[1] Parsonnet, J.; Hansen, S.; Rodriguez, L.; Gelb, A. B.; Warnke,R. A.; Jellum, E.; Orentrich, N.; Vogelman, J. H.; Friedman, G.D. Helicobacter pylori infection and gastric lymphoma.N. Engl. J.Med. 330:1267–1271; 1994.

[2] Hahm, K. B.; Lee, K. J.; Kim, J. H.; Cho, S. W.; Chung, M. H.Helicobacter pylori infection, oxidative DNA damage, gastriccarcinogensis, and reversibility by rebamipide. Dig. Dis. Sci. 43(Suppl.):72S–77S; 1998.

[3] Kabir, S. Detection ofHelicobacter pylori in faeces by culture, PCRand enzyme immunoassay, J. Med. Microbiol., 50:1021–1029;2001.

[4] Klein, P. D.; Graham, D. Y.; Gaillour, A.; Opekun, A. R.; Smith,E. O. Water source as a risk factor for Helicobacter pylori infec-tion in Peruvian children. Lancet 337:1503–1506; 1991.

[5] Mendall, M. A.; Northfield, T. C. Transmission of Helicobacterpylori infection. Gut 37:1–3; 1995.

[6] Sarker, S. A.; Mahalanabis, D.; Hildebrand, P.; Rahaman, M. M.;Bardhan, P. K.; Fuchs, G.; Beglinger, C.; Gyr, K. Helicobacterpylori: prevalence, transmission, and serum pepsinogen II concen-trations in children of a poor periurban community in Bangladesh.Clin. Infect. Dis. 25:990–995; 1997.

[7] Inoue, M.; Sato, E. F.; Park, A. M.; Nishikawa, M.; Kasahara, E.;Miyoshi, M.; Ochi, A.; Utsumi, K. Cross-talk between NO andoxyradicals, a supersystem that regulates energy metabolism andsurvival of animals. Free Radic. Res. 33:757–770; 2000.

[8] Schwille, P. O.; Schellerer, W.; Steiner, H.; Reitzenstein, M. Ratgastric mucosal oxygen tension, ulcer index, plasma gastrin andglucagon following restraint stress. Influence of vagotomy,splanchnicotomy and exogenous secretin. Res. Exp. Med. (Berlin)167:149–158; 1976.

[9] Sorberg, M.; Nilsson, M.; Hanberger, H.; Nilsson, L. E. Morpho-logic conversion of Helicobacter pylori from bacillary to coccoidform. Eur. J. Clin. Microbiol. Infect. Dis. 15:216–219; 1996.

[10] Gribbon, L. T.; Barer, M. R. Oxidative metabolism in noncultu-rable Helicobacter pylori and Vibrio vulnificus cells studied bysubstrate-enhanced tetrazolium reduction and digital image pro-cessing. Appl. Environ. Microbiol. 61:3379–3384; 1995.

[11] Cellini, L.; Robuffo, I.; Di Campli, E.; Di Bartolomeo, S.; Tarabo-relli, T.; Dainelli, B. Recovery of Helicobacter pylori ATCC43504from a viable but not culturable state: regrowth or resuscitation?APMIS 106:571–579; 1998.

[12] Kusters, J. G.; Gerrits, M. M.; Van Strijp, J. A.; Vandenbroucke-Grauls, C. M., Coccoid forms of Helicobacter pylori are the mor-phologicmanifestationof cell death. Infect. Immun.65:3672–3679;1997.

[13] Bode, G.; Mauch, F.; Malfertheiner, P. The coccoid forms ofHelicobacter pylori: criteria for their viability. Epidemiol. Infect.111:483–490; 1993.

[14] Segal, E. D.; Falkow, S.; Tompkins, L. S. Helicobacter pyloriattachment to gastric cells induces cytoskeletal rearrangementsand tyrosine phosphorylation of host cell proteins. Proc. Natl.Acad. Sci. USA 93:1259–1264; 1996.

[15] Wang, X.; Sturegard, E.; Rupar, R.; Nilsson, H. O.; Aleljung,P. A.; Carlen, B.; Willen, R.; Wadstrom, T. Infection of BALB/c A mice by spiral and coccoid forms of Helicobacter pylori.J. Med. Microbiol. 46:657–663; 1997.

