The Apolipoprotein A-I Level Is Downregulated in the Granulosa Cells of Patients with Polycystic Ovary Syndrome and Affects Steroidogenesis Dong-Hee Choi,†,‡ Woo-Sik Lee,‡,§ Miae Won,| Mira Park,| Ho-Oak Park,| Eunju Kim,| Kyoung-Ah Lee,| and Jeehyeon Bae*,| Department of Obstetrics and Gynecology, Bundang CHA Women’s Hospital, Gangnam CHA Hospital, and Department of Biomedical Science, College of Life Science, CHA University, Seongnam, South Korea 463-836 Received January 5, 2010
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder found in women. The etiology of PCOS is still not clear, and there are no available studies on the proteome analysis of granulosa cells (GCs) in PCOS patients. To identify the pathogenic mechanisms and potential diagnostic markers for PCOS, we conducted proteomic profiling of GCs in PCOS patients by two-dimensional gel electrophoresis and liquid chromatography coupled with mass spectrometry (LC-MS/MS) analyses. The proteomic analysis yielded eight downregulated and 12 upregulated proteins in PCOS patients, among which apolipoprotein A-I (ApoA-I) showed significant downregulation in PCOS patients as confirmed by Western blotting. Knockdown of ApoA-I decreased the number of transcripts of steroidogenic enzymes in a granulosa cell line (KGN), while its overexpression generally increased the level of expression of these enzymes. Furthermore, modulation of the expression level of ApoA-I in the granulosa cells altered progesterone production. Therefore, this study suggests that ApoA-I can be useful as a granulosa cell biomarker of PCOS patients and that downregulated ApoA-I may be related to the disturbed production of steroid hormones in PCOS patients. Keywords: PCOS • ApoA-I • ovary • steroidogenesis • granulosa cell • proteomics
Introduction Although polycystic ovary syndrome (PCOS) is the most common endocrine disorder found in women, affecting more than 5% of women of reproductive age,1 little is known about its etiology because of its heterogeneous and complex nature. The phenotypes of PCOS are variable and include polycystic ovaries, hyperandrogenism, hirsutism, glucose dysregulation, and obesity. PCOS patients not only have fertility problems but also are at risk for several other diseases, including diabetes mellitus, obesity, cardiovascular diseases, and endometrial carcinoma.2,3 Numerous investigators have attempted to identify genes or proteins that would improve our understanding of the molecular and cellular mechanisms of PCOS. The results of high-throughput screens of differentially expressed genes (DEGs) in PCOS patients have recently been reported. For instance, attempts to identify a set of DEGs in ovaries biopsied from PCOS patients identified the genes involved in a number of biological functions,4-6 and changes in the transcriptomes have also been reported from microarray * To whom correspondence should be addressed: Department of Biomedical Science, College of Life Science, CHA University, 222 Yatap-dong, Bundang-gu, Seongnam 463-836, South Korea. Phone: 82-31-725-8396. Fax: 82-31-725-7350. E-mail:
[email protected]. † Bundang CHA Women’s Hospital. ‡ These authors contributed equally to this work. § Gangnam CHA Hospital. | CHA University. 10.1021/pr100008e
2010 American Chemical Society
analyses of oocytes, theca, and cumulus cells of women with PCOS.7-9 Proteomic analyses of ovarian tissues and serum from PCOS patients have recently been conducted,10,11 but granulosa cells of PCOS have not been assessed. In addition, no universal single gene or protein with a distinct expression pattern in PCOS patients has been found using high-throughput screening, mainly because of the limited number of studies that have analyzed different ovarian components.4-11 Thus, in this study, we investigated the comparative proteomic signatures of granulosa cells isolated from PCOS patients and from a control group. Granulosa cells are surrounding cells of oocytes and required for the proper development of oocytes. PCOS patients have abnormally increased numbers of follicles,12 and granulosa cells from anovulatory polycystic follicles often exhibit defects in their responses to gonadotropins, including follicle-stimulating hormone (FSH) and luteinizing hormone (LH), differentiation capabilities, and steroidogenic capacities.13-15 By evaluating protein profiles to identify potential diagnostic markers and pathogenic mechanisms of PCOS, we found eight downregulated and 12 upregulated protein spots in PCOS patients. After an immunoblot analysis of the significant spots to validate the proteomics data, we pinpointed apolipoprotein A-I (ApoA-I) because downregulated ApoA-I was observed in a relatively consistent manner between PCOS individuals compared to other spots. Previously, decreased levels of ApoA-I in serum of PCOS patients were Journal of Proteome Research 2010, 9, 4329–4336 4329 Published on Web 04/29/2010
research articles 16-19
observed, but any comparison of ApoA-I expression levels in the granulosa cells of PCOS and normal has not yet been reported. ApoA-I is known to be required for cholesterol transport in ovary,20 and cholesterol is the initial substrate for biosynthesis of steroid hormones in granulosa cells.21 The concerted actions of FSH and LH are obligatory for normal ovarian functions as these hormones control ovarian follicular growth, ovulation, and luteinization.14 The glycoprotein hormones bind to their specific seven-transmembrane G-protein-coupled receptors located on the cell membrane of granulosa cells, and their binding subsequently activates adenylyl cyclase causing an elevation of the level of cyclic AMP (cAMP).22 The increased intracellular cAMP level itself works as a second messenger for the upregulation of target genes, including steroidogenic enzymes.23 Thus, in this study, we further investigated the impact of altered ApoA-I levels in steroidogenesis from granulosa cells and found that changes in the expression level of ApoA-I were capable of affecting the expression of steroidogenic enzymes and the production of the steroid hormone progesterone.
Subjects and Methods Subjects and Collection of Granulosa Cells. This study was performed on patients visiting the Bundang CHA Women’s Hospital and Gangnam CHA Hospital IVF Clinics between September 2008 and August 2009. The protocols and consent forms were approved by the Institutional Review Board (IRB) at the Bundang CHA General Hospital and Gangnam CHA Hospital, and written informed consent was obtained from each patient before each participated in the study. Diagnosis of PCOS was based on the criteria approved by the Rotterdam Consensus Conference. Specifically, all PCOS patients (n ) 11) had oligomenorrhea or amenorrhea, an elevated free androgen index, and polycystic ovaries confirmed using ultrasonography. Fourteen normally menstruating women experiencing infertility due to male, tubal, or uterine factors were recruited as controls. All participants were Korean, and women with a body mass index (BMI) of >25 kg/m2 were excluded to avoid any potential influence of BMI on this study. All of the granulosa cells used in this study were obtained after oocytes had been retrieved from PCOS and control patients undergoing controlled ovarian stimulation for in vitro fertilization (IVF) procedures using standard GnRH antagonist or agonist protocols. Isolation of granulosa cells was performed as described by Salmassi et al.24 with modifications. Briefly, red blood cells were depleted via a Percoll gradient (Sigma, St. Louis, MO), and leukocytes were removed by using Dynabeads CD45 (Invitrogen, Carlsbad, CA). Isolated granulosa cells were washed with cold phosphatebuffered saline (PBS), snap-frozen using liquid nitrogen, and stored at -80 °C. Chemicals. Chemicals were purchased from Sigma unless otherwise indicated. Sample Preparation, Two-Dimensional Gel Electrophoresis (2-DE), Silver Staining, and Imaging. Granulosa cell samples were taken from three PCOS patients with matching controls and were suspended in 50 mM Tris buffer containing 7 M urea, 2 M thiourea, 4% (w/v) CHAPS, and a protease inhibitor cocktail (Roche, Indianapolis, IN). The lysates were homogenized and centrifuged at 12000g for 15 min. Benzonase was added to the lysates, and the protein concentration was determined using the Bradford method (Bio-Rad, Hercules, CA). For 2-DE analysis, 30 µg of pooled protein lysate (10 µg from each person) was diluted with a rehydration solution (8 M urea, 4330
