INFECTION AND IMMUNITY,0019-9567/01/$04.00�0 DOI: 10.1128/IAI.69.11.6749–6754.2001

Nov. 2001, p. 6749–6754 Vol. 69, No. 11

Copyright © 2001, American Society for Microbiology. All Rights Reserved.

Analysis of Cytokines in the Early Development of GastricSecondary Lymphoid Follicles in Helicobacter pylori-Infected

BALB/c Mice with Neonatal ThymectomyKAZUSHIGE UCHIDA,1 KAZUICHI OKAZAKI,1* ANDRAS DEBRECCENI,1 TOSHIKI NISHI,1

HIROSI IWANO,1 MAKI INAI,1 SUGURU UOSE,1 HIROSHI NAKASE,1 MASAYA OHANA,1

CHIKASHI OSHIMA,1 YUMI MATSUSHIMA,1 CHIHARU KAWANAMI,1 HIROSHI HIAI,2

TORU MASUDA,3 AND TSUTOMU CHIBA1

Department of Gastroenterology and Endoscopic Medicine,1 Department of Pathology and Biology of Diseases,2

and Institute of Immunology,3 Graduate School of Medicine, Kyoto University, Kyoto, Japan

Received 29 September 2000/Returned for modification 8 February 2001/Accepted 10 July 2001

Immunological interaction between the host and Helicobacter pylori seems to play a critical role in follicularformation in gastric mucosa. We reported H. pylori-induced follicular gastritis model using neonatally thymec-tomized mice. In this study, we investigated the involvement of various cytokines in this model. BALB/c micewere thymectomized on the third day after birth (nTx). At 6 weeks old, these mice were orally infected with H.pylori. Histological studies showed that follicular formation occurred from 8 weeks after the infection and thatmost of the infiltrating lymphocytes were CD4� and B cells. Neutrophils increased transiently at 1 week afterthe infection. Gamma interferon, interleukin-7 (IL-7), and IL-7 receptor were expressed in the stomach of thenTx mice irrespective of the infection. In contrast, expressions of the tumor necrosis factor alpha, IL-4 andlymphotoxin-� genes were remarkably upregulated by the infection. Our findings suggest that follicularformation may require cooperative involvement of a Th2-type immune response, tumor necrosis factor alphaand lymphotoxin-� in addition to the Th1-type immune response in H. pylori-induced gastritis in nTx mice.

Evidence is accumulating that Helicobacter pylori infection isinvolved in the pathogenesis of gastric ulcers, cancer, and low-grade lymphomas of mucosal-associated lymphoid tissue(MALToma), in addition to atrophic gastritis (16, 27, 37, 39).Among these diseases, a characteristic feature of MALToma isthe development of secondary follicles in the corporal lesionsbefore or at the onset of lymphoma (6). The immunologicalinteraction between the host and H. pylori seems to play acritical role in the pathophysiology of follicular gastritis. How-ever, the mechanisms of host immune reactions to H. pyloriinfection and subsequent follicular formation in the stomachremain unclear.

Use of gene knockout and/or transgenic mice has revealedthe in vivo functions of a number of cytokines or cytokinereceptors in not only normal but also various pathologicalconditions. It is known that lymphotoxin � and � (LT-� and-�), tumor necrosis factor alpha (TNF-�), and TNF receptorsI and II (TNF-R I and II) are involved in constructing thegerminal centers of the Peyer’s patches, spleen, or peripherallymph nodes (19). Moreover, injection of anti-interleukin-7receptor (IL-7R) antibody into pregnant mice has been shownto inhibit the development of germinal centers in Peyer’spatches after birth (1, 10). These findings suggest that multiplefactors seem to be required for the formation of germinalcenters in lymphoid tissue.

BALB/c mice thymectomized on the third day after birth

(nTx) spontaneously develop autoimmune gastritis. In nTxBALB/c mice that show a Th1-predominant reaction in thestomach, CD4-positive Th1 cells often induce autoimmunegastritis, whereas lymphoid follicles never occur (12, 13, 25,31). In our previous study, we showed that infection of BALB/cmice that had autoimmune gastritis with mouse-adapted H.pylori permitted successful colonization of H. pylori in the cor-pus, leading to follicular gastritis similar to human MALToma(26). In the present study, we examined by using this murinemodel the histopathology and cytokine responses in the earlydevelopment of gastric lymphoid follicles to H. pylori infection.

