| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Gene Regulation Group (E.K.L., P.L.C., C.A.A.), Laboratory of Molecular Carcinogenesis, National Institute of Environmental Health Sciences Microarray Center (L.B., P.R.B., C.A.A.), and Biostatistics Branch (L.L.), National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina 27709
Address all correspondence and requests for reprints to: Edward K. Lobenhofer, National Institute of Environmental Health Sciences, P.O. Box 12233 MD2-04, Research Triangle Park, North Carolina 27709. E-mail: lobenho1{at}niehs.nih.gov.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The nongenomic actions of estrogen may also modulate gene expression patterns. For example, several groups demonstrated that estrogen activates the MAPK cascade in breast cancer cells (8, 9). Although the kinetics of the induction are debated, pharmacological inhibition of this pathway abrogates estrogen-induced mitogenesis of estrogen-responsive, breast carcinoma-derived MCF-7 cells (10, 11). This suggests MAPK activity is a requirement for estrogen action; one of the culminating events of the MAPK cascade is activation of the transcription factor elk1, which may lead to further estrogen-associated gene expression changes (12).
Estrogen’s ability to regulate gene expression in cell culture model systems is affected by culturing conditions. The ability of MCF-7 cells to proliferate in response to a physiologically relevant concentration of estrogen is retained in serum-free conditions, albeit at a slower rate than cells stimulated in the presence of charcoal-stripped serum (CSS) (13). Estrogen treatment of MCF-7 cells in the presence of serum growth factors results in increased expression of the progesterone receptor (14, 15). However, Katzenellenbogen and Norman (16) demonstrated that in the absence of serum, a broad range of estrogen concentrations are unable to stimulate progesterone receptor levels. This suggests that a component of serum is necessary but not sufficient to increase progesterone receptor levels. Interestingly, synergy of estrogen with serum components is not necessary for induction of all estrogen-responsive genes. For example cathepsin-D levels are increased in response to estrogen regardless of the presence or absence of serum (17). Additional studies, including the one presented here, that analyze the effect of estrogen in serum-free conditions will allow further elucidation of the estrogen targets vs. those that result from multiple growth factor triggers/synergies.
A significant amount of research has focused on the changes evoked by estrogen that enable a cell to proceed through the G1/S checkpoint. Cumulatively, these studies have demonstrated that increased expression of cyclin D1 is one of the rate-limiting steps for progression into the DNA replication stage of the cell cycle (reviewed in Ref. 18). Cyclin D1 protein then binds to the cyclin-dependent kinase cdk4, generating an active kinase complex. This complex, as well as cyclin E/cdk2, phosphorylate the retinoblastoma (Rb) protein, which is bound to the transcription factor E2F. Phosphorylation of Rb stimulates the dissociation of E2F, enabling E2F to mediate the transcription of responsive genes required for DNA replication (19). These findings suggest that estrogen-induced expression of cyclin D1 is capable of relieving the negative regulation of E2F, thereby permitting replication of the genome. Many E2F target genes with functions implicated in DNA replication have been identified, most recently in microarray-based studies (20, 21, 22). However, increased expression of these genes, with the exception of proliferating cell nuclear antigen (PCNA), in response to estrogen stimulation has not been previously examined.
To clarify the molecular mechanisms underlying estrogen’s ability to drive cell cycle progression, we sought to identify transcriptional changes in genes associated with replication after estrogen stimulation in the absence of serum using cDNA microarray technology. To maximize statistical confidence in the data analysis, we employed a targeted gene chip for these studies (23). Our custom gene chip was spotted with 1,901 human genes containing genes especially selected for hormone-regulated studies. The chip contains genes from classes including transcription factors, kinases, phosphatases, growth factors, metabolism, and cell cycle genes, in addition to over 200 genes implicated in DNA replication and repair and the response to estrogen. This chip provided a powerful tool to study the global expression profile resulting from estrogen stimulation of MCF-7 cells. RNA isolated from MCF-7 cells treated with a physiological concentration of E2, the most prevalent estrogen produced by the ovaries, was harvested at six separate time points to follow the kinetics of regulated gene expression. In addition to revealing clues regarding the mechanisms of estrogen action, this work establishes an expression fingerprint for estrogen-regulated transcriptional events in breast epithelial-derived cells and adds to past studies using oligonucleotide arrays and other molecular techniques to identify estrogen- regulated targets (24, 25, 26, 27, 28). This information can be used as a benchmark to further explore the tissue-specific changes modulated by estrogen, and the action of clinically relevant estrogen antagonists (such as tamoxifen and raloxifene), as well as to serve as a base for the development of an estrogen-associated gene signature that may be used to screen for estrogen mimics (29).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Consistent with previous findings (13, 31, 32), concentrations of E2 in excess of 1 x 10-11 M were able to stimulate a measurable increase in cell number (2.5- to 3-fold) in the presence or absence of CSS. The magnitude of the proliferative response was greater in the presence of serum, indicating the ability of E2 to synergize with growth factors present in the hormone-depleted serum.