Journal of Proteome Research • Vol. 9, No. 9, 2010
Choi et al. 4% CHAPS, 20 mM DTT, 0.5% immobilized pH buffer, and 0.01% bromophenol blue) and was then in-gel rehydrated to pH 3-10 (NL, 7 cm) IPG strips (GE Healthcare, Piscataway, NJ) overnight at room temperature. Isoelectric focusing was conducted for a total running time of 3.5 kVh using a Multiphor II apparatus (GE Healthcare). The strips were equilibrated and alkylated for 20 min in equilibration buffer [1% (w/v) DTT, 6 M urea, 30% (v/v) glycerol, 2% (w/v) sodium dodecyl sulfate (SDS), and 2% acrylamide in 50 mM Tris-HCl buffer (pH 7.0)] to reduce artifacts due to disulfide bond formation. The 2-DE separation was performed using 12% (v/v) sodium dodecyl sulfate-polyacrylamidegelelectrophoresis(SDS-PAGE).Vorum’s method was used for silver staining of two-dimensional (2D) gels, as described by Mortz et al.25 The 2D gel was imaged using a GS-710 imaging-calibrated densitometer (Bio-Rad). Protein spot detection and 2D pattern matching were conducted using ImageMasterTM 2D Platinum V (GE Healthcare). To compare protein spot densities between control and treated samples, we landmarked and normalized at least 10 spots. The quantified spots of candidate proteins were compared with the aid of histograms. To ensure the reproducibility of 2-DE experiments, each sample was analyzed in duplicate. In-Gel Protein Digestion and Identification of Proteins by Liquid Chromatography Coupled with Mass Spectrometry (LC-MS/MS). Protein bands of interest were excised and digested in-gel with sequencing-grade, modified trypsin (Promega, Madison, WI), as previously described.26 The resulting tryptic peptides were separated and analyzed using a reversedphase capillary HPLC system directly coupled to a Finnigan LCQ ion trap mass spectrometer (LC-MS/MS). Both 0.1 mm × 20 mm trapping and 0.075 mm × 130 mm resolving columns were packed with Vydac 218MS low trifluoroacetic acid C18 beads (Vydac, Hesperia, CA) and placed in-line. After the peptides were bound to the trapping column for 10 min with 5% (v/v) aqueous acetonitrile containing 0.1% (v/v) formic acid, the bound peptides were eluted with an 80 min gradient of 80% (v/v) acetonitrile containing 0.1% (v/v) formic acid at a flow rate of 0.2 µL/min. For tandem mass spectrometry, the full mass scan range mode was m/z 450-2000. The individual spectra from MS/MS were converted using TurboSEQUEST (Thermo Quest, San Jose, CA), and then the resulting peak list files were used to query an NCBI nonredundant database (20090513, 8851222 sequences) using MASCOT (http://www. matrixscience.com). Variable modifications of methionine (oxidation) and cysteine (propionamide), deamidation of asparagine and glutamine, a peptide mass tolerance of 2 Da, a MS/ MS ion mass tolerance of 1 Da, an allowance of missed cleavage at 2, mammalian species, and charge states (+1, +2, and +3) were all taken into account. Proteins were considered to be identified with at least two peptides with individual peptide ion scores of identity (p < 0.05). Plasmid Construction. The human ApoA-I (NM 000039) insert DNA was generated after polymerase chain reaction (PCR) with a human ovarian cDNA library using Pfu DNA polymerase (Cosmo, Seoul, Korea) and the primers 5′-CTAGAATTCAAATGAAAGCTGCGGTGCTG and 5′-CTAGTCGACTCACTGGGTGTTGAGCTT. The PCR products were digested with restriction enzymes EcoRI and SalI (Enzynomics, Daejeon, Korea) and cloned into the pCMV HA vector (Clontech, Palo Alto, CA) as shown in Figure 1 of the Supporting Information. Mammalian Cell Culture and Transfection of KGN Cells. Human ovarian granulosa-derived tumor cell line KGN was provided by Drs. Nishi and Yanase (Kyushu University, Fuku-
research articles
