Helicobacter pylori (H. pylori) infection and high intake of various traditional salt-preserved foods are regarded as risk factors for human gastric cancer. We previously reported that Chinese cabbage contains indole compounds, such as indole-3-acetonitrile, a mutagen precursor. 1-Nitrosoindole-3-acetonitrile (NIAN), formed by the treatment of indole-3-acetonitrile with nitrite under acidic conditions, shows direct-acting mutagenicity. In the present study, NIAN administration by gavage to Mongolian gerbils (MGs) at the dose of 100 mg/kg two times a week resulted in three adduct spots (1.6 adducts/108 nucleotides in total), detected in DNA samples from the glandular stomach by 32P-postlabeling methods. Treatment with six consecutive doses of 100 mg/kg of NIAN, two times a week for 3 weeks, induced well—and moderately—differentiated glandular stomach adenocarcinomas in the MGs at the incidence of 31% under H. pylori infection at 54–104 weeks. Such lesions were not induced in MGs given broth alone, broth + NIAN or infection with H. pylori alone. Thus, endogenous carcinogens formed from nitrosation of indole compounds could be critical risk factors for human gastric cancer development under the influence of H. pylori infection.
Gastric cancer is the second most frequent cause of cancer death worldwide.1 Although gastric cancer has become a relatively rare cancer in North America and most Northern and Western European countries, it remains common in East Asia, Eastern Europe, Russia, and selected areas of Central and South America.2Helicobacter pylori (H. pylori) is a well-established major risk factor for gastric cancer,3–5 and the prevalence of H. pylori infection in East Asia countries, including Japan and Korea is reported to be relatively high.6, 7 In addition, the risk of gastric cancer is increased with a high intake of various traditional salt-preserved foods.3 In fact, pickled vegetable consumption is reported to increase gastric cancer risk in Japan and Korea.8–10 In Korea, kimchi, commonly prepared with Chinese cabbage or radish, is a traditional and popular food, which contains high levels of nitrate (median 1550 mg/kg).11 Furthermore, Chinese cabbage is well known as a pickled vegetable commonly consumed in Japan. Moreover, ingestion of nitrate, mainly from food, is suggested to correlate with mortality from gastric cancer.12–14 Ingested nitrate is mainly converted to nitrite by bacteria in the oral cavity after secretion into saliva.15 Carcinogenic N-nitroso compounds can be formed from nitrite and secondary amines under acidic conditions. Furthermore, direct-acting N-nitroso compounds, such as N-methyl-N′-nitro-N-nitrosoguanidine (MNNG)16 and N-methyl-N-nitrosourea (MNU),17 are known to induce cancer in the glandular stomach of experimental animals. Thus, it is suggested that N-nitroso compounds that are formed in the stomach under acidic conditions could be positively associated with the risk of gastric cancer. Nitric oxide, formed by nitric oxide synthase, is also reported to contribute to production of N-nitroso compounds.18
We have previously reported that treatments of various foodstuffs with nitrite under acidic conditions produce direct-acting mutagens towards Salmonella tester strains.19, 20 Among those foodstuffs, Chinese cabbage is shown to contain three indole compounds, indole-3-acetonitrile, 4-methoxyindole-3-acetonitrile and 4-methoxyindole-3-aldehyde as mutagen precursors. 1-Nitrosoindole-3-acetonitrile (NIAN), an N-nitroso-substituted compound formed by treatment of indole-3-acetonitrile with nitrite under acidic conditions, is a direct-acting mutagen in S. typhimurium and Chinese hamster lung cells,20–22 and it is confirmed to form DNA adducts and to induce DNA single-strand scission in the rat glandular stomach.23, 24 Therefore, NIAN could play some role in gastric cancer development, as in the case of the well-known direct-acting mutagens, MNNG and MNU, in animal experiments.16, 17, 25
The Mongolian gerbil (MG) is reported to be susceptible to colonization by H. pylori, and H. pylori infection greatly enhances MNNG or MNU-induced gastric carcinogenesis in MGs.26, 27 Therefore, the MG is considered to be a useful animal model for evaluating the gastric cancer risk of direct-acting N-nitroso compounds, with or without H. pylori infection.
