0013-7227/99/$03.00/0 Endocrinology Copyright © 1999 by The Endocrine Society

Vol. 140, No. 10 Printed in U.S.A.

CCAAT/Enhancer-Binding Protein a Is a Physiological Regulator of Prolactin Gene Expression* KIRSTEN K. JACOB†

AND

FREDERICK M. STANLEY

Departments of Medicine and Pharmacology, New York University School of Medicine, New York, New York 10016 ABSTRACT The sequence 2101/292 of the PRL promoter has been shown to be essential for both basal and hormone-increased PRL gene transcription. It is important to identify transcription factors that bind to this sequence if we are to understand the regulation of the PRL gene. Nuclear proteins, metabolically labeled with 35S were used in gel mobility shift experiments to examine which protein(s) binds to this region of the PRL promoter. An abundant 43-kDa protein binds to the PRL promoter at 2106/287. Two 43-kDa transcription factors were identified in cytosolic extracts of GH4 cells, CCAAT enhancer-binding protein a (C/EBPa) and cAMP response element-binding protein.

T

O UNDERSTAND how hormones regulate the transcription of genes, the transcription factors that mediate both basal transcription and hormone-increased transcription must be identified. Then it is possible to examine how they interact with one another to maintain basal transcription rates and how stimulus-coupled modification of these factors enhances gene transcription. The PRL promoter has been well studied, and a cell type-specific transcription factor Pit-1 has been identified that is required for basal PRL gene expression. Pit-1 has been identified in nuclear extracts of GH cells using deoxyribonuclease I footprinting analysis (1–3), and the Pit-1 complementary DNA has been cloned. This has facilitated studies that have identified the interactions of Pit-1 with DNA and with other transcription factors (4). However, Pit-1 may interact with unidentified, widely expressed factors that bind to 2101/292 of the PRL promoter, as this sequence is also necessary for basal PRL gene transcription (1, 2, 5, 6). The transcription factors that regulate hormone-increased PRL gene expression are beginning to be defined. The estrogen receptor binds the PRL promoter at 21700/21800 and mediates estrogen-increased PRL gene transcription (7). This requires interaction with Pit-1 and is ligand concentration dependent (8). Ets-related transcription factors have been shown to be important for activating the PRL promoter, and several Ets binding sites have been identified in the proximal promoter (4, 9, 9a). A multihormone response element is located at 2100/291. This element is required for Received December 16, 1998. Address all correspondence and requests for reprints to: Dr. Frederick M. Stanley, Department of Medicine, TH 450, New York University Medical Center, 550 First Avenue, New York, New York 10016. E-mail: [email protected]. * This work was supported by NIH Grant DK-43365. † Present address: Amgen, Inc., Thousand Oaks, California 91320.

Both of these bind to the PRL promoter, and both were present in GH4 cell nuclear extract, but only C/EBPa was definitively identified in complexes with PRL promoter DNA. Expression of C/EBPa increased basal PRL gene expression almost 6-fold, whereas expression of Chop10 that can act as an inhibitor of C/EBPa reduced the basal activity of the PRL promoter 60 –75%. Mutational analysis demonstrated that the ability of C/EBPa to increase basal expression of the PRL promoter was dependent on the sequence 2101/292. These data suggest that C/EBPa is an important transcription factor that regulates PRL gene expression. (Endocrinology 140: 4542– 4550, 1999)

the effects of epidermal growth factor (EGF) (10), insulin (5), and agents that increase cAMP (5, 11, 12). This element also binds Ets-related transcription factors (13) and other factors (6). The other factors that bind to this sequence have not been identified, although a 100-kDa protein was shown to bind here (6). CCAAT enhancer-binding proteins (C/EBPs) belong to the bZIP family of transcription factors. These have a conserved C-terminal domain containing a basic DNA-binding domain and a leucine zipper that mediates protein/protein interaction. At least six different genes have been identified that produce C/EBP-related proteins, c/ebpa, c/ebpb, c/ebpg, c/ebpd, celf, and chop10 (14 –16). Several of these have alternate splice variants. Thus, C/EBPa is found in alternately translated 42- and 30-kDa forms. The 42-kDa form is initiated from the first start codon and activates gene transcription, whereas the 30-kDa protein is initiated from the third start codon and has been shown to be inhibitory (17). These proteins can bind to DNA only as homo- or heterodimers, and heterodimerization between stimulatory and inhibitory bZIP proteins has been shown to be important in regulating their activity. The consensus DNA-binding sequence for bZIP proteins is T(T/G)NNGNAA(T/G) (18). C/EBPs are expressed only in terminally differentiated tissues such as liver and adipocytes. C/EBPs have been linked to numerous genes that are regulated by insulin and/or cAMP such as phosphoenolpyruvate carboxykinase (15) and the acetyl coenzyme A carboxylase gene (19). We present evidence in this report that a 43-kDa protein interacts with the PRL promoter at 2106/287. This protein could be C/EBPa, as it binds to this sequence both as an in vitro translated protein and as a factor present in GH4 cell nuclear extract. C/EBPa, expressed from a transfected plasmid, increased both basal and insulin-increased PRLchloramphenicol acetyltransferase (CAT) expression. This

