Biochemical Journal

Research article

The progesterone receptor as a transcription factor regulates phospholipase D1 expression through independent activation of protein kinase A and Ras during 8-Br-cAMP-induced decidualization in human endometrial stromal cells

Ju Hwan Cho , Mee-Sup Yoon , Jun Bon Koo , Yong Seok Kim , Ki-Sung Lee , Jung Han Lee , Joong-Soo Han

Abstract

Decidualization is a biological and morphological process occurring in hES (human endometrial stromal) cells. Previously, we reported that PLD1 (phospholipase D1) plays an important role in cAMP-induced decidualization of hES cells. In the present study, we focused on how PLD1 expression is up-regulated during decidualization. Treatment with PKA (protein kinase A) inhibitors (Rp-cAMP or H89) or a Ras inhibitor (manumycin) partially inhibited PLD1 expression and decidua formation in response to cAMP treatment. Interestingly, dual inhibition of PKA and Ras completely inhibited PLD1 expression and cAMP-induced decidualization. These results suggest that PLD1 expression during decidualization is controlled additively by PKA and Ras. The use of inhibitors showed that extracellular-signal-regulated kinase, a downstream effector of Ras, was required for PLD activation and the morphological changes during decidualization, but not for the increase in PLD1 protein. Next, to investigate the regulator of the PLD1 gene at the transcriptional level, a promoter assay using deletion mutants of the PLD1 promoter was performed; the result indicated that PR (progesterone receptor) was a possible regulator of the PLD1 gene. In addition, chromatin immunoprecipitation assays on the PLD1 promoter identified PR as a transcription factor for PLD1 expression during 8-Br-cAMP-induced decidualization. Taken together, our findings demonstrate that PKA and Ras are novel regulators of PLD1 expression and also identify PR as a transcription factor for PLD1 expression during the decidualization of hES cells.

  • decidualization
  • human endometrial stromal cell (hES cell)
  • phospholipase D1 (PLD1)
  • progesterone receptor (PR)
  • Ras
  • protein kinase A (PKA)

INTRODUCTION

PLD (phospholipase D) has been proposed to play a role in cell differentiation; for example, regulated transcription of PLD1 and/or PLD2 has been observed in C6 glioma cells [1], and sustained activation of PLD has been reported during the differentiation of keratinocytes [23] and myocytes [4]. In addition, we have observed that PLD1 has an important role in decidualization of hES [human ES (endometrial stromal)] cells [5]. In response to various stimuli, PLD generates PA (phosphatidic acid) and choline by hydrolysing PC (phosphatidylcholine) [6,7]. Two major isoforms of PC-specific PLD have been identified in mammals, namely PLD1 and PLD2. PLD1 can be directly activated by three families of proteins: cPKC [conventional PKC (protein kinase C)], Rho GTPases (RhoA, Cdc42 and Rac1), and ARF (ADP-ribosylation factor) GTPase [6,8]. In contrast, PLD2 exhibits high basal activity, although it has also been reported tobe regulated by ARF and PKC [8]. However, the question of how PLD expression is regulated during differentiation, particularly during decidualization of hES cells, has not previously been addressed.

Decidualization is crucial to embryo implantation and the maintenance of pregnancy [9]. The decidualization of hES cells, which is observed in the late secretory stage of the menstrual cycle, is characterized by morphological and functional differentiation [10]. A co-ordinate induction of muscle-specific gene products, such as IGFBP-1 (insulin-like growth factor binding protein-1) and PRL (prolactin), occurs concomitantly with the morphological changes of hES cells [11,12]. These gene products in turn influence both decidualization and trophoblast functions. Progesterone is known to play a role in maintaining the decidual phenotype, but the initiation of decidualization requires a high level of intracellular cAMP and sustained activation of the PKA (protein kinase A) pathway [13,14]. Previously, we reported that 8-Br-cAMP (a cAMP analogue) up-regulated PLD1, and PA, as a product of PLD1, regulated decidualization through PLA2 (phospholipase A2) to maintain decidua forms of hES cells [5,15]. However, the precise mechanism by which cAMP induces PLD1 remains unclear. Signalling through cAMP attenuates ligand-dependent SUMOylation of PR (progesterone receptor) in hES cells [16], and enhances the hormone-dependent transcriptional activity of PR, as well as being able to convert some antiprogestins, such as RU486, into PR agonists [28]. Even though the mechanism by which cAMP potentiates hormone-independent PR activity is not entirely understood, it seemed possible that PR, acting as a transcription factor, regulates the expression of certain genes. Therefore we attempted to elucidate whether PR regulates PLD1 expression by performing deletion-mutant-based promoter assays and PLD1-promoter-sequence-based transcription factor prediction during cAMP-induced decidualization in hES cells. In the present paper, we provide evidence that PLD1 expression is regulated independently by PKA and Ras through PR acting as a transcription factor, and that the up-regulation of PLD1 plays a role in inducing and maintaining the decidualization of hES cells.

