Previous Article | Next Article ![]()
Infection and Immunity, April 2003, p. 2153-2162, Vol. 71, No. 4
0019-9567/03/$08.00+0 DOI: 10.1128/IAI.71.4.2153-2162.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Gastroenterology & Endoscopic Medicine,1 Department of Pathology, Kyoto University, Sakyo, Kyoto, 606-8507, Japan2
Received 30 September 2002/ Returned for modification 25 November 2002/ Accepted 27 December 2002
|
|
|---|
(lymphoid DC marker) or anti-CD11b (myeloid DC marker) and were examined with a confocal microscope. Expression of macrophage inflammatory protein 3
(MIP-3
), which chemoattracts immature DCs, was analyzed by real-time PCR and immunohistochemistry. Follicular dendritic cells (FDCs) were stained with anti-SKY28 antibodies. In noninfected nTx mice, a few myeloid and lymphoid DCs were observed in the bottom portion of the lamina propria, whereas in H. pylori-infected nTx mice, there was an increased influx of myeloid DCs throughout the lamina propria. FDC staining was also observed in the stomachs of members of the infected group. MIP-3
gene expression was upregulated in the infected nTx group, and the immunohistochemistry analysis revealed MIP-3
-positive epithelial cells. These data suggest that H. pylori infection upregulates MIP-3
gene expression in gastric epithelial cells and induces an influx of myeloid DCs in the lamina propria of the gastric mucosa in nTx mice. Myeloid DCs and FDCs might contribute to the development of gastric secondary lymphoid follicles in H. pylori-infected nTx mice. |
|
|---|
DCs are immune regulatory cells that not only secrete chemokines and cytokines but also present antigens for T cells (39, 42, 43). In the intestine, DCs are constitutively present in the Peyer's patches, lamina propria, and mesenteric lymph nodes and have an essential role in the uptake of luminar bacterial antigens (37). Because normal gastric mucosa has no mucosa-associated lymphoid system, very little is known about the role of DCs in the mucosal immune system of the stomach, especially their role in H. pylori-induced chronic follicular gastritis.
In previous studies (33, 46), it was demonstrated that H. pylori infection of BALB/c mice that were thymectomized 3 days after birth (nTx mice) induced follicular gastritis, which resembled human MALT lymphoma. In this model, cooperative involvement of the Th2-type immune response and the Th1 immune response in the development of gastric secondary follicles induced by H. pylori infection was observed. In the present study, to elucidate the involvement of DCs in the development of the follicular formation induced by H. pylori infection, we investigated the presence and characteristics of DCs in the gastric mucosa using the nTx mouse model. In addition, we examined the expression of various chemokines that might influence DCs in the gastric mucosa of nTx mice.
|
|
|---|
Bacterial infection of nTx mice. H. pylori TN2FG4, isolated from a Japanese patient with a duodenal ulcer, was donated by M. Nakao (Pharmaceutical Research Division, Takeda Chemical Industries, Ltd., Osaka, Japan). It was maintained in blood agar base no. 2 with horse serum (5%, vol/vol) containing 2.5 mg of amphotericin B per liter, 5 mg of trimethoprim per liter, 1,250 IU of polymyxin B per liter, and 10 mg of vancomycin per liter. The plates were incubated in a microaerophilic atmosphere at 37°C for 48 h. The inoculated H. pylori strain, TN2FG4, was CagA and VacA positive, as described elsewhere (49). Both nTx and non-nTx mice that were 6 weeks old were orally infected with 108 H. pylori cells as described previously (33). Infected mice were sacrificed 1, 4, 8, and 12 weeks later. Noninfected mice were sacrificed at the same time. Colonization by H. pylori was confirmed by Giemsa staining, culture analysis, and PCR Southern blot analysis for the urease gene by using DNA extracted from the corpus mucosa. The urease gene was amplified 12 weeks after infection for all of the infected mice (33). After decapitation, the sera and stomachs were immediately frozen and stored until they were used. The abnormal hyperimmune status of the nTx mice was confirmed by measuring the serum levels of autoantibodies against parietal cells by an enzyme-linked immunosorbent assay immediately before infection (6 weeks after the mice were thymectomized), as described previously (33).
Histology. The stomach was removed from each mouse, fixed with 4% phosphate-buffered formaldehyde (pH 7.2), and prepared for histologic examination. The sections were stained with hematoxylin and eosin.
Double immunofluorescence staining for DC.
