Arsenite is a cocarcinogen with solar ultraviolet radiation for mouse skin: an animal model for arsenic carcinogenesis.

Although epidemiological evidence shows an association between arsenic in drinking water and increased risk of skin, lung, and bladder cancers, arsenic compounds are not animal carcinogens. The lack of animal models has hindered mechanistic studies of arsenic carcinogenesis. Previously, this laboratory found that low concentrations of arsenite (the likely environmental carcinogen) which are not mutagenic can enhance the mutagenicity of other agents, including ultraviolet radiation (UVR). This enhancing effect appears to result from inhibition of DNA repair by arsenite. Recently we found that low concentrations of arsenite disrupted p53 function and upregulated cyclin D1. These results suggest that the failure to find an animal model for arsenic carcinogenesis is because arsenite is not a carcinogen per se, but rather acts as an enhancing agent (cocarcinogen) with a genotoxic partner. We tested this hypothesis with solar UVR as carcinogenic stimulus in hairless Skh1 mice. Mice given 10 mg/l sodium arsenite in drinking water for 26 weeks had a 2.4-fold increase in yield of tumors after 1.7 KJ/m(2) UVR three times weekly compared with mice given UVR alone. No tumors appeared in mice given arsenite alone. The tumors were mostly squamous cell carcinomas, and those occurring in mice given UVR plus arsenite appeared earlier and were much larger and more invasive than in mice given UVR alone. These results are consistent with the hypothesis that arsenic acts as a cocarcinogen with a second (genotoxic) agent by inhibiting DNA repair and/or enhancing positive growth signaling.

Chemoprevention trials in men with prostate-specific antigen failure or at high risk for recurrence after radical prostatectomy: Application to efficacy assessment of soy protein.

This article discusses the basic elements of chemoprevention trial designs using cohorts of men following radical prostatectomy who either have prostate-specific antigen (PSA) failure indicative of recurrence or are at high risk for recurrence (positive surgical margins, extracapsular extension, seminal vesicle invasion, positive lymph nodes, Gleason score of greater than or equal to 8, preoperative serum PSA less than 20 ng/mL). Two ongoing randomized, double-blind, placebo-controlled clinical trials with soy protein as intervention in these 2 populations are described. In the trial with men at high risk for recurrence, participants started intervention within 4 months after surgery and were followed for up to 2 years; primary endpoints were PSA failure rate and time-to-PSA failure. In the trial with men with PSA failure (PSA 0.1 to 2.0 ng/mL), participants received treatment for 8 months and the primary endpoint is rise in PSA over time. The strengths and limitations of these designs are discussed and interim experience using studies with soy protein as the intervention agent are summarized.

The role of steroid hormones in prostate carcinogenesis.

Carcinoma of the prostate is the most frequently diagnosed malignancy and the second leading cause of death as a result of cancer in men in the United States and in many other Western countries. Notwithstanding the importance of this malignancy, little is understood about its causes. The epidemiology of prostate cancer strongly suggests that environmental factors, particularly diet and nutrition, are major determinants of risk for this disease, and evidence is mounting that there are important genetic risk factors for prostate cancer. Human prostate carcinomas are often androgen sensitive and react to hormonal therapy by temporary remission, followed by relapse to an androgen-insensitive state. These well-established features of prostate cancer strongly suggest that steroid hormones, particularly androgens, play a major role in human prostatic carcinogenesis, but the precise mechanisms by which androgens affect this process are unknown. In addition, the possible involvement of estrogenic hormones is not entirely clear. The purpose of this overview is to summarize the literature about steroid hormonal factors, androgens and estrogens, and prostate carcinogenesis. From these literature observations, a multifactorial general hypothesis of prostate carcinogenesis emerges with androgens as strong tumor promoters acting via androgen receptor-mediated mechanisms to enhance the carcinogenic activity of strong endogenous genotoxic carcinogens, such as reactive estrogen metabolites and estrogen- and prostatitis-generated reactive oxygen species and possible weak environmental carcinogens of unknown nature. In this hypothesis, all of these processes are modulated by a variety of environmental factors such as diet and by genetic determinants such as hereditary susceptibility and polymorphic genes that encode for steroid hormone receptors and enzymes involved in the metabolism and action of steroid hormones.

Multiple pathways of prostate carcinogenesis analyzed by using cultured cells isolated from rats treated with N-methyl-N-nitrosourea and testosterone.

