High salt diets dose-dependently promote gastric chemical carcinogenesis in Helicobacter pylori-infected Mongolian gerbils associated with a shift in mucin production from glandular to surface mucous cells
Article first published online: 27 APR 2006
Copyright © 2006 Wiley-Liss, Inc.
International Journal of Cancer
Volume 119, Issue 7, pages 1558–1566, 1 October 2006
How to Cite
Kato, S., Tsukamoto, T., Mizoshita, T., Tanaka, H., Kumagai, T., Ota, H., Katsuyama, T., Asaka, M. and Tatematsu, M. (2006), High salt diets dose-dependently promote gastric chemical carcinogenesis in Helicobacter pylori-infected Mongolian gerbils associated with a shift in mucin production from glandular to surface mucous cells. Int. J. Cancer, 119: 1558–1566. doi: 10.1002/ijc.21810
- Issue published online: 18 JUL 2006
- Article first published online: 27 APR 2006
- Manuscript Accepted: 29 MAR 2006
- Manuscript Received: 12 AUG 2005
- Ministry of Health, Labour and Welfare, Japan
- Ministry of Education, Culture, Sports, Science and Technology, Japan
- Japan Society for the Promotion of Science
- Mongolian gerbil;
- Helicobacter pylori;
- surface mucous cell mucin;
- gland mucous cell mucin;
Intake of salt and salty food is known as a risk factor for gastric carcinogenesis. To examine the dose-dependence and the mechanisms underlying enhancing effects, Mongolian gerbils were treated with N-methyl-N-nitrosourea (MNU), Helicobacter pylori and food containing various concentrations of salt, and were sacrificed after 50 weeks. Among gerbils treated with MNU and H. pylori, the incidences of glandular stomach cancers were 15% in the normal diet group and 33%, 36% and 63% in the 2.5%, 5% and 10% NaCl diet groups, showing dose-dependent increase (p < 0.01). Intermittent intragastric injection of saturated NaCl solution, in contrast, did not promote gastric carcinogenesis. In gerbils infected with H. pylori, a high salt diet was associated with elevation of anti-H. pylori antibody titers, serum gastrin levels and inflammatory cell infiltration in a dose-dependent fashion. Ten percent NaCl diet upregulated the amount of surface mucous cell mucin (p < 0.05), suitable for H. pylori colonization, despite no increment of MUC5AC mRNA, while H. pylori infection itself had an opposing effect, stimulating transcription of MUC6 and increasing the amount of gland mucous cell mucin (GMCM). High salt diet, in turn, decreased the amount of GMCM, which acts against H. pylori infection. In conclusion, the present study demonstrated dose-dependent enhancing effects of salt in gastric chemical carcinogenesis in H. pylori-infected Mongolian gerbils associated with alteration of the mucous microenvironment. Reduction of salt intake could thus be one of the most important chemopreventive methods for human gastric carcinogenesis. © 2006 Wiley-Liss, Inc.
Helicobacter pylori was discovered in 1983,1 and much scientific evidence has subsequently accumulated pointing to the bacterium acting as a major causative factor for gastric adenocarcinoma development.2, 3, 4, 5 However, various other environmental or host factors are also known to influence gastric carcinogenesis,4, 6 and salt (sodium chloride, NaCl) and salted foods are probably of particular importance, based on evidence from a large number of case–control and other epidemiological studies in man.7, 8, 9, 10, 11 Experimental animal models have also provided support for the conclusion that salt promotes gastric carcinogenesis in the glandular stomach after chemical carcinogen treatment.12, 13, 14 A high concentration of sodium chloride causes damage to the gastric mucosa, cell death and consequent regenerative cell proliferation,15 in the longer term leading to inflammation and diffuse erosion.16 The induced proliferative changes may enhance the effects of food-derived carcinogens17 but questions remain as to the detailed mechanisms involved.
