Nuclear factor of activated T-cell isoform expression and regulation in human myometrium

Human myometrial biopsies were obtained at Caesarean section with informed written consent and institutional Ethics Committee approval (Guy’s and St Thomas’ Hospital NHS Trusts, London, UK; Office of Medical Bioethics, University of Calgary). Biopsies from the upper edge of the lower segment incision were obtained from pregnant women at the time of elective caesarean section (at term prior to labour), none of the women had underlying medical conditions (reasons for elective caesarean section at 37–40 weeks were: maternal request, breech presentation, previous caesarean section, fetal cardiac anomaly detected antenatally, stress incontinence, previous 3^rd degree tear or placenta praevia). In addition, lower segment human myometrium was also obtained from four groups of women at the time of caesarean section (LSCS) under the conditions of preterm no labour (PTNL; 29.2 ± 1.7 weeks’, myometrium samples from n = 5 women, reasons for elective caesarean section were intrauterine growth restriction or preeclampsia/pregnancy induced hypertension), preterm with labour (PTL; 30.0 ± 2.1 weeks’, samples from n = 4 women, reasons for caesarean section were breech presentation, previous caesarean section and increased fetal heart rate detected during labour), term no labour (TNL; 39.5 ± 0.4 weeks’, samples from n = 6 women, reason for caesarean section was previous caesarean section or breech) and term with labour (TL; 39.1 ± 0.6 weeks’, samples from n = 6 women, reasons for caesarean section were presented in early labour and elected to have section, failure to progress and/or fetal distress in labour). Whole myometrial tissue was either snap frozen and stored at −80 °C or used immediately for cell culture.

Myometrial cells were dispersed enzymatically from tissue biopsies as described previously from term non-labouring (TNL) women [39]. Briefly, small segments of myometrium were dissected and chopped into 1–2 mm³ pieces and incubated for 30–40 min in Dulbeccos modified Eagles medium (DMEM) containing 1 mg/ml collagenase 1A, 1 mg/ml collagenase XI plus 0.1 % bovine serum albumin (BSA), penicillin (50 units/ml), and streptomycin (50 mg/ml). Cells were dislodged using a Pasteur pipette and then filtered through a 45 μm sterile filter, and washed twice in DMEM containing 10 % fetal calf serum (FCS) by centrifugation (450 × g 5 min). The cell pellet was suspended in DMEM supplemented with 10 % FCS, penicillin (25 units/ml), and streptomycin (25 mg/ml). Primary myocytes were seeded in T25 culture flasks and incubated at 37 °C in a humidified atmosphere of 95 % air/5 % CO_2. Routine immunofluorescent labeling of cells with alpha-actin and calponin monoclonal antibodies was routinely performed to verify the purity of myocyte cultures. After the first 2 days of culture, media was replaced with DMEM supplemented with 5 % FCS, penicillin (25 units/ml), and streptomycin (25 mg/ml). The medium was changed every 2 days until cells were ~80 % confluent. Cells were used for experimentation at passage 2 (P2) in order to have enough material for in vitro the stretch protocol.

Human myometrial cells (P2) were cultured in six well culture plates in 3 ml DMEM plus 5 % FCS (Corning) until approximately 80 % confluent. Following replenishment of media (24 h prior to experimentation, 5 % FCS), cells were exposed to A23187 Ca²⁺ ionophore (5 μM, Sigma-Aldrich, UK) or vehicle control (0.1 % dimethyl sulfoxide, Sigma-Aldrich, Gillingham, UK) for 6 or 14 h at 37 °C in a humidified atmosphere of 95 % air/5 % CO₂. At the end of the experiment, cells were rinsed with phosphate-buffered saline (PBS) and collected for RNA/protein extraction.

