Establishment of an immortalized human endometrial stromal cell line with functional responses to ovarian stimuli

Endometrial tissue was obtained from the uterus of a 45-year-old woman who had undergone total hysterectomy because of leiomyoma, after her written informed consent and Keio University Hospital approval. Primary human ESCs were purified as described previously [12]. In brief, the tissue specimens were washed in Dulbecco's modified Eagle's medium (DMEM; Sigma-Aldrich, St. Louis, MO) and minced with scissors into pieces less than 1 mm³ in size. The tissues were then gently agitated in tubes for 2 h at 37°C in DMEM with 0.2% (wt/vol) collagenase (Wako Pure Chemical Industries, Osaka, Japan), 0.05% DNase I (Life Technologies, Inc., Gaithersburg, MD), 10% fetal bovine serum (FBS), and 1% antibiotic-antimycotic mixture (Life Technologies, Grand Island, NY). After enzymatic digestion, cell clumps were dispersed by pipetting. Most of the digested human ESCs, presenting as single cells or small aggregates, were filtered sequentially through a 70 μm cell-strainer nylon filter (Falcon 2350; BD Biosciences, Franklin Lakes, NJ) to remove gland cells, layered onto Ficoll-Paque (GE Healthcare Life Sciences, Uppsala, Sweden), and centrifuged at 500 × g for 15 min at 4°C to remove erythrocytes. The isolated human ESCs were collected from the Ficoll interface and resuspended in DMEM with 10% FBS and 1% antibiotic-antimycotic mixture.

Recombinant lentiviruses were prepared using the ViraPower Lentiviral Expression System (Life Technologies), according to the manufacturer's instructions. Full-length cDNA of human telomerase reverse transcriptase isoform 1 (hTERT; GenBank accession number NM_003219) was amplified by PCR from the total RNA of a colon cancer cell line. The first-strand DNA was synthesized using hTERT 3'-noncoding primer 5'-TGACAGGGCTGCTGGTGTCTG and ReverTra Ace reverse transcriptase system (Toyobo, Tokyo, Japan). The cDNA of the N-terminus half of hTERT (1-2304) was amplified using forward primer 5'-CACCATGCCGCGCGCTCCCCGCTGCCGA and reverse primer 5'-GCCTTCTGGACCACGGCATACCGA, and that of the C-terminus half (1977-3399) was amplified using forward primer 5'-CACCGGCACTGTTCAGCGTGCTCAACTACGAG and reverse primer 5'-TCAGTCCAGGATGGTCTTGAAGTC. Each cDNA was ligated to the pLenti6/V5-D TOPO vector. Kozak sequences were introduced according to the instruction manual. The fragment between SpeI (in the vector) and EcoRV sites from the N-terminus half was then recloned in the same sites of the C-terminus half to obtain the full-length hTERT plasmid (pLenti6-hTERT). The DNA sequence of pLenti6-hTERT was completely confirmed by DNA sequence analysis. pLenti6-hTERT was then transfected to near-confluent 293FT cells (Life Technologies) with helper plasmids (pLP1, pLP2, and pLP/VSVG) to produce a lentivirus encoding hTERT (LtV-hTERT). After 2 days of incubation at 37°C, the conditioned mediums of the lentivirus-producing 293FT cells were harvested and centrifuged at 3000 rpm for 15 min at 4°C. The supernatants were separated into aliquots and stored at -80°C until use.

For immortalization, human ESCs were infected with LtV-hTERT in the presence of polybrene (6 mg/ml, Sigma). The cells were grown in phenol red-free DMEM supplemented with 10% dextran-coated charcoal-treated FBS (DCC-FBS; Life Technologies) in the presence of 10 nM 17-b-estradiol (E2; E8875, Sigma) for 7-8 days until uninfected cells died in the presence of 2 mg/ml blasticidine (Life Technologies). Then, the cells were trypsinized and limiting-diluted in 96-well plates. Immortalized clones were obtained in the presence of E2 under selection with blasticidine. The obtained clones were maintained in DMEM supplemented with 10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin, and 2 mg/ml blasticidine at 37°C under 5% CO₂ in a humidified incubator. The clones were tested for marker expression and response to ovarian steroids and cAMP, as described below. Finally, KC02-44D was selected as the most responsive clone to the ovarian stimuli. We did not obtain any clones responsive to these stimuli when primary human ESCs were cultured in the absence of E2.

KC02-44D cells were precultured in phenol red-free DMEM supplemented with 10% DCC-FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin at 37°C over 4 days. They were then stimulated with or without 10 nM E2, 1 μM 6α-methyl-17α-hydroxy-progesterone acetate (MPA; M-1629, Sigma), and/or 0.125 mM dibutyryl cAMP (D0627, Sigma) in combination with the indicated concentrations of cytokines for different time periods according to the experimental procedures. Cell images were acquired using a phase-contrast microscope (Nikon). In the IL-1β experiment, KC02-44D cells were treated with 1.0 ng/ml IL-1β (201-LB-005, R&D Systems, Minneapolis, MN) for 3-7 days, as described in the figure legends.

KC02-44D cells were cultured in 96-well tissue culture plates and characterized by immunofluorescence staining. The cells were fixed with 4% paraformaldehyde (PFA; Wako Pure Chemical Industries) solution for 30 min at room temperature. After washing and permeabilization with phosphate-buffered saline (PBS) containing 0.1% Triton X-100, the cells were blocked with PBS-5% FBS at 4°C. The fixed cells were incubated in PBS-10% FBS containing Cy3-labeled mouse anti-vimentin (C9080, 1:200; Sigma), mouse anti-CD10 (clone SS2/36, 1:25; Dako, Glostrup, Denmark), mouse anti-CD13 (clone WM-47, 1:50; Dako), or mouse anti-cytokeratin (clone MNF116, 1:50; Dako) antibody. After washing in PBS, the cells were incubated with Alexa Fluor 488-labeled anti-mouse secondary antibody (1:500; Molecular Probes, Eugene, OR), washed again, and examined for each marker expression under a fluorescence microscope (BZ-8000; Keyence, Osaka, Japan).

Total proteins from KC02-44D cells were extracted with Cell Lysis Buffer (Cell Signaling Technology, Beverly, MA) containing 1 mM phenylmethanesulfonyl fluoride (Sigma) on ice. The lysate was scraped and sonicated for 5 s by using an ultrasonic disruptor at output 1 (Tomy Digital Biology, Tokyo, Japan). The sonicate was centrifuged at 15,000 rpm for 10 min at 4°C, and the protein contents of the supernatant were measured using the Bradford protein assay (Bio-Rad, Hercules, CA). After the addition of lithium dodecyl sulfate sample buffer (Life Technologies, Grand Island, NY) containing 2-mercaptoethanol (Sigma), the samples were reduced at 70°C for 10 min. Equal amounts of total proteins were then electrophoresed on 4-12% NuPAGE Bis-Tris gels (Life Technologies, Grand Island, NY) and transferred onto polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA), followed by blocking with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T). The membranes were then immunoblotted with the primary antibody against the PR (clone PgR636, 1:800, mouse monoclonal; Dako), IGFBP-1 (06-106, 1:1000, rabbit polyclonal; Upstate, Millipore), or β-actin (1:3000, goat polyclonal; Santa Cruz Biotechnology, Santa Cruz, CA) in TBS-T for 1 h at room temperature. After washing thrice with TBS-T, the membranes were further incubated with horseradish peroxidase-conjugated anti-mouse, anti-rabbit (1:10,000; GE Healthcare, Buckinghamshire, UK) or anti-goat (1:5000; Jackson ImmunoResearch Laboratories, West Grove, PA) secondary antibody in TBS-T for 1 h at room temperature. After washing thrice with TBS-T again, proteins were detected with ECL Plus reagent (GE Healthcare) and visualized with LumiImager (Roche Diagnostics, Penzberg, Germany).

KC02-44D cells were plated at a density of 2 × 10⁵ cells per well in 6-well plates. Two days later, the medium was changed and the cells were stimulated with E2, MPA, and dibutyryl cAMP alone or in combination at the indicated concentrations. The supernatants were collected 6 or 7 days later, and the amounts of IGFBP-1, PGE2, and IL-8 proteins were measured using ELISA kits for IGFBP-1 (DY871, R&D Systems), prostaglandin E2 (PGE2; KGE004, R&D Systems), and IL-8 (GE Healthcare), respectively. Each treatment was performed in triplicate wells.

Total RNA was extracted from cultured KC02-44D cells by using an RNeasy mini kit (Qiagen, Valencia, CA). RT-PCR was carried out with 200 ng total cellular RNA by using a One-Step RT-PCR kit (Qiagen). The primers used for amplification were as follows: human estrogen receptor-alpha (ERα), forward primer 5'-CAGGGGTGAAGTGGGGTCTGCTG-3' and reverse primer 5'-ATGCGGAACCGAGATGATGTAGC-3'; cyclooxygenase-2 (COX-2), forward primer 5'-TGTCTTGACATCCAGATCAC-3' and reverse primer 5'-ACATCATGTTTGAGCCCTGG-3'; β-actin, forward primer 5'-CCCAGGCACCAGGGCGTGATC-3' and reverse primer 5'-TCAAACATGATCTGGGTCAT-3'. Preliminary experiments determined the optimum PCR cycle number within the linear range of amplification for each gene being measured. After PCR amplification, the samples (15 μl aliquots) were electrophoresed in 2% agarose gels, followed by photographic recording of ethidium bromide-stained gels with FAS-III MINI (Toyobo). All data from the RT-PCR analysis represent the results of three independent experiments.

Cell proliferation was determined using a Click-iT EdU Alexa Fluor Imaging kit (Life Technologies, Grand Island, NY), a modified 5-bromo-2'-deoxyuridine (BrdU) system to detect DNA replication. In brief, KC02-44D cells (1.0-1.5 × 10³) were seeded onto 96-well culture plates with phenol red-free DMEM containing 10% DCC-FBS and stimulated with cytokines 48 h later. EdU (5-ethynyl-2'-deoxyuridine) was added to a final concentration of 10 μM in the cell-culture medium and the cells were incubated for 24 h at 37°C. After fixation with 4% PFA and permeabilization with 0.1% Triton X-100 in PBS, the cells were treated with the reaction cocktail containing Alexa Fluor 488 azide. The EdU intensity was measured using an ArrayScan VTi HCS reader (Thermo Scientific Cellomics, Pittsburg, PA). The colocalization bioapplication protocol was used to acquire images with the appropriate filter sets in two separate fluorescence channels: channel 1, Hoechst 33342 (nuclear reference and object identification); channel 2, EdU labeled with Alexa Fluor 488 azide (identification of DNA-replicating S-phase cells in the cell cycle). The cells were imaged with a 20 × objective and 0.63 numerical aperture with exposure times of 0.2 s (Hoechst 33342) and 0.1 s (Alexa Fluor 488 azide). The images were collected as nine fields per well, and the values were expressed as the mean average fluorescence intensity.

All quantitative experiments were conducted at least three times each in triplicate wells. The results were then analyzed using ANOVA and Tukey's test. The data shown in figures was the Mean ± SEM for triplicate wells from one representative experiment.