ToxSci Advance Access originally published online on June 6, 2006
Toxicological Sciences 2006 93(2):357-368; doi:10.1093/toxsci/kfl029
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Estrogenic Effect of Soy Isoflavones on Mammary Gland Morphogenesis and Gene Expression Profile
,2
,1
* Department of Toxicology and Risk Assessment, Danish Institute for Food and Veterinary Research, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark; University Department of Growth and Reproduction, Section GR-5064, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark; and Structural Cell Biology Unit, Department of Medical Anatomy, The Panum Institute, Blegdamsvej 3, DK-2200 Copenhagen N, Denmark
1 To whom correspondence should be addressed. Fax: +(45) 35456054. E-mail: henrik.leffers{at}biobase.dk.
Received March 24, 2006; accepted May 23, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We examined the effect of 17ß-estradiol (E2) and soy isoflavones' exposure on morphogenesis and global gene expression in the murine mammary gland. Three exposure regimens were applied: isoflavones added to the diet throughout either the lactational period (via the dams) or the postweaning period and E2 administered orally during the lactational period. Whole mounts of mammary glands were evaluated both in juvenile and adult animals with respect to branching morphogenesis and terminal end bud (TEB) formation. At postnatal day (PND) 28, we observed a significant increase in branching morphogenesis in all treated groups with the most pronounced effect after E2 exposure. For the E2-treated animals there was also a significant increase in TEB formation. At PNDs 42–43 the postweaning isoflavone and the E2 groups showed a transient reduction in the number of TEBs. A similar response after isoflavone and E2 exposure was further substantiated by changes in gene expression, since the same groups of genes were up- and downregulated, particularly in the E2 and postweaning isoflavone regimen. All changes in gene expression correlated with changes in the cellular composition of the gland, i.e., more and larger TEBs and ducts. The results suggest an estrogenic response of physiological doses of isoflavones on mammary gland development at both the morphological and molecular level, which resembled that induced by puberty.
Key Words: mammary gland; development; estradiol; phytoestrogens; isoflavones.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The natural endogenous estrogen 17ß-estradiol (E2) plays an essential role in the development of various tissues, especially the reproductive organs including the mammary gland (Clarke, 2000
; Hewitt et al., 2000). E2 also plays a critical role in the etiology and progression of breast cancers and other "endocrine-related cancers" (Bernstein, 2002; Mori et al., 1979; Nilsson et al., 2001). Estrogens mediate their action through ligand-dependent activation of estrogen receptors (ER
and ERß). These belong to the nuclear receptor superfamily of transcription factors that regulate transcription in conjunction with coregulatory proteins via binding to DNA elements, located in the promoter regions of target genes (Ray et al., 2005). Environmental estrogen-like compounds have lately received attention as possible mediators of endocrine estrogen activity. Among these are the plant-derived phytoestrogens including the isoflavones, which are nonsteroidal diphenolic compounds that in many aspects have functions resembling those of E2. However, the possible effect of phytoestrogens is complex, as they can have both estrogenic and antiestrogenic effects. The dual effects of isoflavones could partly be explained by their lower estrogenic potency and binding affinities for the two receptors: while E2 binds the ER and ERß equally well, isoflavones preferentially bind ERß (Kuiper et al., 1998), possibly opposing the ER-mediated transcription (Matthews and Gustafsson, 2003; Saunders, 1998). Although phytoestrogens may contribute to the prevention of hormone-dependent cancers, compounds that possess estrogenic activity or can disrupt steroidogenesis may also cause endocrine disruption, particularly when exposure occurs prior to puberty (Newbold et al., 1990; Palmer et al., 2005). Genistein, an isoflavone abundantly present in soybeans, induces an estrogenic response at physiological levels in the uterus and in ER–positive cancer cells in vivo (Jefferson et al., 2002; Matsumura et al., 2005). In addition, isoflavones are reported to mediate estrogen-like adverse effects on the formation of mammary and uterine adenomas (Allred et al., 2004; Day et al., 2001; Luijten et al., 2004).
Animal experiments suggest that exposure to estrogens, including genistein, has an influence on breast cancer risk, conceivably by altering mammary gland development (Cabanes et al., 2004; Hilakivi-Clarke et al., 1998, 1999a; Murrill et al., 1996). At birth, the mammary gland of most mammalian species, including humans, is not completely formed but consists of a small primitive anlage consisting of a few ducts extending from the nipple and growing isometrically in relation to the rest of the body. With the onset of puberty, estrogen level rises and, together with other hormones and growth factors, initiates the development of the mammary gland. In mice, this process begins around postnatal day (PND) 28 and is characterized by increased proliferation at the duct termini, resulting in the formation of the terminal end bud (TEB) (Hovey et al., 2002). These structures are the major sites of proliferation from which ducts elongate and branch dichotomously and sympodially and thereby form the mammary tree. Estrogen signaling is essential for TEB formation and ductal elongation (Korach et al., 1996). Thus, it is likely that the differentiation of the TEB is responsive to environmental estrogens including isoflavones. The TEB may contain immature cells that are potential precursor cells of mammary tumors (Russo and Russo, 1978). Indeed, in rodents the majority of tumors seem to originate from the immature TEB, and the elimination of these structures by differentiation into alveolar structures may reduce the subsequent risk of breast cancer (Russo and Russo, 1978). Animal experiments suggest that estrogenic compounds can alter the course of TEB differentiation; however, the results are contradicting as to whether the differentiation is enhanced or inhibited (Hilakivi-Clarke et al., 1998, 1999b; Murrill et al., 1996). Although the results are controversial, they nevertheless suggest the existence of a "window" in early life where temporal exposure to estrogenic compounds causes alterations in the subsequent mammary gland development.
