Gastric and oesophageal carcinogenesis: models for the identification of risk and protective factors

Bile Acid-Microbiome Interaction Promotes Gastric Carcinogenesis

Bile reflux gastritis (BRG) is associated with the development of gastric cancer (GC), but the specific mechanism remains elusive. Here, a comprehensive study is conducted to explore the roles of refluxed bile acids (BAs) and microbiome in gastric carcinogenesis. The results show that conjugated BAs, interleukin 6 (IL-6), lipopolysaccharide (LPS), and the relative abundance of LPS-producing bacteria are increased significantly in the gastric juice of both BRG and GC patients. A secondary BA, taurodeoxycholic acid (TDCA), is significantly and positively correlated with the LPS-producing bacteria in the gastric juice of these patients. TDCA promotes the proliferation of normal gastric epithelial cells (GES-1) through activation of the IL-6/JAK1/STAT3 pathway. These results are further verified in two mouse models, one by gavage of TDCA, LPS, and LPS-producing bacteria (Prevotella melaninogenica), respectively, and the other by bile reflux (BR) surgery, mimicking clinical bile refluxing. Moreover, the bile reflux induced gastric precancerous lesions observed in the post BR surgery mice can be prevented by treatment with cryptotanshinone, a plant-derived STAT3 inhibitor. These results reveal an important underlying mechanism by which bile reflux promotes gastric carcinogenesis and provide an alternative strategy for the prevention of GC associated with BRG.

Deoxycholic Acid Could Induce Apoptosis and Trigger Gastric Carcinogenesis on Gastric Epithelial Cells by Quantitative Proteomic Analysis

Background. Pathologic duodenogastric reflux can induce or aggravate gastritis because of the presence of bile acids. Bile reflux has been generally considered to be associated with intestinal metaplasia and gastric cancer. However, the pathogenic mechanisms of the effects of bile acids on gastric mucosa are still unknown. Methods. To explore the mechanisms by which bile acids induce gastric mucosal lesions, we examined cell apoptosis in the gastric epithelial cell line GES-1 and investigated the changes in protein profiles of GES-1 cells in response to a bile acid deoxycholic acid using a proteomics approach. Changes in the profiles of the differently expressed proteins were analyzed using the DAVID and STRING programs. Results. We found apoptosis was significantly induced in GES-1 cells by deoxycholic acid. Using liquid chromatographic/tandem mass spectrometric (LC-MS/MS) methods, 134 upregulated proteins and 214 downregulated proteins were identified in the bile acid treated GES-1 cells. Bioinformatics analysis revealed the interactions and signaling networks of these differentially expressed proteins. Conclusion. These findings may improve the understanding of the molecular mechanisms underlying the pathogenicity of bile acids on gastric mucosa.

Bile acids initiate lineage-addicted gastroesophageal tumorigenesis by suppressing the EGF receptor-AKT axis

While bile acids are a risk factor for tumorigenesis induced by reflux disease, the mechanisms by which they contribute to neoplasia remain undefined. Here, we reveal that in gastroesophageal junction (GEJ) cells bile acids activate a tissue-specific developmental program defining the intestinal epithelial cell phenotype characterizing GEJ metaplasia. Deoxycholic acid (DCA) inhibited phosphorylation of EGF receptors (EGFRs) suppressing the proto-oncogene AKT. Suppression of EGFRs and AKT by DCA actuated an intestine-specific cascade in which NF-kappaB transactivated the tissue-specific transcription factor CDX2. In turn, CDX2 orchestrated a lineage-specific differentiation program encompassing genes characterizing intestinal epithelial cells. Conversely, progression from metaplasia to invasive carcinoma in patients, universally associated with autonomous activation of EGFRs and/or AKT, was coupled with loss of this intestinal program. Thus, bile acids induce intestinal metaplasia at the GEJ by activating the lineage-specific differentiation program involving suppression of EGFR and AKT, activating the NF-kappaB-CDX2 axis. Induction of this axis provides the context for lineage-addicted tumorigenesis, in which autonomous activation of AKT corrupts adaptive intestinal NF-kappaB signaling, amplifying tumorigenic programs.