A method similar to that used in the present has been described previously by us and others [40, 41]. Pregnant human myometrial cells (P2) were cultured in six well flexible-bottom culture plates pre-coated with collagen type I (Flexcell International Corp., Hillsborough, USA) in 3 ml DMEM plus 5 % FCS until approximately 80 % confluent. Media was replaced 24 hours before cells were subjected to 25 % tonic mechanical stretch for 6 h using a strain unit (Flexercell FX-4000 Tension system, Flexcell International Corp., Hillsborough, USA) housed in a cell culture incubator (37 °C, 95 % air, 5 % CO₂). Time matched control cells were grown on the same flexible-bottomed culture plates, but were not stretched. In order to test the impact of buffering of intracellular Ca²⁺ whilst stretching, cells were pre-incubated for 1 h with BAPTA-AM (20 μM, Sigma-Aldrich, Gillingham, UK) or vehicle control (0.1 % dimethyl sulfoxide) before commencement of the 6 h tonic mechanical strain protocol. Subsequent to treatments, cells were washed with PBS and RNA extracted.

Total RNA was extracted from human myometrial tissue from term non-labouring (TNL) women (~30 mg per sample) was homogenized using a TissueLyser (Qiagen, Crawley, UK) in Trizol® (Invitrogen, Paisley, UK) as per manufacturer’s recommendations. RNA was extracted from cultured myometrial cells using the RNeasy mini kit (Qiagen, UK) according to the manufacturer’s instructions. RNA samples were quantified using a Thermo Scientific NanoDrop ND-1000 spectrophotometer (Labtech, International Ltd, East Sussex, UK). The absorbance of each sample at 260 nm was used to calculate RNA concentration and an optical density 260/280 ratio was calculated to determine RNA purity. An OD 260/280 ratio of more than 1.8 was considered acceptable [42].

cDNA was synthesised from 500 ng RNA using 0.25 μg random hexanucleotide primers (Promega, Southampton, UK) and 200 IU Superscript III (Invitrogen, Paisley, UK). Quantitative real-time PCR was performed with the use of SYBR Green chemistry (Bioline, London, UK) and a RotorGene 6000 Sequence Detector (Qiagen, Crawley, UK) using primer sets as listed in Table I. The PCR cycling conditions were as follows: an initial denaturation step of 95 °C for 10 min, then 45 cycles of a programme consisting of denaturation (95 °C, 15 s), annealing (60 °C, 30 s) and extension (72 °C, 20 s). Melt curve analysis confirmed a single product, and all products were sequenced to confirm identity. Negative reverse transcription reactions were used routinely to assess genomic DNA contamination. Test samples were run in duplicate in parallel with cDNA standards of known gene copy number abundance (10⁶ to 10¹ copies). Cycle threshold values were used for analysis and abundance data was obtained for test samples by the generation of a standard curve based on the quantified cDNA. Abundance data for NFAT isoform expression for cell studies, Figures 1, 2, 4 and 5 were expressed relative to the geometric mean of a panel of housekeepers (glyceraldehyde 3-phosphate dehydrogenase (GAPDH), β-actin and β-2 microglobulin) as determined by GeNorm software [43] and for tissues only GAPDH was used due to limited cDNA availability and previous assessment which showed GAPDH was the most stable housekeeper for tissue analysis (Fig. 6). All data were log₁₀ transformed.

Myometrial whole cell lysates were prepared by removing culture medium and rising twice with ice-cold PBS. The cells were then aspirated to dryness and 100 μl of lysis buffer (10 mM of Hepes-KOH [pH 7], 1 mM of dithiothreitol, 1 % nonident-P40, and protease-inhibitor cocktail [COMPLETE tablets; Boehringer-Mannheim Biochemicals, Lewes, Sussex, U.K.]) was added to each well and incubated on ice for 5 min. Cells were then scraped and the lysate removed and placed in to a 1.5 ml microfuge tube and centrifuged for 1 minute at 13,000 rpm to remove any cell debris. Tissue and whole cell lysate protein content was quantified using a Thermo Scientific NanoDrop ND-1000 spectrophotometer. Lysates were then diluted 1:1 with x 2 Laemmli sample buffer (Sigma) then boiled at 95 °C for 5 min. Samples were then stored at −20 °C until required, when they were thawed, then re-boiled for 5 min at 95 °C and centrifuged at 12470 g for 1 min prior to electrophoresis.

