Molecular Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Chemical and Biological Aspects of Inflammation and Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Cell Growth & Differentiation
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Itoh, T.
Right arrow Articles by Chen, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Itoh, T.
Right arrow Articles by Chen, S.
Molecular Cancer Research 3:203-218 (2005)
© 2005 American Association for Cancer Research

Cancer Genes and Genomics

Letrozole-, Anastrozole-, and Tamoxifen-Responsive Genes in MCF-7aro Cells: A Microarray Approach

Toru Itoh1, Kim Karlsberg2, Ikuko Kijima1, Yate-Ching Yuan2, David Smith2, Jingjing Ye1 and Shiuan Chen1

1 Department of Surgical Research and 2 Division of Informational Sciences, Beckman Research Institute of the City of Hope, Duarte, California

Requests for reprints: Shiuan Chen, Department of Surgical Research, Beckman Research Institute of the City of Hope, 1450 East Duarte Road, Duarte, CA 91010. Phone: 626-359-8111, ext. 63454; Fax: 626-301-8972. E-mail: schen{at}coh.org


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Antiestrogens and aromatase inhibitors are important drugs in the treatment of estrogen-dependent breast cancer. To investigate the effects of these drugs on gene expression in breast cancer cells, we treated estrogen receptor–positive MCF-7 cells stably transfected with the aromatase gene (known as MCF-7aro cells) with testosterone, 17ß-estradiol, two aromatase inhibitors (letrozole and anastrozole), and an antiestrogen (tamoxifen). We found that testosterone or 17ß-estradiol induced the proliferation of MCF-7aro cells at a rate six times faster than the untreated cells. In addition, the testosterone-induced proliferation of MCF-7aro cells was effectively suppressed by letrozole, anastrozole, or tamoxifen. Microarray analyses on Affymetrix Human Genome U133A GeneChips (Affymetrix, Santa Clara, CA) were carried out using total RNA isolated from the control and treated cells. At the false discovery rate of 0.05 and a minimum fold-change criteria of 1.5, 104 genes were identified that were up-regulated and 109 genes were identified that were down-regulated by both androgen and estrogen. More than 50% of these hormone-regulated genes were counterregulated by all three inhibitors and >90% were counterregulated by at least one of the inhibitors. Comparing the effect of each inhibitor on gene expression, we observed that letrozole and anastrozole are more similar in terms of the genes they affect compared with treatment with tamoxifen. To validate the gene expression profiles identified from microarray analyses, the expression patterns of 13 representative genes were examined by Northern analysis. Finally, the genes identified as statistically significant were classified based on their expression patterns and biological function/pathways. The results of this study provide us with a better understanding of the actions of both aromatase inhibitors and antiestrogens at the molecular level. We believe that the results of this study serve as the first step in identifying unique expression patterns following drug treatment, and that this will ultimately be useful in customizing patient treatment strategies for hormone-dependent breast cancer.


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Estrogens play an important role in breast cancer development. About 60% of premenopausal and 75% of postmenopausal patients have estrogen-dependent carcinomas. The action of estrogens is mediated by estrogen receptors (ER), and the estrogen-bound ER induces the expression of peptide growth factors that are responsible for the cancer cell proliferation (1, 2). In estrogen-dependent breast tumors, control of tumor growth can be achieved by treatment with antiestrogens, which are antagonists that block the binding of estrogens to ER. Indeed, clinical trials using the antiestrogen tamoxifen proved the importance of estrogens in breast tumor development and identified tamoxifen as a breast cancer preventive agent (3, 4).

Aromatase, a cytochrome P450, is the enzyme that synthesizes estrogens by converting C19 androgens to aromatic C18 estrogenic steroids. The expression of aromatase in breast cancer tissues has been shown by enzyme activity measurement (5, 6), immunocytochemistry (7-10), and reverse transcription-PCR analyses (11-13). Cell culture (14, 15) and nude mouse experiments (16) using aromatase-transfected MCF-7 cells showed, in a direct fashion, that aromatase expressed in breast cancer cells plays a role in stimulating the growth of tumors in both an autocrine and a paracrine manner. Tekmal et al. (17) have produced transgenic mice that overexpress int-5/CYP19 under the control of the mouse mammary tumor virus enhancer/promoter. Overexpression of int-5/CYP19 in involuted mammary glands of transgenic females induces hyperplasia in 75% to 80% of ducts and glands that exhibit a range of morphologic abnormalities. The results further indicate that in situ produced estrogen plays a more important role than circulating estradiol in breast tumor promotion. Aromatase inhibitors have been found to be valuable in treating these estrogen-dependent and aromatase-mediated diseases, including breast cancer (18). The Food and Drug Administration has approved the third-generation nonsteroidal aromatase inhibitors letrozole and anastrozole for use as first-line agents against estrogen-dependent cancer. As first-line therapy in postmenopausal women with advanced or metastatic breast cancers, letrozole is more active than tamoxifen, achieving higher response rates, longer time to progression, and early survival advantage while treatment is well tolerated (19). In the same setting, anastrozole and tamoxifen have had at least equivalent activity and tolerability (20-22). A randomized trial of letrozole in postmenopausal women after 5 years of tamoxifen therapy for early-stage breast cancer has been found to significantly improve disease-free survival (23). These studies reveal that aromatase inhibitors are important new drugs in the treatment of hormone-dependent breast cancer.

Whereas aromatase inhibitors, such as letrozole and anastrozole, provide similar clinical efficacy, the observed pharmacologic profiles suggest differences in their molecular actions (24). To better understand the action of tamoxifen and the two aromatase inhibitors, letrozole and anastrozole, at the molecular level, we have carried out a comprehensive gene expression analysis of the effects of testosterone, 17ß-estradiol, letrozole, anastrozole, and tamoxifen on aromatase-overexpressing ER-positive breast cancer cells, MCF-7aro. Cells in our study were cultured in the presence of hormone and/or inhibitor for 1 week because we wanted to model the situation of long-term hormonal therapy in breast cancer patients. The quantitative evaluation of all cellular mRNA populations was done using microarrays. Each set of experiments, consisting of six different treatments, was carried out independently three times, and then statistically combined together. This is the first report that shows the differences and similarities of the effects of letrozole, anastrozole, and tamoxifen on comprehensive gene expression.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Acknowledgements
 References
 
Experimental Design
The aromatase-overexpressing ER-positive breast cancer cell line, MCF-7aro, has been established in our laboratory as described previously (15). Aromatase activity in the MCF-7aro cell line was determined to be 73 ± 6 pmol/mg/h. Because all common breast cancer cell lines used by breast cancer researchers have very low measurable aromatase activity, MCF-7aro has been a valuable tool to evaluate growth responses to aromatase inhibitors. A significant number of in vitro and in vivo experiments using this cell line have been carried out in a few laboratories (25). Therefore, the information generated from the present study will be important for the understanding of the molecular basis of the findings generated from studies using this cell line.