MATERIALS AND METHODS

Mice and thymectomy. Male and female BALB/c Cr.Slc mice (Japan SLC,Shizuoka, Japan) were bred in our animal facility in Kyoto University underspecific-pathogen-free conditions. Neonatal thymectomy was performed on thethird day after birth under ether anesthesia, as described previously (12, 13, 25,26, 31). In a preliminary study, we confirmed the findings reported previously (12,13, 25, 26, 31), in which there were no immunological differences between femaleand male mice; we also confirmed the severity of the autoimmune gastritis andthe immunological findings, including antiparietal autoantibodies and infiltratinglymphocytes (data not shown). The mice were divided into four groups (n � 10in each group): (i) normal (non-nTx) mice without H. pylori infection, (ii) normal(non-nTx) mice with H. pylori infection, (iii) nTx mice without H. pylori infection,and (iv) nTx mice with H. pylori infection.

Bacteria and infection. H. pylori (TN2FG4) isolated from a patient with aduodenal ulcer was donated from M. Nakao (Pharmaceutical Research Division,Takeda Chemical Industries, Ltd., Osaka, Japan). It was maintained in BloodAgar Base No. 2 with horse serum (5% [vol/vol]) containing amphotericin B at2.5 mg/liter, trimethoprim at 5 mg/liter, polymyxin B at 1,250 IU/liter, andvancomycin at 10 mg/liter. The plates were incubated in a microaerophilic at-mosphere at 37°C for 48 h. The inoculated H. pylori strain, TN2FG4, was CagAand VacA positive as described elsewhere (38). Both nTx and non-nTx mice at6 weeks old were orally infected with 108 H. pylori organisms as describedpreviously (26). Infected mice were sacrificed 1, 2, 4, 8, and 12 weeks later.Colonization of H. pylori was confirmed by PCR Southern blot analysis for theurease gene using DNA extracted from the corpus mucosa. The urease gene was

* Corresponding author. Mailing address: Department of Gastroen-terological Endoscopic Medicine, Postgraduate School of Medicine,Kyoto University, Shogoin-Kawaracho, Sakyo-ku, Kyoto City, Kyoto,606-8507, Japan. Phone: 81-75-751-3413. Fax: 81-75-751-3414. E-mail:okak@kuhp.kyoto-u.ac.jp.

6749

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

found to be amplified at 12 weeks after infection (26). Noninfected mice weresacrificed at the same time. The sera and stomachs were stored until use. Weconfirmed the abnormal hyperimmune status by measuring the levels of autoanti-body in serum against the parietal cells by enzyme-linked immunosorbent assay at 6weeks after thymectomy and before infection, as described previously (26).

Histology and immunohistochemistry. The stomach was removed from eachmouse, fixed with 4% phosphate-buffered formaldehyde (pH 7.2), and preparedfor histologic examination. The sections were stained with hematoxylin andeosin. Immunohistochemical staining with monoclonal antibodies (MAbs) wasperformed on freshly frozen sections by the avidin-biotin immunoperoxidasemethod, as described previously (26). Briefly, freshly frozen sections were fixedin acetone for 10 min, rinsed in phosphate-buffered saline (PBS, pH 7.2), andincubated with 10% normal rat serum (as a blocking agent) for 20 min. Theywere incubated for 1 h with the biotin-labeled MAbs as follows. The commer-cially available MAbs to mouse surface markers were rat anti-mouse B220,immunoglobulin M (IgM), CD4, and CD8 MAbs (Pharmingen, San Diego,Calif.). After incubation, the sections were washed with PBS and then incubatedwith ABC complex (Vector Laboratories, Burlingame, Calif.) for another 30 min.Sections washed with PBS were reacted with a fresh mixture of 0.05% 3,3�-diaminobenzidine and 0.005% H2O2 in Tris buffer (0.05 M, pH 7.6) and thenwashed with distilled water. The controls were exposed to normal rat seruminstead of the MAbs. These control samples showed no staining.

Flow cytometry. Infiltrating cells were isolated from the stomach of nTx micewith or without H. pylori infection according to the method reported previously(26). For flow cytometric analysis, 106 viable infiltrating cells were incubated withphycoerythrin or fluorescein isothiocyanate-conjugated MAbs, anti-��-TCR andMAC-1 (Pharmingen) at 4°C for 30 min. The labeled cells were analyzed by flowcytometry with an Epics XL (Coulter, Miami, Fla.).

Semiquantitive RT-PCR. To analyze cytokine gene expressions by reversetranscriptase-PCR (RT-PCR), total RNA was extracted with the RNA extractionsolution Isogen (Nippon Gene, Tokyo, Japan) from the murine stomach. TotalRNA was reverse transcribed into DNA with the SuperScript PreamplificationSystem (Gibco-BRL, Gaithersburg, Md.). The total RNA in the reaction mixturewas heated at 42°C for 50 min and at 70°C for another 15 min and then chilledon ice.

