Effect of dietary alfalfa, pectin, and wheat bran on azoxymethane-or methylnitrosourea-induced colon carcinogenesis in F344 rats.

The effect of dietary alfalfa, pectin, and wheat bran on colon carcinogenesis was studied in female inbred F344 rats. Weanling rats were fed semipurified diets containing 0 or 15% alfalfa, pectin, or wheat bran. At 7 weeks of age, all animals except controls were given azoxymethane (AOM) sc at a dose rate of 8 mg/kg body weight/week for 10 weeks or methylnitrosourea (MNU) intrarectally at a dose rate of 2 mg/rat twice a week for 3 weeks. The AOM-treated group was autopsied 40 weeks and the MNU-treated group 30 weeks after the first injection of the carcinogen. No tumors were observed in the colon or other organs of untreated rats fed the various diets. The animals fed the alfalfa diet and treated with MNU had a higher incidence of colon tumors than did those fed the control diet or the diets containing pectin or wheat bran. The incidence of MNU-induced colon tumors did not differ between the animals fed the control diet or the diets containing pectin or wheat bran. However, the incidence of AOM-induced colon tumors in rats fed diets containing pectin or wheat bran was lower than that in rats fed the control diet or the alfalfa diet. These results thus indicate that the effect of fiber in colon carcinogenesis depends on the type of fiber and, possibly, the fiber's mode of action.

Diet and colon cancer: assessment of risk by fiber type and food source.

How fiber in the diet is related to the development of colon cancer was assessed in a population-based study conducted on 231 cases and 391 controls in Utah between 1979 and 1983. Crude fiber consistently decreased risk associated with colon cancer in both males [odds ratio (OR) = 0.4] and females (OR = 0.5). Dietary fiber, as analyzed by the method of A. S. Bitner, and neutral detergent fiber were not consistently related to colon cancer risk. Of the noncellulose polysaccharides examined, mannose and galactose were protective against cancers in the ascending colon in males (ORs = 0.5 and 0.3, respectively), whereas galactose and uronic acid were protective against cancers in the ascending colon in females (ORs = 0.5). Highest quartiles of intake of fruits and vegetables were also associated with a decreased risk of colon cancer in males (ORs = 0.3 and 0.6, respectively) and in females (ORs = 0.6 and 0.3, respectively) compared with lowest quartile of intake, whereas high intake of grains was not protective.

Reduction of colonic carcinogenesis by wheat bran independent of fecal bile acid concentration.

It has been reported previously that populations with a decreased concentration of fecal bile acids have a lower incidence of colon cancer. We examined the importance of fecal bile acid dilution by wheat bran (WB) in inhibiting colonic tumorigenesis in an experimental animal model. Male F344 rats received oral doses of the colon carcinogen 1,2-dimethylhydrazine [CAS: 540-73-8] and were assigned randomly to groups fed one of four semipurified diets for 26 weeks. The diets were fiber-free (FF), 10% WB, FF + bile salts, or WB + bile salts. The amount of bile salts added was adjusted to produce a fecal bile acid concentration in the group fed WB + bile salts equal to that found in the FF groups. Fecal bile acid concentrations at 12 and 24 weeks in the WB + bile salts group were similar to those in the FF group. Gross and microscopic findings at necropsy revealed a reduced total number and multiplicity of colon tumors in both bran-fed groups. Although the fecal bile acid concentrations of the FF and WB + bile salts groups were equal, the latter showed a significant reduction in tumor yield.

Fiber, steroids, and cancer.