Chinese cabbage, containing nitrate and indole compounds, is commonly consumed in East Asian countries, including Japan, Korea and China, in which gastric cancer mortality is very high. In the present study, DNA adducts were detected with NIAN treatment in the glandular stomach of MGs, and the carcinogenicity of NIAN for gastric cancer in vivo was examined. The results clearly demonstrated that gastric cancer developed with a combination of NIAN administration and H. pylori infection in MGs. Possible involvement of indole compounds and nitrate derived from various foodstuffs, including Chinese cabbage, in gastric cancer development in humans is discussed.
Material and Methods
Indole-3-acetonitrile was purchased from Tokyo Food Techno (Tokyo, Japan), sodium nitrite from Wako Pure Chemical Industries (Osaka, Japan) and ammonium sulfamate from Kanto Chemical (Tokyo, Japan). Brucella broth was obtained from Becton Dickinson (Cockeysville, MD) and horse serum from Nippon Bio-Supply (Tokyo, Japan).
Preparation of NIAN
The chemical structure of NIAN is shown in Figure 1a. Indole-3-acetonitrile in 27 mM citrate-phosphate buffer (pH 3.0) was treated with 50 mM sodium nitrite for 1 hr at room temperature in the dark, as previously reported.21 Nitrosation was stopped by addition of ammonium sulfamate at a final concentration of 50 mM. The reaction solution was filtered and the residue was washed with deionized water, then with n-hexane. The residual paste was dried and stored at −80°C until use. The preparation was >93% pure as judged by its UV absorbance on HPLC.
H. pylori (ATCC 43504; American Type Culture Collection, Manassas, VA) was cultured in brucella broth supplemented with 10% heat-inactivated horse serum for 24 hr at 37°C under microaerobic conditions (5% O2, 10% CO2 and 85% N2), as previously described.28
Specific pathogen-free male, 6-week-old MGs (MGS/Sea, Kyudo, Fukuoka, Japan) were housed in a biohazard room, air-conditioned at 24°C ± 2°C and 55% humidity, on a 12 hr light–dark cycle and were allowed free access to commercial diet (CE-2; CLEA Japan, Tokyo, Japan) and water.
To analyze the formation of DNA adducts in the glandular stomach of MGs by NIAN treatment, NIAN was dissolved in 50% dimethyl sulfoxide (DMSO), and administered to three MGs by gavage of 0.5 ml solution, two times a week at a level of 100 mg/kg body weight. Two further MGs served as a control group receiving the solvent alone (0.5 ml). At 8 hr after administration of NIAN, both groups of animals were sacrificed under ether anesthesia, and their stomachs were resected and stored at −80°C until use. DNA was extracted by a standard procedure with enzymatic digestion of protein and RNA followed by extraction with phenol and chloroform/isoamyl alcohol (24:1, v/v).
The protocol for long-term gastric carcinogenicity in MGs treated with NIAN + H. pylori infection is illustrated in Figure 1b. The animals were randomly divided into four groups (groups A–D). Groups A and C were given 50% DMSO without NIAN (0.5 ml) whereas groups B and D were orally administrated NIAN (0.5 ml, 100 mg/kg body weight) dissolved in 50% DMSO by gavage, two times a week for 3 weeks. At one week after the last administration, the animals of groups C and D were given an intragastric inoculation of H. pylori broth culture (0.5 ml, 0.9 × 108 CFU/animal) whereas animals of groups A and B were given sterilized broth alone (0.5 ml).28
During the experiments, animals which became moribund or emaciated (<80 g body weight) were sacrificed. At 104 weeks after H. pylori infection, all surviving animals were sacrificed under ether anesthesia. At performance of necropsy, all tissues were carefully checked macroscopically and the stomachs and major organs were removed and assessed for macroscopic lesion development. Effective numbers of animals were defined as those surviving until week 54 of the study, when gastric tumors were observed for the first time. In addition, in the H. pylori-infected groups, the animals developing gastritis observed on histological examination were regarded as effective. The percentages of gastritis-bearing animals by the single inoculation of H. pylori were 62% for group C and 76% for group D, being similar to those previously reported.27 All animal experiments were performed according to the “Guidelines for Animal Experiments in the National Cancer Center” and were approved by the Institutional Ethics Review Committee for Animal Experimentation in the National Cancer Center.