The Mongolian gerbil can readily be infected with H. pylori, and the resultant chronic active gastritis, peptic ulcers and intestinal metaplasia resemble lesions apparent in man. We have previously demonstrated that a high-salt diet enhances the promoting effects of H. pylori on gastric carcinogenesis, the two factors acting synergistically but bacterial infection having the greater influence.14 In mice, mucosal damage due to salt is known to enhance H. pylori colonization18 and it is very important to clarify any impact on H. pylori infection itself, which might result from changes in the intragastric mucous microenvironment. Gastric mucus consists of two histochemically different kinds of mucin, surface mucous cell mucin (SMCM) and gland mucous cell mucin (GMCM),19, 20 and H. pylori is found only in surface mucous gel layer (SMGL) or attached to surface mucous epithelial cells, and not in the gland mucous gel layer.21H. pylori thus clearly prefers to colonize the SMCM layer, which deteriorates markedly under conditions of chronic gastritis.22 Our hypothesis for testing in the present study was that salt might change the balance between SMCM and GMCM in the stomach. We therefore evaluated the dose-dependent influence of dietary salt on gastric carcinogenesis in H. pylori-infected Mongolian gerbils treated with a chemical carcinogen, focusing on quantities of gastric surface and GMCMs in the glandular stomach.
Materials and methods
Six-week-old, pathogen-free, male Mongolian gerbils (Seac Yoshitomi, Fukuoka, Japan) were housed in steel cages in an air-conditioned biohazard room with a 12 h/12 h light–dark cycle, as previously described.14 They were given free access to autoclaved water and food, CRF-1 (Oriental Yeast Co. Ltd., Tokyo, Japan) containing 0.32 g sodium per 100 g. The experimental design is shown as Figure 1. Gerbils were divided into 20 groups, Groups 1–10 being given a chemical carcinogen, N-methyl-N-nitrosourea (MNU; Sigma Chemical Co., St. Louis, MO) in their drinking water at the concentration of 30 ppm for alternate weeks from weeks 1 to 10 and the remaining animals serving as noncarcinogen-treated controls. At week 11, gerbils in Groups 1–5 and Groups 11–15 were inoculated with H. pylori (ATCC 43504, American Type Culture Collection, Rockville, MD), while gerbils in the other groups were inoculated with Brucella broth. H. pylori was prepared by the same method as described previously.14 From weeks 12 to 50, the animals in groups 2, 7, 12 and 17 received food including 2.5% sodium chloride, and those in groups 3, 8, 13 and 18 were given 5% NaCl, and groups 4, 9, 14 and 19 received 10% NaCl. The other groups were maintained on normal diet. Groups 5, 10, 15 and 20 were administered 4 ml/kg body weight of saturated sodium chloride solution (˜29%) through gastrogavage tube once a week. At week 50, after 24 hr fasting all animals were sacrificed under deep anesthesia, and had their stomachs resected and blood samples collected from the inferior vena cava. The whole glandular stomachs of Groups 1–10 gerbils and the anterior walls of groups 11, 14, 16 and 19 gerbils were taken for histological examination. The posterior walls of groups 11, 14, 16 and 19 gerbils were divided into fundic and pyloric mucosas, homogenized in 1 ml phosphate-buffered saline (PBS) and subjected to extraction of total RNA and gastric mucins.
Stomachs for histological examination were cut along the greater curvature, and fixed in 10% phosphate-buffered formalin, 70% ethanol, Carnoy's solution or 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. After fixation, tissues were embedded in paraffin, and sectioned (3 μm thickness) for hematoxylin and eosin (H&E), Alcian blue-periodic acid Schiff and galactose-oxidase cold thionine Schiff/paradoxical concanavalin A (GOTS-PCS) (after Ota et al.19) staining. The status of gastritis was graded using criteria for neutrophils, mononuclear cells, intestinal metaplasia and heterotopic proliferative glands, on a four-point scale: 0 (none), 1 (mild), 2 (moderate) and 3 (marked).
Serological examination and H. pylori culture
Blood samples were centrifuged at 5,800 g for 5 min at 4°C to separate serum for the measurement of titers of anti-H. pylori antibodies and gastrin levels as described earlier.14 To assess the influence of salt on H. pylori colonization, cultures of H. pylori were performed from Groups 11 and 14. One hundred microliters aliquots out of 1 ml homogenates from posterior wall of glandular stomach in PBS were seeded on segregating agar plates for H. pylori (Nissui Pharmacertical, Tokyo, Japan) and incubated at 37°C under high humidity with Anaero Pack Campylo (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan) for 5 days. Then numbers of colony forming units were counted.