PCR was performed with a mixture of cDNA, 20 mM Tris-HCl (pH 8.4), 50mM KCl, 2.5 mM MgCl2, a 200 mM concentration of each deoxynucleotidetriphosphate (Perkin-Elmer, Branchburg, N.J.), a 50 pM concentration of eachspecific primer, and 1.0 U of Taq DNA polymerase (AmpliTaq Gold; Perkin-Elmer). The specific primers of mouse TNF-�, IL-4, IL-7, gamma interferon(IFN-�), and �-actin were purchased from Clontech (Palo Alto, Calif.). Thespecific primers of LT-� and IL-7R were synthesized according to the sequencedescribed previously (1, 15). The primer sequences in this study were as follows:for IL-4, 5�-CCAGCTAGTTGTCATCCTGCTCTCCTTTCTCG and 3�-CGTGGTACTTACTCAGGTTCAGGTGTAGTGAC; for IL-7, 5�-GCCTGTCACATCATCTGAGTGAA and 3�-CAGGAGGCATCCAGGAACTTCTG; for IL-7R,5�-CGAGTGAAATGCCTAACTC and 3�-GCGTCCAGTTGCTTTCAC; forIFN-�, 5�-TGCATCTTGGCTTTGCAGCTCTTCCTCATGGC and 3�-CGGTTCAAACTCCAGTTGTTGGTGTCCAGGT, for TNF-�, 5�-TTCTGTCTGAACTTCGGGGTGATCGGTCC and 3�-GTATGAAGCAAATCGGCTGACGGTGTGGG; for LT-�, 5�-TCTCCACCTCTTGAGGGTG and 3�-ACGATCCGTGCTTGCTCTC; and for �-actin, 5�-GTGGGCCGCTCTAGGCACCAA and 3�-CTCTTTGATGTCACGCACGATTTC. Amplification was performed with athermal cycler (GeneAmp PCR System 9600R; Perkin-Elmer) for 35 cycles, eachof which consisted of 20 s at 95°C, 1 min at 55°C, and 1 min at 72°C. The finalcycle induced an extension step for 10 min at 72°C. A 10-ml portion of each PCRproduct was electrophoresed on a 2.0% agarose gel containing ethidium bro-mide. The densities of bands on the gels were measured by an image autoana-lyzing system (Fotodyne, FOTOanalyst, and Archive ECLIPSE; AdvancedAmerican Biotechnology, Fullerton, Calif.) and expressed as the absorbancelevel. Semiquantitative levels of each cytokine were corrected by the �-actindensity of each sample.

Statistical analysis. All results were expressed as mean � the standard devi-ation for each sample, except where noted. The generalized Wilcoxon t test wasused to compare absorbance levels between nTx only and nTx with H. pyloriinfection for each cytokine. A two-tailed P value of 0.05 was used to indicatestatistical significance.

RESULTS

Histopathology and immunohistochemistry. Approximately60% of the nTx mice developed severe inflammation in thebody, with no or weak inflammation in the antral mucosa. We

used the nTx mice that developed inflammation for the fol-lowing experiment. Infiltration of mononuclear cells occurredin the lamina propria of the corpus mucosa as early as 6 weeksafter nTx in association with the production of autoantibodiesagainst parietal cells (data not shown). As the inflammatorycell population expanded along the glands, the destruction ofparietal cells and chief cells became prominent, resulting inglandular atrophy. However, no germinal centers were ob-served, and the diffusely or massively infiltrating cells consistedmainly of lymphocytes and a few other mononuclear cells,including plasma cells and macrophages. These pathologicalfeatures basically remained unchanged throughout the exper-iment (Fig. 1a). These observations are consistent with ourprevious reports (12, 13, 25, 26, 31).

After H. pylori infection, the gastric mucosa of the nTx micedid not show significant alteration for the first 2 weeks (Fig. 1band c). At 4 weeks after infection, clusters of mononuclear cellsappeared at the bottom of the lamina propria in the gastricmucosa (Fig. 1d). A distinct change in lymphoid cell infiltrationwas seen at 8 weeks after infection (14 weeks old, Fig. 1e):unlike noninfected nTx mice, infected mice showed a lymphoidfollicle in the corpus mucosa in 7 of 10 mice (P 0.05 versusnoninfected nTx mice). At 12 weeks after infection (18 weeksold), many secondary follicles developed in 6 of 10 mice (P 0.05 versus noninfected nTx mice) (Fig. 1f), whereas no follic-ular formation was observed in noninfected nTx mice through-out the experiment (Fig. 1a). As we previously reported (26),such a histologic feature observed in these infected mice re-sulted in no significant difference of the mononuclear cell in-filtration from the uninfected mice (data not shown). Therewas little or no inflammation in both the noninfected andinfected normal (non-nTx) mice, as described previously (26).No ��-T-cell-receptor-positive cells could be detected amongthe intraepithelial lymphocytes in the gastric mucosa of eitherinfected or noninfected nTx mice (data not shown). B220- andIgM-positive cells predominantly infiltrated in the mucosa inassociation with the follicle formation 12 weeks after H. pyloriinfection, confirming our previous study (26).

