O6-methylguanine DNA methyltransferase activity in liver from various fish species

O6-Methylguanine DNA methyltransferase (O6-MT) is considered to play an important role in the repair of alkylating carcinogen-induced lesions in a wide range of mammalian species. Fish are used widely in cancer research, one advantage being their high sensitivity to a variety of alkylating agents. To throw light on the mechanisms of DNA repair in the hitherto uninvestigated fish group, O6-MT activity was measured in liver from eight fish species belonging to six classes. Levels of O6-MT activity comparable with mouse values were found in liver of Japanese medaka (Oryzias latipes). Relatively low, but appreciable, levels of O6-MT activity were also observed in the other seven species examined. No adaptive increase in enzyme activity could be established in liver of rainbow trout following chronic dimethylnitrosamine pretreatment.     

Dietary zinc deficiency increases the methylbenzylnitrosamine- induced formation of O6-methylguanine in the esophageal DNA of the rat

Dietary zinc deficiency in combination with environmental exposureto methylbenzylnitrosamine (MBN) is associated with an increasedincidence of esophageal carcinoma in man. The proposed mechanismof MBN-induced esophageal carcinoma is through metabolic activationof MBN to form benzaldehyde, and a carbonium ion which methylatesDNA. MBN is known to methylate DNA forming O⁶-methylguanine(O⁶ MeG) adducts. These adducts can induce guanine to adeninepoint mutations and such mutations are responsible for certaincarcinogen-induced tumors. Rats maintained on a zinc-deficientdiet exhibit an increased incidence of MBN-induced esophagealcarcinoma when compared with ad libitum and pair-fed controls.The caloric restriction of the pair-fed controls was associatedwith a lower incidence of MBN-induced esophageal carcinoma thanwas observed in the ad libilum controls. These differences intumor incidence were associated with alterations in the formationand clearance of MBN-induced esophageal O⁶-MeG. Weanling maleSprague-Dawley rats were raised on egg protein diets containing2.3 p.p.m. zinc (low zinc) or 50 p.p.m. zinc (control). Onegroup of control animals was fed the control diet ad libitumand a second group pair-fed the control diet to match the intakeof the zinc-deficient group. After 3 weeks on the diets theanimals were injected with a single dose of MBN (2.0 mg/kg b.w.)and levels of esophageal O⁶-MeG were determined after 1, 3,6 and 24 h. O⁶-MeG was significantly higher in the zinc-deficientanimals than in controls, with the pair-fed controls demonstratingO⁶-MeG levels lower than the ad libitum controls. Thus, dietaryzinc deficiency results in significantly increased levels ofMBN-induced esophageal O⁶-MeG. and caloric restriction resultsin decreased levels of MBN-induced esophageal O⁶-MeG. Thesechanges in esophageal O⁶-MeG may in part explain the increasedincidence of MBN-induced esophageal carcinoma observed withdietary zinc deficiency.     

Repair of O6-methylguanine in DNA by demethylation is lacking in Mer- human tumor cell strains

The ability of extracts of human tumor cells to demethylateO6-methylguanine (O⁶-MeG) in DNA was assayed using the syntheticDNA polymer poly(dC,dG,m⁶dG). Cell strains proficient in repairof O⁶-MeG in vivo (Mer+ phenotype) contained a methyltransferaseactivity while repair deficient cells (Mer^– phenotype)had little or no activity. Mixing extracts of different Mer^–strains did not result in the appearance of the activity. Extractsof Mer^– cells did not inhibit the activity in extractsof Mer⁺ cells. Both Mer⁺ and Mer^– strains contained methylnitrosourea-damage-specificendonudease activity. The data suggest that the Mer⁺ strainsare deficient in methyltransferase and that this is the fundamentalreason for their hypersensitivity to the cytotoxic effects ofDNA alkyla-tion. The activity was partially purified from aMer⁺ colon carcinoma cell strain. Its kinetics parallel therepair of O⁶-MeG in DNA in vivo and suggest that the activityis inactivated during repair of DNA.     