Fiber is a generic term for dietary components which are not metabolized by intestinal secretions. Fiber encompasses a variety of substances of unique structure having specific physical characteristics and physiological function. There are a number of epidemiological and case-control studies on diet and colon cancer, many of which suggest that dietary fiber or fiber-rich foods may exert a protective action, but the view is not unanimous. One study finds a negative correlation between cancer mortality and pentose-rich dietary fibers. Animal studies have been carried out using rats of different strain, semipurified or commercial diets, and various carcinogens administered in a variety of ways. Again, there is no unanimity. The results depend upon the type of fiber and the route of carcinogen administration. Some type of fiber bind bile acids and salts, and this property has been investigated as a means of explaining the protective action of fiber. Several recent studies in human subjects suggest that susceptibility to colon cancer is related inversely to the concentration of fecal bile acids rather than to the absolute quantity of fecal bile acids or their structure. Future studies require better standardization of protocols so that intradietary effects can be assessed.

Inhibition of chemically induced mammary and colon tumor promotion by caloric restriction in rats fed increased dietary fat.

Tumor promotion associated with increased dietary fat may be inhibited by reduction in total caloric intake. This hypothesis was tested in rats given either 7,12-dimethylbenz(a)anthracene to induce mammary tumors or 1,2-dimethylhydrazine to induce colon tumors. One week after dosage with either carcinogen, the rats were fed semipurified diets that provided 4% fat with ad libitum calories or 13.1% fat with a reduction of calories by 40% from ad libitum intake. Rats treated with 7,12-dimethylbenz(a)anthracene and subjected to caloric restriction weighed 40% less than those fed ad libitum; rats treated with 1,2-dimethylhydrazine were heavier at the onset of caloric restriction and lost weight and weighed approximately 40% less than animals fed ad libitum. At 20 weeks after 7,12-dimethylbenz(a)anthracene administration, rats fed ad libitum had 80% tumor incidence while in those fed restricted calories, 20% had tumors (P less than 0.001). All other measures of mammary tumor growth were significantly reduced in rats given restricted calories. Six months after 1,2-dimethylhydrazine administration, colon tumor incidence was 100% in rats fed ad libitum and 53% in those fed the calorie-restricted diet (P less than 0.001). This reduction of colonic carcinogenesis was seen despite a significant increase in mucosal labeling index following [3H]thymidine autoradiography. This paradoxical finding may be due to the increased fat content of the calorie-restricted diet. These data demonstrate that the tumor-promoting effects of dietary fat can be more than offset by a reduction in total caloric intake and that the promoting effect of fat may be due, at least in part, to its greater caloric density.

Dietary fat versus caloric content in initiation and promotion of 7,12-dimethylbenz(a)anthracene-induced mammary tumorigenesis in rats.

Enhancement of mammary tumor formation by dietary fat may be mediated via increased caloric intake. Three experiments were performed to study this relationship in 7,12-dimethyl-benz(a)anthracene (DMBA)-treated female Sprague-Dawley rats: (a) high- or low-fat isocaloric diets were fed in a crossover design; (b) low-fat, high-calorie and high-fat, low-calorie diets were fed in a crossover design; (c) pair-fed rats were restricted to 60% of the calories of controls with ad libitum access to food beginning 10 days after DMBA administration. The pair-fed rats received daily 60% of calories, the same level of fiber, and 115% more fat than did rats fed ad libitum. Tumor yield but not tumor incidence was greater in rats fed high-fat rather than low-fat isocaloric diets prior to initiation of tumorigenesis. A low-fat, high-calorie diet led to more tumor incidence and yield than was associated with feeding of a high-fat, low-calorie diet. Caloric restriction (although with concomitant intake of more fat) led to complete inhibition of tumor formation. These results indicate that both high-fat and high-calorie diets exhibit cocarginogenic, not merely promotional, properties. Caloric intake may be a greater determinant than dietary fat of a tumor-enhancing regimen. Finally, restriction of caloric intake during promotion markedly suppresses tumor formation, despite the increased fat content of the restricted diet, suggesting a permissive role for calories in tumor formation. The possibility remains that alterations in levels of other dietary components could also have contributed to the observed effects.