Treatment of rats with N-methyl-N-nitrosourea (MNU) and testosterone results in a high incidence of metastasizing dorsolateral prostate tumors. In previous studies, a high frequency (> or = 70%) of a G35 --> A transition mutation at the second position of codon 12 of the Ki-ras oncogene was found in these tumors. This was confirmed in the study reported here, and the frequency of this mutation appeared similar in tumors induced in four different rat strains, regardless of differences in sensitivity among these strains to the induction of prostate cancers by MNU and testosterone: Wistar Furth (62% incidence of grossly visible prostate tumors) > Lobund Wistar (55%) > Fisher 344 (40%) > Copenhagen (37%). A method was developed to isolate and separately culture epithelial and stromal cells from these rat prostate carcinomas. Of 20 primary cell cultures established from histologically confirmed rat prostate carcinomas, 19 (95%) displayed one or more of the following characteristics: the Ki-ras mutation (17 of 20; 85%), anchorage-independent growth in soft agar at early passage (12 of 20; 60%), or tumorigenicity at later passage (eight of eight; 100%). One epithelial cell culture and all five stromal cell cultures established from prostate tumors had none of these characteristics. Epithelial cultures that had the Ki-ras mutation and grew in soft agar constitute the predominant genotype/phenotype (55%), cultures with the mutation that did not grow in soft agar were less frequent (30%), 10% of the cultures had neither characteristic, and only one grew in soft agar but did not have the mutation. These findings suggest that there are at least two and perhaps more different molecular pathways of prostate carcinogenesis in rats treated with MNU plus testosterone. Furthermore, these data suggest that these pathways and the mechanisms determining strain differences in sensitivity to prostate cancer induction are unrelated.

Chemoprevention of rat prostate carcinogenesis by early and delayed administration of dehydroepiandrosterone.

Two in vivo bioassays were conducted to evaluate the efficacy of dehydroepiandrosterone (DHEA) as an inhibitor of prostate carcinogenesis in rats. Prostate adenocarcinomas were induced in male Wistar-Unilever rats by a sequential regimen of cyproterone acetate and testosterone propionate, followed by a single i.v. injection of N-methyl-N-nitrosourea (MNU) and chronic androgen stimulation. In the first experiment, DHEA (1000 or 2000 mg/kg diet) was administered continuously to rats beginning 1 week before MNU exposure. In the second experiment, continuous administration of DHEA (2000 mg/kg diet) was begun either 1 week before, 20 weeks after, or 40 weeks after MNU exposure. Controls received basal diet without added DHEA. Studies were terminated at 13 months after MNU administration, and prostate cancer incidence was determined by histopathological evaluation of step sections of accessory sex glands. In the first study, continuous dietary administration of DHEA beginning 1 week before MNU resulted in a dose-related inhibition of prostate cancer induction. In the second experiment, comparable reductions in prostate cancer incidence were observed in groups exposed to DHEA beginning 1 week before, 20 weeks after, and 40 weeks after carcinogen exposure. These data demonstrate that nontoxic doses of DHEA confer significant protection against prostate carcinogenesis in rats. The efficacy of delayed administration of DHEA suggests that the compound confers protection against later stages of prostate cancer induction and can suppress the progression of existing preneoplastic lesions to invasive disease.

Use of animal models in defining efficacy of chemoprevention agents against prostate cancer.

Animal models are crucial in preclinical efficacy testing of chemoprevention agents. The most feasible, realistic, and potentially effective target for prostate cancer chemoprevention is progression from prostatic intraepithelial neoplasia (PIN) to histologic cancer and from histologic to clinically manifest cancer. There are transgenic mouse models for prostate cancer and models for PIN, but these have not yet been fully developed and evaluated for chemoprevention studies. Human prostate cancer xenografts in mice and transplantable Dunning rat prostate carcinomas can be used to assess tumor growth inhibition. Several Dunning tumors metastasize, enabling detection of inhibition of metastases. Detection of inhibitory effects on de novo prostate cancer development requires induction of a high cancer incidence and similarity of induced tumors to human prostate carcinomas. Transgenic mice with oncogenes expressed in a prostate-specific fashion, combined chronic treatment of NBL rats with estradiol-17beta and testosterone, and sequential treatment of rats with carcinogens such as N-methyl-N-nitrosourea (MNU) and chronic testosterone treatment all lead to a high incidence of prostatic adenocarcinomas. PIN occurs mostly in the former two models, and metastases are frequent in some transgenic models and the MNU-testosterone rat model. The latter model has been applied to chemoprevention agent efficacy testing. In 8 control groups, the carcinoma incidence was 77% in all accessory sex glands combined, 51% for small tumors confined to dorsolateral/anterior prostate, and 25% for large tumors of uncertain origin in the prostate area. This model was predictive of the lack of antiprostate cancer efficacy of N-(4-hydroxyphenyl)all-trans-retinamide in humans. Thus, rats given MNU and chronic testosterone represent a relevant and reliable model for efficacy testing of chemoprevention agents. In conclusion, there are now adequate animal models for prostate cancer proven to be suitable for preclinical chemoprevention studies.