Formation of O6-methylguanine in regenerating rat liver by N-nitrosomethylbenzylamine is not sufficient for initiation of preneoplastic foci

The DNA methylating activities of N-nitrosodimethylamine (NDMA),an initiator of hepatic -glutamyltranspeptidase-positive foci,and N-nitrosomethylbenzylamine (NMBzA), which does not initiate,were studied in regenerating rat liver. Equimolar doses of ¹⁴C-labelledNDMA and NMBzA (33.5 µmol/kg) were administered to maleSprague-Dawley rats 18 h after partial hepatectomy. NDMA andNMBzA both produced 7-methylguanine and O⁶-methylguanine. Theresults suggest that although the formation of O⁶-methylguaninemay be necessary it is not sufficient for initiation of preneoplasticfoci.     

Increased excision of O6-methylguanine from rat liver DNA after chronic administration of dimethylnitrosamine.

Male BDIV rats were given dimethylnitrosamine (2 mg/kg daily) by stomach tube on weekdays for a total of 9 weeks, the final dose being of 14C-labeled material. Control rats received only the labeled dimethylnitrosamine. Liver DNA was isolated at various times later (from 2 to 12 hr), and normal and alkylated purines were determined after hydrolysis in mild acid by chromatography on Sephadex G-10. The levels (measured as dpm/micronmol of parent base) of 7-methylguanine in the DNA of the pretreated rats were the same as or slightly higher than those of the control animals, and the persistence of this product was similar in both groups. This was also true for 3-methyladenine. In contrast, the initial amount of O6-methylguanine in the liver DNA of the pretreated rats was one-third of the amount found in the control rats, and the rate of loss of this product from DNA was higher in the pretreated animals. These differences were reflected in the alkylation product ratios: the 3-methyladenine:7-methylguanine ratios were closely similar in the two groups of animals at all times, whereas the O6-methylguanine:7-methylguanine ratio was initially 3 times higher in the control animals and fell more slowly. DNA synthesis (as measured by the incorporation of [3H]thymidine) was higher in the liver, kidney, and lung of rats receiving dimethylnitrosamine pretreatment. These findings are discussed with respect to the hepatocarcinogenicity of chronically administered dimethylnitrosamine.     

Dietary fish oil inhibits the expression of farnesyl protein transferase and colon tumor development in rodents

Although epidemiological and experimental studies indicate a strong relationship between different dietary fats and risk of colon cancer, the modulating effects of these nutritional factors at the molecular level are not fully elucidated. Activated ras genes have been implicated in the etiology of many human malignancies, including colon cancer. It is well established that the transforming ability of ras-p21 depends on its correct localization in plasma membrane. We have previously demonstrated that ingestion of a relatively higher amount of dietary fish oil leads to reduced plasma membrane levels of ras-p21 with concomitant increase in its cytoplasmic contents during the promotion and progression phases of chemically-induced colon tumorigenesis. In this follow-up experiment, we have found that intake of a high amount of corn oil, one of the most widely used fats in the American diet, enhances the expression of farnesyl protein transferase (FPTase). This enzyme catalyses farnesylation of ras precursors in a critical step during post-translational modification of ras oncoproteins, thereby enabling their anchorage to plasma membrane. In contrast, consumption of high amounts of fish oil, which is rich in omega-3 polyunsaturated fatty acids, reduces the levels of FPTase expression, thus inhibiting post-translational processing of ras precursors resulting in decreased ras function both in colonic mucosa as well as in colon tumors. These results correlate with increased incidence and multiplicity of grossly visibly colon tumors in carcinogen-treated animals fed a high corn oil diet versus decreased incidence and multiplicity of colon tumors in their counterparts fed the high fish oil diet. This dietary inhibition of FPTase may have a practical chemopreventive potential.      

Circumvention of ACNU-resistance in rat glioma cells by pretreatment with O6-methylguanine

The chemotherapy of malignant brain tumors has been, only partially successful yet. Recently major concern is drug resistance, one of possible mechanisms of such drug resistance stems from inducible repair enzyme, especially in case of chloroethylnitrosoureas as ACNU or BCNU. We examined the changes of acquired resistance to ACNU in rat glioma cells by pretreatment with O6-methylguanine, which is a substrate for O6-methylguanine methyltransferase. ACNU-resistant (9L/AC) cells had established after 10 times treatments of ACNU. 9L/AC cells were pretreated with 2 mM O6-methylguanine for 2 hours, and subsequently challenged with increasing doses of ACNU for 2 hours. In vitro colony formation assay the survival fraction of 9L and 9L/AC cells ranged from 0.39 to 0.63 by 2-hour reaction of 1-3 mM O6-methylguanine. Based on the dose-response curve for ACNU in 9L/AC cells, by O6-methylguanine pretreatment (2 mM), ACNU-resistance decreased markedly to one-third, one-fifth, and one-two hundredth at 12, 24, 36 microM ACNU, respectively. In contrast, the survival of 9L cells against ACNU was similar under O6-methylguanine pretreatment or nontreatment condition. Therefore, ACNU-resistance is considerably related to DNA repair enzyme induction, and the substrates may potentiate the cell-killing effect of ACNU in the resistant glioma cells.     

