1 Supplementary Methods Human Endometrial Carcinoma Samples Flash frozen endometrial tissue from type I endometrioid tumours of patients of ages 30–55 was obtained under sterile conditions after hysterectomy at Peking University Oncology Hospital. The samples were graded and staged according to the 1988 International Federation of Gynaecology and Obstetrics (FIGO) systems. None of the patients had received preoperative chemotherapy or radiation. In addition, samples of normal endometrium that were age-matched to the cancers were obtained from patients undergoing hysterectomy for benign gynaecological diseases. All studies were approved by the Ethics Committee of the Peking University Health Science Center. Immunohistochemistry and Image Analysis Paraffin sections of 5 μm from the uteri were made. A standard immunohistochemical technique was used to detect the intensity and distribution of PCNA and Ki-67 immunostaining. A mouse anti-PCNA antibody (18-0110, Zymed Laboratories, Inc., San Francisco, CA) at a 1:150 dilution and a monoclonal mouse anti-human Ki-67 antibody (08-0156, Zymed Laboratories) were used. Quantification of PCNA immunostaining was performed on the digitized images of systematic randomly selected fields of endometrial stroma, from which non-stromal elements (e.g., luminal and glandular epithelium) were interactively removed and analyzed separately. A Leica microscope connected to a computer using Colorvision software (Leica Imaging System Ltd. Cambridge, England) was used for the image analysis. All luminal and glandular epithelia as well as 10 fields of stromal cells were measured separately in each tissue section. By using colour discrimination software, the total area of positively stained cells (brown reaction product) was measured and expressed as a ratio of the total area of cell nuclei (brown reaction product + blue haematoxylin). 2 Animal Experiments and Tissue Collection Adult female CD rats (VAF, Charles River, Beijing, China), ~ 55 days old and weighing approximately 250 g, were housed in a controlled environment at 20°C on an illumination schedule of 12 h of light and 12 h of darkness each day. Standard pellet food and water were provided ad libitum. The rats were anesthetized, ovariectomized, and housed for 14 days before a 48-hour treatment with either 2.5 μg 17β-estradiol (E2) /day (n = 8), 500 μg tamoxifen (TAM)/day (n = 5) (both substances purchased from Sigma, St Louis, MO). The control group (n = 4) received vehicle only (propylene glycol). The animals received injections in the morning of days 1 and 2, and were sacrificed in the morning of day 3. Under animal anaesthesia, the uterus was removed with fat and connective tissues were stripped off. The uteri were weighed and divided; one half was immediately frozen in liquid nitrogen and stored at -70°C until analyzed by real time RT PCR. The other half was immersion-fixed in 4% formaldehyde at 4°C for 12 hours, stored at 4°C in 70% ethanol, and thereafter embedded in paraffin. All studies were approved by the Animal Care Committee of Peking University Health Science Center. Flow Cytometry Cells were collected and resuspended in PBS with 2% glucose and 3% paraformaldehyde. After permeabolization with ethanol, cells were stained with propidium iodide solution (69 M propidium iodide and 38 mM sodium citrate). Cell cycle data were collected with FACScan (Becton Dickinson Immunocytochemistry System) and analyzed with ModFit LT (Verity Software House, Inc., Topsham, ME). Real-Time Quantitative RT-PCR Analysis TaqMan (Perkin-Elmer/Applied Biosystems) RT-PCR analysis was performed with 5'-exonuclease assays using fluorescent nonextendible oligonucleotide probes (TaqMan PCR detector 7700) as described1. Rat uteri were weighed, minced, and homogenized for total RNA extraction using TRIZOL Reagent (Invitrogen Corp., 3 Rockville, MD). Primers and probes used were: human EKLF, forward primer: 5’-CCTGTTTGGTGGTCTCTTCACA-3’, reverse primer: 5’-AGGGTCCATTCGTGGGAAA-3’, probe: 5’-CGCGTGACACAATGCTGGGTGG-3’; rat EKLF, forward primer: 5’-CGCCCCCTTCCTTTTTGA-3’, reverse primer: 5’-CAGTGGCCCCGAGTTCTGT-3’, probe: 5’-TGTCTGGGACCTGGTGCGG-3’; human PKC, forward primer: 5’-ACAAGCAATGCGTCATCAATG-3’, reverse primer: 5’-CGCCCCCTCTTCTCAGTGT-3’, probe: 5’-CCCCAGCCTCTGCGGAATGGA-3’; rat PKC, forward primer: 