Whole cell lysate proteins were separated using 10 % Tris-Glycine precast gels (Invitrogen) using the XCell SureLock™ Mini-Cell system (Invitrogen). Following electrophoresis, proteins were transferred to Immobilion™-P transfer membrane (Millipore) using the XCell SureLock™ Mini-Cell blotting module wet transfer blotting system. After transfer of proteins to the membrane, non-specific sites were blocked by soaking the membrane in 100 % methanol for 10 s and then allowing the membrane to dry completely for approximately 15 min. Membranes were then incubated for at least 5 h with maximum speed of agitation at room temperature or overnight at 4 °C with the appropriate NFATc2 primary antibody (Abcam: AB2722) diluted 1:1000 or β-actin (Abcam: AB6276) diluted 1:5000 in tris-buffered saline with tween (TBS-T: 50 mM Tris, 150 mM NaCl, 0.2 % (v/v) Tween-20, pH 7.4) containing 10 % (w/v) BSA. Membranes were then washed for 3 × 20 min in TBS-T. Following the incubation of the membrane in a 1:10,000 dilution in TBS-T of horseradish peroxidase (HRP)-conjugated goat anti-mouse secondary antibody (BD Pharmigen) for 45 min, the membrane was washed a further 3 × 20 minutes in TBS-T. Immunoreactive proteins were visualised using enhanced chemiluminescence (ECL™) (Amersham) according to the manufacturer’s instructions. Densitometric quantification of immunoreactive bands was carried out using BioRad Molecular Quantity One software, version 4.4.0. To verify equal protein loading, blots were also probed with β-actin antibody.

NFAT mRNA for all isoforms (c1-c4) was detected in human myometrial tissue (n = 6). NFAT isoforms c1-c4 were also expressed in P0 (n = 6) and P2 (n = 6) human myometrial cells, but NFATc1, NFATc2 and NFATc3 expression was suppressed in P0 and P2 human myometrial cells compared to myometrial tissue. NFATc2 expression was lower in P2 compared to P0 cells (95 % CI: 71.4 to 91.7, P < 0.001). (Fig. 1).

Application of the Ca²⁺ ionophore A23187 caused a 20.6 times (95 % CI: 7.6 to 55.6, P < 0.001, n = 6) increase in NFATc2 gene expression over 6 h compared to vehicle control (Fig. 2). There was no significant effect of Ca²⁺ ionophore A23187 treatment on NFATc1, NFATc3 or NFATc4 expression. Incubation of human myometrial cells with Ca²⁺ ionophore A23187 for 14 h significantly increased NFATc2 protein expression (p < 0.05, n = 6) (Fig. 3).

In vitro stretch of cultured (P2) human myometrial cells caused an 11.5 times (95 % CI: 4.3 to 30.9, P < 0.001, n = 6) increase in NFATc2 gene expression over 6 h compared to non-stretched cells (Fig. 4). There was no effect of in vitro stretch on the expression of any other of the NFAT isoforms.

There was a small effect (1.9 fold, 95 % CI: 1.3 to 2.7, P < 0.05, n = 10) of BAPTA-AM on NFATc2 expression in non-stretched cells (change from t = 0 h to t = 6 h) but not in the vehicle control (Fig. 5), but there was no difference between absolute NFATc2 mRNA expression in BAPTA-AM treated cells compared to cells exposed to vehicle at the 6 hour time point. NFATc2 gene expression was significantly increased 11.0 times (95 % CI: 4.6 to 26.4, P < 0.001, n = 10) in stretched cells, but there was a 90 % reduction (95 % CI: 72.3 % to 96.4 %, P < 0.001, n = 10) in the effect of stretch on NFATc2 gene expression in the presence of BAPTA-AM.

NFATc2 mRNA expression was detected in myometrial tissues taken from women delivering preterm and term gestations. There was no significant difference in tissues taken from women prior to or during labour (Fig. 6).