In most of the recently published studies about the effects of hormones or other chemicals on gene expression, cells were cultured in the presence of chemicals for only a few hours to a couple of days before expression analysis (26, 27). These studies have identified early responsive genes whose expression may or may not remain up-regulated after a longer period of treatment. The cells in our study were exposed to chemicals for 1 week because we wanted to identify genes whose expression are modulated under conditions that model breast cancer patients who are treated with medication for a long term. We treated the cells with two steroid hormones: 1 nmol/L testosterone (androgen) or 1 nmol/L 17ß-estradiol (estrogen). Androgen is converted to estrogen by aromatase in MCF-7aro cells, so we expected to see similar effects for the androgen and estrogen treatments. We also cultured the cells with three different inhibitors, two aromatase inhibitors (letrozole and anastrozole) and one ER antagonist (tamoxifen), in the presence of testosterone.

Effects of Hormones and Inhibitors on Cell Proliferation
MCF-7aro cells were cultured in phenol red–free culture medium containing 10% charcoal/dextran-treated fetal bovine serum for 3 days to eliminate the estrogen-like effect of phenol red and the influence of steroid hormones in regular fetal bovine serum. The cells were then cultured with inhibitors and/or hormones for 1 week, followed by the determination of total protein amount, as an indication of cell proliferation (Fig. 1). Under this restricted culture condition (Fig. 1, vehicle), the cells in the vehicle group showed a moderate rate of proliferation. On the other hand, the cells cultured with hormones (Fig. 1, testosterone and 17ß-estradiol) increased more than six times in comparison with vehicle-treated cells. Furthermore, the hormone-treated cells grew at their log phase on day 10, indicating that the cells were still responding well to treatment with estrogen or androgen. In spite of the presence of androgen, the growth curve of letrozole-treated cells was almost identical to that of vehicle-treated cells (Fig. 1, androgen + letrozole), and anastrozole treatment increased the cell number only 2-fold more than vehicle (Fig. 1, androgen + anastrozole). This implies that both chemicals acted as aromatase inhibitors, resulting in very low levels of estrogen and cell proliferation. On the other hand, although both letrozole and anastrozole are aromatase inhibitors, their potencies seem to be different. In terms of the suppression on cell proliferation, letrozole (at 200 nnomol/L) seems to be more efficient than anastrozole (at 1 µmol/L). Culturing the cells with tamoxifen (at 1 µmol/L) also showed a low level of cell growth (Fig. 1, androgen + tamoxifen). Because tamoxifen is an antiestrogen, the estrogen-ER complex could not be formed and failed to lead to cell proliferation. As the treatment controls, MCF-7aro cells were also cultured with aromatase inhibitors or tamoxifen but not testosterone. The cells in these treatment groups showed a low rate of proliferation (data not shown).



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 1. The effect of vehicle, 1 nmol/L testosterone (androgen, A), 1 nmol/L 17 ß-estradiol, androgen + 200 nmol/L letrozole, androgen + 1 µmol/L anastrozole, and androgen + 1 µmol/L tamoxifen on the growth of MCF7/aro cells. Cells were plated on six-well plates and cultured in phenol red–free culture medium including 10% of charcoal/dextran–treated fetal bovine serum for 3 days. Triplicate wells were then treated with the drugs and/or hormones. The cells were harvested with 0.5 N NaOH on days 4, 7, and 10. Points, mean of total protein; bars, SD.

 
Microarray Data Analysis and Identification of Estrogen-Regulated Genes
To investigate the effects of hormones and inhibitors/tamoxifen, gene expression profiling with microarrays was carried out using total RNA from MCF-7aro cells treated with inhibitors/tamoxifen and/or hormones as described above. The data from three sets of six experiments were processed and combined into one as described in Materials and Methods. A flow chart outlining the statistical methods used for analyzing the expression data is shown in Fig. 2. For each comparison between treatments, we calculated the t test statistic by averaging intensities across the three replicates per treatment. The relative intensities of samples treated with vehicle and hormones and those of inhibitors/tamoxifen were calculated against vehicle and androgen, respectively.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 2. Statistical processing of microarray data. The flow chart outlines the statistical methods used for analyzing the expression data.

 
Once background correction and normalization were complete, we began filtering out unwanted and nonsignificant genes. First, we removed 8,869 of the 22,283 genes whose intensities (in one or more of the replicates) were near the level of noise (as described in Materials and Methods). Of the remaining 13,414 genes, 6,549 were up-regulated and 6,865 were down-regulated by estrogen compared with vehicle. Based on our literature search, we found that other laboratories have reported 189 genes induced by estrogen and 197 genes suppressed by estrogen in MCF-7 and its derivatives (26-37). We have learned from a review of the literature that the number of overlapping genes from different reported studies are very low. We compared the collective list from the literature with our list at this stage and found a high degree of similarity. We found that 69.8% of the published up-regulated genes and 67.5% of the down-regulated genes overlapped with our results. The overlapping estrogen up-regulated genes included BRCA1, cyclin D1, IGF-binding protein 4, PDZK1, and TFF1 (pS2), whereas the overlapping estrogen down-regulated genes included BAK1, TGFß3, ERBB2, and ERBB3. It is important to note that we have also been able to identify several estrogen-responsive genes that these other groups did not identify in their microarray studies, including the estrogen up-regulated genes FHL1, H41, and NPY1R and the estrogen down-regulated genes ER{alpha}, NFKBIA, and SH3BGRL. On the other hand, several known estrogen up-regulated genes (such as CBFA2T3, E2IG5, GADD45B, HOXC4, or VEGF) and estrogen down-regulated genes [such as B4-2 (PNRC2), CRADD, E-cadherin, LDLR, RCN2, or SATB1] were present in the published gene list, but not in our list. These differences are likely in part due to variations in experimental conditions. In our study, the cells were treated with aromatase inhibitors or tamoxifen in the presence of androgen, whereas in previous studies the experiments were done in the absence of androgen. Furthermore, because we treated the cells for a longer period of time (1 week compared with only ≤3 days in the other studies), some of the early responsive genes in the published list were not found in our list. The differences could also be due to different criteria of gene selection (e.g., signal intensity and fold change cutoffs). For example, we have found that a few genes with low signal intensity in the vehicle display high fold changes for the treatment samples. Because of low signal intensity, we questioned the significance of such results and, therefore, these genes were excluded from our discussion. In addition, our results were generated from three biological replicates, which is different from several reported analyses that were based on data generated from one or two experiments.

