Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer Int. J. Cancer: 75, 439–443 (1998) r 1998 Wiley-Liss, Inc. EFFECT OF WHEAT BRAN FIBER ON THE DEVELOPMENT OF MAMMARY TUMORS IN FEMALE INTACT AND OVARIECTOMIZED RATS TREATED WITH 7,12-DIMETHYLBENZ(A)ANTHRACENE AND IN MICE WITH SPONTANEOUSLY DEVELOPING MAMMARY TUMORS Maija H. ZILE1*, Clifford W. WELSCH2 and Margaret A. WELSCH1 of Food Science and Human Nutrition, Michigan State University, East Lansing, MI, USA 2Pharmacology and Toxicology, Michigan State University, East Lansing, MI, USA 1Department We examined the effect of consumption of graded increases of dietary fiber (soft white wheat bran) on the development of mammary gland carcinomas in intact female Sprague-Dawley rats during the promotion stage of carcinogenesis, induced with 7,12-dimethylbenz(a)anthracene (DMBA). The percent of rats with mammary carcinomas, the total number of mammary carcinomas and the mean number of mammary carcinomas per rat were reduced significantly at all fiber levels examined compared to rats fed a control diet. Inclusion of 9.6% fiber in the diets of ovariectomized rats that had been treated with a single i.v. dose of 2.5 mg DMBA/100 g body weight 2 weeks prior to removal of the ovaries resulted in a significant decrease of carcinomatous and benign mammary tumors compared to ovariectomized rats fed a control diet. Development of spontaneous mammary carcinomas in virgin C3H/HeOuJ female mice and growth of a transplantable mammary gland tumor in such mice were reduced by inclusion of 9.6% fiber in the diet, a reduction that was significant or just barely missed significance, depending on the source of the fiber. Our observations provide evidence that inclusion of soft white wheat bran in the diet is effective in the suppression of mammary gland tumorigenesis in an array of experimental animal models. Int. J. Cancer 75:439–443, 1998. r 1998 Wiley-Liss, Inc. Breast cancer has a mortality rate of over 50% and is among the major causes of death in women worldwide (Stoll, 1987). While the exact cause(s) and the molecular mechanism(s) involved are not clear, most of the evidence available links breast cancer to the function of ovaries, i.e., the ovarian hormones estrogen and progesterone (Kelsey, 1979; Lubin et al., 1982; Helmrich et al., 1983; reviewed by Sheikh et al., 1994–1995). Epidemiological studies suggest that diet may play an important role in breast cancer (Haenszel and Kurihara, 1968; Gray et al., 1979; Persky et al., 1992). High fat intake has been linked generally to an increase in breast cancer risk (reviewed by Willett and Hunter, 1993), while epidemiological evidence indicates that fiber may have a protective role (reviewed by Rose, 1990). The major support for the latter hypothesis comes from the observation that in Finland, where fat intake is high and similar to that in the United States but fiber intake is much higher than that in the United States, breast cancer incidence is considerably lower than in the United States (reviewed by Cohen et al., 1991). However, a prospective cohort study found no significant association between breast cancer risk and fiber intake (Verhoeven et al., 1997). Nutritional epidemiological investigations clearly can provide only clues to the dietary factors that might participate in breast cancer etiology. Experimental studies are needed to establish a clear relationship between the development of mammary cancer and dietary fiber. Only limited experimental data are available on the potentially beneficial effect of dietary fiber in the prevention of mammary cancer in experimental animals. Cohen et al. (1991) reported that feeding a diet high in fiber (soft white wheat bran) to female F344 rats suppressed the promotional stage of N-methyl-N-nitrosourea (MNU)-induced mammary gland tumorigenesis. The percent of rats with mammary carcinomas, the total number of mammary carcinomas and the mean number of mammary carcinomas per rat were reduced significantly ( p , 0.01) in the fiber-fed rats. This significant reduction in mammary gland