Activated Protein Kinase A Is Required for Differentiation ... - CiteSeerX

0013-7227/97/$03.00/0 Endocrinology Copyright © 1997 by The Endocrine Society

Vol. 138, No. 3 Printed in U.S.A.

Activated Protein Kinase A Is Required for Differentiation-Dependent Transcription of the Decidual Prolactin Gene in Human Endometrial Stromal Cells* ´ N, RALPH TELGMANN, ERIK MARONDE, KJETIL TASKE

AND

BIRGIT GELLERSEN

Institute for Hormone and Fertility Research, University of Hamburg (R.T., B.G.), Hamburg; and the Institute of Anatomy, University of Frankfurt (E.M.), Germany; and the Institute of Medical Biochemistry, University of Oslo (K.T.), Oslo, Norway ABSTRACT Decidualization of human endometrial stromal (ES) cells in vitro is induced by cAMP analogs and ligands that elevate cellular cAMP levels. A marker of this differentiation process is the activation of the decidual PRL (dPRL) promoter. In a primary ES cell culture system we show that relaxin not only acutely but permanently elevates cellular cAMP levels and leads to induction of PRL secretion after 6 days. Northern and Western blot analyses revealed that all regulatory subunit isoforms (RIa, RIb, RIIa, and RIIb) and catalytic subunits Ca and Cb of protein kinase A (PKA) are expressed in ES cells. Transcript levels of PKA subunit isoforms are not altered during decidualization, but in decidualized ES cells, exposed to relaxin for more than 6 days, a significant reduction of RIa protein level occurs, whereas levels of all other forms remain unchanged. Reduction of R subunits might result in a net increase in free C subunit activity. This alteration is not due to a change in the mitotic state of the cells, as proliferating cell nuclear antigen is evenly expressed in undifferentiated and differentiated ES cell cultures. In transient transfections of undifferentiated ES cells, the dPRL promoter is activated by 8-bromo-cAMP

and the C subunit (Cb) of PKA. This induction as well as the differentiation-dependent activity of the dPRL promoter in transfected decidualized cells are effectively abolished by the coexpression of protein kinase inhibitor. We demonstrate that 332 bp of the dPRL promoter are sufficient to mediate full inducibility by cAMP. Activation of the dPRL promoter by cAMP in ES cells occurs in two steps: an initial weak induction within 12 h and a subsequent, much more pronounced induction after 12 h. The secondary induction is not seen with a control construct driven by a consensus cAMP response element (CRE) linked to a minimal promoter and is absent from a uterine cell line that does not express the endogenous dPRL gene. The early response of the dPRL promoter depends upon a noncanonical CRE at position 212, as mutation of this sequence leads to abolition of the early, but not the delayed, induction. The major activation depends upon a different region within 332 bp of the dPRL promoter; is probably indirect, as it follows different kinetics compared to a classical CRE-mediated response; and is specific to ES cells. (Endocrinology 138: 929 –937, 1997)

D

tion in estradiol-primed cultured ES cells. However, stable cAMP analogs alone trigger decidualization much more efficiently (4), whereas progesterone is a rather weak inducer in vitro. The differentiating effect of substances that elevate intracellular cAMP levels [PGE2, hCG, FSH, LH, and relaxin (RLX)] was synergistically enhanced by the presence of progesterone in cell culture experiments (5– 8). The major physiological source of RLX in vivo is the corpus luteum (9, 10) of the mid- to late luteal phase (11). RLX rapidly elevates cAMP levels in ES cells (12). However, its receptor and the subsequent signal transduction pathways are not yet described, but one is via cAMP-dependent protein kinase (PKA) in uterine tissues (12, 13). The level of intracellular cAMP in differentiated ES cells is higher than that in undifferentiated ES cells (14), and endometrial adenylate cyclase activity is altered in the course of the cycle and pregnancy (15). At the molecular level, we had previously shown that dPRL gene transcription is strongly induced by cAMP in transient transfection assays (3). The most prominent signal transduction pathway of cAMP is via PKA (16) and subsequent phosphorylation of nuclear proteins (17). When the pathway is stimulated, membrane-bound adenylate cyclase generates cAMP, which then binds to the R subunits of PKA. PKA consists of two R and two C subunits, the latter being released from the complex after each R subunit has bound two molecules of cAMP, generating a conformational change

ECIDUALIZATION of the human endometrium in the late luteal phase of the menstrual cycle is a profound morphological and biochemical alteration process preparing the endometrium for implantation and maintenance of pregnancy. During this differentiation, endometrial stromal (ES) cells acquire the ability to express and secrete numerous new cellular products (reviewed in Ref. 1). Among the major secretory products, insulin-like growth factor-binding protein-1 and PRL are considered to serve as markers for the onset of decidualization. PRL secretion starts around day 25 of the cycle and remains high throughout pregnancy. Decidual PRL gene transcription is directed by an alternative promoter located at an additional exon (exon 1a) 6 kilobases (kb) upstream of the pituitary promoter and is independent of Pit-1 (2, 3). The molecular basis of decidualization and initiation of decidual PRL (dPRL) gene transcription is not yet understood. Progesterone alone can induce decidualiza-

Received October 14, 1996. Address all correspondence and requests for reprints to: Dr. Birgit Gellersen, Institute for Hormone and Fertility Research, Division of Reproductive Sciences, University of Hamburg, Grandweg 64, 22529 Hamburg, Germany. E-mail: [email protected]. * Presented in part at the 10th International Congress of Endocrinology, San Francisco, CA, June 12–15, 1996. This work is based on the doctoral study by R.T. in the Faculty of Biology, University of Hamburg, and was supported by Deutsche Forschungsgemeinschaft (DFG) Grant Ge 748/1-2.

929

930

PKA AND HUMAN DECIDUAL PRL

in the R subunit dimer (16). The holoenzyme dissociates, and free single C subunits migrate into the nucleus and phosphorylate their target proteins, such as cAMP response element (CRE)-binding protein (CREB) and its modulator CREM (reviewed in Refs. 18 and 19). To date, four R subunit isoforms (RIa, RIb, RIIa, and RIIb) (20 –23) and three C subunit isoforms (Ca, Cb, and Cg) (24, 25) have been described. The isoform composition of PKA can change in response to physiological stimuli or mitotic state (26), and homo- or heterodimers of different R subunits complexed with C form isozymes with different kinetic properties (27, 28). Whereas type I PKA is localized in the cytosol, type II can be attached to subcellular structures such as endoplasmic reticulum or Golgi apparatus membranes by anchoring proteins (for a review, see Ref. 29). It is not known how cAMP mediates activation of the dPRL promoter and what region thereof is responsive. A CRE-like sequence (TGACGTTT) is present at position 212 relative to the transcription start site (3). The same sequence is found in the c-fos promoter and will transduce a cAMP signal via CREB (30, 31). In this study we investigated the role of PKA in decidual-specific PRL gene expression. Materials and Methods Antibodies Primary antibodies against human regulatory PKA subunits RIa (mouse monoclonal), RIb, RIIa, and RIIb (rabbit polyclonals) were generated and applied as previously described (27). Antibody against PKA catalytic subunits, recognizing Ca and Cb (rabbit polyclonal), was a gift from Dr. G. Schwoch (University of Go¨ttingen, Go¨ttingen, Germany) (32). Antiproliferating cell nuclear antigen (PCNA) antibody was purchased from Santa Cruz (Heidelberg, Germany). Secondary antibodies [sheep antimouse Ig peroxidase conjugated (Amersham, Braunschweig, Germany), (goat antirabbit IgG peroxidase conjugated (Sigma, Deisenhofen, Germany)] were used following the manufacturer’s instructions.