Identification of Inhibitor-Responsive Genes
In the next step of our analyses, we focused on inhibitor/tamoxifen-responsive genes. We define inhibitor/tamoxifen-responsive genes as those genes that were up- or down-regulated by androgen and estrogen (at least 1.5-fold regulation compared with vehicle), and counterregulated by at least one of the three of the drugs (compared with the androgen treated sample, no fold change criteria applied). We note that our fold change cutoff of 1.5-fold was specified prospectively as a threshold for detecting genes that were both potentially interesting clinically as well as not being too restrictive for the expression fold changes that we anticipated. This differs from other authors' fold changes that have been proposed in similar experiments. However, our statistical analysis incorporated a measure of general reproducibility and was not solely based on the magnitude of average fold change.

Figure 3A is the result of two-dimensional hierarchical clustering analysis done with the inhibitor/tamoxifen-responsive genes using Gene Tree and Condition Tree clustering within GeneSpring. Each column represents a single gene, with up-regulated genes in red, down-regulated genes in green, and genes not different from control (vehicle for hormones, androgen for inhibitors) in yellow. The dendrogram on the top of this matrix represents similarities in the gene expression patterns as a "gene tree." Each row represents a single experimental sample. The similarities in the expression pattern among experiments are represented as a "condition tree" on the left side of the matrix. A total of 203 genes were clustered into two large groups, hormone up-regulated genes and down-regulated genes, by comparing their levels to those of the vehicle control sample. The expression levels of the 203 genes in the drug-treated samples (testosterone + letrozole, testosterone + anastrozole, or testosterone + tamoxifen) were then compared with those in the sample treated with testosterone alone. On the condition tree, letrozole-responsive genes and anastrozole-responsive genes were clustered together, whereas tamoxifen-responsive genes were clustered on a separate branch. This suggests that the expression patterns of letrozole- and anastrozole-responsive genes are more similar than those of tamoxifen-responsive genes. These results are reasonable if we consider that letrozole and anastrozole are both aromatase inhibitors that inhibit estrogen synthesis, whereas tamoxifen is an ER antagonist that inhibits the formation of the estrogen-ER complex. To further understand the inhibitor/tamoxifen-responsive genes, functional grouping was done. Using various online databases (Affymetrix NetAffx, National Center for Biotechnology Information database, and AmiGo), the inhibitor/tamoxifen-responsive genes with known functions were classified into 16 functional categories (Tables 1 and 2). Upon closer inspection, we found several interesting genes that are regulated differently in letrozole and anastrozole versus tamoxifen. For example, stanniocalcin 1 (STC1), a potential calcium/phosphate-regulating hormone (38), is down-regulated by both hormones and counterregulated by letrozole and anastrozole but not by tamoxifen. NAD(P)H dehydrogenase, quinone 1 (NQO1), which is involved in xenobiotic metabolism, behaves in a similar fashion. On the other hand, SULF1, a sulfatase involved in steroid metabolism, is up-regulated by both hormones and counterregulated by letrozole and anastrozole, but not by tamoxifen. It is possible that some of the genes modulated only by aromatase inhibitors are regulated by estrogen in an ER-independent manner. The expression profiles of several genes in this category have been verified by Northern analysis (discussed below).



View larger version (57K):
[in this window]
[in a new window]
  		FIGURE 3. Inhibitor-responsive genes. A. Two-dimensional hierarchical clustering of inhibitor-responsive genes. MCF7/aro cells were treated with inhibitor and/or hormone (indicated on the right side) and expression data from them were graphically represented as normalized intensities. Each row and column represent drug used for treatment and single gene, respectively. As indicated by color bar, up-regulated genes were in red, down-regulated genes are in green, and genes that did not show any changes compared with control are in yellow. The Gene Tree and the Condition Tree show the similarities in the expression pattern among genes and experiments, respectively. B. Venn diagrams showing the numbers of genes responsive to individual inhibitors in hormone-regulated genes.

 

View this table:
[in this window]
[in a new window]
  		Table 1. The Functional Grouping of Hormone Up-Regulated and Inhibitor(s) Down-Regulated Genes

 

View this table:
[in this window]
[in a new window]
  		Table 2. The Functional Grouping of Hormone Down-Regulated and Inhibitor(s) Up-Regulated Genes

 
Figure 3B shows the number of genes that were affected by individual drugs. The left panel shows the inhibitor/tamoxifen-responsive genes in hormone down-regulated genes. The total number of hormone down-regulated genes was 109, and 102 of these genes were up-regulated by at least one of the drugs. The right panel shows the inhibitor/tamoxifen-responsive, hormone up-regulated genes. A total of 104 genes were up-regulated by hormones, and 102 of these genes were found to be inversely regulated by at least one of the drugs. Interestingly, all or most of the anastrozole-regulated genes were also regulated by letrozole. This supports our observation in Fig. 3A that letrozole and anastrozole have very similar effects on gene expression.

We used a set of strong, robust statistical techniques to analyze our data to reduce possible sources of bias and consequently reduce the probability of false-positive genes in our list. At each step of our analysis, our statistical methods were tailored to disregard genes that were most likely to be noninformative. Our focus was to derive a subset of high-quality, highly expressed genes with the greatest potential to be implicated in our hypothesized cellular processes. Although it is possible that our methods tended to err on excluding genes, we believe that our list of genes will likely be fruitful for further research. To validate our methodology and results, we chose 13 genes to confirm via Northern analysis. This validation set consisted of one known gene (PDZK1) as a control and 12 novel genes (ASCL1, FHL1, H41, Hep27, JAK2, MAOB, MCCC2, NPY1R, PAM, RAI3, SH3BGRL, and STC1). We prefer Northern blot analysis to quantitative PCR because it can provide information regarding the relative expression levels as well as the size of the RNA messages. As shown in Fig. 4, the Northern blot results agreed well with microarray data. Thus, we were able to confirm that our methodology is valid and our results are reliable. Achaete-scute homologue 1 (ASCL1), a potential cell cycle regulator (39), was included because its expression was found to be selectively up-regulated in the presence of androgen plus tamoxifen. Tamoxifen treatment without androgen did not induce the expression of ASCL1 (data not shown). Furthermore, the Northern experiments have confirmed that the expression of Hep27 and MAOB can be significantly up-regulated in the presence of androgen plus letrozole or anastrozole. In addition, using Northern analysis, we confirmed that MCCC2, RAI3, and STC1 were down-regulated by androgen/estrogen and counterregulated by aromatase inhibitors, but not by tamoxifen.