|
). This is important to examine because microarray data are expressed as a ratio of the level of expression of estrogen-treated vs. control cells. Changes in the expression of genes in the control cells, due to cell death for example, would be reflected in the expression ratios, and could be wrongly interpreted as changes mediated by estrogen. The similarity of viability regardless of the presence or absence of 10% CSS (90% without serum, 92% with serum), at the 48-h time point (i.e. the longest time point in the experiment) assured us that the effects we were observing were due to estrogen-induced proliferation not cell death of controls.
Relative Levels of ER
and ERß in MCF-7 Cells
Currently, two distinct forms of the ER have been identified that not only demonstrate different expression patterns in target tissues but also regulate the expression of different genes (33, 34, 35). Therefore, a pervasive question in estrogen-associated mechanistic studies is which receptor is mediating the observed estrogen-induced responses. Though the tissue- specific expression of ER and ERß has been documented, it was reported that different MCF-7 variants do not express the same levels of each receptor (36). For this reason, the relative expression level of each receptor was measured using semiquantitative RT-PCR (Fig. 2). Consistent with previously published findings, T47D and MCF-7 cells express ER, whereas SKBR-3 cells do not (37). Furthermore, our particular T47D variant expresses ERß, whereas SKBR-3 and MCF-7 cells do not demonstrate detectable levels of this receptor. These findings indicate that the changes in the global expression profile of MCF-7 cells stimulated with estrogen in this study result from either the nongenomic actions of estrogen or transcriptional mechanisms involving ER, or both.
|
). (For information pertaining to the level of induction and repression and at which time points these changes were statistically significant, see Table 4, which is published as supplemental data on The Endocrine Society’s Journals Online web site, http://mend. endojournals.org/). These genes were broadly categorized into nine different groups based on their known function, such as the transcription factors myc and myb, apoptotic genes including caspase 7, and genes involved in cellular signaling like PKC. Not surprisingly, several of these categories correlated well with the known phenotype resulting from estrogen stimulation, e.g. increased expression of genes involved in cell cycle progression and DNA replication. Many of these genes were previously identified as being regulated by estrogen stimulation (indicated in Table 1, e.g. cathepsin D, cyclin D1, c-myc, PCNA, etc.); however, many previously unreported genes also appeared to demonstrate estrogen-regulated expression, including two different DNA polymerases (
and 
) as well as DNA ligase I. The induced expression of several genes involved in DNA replication was further validated by one of two different approaches (flap structure endonuclease-1 and Replication Factor C3 by Northern analysis and PCNA and stromal-derived factor-1 by real-time PCR) (data not shown). The expression levels measured by these alternative methods closely resembled the kinetics of expression observed with the microarray results.
|
). The Pearson correlation coefficient was used to quantify this similarity, and the resulting correlation values suggest that there is good agreement between the controls across the six times. This process was also repeated using the 105 limited set of genes that were identified as significantly changed by estrogen exposure. The Pearson correlation coefficient values (see Table 2B) indicate that expression was generally stable as reflected by correlation, r-values, greater than 0.92 for the entire data set and 0.87 for the select gene set.
|
). Cumulatively, the strong correlation of the ethanol-treated samples over time and the lack of any genes being identified as significantly regulated by ethanol indicates that the expression profiles observed in the estrogen-treated hybridizations is a result of the cellular response to estrogen as opposed to the vehicle. (The entire data set for ethanol and estrogen treatments is available at: http://dir.niehs.nih. gov/microarray/datasets/estrogen_ethanol_time_course.txt and http://mend.endojournals.org/).