Comparison of pulmonary O6-methylguanine dna adduct levels and Ki-ras activation in lung tumors from resistant and susceptible mouse strains

The role of O⁶-methylguanine (O⁶MG) DNA adduct formation and persistence in the formation of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK)—induced lung tumors from resistant C57BL/6 and susceptible A/J mice was investigated. In addition, the frequencies of pulmonary tumor formation and Ki-ras activation were defined in C57BL/6 mice treated with NNK or vinyl carbamate (VC), and the role of the p53 gene in pulmonary carcinogenesis in these resistant mice was examined. One day after treatment with 100 mg/kg NNK, O⁶MG adduct concentrations were twofold to eightfold higher in Clara cells and type II cells than in small cells or whole lungs from both mouse strains. The concentrations of O⁶MG in isolated cells decreased at a similar rate in the two strains of mice. Lung tumors were detected by 27 mo of age in 18% of the C57BL/6 mice after a single 100 mg/kg dose of NNK and in 46% of these mice after a single 60 mg/kg dose of VC. In contrast, the tumor incidence in untreated C57BL/6 mice was 4%. Only one of 22 lung tumors from C57BL/6 mice treated with NNK contained an activated Ki-ras gene that was associated with an O⁶MG DNA adduct, whereas previous studies detected activated Ki-ras oncogenes in most of the NNK-induced lung tumors analyzed from susceptible A/J and resistant C3H mice. The small differences in formation and persistence of the O⁶MG adduct in whole lung or isolated lung cells from A/J and C57BL/6 strains do not account for the differences in either susceptibility for tumor formation or activation of the Ki-ras gene between these strains. In contrast to the low number of NNK-induced tumors with Ki-ras mutations in the resistant mice, 11 of 20 lung tumors from VC-treated mice contained activated Ki-ras genes. Neither p53 tumor suppressor gene mutations nor overexpression of the p53 protein were detected in spontaneous or chemically induced lung tumors in C57BL/6 mice. Thus, although Ki-ras activation was detected in some tumors, pathways independent of ras activation and p53 inactivation also appear to be involved in lung tumorigenesis in this resistant mouse strain.     

Apoptosis in malignant glioma cells triggered by the temozolomide-induced DNA lesion O6-methylguanine

Methylating drugs such as temozolomide (TMZ) are widely used in the treatment of brain tumours (malignant gliomas). The mechanism of TMZ-induced glioma cell death is unknown. Here, we show that malignant glioma cells undergo apoptosis following treatment with the methylating agents N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and TMZ. Cell death determined by colony formation and apoptosis following methylation is greatly stimulated by p53. Transfection experiments with O(6)-methylguanine-DNA methyltransferase (MGMT) and depletion of MGMT by O(6)-benzylguanine showed that, in gliomas, the apoptotic signal originates from O(6)-methylguanine (O(6)MeG) and that repair of O(6)MeG by MGMT prevents apoptosis. We further demonstrate that O(6)MeG-triggered apoptosis requires Fas/CD95/Apo-1 receptor activation in p53 non-mutated glioma cells, whereas in p53 mutated gliomas the same DNA lesion triggers the mitochondrial apoptotic pathway. This occurs less effectively via Bcl-2 degradation and caspase-9, -2, -7 and -3 activation. O(6)MeG-triggered apoptosis in gliomas is a late response (occurring >120 h after treatment) that requires extensive cell proliferation. Stimulation of cell cycle progression by the Pasteurella multocida toxin promoted apoptosis whereas serum starvation attenuated it. O(6)MeG-induced apoptosis in glioma cells was preceded by the formation of DNA double-strand breaks (DSBs), as measured by gammaH2AX formation. Glioma cells mutated in DNA-PK(cs), which is involved in non-homologous end-joining, were more sensitive to TMZ-induced apoptosis, supporting the involvement of DSBs as a downstream apoptosis triggering lesion. Overall, the data demonstrate that cell death induced by TMZ in gliomas is due to apoptosis and that determinants of sensitivity of gliomas to TMZ are MGMT, p53, proliferation rate and DSB repair.     