View larger version (66K):
[in this window]
[in a new window]
  		FIGURE 4. Northern blot analyses of hormone/inhibitor–regulated genes. In each panel, the image of signals on X-ray film is shown on the top. The numbers below each panel represent the fold changes determined from the microarray analysis. Veh, vehicle; And, androgen; Est, estrogen; Let, androgen + Letrozole; Ana, androgen + anastrozole; Tam, androgen + tamoxifen. The equal loading of RNAs has been verified by hybridization using the glyceraldehyde-3-phosphate dehydrogenase probe (not shown).

 
Effects of Inhibitors on Gene Expression
We have made several attempts to prepare cells treated with inhibitor only for the identification of the genes that can be regulated by each inhibitor individually. It was technically difficult to produce enough cells for the analysis after a 10-day treatment because these cells did not proliferate well in the absence of androgen or estrogen. We have used our current data to extrapolate the effects of inhibitor treatment alone. We began by selecting those genes with signal intensities larger than 120, and whose expression is 1.5-fold up-regulated or down-regulated by the treatment of inhibitors plus androgen compared with the vehicle sample. We then identified a subset of those genes whose expression in both hormone-treated samples is unchanged compared with the vehicle sample (within 10%). Forty-five genes met these criteria (Table 3). This subset of genes represents the inhibitor-only effects. Whereas we observe that more modulated genes overlap between letrozole and anastrozole than between either aromatase inhibitor with tamoxifen, a good percentage of genes in this category are modified similarly by all three compounds. The majority of the 45 genes are classified as genes involved in signal transduction and transcription processes. However, the functional relationship among these genes is not easily seen.


View this table:
[in this window]
[in a new window]
  		Table 3. The List of Aromatase Inhibitors/Tamoxifen–Modulated Genes

 
Gene Expression Patterns on Pathways
The results shown thus far indicate that there are subtle differences in gene expression profiles in cells treated with tamoxifen, letrozole, or anastrozole. To better understand our results, we have examined the gene expression profiles of three pathways related to apoptosis and discuss the results below.

As shown in Fig. 5, we analyzed our entire data set in the context of three biological pathways using the GeneSpring software. The effects of letrozole, anastrozole, and tamoxifen are indicated by red, blue, and green arrowheads, respectively, and the direction of arrowheads indicates up-regulation or down-regulation or no effect in comparison with androgen. Because treatment with the aromatase inhibitors or tamoxifen all led to the suppression of androgen-dependent proliferation of MCF-7aro cells, changes in the expression of genes in the cell death/apoptosis pathway would be expected. We decided to examine the cell death/apoptosis pathway because induction of apoptosis by aromatase inhibitor treatment of MCF-7aro cells has been reported by Thiantanawat et al. (40). The published data would help us to evaluate the quality of the results generated from this microarray study. Figure 5A shows the apoptosis pathway from GenMAPP (Gene MicroArray Pathway Profiler, University of California, San Francisco, San Francisco, CA; www.genmapp.org). The group of genes labeled "A" (JUN, NFKBIA) includes apoptosis-inducing genes, which were down-regulated by androgen and up-regulated by all three compounds. The genes labeled "C" (BCL2 and BAX) are involved in the mitochondria-mediated apoptosis process. BCL2 is an antiapoptotic gene that was up-regulated by androgen and down-regulated by three drugs. BAX blocks the action of BCL2. Letrozole treatment or tamoxifen treatment induced the expression of BAX, thus suppressing the antiapoptotic action of BCL2. Therefore, the cells treated with inhibitors had higher opportunity to undergo apoptosis compared with androgen-treated cells. However, group B (MDM2, MCL1) genes, which have inhibitory effects on p53-mediated apoptosis, were also down-regulated by androgen and up-regulated by inhibitors. This suggests that treatment with aromatase inhibitors or tamoxifen leads to down-regulation of the p53-mediated apoptosis process. Expression of several other genes (PRF1, TRADD, MAP3K1, MAP2K4, and CASP8) was also affected by the treatment of three drugs, but modified differently by different drugs. In addition, as indicated in Fig. 5A, a number of other proteins are involved in the apoptosis process but their expressions were not affected by the hormones and three drugs.



View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 5. Diagrams of three GenMAPP pathways. A. Apoptosis pathway, B. MAPK cascade. C. Wnt signaling pathway. Colored boxes, genes that were found to have sufficient intensities in the present study. The color of box was determined based on the expression level in androgen samples as shown in Fig. 3A. Arrowheads, letrozole (red), anastrozole (blue), and tamoxifen (green); their directions show the regulation.

 
The mitogen-activated protein kinase (MAPK) cascade, a major signal transduction pathway, mediates apoptosis as well as many other physiologic processes, and is shown in Fig. 5B. Group A (RRAS, MAP3K1) contains Ras/Raf-related genes that mediate this cascade, whereas the genes in group B (MAP2K3, MAP2K4, JUN) mediate apoptosis. These genes in groups A and B were down-regulated by androgen, resulting in decreased activation of these processes; however, in most cases, they were up-regulated by inhibitors, thus augmenting the activation of the MAPK cascade and apoptosis. Group C (MAP2K1, MAP2K2, MAPK3) contains the genes that are located downstream of the Ras/Raf pathway. These genes are down-regulated by androgen. On the other hand, group D genes (MBP, ELK1), which are downstream of group C, were up-regulated by androgen and facilitate cell growth. These results indicate that a single treatment can change the expression of a number of genes in a functional pathway and in different manner. Microarray analysis allows us the unique opportunity to study overall gene expression profiles for thousands of genes and the signature of each treatment within the context of various biological pathways.

Figure 5C shows the Wnt signaling pathway, another important signal transduction pathway. Group A includes PKC genes that were not affected by androgen but were up-regulated by three drugs. The PKC family of proteins are involved in many pathways, including apoptosis. The group B genes (MAPK9, ARHA, RAC1) are also apoptosis-related genes and are up-regulated by inhibitors/tamoxifen. The genes in group C (CTNNB1, APC, CCND3, FOSL1, JUN) are ß-catenin and related genes. The expression of these genes were up-regulated by the treatment of inhibitors/tamoxifen.