Cell culture The uterine sarcoma cell line SKUT-1B (HTB 115, American Type Culture Collection, Rockville, MD) was obtained from the European Collection of Animal Cell Cultures (Salisbury, UK) and maintained in DMEM-Ham’s F-12 (1:1; Life Technologies, Eggenstein, Germany), supplemented with 10% FCS (Serva, Heidelberg, Germany), 50 U/ml penicillin, and 50 mg/ml streptomycin. ES cells from cycling women undergoing hysterectomies for leiomyomas were prepared as previously described (3). Briefly, endometrial tissue was minced thoroughly and digested in DMEM with 0.5 mg/ml collagenase-dispase (Boehringer Mannheim, Mannheim, Germany) and 2.5 mg/ml deoxyribonuclease I (Sigma) for up to 2 h, with gentle pipetting every 20 –30 min. Myometrial contaminations and undispersed material were removed by sieving samples through a nylon stocking and subsequently through a steel sieve (38-mm pore size). ES cells were allowed to attach to six-well plates for 30 – 45 min before epithelial cells were washed away. Cells from tissues containing high amounts of erythrocytes were submitted to centrifugation through a 60% Percoll gradient (1080 3 g for 20 min) after sieving and before the differential attachment. The purity of the cultures was confirmed by immunocytochemistry using antibodies against cytokeratin, a-actin, and vimentin (Dako, Hamburg, Germany). Basal medium ((2/2) condition) contained DMEM-Ham’s F-12 at a 1:1 ratio, 10% FCS that had been depleted of steroids by treatment with dextran-coated charcoal, 100 U/ml penicillin, 100 mg/ml streptomycin, 1 mg/ml insulin (Sigma), and 1029 m 17b-estradiol (E2; Sigma). For treatment with progestin, 2.5 3 1027 m medroxyprogesterone acetate (MPA; Sigma) was added to the basal medium ((1/2) condition). Treatment with RLX ((1/1) condition) was performed by adding 100 ng/ml RLX (3000 U/mg protein; supplied by S. Raiti, National Hormone and Pituitary Program, NIDDK, Baltimore,

Endo • 1997 Vol 138 • No 3

MD) to the (1/2) medium. Medium was changed every 48 –72 h. At confluence, cells were trypsinized and passaged at a ratio of 1:3 into 12-well plates (Nunc, Roskilde, Denmark). Only first passage cells were used for experiments.

Secretion assays, RIA Two parallel sets of confluent cells in 12-well plates were washed twice with PBS and stimulated with E2, E2 plus MPA, E2 plus MPA plus RLX [corresponding to (2/2), (1/2), and (1/1) conditions as described above] in Opti-MEM (Life Technologies) for 30 min. Supernatant was collected for determination of extracellular cAMP content and replaced by (2/2), (1/2), or (1/1) medium for 72 h (see above). After 3 days, one of the two sets was harvested for measurement of intracellular cAMP as follows: cells were washed with PBS twice, incubated with 0.25 mm 3-isobutyl-1-methylxanthine (IBMX) (Sigma) in PBS for 15 min, taken up in 500 ml ice-cold 70% ethanol, and frozen at 220 C for at least 1 h. The remaining set of cells was incubated for another 72 h under the above-mentioned conditions. Media were collected for PRL RIA, and the cells were taken up in ethanol for cAMP measurement. RIA for cAMP from ethanol extractions was performed by evaporating the ethanol for 1 h at 60 C under vacuum. Samples were resuspended in 500 ml DMEM-Ham’s F-12 and then acetylated with 25 ml of a mixture of acetic anhydride and triethylamine (1:2.5; Sigma). Samples in Opti-MEM were acetylated directly. cAMP levels were measured using a commercial RIA (IBL, Hamburg, Germany). Measurement of human (h) PRL was performed by RIA as previously described (3).

Northern blot hybridization RNA from cultures of human ES cells was prepared following the RNA-clean protocol (AGS, Heidelberg, Germany). RNA was submitted to Northern blot hybridization as described previously (33). PKA subunit-specific complementary DNA (cDNA) probes were: 800 bp 59sequence of hRIa (20), 905 bp 59-sequence of hRIb (21), 280 bp 59sequence of hRIIa (22), 519 bp 59-sequence of hRIIb (23), 2.2 kb of hCa (PstI fragment from pUC18-Ca) (25), and 870 bp 59-sequence of hCb (PstI fragment from pBS-T124/Cb10) (25). Probes were labeled with [a-32P]deoxy-ATP following the Random Prime labeling protocol (AGS).

SDS-PAGE, Western blot, and immunodetection Cells were washed with PBS twice and then scraped in protein extraction buffer [Ripa buffer; PBS (pH 7.4), 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and protease inhibitor set (complete, Boehringer)]. Scraped cells were sonicated in an ice-cold water bath. Protein concentration was measured with the DC protein detection kit (Bio-Rad, Munich, Germany). For SDS-PAGE, a modification of a gel system described by Rittenhouse and Marcus (34) was chosen, using 15 mg protein/lane on a 12.5% SDS-polyacrylamide minigel. Western tank blot was performed in ice-cold transfer buffer (20% methanol, 25 mm ethanolamine, 33 mm glycine) for 2 h at 80 V (35) onto polyvinylidene difluoride membranes (Immobilon P, Millipore, Eschborn, Germany). Efficiency of the transfer was tested by staining with Ponceau S (Sigma) or Fount India Ink (Pelikan, Hannover, Germany) before blocking with Blotto (5% nonfat dry milk in PBS). Blots were exposed to primary antibodies overnight at 4 C. Secondary antibodies were applied for 1 h at room temperature, followed by detection with the ECL system (Amersham).

Generation of plasmids Deletion constructs of dPRL-3000/GH (3) were generated by cloning the BamHI fragment of pGEM-dPRL-3000 (position 22999 to 165 of the dPRL promoter) into pBSKSII (Stratagene, Heidelberg, Germany). Nested 59-exonuclease III deletions (ExoIII/Mung Bean Deletion kit, Stratagene) were started at an XbaI/BstXI nick in the pBSKSII polylinker site and allowed to proceed for 1–15 min at 30 C. After religation, i-SacI/BamHI fragments were isolated, inserted into HincII/BamHI-digested vector p0GH (Nichols Institute, Bad Nauheim, Germany), and sequenced. Plasmid dPRL-332/GH was generated by ligating the 2332/ 165 SspI/BamHI fragment of dPRL-3000/GH into HincII/BamHI-digested vector p0GH. The vector pCRE-TK/GH carries a consensus CRE