Flow cytometry. In the infected nTx mice, flow cytometricanalysis demonstrated a transient but prominent increase inthe number of neutrophils after 1 week (Fig. 2a). The percent-ages of infiltrating macrophages and �� T cells did not changeafter infection (Fig. 2b and c).

Semiquantitative cytokine expressions by RT-PCR. Geneexpressions of IFN-�, IL-7, and IL-7R were similarly detectedby RT-PCR in both the infected and noninfected nTx mice(Fig. 3). On the other hand, weak expressions of IL-4, TNF-�and LT-� genes were detected in 1, 6, and 2 of 8 noninfectednTx mice, respectively (Table 1). However, LT-�, TNF-�, andIL-4 gene expressions started to be upregulated in all infectednTx mice from 1, 1, and 2 weeks after infection, respectively(Fig. 3); at 12 weeks after H. pylori infection expressions weresignificantly higher than those of noninfected nTx mice (Table1). In normal (non-nTx) mice either with or without infection,these cytokines could not be detected (data not shown).

DISCUSSION

There are no lymphoid tissues such as lymphoid follicles orintraepithelial lymphocytes in the normal stomach (5). How-

6750 UCHIDA ET AL. INFECT. IMMUN.

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

FIG. 1. Histologic findings of the gastric mucosa in neonatally thymectomized (nTx) BALB/c mice with H. pylori infection. Neonatal thymec-tomy was performed on the 3rd day after birth. nTx mice at 6 weeks old were orally infected with H. pylori and sacrificed 1, 2, 4, 8, and 12 weekslater. (a) At the age of 18 weeks, noninfected nTx mice showed infiltration of mononuclear cells and gland atrophy characterized by loss of parietaland chief cells, with replacement by proliferating epithelial cells. Magnification, 50. Histologic findings of nTx mice without H. pylori infectionwere not changed throughout the experiment. (e and f) In infected nTx mice, follicle formation was observed at 8 weeks after infection in 7 of 10mice (P 0.05 versus noninfected nTx mice) (e), and the body mucosa clearly showed germinal center formation at 12 weeks after infection in6 of 10 mice (P 0.05 versus noninfected nTx mice) (f). The photographs in panels b, c, and d show gastric histology at 1, 2, and 4 weeks afterH. pylori infection in nTx mice, respectively.

VOL. 69, 2001 CYTOKINES IN H. PYLORI-INDUCED FOLLICULAR GASTRITIS 6751

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

ever, some H. pylori-positive patients develop follicular gastri-tis with germinal centers (6), which may be an important pre-disposing condition for MALToma (39). The mechanism bywhich follicular formation occurs in H. pylori-positive chronicgastritis is still unclear. In our histologic study, follicular gas-tritis was the most characteristic feature in the infected nTxBALB/c mice (26). This feature has never been observed in thenoninfected nTx mice or infected normal mice in our previous(12, 25, 26, 31) and the present studies. Moreover, infection ofnormal mice with H. pylori, in contrast to infection of nTx mice,induced little infiltration of mononuclear cells in the gastricmucosa (26). These findings suggest that host factors playimportant roles in the development of gastritis and even fol-licular formation after H. pylori infection. Supporting this idea,recent studies with the H. felis (5, 21) or H. pylori (17) mouse

model demonstrated that inflammation of the stomach wassevere in C57BL/6J but weak in BALB/c mice. The C57BL/6Jmouse is known to be a Th1-dominated strain for intracellularinfections such as those by Leishmania major, whereas theBALB/c mouse is believed to be Th2 dominated (9). Indeed, inthe case of Helicobacter infection, gastric inflammation seems

FIG. 2. Flow cytometric analysis of mononuclear cells infiltratingthe gastric mucosa of nTx mice after H. pylori infection. Infected nTxmice were sacrificed 1, 2, and 4 weeks later. Infiltrating mononuclearcells were collected from the whole stomachs of three mice at eachtime point. (a) The panel shows that neutrophils were transientlyincreased at 1 week after H. pylori infection (P 0.05 versus nonin-fected nTx mice; 2.7, 8.6, 2.9, and 0.3% at 0, 1, 2, and 4 weeks,respectively). Panels b and c show that the ratios of both macrophages(3.4, 4.0, 4.1, and 4.0% at 0, 1, 2, and 4 weeks) and �� T cells (3.5, 1.8,4.7, and 2.9% at 0, 1, 2, and 4 weeks) were not significantly changed byH. pylori infection. The data are representative of three separate ex-periments and are expressed as the percentage of all of the infiltratingcells.