   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Acknowledgements
 References
 
Using the 3H water release assay method, aromatase activity was measured on the particulate fractions from 100 breast tissue specimens, with levels varying from 3 to 120 fmol/mg/h (41). Therefore, a 40-fold range of aromatase activity could be detected in human breast cancer tissue. Immunocytochemical analysis from our laboratory first identified the presence of aromatase in both breast cancer epithelial and stromal cells (7). Our findings have recently been confirmed by immunocytochemical analysis using newly developed antiaromatase monoclonal antibodies 677 and F2 [presented by Dr. H. Sasano (Department of Pathology, Tohoku University School of Medicine, Sendai, Japan) at the Aromatase 2004 meeting]. Dr. Sasano has pointed out that by evaluating 43 cases of breast cancer specimens, aromatase is detected in both breast cancer epithelial and stromal cells. Furthermore, it was found that there is a good correlation between the biochemical activity of aromatase and the immunopositivity of carcinoma cells with antibody 677. Therefore, it can be stated that aromatase is expressed in breast cancer tissue (both cancer and surrounding adipose stromal cells), probably at a higher level than normal breast tissue, as shown by enzyme activity measurement and immunocytochemistry. Several groups have used breast tumor tissues in an attempt to identify genes that can serve as potential molecular markers for breast cancer (42-45). However, breast tumors consist of many kinds of cells including ER-positive and ER-negative cells and aromatase-positive and aromatase-negative cells. These tissues are too complex to investigate the effects of hormone and inhibitor treatments on gene expression. Whereas it is important to understand the molecular action of aromatase inhibitors and tamoxifen in clinical specimens from breast cancer patients, as the first step it is essential to assess the molecular changes induced by these compounds using a defined cell culture system. In this study, we used an aromatase-overexpressing ER-positive breast cancer cell line, MCF-7aro, which has been established in our laboratory (15) and has been used by a number of laboratories to evaluate aromatase inhibitors and their effects using cell culture and in vivo studies (14-16, 46).

We took many steps to ensure high-quality data, including performing three replicates of each treatment. These steps were designed to reduce the false discovery rate and eliminate potential bias from the combined microarray data. As a result of our careful data analysis, we have been able to reproduce our microarray expression results using Northern blot analysis for the 13 genes we chose.

There are many genes known to be estrogen responsive. Interestingly, we found little overlap among the lists of estrogen-modulated genes reported in the previously published papers. However, when we combined the lists of estrogen-modulated genes from the 11 papers and compared it to the list of estrogen-regulated genes from our study, we found that 69.8% and 67.5% of the published up-regulated and down-regulated genes overlap with our results, respectively. Frasor et al. (26) have shown the time course patterns of gene regulation after the estrogen treatment of MCF-7 cells for 0 to 48 hours. According to them, almost all of our genes that were not consistent with published data were categorized as early-responsive genes, meaning they responded to estrogen quickly (within 4-8 hours) and recovered to normal expression level gradually. Because the patients who have hormone-dependent cancers are chronically exposed to hormones over a long period of time, it is very important to investigate the genes that respond to hormones under the long-term exposure.

Hodges et al. (34) have reported that fos, myc, myb, CDC25A, cyclin E, cyclin A2, and STK15 were up-regulated by tamoxifen in MCF-7 cells. In our study, however, CDC25A and cyclin A2 displayed almost no change; fos, myc, myb, and STK15 were down-regulated; and only cyclin E was up-regulated by tamoxifen as well as the aromatase inhibitors. As discussed above, these differences might be the result of differences in experimental conditions (e.g., the length of treatments) and/or the presence (our study) or absence (previous studies) of androgen, as well as different criteria of gene selection (e.g., signal intensity and fold change cutoffs).

We have shown that the inhibitors prevented cell proliferation even in the presence of androgen. In other words, the regulation of gene expression mediated by androgen (i.e., estrogen converted from androgen by aromatase in MCF-7aro) was counteracted by the inhibitors. More than 50% of the hormone-regulated genes were counterregulated by all three inhibitors and >90% were counterregulated by at least one of the inhibitors (Fig. 3B). Letrozole, anastrozole, and tamoxifen down-regulated 92.3%, 75.0%, and 81.7% of hormone up-regulated genes and they up-regulated 85.3%, 78.9%, and 67.0% of hormone down-regulated genes, respectively. In both groups, letrozole showed the best counterregulation of gene expression compared with hormone treatment alone. This is possibly the reason why letrozole-treated cells did not grow very well, and why letrozole showed better activity than tamoxifen in first-line therapy of postmenopausal women with breast cancers (19).

This is the first study to use microarrays to compare the effects of both aromatase inhibitors and antiestrogens at the molecular level. Based on a comparison of expression patterns, we have been able to show both a high correlation between letrozole and anastrozole and a clear difference between aromatase inhibitors and tamoxifen. As indicated in Results, it is possible that some of the genes modulated only by aromatase inhibitors are regulated by estrogen in an ER-independent manner. Additional control experiments will be done to determine whether ER is involved in the regulation of this category of genes.

Thiantanawat et al. (40) have analyzed the genes that are involved in apoptosis pathways and regulated by letrozole and tamoxifen in MCF-7aro cells. According to them, p53, p21, and Bax were up-regulated and Bcl-2, cyclin D1, and c-myc were down-regulated by the chemical treatments. In our data, except for p53 (no change), all genes were similarly regulated by inhibitors. Overall, profiles of the genes involved in the apoptosis pathway showed good agreement with the report of Thiantanawat et al. (40). After treatment with inhibitors, apoptosis-inducing activity (group A) went up, whereas the antiapoptotic gene Bcl-2 was down-regulated (group C). It seems that exposure to these inhibitors can lead cancer cells toward apoptosis. The profiling of the genes involved in the MAPK cascade and Wnt signaling pathway also supports this idea. Both pathways contribute to apoptosis, at least partially, and those apoptosis-related sections of the pathways seem to be consistent with the observation discussed above (group B in both Fig. 5B and C). On the other hand, MAP kinases (Fig. 5B, group C) and protein kinase Cs (Fig. 5C, group A) are involved in cell proliferation as well as apoptosis, suggesting the possibility of mediation of cell growth and proliferation in breast tumors under the chemical treatments.

In summary, we have found that hormonal stimulation of gene expression can be counteracted by treatment with aromatase inhibitors and an antiestrogen, and that these chemicals have their own unique effects on gene expression. In addition, it is clear from our data that the gene expression profiles for letrozole treatment versus anastrozole treatment are more similar than that for tamoxifen treatment.