EXPERIMENTAL

Materials

Materials for tissue culture, such as FBS (fetal bovine serum), penicillin, streptomycin, and DMEM (Dulbecco's modified Eagle's medium, low glucose), were obtained from Gibco-BRL, and progesterone (also called P4), 8-Br-cAMP, manumycin and PD98059 were purchased from Sigma–Aldrich. Rp-cAMP and N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide dihydrochloride (also called H-89) were obtained from Calbiochem and [3H]palmitic acid was from Du Pont/New England Nuclear. pCMV-MEK2A was obtained from Dr Gary L. Johnson (National Jewish Medical Research Center, Denver, CO, U.S.A.), and pcDNA3.1-H-Ras N17 [17] was from Dr David Stokoe (ONYX Pharmaceuticals). Antibodies used were as follows: anti-β-actin, anti-PR, polyclonal anti-ERK (extracellular-signal-regulated kinase) 1/2 and monoclonal anti-p-ERK1/2 (Cell Signaling Technology), and monoclonal anti-pan-Ras (Oncogene Research Products). A polyclonal antibody that recognizes both PLD1 and PLD2 was generously provided by Dr Do Sik Min (Pusan National University, Pusan, Korea). L-α-PBt (phosphatidic butanol) was from Avanti Polar Lipids. The silica gel 60 Å (1 Å=0.1 nm) plates for TLC were purchased from Whatman. All other chemical agents were of analytical grade.

Isolation and culture of hES cells

Human endometria were obtained by hysterectomy from 20 pre-menopausal women, aged 35–44 years, who underwent surgery for non-endometrial abnormalities at Hanyang University Hospital between September 2006 and September 2007. Samples were collected under the protocols approved by the Institutional Review Board of Hanyang University Hospital, and written informed consent was obtained from all participants. A portion of each endometrial specimen obtained was examined histologically. ES cells were isolated as described previously [15]. Briefly, tissue samples were collected in DMEM containing 100 IU/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine and 10% (v/v) FBS. After cleaning and trimming to remove blood clots and mucus, the specimens were minced to fragments of less than 1 mm in size under a laminar flow hood and digested at 37°C for 60 min with 0.25% collagenase I and deoxynuclease (Sigma). The cell suspension was filtered twice through a 40-μm-pore-size sieve (BD Falcon). After enzymatic digestion, most of the stromal cells were present as single cells or small aggregates. The purity of the stromal cells obtained by this method was usually >90%, as determined by immunocytochemical staining for vimentin (a stromal cell marker). The purified stromal cells were washed, and viable cells were counted by dye exclusion using Trypan Blue. The viability of the isolated cells was at least 90% in each experiment.

In vitro decidualization

Aliquots (10 ml) of hES cells were cultured to confluence in 100-mm-diameter culture dishes at 37°C in DMEM supplemented with 10% (v/v) FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere of 5% CO2 in air. To induce in vitro decidualization, cells were exposed to 0.5 mM 8-Br-cAMP for 3–6 days. The culture medium was changed every 3 days, with continuous supplementation with 8-Br-cAMP. Phase-contrast microscopy was used to verify the morphological changes associated with differentiation in response to 8-Br-cAMP.

In vivo PLD assay

PLD activity was determined as described previously by measuring [3H]PBt produced through PLD-catalysed transphosphatidylation in cells labelled with [3H]palmitic acid [15]. Briefly, hES cells cultured in six-well plates were metabolically labelled with 1 μCi/ml of [3H]palmitic acid for 24 h. They were then pretreated with 0.3% 1-butanol for 15 min, quickly washed with ice-cold PBS and suspended in ice-cold methanol. Lipids were extracted according to the method of Bligh and Dyer [18], and [3H]PBt was separated from other phospholipids by TLC on silica gel 60 Å plates, using ethyl acetate/iso-octane/acetic acid/water (26:4:6:20, by vol.) as the solvent system. The regions corresponding to authentic PBt bands were identified with 0.002% primulin in 80% (v/v) acetone, scraped and counted with a scintillation counter.

Western blot analysis

hES cells were suspended by scraping in ice-cold PBS and harvested by microcentrifugation (3000 g, 5 min). They were then resuspended in a buffer solution (20 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 2.5 mM sodium pyrophosphate, 1% Triton X-100, 1 mM PMSF and 1 mM Na3VO4) and disrupted by sonication using an Ultrasonic processor (2 W/cm2, six cycles each of 10 s on /10 s off). Proteins were resolved by SDS/PAGE (10% gel) and transferred to nitrocellulose membranes. The membranes were blocked for 1 h with TTBS (Tris-buffered saline containing 0.01% Tween 20) containing 5% (w/v) non-fat dried skimmed milk powder and incubated for an additional 1 h with primary antibodies (1 μg/ml). Unbound primary antibodies were removed by three washes (10 min each) with TTBS. The blots were then incubated with HRP (horseradish peroxidase)-conjugated secondary antibody (1:2000; New England Biolabs), and specific bands were detected by ECL (enhanced chemiluminescence; GE Healthcare).

Activated Ras affinity precipitation assay

RBD (Raf/Ras-binding domain) fragment (Raf-1 amino acid residues 1–149) fused to GST (glutathione transferase) was purchased from Upstate Biotechnology. The fusion protein was immobilized on glutathione–agarose. The activated Ras affinity precipitation assay was performed as described in the manufacturer's protocol. Briefly, 500 μg of cell extract was incubated with 5 μg of GST–RBD complex for 30 min at 4°C. After washing the agarose beads five times with immunoprecipitation washing buffer (25 mM Hepes, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 10 mM MgCl2, 1 mM EDTA and 2% glycerol), the active Ras (Ras-GTP) bound to the GST–RBD was released by addition of 2× SDS/PAGE loading buffer (25 mM Tris/HCl, pH 8.3, 250 mM glycine and 0.1% SDS). The amount of active Ras was determined by immunoblotting with the anti-pan-Ras monoclonal antibody.