|
. The degree of similarity of the various time points is reflected in the tree (illustrated on the top of the enlarged node). The location of the 1-h time point on its own branch indicates that it is the most dissimilar sample compared with all of the time points. At this early time, the main response to estrogen is the increased expression of immediate early genes, such as c-myc. Maximal expression levels of c-myc are achieved between 1 and 2 h after estrogen stimulation, and by 4 h near basal levels have been restored (43). Therefore, it was expected that this time point would have the least in common with the other time points. The dendrogram clearly illustrates that little difference exists between the 24 and 36 h time points. This is in agreement with cell cycle data, which demonstrates that at these intervals, a large proportion of the cells are actively undergoing DNA synthesis and progressing toward the mitosis phase of the cell cycle (data not shown). Also illustrated in this cluster is the expression profile of MCF-7 cells treated with estrogen for 24 h in the presence of CSS. The overall pattern is very similar to the one observed in the absence of serum for the same time frame. However, the tree reveals that the 24-h sample derived in the absence of serum is more similar to the 12, 36, and 48 h samples also isolated in the absence of serum. This may be due to overall increased levels of induction and repression in the presence of CSS, which would also be consistent with the increased levels of proliferation observed in Fig. 1A.
|
represents a cluster of genes that support this theory. Genes comprising this cluster have been implicated in either cell cycle progression or the mechanics of DNA synthesis (Fig. 3).
The majority of the 100 genes spotted on the microarrays with functions implicated in DNA replication/repair were not found to be regulated by estrogen action at the time points examined. To fully assess the expression levels of these genes, the average log2 calibrated ratio for each time point for all of these genes were subjected to hierarchical clustering and visualized using TreeView (Fig. 4) (42). The preponderance of black hue throughout the time course for many of the genes indicates that estrogen treatment did not effect the basal transcript levels for these genes. Most of these genes are part of the machinery designed to repair the genome as opposed to functioning in normal replication. However, many of the genes involved in the DNA replication fork were found to have induced expression at 12, 24, and 36 h, including a few of the genes that were not included in the 95% confidence level list (e.g. replication factor C4 and C5) (enlarged node). Cumulatively, these experiments indicate that the majority of the genes associated with the DNA replication fork are induced after estrogen stimulation, suggesting an important component of the molecular mechanism driving estrogen-induced cell cycle progression.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Collectively, the microarray analyses revealed many known genes and several expressed sequence tags (ESTs) as estrogen responsive, including many that were previously identified as hormone responsive. Interestingly, a large family of genes associated with the DNA replication fork (replication factor C, PCNA, flap structure endonuclease 1, DNA polymerase , DNA polymerase , and DNA ligase) (reviewed in Ref. 44) were found to be significantly up-regulated 12, 24, and 36 h after estrogen stimulation. These findings corroborate the hypothesis that estrogen-induced DNA synthesis results from a complex transcriptional network regulating the expression of not only genes essential for progression through the G1 checkpoint, but also those required for DNA synthesis, and that the induction of these genes may be initiated solely through estrogen action without the confounding or synergistic effects of serum factors.
The proteins associated with DNA replication have been identified and their actions have been characterized and extensively reviewed (reviewed in Refs. 44, 45). Briefly, the process of replicating the genome commences with a localized separation of the two DNA strands. This enables a helicase enzyme to bind the DNA and move unidirectionally along a DNA strand, dissociating the two strands and forming single-stranded templates. To prevent reassociation of the DNA strands before they are replicated, a complex of 3 proteins (collectively referred to as replication factor A) bind to the single-stranded DNA. Replication of the strands requires an RNA primer to be synthesized, which is performed by DNA polymerase -primase. In semidiscontinuous DNA replication, synthesis of the 5' 
3' strand (leading strand) results from elongation of a single primer, whereas the complementary, 3' 5' strand (lagging strand) is replicated from many different primers. Primer elongation begins with the binding of the multisubunit replication factor C to the primer terminus, where it recruits PCNA. PCNA functions as a sliding clamp by anchoring DNA polymerase or to the DNA. These polymerases synthesize the nascent DNA strand with high fidelity due, in part, to their proofreading exonuclease activity. To generate complete DNA strands, the RNA primers are removed by RNase H1, which nicks the 5' side of the 3' ribonucleotide. This generates a topology that is recognized by flap structure endonuclease-1, which removes the 3' ribonucleotide, producing a gap that is filled by DNA polymerase. The resulting nicked double-stranded DNA is then joined by DNA ligase I.
