.2»‘ we w _ 4-3 MSU RETURNING MATERIALS: P1ace in book drop to LIBRARIES remove this checkout from -_‘—. your record. FINES will be charged if book is returned after the date stamped below. 1? V4"; 5 EXPERIMENTAL COLON CANCER: EFFECTS OF ROUTE OF 1,2-DIMEI‘HYIHYDRAZINE (Mi) AIMINISTRATICN, CABBAGE CONSUMPTION AND PRIOR.EXPOSURE T0 DMH BY Mark John Messina A DISSERTATICN Submitted to Michigan State University in partial fulfillment of the requirements for the degree of [XXI'IOR OF PHIIDSOPHY Department of Food Science and Human Nutrition 1987 ABSTRACT EXPERIMENTAL COLON CANCER: EFFECTS OF ROUTE OF 1,2-DIMETHYLHYDRAZINE (DMH) ADMINISTRATION, CABBAGE CONSUMPTION AND PRIOR EXPOSURE TO DMH BY Mark John Messina Three studies were conducted to examine different aspects of 1,2-dimethy1hydrazine (DMH) induced carcinogenesis. The first study examined the DMH model of carcinogenesis for studying colon tumor liver metastases, a critical problem in cases of human colorectal cancer. The present study is the first to investigate the metastastic potential of intrarectal DMH administration. Findings from this study suggest that in male Sprague-Dawley rats, the dose of DMH influences tumor site while the route of DMH administration influences tumor metastases. However, metastases was primarily limited to the mesentery and therefore, intrarectal administration of DMH is not an effective method for inducing colon tumor liver metastases. Alternative animal models need to be developed. The second study was undertaken because of recommendations by two scientific bodies to increase the consumption of cruciferous vegetables as a means of reducing the risk of cancer. Two separate experiments examined effects of cabbage consumption on DMH induced colon carcinogenesis in CFl male mice. In the first experiment, cabbage consumption significantly increased both colon tumor incidence and number. Results of the second experiment suggested that cabbage exerts its effect on carcinogenesis only during promotion. Cabbage consumption did not induce xenobiotic metabolizing enzymes, indicating they were not involved in the enhanced carcinogenesis. Recommendations to increase cruciferous vegetable consumption may have been made prematurely. The last study was initiated after observing that mice appear to develop an increase in tolerance to DMH toxicity. Findings showed that prior exposure with 30 mg of DMH/kg body weight results in a two fold increase in the medium lethal concentration for DMH. In addition, in comparison to control mice, mice pretreated with DMH exhibited decreased levels of DNA methylation in response to a test dose of DMH. The increased tolerance and decreased methylation in pretreated mice appears to be mediated through a change in DMH metabolism. The altered metabolic pattern may result from the decreased levels of hepatic cytochrome P450 in pretreated mice, indicating this enzyme complex may play a role in DMH metabolism. Findings suggest that the commonly used DMH dosing regimen to initiate tumors may warrant re-evaluation. DEDICATION To my parents, Carmen and Eileen Messina for giving me the love and encouragement I needed to pursue my educational goals. 11 ACKOWLEDGEMENTS I would like to sincerely thank my advisor, Dr. Maurice Bennink, for his guidence and assistance during my degree program. I also wish to thank the members of my doctoral committee: Drs. Justin McCormick, Matthew Zabik, Maija Zile and in particular, Dr. Dale Romsos. In addition, I want to thank Dr. Thomas Mullaney for performing the histologic analyses for this dissertation. Finally, I want to express my graditude to my wife, Ginny Messina. Her constant encouragement and dedication made possible the achievement of this goal. Her role as companion, counselor, and friend during very difficult times was more than could reasonably be expected. She has been and continues to be the most important person in my life. I can only hope that I am able to show my appreciation for all she has given. iii TABLE OF CONTENTS Page LIST OF TABLES .OOOOOOOOOOOOOOOOOOOOOO0.00.00.00.00 Vi LIST OF FIGURES ................................... Vii INTRODUCTION ...0.................................. 1 CHAPTER ONE - REVIEW OF LITERATURE ................ 3 Experimental Colon Cancer ..................... 5 Dimethylhydrazine Metabolism .................. 6 Nutrition and Cancer .......................... 9 Dietary Fat ................................... 9 Dietary Fiber ................................. 15 Bile Acids and Colon Cancer ................... 19 Cruciferous Vegetables ........................ 21 Rational for Conducting Dissertations ......... 28 CHAPTER TWO - EFFECTS OF ROUTE OF 1,2-DIMETHYL- HYDRAZINE (DMH) ADMINISTRATION ON EXPERIMENTAL COLON CANCER .................................... 30 Introduction .................................. 31 Materials and Methods ......................... 32 Results ....................................... 34 Discussion .................................... 37 CHAPTER THREE - EFFECTS OF CABBAGE CONSUMPTION ON 1,2-DIMETHYLHYDRAZINE INDUCED COLON CARCINOGENESIS IN CFl MALE MICE ......... ..... .. . 42 Introduction .................................. 43 Materials and Methods ......................... 45 Results ....................................... 51 Discussion .................................... 53 CHAPTER FOUR - EFFCTS OF PRIOR TREATMENT WITH 1,2- DIMETHYLHYDRAZINE (DMH) ON DMH INDUCED DNA METHYLATION ..............................0...... 60 iv Page Introduction .................................. 61 Materials and MethOds .O....................0.0 62 Resu1ts .0..................................... 67 DiSCUSSion 0................................... 70 SUMMARY AND CONCLUSIONS ........................... 78 REFERENCES .0...................................... 83 LI ST OF TABLES Table Page Chapter 2 1 Tumor frequency in rats given 20 weekly administrations of DMH ........................ 35 2 Tumor metastases in rats receiving 20 weekly administrations of DMH ........................ 36 Chapter 3 3 Compostion of diets ........................... 47 4 Tumor occurance in large intestines of mice treated with DMH and fed one of six diets ..... 52 5 Tumor occurance in large intestines of mice treated with DMH and fed the control or cabbage diet .......................................... 54 6 Effects of dietary treatment on levels of hepatic cytochrome P450 (P450), UDP-glucurono- syltransferase (UDPGT), ethoxycoumarin O- deethylase (ECD), alcohol dehydrogenase (ADH), and hepatic glutathione S-transferase (GST) and intestinal glutathione S-transferase (IntGST) .. 55 Chapte 4 7 Effect of pretreatment with DMH on the medium lethal concentration for DMH in mice ........... 68 8 Percentage of DMH metabolized to azomethane and and carbon dioxide in control DMH-pretreated mice after injection with [ C]DMH ............. 69 9 Effect of pretreatment w'th DMH on the formation of 7-methylguanine and O -methy1guanine in livers of mice administered DMH ................ 71 10 Effect of pretreatment with DMH on hepatic CYtOChrome P450 content .0...................... 72 vi LIST OF FIGURES Flgure Page Chapter 1 1 Metabolism of 1,2-dimethylhydrazine ........... 8 Chapter 3 2 Experimental Design (experiment 2) ............ 49 vii INTRODUCTION In the United States, one of four Americans will develop cancer and one of five will die from it. Cancer is the second leading cause of mortality in this country, killing an estimated 500,000 people annually (1). Health surveys indicate that the threat of cancer is uppermost in the minds of the American public. Although the public fear of cancer has increased in recent years, the age adjusted cancer incidence and mortality rates, with the exception of cancer of the respiratory tract, have remained about the same (1). Two aspects of cancer control have become increasingly evident. First, treatment success depends greatly on early detection. As example, early detection of colon and rectal cancer has been shown to reduce the mortality from these malignancies (2). Second, dietary modification may represent a significant means of cancer prevention. On the basis of epidemiologic data, it has been estimated that 35% of cancer is diet related (3). Consequently, there has been an increased emphasis on identifying dietary components influencing carcinogenesis. Cancer of the colon is one of the types of cancer thought to be influenced by diet. There are approximately 500,000 new cases of colorectal cancer in the world yearly (4). [Epidemiologic findings have indicated that a high fat and low fiber diet is related to an increased risk for colon cancer (5, 6), whereas several case control studies have found consumption of cruciferous vegetables to be associated with a decreased risk (6-8). To enhance our understanding and to establish a causal relationship between a particular dietary component and colon cancer, animal models have been developed. For inducing experimental colon cancer, 1,2-dimethylhydrazine (DMH) (9) and DMH derivatives (10) are frequently used. Studies of experimentally induced colon cancer may provide insights how to significantly reduce both the incidence of and mortality from human colon cancer. CHAPTER 1 ... REVIEW OF LITERATURE REVIEW OF LI TERATURE Epidemiologists believe that 60 to 90% of human cancer is related to enviromental factors (11); therefore, a significant percentage of cancer may be potentially avoidable. Interest in the relationship between diet and cancer has been stimulated by epidemiologic studies in which large differences in cancer incidence rates have been found among countries (12, 13). Findings from these studies, along with data from migration studies, have suggested that these differences are largely diet related. For instance, the 0.5. population has a low incidence of stomach cancer and a high incidence of colon and breast cancer (1). In contrast, the Japanese have a high incidence of stomach cancer and a low incidence of colon and breast cancer (14). Correlations have been found between per capita fat intake and the national rate of breast cancer mortality and between per capita fat intake and mortality from colon cancer (12). One problem with stating conclusions based on these data, is that there are many differences among countries, not just differences in dietary customs. For example, the association that exists between per capita fat intake and the incidence of breast and colon cancer, can also be seen between gross national product and breast and colon cancer (12). The extent to which genetic disposition contributes to S differences in cancer incidence among countries can generally be determined by studies of migrant populations and of secular trends within countries. In general, populations migrating from an area with one pattern of cancer incidence rates will acquire incidence rates characteristic of their new location (15). In the case of Japanese immigrants to the United States, within approximately two generations, the colorectal cancer incidence approaches that of the United States level (16, 17). The rapid rise in colorectal cancer incidence among immigrants rules out the possibility that genetics is a main factor for differences among colon cancer among countries. The increase in cancer incidence closely paralleles adoption of the host countries dietary patterns. EXPERIMENTAL COLON CANCER Colonic tumors induced by DMH, or DMH derivatives, azoxymethane and methylazoxymethanol, are the most commonly used experimental models of colon cancer. These compounds, when injected subcutaneously, induce primarily large intestinal tumors which are similar in nature to those seen in human colon cancer (18). Therefore, the DMH model is in most respects, a good model for studying human colon cancer (19). The major difference between DMH induced colon cancer and human colon cancer, is the lack of metastases. In cases of human colonic cancer, approximately 1/2 of the patients have evidence of lymphatic metastases and 1/3 have evidence of hematogenous metastases at the time of presentation. With respect to lymphatic metastases, the usual pattern is an orderly progression of tumor through the lymphatic network from the pericolonic lymph nodes and progressively up the perivascular lymphatics (20). The usual route of hematogenous metastases is initially to the liver and subsequently to the lung (21). Most studies using DMH to induce colon cancer have produced a very low incidence of metastases. The metastases that do occur are generally found in the regional lymphatics or on the peritoneal surface. Whereas liver metastases are commonly found in the human disease, they are rarely, if ever, found in DMH models. Human colon tumor liver metastasis is a critical problem in the treatment of colon cancer. Thus far, the DMH model of carcinogenesis does not appear suitable for studying metastasis. Presently, attempts are being made to develop animal models designed for the study of colon tumor liver metastases (22, 23). Dimethylhydrazine Metabolism In order to exerts its effects, DMH must be metabolically activated. It is believed that DMH induces carcinogenesis via DNA methylation (24, 25). DMH metabolism has been studied extensively, although a complete understanding of its mode of action has not yet been achieved. As intially proposed by Druckey, in 1969, the procarcinogen DMH (figure 1), is activated in vivo to azomethane, azoxymethane and methylazoxymethanol to yield formaldehyde and the reactive methyldiazonium ion (24, 25). The spontanous breakdown of the methyl diazonium ion results in the formation of a methyl carbonium ion, which is thought to be the actual methylating species. After the injection of labeled DMH in vivo, radioactive azomethane and carbon dioxide are exhaled, the latter compound presumably resulting from the spontaneous breakdown of formaldehyde (26, 27). Labeled azoxymethane and methylazoxymethanol are excreted in the urine (28) and labeled methyl groups bind to DNA and other cellular macromolecules (26, 27). Although methylazoxymethanol breaks down spontanously (29, 30), it is believed that an enzymatic conversion must also be involved to account for the narrow range of organotropism of this carcinogen. Two theories explaining DMH organospecificity have been proposed. Weisburger suggested that DMH is metabolised to methylazoxymethanol in the liver, conjugated with glucuronic acid, transported via the bile to the gut where it undergoes hydrolysis by intestinal flora, thereby directly exposing the colon to this carcinogen (31). However, evidence strongly refuting this hypothesis comes from studies showing that the biliary metabolites of DMH account for approximately only 1% of the total administered dose (32) and that these metabolites incubated with or Figure l. METABOLISM OF 1,2-DIMETHYLHYDRAZINE DI MEI'HYIHYDRAZ INE AZOMETHANE AZOXYME’I'HANE alcohol dehydrogenase ? MEI‘HYLAZOXYMETHANOL -—-——~ ME‘I‘HYIAZOXYFORMAIDEHYDE \ FORMAIDEHYDE CARBCN DIOXIDE \ v METHYLDIAZONIUM ION «¢--- METHYLAZOXYFORMIC ACID METHYL CARBONIUM ION METHYLATION OF MACROMOLECULES (From Fiala, ES., Inhibition of Tumor Induction and Develop- ment, Zedeck, MS and Lipkin M, Eds., Plenum Press, New York. 1981, 37.) without;?