| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Division of Comparative Medicine, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 [J. G. F., B. J. S., C. A. D., M. T. W., M. I.], and University of Massachusetts Medical Center, Worcester, Massachusetts 01605 [T. C. W.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
IFN, interleukin (IL)-1, IL-10, and IL-4 was quantified by PCR. Sera were also evaluated for H. felis antibody by ELISA. Antral inflammation increased significantly with time in infected mice. There was a significant, protective effect on the development of preneoplastic fundic lesions and invasive carcinoma attributable to the deletion of one p53 allele (P < 0.05). Submucosal invasive foci were observed in 9 of 11 WT-infected mice ranging from 13 to 15 months p.i.; invasion of adjacent submucosal blood vessels by glandular epithelia also was present in 5 of these mice. None of these lesions were observed in 33 p53+/- mice, infected or not, at any time p.i. p53+/- mice had significantly higher helicobacter colonization consistent with a Th2 host response. In sera from WT mice, IgG2a, considered a proinflammatory Th1 response, continued to rise throughout the 15-month study (P < 0.004). In contrast, IgG2a levels of the p53+/- mice were 50–60% lower than those of the WT mice at each time point (P range, <0.012 to 0.002) and did not progress in magnitude between 12 and 15 months of chronic H. felis infection (P = 0.167). mRNA levels for IFN and IL-1 were significantly up-regulated in WT mice infected with H. felis (P < 0.05) but were slightly elevated or were at background levels in p53+/- mice. IL-10 and IL-4 mRNA expression was not significantly different from control samples. Our results support the hypothesis that germ-line deletion of one p53 allele results in a down-regulated Th1 response to gastric helicobacter infection, possibly because of T-cell senescence, which may indirectly protect against the development of gastric cancer and other epithelial-derived neoplasms associated with chronic inflammation. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gastric cancer is among the leading causes of cancer-related deaths in the world. Although declining in the United States, gastric cancer remains the second leading cause of cancer-related mortality and the 14th leading cause of death worldwide. As is the case for many other tumors, p53 appears to be the most commonly mutated tumor suppressor gene in gastric cancer. p53 mutations have been detected in over 60% of advanced gastric cancers, as well as in intestinal metaplasia (38%), gastric dysplasia (60%), and early gastric cancer; and these targeted mutations are associated with a generally worse prognosis (2, 3, 4, 5, 6, 7) .
Although p53 mutations are common in most sporadic gastric cancers, gastric cancer does not seem to be significantly increased in patients with germ-line p53 mutations (Li-Fraumeni syndrome). Most large studies have noted only a limited spectrum of tumors in patients with germ-line p53 mutations. Excess incidence of tumors in these patients are generally confined to six cancer types, breast carcinomas, soft tissue sarcomas, osteosarcomas, leukemia, brain tumors, and adrenocortical carcinomas, whereas additional component tumors have failed to be implicated (8) . Given the likely importance of p53 mutations in gastric cancer progression, the absence of gastric tumors as part of the Li-Fraumeni syndrome remains puzzling.
Unlike most other tumors, gastric cancer is characterized by a strong association with chronic Helicobacter pylori infection. This bacterial pathogen induces gastric cancer, primarily by inducing chronic, persistent inflammatory response that accelerates remodeling of the gastric epithelium and glandular loss (gastric atrophy) followed by metaplasia, dysplasia, and progression to gastric cancer (9 , 10) . The development of neoplasia has in fact been associated with a number of chronic inflammatory conditions, although the precise relationship between inflammation and tumor development remains largely obscure at the molecular level.
In a previous study, we infected WT4 and p53+/- hemizygous knockout mice with Helicobacter felis and followed the mice for up to 1 year postinoculation (11) . These studies demonstrated that H. felis infection of p53+/- mice resulted in increased but not statistically significant gastric epithelial proliferation (bromodeoxyuridine labeling) when compared with infected WT control mice. Similar degrees of inflammation and colonization were observed in the two helicobacter-infected groups, and after 1 year of observation, progression to gastric cancer was not found (11) . However, the possibility that neoplasia might develop after a longer observation period could not be excluded.
To test this possibility, we undertook a 15-month study of H. felis-infected p53+/- mice. Surprisingly, this study revealed that p53 mutations seem to protect against the development of atrophy and cancer. This protection appears to be associated with down-regulation of immune responsiveness, particularly Th1-associated pathology secondary to H. felis infection.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bacteria.