5’-GGCGACTGTCCGTAGAAATCTG-3’, reverse primer: 5’-GGAAAGGGAGCCCATGAAGT-3’, probe: 5’-ACTGGGATCGGACGACACGGAA-3’; human PAX2, forward primer: 5’-CTGGGCAGCAACGTGTCA-3’, reverse primer: 5’-GAGTGGTGCTCGCCATGTC-3’, probe: 5’-ACAGACATACCCCGTTGTGACTGGTCG-3’; rat PAX2, forward primer: 5’-CTGGAAGCTTTGGATCGAGTCT-3’, reverse primer: 5’-TGTGCTCTGATGCTTGGAAGA-3’, probe: 5’-TGAGCGTCCTTCCTATCCCGA-3’; hKID, forward primer: 5’-CCCTCCTGGCCTCACTGAT-3’, reverse primer: 5’-TGCGCCAGGGTCTTGGT-3’, probe: 5’-TACAACTTCGTCCTGTTCCCCG-3’; rat hKID homolog, forward primer: 5’-GGCCCTGCGGTCATTG-3’, reverse primer: 5’-GTGAGCGGTCCTACCCAGAA-3’, probe: 5’-TGGGAAGTTTGCAGTCCATTGG-3’; human cyclin F, forward primer: 5’-CCAGA GGCGTTCGTATGTGA-3’, reverse primer: 5’-AAAGAATGGCCCAGAAATTCAG-3’, probe: 5’-CCACAGATGGCGTCAATGTGAACA-3’; rat cyclin F, forward primer: 5’-GGAAAGTGACAGAAAGGTGGATATG-3’, reverse primer: 5’-GCGGCGTAGTCCCTGAGTT-3’, probe: 5’-CTGACCCCGGACGATGCCTCC-3’; human RAR1, forward primer: 5’-TGTTCACAAGCCATTAGGGAAA-3’, reverse primer: 4 5’-ACACTAGAGTTACATGGTCCAGTGGTA-3’, probe: 5’-TTCATGGGATAATTAGCAGGCTGG-3’; rat RAR1, forward primer: 5’-GGCCTCTGGGACAAATTCAG-3’, reverse primer: 5’-CTTGGCGAACTCCACGATCT-3’, probe: 5’-AGCTGGCCACCAAGTGCATTATT-3’. Gene Silencing by Retroviral siRNA System Recombinant retroviruses carrying specific oligonucleotides for generating siRNAs to the 35 genes described in this report were prepared using the BD™ Knockout RNAi System (BD Biosciences) according to the manual provided by the vendor. The mRNA sequences representing the 35 genes and their ORF (open reading frame) information were obtained from RefSeq (http://www.ncbi.nlm.nih.gov/RefSeq/). Sequences were masked to remove repetitive sequences using RepeatMasker (http://ftp.genome.washington.edu/cgi-bin/RepeatMasker), and vector contamination was masked by searching with NCBI BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) against UniVec (http://www.ncbi.nih.gov/VecScreen/UniVec.html). Three unique 19-mers for each target were selected, where possible (1) to contain no stretches of four or more consecutive T or A residues (to avoid premature pol III transcription termination signals); (2) to have 40–70% overall GC content; (3) to lie within the coding sequence of the target gene; (4) to begin with an AA dimer in the 5' flanking sequence; (5) to end just prior to a TT, TG or GT doublet in the 3' flanking sequence; (6) to share minimal sequence identity with other genes; (7) to target all transcript variants represented by RefSeq mRNAs; and (8) to not overlap with other 19-mers selected from the same target sequence. Specific oligonucleotides to the 35 genes were synthesized and annealed into double strands. After purification and restriction digestion, the oligonucleotides were ligated into the linearized RNAi-Ready pSIREN vector (BD Biosciences). The resultant plasmid was transformed into BD Fusion-Blue™ Competent Cells, purified, and co-transfected with the pVSV-G vector into the GP2-293 packaging cell line (BD Biosciences) using LipofectAMINE (Invitrogen). Three days after transfection, cell lysates were obtained from the 5 293 cells. The cell lysates were added to 293 cells again, and when most of the cells had been killed by virus infection and detached, cell lysates were collected and filtered through a 0.45 m cellulose acetate filter. The viral titers were determined according to the procedure provided by the vendor (BD Biosciences). For infection, cells were incubated with viruses at a MOI of 10 for 6 h before fresh media were added and cultures were continued for another 48 h. For generation of PAX2 expression-silenced ECC-1 cells to be transplanted into nude mice, infected cells were grown in the presence of 2 g/ml of puromycin for three weeks. Individual drug-resistant clones were collected, pooled, and expanded. The oligonucleotide sequences synthesized for EKLF RNAi were: forward sequence, 5’-GGATCCCGAGGCAGGCAGCGCACACGTTCAAGAGACGTGTGCGCTGCC TGCCTCTTTTTGGAGA -3’; reverse sequence, 5’- GAATTCTCCAAAAAGAGGCAGGCAGCGCACACGTCTCTTGAACGTGTGCG CTGCCTGCCTCGGG-3’. In synthesizing oligonucleotides for other genes, the underlined sequences were replaced by the following sequence and its reverse complement sequence: NF I/B: GAAGCACCCGTGCTGTGTC RPIP8: GCCACCATCCTCACCGGAA EREG1: GGTGCCCTGGGAGACGGAG MYL7: GGGCCCCATCAACTTCACC Cadherin-8: GCAAACCTCTGGAGCCTCC NTAK: CGAGACAGCCAAGTCCTAT Nebulette: CTTCAACGGGGAACTGCAA DLC-1: GGAACTGTCCAGCTTCAGC PKC: CAAGGCTTCCAGTGCCAAG TRIP9: GATGTGGAGATGGTCCGGC gb=AL109681: CGTATATGTCTAGTATCCC SH-PTP3: GAGAAGAATGGAGATGTCA Flt3 ligand: CCTGCAGGACGAGGAGCTC Phosphodiesterase I alpha: CTAATGTGGATGATATTAC 6 MEA6: GACTCTGACTGAACGCTTG Myelin proteolipid protein: CTGCCTCTTTCTTCTTCCT PAX2: GCACTTGCGAGCTGACACC IGF II: GTTCTTCCAATATGACACC gb=AL049390: GAGAAGTGATTTCTGAAGC Carboxylesterase: CCCAGGCTCTGTGACCATC hKID: GTTCACAGTCCACTGGGTC Heme oxygenase: CATTGCCAGTGCCACCAAG GABA-B R1a receptor: CCGGCAGCGTTTCCCCACT OPCML: CACCAATGCCAGCATCACA Cyclin F: GACCTCATTCCCTGCGTCT AILIM: GTTCTGGTTACCCATAGGA ENaC: GAATTTAACTATCGCACCA MYOD1: CCCCAACCCGATATACCAG KCNQ3: GTACCGGCGCATCCAAACT RAR1: CTGGCCACCAAGTGCATTA PRGP2: CCTGGAACGGGAGTGTCTG BK-2: CATCTTTGTCCTCAGCGTC gb=3413939: GTGTGCGTGGCCACCTGAA gb=3882154: CATTCTCAACAATGAGGCC ER: GCTACTGTTTGCTCCTAAC ER: CACCTGGGCACCTTTCTCC Athymic Mouse Model ECC-1 endometrial tumours were developed according to the procedures described2. Briefly, about 108 ECC-1 cells/0.1 ml were injected bilaterally into mammary fat pads of 4-5 week old ovariectomized athymic mice (BALB/c, Charles River, Beijing, China). E2 was administered via a 1.0 cm silastic capsule implanted subcutaneously to stimulate tumour growth2. When ECC-1 tumours reached 1.0 cm2, a single tumour was harvested, and 1.0 mm3 pieces were bitransplanted into the anterior axillary 7 region of 22 athymic mice, which were treated as follows: 0.3 cm E2 capsules (n=8); 0.5 mg of tamoxifen capsules (n=8); or untreated control (n=6). Silastic E2 capsules (Sigma Chemical Co.) were made as described3; they were implanted subcutaneously and replaced every 8 weeks. Tamoxifen was dissolved in ethanol and suspended in a solution of 90% carboxymethylcellulose (1% carboxymethylcellulose in double-distilled water) and 10% polyethylene glycol 400/Tween 80 (99.5% polyethylene glycol 400 and 0.5% Tween 80). The ethanol was evaporated under nitrogen before use. Tamoxifen were administered by gavages at 0.5 mg per mouse per day, 5 days/week. Tumour measurements were performed weekly using Vernier callipers. The cross-sectional area was calculated using the formula: length x width/4 x . All studies were approved by the Animal Care Committee of Peking University Health Science Center. Dual Immunofluorescent Staining For ER detection, paraffin-embedded tissue sample sections were incubated with a mouse anti-human ER monoclonal antibody and with a biotinylated goat anti-mouse antibody (Biosource International, Camarillo, CA), and FITC-avidin (Vector Laboratories, Burlingame, CA). For PAX2 detection, sections were incubated with a mouse monoclonal antibody and with a biotinylated goat anti-mouse antibody and Texas red-avidin (Biosource International). Sections were mounted with Vecta-shield mountant for fluorescence (Vector Laboratories) and stored in the dark at 4°C. Fluorescence staining was examined using an Olympus BX 40 fluorescence microscope fitted with filters to detect both Texas Red (BP 545–580) and FITC (BP 450–480) fluorescence simultaneously as well as each of the two fluorochromes separately. Controls (adjacent sections to each tumour sample) were treated and stained in the same way as the test sections. Chromatin Immunoprecipitation (ChIP) ChIP experiments were performed according to the protocol described previously1. The following primers were used to amplify a 605 bp fragment within the regulatory 8 region of PAX2 gene: forward: 5’-AGAGAGACACACACCGGGAG-3’, and reverse: 5’-GATGAGTGAACGCCCGGCT-3’; and a 331 bp fragment in the promoter region of EREG1 gene: forward: 5’-CAGTAGGCCACCGAAGAACAC-3’, and reverse: 5’-CCCTAATACCCGCGGTTAGTG-3’. Reference 1. Shang, Y., Hu, X., DiRenzo, J., Lazar, M. A. & Brown, M. Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103, 843-52. (2000). 2. Dardes, R. C., Bentrem, D., O'Regan, R. M., Schafer, J. M. & Jordan, V. C. Effects of the new selective estrogen receptor modulator LY353381.HCl (Arzoxifene) on human endometrial cancer growth in athymic mice. Clin Cancer Res 7, 4149-55 (2001). 3. Robinson, S. P. & Jordan, V. C. Antiestrogenic action of toremifene on hormone-dependent, -independent, and heterogeneous breast tumor growth in the athymic mouse. Cancer Res 49, 1758-62 (1989).