FIG. 3. RT-PCR of various cytokine mRNAs in the gastric mucosaafter H. pylori infection. IFN-� mRNA was present in nTx mice, but itsexpression was not altered by H. pylori infection. LT-�, TNF-�, andIL-4 gene expressions were upregulated in all infected nTx-mice at 1,1, and 2 weeks after infection, respectively. There were no differencesin the gene expression of the IL-7 and IL-7R genes between infectedand noninfected nTx mice. N, noninfected normal (non-nTx) mice;nTx, noninfected nTx mice at 18 weeks old. The designations 1w, 2w,4w, 8w, and 12w refer to 1, 2, 4, 8, and 12 weeks, respectively, afterinfection in nTx mice.

TABLE 1. Comparison of cytokine gene expressions in thestomachs of nTx mice with or without H. pylori infection

as determined by RT-PCR

Cytokine GroupaPositive prevalence

(no. of animals positive/total no. of animals)

Relative intensityof mRNAb � SD

IFN-� nTx 8/8 469.00 � 109.00nTx�Hp 10/10 488.95 � 96.23

IL-7 nTx 8/8 153.45 � 17.97nTx�Hp 10/10 109.33 � 33.26

IL-7R nTx 8/8 404.73 � 43.70nTx�Hp 10/10 519.42 � 61.39

IL-4 nTx 1/8 30.15 � 32.01nTx�Hp 8/10 168.80 � 17.57

TNF-� nTx 6/8 91.34 � 32.90nTx�Hp 7/7 237.68 � 39.73

LT-� nTx 2/8 65.93 � 36.97nTx�Hp 9/10 181.29 � 41.61

a nTx, thymectomized mice without H. pylori infection; nTx�Hp, thymecto-mized mice with H. pylori infection.

b Values are means � the standard deviations. The densities of the bands weremeasured by an image autoanalyzing system (Fotodyne, FOTOanalyst, and Ar-chive ECLIPSE) and corrected by using the �-actin density of each sample.Samples were obtained from the mouse stomachs 12 weeks after H. pylori infec-tion. For the last three pairs of values (i.e., for IL-4, TNF-�, and LT-�), thevalues for the infected animals were significantly different (P 0.05) from thevalues for the noninfected nTx mouse group as calculated by using the Wilcoxont test.

6752 UCHIDA ET AL. INFECT. IMMUN.

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

to be predominantly associated with Th1 response (11, 22). Itappears reasonable that H. pylori infection induced severe gas-tritis in nTx BALB/c mice, because the microenvironment inthe stomach changes to be Th1 dominant by nTx (13, 25).

At the early stage of H. pylori infection in this study, neu-trophils in the gastric mucosa transiently increased at 1 weekand then decreased in a time-dependent manner. These find-ings confirmed that neutrophils are mainly involved in theacute phase of the reaction against H. pylori infection and notin the chronic phase (4, 23). Macrophages and �� T cells arethought to play important roles in the local immune responsesto bacterial antigens, including H. pylori (30, 34, 35). However,the distribution of macrophages and �� T cells has been poorlyexplored in the H. pylori-infected gastric mucosa (8, 30, 33, 34).Our previous and present studies have shown that macro-phages mainly infiltrated at the bottom of gastric mucosa ofnTx mice after H. pylori infection, although their numbers andratios did not significantly change (12, 25, 26, 31). On the otherhand, �� T cells in the gastric mucosa were not changed afterinfection, which suggested that �� T cells were not mainlyinvolved in the early stage of H. pylori-infected gastritis.

Among the infiltrating mononuclear cells, B cells increasedin numbers, along with formation of the germinal center at 12weeks after H. pylori infection in the nTx mice. Th2 cytokinessuch as IL-4 take important roles of maturation and prolifer-ation of B cells (2). IL-4 knockout mice lack intestinal Peyer’spatches, which suggests that the Th2 cytokines are involved inthe development of constitutive lymph follicles (36). However,the actual roles and direct linkage of the Th2 responses toacquired follicular formation are still not established. In theprevious study, we showed that Th2 cytokines as well as Th1cytokines were involved in the development of follicular gas-tritis after long-standing H. pylori infection with respect toTh1/Th2 ratio (26). This suggested that similar to constitutivegerminal centers (32), a synergistic effect of Th1 and Th2cytokines may be important in the formation of secondaryfollicles in the H. pylori-infected stomach. However, it was stillunclear whether additional molecules besides Th1 and Th2cytokines are involved in these characteristic histologicchanges or not. Recently, it has been shown that, in addition toTh2 cytokines, several molecules, such as TNF-�, TNF-R, LT-�/�, IL-7, and IL-7R, are involved in the development ofsplenic follicular dendritic cell clusters, lymph nodes, Peyer’spatches, or the primary germinal centers (3, 10, 14, 15, 19, 28,29). However, the mechanisms of organogenesis in these lym-phoid organs are not identical. TNF signaling through theTNF-R is essential for the formation of secondary lymphoidtissues (29). IL-7- or IL-7R-deficient mice lack the Peyer’spatches but possess the cecal and colonic patches (37). Incontrast, LT-�-deficient mice completely lack germinal centersin the gastrointestinal tract, suggesting that LT-�/� may have acrucial role in the formation of germinal centers in the gut-associated lymphoid tissue (3, 14). The present study showedthat in addition to IL-4, a Th2-type cytokine, gene expressionsof TNF-� and LT-� were upregulated in the H. pylori-infectednTx mice compared to those in noninfected nTx mice, whileneither the expression of the IL-7R gene nor the expression ofthe IL-7 gene was changed. Taken together, these results showthat Helicobacter infection can induce follicular gastritis only inmice with autoimmune gastritis, which requires Th1 cytokine-