H. felis (ATCC 49179) was used for oral inoculation as described previously (12)
. The organism was grown for 48 h at 37°C under microaerobic conditions on 5% lysed horse blood agar. The bacteria were harvested and inoculated (at a titer of 1010 organisms per ml) into brain heart infusion broth with 30% glycerol added. The bacterial suspension was frozen at -70°C. Prior to use, aliquots were thawed, analyzed for motility, and cultured for evidence of aerobic or anaerobic bacterial contamination.
Experimental Infection.
Of the 46 TSG-p53 mice, 33 (15 male, 18 female) were inoculated with H. felis and 13 (6 male and 7 female) served as controls. Of the 30 wild-type C57BL/6 mice, 20 (10 male, 10 female) were inoculated with H. felis and 10 (5 male and 5 female) were controls. Brain heart infusion broth containing 
1010 colony-forming units of H. felis per ml was used as inoculum. The inocula (0.5 ml) were delivered by gastric intubation into each test mouse three times at 2-day intervals by using a sterile oral catheter. At 8–11.5 and 12–15 months p.i., H. felis-infected p53 and wild-type mice and uninfected control mice were killed with CO2 and necropsied (Table 1)
.
|
Statistical Evaluation.
Histological scores were statistically analyzed using nonparametric Wilcoxin rank-sum (P < 0.05) testing to determine significant differences among groups. ELISA values were analyzed using Student’s t tests. Correlation between the humoral immune response and the assessment of lesions and number of H. felis organisms was performed using a Bonferroni correction.
ELISA for Anti-H. felis IgG2a and IgG1 in Serum.
Sera were collected from all of the mice prior to dosing with H. felis and then at 2 and 4 months p.i. and at necropsy. An outer membrane antigen preparation of H. felis was obtained by methods previously described for preparing Helicobacter hepaticus antigen (16)
. Briefly, H. felis was cultured in trypticase soy broth containing 5% fetal bovine serum for 48 h under microaerobic conditions as detailed above. After three washes in PBS and examination for bacterial contaminants using Gram’s stain and phase microscopy, the pellet was resuspended in 4 ml of 1% N-octyl-ß-glucopyranoside (Sigma Chemical Co.) for 30 min at room temperature. Insoluble material was removed by ultracentrifugation at 100,000 x g for 1 h. After dialysis against PBS for 24 h at 4°C, supernatant protein concentration was measured by the Lowry technique (Sigma Chemical Co.). For serum IgG isotype measurement, 96-well plates were coated with 100 µl per well of 10 µg/ml H. felis protein in carbonate buffer (pH 9.6) overnight at 4°C. Biotinylated secondary antibodies were monoclonal rat antimouse antibodies produced by clones G1-6.5 and R19-15 (PharMingen, San Diego, CA) for detecting IgG1 and IgG2a, respectively. Incubation with extravidin peroxidase (Sigma Chemical Co.) was followed by 2,2'-azino-di-(3-ethylbenzthiazoline-6-sulfonate substrate (Kirkegaard and Perry Laboratories, Gaithersburg, MD) for color development. Absorbance (A) development at 405
was recorded by an ELISA plate reader (Dynatech MR7000; Dynatech Laboratories, Inc., Chantilly, VA). Serum IgG1 and IgG2a results are reported as absorbance values at a sample dilution of 1:100.
Cytokine mRNA Expression in Gastric Samples.
mRNA was processed from gastric tissue using the SNAP Total RNA Isolation kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Samples were obtained from 11-to-15-month-old control C57BL/6 (n = 4) and control p53+/- mice (n = 5) and from age-matched C57BL/6 (n = 7) and p53+/- mice (n = 8) infected with H. felis. mRNA expression for IFN, IL-1, IL-4, IL-10, and GAPDH were determined by real-time quantitative PCR [PE Applied Biosystems Sequence Detection system (Model 7700); Applied Biosystems, Foster City, CA] using proprietary primer and probe kits as supplied by Applied Biosystems. Data were first normalized to GAPDH mRNA expression and then reported as percentage of increase in cytokine gene expression in comparison with control samples from uninfected mice using derivative calculations as described by the equipment manufacturer (Applied Biosystems).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Atrophy and lymphoid follicle development scores for the fundus of uninfected mice of both strains were zero.
|
. There were no significant differences in antral lymphoplasmacytic infiltration or lymphoid follicle development between WT and p53+/- mice at the early time point. At the later time point, WT mice had significantly greater lymphoplasmacytic inflammation than did the p53+/- mice (P = 0.0469). There were also no significant differences between the infected groups in granulocytic infiltrates within the fundus or antrum at either time point.