Effect of chronic administration of dimethylnitrosamine on the excision of O6-methylguanine from rat liver DNA.

Rats were exposed chronically to unlabelled N,N-dimethylnitrosamine (25 ppm in the drinking water) then given a single dose of N-[3H]methyl-N-nitrosourea (10 mg/kg body weight). The rates of loss of tritium-labeled 7-methylguanine, O6-methylguanine and 3-methyladenine from the liver DNA in control and dimethylnitrosamine-treated rats were found not to be significantly different. Thus, under the conditions used, inhibition of the O6-methylguanine excision repair system does not seem to be a factor in the induction of liver tumours by chronic DMN application.     

Ubiquitous presence of O6-methylguanine in human peripheral and cord blood DNA.

O6-Methylguanine (O6-meG) is a powerful premutagenic lesion that can arise from exposure to methylating agents. Although it has been reported to occur in human DNA, no systematic epidemiological analysis of its occurrence in populations suffering general environmental exposure is available. We report here results from a study of the presence of O6-meG in maternal and cord blood leukocyte DNA of women not knowingly exposed to methylating agents. Using a modification of an already existing method capable of detecting the lesion at levels as low as 16 nmol/molG, the adduct was detected in 31 of 36 maternal and 30 of 36 cord samples, at levels ranging up to 192 nmol/molG. Adduct levels in maternal blood DNA were significantly higher than those in cord blood DNA (P < 0.05), and there was a strong correlation between adduct levels in the two tissues (P < 0.001). In bivariate analysis, no significant association of adduct levels in either tissue and residence air pollution, active and passive smoking status, or eating habits was found. However, intake of fruits/vegetables and of vitamin supplements showed nonstatistically significant trends toward being associated with lower adduct levels in both maternal and cord blood DNA. The same trend was observed after multivariate analysis where all the above variables were controlled for. These findings indicate that premutagenic methylation DNA damage is commonplace in individuals not known to have suffered excessive exposure to environmental methylating agents or their precursors and are compatible with an endogenous origin of this damage, possibly associated with endogenous nitrosation processes.     

Non-steroidol anti-inflammatory drug effect on crypt cell proliferation and apoptosis during initiation of rat colon carcinogenesis.

Sustained use of non-steroidal anti-inflammatory drugs (NSAIDs) may prevent colorectal cancer. However, the optimal drug, period of efficacy and mechanism(s) of action are unknown. Experiments were undertaken to determine which of several NSAIDs would modulate colon crypt cell proliferation or apoptosis when given during the initiation phase of 1,2-dimethylhydrazine (DMH)-induced rat colon cancer. Colon crypts located both away from and over an aggregate of lymphoid nodules (ALN) were examined. Rats were injected with aspirin, indomethacin, nabumetone, sodium salicylate, 16,16-dimethyl prostaglandin E2 or saline for 3 days and DMH or DMH vehicle on day 4 of each week for 8 weeks, then killed 3 days after the last DMH injection. At the time of killing, DMH had significantly increased crypt cell proliferation but not apoptosis. There was significantly more cell proliferation and apoptosis in crypts over the ALN than away from the ALN. Aspirin and salicylate increased proliferation and apoptosis in crypts over the ALN. Finally, the distributional peaks of cell proliferation and apoptosis were shifted significantly closer together after DMH. Thus, DMH increases proliferation and alters the distribution of proliferating and apoptotic cells in colon crypts early in carcinogenesis. Aspirin may suppress tumour incidence via salicylate by enhancing apoptosis in carcinogen-initiated cells.     

Bladder tumor contains higher N7-methylguanine levels in DNA than adjacent normal bladder epithelium.