Luciferase promoter assay

The PLD1 reporter plasmid contained a human PLD1 promoter fragment (PLD1p −1874/+9) cloned upstream of the luciferase cDNA in pGL3-Basic vector (Promega). hES cells were transiently transfected with pGL3-Basic, the promoter-less vector or the PLD1 reporter plasmid (pGL3-PLD1p −1874/+9) using a Nucleofector™ kit (Amaxa). To correct for differences in transfection efficiency, cells were co-transfected with a pCMV–β-gal (β-galactosidase) construct. At 1 day after transfection, the cells were treated with 0.5 mM cAMP for 2 days to induce decidualization. Luciferase expression was analysed with a firefly-luciferase assay system (BD Pharmingen) using a microplate luminometer (Berthold). Relative firefly-luciferase activity was normalized by the β-gal activity in the corresponding cell lysate.

Construction of deletion mutants of the PLD1 promoter

Successive 5′-deletion mutants of the PLD1 promoter were constructed by PCR using the following sense primers: for −1735, 5′-CTCACAGCAACCTCCTCCTC-3′; for −1344, 5′-CAGTTACGTGTGTGCACAAGAC-3′; for −1137, 5′-ATGTCAACAAGGAAAGAGTGAGG-3′; for −887, 5′-ATGCTTTAAAGTGGCTGACG-3′; for −668, 5′-GAACGGTTCCATAGGTGGAATG-3′; for −354, 5′-AGTTGTGGCTACTATCCGCAG-3′; for −204, 5′-GGGAAAAAGAGAACAAGGAACGC-3′; and for −74, 5′-GAATCCGGGCGGAGAGTTCCTG-3′. The same antisense primer, 5′-GGGAGGGAGGGAGAGAGGGC-3′, was used in all of the PCRs to construct each of the deletion mutants. After construction, each deletion mutant was confirmed by direct sequencing.

RNA extraction and RT (reverse transcription)–PCR analysis

Total cellular RNA was prepared using Tri Reagent (Molecular Research Center) and following the recommendations of the manufacturer. A SuperScript® kit (Introgen Therapeutics) was used for the cDNA synthesis. The PCR reactions were carried out using standard protocols. Optimal MgCl2 concentrations and cycle numbers in the linear amplification range were first determined. Primer sequences (sense and antisense) were as follows: IGFBP-1, 5′-GAGAGCACGGAGATAACTGAGG-3′ and 5′-AGGATCCTCTTCCCATTCCA-3′; PLD1, 5′-TCTGGAGGACACAGAATACC-3′ and 5′-CTTCTTCCCCACCTTAATTT-3′; PR-A, 5′-ACCCAGCGGGTGTTGTCCCC-3′ and 5′-GCGGACCCGCAGACTCCTCG-3′; PR-B, 5′-GGTCCGGGGCCCCAGTGAAG-3′ and 5′-GCGACCCTGGGCGCTGAGAA-3′; and β-actin, 5′-CCCAGGCACCAGGGCGTGATC-3′ and 5′-TCAAACATGATCTGGGTCAT-3′. RT–PCR products were analysed with ethidium bromide-containing agarose gels.

ChIP (chromatin immunoprecipitation) assay

Nuclear isolation, cross-linking and ChIP assays were performed as described previously [19]. Precipitated DNA fragments were analysed by PCR using primers against the relevant human PLD1 promoters. Two sets of primers were used as follows (sense and antisense): set 1, 5′-GTATTCTGCCTTCCAGTAACAT-3′ and 5′-CAATCATTTATGTCAACAAGGA-3′ (product size, 146 bp); and set 2, 5′-TGCACAAGACTTATCAGACCA-3′ and 5′-TCCTCTTAGGCAATACTCTGC-3′ (product size, 126 bp). The PCR conditions comprised an initial polymerase activation step at 95°C for 5 min, followed by 33 cycles of denaturation at 95°C for 1 min, annealing at 58°C for 1 min and extension at 72°C for 1 min.

siRNAs (small interfering RNAs)

The single siRNA for PR (5′-GAGAUGAGGUCAAGCUACA-3′; this sequence targeted both PR-A and PR-B) and the negative control (5′-CCUACGCCACCAAUUUCGU-3′) were purchased from Bioneer. The oligonucleotide siRNAs were transfected using siPORT™ NeoFX (Ambion). At the indicated times after transfection, cell lysates were assayed for gene silencing by RT–PCR or Western blotting.

Statistical analysis

All data are presented as means±S.E.M.; P<0.05 was considered statistically significant. Comparisons between groups were analysed using ANOVA, with a Dunnett's test or unpaired Students' t test as indicated in the legends to the Figures.