The aim of the present study was to investigate the possible effects of exposure to E2 and isoflavones on mammary gland morphogenesis and gene expression profile in the juvenile mammary gland of FVB mice. It is generally claimed that isoflavones at low concentrations in vitro have pure estrogenic effects, while at higher concentrations the antiestrogenic effects dominate. Here we examined the in vivo effects of isoflavones administered at concentrations resulting in plasma concentrations comparable to the concentrations found in blood from humans consuming a traditional Asian diet (Adlercreutz et al., 1993) and compared it to the effect of E2 at a concentration known to induce an estrogenic response in vivo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Diet and test compounds.
The animals were fed a semisynthetic diet based on casein (18%) and carbohydrates (cornstarch, potato starch, dextrin, and sucrose) (68%).
Prevastein, a concentrated soy isoflavone product, was obtained from Central Soy Company, Inc. (Port Wayne, IN). According to the manufacturer, the batch used in this study contained 46.19% (wt/wt) of isoflavones, of which 66.5% was genistein, 32.3% daidzein, and 1.2% glycitein. The other constituents in Prevastein were carbohydrates (29%), proteins (17.6%), crude fiber (0.5%), and fat (< 0.1%). Two types of diets were prepared: a control diet and control diet to which was added 0.06% wt/wt of Prevastein (270 mg Prevastein/kg diet). 17ß-estradiol acetate was purchased from Sigma Chemical Co. (St Louis, MO).
Animals, housing, and clinical observations.
FVB mice were obtained from Taconic M&B (Berlin, Germany) and housed one per cage. At weaning, the pubs were housed four per cage. In order to prevent the occurrence of anestrus, four male NMRI mice approximately 4 weeks old were placed in the same room as FVB female mice. The males were obtained from M&B Research (Ry, Denmark) and housed one per cage and fed Altromin 1324 diet. All mice were kept under controlled environmental conditions (temperature 21 ± 1°C, relative humidity 55 ± 5%, 12/12-h light/dark cycle, air changed 10 times per hour) and had free access to food and water acidified to pH 3.0 by citric acid (to prevent growth of microorganisms). All mice were observed at least b.i.d. for any abnormalities in clinical appearance. The Danish Experimental Animal Inspectorate approved the animal studies and all procedures related to the handling of animals.
Experimental design.
In all, 35 dams resulting in 144 female pups were included in the study. The pregnant mice were kept on the control diet until delivery, where they were allocated to two treatment groups receiving either the control diet (25 dams) or control diet to which was added 270 mg Prevastein/kg diet (10 dams). At weaning, the litters of dams that received the control diet were randomized into three treatment groups fed control diet, Prevastein-added diet (270 mg/kg diet), or control diet with E2 (2.5 mg/kg body weight) administered orally as a droplet consisting of approximately 5 µl of 5 mg/ml 17ß-estradiol acetate (10% DMSO in sesame oil) during the lactational period (on PNDs 10, 12, 14, 16, 18, and 20). The litters from dams that received the Prevastein-containing diet were shifted to control diet. As a result, the exposure regimens of litters were (I) control diet (42 female litters); (II) 270 mg isoflavone/kg diet throughout lactation—from PNDs 0 to 21 (38 female litters); (III) 270 mg Prevastein/kg diet from weaning (PND 21) until termination (38 female litters); and (IV) E2 during the lactational period (26 female litters). Body weights of the litters were recorded once weekly. Body weight and feed consumption of female pups were recorded weekly from PND 21 until terminal sacrifice on PNDs 28, 42–43, and 70–72.
Mammary glands and blood samples were collected at PNDs 28, 42–43, and 70–73. Animals were sacrificed while in estrus, as determined by vaginal smear, except at PND 28. All mice were anesthetized by intraperinatal injection of a pentobarbital solution (60 mg/kg body weight), and the arterial system was flushed with 10 ml ice-cold sterile PBS.
Isoflavone and estradiol concentrations in plasma.
Heparinized blood samples were centrifuged at 1500 x g for 10 min at 4°C to separate plasma from RBC. Concentrations of daidzein, genistein, and eqoul were quantified in five samples per group by time-resolved fluoroimmunoassay as previously described (Thomsen et al., 2005). Estradiol concentrations were measured in three to five animals per group by the DELFIA Estradiol Kit from Wallac Oy (Turku, Finland) according to the manufacturer's instructions.
Mammary whole-mount preparation and morphometric analysis.
Mammary whole mounts were prepared from the fourth abdominal gland as previously described (Thomsen et al., 2005). In brief, glands were spread on slides, fixed in Carnoys fixative, rehydrated, stained with carmine alum (2 g/l), and cleared in xylene. After mounting, the whole mounts were examined under a light microscope equipped with ocular micrometer. At PND 28, the number of glands analyzed were 12 from group I, 10 from groups II and III, and 8 from group IV. At PNDs 42–43, 16 glands from group I, 14 from groups II and III, and 10 from group IV were analyzed. At PNDs 70–73, the number of glands analyzed were 14 from groups I–III and 10 from group IV. The length, branching morphogenesis, and stage of gland differentiation occurrence of TEBs, terminal ducts, and alveolar buds in the area most distal to the nipple, based on the classification by Russo and Russo (1978), were analyzed. Data are presented as means ± standard deviations (SDs). Before being subjected to further analysis, all data were tested for normal distribution by Shapiro-Wilks test, and the homogeneity of variance among the groups was evaluated by judgment of standardized residuals plot. All data on mammary gland morphology were analyzed by one-way analysis of variance followed by Student's t-test for the means (unpaired and two tailed) for pairwise comparison. Values of p < 0.05 were considered statistically significant. The statistical analyses were performed using Statistical Analysis System software (release 6.12, 1996, SAS Institute Inc., Cary, NC).
