Communicated by: Shunsuke Ishii
Genome-wide analysis of changes in early gene expression induced by oestrogen
Version of Record online: 28 MAY 2002
Genes to Cells
Volume 7, Issue 5, pages 497–507, May 2002
How to Cite
Watanabe, H., Suzuki, A., Mizutani, T., Khono, S., Lubahn, D. B., Handa, H. and Iguchi, T. (2002), Genome-wide analysis of changes in early gene expression induced by oestrogen. Genes to Cells, 7: 497–507. doi: 10.1046/j.1365-2443.2002.00535.x
- Issue online: 28 MAY 2002
- Version of Record online: 28 MAY 2002
- Received: 9 January 2002 Accepted: 22 February 2002
Background: The sex hormone 17β-oestradiol (E2) has profound effects on many aspects of reproduction, development, as well as behaviour. Although the oestrogen receptor is well characterized on a molecular level, relatively few genes affected by E2 have been identified, and the mechanisms underlying the physiological changes caused by E2 are largely unknown. In order to identify oestrogen-regulated genes in vivo, early uterine gene expression profiles were developed using DNA microarrays.
Results: Ovariectomized mice were exposed to 17β-oestradiol for 6 h, and mRNA expression analysis for 9977 genes was performed. Although a large number of genes was affected by oestrogen administration, the genes that showed higher reproducibility in repetitive experiments were selected and further examined. For most of the selected genes, expression was induced in a dose-dependent manner, and gene expression was not altered following oestrogen treatment in oestrogen receptor-α (ERα)-deficient mice. In combination with the estimation of gene expression levels using quantitative PCR, it was revealed that multiple genes related to sterol biosynthesis, tRNA synthesis, RNA processing, and growth signalling were activated. Based on the microarray data, we selected additional genes related to sterol biosynthesis and tRNA synthesis and confirmed that these genes are also activated by oestrogen.
Conclusion: Genes suggesting a basis for the drastic uterotrophic effect observed several days following oestrogen administration were identified. These findings not only reveal the diverse effect of oestrogen signalling on transcript levels in vivo but also demonstrate the ability of DNA microarrays to identify cellular pathways affected by oestrogen.
Steroid hormones exert profound effects on many aspects of developmental and physiological processes. These steroid hormones bind to nuclear hormone receptors that directly bind to target DNA sequences thereby regulating gene expression. The sex hormone 17β-oestradiol (E2) is a well-characterized steroid hormone and the most potent endogenous oestrogen. E2 is known to control the development of the reproductive tract and mammary gland, regulate the oestrus cycle, control lactation, and exert regulatory influences important for bone, liver, and cardiovascular systems, as well as behaviour. The critical regulatory functions of oestrogen are mediated by oestrogen receptors (ERs) and at least two types of oestrogen receptors, ERα and ERβ, are known in mammals (Green et al. 1986; Kuiper et al. 1996). The molecular mechanisms of ER transactivation have been well characterized. ER directly binds to oestrogen-responsive elements (EREs) within enhancer regions of target genes and functions as a transcription factor by recruiting cofactors in a ligand-dependent manner. In addition to the so-called genomic action of the oestrogen receptor, a non-genomic action of oestrogen is also known (Kousteni et al. 2001). Despite the wide variety of oestrogen responses and great progress in characterizing the molecular biology of ER in vitro, relatively few genes affected by E2 have been identified, and the mechanisms underlying the physiological changes caused by E2 are largely unknown. As the uterus is a prominent E2 target organ which also shows drastic changes during the oestrus cycle, it has long been used as a standard system for examining oestrogenic activities in vivo. In spite of drastic changes in the uterus after oestrogen administration, such as an increase in uterine wet weight, glucose metabolism, and histamine levels, little is known about overall changes in gene expression, and the number of uterine genes known to be regulated by oestrogen is amazingly low. As a consequence of this limitation, it has been difficult to elucidate the early molecular events contributing to uterotrophic changes as stimulated by oestrogen during the acute uterine response. In order to clarify the mechanisms of oestrogen function, it is essential to identify the genes affected by oestrogen, especially during the early stages of the uterine response. Since ER is a transcription factor, analysis of changes in mRNA expression patterns caused by E2 can be a powerful tool for understanding the molecular mechanisms underlying oestrogen action. Thus, genome-wide gene expression profiling following E2 administration is essential to understand fully the dramatic effects of oestrogen on the uterus. To analyse genome-wide gene expression changes, we employed a DNA microarray technique. DNA microarray technology has recently been developed and successfully applied to genome-wide analysis of gene expression as influenced by various stimuli, such as serum (Iyer et al. 1999), hormones (Feng et al. 2000), or chemicals (Marton et al. 1998). High-density oligonucleotide arrays (Lockhart et al. 1996) are especially suitable for genome-wide mapping of gene expression because a large number of genes can be analysed at one time, and it is readily scalable to the simultaneous monitoring of tens of thousands of genes. In this study, gene expression changes following oestrogen treatment were profiled using DNA microarrays, and the genes representing early uterotrophic effects were determined.