PKA AND HUMAN DECIDUAL PRL in front of the thymidine kinase (TK) promoter fragment 281/152 (36) and was constructed in two steps: two oligonucleotides (59-CAAATTGACGTCATGGTAAGAGCT-39 and 59-CTTACCATGACGTCAATTTGAGCT-39) were annealed to give rise to a fragment containing the CRE sequence from the human a-subunit gene (italicized) (37) flanked by SstI overhangs (boldface). This double stranded oligonucleotide was phosphorylated and ligated into SstI-digested vector pT81luc (36) to create pCRE-TK/luc. The CRE sequence linked to the TK promoter was excised from this vector with HindIII/BglII, and the fragment was ligated into p0GH digested with HindIII/BamHI, giving rise to pCRE-TK/GH. Plasmid CRE/236rPRL/luc3 was generated based upon pCRE/236rPRL/ luc (a gift from Dr. M. G. Rosenfeld, Howard Hughes Medical Institute, San Diego, CA) (38), which contains two composite CREs (italicized) (59-TTGGCTGACGTCAGAGAGAGGCCGGCCCCTTACGTCAGAGGCGAG-39) linked to the minimal rat (r) PRL promoter fragment 236/ 134 and the luciferase gene. This sequence was amplified by PCR using Pfu-polymerase (Stratagene, Heidelberg, Germany), upstream primer (59-CTTGGCTGACGTCAGAGAGAG-39), and antisense primer GL2 (Promega, Madison, WI) located in the luciferase gene. The PCR product was cut at the HindIII site located at position 134 of the rPRL promoter, and the resulting blunt end/HindIII fragment was inserted into vector pGL3-Basic (Promega), digested with SmaI/HindIII, to generate pCRE/236rPRL/luc3. Plasmids dPRL-332wt/luc3 (wt, wild type), dPRL-332MUT/luc3, and dPRL-332CRE/luc3 were constructed by amplifying portion 2332/ 165 of the dPRL promoter carrying either the native imperfect CRE (wt) sequence 59-TGACGTTT-39, mutated sequence 59-TTATGATT-39 (MUT), or perfect palindrome 59-TGACGTCA-39 (CRE) at position 212 via PCR. The PCR strategy was as follows. The upstream primer covers the dPRL region from 2332 to 2315, introducing a BamHI site (boldface; 59-AGGATCCATTATGTTCTGAGGGCTG-39) at the 59-end of the PCR product. The downstream primer for creation of dPRL-332wt/luc3 spans dPRL 154 to 180 relative to the transcription start site, inserting a BamHI site at position 165 by exchange of two nucleotides. The PCR product of these two primers using the genomic hPRL clone 8b-1 (3) as template was purified and briefly digested with BamHI. This resulted in blunt-end/BamHI-digested PCR product, which was ligated into i-SacI/ BglII-digested vector pGL3-Basic. Plasmid dPRL-332MUT/luc3 was generated by PCR, using dPRL-332wt/luc3 as template, the abovementioned upstream primer, and a downstream primer (59-TGC CAA GCT TAC TTA GAT CGC GGA TCC GAT TCT TCT TGG TGT CTC TGT CTT TGA GGG TAC TTC TGG AAT GAA GGT TCT TAT GAC CTA CTT TAT AGA ATC ATA AGA GGA TT-39), including a HindIII site (italicized; corresponding to the HindIII site in the multiple cloning site of pGL3-Basic), a BamHI site (italicized; corresponding to the BamHI site introduced by the downstream primer in dPRL-332wt/luc3), and a mutated sequence in the CRE site (boldface). The PCR product was digested with HindIII giving rise to a fragment spanning from the HindIII site at position 256 of the dPRL promoter to the HindIII site in pGL3-Basic. This fragment was used to replace the respective HindIII fragment from dPRL-332wt/luc3, giving rise to dPRL-332MUT/luc3. Plasmid dPRL-332CRE/luc3 was created following the same strategy, but with a downstream primer introducing a perfect palindromic CRE

FIG. 1. Inducibility of deletion constructs of the dPRL promoter by 8-BrcAMP. Deletion constructs of dPRL3000/GH linked to hGH as a reporter gene were transfected into undifferentiated ES cells that were kept in the absence or presence of 0.5 mM 8-BrcAMP for 72 h. Reporter gene activity in the presence of 8-Br-cAMP is depicted as the fold induction over unstimulated controls. One representative experiment of five similar experiments is shown.

931

sequence (TGAC GTCA). Plasmids pRSV-Cb, pRSV-Cbmut (39) and pRSV-PKI, pRSV-PKImut (40) were gifts from Dr. R. Maurer (University of Oregon, Portland, OR).

Transient transfection and reporter gene assays For transfections, ES cell cultures were used at 75– 85% confluency. Cultures were transfected in triplicate in 12-well plates by calcium phosphate coprecipitation (Promega) for 16 h without subsequent glycerol shock. When the hGH gene was used as a reporter (41), medium was harvested 72 h after removal of the transfection mix, and hGH secretion was measured using a chemiluminescent assay (Nichols Institute) as previously described (3). Cells received equimolar amounts of reporter gene constructs (maximum, 2.5 mg DNA/well); inert vector pGL2-Basic (Promega) was added to expose cells to equal amounts of transfected DNA. For luciferase reporter assays, cells were harvested 24 h after removal of the transfection mix, and reporter gene activity was measured with the luciferase assay system (Promega). Transfection of SKUT-1B cells was performed as described previously (3).

Results 332 bp of the dPRL promoter are sufficient to mediate full inducibility by cAMP

To locate the region of the dPRL 59-flanking region that is responsible for mediating inducibility by cAMP, we transfected ES cells with a series of constructs made by 59-deletions of the upstream region, linked to hGH as reporter gene, and kept the cells in the absence or presence of 0.5 mm 8-bromo-cAMP (8-Br-cAMP) for 72 h (Fig. 1). As a negative control, the promoterless construct p0GH was used. A 3000-bp portion of the promoter was induced 16.4-fold by the cAMP analog. The smallest portion that was able to mediate full inducibility was 332 bp long (18.4-fold induction). Further truncation of the promoter reduced inducibility noticeably (dPRL-270/GH, 6.5-fold; dPRL-102/GH, 9.8-fold; dPRL-56/GH, 2.9-fold induction). cAMP treatment caused a general stimulation of reporter gene expression, as even the promoterless construct p0GH was induced 3-fold. Activity of the dPRL promoter in differentiated ES cells is PKA dependent

PKI is a pseudosubstrate of PKA, and a PKI nuclear export signal serves to actively transport PKI:C complexes out of the nucleus (42). To assess whether coexpression of PKI would inhibit cAMP-induced activation of the dPRL promoter, we transfected undifferentiated, PRL-negative ES cells with