Consistent with the proliferative phenotype of MCF-7 cells treated with estrogen, a large number of DNA replication fork-associated genes were induced by estrogen. Regulated expression was observed for several of the replication factor C subunits, PCNA, DNA polymerase 1 (the catalytic subunit), DNA polymerase , flap structure endonuclease-1, and DNA ligase I. RNase H1 was not present on the microarray chips, so the expression pattern was not determined. Despite previous data that E2 regulates the transcriptional activation of DNA polymerase (46), several components of DNA polymerase -primase complex remain at, or near, basal levels throughout the time course. The apparent discrepancy may be the result of the time points that were selected for analysis as Samudio et al. (46) demonstrated increased expression of DNA polymerase 6 h after estrogen stimulation with near basal levels restored by 12 h. Therefore, the gene may not appear to be regulated in our study as a result of the time points that were selected (provided transcript levels increase between 4 and 6 h). Alternatively, the activity of DNA polymerase may not be regulated at the transcriptional level. Support for this theory comes from the recent work by Schub et al. (47) demonstrating that the polymerase (180 kDa) and regulatory (70 kDa) are phosphorylated in a cell cycle-dependent manner by the S phase cyclin, cyclin A, and its associated kinase cdk2. Cyclin A/cdk2 complex formation and kinase activation in response to estrogen has been demonstrated previously (48). Phosphorylation of p70 subunit of DNA polymerase correlates with maximal DNA polymerase -primase activity, whereas phosphorylation of both p70 and p180 subunits inhibits primer formation. This suggests that estrogen-induced initiation of DNA replication may result from activation of kinase activities rather than a direct transcriptional effect on the subunits comprising DNA polymerase -primase.
A similar scenario may explain the lack of regulated expression of the replication factor A subunits. Estrogen stimulated a slight increase in expression for the smallest subunit (p14), but both the 32- and 70-kDa subunits were unaffected (see Fig. 4 and data not shown). p32 is a substrate for cdk1 when this kinase is complexed with cyclin A or B and is phosphorylated at the end of G1 and through the end of the S phase of the cell cycle (49, 50, 51). Phosphorylated p32 is found in complexes with the other replication factor A subunits and associated with chromatin (51). Therefore, it may be the phosphorylation of p32 rather than the enhanced transcription of replication factor A subunits that enables estrogen to stimulate activity at the DNA replication fork.
While many genes associated with the replication fork are induced transcriptionally as a result of estrogen treatment, this does not necessarily imply that these genes are regulated according to the classical model (ER binding to an ERE). Alternatively, these genes may be transcribed via one of the many transcription factors induced by estrogen (Table 1), the nongenomic actions of estrogen culminating in the activation of a transcription factor, or by increased expression of a protein that has downstream effects on the activation of a transcription factor. Increased activation of the transcription factor E2F-1 has been shown to result from estrogen stimulation (6, 52, 53). (Unfortunately, E2F-1 is not present on the ToxChip version 1.0 microarray chip). As detailed in the Introduction, increased expression of cyclin D1 can result in phosphorylation of Rb, which causes dissociation of E2F from the transcriptionally inactive Rb-E2F complex. Previous studies have demonstrated that E2F’s transcriptional activity results in increased expression of genes involved in DNA synthesis (reviewed in Ref. 54). Using oligonucleotide arrays, Ishida et al. (22) recently demonstrated that PCNA, flap structure endonuclease 1, an EST similar to a replication factor C subunit, and DNA ligase I are transcriptionally responsive to E2F. Additional work by Ren et al. (20) has also implicated replication factor C (activator 1) 3, the catalytic subunit of DNA polymerase delta 1, and minichromosome maintenance (mcm3) as E2F target genes. Searches of the public sequence databases revealed E2F sites in the promoter or other regulatory elements for 4 genes implicated in DNA replication (PCNA, DNA polymerase , mcm3, and mcm7) and found to be estrogen-regulated in the microarray data (55). As the promoters for the remaining genes are defined, it will be interesting to see whether they also are E2F regulated.
The preponderance of E2F-regulated genes in this data set suggests a molecular mechanism by which estrogen is capable of inducing DNA replication. In a recent study (24), Soulez and Parker treated ZR75–1 cells, another hormone-responsive breast cancer cell line, with estrogen in the presence of cyclohexamide, which allowed a means to measure the direct transcriptional effects of estrogen. Cyclin D1 was found to be induced with 6 h; however, no DNA replication fork genes were significantly regulated by estrogen in the presence of the protein synthesis inhibitor, indicating that these transcriptional events may be an indirect or secondary effect of estrogen stimulation. However, examining the kinetics of expression of these genes (Fig. 5) indicates cyclin D1 accumulates before the increased expression of the DNA replication fork genes. Therefore, our data, coupled with the findings of Soulez and Parker, suggest that increased expression of cyclin D1 is a direct effect of estrogen stimulation. Accumulation of cyclin D1 initiates a cascade of reactions, resulting in the liberation of E2F from Rb and enabling transcriptional activation of the DNA replication fork genes (Fig. 6).
|
|
-mediated events (either transcriptional or nongenomic).