In order to get an overview of genome-wide gene expression patterns affected by oestrogen in the uterus, total uterine RNA was prepared 6 h after oestrogen administration (5 µg/kg b.w.) and changes in the gene expression of 9977 genes were examined using oligo-DNA microarrays. The scanned data were analysed, and fold changes of gene expression levels were determined based on fluorescence intensities representing the amount of each mRNA. To confirm reproducibility, experiments were repeated independently three times and the fold changes were calculated. Although the number of oestrogen-affected genes was different in each experiment, generally 5% of all genes were activated or repressed more than three-fold by oestrogen. In order to obtain a set of genes whose response was highly reproducible, we applied the Student’s t-test. In this study, we considered gene expressions whose P-values were less than 0.05 as significant. By selecting the genes whose P-values were less than 0.05, 616 genes were selected as oestrogen-affected genes, of which 299 genes were activated and 317 genes were repressed. The mean values of fluorescence intensities of the selected genes in the three experiments were calculated and plotted as shown in Fig. 1. The mean values of the fold change between E2 and oil were calculated, and genes induced more than three-fold were selected for further analysis. Out of 94 selected genes, 45 genes were listed as known genes in GenBank (Table 1) and 49 genes were ESTs. Among the selected genes, we found that the small proline-rich protein 2F can be a sensitive marker of oestrogen administration in the uterus. Interestingly, multiple genes related to growth signalling, receptors, RNA maturation, translation, and cholesterol synthesis were found in a very limited number of the selected genes. On the other hand, 38 known genes were selected as genes whose expression is repressed more than three-fold and having P-values less than 0.05 (Table 2). It is notable that multiple genes related to detoxification and transcription are listed as repressed genes. Among the down-regulated genes, one-fifth of the genes were related to oxidative stress and detoxification. These included Glutathione S-transferaseθ2, Selenium binding protein 1, Thioether S-methyltransferase, Microsomal expoxide hydrolase, Xeroderma pigmentosum, Complementation group C, 8-oxo-dGTPase, and P glycoprotein 3. To further confirm the expression of the genes induced by oestrogen, we examined the dose dependency of the gene expression pattern by changing the quantities of oestradiol administrated to mice: 0.05, 0.5, 5.0 and 50 µg/kg b.w. of oestrogen were administrated to mice, processed, and the gene expression pattern analysed using DNA microarrays. For each dose, two independent experiments were performed and mean values were calculated. All selected genes showed gene activation in a dose-dependent manner between 0.5 µg/kg and 5 µg/kg b.w., although several genes did not show clear dose–responses at the highest (50 µg/kg b.w) and/or lowest (0.05 µg/kg b.w) dose. Figure 2A indicates the normalized dose–response curve of the selected induced genes. Since the doses between 0.5 and 50 µg/kg are optimal concentrations of oestrogen to evoke a uterotrophic effect, this dose–response supports the idea that the selected genes are contributing to the uterotrophic effect. Similar to the oestrogen-induced genes, the repressed genes showed a dose dependency, as demonstrated in Fig. 2B. In order to examine whether the activation of selected genes depends on ERα, we analysed the gene expression pattern of the selected genes in ERα-deficient mice (αERKO) after oestrogen administration. Total uterine RNA from αERKO mice was prepared 6 h after E2 administration (5 µg/kg b.w.) and processed in the same manner as wild-type tissue. The fluorescence intensities of each gene were compared with that of an oil-treated control and fold changes were calculated. Most of the selected genes were not activated by E2 in αERKO (Figs 2 and 3). Only five genes—IGF-I, three kallikrein genes (kallikrein, potential kallikrein, epidermal growth factor binding protein), and the early growth response 1(EGR-1) gene—were activated more than two-fold by oestrogen administration in αERKO cells. All five genes were activated to much greater extents in wild-type mice. The dose dependency and the oestrogen receptor dependency of the selected genes supported our hypothesis that these are in fact oestrogen-activated genes.