Detection of DNA adducts by 32P-postlabeling method
Calf thymus DNA (0.5 mg, Sigma, St. Louis, MO) treated with NIAN (3 mg) for 12 hr under neutral conditions was used for authentic NIAN-DNA adducts.23 DNA samples from the glandular stomach of MGs and calf thymus DNA samples were digested with micrococcal nuclease and phosphodiesterase II, and subjected to 32P-postlabeling analysis using the same procedure as described previously23 except with solvent systems for two-dimensional development. The solvent system consisted of buffer A (4.0 M lithium formate, 7.7 M urea, pH 3.5) from bottom to top, and buffer B (0.90 M lithium chloride, 0.45 M Tris-HCl, 7.7 M urea, pH 8.0) from left to right, followed by 1.7 M sodium phosphate buffer, pH 6.0, from left to right, with 3.5 cm filter paper. Adducts were detected with a Bio-Image Analyzer (BAS 3000; Fuji Photo Film, Tokyo, Japan) after exposing the TLC sheets to Fuji imaging plates. Relative adduct labeling was determined by the methods of Reddy et al.,29 and values were calculated as averages using data from three assays.
All excised stomachs were opened along the greater curvature and washed twice with saline, then fixed in 10% neutral-buffered formalin. The fixed stomachs were sliced along the longitudinal axis into 9–12 strips of equal width, and routinely processed to sections stained with hematoxylin and eosin (H&E). The degree of chronic active gastritis was graded according to criteria modified from the Updated Sydney System,30 by scoring the infiltration of neutrophils and mononuclear cells. Other organs, in which macroscopic lesions were observed, were also fixed in 10% neutral-buffered formalin and routinely processed to sections stained with H&E for histological examination.
The significance of differences in quantitative data for gastric inflammation, gastric adenocarcinoma and tumors of other organs was analyzed by Fisher's exact test. Data for stomach wet weight and inflammation score were examined using Tukey's multiple comparison test. Significance was concluded at p < 0.05.
DMSO: dimethyl sulfoxide; H&E: hematoxylin and eosin; H. pylori: Helicobacter pylori; MG: Mongolian gerbil; MNNG: N-methyl-N′-nitro-N-nitrosoguanidine; MNU: N-methyl-N-nitrosourea; NIAN: 1-nitrosoindole-3-acetonitrile.
DNA adduct formation by NIAN administration in the glandular stomach of MGs
To confirm the formation of NIAN-DNA adducts in the glandular stomach of MGs, NIAN was injected two times a week at a dose of 100 mg/kg by gavage, and then analyzed by 32P-postlabeling method. Three adduct spots were observed in DNA samples derived from NIAN-treated animals (Fig. 2a). The adduct levels were 0.3 for adduct 1, 1.1 for adduct 2, 0.2 for adduct 3 and 1.6 adducts/108 nucleotides in total. This TLC pattern was similar to that in the in vitro reaction of calf thymus DNA with NIAN (total adduct level of 4.8 adducts/107 nucleotides, Fig. 2b). In the case of DNA samples derived from control animals, no adduct spots were seen on the TLC sheets (Fig. 2c).
Macroscopical and microscopical observation of H. pylori-induced gastritis in MGs
MGs were sacrificed until 104 weeks after H. pylori infection, and gastric disorders were analyzed. Stomach wet weights and gastric inflammation scores are shown in Table 1. Macroscopically, edematous thickening with hemorrhagic spots was apparent in the gastric mucosa in H. pylori-infected MGs (groups C and D), but not in animals uninfected with H. pylori (groups A and B). The stomach wet weight, reflecting edematous thickening, in animals infected with H. pylori (groups C and D) was significantly increased compared with that of animals not infected with H. pylori (groups A and B) (p < 0.01). No significant differences of stomach wet weight were detected between groups A and B and also between groups C and D.