In addition to the pathways altered by estrogen action, this study also provides a data source to which future expression profile studies using estrogen can be compared. For example, the tissue-specific pattern of expression resulting from estrogen stimulation, the different transcriptional regulation emanating from the and ß isoforms of the ER. Additionally, the identical series of experiments could be repeated in a different estrogen-responsive breast cancer cell line (e.g. T47D or ZR-751) and compared with this data set to further identify the specific expression changes modulated by estrogen. Alternatively, this data set could also be compared with microarray data examining the expression changes resulting from estrogen mimics or clinically relevant estrogen antagonists (such as tamoxifen and raloxifene) to better understand the mechanisms of action of these compounds and their resulting phenotype.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Proliferation Assay
Proliferation of MCF-7 cells was measured using a modified E-SCREEN assay (59). Briefly, 5 x 104 cells were seeded in 12-well dishes. After sufficient time for cells to adhere (24 h), the cells were washed twice with PBS-CMF and maintained in starvation medium for an additional 24 h. Cells were stimulated (in triplicate) with various concentrations of E2 in starvation medium containing 10% CSS or in the complete absence of serum. Five days after treatment, cells were harvested by trypsinization and relative cell numbers from each treatment condition were obtained using a Coulter counter (Beckman Coulter, Inc., Miami, FL).
Viability Assay
Cell viability was assessed by a dye exclusion technique. Cells were seeded in six-well dishes before treatment with either 10% CSS containing medium or serum-free medium. Detached and attached cells for each condition were pooled and pelleted. Cell pellets were resuspended in a small volume of PBS and stained with erythrosin B (0.075% final concentration). The percentage of cells excluding the dye (viable) was determined by microscopic examination (500 total cells counted in two separate experiments).
RNA Isolation
For microarray hybridizations, approximately 5 x 107 cells were resuspended in 4 ml of Buffer RLT (QIAGEN, Valencia, CA) before lysing using 3x 30-sec bursts of sonication in a Sonic Dismembrator (Fisher Scientific Model 60, Pittsburgh, PA) set at level 8. Total RNA was isolated using QIAGEN RNeasy Midi columns after the manufacturer’s protocol. Eluted RNA was concentrated (
9 µg/µl) using Microcon-30 columns (Millipore Corp., Bedford, MA).
RT-PCR Analysis of ER and ERß Expression
RNA was extracted from 1 x 106 cells using a combination of QiaShredder columns to lyse cells and QIAGEN RNeasy Mini columns to isolate total RNA. cDNA synthesis and amplification of ER, ERß, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) messages was performed using the ProSTAR HF Single Tube RT-PCR System (Stratagene, La Jolla, CA) after the manufacturer’s protocol. All primers were previously described: Burow et al. (36) detailed the ERß and GAPDH primers, whereas Pasquali et al. (60) reported the use of the ER primers.
Real-Time PCR
RNA (2 µg) from all time points and treatment conditions was reverse transcribed 250 U MultiScribe RTase (Applied Biosystems, Foster City, CA) in a 200 µl reaction mix containing 5.5 mM magnesium chloride, 1x PCR Buffer II (Applied Biosystems), 500 µM each deoxynucleotide triphosphate, 2.5 µM random hexamer (Applied Biosystems), 80 U RNase inhibitor (Applied Biosystems). The reaction was incubated at 25 C for 10 min, 48 C for 30 min, and finally 95 C for 5 min. The resulting cDNA (10 µl) was amplified in triplicate using gene specific primers (listed below) using the SYBR Green PCR Core Reagents (Applied Biosystems) after the manufacturer’s recommended protocol. The amplification reaction was performed in an ABI Prism 7700 Sequence Detector (Applied Biosystems) with the following cycling conditions: an initial incubation of 95 C for 10 min followed by 50 cycles of 95 C for 15 sec and 60 C for 1 min. The threshold cycle for each amplification was determined using Sequence Detection System software (Applied Biosystems). The levels of expression for each gene were normalized using 36B4 levels after the comparative threshold cycle method (61, 62).
