HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Hilakivi-Clarke, L Articles by Pestell, R G Search for Related Content PubMed PubMed Citation Articles by Hilakivi-Clarke, L Articles by Pestell, R G

Lombardi Comprehensive Cancer Center, Department of Oncology, 3970 Reservoir Road NW, Box 571468, Washington DC 20057-1468, USA

(Requests for offprints should be addressed to R G Pestell; Email: pestell{at}georgetown.edu)

	Abstract

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
In the USA, breast cancer accounts for approximately 30% of all cancers diagnosed in women and is the second leading cause of cancer death in women. An understanding of the molecular genetic events governing breast cancer lead to both prevention and intervention strategies in an attempt to reduce mortality and morbidity from breast cancer. The last three decades of medical research examining the molecular pathogenesis of cancers have provided compelling evidence for the universal disruption of the cell cycle in human tumors. The importance of cell cycle control in human cancer was recognized by the recent award of the Nobel Prize to Drs Nurse and Hartwell for their discovery of the cyclins. More recent studies have demonstrated a critical interface between hormonal signaling and the cell cycle. In parallel, epidemiological studies have identified as being associated with breast cancer important dietary and environmental components that regulate hormonal signaling. This review describes the intersection of these two fields of study, which together imply a role for dietary prevention and intervention in human breast cancer perhaps through altering cell cycle components.

	Introduction

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Breast cancer accounts for approximately 30% of all cancers diagnosed in women in the USA. As the second leading cause of cancer death in women, breast cancer accounts for approximately 15% of all female cancer deaths (Zafonte et al. 2000). A comprehensive analysis of cancer mortality and morbidity over the last 25 years has demonstrated a reduction in breast cancer mortality in the 1990s. However, the benefit of reduction in death is greater in white women than in black women. The decline in mortality is likely due to improved screening and early detection, risk factor reduction, and perhaps improved treatment of early stage disease. Risk factors proposed for breast cancer include family history and lifestyle elements, such as diet and lack of exercise.

Over the last two decades, a compelling body of evidence has identified a disruption of cell cycle control mechanisms as a common pathway in human cancer. The importance of the cell cycle to human disease is evidenced by the recent award of the Nobel Prize to Drs Hartwell and Nurse for their discovery of the cyclins in the late 1990s. The dysregulation of cell cycle control in cancer is distilled into a model in which disruption of two parallel pathways is required for the occurrence of cancer (Fig. 1). These two parallel pathways function in such a manner that inactivation of one component of the pathway is sufficient for disruption of the pathway’s activity. In this regard, the p16^INK4a, retinoblastoma protein (pRb), cyclin D1, cdk4 pathway is frequently inactivated in human breast cancer due to overexpression of cyclin D1. In melanoma, loss of p16^INK4a is considered a key initiating event. Thus, distinct tumor types are thought to result from inactivation of distinct components of a common signaling pathway.

View larger version (13K): [in this window] [in a new window]   Figure 1 Model for dysregulation of cell cycle proteins (Weinberg 1995, Sherr & Roberts 1999). There are two parallel arms that include either the p16INK4a (INK4a) pathway or the p19ARF pathway, and inactivation of one arm can occur at any point in the respective pathway. Overexpression of cyclin D1 (D1), mutations in CDK4, or mutations in pRB (Rb), for example, will inactivate one arm of the signaling pathway and lead to unchecked cell cycle progression. Inactivation of both arms may therefore be one of the keys to tumorigenesis.

	Breast cancer and the mammalian cell cycle

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

The mammalian cell cycle has been divided into a series of sequential phases. The G₁, S, G₂, and M phases are sequentially transitioned in response to growth factor or oncogenic stimulation. The DNA synthetic (S phase) and mitotic (M phase) phases are preceded by gap phases (G₁, G₂). During the transition of the cell cycle, distinct checkpoints are inactivated. These checkpoints provide mechanisms by which the intracellular compartment senses a favorable growth factor environment and continually transduces this information to ensure genetic integrity during cellular replication. The presence of DNA damage and the integrity of mitotic spindles are assessed during transition between these phases of the cell cycle. Cell cycle progression is orchestrated by the relative activity of a family of serine threonine kinases (Fig. 2). The cyclins encode co-regulatory subunits of a holoenzyme that phosphorylates and inactivates a number of substrates including pRb. Cyclin D1 encodes the labile regulatory subunit of the cyclin cdk4/6 holoenzyme. Phosphorylation of the pRb protein is thought to change the confirmation of pRb which, in turn, alters further upon phosphorylation by cyclin E/cdk2 complexes. It is considered that the phosphorylation of the pRb protein by the cyclin D1/cdk4/6 holoenzymes is required for transition through the G₁/S phase of the cell cycle.

View larger version (24K): [in this window] [in a new window]   Figure 2 Schematic representation of the mammalian cell cycle. Competence factors such as platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF) promote entry into the early G1 phase. Sequential treatment with progression factors, insulin-like growth factor (IGF) or epidermal growth factor (EGF), promote progression through the G1 phase restriction point. Competence of the cyclin D1/Cdk4 complex is induced by mitogens. The cyclin D1/Cdk4 complex phosphorylates the pRB protein leading to sequential phosphorylation by cyclin E/Cdk2 and release of free E2F. The phosphorylation (p) of pRB and relief of transcriptional repression by pRB induces genes involved in the induction of S phase entry. Reprinted with permission from Pestell et al.1999; The Endocrine Society.

The cyclin D1 gene is shown to be overexpressed in 30–50% of human breast cancers. Overexpression of cyclin D1 correlates in several studies with poor prognosis. Indeed, the correlation of cyclin D1 with estrogen receptor (ER)--positive tumors is said to convert ER-positive tumors from good to poor prognosis (Kenny et al. 1999). The cyclin D1 gene is shown to physically associate with the ER and induce ER signaling in a cdk-independent manner (Zwijsen et al. 1997). Immuno-neutralizing experiments have shown that cyclin D1 is required for estrogen-induced cellular proliferation (Lukas et al. 1996). Cyclin D1 enhances ER signaling through functioning as a coactivator-like protein, analogous to the p160 steroid receptor coactivator (SRC) of the ER, in cultured cells.

Subsequent studies have shown the cyclin D1 gene regulates a number of transcription factors in a cdk-independent manner (Table 1). Two key publications have provided strong evidence for a biological role for two of these molecular targets. In this regard, cyclin D1 was shown to inhibit the activity of two gene products involved in fat metabolism, the CCAAT/enhancer binding protein (CEBP)-ß (Lamb et al. 2003) and peroxisome proliferator-activated receptor (PPAR)- proteins (Wang et al. 2003b). CEBPß and PPAR function in a common molecular pathway of adipocyte differentiation. CEBPß induces PPAR and PPAR in turn, co-ordinates the expression of a number of genes involved in adipocyte differentiation. In these recent studies, cyclin D1 was shown to inhibit the functional activity of either CEBPß (Lamb et al. 2003) or PPAR (Wang et al. 2003b).