Caloric restriction and 7,12-dimethylbenz(a)anthracene-induced mammary tumor growth in rats: alterations in circulating insulin, insulin-like growth factors I and II, and epidermal growth factor.

Caloric restriction (CR) inhibits many neoplastic diseases in rodents, yet the biochemical mechanism(s) for these effects are poorly understood. We have examined the effects of ad libitum (AL) feeding with 25 or 40% CR on the promotion of 7,12-dimethylbenz(a)anthracene-induced mammary tumorigenesis in virgin female Sprague-Dawley rats. Further, we have also studied the influence of chronic CR on temporal alterations in circulating insulin, insulin-like growth factor I/somatomedin C, insulin-like growth factor II/multiplication-stimulating activity, and epidermal growth factor levels at 0, 1, 3, 5, 11, and 20 weeks in carcinogen- and vehicle-treated animals. Tumor incidence and multiplicity were markedly inhibited (P less than 0.05) with increasing CR. Fasting serum insulin-like growth factor I/somatomedin C levels exhibited a significant acute decline with CR at 1 and 3 weeks, but were comparable to AL-fed controls throughout the remainder of the 5-month study, despite continued differences in weight gain between AL and CR rats. Levels of insulin-like growth factor II/multiplication-stimulating activity exhibited no discernible pattern in relation to CR. Serum insulin levels showed age-dependent increases, but were affected by increasing CR at all time points. Insulin levels were significantly (P less than 0.05) reduced in 40% CR rats from 3 weeks onward compared to controls, while 25% CR resulted in nonsignificant (P less than 0.07) reductions throughout the study. No significant differences in growth factor levels were observed between 7,12-dimethylbenz(a)anthracene- and vehicle-treated rats. Circulating epidermal growth factor was not detectable in any treatment group regardless of the nature or duration of the dietary regimen, time of blood collection, or subsequent tumor-bearing status. These data suggest that decreased serum insulin-like growth factor I/somatomedin C and insulin levels with CR and their complex interactions in vivo may play a role in the inhibition of mammary tumor promotion by CR.

Growth factor binding to 7,12-dimethylbenz(a)anthracene-induced mammary tumors from rats subject to chronic caloric restriction.

Caloric restriction (CR) inhibits tumorigenesis in rodents. To understand the basis for this effect the binding of insulin, insulin-like growth factor I/somatomedin C (IGF-I/Sm-C), insulin-like growth factor II/multiplication stimulating activity (IGF-II/MSA), and epidermal growth factor were examined to membrane preparations of 7,12-dimethylbenz(a)anthracene-induced mammary adenocarcinomas and several normal tissues from female Sprague-Dawley rats. Animals were fed ad libitum (AL) or 25% and 40% calorically restricted diets. Large, palpable (LP) and small, less than or equal to 100 mg, nonpalpable (SNP) tumors were evaluated. Growth factor binding to tumors was differentially affected by CR. IGF-I/Sm-C binding was comparable for AL-LP, AL-SNP, and 25% CR-LP tumors, but elevated in 25% CR-SNP tumors. Scatchard analysis revealed high and low affinity IGF-I/Sm-C binding sites, with AL-SNP and 25% CR-SNP tumors exhibiting similar levels of high affinity sites and at a greater concentration than AL-LP and 25% CR-LP tumors. Insulin binding to mammary tumors was low, i.e., 8- to 13-fold lower than IGF-I/Sm-C binding. The 25% CR-LP and SNP tumors bound 2- to 5-fold more insulin than corresponding AL-LP and SNP tumors. Binding of IGF-II/MSA to these tumor preparations was high, approximately 11- to 25-fold greater than insulin binding, and was unaffected by CR or tumor size. The binding of epidermal growth factor was not detected in any tumor preparations. Receptor binding studies were confirmed with covalent cross-linking and sodium dodecyl sulfate-polyacrylamide gel electrophoresis analyses. Normal tissues exhibited tissue- and growth factor-specific alterations in binding with host CR. Thus, alterations in growth factor binding were not tumor specific, but were less pronounced than in mammary tumors. These findings suggest alterations in IGF-I/Sm-C and insulin binding properties to tumors in relation to CR and tumor size may contribute, in part, to the inhibitory effects of CR on tumorigenesis.