Extraction of RNA and reverse transcription
One hundred micorliters aliquots of antrum homogenates were dissolved in ISOGEN (Nippon Gene Co. Ltd, Tokyo, Japan), total RNA was extracted according to the manufacturer's instructions, and 5 μg was applied for synthesis of first strand cDNAs with ThermoScript reverse transcriptase (Invitrogen Corp., Carlsbad, CA, USA), according to the manufacturer's guideline. The remaining homogenized antrum served for extraction of mucins.
Sequencing of MUC5AC and MUC6 cDNA
To obtain partial cDNA sequences for MUC5AC in Mongolian gerbils, the most homologous regions were selected by comparison of the rat (Gene bank accession number, U83139) and mouse (L42292) (blue letters in Fig. 3) sequences. To determine MUC6 cDNA sequences, those of rat (XM_215127) and mouse (NM_181729) were compared for selection of two primer pairs (blue and green primer pairs in Fig. 4). After successful RT-PCR amplification using total RNA from gerbils' stomach as a template, sequencing was performed with a BigDye Terminator ver. 3.1 Cycle Sequencing kit (Applied Biosystems, Foster, CA, USA) using ABI Prism 3100 (Applied Biosystems), according to the manufacturer's instructions.
Quantitative real-time RT-PCR
After determination of gerbil's MUC5AC and MUC6 cDNA sequences, new pairs of PCR primers (underlined sequences in Figs. 3 and 4, respectively) were chosen for quantitative RT-PCR using OLIGO software (Molecular Biology Insights, Inc., Cascade, CO, USA). Real-time quantitative RT-PCR was utilizing the LightCycler system (Roche Diagnostics, Mannheim, Germany) with a QuantiTect SYBR Green PCR kit (QIAGEN, Hilden, Germany) as previously described.23, 24, 25 Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Gene bank accession number AB040445) was used as an internal control (forward primer, 5′-AACGGCACAGTCAAGGCTGAGAACG-3′; reverse primer, 5′-CAACATACTCGGCACCGGCATCG-3′). The amplification schedule was an initial 15 minutes preheating at 95°C followed by 50 rounds of 95°C for 60 sec, 60°C for 60 sec, and 72°C for 60 sec. Second derivative maximum points obtained with LightCycler software ver. 3.5 (Roche Diagnostics) for MUC5AC and MUC6 were normalized to that for GAPDH in each animal. Averages and SDs or SEs were calculated and the values were statistically compared with the Mann–Whitney U-test. Relative values were expressed as percentages of control group 16 data as reported earlier.23, 24, 25
Extraction of gastric mucin
Gastric mucin was collected from the antral posterior wall of the stomach in groups 11, 14, 16 and 19 for extraction by the ethanol precipitation method, according to Kakei et al.26 Briefly, homogenates were mixed with 50 mM Tris–HCl, pH 7.2 and boiled for 3 min. Then, Triton X-100 was added to obtain a final concentration of 2%, followed by incubation at 37°C for 1 hr and centrifugation at 4°C and 10,000 rpm for 30 min. Potassium acetate and ethanol were added to supernatant samples to give the final concentrations of 1% and 70%, respectively. After storage for 1 hr at 0°C, the solutions were centrifuged at 4°C at 10,000g for 20 min and the precipitates were collected. These processes were repeated three times for purification of mucins. Finally, the collected precipitates were dissolved in 50 mM Tris–HCl, pH 7.2 and protein concentrations were measured using a DC protein assay kit (Bio-Rad Laboratories, Hercules, CA).