The gene expression profiles in MCF-7aro cells following different treatments are being examined carefully and extensively in our laboratory. Whereas one could argue that our results were generated from a single cell line, this approach allows us to compare the effects of tamoxifen, letrozole, and anastrozole under identical experimental conditions. It is our goal to identify unique gene expression profiles in aromatase-positive and ER-positive breast cancer cells in the presence of androgen or estrogen, with or without aromatase inhibitors/antiestrogens. As discussed above, MCF-7aro cells have been utilized in a number of laboratories to evaluate the effects of aromatase inhibitors versus tamoxifen in vitro and in vivo. Several clinical trials have been designed based on results generated using the MCF-7aro–induced tumor model. Therefore, results generated from this microarray study are potentially very valuable for the patient-designed treatment of breast cancer. For example, in the case of tumors with gene expression patterns that have been found to be modulated better by letrozole than the other two drugs, letrozole would be chosen as the drug of choice for treatment.

Results generated from gene microarray analysis of clinical tumor specimens are much more complicated to analyze and interpret. To better understand the results generated from clinical tissue, we feel that it is important to first obtain information from a model system, such as the study reported here. The results produced from this study will serve as a basis to help evaluate the results generated from clinical samples later.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Acknowledgements
 References
 
Cell Culture, Growth Curve, and RNA Extraction
A stable aromatase-expressing ER-positive MCF-7 cell line (MCF-7aro) has been established in our laboratory by aromatase cDNA transfection and G-418 (neomycin) selection (15). The cells were maintained routinely in Eagle's MEM with nonessential amino acids, sodium pyruvate, 10% fetal bovine serum, and 100 µg/mL G-418 at 37°C, 5% CO2. To eliminate the influence of steroid hormones in culture medium, for hormone/inhibitor treatment, cells were cultured in phenol red–free Eagle's MEM with nonessential amino acids, sodium pyruvate, and 10% charcoal/dextran–treated fetal bovine serum for 3 days before addition of inhibitor and/or hormone to culture medium. In our experiments, we cultured the cells under the following six conditions: with DMSO (vehicle), with 1 nmol/L testosterone (androgen; Sigma Chemical, St. Louis, MO), with 1 nmol/L 17ß-estradiol (estrogen; Sigma Chemical), with androgen + 200 nmol/L letrozole (aromatase inhibitor; Novartis, Basel, Switzerland), with androgen + 1 µmol/L anastrozole (aromatase inhibitor; Zeneca Pharmaceuticals, Macclesfield, United Kingdom), and with androgen + 1 µmol/L tamoxifen (ER inhibitor; Sigma Chemical). The concentrations of hormones, letrozole, anastrozole, and tamoxifen used in this study were determined from dose-response experiments previously done in our laboratory (data not shown). The cells were cultured for 7 days in culture media containing inhibitor and/or hormone that were refreshed every 3 days. For cell growth analysis, cells were harvested every 3 days with 0.5 mol/L NaOH and the protein concentration was measured using Bio-Rad Protein Assay reagent (Bio-Rad, Hercules, CA). For microarray and Northern blot analyses, total RNA was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA) after 7 days of culture according to the manufacturer's instruction.

Microarray
For the microarray analyses, all six experiments were done in triplicate to increase the precision of estimation. Each set of six experiments was done independently and all 18 total RNA samples were extracted as described above. Synthesis and labeling of cRNA probes, hybridization on GeneChips, and signal detection were carried out by the Microarray Core Facility at the University of California, Irvine. Briefly, total RNA obtained from hormone/inhibitor-treated cells was reverse-transcribed and double-stranded cDNA was synthesized. The cDNA was used as a template for synthesis of biotin-labeled cRNA probe. Hybridization onto Affymetrix Human Genome U133A GeneChips (Affymetrix) was done at 45°C for 16 hours. After washing, the GeneChips were stained with streptavidin phycoerythrin, and scanned and analyzed using Affymetrix Microarray Suite version 5.0 (MAS, Affymetrix). Further analyses were done using GeneSpring version 6.1 (Silicon Genetics, Redwood City, CA) and Excel 2000 (Microsoft, Redmond, WA).

Statistical Processing
We exported the raw expression data, with no normalization or scaling, to multiple .CEL files. The Affymetrix raw, nonnormalized .CEL files were uploaded to the R-Project Bioconductor statistical tools package (47, 48). The affy package within BioConductor was used to carry out the normalization and background noise subtraction. Our background correction was based on the RMA method in the Bioconductor affy package. The probe-level correction was to used perfect matches only (pmonly method in the affy package). We applied quantile normalization simultaneously across all arms of the experimental groups (49). The quantile normalization forces all probe intensities to conform to the same distribution for each array among the set of arrays. After the normalized expression values were derived, we specified average difference (avgdiff in the affy package) as the summary method. We did comparisons among the treatment and control groups. We excluded all genes with intensity values <120 and fold changes between 0.67 and 1.50 for each individual comparison. These genes were excluded to reduce the overall false discovery rate for each of the statistical treatment comparisons. The experimental design of this study is a balanced design with three replications for each of the treatment and control groups. Each of the expression-level comparisons was determined to be up-regulated or down-regulated relative to the control. For each of the differences in the mean expression levels, we applied standard ANOVA tests to derive the test statistic. To further reduce the overall false discovery rate, we applied the Benjamini and Yuketieli adjustment for multiple-hypothesis comparisons (50). Two-dimensional hierarchical clustering analysis was done using Gene Tree and Condition Tree clustering methods within GeneSpring. The genes with the most statistically significant fold changes and functional roles were considered for further study.

Northern Blot Analysis
Probes for Northern analyses were designed based on the sequence of each probe set on the Affymetrix GeneChips and from National Center for Biotechnology Information's Genbank. The target fragment of each gene was amplified by reverse transcription-PCR, run on agarose gels, purified from gels, and inserted into TA-vectors using the TOPO TA Cloning kit (Invitrogen). After confirmation of the nucleotide sequences, the TA vectors were used as templates for PCR to obtain enough target DNA fragments. Fifty nanograms of the PCR product were used for 32P labeling using Prime-It II Random Primer Labeling kit (Stratagene, Cedar Creek, TX). Total RNA (20 µg/20 µL per lane) obtained from hormone/inhibitor–treated cells was mixed with 5 µL RNA Sample Loading Buffer (Sigma-Aldrich, St. Louis, MO), heated at 65°C for 5 minutes, and chilled on wet ice for 2 minutes. Samples were run on a denaturing gel containing 1% agarose, 2.2 mol/L formaldehyde, and 1x MOPS buffer [20 mmol/L MOPS, 5 mmol/L sodium acetate, 1 mmol/L EDTA (pH 7.0)] at 50 mV for 2.5 hours in 1x MOPS running buffer at room temperature, and were then transferred to Zeta-Probe GT Membrane (Bio-Rad) in 20x SSC [333 mmol/L NaCl, 333 mmol/L sodium citrate (pH 7.0)] using a TurboBlotter System (Schleicher & Schuell, Germany) for 4 hours. The membrane was washed in 2x SSC and samples were immobilized onto the membrane by UV irradiation. After prehybridization with ExpressHyb (BD Bioscience Clontech, Palo Alto, CA) in a hybridization bag at 65°C for 1 hour, the prehybridization solution was discarded and 32P-labeled probe, diluted in fresh ExpressHyb, was poured onto the membrane. Hybridization was done at 65°C for 1 hour in a hybridization oven and the membrane was washed thrice with 2x SSC/0.05% SDS at room temperature for 10 minutes and twice with 0.1x SSC/0.1% SDS at 50°C for 20 minutes, followed by exposure to X-ray film at –80°C. The films were developed using a Konica SRX-101A Film Processor (Konica, Tokyo, Japan) and the density of each band was measured by using a GS-700 Imaging Densitometer and Quantity One software (Bio-Rad). The equal loading of RNA samples for the Northern analyses was confirmed by the hybridization of each membrane using a glyceraldehyde-3-phosphate dehydrogenase probe.