Schistosoma haematobium-infected patients are more likely to develop bladder cancer and be more exposed to carcinogenic N-nitroso compounds than uninfected patients. As N7-methylguanine is a marker of exposure to methylating agents of this type, we have measured N7-methyldeoxyguanosine 3'-monophosphate (N7-MedGp) by (32)P postlabeling. DNA was isolated from 42 paired normal and tumor tissue of Egyptians with bladder cancer. N7-MedGp was detected in DNA from 93% of the tumors and 74% of the normal bladder tissue samples. Adduct levels were highly variable and ranged from 0.04 to 6.4 and from 0.02 to 0.72 micromol/mol deoxyguanosine 3'-monophosphate (dGp) in tumor and normal DNA, respectively. N7-MedGp levels in normal and tumor DNA were highly correlated with one another (P = 0.007). The mean difference (95% confidence interval) in adduct levels between tumor and normal DNA was 0.21 (0.13-0.32) micromol/mol dGp and this was statistically significant (P < 0.001). The adduct ratio (tumor DNA/normal DNA) varied between 0.2 and 136 (median, 4.6). N7-MedGp levels were not associated with gender, age, or the presence of schistosomiasis. However, lower N7-MedGp levels were found in normal DNA from individuals lacking the GSTM1 gene (P = 0.03) but not the GSTT1 gene or in subjects with the Ile105Val GSTP1 polymorphism. These results show that exposure to methylating agents is widespread and suggest that such exposure may play a role both in tumor initiation and progression.     

Dietary fat and fiber differentially alter intracellular second messengers during tumor devolopment in rat colon

The effect of fat, fiber and carcinogen on clonic epithelialintracellular second messengers 1,2-disacyl -sn-glycerol (DAG),ceramide, and the steady-state level of phospholipase C (PLC-     

N-nitrosodimethylamine-derived O6-methylguanine in DNA of monkey gastrointestinal and urogenital organs and enhancement by ethanol

N-nitrosodimethylamine (NDMA) is a human cancer initiator suspect. Ethanol, a cancer risk factor, may synergize with nitrosamines by suppressing hepatic clearance, to increase internal exposure. A limitation to these hypotheses is lack of activation of NDMA by many rodent tissues. However, systematic primate studies are lacking. Patas monkeys were utilized to investigate NDMA activation by primate tissues in vivo, generating the promutagenic DNA lesion O⁶-methylguanine (O⁶-meG). Adult monkeys received 0.1 mg/kg NDMA by gavage, in some cases preceded by ethanol. Four hours after NDMA only, O⁶-meG was detected in DNA from all tissues. Levels were highest in gastric mucosa and liver and were only about 50% lower in DNA from white blood cells, esophagus, ovary, pancreas, urinary bladder and uterus. With ethanol co-exposure, amounts of O⁶-meG increased at least 2-fold in all tissues except liver. The largest effect was in esophagus (17-fold increase), followed by ovary, large intestine, urinary bladder, spleen and cerebellum (9- to 13-fold increases), and uterus, cerebrum and brain stem (7- to 8-fold increases). Alkylguanine alkyltransferase activities varied over a 30-fold range and were highest in liver and stomach. Thus primate tissues, especially those of the gastrointestinal and urogenital organs, are sensitive targets for DNA adduct damage due to NDMA, and ethanol co-exposure leads to striking increases in adducts. Our data support epidemiology implicating nitrosamines in causation of cancers of stomach and other organs, and alcohol as enhancing internal exposure to nitrosamines. © 1996 Wiley-Liss, Inc.     

O6-methylguanine accumulates in DNA of mammary glands after administration of N-methyl-N-nitrosourea to rats.

N-Methyl-N-nitrosourea (MNU) induces mammary carcinoma in female rats when given intravenously. After a single intravenous dose of N-methyl-N-nitrosourea (5 mg/100 g body wt.), we were unable to detect a shift of rat mammary gland DNA on an alkaline sucrose gradient. However, the alkylated products in DNA, 7-methylguanine and O6-methylguanine, were determined at various times following treatment with N-methyl-N-nitrosourea. O6-Methylguanine was removed from the DNA at a slower rate than 7-methylguanine and increased in the DNA with a second injection of N-methyl-N-nitrosourea. 3-Methyladenine was not detected in DNA from the mammary gland of the rat. These data support previous work with brain and bladder that suggest the persistence of O6-methylguanine in DNA might be involved in the induction of cancer by N-methyl-N-nitrosourea.     