|37.3||AJ005564||small proline-rich protein 2F||46.1||C|
|20.4||AJ250491||receptor activity modifying protein 3||19.5||C|
|19.4||X04480||insulin-like growth factor 1||15.2||A (1)|
|9.4||Y11666||hexokinase II, exon 1||5.5||B (2)|
|6.8||M13500||potential kallikrein||5.0||B (3)|
|6.0||AJ010108||cytosolic adenylate kinase||12.7||C|
|6.0||M35970||tumor metastatic process-associated protein (NM23)||8.4||A (4)|
|5.5||AF025653||capping enzyme (MCE)||1.9||C|
|5.4||M17979||epidermal growth factor binding protein type A||NT||B (3)|
|5.3||AF000236||chemokine orphan receptor 1||4.8||C|
|5.0||L02526||protein kinase, mitogen activated, kinase 1, p45||2.0||C|
|4.8||M62766||3-hydroxy-3-methylglutaryl-coenzyme A reductase||5.0||A (5)|
|4.7||J04627||NAD-dependent methylenetetrahydrofolate dehydrogen methenyltetrahydrofolate cyclohydrolase||30.3||C|
|4.6||AB017189||4F2/CD98 light chain||12.9||C|
|4.5||L40406||heat shock protein, 105 kDa||9.6||C|
|4.4||AF034610||nuclear autoantigenic sperm protein||2.0||C|
|4.2||AF096875||type 2 deiodinase||2.5||C|
|4.2||U08110||RAN GTPase activating protein 1||4.0||C|
|4.1||M96163||serum inducible kinase (SNK)||9.9||C|
|4.0||U14648||splicing factor, arginine/serine-rich 10||2.3||C|
|4.0||AF009246||ras-related protein (DEXRAS1)||5.3||C|
|3.7||U12922||RAB geranylgeranyl transferase, b subunit||3.9||C|
|3.4||U28423||protein kinase inhibitor p58||5.2||C|
|3.3||AB012276||ATFx, partial cds||4.0||C|
|3.3||Z18272||procollagen, type VI, alpha 2||9.9||C|
|3.3||AB003502||guanine nucleotide regulatory protein||4.1||C|
|3.3||AF016544||spermatogenesis associated factor||4.4||C|
|3.3||AF064749||type VI collagen alpha 3 subunit||0.4||C|
|3.2||D14336||RNA polymerase I associated factor (PAF53)||96.3||C|
|3.2||M28845||Early growth response 1||5.5||A (6)|
|3.2||U43327||Laminin, gamma 2||3.4||B (7)|
|3.1||D17666||mitochondrial stress-70 protein (PBP74/CSA)||5.2||C|
|3.1||X59769||Interleukin 1 receptor, type 11||17.5||C|
|3.1||U90446||RNAse L inhibitor (Mu-RLI)||4.6||C|
|AF123263||phenylalanyl-tRNA synthetase beta chain||2.7|
|−10||X98056||glutathione S-transferase, theta 2||C|
|−9.2||X04120||intracisternal A particles, Thbd linked||C|
|−8.7||U89491||microsomal expoxide hydrolase (Eph1)||C|
|−6.8||AB012808||mBOCT (organic cation transporter)||C|
|−6.8||U42467||leptin receptor||D (1)|
|−6.2||M32032||selenium binding protein 1||C|
|−5.8||X16670||type 11B intracisternal A-particle (RIP) element encoding integrase||C|
|−5.6||U12791||3-hydroxy-3-methylgiutaryl coenzyme A synthase 2||C|
|−5.6||AJ132192||HS1 binding protein 3||C|
|−5.5||L31532||B-cell leukaemia/lymphoma 2 major histocompatibility complex region NG27, NG28, RPS28, NADH oxidoreductase, NG29||D (2)|
|−5.1||AF110520||KIFC1, Fas-binding protein, BING1, tapasin, RalGDS-like, KE2, BING4, beta 1,3-galactosyl transferase, and RPS18 genes||C|
|−4.3||U51167||isocitrate dehydrogenase 2||C|
|−4.2||U61183||yolk sac gene 2||C|
|−4.1||X70854||BM-90/fibulin extracellular matrix glyroprotein||C|
|−4.1||U04710||insulin-like growth factor 2 receptor||D (3)|
|−4||X94404||b3 gene for alpha3 subunit of l-type Ca2+ channel||D (4)|
|−3.9||M70642||fibroblast inducible secreted protein||C|
|−3.8||X99572||c-fos induced growth factor||C|
|−3.8||AB023957||EIG 180 (ethanol induced gene)||C|
|−3.7||U77364||homeodomain-containing transcription factor (Hoxd4)||C|
|−3.6||X92397||Norre disease homologue||C|
|−3.6||A8010031||outer arm dynein light chain 4||C|