The role of stromal cells in prostate cancer development and progression.

Prostate cancer is one of the most common malignancies in males, and tumor progression critically determines its clinical significance. Prostatic stromal cells may be critically involved in growth and progression of prostate cancer. There is substantial evidence that the stromal component of the embryological tissue of origin, the urogenital sinus, is essential in directing outgrowth and prostatic differentiation of the epithelial anlage of the prostate. The presence of a stromal androgen receptor is required for this effect, and humoral factors, such as keratinocyte growth factor, have been shown to be able to mediate it in a paracrine fashion. The adult prostate is also under control of multiple steroid hormone and paracrine peptide factors, and there is evidence that the prostatic stroma plays a major role in mediation of androgen effects on prostatic epithelium. Normal seminal vesicle mesenchyme can cause differentiation of the Dunning R3327H prostate carcinoma. Normal rat prostatic fibroblasts influence the in vivo and soft agar growth of epithelial cells derived from chemically/hormonally induced rat prostate carcinomas, as do fibroblasts that are isolated from these tumors. Both growth-enhancing and growth-inhibiting effects were observed, apparently depending on the stage of progression of both cell types as well as on whether fibroblasts were derived from the same or a different tumor than the epithelial cells. These findings indicate that stromal cells critically influence epithelial prostate cancer growth, and they suggest that these effects can significantly vary in different tumors as well as in different stages of tumor progression.

Chemoprevention of rat prostate carcinogenesis by 9-cis-retinoic acid.

A chemoprevention study was conducted to evaluate the activity of 9-cis-retinoic acid (9-cis-RA) as an inhibitor of prostate carcinogenesis in male Wistar-Unilever (HsdCpb:Wu) rats. After pretreatment with a sequential regimen of cyproterone acetate (50 mg/kg/day for 21 days) and testosterone propionate (100 mg/kg/day for 3 days), groups of 40 rats received a single i.v. injection of N-methyl-N-nitrosourea (MNU; 30 mg/kg body weight). Beginning 2 weeks after carcinogen administration, rats received chronic exposure to testosterone administered in s.c. implanted silastic capsules. The study was terminated at 13 months after MNU administration, and prostate cancer incidence was determined by histopathological evaluation of step sections of accessory sex glands. Continuous dietary administration of 9-cis-RA at 100 mg/kg diet or 50 mg/kg diet beginning 1 week before MNU administration reduced cancer incidence in the dorsolateral + anterior prostate from 65% in dietary controls to 18 and 20%, respectively (P < 0.001 for both comparisons). Similarly, these dose levels of 9-cis-RA reduced the incidence of cancer in all accessory sex glands from 79% in dietary controls to 48 and 33% (P < 0.01 for both comparisons), respectively. Chronic dietary administration of 9-cis-RA induced no gross or organ-specific toxicity in any animal and did not suppress group mean body weight gain. The potent anticarcinogenic activity of 9-cis-RA in the rat prostate, when considered with its apparent lack of toxicity in rodents, suggests that this and other ligands for the retinoid X receptor merit consideration for evaluation in clinical prostate cancer chemoprevention trials.

Influence of N-methyl-N-nitrosourea, testosterone, and N-(4-hydroxyphenyl)-all-trans-retinamide on prostate cancer induction in Wistar-Unilever rats.