-glucuronidase are non-mutagenic (33). Perhaps more damaging to the credibility of this scheme, are studies showing that segments of the colon surgically removed from the fecal stream are as equally subject to DMH induced carcinogenesis as is the intact colon (34). This suggests that DMH and/or its metabolites reach the colon via the blood stream. The second theory explaining the organospecificity of DMH, focuses on the specific metabolic capability of the colon. Cytosolic fractions of various rat tissues as well as solutions of purified horse liver alcohol dehydrogenase (ADH) are able to catalyze the reduction of NAD using methylazoxymethanol as a substrate (35). The ability to stimulate methylazoxymethanol reduction of NAD in various tissues is reported to be directly correlated with their sensitivity to methylazoxymethanol-acetate induced carcinogenesis. The protein responsible for this reaction is reportedly similar to ADH. Schoental, in 1973 proposed that ADH may catalyse the oxidation of methylazoxymethanol to methylazoxyformaldehyde (figure 1), but this compound has not yet been isolated (36). ADH would contribute to the organospecificity of DMH only if there was a colon specific isoenzyme, since ADH is ubiquitous. However, a recent report by Fiala et al., showing that methylazoxymethanol is metabolized to a methylating species equally in two strains of deer mice, one with and the other without ADH, indicates that, at least in this animal, ADH 10 is not important in DMH metabolism (37). Furthermore, pyrazole, a known inhibitor of both DMH metabolism and ADH, was equally effective in both types of mice. It is clear from this discussion, that the enzyme(s) responsible for the complete metabolism of DMH are as yet unknown, as is an explanation for the organospecificity. NUTRITION AND COLON CANCER The colon and rectum are among the most frequent sites of cancer development in affluent societies with a Westernized life-style. Internationally, the U.S., Canada, the countries of northwestern Europe, Australia, and New Zealand have the highest colon cancer incidence rates (38). Although numerous dietary factors have been examined for their relationship with colon cancer, two components, dietary fat and fiber, appear to be the strongest candidates responsible for an increase in the incidence of colon cancer. DIETARY FAT Epidemiology Wynder and Shigematsu were the first to suggest that dietary fat intake may be related to the international variations in colon cancer (39). Several years later, Armstrong and Doll, found a strong correlation (r = 0.85 11 for both men and women) between total meat intake and the world wide incidence of colon cancer (12). Similarly, in a study involving 38 countries, animal fat, total meat and red meat, but not vegetable fat, were strongly correlated with colon cancer incidence (40). In Japan, recent increases in fat consumption (41) have been associated with a striking increase in rates of colon cancer (42). Seventh Day Adventists, who consume little or no meat, were found to have 1/7 the colon cancer incidence of the control population, a finding which supports the notion of a fat- associated risk for colon cancer (43). Consumption of meat, an important source of fat, was associated with large-bowel cancer among Japanese Hawaiian colon cancer patients in a study by Haenszel et a1. (44), although no such relationship was found by this investigator in Japan (45). Potter and McMichael, in an Australian study, also found a positive correlation between colorectal cancer and protein intake (2-3 fold increased risk) (46). In addition to protein, caloric intake, which the authors suggest may be indicative of a high fat intake, was also related to an increased risk. Kune et al., comparing the dietary intake of 715 colorectal cancer patients with that of 727 age matched controls in Melbourne (Australia), found total fat intake to be positively associated with colon cancer incidence in both males and females (6). However, beef intake was a risk factor only in males, while a high intake of pork and fish was 12 protective in both sexes. A recent Canadian study reported a higher intake of total fat, saturated fat, and cholesterol in colon cancer patients, although patients also consumed more total calories (47). Tajima and Tominaga, in a study involving 93 cases of colo-rectal cancer and 186 controls at the Aichi Cancer Center Hospital in Japan, found that eating a Western style breakfast for 10 years or more made a significant contribution to the risk of colon cancer, but decreased the relative risk for stomach and rectal cancer (48). Interestingly, a Puerto Rican study found patients with colon cancer had a higher intake of both fat and fiber than controls (49). In contrast to the above findings, several studies have found neither fat or meat intake to be a risk factor for colon cancer. Phillips and Snowdon examined the consumption pattern for 21 different foods and for coffee and percent of desirable weight in relation to mortality rates from colon cancer in over 25,000 Seventh-Day Adventists (50). They found that consumption of meat or poultry was not significantly related to death from either colon or rectal cancer. However, egg consumption did show a positive relationship with colon cancer. Case-control studies of colon cancer in Israel (8) and the United States (51), were unable to establish an increased risk for fat or meat intake (Israeli study) or for meat intake (American study). In fact, Graham et al., found that when comparing 13 190 colon cancer patients with 600 controls, frequent meat eaters had a relative risk for colon cancer of about 1/3 that of those who rarely ate meat (7). Similarly, Stemmerman et al., found that saturated fat intake, rather than being a risk factor, was found to be associated with a decreased risk for colon cancer (52). In summary, the preponderance of the epidemiologic data suggests that fat and/or meat intake is associated with an increased risk for colon cancer (5, 6, 12, 39-44, 46-49). However, several studies do not support this relationship (7, 8, 50-52). The inconsistency among these studies is most likely due to the complexity involved in conducting epidemiologic studies, with many uncontrolled or poorly controlled variables. At the present time, based on the available information, it is appropriate to conclude that dietary fat intake is a risk factor for colon cancer. Animal studies Early animal studies supported a positive association between dietary fat and experimental colon cancer. Reddy et al., found that rats fed a 5% corn oil diet had a greater incidence of DMH induced colon tumors than those fed 5% lard (53). When fed at the 20% level, overall tumor incidence was significantly enhanced, but the type of fat was not a factor. Dietary fat in rodents has also been shown to enhance colon cancer induced by azoxymethane, methylazoxymethanol, and methylnitrosourea (54, 55). 14 However, as has been pointed out in a number of reviews, these early studies employed diets which were not isocaloric and therefore, differences may not have been specifically related to fat intake per se, but rather to an increased caloric intake (56), or to differences in vitamin, mineral or protein intake. Consumption of a diet high in beef fat increased large intestinal tumor incidence when fed after azoxymethane administration, i.e, promotion (55). However, if animals were fed this diet before or during carcinogen administration, tumor incidence was not effected. Reddy and Maeura observed an increase in azoxymethane induced colon carcinogenesis in rats consuming diets containing 20% corn or safflower oil, as compared with animals consuming diets containing 5% oil (57). Reddy and Maruyama, investigated the effects of diets comprised of 5%, 13.6% or 23.5% corn oil or lard during or following the administration of azoxymethane (58). Corn oil increased tumorigenesis only when fed after the carcinogen was administered. When fed prior to and during azoxymethane administration, there was no effect. In contrast, both the 13.6% and 23.5% lard diets increased tumorigenesis when fed prior to and during carcinogen administration, i.e., initiation. The highest level of lard also enhanced tumorigenesis when fed during promotion. The effect of fat during initiation may be attributable in part, to the influence of fat on the metabolism of DMH by colonic microsomes (59). 15 In contrast to much of the above data, Nauss et al., found that diets composed of either 24% corn oil, beef tallow or Crisco had no effect on DMH induced carcinogenesis (60), a finding which concurs with work of Glauert and Bennink (61). In summary, findings from animal studies, in general have found fat intake to be positively correlated with colon cancer (53-55, 57, 58). Although, as one might expect, there are studies which do not support this relationship, but they are relatively few in number (60, 61). DIETARY FIBER The relationship between fiber intake and colon cancer is particularly complex because there are several types of fiber which differ with respect to their effect on this disease. Overall, the data indicates that the less fermentable and more insoluble fibers such as lignin, cellulose and hemicellulose offer protection, whereas easily fermentable and very soluble fiber either offer no protection or are associated with tumor enhancement. Epidemiology Burkitt suggested that countries consuming a diet rich in fiber have a low incidence of colon cancer, whereas those eating refined carbohydrates, with little fiber have 16 a higher incidence of this disease (62). According to a recent comprehensive review by Jacobs, 69% of the human epidemiologic studies have demonstrated an inverse association between high fiber diets and colonic cancer (63). Of greater significance, is the fact that when these studies were catagorized according to the type of fiber, an inverse association between cereal fiber and colon cancer was found in 80% of the cases, whereas fruit and vegetable consumption showed no association in 75% of the cancers. However, two British studies found that the components of fiber containing foods exhibiting the highest negative correlation with colon cancer include the pentose fraction and uronic acids (64, 65). A comparison between the incidence of colon cancer and fiber consumption in two Scandanavian countries, found that individuals from Denmark, residents of which have a four fold higher colon cancer incidence then Finland, consumed approximately 50% less total fiber (66). A follow up study found a very strong inverse relationship between cereal consumption and colon cancer incidence in four different areas in Denmark and Finland (67). Similarly, cereal fiber was found to be the only source of fiber negatively associated with colon cancer incidence among 38 countries (40). Cereal fiber had a higher negative correlation with mortality from colon cancer than that of crude fiber. Finally, Walker et al., observed a negative association between fiber intake and the incidence of colon cancer 17 among South Africans (68). Generally, findings from case control studies have not found fiber to be protective against colon cancer (47, 69— 71). Among those case control studies in which fiber has been shown to be protective, are reports showing an inverse relationship between vegetable consumption and colon cancer (45, 72). The combination of a high fiber-high vegetable diet was found to be associated with a decreased risk for colon cancer by Kune et a1 (6). However, several studies have failed to find an association between vegetable consumption and colon cancer (69, 73). Furthermore, in one study, colon cancer patients consumed more vegetables than controls (44), while in another study, there was a strong indication that cabbage consumption increased the relative risk for colon cancer (47). The contrasting findings between case control studies and those studies comparing colon cancer incidence and fiber consumption among countries indicates the uncertainity surronding this issue. Although the association between fiber intake and colon cancer has been discussed for many years, it appears the basis for this contention is quite weak. Until new findings are presented, it is not possible to say with a high degree of certainty that fiber is protective against colon cancer. Animal Studies One advantage to using animal models is that it allows 18 for testing various fiber fractions independently. However, in doing this, important interelationships among dietary fibers may be overlooked. Additionally, interpretation of the data becomes difficult when there are differences among species. Generally, wheat bran has been shown to reduce experimental colon cancer in mice (74) and rats (75-78), but Clapp et al., found that in mice consuming one of four kinds of wheat bran, tumor incidence ranged from 44% to 72% as compared to an 11% tumor incidence in control mice (no fiber) (79). Furthermore, Jacobs has recently shown that wheat bran supplementation enhances colon carcinogenesis in Sprague-Dawley rats and attributes this effect to a decreased colonic pH, resulting from an increase in bacterial fermentation (80, 81). In general, the addition of cellulose to the diets of rats has had a protective effect on DMH induced carcinogenesis (82, 83), although Ward et al., reported that the addition of cellulose had no effect on azoxymethane induced colon cancer (84). Freeman et al., found that rats consuming a 4.5% or a 9% cellulose diet had approximately one half the tumor frequency and tumor number as the fiber free group (82). No difference in these parameters was noted between the two cellulose groups. Cellulose exerted its inhibitory effect on tumor frequency during the period of carcinogen administration. In this study, dietary pectin had no effect on carcinogenesis, 19 although several studies have found that pectin enhances DMH induced carcinogenesis (85, 81). In summary, animal studies investigating the relationship between fiber intake and experimental colon cancer do not support a protective effect of fiber. For each fiber, almost without exception, are studies showing opposite effects on colon carcinogenesis. BILE ACIDS AND COLON CANCER Diet may influence colon carcinogenesis by altering bile acid concentration in the large intestine. Bile acids have a range of effects on the inteStinal cells of the colon, including hyperplasia (86) and disruption of cell membrane (87). Fecal bile acid excretion is quantitatively related to dietary fat intake (88) and it has been suggested that dietary fiber may exert a protective effect against colon cancer by diluting the concentration of bile acids (89). When administered intrarectally following a dose of the colon carcinogen N-methyl-N'-nitrosoguanidine, sodium deoxycholate, lithocholic acid, taurodeoxycholic acid, sodium cholate, and sodium chenodeoxycholate all increase tumor frequency (90-92). Fecal bile acid levels have been shown to be increased in populations with high rates of colon cancer (93, 94), in individual patients with colon cancer (95, 96) and in patients with colonic polyps (97). Fecal bile acid concentration in Denmark, which has 20 a high incidence of colon cancer, was found to be higher then in Finland, which has a low incidence of colon cancer (67). Of interest, was the finding that daily fat intake was not related to colon cancer incidence. The protective factor in this study, appeared to be the increased fecal bulk of the Finland population. In contrast to the above data, findings from several studies have failed to show a relationship between bile acids and colon cancer (66, 98). Tanida et al., found that the amount of dysplasia and the number of polyps present in the colon, correlated with a higher level of bile acid excretion (99). However, patients with the largest polyps excreted less total bile acids (99). Consequently, the authors of this study concluded that the contribution of bile acids in the development of adenomatous polyps in Japanese subjects was insignificant. Finally, Breure et al., determined total and individual bile acid concentration and found no difference between 23 patients with colonic carcinoma and 21 controls (100). Findings presented generally suggest that high fecal bile acid concentrations correlated with increased colon cancer incidences. Fecal bile acid concentration is influenced by both fat and fiber intake - increasing dietary fat increases bile acid concentration while increasing dietary fiber decreases bile acid concentration. 