RNA purification.
RNA was isolated from the third mammary gland. Glands from all sacrificed animals were instantly isolated, and muscle tissue was removed. Glands were then immediately snap frozen in liquid nitrogen and stored at – 80°C until use. RNA was purified using Nuclospin Kit (Macherey-Nagel, Düren, Germany) with on-column DNAse treatment, and the quality was confirmed by gel electrophoresis.
Microarray analysis.
The quality of RNA samples was investigated by gel electrophoresis, and only RNA samples of high quality were used for subsequent analysis. The RNA samples for microarray analysis were selected, in accordance to the morphometric data, to be representative for the group. RNA from five animals belonging to the unexposed control group (group I) was amplified by reverse transcription (RT) and in vitro transcription into antisense RNA (aRNA) using the RiboAmp RNA amplification kit (Arcturus GmbH, Moerfelden-Walldorf, Germany) and pooled. This sample served as the common control for the consecutive eight cDNA array analysis. RNA samples from three animals from each of groups II and III and two animals from group IV were likewise amplified with the RiboAmp kit. The antisense RNA (aRNA) samples were then used for cDNA synthesis with an indirect aminoallyl labeling and coupled to Cy3 (exposed animals) or Cy5 (unexposed controls) and hybridized to a 15K cDNA microarray containing 15,264 sequence-verified mouse ESTs (National Institute of Aging, Washington, DC) and spotted in duplicates (UHN-Microarray-Center, Toronto, Canada). Spots were quantified using histogram segmentation. Spots of good quality bypassing shape and signal-to-noise filtering were normalized by framed median ratio centering linear regression using the ChipSkipper software from European Molecular Biology Laboratory (Heidelberg, Germany). Expression ratios were, prior to hierarchical clustering, log2 transformed. In all instances, average linkage and Euclidean distances were used. The normalized data were subsequently analyzed by Statistical Analysis of Microarrays (SAM) and hierarchical clustering in the MeV software (The Institute for Genomic Research; http://www.tm4.org/). Data conforms to MIAME (Minimum Information About a Microarray Experiment) standards outlined by the Microarray Gene Expression Data Society.
RT–polymerase chain reaction.
The cDNA was prepared from 2 µg RNA. Primers were designed to amplify across introns, and all oligonucleotides were purchased from DNA Technology (Aarhus, Denmark). The polymerase chain reaction (PCR) included 5 min at 95°C followed by 30 cycles with hybridization for 1 min at 65°C and elongation for 2 min at 72°C. The primer sequences were as follows: Clu, forward: 5'-GTGACCACCCATTCCTCTGA; Clu, reverse: 5'-AGAGCAGCAAGTGCAGGCAT; CK18, forward: 5'-CATCAACTTGGAGAACAGCCT; CK18, reverse: 5'-GTGCCTCAGAACTCTGGTGT; Crk, forward: 5'-ACGTGAACTGTTTGGTTG-GATT; Crk, reverse: 5'-CTGTCAAGAAAGTGAATAGTGT; Mcmd5, forward: 5'-TGTCCAGGACTTCACCAAACA; Mcmd5, reverse: 5'-CTGTTCCTGCTCCAGACTCA; AP-2
, forward: 5'-AA-CTTGAAGAGGGTAGGCACA; AP-2, reverse: 5'-ATCCACACGTCACCCAACACA; ß-actin, forward: 5'-TTGACAAAACCTAACTTGCGCA; and ß-actin, reverse: 5'-TGCGCAAGTTAGGTTTTG-TCAA.
Real-time PCR.
Real-time PCR was performed using IQ SYBRGreen supermix PCR Reagents (Bio-Rad Laboratories, Hercules, CA) on the iCycler Thermal Cycler according to the manufacturer's instructions and optimized with regard to temperature and template concentration. Melting curves verified that there was only one PCR product in all reactions. This analysis included samples from three to five animals per group. A previously published primer set for ER was used (Waters et al., 2001). Primer sequences were ER, forward, GCCTCTGGCTACCATTATGG and ER, reverse, GCACAAGCGTCAGAGAGATG (product 241 base pairs). The 2–
Ct method was used to calculate the gene expression level relative to the control group (Livak and Schmittgen, 2001).
In situ hybridization.
Tissue sections for in situ hybridization (ISH) were prepared from the fourth abdominal mammary gland; a minimum of three samples from each treatment group and controls were investigated. ISH was performed as described previously (Nielsen et al., 2003). In short, mammary glands were fixed in PFA (4%): 4- to 8-µm sections were placed on Super-Frost Plus slides (Menzel GmbH, Braunschweig, Germany); deparaffinized and PFA (4%) fixed; and treated with Proteinase K (P-2308; Sigma), postfixed (4% PFA), prehybridized for1 h at 50°C, and incubated overnight with the labeled probe (50°C). Visualization was made with Anti-Digoxigenin-AP (Roche Diagnostics GmbH, Mannheim, Germany) and the chromogens BCIP (Sigma B-8503) and NBT (Sigma N-6876).
Immunohistochemistry.