The influence of chemical carcinogen, hormonal stimulation, and chronic dietary administration of the synthetic retinoid, N-(4-hydroxyphenyl)-all-trans-retinamide (4-HPR), on the induction of prostate cancer in male Wistar-Unilever rats was determined. Three different tumor induction regimens were used: (a) a single i.v. dose of 50 mg of N-methyl-N-nitrosourea (MNU) per kg body weight, followed by chronic androgen stimulation via s.c. implantation of two silastic capsules containing 40 mg testosterone each; (b) a single i.v. dose of 50 mg of MNU per kg body weight (no testosterone treatment); and (c) chronic androgen stimulation with implanted testosterone capsules (no MNU treatment). In a fourth series of animals, the incidence of spontaneous prostate tumors was determined in groups of rats receiving neither carcinogen nor hormone stimulation. Within each series, parallel groups of animals were fed a control (vehicle-supplemented) diet or control diet supplemented with 4-HPR beginning 1 day after carcinogen administration; retinoid administration was continuous until termination of the study at 450 days. The incidence of accessory sex gland cancer in rats treated sequentially with MNU + testosterone was >60%, in comparison with cancer incidences of <20% in rats receiving MNU only and <5% in rats treated with testosterone only. No spontaneous accessory sex gland tumors were observed in rats receiving no carcinogen and no testosterone. Tumor induction in the accessory sex glands by MNU + testosterone was relatively specific for the prostate: the incidence of carcinoma of the dorsolateral/anterior prostate was more than 5-fold greater than the incidence of cancer present only in the seminal vesicle. 4-HPR conferred no protection against cancer induction in the prostate by any regimen of MNU and/or testosterone. These results demonstrate the importance of both carcinogen exposure and hormone stimulation on the induction of neoplasia in the prostate of Wistar-Unilever rats.

Influence of acid aerosol droplet size on structural changes in the rat lung caused by acute exposure to sulfuric acid and ozone.

To investigate whether aerosol droplet size influences structural changes in the lung produced by short-term, concomitant exposure to ozone and sulfuric acid, groups of 10 rats were exposed 4 hr/day for 2 days to filtered air, 0.6 ppm ozone, 0.5 mg/m3 fine (aerosol mass median diameter (MMD) = 0.3 microm) or ultrafine (MMD = 0.06 microm) sulfuric acid, or a mixture of ozone and 0.5 mg/m3 fine or ultrafine sulfuric acid. The volume percentage of total parenchyma containing markedly to severely injured alveolar septae was measured morphometrically. There were no differences between the ultrafine or fine acid exposure groups and the sham group for any of the morphologic endpoints. Volume percentage of markedly to severely injured tissue was increased in the ultrafine, but not fine, mixture animals when compared with the ozone-only group. In addition, a synergistic interaction between ozone and ultrafine, but not fine, sulfuric acid was found for this endpoint. The bromodeoxyuridine cell labeling index in the periacinar region was greater in the rats exposed to the fine sulfuric acid and ozone mixture than that in rats exposed to ozone alone, and a synergistic interaction between ozone and fine sulfuric acid was found for this end point. None of the exposures produced any changes in ventilatory parameters. Thus, acid aerosol droplet size was found to influence the effect of sulfuric acid in modifying ozone-induced structural changes in the rat lung.

Hormonal factors in carcinogenesis of the prostate and testis in humans and in animal models.

The etiology of human testicular tumors is poorly defined. With the possible exception of prenatal estrogen exposure, no specific chemical exposures have been associated with testicular cancer risk in men. Prenatal as well as postnatal estrogen treatments induce testicular tumors in some mouse strains, but not in other mouse strains or in rats. Prenatal estrogen exposure also causes cryptorchid testes in mice and possibly rats. Cryptorchidism is a consistent risk factor for testicular cancer in men, and estrogen- or surgically-induced cryptorchidism is associated with Leydig cell tumorigenesis in mice. In rats, however, surgically induced cryptorchidism inhibits Leydig cell tumor formation. Overall, it appears that the mouse is the most appropriate species as animal model for testicular tumorigenesis in humans. Any of the following hormonal exposures can cause testicular tumor formation in rodents: 1) chronic exposure to estrogenic compounds of adult mice and hamsters; 2) prenatal exposure to estrogenic compounds of mice and possibly humans; and 3) any treatment or condition that induces cryptorchidism in mice and humans. The mechanisms whereby these treatments or conditions may cause testicular tumorigenesis are poorly understood. Undefined local testicular factors appear to be dominant in tumorigenesis in cryptorchid human and rodent testes. Pituitary factors, most likely LH and perhaps prolactin, play a critical but poorly defined role in estrogen-induced and spontaneous testicular tumorigenesis in rodents. In the mouse, estrogen receptor-mediated mechanisms seem to be involved in induction of testicular tumors by prenatal estrogen exposure, and a direct, perhaps estrogen receptor-mediated, inhibiting effect of estrogens on the action of müllerian inhibiting substance is probably central in the induction of cryptorchidism in this species.