Table 1. H. pylori infection induced-gastritis in MGs
Microscopically, gastritis, featuring infiltration of many inflammatory cells, and hyperplastic change of glandular epithelium, and erosion were observed in the pyloric regions of the animals infected with H. pylori (groups C and D) (Fig. 3). Heterotopic proliferative glands, whose development is related to severe gastritis in H. pylori-infected MGs, were sometimes observed in H. pylori-infected groups (groups C and D). No gastritis was found in animals not infected with H. pylori (groups A and B). The gastric inflammation score in H. pylori-infected animals was significantly increased compared with that of animals uninfected with H. pylori (p < 0.01). There were no significant differences of gastric inflammation score between groups C and D.
Development of glandular stomach adenocarcinomas in MGs treated with both NIAN and H. pylori
The observed incidences of glandular stomach adenocarcinomas are shown in Table 2. Glandular stomach adenocarcinomas, histologically featuring tubular structures with cellular atypia infiltrating into the muscle layer, were found in eight animals treated with both NIAN and H. pylori (8/26 = 31%) at 54–104 weeks. All adenocarcinomas were observed in the pyloric mucosa and located in the lesser curvature of the stomach, where macroscopically severe edematous thickening was also seen (Fig. 4a). The observed adenocarcinomas in seven animals were of well differentiated (Fig. 4b), and a moderately differentiated lesion was observed in one animal (Fig. 4c). In the animals treated with broth alone, broth + NIAN and H. pylori alone (groups A, B and C), no glandular stomach adenocarcinomas were observed. The incidence of glandular stomach adenocarcinomas in group D was significantly higher than that in groups A, B and C (p < 0.05, p < 0.01 and p < 0.05, respectively).
Table 2. Incidence of glandular stomach adenocarcinoma in MGs
Irrespective of NIAN treatment and H. pylori infection, skin tumors, which histologically were well to poor differentiated squamous cell carcinomas, sebaceous carcinomas and melanomas, were found in one animal (1/15 = 7%) in group A, three animals (3/22 = 14%) in group B, two animals (2/18 = 11%) in group C and five animals (5/26 = 19%) in group D. A hemangioma was also observed in a kidney of one animal in group D (1/26 = 4%). No significant differences were apparent in these tumor incidences among groups A–D.
In the present study, NIAN was found to induce glandular stomach adenocarcinomas in MGs in combination with H. pylori infection. NIAN-DNA adducts were also detected in the glandular stomach of MGs after treatment with NIAN, although clarification of their chemical structure(s) has yet to be performed. DNA adducts observed in the glandular stomachs of NIAN-treated MGs probably contain an indole-3-acetonitrile moiety. However, it is further likely that NIAN would act as an NO donor under aqueous conditions, thereby causing DNA modifications.31–33 In fact, Lucas et al. demonstrated that NIAN can efficiently transfer nitroso groups to nucleophilic targets in purine nucleotides, causing N-nitrosation, deamination and the formation of a novel guanine analog, oxanine.33
Glandular stomach adenocarcinomas induced by NIAN treatment plus H. pylori infection were located in the pyloric region, similar to MNNG or MNU treatment plus H. pylori infection-induced glandular stomach adenocarcinomas in MGs.26, 27 Meanwhile, no glandular stomach cancers were observed in the groups of H. pylori-infected MGs without NIAN treatment, which is consistent with previous studies,26, 27 nor in the group treated with only NIAN. These findings indicated that H. pylori is a strong promoter of gastric carcinogenesis. Histological examination revealed that the tumors developed by NIAN + H. pylori were of well or moderately differentiated adenocarcinomas. Well or poorly differentiated adenocarcinomas and signet ring cell carcinomas were observed in H. pylori-infected MGs treated with MNNG or MNU.26, 27 Further studies are required to clarify the histological variety of stomach adenocarcinomas induced by NIAN, MNNG or MNU, since the type of cancer might depend on the genotoxic action of chemical carcinogens, rather than the effects of H. pylori infection.27 In addition, tumors were observed in skin and kidney, which were suspected to spontaneously develop. The MGs have been reported to develop spontaneous skin tumors such as sebaceous and squamous cell carcinoma.34