The immunohistochemistry analysis was performed with serially cut frozen sections by using antibodies (Abs) reactive to DCs (for the pan-DC marker, fluorescein isothiocyanate-anti-CD11c; for the linear DC marker, biotin-anti-CD11b and -CD8
; Pharmingen, San Diego, Calif.), to major histocompatibility complex class II (biotin-anti-Iab; Pharmingen), to costimulatory molecules (biotin-anti-CD80 and -CD86; Pharmingen), and to macrophage inflammatory protein 3
(MIP-3
) (biotin-anti-MIP-3
; R&D Systems, Minneapolis, Minn.).
Briefly, freshly frozen sections were fixed in acetone for 10 min, rinsed in phosphate-buffered saline (PBS) (pH 7.2), and incubated with 10% normal rat, hamster, or mouse serum (as a blocking agent) for 20 min. Sections were incubated for 1 h with biotin-labeled Abs. After incubation, the sections were washed with PBS and then incubated with Texas Red avidin (Vector Laboratories, Inc., Burlingame, Calif.) for 30 min. The sections were washed with PBS and incubated with 10% normal hamster serum (as a blocking agent) for 20 min and incubated for 1 h with the fluorescein isothiocyanate-labeled monoclonal Abs. After incubation, the sections were washed with PBS and then washed with distilled water and mounted on glass slides with Vectashield (Vector Laboratories, Inc.). Sections were viewed with a Zeiss Axioplan/Bio-Rad MRC 1024 confocal laser microscope with a 10x objective. Control sections were exposed to normal rat serum instead of monoclonal Abs. There was no staining in the control sections.
Immunohistochemistry analysis for follicular dendritic cells (FDCs). Frozen sections prepared from each stomach were stained by the avidin-biotin complex method. Anti-mouse SKY28 (12) Ab, kindly provided by N. Imazeki (National Defense Medical College, Tokorozawa, Japan), or isotype-matched control goat immunoglobulin G was used as the primary Ab. Rapid immunohistochemical staining of the primary Ab was performed by using a Vectastain universal quick kit (Vector Laboratories, Inc.) according to the manufacturer's instructions. Localization of the antigens was visualized by incubation with a diaminobenzidine solution, and the antigens were counterstained with methyl green.
Measurement of the number of DC subpopulations. Myeloid DCs and lymphoid DCs were quantified in four different areas of the gastric mucosa (surface epithelium, upper portion of the lamina propria, middle portion of the lamina propria, bottom portion of the lamina propria) of both 14-week-old nTx mice that were not infected with H. pylori and 14-week-old nTx mice that were infected (8 weeks postinfection). The total number of positive cells in five continuous gastric gland units was determined for five different regions in one frozen section. Each group consisted of five animals, and the mean ± standard error of the mean was calculated for each group.
Chemokine and cytokine expression.
Total RNA was extracted from the stomachs of 10 mice in each group by using the single-step guanidium thiocyanate-phenol-chloroform method. The extracted RNA preparations were reverse transcribed with MultiScribe reverse transcriptase (PE Applied Biosystems, Foster City, Calif.). The resultant cDNAs (50 ng/reaction mixture) were analyzed for expression of gamma interferon (IFN-
), interleukin-4 (IL-4), MIP-3
, B-lymphocyte chemoattractant (BLC), thymus- and activation-regulated chemokine (TARC), and eotaxin genes by using the TaqMan PCR assay with an ABI Prism 7700 sequence detection system (Perkin-Elmer, Foster City, Calif.). The reaction mixtures were incubated for 2 min at 50°C, denatured for 10 min at 95°C, and subjected to 40 two-step amplification cycles consisting of annealing and extension at 60°C for 1 min followed by denaturation at 95°C for 15 s.
The primers and probes for cytokines and chemokines designed with the PRIMER-EXPRESS software (PE Applied Biosystems) are shown in Table 1. Reaction mixtures were incubated for 2 min at 50°C, denatured for 10 min at 95°C, and subjected to 45 amplification cycles consisting of annealing and extension at 60°C for 1 min followed by denaturation at 95°C for 15 s. All TaqMan PCR data were captured by using the Sequence Detector software (PE Applied Biosystems). The template concentration in each reaction mixture was determined by comparison with a gene-specific standard curve constructed by using the cDNA of positive control samples and normalized by dividing the number of copies of the target gene per nanogram by the number of copies of a housekeeping gene (glyceraldehyde-3-phosphate dehydrogenase [GAPDH] gene, amplified by using TaqMan rodent GAPDH control reagents; PE Applied Biosystems) per nanogram and was expressed as a percentage (mean ± standard error).