[16] Nakamura, A.; Park, A. M.; Nagata, K.; Sato, E. F.; Kashiba, M.;

Transformation of H. pylori 1133

Ogata, T.; Inoue, M. Oxidative cellular damage associated withtransformation of Helicobacter pylori from a bacillary to a coc-coid form. Free Radic. Biol. Med. 28:1611–1618; 2000.

[17] Park, A. M.; Nagata, K.; Sato, E. F.; Tamura, T.; Shimono, K.;Inoue, M. Mechanism of strong resistance of Helicobacter pylorirespiration to nitric oxide. Arch. Biochem. Biophys. 411:129–135;2003.

[18] Ferraro, R. L.; Hazell, S. L.; Lee, A. The urease enzymes ofCampylobacter pylori and a related bacterium. J. Med. Microbiol.27:33–40; 1988.

[19] Crapo, J. D.; McCord, J. M.; Fridovich, I. Preparation and assayof superoxide dismutases. Methods Enzymol. 53:382–393; 1978.

[20] Sugioka, K.; Nakano, M.; Kurashige, S.; Akuzawa, Y.; Goto, T. Achemiluminescent probe with a Cypridina luciferin analog, 2-methyl-6-phenyl-3,7-dihydroimidazo[1,2-a]pyrazin-3-one, speci-fic and sensitive for O2

� production in phagocytizing macro-phages. FEBS Lett. 197:27–30; 1986.

[21] Imada, I.; Sato, F. E.; Miyamoto, M.; Ichimori, Y.; Minamiyama,Y.; Konaka, T.; Inoue, M. Analysis of reactive oxygen speciesgenerated by neutrophils using a chemiluminescence probe L-012. Anal. Biochem. 271:53–58; 1999.

[22] Levine, R. L.; Williams, J. A.; Stadtman, E. R.; Shacter, E. Car-bonyl assays for determination of oxidatively modified proteins.Methods Enzymol. 233:346–357; 1994.

[23] Wang, G.; Wilson, T. J.; Jiang, Q.; Taylor, D. E. Spontaneousmutations that confer antibiotic resistance in Helicobacter pylori.Antimicrob. Agents Chemother. 45:727–733; 2001.

[24] Choi, J. H.; Yu, B. P. Brain synaptosomal aging: free radicalsand membrane fluidity, Free Radic. Biol. Med., 18:133–139;1995.

[25] Berlett, B. S.; Stadtman, E. R. Protein oxidation in aging, disease,and oxidative stress. J. Biol. Chem. 272:20313–20316; 1997.

[26] Beckman, K. B.; Ames, B. N. Oxidative decay of DNA. J. Biol.Chem. 272:19633–19666; 1997.

[27] Kato, Y.; Uchida, K.; Kawakishi, S. Oxidative fragmentation ofcollagen and prolyl peptide by Cu(II)/H2O2: conversion of prolineresidue to 2-pyrrolidone. J. Biol. Chem. 267:23646–23651; 1992.

[28] Davies, J. A. Protein damege and degradation by oxygen radicals.I. general aspects. J. Biol. Chem. 262:9895–9901; 1987.

[29] Davies, J. A. Protein damage and degradation by oxygen radicals.II. Modification of amino acid. J. Biol. Chem. 262:9902–9907;1987.

[30] Yim, M. B.; Berlett, B. S.; Chock, P. B.; Stadtman, E. R. Man-ganese(II)-bicarbonate-mediated catalytic activity for hydrogenperoxide dismutation and amino acid oxidation: detection of freeradical intermediates. Proc. Natl. Acad. Sci. USA 87:394–398;1990.

[31] Demple, B.; Halbrook, J. Inducible repair of oxidative DNAdamage in Escherichia coli. Nature 304:466–468; 1983.

[32] Farr, S. B.; Natvig, D. O.; Kogoma, T. Toxicity and mutagenicityof plumbagin and the induction of a possible new DNA repairpathway in Escherichia coli. J. Bacteriol. 164:1309–1316; 1985.

[33] Christman, M. F.; Morgan, R. W.; Jacobson, F. S.; Ames, B. N.Positive control of a regulon for defenses against oxidative stressand some heat-shock proteins in Salmonella typhimurium. Cell41:753–762; 1985.

[34] Greenberg, J. T.; Monach, P.; Chou, J. H.; Josephy, P. D.; Demple,B. Positive control of a global antioxidant defense regulon acti-vated by superoxide-generating agents in Escherichia coli. Proc.Natl. Acad. Sci. USA 87:6181–6185; 1990.