|−3.5||X81202||glycine receptor, beta subunit||C|
|−3.4||M24417||P glycoprotein 3||D (5)|
|−3.3||Y12650||hereditary haemochromatosis-like protein||C|
|−3.3||U27398||Xeroderma pigmentasum, complementation group C||C|
|−3.3||M97516||alpha-2 adrenergic receptor||D (6)|
|−3.3||M60523||inhibitor of DNA binding 3||C|
|−3.2||M96265||galactose-1-phosphate undyl transferase (GALT)||C|
|−3.2||Y10007||fibroblast activation protein||C|
|−3.1||U07861||zinc finger protein 101||C|
In order to confirm the activation of the genes by E2, we also employed reverse transcription based quantitative-real-time-PCR (Q-RT-PCR). Total RNA was prepared from oil- or oestradiol-treated mouse uteri, and quantitative PCR was performed using the primers listed in Table 3. Fold changes calculated by Q-RT-PCR are shown in Table 1 together with the results of DNA microarray analysis. Forty-four genes were activated more than two-fold and 36 genes were found to be activated more than three-fold, and only one gene was repressed by oestradiol. Although there were differences in the fold change values between analysis by DNA microarrays and Q-RT-PCR, our PCR results confirmed that most of the oestrogen-induced genes selected using DNA microarrays were actually activated. Generally, fold changes were underestimated by DNA microarray analysis.
Since our study demonstrated that genes related to sterol synthesis (3-hydroxy-3-methylglutaryl-coenzyme A reductase, squalene epoxidase, and squalene synthetase) are oestrogen-activated genes, we examined the expression of five additional genes related to sterol biosynthesis using Q-RT-PCR. We found that all five mouse genes examined—mevalonate kinase, isopentenyl-diphosphate delta-isomerase, farnesyl pyrophosphate synthase, farnesyl-diphosphate farnesyltransferase, and lanosterol 14 alpha-demethylase—were activated more than two-fold following oestrogen stimulation (Fig. 4). However, a gene related to retinol metabolism geranylgeranyl pyrophosphate synthetase that uses an intermediate metabolite of cholesterol synthesis was not activated.
Although the number of tRNA ligase genes listed in GenBank is rather limited, a number of tRNA ligases were also activated more than two-fold following E2 stimulation (Table 1). Combined with the activation of PAF53, a gene that activates RNA polymerase I (Hanada et al. 1996), our result suggests that translational machinery was up-regulated after oestrogen administration in the uterus.
In the present study, we found multiple genes activated by oestrogen using DNA microarrays. Although it is impossible to select all oestrogen-affected genes using this method, we showed that the selection of genes based on a conventional Student’s t-test gave reliable results that could be confirmed by Q-RT-PCR. The dose–response analysis and the use of ER-deficient mice also supported the reliability of the selected genes. Differences in values between DNA microarray analysis and Q-RT-PCR may result from differences in the experimental systems.
Our result showing that the genes induced by oestrogen are not activated in αERKO mice indicates that ERα is responsible for their activation; however, this does not necessarily indicate that they are directly affected by oestrogen. It remains possible that some of the genes could be activated by the non-genomic action of ERα. Further analysis of each gene is necessary to elucidate whether it is activated directly or indirectly by the oestrogen receptor.
In our study, we examined oestrogen’s effect on wild-type and αERKO mice but not on βERKO mice because βERKO mice did not show a severe effect on the uterus. αERKO is known to cause severe reproductive and behavioural phenotypes, including infertility of both male and female mice and the absence of breast tissue development; however, βERKO mice have lower ovarian efficiency but develop normally and are grossly and histologically indistinguishable as young adults from their littermates (Krege et al. 1998). The fact that the selected genes were not affected in αERKO mice suggests that these genes are related to drastic oestrogen-induced changes in the uterus.
In the case of sterol biosynthesis and tRNA ligase, we showed that some genes involved in these processes but not present on the microarray were also activated by Q-RT-PCR. Thus, the following strategies: (i) selection of genes using microarrays, (ii) defining subgroups within the selected genes, and (iii) expanded expression analysis based on pathway or functional groups, can be powerful tools for understanding the effects of oestrogen on a transcriptional level.
Interestingly, we found that there were several functionally closely related subgroups in the oestrogen-activated genes. They were sterol biosynthesis, tRNA ligase, RNA maturation, receptors and growth factors. As the genes related to RNA maturation, the mRNA capping enzyme, and splicing factors, RanGTPase and its activating protein that are necessary components in the active transport of proteins and RNA were activated.
One-fifth of oestrogen-repressed genes are related to anti-oxidation and detoxification. Although there are many studies relating cancer and oxidative stress, there is no study showing that the genes related to oxidative stress are down-regulated by oestrogen at the transcriptional level. This finding may be crucial for understanding hormone-induced neoplasms.
In the sterol biosynthesis pathway, only 3-hydroxy-3-methylglutaryl-coenzyme A reductase is known to be regulated by oestrogen in rats, and an exceptional oestrogen-responsive element is found in its promoter region (Di Croce et al. 1999). As to the cholesterol synthesis pathway, the sterol response element binding protein (SREBP) is known as a common transcriptional activator (Yokoyama et al. 1993). It has been reported that geranylgeranyl pyrophosphate synthetase is also activated by SREBP (Sakakura et al. 2001), but our results showed that this gene is not activated by oestrogen (1.3-fold by Q-RT-PCR), suggesting that there is another mechanism of gene activation related to cholesterol synthesis in response to oestradiol. Although we examined whether ERE sequences are located in the promoter region of cholesterol synthesis genes, based on the human genome database, canonical ERE sequences were not found.
Interestingly, 3-hydroxy-3-methylglutaryl coenzyme A synthase 2 was found to be repressed by oestrogen. This gene product is located in the mitochondria and is known as the first enzyme of ketogenesis, a metabolic pathway that provides lipid-derived energy. Thus, it was suggested that oestrogen down-regulates ketogenesis and activates sterol synthesis. This effect of oestrogen is the reverse of the effect of leptin (Liang & Tall 2001), and the overall gene expression changes by oestrogen contrasted significantly with the gene expression changes caused by leptin. Although not all gene expression profiles are consistent with an oestrogen effect (for example, urea metabolism was not affected in our study though it was affected by leptin), the behaviour of the selected genes in our study showed the reverse effect following leptin treatment. That is, we found that oestrogen down-regulates the expression of the leptin receptor and genes related to oxidative stress and activates the genes related to cholesterol synthesis. On the other hand, when leptin was added to ob/ob mutant mice, the genes related to fatty acid and cholesterol synthesis were repressed and the genes related to fatty acid oxidation and oxidant defence were activated. Although they were not selected because of higher P-values, the genes related to fatty acid oxidation [the very long chain acyl-CoA dehydrogenase, acetyl coenzyme A dehydrogenase (short chain and medium chain)] were repressed 2.4- to 7-fold. Thus shutting off the leptin signal by repressing its receptor gene expression and switching the pathway from fatty acid oxidation to synthesis may be an important process related to the initial stage of oestrogen administration.
The insulin-like growth factor I is known to be activated by oestrogen at the transcriptional level (Umayahara et al. 1994). In addition, three kallikrein genes (kallikrein, potential kallikrein, and epidermal growth factor binding protein) were activated by oestrogen. Since kallikrein is known to degrade insulin-like growth factor binding proteins such as IGFBP-3 (Rajah et al. 1995), simultaneous transcription induction of IGF-I and kallikrein may be important for an effective response to oestrogen. Interestingly, IGF-I and the early response gene (ERG) 1 were weakly activated by oestrogen even in ERα-deficient mice.
It is known that activation of EGR-1 is a non-nuclear action of ER (de Jager et al. 2001) and that the IGF-I gene is not regulated through the oestrogen response element in its transcriptional regulatory region. Although EGR-1 and IGF-I were activated much more in wild-type cells, activation of these genes in receptor-deficient mice might reflect other activation mechanisms not directly related to ER.
Small increases of mRNA related to the chemokine orphan receptor (RDC1) and the receptor activity modifying protein (RAMP3) suggested a role for CGRP and/or adrenomedullin in oestrogen responsiveness. Although it has been reported that the concentration of CGRP is elevated by oestrogen (Gangula et al. 2000) and the concentration of adrenomedullin is elevated in pregnancy (Jerat & Kaufman 1998), there is little information about these receptors. In various AM/CGRP receptors, these two genes may play an important role in the uterine oestrogen response.
Of all the genes examined, small proline-rich proteins type 2 showed the highest level of activation by oestrogen. Although type 1 is known to be expressed mainly in squamous tissues, there is little known about the type 2 small proline-rich protein. This gene can serve as a sensitive indicator of oestrogen in uterus because it was activated in a dose-dependent manner and could not be activated in ERα-deficient mice.
Although the genes selected in this study need to be understood in greater detail—e.g. to determine whether the rate-limiting factor of the pathway is actually activated by oestrogen and whether the genes have EREs in their promoters, this study demonstrated that DNA microarrays can identify on a transcriptional level a candidate list of genes involved in the uterotrophic effect of oestrogen.
Overall, analysis of the oestrogen-regulated genes revealed at a transcriptional level the switch from fatty acid degradation to synthesis, the change of receptors (i.e. AM/CGRP receptors and leptin receptors), the change of growth hormone activity (activation of IGF-I and kallikrein), and the induction of RNA and protein synthesis. Thus, the importance of the transcriptional switch to biomolecule synthesis (RNA, proteins, and fatty acids) is an important determinant of the uterotrophic effects of oestrogen. Although it takes several days to estimate oestrogen levels using conventional uterotrophic tests, our study identifies genes that may be related to the uterotrophic effect and whose up-regulation can be detected only 6 h after oestrogen administration. Activation of these genes suggests a mechanism for the drastic uterotrophic effects following oestrogen administration, and these findings reveal the diverse effects of oestrogen via ERα activation on cellular transcript levels in vivo.
Animals were housed under a 12-h light, 12-h dark cycle. ERα knockout (αERKO) mice and their wild-type counterparts (C57/BL6/J background) (Lubahn et al. 1993) were ovariectomized at 8 weeks of age. After 2 weeks, the ovariectomized mice were injected with 17β-oestradiol (Sigma; 5 µg/kg b.w.) or oil and whole uteri (n = 4) were collected. All animal experiments were approved by the institutional animal care committee.
Preparation of labelled cRNA and hybridization
Total uterine RNA was extracted using the TRIZOL reagent (InVitrogen, Tokyo, Japan) and purified using the RNeasy total RNA purification kit (Qiagen, Tokyo, Japan). Total RNA (10 µg) was converted into double stranded cDNA using the SuperScript Choice System (InVitrogen, Tokyo, Japan) except that the T7-(dT)24 primer (APB, Tokyo, Japan) was used for first-strand cDNA synthesis. Biotin-labelled cRNA was synthesized using the ENZO BioArray HighYield RNA transcript labelling kit (APB, Tokyo, Japan). The cRNA was purified using RNeasy (Qiagen, Tokyo, Japan) and fragmented in fragmentation buffer (40 mm Tris (pH 8.1), 100 mm KOAc, and 30 mm MgOAC) by heating to 94 °C for 35 min. Fragmented cRNA was mixed with hybridization buffer containing 100 mm MES, 1 m NaCl, 20 mm EDTA and 0.01% Tween 20 and control oligonucleotides. The quality of cRNA was assessed by analysis with the Test 2 array (Affymetrix, APB, Tokyo, Japan) containing housekeeping genes. All preparations met Affymetrix’s recommended criteria for use on their expression arrays.
After checking the quality of the cRNAs, 15 µg of cRNA was hybridized to the high-density oligonucleotide arrays (Mouse U74A, Affymetrix, APB, Tokyo, Japan) for 16 h at 45 °C. Arrays were then washed, stained with streptavidin-phycoerythrein (Molecular Probes), and scanned with an argon-ion laser confocal scanner (APB, Tokyo, Japan).
Scanned data were analysed with GeneChip software (Affymetrix, APB, Tokyo, Japan) and detailed methods for data analysis have been described (Lockhart et al. 1996). Briefly, each gene is represented by the use of 20 perfectly matched (PM) and one-base-mismatched (MM) 25-base oligonucleotides. As the MM probes are used to detect background levels and cross-hybridization signals, the relative level of gene expression is represented by differences in the levels of fluorescent intensity between PM and MM, called Average Difference. To normalize data, Average Differences were adjusted to produce an average intensity that equalled 2500.
Three independent experiments were performed, and values for the mean and standard deviation of the three replicate Average Difference scores were calculated for each gene. For the validation of genes affected by E2 administration, a comparison between oil-treated and E2-treated groups was performed using the Student’s t-test (P < 0.05 considered significant). The fold change of each gene was calculated by GeneChip software (Affymetrix, APB, Tokyo, Japan) based on the ratio of Average Differences between E2 and oil.
Quantitative real time-PCR
Total RNA was purified as described above. cDNA was synthesized from purified total RNA by Superscript II RT (InVitrogen, Tokyo, Japan) with random primers at 42 °C for 60 min. PCR reactions were performed in the PE Prism 5700 sequence detector (PE Biosystems, Tokyo, Japan) using SYBR-Green (Molecular Probes, Eugene, OR, USA) in the presence of appropriate primers according to the manufacturer’s instructions. The assay uses fluorescence emitted by SYBR-Green to quantify double-stranded DNA produced during the PCR reaction. The use of a sequence detector allows continuous measurement of the fluorescent spectra in all 96 wells of the thermal-cycler during PCR amplification. Therefore, amplification reactions were monitored in real time. The model 5700 software constructed amplification plots from extension phase fluorescent emission data collected during PCR amplification. Cτ (threshold) values were calculated by determining the point at which fluorescence exceeds a threshold limit (usually 10 times the SD of the baseline), and the primers were chosen to amplify short PCR products less than 100 base pairs. The primers used in the PCR reactions are listed in Table 3.
Amplification reactions were performed using a series of diluted cDNAs, 0.625 U AmpliTaq Gold and 0.25 U AmpErase uracil N-glycosylase in a buffer containing 50 mm KCl, 10 mm Tris-HCl, 10 mm EDTA, 20 µm d-ATP, dCTP, dGTP, and 40 µm dUTP; 0.7 mm MgCl2, and 15 nm of each primer. For the standard curves of the genes, serial dilutions of a known amount of a cDNA sample were used. The Cτ values of each gene were plotted on these standard curves to obtain the amount of copies present in the initial cDNA sample. Each PCR amplification was performed in triplicate using the following conditions: 2 min at 50 °C and 10 min at 95 °C, followed by a total of 40 two-temperature cycles (15 s at 95 °C and 1 min at 60 °C). Gene expression levels were normalized by the expression levels of 28S (X00525) and 18S (X00686) ribosomal RNA. Gel electrophoresis and melting curve analyses were performed to confirm correct amplicon size and the absence of nonspecific bands.
This work was supported in part by a research grant from Core Research for Evolutional Science of Japan Science and Technology Cooperation, a Grant-in-Aid for Scientific Research of the Japan Society for the Promotion of Science, a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, and a grant for R&D projects in Cooperation with Academic Institutions from New Energy and Industrial Technology and Development Organization (NEDO).