|
View this table: [in a new window] |
TABLE 1. Real-time PCR primers and primers and probes
|
|
|
|---|
![]() View larger version (88K): [in a new window] |
FIG. 1. Confocal microscopic analysis of myeloid and lymphoid DCs in the stomach and hematoxylin and eosin staining of non-nTx (normal) and nTx mice. Two-color staining was performed with 6-µm frozen stomach sections by using Abs against CD11c (green) in combination with anti-CD11b (red) (b, e, h, and k) or anti-CD8 (red) (c, f, i, and l). Sections were prepared from 6-week-old noninfected normal mice (a to c), 6-week-old noninfected nTx mice (d to f), 14-week-old noninfected nTx mice (g to i), and 18-week-old noninfected nTx mice (j to l). DCs positive for either CD11b or CD8 were stained yellow. Sections of the same animals were stained with hematoxylin and eosin (a, d, g, and j) to visualize stomach morphology, and the presence of gastritis in nTx mice was confirmed. There was no apparent gastritis (a) and no DC staining (b and c) in normal mice. The histologic findings for the stomachs of nTx mice were typical of AIG, with a loss of parietal cells, proliferation of pit cells, and moderate to severe lymphocyte infiltration (d, g, and j). In 6-week-old nTx mice, staining of a few myeloid DCs was observed at the bottom of the lamina propria (e), whereas there was no lymphoid DC staining (f). In 14- and 18-week-old nTx mice, some staining of both myeloid and lymphoid DCs was observed in the bottom portion of the lamina propria.
|
For H. pylori-infected nTx mice, the histologic findings were similar to those obtained for noninfected nTx mice for the first 12 weeks after infection (Fig. 2a, d, and g). At 12 weeks after infection, however, follicle formation was observed (Fig. 2j). On the other hand, lymphoid DCs were observed in the bottom portion of the lamina propria and submucosal area (Fig. 2c, f, i, and l) after 1 week of infection; this pattern continued during the observation period, but it was not different from that in noninfected nTx mice (Table 2). In contrast, H. pylori infection remarkably enhanced the influx of myeloid DCs in nTx mice (Fig. 2 and Table 2). At 1 week after infection, myeloid DCs were observed throughout the lamina propria, and the number of these cells increased thereafter (Fig. 2b, e, and h). Twelve weeks after infection, myeloid DCs were induced around the follicles (Fig. 2k). Eight weeks after infection, the number of myeloid DCs in the infected nTx mice was significantly greater than that in noninfected nTx mice, and moreover, the number of myeloid DCs in H. pylori-infected nTx mice was also significantly higher than the number of lymphoid DCs in the lamina propria of the same mice except for the bottom portion (Table 2). In the stomachs of non-H. pylori-infected nTx mice, there was not a significant difference between the number of myeloid DCs and the number of lymphoid DCs (Table 2). There was no correlation between the number of myeloid DCs and H. pylori colonization (data not shown).
![]() View larger version (91K): [in a new window] |
FIG. 2. Confocal microscopic analysis of myeloid and lymphoid DCs in the stomach and hematoxylin and eosin staining of H. pylori-infected nTx mice. The histologic findings for the stomachs from H. pylori-infected nTx mice were mild to severe gastritis (a, d, and g [1, 4, and 8 weeks after infection, respectively]). At 12 weeks after infection (18-week-old mice), there was follicle formation (j). Staining of lymphoid DCs (CD11c and CD8 double positive; yellow) was observed in the bottom portion of the lamina propria and submucosal area throughout the experiment (c, f, i, and l). In contrast, there were myeloid DCs (CD11c and CD11b double positive; yellow) throughout the lamina propria, and the number increased linearly during the observation period (b, e, and h). In H. pylori-infected nTx mice at 12 weeks after infection (18 weeks old), myeloid DCs were induced around the follicle (k).
|
|
View this table: [in a new window] |
TABLE 2. Numbers of DCs in the gastric mucosa of nTx mice that were infected or not infected with H. pylori
|
![]() View larger version (20K): [in a new window] |
FIG. 3. Confocal microscopic analysis of activated DCs in the stomachs of H. pylori-infected and noninfected nTx mice (12 weeks after infection, 18 weeks old). Two-color staining was performed with sections by using Abs against CD11c (green) in combination with anti-IAb (red) (a and d), anti-CD80 (red) (b and e), or anti-CD86 (red) (c and f). In the stomachs of noninfected nTx mice, there was no staining of double-positive cells (a, b, and c). In contrast, there were a few double-positive cells in the bottom portions of the lamina propria in the stomachs of H. pylori-infected nTx mice (yellow) (d, e, and f) (the small boxes were magnified to produce the large boxes).