The cyclin-dependent kinase inhibitors

The cyclin-dependent kinases are inhibited by two families of inhibitors. The cyclin inhibitor protein/kinase inhibitor protein (CIP/KIP) family includes the p21^CIF1/WAF1, the p27^KIP1, and the p57^KIP2 family. The INK4 inhibitor group includes p16^INK4a, p15^INK4b, p18^INK4c, and p19^INK4d. The INK4 proteins inhibit the catalytic domains of cdk4 and cdk6. Several additional functions of the p16^INK4a protein have been identified, including the regulation of RNA polymerase II (polII) carboxyl terminal kinase domain (CTD) (Nishiwaki et al. 2000). The RNA pol II CTD plays an important role in regulating expression of a number of genes suggesting that p16^INK4a may play a broad role in co-ordinating gene expression. In addition, p16^INK4a regulates cellular motility through an effect on cdk4 at the cell membrane (Fahraeus & Lane 1999). The CIP/KIP family of proteins inhibit the cyclin E/cdk2 kinase family. The role of the CIP/KIP proteins in cell cycle regulation appears to be dose dependent, inhibiting cyclin-dependent kinase activity in cell cycle progression at higher concentrations but serving as a chaperone or assembly protein enhancing cyclin-dependent kinase activities at lower concentrations (Sherr & Roberts 1999). The functions of the CIP/KIP family are, in turn, co-ordinated by additional proteins that regulate the subcellular distribution of the CIP/KIP proteins including jun-activation binding protein-1 (JAB1).

The p27^KIP1 protein was characterized as a protein homologous to the p21^CIP1 tumor suppressor. The abundance of p27^KIP1 is regulated primarily at the post-translational level through an SCF (composed of SKIP, CUL, and F box proteins) complex. The substrate specificity of the ubiqitin ligase or SCF is determined by the F box protein which binds the substrate. The F box protein that binds p27^KIP1 is called Skip2. The abundance of Skip2 may therefore contribute to the destruction of p27^KIP1. Skip2 is frequently overexpressed in tumor cell lines and collaborates with Ras in cellular transformation.

	Hormonal control of the cell cycle

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Steroid hormones induce cellular proliferation in the breast

A comprehensive analysis of cell cycle control proteins in this process has been reviewed (Pestell et al. 1999, Fu et al. 2002, Wang et al. 2003a). Estrogens are known to stimulate cell cycle progression, particularly in early G₁, and the induction of cellular proliferation correlates with the induction of cyclin D1 expression (Lippman & Bolan 1975, Lippman et al. 1976, Leung & Potter 1987, Altucci et al. 1996, Foster & Wimalasena 1996, Musgrove et al. 1996). Estrogen conveys several distinct effects on components of the cell cycle and breast epithelial cells. Cyclin D1 is required for estrogen-induced cellular proliferation in MCF-7 and several other cell types. Two related proteins, ER and ERß, function as transcription factors to regulate expression of target genes and modulate the effects of estrogens. Estrogens induce the expression of cyclin D1 through ER. Conversely, estrogens inhibit cyclin D1 expression through ERß (Liu et al. 2002). ERß completely inhibits cyclin D1 gene activation by estrogen and ER. The opposing action and dominance of the ERß over ER in regulating cyclin D1 gene expression implies a potential role for ERß as a modulator of the proliferative effects of estrogens (Liu et al. 2002).

ER function determines normal mammary gland duct growth, angiogenesis, and somatogenesis (Davis et al. 1994, Eddy et al. 1996, Johns et al. 1996, Korach et al. 1996), together with other hormones and growth factors. Estrogen also functions through membrane tyrosine kinase receptor signaling pathways (Migliaccio et al. 1996). Furthermore, peptide growth factors induce ER activity independently of ligand. Thus, the ER can be activated through phosphorylation of serine 118 or 167. The Ras/ Raf/MAP kinase pathway enhances the recruitment of the ER coactivators, including amplified in breast cancer-1 (AIB1). As peptide growth factors induce both ER activity and cyclin D1 abundance, it has been anticipated that ER and cyclin D1 may be regulated co-ordinately.

Cyclin D1 expression is reduced by estrogen antagonists in T47D human breast cancer cells (Fu et al. 2002). Cellular proliferation induced by estrogens in several breast cancer cell lines correlates with induction of cyclin D1 mRNA, protein, and cyclin D1-dependent kinase activity. Cyclin D1 enhances activity of a chimaeric ER, in which the ligand binding domain of the ER was replaced by the Gal4 DNA binding domain (Neuman et al. 1997, Zwijsen et al. 1997). It has been proposed that the overexpression of cyclin D1, frequently seen in ER-positive breast tumors, may function to promote ER activation of target genes in the presence of low estrogen concentrations. Analysis of murine cyclin D1^{–/ –} cells has suggested that cyclin D1 may perform a chaperone function bringing together multi-molecular complexes that in turn regulate the activity of the ER (R G Pestell & C Wang, unpublished observations).

The cyclin-dependent kinase inhibitors also play an important role in estrogen-induced mitogenesis. p16^INK4a inhibits estrogen-induced DNA synthesis in MCF-7 cells (Lukas et al. 1996, D’Amico et al. 2003), in part through repression of cyclin D1 expression (D’Amico et al. 2003). Estrogens also alter subcellular localization of the cyclin-dependent kinase inhibitors. The relative distribution of p21^CIP within cyclin E/cdk2 complexes is regulated by estradiol (Planas-Silva & Weinberg 1997). Estrogens can also reduce the amount of p21^CIP and p27^KIP bound to cyclin E/cdk2 (Prall et al. 1997). Estrogen induces both cyclin E/cdk4 and cyclin E/cdk2 activity. As Myc is induced by estrogens and Myc is sufficient for the induction of cyclin E/cdk2 activity, c-Myc may contribute to the enhanced DNA synthesis observed with estrogen treatment. As Myc and cyclin D1 collaborate in several models of oncogenesis, the proliferative activity of estrogens in human breast cancer may be in part regulated by both cyclin D1 and c-Myc.

Nuclear receptors in breast cancer

Estrogen receptor
The important role for ER activity in breast cancer onset and progression is underscored by the efficacy of ER antagonists as adjuncts for breast cancer prevention in high risk individuals and the efficacy of adjuvant therapy for treatment of patients with ER-positive breast cancer (Jordan 1998). The ER status thus forms part of the stathmin (TMN) staging classification category 1 (Fitzgibbons et al. 1998). Estrogens influence normal proliferation, differentiation, and physiology of breast tissue and the development and progression of breast cancer (Hoskins & Weber 1994, Korach 1994, Eisen & Weber 1998, Gustafsson 1998, Jordan 1998). The presence of ER immuno-reactivity serves as an important prognostic indicator of high survival rates and lower relapse risk (Jordan & Morrow 1999, Brodie 2003). Curiously, approximately 50% of ER-positive tumors fail to respond to anti-estrogen therapy, suggesting that important additional components play a role in ER antagonist therapies.