mediated process, and not in normal BALB/c mice. This indi-cated that the cytokine milieu induced by autoimmune gastritismight have an effect on the development of gastric secondaryfollicles in chronic Helicobacter infection. Recently, it has beenreported that a CXC chemokine called B-cell-attracting che-mokine 1 (BCA-1) in humans (18) and B-lymphocyte chemoat-tractant (BLC) in mice (7) is important for the development ofB-cell areas of secondary lymphoid tissues. Mazzucchelli et al.investigated whether BCA-1 was induced in chronic H. pylorigastritis and involved in the formation of lymphoid follicles andMALToma (20). Although we did not investigate BLC in thepresent study, LT and TNF were shown to have a regulatoryrole in the upstream of BLC (24). Therefore, it is reasonablethat LT and TNF are also key molecules in the development ofthe lymphoid follicles in the stomach.

In conclusion, the cytokine milieu induced by autoimmunegastritis might and cooperative roles of LT-�, TNF-�, and Th1and Th2-type immune responses may be involved in the devel-opment of gastric secondary follicles by H. pylori infection.Further studies are needed to clarify the precise mechanismfor the acquired formation of germinal centers in the stomachby H. pylori infection.

ACKNOWLEDGMENTS

This work was supported by a grant-in-aid for scientific research (C)from the Ministry of Culture and Science of Japan (09670543), agrant-in-aid for Research for the Future from The Japan Society forthe Promotion of Science (JSPS-RFTF97I00201), and Research Fundsfrom the Japanese Foundation for Research and Promotion of Endos-copy (JFE-1997).

REFERENCES

1. Adachi, S., H. K. Yoshida, K. Honda, K. Maki, K. Saijo, K. Ikuta, T. Saito,and S. Nishikawa. 1998. Essential role of IL-7 receptor alpha in the forma-tion of Peyer’s patch anlage. Int. Immunol. 10:1–6.

2. Choe, J., H. Kim, R. J. Armitage, and Y. S. Choi. 1997. The functional roleof B cell antigen receptor stimulation and IL-4 in the generation of humanmemory B cells from germinal center B cells. J. Immunol. 159:3757–3766.

3. De Togni, P. D., J. Goellner, N. H. Ruddle, P. R. Streeter, A. Fick, S.Mariathasan, S. C. Smith, R. Carlson, L. P. Shornick, and J. Strauss-Schoenberger. 1994. Abnormal development of peripheral lymphoid organsin mice deficient in lymphotoxin. Science 264:703–707.

4. Dixon, M. F. 1989. Campylobacter pylori and chronic gastritis, p. 106–116. InB. J. Rathbone and R. V. Heatley (ed.), Campylobacter pylori and gastrodu-odenal disease. Blackwell Scientific, Oxford, England.

5. Enno, A., J. L. O’Rourke, C. R. Howlett, A. Jack, M. F. Dixon, and A. Lee.1995. MALToma-like lesions in the murine gastric mucosa after long-terminfection with Helicobacter felis—a mouse model of Helicobacter pylori-in-duced gastric lymphoma. Am. J. Pathol. 147:217–222.

6. Genta, R. M., H. H. Wentzell, and D. Y. Graham. 1993. Gastric lymphoidfollicles in Helicobacter pylori infection: frequency, distribution and responseto triple therapy. Hum. Pathol. 342:577–583.

7. Guan, M. D., V. N. Ngo, K. M. Ansel, E. H. Ekland, J. G. Cyster, and L. T.Williams. 1998. A B-cell-homing chemokine made in lymphoid folliclesactivates Burkitt’s lymphoma receptor-1. Nature 391:799–803.

8. Hatz, R. A., G. Meimarakis, E. Bayerdirffer, M. Stolte, T. Kirchner, and G.Enders. 1996. Characterization of lymphocytic infiltrates in Helicobacter py-lori-associated gastritis. Scand. J. Gastroenterol. 31:222–228.