Epidemiological studies have indicated that nitrate intake increases gastric cancer risk, and major sources are vegetables including Chinese cabbage, spinach and parsley.14 Indole-3-acetonitrile, a precursor of NIAN, is distributed widely in cruciferous vegetables including Chinese cabbage and sprouts.35 Furthermore, fava beans (Vicia faba), which are commonly consumed in Colombia, give rise to a potent mutagen in the presence of nitrite under acidic conditions.36 The nitrosatable precursor of the mutagen in fava beans and the major product of nitrosation are reported to be an indole compound, 4-chloro-6-methoxyindole and an N-nitroso compound, 4-chloro-2-hydroxy-N1-nitroso-indolin-3-one oxime, respectively.37 Other indole compounds are also reported to produce direct-acting mutagens after nitrite treatment under acidic conditions.38, 39 In general, conversion of indole derivatives to nitrosated forms in vitro is known to be rapid and efficient at physiologically feasible nitrite concentrations with the low pH of the human stomach.37 Thus, it is conceivable that nitrosation of indole compounds such as indole-3-acetonitrile probably occurs in human stomach. On the other hand, nitric oxide is suggested to be produced by activated macrophages in inflamed organs with H. pylori infection.18 Therefore, nitrosation of indole compounds could be mediated by both acid catalysis and inflammatory responses in the human stomach.18, 20, 37–40 On the basis of the conversion rate of NIAN from indole-3-acetonitrile under physiological conditions, the dose of NIAN used in the present study appears about 500–1000 fold the expected human exposure to NIAN via fresh or pickled Chinese cabbage. However, humans continually consume various kinds of foods containing indole compounds and nitrate during ordinary life. Thus, it is probable that the total amount of nitroso-indole compounds would be much closer to the dose of NIAN used in the present study. Moreover, it has been reported that low doses of chemical carcinogens, such as MNNG and MNU, could induce glandular stomach cancers in rodents under inflammation conditions including NaCl treatment and H. pylori infection, but hardly induce glandular stomach cancer without NaCl treatment and H. pylori infection. Therefore, the continuous intake of indole compounds and nitrate may play an important role for gastric carcinogenesis in East Asian countries still with a high salt consumption and H. pylori infection rate.
Gastric cancer is tending to decline in most countries.41–43 One of the explanations for this tendency is the reduced prevalence of H. pylori infection.42 Changes in dietary habits, mainly being lower salt consumption, could be also related to reduced gastric cancer incidence. However, the gastric cancer prevalence in East Asian countries, such as Japan and Korea, is still high.2 At present, we have not succeeded in detecting NIAN in human bodies nor the exposure levels of the precursor, indole compounds for humans. Thus, it is necessary to estimate the human exposure levels to nitroso-indole compounds including NIAN, and to study further animal experiments and epidemiological analyses for clarification of contribution of nitroso-indole compounds under H. pylori infection in humans gastric carcinogenesis.
In conclusion, the present study demonstrated that NIAN can induce gastric cancer in H. pylori-infected MGs. It is noteworthy that nitrosatable precursors widely exist in foods. Thus, it is suggested that N-nitroso indole compounds including NIAN might contribute to the frequent development of gastric cancer in East Asian countries such as Japan and Korea in which the prevalence of H. pylori infection is relatively high. Further studies of interaction with other dietary elements appear warranted to promote the prevention of human gastric cancer.
The authors thank Dr. Nobuo Takasuka, Naoaki Uchiya and Yusaku Hori for their expert technical assistance. S.T. is presently the recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research.