|
gene expression (0.3 ± 0.1 versus 3.1 ± 0.9 copies/GAPDH gene copy; P < 0.05) compared with non-nTx normal mice, there was no significant difference between the levels of IFN-
gene expression in the H. pylori-infected and noninfected nTx mouse groups (3.2 ± 0.8 versus 3.1 ± 0.9 copies/GAPDH gene copy) (Fig. 4a). On the other hand, IL-4 gene expression was not affected by the nTx procedure in noninfected mice (0.03 ± 0.01 versus 0.1 ± 0.1 copies/GAPDH gene copy). IL-4 gene expression, however, was significantly enhanced by H. pylori infection in the nTx mice (0.8 ± 0.1 versus 0.1 ± 0.1 copies/GAPDH gene copy; P < 0.01) (Fig. 4b).
![]() View larger version (30K): [in a new window] |
FIG. 4. Profiles of cytokine and chemokine mRNA expression in the gastric mucosa of non-H. pylori-infected normal mice (non-nTx, 14 weeks old), non-H. pylori-infected nTx mice (14 weeks old), and H. pylori-infected nTx mice (8 weeks after infection, 14 weeks old), as determined by real-time PCR. Ten mice from each group were analyzed. The data are expressed as the means ± standard errors for the number of copies of the mRNA per GAPDH gene copy. One asterisk and two asterisks indicate that the data were significantly different at P values of <0.05 and <0.01, respectively. NS, not significant; Hp, H. pylori infected.
|
Analysis of MIP-3
expression.
We analyzed expression of the gene encoding MIP-3
, which chemoattracts immature DCs (2, 22), in the gastric mucosa of non-H. pylori-infected non-nTx mice (control, 14 weeks old), non-H. pylori-infected nTx mice (14 weeks old), and H. pylori-infected nTx mice (8 weeks postinfection, 14 weeks old). Although nTx alone did not have an effect, H. pylori infection significantly enhanced MIP-3
mRNA expression in nTx mice (4.0 ± 1.4 versus 0.06 ± 0.03 copies/GAPDH gene copy; P < 0.05) (Fig. 4e).
In the gastric mucosa of control mice, there was no staining of DCs or MIP-3
14 weeks after birth (Fig. 5a). MIP-3
staining, however, was observed after H. pylori infection in the gastric mucosa of nTx mice, as well as non-nTx mice (Fig. 5b and d). On the other hand, in the stomachs of non-H. pylori-infected nTx mice, only DCs were stained (Fig. 5c).
![]() View larger version (122K): [in a new window] |
FIG. 5. Confocal microscopic analysis of DCs and MIP-3 staining in the gastric mucosa of non-H. pylori-infected normal mice (non-nTx) (a), H. pylori-infected normal mice (non-nTx, 14 weeks old) (b), non-H. pylori-infected nTx mice (14 weeks old) (c), and H. pylori-infected nTx mice (8 weeks after infection, 14 weeks old) (d). Two-color staining of the sections was performed with Ab against CD11c (green) in combination with anti-MIP-3 (red).
|
![]() View larger version (71K): [in a new window] |
FIG. 6. Immunohistochemical staining of FDCs in the gastric mucosa of nTx mice (original magnification, x400). Gastric tissues were incubated with isotype-matched control goat immunoglobulin G (a) and anti-mouse SKY28 Ab (b and c). Positive cells (arrows) were observed in the stomachs of H. pylori-infected nTx mice (8 weeks after infection, 14 weeks old) (c), whereas there were no positive cells in the stomachs of noninfected nTx mice (14 weeks old) (b).
|
|
|
|---|
Different DC subsets, such as lymphoid and myeloid DCs, induce distinct immune responses (Th1 and Th2) (30). In mice, freshly isolated CD8
+ (lymphoid) and CD8
- (myeloid) DCs from spleens (25, 36) or Peyer's patches (13) induce Th1 and Th2 responses, respectively. CD8
+ DCs secrete IL-12, an essential cytokine for inducing the Th1-type immune response, whereas CD8
- DCs secrete IL-4 and IL-10, which elicit Th2 immunity (14, 25, 35, 41). Neonatal thymectomy induces AIG in BALB/c mice, and Th1 cells predominantly infiltrate the gastric mucosa, while Th2 cells reside mainly in the regional lymph nodes and not in the lamina propria in nTx mice (17). In contrast, it was demonstrated previously that secondary lymph follicle formation induced by H. pylori infection in nTx mice was associated with activation of both the Th1 and Th2 immune responses (46). Moreover, we demonstrated in the present study that in addition to the level of IL-4 gene expression, the levels of expression of the genes encoding the CC chemokines eotaxin and TARC, both of which evoke chemotaxis of the Th2-type CD4+ T lymphocytes (22, 39), were significantly increased in the stomachs of H. pylori-infected nTx mice. These data suggest that myeloid DCs have an important role in the induction of the Th2 response by H. pylori infection in nTx mouse stomachs.