RESULTS

PKA mediates PLD1 expression during 8-Br-cAMP-induced decidualization

Decidualization of hES cells was induced by incubating subconfluent cells in medium containing 0.5 mM 8-Br-cAMP and replacement with the same medium every 3 days. The cells were transformed in 2–3 days into large polygonal cells with enlarged nuclei and an increased amount of cytoplasm [5]. A major downstream factor reacting with cAMP is cAMP-dependent PKA [20], and activated PKA is required for differentiation-dependent transcription of the decidual PRL and IGFBP1 genes in hES cells [14,21]. In addition, we have reported previously that both PLD1 expression and PLD activity were increased, but another isotype of PLD, PLD2, was not involved during 8-Br-cAMP-induced decidual differentiation in hES cells [5,15]. Therefore we first tested whether PKA is involved in PLD expression and activity during decidualization. In the present study, two different inhibitors, H-89 and Rp-cAMP, were used for the inhibition of PKA. Both inhibitors had the same effect on the prevention of 8-Br-cAMP-induced decidualization, reduction of IGFBP1 mRNA expression and PLD1 expression (Figures 1A, 1B and 1D). However, as de la Rosa et al. [22] reported that H-89 had a non-specific effect on the endoplasmic reticulum-associated Ca2+ ATPase, we used Rp-cAMP, which binds competitively to the regulatory subunit of PKA and blocks cAMP-induced signalling, for further investigation. Next, we checked whether PKA affected PLD expression and activity during decidualization. PLD1 promoter activity was measured in the presence and absence of Rp-cAMP. After 2 days of cAMP-induced decidual differentiation, the reporter activity increased 11-fold in PGL3-PLD1-transfected hES cells when treated with cAMP; however, Rp-cAMP treatment suppressed the reporter activity by up to 73% (Figure 1C). At the same time, PLD1 expression was also decreased by PKA inhibition compared with the expression in cAMP-treated hES cells (Figure 1D), indicating that PKA regulates PLD1 expression during decidualization. H-89 treatment also strongly inhibited the cAMP-induced increase in PLD activity (Figure 1E), suggesting that the decrease in PLD1 expression accounts for the decrease in basal PLD activity. These results show that PKA up-regulates both the activity and expression of PLD1 during decidualization. However, Rp-cAMP did not completely inhibit the increase of PLD1 promoter activity, indicating that PLD1 expression may be controlled by some additional mechanism.

Figure 1 PKA is involved in cAMP-induced PLD1 expression during decidualization

(A and B) hES cells were treated with 0.5 mM 8-Br-cAMP in the presence or absence of 10 μM H-89 (H89) or Rp-cAMP. After culturing for 3 days, the cells were photographed under a microscope (A) and RNA was isolated from the hES cells. RT–PCR was performed to analyse IGFBP-1 (IGFBP1) as a marker of decidual cells (B). (C) hES cells were transiently transfected with a PLD1 reporter plasmid (pGL3-PLD1p −1874/+9) together with pCMV–β-gal, using a Nucleofector™ kit (Amaxa). The cells were incubated in the presence or absence of 10 μM Rp-cAMP (a PKA inhibitor) during 8-Br-cAMP-induced decidualization for 2 days. Luciferase activity was measured, and the results are presented as fold inductions over the control in triplicate measurements. (D) hES cells were allowed to differentiate into decidual cells with or without 10 μM H-89 or Rp-cAMP (PKA inhibitors) for 3 days in the presence of 8-Br-cAMP, and cell lysates were analysed by Western blotting for PLD1. (E) After incubation for the indicated periods with or without 10 μM Rp-cAMP during cAMP-induced decidualization, cells were labelled with [3H]palmitic acid and PLD activity was measured as described in the Experimental section. Untreated cells were kept cultured in the medium (DMEM with 10% FBS) until harvest and used as a control (CON). *Differences between treatments with and without Rp-cAMP (P<0.05) were assessed by Student's t test.

Ras mediates cAMP-induced PLD expression

To identify other upstream activators of PLD1 up-regulation, we investigated Ras activation during decidualization. There is considerable evidence of cross-talk between cAMP signalling and Ras signalling, and also that Ras signalling plays a predominant role in developmental processes [23]. Therefore we investigated whether Ras is activated during decidualization. To detect activated Ras, we used a Ras activity assay kit, which is based on the fact that only activated Ras (Ras-GTP) can bind to the RBD of Raf-1 [24]. Active GTP-bound Ras was pulled down from cell lysates with GST–Raf-RBD coupled to glutathione–agarose, and the fraction of activated Ras was determined by immunoblotting with an anti-Ras antibody. As shown in Figure 2(A), Ras was strongly activated during cAMP-induced decidualization. To determine whether this Ras activation affects PLD expression, we examined the effect of manumycin, a Ras inhibitor, on the activity of the PLD1 promoter. Manumycin treatment reduced luciferase reporter activity by up to 82% (Figure 2B) and, simultaneously, PLD1 expression was suppressed by manumycin treatment (Figure 2C), implying that Ras is also involved in PLD1 expression during decidualization.

Figure 2 Ras regulates PLD1 expression during decidualization

(A) hES cells were harvested after 3 days of differentiation. The fraction of active Ras was determined by affinity precipitation assays as described in the Experimental section. The experiments were performed three times and a representative result is shown. (B) hES cells were transiently transfected as described in the legend to Figure 1. They were then permitted to differentiate with or without 10 μM manumycin (a Ras inhibitor) during 8-Br-cAMP-induced decidualization for 2 days. Luciferase activity was measured, and the results are presented as fold inductions over the control in triplicate measurements. *Differences between treatments with and without manumycin (P<0.05) were assessed by Student's t test. (C) hES cells were differentiated into decidual cells with or without 10 μM manumycin for 3 days in the presence of 0.5 mM 8-Br-cAMP, and cell lysates were analysed by Western blotting for PLD1. Untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest and used as a control.