932


dPRL-promoter constructs dPRL-3000/GH and dPRL-332/ GH. Cotransfections were performed with expression vectors for PKA-Cb (pRSV-Cb), an inactive PKA-Cb mutant (pRSV-Cbmut), PKIa (pRSV-PKI), or an inactive PKI mutant (pRSV-PKImut) (39, 40) (Fig. 2). In the absence of PKA-Cb, all promoter constructs remained silent. Coexpression of PKA-Cb induced dPRL constructs dPRL-3000/GH (19.8-fold induction) and dPRL-332/GH (16.7-fold induction) even more strongly than a control construct (pCRE-TK/GH, 5.6fold induction). Cotransfection of an expression vector for PKI effectively suppressed PKA-Cb-induced activation of the dPRL promoter. In contrast, mutant Cb or mutant PKI had no effect. We have shown previously that differentiation enables ES cells to use the transfected dPRL promoter (3). We next wanted to assess whether the capability of decidualized ES cells to use the promoter is dependent upon activated PKA. In a parallel set of experiments, we transfected both in vitro decidualized ES cells (1/1 cells) and undifferentiated ES cells (2/2 cells) with the above-mentioned dPRL promoter constructs (Fig. 3). Undifferentiated cells were stimulated with 0.5 mm 8-Br-cAMP for 72 h. Both sets of cultures were cotransfected with either expression vector pRSV-PKI or pRSV-PKImut. In undifferentiated cells, cAMP-induced activation of the dPRL promoter constructs was effectively abolished by the presence of PKI (dPRL-3000/GH, 90% reduction; dPRL-332/GH, 87% reduction; Fig. 3B). Similarly, differentiation-dependent activity of dPRL-3000/GH and dPRL-332/GH in decidualized ES cells in the absence of exogenously added cAMP analog was significantly suppressed by PKI (73% and 70% reduction, respectively; Fig.

FIG. 2. Activation of the dPRL promoter in undifferentiated ES cells by PKA. Undifferentiated ES cells were transfected with dPRL promoter constructs dPRL-3000/GH or dPRL-332/GH, promoterless vector p0GH, or control vector pCRE-TK/GH. Cotransfections were performed with expression vectors for PKA-Cb (pRSV-Cb) and PKI (pRSV-PKI) or inactive mutants thereof (pRSV-Cb/mut and pRSVPKI/mut). Reporter gene activity was measured 72 h posttransfection (mean 6 SD; n 5 3).

Endo • 1997 Vol 138 • No 3

3A). This indicates that differentiation-dependent dPRL gene transcription depends at least partially upon active PKA. Relaxin causes a persistent elevation of intracellular cAMP

RLX is known to induce the decidualization process (7), and it acutely raises intracellular cAMP levels in cultured ES cells (12). As decidualized ES cells depend upon active PKA for dPRL expression, we investigated the time course of RLX-induced cAMP generation over a period of 6 days (Fig. 4). We exposed undifferentiated ES cells to control medium (2/2), MPA alone (1/2), or MPA plus RLX (1/1) and measured cAMP formation after 30 min, 3 days, and 6 days. In the same cultures, PRL secretion was measured after 6 days of treatment. RLX not only caused a drastic rapid increase in cAMP production after 30 min, but also resulted in a permanently elevated cAMP level for the entire period of the experiment (2.8-fold increase after 3 days, 3.1-fold increase after 6 days compared to (2/2) conditions). After 6 days, cells treated with RLX and MPA produced significantly more PRL than cells treated with MPA alone. MPA alone had no measurable effect on cAMP formation, and PRL was barely detectable in MPA-stimulated cultures after 6 days. No differences in cGMP formation were observed among the different treatments (data not shown). Transcripts for all PKA subunit isoforms are present in undifferentiated and differentiated ES cells

The expression of PKA and its isoform composition might be altered in response to different cellular cAMP levels, so as a next step we investigated the distribution of the PKA subunit isoforms in undifferentiated and differentiated ES cell cultures (Fig. 5). Total RNA from undifferentiated ES cells kept under (2/2) conditions and from differentiated ES cells maintained under (1/1) conditions was subjected to Northern blot hybridization using subunit-specific cDNA probes. Transcripts of all R subunits (RIa, RIb, RIIa, and RIIb) and C subunits (Ca and Cb) were present in human ES cells. For RIa, two transcripts could be detected, 3.0 and 1.5 kb in size. The larger and predominant form represents the RIa transcript, which is ubiquitous in somatic cells; the smaller transcript has been detected in human testis (20). RIb messenger RNA (mRNA) had a size of 2.7 kb. For RIIa, a single transcript of 7.0 kb was detectable, but no smaller forms, like those described for germ cells, were present (22, 43). The probe against RIIb mRNA detected a transcript of 3.3 kb. The mRNA level of Ca (2.8 kb) far exceeded that of Cb (4.4 kb), but Cb mRNA was detectable. However, the respective levels of any of the isoforms were not altered over the course of the differentiation process. Regulatory subunit RIa protein is decreased in differentiated ES cells

To assess whether ES cells counteract permanently elevated cAMP by down-regulating C subunits or up-regulating R subunits at the protein level, we subjected protein extracts from ES cultures kept under (2/2), (1/2), and (1/1) conditions to Western blot analysis (Fig. 6). For the C subunits, a single band of approximately 40 kDa was de-


933

FIG. 3. Comparison of the effect of PKI on differentiation-dependent and cAMP-stimulated dPRL promoter usage. A, In vitro decidualized ES cells were transfected with the same reporter gene constructs as those in Fig. 2 in the absence or presence of active PKI. B, Undifferentiated ES cells from the same individual as in A, kept in basal medium plus 0.5 mM 8-Br-cAMP, were subjected to the same transfection regime as that in A. Reporter gene activity was measured after 72 h (mean 6 SD; n 5 3).

FIG. 4. Induction of cAMP generation and PRL secretion in response to MPA and RLX. Undifferentiated ES cells were exposed to basal medium (2/2), MPA (1/2), or MPA plus RLX (1/1) as outlined in Materials and Methods. A, cAMP concentration was measured in the supernatant after 30 min. B and C, Cells were harvested for measurement of intracellular cAMP after 3 or 6 days, respectively. D, PRL secretion was determined after 6 days as the amount of hormone accumulated in the supernatant during the last 3 days of the experiment (n.d., below detection limit; mean 6 SD; n 5 3, representative of three similar experiments).

FIG. 5. Northern blot analysis of PKA subunit transcripts in cultured ES cells. ES cells were maintained in basal medium (2/2 conditions) or in the presence of MPA plus RLX (1/1 conditions) for 15 days. Northern blots containing 30 mg total RNA were hybridized with subunit-specific cDNA probes as described in Materials and Methods. Hybridization for RIIa and RIIb was performed simultaneously, as previous experiments had shown the transcripts to be sufficiently different in size. Even loading was confirmed by ethidium bromide staining of the gel.

tected. The antibody used does not discriminate between Ca and Cb. The difference in Mr of Ca and Cb is only 122 Da, so both isoforms migrate in a single complex during normal SDS-PAGE (44). It is, therefore, not possible to judge the level of the individual subunit isoforms, but the total amount of

C subunit protein is not altered over the course of decidualization. Proteins for all R subunits were present, no differences were detectable between undifferentiated cells (2/2) and cells treated with MPA alone (1/2). However, decidualized, PRL-secreting ES cells kept in the presence of

934

Endo • 1997 Vol 138 • No 3


FIG. 6. Western blot analysis of PKA subunit isoforms and PCNA in cultured ES cells. ES cells were maintained under basal conditions (2/2), in the presence of MPA (1/2), or with MPA and RLX (1/1) for 12 days. Cultures secreted undetectable amounts or 6 or 70 ng/ml PRL/3 days, respectively, at the time of harvest. Western blots were subjected to immunodecoration using antibodies against the indicated PKA subunit isoforms (upper five panels) and an antibody against PCNA (bottom panel).

MPA plus RLX (1/1) displayed a markedly reduced level of R subunit RIa, whereas the other isoforms were not affected. The PKA subunit composition can be dependent upon the mitotic state of the cell. We, therefore, submitted Western blots of ES cells to immunodetection using an antibody to PCNA, which is an auxiliary protein of DNA polymerase d (45) and serves as a marker for proliferation in most cells. PCNA levels were not altered by any of the treatments, suggesting that decidualization does not arrest the cell cycle of ES cells and that the reduced level of RIa protein is, therefore, not cell cycle dependent. Activation of the dPRL promoter by cAMP is cell-specific and delayed in human ES cells

Having shown that RLX up-regulated cAMP during differentiation and that differentiating ES cells maintained high levels of PKA, we wanted to examine in more detail the effect of cAMP on the dPRL promoter during differentiation. We linked the dPRL promoter fragment dPRL-332 to luciferase as a reporter gene to be able to detect early cAMP effects in transient transfection assays. The luciferase reporter gene system, in contrast to the hGH reporter gene system, allowed detection of promoter utilization in ES cells within less than 24 h. To investigate the contribution of the imperfect CRE at position 212 to the cAMP inducibility of the dPRL promoter, we altered the nonpalindromic sequence in the context of the 332-bp fragment to either a perfect palindrome (dPRL332CRE/luc3) or a mutation that abolishes CRE activity (dPRL-332MUT/luc3) and compared it with the wild type (dPRL-332wt/luc3) in a stimulation transfection assay. Undifferentiated ES cells and the uterine sarcoma cell line SKUT-1B, which does not express endogenous dPRL, were transfected, and cells were lysed 24 h after transfection.

Transfected cells were treated with 0.5 mm 8-Br-cAMP for the last 6, 12, or 24 h. The control plasmid in this experiment (pCRE/236rPRL/luc3) contained two CREs linked to a minimal PRL promoter fragment (Fig. 7). In ES cells, as in SKUT-1B cells, activation of the control plasmid peaked within 12 h of stimulation. The basal unstimulated activity of the control plasmid (pCRE/236rPRL/ luc3) was higher than that of dPRL-332wt/luc3 in both ES cells and SKUT-1B cells (1.6- and 4.8-fold, respectively; not shown), which might explain the relatively weak additional induction by cAMP (2.8- and 2.9-fold, respectively). In SKUT-1B cells none of the dPRL promoter constructs was activated by cAMP. In contrast, in ES cells an initial weak activation of dPRL-332wt/luc3 and dPRL-332CRE/luc3 within the first 12 h of stimulation was followed by a delayed, but much stronger, induction, with no differences between the wild-type and the perfect palindromic sequence. Mutation of the CRE fully abolished the early, but not the delayed, response. We, therefore, suggest that cAMP activation of the dPRL promoter is cell specific, indirect, and predominantly mediated by sequences outside of the nonpalindromic CRE. Discussion

Decidualization describes the complex morphological and biochemical differentiation process that ES cells undergo toward the end of the menstrual cycle in preparation for blastocyst implantation. Even though a variety of factors have been identified that trigger decidualization in vitro, such as gonadotropins, RLX, PGE2, CRF, cAMP analogs, progesterone, and androgens (5– 8, 46 –50), little is known about the molecular basis of the induction of decidual-specific genes. We have previously shown that progesterone, which is produced by the corpus luteum during the secretory phase of the cycle, does not regulate dPRL gene transcription. cAMP, however, effectively induces the dPRL promoter within 72 h (3). cAMP signaling to the nucleus is transduced almost exclusively by binding of cAMP to R subunits of PKA and subsequent phosphorylation of nuclear target proteins by released and translocated C subunits (17). To investigate the role of PKA in induction of the dPRL promoter, we introduced an expression vector for PKA-Cb into undifferentiated ES cells. This vector strongly induced reporter constructs carrying at least 332 bp of the dPRL promoter. In vitro decidualized ES cells have acquired the ability to use transfected dPRL promoter constructs concomitantly to their capability to express the endogenous PRL gene (3). When we introduced PKIa into decidualized ES cells, the activity of the dPRL promoter was repressed. As PKIa exclusively inhibits PKA, but not cGMP-dependent protein kinase (40), these data strongly suggest that decidualized ES cells require active PKA for ongoing expression of the dPRL gene. Maintenance of the activated PKA during ES cell differentiation to drive dPRL expression requires permanent presence of sufficient levels of cellular cAMP. RLX has been shown to rapidly elevate cAMP levels in ES cells (12) and to induce PRL secretion, synergistically with MPA (8). We investigated the time course of cellular cAMP generation by ES cells in response to MPA and RLX. Treatment of undiffer-


935

FIG. 7. Kinetics of 8-Br-cAMP-mediated activation of wild-type and mutated dPRL-332 constructs. A, The putative CRE, located at position 212 of the dPRL promoter, was rendered into a perfect palindrome (dPRL-332CRE/ luc3; open square) or into an inactive mutant (dPRL-332MUT/luc3; open circle). Wild-type (dPRL-332wt/luc3; open triangle) and mutated fragments were linked to the luciferase reporter gene. The positive control vector for cAMP induction, pCRE/236rPRL/luc3 (filled square) contains two CREs linked to a minimal promoter as described in Materials and Methods. B, The above constructs were transfected into undifferentiated ES cells. Cells were harvested for reporter gene assay 24 h post transfection. Triplicate dishes had been exposed to 0.5 mM 8-Br-cAMP for the last 6, 12, or 24 h; controls had received no cAMP. C, The uterus sarcoma cell line SKUT-1B was subjected to the same transfection/stimulation regime as ES cells. Reporter gene activation by 8-BrcAMP is expressed as fold induction compared to unstimulated control values.

entiated ES cells with MPA plus RLX led to a drastic acute increase in cAMP generation after 30 min, as well as a permanently elevated intracellular level of the second messenger after 3 and 6 days, compared to control cells or cells treated with MPA alone. Concomitantly, the presence of RLX effected a significant induction of PRL production, whereas PRL was barely detectable in cells exposed to MPA alone after 6 days. Treatment of ES cells with vasoactive intestinal peptide, which only results in a brief spurt of cAMP formation, does not lead to induction of PRL secretion (our unpublished observation). Thus, a persistently elevated cellular cAMP content seems to be critical for efficient dPRL gene expression. This is supported by the finding that dPRL production induced by PGE2 plus progesterone in ES cells is suppressed by all-trans retinoic acid or interleukin-1b; this effect is accompanied by reduced levels of cellular cAMP (51, 52). During in vitro decidualization, ES cells are exposed to a cAMP stimulus for days to weeks. Numerous studies in other tissues and cell lines have dealt with the response of the cellular PKA system to permanently elevated cAMP, mostly achieved by treatment of the tested cells with forskolin (reviewed in Refs. 16, 18, and 19). It has been described in various systems that the permanent stimulus is counteracted by either decreasing the amount of C subunits or by elevating the level of R subunits of PKA (53). In isoprenaline-stimulated rat parotid gland cells and in rat pituitary GH3 cells, elevation of cAMP leads to degradation of C subunits, whereas R subunit levels remain unchanged (54, 55). In rat hepatocytes, a decrease of C subunits is accompanied by an increase in R subunit transcripts upon stimulation by cAMP (56). In all those cases, the net effect is an increase in the R/C ratio. In B lymphoid cells, dissociation of the holoenzyme