Ultrastructural modifications of intestinal and colonic mucosa induced by free or bound bile acids.

There is substantial evidence that bile acids may enhance the colon tumorigenesis induced by chemical carcinogens and that agents stimulating increased bile acid excretion may show similar promoting or enhancing activity. To test the premise that these agents might modify topographical ultrastructure of the small intestine and colon in the absence of carcinogens, rats were fed for 6 weeks on chemically defined diets containing 2% levels of three commercial bile acid sequestrants or 15% levels of wheat brain, cellulose, pectin, or alfalfa. Major qualitative and quantitative deviations from normal morphology were observed with each of the three sequestrants. Similar but less dramatic modifications occurred with diets containing alfalfa or pectin, both of which either "bind" bile acids in vitro or result in increased bile acid excretion. Bran and cellulose which neither "bind" bile acids nor increase their fecal excretion, were without significant effects on intestinal or colonic morphology. The morphological deviations observed with bile acid sequestrants were shown to be a direct response to free or bound bile acids by comparing the morphological modifications resulting from daily intracolonic infusions of free bile acids, sequestrant-bound bile acids, or the sequestrant alone.

Biochemical alterations in 7,12-dimethylbenz[a]anthracene-induced mammary tumors from rats subjected to caloric restriction.

Caloric restriction reduces the incidence and progression of a broad spectrum of neoplastic diseases, yet little is known about the biochemical and molecular mechanisms involved. Profiles of enzyme activities of importance in cellular energy utilization were examined in 7,12-dimethylbenz[a]anthracene-induced (DMBA) mammary adenocarcinomas from rats fed ad libitum or calorically restricted diets. The diets provided equal nutrients except for fewer carbohydrate-derived calories; graded caloric restriction was 10, 20, 30 and 40%. The specific activities of hexokinase, pyruvate kinase, lactate dehydrogenase, glucose-6-phosphate dehydrogenase, malic enzyme and fructose-1,6-bisphosphatase were all elevated to varying degrees in both large palpable and small, non-palpable tumors from calorically restricted hosts compared to activities in tumors from ad libitum-fed rats. Phosphofructokinase activity was increased in palpable tumors from calorically restricted hosts but markedly reduced in non-palpable tumors. These results suggest adaptive or compensatory alterations in tumor enzyme profiles in response to the altered nutritional state of the host.

Stearic acid metabolism and atherogenesis: history.

Studies conducted in dogs, rats, and hamsters show that stearic acid or stearic acid-rich glycerides are absorbed less efficiently than are lauric, myristic, and palmitic acids or their triglycerides. This observation may explain in part why stearic acid is less cholesterolemic than saturated fatty acids of shorter chain length. In rabbits, cocoa butter or other fats rich in stearic acid are less atherogenic than other saturated fatty acids. This finding is true for both cholesterol-containing and cholesterol-free diets.

Herman Award Lecture, 1992: lipid nutrition--a personal perspective.

In the 1950s excess dietary fat was thought to be a risk factor for development of atherosclerosis. Over the ensuing years we have examined and given advice concerning first, saturation and unsaturation, and later, specific fatty acids such as oleic and stearic acids. Current work is further refining the delineation of fatty acids that might be harmful. Eventually, we will probably again cite excess fat as the problem.

Fiber, lipids, and atherosclerosis.