Bile acids induce ectopic expression of intestinal guanylyl cyclase C Through nuclear factor-kappaB and Cdx2 in human esophageal cells

BACKGROUND & AIMS

Although progression to adenocarcinoma at the gastroesophageal junction reflects exposure to acid and bile acids associated with reflux, mechanisms mediating this transformation remain undefined. Guanylyl cyclase C (GC-C), an intestine-specific tumor suppressor, may represent a mechanism-based marker and target of transformation at the gastroesophageal junction. The present studies examine the expression of GC-C in normal tissues and tumors from esophagus and stomach and mechanisms regulating its expression by acid and bile acids.

METHODS

Gene expression was examined by reverse-transcription polymerase chain reaction, promoter analysis, immunohistochemistry, immunoblotting, and functional analysis. Promoter transactivation was quantified by using luciferase constructs and mutational analysis. DNA binding of transcription factors was examined by electromobility shift analysis.

RESULTS

GC-C mRNA and protein were ectopically expressed in approximately 80% of adenocarcinomas arising in, but not in normal, esophagus and stomach. Similarly, in OE19 human esophageal cancer cells, deoxycholate and acid induced expression of GC-C. This was associated with the induction of expression of Cdx2, a transcription factor required for GC-C expression. In turn, induction of Cdx2 expression by deoxycholate was mediated by binding sites in the proximal promoter for nuclear factor kappaB (NF-kappaB). Furthermore, deoxycholate increased NF-kappaB activity, associated with nuclear translocation and Cdx2 promoter binding of the NF-kappaB subunit p50. Moreover, a dominant negative construct for NF-kappaB prevented deoxycholate-induced p50 nuclear translocation and activation of the Cdx2 promoter.

CONCLUSIONS

Transformation associated with reflux at the gastroesophageal junction reflects activation by bile acid and acid of a transcriptional program involving NF-kappaB and Cdx2, which mediate intestinal metaplasia and ectopic expression of GC-C.

Riboflavin deficiency and esophageal cancer: a case control-household study in the Caspian Littoral of Iran

BACKGROUND

In a case-household-control-household study in two very high and low esophageal cancer (EC) risk regions of the Caspian Littoral of Iran, a total of 21 cases (12 subjects from the high risk and 9 subjects from the low-risk region) with a total of 91 household members (57 subjects from the high risk and 34 subjects from the low-risk region) were investigated. Cases were matched for sex and age (+/-5 years) with non-blood relative controls.

METHODS

A standard 24-h dietary recall questionnaire was used to estimate riboflavin intake. The erythrocyte glutathione reductase activity coefficient (EGR-AC) was measured to assess riboflavin status. The Student t-test was used to test differences, and chi2 analysis was applied to test associations. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were also calculated.

RESULTS

Results indicated that in both regions, the mean daily intake of riboflavin for cases was less than that of the controls (0.66+/-0.43 mg/day versus 0.82+/-0.37 mg/day) whereas for their households, it was virtually the same. Both cases and control households showed riboflavin deficiency in two regions, with higher deficiency in the high risk area. Statistical analysis revealed significant differences between the two regions for EGR-AC (P<0.001). Odd ratios indicated that the risk of developing EC for persons living in riboflavin-deficient households was more than twice of non-deficient households.

CONCLUSION

Therefore, this study suggests that riboflavin deficiency may play an important role in the etiology of esophageal cancer in the Caspian Littoral of Iran.

Effect of salt on cell proliferation and N-methyl-N'-nitro-N-nitrosoguanidine penetration to proliferative cells in the forestomach of rats