leads to destabilization and degradation of both C and RIa (57). By Northern blot analysis we found that transcripts for all R and C subunits are present in ES cells, but no difference in transcript levels could be detected between undifferentiated ES cells and decidualized cells that had been exposed to RLX for 2 weeks. Thus, at the mRNA level, a permanently elevated cAMP content does not alter PKA expression in fully differentiated ES cells. Western blot analysis, however, revealed a significant reduction of RIa protein in RLX-treated cells. Whether this reduction of RIa is due to a more rapid degradation of free RIa dissociated by the increased cAMP levels or due to a translational down-regulation cannot be answered at present. No difference in the C subunit protein levels could be detected, and the absence of a coordinated degradation of RIa and C may be due to compartmentalization of C or redistribution of R:C complexes. As a consequence of reduced RIa protein, the R/C ratio is decreased, and signal transduction to the nucleus is maintained. The endometrial PKA system, therefore, does not desensitize to the sustained cAMP stimulus, but, rather, enhances the cAMP effect by reducing regulatory subunit isoform RIa. RIa is the dominant R subunit in proliferating tissues (26), so a loss of RIa protein could be due to a halt of proliferation after differentiation of ES cells. However, the mitotic marker PCNA, appearing in early G1 and S phase of the cell cycle (45, 58), was evenly expressed under control and differentiating culture conditions, suggesting that decidualization does not arrest the cell cycle of human ES cells. The dPRL promoter contains a putative CRE at position 212. Our studies show that this element is probably not the main mediator of the cAMP signal in ES cells. The classical response to cAMP stimulation is elicited within hours, as

936

Endo • 1997 Vol 138 • No 3


CREB is phosphorylated approximately 30 min after stimulation of the pathway and subsequently activates transcription at responsive gene promoters (59). In transfection assays with ES cells as well as SKUT-1B cells, control plasmid CRE/ 236rPRL/luc3 was fully activated within 12 h of stimulation by cAMP and began to decline thereafter. In ES cells the initial response of the wild-type dPRL-332 promoter was mediated by the putative CRE, as it was suppressed when the sequence was mutated (dPRL-332MUT), but the response was comparably weak and could not be enhanced by rendering the sequence into a palindromic CRE as in dPRL332CRE. The much stronger secondary induction occurred in a delayed manner after 12 h, following the weak initial induction. In contrast to the latter, the pronounced subsequent induction was not suppressed by destruction of the CRE sequence in dPRL-332MUT. This effect was cell specific, as SKUT-1B cells, which do not use the endogenous dPRL promoter, failed to activate any of the dPRL promoter constructs during the 24 h of the experiment. The putative CRE sequence in the dPRL promoter (TGACGTTT) has been shown to mediate cAMP activation in the c-fos protooncogene promoter (30, 31); however, the promoter context appears to be important for a potential response element to become functional. Even the presence of a palindromic CRE sequence does not imply that it is unavoidably activated by the PKA pathway. The perfect CRE palindrome in the synapsin I promoter was shown not to be regulated by cAMP (60). In the surfactant protein A promoter, rendering of a putative CRE (CREsp-a; TGACCTCA) into a perfect palindrome led to a significant reduction of activation by cAMP (61). Our results suggest that the strong secondary activation requires ES cell-specific factors that are induced in the course of prolonged elevation of cAMP, which, in concert with a factor(s) interacting with the CRE, direct efficient dPRL gene expression. In deletion studies we show that 332 bp of the promoter are sufficient to mediate inducibility of the dPRL promoter by cAMP within 72 h. These studies suggest that region 2270/ 2332 is of major importance for mediating this response in ES cells. The dPRL promoter is activated not only in ES cells, but also in lymphoblastoid cells (2). In the T cell line Jurkat, Berwaer et al. (62) identified a lymphoid-specific element at position 2212/2375 by footprint analysis. It is not known at present whether similar factors from decidualized and lymphoid cells bind to this region. We earlier reported, however, that dPRL promoter induction in lymphoid cells differs from that in ES cells, as in the latter utilization of the transfected dPRL promoter and activation of the endogenous dPRL gene are strictly correlated. This correlation is absent in lymphoid cells (3). To our knowledge the recent description of a new Sp3 form that suppresses transcriptional repression of the insulin-like growth factor-binding protein-1 promoter in the decidua is the first report of an endometrial-specific factor involved in activation of a decidual gene (63). In the dPRL promoter, however, no Sp1/Sp3 binding motif is present (3). Taken together, our results suggest that dPRL gene transcription in ES cells is dependent upon permanently activated PKA, as achieved by permanently elevated levels of cellular cAMP, which leads to an elevated kinase activity and a decreased R/C subunit ratio. The endometrial PKA system

does not desensitize to the chronic stimulus. This is supported by our finding that in in vitro decidualized ES cells, transcripts for the nuclear factor ICER (inducible cAMP early repressor) are constantly up-regulated (64), whereas ICER is normally rapidly down-regulated via an autofeedback mechanism (65). The major transcriptional response of the dPRL promoter to cAMP is not immediate, but delayed, and is mediated by a region different from the CRE at position 212, probably located between 2270/2332 of the promoter. Therefore, we suggest the presence of an intermediate, ES cell-specific factor(s), whose activity is linked to the PKA pathway. Whether this factor is activated or de novo synthesized in response to cAMP remains to be elucidated. Acknowledgments The authors are indebted to Dr. T. Lo¨ning, Department of Gynecological Histopathology, University Hospital Hamburg, for providing endometrial tissue; Rita Kempf for excellent technical support; Dr. R. Maurer, University of Oregon (Portland, OR), and Dr. M. G. Rosenfeld, Howard Hughes Medical Institute (San Diego, CA), for the supply of plasmids; and Dr. G. DiMattia, London Regional Cancer Center (London, Ontario, Canada), for helpful discussions and critical revision of the manuscript.

References 1. Tang B, Guller S, Gurpide E 1994 Mechanism of human endometrial stromal cells decidualization. Ann NY Acad Sci 734:19 –25 2. DiMattia GE, Gellersen B, Duckworth ML, Friesen HG 1990 Human prolactin gene expression. The use of an alternative noncoding exon in decidua and the IM-9-P3 lymphoblast cell line. J Biol Chem 265:16412–16421 3. Gellersen B, Kempf R, Telgmann R, DiMattia GE 1994 Nonpituitary human prolactin gene transcription is independent of Pit-1 and differentially controlled in lymphocytes and in endometrial stroma. Mol Endocrinol 8:356 –373 4. Tang B, Guller S, Gurpide E 1993 Cyclic adenosine 39,59-monophosphate induces prolactin expression in stromal cells isolated from human proliferative endometrium. Endocrinology 133:2197–2203 5. Tang B, Gurpide E 1993 Direct effect of gonadotropins on decidualization of human endometrial stromal cells. J Steroid Biochem Mol Biol 47:115–121 6. Frank GR, Brar AK, Cedars MI, Handwerger S 1994 Prostaglandin E2 enhances human endometrial stromal cell differentiation. Endocrinology 134:258 –263 7. Huang JR, Tseng L, Bischof P, Ja¨nne OA 1987 Regulation of prolactin production by progestin, estrogen, and relaxin in human endometrial stromal cells. Endocrinology 121:2011–2017 8. Zhu HH, Huang JR, Mazella J, Rosenberg M, Tseng L 1990 Differential effects of progestin and relaxin on the synthesis and secretion of immunoreactive prolactin in long term culture of human endometrial stromal cells. J Clin Endocrinol Metab 71:889 – 899 9. Stewart DR, Celniker AC, Taylor Jr CA, Cragun JR, Overstreet JW, Lasley BL 1990 Relaxin in the peri-implantation period. J Clin Endocrinol Metab 70:1771–1773 10. Hansell DJ, Bryant-Greenwood GD, Greenwood FC 1991 Expression of the human relaxin H1 gene in the decidua, trophoblast, and prostate. J Clin Endocrinol Metab 72:899 –904 11. Ivell R, Hunt N, Khan-Dawood F, Dawood MY 1989 Expression of the human relaxin gene in the corpus luteum of the menstrual cycle and in the prostate. Mol Cell Endocrinol 66:251–255 12. Fei DTW, Gross MC, Lofgren JL, Mora-Worms M, Chen AB 1990 Cyclic AMP response to recombinant human relaxin by cultured human endometrial cells—a specific and high throughput in vitro. bioassay. Biochem Biophys Res Commun 170:214 –222 13. Meera P, Anwer K, Monga M, Oberti C, Stefani E, Toro L, Sanborn BM 1995 Relaxin stimulates myometrial calcium-activated potassium channel activity via protein kinase A. Am J Physiol 269:C312–C317 14. Bergamini CM, Pansini F, Bettochi Jr S, Segala V, Dallochio F, Bagni B, Mollica G 1984 Hormonal sensitivity of adenylate cyclase from human endometrium: modulation by estradiol. J Steroid Biochem Mol Biol 22:299 –303 15. Tanaka N, Miyazaki K, Tashiro H, Mitzutani H, Okamura H 1993 Changes in adenylyl cyclase activity in human endometrium during the menstrual cycle and in human decidua during pregnancy. J Reprod Fertil 98:33–39 16. Døskeland SO, Maronde E, Gjertsen BT 1993 The genetic subtypes of cAMPdependent protein kinase–functionally different or redundant? Biochim Biophys Acta 1178:249 –258 17. Hagiwara M, Brindle P, Harootunian A, Armstrong R, Rivier J, Vale W, Tsien


18. 19. 20.

21.

22. 23.

24. 25.

26.

27.

28.

29. 30. 31.

32. 33. 34. 35. 36. 37.

38. 39. 40. 41.

R, Montminy MR 1993 Coupling of hormonal stimulation and transcription via the cyclic AMP-responsive factor CREB is rate limited by nuclear entry of protein kinase A. Mol Cell Biol 13:4852– 4859 Spaulding SW 1993 The ways in which hormones change cyclic adenosine 39,59-monophosphate-dependent protein kinase subunits, and how such changes affect cell behaviour. Endocr Rev 14:632– 650 Meyer TE, Habener JF 1993 Cyclic adenosine 39,59-monophosphate response element binding protein (CREB) and related transcription-activating deoxyribonucleic acid-binding proteins. Endocr Rev 14:269 –290 Sandberg M, Taskeń K, Øyen O, Hansson V, Jahnsen T 1987 Molecular cloning, cDNA structure and deduced amino acid sequence for type I regulatory subunit of cAMP-dependent protein kinase from human testis. Biochem Biophys Res Commun 149:939 –945 Solberg R, Taskeń K, Keiserud A, Jahnsen T 1991 Molecular cloning, cDNA structure and tissue-specific expression of the human regulatory subunit RIb of cAMP-dependent protein kinases. Biochem Biophys Res Commun 176:166 –172 Øyen O, Myklebust F, Scott JD, Hansson V, Jahnsen T 1989 Human testis cDNA for the regulatory subunit RIIa of cAMP-dependent protein kinase encodes an alternate amino-terminal region. FEBS Lett 246:57– 64 Levy FO, Øyen O, Sandberg M, Taskeń K, Eskild W, Hansson V, Jahnsen T 1988 Molecular cloning, complementary deoxyribonucleic acid structure and predicted full-length amino acid sequence of the hormone-inducible regulatory subunit of 39,59-cyclic adenosine monophosphate-dependent protein kinase from human testis. Mol Endocrinol 2:1364 –1373 Maldonado F, Hanks SK 1988 A cDNA clone encoding human cAMP-dependent protein kinase catalytic subunit Ca. Nucleic Acids Res 16:8189 – 8190 Beebe SJ, Øyen O, Sandberg M, Frøysa A, Hansson V, Jahnsen T 1990 Molecular cloning of a tissue specific protein kinase (Cg) from human testis– representing a third isoform for the catalytic subunit of cAMP-dependent protein kinase. Mol Endocrinol 4:465– 475 Tortora G, Pepe S, Bianco C, Baldassarre G, Budillon A, Clair T, Cho-Chung YS, Bianco AR, Ciardiello F 1994 The RIa subunit of protein kinase A controls serum dependency and entry into cell cycle of human mammary epithelial cells. Oncogene 9:3233–3240 Taskeń K, Skålhegg BS, Solberg R, Andersen KB, Taylor SS, Lea T, Blomhoff HK, Jahnsen T, Hansson V 1993 Novel isoforms of cAMP-dependent protein kinase exist in human cells due to formation of RIa-RIb heterodimeric complexes. J Biol Chem 268:21276 –21283 Otten AD, Parenteau LA, Døskeland SO, McKnight GS 1991 Hormonal activation of gene transcription in ras-transformed NIH3T3 cells overexpressing RIIa and RIIb subunits of the cAMP dependent protein kinase. J Biol Chem 266:23074 –23082 Scott JD, McCartney S 1994 Localization of A-kinase through anchoring proteins. Mol Endocrinol 8:5–11 Sassone-Corsi P, Visvader J, Ferland L, Mellon P, Verma IM 1988 Induction of proto-oncogen fos transcription through the adenylate cyclase pathway: characterization of a cAMP-responsive element. Genes Dev 2:1529 –1538 Boutillier A-L, Barthel F, Roberts JL, Loeffler J-P 1992 b-Adrenergic stimulation of cFOS via protein kinase A is mediated by cAMP regulatory element binding protein (CREB) and tissue-specific CREB-independent mechanisms in corticotrope cells. J Biol Chem 267:23520 –23526 Schwoch G, Hamann A, Hilz H 1980 Antiserum against the catalytic subunits of cyclic adenosine 39,59-monophosphate dependent protein kinases. Biochem J 192:223–230 Gellersen B, Kempf R, Hartung S, Bonhoff A, DiMattia GE 1992 Posttranscriptional regulation of the human prolactin gene in IM-9-P3 cells by retinoic acid. Endocrinology 131:1017–1025 Rittenhouse J, Marcus F 1983 Peptide mapping by polyacrylamide gel electrophoresis after cleavage at aspartyl-prolyl peptide bonds in sodium dodecyl sulfate-containing buffers. Anal Biochem 138:442– 448 Szewczyk B, Kozloff LM 1985 A method for the efficient blotting of strongly basic proteins from sodium dodecyl sulfate-polyacrylamide gels to nitrocellulose. Anal Biochem 150:403– 407 Nordeen SK 1988 Luciferase reporter gene vectors for analysis of promoters and enhancers. BioTechniques 6:454 – 457 Bokar JA, Keri RA, Farmerie TA, Fenstermaker RA, Andersen B, Hamernik DL, Yun J, Wagner T, Nilson JH 1989 Expression of the glycoprotein hormone a-subunit gene in the placenta requires a functional cyclic AMP response element, whereas a different cis-acting element mediates pituitary-specific expression. Mol Cell Biol 9:5113–5122 Lin C, Lin SC, Chang CP, Rosenfeld MG 1992 Pit-1-dependent expression of the receptor for growth hormone releasing factor mediates pituitary cell growth. Nature 360:765–768 Maurer R 1989 Both isoforms of the cAMP-dependent protein kinase catalytic subunit can activate transcription of the prolactin gene. J Biol Chem 264:6870 – 6873 Day RN, Walder JA, Maurer RA 1989 A protein kinase inhibitor gene reduces both basal and multihormone-stimulated prolactin gene transcription. J Biol Chem 264:431– 436 Selden RF, Howie KB, Rowe ME, Goodman HM, Moore DD 1986 Human