Since manumycin may have nonspecific effects, we set out to check the role of Ras using a DN-Ras (dominant-negative Ras; Ras-Asn17). Transfection of hES cells with Ras-Asn17 resulted in inhibition of the decidualization-associated morphological changes (Figure 3A) and reduced the cAMP-induced increase in PLD1 promoter activity (Figure 3B) markedly. Moreover, transfection of Ras-Asn17 reduced basal PLD1 promoter activity by up to 70% in the absence of 8-Br-cAMP treatment, suggesting that inhibition of Ras was sufficient to decrease PLD1 promoter activity. Ras-Asn17 also inhibited Br-cAMP-induced PLD1 expression (Figure 3C). Then, to confirm the involvement of Ras in PLD up-regulation, hES cells were transfected with Ras-Asn17 or H-Ras, and were incubated without 8-Br-cAMP for 3 days. As shown in Figure 3(D), H-Ras overexpression, by itself, was sufficient to substantially increase PLD1 expression. Taken together, these results establish that Ras contributes to PLD1 expression.

Figure 3 DN-Ras blocks PLD1 expression and decidua formation

(A) hES cells were transfected with DN-Ras and differentiated with 8-Br-cAMP for 3 days, then the morphology of the cells was photographed under a light microscope. (B) hES cells were co-transfected transiently as described in the legend to Figure 1. They were incubated with 0.5 mM 8-Br-cAMP for 2 days after transfection. Luciferase activity was measured, and the results are presented as fold inductions over the control in triplicate measurements. *Differences between pcDNA3.1 and DN-Ras with cAMP (P <0.05) were assessed by Student's t test. (C) DN-Ras-transfected cells were differentiated into decidual cells for 3 days in the presence of 8-Br-cAMP, and cell lysates were analysed by Western blotting for PLD1 and Ras. (D) Cells were transfected with DN-Ras or H-Ras, and proteins were isolated from hES cells after culture for 3 days without 8-Br-cAMP. Untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest and used as a control.

ERK1/2 is involved in 8-Br-cAMP-mediated PLD activation

The Ras/Raf/MEK [MAPK (mitogen-activated protein kinase)/ERK kinase]/ERK pathway is another ancient signalling pathway [23]. Moreover, MAPK is one of the targets of cAMP. When cAMP stimulation activates ERK, it can induce cell differentiation as well as proliferation [25]. In addition, we previously reported that, in undifferentiated hES cells, exposure to 8-Br-cAMP stimulated PLD activity by activating ERK [15]. Consequently, we questioned whether Ras induces PLD expression during decidualization by activating ERK. As shown in Figure 4(A), ERK1/2 was activated progressively during cAMP-induced decidualization. To address a possible role of ERK in decidualization, we used a dominant-negative form of MEK (DN-MEK) to inhibit ERK1/2. When ERK1/2 was blocked by DN-MEK, transfection decidualization was substantially reduced (Figure 4B), and the increase in PLD activity was effectively blocked as well (Figure 4C). Exposure to PD98059, a MEK inhibitor, also decreased the number of decidual cells and PLD activity (results not shown). However, surprisingly, the induction of PLD1 protein was not affected by transfection of DN-MEK (Figure 4D) or by treatment with PD98059 (results not shown). These results indicate that ERK regulates PLD1 activation, but not PLD1 expression, and suggest that Ras regulates PLD1 expression through regulators other than ERK.

Figure 4 ERK activation regulates PLD activation, but not PLD1 expression

(A) hES cells were differentiated for different periods in the presence of 8-Br-cAMP, and then ERK and p-ERK proteins were detected in cell lysates. (BD) DN-MEK-transfected hES cells were differentiated with 8-Br-cAMP for 2 days and photographed under a light microscope (B), assayed for in vivo PLD activity (C), and Western blotted for p-ERK and PLD1 (D) as described in the Experimental section. *Differences between pcDNA and DN-MEK with cAMP (P<0.05) were assessed by Student's t test. Control (0 day): untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest.

PKA and Ras mediate PLD1 expression independently

To determine the relationship between the effects of PKA and Ras on PLD1 expression, we investigated Ras activation after blocking PKA. As shown in Figure 5(A), Rp-cAMP treatment did not inhibit 8-Br-cAMP-induced Ras activation during decidualization, suggesting that PKA is not involved in Ras activation. To see whether Ras and PKA regulate PLD1 expression independently, we treated cells simultaneously with manumycin and Rp-cAMP. As seen in Figure 5(B), PLD1 promoter activity was reduced by 70% by manumycin alone, and by 69% by Rp-cAMP, whereas the two agents together further inhibited PLD1 promoter activity, indicating that Ras and PKA regulate independently in activating the PLD1 promoter. Consistent with these results, PLD1 expression in response to 8-Br-cAMP was only completely blocked when both agents were present (Figure 5C). In order to confirm that PKA and Ras are indispensable for decidualization, we investigated the morphological changes after treatment with manumycin or/and Rp-cAMP. As shown in Figure 5(D), inhibition of decidua formation was more pronounced when hES cells were co-treated with both manumycin and Rp-cAMP, as compared with single-treatment with each inhibitor. These results showed that inhibition of PLD1 expression by manumycin or/and Rp-cAMP blocks decidualization. Taken together, PKA and Ras are requisite for decidualization and each independently regulates PLD1 expression.

Figure 5 PKA and Ras regulate PLD1 expression independently

(A) hES cells were harvested after 3 days of differentiation with or without 10 μM Rp-cAMP. The fraction of active Ras was determined by an affinity precipitation assay as described in the Experimental section. The experiment was repeated three times with similar results, and a representative result is shown. (B) hES cells were transfected as described in the legend to Figure 1. They were differentiated with or without manumycin (10 μM) or Rp-cAMP (10 μM) in the presence of 8-Br-cAMP for 2 days, and then luciferase activity was measured; the results are presented as fold inductions over the control in triplicate measurements. *Data were analysed by one-way ANOVA with post-hoc testing by Dunnett's test (P<0.05). NS, not significant. (C and D) hES cells were treated with manumycin or Rp-cAMP in the presence of 8-Br-cAMP for 3 days. The cell lysates were analysed by Western blotting for PLD1 (C) and the cells were photographed under a microscope (D) as described in the Experimental section. Untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest and used as a control.

PR acts as a transcription factor for PLD1 during decidualization

To identify the transcription factor responsible for PLD1 expression during 8-Br-cAMP-induced decidualization, we first constructed deletion mutants of the promoter of PLD1 (Figure 6A, left-hand side). As shown in Figure 6(A) (right-hand-side), promoter activity was dramatically decreased when the 5′-deletion extended to position −1137 of the promoter when hES cells were transiently transfected with luciferase reporter derivatives and cultured for 2 days in the presence of 8-Br-cAMP. On the basis of this result, several transcription factor candidates were identified by Signal Scan (http://www-bimas.cit.nih.gov/molbio/signal/) using the sequence from −1874 to −1136. One of the strongest candidates was the PR, since it has strong transcriptional activity in the presence of elevated cAMP levels [26]. Therefore we checked whether expression of either of the two isotypes of PR (PR-A and -B) changed during 8-Br-cAMP-induced decidualization. Indeed, expression of both PR-A and -B was found to increase progressively during decidualization along with IGFBP-1 and PLD1. However, progesterone (P4) treatment along with cAMP had no synergistic effect on PR, IGFBP1 and PLD1 expression during decidualization (Figure 6B). As shown in Figure 6(C) (upper panel), a binding site for PR is present in the PLD1 promoter (−1265 to −1258) and ChIP assays revealed significant and specific binding of PR to the promoter (Figure 6C, lower panel). Furthermore, PR siRNA reduced binding of PR to the PLD1 promoter (Figure 6C, lower panel). These results demonstrate that PR can bind to and activate the PLD1 promoter.

Figure 6 PR is important for the regulation of PLD1 expression during decidualization

(A) Schematic representation of full-length PLD1 promoter and deletion mutants. A palindromic sequence bound by PR is located at −1265 to −1258 (TGTTACT) in the PLD1 promoter region [this PR-binding site was predicted using Signal Scan (http://www-bimas.cit.nih.gov/molbio/signal/)]. Each construct was cloned into the pGL3-Basic reporter vector (left) and hES cells were co-transfected transiently with the various constructs and pCMV–β-gal. The transfected hES cells were differentiated in the presence of 8-Br-cAMP for 2 days. Luciferase activity was measured, and the results represent fold inductions over the control in triplicate measurements. *Differences between −1344 and −1137 were assessed by Students t test (P<0.05). (B) RNA was isolated from the cells at the indicated times and RT–PCR was performed to analyse IGFBP1, PLD1, PR-A/-B and β-actin. Control (0 day): untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest and used as a control. P4, progesterone. (C) ChIP analysis of binding of PR to the PLD1 promoter with and without PR knockdown. To detect the ChIP DNA fragment using anti-PR, two pairs of primers containing the PR-binding motif in the PLD1 promoter (bidirectional arrows) were used, as indicated in the upper panel of (C). hES cells were differentiated in the presence of 8-Br-cAMP for 2 days with or without transient transfection of PR siRNA or control siRNA and cells were collected for ChIP assays using the anti-PR antibody (lower panel; RT-PCR results). *Differences between PR-binding-motif-containing (full length, −1735, and −1344) constructs and the PR-binding-motif-deleted (−1137) mutant of PLD1 promoter (P<0.05) were measured by Student's t test.

Next, we tested whether PR in fact regulates PLD1 expression. As shown in Figure 7, inhibition of PR with PR siRNA resulted in decreased PLD1 expression (Figure 7A), blocked the morphological change associated with decidualization (Figure 7B) and dose-dependently decreased promoter activity (Figure 7C). These results indicate that PR acts as a transcription factor for PLD1 expression during decidualization. Finally, we enquired whether PR was regulated through PKA and/or Ras, and found that 8-Br-cAMP-induced expression of PR (PR-A and PR-B) and PLD1 was inhibited by exposure to Rp-cAMP and by transfection of DN-Ras (Figure 7D). These results indicate that PLD1 expression is regulated by PR as a transcription factor through the independent activations of PKA and Ras during 8-Br-cAMP-induced decidualization in hES cells.