The influence of dietary fiber on lipid metabolism and atherosclerosis in animals and on lipid metabolism in man is reviewed. Pectin, guar gum, and lignin lower serum and liver cholesterol in cholesterol-fed rats. Agar increases liver cholesterol in rats. Bran has no effect on serum lipid levels in rats or monkeys. Vegetarians have long been known to exhibit cholesterol levels lower than those of comparable populations who subsist on a mixed diet. Pectin and guar gum lower cholesterol levels in man whereas cellulose and bran have no effect. Rabbits fed a semipurified diet containing saturated fat become atherosclerotic, but addition of the same fat to laboratory ration has no effect; it has been shown that the residue in laboratory ration is the cause of the difference. Semipurified diets containing cellulose are more atherogenic than those containing wheat straw or alfalfa. The semipurified diets also cause aortic sudanophilia or atherosclerosis in baboons and vervet monkeys. One possible mechanism of hypolipemic action of fiber involves the binding of bile acids, which would result in reduced absorption of cholesterol, resulting in lower levels of serum cholesterol.

Influence of vegetable protein on gallstone formation in hamsters.

The lithogenic diet generally used for production of gallstones in hamsters contains 20% casein. In four separate experiments the 20% casein diet was compared with one containing 20% soy protein. All other components were kept constant. Two of the four experiments were of 45-day duration, one of 70 days and one of 100 days. When results of all four experiments were averaged, 57.5 +/- 3.6% of hamsters fed casein exhibited cholesterol gallstones whereas only 14.0 +/- 1.3% of soy protein-fed hamsters had gallstones. Thus, soy protein has a significant inhibitory effect on gallstone formation in hamsters. When soy protein was fed to hamsters with preestablished gallstones there was evidence of gallstone dissolution.

Gallstone formation in hamsters: effect of varying animal and vegetable protein levels.

The lithogenic diet routinely used for production of gallstones in hamsters contains 20% casein. It has previously been shown that replacement of casein by soy protein significantly decreases gallstone formation. In this study hamsters were fed a lithogenic diet containing casein (C), soy isolate (S), C/S 3:1, C/S 1:1, and C/S 1:3. The percentages of hamsters with gallstones on these five diets were: 44, 12, 38, 23, and 15. Biliary cholesterol levels and lithogenic index both decreased significantly with increasing levels of soy protein. Dilution of casein with soy protein progressively decreases lithogenicity.

Bile acid metabolism and fiber.

Studies in experimental animals and in humans have shown a definite interaction of dietary fiber with bile acid metabolism. In experimental animals, some types of fiber have been shown to increase bile acid excretion as well as increase pool size and turnover of bile acids. In man, increases in bile acid excretion have been observed but not consistently. Alternations in excretion and biliary bile acid levels indicate an influence of dietary fiber in bile acid metabolism. In vitro experiments have attempted to determine the nature of the fiber-bile acid interaction. Some types of fiber and some components of fiber have been shown to bind bile acids and bile salts in appreciable quantities. The capacity for various types of fiber to bind bile acids or bile salts is quite variable, and a high capacity seems to correspond with a hypocholesteremic effect. A great deal of research is still needed for elucidation of these interactions of fiber and bile acid metabolism and the resultant beneficial effects on cholesterol metabolism and the disease-related abnormalities in cholesterol metabolism.

Diet, nutrition intake, and metabolism in populations at high and low risk for colon cancer. Relationship of diet to serum lipids.