[35] Tsaneva, I. R.; Weiss, B. soxR, a locus governing a superoxideresponse regulon in Escherichia coli K-12. J. Bacteriol.172:4197–4205; 1990.

[36] Wu, J.; Weiss, B. Two divergently transcribed genes, soxR andsoxS, control a superoxide response regulon of Escherichia coli.J. Bacteriol. 173:2864–2871; 1991.

[37] Tomb, J. F.; White, O.; Kerlavage, A. R.; Clayton, R. A.; Sutton,G. G.; Fleischmann, R. D.; Ketchum, K. A.; Klenk, H. P.; Gill, S.;Dougherty, B. A.; Nelson, K.; Quackenbush, J.; Zhou, L.; Kirk-ness, E. F.; Peterson, S.; Loftus, B.; Richardson, D.; Dodson, R.;Khalak, H. G.; Glodek, A.; McKenney, K.; Fitzegerald, L. M.;Lee, N.; Adams, M. D.; Venter, J. C. The complete genome se-quence of the gastric pathogen Helicobacter pylori. Nature388:539–547; 1997.

[38] Alm, R. A.; Ling, L. S.; Moir, D. T.; King, B. L.; Brown, E. D.;Doig, P. C.; Smith, D. R.; Noonan, B.; Guild, B. C.; deJonge,B. L.; Carmel, G.; Tummino, P. J.; Caruso, A.; Uria-Nickelsen,M.; Mills, D. M.; Ives, C.; Gibson, R.; Merberg, D.; Mills, S.D.; Jiang, Q.; Taylor, D. E.; Vovis, G. F.; Trust, T. J. Genomic-sequence comparison of two unrelated isolates of the humangastric pathogen Helicobacter pylori. Nature 397:176–180;1999.

[39] Marshall, D. G.; Dundon, W. G.; Beesley, S. M.; Smyth, C. J.Helicobacter pylori: -a conundrum of genetic diversity. Microbi-ology 144:2925–2939; 1998.

[40] Blaser, M. J.; Berg, D. E. Helicobacter pylori genetic diversityand risk of human disease. J. Clin. Invest. 107:767–773; 2001.

[41] Oya, Y.; Yamamoto, K.; Tonomura, A. The biological activity ofhydrogen peroxide: I. Induction of chromosome-type aberrationssusceptible to inhibition by scavengers of hydroxyl radicals inhuman embryonic fibroblasts. Mutat. Res. 172:245–253; 1986.

[42] Farr, S. B.; Touati, D.; Kogoma, T. Effects of oxygen stress onmembrane functions in Escherichia coli: role of HPI catalase. J.Bacteriol. 170:1837–1842; 1988.

[43] Donelli, G.; Matarrese, P.; Fiorentini, C.; Dainelli, B.; Taraborelli,T.; Di Campli, E.; Di Bartolomeo, S.; Cellini, L. The effect ofoxygen on the growth and cell morphology of Helicobacter py-lori. FEMS Microbiol. Lett. 168:9–15; 1998.

[44] Wang, X.; Sturegard, E.; Rupar, R.; Nilsson, H. O.; Aleljung, P.A.; Carlen, B.; Willen, R.; Wadstrom, T. Infection of BALB/c Amice by spiral and coccoid forms of Helicobacter pylori. J. Med.Microbiol. 46:657–663; 1997.

[45] Chan, W. Y.; Hui, P. K.; Leung, K. M.; Chow, J.; Kwok, F.; Ng, C.S. Coccoid forms of Helicobacter pylori in the human stomach.Am. J. Clin. Pathol. 102:503–507; 1994.

[46] Ogata, M.; Araki, K.; Ogata, T. An electron microscopic study ofHelicobacter pylori in the surface mucous gel layer. Histol. His-topathol. 13:347–358; 1998.

ABBREVIATIONS

DNP—dinitrophenol

LO12—8-amino-5-chloro-7-phenyl-pyridopyridazine

MCLA—2-methyl-6-[f-methoxyphenyl]-3,7-dihydroi-

midazol[1,2-a]pyrazin-3-one

ROS—reactive oxygen species

SOD—superoxide dismutase

TMPD—N,N,NV,NV-tetramethyl-p-phenylenediamine