Figure 7 Knockdown of PR by siRNA inhibits PLD1 expression and 8-Br-cAMP-induced decidualization

(A and B) hES cells were differentiated into decidual cells for 3 days in the presence of 0.5 mM 8-Br-cAMP after transient transfection with siPR or scrambled siRNA (Scramble). Proteins were isolated and Western blotting was performed to detect PLD1 and PR (A), and the cells were photographed under a light microscope (B). (C) hES cells were co-transfected transiently with either the promoter-less vector (pGL3-Basic), full-length PLD1 promoter (−1874) or a deletion mutant (−1137) of the PLD1 promoter and pCMV–β-gal along with siPR or scrambled siRNA. They were differentiated in the absence or presence of 8-Br-cAMP for 2 days. Luciferase activity was measured, and the results represent fold inductions over the control (untreated) in triplicate measurements. *P<0.05, difference in PLD1 promoter activity between the addition or not of siPR (200 nM); both differences between the −1874 constructs were measured by Student's t test. (D) hES cells were differentiated into decidual cells for 3 days in the presence of 8-Br-cAMP together with 10 μM Rp-cAMP (PKA inhibitor) and/or after DN-Ras transfection. RNA was isolated, and RT–PCR was performed to analyse IGFBP-1 as a decidual marker (upper panel). The cell lysates were analysed by Western blotting for PLD1 and PR (lower panel). Untreated cells were cultured in the medium (DMEM with 10% FBS) until harvest and used as a control.

DISCUSSION

Decidualization involves dramatic changes in the biochemical properties as well as morphological appearance of ES cells in order for implantation to occur [10,11]. The decidualized ES cell becomes rounded, acquires myofibroblast characteristics and secretes a variety of phenotypic antigens, including PRL and IGFBP-1 [27,28]. Previously, we reported that PLD1 regulates the decidualization-specific phenotype on the basis of the following observations: knockdown of PLD1 by siRNA, or blocking of PLD activity with 0.3% 1-butanol decreased PRL/IGFBP-1 expression, whereas overexpression of PLD1 or treatment with PA (an end-product of PLD) increased PRL/IGFBP-1 expression and morphological change [5]. According to these results, we suggested that increased PLD activation was mainly due to the increased PLD1 expression in cAMP-treated hES cells. In the present study, we investigated how PLD1 expression is regulated during cAMP-induced decidualization.

The PKA signal transduction pathway mediates most or all of the actions of cAMP in inducing decidualization [13,14,26,29,30]. cAMP signalling to the nucleus is transduced almost exclusively by binding of cAMP to the R subunit of PKA and subsequent phosphorylation of nuclear target proteins by the released and translocated C subunit [31]. Sustained elevation of intracellular cAMP significantly reduces R (RIα) subunit expression during decidualization in ES cells [14]. As a consequence, the R/C subunit ratio falls, the endometrial PKA system does not desensitize to the sustained cAMP stimulus and PKA can be continuously active. In the present study, we demonstrated that PKA inhibitors markedly reduced the formation of decidual cells in response to 8-Br-cAMP (Figure 1A), accompanied by a decrease in PLD1 expression and PLD activity (Figures 1D and 1E). However, both PLD1 expression and promoter activity were not decreased to the basal level. These results indicated that not only PKA but also other signalling molecules may be involved in the expression of PLD1 during decidualization in hES cells.

Even though an increased intracellular concentration of cAMP results in activation of PKA, which in turn phosphorylates gene-regulatory transcription factors, several studies demonstrate that Ras is reported to mediate some effects of cAMP (e.g. [23]). This mechanism occurs independently of PKA signalling: cAMP-stimulated progesterone expression occurs through direct activation of the Ras protein in an in vitro steroidogenic cell model [32]. A Ras-specific activator called CNrasGEF [cyclic nucleotide ras GEF (guanine-nucleotide-exchange factor)] is activated by cAMP in melanocytes, and Rap1, a member of the Ras family of GTPases, is activated by cAMP during endothelial cell–cell contact [33,34]. In addition, Liu et al. [35] and Lee and Han [36] reported that an exchange protein directly activated by cAMP [Epac2, a GEF for Rap (a Ras-like small GTPase)] is not only a cAMP sensor, but also a Ras effector in PC12 cells and in Xenopus laevis. They showed that coincident detection of cAMP and Ras signals is essential for Epac2 to activate Rap1 in a temporally and spatially controlled manner. These results suggest that there may be a PKA-independent route by which cAMP activates Ras. We observed that Ras was activated during cAMP-induced decidualization (Figure 2A) and that inhibition by DN-Ras resulted in a decrease of decidua-forming cells (Figure 3A) as well as PLD1 expression (Figure 3C). However, Rp-cAMP, a PKA inhibitor, did not affect Ras activation (Figure 5A), whereas manumycin, a Ras inhibitor, reduced decidualization, PLD1 promoter activity, and PLD1 expression in cAMP-treated hES cells. Moreover, these reductions were completely abolished when Ras and PKA were co-inhibited by their specific inhibitors (Figures 5B–5D). These results suggest that Ras regulates PLD1 expression and decidualization in a PKA-independent manner.

Previously, Kang et al. [37] reported that phorbol ester up-regulates PLD1 expression through a PKC/Ras/ERK/NF-κB (nuclear factor κB)-dependent pathway in colon cancer cells. They showed that ERK is a critical regulator of PLD1 expression, whereas PKA is not, which is different from our present model. However, their report strongly supports the idea that Ras can be an important regulator of PLD1 expression. Ras activates several signalling pathways through a number of effectors [38]. One of them is the protein kinase Raf/MEK/ERK pathway [23]. In the present study, we also observed that ERK1/2 was activated during decidualization (Figure 4A), in agreement with previous observations [39,40]. Previously, we suggested that treatment with 8-Br-cAMP increased PLD activity through ERK1/2 [15]. Therefore, to evaluate whether ERK activates PLD to induce decidualization, we transfected hES cells with DN-MEK to block ERK activation and found that DN-MEK expression reduced decidualization (Figure 4B) and PLD activity (Figure 4C), but did not affect the expression level of PLD1 protein (Figure 4D), suggesting that ERK promotes decidualization by activating PLD1, but not by increasing its expression. A previous study demonstrated that rapid activation of ERK1/2-dependent intracellular signalling cascades by growth factors and oestrogens is involved in the migration of normal ES cells [41]. However, it is still unclear how ERK/PLD activation influences decidualization.

To identify the transcription factor involved in PLD1 expression during 8-Br-cAMP-induced decidualization in hES cells, we constructed deletion mutants of the PLD1 promoter, and assays using these deletions revealed several transcription factor candidates, PR notably among them (Figure 6A). PLD1 promoter activity was dramatically decreased when cells were transiently transfected with a 5′-deletion mutant extended to position −1137. However, promoter activities between −1137 and −204 were still higher than the control. This result indicated that another regulatory element(s) may exist between −204 and −1. As shown in Supplementary Figure S1 (at http://www.BiochemJ.org/bj/436/bj4360181add.htm), these higher promoter activities were diminished by deletion of the proximal promoter region (−1 to −74) which contains binding sites to transcriptional regulatory factors, such as AP (activator protein)-1, AP-2 and Sp1 (stimulating protein-1); seven putative binding sites for Sp1 are located in this proximal promoter region. The region between −204 and −74 also contained four binding sites for Sp1 and three binding sites for GR (glucocorticoid receptor). Gao et al. [42,43] reported that dexamethasone treatment has no effect on PLD1 promoter activity, but Sp1 is important in Ras-induced PLD1 expression. This result indicated that PR may regulate PLD1 expression with Sp1, not with GR, during decidualization of hES cells. PR is a member of the superfamily of ligand-activated transcription factors that bind to regulatory regions of their target genes. Two isoforms, PR-A and PR-B, exist, and PR-B differs from PR-A in that it contains an additional 164 amino acids at its N-terminus (Figure 7A) [44]. In general, the PR-A isoform is transcriptionally much less active and functions as a dominant inhibitor of transcription by PR-B and various other steroid receptors. However, we did not check which isoform of PR is important for up-regulation of PLD1 expression, since both are increased during decidualization (Figure 6B). The results from the ChIP assay with PR siRNA indicate that PR can act as a transcription factor for PLD1 (Figure 6C). Consequently, expression of PLD1 and decidualization of hES cells were inhibited by PR siRNA.

Hormone binding is not an absolute requirement for the activation of steroid receptors, even though PR binding with its ligand is the most important initiation step in the induction of a conformational change in the receptor, such as phosphorylation, dimerization and SUMOylation [4547]. For instance, elevated intracellular cAMP levels and activation of the PKA pathway can induce ligand-independent activation of cPR (chicken PR), AR (androgen receptor) and ER (oestrogen receptor) [4850]. PR appears to be activated by cAMP in a progesterone-independent fashion in the present system. Progesterone treatment did not increase expression of IGFBP1, PLD1 and PR. Moreover, co-treatment of progesterone with cAMP had no synergistic effect on the expression of these genes (Figure 6B). PLD1 promoter activity decreased in parallel with down-regulation of PR by PR siRNA, and expression of PR (both PR-A and PR-B) was decreased by inhibition of PKA or Ras (Figures 7C and 7D).

In conclusion, we have shown that PKA and Ras are closely involved in regulating PLD1 expression during 8-Br-cAMP-induced decidualization of hES cells. Thereafter, the ligand-independent PR acting as a transcription factor stimulates PLD1 expression.

AUTHOR CONTRIBUTION

Ju Hwan Cho and Mee-Sup Yoon, essentially through equal contribution, designed and performed the experiments, and analysed and interpreted the results. Jun Bon Koo assisted with the experiments. Yong Seok Kim and Ki-Sung Lee contributed reagents, materials and analytical tools. Joong-Soo Han and Jung Han Lee devised the overall project, interpreted the results and wrote the paper.

FUNDING

This work was supported by the Korea Healthcare Technology R&D Project, the Ministry for Health, Welfare and Family Affairs, Republic of Korea [grant number A084985], a National Research Foundation of Korea (NRF) grant funded by the Korean Government (Ministry of Education, Science and Technology; MEST) [grant number 2010-0029503], and partly by the research fund of Hanyang University [grant number HY-2006-C].

Abbreviations: AP, activator protein; ARF, ADP-ribosylation factor; β-gal, β-galactosidase; ChIP, chromatin immunoprecipitation; DMEM, Dulbecco's modified Eagle's medium, low glucose; DN-MEK, dominant-negative form of MEK; DN-Ras, dominant-negative Ras; ERK, extracellular-signal-regulated kinase; FBS, fetal bovine serum; GEF, guanine-nucleotide-exchange factor; GR, glucocorticoid receptor; GST, glutathione transferase; hES cell, human ES (endometrial stromal) cell; HRP, horseradish peroxidase; IGFBP-1, insulin-like growth factor binding protein-1; MAPK, mitogen-activated protein kinase; MEK, MAPK/ERK kinase; PA, phosphatidic acid; PBt, phosphatidic butanol; PC, phosphatidylcholine; PKA, protein kinase A; PKC, protein kinase C; PLD, phospholipase D; PR, progesterone receptor; PRL, prolactin; RBD, Raf/Ras-binding domain; RT, reverse transcription; siRNA, small interfering RNA; Sp1, stimulating protein-1; TTBS, Tris-buffered saline containing 0.01% Tween 20

References

View Abstract