Quantification of gastric mucin
PVDF membranes (Immobilon-P, Millipore, Bedford, MA) were soaked in 100% methanol for 15 sec, transferred into milliQ water for 2 min, and then set in a Minifold II slot blotter (Schleicher & Schuell, Dassel, Germany). Extracted mucin was applied to the membranes with aspiration followed by drying at 37°C for 1 hr. The prepared PVDF membranes were then stained with GOTS, as described previously,27 or immunostained with anti-mucin monoclonal antibody (MoAb) (HIK1083, Kanto Chemical Co., Inc., Tokyo, Japan), according to the manufacturer's protocol (Millipore Technical Publication on the internet at http://www.millipore.com/publications.nsf/docs/rp562). Quantitative analysis of stained membranes was performed on a Macintosh computer, using the NIH Image program ver. 1.62 (National Institutes of Health, Bethesda, MD). All values were normalized to the protein concentration.
The Fisher's exact test was used to assess the incidence of gastric adenocarcinomas. Mann–Whitney two sample statistics and Kruskal–Wallis equality-of-populations rank tests were applied to establish the significance of differences in body weight distributions, MNU intake, salt intake, titers of anti-H. pylori antibodies, serum gastrin, numbers of colony forming units and gastric mucin. p-Values < 0.05 were considered to be statistically significant.
Total MNU intake, total salt intake per animal and average body weights in each experimental group
Data for total MNU intake, total salt intake per animal and average body weights at week 50 are shown in Table I. There were no significant differences in MNU intake among Groups 1–10. Total salt intake of each group essentially corresponded to the proportion of sodium chloride in their food. The average body weight for the 10% salt diet group was significantly lower than in the normal, 2.5% salt and 5% salt diet groups under the same MNU and H. pylori conditions.
|Groups||Effective number||Body weight (g) (AVE ± SD)||MNU intake (ml) (AVE ± SD)||Total salt intake (g) (AVE ± SD)||Number of carcinomas||Histological types of carcinoma||Incidence (%)||Anti-Hp Ab titer (AI) (AVE ± SD)||Serm gastrin (pg/ml) (AVE ± SD)|
|1||40||88.3 ± 8.1||356.1 ± 23.8||9.7 ± 0.7||6||4||2||0||0||15f||0.541 ± 0.172||342.0 ± 87.0|
|2||24||107.4 ± 7.0||341.1 ± 28.6||36.5 ± 1.6||8||5||3||0||0||33f||0.578 ± 0.167||328.5 ± 140.4|
|3||25||97.8 ± 8.6||328.4 ± 18.3||77.7 ± 4.7||9||8||0||1||0||36f||0.513 ± 0.128||333.7 ± 102.3|
|4||30||75.7 ± 15.9a||329.7 ± 31.3||153.8 ± 24.9||19||17||0||1||1||63e,f||0.486 ± 0.173||402.5 ± 150.2|
|5||21||82.8 ± 11.7||332.6 ± 52.3||12.5 ± 1.1||5||5||0||0||0||24||0.552 ± 0.169||287.7 ± 89.1|
|6||27||95.3 ± 7.7||346.7 ± 56.3||9.7 ± 0.8||0||0||0||0||0||0||0.119 ± 0.094||260.6 ± 91.1|
|7||25||104.0 ± 11.9||344.1 ± 37.0||37.5 ± 2.2||0||0||0||0||0||0||0.090 ± 0.056||192.8 ± 47.0|
|8||25||104.4 ± 7.6||347.5 ± 40.5||80.6 ± 3.7||1||1||0||0||0||4||0.080 ± 0.069||216.8 ± 62.0|
|9||27||85.1 ± 10.3b||339.5 ± 22.1||165.2 ± 17.3||0||0||0||0||0||0||0.068 ± 0.039||216.4 ± 76.2|
|10||20||85.6 ± 7.6||348.9 ± 18.3||13.1 ± 0.7||0||0||0||0||0||0||0.104 ± 0.097||209.8 ± 46.5|
|11||20||89.7 ± 12.8||0||11.8 ± 2.1||0||0||0||0||0||0||0.308 ± 0.116i,j||201.8 ± 85.3|
|12||9||103.0 ± 6.5||0||34.7 ± 1.3||0||0||0||0||0||0||0.359 ± 0.103j||549.9 ± 209.7k|
|13||9||105.8 ± 7.6||0||76.7 ± 2.8||0||0||0||0||0||0||0.413 ± 0.071g,j||408.5 ± 89.5l|
|14||19||69.5 ± 15.3c||0||138.8 ± 24.2||0||0||0||0||0||0||0.528 ± 0.111h,j||454.4 ± 77.7m|
|15||11||84.0 ± 15.4||0||17.2 ± 4.4||0||0||0||0||0||0||0.165 ± 0.051||209.6 ± 35.8|
|16||17||88.1 ± 14.7||0||11.2 ± 1.9||0||0||0||0||0||0||0.060 ± 0.078||253.2 ± 78.3|
|17||7||113.4 ± 7.4||0||37.8 ± 0.3||0||0||0||0||0||0||0.034 ± 0.016||229.0 ± 73.7|
|18||7||97.0 ± 10.9||0||77.7 ± 1.5||0||0||0||0||0||0||0.034 ± 0.013||205.8 ± 47.5|
|19||18||75.6 ± 9.8d||0||172.2 ± 5.2||0||0||0||0||0||0||0.039 ± 0.008||258.0 ± 127.8|
|20||11||88.0 ± 16.2||0||16.1 ± 4.0||0||0||0||0||0||0||0.038 ± 0.020||219.2 ± 29.4|
We did not monitor blood pressure or examine brain and cerebral vessels histopathologically because no gerbils showed neurologically abnormal symptoms in the present experiments.