Serum cholesterol, high-density lipoprotein cholesterol, and triglycerides of three groups of Seventh-day Adventists (SDAs)--true vegetarians, lacto-ovo vegetarians, and nonvegetarians--and the general population were measured and related to age, sex, and diet. True vegetarian SDAs had the lowest cholesterol levels while cholesterol levels of the other three groups were similar. Triglyceride levels were highest in the true vegetarian SDAs and lowest in general population. Cholesterol levels rose with increasing age while percentage of high-density lipoprotein cholesterol fell. Female SDAs in all three dietary groups had higher cholesterol levels than males. In all four groups percentage of high-density lipoprotein cholesterol was higher in females than in males. In the male subjects, cholesterol levels varied linearly and inversely with carbohydrate intake. Protein and fat intake in true vegetarian SDAs and lacto-ovo vegetarian SDA groups were lower than in the nonvegetarian SDA and general population groups. The differences in protein and fat intake were reflected in elevations in serum cholesterol levels. In the female subjects, cholesterol levels were also inversely correlated with carbohydrate intake but not as distinctly as in the males. True vegetarian SDA females ingested less protein and fat than the other three groups and exhibited lower serum cholesterol levels. Cholesterol levels in lacto-ovo vegetarian SDA, nonvegetarian SDA, and general population female subjects were similar and did not reflect differences in protein or fat intake. The true vegetarian dietary lifestyle resulted in lowest cholesterol levels, however, if it was breeched to the least extent, cholesterol levels rose. Thus, intake of skim, low fat, or whole milk resulted in similar serum cholesterol levels as did low or high egg consumption.

Diet, nutrition intake, and metabolism in populations at high and low risk for colon cancer. Binding of bile salts to dietary residues.

The binding of glycocholic and glycochenodeoxycholic acids to dried defatted residues of homogenized diets consumed by four dietary groups has been measured. The four groups were true vegetarian Seventh-day Adventists (SDA), lacto-ovo vegetarian SDA, nonvegetarian SDA, and the general population. Glycocholic acid was bound to the same extent (7 to 9%) by all four dietary residues. The true vegetarian SDA residues bound significantly more glycochenodeoxycholic acid than those of either lacto-ovo vegetarian SDA (p less than 0.001) or nonvegetarian SDA (p less than 0.001). The general population residue bound significantly more glycochenodeoxycholic acid than did those of nonvegetarian SDA (p less than 0.001). Binding of glycochenodeoxycholic acid was significantly correlated with dietary neutral detergent fiber (p less than 0.001), hemicellulose (p less than 0.01) and cellulose (p = 0.01). There was no correlation between the binding of glycochenodeoxycholic acid and dietary lignin, pectin, or cutin.

Diet, nutrition intake, and metabolism in populations at high and low risk for colon cancer. Comparison of serum hexosaminidase levels.

Elevated levels of serum hexosaminidase (beta-2-acetamido-2-deoxy-D-glucoside; N-acetamidodeoxy-D-glucohydrolase, EC 3.2.1.30)(HEX) have been found in patients with cancer. In view of the reported low levels of colon cancer among Seventh-day Adventists (SDA) we determined levels of total HEX, and its heat-labile (HEX-A) and heat-stable (HEX-B) isozymes in vegetarian SDA, lacto-ovo vegetarian SDA, nonvegetarian SDA, and the general public. Lowest levels of total HEX and highest percentage of HEX-B were found in vegetarian SDA. The levels in the other three groups were comparable, and these differences were seen at all age levels. Female subjects exhibited significantly lower total HEX levels than the males but the percentage of HEX-B was the same.

An improved method for oxidation of chromium(III) oxide-containing fecal samples by using sodium peroxide fusion.

A safer method of oxidation of Cr2O3-containing fecal samples from transit-time studies was developed using sodium peroxide to replace perchloric acid as the oxidizing agent. The percentage recovery of Cr2O3 with this method was compared with that of perchloric acid method for samples containing quantities of fecal ash and Cr2O3 typical of those from rodent transit-time studies. Both methods gave relatively constant percentage recoveries for Cr2O3 contents from 0.4 to 10 mg. Over this range, mean (+/- SD) percentage recoveries of Cr2O3 for sodium peroxide fusion and the perchloric acid method were 75.5 +/- 4.3 and 89.9 +/- 2.5, respectively. As long as percentage recovery is constant, the transit time as determined by calculation of the time of 80% excretion of the total recovered Cr2O3 is not affected. Sodium peroxide fusion provides a useful and safer alternative to perchloric acid oxidation in transit-time studies using Cr2O3 as a nonabsorbable marker.