We have studied the effect of intragastric instillation of 4.5 M NaCl on cell proliferation and carcinogen penetration into the forestomach of Wistar rats at different time intervals after salt exposure. Cells in the S-phase were labelled by incorporation of bromodeoxyuridine and located after immunohistochemistry. N-[3H]Methyl-N'-nitro-N-nitrosoguanidine ([3H]MNNG) was used as the carcinogen and penetration of [3H]MNNG to proliferative cells was determined by autoradiography. The number of cells in S-phase per millimetre epithelium length 12 h and 24 h after salt exposure (32.2 +/- 11.9 and 20.6 +/- 7.4) was significantly higher than in the control animals (9.4 +/- 3.6). No increase in cell proliferation occurred during the first 2 h after salt exposure. Microscopy also revealed oedema in the lamina propia. The forestomach blood flow was not influenced by the application of hypertonic saline. [3H]MNNG at a concentration of 40 micrograms/ml did not penetrate to the proliferative cells in the forestomach and no effect of salt pretreatment on carcinogen penetration was seen. The low penetration of [3H]MNNG to proliferative cells in the forestomach epithelium may explain why this concentration of MNNG given in drinking water over several weeks usually does not induce squamous cell carcinomas in the forestomach. The previously observed cocarcinogenic effect of salt on squamous cell cercinogenesis in the upper gastrointestinal tract could be due to the observed increase in cell proliferation after salt exposure.

Gastric carcinogenesis in rats given hypertonic salt at different times before a single dose of N-methyl-N'-nitro-N-nitrosoguanidine

A 1-ml dose of 4.5 M NaCl was given intragastrically to male Wistar rats at 10 min, 1 h, 4 h, 12 h, 24 h or 48 h before a single intragastric dose of 250 mg/kg N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). After 52 weeks the incidence of forestomach squamous cell carcinoma was 21% in control animals receiving MNNG alone. The cancer incidence in the forestomach varied with the time elapsed between application of NaCl and MNNG, and was significantly increased in animals pretreated with NaCl 4 h (43%), 12 h (54%) and 24 h (41%) before MNNG. These results show that salt has a cocarcinogenic effect on initiation of forestomach carcinogenesis in rats, and that this effect depends on the time interval between pretreatment with NaCl and application of MNNG. Gastric adenocarcinomas occurred more frequently in the antrum (78%) than in the corpus (22%). The incidence of gastric adenocarcinoma in animals pretreated with salt before application of MNNG (11%-22%) was not significantly influenced by the time elapsed between pretreatment with salt and application of MNNG, and did not differ from animals receiving MNNG alone (18%). The lack of a cocarcinogenic effect of NaCl on glandular gastric carcinogenesis might be due to the use of dimethyl/sulfoxide as solvent for MNNG.

[Toxicology of the synthetic antioxidants BHA and BHT in comparison with the natural antioxidant vitamin E]

The toxicology of the food preservatives butylhydroxyanisole (BHA) and butylhydroxytoluene (BHT) as well as the naturally occurring vitamin E (alpha-tocopherol) is described. In high dosages all three compounds induce in animals impairment of blood clotting, which can be explained by an antagonism with vitamin K. Specific toxic effects to the lung have only been observed with BHT. The other described toxic effects of BHA and BHT are less characteristic and often occur only after high dosage and long-term treatment. However, BHA induces in animals tumours of the forestomach, which are dose dependent, whereas BHT induces liver tumours in long-term experiments. Because there is no indication of genotoxicity of BHA and BHT, all published findings agree with the fact that BHA and BHT are tumour promoters. In contrast to BHA and BHT, vitamin E is not carcinogenic. On the other hand, all three antioxidants have also anticarcinogenic properties. The intake of the necessary high doses as for these effects are, however, contraindicated with BHA and BHT because of their carcinogenic effects. The present overview concludes that the concentrations of BHA and BHT nowadays used in food, drugs and cosmetics are probably harmless. In addition, vitamin E can also be used in higher doses without the occurrence of adverse effects.