Induction at high incidence of ductal prostate adenocarcinomas in NBL/Cr and Sprague-Dawley Hsd:SD rats treated with a combination of testosterone and estradiol-17 beta or diethylstilbestrol.

This study determined the incidence of prostate adenocarcinoma following long-term treatment of NBL and Sprague-Dawley rats with estradiol-17 beta or diethylstilbestrol (DES) plus testosterone and it defined the origin of these tumors. NBL and Sprague-Dawley rats were treated with two Silastic tubing implants (i.d. 1.6 mm, o.d. 3.2 mm) containing a 2 cm long filling of testosterone and one implant containing a 1 cm long filling of estradiol-17 beta or DES. Control animals received empty implants. Treated animals were killed when moribund and controls were killed at 91 (NBL) or 75 (Sprague-Dawley) weeks after initiation of treatment and accessory sex glands were sampled for histopathological examination of multiple step sections. Prostatic adenocarcinoma occurred in 100% of NBL rats after treatment with estradiol-17 beta or DES plus testosterone for 44 and 59 weeks (group means) respectively. Adenocarcinoma incidences were lower in Sprague-Dawley rats. The adenocarcinomas were small, microscopic, invasive tumors and they were spatially closely associated with the periurethral ducts of the dorsal, lateral and/or anterior (= coagulating gland) prostate, but never with the ducts of the ventral lobe and seminal vesicles. One adenocarcinoma was of uncertain origin. Duct-acinar dysplastic lesions occurred in the periphery of the dorsal and lateral prostate of all hormone-treated NBL and many Sprague-Dawley rats, but did not appear to give rise to carcinoma. Although some adenocarcinomas were contiguous with dysplastic ducts of the peripheral dorsolateral prostate, the main mass of these neoplasms was located in the periurethral area. Also, most adenocarcinomas were only connected with the periurethral ducts, in which atypical hyperplasia occurred following hormone treatment for 36 weeks or longer. Thus atypical hyperplasia of the periurethral prostate ducts, but not peripheral duct-acinar dysplasia, appeared to be the likely precursor of the induced carcinomas. Testosterone plus DES, but not estradiol-17 beta, induced marked dysplasia-like lesions in the acini of the ventral prostate of all NBL and many Sprague-Dawley rats. These lesions had progressed to carcinoma in situ (or adenoma) in 46% of NBL rats.

Induction of benign prostatic hyperplasia in intact dogs by near-physiological levels of 5 alpha-dihydrotestosterone and 17 beta-estradiol.

Benign prostatic hyperplasia was induced in mongrel dogs treated for 60 days with one silastic implant containing 17 beta-estradiol and four containing 5 alpha-dihydrotestosterone. The condition was characterized by (1) a marked increase of the stromal elements, particularly the stromal septa between the individual glands, (2) a slight increase in prostatic volume, and (3) a morphology that resembled spontaneous complex benign prostatic hyperplasia in the dog. Other groups of animals that remained untreated or received only 17 beta-estradiol or only 5 alpha-dihydrotestosterone did not develop this condition. Prostate volumes decreased by 14% in the estrogen-treated dogs, whereas they increased in the androgen-treated animals by 6% compared to pretreatment prostate volumes. The morphology of the epithelium of the prostates of androgen-treated animals was not different from that of controls despite the increase in prostate volume. The serum 17 beta-estradiol and 5 alpha-dihydrotestosterone concentrations were increased from 25 +/- 2 (mean +/- SEM) and 256 +/- 42 pg/mL, respectively, in control dogs to 52 +/- 37 and 562 +/- 37 pg/mL, respectively, in the dogs treated with the hormone combination. Thus, hormone concentrations were two- to three-fold higher than control values, and the ratio of estradiol-17 beta to 5 alpha-dihydrotestosterone was increased by up to 19%. These data demonstrate that treatment of dogs with low levels of estrogen and androgen may be an excellent model for the study of spontaneous complex benign prostatic hyperplasia in aging men.