   Acknowledgements
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
References
 
We thank Fun Shan Chen and Dr. Xiwei Wu for their comments and suggestions.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Grant support: NIH grant CA44735 (S. Chen) and Flanigan Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: T. Itoh and K. Karlsberg contributed equally to this work.

Received 7/ 6/04; revised 2/22/05; accepted 2/23/05.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 

  1. Lippman ME, Dickson RB, Kasid A, et al. Autocrine and paracrine growth regulation of human breast cancer. J Steroid Biochem Mol Biol 1986;24:147–54.
  2. Bates SE, Davidson NE, Valverius EM, et al. Expression of transforming growth factor {alpha} and its messenger ribonucleic acid in human breast cancer: its regulation by estrogen and its possible functional significance. Mol Endocrinol 1988;2:543–55.[Abstract]
  3. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371–88.[Abstract/Free Full Text]
  4. Fisher B, Dignam J, Wolmark N, et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet 1999;353:1993–2000.[CrossRef][Medline]
  5. James VH, McNeill JM, Lai LC, Newton CJ, Ghilchik MW, Reed MJ. Aromatase activity in normal breast and breast tumor tissues: in vivo and in vitro studies. Steroids 1987;50:269–79.[CrossRef][Medline]
  6. Miller WR, O'Neill J. The importance of local synthesis of estrogen within the breast. Steroids 1987;50:537–48.[CrossRef][Medline]
  7. Esteban JM, Warsi Z, Haniu M, Hall P, Shively JE, Chen S. Detection of intratumoral aromatase in breast carcinomas. An immunohistochemical study with clinicopathologic correlation. Am J Pathol 1992;140:337–43.[Abstract]
  8. Sasano H, Nagura H, Harada N, Goukon Y, Kimura M. Immunolocalization of aromatase and other steroidogenic enzymes in human breast disorders. Hum Pathol 1994;25:530–5.[CrossRef][Medline]
  9. Santen RJ, Martel J, Hoagland M, et al. Stromal spindle cells contain aromatase in human breast tumors. J Clin Endocrinol Metab 1994;79:627–32.[Abstract]
  10. Lu Q, Nakmura J, Savinov A, et al. Expression of aromatase protein and messenger ribonucleic acid in tumor epithelial cells and evidence of functional significance of locally produced estrogen in human breast cancers. Endocrinology 1996;137:3061–8.[Abstract]
  11. Bulun SE, Price TM, Aitken J, Mahendroo MS, Simpson ER. A link between breast cancer and local estrogen biosynthesis suggested by quantification of breast adipose tissue aromatase cytochrome P450 transcripts using competitive polymerase chain reaction after reverse transcription. J Clin Endocrinol Metab 1993;77:1622–8.[Abstract]
  12. Zhou C, Zhou D, Esteban J, et al. Aromatase gene expression and its exon I usage in human breast tumors. Detection of aromatase messenger RNA by reverse transcription-polymerase chain reaction. J Steroid Biochem Mol Biol 1996;59:163–71.[CrossRef][Medline]
  13. Harada N. Aberrant expression of aromatase in breast cancer tissues. J Steroid Biochem Mol Biol 1997;61:175–84.[CrossRef][Medline]
  14. Santner SJ, Chen S, Zhou D, Korsunsky Z, Martel J, Santen RJ. Effect of androstenedione on growth of untransfected and aromatase-transfected MCF-7 cells in culture. J Steroid Biochem Mol Biol 1993;44:611–6.[CrossRef][Medline]
  15. Sun XZ, Zhou D, Chen S. Autocrine and paracrine actions of breast tumor aromatase. A three-dimensional cell culture study involving aromatase transfected MCF-7 and T-47D cells. J Steroid Biochem Mol Biol 1997;63:29–36.[CrossRef][Medline]
  16. Yue W, Zhou D, Chen S, Brodie A. A new nude mouse model for postmenopausal breast cancer using MCF-7 cells transfected with the human aromatase gene. Cancer Res 1994;54:5092–5.[Abstract/Free Full Text]
  17. Tekmal RR, Ramachandra N, Gubba S, et al. Overexpression of int-5/aromatase in mammary glands of transgenic mice results in the induction of hyperplasia and nuclear abnormalities. Cancer Res 1996;56:3180–5.[Abstract/Free Full Text]
  18. Winer EP, Hudis C, Burstein HJ, et al. American Society of Clinical Oncology technology assessment on the use of aromatase inhibitors as adjuvant therapy for women with hormone receptor-positive breast cancer: status report 2002. J Clin Oncol 2002;20:3317–27.[Abstract/Free Full Text]
  19. Mouridsen H, Gershanovich M, Sun Y, et al. Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: results of a phase III study of the International Letrozole Breast Cancer Group. J Clin Oncol 2001;19:2596–606.[Abstract/Free Full Text]
  20. Nabholtz JM, Buzdar A, Pollak M, et al. Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial. Arimidex Study Group. J Clin Oncol 2000;18:3758–68.[Abstract/Free Full Text]
  21. Bonneterre J, Thurlimann B, Robertson JF, et al. Anastrozole versus tamoxifen as first-line therapy for advanced breast cancer in 668 postmenopausal women: results of the Tamoxifen or Arimidex Randomized Group Efficacy and Tolerability study. J Clin Oncol 2000;18:3748–57.[Abstract/Free Full Text]
  22. Bonneterre J, Buzdar A, Nabholtz JM, et al. Anastrozole is superior to tamoxifen as first-line therapy in hormone receptor positive advanced breast carcinoma. Cancer 2001;92:2247–3.[CrossRef][Medline]
  23. Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003;349:1793–802.[Abstract/Free Full Text]
  24. Buzdar AU. Pharmacology and pharmacokinetics of the newer generation aromatase inhibitors. Clin Cancer Res 2003;9:468–72S.