Northern Blotting
RNA (5 µg) was electrophoresed on a 1% agarose-2.2 M formaldehyde gel then transferred onto a Nytran N membrane (Schleicher \|[amp ]\| Schuell, Inc., Keene, NH). Blots were cross-linked using UV irradiation and hybridized with 32P-labeled gene-specific probes generated by random priming. 36B4 Plasmid (kindly provided by Donald P. McDonnell) was digested with PstI and the 700-bp fragment was used in the labeling reaction. Flap structure-specific endonuclease 1 (Unigene CloneID 49950) was digested with NotI and HindIII and the resulting 1422-bp fragment was used in the labeling reaction. Replication factor C3 (38 kDa) (Unigene CloneID 256260) was digested with NotI and EcoRI and the resulting 1473-bp fragment was used in the labeling reaction). After overnight hybridizations, blots were washed for 10 min at 65 C with 2x SSC (150 mM NaCl and 15 mM Na3C6H3O7, pH 7.0) containing 1% SDS and then with 0.1x SSC/0.1% SDS before exposing a PhosphorImager screen (Molecular Dynamics, Inc., Sunnyvale, CA). IMAGEQuant (Molecular Dynamics, Inc.) was used to quantify the expression levels of the gene of interest and loading inconsistencies were corrected using detected levels of 36B4 expression.
cDNA Microarray
cDNA microarray analysis was conducted on two replicate cultures. A sufficient quantity of total RNA (35 µg for Cy3-labeled probes and 75 µg for Cy5 were used for microarray analysis of the first biological replicate; 30 µg was used for either label during the second replicate) in a volume of 17 µl was combined with 2 µl 500 µg/ml oligo(dT) (12, 13, 14, 15, 16, 18) primer (Amersham Pharmacia Biotech, Piscataway, NJ) and 1 µl RNasin (10 U/µl) (Invitrogen, Carlsbad, CA) before heating for 10 min at 70 C. After chilling on ice for 2 min, 9 µl 5x buffer (250 mM Tris-HCl, pH 8.3; 375 mM KCl; 15 mM MgCl2) (Invitrogen), 5 µl 0.1 M dithiothreitol (Invitrogen), 4 µl 25 nM FluoroLink Cy3-deoxyuracil triphosphate or Cy5-deoxyuracil triphosphate (Amersham Pharmacia Biotech), 1.2 µl [25 mM deoxy (d)-ATP, 25 mM dGTP, 25 mM dCTP, 15 mM deoxythymidine triphosphate] deoxynucleotide triphosphate mix (Amersham Pharmacia Biotech), and 2 µl SuperScript II Reverse Transcriptase (Invitrogen). Samples were incubated at 42 C for 1.5 h prior to adding an additional 2 µl of SuperScript II Reverse Transcriptase and incubating another 1.5 h. After cDNA synthesis, 30 µl of 0.1 M NaOH was added and incubated at 70 C for 30 min to degrade the RNA. The pH was then returned to neutral by adding 30 µl of 0.1 M HCl. Cy3- and Cy5-labeled probes were pooled and unincorporated label was removed by using a Microcon-30 filter (Millipore Corp.). Before recovering the probes 10 µg/µg of RNA of human COT1 DNA (Invitrogen), and 20 µg of yeast tRNA (Invitrogen) were added to limit nonspecific binding of the probe during the hybridization. Probes were added to a hybridization solution (3x SSC, 2x Denhardt’s, and 0.3x SSC), boiled for 2 min, and purified through a 0.45 µM filter (Millipore Corp.). The purified solution was then applied to a glass cDNA microarray slide, covered with a coverslip, and incubated for 16 h in a humidified chamber at 65 C. These custom cDNA chips (ToxChip version 1.0) contain 1901 known human genes and ESTs, categorized based on their cellular function: apoptosis, cell cycle control, DNA replication and repair, heat shock proteins, oncogenes and tumor suppressors, kinases, phosphatases, transcription factors, etc. (reviewed in Ref. 63). (The database of genes present on ToxChip version 1.0 can be searched at http://dir.niehs.nih.gov/microarray/chips.htm). Slides were inverted in 0.5x SSC, 0.01% SDS for 3 min in order to remove the coverslip passively and minimize spot damage. The slide was washed for an additional 3 min in a fresh 0.5x SSC, 0.01% SDS. The slide was then washed twice with 0.06x SSC for 3 min before spinning for 3 min at 1000 rpm. A GenePix 4000A-microarray scanner (Axon Instruments, Inc., Union City, CA) was used to scan and generate image files of the arrays. Each RNA was hybridized to four arrays, two with each dye orientation. Because each time point was also biologically replicated a total of eight arrays for each time point was analyzed.