Stimulation of transfer of methyl groups from O6-methylguanine in DNA to protein by rat liver extracts in response to hepatotoxins

An enzymatic activity present in rat liver extracts catalyzesthe transfer of methyl groups from O⁶-methylguanine in DNA toprotein. This activity was stimulated by treatment of rats withthioacelamide, carbon tetrachioride, 1,2-dimethylhydrazine,diethylnitrosamine, dimethylnitrosamine and by partial hepatectomybut not by treatment with N-methyl-N-nitrosourea or streptozotocin.These results suggest that an enhancement of this activity accompaniesthe in crease in cell division brought about by these agentsand is not necessarily a specific response to the presence ofalkylated bases in DNA.     

Depletion of O6-alkylguanine-DNA alkyltransferase activity in mammalian tissues and human tumor xenografts in nude mice by treatment with O6-methylguanine

Summary We have previously shown that exposure of cells in culture to O⁶-methylguanine significantly reduces their level of the repair protein, O⁶-alkylguanine-DNA-alkyltransferase (AGT), thus rendering cells more sensitive to the cytotoxic effects of chemotherapeutic chloroethylating agents. Experiments were carried out in mice to determine whether the AGT content of tissues and tumors could be reduced by in vivo treatment with O⁶-methylguanine. There was a dose-dependent decrease in AGT activity in liver tissues of CD-1 mice to 24% of basal levels after four hourly intraperitoneal injections of O⁶-methylguanine (110 mg/kg). Although the decline in AGT activity in the liver was reversible, the activity remained at 75% of basal levels for up to 25 h after the final injection. The effect of O⁶-methylguanine treatment on AGT activity was measured in mouse tissues as well as human colonic carcinoma tumors (HT29 and BE) grown in Swiss athymic nude mice. The activity in the liver, kidney, and spleen of these mice decreased to 33%–35% of control levels, whereas the activity in HT29 tumors was likewise diminished to 25% of control levels after four hourly injections of O⁶-methylguanine (100 mg/kg). There was no enhancement of the tumoricidal effectiveness of chloroethylating agents on the HT29 tumor after O⁶-methylguanine treatment, probably due to a disproportionately higher level of AGT in human tissue than in murine tissue. However, these studies suggest that O⁶-methylguanine can be given in vivo to examine the role of the AGT protein in protecting against the toxic and carcinogenic effects of alkylating agents.Abbreviations AGT O⁶-alkylguanine-DNA alkyltransferase - MeCCNU 1-(2-chloroethyl)-3-(4-methylcyclohexyl)-1-nitrosourea - Clomesone 2-chloroethyl(methylsulfonyl) methanesulfonate - HPLC high-pressure liquid chromatography     

Dietary resistant starch type 3 prevents tumor induction by 1,2-dimethylhydrazine and alters proliferation, apoptosis and dedifferentiation in rat colon

Some epidemiological and experimental studies suggest that consumption of resistant starch is preventive against colon cancer. Resistant starch leads to a fermentation-mediated increase in the formation of short-chain fatty acids, with a particularly high butyrate fraction in large bowel. Butyrate is considered to be protective against colon cancer because it causes growth arrest and apoptosis and regulates expression of proteins involved in cellular dedifferentiation in various tumor cell lines in culture. We sought to investigate these processes under conditions of a carcinogenicity experiment in vivo. In the present study, 1,2-dimethylhydrazine-treated Sprague-Dawley rats were fed standard diet (n=12) or diet with 10% hydrothermally modified Novelose 330, a resistant starch type 3 (RS3), replacing digestible starch (n=8). After 20 weeks tumor number, epithelial proliferation, apoptosis, immunoreactivity of carcinogenesis-related proteins [protein kinase C-delta (PKC-delta), heat shock protein 25 (HSP25) and gastrointestinal glutathione peroxidase (GI-GPx)], as well as mucin properties were evaluated in proximal and distal colon in situ. No tumors developed under RS3 diet, compared to a tumor incidence of 0.6+/-0.6 (P<0.05) under the standard diet. RS3 decreased the number of proliferating cells, the length of the proliferation zone and the total length of the crypt in the distal colon, but not proximal colon, and enhanced apoptosis in both colonic segments. It induced PKC-delta and HSP25 expression, but inhibited GI-GPx expression in the epithelium of distal colon. RS3 increased the number of predominantly acidic mucin containing goblet cells in the distal colon, but had no effect on the goblet cell count. We conclude that hydrothermally treated RS3 prevented colon carcinogenesis, and that this effect was mediated by enhanced apoptosis of damaged cells accompanied by changes in parameters of dedifferentiation in colonic mucosa.