On the other hand, Helicobacter infection upregulates IFN-
expression but not IL-4 expression in the stomachs of C57BL/6 mice (29). The differences between mouse strains might be due to differences in the genetic backgrounds of the Th1-dominated C57BL/6 mice and the Th2-dominated BALB/c mice (10).
MIP-3
, which chemoattracts CCR6-expressing immature DCs (2, 22), is abundantly expressed in mouse and human inflammatory enteric mucosa (15, 45). Iwasaki and Kelsall (13) recently demonstrated that in Peyer's patches a subset of CD11b+ myeloid DCs that is positive for CCR6 lines the epithelial layer of the dome and is the only DC population that responds to MIP-3
. Therefore, we examined whether MIP-3
is involved in the preferential increase in the number of myeloid DCs in the gastric mucosa in our mouse model, and our immunohistochemistry analysis revealed that MIP-3
was present in the epithelial cells of the stomachs of H. pylori-infected nTx mice, whereas there was no MIP-3
in non-H. pylori-infected nTx or non-nTx mice. These findings indicate that H. pylori infection is responsible for MIP-3
induction in the epithelial cells and that MIP-3
expressed in gastric epithelial cells might have an important role in the migration of myeloid DCs into the upper portion of the lamina propria. MIP-3
expression was observed not only in nTx mice but also in non-nTx mice that were infected with H. pylori, in which we did not observe significant infiltration of myeloid DCs into the lamina propria or secondary lymph follicles. These results suggest that MIP-3
is necessary, but not sufficient, for the preferential infiltration of myeloid DCs into the lamina propria with resulting follicular formation induced by H. pylori infection and that, in addition to MIP-3
, other factors are required for such responses to H. pylori infection in nTx mice. In human tonsils, MCP-4 molecules expressed on basal epithelial cells lining the blood vessels have an important role in mobilizing circulatory DCs into the inflammatory site of a tonsil (47). Although similar molecules have not been identified yet in mice, H. pylori infection might induce an unknown chemokine similar to MCP-4, which chemoattracts myeloid DCs from the bloodstream to the inflamed gastric mucosa.
In the present study, activated DCs were present in the lamina propria of H. pylori-infected nTx mice, whereas there were no such cells in noninfected nTx mice. These findings suggest that DCs were activated by H. pylori infection. It should be noted that the activated DCs were localized in the bottom portion of the lamina propria. Thus, the activated DCs might migrate to the regional lymph nodes. Another interesting observation is that localization of the activated DCs were superimposed on lymphoid DCs. Recently, Martinez del Hoyo et al. (26) reported that lymphoid and myeloid DCs represent different maturation or differentiation stages of the same DC population and that lymphoid DCs represent the late stage of DC differentiation. Therefore, although the number of lymphoid DCs was not increased by H. pylori infection, an interesting question is whether the lymphoid DCs distributed over the bottom portion of the lamina propria in H. pylori-infected nTx mice are mature DCs.
In previous studies (33, 46), increases in the number of B cells were observed along with the formation of germinal centers induced by H. pylori infection in nTx mice. One candidate factor that could attract B cells to follicles is the potent chemoattractant BLC, which is constitutively expressed in lymphoid tissues (9, 21). Mice lacking CXCR5, the BLC receptor, exhibit a defect in the development of B-cell follicles in the spleen, Peyer's patches, and inguinal lymph nodes (6). Thus, a BLC-CXCR5 interaction might have an important role in homing B cells into secondary lymphoid organs. The present study demonstrated that expression of the BLC gene was significantly greater in the stomachs of H. pylori-infected nTx mice than in the stomachs of noninfected nTx mice. In addition, we confirmed that FDCs, which secrete BLC (9, 28), appeared in the stomachs of H. pylori-infected nTx mice but not in the stomachs of noninfected nTx mice. These findings suggest that BLC has a critical role in maintaining secondary lymphoid follicles.
In summary, the present study demonstrated that H. pylori infection induces MIP-3
expression in gastric epithelial cells and promotes preferential migration of myeloid DCs into the lamina propria in nTx mice. This effect, coordinated with FDCs, might promote Th2 immune responses, which might then facilitate the development of secondary lymphoid follicles in H. pylori-infected gastric mucosa.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»