Incidences of glandular stomach adenocarcinomas
The incidences of tumors are summarized in Table I. The value for adenocarcinomas in Group 4 was significantly higher than that in Group 1 (p < 0.01), while there were no significant differences between the incidences in Groups 1 and 5. The trend for increase in the adenocarcinoma incidence from Groups 1 through 4 was revealed to be statistically significant (p < 0.01). Both differentiated and undifferentiated adenocarcinomas were found in Groups 3 and 4, while only differentiated lesions were observed in Groups 1, 2 and 5 (Fig. 2). In the groups with no H. pylori infection treated with MNU (Groups 6–10), only one glandular stomach adenocarcinoma was detected, in Group 8. In other groups without MNU, no gastric adenocarcinomas were observed. All of the glandular stomach adenocarcinomas generated in the present study developed in the pyloric gland area. The gastric mucosa of most of the H. pylori-infected gerbils was swollen and edematous with many polypoid lesions. Such macroscopic findings were not recognized in the stomachs of non-H. pylori-infected animals.
Status of gastritis
Data for the status of gastritis in each group are summarized in Table II. H. pylori-infected groups showed significantly higher scores for infiltration of neutrophils and mononuclear cells and development of heterotopic proliferative glands than in any noninfected groups. Neutrophils scores in H. pylori with high salt diet groups (Groups 12–14) were significantly higher than in the corresponding normal diet group (Group 11; p < 0.05). Although the H. pylori infection increased the scores for mononuclear cells (p < 0.05), high salt intake caused no significant difference. With regard to intestinal metaplasia, there were no significant differences among the H. pylori-infected groups although lesions were few. There were significant differences in scores for heterotopic proliferative glands between MNU + H. pylori + high salt diet (Groups 2–4) and MNU + H. pylori + normal diet (Group 1; p < 0.01), and between H. pylori + high salt diet (Groups 12–14) and H. pylori plus normal diet (Group 11; p < 0.05).