|
). Infected mice of both genotypes experienced fundic hyperplasia; however, C57BL/6 mice consistently demonstrated extensive areas of remarkably abnormal mucosa that were not observed in p53+/- mice. Areas with the most severe hyperplasia and metaplastic alterations in C57BL/6 mice often formed multiple cytic chambers with extension into the submucosa (Fig. 2I)
. The degree of loss of the normal mucosal architecture and cellular atypia associated with these areas was suggestive of invasive carcinoma; however, diverticulosis or gastritis cystica profunda of noncarcinomatous lesions could not be ruled out based on morphology alone. Five of the nine cases with invasion into the submucosal stroma also demonstrated well-demarcated intravascular cellular aggregates within adjacent submucosal vessels (Fig. 2G)
. The epithelial character of the intravascular cellular aggregates was verified by positive immunohistochemical identification using a pancytokeratin antibody (Fig. 2H)
, which confirmed that these foci represented extension of the abnormal mucosal tissue into the adjacent vasculature. p53+/- mice did not develop preneoplastic lesions or lesions characterized by invasion into the submucosal stroma or vasculature.
Colonization of the pylorus and fundus at 8–11.5 months p.i. was similar for both WT and p53+/- mice. At 12–15 months p.i., however, the difference in colonization of H. felis within pyloric glands was significant (P < 0.019) with greater colonization in the p53+/- mice (Fig. 2, E and F)
. Colonization of the fundic glands (P = 0.0480) also differed significantly at this later time point.
Serum IgG2a and IgG1 Humoral Responses to H. felis.
Anti-H. felis serum IgG2a and IgG1 from infected mice significantly increased to above control levels by 2 months p.i. (P < 0.006; Fig. 3
). In sera from WT C57BL/6 mice, the IgG2a response, which is associated with a proinflammatory Th1 immune response in mice, continued to rise through 11–15 months p.i. (P < 0.004). In contrast, the IgG2a response of the p53+/- mice was 50–60% lower than that of the infected WT mice at each time point (P range, <0.010–0.002) and did not progress in magnitude between 2 months and 11–15 months of chronic H. felis infection (P = 0.16; Fig. 3A
). The IgG1 response, associated with anti-inflammatory Th2 immune responses in mice, was of the same magnitude in WT and p53+/- at 2 months p.i. but was 40% lower in p53+/- mice at 4 months (P < 0.02) and 50% lower at necropsy at 11–15 months p.i. (P < 0.003; Fig. 3B
).
|
Cytokine mRNA Expression in Gastric Samples.
Proinflammatory IFN and IL-1 mRNA expression in H. felis-infected C57BL/6 mice were both significantly higher than in p53+/- H. felis-infected mice (P < 0.05). IFN mRNA levels in H. felis-infected C57BL/6 mice were 29-fold higher (95% CI, 27.8–30.2) compared with uninfected controls and were 4.4-fold higher (95% CI, 3.5–5.2) in p53+/- H. felis-infected mice compared with uninfected controls. IL-1 mRNA expression levels in H. felis-infected C57BL/6 were 3.5-fold higher (95% CI, 3.0–4.1) and were at background levels in p53+/- H. felis-infected mice (0.72 fold; 95% CI, 0.2–1.3) when compared with control samples. mRNA expression for the Th2-associated anti-inflammatory cytokines of IL-4 and IL-10 were only slightly elevated or were unchanged, respectively, in H. felis-infected C57BL/6 or p53+/- mice compared with controls, without statistical significance (data not shown).
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A number of observations support the current paradigm that H. pylori-mediated progression to gastric atrophy and cancer depends to a large extent on a strong proinflammatory Th1 immune response. In mouse models, chronic infection with H. felis in the C57BL/6 background results in an intense Th1 response, as demonstrated using serology for the IgG2a humoral response and reverse transcription-PCR analysis of tissue cytokines, and is associated with marked atrophy of the glandular epithelium (13) . In the present study, we show that such atrophy progresses in 80% of infected C57BL/6 mice to invasive lesions with penetration into the submucosa and, in selected cases, with invasion into endothelial lined vessels.
The literature suggests that total IgG systemic antibody responses to gastric helicobacter infections are not predictive of the severity of gastric lesions. However, in our study, the relationship between the Th1 response to H. felis infection and the severity of gastritis was supported by the positive correlation between the Th1-associated IgG2a response and the extent of mononuclear inflammatory infiltrate and atrophy in the fundus. In addition, the IgG2a response was significantly negatively correlated with the number of colonizing H. felis, as reported previously (13 , 18) .
Infected WT mice had a high IFN response, which is known to promote MHC Class I (endogenous) and Class II (exogenous) antigen presentation. Thus, increased IFN promotes chronic inflammation in the WT mice in response to H. felis infection with an associated higher risk of genetic mutations inherent in the hyperplastic response. Increased antigen presentation of helicobacter antigens (Class II) and self-antigens (Class I) would help promote tissue damage and secondary DNA damage from inflammatory mediators (19)
. This Th1 response to H. felis becomes polarized toward a Th2 response when mice are concurrently infected with the murine nematode Heligomosomoides polygyrus. This Th2 polarization is associated with a marked reduction in gastric atrophy despite high H. felis colonization (13)
. There is a general correlation in selected inbred mouse strains between Th1 responses [IL-1, IFN, tumor necrosis factor (TNF)] and progression to gastric atrophy in response to helicobacter infection. In support of this Th1/gastric cancer paradigm, recent reports by El-Omar et al. (20)
have indicated that the development of atrophy and achlorhydria in human patients is associated with an IL-1ß gene polymorphism that would be expected to result in higher levels of IL-1ß production.
The finding of attenuated gastric pathology secondary to H. felis infection in the p53+/- may in part be explained by accelerated aging of the immune system in association with the altered p53 status. Induction of p53 expression has been reported to be impaired in activated T cells from elderly individuals (21) . An age-related decline in immune function has been associated with an accumulation of memory phenotype T cells, potentially resulting in a Th1/Th2 imbalance. Age-related alterations in production of cytokines include higher IL-4 and IL-10, with lower IL-2 production in aged WT mice (22, 23, 24) . This Th2 shift, which is associated with senescence, appears to be even more exaggerated in p53 mutant mice. A recent study reported that the accumulation of memory T cells was spontaneously accelerated in p53-/- mice (25) . In addition, measurement of induced cytokine production in p53-/- mice showed higher expression of IL-4, IL-6, and IL-10 after stimulation of the T-cell receptor and in response to antigenic stimulation, whereas expression of IL-2 remained unaltered (25) . We did not use p53-/- mice in this study because of their limited life span associated with a high incidence of spontaneous tumors, particularly lymphoma (26) .
Although a functional decline in activity of the immune system observed with aging might confer an increased risk for certain forms of tumor because of a lack of immune surveillance, our data suggest that immune senescence may actually protect against progression to gastric atrophy and cancer when the lesions are secondary to Helicobacter spp. infection. In addition, our data support the hypothesis that cells with a deleted p53 allele may have very different consequences depending on the cell type carrying such mutations. The finding of increased proliferative index at earlier time points in the infected p53+/- mice supports the hypothesis that p53-targeted disruptions in the glandular epithelial compartment may predispose to gastric adenocarcinoma. However, the altered pattern of inflammation and lack of invasive gastric lesions in these animals are consistent with a protective role conferred by a p53 gene deletion in the lymphocyte population. Thus, with respect to the Li-Fraumeni syndrome, in which p53 mutations are germ-line and present in all cell types, the notion of an "immune phenotype" might explain why certain types of tumors, such as gastric cancer, are not clearly found in excess in such individuals. Our data supports the hypothesis that germ-line p53 mutations do not accelerate tumorigenesis in those tissues in which neoplastic progression is heavily dependent on Th1 inflammatory responses to helicobacter infection and may, in fact, be protective against such tumors.
The association between germ-line mutations in tumor suppressor genes and altered immune responses, noted for p53, may, in fact, represent a more general paradigm that applies to other tumor suppressor genes. For example, in a recent study, we showed that Apc 1638 mice with a a targeted disruption of the Apc allele had a similar decrease in epithelial proliferation, immune responsiveness, and inflammation compared with WT mice (18) . The finding of higher bacterial and urease scores in the Apc 1638 mice observed at earlier (4.5–7.5 months) time points also suggested a Th2 shift (13 , 18) . The serum IgG levels of H. felis in APC+/- mice were diminished when compared with those in WT mice as was the Th1 immune response (IgG2a) (18) . Thus, alterations in immunological responses to bacterial pathogens affecting the gastric mucosa may be a general phenoenon associated with germ-line mutations in tumor suppressor genes that warrants further study.