Androgen receptor
Androgens induce the expression of endogenous estrogen-responsive target genes and are capable of activating an ER reporter gene in MCF-7 cells. Androgens induce proliferation of ER-positive cells (Poulin & Labrie 1986, Najid & Habrioux 1990, Boccuzzi et al. 1992, Birrell et al. 1995). A subset of cell lines is exquisitely sensitive to the proliferative effects of androgens, including the Shianogi line (Stanley et al. 1977). The mechanisms by which androgens induce breast cellular proliferation may include induction of ER. Thus, the androgens dehydroepian-drosterone, 5-androstene-3ß,17ß-diol, testosterone, and dihydrotestosterone activate ER reporter activity and induce breast cancer cell proliferation (Maggiolini et al. 1999). The effectiveness of aromatase inhibitors in blocking the growth of ER-positive tumors may be in part due to their ability to block androgen induction of ER activity.

PPAR
The PPARs are ligand-activated nuclear receptors including PPAR, PPAR, and PPAR. Their modular structure resembles other nuclear hormone receptors with an N-terminal activation and DNA binding domain and a C-terminal ligand binding domain (Rosen & Spiegelman 2001b). PPAR was cloned as a transcription factor promoting fat cell differentiation. The PPAR ligands include eicosanoids such as 15-deoxy- 12,14-prostaglandin J2 (15d-PGJ₂), and synthetic ligands including the thiazolidinedione (TZD) class. It has been estimated that approximately 2 million Americans take anti-diabetic agents which are PPAR ligands. There is some evidence that breast cancer risk is reduced in individuals taking PPAR ligands (Michels et al. 2003). These epidemiological data suggest that the mechanisms by which PPAR ligands may function are of broad importance in cancer.

PPAR is expressed in breast, prostate, and colonic epithelium. Several lines of evidence have implicated PPAR as a candidate tumor suppressor (Rosen & Spiegelman 2001a, Koeffler 2003). Heterozygous mutations of PPAR were identified in four of 55 patients with colon cancer (Sarraf et al. 1999). A translocation between paired box homeotic gene 8 (PAX-8) and PPAR was also identified in follicular thyroid cancer which appeared to serve as a dominant inhibitor of endogenous PPAR expression (Kroll et al. 2000). Addition of PPAR ligands to cultured tumor cell lines derived from breast, prostate, or colonic cancer inhibits cellular proliferation (Brockman et al. 1998, Elstner et al. 1998a, Mueller et al. 1998, Ricote et al. 1998). The role of PPAR ligands in tumor onset in vivo is unresolved. Studies of murine familial adenoma models of adenomatosis polyposis coli using Apc^min mice demonstrated that PPAR ligands may either promote (Saez et al. 1998) or inhibit colon polyp formation (Niho et al. 2003). As the same murine model of tumor formation was used in both studies, the explanation for these disparate findings warrants further analysis.

The relative importance of PPAR signaling in breast cancer onset and progression is an area of substantial current interest. PPAR expression has been observed in a subset of human breast cancers and benign breast disease (Wang et al. 2003b). In breast cancer cell lines, PPAR activation by TZDs or 15d-PGJ₂ induced cellular differentiation (Elstner et al. 1998a, Mueller et al. 1998). Both TZDs and 15d-PGJ₂ inhibit cell cycle progression through direct repression of cyclin D1 gene expression (Wang et al. 2001). The relative importance of PPAR agonists as chemopreventive agents in humans remains to be explored. The presence of ß-catenin activation in the subset of breast cancers (Lin et al. 2000) and the failure of PPAR ligands to impact tumor progression in the presence of mutant ß-catenin in some studies on colon cancer (Saez et al. 1998) suggest that PPAR ligands may be important adjuncts to prevent rather than treat human breast cancer.

	Diet and the cell cycle

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Diet has been suggested to contribute to the etiology of 30–50% of newly diagnosed breast cancers (Willett 2001). Animal studies strongly support the ability of dietary components either to increase or reduce breast cancer risk. For example, n-6 polyunsaturated fatty acids (PUFAs) present in vegetable oils pre-initiate (i.e. increase susceptibility to carcinogens and other cancer-initiating factors) and promote carcinogen-induced mammary tumorigenesis (Welsch 1992, Hilakivi-Clarke et al. 1997). In contrast, n-3 PUFAs present in fish, flaxseed and canola oil reduce the growth of ER-negative human breast cancer cells implanted in nude mice (Rose 1997). Other dietary components that have been suggested to reduce the risk of breast cancer include the phytoestrogen genistein (Bouker & Hilakivi-Clarke 2000), vitamins (such as A, D, and E), and other components in fruits and vegetables (Smith-Warner et al. 2001, Riboli & Norat 2003). Restricted energy intake has also been shown to reduce mammary tumorigenesis in animal models (Zhu et al. 2002). Despite the compelling evidence from experimental models, studies in human populations that have attempted to link a particular dietary factor to increased or reduced breast cancer risk have generated contradictory data (Willett 2001).

Most human studies have assessed dietary intakes at the time of breast cancer diagnosis or shortly before. As breast cancer is a multi-step process, and there is likely to be a considerable time-lapse between initiation and diagnosis of this disease, studies which assess diet only at or around the time of diagnosis are limited and may miss or wrongly identify dietary components involved in increasing breast cancer risk. For instance, dietary components that affect tumor cell proliferation (e.g. alcohol, phytoestrogens, some fats) might be identified as affecting cancer risk. Those that affect susceptibility to malignant transformation (e.g. anti-oxidant vitamins) may not be different between the cases and controls at the time of diagnosis, although when breast cancer was initiated the cases theoretically could have been consuming fewer vitamins than the controls. It is not uncommon that, as an adult, individuals adopt a healthier diet that includes more fruits and vegetables compared with childhood dietary preferences. Interestingly, the evidence in humans linking diet to breast cancer is strongest for alcohol and perhaps phytoestrogens and fats, but weak for vitamins.

Dietary factors have several different mechanisms of action, depending on dose or timing of exposure, that could ultimately affect breast cancer risk. The phytoestro-gen genistein, present in soy products, is an example. Genistein is a weak estrogen that binds to both ER and ERß, and at physiological concentrations (<1 µM) causes cell proliferation (Wang et al. 1996, Zava & Duwe 1997). This genistein-induced increase in proliferation has been observed in vitro in human breast cancer cells, in vivo in animals, and also in normal human breast epithelium (Bouker & Hilakivi-Clarke 2000). Conversely, pharmacological doses (>10 µM) of genistein cause changes that are potentially protective, including inhibition of tyrosine kinase activity, angiogenesis, and induction of apoptosis (Kim et al. 1998, Messina 1999). It is unlikely that soy consumption produces genistein exposures at pharmacological levels, since Asians who, on average, consume high levels of soy compared with Caucasians have blood concentrations around 1–4 µM (Xu et al. 1994). It is not clear whether the concentrations of genistein in the breast tissue reflect those in the circulation, or are lower or higher than in the blood.

Timing of exposure

Timing of hormonal and dietary exposures has important ramifications for breast cancer risk. There are three periods during a female’s life-time when her breast epithelium undergoes extensive growth and is highly sensitive to estrogens: fetal life, puberty, and pregnancy. In rodents, maternal exposure during pregnancy to estrogenic compounds, including dietary compounds, increases breast cancer risk among female offspring (Walker 1984, 1990, Hilakivi-Clarke et al. 1997, 1999). Thus, some breast cancers may be pre-initiated during fetal life by an exposure to high levels of estrogens that may imprint the mammary gland in a manner that increases later susceptibility to breast cancer (Trichopoulos 1990, Hilakivi-Clarke et al. 2002a). Persistent changes in the mammary glands of rodents exposed to a high estrogenic environment in utero include increased number of targets for malignant transformation (Hilakivi-Clarke et al. 1997) and altered expression of ER and tumor suppressor genes BRCA1 and p53 (Yu et al., unpublished data).

Prepubertal estrogenic exposures paradoxically reduce breast cancer risk (Cabanes et al. 2004), perhaps through differentiation of the mammary gland structures (terminal end buds; TEBs) that are known to be the sites for malignant transformation. The gland of an animal exposed to estradiol or genistein during prepuberty has been found to contain fewer TEBs and more lobulo-alveolar units (Cabanes et al. 2004). These changes are accompanied by long-lasting alterations in the expression of ER and ERß (Cabanes et al. 2004).

Prepubertal exposure to estradiol or genistein leads to a long-lasting upregulation of BRCA1 mRNA in the rat mammary gland (Cabanes et al. 2004), suggesting an increase in DNA repair capacity. The tumor suppressor BRCA1 interacts with estrogens and the ER, at least in normal mouse mammary gland and in human breast cancer cell lines (Gudas et al. 1995, Marquis et al. 1995, Spillman & Bowcock 1996, Fan et al. 1999, 2000, 2001). Loss of wildtype BRCA1 is linked to inherited breast cancers (Garland et al. 1998, Satagopan et al. 2001), probably because this gene participates in DNA damage repair and recombination processes related to maintenance of genomic integrity, control of cell proliferation, and regulation of gene transcription (Rosen et al. 2001, Welcsh et al. 2002). BRCA1 regulates the expression of several genes implicated in breast cancer, including cyclin D1, c-myc, and components of the Jak-Stat pathway (Welcsh et al. 2002). In addition, BRCA1 inhibits the signaling of the ligand-activated ER (Fan et al. 1999).

Pregnancy is closely linked to changes in breast cancer risk with early first pregnancy reducing and later first pregnancy increasing the risk (MacMahon et al. 1970). Higher estrogen levels during pregnancy increase the mother’s breast cancer risk (Richardson et al. 1998, Peck et al. 2002). The proposed mechanism is that the breast tissue of older first time mothers are more likely to have acquired malignant cells that are stimulated by high pregnancy hormonal environment. Collectively, these findings indicate that although estrogens stimulate the proliferation of ER-positive human breast cancer cells, their effect on the normal breast depends on the timing of exposure.

	Dietary components that modify the cell cycle

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Several dietary factors have been demonstrated to alter the normal cell cycle of non-malignant and malignant breast cells. Herein, we address the role of body mass index (BMI), energy restriction, dietary fat, soy, and vitamin A in affecting breast cancer risk and cell cycle-related end points.

Body weight

Excess weight and obesity are major problems in the Western world because of their link to several serious health problems, including cardiovascular diseases, diabetes, and cancer. According to the 1999 National Health and Nutrition Examination Survey, it was estimated that 61% of Americans are overweight (BMI 25–30 kg/m²) and 25% are obese (BMI>30 kg/m²) (Flegal et al. 2002). BMI affects breast cancer risk, but the effects depend on the developmental stage of an individual. Further, multiple pathways have been identified that could mediate the effects of obesity on breast cancer risk.

Adrenal androgens are aromatized to estrogens in adipose tissue. Most circulating estrogens present in postmenopausal women originate from adipose cells, and approximately one-third of estrogens in premeno-pausal women come from these cells. A high fat and/or a high total caloric intake increases the levels of circulating free estrogens (Goldin et al. 1982, Rose et al. 1987, 1993, Bennett & Ingram 1990) and Wu et al. (1999) have reported that women who reduce their total fat consumption lower their circulating estradiol levels. In postmenopausal women, BMI correlates with elevated expression of cyclin D1 and bcl-2 mRNA in the mammary tissue (Suga et al. 2001), indicating that high BMI may increase breast cancer risk by increasing serum estrogens, modulating cell cycle, and inhibiting apoptosis.

Besides estrogens, other hormones and growth factors that are altered by BMI include leptin, adiponectin, insulin, and IGFs. Further, PPAR has a key role in adipocyte differentiation and in mediating high fat-induced obesity (Kadowaki et al. 2003). Leptin is expressed by the adipose cells as well as epithelial cells: the levels of this hormone are closely linked to BMI, regardless of a woman’s age or reproductive status (Huang & Li 2000). Leptin has been linked to increased breast cancer risk (Marttunen et al. 2000, Tessitore et al. 2000, Ozet et al. 2001, Hu et al. 2002). Leptin also alters the rate of cell proliferation, and modifies the expression of genes linked to the cell cycle, such as cyclin D1, PPAR, and ER and ERß (Okumura et al. 2002). Adiponectin is secreted exclusively by adipose tissue and it is present at high levels in the serum. However, it is reversely associated with BMI (Arita et al. 1999, Stefan et al. 2002). Reduced adiponectin levels have recently been linked to increased breast cancer risk (Miyoshi et al. 2003), and thus this hormone might act as a tumor suppressor. Extensive literature links IGF-I to breast cancer. In addition to being linked to obesity, all these hormones and growth factors interact with each other.

Birth weight
Body weight at birth may alter later risk of developing breast cancer. Women who had a high birth weight have an elevated breast cancer risk in most (Michels et al. 1996, Sanderson et al. 1996) (but not all studies, Potischman & Troisi 1999), particularly for premenopausal breast cancer (Michels et al. 1996, Sanderson et al. 1996). The increase in breast cancer risk by high birth weight is profound in twins (Sparano & Winer 2001), suggesting genetic modifiers of high birth weight on breast cancer risk.

Birth weight has been linked to an elevated estrogenic environment during pregnancy (Gerhard et al. 1987). In human populations, size at birth correlates with increased cellular proliferation, as mammary epithelial area/and mammographic density are increased (Ekbom et al. 1995). In animal studies (Hilakivi-Clarke et al. 1997), the results indicate that maternal exposures to a high fat diet or estradiol during pregnancy increased the density of the offspring’s mammary epithelium. Thus, one mechanism by which high birth weight increases later breast cancer risk could be through an increase in epithelial cell proliferation.

Prepubertal BMI
High body mass during childhood is associated with a reduction in breast cancer risk (Le Marchand et al. 1988, Magnusson et al. 1998, Berkey et al. 1999, Hilakivi-Clarke et al. 2001, Swerdlow et al. 2002). This is puzzling in light of the fact that overweight and obese girls reach puberty earlier than lean girls, and early puberty is associated with increased breast cancer risk. A monozygotic twin that reached puberty first has a fivefold higher breast cancer risk than her twin sister with later puberty onset (Hamilton & Mack 2003), suggesting environmental modifiers affecting puberty onset and breast cancer risk. Since in animal studies in utero exposures to estrogenic compounds accelerate puberty onset and increase susceptibility to carcinogen-induced mammary tumors (Hilakivi-Clarke et al. 1997), early puberty may be a marker of high in utero estrogen exposure rather than directly increasing breast cancer risk.

High body mass during childhood may be linked to increased levels of circulating estrogens. Girls with elevated low-density lipoprotein cholesterol levels exhibited a reduction in serum estradiol levels after their dietary fat intake was reduced and fiber intake increased (Dorgan et al. 2003). In heifers, elevated energy consumption during the prepubertal period reduced the growth of mammary parenchyma (Hull & Harvey 2002) and increased breast adipose tissue mass, a predictor of reduced breast cancer risk (Oza & Boyd 1993, Boyd et al. 2001). Similarly, women who had high BMI at puberty have a persistently lower mammographic density (McCormack et al. 2003).

BMI during reproductive years and postmenopause
Postmenopausal breast cancer risk is elevated in individuals who were obese either during the pre- or post-menopausal years, or both (Kabuto et al. 2000). Paradoxically, obesity reduces premenopausal breast cancer risk (Potischman et al. 1996, Cleary & Maihle 1997, Huang et al. 1997, Trentham-Dietz et al. 1997). We have studied excessive weight gain during pregnancy and found that it increases postmenopausal breast cancer, and the effect is independent from the body weight at the time of diagnosis (Kinnunen et al. 2004).

High BMI during the postmenopausal years correlates with increased serum estrogens (Toniolo et al. 1995, Hankinson et al. 1998), although this may not be true for premenopausal women (Trichopoulos et al. 1983). In premenopausal women, estrogens derived from an excessive amount of adipose tissue are likely to inhibit the pituitary–gonadal axis, potentially causing a reduction in estrogens to be released from the ovaries. However, the total estrogen levels are not altered, because the loss of ovarian estrogen production is compensated for by adipose-derived estrogens.

Energy restriction

Dietary caloric restriction in monkeys and rodents extends lifespan, slows the aging process (Roth et al. 2000), and reduces mammary tumor risk in rodents (Thompson et al. 1999, Jiang et al. 2003). Energy restriction reduces oxidative tissue damage and mitochondrial free radical generation in rodents (Merry 2002). Energy restriction reduces mammary tumor cell proliferation via G1 cell cycle arrest, possibly through a reduced expression of cyclin D1, Cdk4 and increased expression of p21 and p27 (Jiang et al. 2003) perhaps due to increased adrenocortical steroid secretion (Zhu et al. 2003).

It is not known whether the health benefits of energy restriction extend to humans. In Norway (Robsahm & Tretli 2002) and The Netherlands (Dirx et al. 1999), reduction in energy intake in adolescents and women during World War II (1940–44) and the Hunger Winter (1944–45), increased breast cancer risk later in life (Robsahm & Tretli 2002). Consistent with recent data are the findings that low childhood BMI increased breast cancer risk (Le Marchand et al. 1988, Magnusson et al. 1998, Luo & Miller 2000, Hilakivi-Clarke et al. 2001, Swerdlow et al. 2002).

No studies have directly addressed the role of energy restriction during adult life in affecting breast cancer risk. However, women with anorexia nervosa are at a 20% reduced risk of developing a cancer in general (Mellemkjaer et al. 2001). Studies that have assessed longevity in thin individuals have generated conflicting findings (Lee et al. 2001), possibly because many severe illnesses cause weight loss.

Polyunsaturated fatty acids

High dietary fat intake has been linked to the promotion of breast cancer in animal models (Freedman et al. 1990, Welsch 1992), most case–control studies (Howe et al. 1990, Richardson et al. 1990, Van’t Veer et al. 1990), but not in prospective cohort studies (Welsch 1987, Willett et al. 1992, van den Brandt et al. 1993, Willett & Hunter 1994), with few exceptions (Cho et al. 2003a).

Diets are composed of several types of dietary fats, including mono-unsaturated fatty acids (olive and canola oils), PUFAs (vegetable oils, n-6 PUFA and fish, n-3 PUFA), and saturated fats (dairy products and meat). Saturated fats may be associated with increased breast cancer risk in women (Willett 1997, Cho et al. 2003a), but animal studies have generally been inconclusive (Rose 1997). Diets high in n-6 PUFAs are associated with elevated breast cancer risk in animal studies (Freedman et al. 1990, Welsch 1992), but not humans (Willett 2001, Cho et al. 2003a). n-3 PUFAs are protective in some (Eid & Berry 1988, Caygill et al. 1996, Klein et al. 2001), but not all studies (Holmes et al. 2003, Stripp et al. 2003). In nude mice, dietary n-3 PUFA reduces growth and metastasis of human breast cancer xenografts (Rose et al. 1995) and spontaneous or carcinogen-induced rodent mammary tumors (Karmali et al. 1984, Hirose et al. 1990, Fay et al. 1997).

A diet high in n-6 PUFA corn oil, or a low or high fat n-3 PUFA menhaden oil increases serum estradiol levels in rats and mice when compared with a low fat n-6 PUFA standard AIN93 laboratory diet (Hilakivi-Clarke et al. 1996a,b, 1997, 2002b). Consumption of a high fat diet may increase the levels of adipose tissue, and thereby increase aromatization of estrogens. n-6 PUFA consumption may also directly enhance aromatization (Richards & Brueggemeier 2003). It is not clear why n-3 PUFA diet increases serum estrogen levels. PPAR is either activated (Rubin et al. 2000) or inhibited by n-3 PUFAs (Thoennes et al. 2000), potentially reducing aromatase activity. However, PPAR agonists can increase arachidonic acid release from the liver (Levine 2001), which could stimulate aromatization.

n-3 and n-6 fatty acids impact cell proliferation and differentiation; however, some studies report inconsistent data (Finstad et al. 1994, Rudolph et al. 2001). The effects of these fatty acids on the cell cycle suggest that n-6 PUFAs increase cyclin D1 mRNA in T47D breast cancer cells (Razanamahefa et al. 2000), the n-3 PUFA docosahexaenoic acid reduces cyclin D1-, E-, and A-associated kinase activity in HT-29 colon cancer cells (Chen & Istfan 2001), and the n-3 PUFA, eicosapentae-noic acid, reduces cyclin D1 and cyclin E in NIH 3T3 cells (Palakurthi et al. 2000).

Prepubertal exposure to PUFAs and breast cancer risk
Maternal exposure to a diet containing high levels of n-3 PUFAs reduces an offspring’s risk of mammary tumor-igenesis (Hilakivi-Clarke et al. 2002b). In contrast, prepubertal exposure to a high fat n-3 PUFA diet increases carcinogen-induced mammary tumorigenesis, while a prepubertal low fat n-3 PUFA exposure provides a protection against breast cancer (Olivo et al., unpublished data).

Identifying the pathways and timing of exposure by which in utero or prepubertal exposures to n-3 PUFAs may decrease breast cancer risk is complicated, because multiple potential mechanisms exist (Welsch 1987). We have found that changes induced by having been exposed to a low fat n-3 PUFA diet during prepuberty include an increase in lipid peroxidation, but also an increase in the expression of oxidative damage repair genes, apoptosis and cyclo-oxygenase 2. Animals fed a high fat n-3 PUFA diet prepubertally which increases their breast cancer risk also exhibit increased lipid peroxidation, but this is accompanied by an increase in DNA damage, cell proliferation, cyclin D1 expression, and phosphorylated Akt, and reduced apoptosis and BRCA1 expression (Olivo et al. unpublished data). Thus, high levels of n-3 PUFAs fed in a high fat context before the onset of puberty can cause multiple changes in the mammary gland, including disruption of the cell cycle. In contrast, a prepubertal exposure to a low fat n-3 PUFA diet may be protective because of increased oxidative damage repair and apoptosis.

Genistein

Soybeans contain large amounts of the isoflavones daidzein and genistein (Barnes et al. 1994, Adlercreutz 1995). Genistein or equol, a metabolite of daidzein, may play significant roles in the bioactivity of soy (Lampe 2003). It has been proposed that high soy intake contributes to low breast cancer incidence among Asian women (Adlercruetz et al. 1996), and our recent meta-analysis showed that a high soy intake reduces risk of developing premenopausal breast cancer (Trock et al. 2001).

Timing of exposure to soy plays a key role in determining its effect on the breast with soy exposure during childhood significantly reducing breast cancer risk (Shaag et al. 2001, Wu et al. 2002). Rats exposed to genistein during prepuberty have a reduced risk of developing carcinogen-induced mammary tumors (Murrill et al. 1996, Hilakivi-Clarke et al. 1998, Cotroneo et al. 2002), yet genistein either via injections or dietary soy during adulthood does not affect mammary tumorigenesis (Constantinou et al. 1996, Cohen et al. 2000). In ovariectomized animals that mimic postmenopause, genistein promotes the growth of human MCF-7 breast cancer cells (Allred et al. 2001, Ju et al. 2001).

In vitro, genistein has mitogenic effects at low, physiological doses (0.01–1 µM) and anti-proliferative effects at higher, pharmacological doses (>10 µM) (Wang et al. 1996, Hsieh et al. 1998). However, when estradiol is present, genistein may not induce cell proliferation (Trock et al. 2001). It is not clear whether genistein inhibits estrogens actions in humans, since exposure to soy induces breast proliferation in premenopausal, but not postmenopausal women (Petrakis et al. 1996, McMichael-Phillips et al. 1998).

Animal studies have investigated whether an exposure to genistein either in utero or during prepuberty alters the expression of nuclear receptors in the mammary gland or other tissues. At the time of genistein administration, ER expression is reduced (Cotroneo et al. 2002), a result that is in accordance with findings obtained in vitro in cells treated with estradiol (Saceda et al. 1988). Both fetal and prepubertal genistein administration causes an increase in ER expression in estrogen-responsive tissues (Jefferson et al. 2002), and prepubertal genistein exposure also up-regulates ERß in the mammary gland (Y Zhu, R Clarke, S de Assis, J S Miller & L Hilakivi-Clarke, unpublished data). It is plausible that genistein’s ability to bind preferentially to ERß (Kuiper et al. 1997) might be more important than its effect on ER expression in affecting breast cancer risk.

Genistein can directly bind to PPAR to either inhibit it at physiological doses (<1 µM) or stimulate at pharmacological (>5 µM) doses in mesenchymal progenitor cells (Dang et al. 2003), in adipose cells (Dang et al. 2003), and in murine macrophages (Mezei et al. 2003).

Physiological doses of genistein and another phytoestrogen, zearalenone, increased the expression of cyclin D1 and Cdk2 in MCF-7 cells in vitro (Dees et al. 1997) and in vivo (Ju et al. 2002), but higher doses (100–200 µM) inhibited cyclin D1 and cyclin E (Agarwal 2000). The effects of genistein on cyclin D1 are consistent with genistein’s reported ability to both stimulate and inhibit cell proliferation, depending upon the dose.

Vitamin A and retinoids

Since high dietary intake of fruits and vegetables is consistently linked to reduced breast cancer risk (Smith-Warner et al. 2001, Riboli & Norat 2003), it seems reasonable that the vitamins present at high levels in these foods contribute to this effect. These vitamins, including A, C, and E, have several potentially beneficial effects on biological pathways related to preventing oxidative damage, inhibiting cell proliferation, and inducing differentiation. Surprisingly, the evidence linking each of these vitamins to reduced breast cancer risk is weak.

Retinoid receptor function
Retinoid receptors are members of the nuclear steroid/ thyroid hormone receptor family that play a key role in the control of cell growth and differentiation (reviewed in Mangelsdorf et al. 1993, Fontana & Rishi 2002, Fu et al. 2002). Two distinct classes of retinoid receptors exist: the retinoic acid receptors (RARs , ß, ) and the retinoid X receptors (RXRs , ß, ). Like other nuclear receptors, these molecules homo- and heterodimerize in response to ligand binding and function as transcription factors at specific promoter elements or indirectly via other transcriptional regulators. In addition, RXRs can hetero-dimerize with vitamin D and thyroid hormone receptors as well as PPAR (Dawson et al. 2000).

Activation of retinoid receptors has been shown to be growth inhibitory in breast cancer cells as well as other cell types (see Amos & Lotan 1990, Seewaldt et al. 1997, 1999), and consequently retinoids have been explored as potential therapeutic and chemopreventive agents. Retinoid receptor activation inhibits cell cycle progression, delaying the transition between G₀/G₁ and S phases (Seewaldt et al. 1997, 1999). RARs and RXRs interact with many different downstream effectors, resulting in cell type-specific mechanisms of growth inhibition and apoptosis. In breast cancer cells, the transcriptional regulator CBP/p300 is up-regulated by trans retinoic acid, functions as a coactivator for RARs, and is required for sensitivity to retinoid-mediated growth arrest (Dietze et al. 2003). Other transcriptional regulators implicated in retinoid responses include the Drosophila hairy and enhancer of split homologue transcriptional repressor and SOX-9 (Muller et al. 2002). The anti-apoptotic factor Bcl-2 is down-regulated by retinoic acid in MCF-7 cells, leading to enhanced caspase activation (Pratt et al. 2003). IGF signaling also appears to be altered in response to retinoids. Retinoic acid induces down-regulation of insulin receptor substrate-1 and Akt, resulting in growth arrest and apoptosis (del Rincon et al. 2003).

Heterodimerization of RXR and PPAR may contribute to retinoid anti-proliferative signaling in breast cancer. Simultaneous activation of both receptors inhibits breast cancer cell proliferation in vitro and in vivo (Mukherjee et al. 1997, Elstner et al. 1998b). Inhibition of cyclin D1 expression by both RXR and PPAR stimulation has been demonstrated in a pancreatic cancer cell line (Suh et al. 1999). RXR/PPAR induce apoptosis by inhibiting nuclear factor-kB-dependent survival pathways (Dubuquoy et al. 2002) and repression of the P450 aromatase gene, thereby decreasing estrogen synthesis in some (Rubin et al. 2002) but not all studies (Mu et al. 2001).

Dietary intake of vitamin A and breast cancer

Most studies, however, have failed to identify a clear association between breast cancer and dietary vitamin A or carotenoid intake (Kushi et al. 1996, Jarvinen et al. 1997, Verhoeven et al. 1997, Michels et al. 2001) (reviewed in Fairfield & Fletcher 2002). Recent studies suggest that increased intake of vitamin A or carotenoids may prevent breast cancer (Bohlke et al. 1999, Zhang et al. 1999, Toniolo et al. 2001), particularly in premenopausal but not postmenopausal women, women who consume alcohol or women who are at high inherited risk for this disease. Vitamin A may be protective for premenopausal women who smoke, but not in non-smokers (Cho et al. 2003b).

In rats, vitamin A-deficient and -supplemented rats develop more carcinogen-induced mammary tumors than animals fed vitamin A-adequate diets (Metz et al. 2002).

Confounding epidemiological variables include the failure of serum levels of retinoids and carotenoids to correlate with consumption (reviewed in Cheung et al. 2003). There are some 600 carotenoid plant pigments, 50 of which may be converted into vitamin A; whether it is strictly vitamin A or other carotenoid metabolites that function to reduce breast cancer risk has not been determined. Similarly, women who consume high levels of carotenoids and vitamin A are also likely to obtain additional nutrients from other fruits and vegetables, and it may be that these contribute significantly to any observed reduction in breast cancer risk (Fairfield & Fletcher 2002).

	Conclusions

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Several dietary factors modify the cell cycle. Dietary factors that up-regulate cyclin D1 and ER and down-regulate PPAR and BRCA1 might play a role in increasing breast cancer risk (Fig. 3). Dietary factors that have opposing effects on these genes could reduce breast cancer risk. In utero exposures to genistein and perhaps n-6 PUFA mediate their effects on the breast partly by affecting these genes and increase breast cancer susceptibility (Ju et al. 2002, Dang et al. 2003). Prepubertal exposure to genistein or a low fat n-3 PUFA diet up-regulates BRCA1 (Cabanes et al. 2004, S Olivo, L Hilakivi-Clarke, Y Zhu, R G Lee, A Cabanes, G Khan, A Zwart, Y Wang & R Clarke, unpublished data) and other genes that lead to increased ability to repair DNA damage, or cause apoptosis and differentiation. Further studies will determine whether the timing of exposure to vitamins, such as vitamin A and E which induce differentiation and have anti-oxidant properties, modifies later breast cancer risk, and whether changes in susceptibility to malignant transformation are related to diet-induced changes in the cell cycle.

View larger version (19K): [in this window] [in a new window]   Figure 3 Examples of dietary factors that increase breast cancer risk, perhaps by up-regulating cyclin D1 and down-regulating PPAR, and dietary factors that reduce breast cancer risk by having the opposite effects on the expression of these two genes.

	References

Top Abstract Introduction Breast cancer and the... Hormonal control of the... Diet and the cell... Dietary components that modify... Conclusions References

 
Adlercreutz H 1995 Phytoestrogens: epidemiology and a possible role in cancer protection. Environmental Health Perspectives 103 103–112.

Adlercruetz CH, Goldin BR, Gorbach SL, Hockerstedt KAV, Watanabe S, Hamalainen E, Markkanen MH, Makela TH, Wahala KT, Hase TA & Fotsis T 1996 Soybean phytoestrogen intake and cancer risk. Journal of Nutrition 125 757S–770S.

Agarwal R 2000 Cell signaling and regulators of cell cycle as molecular targets for prostate cancer prevention by dietary agents. Biochemical Pharmacology 60 1051–1059.[CrossRef][ISI][Medline]

Allred CD, Allred KF, Ju YH, Virant SM & Helferich WG 2001 Soy diets containing varying amounts of genistein stimulate growth of estrogen-dependent (MCF-7) tumors in a dosedependent manner. Cancer Research 61 5045–5050.[Abstract/Free Full Text]

Altucci L, Addeo R, Cicatiello L, Dauvois S, Parker MG, Truss M, Beato M, Sica V, Bresciani F & Weisz A 1996 17beta- Estradiol induces cyclin D1 gene transcription, p36D1-p34cdk4 complex activation and p105Rb phosphorylation during mitogenic stimulation of G(1)-arrested human breast cancer cells. Oncogene 12 2315–2324.[ISI][Medline]

Amos B & Lotan R 1990 Retinoid-sensitive cells and cell lines. Methods in Enzymology 190 217–225.[Medline]

Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsubara K, Muraguchi M, Ohmoto Y, Funahashi T & Matsuzawa Y 1999 Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochemical and Biophysical Research Communications 257 79–83.[CrossRef][ISI][Medline]

Barnes S, Peterson G, Grubbs C & Setchell K 1994 Potential role of dietary isoflavones in the prevention of cancer. In Diet and Cancer: Markers, Prevention, and Treatment, pp 135–147. Ed. MM Jacobs. New York: Plenum Press.

Bennett FC & Ingram DM 1990 Diet and female sex hormones concentrations: an intervention study for the type of fat consumed. American Journal of Clinical Nutrition 52 808–812.[Abstract/Free Full Text]

Berkey CS, Frazier AL, Gardner JD & Colditz G 1999 Adolescence and breast carcinoma risk. Cancer 85 2400–2409.[CrossRef][ISI][Medline]

Birrell SN, Bentel JM, Hickey TE, Ricciardelli C, Weger MA, Horsfall DJ & Tilley WD 1995 Androgens induce divergent proliferative responses in human breast cancer cell lines. Journal of Steroid Biochemistry and Molecular Biology 52 459–467.[CrossRef][ISI][Medline]

Boccuzzi G, Brignardello E, di Monaco M, Forte C, Leonardi L & Pizzini A 1992 Influence of dehydroepiandrosterone and 5- en-androstene-3 beta,17 beta-diol on the growth of MCF-7 human breast cancer cells induced by 17 beta-estradiol. Anticancer Research 12 799–803.[ISI][Medline]

Bohlke K, Spiegelman D, Trichopoulou A, Katsouyanni K & Trichopoulos D 1999 Vitamins A, C, and E and the risk of breast cancer: results from a case-control study in Greece. British Journal of Cancer 79 23–29.[CrossRef][ISI][Medline]

Bouker KB & Hilakivi-Clarke L 2000 Genistein: does it prevent or promote breast cancer? Environmental Health Perspectives 108 701–708.[ISI][Medline]

Boyd NF, Martin LJ, Stone J, Greenberg C, Minkin S & Yaffe MJ 2001 Mammographic densities as a marker of human breast cancer risk and their use in chemoprevention. Curren Oncology Reports 3 314–321.

van den Brandt PA, Van’t Veer P, Goldbohm RA, Dorant E, Volovics A, Hermus RJJ & Sturmans F 1993 A prospective cohort study on dietary fat and the risk of postmenopausal breast cancer. Cancer Research 53 75–82.[Abstract/Free Full Text]

Brockman JA, Gupta RA & Dubois RN 1998 Activation of PPARgamma leads to inhibition of anchorage-independent growth of human colorectal cancer cells. Gastroenterology 115 1049–1055.[CrossRef][ISI][Medline]

Brodie A 2003 Aromatase inhibitor development and hormone therapy: a perspective. Seminars in Oncology 30 12–22.[CrossRef]

Cabanes A, Wang M, Olivo S, de Assis S, Gustafsson JA, Khan G & Hilakivi-Clarke L 2004 Prepubertal estradiol and genistein exposures up-regulate BRCA1 mRNA and reduce mammary tumorigenesis. Carcinogenesis 25 741–748.[Abstract/Free Full Text]

Caygill CPJ, Charlett A & Hill MJ 1996 Fat, fish, fish oil, and cancer. British Journal of Cancer 74 159–164.[ISI][Medline]

Chen ZY & Istfan NW 2001 Docosahexaenoic acid, a major constituent of .sh oil diets, prevents activation of cyclindependent kinases and S-phase entry by serum stimulation in HT-29 cells. Prostaglandins Leukotrienes and Essential Fatty Acids 64 67–73.[CrossRef][ISI][Medline]

Cheung B, Yan J, Smith SA, Nguyen T, Lee M, Kavallaris M, Norris MD, Haber M & Marshall GM 2003 Growth inhibitory retinoid effects after recruitment of retinoid X receptor beta to the retinoic acid receptor beta promoter. International Journal of Cancer 105 856–867.

Cho E, Spiegelman D, Hunter DJ, Chen WY, Stampfer MJ, Colditz GA & Willett WC 2003a Premenopausal fat intake and risk of breast cancer. Journal of the National Cancer Institute 95 1079–1085.[Abstract/Free Full Text]

Cho E, Spiegelman D, Hunter DJ, Chen WY, Zhang SM, Colditz GA & Willett WC 2003b Premenopausal intakes of vitamins A, C, and E, folate, and carotenoids, and risk of breast cancer. Cancer Epidemiology Biomarkers and Prevention 12 713–720.[Abstract/Free Full Text]

Cleary ML & Maihle NJ 1997 The role of body mass index in the relative risk of developing premenopausal breast cancer. Proceedings of the Society for Experimental Biology and Medicine 216 28–43.[Abstract]

Cohen LA, Zhao Z & Pittman BSJA 2000 Effect of intact and isoflavone-depleted soy protein on NMU-induced rat mammary tumorigenesis. Carcinogenesis 21 929–935.[Abstract/Free Full Text]

Constantinou AI, Mehta RG & Vaughan A 1996 Inhibition of N-methyl-N-nitroisourea-induced mammary tumors in rats by the soybean isoflavones. Anticancer Research 16 3293–3298.[ISI][Medline]

Cotroneo MS, Wang J, Fritz WA, Eltoum IE & Lamartiniere CA 2002 Genistein action in the prepubertal mammary gland in a chemoprevention model. Carcinogenesis 23 1467–1474.[Abstract/Free Full Text]

D’Amico M, Wu K, Di Vizio D, Reutens AT, Stahl M, Fu M, Albanese C, Russell RG, Muller WJ, White M, Negassa A, Lee H-W, DePinho RA & Pestell RG 2003 The role of Ink4a/Arf in ErbB2 mammary gland tumorigenesis. Cancer Research 63 3395–3402.[Abstract/Free Full Text]

Dang ZC, Audinot V, Papapoulos SE, Boutin JA & Lowik CW 2003 Peroxisome proliferator-activated receptor gamma (PPARgamma) as a molecular target for the soy phytoestrogen genistein. Journal of Biological Chemistry 278 962–967.[Abstract/Free Full Text]

Davis VL, Couse JF, Goulding EH, Power SG, Eddy EM & Korach KS 1994 Aberrant reproductive phenotypes evident in transgenic mice expressing the wild-type mouse estrogen receptor. Endocrinology 135 379–386.[Abstract]

Dawson MI, Hobbs PD, Jong L, Xiao D, Chao WR, Pan C & Zhang XK 2000 sp2-bridged diaryl retinoids: effects of bridge-region substitution on retinoid X receptor (RXR) selectivity. Bioorganic and Medical Chemistry Letters 10 1307–1310.

Dees C, Foster JS, Ahamed S & Wimalasena J 1997 Dietary estrogens stimulate human breast cells to enter the cell cycle. Environmental Health Perspectives 105 (Suppl 3) 633–636.