9. Heinzel, F. P., M. D. Sadic, B. J. Holaday, R. L. Coffman, and R. M.Locksley. 1989. Reciprocal expression of interferon gamma or interleukin 4during the resolution or progression of murine leishmaniasis. Evidence forexpansion of distinct helper T cell subsets. J. Exp. Med. 169:59–72.

10. Hikida, M., Y. Nakayama, Y. Yamashita, Y. Kumazawa, S. Nishikawa, andH. Ohmori. 1998. Expression of recombination activating genes in germinalcenter B cells: involvement of interleukin 7 (IL-7) and the IL-7 receptor. J.Exp. Med. 188:365–372.

11. Itoh, T., Y. Wakatsuki, M. Yoshida, T. Usui, Y. Matsunaga, S. Kaneko, T.Chiba, and T. Kita. 1999. The vast majority of gastric T cells are polarized toproduce T helper 1 type cytokines upon antigenic stimulation despite theabsence of Helicobacter pylori infection. J. Gastroenterol. 34:560–570.

12. Jones, C. M., J. M. Callaghan, P. M. Gleeson, Y. Mori, T. Masuda, and B. H.

VOL. 69, 2001 CYTOKINES IN H. PYLORI-INDUCED FOLLICULAR GASTRITIS 6753

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

Toh. 1991. The parietal cell autoantibodies recognized in neonatal thymec-tomy-induced murine gastritis are the � and � subunits of the gastric protonpump. Gastroenterology 101:287–294.

13. Katakai, T., K. J. Mori, T. Masuda, and A. Shimizu. 1998. Differentiallocalization of Th1 and Th2 cells in autoimmune gastritis. Int. Immunol.10:1325–1334.

14. Koni, P. A., R. Sacca, P. Lawton, J. L. Browning, N. H. Ruddle, and R. A.Flavell. 1997. Distinct roles in lymphoid organogenesis for lymphotoxin �and � revealed in lymphotoxin �-deficient mice. Immunity 6:491–500.

15. Korner, H., M. Cook, D. S. Riminton, F. A. Lemckert, R. M. Hoek, B.Ledermann, F. Kontgen, B. Fazekas de St. Groth, and J. D. Sedgwick. 1997.Distinct roles for lymphotoxin-alpha and tumor necrosis factor in organo-genesis and spatial organization of lymphoid tissue. Eur. J. Immunol. 27:2600–2609.

16. Kuipers, E. J., A. M. Uyterlinde, A. S. Pena, R. Roosendaal, G. Pals, G. F.Nelis, H. P. Festen, and S. G. Meuwissen. 1995. Long-term sequelae ofHelicobacter pylori gastritis. Lancet 345:1525–1528.

17. Lee, A., J. L. O’Rourke, M. C. De Ungiria, B. Robertson, G. Daskalopulos,and M. F. Dixon. 1997. A standardized mouse model of Helicobacter pyloriinfection: introducing the Sydney strain. Gastroenterology 112:1386–1397.

18. Legler, D. F., M. Loetscher, R. S. Roos, I. Clark-Lewis, M. Baggiolini, and B.Moser. 1998. B cell-attracting chemokine 1, a human CXC chemokine ex-pressed in lymphoid tissues, selectively attracts B lymphocytes via BLR1/CXCR5. J. Exp. Med. 187:655–660.

19. Matsumoto, M., S. Marianthasan, H. M. Nahm, F. Baranyay, J. J. Peschon,and D. D. Chaplin. 1996. Role of lymphotoxin and the type I TNF receptorin the formation of germinal centers. Science 271:1289–1291.

20. Mazzucchelli, L., A. Blaser, A. Kappeler, P. Schrli, J. A. Laissue, M. Bag-giolini, and M. Uguccioni. 1998. BCA-1 is highly expressed in Helicobacterpylori-induced mucosa-associated lymphoid tissue and gastric lymphoma. J.Exp. Med. 104:R49–R54.

21. Mohammadi, M., R. Redkine, J. Nedrud, and S. Czinn. 1996. Role of thehost in pathogenesis of Helicobacter-associated gastritis: H. felis infection ofinbred and congenic mouse strains. Infect. Immun. 64:238–245.

22. Mohammadi, M., J. Nedrud, R. Redline, N. Lycke, and S. J. Czinn. 1997.Murine CD4 T-cell response to Helicobacter infection: TH1 cells enhancegastritis and TH2 cells reduce bacterial load. Gastroenterology 113:1848–1857.

23. Mooney, C., J. Keenan, D. Munster, I. Wilson, R. Allardyce, P. Bagshaw, B.Chapman, and V. Chadwick. 1991. Neutrophil activation by Helicobacterpylori. Gut 32:853–857.