|Groups||Neutrophils||Mononuclear cells||Intestinal metaplasia||Heterotopic proliferative glands|
|1||1.38 ± 0.74||1.95 ± 0.75||0.28 ± 0.60||1.78 ± 0.80|
|2||1.46 ± 0.78||2.21 ± 0.93||0.29 ± 0.46||2.17 ± 0.82c,f|
|3||1.36 ± 0.76||2.12 ± 0.88||0.40 ± 0.49||1.92 ± 0.81f|
|4||1.25 ± 0.88||2.37 ± 0.96||0.41 ± 0.67||2.09 ± 1.03d,f|
|5||1.19 ± 0.81||2.14 ± 0.85||0 ± 0||1.19 ± 0.98e|
|6||0.04 ± 0.19||0.26 ± 0.45||0.04 ± 0.19||0 ± 0|
|7||0.04 ± 0.20||0.64 ± 0.57||0 ± 0||0 ± 0|
|8||0.04 ± 0.20||0.68 ± 0.63||0 ± 0||0.04 ± 0.20|
|9||0 ± 0||0.37 ± 0.84||0 ± 0||0 ± 0|
|10||0 ± 0||0.35 ± 0.75||0 ± 0||0 ± 0|
|11||0.60 ± 0.68||2.85 ± 0.67||0.10 ± 0.30||1.85 ± 0.99|
|12||1.11 ± 0.60a||2.89 ± 0.33||0.33 ± 0.71||2.44 ± 0.53g|
|13||0.78 ± 0.67a||2.89 ± 0.33||0.22 ± 0.67||2.44 ± 0.53g|
|14||1.16 ± 0.83a,b||2.89 ± 0.32||0.47 ± 1.12||2.47 ± 0.77g,h|
|15||0.18 ± 0.40||2.45 ± 1.04||0 ± 0||0.55 ± 1.04i|
|16||0 ± 0||0.18 ± 0.39||0 ± 0||0 ± 0|
|17||0 ± 0||0.43 ± 0.53||0 ± 0||0 ± 0|
|18||0 ± 0||0.29 ± 0.49||0 ± 0||0 ± 0|
|19||0 ± 0||0.17 ± 0.38||0 ± 0||0 ± 0|
|20||0 ± 0||0.64 ± 0.50||0 ± 0||0 ± 0|
Titers of anti-H. pylori antibodies are summarized in Table I. All H. pylori-infected groups demonstrated significantly higher values than the noninfected groups. MNU and H. pylori treated animals (Groups 1–5) showed no significant differences in the antibody titers. Among the groups treated with H. pylori and various concentrations of salt diets (Groups 11–14), the titers increased along with the salt intake (p < 0.0001). With H. pylori culture, the number of colony forming units with Group 14 was larger than that for Group 11 [7147.2 ± 1750.3 vs 2692.6 ± 978.0 (average ± SE), p < 0.05]. Group 15 was revealed to have a significantly lower titer than that of Group 11 (p < 0.001). The data for serum gastrin are also included in Table 1. Among H. pylori-infected groups, Group 12 (p < 0.001), Group 13 (p < 0.005) and Group 14 (p < 0.0001) showed significant elevation as compared to Group 11. There were also significant differences between Groups 12 and 17 (p < 0.005), Groups 13 and 18 (p < 0.005) and Groups 14 and 19 (p < 0.0001).
Cloning of gerbil MUC5AC
Partial nucleotide sequences of MUC5AC in the Mongolian gerbil, rat and mouse are aligned in Figure 3a. Homology was found to be 77.4% between those of the gerbil and rat. With the mouse, the figure was 78.8%. Comparing amino acid sequences (Fig. 3b), homology was found to be 67.1% between the gerbil and rat, and 72.7% for the gerbil and mouse.
Cloning of gerbil MUC6
Partial nucleotide and amino acid sequences of MUC6 in the gerbil are shown with those of the rat and mouse (Fig. 4). Homologies for nucleotide sequences were 91.2% and 92.0% with the rat and mouse, respectively, and those for peptide sequences were 93.0% and 91.5%, respectively.
Expression of MUC5AC and MUC6 mRNA
Data for transcription levels of MUC5AC and MUC6 are shown in Figures 5a and 5b. The values are expressed relative to the Control level, significant increases in MUC6 mRNA being observed with bacterial infection (p < 0.05).
Observation of the SMGL
Figure 6a shows the normal stomach of a Mongolian gerbil stained with GOTS-PCS, featuring a SMGL consisting of alternate thin laminates of SMCM and pyloric GMCM, the former stained blue and the latter brown. High salt diet increased the blue SMCM (Group 19, Fig. 6b) and H. pylori infection upregulated the brown GMCM (Group 11, Fig. 6c). Both kinds of mucin were found to be increased and SMGL was partially disturbed in Group 14 (Fig. 6d).