Data Analysis
Signal intensities from the image files were quantified and normalized using IPLabs image processing software (Scanalytics, Inc., Fairfax, VA) with the ArraySuite extensions (National Human Genome Research Institute) that are based on the algorithm previously described by Chen et al. (40). This software package identifies genes that demonstrate statistically significant expression changes for a user-defined confidence level. Genes identified as being up- or down-regulated at the 95% confidence level were stored in Microarray Project Systems, a database that is used to manage and interpret gene expression data (41). (For information pertaining to the fold-cutoff that was used for all 60 hybridizations, see Table 5 in the supplemental data.) To increase the statistical confidence in the data, only the genes found to be altered at the 95% confidence level in at least four out of the eight hybridizations for a given time point were considered for further analysis. Furthermore, the average fold-induction of all eight hybridizations for these genes needed to exceed 1.3 to be included for further analysis. To identify genes in this list that demonstrated a bias for a particular fluorophore or highly variable expression, the coefficient of variation (CV) for each gene was calculated. The CV (SD/absolute value of the calibrated ratio) was computed using the log2 ratio intensity values of the genes detected as differentially expressed at a given confidence level. Genes with a CV value greater than 0.43 were eliminated from the list before clustering. (For information on the level of induction and repression, and at which time points the changes were statistically significant, see Table 4 in the supplemental data.)
Correlation Analysis
To assess the stability of gene expression for the vehicle-control treated cells, the pixel intensity values for the control samples were compiled from each array. After transforming to the log2 scale, each value was standardized using the mean and standard deviation of all log intensity values on that array using JMP (SAS Institute, Inc., Cary, NC). Averaging the standardized log intensities for replicates gave a 1900 x 6 matrix of values, where the rows represent all genes on the array and the columns represent the six time points. The similarity between the time points was then measured using Pearson correlation. The same procedure was carried out for the smaller set of genes that were identified as differentially expressed.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication January 3, 2002. Accepted for publication March 12, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
receptor chimeras. Endocrinology 139:4513–4522 and ER ß. J Steroid Biochem Mol Biol 69:165–175[CrossRef][Medline]
, ß expression and signaling correlate with apoptotic sensitivity of MCF-7 breast cancer cell variants. Int J Oncol 16:1179–1187[Medline]
gene expression in MCF-7 cells by 17ß-estradiol. Endocrinology 142:1000–1008-primase cooperate to regulate the initiation of DNA replication in vitro. J Biol Chem 276:38076–38083This article has been cited by other articles:
 |
|
 |
|
  F. Wu, I. Ivanov, R. Xu, and S. Safe Role of SP transcription factors in hormone-dependent modulation of genes in MCF-7 breast cancer cells: microarray and RNA interference studies J. Mol. Endocrinol., January 1, 2009; 42(1): 19 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Dip, S. Lenz, J.-P. Antignac, B. Le Bizec, H. Gmuender, and H. Naegeli Global gene expression profiles induced by phytoestrogens in human breast cancer cells Endocr. Relat. Cancer, March 1, 2008; 15(1): 161 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Higgins, S. Liu, M. Abdelrahim, K. Vanderlaag, X. Liu, W. Porter, R. Metz, and S. Safe Vascular Endothelial Growth Factor Receptor-2 Expression Is Down-Regulated by 17{beta}-Estradiol in MCF-7 Breast Cancer Cells by Estrogen Receptor {alpha}/Sp Proteins Mol. Endocrinol., February 1, 2008; 22(2): 388 - 402. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Shioda, J. Chesnes, K. R. Coser, L. Zou, J. Hur, K. L. Dean, C. Sonnenschein, A. M. Soto, and K. J. Isselbacher Importance of dosage standardization for interpreting transcriptomal signature profiles: Evidence from studies of xenoestrogens PNAS, August 8, 2006; 103(32): 12033 - 12038. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Buterin, C. Koch, and H. Naegeli Convergent transcriptional profiles induced by endogenous estrogen and distinct xenoestrogens in breast cancer cells Carcinogenesis, August 1, 2006; 27(8): 1567 - 1578. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. E. Harvell, J. K. Richer, D. C. Allred, C. A. Sartorius, and K. B. Horwitz Estradiol Regulates Different Genes in Human Breast Tumor Xenografts Compared with the Identical Cells in Culture Endocrinology, February 1, 2006; 147(2): 700 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Frasor, J. M. Danes, C. C. Funk, and B. S. Katzenellenbogen Estrogen down-regulation of the corepressor N-CoR: Mechanism and implications for estrogen derepression of N-CoR-regulated genes PNAS, September 13, 2005; 102(37): 13153 - 13157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Laganiere, G. Deblois, C. Lefebvre, A. R. Bataille, F. Robert, and V. Giguere From the Cover: Location analysis of estrogen receptor {alpha} target promoters reveals that FOXA1 defines a domain of the estrogen response PNAS, August 16, 2005; 102(33): 11651 - 11656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Laganiere, G. Deblois, and V. Giguere Functional Genomics Identifies a Mechanism for Estrogen Activation of the Retinoic Acid Receptor {alpha}1 Gene in Breast Cancer Cells Mol. Endocrinol., June 1, 2005; 19(6): 1584 - 1592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.D. Swartz, C.A. Afshari, L. Yu, K.E. Hall, and D. Dixon Estrogen-induced changes in IGF-I, Myb family and MAP kinase pathway genes in human uterine leiomyoma and normal uterine smooth muscle cell lines Mol. Hum. Reprod., June 1, 2005; 11(6): 441 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J G Moggs, T C Murphy, F L Lim, D J Moore, R Stuckey, K Antrobus, I Kimber, and G Orphanides Anti-proliferative effect of estrogen in breast cancer cells that re-express ER{alpha} is mediated by aberrant regulation of cell cycle genes J. Mol. Endocrinol., April 1, 2005; 34(2): 535 - 551. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. X. Jin, Y.-W. Leu, S. Liyanarachchi, H. Sun, M. Fan, K. P. Nephew, T. H.-M. Huang, and R. V. Davuluri Identifying estrogen receptor {alpha} target genes using integrated computational genomics and chromatin immunoprecipitation microarray Nucleic Acids Res., December 17, 2004; 32(22): 6627 - 6635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Tang, S. L. Tan, S. K. Ramadoss, A. P. Kumar, M.-H. E. Tang, and V. B. Bajic Computational method for discovery of estrogen responsive genes Nucleic Acids Res., December 1, 2004; 32(21): 6212 - 6217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Eward, E. C. Obermann, S. Shreeram, M. Loddo, T. Fanshawe, C. Williams, H.-I. Jung, A. T. Prevost, J. J. Blow, K. Stoeber, et al. DNA replication licensing in somatic and germ cells J. Cell Sci., November 15, 2004; 117(24): 5875 - 5886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Bourdeau, J. Deschenes, R. Metivier, Y. Nagai, D. Nguyen, N. Bretschneider, F. Gannon, J. H. White, and S. Mader Genome-Wide Identification of High-Affinity Estrogen Response Elements in Human and Mouse Mol. Endocrinol., June 1, 2004; 18(6): 1411 - 1427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. E. Cunliffe, M. Ringner, S. Bilke, R. L. Walker, J. M. Cheung, Y. Chen, and P. S. Meltzer The Gene Expression Response of Breast Cancer to Growth Regulators: Patterns and Correlation with Tumor Expression Profiles Cancer Res., November 1, 2003; 63(21): 7158 - 7166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Liu, Z. Zhang, W. Gladwell, and C. T. Teng Estrogen Stimulates Estrogen-Related Receptor {alpha} Gene Expression through Conserved Hormone Response Elements Endocrinology, November 1, 2003; 144(11): 4894 - 4904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. RHEN, S. GRISSOM, C. AFSHARI, and J. A. CIDLOWSKI Dexamethasone blocks the rapid biological effects of 17{beta}-estradiol in the rat uterus without antagonizing its global genomic actions FASEB J, October 1, 2003; 17(13): 1849 - 1870. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Frasor, J. M. Danes, B. Komm, K. C. N. Chang, C. R. Lyttle, and B. S. Katzenellenbogen Profiling of Estrogen Up- and Down-Regulated Gene Expression in Human Breast Cancer Cells: Insights into Gene Networks and Pathways Underlying Estrogenic Control of Proliferation and Cell Phenotype Endocrinology, October 1, 2003; 144(10): 4562 - 4574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. K. Kinyamu and T. K. Archer Estrogen Receptor-Dependent Proteasomal Degradation of the Glucocorticoid Receptor Is Coupled to an Increase in Mdm2 Protein Expression Mol. Cell. Biol., August 15, 2003; 23(16): 5867 - 5881. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) or complete serum starvation (No Serum, 
). To assess the mitogenic potential of estrogen in these three different serum conditions, cells were then stimulated with various concentrations of E2 for 5 d. Total cell counts were assessed using a Coulter counter and are represented as an average of three independent samples. (The physiological range of total E2 in women is between 50 to 350 pg/ml (1.8 x 10-10 M to 1 x 10-9 M). B, To assess viability, MCF-7 cells were cultured in either complete medium containing 10% FCS (Log), phenol red free medium containing 10% CSS, or phenol red free medium without any serum additions (No Serum). After the indicated period of time, detached and attached cells for each condition were pooled and pelleted. Cell pellets were resuspended in a small volume of PBS and stained with erythrosin B. The percentage of cells excluding the dye (viable) was determined by microscopic examination (500 total cells counted in two separate experiments).