24. Ngo, V. N., H. Korner, M. D. Gunn, K. N. Schmidt, D. S. Riminton, M. D.Cooer, J. L. Browning, J. D. Sedgwick, and J. G. Cyster. 1999. Lymphotoxin�/� and tumor necrosis factor are required for stromal cell expression ofhoming chemokine in B and T cell area of the spleen. J. Exp. Med. 189:403–412.

25. Nishio, A., M. Hosono, Y. Watanabe, M. Sakai, M. Okuma, and T. Masuda.1994. A conserved epitope on H�/K�-adenosine-triphosphatase of parietalcells discerned by a murine gastritogenic T cell clone. Gastroenterology107:1408–1414.

26. Oshima, C., K. Okazaki, Y. Matsushima, M. Sawada, T. Chiba, K. Taka-hashi, H. Hiai, T. Katakai, S. Kasakura, and T. Masuda. 2000. Induction offollicular gastritis following postthymectomy autoimmune gastritis in Heli-cobacter pylori-infected BALB/c mice. Infect. Immun. 68:100–106.

27. Parsonnet, J., G. D. Friedmann, D. P. Vandersteen, Y. Chang, J. H. Vo-gelman, N. Orentreich, and R. K. Sibley. 1991. Helicobacter pylori infectionand risk of gastric carcinoma. N. Engl. J. Med. 325:1127–1131.

28. Pasparakis, M., L. Alexopoulou, V. Episkopou, and G. Kollias. 1996. Im-mune and inflammatory responses in TNF alpha-deficient mice: a criticalrequirement for TNF alpha in the formation of primary B cell follicles,follicular dendritic cell networks and germinal centers, and in the maturationof the humoral immune response. J. Exp. Med. 184:1397–1411.

29. Pasparakis, M., L. Alexopoulou, M. Grell, K. Pfizenmaier, H. Bluethmann,and G. Kollias. 1997. Peyer’s patch organogenesis is intact yet formation ofB lymphocyte follicles is defective in peripheral lymphoid organs of micedeficient for tumor necrosis factor and its 55-kDa receptor. Proc. Natl. Acad.Sci. USA 94:6319–6323.

30. Roark, C. E., M. K. Vollmer, P. A. Campbell, W. K. Born, and R. L. O’Brien.1996. Response of a �� T cell receptor invariant subset during bacterialinfection. J. Immunol. 156:2214–2220.

31. Sakaguchi, S., K. Fukuma, K. Kuribayashi, and T. Masuda. 1985. Organ-specific autoimmune diseases in mice by elimination of T cell subset. 1.Evidence for the active participation of T cells in natural self-tolerance:deficit of a T cell subset as a possible cause of autoimmune disease. J. Exp.Med. 161:72–87.

32. Secord, E. A., L. V. Rizzo, E. W. Barroso, D. T. Umetsu, G. J. Thorbecke, andR. H. DeKruyff. 1996. Reconstitution of germinal center formation in nudemice with Th1 and Th2 clones. Cell. Immunol. 174:173–179.

33. Seifarth, C., K. Deusch, K. Reich, and M. Classen. 1996. Local cellularimmune response in Helicobacter pylori associated type B gastritis—selectiveincrease of CD4� but not of �� T cells in the immune response to H. pyloriantigens. Z. Gastroenterol. 34:215–224.

34. Seifarth, C., A. Funk, K. Reich, I. Dahne, M. Classen, and K. Deusch. 1995.Selective increase of CD4� and CD25� T cells but not of �/� T cells in H.pylori-associated gastritis, p. 931–935. In J. Mestecky (ed.), Advances inmucosal immunology. Plenum Press, Inc., New York, N.Y.

35. Stolte, M., and S. Eidt. 1996. Helicobacter pylori and the evolution of gastritis.Scand. J. Gastroenterol. 214:13–16.

36. Vajdy, M., M. H. Kosco-Vibois, M. Kopf, G. Kohler, and N. Lycke. 1995.Impaired mucosal immune respose in interleukin 4-targeted mice. J. Exp.Med. 181:41–53.

37. Walsh, J. H., and W. L. Peterson. 1995. The treatment of Helicobacter pyloriinfection in the management of peptic ulcer disease. N. Engl. J. Med. 333:984–991.

38. Watanabe, T, M. Tada, H. Nagai, S. Sasaki, and M. Nakao. 1998. Helico-bacter pylori infection induces gastric cancer in Mongolian gerbils. Gastro-enterology 115:1–8.

39. Wotherspoon, A. C., C. Ortiz-Hidalgo, M. R. Falzon, and P. G. Isaacson.1991. Helicobacter pylori-associated gastritis and primary B-cell gastric lym-phoma. Lancet 338:1175–1176.

Editor: R. N. Moore

6754 UCHIDA ET AL. INFECT. IMMUN.

on July 24, 2018 by guesthttp://iai.asm

.org/D

ownloaded from