Levels of gastric mucins
For quantitative analysis, gastric mucins blotted on membranes were stained with GOTS or with anti-Mucin MoAb and results for SMCM and GMCM are shown in Figures 7a and 7b, respectively. The amount of GOTS-positive SMCM was upregulated to 238.2 ± 40.4% of the control level in gerbils eating diets supplemented with 10% NaCl (Group 19; p < 0.005). H. pylori infection did not affect the production of SMCM (Group 11, 109.8 ± 27.3%) but when H. pylori-infected animals were given the high salt diet, SMCM was again upregulated to 210.4 ± 52.2%, although without statistical significance (Fig. 7a). In contrast, the amount of GMCM was about 6-fold upregulated with H. pylori infection (Group 11, 595.2 ± 104.8%) compared to the control group (Group 16, 100 ± 17.6%; p < 0.0001) with the normal diet and more than 6-fold elevated with the high salt diet (Group 14, 405.8 ± 82.9; Group 19, 60.0 ± 18.9%; p < 0.0001). Considering the effect of the high salt diet, the amount of GMCM in Group 19 was significantly lower than that of the controls (Group 16; p < 0.05). A similar tendency was noted in H. pylori-infected groups (Groups 11 vs. 14, p < 0.05).
Our present study clearly demonstrated that salt promotes gastric carcinogenesis in the H. pylori-infected glandular stomach of Mongolian gerbils in a dose-dependent fashion, in line with our earlier findings.14 In rats, when given alone, NaCl has no apparent carcinogenicity but, when administered with N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) or 4-nitroquinoline 1-oxide (4-NQO), it promotes gastric carcinogenesis in the rat glandular stomach,12 in a dose-dependent fashion.13 In humans, Tsugane et al.16 reported that restriction of salt and salty food may decrease gastric cancer risk in Japan from an epidemiological study. In Japan, foods containing salt at up to 12% are commonly consumed, such as pickled vegetables (salt content: 1–10%) and salted fish roe or fish preserves (6–12%).16 Thus the present experiment can be considered to well mimic the human situation. Taking into account the present and previous data, we considered that salt is a promoter of gastric adenocarcinoma, especially in combination with H. pylori, and that avoidance of excess consumption is important for the prevention of gastric adenocarcinomas.
Tatematsu et al.28 reported that oral administration of porcine gastric mucin suppressed promotion effect of glandular stomach cancer by saturated NaCl solution but did not prevent induction of gastric cancer by MNNG in rats. Exogenous mucin might reinforce the mucosal barrier against damage by salt, possibly by inhibiting toxic factors like pepsin or free acid in the gastric juice. Ota et al.19, 20 demonstrated that gastric mucin consists of two histochemically different kinds, SMCM and GMCM, in human stomach. In gerbils, these two kinds of gastric mucins could be separated histochemically with GOTS-PCS. Hidaka et al.21 reported that H. pylori exists only in the SMGL or attached to surface mucous epithelial cells, but not in the gland mucous gel layer, indicating that GMCM contains some element(s) not conducive to the bacterium. Matsuzwa et al.29 has provided evidence that GMCM is upregulated as a defensive reaction against H. pylori infection and Kawakubo et al.30 showed that the GMCM have terminal α1,4-GlcNAc residues attached to core O-glycans, which inhibit biosynthesis of cholesteryl-α-D-glucopyranoside, a major cell wall component of H. pylori. We here confirmed the amount of GMCM to be increased in H. pylori-infected gerbils and, furthermore, revealed that it was regulated in transcriptional level, although the mechanisms remain to be clarified. In consideration of the available evidence, enhancement of gastric carcinogenesis by salt may involve two pathways. One is via direct effects on mucosal cells resulting in acceleration of cell proliferation following mucosal injury. The other may be indirect by altering the mucous microenvironment and aiding H. pylori colonization and growth. Salt may thus influence the balance of the two types of gastric mucins in addition to the proposed destruction of the mucus barrier and decrease in mucus viscosity facilitating carcinogen access to the mucosa.12, 28 The present slot blot analysis showed that the high salt diet upregulated SMCM significantly in H. pylori-negative animals, and had similar effect in H. pylori-positive gerbils, despite the lack of significant change in MUC5AC, the core-protein of SMCM. At the same time, high salt diet downregulated GMCM regardless of H. pylori infection.