42. 43.

44. 45. 46. 47. 48. 49. 50. 51. 52.

53. 54. 55.

56.

57.

58. 59.

60. 61.

62. 63.

64. 65.

937

growth hormone as a reporter gene in regulation studies employing transient gene expression. Mol Cell Biol 6:3173–3179 Wen W, Meinkoth JL, Tsien RY, Taylor SS 1995 Identification of a signal for rapid export of proteins from the nucleus. Cell 82:463– 473 Øyen O, Myklebust F, Scott JD, Cadd GG, McKnight GS, Hansson V, Jahnsen T 1990 Subunits of cyclic adenosine 39,59-monophosphate-dependent protein kinase show differential and distinct expression patterns during germ cell differentiation: alternative polyadenylation in germ cells gives rise to unique smaller-sized mRNA species. Biol Reprod 43:46 –54 Gamm DM, Baude EJ, Uhler MD 1996 The major catalytic subunit isoforms of cAMP-dependent protein kinase have distinct biochemical properties in vitro and in vivo. J Biol Chem 271:15736 –15742 Bravo R, Frank R, Blundell PA, Macdonald B, Macdonald H 1987 Cyclin/ PCNA is the auxiliary protein of polymerase-d. Nature 326:515–517 Ferrari A, Petraglia F, Gurpide E 1995 Corticotropin releasing factor decidualizes human endometrial stromal cells in vitro. Interaction with progestin. J Steroid Biochem Mol Biol 54:251–255 Tabanelli S, Tang B, Gurpide E 1992 In vitro decidualization of human endometrial stromal cells. J Steroid Biochem Mol Biol 42:337–344 Daly DC, Maslar IA, Riddick DH 1983 Prolactin production during in vitro decidualization of proliferative endometrium. Am J Obstet Gynecol 145:672– 678 Chen G, Huang JR, Mazella J, Tseng L 1989 Long-term effects of progestin and RU 486 on prolactin production and synthesis in human endometrial stromal cells. Hum Reprod 4:355–358 Narukawa S, Kanzaki H, Inoue T, Imai K, Higuchi T, Hatayama H, Kariya M, Mori T 1994 Androgens induce prolactin production by human endometrial stromal cells in vitro. J Clin Endocrinol Metab 78:165–168 Brar AK, Kessler CA, Meyer AJ, Cedars MI, Jikihara H 1996 Retinoic acid suppresses in-vitro decidualization of human endometrial stromal cells. Mol Hum Reprod 2:185–193 Brar AK, Frank GR, Kessler CA, Meyer AJ, Cedars MI, Activation of the cAMP pathway during progesterone-dependent decidualization of the human endometrium. 10th International Congress of Endocrinology, San Francisco CA, 1996, P3– 455 (Abstract) Houge G, Cho-Chung YS, Døskeland SO 1992 Differential expression of cAMP-kinase subunits is correlated with growth in rat mammary carcinomas and uterus. Br J Cancer 66:1022–1029 Schwoch G 1987 Selective regulation of the amount of catalytic subunit of cyclic AMP-dependent protein kinases during isoprenaline-induced growth of the rat parotid gland. Biochem J 248:243–250 Richardson JM, Howard P, Massa JS, Maurer RA 1990 Post-transcriptional regulation of cAMP-dependent protein kinase activity by cAMP in GH3 pituitary tumor cells. Evidence for increased degradation of catalytic subunit in the presence of cAMP. J Biol Chem 265:13635–13640 Houge G, Vintermyr OK, Døskeland SO 1990 The expression of cAMPdependent protein kinase subunits in primary rat hepatocyte cultures. Cyclic AMP down-regulates its own effector system by decreasing the amount of catalytic subunit and increasing the mRNAs for the inhibitory subunits of cAMP-dependent protein kinase. Mol Endocrinol 4:481– 488 Taskeń K, Andersen KB, Skålhegg BS, Taskeń KA, Hansson V, Jahnsen T, Blomhoff HK 1993 Reciprocal regulation of mRNA and protein for subunits of cAMP-dependent protein kinase (RIa and Ca) by cAMP in neoplastic B cell line (Reh). J Biol Chem 268:23483–23489 Xiong Y, Zhang H, Beach D 1992 D type cyclins associated with multiple protein kinases and DNA replication and DNA repair factor PCNA. Cell 71:505–514 Hagiwara M, Alberts A, Brindle P, Meinkoth J, Feramisco J, Deng T, Karin M, Shenolikar S, Montminy M 1992 Transcriptional attenuation following cAMP induction requires PP-1-mediated dephosphorylation of CREB. Cell 70:105–113 Hoesche C, Bartsch P, Kilimann MW 1995 The CRE consensus sequence in the synapsin I gene promoter region confers constitutive activation but no regulation by cAMP in neuroblastoma cells. Biochim Biophys Acta 1261:249 –256 Michael LF, Alcorn JL, Gao E, Mendelson CR 1996 Characterization of the cyclic adenosine 39,59-monophosphate response element of the rabbit surfactant protein-A gene: evidence for transactivators distinct from CREB/ATF family members. Mol Endocrinol 10:159 –170 Berwaer M, Martial JA, Davis JRE 1994 Characterization of an up-stream promoter directing extrapituitary expression of the human prolactin gene. Mol Endocrinol 8:635– 642 Gao JG, Tseng L 1996 Distal Sp3 binding sites in the hIGBP-1 gene promoter suppress transcriptional repression in decidualized human endometrial stromal cells: identification of a novel Sp3 form in decidual cells. Mol Endocrinol 10:613– 621 Gellersen B, Kempf R, Telgmann R 1997 Human endometrial stromal cells express novel isoforms of the transcriptional modulator CREM and upregulate ICER in the course of decidualization. Mol Endocrinol 11:97–113 Molina CA, Foulkes NS, Lalli E, Sassone-Corsi P 1993 Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early response repressor. Cell 75:875– 886