21 CRUCIFEROUS VEGETABLES AND CANCER A select committee on Diet, Nutrition and Cancer, sponsored by the National Research Council, in 1982 recommended increased consumption of cruciferous vegetables as a means to reduce the incidence of cancer (101). At the present time, this position is endorsed by the American Cancer Society (102). The human data supporting an inverse association between cabbage consumption and colon cancer comes from case control studies (6-8, 45, 51, 72, 103). The only study comparing cabbage consumption and colon cancer incidence among countries failed to find an association (40). Epidemiology In a study of 190 male and female colon cancer patients and 600 age matched controls, an increased risk for colon cancer was found to be associated with a decreased consumption of vegetables, in particular cabbage, but also including brussels sprouts and broccoli (7). The relative risk of those individuals consuming the highest amounts of cabbage as compared to those consuming little or none, was approximately one-third. No relationship between fiber and colon cancer was noted. An earlier study by this same investigator, also found an increased risk for both male and female colon cancer patients as the frequency of cabbage consumption decreased (51). These results are in 22 agreement with Bjelke, who found less cabbage consumption among colorectal cancer patients in comparison to age and sex matched controls (103). In an Australian study, not only was a high fiber-high vegetable diet protective, but cruciferous vegetable consumption as well (6). Modan et al., in an Israeli study, determined the frequency of consuming 243 different foods and beverages in 198 colon cancer and 77 rectal cancer cases and a comparable number of controls (8). Cancer patients consumed significantly less from the fiber group, which included 73 different foods with a high fiber content. Cabbage, in particular, was one of only 10 foods which, when evaluated separately, was consumed significantly less frequently than both the "neighborhood" and surgical controls. Manousos et al., found that colon cancer patients reported significantly less consumption of vegetables, especially cabbage, beets, lettuce and spinach (72). There was an eight fold increased risk between those consuming a high meat-low vegetable diet and those consuming a high vegetable-low fat diet. Haenzel et al., in a study involving 588 patients with colorectal cancer and 1176 hospitalized controls, found a negative association between consumption of hakusal (cabbage) and colon cancer in Japan (45). However, these authors found no such correlation in Hawaiin subjects (44). Similarly, Miller et al., reported that cruciferous vegetables had only a minor protective effect against colon 23 cancer in females and no protective effect for males (104). Finally, Tajima and Tominaga, reported a two fold increase in relative risk for colon cancer in patients consuming cabbage frequently (3-4 times/week), although the risk associated with cabbage consumption was not statistically significant (48). In summary, most case controls studies have found cabbage consumption to be protective against colon cancer (6-8, 45, 51, 72, 103), whereas only a few studies failed to find a correlation between cabbage consumption and colon cancer (44, 48, 104). The only study comparing cabbage consumption and colon cancer incidence among countries, did not find an association (40). In conclusion, on the basis of the human data, it appears cabbage consumption is associated with a decreased relative risk for colon cancer. Animal Studies Several animal studies have shown diets containing cruciferous vegetables and/or compounds isolated from cruciferous vegetables to reduce experimentally induced cancers. Wattenberg et al., found that when rats were fed a diet containing 25% cauliflower one week following dimethylbenzanthracene administration they had a decreased mammary tumor frequency in comparison to animals consuming the control diet (105). Boyd et al., determined the level of Oxfetoprotein, which is used as a marker for cancer, in Fischer 344 rats 5 weeks after the administration of the 24 liver carcinogen, aflatoxin B1 (106). Rats consuming a 20% freeze dried cauliflower diet, had one half the level of fetoprotein as rats administered aflatoxin B1 and consuming the basal diet. Similarly, the number of aflatoxin B1 induced hepatic gamma-glutamyl transaminase foci, an indication of preneoplasia, was significantly lower in male Fischer rats fed a 20% brussels sprouts diet for 3 weeks prior to and for two weeks during carcinogen administration in comparison to the basal group without brussel sprouts (107). In addition to using the whole vegetable, glucosinolate derivatives, compounds isolated from cruciferous vegetables which are thought to possess anti-carcinogenic properties, have also been shown to inhibit carcinogenesis (108, 109). Hydrolysis of indolylmethyl glucosinolate, the parent compound, produces three glucosinate derivatives: indole-3- carbinol (I3C), indole-3-acetonitrile and 3,3'- diindolymethyl (110). 13C is the most frequently tested derivative. 13C has been found to significantly reduce the frequency of dimethylbenzanthracene mammary cancer in Sprague Dawley female rats (109), benzo(a)pyrene induced gastric cancer in female ICR/Ha mice (111) and aflatoxin B1 induced liver cancer in fingerling rainbow trout (112). However, when 13C was fed to fingerling trout during promotion only, the frequency of liver cancer was enhanced. When this compound was fed during initiation only, tumor frequency was inhibited (113). 25 When looking specifically at the effect of cabbage on experimentally induced cancers and/or the effect of 13C on DMH induced colon cancer, it becomes readily apparent that very little work has been done in this area. There are only two published studies in which the effect of cabbage consumption on DMH induced carcinogenesis was examined (114, 115). However, these studies provide very little insight into this relationship. Srisangnam et al., employed a strain of mice resistant to DMH induced colon carcinogenesis, with the result that no colon tumors were produced (114). In the study by Srisangnam et al., tumors were found primarily in the spermatic cord, and to a lesser extent in the kidney and liver. The incidence of spermatic tumors was higher in mice consuming diets composed of either 10% and 20% dried cabbage, while tumor incidence in the 40% cabbage group was reduced. However, in no case were differences between the cabbage groups and controls statistically significant. Temple et al., found cabbage consumption had no effect on DMH induced colon carcinogenesis (115). However, in this study, diets contained only 13% fresh weight cabbage, a level which may have been too low to exert an effect. Although several case control studies suggested cabbage was protective against colon cancer, Pence et al., found that 13C supplementation enhanced DMH induced colon carcinogenesis in male Fischer 344 rats (116). In this study, a 2 x 4 factorial design was used to examine the 26 main and interactive effects of diets containing 15% wheat bran, 1% cholesterol with cholic acid, 20% beef tallow and 0.1% 13C. Test diets were fed throughout the experiment. Wheat bran decreased tumorigenesis while 13C was found to be the single main effect most responsible for increasing both tumOr incidence and number. 13C was also found by Autrup et al., to enhance the binding of DMH to colon DNA (117). In this study, nontumorous human colon tissues were placed in culture for 24 hr after which time labeled DMH was added along with one of the test compounds under investigation. After a 24 hr incubation period, the amount of radioactive DNA was determined. The findings indicated that 13C enhanced binding to DNA by 66% over explants incubated without 13C. Diets composed of approximately 10% cabbage fed to Syrian hamsters treated with a single dose of the pancreatic carcinogen, N-nitrosobis(2-oxopropyl)amine administered at 8 weeks of age, resulted in an increase in tumor number (118). Animals fed a high fat-cabbage diet had twice the number of tumors as animals fed a high fat diet without cabbage. The effect of consuming a 25% cabbage diet on aflatoxin B induced liver carcinogenesis 1 and liver levels of‘*~fetoprotein was examined by Boyd et al (119). In the cabbage fed group, total tumor frequency and‘41fetoprotein level was significantly lower in comparison to the control group. However, the authors of this study acknowledged two confounding variables which may 27 have biased the results; cabbage fed animals consumed less food and had significantly lower body weights. Finally, results from an unpublished study by Wattenberg, cited in another paper by this author, show that Sprague Dawley rats placed on a diet composed of 10% cabbage one week after administration of dimethylbenzanthracene, had a decreased mammary tumor incidence in comparison to the basal group (93% vs 64%) (105). Several authors have suggested that cruciferous vegetables and/or glucosinolate derivatives exert their effects on carcinogenesis through enzyme induction. For example, Pence et al., related the effect of I3C on DMH induced carcinogenesis to an increase in aryl hydrocarbon hydroxylase (AHH) (116). Wattenberg has attributed the inhibitory effect of cruciferous vegetables to an increase in AHH (109) and also to an increase in glutathione S- transferase (111). The decrease incw fetoprotein levels in rats administered aflatoxin B1 and consuming a diet containing 25% cauliflower was attributed to an increase in hepatic aminopyrine N-demethylase (106). However, in contrast, Nixon et al., found that 13C supplementation inhibited aflatoxin induced liver cancer without increasing mixed function oxidase (112). Procarcinogens require metabolism or activation to become carcinogens. In many cases, increasing hepatic cytochrome P450 leads to enhanced carcinogenesis. Frequently, when cytochrome P450 is increased, other 28 enzymes, referred to as phase II enzymes, are also. elevated. Phase II enzymes, such as glutathione S- transferase and UDP-glucuronosyl transferase act to increase the excretion of xenobiotics, a process which is generally viewed as protective to the organism. Consequently, induction of enzymes that results from the consumption of cruciferous vegetable can not be simply viewed as either beneficial or harmful. The effect of enzyme induction on carcinogenesis depends on the specific carcinogen and the extent to which the relevant enzymes are effected. For a thorough discussion of the relationship between cytochrome P450 (phase I) and phase II enzymes, and carcinogenesis, the reader is referred to a review by Wattenberg (105). In summary, animal studies have shown cabbage, cruciferous vegetables and/or glucosinolates to be protective against a number of non-colonic, experimentally induced cancers (105-109, 111, 112). There is however, some evidence indicating that carcinogenesis can also be enhanced (116, 118, 119). Furthermore, there is evidence indicating that I3C supplementation enhances experimental colon cancer (116). RATIONALE FOR CONDUCTING DISSERTATION Each of the three studies which comprise the present work, was undertaken so as to improve our understanding of 29 DMH induced colon carcinogenesis. The first study, "Effects of Route of 1,2-Dimethylhydrazine (DMH) Administration on Experimental Colon Cancer" had two aims: l) to determine the feasability of using intrarectal administration of DMH as a means for inducing colon tumor liver metastases and 2) to determine the extent to which route of administration influences DMH induced carcinogenesis. As pointed out in the literature review, researchers are actively attempting to develop an animal model suitable for studying colon tumor liver metastases. The second study, "Effects of Cabbage Consumption on 1,2-Dimethylhydrazine Induced Colon Carcinogenesis in CF1 Male Mice" was initiated because of the public recommendations by two scientific organizations to increase cruciferous vegetable consumption. Much of the justification for these recommendations is based on case control studies showing an inverse association between cabbage consumption and colon cancer. However, no animal studies have been done which either clearly prove or disprove this relationship. Finally, the third study, "Effects of Prior Treatment with 1,2-Dimethylhydrazine (DMH) on DMH Induced DNA Methylation" was undertaken as a result of observations about DMH toxicity made during the cabbage and colon cancer experiments. A complete understanding of the mode of action of DMH has not yet been achieved. Increasing our understanding of DMH metabolism, may provide insights into the nature of colon cancer. CHAPTER 2 .... EFFECTS OF ROUTE OF 1,2-DIMETHYLHYDRAZINE (DMH) ADMINISTRATION ON EXPERIMENTAL COLON CANCER 30 INTRODUCTION 1,2-dimethylhydrazine dihydrochloride (DMH) is frequently used to induce experimental colon cancer in both mice (9, 120) and rats (121, 122). The type of tumor induced by DMH resembles that seen in human colon cancer (18). However, the multiple liver metastases which occurs in approximately 20% of the human colorectal cancer cases has generally not been reported in laboratory animals administered DMH (123). Presently, researchers are attempting to develop an animal model to study the spread of colon cancer to the liver. It has not been established whether the route of DMH administration influences metastases, although there is evidence suggesting it may modulate tumor number and location (124-127). For example, Toth et al. found that Swiss mice injected subcutaneously with DMH developed colon tumors (127); colon tumors did not develop in response to DMH administered intrarectally (128) or orally via the drinking water (126). Differences in tumor location and incidence were shown by Toth et al. to be a function of route, rather than the amount of DMH absorbed (130). A second factor influencing DMH induced carcinogenesis is the presence or absence of intestinal microflora. Reddy et al., reported a 21% colon tumor incidence in female germ-free Fischer rats administered DMH in comparison to a 93% tumor incidence in conventional animals (130). In contrast, when a DMH 31 32 derivative, azoxymethane, was administered, germ-free animals had an increased colon tumor incidence (93%) compared to conventional animals (60%) (130). It is difficult to ascertain the means by which the intestinal bacteria influence colon carcinogenesis by comparing germ-free and conventional animals, since the lack of microflora profoundly affects the intestinal mucosa in a variety of ways (131). Changes in both colon structure and/or metabolism could have an impact on colon carcinogenesis. Additional consideration should be given to the likelihood that some bacterial metabolism of DMH is feasible, given the extensive metabolic capability of the intestinal microflora (132). We studied the influence of route of DMH administration on DMH induced carcinogenesis and tumor metastases to determine the feasibility of using DMH to induce colon tumor liver metastasis. Three groups of rats were either injected subcutaneously with DMH or given intrarectal instillations over a period of 20 weeks. Eight weeks following the final DMH administration, necropsies were performed on all animals and tumor type and location were noted. MATERIALS AND METHODS Animal conditions.-Male Sprague-Dawley rats (Harlan Sprague Dawley, Inc., Madison, WI) weighing approximately 200 g were placed into one of four groups. Animals were housed at 23'C in solid-bottom plastic cages (3 rats per cage) with wood 33 shavings for bedding. Rats were fed stock diet (Wayne Rodent Blox, Wayne Pet Food Division Continental Grain Co., Chicago, IL) and water ad libitum thoughout the experiment. Room lights were on from 0700 to 1900 hours daily. Experimental design.-One group of animals was injected subcutanously with 20 mg of DMH/kg body weight; two other groups were administered intrarectally 20 and 40 mg of DMH/kg body weight, respectively. The control group received vehicle only. Immediately prior to administration, DMH was dissolved in 0.1 M sodium phosphate buffer and the pH adjusted to 7.4. All animals were administered 200 ul of either the DMH solution or vehicle (buffer only) between 0800 and 1000 hours once per week for 20 weeks. Subcutaneous injections were given behind the neck. Intrarectal instillations were given by inserting a plastic tube attached to a tuberculin syringe 8 cm through the anus. Prior to administration, fecal matter was removed by gently massaging the anal area. To aid in absorption, after withdrawing the plastic tube, rats were held with head down at a 45° angle for 30 seconds to limit rectal expulsion of the DMH solution. Eight weeks following the final administration of DMH, the experiment was terminated and animals were killed by C02 asphyxiation. Post-mortem examinations were performed on all rats. Location of all macroscopic tumors in the colon and rectum were recorded. Viscera were carefully examined for evidence of metastases and representative samples of all 34 affected organs were examined histologically. Tissues were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 6 microns and stained with hematoxylin and eosin. RESULTS No difference in body weight was observed among the groups (means i SD = 503 i 28, 450 i 31, 498 i 38, 468 + 29 for the control, 20 mg subcutaneous, and 20 and 40 mg intrarectal groups, respectively). Tumors were not observed in the control group. Almost without exception, intestinal tumors in all three groups receiving DMH were classified histologically as carcinomas. The incidence of large intestinal tumors in the 20 mg intrarectal group was lower (P<0.05) than in each of the other two groups, which had a similar colon tumor incidence (table 1). Additionally, the number of colon tumors per tumor bearing rat was lower (P<0.05) in the 20 mg intrarectal group than in the 20 mg subcutaneous group, but not in comparison to the 40 mg intrarectal group. No difference in the incidence of small intestinal tumors or number was noted among the groups (table 1). In the 40 mg intrarectal group (table 2), the incidence of tumor metastases (53%) was greater (P<0.05) than in the 20 mg intrarectal group (0%), but not statistically different from the 20 mg subcutanous group (27%). In the 40 mg 35 .mo.ova azocm .eoumemcocw we om eoew ucmtmcewu Spaceuwcwcmwm o .om H memos ucommcaoc mm:_e> a .mwmmcucmcea cw :302m azocm comm :_ mung mo Lopez: m>PuomeCm e -- --- o o -- --- o o Amav .oeocou Pepomemeuc_ OH m.o “ «.1 k es mm m.H n N.N ea 6mm AmHv as as _euomeccucw e e.o « m.a m mm s 9.0 « m.a 4 mm ANHV as om msomceoaonzm oh 3.1 « o.~ m we mm w.~ a uk.s HE ooze ABEL me am some mcwcemo .oz Nev new; mercemn .oz ~m¢ «mecca _epo» cogs» ewe mucouwocfi _eooe Loss» cog mocmcwocm messap mcoe:p mcwpmmucw __eEm mcwummucw mate; :20 Co meowpetumw:_eSe x_xmmz om :m>wm moat cw mocmscmee Leash .H mynep 36 Table 2. TUmor metastases in rats given 20 weekly administrations of DMH Incidence of Groupa metastases Description (%) No. 20 mg (11) 27 3 Primarily mesentery, pancreatic subcutaneous lymph nodes and pancreas 20 mg (12) 0 0 intrarectal 40 mg (15) 53b 8 Primarily mesentery, abdominal intrarectal cavity, hepatic lymph nodes and pancreas a Effective number of rats in each group shown in parenthesis. Significantly different from 20 mg intrarectal group P<0.05. 37 intrarectal group, 8 of 14 animals with colon tumors had metastases; 4 animals had metastases which originated from the small intestine, 3 from the large intestine and in one case it was not possible to determine the site of origin. In the 20 mg subcutaneous group, 3 animals exhibited varying degrees of metastases. Metastases in two of these animals originated from the colon while in the third, it was not possible to determine the site of origin. No metastases was observed in the 20 mg intrarectal group. Metastases were found primarily in the mesentery and mesenteric lymph nodes and to a lesser extent in the pancreas, lung, spleen, diaphragm, thorax and hepatic lymph nodes. One animal in the subcutaneous group developed tumor nodules in the liver; in the intrarectal group, one animal developed tumor nodules in the hilus of the liver. Overall, metastases originating from the colon were more wide spread than from the small intestine. DISCUSSION This experiment was undertaken in order to compare two different routes of DMH administration on DMH induced carcinogenesis and tumor metastases. There exists a need for developing an animal model suitable for studying colon tumor liver metastases, since in cases of human colorectal cancer this represents a critical secondary medical complication. The basis for using intrarectal instillations was two fold: 38 1) data on metastases induced by DMH administered intrarectally are lacking and 2) prior to absorption, DMH would first be exposed to the intestinal microflora and to the colon before undergoing hepatic metabolism. These two factors may influence colon carcinogenesis. In the present study, subcutaneous injection was more effective in producing colon tumors than intrarectal administration. Colon tumor number and incidence in the 20 mg subcutaneous group were significantly greater (P<0.05) then in the 20 mg intrarectal group (table 1). No difference in colon tumor number was noted between the 40 mg intrarectal and 20 mg subcutaneous groups, even though the intrarectal group received twice the level of carcinogen. In fact, the number of colon tumors per tumor bearing rat in the 20 mg subcutaneous group was greater than that of the 40 mg intrarectal group, although this difference was not statistically significant. Clearly, these findings indicate the greater efficiency of subcutaneous injection to induce colon tumors. In contrast to the findings here, data taken from two separate studies by Reddy et al., indicated no difference between colonic tumor incidence in female Fischer rats induced by intrarectal and subcutaneous administration (130, 133). However, DMH induced carcinogenesis has been shown to differ among species, strain and sex (134), a fact which may have contributed to the difference between the studies by Reddy et al. and the present one. The difference between the two routes of administration in this study may 39 have been due to a greater amount of the DMH being made available to the animal through subcutaneous injection. Colonic absorption of DMH has been shown to be highly pH dependent (135). In order to maximize intrarectal DMH absorption in this study, the pH of the DMH solution was carefully adjusted to pH 7.4. According to Toth el., route of administration, rather then the amount of DMH administered is the critical factor for determining tumor location (129). However, the present study suggests that, at least in Sprague-Dawley rats, the dose of DMH also influences tumor site. As the dose of DMH administered intrarectally was increased from 20 to 40 mg/kg, small intestinal tumor number increased 2.5 fold while large intestinal tumor number increased over 6 fold. A similar trend was observed with respect to tumor incidence. As the dose was increased, the incidence of small intestinal tumors increased slightly less than two fold while the incidence of large intestinal tumors increased almost threefold. The importance of dose is also supported by comparing the 20 mg intrarectal group and the 20 mg subcutaneous groups. The 20 mg subcutaneous group had only 2.5 times the number of small intestinal tumors as the 20 mg intrarectal group, but over 10 times the number of large intestinal tumors. It is probable, that because of poorer absorption in the intrarectal group, the amount of DMH made available to rats in the subcutaneous group was greater, even though the administered dose was identical. It must be noted however, that this last 40 comparison is also consistent with the notion that route of administration influences tumor site. Support for the importance of dose also comes from the observation that only in the 20 mg intrarectal group were there animals with small intestinal tumors without also having colon tumors. These data suggest that the colon is relatively more responsive to an increased dose of DMH than is the small intestine. As the dose of DMH was increased, the ratio of large intestinal to small intestinal tumor incidence and number increased dramatically. Clearly, because of the limited number of animals in each group and because only 3 different groups were involved in this study, any conclusions based on these data must be drawn with caution. In contrast to the importance of dose in determining tumor site, tumor metastases appears to be influenced by the route of DMH administration. Of the 5 animals with small intestinal tumors in the 20 mg subcutaneous group, small intestinal tumor metastases possibly occured in one animal. In contrast, of the 7 animals with small intestinal tumors in the 40 mg intrarectal group, small intestinal tumor metastases was observed in 4 and possibly 5 animals. As to why small intestinal tumors metastasized when induced by intrarectal administration and not by subcutaneous injection, is not clear. Possibly, there is a difference in DMH metabolism between these two forms of administration. The colon has been shown to be capable of metabolizing DMH (136) and it is feasible that the intestinal microflora may also 41 play a role in this regard (132); intrarectal instillation may have provided more of an opportunity for this to occur. The incidence of DMH induced metastases is often cited as being approximately five percent (137); although, several researchers have reported much higher rates (122, 138-140). Metastases in these studies generally did not involve the liver, as was the case in the present study. An alternative means of inducing colon tumor liver metastases in experimental animals, such as injecting human and murine colon tumor cell lines into the spleen of nude mice have produced promising results (141, 142). In conclusion, results suggest that the dose level of DMH is a factor in determining tumor location and that the route of DMH administration influences small intestinal tumor metastases. Intrarectal administration of 40 mg of DMH/kg produced the greatest incidence of metastases but offers no advantage over subcutaneous injection for studying colon tumor liver metastases. CHAPTER 3 .... EFFECTS OF CABBAGE CONSUMPTION ON 1,2- DIMETHYLHYDRAZINE (DMH) INDUCED COLON CARCINOGENESIS IN CF1 MALE MICE 42 INTRODUCTION Colon cancer is a major form of cancer among industrialized countries (143). In the United States, colon cancer kills an estimated 50,000 people annually (1). Both genetic and environmental factors contribute to the incidence of this disease, with diet being cited as the most important environmental factor (3, 144). Among those dietary factors thought to be protective against colon cancer is cruciferous vegetables (146). Wattenberg, who conducted much of the initial animal work, found that supplementation with compounds isolated from cruciferous vegetables reduced 7,12- dimethylbenz(a)anthracene (DMBA) induced mammary cancer and benzo(a)pyrene induced stomach cancer (109). Diets containing cauliflower or cabbage were also found to reduce DMBA induced breast tumor incidence and number (105). In 1982, a report by the National Research Council, Committee on Diet, Nutrition and Cancer, recommended increased consumption of cruciferous vegetables as a means to decrease human cancer incidence (100). The American Cancer Society also supports this position (101). The strongest epidemiologic data between cruciferous vegetables and cancer is between cabbage consumption and colon cancer. Several case control studies have found a decreased risk for colon cancer with an increase in cabbage consumption (6-8, 45, 51, 71, 103), although other studies did not find such a relationship (44, 48, 104). 43 44 The limited work examining the effect of cabbage consumption on 1,2-dimethylhydrazine (DMH) induced experimental colon cancer does not support the epidemiologic data. One study, used a strain of mice resistant to DMH induced colon carcinogenesis, with the result that no colon tumors were induced (114). In a second study, the lack of significant effect due to cabbage consumption, may have been because of the low level of cabbage employed in this study (115). More significantly, diets supplemented with indole-3- carbinol (I3C), a glucosinolate found in cabbage, increased DMH induced colon carcinogenesis in male F344 rats (116). Also, when 13C was incubated along with labeled DMH, the amount of radioactivity associated with DNA in explants of human colon tissue increased (117). In addition to the work with DMH, cabbage consumption by Syrian hamsters has been shown to enhance experimentally induced pancreatic cancer (118). It is apparent that recommendations to increase cruciferous vegetable consumption were made without adequate supporting data from animal studies. The purpose of this study was to examine the effect of cabbage consumption on DMH induced colon carcinogenesis in CF1 male mice. Two separate experiments were conducted; findings from this study do not support a protective role for cabbage against colon carcinogenesis in the DMH model. 45 MATERIALS AND METHODS Animal conditions.-CF1 male mice were used for all experiments and were housed in stainless steel hanging wire cages. In the first colon cancer experiment, five week old mice weighing 10—12 g were housed 3-4 per cage. Otherwise, 8- 10 week old mice were housed individually and weighed approximately 30 grams. The animal room was temperature (23°C) and humidity controlled and was on a 12 hr light and dark cycle. Food and water were fed ad libitum. Cabbage preparation.-Cabbage was incorporated into the diets in the following manner. After removing the outer leaves, the cabbage head (green cabbage, California grown) was cut into quarters and the center core discarded. The cabbage was finely shredded, steam blanched for 2 minutes, cooled by cold running water and placed in a refrigerator. The cabbage was then dried in a convection oven for 15 hrs at 55°C. Dehydration was complete within 7 days after arrival. The dried cabbage was ground into a fine powder and incorporated into diets at two month intervals. Diets and the powdered cabbage were kept refrigerated. Freeze dried cabbage was prepared similarly to the oven dried cabbage, except that after blanching, the cabbage was stored at —20°C and then lyophilized for 48 hours. Diet formulation.-Cabbage, on a dry weight basis was 46 estimated to be 30% fiber (Paul and Southgate, 146). One- half of the fiber was assumed to be fermentable. For calculation purposes, we assummed the non-fiberous part of the cabbage to contain 4 kcal/gram. Therefore, each gram of cabbage on a dry weight basis, after discounting the fiber content, contained 2.8 kcal. The kcal and non-fermentable fiber gained through the addition of cabbage was balanced by removing appropriate amounts of sucrose and cellulose from the control diet. Each of the diets, was formulated such that the ratio of protein, fat, minerals and vitamins to kcal was the same and except for the positive control group, such that there was an equal amount of non-fermentable fiber in all groups (table 1). No correction was made for the negligible amount of vitamins, minerals, or protein in the cabbage, nor for the potential energy arising from the absorption and metabolism of organic acids produced by the bacterial fermentation of the fiber. Carcinogen administration.-l,2-Dimethylhydrazine dihydrochloride (Aldrich Chemical Co.) was dissolved in 1 mM EDTA and neutralized with sodium bicarbonate immediately prior to administration. For both colon cancer experiments, each animal received 8 weekly subcutaneous injections of 30 mg of DMH/kg body weight in approximately 200 ul. Colon cancer experiments.—In the first experiment, mice were fed one of 6 diets starting three weeks prior to carcinogen 47 Composition of Diets Table 1. g dried cabbage/100 grams diet 2.6.42. 29-_1 13.4 0 0% fib/cab Ingredient 7987083242 oooooooooo 3758164100 2 11 2 7984013242 o 0.000.... 3754204100 2 12 2 7981043242 coo-cocoo- 3750334100 2 13 1 7988073242 oooooooooo 3755464100 2 13 7985003242 0 o o o o o o o o 0 3751504100 2 14 0078005343 oooooooooo 5863004100 2 14 Beef Tallow Sucrose Methionine Choline Cellulose Casein Corn oil Cabbage Mineral* ‘Vitamins* 100.0 100.0 100.0 100.0 100.0 100.0 total *AIN-76 TM Mineral and Vitamin mixes 48 administration (table 1). Nine months following the initial injection of DMH, animals were killed by CO asphyxiation. 2 In the second experiment, four groups of mice and two diets, the control and the 13.4% cabbage diet were employed. The experimental protocol is shown in figure 1. Groups 1 and 4, starting three weeks prior to carcinogen administration, were fed the control and cabbage diet, respectively. Groups 1 and 4 remained on these diets throughout the study. Group 2, the - L initiation group, and group 3, the promotion group, received the cabbage and control diets respectively, starting three weeks prior to carcinogen administration. Fourty-eight hours ! following the final injection of DMH, groups 2 and 3 were switched to the control and cabbage diets for the remainder of the experiment. Six months following the first DMH injection, mice were killed by CO asphyxiation. For both 2 experiments, immediately after killing, the large intestine, from the ileocecal junction to the anus was removed. Colons were opened longtitudinally, flushed with saline, pinned flat and placed in neutral buffered formalin. The sites of suspected tumors were recorded. Macroscopically abnormal tissue was embedded in paraffin, sectioned at 6 microns and stained with hematoxylin and eosin. Colons were coded such that tumor identification by the pathologist was done without knowledge of the dietary treatment. Enzyme experiment.-Group 1, received the control diet and group 2, the 13.4% oven dried cabbage diet. Group 3 received 49 Figure 1 Engaerimental Design (Ebcperiment 2) HEBEND WControl diet m Cabbage diet Group 1 Group 2 Group 3 Group 4 ‘ Weekly 5. c. injection of 30 mg of 1. 2-dimethylhydrazine/kg W v , W _ _ _ W _""_. / AW _ _ .. HH‘HH , I 7’ \\ ‘ — .7 . j .l . ’\ \ \ '7” / 7!.|/\/1\‘\ \ \‘ ‘ \x,‘ ‘\\ , \ /‘. /\ \w W - AMA/W ’ . W 11111111 ([\'\/ XV/ /'/\//\//"' ’_\’é"f I. >\' 11mm 6 months 50 the same diet as group 2, except that the cabbage was freeze dried, as described under cabbage preparation. To induce enzymes in the positive control, group 4, mice received an i.p. injection of 80 mg of sodium phenobarbitol/kg body weight for 3 consecutive days prior to being killed (147). Group 4 was fed the control diet. All animals were fed their respective diets for three weeks. Immediately after being killed, livers were removed, rinsed with cold saline and homogenized in four volumes of buffer (0.1 M tris buffer, (pH 7.4), 1 mM EDTA, 1 mM dithiothreitol and 250 mM sucrose). After centrifuging the crude homogenate for 20 min at 9,000 x g at 4'C, aliquots of the supernatant (S9) were removed and frozen at -20'C until analyzed. The remaining supernatant was then centrifuged at 100,000 x g for 75 min at 4'C. After removing the cytosol, the microsmal pellet was resuspended in a 10 mM tris HCl (pH 7.4), 1 mM EDTA and 20% glycerol buffer. The intestinal S9 fraction was obtained in the following manner. The large intestine was placed on a glass plate on ice with the mucosal side up. Mucosal cells were collected by gently scraping the mucosal surface with the edge of a microscope slide 3 times. The mucosal cells from 2 animals were combined and homogenized in 9 volumes of the tris-sucrose buffer. Cytosolic, microsomal and 89 fractions were frozen at -20°C until analyzed. Hepatic cytochrome P450 content was determined by the method of Omura and Sato, using 91 mM-1 cm.1 as the molar 51 extinction coefficient for the dithionite-reduced cytochrome P450 complex (148). Liver alcohol dehydrogenase activity was measured in a Beckman Lambda 4B spectrophotometer by monitoring the rate of NAD conversion to NADH at 340 nm (149). Microsomal ethoxycoumarin O-deethylase activity was determined by a modification of the fluorimetric method of Greenlee and Poland, as described by Chang and Bjeldanes (150). The concentration of 7-OH-coumarin was determined fluorometrically (excitation 368 nm, emmission 456 nm) using a Varian SF-330 spectrofluorimeter. Sample quantification was based upon a calibration curve of 7-OH-coumarin, which was exposed to the identical procedure as samples. Glutathione S-transferase activity was measured colorimetrically using O-dinitrobenzene as substrate (151). Finally, UDP-glucuronosyltransferase activity was assayed spectrophotometrically using 2-aminophenol as substrate (152). Microsomal, cytosolic, and S9 protein concentration was determined by a modification of the Lowry procedure as described by Markwell et al (153). RESULTS In the first colon cancer experiment (table 2), the 20.1% cabbage fed group had more tumors per mouse (5.7 i.5.3 vs 2.4 i 2.1) and the 26.8% cabbage fed group a higher tumor incidence (96% vs 72%) than the control group (P<0.05). No other differences among the groups were noted when individual 52 Table 2. Tumor occurance in large intestines of mice treated with DMH and fed one of 6 diets Mice*with All colon tumors a Body colon tumors Per Per tumor Group ‘Wt(g) No Percent mouse bearing mouse Positive control (13) 45:6.8 10 77 2.2: 1.9 2.8 1 1.7 0% Cabbage (31) 44:5.4 22 72 2.4: 2.1 3.4 i 1.7 6.7% Cabbage (18) 45:4.3 17 94 4.1: 3.4 4.3 1 3.4 13.4% Cabbage (11) 41: 8.3 11 100 4.1: 2.4 4.1 i 2.4 20.1% C C Cabbage (19) 42: 5.1 16 84 5.7 i 5.3 6.4 i 5.2 26.5% C Cabbage (24) 40 14.4 23 96 4.3 i 3.7 4.5 _+_ 3.6 Combinedb c c c cabbage (72) 4215.4 67 93 4.6 i 3.9 4.9 + 3.9 aPositive control lacked both cellulose and cabbage. Percent cabbage based on dry wt, g/100 g diet (for diet descriptions Effective number of mice shown in parentheses. values represent means 1 SD. see table 1). C alues represent all cabbage consuming mice. Significantly different from control group P<0.05. 53 comparisons were made. Approximately 15% and 85% of the total tumors were classified as adenocarcinomas and adenomas, respectively. There was no difference in this parameter among groups. Because no difference in tumor number or incidence was observed among the cabbage groups, these groups were combined and compared to the control group. Cabbage fed mice had a higher tumor incidence (93% vs 72%) and more tumors per mouse (4.6 i 2.4) than control mice (P<0.05). In the follow up experiment (table 3), there were no statistically significant differences among the groups. However, there was an indication of an increased tumor number in groups 3 and 4. Approximately 10% and 90% of the total tumors were classified as adenocarcinomas and adenomas, respectively. There was no difference in this parameter among groups. Cabbage consumption did not influence enzyme activity (table 4). In comparison to the control group, mice administered sodium phenobarbitol had increased levels of hepatic UDP-glucuronosyltransferase (P<0.05), glutathione S- transferase (P<0.01), cytochrome P450 (P<0.01) and ethoxycoumarin O-deethylase (P<0.01). DISCUSSION Our studies demonstrate that cabbage consumption enhances DMH induced colon carcinogenesis and that cabbage likely mediates its effect during promotion. In the first 54 Table 3. Tumor occurance in large intestines of mice treated with [NH and fed the control or cabbage diet Mice with All colon tumors a Body colon tumors Per Per tumor Group Wk(g) No. Percent mouse bearing mouse Group 1 (control) (33) 43 15.4 13 40 2.2: 1.5 0.9: 1.5 Group 2 Initiation (21) 42 14.7 11 52 2.1: 1.1 1.1: 1.3 Group 3 Promotion (33) 42 :5.7 16 49 3.8: 2.9 1.8: 2.8 Group4 Init-Prom (30) 41 15.3 15 50 3.2: 2.1 1.6: 2.2 aMice consumed either the control or 13.4% oven dried cabbage diet (table 1). Groups 1 and 4 consumed the control and and cabbage diets respectively throughout the experiment. Groups 2 and 3 consumed the cabbage and control diets respectively three weeks prior to and during the 8 weeks of carcinogen administration. Fourty-eight hours following the final injection of DMH, groups 2 and 3 were switched to the control and cabbage diets. Effective number of mice shown in parentheses. values represent means :_SD. There were no significant differences among groups (P>0.05) 55 .mo.ova azoom _ooucoo on» gene ucmovoeec x—ucoowu_cmemo .Ho.ovo azoom _ooucoo on» son» ucmowuu_o »_ucouwe_cm_m .e mo: :much too can :mwucH oomoxm o maommo __o to» m we) azoom _opwooooococa e:_oom no» meoPHocpaomumo Co topazz .o mo; pmuucH to» use NH mo; pmwucm ugmoxm mxommo __o Lou moaoom mmooooo omega m~mmoo one mmoooou omega cm>o ._ooucou no» meowuocweCmumo Co amassz .om H mcome moo mm:_o> .muooumozm mo mcmNcmooopwcpouo a:_m: .:PE\:quooa me\ow5com mm_oe: u Imoucu oco emu m:Pe\crmuooa me\omeLou zoov as a Nob em a one ee.o a Nu.o mm.o « ma.e em.o a me.u eo.o n 94.3 N aaoeo A_otoeoov mm H was on a one om.o n bu.o em.o n bN.e mu.o a be.b No.3 4 om.o u aaoee pmopco hmw :94 com puma: omen msooo cw omm_p_>wuoo machm Ahmwpcav wmoomumcoopum mco_zuop:_m _ocwummpcp uco Apmov uwuoam; oco .Azom_ co newswomou Aooumeo Co muomuwm .e m_oop 56 experiment, in comparison to the control group, the 26.8% cabbage group had a significantly higher tumor incidence and the 20.1% cabbage group significantly more tumors per mouse. Overall, cabbage fed mice had significantly more tumors per mouse and a higher tumor incidence than control mice. Because there was difference among the cabbage groups in experiment 1, the 13.4% cabbage diet was chosen as an intermediate level and used in the follow up experiment. To decrease tumor incidence, animals were killed 6 months, rather then 9 months (as in experiment 1), following the first injection of DMH. Although there were no statistically significant differences in tumor frequency among the groups in experiment 2, there was an indication of an increase in tumor number in group 3, which received the cabbage diet after carcinogen administration i.e., during promotion, and in group 4, which received cabbage throughout the experiment i.e., during initiation and promotion. These data conflict with findings from case control studies showing a protective effect of cabbage against colon cancer (6-8, 45, 51, 71, 103). However, the present study agrees with reports that 13C enhances DMH induced colon carcinogenesis in F344 rats (116) and increases DNA methylation in explants of colon tissue (117). Additional support, although less convincing because of a lack of statistical significance, comes from findings by Srisangnam et al., that mice consuming diets comprised of 10% and 20% cabbage had an increased number of DMH induced spermatic 57 tumors (114) and by Temple and El-Khatib, that cabbage fed female mice had twice the incidence of DMH induced colon tumors as controls (115). Support can also be found for the the suggestion of the present study that cabbage exerts its tumor enhancement effect during the promotion phase of carcinogenesis. Nixon et al., found that 13C supplementation during promotion enhanced aflatoxin B1 induced liver cancer in fingerling trout (154). And Birt et al., observed that I3C increased ornithine decarboxylase activity in mouse epidermis treated 1 with 12-O-tetradecanoylphorbo1-13-acetate (155). This study demonstrates that cabbage consumption has no effect on several xenobiotic metabolizing enzymes in liver or large intestine of CF1 male mice. These enzymes have been cited for their role in carcinogenesis (156, 157), but a specific role in DMH induced carcinogenesis has not been established. However, cytochrome P450 has been implicated in DMH metabolism (37) and Pence et al., attributed the enhancing effect of 13C on DMH induced carcinogenesis to an increase in aryl hydrocarbon hydroxylase (AHH) (116). AHH was not determined, although cytochrome P450 content, which was measured, often parallels AHH activity. The lack of effect of cabbage indicates these enzymes did not play a role in the enhanced carcinogenesis in cabbage fed mice, an observation in line with our finding that cabbage likely influences DMH induced colon carcinogenesis only during promotion. Other studies have also reported changes in S8 carcinogenesis in response to consumption of cruciferous vegetable and/or I3C consumption without observing a change in enzyme activities (112, 114, 158-160). The lack of enzyme induction by cabbage feeding agrees with findings from several studies. Diets fed to C57Bl6 mice containing 20% cabbage failed to increase hepatic cytochrome P450 (115, 161), AHH or epoxide hydrolase (161). Uotila et a1, fed diets containing 2.5% and 100% cabbage to Sprague— Dawley rats and saw no effect on either ethoxycoumorin O- deethylase or UDP-glucuronosyltransferase at either level, although there was a 60% increase in epoxide hydrolase activity at the 100% cabbage level (162). Liver glutathione S-transferase activity did increase significantly (3.5 fold) in response to a 50% cabbage diet fed to Coturnix quail (163), but this effect was not observed in ICR/Ha mice fed 20% cabbage (164). The finding that cabbage consumption does not reduce and may actually enhance experimentally induced colon cancer is significant because of public recommendations to increase the consumption of cruciferous vegetables. The effects of cabbage and/or glucosinolates on carcinogenesis appears to be carcinogen specific. Consequently, making recommendations based on the results with just a few carcinogens affecting only a limited number of tissues was premature. Also, since cabbage consumption appears to enhance the promotion phase of colon carcinogenesis, which is less carcinogenic specific, the consumption of large amounts of cabbage is contrainicated 59 because of the potential to promote colon cancer. Although cabbage is not likely to produce harmful effects in the amounts generally consumed, toxic effects are produced when cabbage is consumed in large amounts. Allyl isothiocyanate, a naturally occuring compound found in cabbage was found to be fetotoxic in Holtzman rats (166), to cause transitional cell papillomas in urinary bladders of male F344 rats (166) and to be mutagenic in the Ames test (167). Metabolites of glucosinolates have produced poor growth, hepatic and renal lesions (168), decreased B-carotene levels (169) and decreased protein synthesis in laboratory animals (170). Finally, the consumption of large amounts of cabbage has been responsible for development of goiter in humans (171). Because public recommendations to increase cruciferous vegetable consumption conflict with much of the animal work, continued research of this issue is needed. CHAPTER 4 .... EFFECTS OF PRIOR TREATMENT WITH 1,2- DIMETHYLHYDRAZINE (DMH) ON DMH INDUCED DNA METHYLATION 6O INTRODUCTION 1,2-Dimethylhydrazine dihydrochloride (DMH) is frequently used to induce experimental colon cancer in both mice (9, 120) and rats (121, 122). It is believed that DMH induces tumors through DNA methylation. In order to exert its effects, DMH must be metabolically activated. As initially proposed by Druckey et al., the parent compound DMH is converted into a methyl carbonium ion, the ultimate carcinogenic agent (24, 25). In the metabolism of DMH, two gaseous metabolites, azomethane and carbon dioxide are produced. These gases provide a convenient method for evaluating DMH metabolism (172). Several compounds have been shown to inhibit the metabolism of DMH or of DMH derivatives (173-175). Inhibition of metabolism, leads to a reduction in DNA methylation (173, 176) and tumorigenesis (177, 178). Although there exists a considerable body of knowledge on DMH metabolism, a complete understanding remains elusive. We observed in preliminary studies, that pretreatment with DMH decreases the toxicity of a subsequent treatment with this carcinogen. We found we could produce almost 100% survival in mice given a dose of DMH two to three times the normal lethal dose, if these mice were first exposed to a series of gradually increasing doses of DMH. In order to evaluate the mechanisms underlying this phenonomon, we offered two hypotheses: 1) sensitive mice 61 62 are eliminated after the initial injections of DMH and 2) mice adapt metabolically to reduce DMH toxicity. To test the first hypothesis, we determined the medium lethal concentration (LCSO) for DMH in two groups of mice, one of which was previously exposed to DMH. To test the second hypothesis, we examined the effect of pretreatment on DMH metabolism. Because alterations in DMH metabolism should be reflected by changes in the amount of DNA methylation, we also examined levels of 7-methylguanine and 06- methylguanine in response to a test dose of DMH. Finally, we measured levels of hepatic cytochrome P450 to determine whether this enzyme complex changes in response to pretreatment. MATERIALS AND METHODS Animal conditions-For all experiments, 8-10 week old male CF1 mice (Harlan Co., Madison, WI) weighing approximately 30 grams were housed individually in stainless steel hanging wire cages. The animal room was temperature (23°C) and humidity controlled and was on a 12 hour light and dark cycle. All mice were fed a semipurified diet (modified AIN- 76 diet) and water ad libitum (179). The composition of the diet when expressed as g/lOO diet was as follows: casein, 23.7%; DL-methionine, 0.4%; sucrose, 41.5%; cellulose, 5.0%; corn oil, 7.9%; beef tallow, 15.8%; mineral mix, 4.3%; vitamin mix 1.2% and choline chloride, 63 0.2%. Choline chloride and DL-methionine were purchased from Sigma Chemical Co. (St. Louis, MO) and casein, cellulose, mineral and vitamin mix from 0.8. Biochemical Co. (Cleveland, OH). Carcinogen administration. For all injections, DMH (Aldrich Chemical Co.; Milwaukee, WI.) was dissolved in 1 mM EDTA and neutralized with sodium bicarbonate immediately prior to administration. DMH was injected s.c. in a volume of approximately 200 ul between 0800 and 1000 hours. -The LC Medium lethal concentration (LC for DMH was 50’ ° so determined in two groups of mice. One group, composed of 90 mice, referred to as pretreated-30 (PT-30), was given a s.c. injection of 30 mg of DMH/kg body weight one week prior to conducting the LC The second group, containing 50' the control mice, was not previously exposed to DMH. Eighty-five of the surviving eighty-nine PT-30 mice were divided equally into 5 groups and given a s.c. injection of DMH at a level of either 40, 55, 70, 85 or 100 mg/kg body weight. Fifty control mice were divided equally into 5 groups and given a s.c. injection of DMH at a level of either 28, 32, 36, 40 or 44 mg/kg body weight. One animal from each dose was injected with DMH before a second animal at that same level was injected. Animals alive 7 days following the test dose were counted as survivors. The LCSO for DMH was calculated using the probit transformation 64 least likelihood method (180). DMH metabolism. DMH metabolism was studied in three groups of mice: the control group, a second group referred to as pretreated-20 (PT-20) and a third group, referred to as pretreated-100 (PT-100). PT-20 mice received three weekly s.c. injections of 20 mg of DMH/kg body weight. PT—lOO mice received four weekly stepwise s.c. injections of 30, 50, 75 and 100 mg of DMH/kg body weight. The final injection of DMH during the pretreatment period, for both the PT-20 and PT-100 mice was given 7—10 days prior to studying DMH metabolism. DMH metabolism was studied in control and PT-100 mice at two levels of DMH. Mice were injected with either 20 mg of [14C]DMH (250 uCi/mmol, New England Nuclear, Boston, MA) or 100 mg of [14C]DMH (50 uCi/mmol) per kg body weight. PT-20 mice received only the lower dose of [14C]DMH. Immediately after injection of the labeled DMH, mice were placed in an air tight chamber. Exhaled air was drawn by vacuum through a series of four flasks for a period of six hours. The collection system was essentially that as described by Fiala et al. (172). The first two flasks, designed for azomethane collection, each contained 350 ml of pure ethanol. The second flask was surrounded by dry ice. Aliquots of 0.4 ml were taken from each of these two flasks every 60 minutes for the first three hours. Contents of the second flask were replaced with fresh pre- cooled ethanol each hour. The third and fourth flasks were 65 for carbon dioxide collection and each contained 350 ml of 1 M NaOH. At the end of the 6 hour collection period, 0.2 ml aliquots were taken from the flasks containing NaOH. In all cases, aliquots were immediately placed in glass scintillation vials containing 15 ml ACS scintillation fluid (Amersham Co, Arlington Heights, IL). Total radioactivity was determined using a Beckman liquid scintillation counter. Results, based on radioactivity, are expressed as the percentage of total DMH exhaled as azomethane and carbon dioxide. DNA methylation. Nine hours following a s.c. injection of 100 mg of DMH/kg body weight, PT-100 and control mice were killed via carbon dioxide asphyxiation. Livers were immediately removed, rinsed in cold saline and stored at -80°C. DNA was isolated by a modification of the method of Marmur (181). Livers were thawed and homogenized in 8 volumes of saline citrate buffer (0.15 M NaCl, 0.015 M trisodium citrate pH 7.0) and then centrifuged at 8,700 x g for 15 min at 0°C. The resulting pellet was dispersed in 8 volumes of 1 M NaCl and after adding 10% sodium dodecyl sulfate to a final concentration of 0.07% was placed on ice for 45 min. After the addition of 1/2 volume of chloroform:isoamylalcohol (10:2), tubes were vortexed for 7 min and then centrifuged for 10 min at 12,000 x g at 0°C. After repeating the chloroform:isoamylalcohol extraction step, 50 ug of RNAse/ml (bovine pancreatic ribonuclease (Type l-A) Sigma Chemical Co., St. Louis, MO) was added to 66 the aqueous layer. A third extraction step followed the 30 min RNA digestion. The DNA was precipitated by the addition of 2 volumes of ethoxyethanol and then washed three times in 2 ml 95% ethanol. After drying under nitrogen for 30 min, the DNA was frozen at -20°C. DNA methylation was quantitated using a modification of a procedure described by Herron and Shank (182). The isolated DNA was hydrolyzed in 1 N HCl at 37°C for 18 hours. The hydrolyzed DNA was solubilized in mobile phase (5 mg/ml) and approximately 0.2 mg DNA per injection was separated via HPLC using a strong cation exchange column (Partisil - 10 SCX, 25 cm x 4.5 mm id, Whatman Inc. Clifton, New Jersey), a mobile phase of 0.1 M ammonium phosphate, pH 2, and a flow rate of 2.0 ml per minute. Elution of the methylated bases, 7-methylguanine and 06- methylguanine was monitored spectrofluorimetrically (excitation 295, emission 370) and guanine was detected spectrophotometrically at 276 nm. Calibration curves of concentration versus peak area for guanine, 7-methylguanine and 06-methylguanine were used to establish regression equations for sample calculation. Hepatic cytochrome P450. Immediately after carbon dioxide asphyxiation, livers from control and PT-100 mice were removed and homogenized in 4 volumes of buffer (0.1 M tris buffer (pH 7.4), 1 mM EDTA, 1 mM dithiothreitol and 250 mM sucrose). After centrifuging the crude homogenate for 20 67 minutes at 9,000 x g at 4°C, the resulting supernatant was centrifuged at 100,000 x g for 75 minutes at 49C. The microsomal pellet was resuspended in a 10 mM tris buffer (pH 7.4) containing 1 mM EDTA and 20% glycerol. Hepatic cytochrome P450 content was determined by the method of Omura and Sato using 91 mM-1 cm"1 as the molar extinction coefficient for the dithionite-reduced cytochrome P450 complex (148). Microsomal protein concentration was determined by a modification of the Lowry procedure as described by Markwell et al (153). RESULTS Mice which received one injection of 30 mg of DMH/kg body weight 7 days prior to being tested for the LCSO' had a 2 fold higher LC 0 (P<0.05) than the control mice (table 5 1). The LC for PT-30 mice was 77 mg/kg i_1 versus 37 50 mg/kg :.1 for control mice. Mice receiving a lethal dose of DMH generally died between 36 and 48 hours following the DMH injection. DMH metabolism, as indicated by azomethane and carbon dioxide exhalation, is shown in table 2. In mice previously treated with DMH, the amount of exhaled azomethane increased and the amount of carbon dioxide derived from DMH decreased relative to control mice. When administered a test dose of 20 mg of [14C1DMH/kg body weight, PT-20 and PT-100 mice exhaled approximately 1.5 68 Table 1. Effect of pretreatment with DMH on the medium lethal concentration for DMH in mice Group3 and Dose b c Treatment (mg DMH/kg Bwt) Mortality (%) LC50 28 0 (0/10) 32 0 (0/10) Control 36 30 (3/10) 37 : 1 40 90 (9/10) 44 90 (9/10) 40 0 (0/17) 55 29 (5/17) PT-30 7o 47 (8/17) 77d: 1 85 58 (10/17) 100 70 (12/17) a mice received a s.c. Control mice were not previously exposed to DMH. PT-30 injection of 30 mg DMH/kg body weight one week prior to the LC determination. Percentage of mice that died. églue in parenthesis is the number of mice that died per total in group. Calculated medium lethal concentration (mg DMH/kg body weight) + 95% confidence interval. SignificEntly different from controls P<0.05. 69 Table 2. Percentage of DMH metabolized to azomethane and carbon dioxide in contigl and DMH—pretreated mice after injection with [ C]DMH Groupsa and [14C1DMH Azomethane Carbon Dioxideb Treatment (mg/kg Bwt) production (%) production (%) Expt 1. (N=5) Control 20 14 3.2.5 37 i 5.8 PT-20 20 23° 3: 1.2 25d i 1.8 ‘ PT-lOO 20 zed : 6.6 22d _+_ 1.6 “ Expt 2. (N=6) f Control 100 21 i 1.9 20 iL2.0 : PT-lOO 100 34C 11.5 9C :1.7 E aControl mice were not previously exposed to DMH. Prior to the test dose, PT-20 mice received three weekly injections of 20 mg of DMH/kg and PT-100 mice 4 weekly stepwise doses of 30, 50, 75 and 100 mg of DMH/kg body weight. Values represent meansjt SD Significantly different from controls P<0.05. Significantly different from controls P<0.01. 7O (P<0.05) and 2.0 times (P<0.05) more azomethane respectively, than control mice. The amount of [14C]DMH metabolized to carbon dioxide was decreased by approximately 33% and 40% in PT-20 and PT-100 mice respectively, (P<0.01). When administered 100 mg of [1 azomethane (P<0.01) and approximately 50% less carbon 4C]DMH/kg body weight, PT-lOO mice exhaled 1.5 times more dioxide derived from DMH than control mice (P<0.01). Levels of 7-methylguanine and 06-methylguanine were determined in the livers of control and PT-lOO mice after receiving a s.c. injection of 100 mg of DMH/kg body weight. Livers of PT-lOO mice had 50% less 7-methylguanine and 06-methylguanine than control mice (P<0.01, table 3). The ratio of the two bases was similar in both groups. PT-lOO mice had approximately 33% less hepatic cytochrome P450/mg microsomal protein than control mice (P<0.05) (table 4). Additionally, the ratio of liver weight to body weight was reduced by about 20% (P<0.05). Because of the smaller livers and lowered cytochrome P450/mg of microsomal protein, total cytochrome P450 content in PT-lOO mice was decreased by approximately 50% (P<0.05). DISCUSSION DMH toxicity was investigated in this study after observing in preliminary studies, that pretreatment with DMH decreases the toxicity of a subsequent treatment with 71 Table 3. Effect of pretreatment 'th DMH on the formation of 7-methylguanine and O -methy1guanine in livers of mice administered DMH Group anda No. of 7MGua:Gua.bC OGMGuazGuabc 7MGua:O6MGua Treatment mice (ngzug) (ng:ug) Control 6 5.80 :_0.61 1.04 i_0.13 5.6 :_0.7 PT-lOO 8 2.77‘51 1 0.40 0.49d : 0.06 5.7 + 0.9 aControl mice were not previously exposed to DMH. Prior to test dose, PT-lOO mice received received 4 weekly stepwise doses of 30, 50, 75 and 100 mg of DMH/kg body bweight. 6 6 7MGuazGua = 7-methy1guanine:guanine; O MGuazGua = O - methylguanine:guanine. SValues represent means 1 SD Significantly different from controls P<0.01. 72 Table 4. Effect of pretreatment gith DMH on hepatic cytochrome P450 content b d Group and Liver wt:Bwt cytochrome P4500 Total P450 Treatment (9/ 100 9) Control 6.6 i 0.60 0.78 i 0.11 4.41 i 0.61 PT-lOO 5.29: 0.60 0.519: 0.09 2.366: 0.41 :Values represents means: SD (N=9) Control mice were not previously exposed to DMH. Prior to the assay, PT-lOO mice received 4 weekly stepwise Cdoses 30, 50, 75 and 100 mg of DMH/kg body weight. nmoles hepatic cytochrome P450/mg microsomal protein. dTotal nmoles hepatic cytochrome P450. eSignificantly different from controls P<0.05. 73 this carcinogen. The two fold increase in the LCso for DMH in PT-30 mice confirmed these observations. The LC50 of 37 mg/kg for control mice is similar to that reported by Chang of 35 mg/kg for CF1 female mice (183). Because only one of ninety mice died during pretreatment, we rejected our hypothesis that the decreased toxicity was due to the elimination of sensitive mice. Rather, because of the increased radioactive azomethane and decreased carbon dioxide exhalation in PT-20 and PT-lOO mice, we accepted our second hypothesis. i.e., that mice adapt metabolically to reduce DMH toxicity. As the test dose of [14C1DMH was increased from 20 mg to 100 mg/kg body weight, the percentage of [141DMH metabolized to azomethane increased while the percentage metabolized to carbon dioxide decreased. The effect of dose on DMH metabolism, as seen in the present study, has been reported by other investigators (172, 173). The decreased carbon dioxide production in PT-ZO and PT- 100 mice suggested that these mice would exhibit less DNA methylation in response to a given dose of DMH. In the metabolism of DMH, as initially proposed by Druckey et al., the amount of carbon dioxide derived from DMH reflects methyl carbonium ion production (24, 25). The methyl carbonium ion is thought to be the agent responsible for DNA methylation and causation of tumors. Therefore, we determined the levels of 7-methylguanine and 06- methylguanine. 06-methylguanine is thought to be a 74 critical lesion for producing mutagenesis/carcinogenesis via DNA methylation (184-186) and was shown by Cooper et al., to correlate with the degree of sensitivity to DMH induced carcinogenesis in mice (187). However, this contrasts with findings by James and Autrup, where no difference in the amount of 06—methylguanine was noted between two strains of mice with different degrees of senstivity to DMH (188). The target organ for DMH induced carcinogenesis is the colon, although overall, initially, more methylation occurs in liver (190, 191). In agreement with our DMH metabolism data, which indicated a reduced metabolism to the active methylating species, PT-lOO mice had 50% less methylation than controls (P<0.01). The reduced levels of 7-methylguanine and 06- methylguanine are likely the result of a decrease in initial methylation and not due to enhanced DNA repair in PT-lOO mice, because 06-methylguanine-DNA methyltransferase, the protein responsible for removing the methyl group from 06-methylguanine, has thus far been shown not to be inducible in mice (189). Our data suggest that: 1) the increased tolerance to DMH toxicity may be a result of a decrease in macromolecule methylation and 2) pretreatment may reduce the carcinogenic effect of DMH. We did not determine colonic DNA methylation in this study. However, James and Autrup found less methylation in the colon of mice previously treated with DMH (188). There was approximately 50% less colonic methylation in animals 75 given 4 weekly doses of 45 mg of DMH/kg body weight and then administered a test dose at this same level in comparison to controls, which received only the test dose. If the conversion of azomethane to methyl carbonium ion were inhibited, azomethane would accumulate, resulting in an increased exhalation, as was seen in PT-20 and PT-lOO mice. Fiala et al., suggested that cytochrome P450 may be involved in DMH metabolism (37). In addition to the suggestion by Fiala et al., Autrup et al., found that phenobarbitol added to colon tissue explants along with [14C1DMH increased the amount of radioactivity associated with DNA (117). The significance of this finding may be considerable, given that in this same system, disulfiram decreased the amount of radioactivity associated with DNA. Phenobarbitol has been found to increase (30) and disulfiram to decrease (177) DMH induced carcinogenesis. Phenobarbitol administration in vivo increases cytochrome P450 content (147). Therefore, we decided to measure hepatic cytochrome P450 content. The level of cytochrome P450 in PT-lOO mice was approximately 50% lower in comparison to controls (P<0.05, table 4). These data suggest that the increased azomethane exhalation in PT-IOO mice may have been a result of a decreased cytochrome P450 content. In summary, the findings from this study are: 1) mice pretreated with DMH are less sensitive to DMH toxicity, as indicated by an increased LC for DMH; 2) administering 50 76 gradually increasing doses of DMH increases tolerance more than one single dose; 3) pretreatment alters DMH metabolism, as exhibited by an increase in the percentage of azomethane and a decrease in the percentage of carbon dioxide derived from DMH; 4) in comparison to control mice, mice previously treated with DMH exhibit decreased levels of hepatic DNA methylation in response to a test dose of DMH and 5) prior treatment with DMH decreases hepatic cytochrome P450, a finding which suggests this enzyme complex may be involved in DMH metabolism. The reduced DNA methylation in mice previously treated with DMH suggests that the latter doses of this compound may have a reduced tumor initiating effect. When administered 20 mg of [14C1DMH/kg body weight, radioactive carbon dioxide production in PT-20 mice (25%) was only slightly higher than in PT-lOO mice (22%) (table 2). Therefore, since carbon dioxide production is indicative of potential methylation, it is likely PT-ZO mice, which underwent a pretreatment regimen similar to that which is used to initiate tumors would also exhibit less DNA methylation in response to a test dose of DMH. Temple and El-Khatib, after observing an increased tolerance to DMH toxicity in Swiss mice (no data provided; 192), used a DMH dosing regimen to initiate tumors which consisted of s.c. injections of 23, 31, 42 and 4 doses of 56 mg/kg, rather than serial injections at the same level (115). However, before reaching conclusions about the optimum DMH dosing 77 regimen for tumor initiation; studies need to precisely determine the relationship between the total amount of DMH administered and the extent of colonic DNA methylation. If the number of DMH injections can be decreased without significantly reducing the initiating effects of this carcinogen, then a clearer separation between initiation and promotion can be achieved and the research value of the 1‘ DMH model of carcinogenesis will be enhanced. SUMMARY AND CONCLUS I ONS 78 SUMMARY AND CONCLUSIONS The primary objective of the first study, was to establish whether intrarectal instillation of DMH could be used as a model to study colon tumor liver metastases in Sprague-Dawley rats. DMH, when injected subcutanously, induces only very limited colon tumor metastases. Intrarectal administration of DMH had not been examined in this regard. Findings from this study suggest that route of DMH administration influences tumor metastases. In the group of rats administered 40 mg of DMH/kg body weight, tumor metastases occured in 8 of the 15 animals (53%). In contrast, in the group of rats injected s.c. with 20 mg of DMH/kg body weight, tumor metastases occured in only 3 of the 11 (27%) animals with colon tumors. The higher incidence of metastases was not a function of dose, since colon tumor frequency was similar in both groups. The enhanced metastases in the intrarectal group, was due to a higher rate of small intestinal tumor metastases. The clinical significance of this finding would appear to be limited since the incidence of human small intestinal cancer is rare. More significantly, metastases, when induced by either intrarectal administration or subcutaneous injection was primarily limited to the mesentery. Therefore, neither form of DMH administration provides a suitable method for studying colon tumor liver metastases. The finding that route of administration influences metastases raises the possibility that DMH metabolism when 79 80 given by intrarectal instillation differs from that when injected subcutanously. Intrarectal administration may provide an opportunity for bacterial metabolism of DMH. In addition, DMH is first be exposed to the colon, which is capable of metabolizing DMH, before undergoing hepatic metaboliSm (136). Clearly, subcutaneous administration does not reflect the manner in which environmental carcinogens are absorbed. This study emphasizes the potential role of extrahepatic carcinogen metabolism. In the second study, effects of cabbage consumption on DMH induced colon carcinogenesis was investigated in CF1 male mice. The need to examine this issue was clear. Recommendations to increase cruciferous vegetable consumption as a means of reducing cancer risk were based primarily on case control studies, which indicated in particular, that cabbage was protective against colon cancer (6-8, 45, 51, 71, 103). Prior to the present study, there were no studies in which the effect of cabbage consumption on experimently induced colon cancer could be adequately assessed. Two separate experiments were conducted. Results suggest that cabbage enhances, rather then inhibits carcinogenesis. In comparison to controls (0% cabbage), cabbage consuming mice had a significantly greater (P<0.05) tumor incidence (93% vs 72%) and more tumors per mouse (4.6 i 3.9 vs 2.4 i 2.1). Findings from the follow up experiment, indicate that cabbage consumption appears to mediate its influence only during the promotion phase of carcinogenesis, a suggestion which is 81 supported by the literature (154, 155).' Cabbage consumption had no effect on xenobiotic metabolizing enzymes, indicating that these enzymes did not play a role in the enhanced carcinogenesis in cabbage fed groups. The significance of this study is considerable for at least two reasons. First, the most pronounced effect of diet on human carcinogenisis is likely via its influence on promotion. Consequently, dietary components influencing the promotion phase of carcinogenesis, as appears to be the case with cabbage, are generally of greater importance than those influencing initiation. Second, if results of this study are confirmed, then public recommendations to increase cruciferous vegetable consumption may be unwarranted. Research into the manner in which cabbage influences the promotion phase of the DMH model of carcinogenesis needs to be conducted to determine the applicability of these findings to human colon cancer. The last study was initiated after observing an increased tolerance to DMH toxicity in CF1 male mice previously exposed to this carcinogen. Results indicate that with pretreatment, DMH metabolism is altered such that DNA methylation is reduced. The reduced methylation indicates that the carcinogenic effect of DMH is also deminished, since DMH is thought to induce tumors via methylation. The implications of this finding are significant. Because repeated injections of DMH are commonly used to induce tumors, the initiation and promotion phases of carcinogenesis 82 in the DMH model can not be clearly separated. Consequently, the DMH model is somewhat inadequate for testing dietary effects on promotion only. It may be possible, based on findings of this work, to reduce the number of injections of DMH without a corresponding decrease in tumor initiation. This will provide a better opportunity to evaluate the effects of dietary treatments on the promotion phase of experimentally induced colon cancer. In addition to the reduced level of methylation in response to pretreatment with DMH, animals also exhibited altered DMH metabolism. The percentage of DMH metabolized to azomethane increased while the percentage metabolized to carbon dioxide decreased with pretreatment. The decreased hepatic cytochrome P450 in pretreated mice suggest this enzyme complex may play a role in the altered DMH metabolism, and thus, indirectly in the decreased DNA methylation. The finding that hepatic cytochrome P450 may be involved in DMH metabolism may lead to an improved understanding of how certain dietary factors influence DMH induced colon carcinogenesis. REFERENCES 10. 11. 12. 13. REFERENCES Silverberg, E. Cancer statistics. CA 35(1985):5-21. Ninawer SJ, Miller G, Sherlock P. Risk and screening for colorectal cancer. Adv Int Med 1984;30:471-96. Doll R, Peto R. The causes of cancer:quantitative estimates of avoidable risks of cancer in the United States today. JNCI 1981;66:1191-308. Muir CS, Parkin DM. "The Wbrld Cancer Burden: Prevent or Perish (editorial)." Br Med J 1985;1:5-6. Carroll, KK and Khor, HT. Dietary fat in relation to tumorigenesis. Prog Biochem Pharmacol 1975;10:308. Kune S, Kune GB, watson LF. Case-control study of dietary etiological factors: The Melbourne colorectal cancer study. Nutr Cancer 1987;9:21-42. Graham 8. Results of Case-control studies of diet and cancer in Buffalo, New York. Cancer Res 1983;43:24098- 2413s. Modan B, Barell‘V, Lubin F. Modan M, Greenberg RA, Graham S. Lowaiber intake as an etiologic factor in cancer of the colon. JNCI 1975;55:15—18. Deschner EB, Long FC. Colonic neoplasms in mice produced with six injections of 1,2—dimethylhydrazine. Oncology 1977;34:255-57. Wfilliamson RCN, Bauer FLR, Oscarson JEA, Ross JS, Malt RA. Promotion of azoxymethane-induced colonic neoplasia by resection of the proximal small bowel. Cancer Res 1978;38:3212—3217. Miller, AB, Howe GR, Jain M, KJP Craib, Harrison L. Food items and selected food groups as risk factors in a case control of study of diet ad colo—rectal cancer. Int J Cancer 1983;32:1555—1610. Armstrong B, Doll R. Environmental factors and cancer incidence and mortality in different countrieS'with special reference to dietary practices. JNCI 1975;15:617-31. Schrauzer GN, White DA, Schneider CJ. Cancer mortality correlation studies III. Statistical associations with dietary selenium intakes. Bioinorg Chem 1977;7:23-31. 83 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 84 Segi M. "Age-adjusted death rates for cancer for selected sites in 46 countries in 1975" (1980). Segi Institute of Cancer of Epidemiology, Nagoya. Staszewski J, Haenszel‘W. Cancer mortality among the Polish-born in the United States. JNCI 1965:35;291-7. Haenszel W, Kurihara M, Segi M, Lee RKC. Stomach cancer among Japanese in Hawaii JNCI. 1972;49:969-88. Buell K. Changing incidence of breast cancer in Japanese- American women. JNCI 1973;51:1479-83. Pozharisski KM, Likhachev AJ, Klimashevski VG, Shaposhnikov JD. Experimental intestinal cancer research with special reference to human pathology. Adv Cancer Res 1979;30:165- 171 O Ahnen DJ. Are animal models of colon cancer relevant to human disease. Dig Dis Sci 1985;30:1035-1065. Grinnel RS. Lymphatic metastases of carcinoma of colon and rectum. Ann Surg 1950;131:494-506. Dionne L. The pattern of blood-borne metastases in carcinoma of the rectum. Cancer 1965;18:775-781. Dunnington DJ, Buscarino C, Gennaro D, Greig R, Poste G. Characterization of an animal model of metastatic colon carcinoma. Int J Cancer 1987:248-254. Sharkey FE, Fogh J. Metastasis of human tumors in athymic nude mice. Int J Cancer 1986;24:733-738. Druckey H. Production of colonic carcinomas by 1,2- dimethylhydrazine and azoxyalkanes, in‘W.J. Burdette (Ed), Carcinoma of the Colon and Antecedent Epithelium C.C. Thomas, Springfield. 1970 267—279 Preussmann R, Druckey H, Ivankovic S, von Hodenberg A. Chemical Structure and carcinogenicity of aliphatic hydrazo, azo and azoxy compounds and of triazine, potential in vivo alkylating agents. Ann NY Acad Sci 1969;163:697. Fiala EM, Bobotas G, Kulakis C, wattenberg LW,‘Weisburger JH. Effects of Disulfiram and related compounds on the metabolism in vivo of the colon carcinogen, 1,2- dimethylhydrazine. Biochemical Pharmacol 1977;26:1763- 1768. Harbach PR, Swenberg JA. Effects of selenium on 1,2- dimethylhydrazine metabolism and DNA alkylation. Carcinogenesis l981:2:575-580. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 85 Fiala ES, Bobotas G, Kulakis C, Weisburger JH. Seperation of 1,2-dimethylhydrazine metabolites by high-pressure liquid chromatography. J Chromatogr 1975;117:181-185. Feinberg A, Zedeck MA. Production of a highly reactive alkylating agent from the organospecific carcinogen methylazoxymethanol by alcohol dehydrogenase. Cancer Res 1980;40:3336-3350. Fiala ES. Investigations into the metabolism and mode of action of the colon carcinogen 1,2-dimethylhydrazine. Cancer 1975;36:2407-2412. Weisburger JH. Colon carcinogens: their metabolism and mode of action. Cancer 1971;28:60-71. Hawks A, Magee EN. The alkylation of nucleic acids of rat and mouse in vivo by the carcinogen 1,2-dimethylhydrazine. Br J Cancer l974:30:440-446. Moriya M, Ohta T, Sugiyama F, Miyazawa T, Shirasu Y. Assay for mutagenicity of bile in Sprague-Dawley rats treated subcutanously with intestinal carcinogens. JNCI 1979;63:977-981. Rubio CA, Nylander G. Further studies on the carcinogenesis of the colon of the rat‘with special reference to the absence of intestinal contents. Cancer 1981;48:951-956. Zedeck MS. Experimental colon carcinogenesis. In Gastrointestinal Tract Cancer, Lipkin, M. and Good, R.A. Eds., Plenum Press, New York. 1978;343-49. Schoental R. The mechanisms of action of the carcinogenic nitroso and related compounds. Br J Cancer 1973;28:436- 941. Fiala EM, Caswell N, Sohn OS, Felder MR, McCoy DG, “bisburger JH.‘N0neAlcohol dehydrogenase-mediated metabolism of methylazoxymethanol in the Deer Mouse, Peromyscus maniculatus. Cancer Res 1984;44:2885-2891. waterhouse J, Muir C, Shanmugaratman K. Powel J. (editors) 1982 Cancer Incidence in Five Continents Vbl. IV IARC. Scientific Publications No. 42. International Agency for Research on Cancer, Lyon). wynder EL, Shigematsu T. Environmental factors of cancers of the colon and rectum. Cancer 1967;20:1520—1530. McKeown-Eyssen GE, Bright-See E. Dietary factors in Colon Cancer: International Relationships. Nutr Cancer 1984;6:160-170. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 86 Hirayama T. Epidemiology of cancer of the stomach with special reference to its recent decrease in Japan. Cancer Res 1975;35:3460—3463. Lee JAH. Recent trends of large bowel cancer in Japan compared to the United States and England and.Wbles. Int J Epidemiol 1976;5:187-194. Jensen OM. Cancer risk among Danish male Seventh Day Adventists and other temperance society members. JNCI 1983;70:1011-1014. Haenszel W; Berg JW, Segi M, Kurhiara M, Locke FB. Large bowel-cancer in Hawaiian Japanese. JNCI 1973;51:1765-1779. Haenszel W, Locke FB, Segi M. A case-control study of large bowel cancer in Japan. JNCI 1980;64:17-22. Potter JD, McMichael AJ. Diet and Cancer of the colon and rectum: A case-control study. JNCI 1986;76:557-569. Jain M, Cook GM, Davis FG, Grace MG, Howe GR, Miller AB. A case-control study of diet and colo-rectal cancer. Int J Cancer 1980;26:757-768. Tajima K, Tominaga S. Dietary habits and gastro-intestinal cancers: A comparative case-control study of stomach and large intestine cancers in Nagoya, Japan. Jpn J Cancer Res 1985;76:705-716. Martinez I, Torres R, Frias Z, Colon JR, Fernandez N. Factors associated with adenocarcinomas of the large bowel in Puerto Rico. In: Birch JM, et. Advances in medical oncology, research and education. vol 3. New YOrk: Pergamon Press. 1979:45-53. Phillips RL, Snowdon DA. Dietary relationships with fatal colorectal cancer among Seventh-Day Adventists. JNCI 1985;74:307-317. Graham S, Dayal H, Swanson M, Mittelman G, Wilkinson G. Diet in the epidemiology of cancer of the colon and rectum. JNCI 1978;61:709-714. Stemmerman GN, Nomura AMY, Heilbrun L. Dietary fat and the risk of colorectal cancer. Cancer Res 1984;44:4633-4637. Reddy BS, Narisawa T, WUkusich D, weisburger JH, wynder EL. Effect of quality and quantity of dietary fat and dimethylhydrazine in colon carcinogenesis in rats. Proc Soc Exp Biol Med 1976;151:237-239. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 87 Reddy BS, watanabe K, weisburger JH. Effect of high fat diet on colon carcinogenesis in F344 rats treated with 1,2- dimethylhydrazine, methylazoxymethanol acetate, or methylnitrosourea. Cancer Res 1977;37:4156-4159. Bull AW, Soullier BK, Wilson PS, Hayden MT, Nigro ND. Promotion of azoxymethane-induced intestinal cancer by high fat diet in rats. Cancer Res 1979;39:4956-4959. Pariz MW. Calorie Restriction, ad Libitum Feeding, and Cancer. Proc Soc EXp Biol Med 1986;183:293-298. Reddy BS, Maeura Y. Tumor promotion by dietary fat in azoxymethane-induced colon carcinogenesis in female F344 rats:influence of amount and source of dietary fat. JNCI 1984;72:745-52. Reddy BS, Maruyama H. Effect of different levels of dietary corn oil and lard during the initiation phase of colon carcinogenesis in F344 rats. JNCI 1986;77:815-822. wargovich MJ, Felkner IC. Metabolic activities of DMH by colonic microsomes: a process influenced by type of fat. Nutr Cancer 1982;4:146-52. Nauss KM, Locninskar M, Newberne PM. Effect of alterations in the quality and quantity of dietary fat on 1,2- dimethylhydrazine-induced colon tumorigenesis in rats. Cancer Res 1983:43:4083-4090. Glauert HP, Bennink MR, Sander CH. Enhancement of 1,2- dimethylhydrazine induced colon carcinogenesis in mice by dietary agar. Food Chem Toxicol 1982;19:281-286. Burkitt DP. Epidemiology of Cancer of the Colon and Rectum. Cancer 1971;28:3-13. Jacobs LR. Relationship between dietary fiber and cancer: metabolic, physiologic, and cellular mechanisms. Proc Soc EXp Biol Med 1986;183:299-310. Bingham S,‘Williams DRR, Cole TJ, James‘WPT. Dietary fiber and regional large-bowel cancer mortality in Britain. Br J Cancer 1979;40:456-463. Bingham S, Williams DRR, Cummings JH. Dietary fibre consumption in Britain: New estimates and their relation to large bowel cancer mortality. Brit J Cancer 1985;52:399- 402. Report from the international agency for research on cancer. Dietary fibre, transit-time, faecal bacteria, steriods, and colon cancer in two Scandinavian populations. Lancet;1977:207-211. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 88 Jensen OM, MacLennan R, Wehrendorf J. (IARC‘Large Bowel Cancer Group). Large bowel cancer in Scandanavia in relation to diet and fecal characteristics. Nutr Cancer 1982 ; 4 : 5-19 . Walker ARP, Walker BF, Walker AJ. Faecal pH, dietary fiber intake, and proneness to colon cancer in four South African populations. Br J Cancer 1986;53:489-495. Phillips RL. Role of life-style and dietary habits in risk of cancer among Seventh-Day Adventists. Cancer Res 1975 ; 35 : 3513-3522. Dales LG, Friedman GD, Ury Hk, Grossman S, Williams SR. A case-control study of relationships of diet and other traits to colorectal cancer in American blacks. Am J Epidemiol 1978 ; 109 : 132-144 . Kuratsune M, Honda T, Englyst HN, Curmu'ngs JH. Dietary fiber in Japanese diet as investigated in connection with colon cancer risk. Jpn J Cancer Res 1986;77:736-738. Manousos 0, Day NE, Trichopoulos D, Gerovassilis F, Tzonou A, Polychronopoulou A. Diet and colorectal cancer: A case- control study in Greece. Int J Cancer 1983;32:1-5. Bristol JB, Emmett PM, Heaton KW, Williamson RCN. Sugar, fat, and the risk of colorectal cancer. Br Med J 1985;291:1467-1470. Chen‘W-F, Patchefsky AS, Goldsmith HS. Colonic protection from dimethylhydrazine by a high fiber diet. Surg Gynecol Obstet 1978;147:503-506. Wfllson RB, Hutcheson DP, Widerman. Dimethylhydrazine- induced colon tumors in rats fed diets containing beef fat or corn oil with or without wheat bran. Am J Clin Nutr 1977;30:176-181. Watanabe K, Reddy BS, Weisburger JH, Kritchevsky D. Effect of dietary alfalfa, pectin and wheat bran on azoxymethane or methylnitrosourea-induced colon carcinogenesis in F344 rats. JNCI 1979;63:141-145. Barbolt TA, Abraham R. The effect of bran on dimethylhydrazine induced colon carcinogenesis in the rat. Proc Soc Exp Biol Med 1978;157:656-659. Barbolt TA, Abraham R. Dose-response, sex difference, and the effect of bran in dimethylhydrazine induced intestinal tumorigenesis in rats. Toxicol Appl Pharmacol 1980;55:417- 422. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 89 Clapp NK, Henke MA, London JF, Shock TL. Enhancement of 1,2-dimethylhydrazine-induced large bowel tumorigenesis in Balb/c mice by corn, soybean, and.wheat brans. Nutr Cancer 1984:6:77-85. JaCObs LR. Enhancement of rat colon carcinogenesis by wheat bran consumption during the stage of 1,2-dimethylhydrazine administration. Cancer Res 1983;43:4057-4061. Jacobs LR, Lupton.JR. Relationship between colonic luminal pH, cell proliferation, and colon carcinogenesis in 1,2- dimethylhydrazine treated rats fed a high fiber diet. Cancer Res 1986;46:1727-1734. Freeman HJ, Spiller GA, Kim YS. A double-blind study on the effect of purified cellulose dietary fiber on 1,2- dimethylhydrazine-induced rat colonic neoplasia. Cancer Res 1978;38:2912-2917. Freeman HJ, Spiller GA, Kim YS. A double blind study on the effects of differing purified cellulose and pectin fiber diets on 1,2-dimethylhydrazine-induced rat colonic neoplasia. Cancer Res 1980;40:2661-2665. ward JM, Yamamoto RS,‘Weisburger JH. Cellulose dietary bulk and azoxymethane-induced intestinal cancer. JNCI 1973;51:713-715. Bauer HG, Asp N-G, Oste R, Dahlqvist A, Fredlund PE. Effect of dietary fiber on the induction of colorectal tumors and fecal B-glucuronidase activity in the rat. Cancer Res 1979;3752-3756. Deschner EE, Cohen BI, Raicht RF. Acute and chronic effect of dietary cholic acid on colonic epithelial proliferation. Digestion 1981;21:290-296. vahoundy GV, Cassidy MM, Lightfoot F. Ultrastuctural modifications of intestinal and colonic mucosa induced by free or bound bile acids. Cancer Res 1981;41:3764-3765. Reddy BS, Mastromarino, wynder DL. Diet and metabolism: large bowel cancer. Cancer 1977;39:1815-1819. Mitchel WD, Eastwood MA. Dietary fiber and colon function. In: Fiber in Human Nutrition. Ed. GA Spiller, RJ Amen. Plenum Press, New York, NY. 1976. 185-206. Narisawa T, Magadia NE, weisburger JH,‘Wynder EL. Promoting effects of bile acids on colon carcinogenesis after intrarectal instillation of N-methyl-N'- nitrosoguanidine in rats. JNCI 1974;53:1093-1097. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 90 Reddy BS, Narisawa T, weisburger JH, wynder EL. Promoting effects of sodium deoxycholate on colon adenocarcinomas in germ-free rats. JNCI 1976;56:441-442. Reddy BS, watanabe K,‘Weisburger JH, Wynder EL. Promoting effect of bile acids on colon carcinogenesis in germ-free and conventional rats. Cancer Res 1977;37:3238-3242. Reddy BS. Diet and excretion of bile acids. Cancer Res 1981;41:3766-3768. Hill MJ, Crowther JS, Drasar BS, Hawksworth.GB, Aries‘V, Wfilliams RED. Bacteria and aetiology of cancer of large bowel. Iancet 1971;1:95-100. Hill MJ, Drasar BS, William RED. Fecal bile-acids and clostridia in patients with cancer of large bowel. Lancet 1975;35:3403-3406. Reddy BS, Mastromarino A, wynder EL. Further leads on metabolic epidemiology of large bowel cancer. Cancer Res 1975;35:3403-3406. Reddy BS, Wynder EL. Metabolic epidemiology of colon cancer: fecal bile acids and neutral sterols in colon cancer patients and patients with adenomatous polyps. Cancer 1977;39:2533-2539. Mudd DG, McKelvey STD, Norwood W, Elmore DT, Roy AD. Faecal bile acid concentrations of patients with carcinoma or increased risk of carcinoma in the large bowel. Gut 1980;21:587-590. Tanida N, Hikasa Y, Shimoyama T, Setchell KDR. Fecal bile acid profiles of the large bowel: special reference to distribution, multiplicity, size and degree of dysplasia of the polyps. Jpn J Cancer Res 1985;76:104-112. Breuer NF, Domnes, P, Jaekel S, Goebell, H. Fecal Bile Acid Excretion Pattern in Colonic Cancer Patients. Dig Dis Sciences 30:852-859. National Research Council Diet.NUtrition and Cancer; National Academy Press;‘WaShington, DC. 1983. Anonomous. Taking Control-10 steps to a healthier life and reduced cancer risk. 1985. Publ Amer Cancer Society, Inc. Bjelke, E. Epidemiologic studies of cancer of the stomach, colon, and rectum with special emphasis on the role of diet. Thesis, University of Minnesota. Ann Arbor, Midh.: University Microfilms, 1973. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 91 Miller EC. Some current perspectives on chemical carcinogenesis in humans and experimental animals. Cancer Res 1978;38:1479-1496. Wattenberg EW. Inhibition of neoplasia by minor dietary constituents. Cancer Res 1983;43:24485-2453s. Boyd JN, Sell S, Stoewsand GS. Inhibition of aflatoxin- induced serum afetoprotein in rats fed cauliflower. Proc Soc EXP Biol Med 1979;161:473-475. Godlewski CE, Boyd JN, Sherman WK, Anderson JL, Stoewsand GS. Hepatic glutathione S-transferase activity and aflatoxin B induced enzyme altered foci in rats fed fractions of brussels sprouts. Cancer Lett 1985;28:151- 157. ’ ‘Wattenberginw. Studies of polycyclic hydrocarbon hydroxylases of the intestine possibly related to cancer. Cancer 1971;28:99-102. Wattenberg M, Loub WD. Inhibition of polycyclic aromatic hydrocarbon-induced neoplasia by naturally occurring indoles. Cancer Res 1978;38:1410-1413. 'Virtanen AI. Studies on organic sulphur compounds and other labile substances in plants. Phytochemistry 1965;4:207-228. wattenberg LW, Coccia JG, Lam LK. Inhibitory effects of phenolic compounds on benzopyrene-induced neoplasia. Cancer Res 1980;40:2820-2823. Nixon JE, Hendricks JD, Pawlcwski NE, Percira CB, Sinnhuber RD, Bailey GS. Inhibition of aflatoxin B carcinogenesis in rainbow trout by flavone and indole unds. Carcinogenesis 1984;5:615-619. Bailey GS, Hendricks JD, Shelton DW, Nixon JE, Pawlowski NE. Enhancement of carcinogenesis by the natural anticarcinogen indole-3-carbinol. JNCI 1987;78:931-934. Srisangnam C, Hendricks, DG, Sharma RP, Salunkhe DK, Mahoney AW. Effects of dietary cabbage (Brassica Oleracea L.) on the tumorigenicity of 1,2-dimethylhydrazine in mice. J Food Saf 1980;4:235-245. Temple NJ, El-Khatib SM. Cabbage and vitamin E: their effect on colon tumor formation in mice. Cancer Lett 1987;35:71-77. Pence BC, Buddingh F, Yang SP. Multiple dietary factors in the enhancement of dimethylhydrazine carcinogenesis: main effect of indole-3-carbinol. JNCI 1986;77:269-276. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 92 Autrup H, Harris CC, Schwartz RD, Trump BF, Smith L. Metabolism of 1,2-dimethylhydrazine by cultured human colon. Carcinogenesis 1980;1:375-380. Birt DF, Tibbels MG, Pour PM. Enhancement of pancreatic carcinogenesis in Syrian Hamsters fed dried cabbage in a high fat diet. FASEB 1986;45:5361 (Abstract). Boyd JN, Babish JG, Stoewsand GS. Modification by beet and cabbage diets of alfltoxin B -induced rat plasma foetoprotein elevation, hepatic tumorigenesis, and mutagenicity of urine. Food Chem Toxicol 1982;20:47-52. Deschner DD, Long FC, Maskens AP. Relationship between dose, time and tumor yield in mouse dimethylhydrazine- induced colon tumorigenesis. Cancer Lett 1979;8:23-28. Shirai T, Nakanowatari J, Kurata Y, FukuShima S, Ito N. Different dose response relationships in the induction of different types of colonic tumors in Wistar rats by 1,2- dimethylhydrazine. Jap J Cancer Res 1983;74:21-27. Martin MS, Martin F, Midhiels R, Bastien H, Justrabo E, Bordes M, Viry B. An experimental model for cancer of the colon and rectum. Digestion 1973;8:22-34. Nielson J, Balslev F, Jensen HE. Carcinoma of the rectum with liver metastases: Prognosis and operative indications. Acta Chir Scand 1973;139:479-481. Izumi K, Otsuka H, Furuya K, Akagi A. Carcinogenicity of 1,2-dimethylhydrazine dihydrochloride in Balb/c mice. Virchows Arch[A] 1979;384:263-267. SChiller CM, Curley WH, McConnel EE. Induction of colon tumors by a single oral dose of 1,2-dimethylhydrazine. Cancer Lett 1980;11:75-79. Toth B, Wilson RB. Blood vessel tumorigenesis by 1,2- dimethylhydrazine dihydrochloride (symmetrical). Am J Path 1971;64:585-600. Toth B, Makick 1M, Shimizu H. Production of intestinal and other tumors by 1,2-dimethylhydrazine dihydrochloride in mice. Am J Path 1976;84:69-86. Toth B, Patil K. The carcinogenic effect of intracolonic administration of 1,2-dimethylhydrazine dihydrochloride in mice. Anticancer Res 1983;3:167-170. Toth B, Patil K. A carcinogenicity dose response study by continuous administration of 1,2-dimethylhydrazine dihydrochloride in mice. Anticancer Res 1982;2:365-368. 130. 131. 132. 133. 134 135. 136. 137. 138. 139. 140. 141. 93 Reddy BS, Narisawa T, wright P, VUkusiCh D, weisburger JH, wynder EL. Colon carcinogenesis with azoxymethane and dimethylhydrazine in germ-free rats. Cancer Res 1975;35:287-290. Abrams GD. Impact of the intestinal microflora on intestinal structure and function. In Human intestinal microflora in health and disease. Ed. D. Hentges. Academic Press, New York. 1983. 13:291-310. Goldman P. Biochemical pharmacology and toxicology involving the intestinal microflora. In Human intestinal microflora in health and disease. Ed. D. Hentges. Academic Press, New'YOrk. 1983. 11:241-263. Reddy BS, Marisawa T, weisburger JH. Colon carcinogenesis in germ-free rats with intrarectal 1,2-dimethylhydrazine and subcutaneous azoxymethane. Cancer Res 1976;36:2874- 2876. Shamsuddin AKM. In vivo induction of colon cancer dose and animal species. In EXperimental colon carcinogenesis. Ed. H. Autrup, GM Williams. Boca Raton, Florida:CRC Press. 1983 3:51-62. Meshkinpour H, Holland D, Harmon D, Koyama S. Colonic absorption of 1,2-dimethylhydrazine (DMH) in the rat. Gen Pharmacol 1985;16:61-64. Autrup H, Schwartz RD, Essigman JM, Smith L, Trunp BF, Harris CC. Metabolism of aflatoxin B , benzo(a)pyrene, and 1,2-dimethy1hydrazine by cultured at and human colon. Teratogenesis Carcinog Mutagen 1980;1:3-11. Rodgers AE, Nauss KM. Rodent models for carcinoma of the colon. Dig Dis Sci 1985;30:875-1023. Mooney B, Taylor I. The effect of hepatic arterial ligation on spontaneously developing colorectal liver metastases in the rat. Clin Oncol 1982;8:231-235. Metzger U, Meier J, Uhlschimid G, Wei H. Influence of various prostaglandin synthesis inhibitors on DMH-induced rat colon cancer. Dis Colon Rectum 1984;27:366-369. Hamada S, Hirayama R, Matsumura J, Takagi Y, Miyanaga T, Aoki N, Mishima Y. Longer survival and fewer metastases by levamisole and tegafur in 1,2-dimethylhydrazine-induced murine colonic cancers. Bull Tokyo Med Dent Univ 1985;32:67-75. Kozlowski JM, Fidler IJ, campbell D. Metastic behavior of human tumor cell lines grown in nude mouse. Cancer Res 1984;44:3522-3529. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151 152. 153. 154. 94 Giavazzi R, Jessep JM, Campbell DE. Experimental nude mouse model of human colorectal cancer liver metastases. JNCI 1986;77:1303-1308. Doll R. The epidemiology of cancer. Cancer 1980;45:2475- 2485 Byers T, Graham S. The epidemiology of diet and cancer. Adv Cancer Res 1984;41:1-69. Graham, S. Alimentory factors in the epidemiology of gastric cancer. Cancer 1972;30:927-938. Paul AA, Southgate PAT. Ed. 1978 McCance and‘Widdowson's The composition of foods. Elsevier NOrth-Holland Biochemical Press., New'YOrk. Shertzer HG. Indole-3-carbinol and indole-3-acetonitrile influence on hepatic microsomal metabolism. Toxicol Appl Pharmacol 1982;64:353-361. Omura T, Sato R. The carbon-monoxide-binding pigment of liver microsomes. 1. Evidence of its hemoprotein nature. J Biol Chem 1964;239:2370-2378. Braggins TJ, Crow KB. The effects of high ethanol doses on rats of ethanol oxidation in rats. Eur J Biochem 1981;119:633-640. Chang Y, Bjeldanes LF. Effects of dietary R-goitrin on hepatic and intestinal glutathione S-transferase, microsomal epoxide hydratase and ethoxycourmarin O- deethylase activities in the rat. Food Chem Toxic 1985;23:905-909. Asaoka A, Takahashi K. A colorimetric assay of glutathione S-transferase using O-dinitrobenzene as a substrate. J Biochem 1983;94:1685-1688. Dutton GJ, Leakey JEA, Polland MR. Assays for UDP- glucuronyltransferase Activities. Methods Enzymol 1981;77:383-391. Markel MAK, Hass SM, Bieber LL, Tolbert.NE. A modification of Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 1978;87:206-210. Bailey GS, Hendricks JD, Shelton DW, Nixon JE, Pawlowski NE. Enhancement of carcinogenesis by the natural anticarcinogen indole-3-carbinol. JNCI 1987;78:931-934. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 95 Birt DF, walker B, Tibbels MG, Bresnick E. Anti- mutagenesis and anti-promotion of apigenin, rdbinetin and indole-3-carbinol. Carcinogenesis 1986;7:959-963. Sparnins DI” wattenberg DW. Enhancement of glutathione S- transferase activity of the mouse forestcmach by inhibitors of benzo(a)pyrene-induced neoplasia of the forestcmach. JNCI 1981;66:769-771. Cooper AH, Burns JJ. Metabolic interactions among environmental chemicals and drugs. Science 1972;178:576- 586. Goeger DE, Shelton DW, Hendricks JD, Bailey GS. Mechanisms of anti-carcinogenesis by indole-3-carbinol: effect on the distribution and metabolism of aflatoxin B1 in rainbow trout. Carcinogenesis 1986;7:2025-2031. Shertzer HG. Protection by indole-3-carbinol against covalent binding of benzo(a)pyrene metabolites to mouse liver DNA and protein. Food Chem Toxic 1983;21:31-35. Godlewski CE, Boyd JN, Sherman WK, Anderson JL, Stoewsand GS. Hepatic glutathione S-transferase activity and aflatoxin B -induced enzyme altered foci in rats fed fractions of brussels sprouts. Cancer Lett 1985;28:151- 157. Hendrich S, Bjeldanes LF. Effects of cabbage, brussels sprouts, illicium verum, sdhizandra chinensis and alfalfa on the benzo(a)pyrene metabolic system in mouse liver. Food Chem Toxic 1983;21:479-486. Uotila P, Aitio A. The effects of dietary vegetables on rat intestinal and hepatic metabolism of xenobiotics. Res Commun Chem Path Pharmacol 1979;26:425-428. Stoewsand GS, Anderson.J1n Iisk DJ. Changes in liver glutathione S-transferase activities in Coturniz quail fed municipal sludge-grown cabbage with reduced levels of glucosinolates (42315). Proc Soc EXp Biol Med 1986;182:95- 99. Sparnins VL, Venegas PL, Wattenberg M. Glutathione S- transferase activity: enhancement of compounds inhibiting Chemical carcinogenesis and by dietary constituents. JNCI 1982;68:493-496. Nishie K, Daxenbicher M. Toxicology of glucosinolates, related compounds (nitriles, R-goitrin, isothiocyanates) and vtamin U found in cruciferae. Food Chem Toxicol 1980;18:159-172. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 96 Dunnink JK, Prejean JD, Haseman J, Thompson RB, Giles HD, McConnell EE. Carcinogenesis bioassay of allyl isothiocyanate. Fund Appl Toxicol 1982;2:114-120. Neudecker T, Henschler D. Allyl isothiocyanate is mutagenic in Salmonella typhimurium. Mutat Res 1985;156:33- 37. Gould H, Gumbonann MR, DaxenbiChler ME. Pathological changes in rats fed crambe meal glucosinolates hydrolytic products, LS-l cyano 2-hydroxy 3,4 epithiobutanes (erythro and threo) for 90 days. Food Chem Toxicol 1980;18:619-625. Rahman MK. Effect of allyl isothiocyanate (AITC) on B- carotene metabolism. Bangladesh Pharm J 1977;6:16-21. Iebdova-Svobodova S. Effect of isothiocyanate on the synthesis of proteins. Naturwissenschaften 1965;52:430- 435. Kelly FC, Snedden WW. Prevalence and geographical distribution of endemic goitre in Endemic goitre,‘W0rld Health Organ. Monog. Ser. No. 44. Wbrld Health Organization, Geneva 191. Fiala ES, Bdbotas G, Kulakis C, Weisburger JH. Brief communication: Detection and estimation of azomethane in expired air of 1,2-dimethylhydrazine-treated rats. JNCI 1976;56:1271-1277. P999, AE. Inhibition of the alkylation of nucleic acids and of the metabolism of 1,2-dimethylhydrazine by aminoacetonitrile. Chem Biol Interac 1978;23:273-279. Fiala ES, Kulakis C, Christiansen G, weisburger JH. Inhibition of the Metabolism of the colon carcinogen, aoxymethane, by pyrozole. Cancer Res. 1978;38:4515-4521. Fiala EM, Stathopoulos C. Metabolism of methylazoxymethanol acetate in the F344 rat and strain-2 guinea pig and its inhibition by pyrazole and disulfiram. J Cancer Res Clin Oncol 1984;108:129-134. Swenberg JA, Cooper HK, Budheler J, Kleihues P. 1,2- dimethylhydrazine-induced methylation of DNA bases in various rat organs and the effect of pretreatment‘with disulfiram. Cancer Res 1979;39:465-467. Wattenberg IN. Brief Carmunication: Inhibition of dimethylhydrazine-induced neoplasia of the large intestine by disulfiram. JNCI 1975;54:1005-1006. 178. 179. 180. 181. 182. 183. 184 O 185. 186. 187. 188. 189. 97 Wattenberg DW, Fiala ES. Inhibition of 1,2- dimethylhydrazine-induced neoplasia of the large intestine in female CF1 mice by carbon disulfide:Brief Conmunication. JNCI 1978 ; 60 : 1515-1517 . American Institute of'NUtrition. Report of the American Institutes of NUtrition Ad H0c Committee on Standards for NUtritional Studies. J Nutr 1977;107:1340-1348. Hubert, JJ. Methods for determining the DD Kendall/Hunt Publishing Co. 1984 p50-62. 50 in Bioassays. Marmur J. A procedure for the isolation of deoxyribonucleic acid from micro-organisms. J Mol Biol 1961;3:208-218. Herron DC, Shank RC. Quantitative high-pressure liquid chromatographic analysis of methylated purines in DNA of rats treated with chemical carcinogens. Anal Biodhemistry 1979;100:58-63. . Chang WWL. Histogenesis of symmetrical 1,2- dimethylhydrazine-induced neoplasms of the colon in the mouse. JNCI 1978;60:1405-1418. Pegg AE. Methylation of the 06-position of guanine in DNA is the most likely initiating event in carcinogenesis by methylating agents. Cancer Invest 1984;2:223-231. Doniger J, Da RS, DiPaolo JA. Quantitative assessment of the role of O -methylguanine in the initiation of carcinogenesis by methylating agents. Proc Natl Acad Sci 1985;82:421-425. Maher VM, Domoradzki g, Corner RC, McCormick JJ. Correlation between O -alkylguanine-DNA alkyltransferase activity and resistance of human cells to the cytotoxic and mutagenic effects of methylating and ethylating agents. In Biochemical and Molecular Epidemiology of Cancer Ed. CC Harris. Alan R Liss, New York. 1986;411-418. Cooper HK, Bucheler J, Kleihues P. DNA alkylation in mice with genetically different susceptibility to 1,2- dimethylhydrazine-induced colon carcinogenesis. Cancer Res 1978;38:3063-3065. James JT, Autrup H. Methylated DNA adducts in the large intestine of ICR/Ha and C57BL/Ha mice given 1,2- dimethylhydrazine. JNCI 1983;70:541-546. Pegg AE, Dolan , Scicchitano D, rimoto K. Studies of the repair of O -alky1guanine and O -alkylthymine in DNA by alkyltransferases from marmalian cells and bacteria. Environ Health Perspect 1985;62:109-114. 190. 191. 192. 98 Rodgers KJ, Pegg AE. Formation of 06-methylguanine by alkylation of rat liver, colon, and kidney DNA following administration of 1,2-dimethylhydrazine. Cancer Res 1977;37:4082-4087. HarbaCh PR, Swenberg JA. Effects of selenium on 1,2- dimethylhydrazine metabolism and DNA alkylation. Carcinogenesis 1981;2:575-579. Temple NJ, El-Khatib SM. Toxicity of 1,2-dimethylhydrazine in mice: effect of diet and development of tolerance. J Food Saf 1985;7:77-81. "'l M. .9 PHMK"