In the present study, there were no statistical differences of the incidences of adenocarcinomas between controls and animals receiving saturated sodium chloride solution (˜29%) through gastrogavage tube once a week. Charnley et al.31 reported that the proportion of S-phase cell was increased from 6 till 24 hr after administration of saturated NaCl solution on the rat gastric mucosa and decreased to the normal level after 48 hr. Furihata et al.15 also documented that gastric tissue damage occurred within one minute of administration of hypertonic NaCl solution and then the number of S-phase cell increased until 17 hr. Thus, although the saturated NaCl solution damaged the gastric mucosa temporarily, the damage would be restored within 1 or 2 days after the administration. We therefore consider that continuous exposure of the gastric mucous cells to high levels of salt could be an important factor for gastric carcinogenesis.
From evaluation of gastritis in the present study, we demonstrated that high salt diet exerts a promoting influence regarding neutrophil infiltration and induction of heterotopic proliferative glands with H. pylori infection. The results for carcinoma incidence and gastritis support the hypothesis that continuous salt intake promotes gastric carcinogenesis because of exacerbation of gastritis caused by the previously referred to change of the gastric mucus environment in the H. pylori-infected stomach.
From the serological results, titers of anti-H. pylori antibodies increased significantly with salt intake among H. pylori-infected groups (Groups 11–15; p < 0.05), in agreement with human findings.32 Fox et al.18 demonstrated that a high salt diet enhances colonization of H. pylori and our results reconfirmed this and point to salt intake enhancing H. pylori infection, and that consequent increase in inflammation is responsible for promotion of gastric carcinogenesis in a dose-dependent fashion. Inversely, Group 15 demonstrated a significantly lower titer than Group 11, so intermittent administration of concentrated NaCl solution may hinder H. pylori growth. On the other hand, there was no significant difference in titers of anti-H. pylori antibodies among groups receiving both MNU and H. pylori (Groups 1–5), as we reported earlier.14 Waynforth et al.33 pointed out that MNU reduces humoral immune responses and Brodt et al.34 also showed that chemical carcinogens can induce immunosuppression. Thus MNU may interfere with the humoral immune response of gerbils against H. pylori.
Our results also showed that salt intake induces hypergastrinemia in H. pylori-infected gerbils, although the role of gastrin in gastric carcinogenesis is still unclear and experimental studies have provided no definite consensus. Peek et al.35 suggested that epithelial cell growth in H. pylori-colonized mucosa may be mediated by a gastrin-dependent mechanism and Wang et al.36 reported that chronic hypergastrinemia and Helicobacter felis infection synergistically contribute to gastric carcinogenesis in a transgenic hypergastrinemic mouse model. On the other hand, Zavros et al.37 described neoplastic transformation in antral gastric mucosa in a gastrin-deficient mouse model. The serum gastrin level could not account for the dose-dependence of gastric adenocarcinoma development in the present study. Takashima et al.38 suggested that H. pylori infection can upregulate serum gastrin levels in Mongolian gerbils, mediated by interleukin (IL)-1β. In this context it is of interest that an IL-1β polymorphism is known to be linked with H. pylori infection and gastric cancer risk.39, 40
In conclusion, the present study showed dose-dependent enhancing effects of salt on gastric carcinogenesis in Mongolian gerbils treated with MNU and H. pylori. One of the most important mechanisms could be exacerbation of H. pylori infection through upregulation of SMCM and downregulation of GMCM. The results clearly provide evidence suggesting that reduction of salt intake may decrease gastric cancer risk in humans, compatible with previous epidemiological findings.
The authors thank Dr. Jun Nakayama, Shinshu University, for his collegial advice and Ms. Hisayo Ban for expert technical assistance.
- 8Salt and stomach cancer. In: ReedPI, HillMJ, eds. Gastric carcinogenesis. Amsterdam: Excerpta Medica, 1988. p 105–26., .
- 24Down regulation of a gastric transcription factor, Sox2, and ectopic expression of intestinal homeobox genes, Cdx1 and Cdx2: inverse correlation during progression from gastric/intestinal-mixed to complete intestinal metaplasia. J Cancer Res Clin Oncol 2004; 130: 135–45., , , , , , , , .