In conclusion, our results suggest two distinct roles for p53 within the gastric epithelium. Deletion of one p53 allele leads initially to increased proliferation within the epithelial compartment, which by itself might be associated with a heightened cancer risk. However, also significantly over time, p53 mutations, presumably within the lymphocyte compartment, result in immune senescence and a decreased immune responsiveness, particularly in the progression of gastric lesions. The decrease in inflammation leads to the slowing of progression to gastric atrophy and a decreased progression to cancer. However, further validation of this hypothesis will require segregating the influences of the p53 genotype on the gastric epithelial and immune cell populations, either through application of immune cell reconstitution experiments or targeted gene knockout studies. Although the helicobacter-associated gastric tumors that we have described in this mouse model do not fulfill all of the histological criteria of gastric carcinoma in humans, such as penetration of carcinoma into the muscular layers and serosa with metastasis to regional lymph nodes, our findings will provide a useful in vivo model to study mechanisms involved in gastric carcinogenesis.
|
FOOTNOTES |
|---|
1 Supported by NIH Grants R01 A137750, T32 RR07036, and P01 CA26723. 
2 To whom requests for reprints should be addressed, at Division of Comparative Medicine, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Building 16, Room 825C, Cambridge, MA 02139. Phone: (617) 253-1757; Fax: (617) 258-5728; E-mail: jgfox{at}mit.edu
3 Present address: Bristol-Myers Squibb, Lead Safety Assessment, 5 Research Parkway, Wallingford, CT 06492.
4 The abbreviations used are: WT, wild type; p.i., postinfection/postinoculation; IL, interleukin; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; CI, confidence interval.
Received 1/ 8/01. Accepted 12/ 3/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Sayi, E. Kohler, I. Hitzler, I. Arnold, R. Schwendener, H. Rehrauer, and A. Muller The CD4+ T Cell-Mediated IFN-{gamma} Response to Helicobacter Infection Is Essential for Clearance and Determines Gastric Cancer Risk J. Immunol., June 1, 2009; 182(11): 7085 - 7101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Cai, C. Stoicov, H. Li, J. Carlson, M. Whary, J. G. Fox, and J. Houghton Overcoming Fas-Mediated Apoptosis Accelerates Helicobacter-Induced Gastric Cancer in Mice Cancer Res., December 1, 2005; 65(23): 10912 - 10920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Nambiar, S. Kirchain, and J. G. Fox Gastritis-associated Adenocarcinoma and Intestinal Metaplasia in a Syrian Hamster Naturally Infected with Helicobacter Species Vet. Pathol., May 1, 2005; 42(3): 386 - 390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ihrig, M. T. Whary, C. A. Dangler, and J. G. Fox Gastric Helicobacter Infection Induces a Th2 Phenotype but Does Not Elevate Serum Cholesterol in Mice Lacking Inducible Nitric Oxide Synthase Infect. Immun., March 1, 2005; 73(3): 1664 - 1670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Houghton, C. Stoicov, S. Nomura, A. B. Rogers, J. Carlson, H. Li, X. Cai, J. G. Fox, J. R. Goldenring, and T. C. Wang Gastric Cancer Originating from Bone Marrow-Derived Cells Science, November 26, 2004; 306(5701): 1568 - 1571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Rad, A Dossumbekova, B Neu, R Lang, S Bauer, D Saur, M Gerhard, and C Prinz Cytokine gene polymorphisms influence mucosal cytokine expression, gastric inflammation, and host specific colonisation during Helicobacter pylori infection Gut, August 1, 2004; 53(8): 1082 - 1089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Rogers and J. G. Fox Inflammation and Cancer I. Rodent models of infectious gastrointestinal and liver cancer Am J Physiol Gastrointest Liver Physiol, March 1, 2004; 286(3): G361 - G366. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Eguchi, N. Herschenhous, N. Kuzushita, and S. F. Moss Helicobacter pylori Increases Proteasome-mediated Degradation of p27kip1 in Gastric Epithelial Cells Cancer Res., August 1, 2003; 63(15): 4739 - 4746. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Erdman, T. Poutahidis, M. Tomczak, A. B. Rogers, K. Cormier, B. Plank, B. H. Horwitz, and J. G. Fox CD4+ CD25+ Regulatory T Lymphocytes Inhibit Microbially Induced Colon Cancer in Rag2-Deficient Mice Am. J. Pathol., February 1, 2003; 162(2): 691 - 702. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