  25. Chen S, Zhang F, Sherman MA, et al. Structure-function studies of aromatase and its inhibitors: a progress report. J Steroid Biochem Mol Biol 2003;86:231–7.[CrossRef][Medline]
  26. Frasor J, Danes JM, Komm B, Chang KC, Lyttle CR, Katzenellenbogen BS. 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 2003;144:4562–74.[Abstract/Free Full Text]
  27. Frasor J, Stossi F, Danes JM, Komm B, Lyttle CR, Katzenellenbogen BS. Selective estrogen receptor modulators: discrimination of agonistic versus antagonistic activities by gene expression profiling in breast cancer cells. Cancer Res 2004;64:1522–33.[Abstract/Free Full Text]
  28. Charpentier AH, Bednarek AK, Daniel RL, et al. Effects of estrogen on global gene expression: identification of novel targets of estrogen action. Cancer Res 2000;60:5977–83.[Abstract/Free Full Text]
  29. Ghosh MG, Thompson DA, Weigel RJ. PDZK1 and GREB1 are estrogen-regulated genes expressed in hormone-responsive breast cancer. Cancer Res 2000;60:6367–75.[Abstract/Free Full Text]
  30. Finlin BS, Gau CL, Murphy GA, et al. RERG is a novel ras-related, estrogen-regulated and growth-inhibitory gene in breast cancer. J Biol Chem 2001;276:42259–67.[Abstract/Free Full Text]
  31. Inoue A, Yoshida N, Omoto Y, et al. Development of cDNA microarray for expression profiling of estrogen-responsive genes. J Mol Endocrinol 2002;29:175–92.[Abstract]
  32. Seth P, Krop I, Porter D, Polyak K. Novel estrogen and tamoxifen induced genes identified by SAGE (Serial Analysis of Gene Expression). Oncogene 2002;21:836–43.[CrossRef][Medline]
  33. Hayashi S, Sakamoto T, Inoue A, Yoshida N, Omoto Y, Yamaguchi Y. Estrogen and growth factor signaling pathway: basic approaches for clinical application. J Steroid Biochem Mol Biol 2003;86:433–42.[CrossRef][Medline]
  34. Hodges LC, Cook JD, Lobenhofer EK, et al. Tamoxifen functions as a molecular agonist inducing cell cycle-associated genes in breast cancer cells. Mol Cancer Res 2003;1:300–11.[Abstract/Free Full Text]
  35. Coser KR, Chesnes J, Hur J, Ray S, Isselbacher KJ, Shioda T. Global analysis of ligand sensitivity of estrogen inducible and suppressible genes in MCF7/BUS breast cancer cells by DNA microarray. Proc Natl Acad Sci U S A 2003;100:13994–9.[Abstract/Free Full Text]
  36. Oesterreich S, Deng W, Jiang S, et al. Estrogen-mediated down-regulation of E-cadherin in breast cancer cells. Cancer Res 2003;63:5203–8.[Abstract/Free Full Text]
  37. Wang DY, Fulthorpe R, Liss SN, Edwards EA. Identification of estrogen-responsive genes by complementary deoxyribonucleic acid microarray and characterization of a novel early estrogen-induced gene: EEIG1. Mol Endocrinol 2004;18:402–11.[Abstract/Free Full Text]
  38. Yoshiko Y, Aubin JE. Stanniocalcin 1 as a pleiotropic factor in mammals. Peptides 2004;25:1663–9.[CrossRef][Medline]
  39. Hu Y, Wang T, Stormo GD, Gordon JI. RNA interference of achaete-scute homolog 1 in mouse prostate neuroendocrine cells reveals its gene targets and DNA binding sites. Proc Natl Acad Sci U S A 2004;101:5559–64.[Abstract/Free Full Text]
  40. Thiantanawat A, Long BJ, Brodie AM. Signaling pathways of apoptosis activated by aromatase inhibitors and antiestrogens. Cancer Res 2003;63:8037–50.[Abstract/Free Full Text]
  41. Miller WR, Mullen P, Sourdaine P, Watson C, Dixon JM, Telford J. Regulation of aromatase activity within the breast. J Steroid Biochem Mol Biol 1997;61:193–202.[CrossRef][Medline]
  42. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000;406:747–52.[CrossRef][Medline]
  43. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001;98:10869–74.[Abstract/Free Full Text]
  44. van 't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–3.[CrossRef][Medline]
  45. Amatschek S, Koenig U, Auer H, et al. Tissue-wide expression profiling using cDNA subtraction and microarrays to identify tumor-specific genes. Cancer Res 2004;64:844–3.[Abstract/Free Full Text]
  46. Eng ET, Ye J, Williams D, et al. Suppression of estrogen biosynthesis by procyanidin dimers in red wine and grape seeds. Cancer Res 2003;63:8516–22.[Abstract/Free Full Text]
  47. Ihaka R, Gentleman R. R: A language for data analysis and graphics. J Comput Graph Stat 1996;5:299–314.[CrossRef]
  48. R Development Core Team. R: A language and environment for statistical computing. Vienna (Austria): R Foundation for Statistical Computing; 2003.
  49. Irizarry RA, Hobbs B, Collin F, et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 2003;4:249–64.[Abstract]
  50. Benjamini Y, Yekutieli D. The control of the false discovery rate in multiple hypothesis testing under dependency. Ann Stat 2001;29:1165–88.[CrossRef]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
S. Masri, S. Phung, X. Wang, X. Wu, Y.-C. Yuan, L. Wagman, and S. Chen
Genome-Wide Analysis of Aromatase Inhibitor-Resistant, Tamoxifen-Resistant, and Long-Term Estrogen-Deprived Cells Reveals a Role for Estrogen Receptor
Cancer Res., June 15, 2008; 68(12): 4910 - 4918.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
M. Dolled-Filhart, L. Ryden, M. Cregger, K. Jirstrom, M. Harigopal, R. L. Camp, and D. L. Rimm
Classification of breast cancer using genetic algorithms and tissue microarrays.
Clin. Cancer Res., November 1, 2006; 12(21): 6459 - 6468.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Itoh, T.
Right arrow Articles by Chen, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow