| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Global Business Intelligence Center (E.J.); Department of Molecular Design and Informatics (J.P.); and Target Discovery Unit (C.v.R., C.v.d.H., J.W.), Department of Pharmacology (S.M.), NV Organon, 5340 BH Oss; Center of Reproductive Medicine (J.S.E.L., B.C.J.M.F.), Erasmus Medical Center, 3000 CA Rotterdam; Department of Obstetrics and Gynecology (H.B.R.D.), Flevohospital, 1315 RA Almere; and Department of Reproductive Medicine (B.C.J.M.F.), University Medical Center Utrecht, 3508 GA Utrecht, The Netherlands
Address all correspondence and requests for reprints to: Erik Jansen, NV Organon, PO Box 20, 5340 BH Oss, The Netherlands. E-mail: erik.jansen{at}organon.com.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In PCOS ovaries, growth of early antral follicles is typically arrested at the 5- to 10-mm stage, resulting in ovaries with multiple follicular structures less than 10 mm in diameter (6, 7). The volume and density of the ovarian stroma are increased (8), and there is rarely evidence of recent ovulations. The number of primordial follicles is similar compared with normal ovaries. An increased amount of ripening as well as atretic 2- to 10-mm follicles can be observed (9, 10), which is supported by recent observations indicating elevated serum concentrations of anti-Müllerian hormone, a specific marker for preantral and early antral ovarian follicles, in these patients (11). Studies involving PCOS theca cells in culture have documented a biochemical (12) and molecular (13) phenotype distinctly different from cells derived from regularly cycling women.
The existing literature indicates familial clustering of PCOS (14). The mode of inheritance of the disorder is still uncertain. Although initially a single autosomal-dominant pattern of transmission was proposed, recent studies are indicative of a more complex mode of inheritance (15, 16). Genetic studies have been hampered due to heterogeneity in the phenotype of PCOS patients. Moreover, the availability of only small numbers of sibpairs and the lack of an unambiguous male phenotype along with the absence of an appropriate animal model have complicated the elucidation of the genetic basis of PCOS. Most genetic studies undertaken so far have taken a candidate gene approach in cultures of isolated cells, focusing on genes involved in folliculogenesis [myeloid cell leukemia-1 (17), growth differentiation factor-9 (18), or plasma protein A (19)], steroidogenesis [such as steroid acute regulatory protein (20), or cytochrome P450 17 (21, 22)] and genes involved in insulin signaling (23, 24, 25). As stated earlier, it seems highly likely that a multigenic defect is to be held responsible for PCOS. However, to date, there are no studies that describe genes specifically linked to PCOS susceptibility.
Today’s technology (e.g. microarray analysis) enables us to generate a gene expression profile in any given tissue or cell type of interest (26, 27). It is therefore possible to assess the expression pattern of tens of thousands of genes in a single experiment. This experimental approach represents a new method to define the molecular phenotype of a disease. Moreover, this technology has recently been used successfully to identify altered gene expression in other disease conditions such as multiple sclerosis (28) and type 2 diabetes (29). The current study describes the first analysis of differences in gene expression profiles in ovarian tissue obtained from PCOS patients, female-to-male transsexuals, and normoovulatory individuals. Abnormal pathways are identified in PCOS, which may potentially lead to the identification of candidate genes for subsequent evaluation. Consequently, these transcription profiles might reveal novel critical information regarding the molecular basis of ovarian dysfunction in PCOS patients.
Because there is no suitable animal model, we took advantage of several interesting observations in humans that were reported more than a decade ago. It has been described that the ovarian morphology (30) and endocrine profile (7) in long-term androgen-treated female-to-male transsexuals (TSX) resembles PCOS, providing supportive evidence that androgens play a critical role in the pathogenesis of PCOS. However, these similarities between PCOS and TSX ovaries were described at the level of morphology and endocrinology. No studies address changes in both PCOS and TSX at the molecular level. The TSX may represent a unique human model for the study of PCOS despite the fact that TSX ovaries are smaller with more atretic follicles (presumably due to suppressed gonadotropins resulting from exogenous androgen administration).
The comparison of gene expression profiles between PCOS and normal ovaries may also reveal novel information regarding genes associated with early and later stages of antral follicle development, ovulation, and corpus luteum formation under normal conditions. Moreover, the identification of pathways that are abnormal in PCOS may contribute to the understanding of the pathogenesis of PCOS and might reveal novel avenues for therapeutic intervention.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The PCOS patient group consisted of six women with a median age of: 29.8 yr (range, 27.3–32.4). Their median BMI was 20.9 kg/m2, ranging from 17.7 to 33.5 kg/m2. PCOS women had a median bleeding interval of 160 d ranging from 90–199 d. Three of six women (50%) were amenorrheic, whereas the other three were oligomenorrheic. LH serum levels were elevated (median, 12.8 IU/liter; range, 5.8–18.1), whereas FSH concentrations were within the normal range (median, 5.5 IU/liter; range, 4.5–7.5). Median T levels were 2.6 nmol/liter (range, 1.1–3.2) being within the normal range. The free androgen index (FAI) was elevated (i.e. 5.5; range, 1.1–13.5). The mean ovarian volume in PCOS patients was increased (i.e. 20.6 ml; range, 11.7–53.0) as was the total follicle number (median, 42; range, 22–60). Ovarian biopsies were collected randomly because these patients did not have a regular menstrual cycle. Histological evaluation of two samples revealed the characteristic histological PCOS picture showing multiple small subcortical follicles and an increased stromal hyperplasia (data not shown). In two of six PCOS women, both biopsies were analyzed whereas in the remaining four patients only one biopsy, randomly chosen, was processed for microarray analysis.
The TSX group consisted of 15 female-to-male transsexuals (TSX) aged between 20 and 43 yr with a median age of 29.5 yr. Their median BMI was 24.2 kg/m2 ranging from 20–33 kg/m2. TSX subjects had a median menstrual cycle length of 28 d ranging from 26–32 d. Median LH and FSH serum concentrations, before testosterone (T) treatment was initiated, were 4.2 IU/liter (range, 1.1–7.1) and 2.5 IU/liter (range, 2.5–7.0), respectively. T levels before T therapy were within the normal range (i.e. 1.8 nmol/liter; range, 1.0–2.6). Ovaries were collected at random because these patients did lack regular menstrual periods due to the exogenous androgen therapy. Histological evaluation revealed multiple cystic follicles and pronounced theca cell hyperplasia in all ovaries (data not shown). In seven of 15 TSX subjects both ovaries were analyzed, whereas in the remaining eight patients only one ovary, randomly chosen, was processed for microarray analysis.
There was a significant difference in age (P < 0.001) between normoovulatory controls, on one hand, and PCOS women and TSX individuals, on the other. There was no significant difference in BMI between the three groups. Comparing endocrine data from PCOS patients with those in TSX subjects revealed a significant difference (P < 0.001) in LH levels, whereas FSH, T and 17ß-estradiol (E2) levels were similar. The FAI was significantly (P < 0.001) higher in PCOS women compared with TSX subjects.
Microarray Data Analysis
We compared the gene expression profiles of the luteal- and follicular-phase samples of the normal women. The principal component analysis indicated that there are no major differences between the two samples. The sets do not form two different groups, i.e. they do not constitute two separate entities within the control sample group (data not shown).
In addition, to substantiate the relative normality of our control samples, we compared our gene array data with the gene array data as present in the so-called GeneExpress datasuite from GeneLogic, Gaithersburg, MD (www.genelogic.com). The resulting gene expression profiles matched very well with the data obtained within our controls (i.e. comparable list of regulated genes).
To assess whether we would be able to identify expression profiles indicative of a molecular signature of PCOS ovaries, we performed a principal component analysis (PCA). The PCA was performed for all ovary samples using the normalized fluorescence intensity data from 500 representative transcripts. PCA reduces redundant variables in complex data sets to principal components that represent the majority of the variability in the data (31). When the PCA data are plotted, the eight PCOS samples cluster together within a discrete region that is clearly distinct from the 11 normal ovary samples (Fig. 1
). In addition, the 22 TSX samples are clustered close to the PCOS samples. A distinctly different gene expression profile of one particular PCOS individual can be observed in Fig. 1. Interestingly, this patient is the only patient that conceived spontaneously after bilateral ovarian biopsies were taken. She also experienced a severe ovarian hyperstimulation syndrome twice during previous ovulation induction using gonadotropins in a low-dose step-up protocol.
|
). The two diagonals indicate the 1.8-fold cutoff in differential expression. Briefly, the data show that 78% of the regulated genes are expressed at a lower level in PCOS (Fig. 2A), compared with the normal samples. In addition, in TSX ovaries 66% of the regulated genes are also expressed at a lower level compared with normal samples (Fig. 2B). Moreover, TSX samples do show a similar expression pattern as PCOS samples.
|
0.01. These fold-change levels of differential expression are average ratios resulting from consistent changes in many samples. Due to redundancy of the oligonucleotide probes on the U133A and U133B chips, the 230-probe fragments represented 189 genes that are differentially expressed in PCOS as compared with the controls. Several structure-function categories were represented in the list of differentially expressed genes including, for example, membrane receptors, transcription factors, and extracellular matrix (ECM) components. Several genes could be classified into specific signal transduction cascades or gene networks. Some of these genes and their role in ovarian function, such as nuclear growth factor I-B (NGFI-B) (32), CCAAT enhancer binding protein ß (C/EBPß) (33), p21Cip (34), and serum and glucocorticoid-regulated kinase (SGK) (35), have been described previously. In Table 1 we have listed all the genes that are differentially expressed in PCOS ovaries, and we have also included the corresponding expression profile of TSX ovaries. The data show that the majority of PCOS-deregulated genes are also deregulated in TSX, confirming the PCA data. When analyzing the regulated genes for their chromosomal location, it becomes apparent that multiple chromosomes are more frequently involved. When corrected for chromosome size and each chromosome’s individual total predicted gene content, the PCOS-regulated genes are most frequently found on chromosome 6, 8, 14, 19, 21, 22, and, to a somewhat lesser extent, on chromosome 14 (Fig. 3). Moreover, chromosome 19 displays the highest relative score.
|
|
, it is striking to see that many genes, such as Jun D, Fra-2, and EGR-1, which have a lower expression in PCOS and TSX ovaries, are positive regulators of cell proliferation. As to be expected, several genes, such as SGK and p21Cip, which are activated by the LH surge or which are predominantly expressed in preovulatory follicles, are decreased in PCOS ovaries.
Microarray Data Validation
Using real-time quantitative PCR, we have validated five gene transcript expression levels that were altered at least 1.8-fold in microarray analysis (Fig. 4). For each of the five genes, mRNA levels in qualitative PCR (Q-PCR) showed trends similar to those of the microarray experiment in both PCOS and TSX samples. In addition, because it has been shown that sFrp4 is up-regulated in luteinized granulosa cells (36) and sFrp4 was not found in our list of regulated genes (Table 1), we performed a Q-PCR experiment to verify this. The results indicate no differential expression in normal ovaries compared with PCOS and TSX (data not shown). Therefore, the deselection of sFrp4 is not the result of too-stringent selection and cutoff criteria that we applied in generating the list of regulated genes but seems to represent a real phenomenon.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The present study shows that the majority of the differentially expressed genes are down-regulated in PCOS ovaries, suggesting that several processes that are activated by these genes are in a "off-state." This may not be surprising because it is consistent with a less active state of PCOS ovaries with respect to the later follicular stages (i.e. dominant and preovulatory) as compared with controls. Accordingly, we identified several down-regulated genes from which it is well established that they are either induced by LH, or their expression levels are elevated in the post-LH surge and/or periovulatory window. These represent markers such as C/EBPß, EGR-1, NGFI-B, Jun D, and ADAMTS-1. Their relative underexpression is in accordance with the absence of ovulation in PCOS. Remarkably, androgen-regulated genes such as Pem (37) and OTEX (38) were not differentially expressed in the present study. This might be explained by a stable status of PCO, exposed to elevated androgen levels for an extended period of time. Such a condition could result in a decline of the androgen response at the level of gene transcription. Again, the possibility cannot be excluded that observed differences might be due to differences in cell populations studied, because the present samples constitute a mixture of follicular (theca, granulosa, oocyte) and stromal components. For example, the selection of ECM genes that we identified as being differentially expressed may result from genes expressed in the stromal/epithelial compartment and therefore might not be differentially expressed in the in vitro theca cell cultures. Finally, theca cells might behave differently in vitro compared with the in vivo situation in which they are embedded and among other endocrine cells.
By comparing the transcriptome of PCOS ovaries and normal ovaries, we set out to identify gene networks and signal transduction pathways that are abnormal in PCOS. These deviant pathways may play important roles in arrested follicle development and hence represent a key feature of the PCOS phenotype. This approach might lead to a series of candidate genes, which might subsequently be evaluated and validated using other genetic approaches. When browsing the table of genes with altered expression in PCOS ovaries, several aberrant pathways can be identified, and these are discussed below.
Wnt Signaling
At least four genes that are linked to Wnt signaling show altered expression in PCOS: 1) Wnt-1-responsive Cdc42 homolog (up); 2) dickkopf homolog 3 (down); 3) Wnt5A (up); and 4) carboxypeptidase Z (up). Moreover, dickkopf is a Wnt signaling inhibitor (39), and very recently it has been shown that carboxypeptidase Z also inhibits Wnt signaling (40). The spatial and temporal distribution of activating and inhibitory factors determines whether Wnt signaling is activated or repressed in PCOS ovaries. In a recent microarray study using theca cells from PCOS and controls, Wnt signaling was also shown to be altered (13). Moreover, several reports demonstrate that Wnt signaling factors are expressed and hormonally regulated in the adult rat ovary (36, 41, 42).
ECM Components and ECM Remodeling Factors
At least 10 genes that are linked to the ECM are differentially expressed in PCOS: fibulin 1; biglycan; nidogen 2; thrombospondin 1 (THBS1); a disintegrin and metalloprotease with thrombospondin motifs-1 (ADAMTS1); heparan sulfate (glucosamine) 3-O-sulfotransferase 1; UDP-Gal:ßGlcNAc ß 1,4-galactosyltransferase, polypeptide 6 (B4GALT6); osteomodulin; collagen type IV and ADAM12. Nidogen 2 is a potential ligand of fibulin 1 (43). All factors, except fibulin, ADAMTS1, and collagen type IV are up-regulated in PCOS ovaries. In TSX ovaries, only B4GALT6, osteomodulin, and ADAM12 are up-regulated. It is tempting to speculate that the altered expression of these genes in PCOS is related to the process of cyst formation and/or physical properties of the thick ovarian capsule. In addition, THBS1 expression is associated with local neovascularization processes in the ovary and its expression is high in early antral follicles. An increased amount of these follicles is present in PCOS ovaries compared with controls (10, 11).
Immune System Components
At least 17 genes that are linked to the immune system show altered expression in PCOS: Ig 
joining 3; Ig 
constant; major histocompatibility complex (MHC), class II, DR
; MHC class II, DP 1; MHC class II, DR ß 1 and 5; MHC class II, DQ 1; MHC class II, DQ ß 1; MHC class II, DM ß; MHC class II, DM ; CD74; T cell receptor ß locus; complement component C3; suppressor of cytokine signaling 3; transcription factor 8; chemokine ligand 3. All factors are expressed at a lower mRNA level in PCOS ovaries. TSX ovaries display a similar profile, although all but one of the MHC genes are not regulated. It has become apparent that in the ovary the immune system contributes to the regulation of gonadal function (44, 45). Immune cells are not detectable in the granulosa layer until the LH surge occurs. MHC class II DR is highly expressed in human large luteinized cells (46). It has been proposed that luteinized cells communicate with T cells during formation and subsequent demise of the corpus luteum. Moreover, the actual rupture process of the follicle displays characteristics of an inflammatory reaction. In addition, it has been shown that MHC class II molecules (e.g. HLA DR) are expressed by granulosa cells after luteal transformation (47, 48, 49). The observed lower expression of these immune system components in PCOS and TSX ovaries relates to the fact that these periovulatory processes are almost absent in ovaries of PCOS patients.
Apoptosis and Stress Response
A large number of genes that have a lower expression in both PCOS and TSX ovaries are positive regulators of cell proliferation. We identified the phosphatase DUSP1 (dual specificity phosphatase 1) as being strongly down-regulated in PCOS ovaries as well as TSX ovaries. This DUSP1 expression profile is comparable to what has been published on NGFI-B displaying a rapid and transient stimulation by human chorionic gonadotropin (32). The downstream substrates of DUSP1 are members of p38 MAPK, the Erk and c-Jun N-terminal kinase pathways. Because these are inactivated by the phosphatase activity of DUSP1, the reduced DUSP1 mRNA levels in PCOS ovaries might therefore (transiently) activate p38MAPK, Erk, and c-Jun N-terminal kinase signaling. Activation of these pathways is required to mediate apoptosis induced by proinflammatory cytokines such as TNF (50). Interestingly, TNFRSF25, a member of the TNF receptor superfamily involved in inducing apoptosis (51), is expressed at a higher level in PCOS ovaries.
Heat Shock Proteins (HSPs)
Five genes encoding heat shock proteins are expressed at a reduced level in PCOS: DnaJ (HSP40) homolog, subfamily B, member 1; heat shock 70-kDa protein 6 (HSP70B); heat shock 70-kDa protein 1A; heat shock 70-kDa protein 1B; and heat shock 90-kDa protein 1 . The 70-kDa stress protein family (HSP70) plays important roles in a variety of physiological processes, including protein chaperoning, steroidogenesis, and protection against apoptosis and is a general mediator of cellular stress responses. In summary, some apoptotic processes might be activated in PCOS ovaries as compared with controls. Indeed it has been suggested that the number of atretic follicles is increased in PCOS patients (52). In addition, induction of HSP70 is associated with inhibition of hormone-sensitive, LH-induced steroidogenesis via inhibition of steroid acute regulatory protein synthesis. The down-regulated HSP70 in PCOS therefore does not interfere with steroid synthesis. This is consistent with the observation that androgenic steroid production is even higher in PCOS as compared with control individuals (12).
Peroxisome Proliferator-Activated Receptor (PPAR)
and Histone Deacetylase 3 (HDAC3) in Proliferation/Differentiation
HDAC3 has an important function in regulating PPAR activity (53). HDAC inhibition is more effective in suppressing growth and inducing differentiation when treatment is combined with PPAR agonists. Actually, in PCOS ovaries we identified HDAC3 as down-regulated and PPAR as up-regulated. This might result in locally enhanced differentiation-promoting activity in PCOS ovaries and may be involved in the process of follicle arrest in PCOS. Expression of both genes is not altered in TSX. In this context it is important to note that a commonly used antiepileptic drug, valproic acid (VPA), has both HDAC-inhibitory activity and agonistic activity on PPAR (54, 55). Interestingly, VPA treatment is associated with PCOS-like features because these patients develop hyperandrogenism and PCO during VPA treatment (56, 57). Moreover, VPA does increase androgen synthesis in ovarian theca cells (58). Finally, PCOS-like features disappear when VPA therapy is discontinued. In addition, two independent studies show that polymorphism in the PPAR gene might be associated with PCOS (59, 60, 61).
Until now the etiology of PCOS remains unknown. When the differentially regulated genes in PCOS and TSX are analyzed for their chromosomal location, it becomes apparent that chromosome 19 displays the highest relative score. This is intriguing, because several studies provide support for a genetic basis for PCOS (62, 63) and demonstrated linkage to chromosome 19p13.3. Our microarray experiments revealed two genes that are located in this chromosomal interval: complement component 3 and growth arrest and DNA damage-inducible ß (GADD45B). However, we do not yet have functional links that associate one of these genes with PCOS. The identification of pathways that are abnormal in PCOS may contribute to understanding the pathophysiology of PCOS and the identification of causative molecular factors and therefore might reveal novel leads for therapeutic intervention.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In normoovulatory controls uni- or bilateral oophorectomy was performed because subjects were either carriers of a known gene mutation (BRCA 1 or 2) or their family history indicated a severely increased chance of ovarian cancer (64) or because of other gynecological conditions. Only women with regular menstrual cycles (21–35 d) and younger than 40 yr were included in this study. The BMI (kg/m2) was recorded in all patients. Women with PCO, defined as more than 12 follicles smaller than 10 mm present in at least one ovary, detected by ultrasound, were excluded. Similarly, if ovarian cancer was shown to be present in one or both ovaries, patients were excluded.
Women between 20 and 40 yr of age and having PCOS were recruited from the infertility clinic of the Erasmus Medical Center. PCOS was defined according to the latest PCOS consensus (1, 2). Briefly, women should have irregular menstrual cycles and/or clinical or endocrine hyperandrogenism [FAI (T x 100/sex hormone binding globulin (SHBG) > 4.5] and/or PCO on ultrasound. All patients with at least two of the four aforementioned criteria were considered to have PCOS and were included. All patients had a long standing history of infertility, and all underwent ovulation induction without achieving a pregnancy. All patients previously developed moderate to severe ovarian hyperstimulation syndrome. In all PCOS patients uni- or bilateral ovarian biopsies were performed.
TSX individuals underwent hysterectomy and bilateral oophorectomy because of their sex reversal. Only subjects aged between 20 and 40 yr with a history of regular menstrual cycles (21–35 d) and without a previous history of hyperandrogenemia were included. In all TSX subjects, 250 µg T (Sustanon, Organon, Oss, The Netherlands) had been administered im once biweekly during at least half a year before surgery was performed.
Interventions
Standardized initial screening (clinical investigation, transvaginal ultrasound, and fasting blood withdrawal) was performed between 0900–1100 h, as previously described (65). For sonographic imaging we used a 6.5-MHz vaginal transducer (model EUB-415, Hitachi Medical Corp., Tokyo, Japan). The ovaries were localized and scanned as described previously (65). Ultrasonography as well as endocrinological screening was only performed in PCOS patients and TSX individuals. In PCOS patients blood withdrawal was performed on a random day. In TSX individuals blood was withdrawn before T treatment was initiated.
Bilateral oophorectomy in normoovulatory controls was performed laparoscopically under general anesthesia. Each ovary was removed using ultrasonic scissors. After cutting the suspending ligaments, one ovary was removed from the peritoneal cavity through a small incision in the lateral abdominal wall. The remaining ovary was thereafter removed by a similar procedure. In TSX individuals oophorectomy was performed either via the abdominal or vaginal route using standard gynecological techniques. Ovarian biopsies from both ovaries were taken in PCOS patients during laparoscopy. Biopsies constituted minimal wedges taken at the equatorial plane at the antimesovarian edge of each ovary. Care was taken that both cortical and stroma components were collected. Generally, biopsies constituted about one tenth of each ovary and had a pyramidal shape with the top of the pyramid being located at the ovarian hilus. Hemostasis was achieved using bipolar scissors. All tissue samples were processed immediately after collection. Biopsies were snap frozen in liquid nitrogen immediately after collection, whereas whole ovaries were cut in quarters. A randomly selected quarter was subsequently snap frozen. Sections for the remaining tissue samples were histologically processed and microscopically evaluated according to standard procedures (30).
Hormone Assays
Blood samples were obtained by venapuncture and processed within 2 h after withdrawal. Serum was stored at –20 C and assayed for LH, FSH, androstenedione (AD), T, SHBG, dehydroepiandrosterone (DHEA), dehydroepiandrosteronesulfate (DHEAS), and E2. Serum levels of LH, FSH, and SHBG were assessed using luminescence-based immunoassays (Immulite, Diagnostic Products Corp., Los Angeles, CA), whereas serum E2, T, AD, DHEA, and DHEAS levels were measured using coated tube RIAs provided by the same supplier. Intra- and interassay coefficients of variation were less than 5% and 15% for LH, less than 3% and 8% for FSH, less than 8% and 11% for AD, less than 3% and 6% for DHEA and DHEAS, less than 3% and 5% for T, less than 5% and 7% for E2, and less than 4% and 5% for SHBG, respectively.
RNA Sample Preparation
Total RNA was extracted from the biopsies using TRIZOL reagent (Invitrogen, Breda, The Netherlands) as per the manufacturer’s instructions. After TRIZOL purification, RNA was repurified using the RNeasy Mini Kit (QIAGEN, Venlo, The Netherlands) according to the manufacturer’s protocol. Total RNA was quantified by OD260/OD280 measurement and with the RiboGreen RNA Quantitation Kit (Molecular Probes, Leiden, The Netherlands). RNA quality was assessed on a 6000 nanochip using the Agilent 2100 bioanalyzer (Agilent Technologies, Waldbronn, Germany).
cDNA Synthesis and Q-PCR
To 1 µg total RNA comprising a pooled sample (either PCOS, TSX, or normal), 0.5 µg Oligo(dT)15 primer (Promega, Leusden, The Netherlands) and 1 µg random hexamers pd(N)6 (Amersham Biosciences, Roosendaal, The Netherlands) were added in a total volume of 12 µl. The mixture was heated at 70 C for 10 min and quickly chilled on ice for 5 min. cDNA was synthesized in a total volume of 20 µl containing 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM dithiothreitol, 0.5 mM deoxynucleotide triphosphates and 200 U Superscript II RNase H– Reverse Transcriptase (Invitrogen). After incubation for 1 h at 37 C, the enzyme was inactivated by heating for 10 min at 95 C. The cDNA was diluted to a concentration equivalent to 2 ng/µl equivalent RNA.
Quantitative-RT-PCR was performed using cDNA equivalent to 10 ng RNA in a total of 25 µl PCR mix. The total mix contained cDNA, 300 nM forward primer, 300 nM reverse primer, and 1x SYBRgreen PCR Master Mix. The 2x SYBRgreen PCR Master Mix [Applied Biosystems (ABI), Nieuwekerk a/d IJssel, The Netherlands] is optimized for SYBRgreen reactions and contains SYBRgreen I Dye, AmpliTaq Gold DNA Polymerase, deoxynucleotide triphosphates with dUTP, passive reference, and optimized buffer components. The Q-PCR was performed in a MicroAmp optical 96-well reaction plate (Applied Biosystems) with an ABI prism optical adhesive cover in the ABI PRISM 7900HT sequence detection system. The program used was 10 min at 95 C (100% ramp), 40 cycles of 15 sec at 95 C (100% ramp), and 1 min at 60 C (100% ramp), followed by a dissociation curve step of 15 sec at 95 C (100% ramp), 15 sec at 60 C (100% ramp), and 15 sec at 95 C (2% ramp).
The following forward (F) and reverse (R) primer pairs were used: DUSP1–354F, 5'-CAACGAGGCCATTGACTTCATA-3'; DUSP1–422R, 5'-GCCTGGCAGTGGACAAACA-3'; Wnt5A-189F, 5'-CATGAACCTGCACAACAACGA-3'; Wnt5A-296R, 5'-CAGCATGTCTTCAGGCTACATGA-3'; ADAMTS-1–75F, 5'-GCAGAGCACTATGACACAGCAAT-3'; ADAMTS-1–158R, 5'-ATCAGCCATCCCAAGAGTATCAC-3'; TPA-170F, 5'-GGGCACAGTGCCACTCAGT-3'; TPA-272R, 5'-TGGCACACGAAATCTGAGAAGT-3'; PAI-240F, 5'-GGGTGAAGACACACACAAAAGGT-3'; PAI-352R, 5'-CTTCCACTGGCCGTTGAAGTA-3'; Sfrp4–409F, 5'-GGACCTCCCGGAGGATGTTA-3'; Sfrp4–477R, 5'-TCAAGAGGCCTTTCCTGTACCA-3'.
Microarray Hybridization, Data Processing, and Statistical Analysis
GeneChips HG_U133A and HG_U133B (Affymetrix, High Wycombe, UK), together encompassing more than 45,000 human DNA fragments, were hybridized at the Organon Gene Chip Platform (Newhouse, UK). Details of chip content are available at the NetAffx analysis center (http://www.affymetrix.com/analysis/index.affx) (66). Biotin-labeled cRNA, which was generated from eight different PCOS ovary samples, 22 TSX ovary samples, and 11 different normal ovary samples was fragmented and used for hybridizations according to Affymetrix protocols. Hybridized chips were scanned and data automigrated into Rosetta Resolver (Rosetta Biosoftware, Kirkland WA). A Chip Quality Report was evaluated for abnormal glyceraldehyde-3-phosphate dehydrogenase 3'-5' ratios, average background, and glyceraldehyde-3-phosphate dehydrogenase absolute signals.
PCA was performed with a guide ("Analyze experiments using PCA") within Spotfire DecisionSite 7.2 (Spotfire, Göteborg, Sweden). A representative set of 500 fragments was selected by K-means clustering. The resulting table of 500 rows (genes) and 41 columns (ovary samples) was transposed and PCA was run to detect and reduce the number variables to three principal components which represent the majority of the variability in the dataset. A three-dimensional scatterplot was produced to visualize the differences in sample type (i.e. PCOS or TSX or normal) based on each sample’s gene expression profile.
Rosetta Resolver allows normalization of sample data after selection of the appropriate set of samples (e.g. PCOS and normal ovary samples) for calculation of one-way ANOVA. A one-way ANOVA with build ratio was calculated to identify changes in expression levels between the two sample sets. Three criteria were used to define genes that had altered mRNA abundance in PCOS or TSX: 1) the absolute fold change had to be 1.80, or 2) the corresponding P value for the fold change had to be 0.01 or less, and 3) the number of present calls within a sample group must be sufficiently high. All Affymetrix Microarray Suite version 5.0 (MAS5)-processed expression signals contain a corresponding P value for the significance of expression. A P value of 0.05 or less was considered as "present" and higher levels as "absent." For up-regulated genes in PCOS samples, at least six of eight samples should have present calls and for down-regulated genes in PCOS at least eight of 11 normal samples should have present calls. The TSX sample group must contain at least 16 present calls for the TSX up-regulated genes. With the list of PCOS deregulated genes we searched gene by gene in the Ovarian Kaleidoscope database (http://ovary.stanford.edu/) (67) for additional information on ovarian biology that might be relevant in our studies. This database provides information regarding the biological function, expression pattern, and regulation of genes expressed in the ovary. When applicable, this supplementary information is appended.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: AD, Androstenedione; ADAMTS-1, a disintegrin and metalloprotease with thrombospondin motifs-1; BMI, body mass index; C/EBPß, CCAAT enhancer binding protein ß; DHEA, dehydroepiandrosterone; DHEAS, dehydroepiandrosteronesulfate; DUSP1, dual specificity phosphatase 1; E2, estradiol; ECM, extracellular matrix; FAI, free androgen index; HDAC, histone deacetylase; MHC, major histocompatibility complex; NGF, nuclear growth factor; PCA, principal component analysis; PCO, polycystic ovaries; PCOS, polycystic ovary syndrome; PPAR, peroxisome proliferator-activated receptor; SGK, serum and glucocorticoid-regulated kinase; SHBG, sex hormone-binding globulin; T, testosterone; TSX, androgen-treated female-to-male transsexual; VPA, valproic acid.
Received for publication February 20, 2004. Accepted for publication August 5, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-hydroxylase and steroidogenic acute regulatory protein gene promoters in normal and polycystic ovary syndrome theca cells. J Clin Endocrinol Metab 85:2304–2411-hydroxylase/C(17–20) lyase promoters. Fertil Steril 77:274–280[CrossRef][Medline]
and inhibitors of histone deacetylase in adenocarcinoma of the lung. Clin Cancer Res 8:1206–1212
). Toxicol Appl Pharmacol 160:238–249[CrossRef][Medline]
gene. J Clin Endocrinol Metab 87:772–775 gene in women with polycystic ovary syndrome. Hum Reprod 18:540–543 polymorphisms in polycystic ovary syndrome. J Clin Endocrinol Metab 88:5887–5892This article has been cited by other articles:
 |
|
 |
|
  V. M. Victor, M. Rocha, C. Banuls, M. Sanchez-Serrano, E. Sola, M. Gomez, and A. Hernandez-Mijares Mitochondrial Complex I Impairment in Leukocytes from Polycystic Ovary Syndrome Patients with Insulin Resistance J. Clin. Endocrinol. Metab., September 1, 2009; 94(9): 3505 - 3512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Dissen, C. Garcia-Rudaz, A. Paredes, C. Mayer, A. Mayerhofer, and S. R. Ojeda Excessive Ovarian Production of Nerve Growth Factor Facilitates Development of Cystic Ovarian Morphology in Mice and Is a Feature of Polycystic Ovarian Syndrome in Humans Endocrinology, June 1, 2009; 150(6): 2906 - 2914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Legro and J. M. McAllister Heirarchical Clustering and Beyond in PCOS Endometrium: Brave New World J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1084 - 1085. [Full Text] [PDF]
|
|
|
|
|
|
J. Y. Kim, H. Song, H. Kim, H. J. Kang, J. H. Jun, S. R. Hong, M. K. Koong, and I. S. Kim Transcriptional Profiling with a Pathway-Oriented Analysis Identifies Dysregulated Molecular Phenotypes in the Endometrium of Patients with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1416 - 1426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kenigsberg, Y. Bentov, V. Chalifa-Caspi, G. Potashnik, R. Ofir, and O. S. Birk Gene expression microarray profiles of cumulus cells in lean and overweight-obese polycystic ovary syndrome patients Mol. Hum. Reprod., February 1, 2009; 15(2): 89 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Franks, J. Stark, and K. Hardy Follicle dynamics and anovulation in polycystic ovary syndrome Hum. Reprod. Update, May 22, 2008; (2008) dmn015v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.O. Goodarzi, N. Xu, J. Cui, X. Guo, Y.I. Chen, and R. Azziz Small glutamine-rich tetratricopeptide repeat-containing protein alpha (SGTA), a candidate gene for polycystic ovary syndrome Hum. Reprod., May 1, 2008; 23(5): 1214 - 1219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Das, O. Djahanbakhch, B. Hacihanefioglu, E. Saridogan, M. Ikram, L. Ghali, M. Raveendran, and A. Storey Granulosa Cell Survival and Proliferation Are Altered in Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2008; 93(3): 881 - 887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Natesampillai, J. Kerkvliet, P. C. K. Leung, and J. D. Veldhuis Regulation of Kruppel-like factor 4, 9, and 13 genes and the steroidogenic genes LDLR, StAR, and CYP11A in ovarian granulosa cells Am J Physiol Endocrinol Metab, February 1, 2008; 294(2): E385 - E391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. E. Beshay, J. C. Havelock, R. Sirianni, P. Ye, T. Suzuki, W. E. Rainey, and B. R. Carr The Mechanism for Protein Kinase C Inhibition of Androgen Production and 17{alpha}-Hydroxylase Expression in a Theca Cell Tumor Model J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4802 - 4809. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. R. Dunning, M. Lane, H. M. Brown, C. Yeo, R. L. Robker, and D. L. Russell Altered composition of the cumulus-oocyte complex matrix during in vitro maturation of oocytes Hum. Reprod., November 1, 2007; 22(11): 2842 - 2850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Skov, D. Glintborg, S. Knudsen, T. Jensen, T. A. Kruse, Q. Tan, K. Brusgaard, H. Beck-Nielsen, and K. Hojlund Reduced Expression of Nuclear-Encoded Genes Involved in Mitochondrial Oxidative Metabolism in Skeletal Muscle of Insulin-Resistant Women With Polycystic Ovary Syndrome Diabetes, September 1, 2007; 56(9): 2349 - 2355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Russell and R. L. Robker Molecular mechanisms of ovulation: co-ordination through the cumulus complex Hum. Reprod. Update, May 1, 2007; 13(3): 289 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Wood, D. A. Dumesic, D. H. Abbott, and J. F. Strauss III Molecular Abnormalities in Oocytes from Women with Polycystic Ovary Syndrome Revealed by Microarray Analysis J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 705 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Corton, J. I. Botella-Carretero, A. Benguria, G. Villuendas, A. Zaballos, J. L. San Millan, H. F. Escobar-Morreale, and B. Peral Differential Gene Expression Profile in Omental Adipose Tissue in Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 328 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Hughes, M. Elgasim, R. Layfield, and W. Atiomo Genomic and post-genomic approaches to polycystic ovary syndrome--progress so far: Mini Review Hum. Reprod., November 1, 2006; 21(11): 2766 - 2775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Natesampillai, M. E. Fernandez-Zapico, R. Urrutia, and J. D. Veldhuis A Novel Functional Interaction between the Sp1-like Protein KLF13 and SREBP-Sp1 Activation Complex Underlies Regulation of Low Density Lipoprotein Receptor Promoter Function J. Biol. Chem., February 10, 2006; 281(6): 3040 - 3047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Havelock, A. L. Smith, J. B. Seely, C. A. Dooley, R. J. Rodgers, W. E. Rainey, and B. R. Carr The NGFI-B family of transcription factors regulates expression of 3{beta}-hydroxysteroid dehydrogenase type 2 in the human ovary Mol. Hum. Reprod., February 1, 2005; 11(2): 79 - 85. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