- 1990) Multidrug resistance gene expression is controlled by steroid hormones in the secretory epithelium of the uterus. Mol. Reprod. Dev. 25, 101–109. , , , , & (
- 1995) The influence of 17-beta-oestradiol and the natural oestrous cycle on alpha-adrenoceptor-mediated responses of the cardiovascular system in the rat. J. Pharm. Pharmacol. 47, 656–660. & (
- 1999) The promoter of the rat 3-hydroxy-3-methylglutaryl coenzyme A reductase gene contains a tissue-specific estrogen-responsive region. Mol. Endocrinol. 13, 1225–1236. , , , , & (
- 2000) Thyroid hormone regulation of hepatic genes in vivo detected by complementary DNA microarray. Mol. Endocrinol. 14, 947–955. , , & (
- 2000) Pregnancy and sex steroid hormones enhance circulating calcitonin gene-related peptide concentrations in rats. Hum. Reprod. 15, 949–953. , & (
- 1986) Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 320, 134–139. , , , et al. (
- 1996) RNA polymerase I associated factor 53 binds to the nucleolar transcription factor UBF and functions in specific rDNA transcription. EMBO J. 15, 2217–2226. abs.html. , , , et al. (
- 1991) Insulin-like growth factor receptor gene expression in the rat ovary: divergent regulation of distinct receptor species. Mol. Endocrinol. 5, 1799–1805. , , , et al. (
- 1989) Hormonal regulation of human endometrial stromal cells in culture: an in vitro model for decidualization. Fertil. Steril. 52, 761–768. , , , & (
- 1999) The transcriptional program in the response of human fibroblasts to serum. Science 283, 83–87. , , , et al. (
- 2001) Mechanisms of estrogen receptor action in the myocardium. Rapid gene activation via the ERK1/2 pathway and serum response elements. J. Biol. Chem. 276, 27873–27880. , , , , & (
- 1998) Effect of pregnancy and steroid hormones on plasma adrenomedullin levels in the rat. Can. J. Physiol. Pharmacol. 76, 463–466. & (
- 1992) Rat brain glycolysis regulation by estradiol-17 beta. Biochim. Biophys. Acta 1133, 301–306. & (
- 2001) Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104, 719–730. , , , et al. (
- 1998) Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc. Natl. Acad. Sci. USA 95, 15677–15682. , , , et al. (
- 1996) Cloning of a novel receptor expressed in rat prostate and ovary. Proc. Natl. Acad. Sci. USA 93, 5925–5930. , , , & (
- 2001) Transcriptional profiling reveals global defects in energy metabolism, lipoprotein, and bile acid synthesis and transport with reversal by leptin treatment in ob/ob mouse liver. J. Biol. Chem. 276, 49066–49076. & (
- 2002) Activation of Antimetastatic Nm23–H1 Gene Expression by Estrogen and Its alpha-Receptor. Endocrinology 143, 467–475. , , & (
- 2001) Leptin receptor 5′untranslated regions in the rat: relative abundance, genomic organization and relation to putative response elements. Mol. Cell Endocrinol. 172, 37–45. , , , , & (
- 1996) Expression monitoring by hybridization to high-density oligonucleotide arrays. Nature Biotechnol. 14, 1675–1680. , , , et al. (
- 1993) Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc. Natl. Acad. Sci. USA 90, 11162–11166. , , , , & (
- 1998) Drug target validation and identification of secondary drug target effects using DNA microarrays. Nature Med. 4, 1293–1301. , , , et al. (
- 1987) Estrogen induces insulin-like growth factor-I expression in the rat uterus. Mol. Endocrinol. 1, 445–450. , & (
- 1995) 17beta-Estradiol inhibits the voltage-dependent 1-type Ca2+ currents in aortic smooth muscle cells. Eur. J. Pharmacol. 294, 625–635. , , , , & (
- 2000) 17beta-estradiol inhibits apoptosis in MCF-7 cells, inducing bcl-2 expression via two estrogen-responsive elements present in the coding sequence. Mol. Cell. Biol. 20, 2890–2901. , , , (
- 1995) Insulin-like growth factor binding protein (IGFBP) proteases: functional regulators of cell growth. Prog. Growth Factor Res. 6, 273–284. , , , , , (
- 2001) Sterol regulatory element-binding proteins induce an entire pathway of cholesterol synthesis. Biochem. Biophys. Res. Commun. 286, 176–183. , , , et al. (
- 1994) Estrogen regulation of the insulin-like growth factor I gene transcription involves an AP-1 enhancer. J. Biol. Chem. 269, 16433–16442. , , , et al. (
- 1998) Enhanced expression of the 8-oxo-7,8-dihydrodeoxyguanosine triphosphatase gene in human breast tumor cells. Cancer Lett. 125, 123–130. , , (
- 1993) SREBP-1, a basic-helix-loop-helix-leucine zipper protein that controls transcription of the low density lipoprotein receptor gene. Cell 75, 187–197. , , , et al. (