tumorigenesis was observed in the rats fed a high-fat diet; in animals fed a low-fat diet, no significant effect of dietary fiber on mammary gland tumorigenesis was observed. During the same year, Arts et al. (1991) reported that high levels of dietary fiber (wheat bran) fed to female F344 rats commencing 3 weeks before MNU treatment and for the duration of the study resulted in a significant reduction ( p , 0.01) in mean mammary carcinoma weight; however, the percent of rats with mammary carcinomas, the mean number of mammary carcinomas per rat and the mean latency period of mammary carcinoma appearance were not influenced significantly by the high-fiber diet. Similarly, Fisher et al. (1985) reported the lack of a significant effect of high levels of dietary fiber (white bread, wholemeal bread or bran) on the spontaneous development of mammary fibroadenomas or mammary carcinomas in female Wistar rats fed the fiber-supplemented diets throughout their life-span. From these reports, it is reasonable to conclude that a suppressive role for dietary fiber in the development and/or growth of mammary gland tumors in experimental animals has not been established. More research is required to clarify the role of dietary fiber in reducing the risk of mammary cancer. In the studies reported here, we have undertaken to examine further the potentially beneficial role of dietary fiber in the suppression of mammary cancer in carcinogen-treated female rats. We also address the aspect of dose responsiveness. In addition, we examined the effect of dietary fiber in ovariectomized rats to test the ‘‘estrogen hypothesis,’’ which has been used to explain the mechanism by which dietary fiber suppresses experimental mammary gland tumorigenesis and has been implied when interpreting epidemiological studies (Rose, 1990; Adlercreutz et al., 1986). If dietary fiber can significantly suppress the development of mammary cancers in ovariectomized animals, i.e., in animals that develop ovarian hormone-independent mammary tumors, it may be concluded that dietary fiber can act via a mechanism independent of estrogen activities. Furthermore, we report on the effect of dietary fiber on the development of spontaneous mammary cancer using C3H mice, the most commonly used animal model for the study of spontaneous mammary tumorigenesis. Data also are presented on the effect of dietary fiber on a transplantable murine mammary cancer cell line. Contract grant sponsor: Kellogg Corporation. *Correspondence to: Department of Food Science and Human Nutrition, 234 Trout Bldg., Michigan State University, East Lansing, MI 48824, USA. Fax: (517) 353-8963. E-mail: [email protected] Received 15 July 1997 ZILE ET AL. 440 TABLE II – DIET COMPOSITION FOR MICE MATERIAL AND METHODS Animals and diets Female Sprague-Dawley rats were purchased from Harlan Sprague-Dawley (Indianapolis, IN). Animals were housed 3 per cage in plastic boxes in an environmentally controlled room (light 14 hr, temperature 24°C, relative humidity 40%). Body weights were recorded weekly. Water and diets were provided ad libitum. The composition of the diets is described in Table I. Two sources of fiber were used: soft white wheat bran from the American Association of Cereal Chemists (St. Paul, MN; Minnesota fiber) and soft white wheat bran from King Milling Co. (Lowell, MI; Michigan fiber). The ‘‘low-fiber’’ group received the Minnesota fiber at 5% of the diet. ‘‘Intermediate-fiber’’ groups received the Minnesota fiber or the Michigan fiber at 9.6% of the diet. The ‘‘high-fiber’’ group received the Minnesota fiber at 17.5% of the diet. Female virgin C3H/HeOuJ mice were obtained from the Jackson Laboratory (Bar Harbor, ME). Mice were housed in plastic boxes, 5 per box, in an environmentally controlled room. Body weight was recorded weekly; water and diets were provided ad libitum. The composition of the diets is given in Table II. In these studies, the Minnesota fiber and the Michigan fiber were examined only at the intermediate level, i.e., 9.6% of the diet. TABLE I – DIET COMPOSITION FOR RATS Diet Control Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total Low-level fiber Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total Intermediate-level fiber Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total High-level fiber Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total % Weight kcal/rat/day g/rat/day 20.00 20.17 0.35 33.08 16.09 4.13 1.18 5.00 — 100.00 17.66 7.91 0.14 12.98 6.31 — — — — 45.00 1.96 1.98 0.04 3.25 1.58 0.41 0.12 0.49 — 9.83 18.94 19.13 0.39 31.40 15.27 3.96 1.16 4.73 5.00 99.98 17.64 7.92 0.16 13.00 6.32 — — — — 45.04 1.96 1.98 0.04 3.25 1.58 0.41 0.12 0.49 0.52 10.35 18.05 18.22 0.37 29.90 14.54 3.77 1.10 4.51 9.57 100.03 17.64 7.92 0.16 13.00 6.32 — — — — 45.04 1.96 1.98 0.04 3.25 1.58 0.41 0.12 0.49 1.04 10.87 16.46 16.62 0.34 27.29 13.27 3.44 1.01 4.11 17.46 100.00 17.64 7.92 0.16 13.00 6.32 — — — — 45.04 1.96 1.98 0.04 3.25 1.58 0.41 0.12 0.49 2.08 11.91 Diet Control Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total Intermediate-level fiber Fat (corn oil) Casein d,l-methionine Dextrose Sucrose AIN mineral mix AIN vitamin mix Celufil Fiber Total % Weight kcal/mouse/day g/mouse/day 20.00 20.17 0.35 33.08 16.09 4.13 1.18 5.00 — 100.00 7.06 3.17 0.06 5.19 2.52 — — — — 18.00 0.78 0.79 0.02 1.30 0.63 0.16 0.05 0.20 — 3.93 18.05 18.22 0.37 29.90 14.54 3.77 1.10 4.51 9.57 100.03 7.06 3.17 0.06 5.19 2.52 — — — — 18.00 0.78 0.79 0.02 1.30 0.63 0.16 0.05 0.20 0.41 4.34 Diet components The components of the diets were as follows: dextrose and AIN-76 Mineral Mix (Amersham, Arlington Heights, IL); casein (vitamin-free), sucrose, corn oil and AIN-76 vitamin mix (Dyets, Bethlehem, PA); celufil (USB, Cleveland, OH). Rationale for the selection of experimental diets Rat diets were iso-energetic; each diet provided each rat the same quantity of fat, protein, dextrose, sucrose, vitamins, minerals and celufil. Diets were constructed based on an average consumption of 45 kcal/rat/day. Minerals, vitamins, celufil and fiber (Minnesota fiber or Michigan fiber) were designated as being of zero caloric value. This is not entirely correct as celufil and the fiber used in these studies have some caloric value, though it is not known precisely. Mouse diets were also iso-energetic; each diet provided each mouse with the same quantity of fat, protein, dextrose, sucrose, vitamins, minerals and celufil. Diets were constructed based on an average consumption of 18 kcal/mouse/day. Experimental design and methods Study 1 was designed to determine whether or not graded increases in dietary fiber consumption (Minnesota fiber) could suppress the development of 7,12-dimethylbenz(a)anthracene (DMBA)-induced mammary gland tumorigenesis (promotion stage) in female Sprague-Dawley rats in a dose-related manner. Female Sprague-Dawley rats were obtained at 36 days of age and fed the control diet. At the age of 50 days, all animals were injected i.v. with a single dose of DMBA (Upjohn, Kalamazoo, MI), 2.5 mg/100 g body weight (bw). At the age of 57 days, animals were divided into 5 groups, 35 rats/group. One group was continued on the control diet. The 2nd group was provided a diet containing 5% Minnesota fiber (low fiber). The 3rd group was provided a diet containing 9.6% Minnesota fiber (intermediate Minnesota fiber), while a 4th group received 9.6% Michigan fiber (intermediate Michigan fiber). The 5th group was provided with 17.6% of Minnesota fiber in the diet (high fiber). Mammary tumor development was assessed by determining the following parameters: percent of rats/group with mammary carcinomas and mean number of mammary carcinomas/rat. All animals were killed at 141 days of age, 13 weeks after carcinogen treatment. Study 2 was designed to determine whether or not high levels of dietary fiber can suppress the development of DMBA-induced mammary gland tumorigenesis (promotion stage) in female Sprague- DIETARY FIBER AND MAMMARY TUMORS Dawley rats ovariectomized 2 weeks after carcinogen treatment. The mammary carcinomas that develop in ovariectomized rats are ovarian hormone-independent. This study was designed to examine the ‘‘estrogen hypothesis’’ as an explanation for the mechanism by which dietary fiber suppresses experimental mammary gland tumorigenesis. If dietary fiber can significantly suppress the development of mammary carcinomas in this model system, it can be concluded that dietary fiber can act via a mechanism independent of estrogen activities. Female Sprague-Dawley rats were obtained at 36 days of age and fed the control diet. At the age of 50 days, rats were injected i.v. with a single dose of DMBA, 2.5 mg/100 g bw. At 64 days of age, all rats were bilaterally ovariectomized. At the age of 71 days, rats were divided into 3 groups, 59 to 60 rats/group. The first group was continued on the control diet, the second group was fed 9.6% Minnesota fiber in the diet (intermediate Minnesota fiber) and the third group was fed 9.6% Michigan fiber in the diet (intermediate Michigan fiber). Mammary tumor development was assessed by determining the following parameters: percent of rats/group with mammary tumors and mean number of mammary tumors/rat. The histopathology of each mammary tumor was determined as in this model system (DMBA followed by ovariectomy, termination of study 29 weeks after DMBA treatment) many benign as well as carcinomatous mammary tumors are being observed. In DMBA-treated intact female rats, at 13 weeks after DMBA treatment (Study 1), virtually all of the mammary tumors were carcinomas. All animals were killed at 253 days of age, 29 weeks after carcinogen treatment. Study 3 was designed to determine whether or not the 9.6% dietary fiber level could suppress the development of spontaneous mammary gland carcinomas in female C3H mice. Virgin female C3H mice (C3H/HeOuJ) were obtained at the age of 21 days and fed the control diet for 2 weeks, then divided into 3 groups of 80 mice each. One group was continued on the control diet, the second group received 9.6% Minnesota fiber in the diet (intermediate Minnesota fiber) and the third group was fed a diet with 9.6% Michigan fiber (intermediate Michigan fiber). Mammary carcinoma development was assessed by determining the following parameters: percentage of mice in each group with mammary carcinomas and mean number of mammary carcinomas/mouse. All animals were killed at 330 days of age, 300 days after the commencement of feeding the fiber diets. (After this study began, we were informed by the staff at the Jackson Laboratories that the activity of the mouse mammary tumor virus [MTV] had recently and unexpectedly declined substantially in their female virgin C3H/HeOuJ strain. Thus, as will be seen below, the percent of these mice with mammary carcinomas, i.e., 24% in the control group, is reduced sharply from past incidence levels for this age.) Study 4 was designed to determine if the growth of a transplantable mammary tumor could be suppressed by providing 9.6% fiber in the diet of host mice. Virgin female C3H/HeOuJ mice were obtained at the age of 21 days and fed the control diet for 10 days, after which they were divided in 3 groups of 67 to 75 mice/group and fed either the control diet or a diet with 9.6% Minnesota fiber (intermediate Minnesota fiber) or a diet with 9.6% Michigan fiber 441 (intermediate Michigan fiber) for 10 months. Transplantable mouse mammary tumor cells (ATCC-CRL-1637) were obtained from the ATCC (Rockville, MD), grown in culture for 2 weeks, harvested and implanted s.c. (5 3 105 cells/mouse) in the scapular region of each mouse at 10 months of age. The transplanted mammary tumor cells were allowed to grow for 1 month, at which time the study was terminated and the tumors excised and weighed. Statistical analysis Mean number of mammary tumors per animal was analyzed by 1-way ANOVA. Mean separations were performed using Duncan’s multiple range test; x2 analysis was used to determine percent of animals with mammary tumors. For mean values to be significantly different, p had to be equal to or less than 0.05. Detection of mammary tumors In Studies 1 and 2, mammary tumor palpations began 1 month after DMBA treatment and were performed bi-weekly until study termination. When tumors reached 2.0 cm in diameter, tumors were excised surgically from the lightly anesthetized animal and the animal was placed back on the experiment. In Study 3, mammary tumor palpation began at 4 months of age and continued monthly until termination of study. At the termination of studies, all animals were killed and palpable as well as non-palpable mammary tumors were excised. RESULTS Effect of dietary fiber on the promotion stage of DMBA-induced mammary gland tumorigenesis of intact rats The percent of rats with mammary carcinomas, the total number of mammary carcinomas and the mean number of mammary carcinomas per rat are shown in Table III. There was a striking decrease (44%) in the total number of mammary carcinomas when rats were fed a diet containing the 9.6% and 17.5% levels of the Minnesota fiber. Indeed, the mean number of mammary carcinomas per rat was reduced significantly at all levels of fiber tested, and this reduction was greater when fiber level was increased. Reductions of mean mammary carcinomas per rat were 22% at the low (5%) fiber level and 43% and 45% at the medium (9.6%) and high (17.5%) levels, respectively. While the percentage of rats with mammary tumors in all fiber-fed groups (89–94%) was numerically less than the controls (100%), the relatively large number of mammary tumors that developed in each group does not allow for this parameter to be significant. Feces were collected from 12 rats in each group every 96 hr. Rats were housed individually in metabolism cages. Dry fecal weight and fecal volume (not packed) were increased in a dose-dependent manner as the dietary fiber content was increased. From a control group level of 61.0 g of dry feces, this parameter increased by 18% (to 72.3 g), 32% (to 80.3 g) and 35% (to 82.1 g) when Minnesota fiber was fed at dietary levels of 5%, 9.6% and 17.5%, respectively. This proportional increase in fecal dry weight was also evident when calculated per rat per 24 hr. TABLE III – EFFECT OF WHEAT BRAN FIBER ON DEVELOPMENT OF MAMMARY CARCINOMAS IN INTACT FEMALE SPRAGUE-DAWLEY RATS TREATED WITH DMBA Group Number of rats per group Mean initial body weight (g) Mean final body weight (g) Rats with mammary carcinomas (%) Total number of mammary carcinomas Mean number of mammary carcinomas per rat (6S.E.) Control Low Minnesota fiber Intermediate Minnesota fiber High Minnesota fiber Intermediate Michigan fiber 35 35 35 35 35 194 189 190 192 189 271 265 258 252 250 100 89 94 89 94 203 156 115 113 144 5.8 6 0.81 4.5 6 0.53 3.3 6 0.44 3.2 6 0.43 4.1 6 0.52 All rats were treated with DMBA at 50 days of age. Fiber diets were begun 1 week later and continued daily for 12 weeks until experiment termination.–1/2p , 0.05.–1/4p , 0.02.–1/3p , 0.01.–1/2/3/4p , 0.001. ZILE ET AL. 442 TABLE IV – EFFECT OF WHEAT BRAN FIBER ON DEVELOPMENT OF MAMMARY TUMORS IN OVARIECTOMIZED SPRAGUE-DAWLEY RATS TREATED WITH DMBA Number Mean Mean Rats with Total number Mean number of Histopathology: total number of mammary tumors of rats initial body final body mammary of mammary mammary tumors per group weight (g) weight (g) tumors (%) tumors per rat (6S.E.) Carcinomas Adenomas Adenofibromas Fibromas Group Control Intermediate Minnesota fiber Intermediate Michigan fiber 60 59 59 194 197 201 362 346 342 681 492 462 1.6 6 0.25 0.8 6 0.13 0.9 6 0.24 94 45 55 606 387 467 188 29 39 1410 511 611 2 — — All rats were treated with DMBA at 50 days of age; each rat was ovariectomized at 64 days of age. Fiber diets were begun 1 week after ovariectomy and continued daily for 26 weeks, at which time the experiment was terminated.–1/2p , 0.001.–3/4p , 0.032.–3/5p , 0.004.–6/7p , 0.01.–8/9p , 0.002.–10/11p , 0.05. TABLE V – EFFECT OF WHEAT BRAN FIBER ON SPONTANEOUSLY DEVELOPING MAMMARY CARCINOMAS IN VIRGIN C3H/HeOuJ MICE Group Control Intermediate Minnesota fiber Intermediate Michigan fiber Number of mice Mean number of Number Mean initial Mean final with spontaneous spontaneous of mice body weight (g) body weight (g) mammary mammary carcinomas per group carcinomas (%) per mouse (6S.E.) 80 80 80 20 19 20 44 44 45 0.26 6 0.064 0.08 6 0.036 0.20 6 0.055 19 (24)1 6 (8)3 14 (18)2 Fiber diets were begun at 1 month of age and continued daily until experiment termination at 11 months of age.–1/2p , 0.15.–1/3p , 0.001.–4/5p . 0.60.–4/6p , 0.02. Similarly, fecal volume increased from the control volume of 163 ml to 188 ml, 240 ml and 291 ml, respectively. Effect of dietary fiber on DMBA-induced mammary gland tumorigenesis of ovariectomized rats The percent of rats with mammary tumors, the total number of mammary tumors, the mean number of mammary tumors per rat and the tumor histopathology are shown in Table IV. Results represent 2 experimental groups at the 9.6% (intermediate) dietary fiber level. With both the Minnesota fiber and the Michigan fiber there was a significant reduction in the number of rats with mammary tumors (27% and 32%, respectively), as well as in total number of mammary tumors (52% and 41%, respectively) and in mean number of mammary tumors per rat. Histopathological examination revealed that the majority of mammary tumors in ovariectomized rats were carcinomas. Ovariectomized rats that had been provided with 9.6% (intermediate level) of Minnesota fiber had 36% fewer carcinomas, and rats fed 9.6% (intermediate level) of the Michigan fiber had 22% fewer carcinomas. The total number of benign mammary tumors (adenomas, adenofibromas) also was reduced significantly in the ovariectomized rats fed the fiber diets compared to control rats. Mean uterine wet weights (g 6 S.E.) in rats from the ovariectomized controls, the Minnesota fiber group and the Michigan fiber group were 0.109 6 0.009, 0.131 6 0.015 and 0.112 6 0.012, respectively. These differences were not significant. Effect of dietary fiber on spontaneously developing mammary carcinomas in C3H/HeOuJ mice The number and percent of mice with mammary carcinomas and the mean number of spontaneous mammary carcinomas per mouse are shown in Table V. Results represent 2 experimental groups at the intermediate (9.6%) dietary fiber level. Both dietary fiber groups (Minnesota and Michigan) caused a numerical reduction in both of these parameters of mammary tumorigenesis compared to control values; only the reduction in tumorigenesis by the Minnesota fiber was significant. Effect of dietary fiber on growth of transplantable mammary tumors in C3H/HeOuJ mice This study represents 2 experimental groups at the intermediate (9.6%) dietary fiber level and the effect of fiber on the weight of the transplanted mammary tumors at experiment termination (Table VI). A statistically significant reduction was observed only in mice TABLE VI – EFFECT OF WHEAT BRAN FIBER ON GROWTH OF TRANSPLANTABLE MAMMARY TUMORS IN FEMALE C3H/HeOuJ MICE Mean weight of transplanted Number of per mouse mice per group mammary(gtumors 6 S.E.) Group Control Intermediate Minnesota fiber Intermediate Michigan fiber 68 75 67 1.26 6 0.131 1.00 6 0.083 0.81 6 0.092 Fiber diets were begun at 1 month of age and continued daily until experiment termination at 11 months of age. Transplantable mammary tumor cells (ATCC-CRL-1637) were implanted s.c. in scapular region of each mouse at 10 months of age, excised and weighed at experiment termination at 11 months of age.–1/2p , 0.01.–1/3p , 0.15. fed the Michigan fiber; Minnesota fiber did not reduce significantly ( p , 0.15) the weight of the transplanted tumors. DISCUSSION Since breast cancer is of major consequence to womens’ health, it is important to examine every potential approach that might lead to a lowering of the incidence of this disease. Decreasing cancer risk by modulation of dietary behavior is an attainable goal. The health benefits of increased dietary fiber intake are generally not disputed, particularly in regard to bowel health, while the link to lowering heart disease has not been established clearly (Rimm et al., 1996; Katan, 1996). Evidence from epidemiological studies suggests that breast cancer risk also may be lowered by an increased dietary fiber consumption (Rose, 1990). There are several studies that have addressed this important question using controlled experimental animal models (Fisher et al., 1985; Cohen et al., 1991; Arts et al., 1991), but the overall results have been inconclusive. In our studies, we have addressed several questions concerning the relationship of dietary fiber and mammary tumor development in experimental animals. One observation is that the total number of mammary tumors as well as the number of mammary tumors per rat were numerically decreased as dietary soft white wheat bran was increased from 5% to 9.6% and 17.5% (Table III). Another aspect that we addressed concerns the potential mechanism of action of fiber in suppressing mammary tumor development. DIETARY FIBER AND MAMMARY TUMORS Adlercreutz et al. (1975, 1986) have proposed that the lower risk of breast cancer in women with high fiber intake is related to a lower level of estrogens in their circulation as a consequence of an enhanced removal of conjugated estrogens by fiber in the bowel, with microflora playing an important role in the enterohepatic recirculation of estrogen (see also Gorbach, 1984; Schultz and Howie, 1986). Another mechanism proposed by Adlercreutz et al. (1986) is that dietary fiber may be fermented by microflora in the bowel to form lignans, which may act as regulators of estrogen activity. These mechanisms are based on the concept that dietary fiber suppresses mammary (breast) tumorigenic processes by modulation of estrogen activity. We have addressed this question by examining the effect of dietary fiber in ovariectomized rats since ovaries are the major source of estrogen in rats. We observed a highly significant reduction in total mammary tumors, carcinomatous and benign, with 9.6% dietary fiber in ovariectomized rats. Indeed, the strong suppression of mammary tumorigenesis by dietary fiber was comparable quantitatively in the intact and ovariectomized carcinogen-treated rats. Mammary tumors that arise in ovariectomized animals are ovarian hormone-independent for growth processes. That dietary fiber can suppress the developmental growth of such tumors suggests that mechanisms other than modulation of estrogen activities also must be considered to explain the beneficial effect of fiber. Our demonstration that uterine weight in the ovariectomized rats was not affected by the dietary fiber provides evidence that the fiber used in these studies does not contain or generate agonistic or antagonistic estrogenic activities. 443 A mechanism that warrants increased attention is that dietary fiber consumption results in a considerable increase in fecal bulk and weight, as observed in our studies; decreased fecal transit time also is associated with dietary fiber consumption. The above factors may result in an increased excretion of high calorie-generating dietary components as well as an enhanced elimination of dietary and bowel components from microbial activities that might have tumor growth-promoting activities. Reports suggest that dietary fiber may have immunoregulatory effects on the intestinal immune system (Tappenden et al., 1995; Lim et al., 1997) and, thus, may enhance systemic immune defenses. In all of these experiments, the feeding of dietary fiber did not cause any apparent adverse health effects or any significant changes in body weight gains of the animals. Our salient conclusions are that dietary fiber at the 9.6% level is very effective in suppressing the development of mammary gland tumors in an impressive array of animal models for this disease, i.e., in carcinogen-treated intact rats (develops primarily ovarian hormone-dependent mammary tumors), in carcinogen-treated ovariectomized rats (develops ovarian hormone-independent mammary tumors, carcinomatous and benign), in rodents that develop mammary tumors spontaneously (C3H/HeOuJ mice, MTV-enhanced) and in a transplantable mammary tumor model (C3H/HeOuJ mice/ATCC-CRL-1637 cells). It is hoped that our results will encourage continued study of the relationship between dietary fiber and tumorigenic processes and facilitate a mechanistic understanding of this relationship. REFERENCES ADLERCREUTZ, H., FOTSIS, T., BANNWART, C., WÄHÄLÄ, I., MÄKELÄ, T., BRUNOW, G. and HASE, T., Determination of urinary lignans and phytoestrogen metabolites, potential antiestrogens and anticarcinogens, in urine of women on various habitual diets. J. Steroid Biochem., 25, 791–797 (1986). ADLERCREUTZ, H., MARTIN, F., TIKKANEN, M.J. and PULKKINEN, M., Effect of ampicillin administration on the excretion of twelve estrogens in pregnancy urine. Acta. Endocrinol., 80, 551–557 (1975). ARTS, C.J.M., DE BIE, A.T.H.J., VAN DEN BERG, H., VAN’T VEER, P., BUNNIK, G.S.J. and THIJSSEN, J.H.H., Influence of wheat bran on NMU-induced mammary tumor development, plasma estrogen levels and estrogen excretion in female rats. J. Steroid Biochem. Mol. Biol., 39, 193–202 (1991). COHEN, L.A., KENDALL, M.E., ZANG, E., MESCHTER, C. and ROSE, D.P., Modulation of N-nitrosomethylurea-induced mammary tumor promotion by dietary fiber and fat. J. nat. Cancer Inst., 83, 496–501 (1991). FISHER, N., BERRY, C.S., FEARN, T., GREGORY, J.A. and HARDY, J., Cereal dietary fiber consumption and diverticular disease: a life span study in rats. Amer. J. clin. Nutr., 42, 788–804 (1985). GORBACH, S.L., Estrogens, breast cancer, and intestinal flora. Rev. infect. Dis., 6 (Suppl. 1), S85–S90 (1984). GRAY, G.E., PIKE, M.C. and HENDERSON, B.E., Breast cancer incidence and mortality rates in different countries in relation to known risk factors and dietary practices. Brit. J. Cancer, 39, 1–7 (1979). HAENSZEL, W. and KURIHARA, M., Studies of Japanese migrants. I. Mortality from cancer and other diseases among Japanese in the United States. J. nat. Cancer Inst., 40, 43–68 (1968). HELMRICH, S.P. and 13 OTHERS, Risk factors for breast cancer. Amer. J. Epidemiol., 117, 35–45 (1983). KATAN, M.B., Forever fiber (a review). Nutr. Rev., 54, 253–254 (1996). KELSEY, J.L., A review of the epidemiology of human breast cancer. Epidemiol. Rev., 1, 74–109 (1979). LIM, B.O., YAMADA, K., NONAKA, M., KURAMOTO, Y., HUNG, P. and SUGANO, M., Dietary fibers modulate indices of intestinal immune function in rats. J. Nutr., 127, 663–667 (1997). LUBIN, J.H., BURNS, P.E., BLOT, W.J., LEES, A.W., MAY, C., MORRIS, L.E. and FRAUMENI, J.F., JR., Risk factors for breast cancer in women in northern Alberta, Canada, as related to age at diagnosis. J. nat. Cancer Inst., 68, 211–217 (1982). PERSKY, V.W., CHATTERTON, R.T., VAN HORN, L.V., GRANT, M.D., LANGENBERG, P.A. and MARVIN, J., Hormone levels in vegetarian and nonvegetarian teenage girls: potential implications for breast cancer risk. Cancer Res., 52, 578–583 (1992). RIMM, E.B., ASCHERIO, A., GIOVANNUCCI, E., SPIEGELMAN, D., STAMPFER, M.J. and WILLETT, W.C., Vegetable, fruit, and cereal fiber intake and risk of coronary heart disease among men. J. Amer. med. Ass., 275, 447–451 (1996). ROSE, D.P., Dietary fiber and breast cancer. Nutr. Cancer, 13, 1–8 (1990). SCHULTZ, T.D. and HOWIE, B.J., In vitro binding of steroid hormones by natural and purified fibers. Nutr. Cancer, 8, 141–147 (1986). SHEIKH, M.S., GARCIA, M., PUJOL, P., FONTANA, J.A. and ROCHEFORT, H., Why are estrogen-receptor negative cancers more aggressive than estrogenreceptor-positive breast cancers? Invasion Metastasis, 14, 329–336 (1994– 1995). STOLL, B.A., Prognostic indices in breast cancer. In: B.A. Stoll (ed.), Pointers to cancer prognosis, pp. 195–229, Nijhoff, Dordrecht (1987). TAPPENDEN, K.A., PRATT, V.C., GORUK, S.D., FIELD, C.J. and MCBURNEY, M.C., Short chain fatty acid (SCFA) supplementation of total parenteral nutrition (TPN) improves whole body response to intestinal resection. FASEB J., 9, 862A (1995). VERHOEVEN, D.T.H., ASSEN, N., GOLDBOHM, R.A., DORANT, E., VAN’T VEER, P., STURMANS, F., HERMUS, R.J. and VAN DEN BRANDT, P.A., Vitamins C and E, retinol, beta-carotene and dietary fiber in relation to breast cancer risk: a prospective cohort study. Brit. J. Cancer, 75, 149–155 (1997). WILLETT, W. and HUNTER, D.J., Diet and breast cancer. Contemp. Nutr., 18, 1–4 (1993).