Immunohistochemistry (IHC) was investigated in tissue sections prepared from the fourth abdominal mammary gland on a minimum of three samples from each treatment group and controls. Commercially available polyclonal rabbit anti–AP-2 (H-77): (sc-8977), goat anti-CrkI/II (S-20): (sc-17989), and anti-ER (MC-20): (sc-542) antibodies from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) were used. The immunohistochemical staining with AP-2 and CrkI/II was performed using a standard indirect peroxidase method, where deparaffinized and rehydrated sections were heated in 10 mmol/l citrate buffer, pH 6.0. Subsequently, the sections were treated with 0.5% H2O2 to inhibit the endogenous peroxidase, followed by blockade of unspecific binding sites with nonimmune human serum 1:4 in TBS (Rigshospitalet, Copenhagen, Denmark) in the case of the CrkI/II antibody and normal goat serum (Zymed Histostain kit, code 95-6543B, San Francisco, CA) for the Ap-2 and ER. The primary antibodies polyclonal rabbit anti–AP-2, goat anti-CrkI/II, and anti-ER were diluted 1:200, 1:100 in TBS, and 1:600 in PBS, respectively. Subsequently, a biotinylated goat anti-rabbit (Zymed 95-6143B), a peroxidase-conjugated streptavidin-biotin complex (Zymed 40880848), and anti-rabbit DAKO EnVision + System peroxidase (K3954) were applied, respectively. The sections were thoroughly washed between all steps. The bound AP-2 and CrkI/II antibodies were visualized using aminoethyl carbazole substrate (Zymed 00-2007), whereas ER was visualized with 3,3'-diaminobenzidine + chromogen. Sections were lightly counterstained with Mayer's hematoxylin.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals were enrolled into the following treatment groups: (I) control diet, (II) 270 mg isoflavone/kg diet throughout lactation (via the dams), (III) 270 mg isoflavones/kg diet from weaning until termination, (IV) E2 (2.5 mg/kg body weight) during lactation. Mammary glands were collected and morphometrically evaluated at the beginning of puberty, midpuberty, and late puberty corresponding to PNDs 28, 42–43, and 70–73, respectively. Gene expression analysis and verification was performed on tissues collected on PND 28.
Plasma Levels of Isoflavones and Estradiol
At PND 28, the plasma concentration of isoflavone metabolites in animals from group II is shown in Table 1 and confirms to increased plasma concentration of isoflavones compared to animals on control diet (groups I, III, and IV) at this age (Table 1). Plasma concentrations at PND 21 for group I was not measured since sacrifice was not performed at this age. However, data from previous studies, using identical route of administration of the same batch of Prevastein in mice, show a similar level of isoflavones in plasma following lactational exposure and postweaning exposure (Luijten et al., 2004; Thomsen et al., 2005). The plasma levels of estradiol were comparable among all exposure regimens, as shown in Table 1.
|
Effects on Mammary Gland Morphology
Examples of whole mounts of mammary glands at PND 28 from FVB virgin mice from the control group and the three exposure regimens are presented in Figure 1. The mammary trees from E2-treated animals were shorter, more branched, and contained more TEBs than glands from animals on the control diet, whereas the isoflavone-treated animals seemed only to differ from controls by increased branching. Morphometric analysis of the mammary epithelium on PNDs 28, 42–43, and 70–72 showed no statistically significant difference in the relative length of the mammary tree in the three exposure trials compared to the controls, except for significantly shorter (p < 0.05) glands in the E2 exposure group (group IV) at PNDs 42–43 (Fig. 2). A significant (p < 0.05) increase in overall branching in the juvenile mammary gland (PND 28) was observed in all exposed groups (groups II–IV). At PNDs 42–43, increased branching was only observed in glands from the E2-treated animals (group IV). In the juvenile mammary gland, the number of TEBs was only significantly affected in the E2 treatment (group IV), although a similar effect was found after postweaning isoflavone treatment (group III). At PNDs 42–43, a statistically significant reduction in the number of TEBs was evident after postweaning isoflavone treatment and E2 treatment (groups III and IV) compared to controls at this time point. This effect was, however, transient as no significant difference between control and treated glands could be observed in the adult animals (PNDs 70–73). Accordingly, if the number of immature cells located in the TEB is central to the etiology of mammary tumorigenesis, the consequence of isoflavone or E2 exposure would depend on the specific timing of tumor initiation.
|
|
Gene Expression Profiling
To complement the morphological changes in the mammary gland, we used a total of eight cDNA microarrays from the National Institute of Aging each containing 15,264 sequence-verified mouse ESTs, to screen for differences in gene expression in mammary tissue from the control and the three exposure groups. In all, eight microarrays were performed. RNA from the unexposed control group was cohybridized with three independent RNA samples from animals from groups II and III and two from animals from group IV. The level of plasma estradiol was comparable between the groups and among the samples chosen for the microarray, with 0.44nM as the lowest and 0.66nM as the highest concentration.
Hierarchical clustering of genes differentially expressed by more than twofold combined with the tree constructed from hierarchical clustering (with average linkage and Euclidean distances based on all 15,264 genes) revealed that the three treatments resulted in very similar gene expression profiles (Fig. 3).
|
We used SAM to identify differentially expressed genes. SAM identifies genes with statistically significant changes in expression by assimilating a set of gene-specific t-tests. One-class SAM of all 15K genes identified more than 100 genes that were significantly differentially expressed in all eight expression arrays (with a delta value of 6.1, equal to a false discovery rate of zero genes). The top 20 regulated genes are listed in Table 2, together with the biological pathway as suggested by SOURCE (http://source.stanford.edu). Two-class SAM was used for pairwise comparison of the different treatments. No genes were found to differ in expression level when comparing the postweaning isoflavone exposure to the E2 exposure, and only very few genes were differently expressed when comparing the lactational isoflavone exposure to the E2 exposure, all differences are listed in Table 3. This strongly suggests a similar genetic response to E2 and isoflavone treatment.
|
|
For verification, 10 genes regulated in all three treatments were further studied by RT-PCR (the selected genes are indicated in Fig. 3). A significant sample-to-sample variation in expression levels was observed, especially among the control samples. For most genes, the differential expression could be verified for the samples used for microarray analysis; however, when additional samples were analyzed, we often observed similar expression levels in some control samples and in the treated samples (Fig. 4A; Clusterin). Nevertheless, for most of the analyzed genes, RT-PCR confirmed the differential expression (Figs. 4A and 4B). Since the changes observed after treatment closely resemble the changes normally induced at puberty, the variation was likely caused by small differences in the timing of the onset of puberty in the control group.
|
Identification of the Expressing Cell Types
We used IHC and ISH to identify the cells that expressed some of the differentially expressed genes. A minimum of three samples from each treatment group and controls were investigated. Antibodies against Crk and AP-2
were applied to sections of mammary tissue from both controls and treated animals at PND 28. The CrkI/II antibody stained epithelia cells with a slightly stronger staining in cap/transition cells as compared to body cells and fibroblasts surrounding the TEBs (examples are presented in Figs. 5A and 5B). In addition, a subset of adipocytes were also stained albeit with a lower intensity. There was no difference in the intensity of the staining, but TEBs from the control animals were smaller and less developed compared to TEBs from E2- and isoflavone-treated animals (Fig. 5). IHC with antibodies against AP-2 showed expression in cap/transition and body cells of the TEBs, in ductal epithelium and also in some fibroblasts. However, the staining of all cell types showed a mixed pattern with both positive and negative cells. The staining was mainly nuclear (as expected); however, we also consistently observed some cytoplasmic staining. As for the Crk antibody, there were no significant differences in staining intensity between treated and control samples. Thus, the higher expression in the treated groups, found in the microarray, most likely reflected a change in cellularity (i.e., the cell-type composition) rather than an increased expression in existing cells (Fig. 5).
|
The expression of Clusterin (apolipoprotein J) was investigated by ISH. The most intense staining was found in cap/transition cells in the TEBs, but Clusterin was also expressed at the extending edge of lateral branches (an example is shown in Fig. 6). A low expression could be detected in epithelial body cells, myoepithelia cells, and in adipose tissue. In addition, Clusterin was also expressed in the lymph node, but the staining was not as strong as in cap/transition cells. Again, in agreement with the results from the IHC staining, there was no difference in staining intensity between mammary glands from treated and control animals, suggesting that the increased expression in treated samples was caused by a higher number of TEB cells and lateral branches in mammary glands from treated animals. Together, these results strongly suggest that the altered gene expression was caused by an increase in the number of epithelial cells expressing these genes and not by a specific up- and downregulation within the cells.
|
Gene Expression of ER
It is suggested that exposure to an increased level of E2 result in a tissue-specific downregulation of its receptors, probably due to autoregulative mechanisms (Hatsumi and Yamamuro, 2006
). Because of the low expression level of ER, the expression of this gene could not be determined from the microarray data. Thus, real-time PCR was used to evaluate the expression of ER in animals treated with isoflavones and E2 during the prepubertal period, relative to the control group. The real-time PCR data for groups II and III were normalized to the control group and are presented as 2–Ct values. The result indicated a downregulation of more than twofold for both treatments (Fig. 7A). Staining of mammary glands with an ER antibody was used to confirm the real-time PCR data. The ER antibody stained both luminal epithelial cells of the duct and body cells of the TEB (Fig. 7B). The staining of cells in the control glands seemed qualitatively to be slightly more intense than in the treated glands, supporting the conclusions from real-time PCR data.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results suggest that isoflavones given at physiological concentrations (resulting in serum levels of approximately 500nM) cause analogous morphological and gene expression changes, as treatment with E2 at doses known to induce an estrogenic response in vivo (2.5 mg/kg body weight).
Early postnatal exposure to 17ß-estradiol acetate (E2) and isoflavones affected the pattern of mammary gland morphogenesis in the juvenile gland by inducing an increased branching of the ductal tree in the juvenile gland, suggesting an early growth stimulatory effect. Also, when mice were exposed to E2 during PNDs 10–20, the number of proliferative TEBs was increased at PND 28. An increase in TEBs was found following postweaning isoflavone exposure (PNDs 21–28) as well, although it did not reach statistical significance. At PNDs 42–43, the number of the proliferative TEB was reduced by all treatments, with statistical significance after postweaning isoflavone treatment and E2 treatment, supporting the notion of accelerated or induced onset of pubertal maturation, which in theory may result in a reduced cancer risk (Murrill et al., 1996). However, this effect was transient, as no reduction in the TEB number was observed at adulthood (PNDs 70–73). This could suggest that the treatment did not as such induce the formation of additional TEBs but rather enhanced the normal development. If the number of TEBs, i.e., the number of immature proliferating cells, is an important aspect in the etiology of mammary tumorigenesis, then the effect of environmental estrogens may depend on the specific timing of tumor initiation: a protective effect may be expected if the tumor transformation is initiated late in puberty but the opposite may be true if the tumor transformation is initiated at an earlier point in life.
To complement the findings of morphological alterations in the juvenile gland (PND 28) and to clarify whether exposure to isoflavones and E2 induced distinct pathways, we compared the global gene expression profiles in the different exposure groups. We found that the gene expression profiles of the isoflavone exposures were remarkably similar to those induced by E2. The result is in accordance with a previous study, where gene expression after exposure to genistein and synthetic and physiologic estrogens was investigated in immature mouse uterus. The authors showed that the different types of estrogens induced similar genetic responses and that the differences in transcriptional responses probably were caused by dose-dependent variations in magnitude and kinetics of gene expression rather than induction of distinct pathways (Moggs et al., 2004). Moreover, ER seemed downregulated (
50%) after exposure to both isoflavone and E2 during the lactational period. This effect was previously observed in the mammary gland after exposure to exogenous E2, probably as a consequence of tissue-specific autoregulation (Hatsumi and Yamamuro, 2006).
Having identified a set of differentially expressed genes, we sought to clarify if the effects of isoflavones and E2 on gene expression were caused by the induction of genes in existing cell types or by an altered number of the cells expressing the genes, i.e., changes in cellularity. Based on their relevance to mammary gland development, three of the upregulated genes (AP-2, Crk, and Clusterin) were selected for investigation of the cell types expressing the genes. AP-2 was expressed in ductal epithelia cells and myofibroblasts in the adult mammary gland in a pattern similar to what has previously been described (Zhang et al., 2003). The AP-2 family of transcription factors might participate in direct activation of ER-, Insulin-like Growth Factor Receptor, and ErbB-mediated proliferative signaling and may also participate in regulating mammary gland morphogenesis (Hoei-Hansen et al., 2004; Turner et al., 1998; Zhang et al., 2003); AP-2 is significantly upregulated in early-stage breast tumors and in testicular carcinomas (Hoei-Hansen et al., 2004; Turner et al., 1998). The highest expression of Crk was observed in cap/transition cells in the TEBs, but it was also expressed in body cells and luminal epithelial cells. Crk is involved in epithelial invasion and morphogenesis and is a mediator of ErbB signaling, and its expression pattern is thus in accordance with its putative function in cell migration (Lamorte et al., 2002). For both Crk and AP-2, we could not detect a change in staining intensity between mammary glands from treated and control animals. However, the TEBs from the treated glands (groups III and IV) were significantly larger in size and seemed further developed compared to the controls (Fig. 5). Finally, we used ISH to determine the precise expression pattern of Clusterin (Apo J), which, in the microarray data showed a significantly higher expression in all treatment groups as compared to controls. ISH showed that the Clusterin mRNA was localized to cap/transition cells of the TEBs and to the leading edge of lateral branches. As for Ap-2 and Crk, the staining intensity among controls and treated animals was similar, but there were more and larger TEBs in the treated glands. Clusterin is involved in tissue remodeling and in promoting and preventing apoptosis (Trougakos and Gonos, 2002), and Clusterin is highly expressed in bladder transitional cell carcinoma (Miyake et al., 2002) and in human breast carcinoma, whereas it is barely detectable in normal adult breast tissue (Redondo et al., 2000).
Together, the observations suggest that the increased expression of these genes probably reflects (1) an increase in the number of cells in the TEBs, (2) an increase in lateral branches, and (3) the presence of more luminal epithelial cells in the mammary gland. That changes in cellularity is the cause of the changes in gene expression is supported by the increased expression in all experiments of four cytokeratins (CK7, CK8, CK18, and CK19). The cytokeratins are markers for luminal epithelial cells (Petersen et al., 2003; Stingl et al., 2005), and an increased total ductal volume and, thus, a higher percentage of ductal epithelia cells can explain their increased expression.
Interestingly, three of the identified genes, CK19, Crk, and AP-2, are referred to as stem cell markers and are also highly expressed in mammary cancer cells. In the human mammary gland, CK19 is expressed in a subpopulation of cells with stem cell characteristics located within the luminal epithelial compartment and is also expressed in the majority of breast cancer cells arising from these compartments (Petersen et al., 2003). It has been proposed that exposure to hormones in utero or during the immediate early postnatal growth may modulate the number of stem cells and their proliferative potential (Baik et al., 2005; Trichopoulos, 1990). The increase in the number of cells expressing CK19, Crk, and AP-2 may imply an increase in mammary stem cells introduced by early exposure to E2 and isoflavones. Since mammary stem cells/progenitor cells are suggested to be the cellular origin of at least a subset of breast cancers (Petersen et al., 2003), an increase in their number may result in an increased breast cancer risk.
In conclusion, the changes in gene expression after E2 and isoflavone exposure most likely reflect altered cellularity, whereas an increase in ductal epithelium or in the number and volume of TEBs would result in an apparent upregulation of all genes that are highly expressed in these specific structures. Hence, the working hypothesis that estrogenic compounds could enhance mammary gland development by promoting an earlier proliferation in the mammary gland is supported by the results herein, as evident by increased branching and increased number of TEBs. In addition, the notion that this early event should lead to increased glandular differentiation was also supported, as we observed a significant transient reduction in TEB number at midpuberty (PNDs 42–43). This reduction in TEB number implies that the further differentiation of the immature cells of the TEB into luminal epithelial cells is enhanced by both isoflavone and E2 treatment, which in theory could reduce the subsequent breast cancer risk. However, as a consequence of the initial increase in the number of proliferating cells (at PND 28), the potential effect of these compounds on breast cancer risk may depend on the specific timing of tumor initiation.
The appearance of TEBs and initiation of ductal branching observed in the treated groups are also indicators of the onset of puberty. Thus, the observed effects could be caused by treatment-induced puberty, resulting in a more developed stage of the gland. Thus, it remains to be elucidated whether the alterations in morphology and gene expression pattern are caused directly by the E2 and isoflavone exposure or whether these compounds induced early puberty.
|
NOTES |
|---|
2 Present address: Molecular Genetics, Novo Nordisk A/S, Novo Allé, DK-2880 Bagsværd, Denmark.
3 Present address: Maxygen Aps, Agern Alle 1, DK-2970 Hørsholm, Denmark.
|
ACKNOWLEDGMENTS |
|---|
The authors thank Pernille Timmerby, Brian Vendelbo, and Malene Dalgaard for their skilful technical assistance. The study was supported by grants from the Commission of the European Communities: FAIR program (CT95-0894) and EU-FW5 (QLK1-2000-00266).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adlercreutz, H., Markkanen, H., and Watanabe, S. (1993). Plasma concentrations of phyto-oestrogens in Japanese men. Lancet 342, 1209–1210.[CrossRef][ISI][Medline]
Allred, C. D., Allred, K. F., Ju, Y. H., Goeppinger, T. S., Doerge, D. R., and Helferich, W. G. (2004). Soy processing influences growth of estrogen-dependent breast cancer tumors in mice. Carcinogenesis 15, 1649–1657.
Baik, I., Devito, W. J., Ballen, K., Becker, P. S., Okulicz, W., Liu, Q., Delpapa, E., Lagiou, P., Sturgeon, S., Trichopoulos, D., et al. (2005). Association of fetal hormone levels with stem cell potential: Evidence for early life roots of human cancer. Cancer Res. 1, 358–363.
Bernstein, L. (2002). Epidemiology of endocrine-related risk factors for breast cancer. J. Mammary Gland Biol. Neoplasia 7, 3–15.[CrossRef][ISI][Medline]
Cabanes, A., Wang, M., Olivo, S., DeAssis, S., Gustafsson, J. A., Khan, G., and Hilakivi-Clarke, L. (2004). Pre-pubertal estradiol and genistein exposures up-regulate BRCA1 mRNA and reduce mammary tumorigenesis. Carcinogenesis 5, 741–748.
Clarke, R. (2000). Introduction and overview: Sex steroids in the mammary gland. J. Mammary Gland Biol. Neoplasia 5, 245–250.[CrossRef][ISI][Medline]
Day, J. K., Besch-Williford, C., McMann, T. R., Hufford, M. G., Lubahn, D. B., and MacDonald, R. S. (2001). Dietary genistein increased DMBA-induced mammary adenocarcinoma in wild-type, but not ER alpha KO, mice. Nutr. Cancer 39, 226–232.[CrossRef][ISI][Medline]
Hatsumi, T., and Yamamuro, Y. (2006). Downregulation of estrogen receptor gene expression by exogenous 17ß-estradiol in the mammary glands of lactating mice. Exp. Biol. Med. 231, 311–316.
Hewitt, S. C., Couse, J. F., and Korach, K. S. (2000). Estrogen receptor transcription and transactivation: Estrogen receptor knockout mice: What their phenotypes reveal about mechanisms of estrogen action. Breast Cancer Res. 2, 345–352.[CrossRef][ISI][Medline]
Hilakivi-Clarke, L., Cho, E., and Clarke, R. (1998). Maternal genistein exposure mimics the effects of estrogen on mammary gland development in female mouse offspring. Oncol. Rep. 5, 609–616.[ISI][Medline]
Hilakivi-Clarke, L., Cho, E., Onojafe, I., Raygada, M., and Clarke, R. (1999a). Maternal exposure to genistein during pregnancy increases carcinogen-induced mammary tumorigenesis in female rat offspring. Oncol. Rep. 6, 1089–1095.[ISI][Medline]
Hilakivi-Clarke, L., Onojafe, I., Raygada, M., Cho, E., Skaar, T., Russo, I., and Clarke, R. (1999b). Pre-pubertal exposure to zearalenone or genistein reduces mammary tumorigenesis. Br. J. Cancer 80, 1682–1688.[CrossRef][ISI][Medline]
Hoei-Hansen, C. E., Nielsen, J. E., Almstrup, K., Brask, S., Graem, N., Skakkebaek, N. E., Leffers, H., and Rajpert-De Meyts, E. (2004). Transcription factor AP-2 is a developmentally regulated marker of testicular carcinoma in situ and germ cell tumors. Clin. Cancer Res. 10, 8521–8530.
Hovey, R. C., Trott, J. F., and Vonderhaar, B. K. (2002). Establishing a framework for the functional mammary gland: From endocrinology to morphology. J. Mammary Gland Biol. Neoplasia 7, 17–38.[CrossRef][ISI][Medline]
Jefferson, W. N., Padilla-Banks, E., Clark, G., and Newbold, R. R. (2002). Assessing estrogenic activity of phytochemicals using transcriptional activation and immature mouse uterotrophic responses. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 25, 179–189.
Korach, K. S., Couse, J. F., Curtis, S. W., Washburn, T. F., Lindzey, J., Kimbro, K. S., Eddy, E. M., Migliaccio, S., Snedeker, S. M., Lubahn, D. B., et al. (1996). Estrogen receptor gene disruption: Molecular characterization and experimental and clinical phenotypes. Recent Prog. Horm. Res. 51, 159–188.[ISI][Medline]
Kuiper, G. G., Lemmen, J. G., Carlsson, B., Corton, J. C., Safe, S. H., van der Saag, P. T., van der Burg, B., and Gustafsson, J. A. (1998). Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139, 4252–4263.
Lamorte, L., Rodrigues, S., Naujokas, M., and Park, M. (2002). Crk synergizes with epidermal growth factor for epithelial invasion and morphogenesis and is required for the met morphogenic program. J. Biol. Chem. 277, 37904–37911.
Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2 (-Delta Delta C(T)) Method. Methods. 25, 402–408.[CrossRef][ISI][Medline]
Luijten, M., Thomsen, A. R., van den Berg, J. A., Wester, P. W., Verhoef, A., Nagelkerke, N. J., Adlercreutz, H., van Kranen, H. J., Piersma, A. H., Sorensen, I. K., et al. (2004). Effects of soy-derived isoflavones and a high-fat diet on spontaneous mammary tumor development in Tg.NK (MMTV/c-neu) mice. Nutr. Cancer 50, 46–54.[Medline]
Matsumura, A., Ghosh, A., Pope, G. S., and Darbre, P. D. (2005). Comparative study of oestrogenic properties of eight phytoestrogens in MCF7 human breast cancer cells. J. Steroid Biochem. Mol. Biol. 94, 431–443.[CrossRef][ISI][Medline]
Matthews, J., and Gustafsson, J. A. (2003). Estrogen signaling: A subtle balance between ER alpha and ER beta. Mol. Interv. 3, 281–292.
Miyake, H., Gleave, M., Kamidono, S., and Hara, I. (2002). Overexpression of clusterin in transitional cell carcinoma of the bladder is related to disease progression and recurrence. Urology 59, 150–154.[CrossRef][ISI][Medline]
Moggs, J. G., Ashby, J., Tinwellm, H., Limm, F. L., Moore, D. J., Kimber, I., and Orphanides, G. (2004). The need to decide if all estrogens are intrinsically similar. Environ. Health Perspect. 112, 1137–1142.[ISI][Medline]
Mori, T., Nagasawa, H., and Bern, H. A. (1979). Long-term effects of perinatal exposure to hormones on normal and neoplastic mammary growth in rodents: A review. J. Environ. Pathol. Toxicol. 3, 191–205.[ISI][Medline]
Murrill, W. B., Brown, N. M., Zhang, J. X., Manzolillo, P. A., Barnes, S., and Lamartiniere, C. A. (1996). Pre-pubertal genistein exposure suppresses mammary cancer and enhances gland differentiation in rats. Carcinogenesis 17, 1451–1457.
Newbold, R. R., Bullock, B. C., and McLachlan, J. A. (1990). Uterine adenocarcinoma in mice following developmental treatment with estrogens: A model for hormonal carcinogenesis. Cancer Res. 50, 7677–7681.
Nielsen, J. E., Hansen, M. A., Jørgensen, M., Tanaka, M., Almstrup, K., Skakkebaek, N. E., and Leffers, H. (2003). Germ cell differentiation-dependent and stage-specific expression of LANCL1 in rodent testis. Eur. J. Histol. 47, 215–222.
Nilsson, S., Makela, S., Treuter, E., Tujague, M., Thomsen, J., Andersson, G., Enmark, E., Pettersson, K., Warner, M., and Gustafsson, J. A. (2001). Mechanisms of estrogen action. Physiol. Rev. 81, 1535–1565.
Palmer, J., Wise, L. A., Robboy, S. J., Titus-Ernstoff, L., Noller, K. L., Herbst, A. L., Troisi, R., and Hoover, R. N. (2005). Hypospadias in sons of women exposed to diethylstilbestrol in utero. Epidemiology 16, 583–586.[CrossRef][ISI][Medline]
Petersen, O. W., Gudjonsson, T., Villadsen, R., Bissell, M. J., and Ronnov-Jessen, L. (2003). Epithelial progenitor cell lines as models of normal breast morphogenesis and neoplasia. Cell Prolif. 36,(Suppl), 33–44.[Medline]
Ray, G., Banerjee, S., Saxena, N. K., Campbell, D. R., Van Veldhuizen, P., and Banerjee, S. K. (2005). Stimulation of MCF-7 tumor progression in athymic nude mice by 17beta-estradiol induces WISP-2/CCN5 expression in xenografts: A novel signaling molecule in hormonal carcinogenesis. Oncol. Rep. 13, 445–448.[ISI][Medline]
Redondo, M., Villar, E., Torres-Munoz, J., Tellez, T., Morell, M., and Petito, C. K. (2000). Overexpression of clusterin in human breast carcinoma. Am. J. Pathol. 157, 393–399.
Russo, I. H., and Russo, J. (1978). Developmental stage of the rat mammary gland as determinant of its susceptibility to 7,12-dimethylbenz[a]anthracene. J. Natl. Cancer Inst. 61, 1439–1449.[ISI][Medline]
Saunders, P. T. (1998). Oestrogen receptor beta (ER beta). Rev. Reprod. 3, 164–171.[Abstract]
Stingl, J., Raouf, A., Emerman, J. T., and Eaves, C. J. (2005). Epithelial progenitors in the normal human mammary gland. J. Mammary Gland Biol. Neoplasia 10, 49–59.[CrossRef][ISI][Medline]
Thomsen, A. R., Mortensen, A., Breinholt, V. M., Lindecrona, R. H., Peñalvo, J. L., and Sørensen, I. K. (2005). Influence of Prevastein®, an isoflavone-rich soy product on mammary gland development and tumerigenesis. Nutr. Cancer 52, 176–188.[CrossRef][ISI][Medline]
Trichopoulos, D. (1990). Hypothesis: Does breast cancer originate in utero? Lancet 21, 939–940.
Trougakos, I. P., and Gonos, E. S. (2002). Clusterin/apolipoprotein J in human aging and cancer. Int. J. Biochem. Cell Biol. 34, 1430–1448.[CrossRef][ISI][Medline]
Turner, B. C., Zhang, J., Gumbs, A. A., Maher, M. G., Kaplan, L., Carter, D., Glazer, P. M., Hurst, H. C., Haffty, B. G., and Williams, T. (1998). Expression of AP-2 transcription factors in human breast cancer correlates with the regulation of multiple growth factor signalling pathways. Cancer Res. 58, 5466–5472.
Waters, K. M., Safe, S., and Gaido, K. W. (2001). Differential gene expression in response to methoxychlor and estradiol through ERalpha, ERbeta, and AR in reproductive tissues of female mice. Toxicol. Sci. 63, 47–56.
Zhang, J., Brewer, S., Huang, J., and Williams, T. (2003). Overexpression of transcription factor AP-2alpha suppresses mammary gland growth and morphogenesis. Dev. Biol. 256, 127–145.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||