Induction of a DNA adduct detectable by 32P-postlabeling in the dorsolateral prostate of NBL/Cr rats treated with estradiol-17 beta and testosterone.

Treatment with estradiol-17 beta and testosterone induces epithelial dysplasia and, subsequently, adenocarcinoma in the dorsolateral prostate of NBL rats. The purpose of this study was to determine whether this carcinogenic effect is mediated by genotoxicity. Analogous to adducts produced by estrogens in the male hamster kidney, a target of estrogen carcinogenicity, induction of DNA adducts detectable by 32P-postlabeling was investigated in the prostate target tissue. NBL rats were treated with separate Silastic tubing implants containing testosterone and estradiol-17 beta. Control animals received empty implants. Animals were killed at 8, 16 and 24 weeks after initiation of treatment, and accessory sex glands were sampled for adduct analysis. DNA of the dorsolateral and ventral prostate and the coagulating gland (= anterior prostate) was isolated and analyzed by nuclease P1-enhancement of the 32P-post-labeling assay. DNA adducts were quantitated by Cerenkov counting. An adduct occurred selectively in DNA of the dorsolateral prostate of rats treated with estradiol plus testosterone for 16 or 24 weeks with relative adduct level values of approximately 10 x 10(9), but not in DNA of the ventral or anterior prostate. The adduct was not present in DNA of prostate tissue of rats treated for 8 weeks or in DNA of control tissues. This adduct was unique with respect to chromatographic location and has not been observed before in any tissue of control or hormone-treated animals. Neither the structure of the treatment-induced adduct nor the mechanism of its formation is known. However, the selective occurrence of this adduct in the tissue of origin of the carcinomas and its appearance coinciding with putative preneoplastic lesions and preceding carcinoma development suggests a causal relation between adduct formation and prostate cancer development in testosterone plus estradiol-17 beta-treated rats.

Dietary fat affects plasma prolactin in female F344 rats under conditions of ether stress.

The influence of amount and type of dietary fat on circulating concentrations of prolactin and estradiol-17 beta in female F344 rats from which blood was sampled by decapitation under ether anesthesia was compared with that in rats from which blood was collected without anesthesia. The animals were fed isonutrient (adjusted for differences in energy density) semipurified diets containing 5% or 20% (by weight) sunflower seed oil or lard. Blood was sampled by decapitation with or without standardized ether anesthesia during the afternoon of proestrus-estrus or the morning of metestrus-diestrus, as determined by examination of vaginal smears. Plasma hormone concentrations were measured by radioimmunoassay. Prolactin levels were lower during proestrus-estrus in rats fed a low-fat diet than in animals fed a high-fat diet, statistically independent of the type of dietary fat, but only when blood was sampled by decapitation under ether anesthesia [p = 0.0384, 2-way analysis of variance (ANOVA)]. No such difference was found in rats decapitated without anesthesia. This effect of amount of dietary fat on prolactin in proestrus-estrus animals anesthetized with ether was predominantly present in animals fed polyunsaturated fat (p < 0.05, 1-way ANOVA and Tukey's test) and was statistically not significant in rats fed saturated fat diets. During metestrus-diestrus, prolactin levels were significantly lower in animals fed a high-saturated fat diet than in those fed low-saturated fat, low-unsaturated fat, or high-unsaturated fat diets, independent of the blood sampling conditions (p < 0.05, 2-way ANOVA and Tukey's test). No consistent effects on estradiol-17 beta levels were found in type or amount of dietary fat or in presence or absence of ether anesthesia before decapitation. Growth, apparent digestibility of fat, and caloric intake were similar in all four dietary groups, but food consumption was higher and food conversion efficiency was lower in animals fed low-fat diets than in those fed high-fat diets. This study confirms the hypothesis that effects of dietary fat, particularly polyunsaturated fat, on circulating prolactin occur only during (ether) stress. Because stress is a frequent and normal phenomenon, this observation implies that the mammary glands of animals with a high dietary intake of polyunsaturated fat are frequently exposed to higher circulating prolactin concentrations than rats fed a low-fat diet, which may be a major mechanism by which dietary fat enhances rat mammary carcinogenesis.