Butylated hydroxyanisole in perspective

Butylated hydroxyanisole (BHA) is a synthetic food antioxidant used to prevent oils, fats and shortenings from oxidative deterioration and rancidity. This review depicts the current knowledge on BHA. The physical and chemical characteristics of BHA are summarized and its function as a food antioxidant is made clear. The toxicological characteristics of BHA and its metabolic fate in man and animal are briefly reviewed. Special emphasis is laid on the carcinogenicity of BHA in the forestomach of rodents and to related events in the forestomach and other tissues in experimental animals. At present there is sufficient evidence for carcinogenicity of BHA, but there is hardly any indication that BHA is genotoxic. Therefore risk assessment for this epigenetic carcinogen is based on non-stochastic principles. However, the mechanism underlying the tumorigenicity of BHA is not known. In the last part of this review an attempt is made to unravel the unknown mechanism of carcinogenicity. It is hypothesized that BHA gives rise to tumor formation in rodent forestomach by inducing heritable changes in DNA. Evidence is being provided that reactive oxygen species, in particular hydroxylradicals, may play a crucial role. The key question with respect to risk assessment for BHA is whether or not the underlying mechanism is thresholded, which is important for the choice of the appropriate model to assess the risk, if any, for man and to manage any potential risk.

Antioxidants--carcinogenic and chemopreventive properties

Synthetic and naturally occurring antioxidants have a wide variety of biological actions in rodents in addition to their primary antioxidant activity. Some of the included biological effects are of direct interest in relation to studies of carcinogenicity and/or modulation of carcinogenesis. Since the synthetic antioxidant BHA was first found to exert carcinogenic potential in rat and hamster forestomach epithelium, many other synthetic and naturally occurring antioxidants have been examined for their ability to induce proliferative activity in the alimentary canal. These studies have revealed that caffeic acid and sesamol are also tumorigenic for rat forestomach epithelium, whereas catechol and p-methylcatechol induce neoplasia in rat glandular stomach epithelium. Although the proliferative response is very rapid, with inflammation and ulceration, it takes a very long time before carcinomas develop. The proliferative lesions in the forestomach induced by BHA or caffeic acid are largely reversible, in contrast to those induced by genotoxic carcinogens, which generally persist and develop into cancer. Therefore, chronic irritation is considered to be responsible for the induction of stomach cancer by antioxidants. Butylated hydroxyanisole can undergo oxidative metabolism in vitro, and some of the metabolites formed have the potential for binding to proteins. Neither BHA nor its metabolites binds to DNA in vivo, but protein binding in the forestomach was greater than 10 times higher than that in the glandular stomach. It is thus conceivable that BHA is oxidatively metabolized in the forestomach epithelium (possibly entering into redox cycling), and reactive metabolites including semiquinone radicals or active oxygen species are responsible for the carcinogenesis by a mechanism involving binding to macromolecules. Many antioxidants have been shown to modify carcinogenesis, and as a rule, they inhibit the initiation stage by reducing the interaction between carcinogen and DNA. However, both promotion and inhibition have been reported for second-stage carcinogenesis, depending on the organ site, species of animal, or initiating carcinogen. They can also block reaction of amine and nitrite to form nitrosamines or reduce TPA promotion of skin carcinogenesis. Generally high doses of antioxidants are required for carcinoma induction or modification of chemical carcinogenesis. The significance of the reported tumorigenicity and strong promoting activity of antioxidants for forestomach epithelium of animals to the development of human cancer appears limited mainly because humans do not have a forestomach. The carcinogenic and strong promoting activities of catechol and its structurally related compounds on rat glandular stomach epithelium are of greater concern because this tissue is directly analogous to human gastric epithelium.(ABSTRACT TRUNCATED AT 400 WORDS)

Gastric carcinogenesis: a model for the identification of risk factors

In a series of extensive studies on gastric carcinogenesis, we have used Sprague-Dawley rats to examine the morphologic, histochemical, and biochemical effects of risk and protective factors on N-methyl-N'-nitro-N-nitroso guanidine (MNNG)-induced tumors in an attempt to link early observations with the end-point lesion, gastric cancer. We have observed that the putative risk factors sodium chloride (NaCl); a mixture of bile acids; aspirin; alcohol; and nitrite enhance MNNG-induced neoplasia of the gastric mucosa. On the other hand butylated hydroxyanisol (BHA), Se and difluromethylornithine (DFMO) were protective and inhibited the induction of gastric mucosal neoplasia. In most cases, early changes detected by a number of criteria correlated with the end-point, gastric neoplasia. This model appears to be useful in screening and evaluating chemicals for risk for or protection against gastric cancer.