Downregulated Apolipoprotein A-I in PCOS oka, Japan). The cell line was cultured in DMEM and Ham’s F-12 medium (Welgene, Seoul, Korea), supplemented with 10% fetal bovine serum (FBS) (Welgene) and 1% penicillin-streptomycin (Welgene), at 37 °C in a 5% CO2 environment. Cells (3 × 106) were electroporated with the plasmid encoding ApoA-I using a MicroPorator MP-100 (Digital-Bio Technology, Seoul, Korea) according to the manufacturer’s protocol and were then incubated on 100 mm dishes for 24 h. RNA Interference. Small inhibitory RNA (siRNA) duplexes targeting ApoA-I mRNA (5′-CCCUGGACGACUUCCAGAAGAAdTdT-3′) and a control siRNA (5′-CCUACGCCACCAAUUUCGUdTdT-3′) duplex were purchased from Bioneer (Daejeon, Korea). Cells (3 × 106) were electroporated with 100 nM siRNA duplex using a MicroPorator MP-100 and were incubated on 100 mm dishes for 24 h. Knockdown of ApoA-I was confirmed by Western blotting using an anti-ApoA-I antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Western Blot Analysis. Protein extracts of granulosa cells were prepared with NP-40 lysis buffer [50 mM Tris-HCl (pH 8.0), 0.15 M NaCl, and 1% NP-40] containing a 10% (v/v) protease inhibitor cocktail, and lysates were centrifuged, followed by incubation. For the quantitative protein analysis, a standard curve was established with a standard BSA solution using a BCA (bicinchoninic acid) protein assay kit (Pierce, Rockford, IL). Aliquots of protein extracts were boiled in SDS sample buffer, separated by SDS-PAGE, and electroblotted onto PVDF membranes (Amersham, Little Chalfont, Buckinghamshire, U.K.). The membrane was then blocked with 5% nonfat milk in PBS-Tween 20, followed by overnight incubation with polyclonal anti-ApoA-I antibodies. Western blot analysis of GAPDH was performed using an anti-GAPDH antibody (Ab Frontier, Seoul, Korea) as a loading control. The membranes were washed with PBS-Tween 20 before being incubated with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibodies (Santa Cruz Biotechnology) at room temperature for 1 h prior to visualization using enhanced chemiluminescence (Ab Frontier). Proteins were detected with the LAS image program (Fuji, Tokyo, Japan), with the relative protein levels in each sample normalized to GAPDH. Real-Time PCR. Total RNA was extracted from KGN cells using an Intron Total RNA Extraction Kit (Intron, Seungnam, Korea). The concentration and quality of RNA were determined with an ND-1000 spectrophotometer (NanoDrop, Waltham, MA). Reverse transcription to cDNA was conducted using the SuperScriptIII first-strand synthesis system kit (Invitrogen) based on the manufacturer’s instructions. All cDNAs used in the real-time PCR were normalized with GAPDH. A quantitative real-time PCR was performed using a SYBR Green I kit (Qiagen, Hilden, Germany) and using the housekeeping gene GAPDH as a control. Nucleotide sequences of primers used are listed in Table 1 of the Supporting Information. Gene expression was quantified by the delta delta CT method, and real-time PCR was performed in a Rotor-gene 3000 (Corbett Research, Sydney, Australia). Measurement of Hormones. ApoA-I-overexpressed or ApoAI-knockdown KGN cells were incubated in medium containing either cholesterol (1 µM) (Sigma) or forskolin (10 µM) (Calbiochem, Nottingham, U.K.) for 24 h. As a control, the cells were exposed to the vehicle solvent with a final volume of 0.01% dimethyl sulfoxide. The media were harvested, and the levels of progesterone, estradiol, and testosterone were measured by a chemiluminescence immunoassay (CLIA) using ADVIA Centaur (Bayer Healthcare, Chicago, IL).
Table 1. Clinical Profiles of PCOS Patients and Controls parameter
age (years) BMI (kg/m2) basal estradiol level (pg/mL) fasting glucose level (mg/dL) luteinizing hormone level (IU/L)a FSH level (IU/L) systolic blood pressure (mmHg) diastolic blood pressure (mmHg) total cholesterol level (mg/dL) a
control (n ) 14)
33.1 ( 5.6 17.3 ( 4.4 21.4 ( 16.3
PCOS (n ) 11) p value
31.4 ( 3.6 0.443 21.3 ( 2.4 0.066 27.7 ( 11.9 0.724
not determined 100.2 ( 34.7 4.3 ( 1.6 7.3 ( 2.3 122.5 ( 9.5 74.9 ( 7.5 182.3 ( 38.5
9.5 ( 4.0
0.037
6.2 ( 1.2 0.341 124.5 ( 10.4 0.975 74.8 ( 9.3
0.618
181.4 ( 31.5 0.493
Significantly different values between two groups (p < 0.05).
Statistical Analysis. The significance of patient values compared to controls was determined by the Student’s t-test. Twotailed p values of