4542

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

C/EBP REGULATION OF THE PRL GENE

suggested that C/EBPa may be a physiological regulator of basal and insulin-increased PRL gene expression. Chop10, which inhibits bZIP transcription factors such as C/EBPa by forming nonproductive associations with them, inhibits both basal and insulin- or EGF-increased PRL gene expression. Finally, mutations of the PRL promoter that eliminate the sequence at 2101/292 that was shown to be important for basal PRL gene transcription also eliminate the effects of C/EBPa on PRL gene expression. Together, these data suggest that C/EBPa may be the factor that regulates basal PRL gene expression by binding to 2101/292 and also that it participates in insulin-increased PRL gene expression through interaction with other transcription factors. Materials and Methods Materials [32P]Deoxy-CTP (3000 Ci/mmol), [32P]ATP (3000 Ci/mmol), and Trans 35S label were obtained from ICN Biochemicals, Inc. (Costa Mesa, CA). Trans 35S label is a metabolic labeling reagent derived from 35Slabeled Escherichia coli and contains 70% [35S]methionine and 15% [35S]cysteine. All enzymes and linkers were obtained from either New England Biolabs, Inc. (Beverley, MA) or from Pharmacia Biotech (Piscataway, NJ) and, unless otherwise indicated, were used under conditions recommended by the suppliers. Oligonucleotides were purchased from Operon Technologies (Alameda, CA). Duplex poly(dI-dC) was obtained from Pharmacia Biotech. Antibodies to C/EBPa and C/EBPd were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibody to cAMP response element-binding protein (CREB) and phospho-CREB were gifts from Dr. Michael Greenberg (Harvard University, Boston, MA). Antibodies to C/EBPb and Chop10 were gifts from Dr. D. Ron (New York University School of Medicine, New York, NY). Horseradish peroxidase-labeled secondary antibodies to mouse and rabbit IgG were obtained from Kirkegaard & Perry Laboratories (Gaithersburg, MD). Reagents used for gel electrophoresis were purchased from Fisher Scientific (Fairlawn, NJ). Acetyl coenzyme A and silica gel plates were obtained from Sigma Chemical Co. (St. Louis, MO). DMEM containing 4.5 g/liter glucose (DMEM) and iron-supplemented calf serum were obtained from HyClone Laboratories, Inc. (Logan, UT). Triton X-100, reagents for enhanced chemiluminescence, and bicinchoninic acid reagent were purchased from Pierce Chemical Co. (Rockford, IL). All other reagents were of the highest purity available and were obtained from Sigma Chemical Co., Behring Diagnostics, Calbiochem (La Jolla, CA), Bio-Rad Laboratories, Inc. (Richmond, CA), Eastman Kodak Co. (Rochester, NY), Fisher Scientific, or Roche Molecular Biochemicals (Branchburg, NJ).

Plasmids The construction of pPRL-CAT plasmids containing 2173/175 of rat PRL 59-flanking DNA was described previously (20). The human insulin expression vector, pRT3HIR2, was a gift from Dr. J. Whittaker (Stony Brook, NY). The expression vector for C/EBPa was a gift from Dr. S. L. McKnight (Tularik, South San Francisco, CA). The expression vector for CREB was obtained from Dr. M. R. Montminy (The Salk Institute, San Diego, CA). The construction and sequence of the PRL promoter mutants used in Fig. 7 were described previously (5). The sequence of the wild-type promoter is 2106 TCTTAATGAC GGAAATAGAT GATTGGGAGG GGAAGAGGAT GCCTGATTAT 257. The LS(2101/ 292)CAT has the sequence 2106 TCTTAgaaga tcttcTAGAT GATTGGGAGG GCCTGATTAT 257. The sequence of the double linker scanning mutant, LS(296/87, 276/267)CAT, is 2106 TCTTAATGAC GaAgATcttc GATTGGGAGG GaAgatcttc GCCTGATTAT 257. The bases in lowercase have been mutated.

Western immunoblot analysis GH4 cells were harvested after 48 h with or without 1 mg/ml insulin, and nuclei were prepared as previously described (21). A nuclear extract was prepared by disrupting the nuclei with 400 mm KCl in a buffer

4543

containing 15% glycerol, 25 mm Tris (pH 8), 10 mm b-mercaptoethanol, 0.5 mm EDTA, and 0.05% Triton X-100. SDS-PAGE was performed using 10% gels (22). The proteins were then blotted to nitro-cellulose membranes (Micron Separations, Westboro, MA) in Towbin’s buffer (25 mm Trisma base, 192 mm glycine, and 20% methanol). Immunoblotting using enhanced chemiluminescence was performed as described by the manufacturer (Pierce Chemical Co.). 35

S-Labeling of GH4 cell nuclear proteins

GH4 cells were depleted of methionine and cysteine by incubation for 2 h in F-10 medium without cysteine and methionine and with 10% dialyzed calf serum. This medium was then replaced with medium consisting of cysteine/methionine-free F-10 to which 1/10th the normal concentration of cysteine and methionine had been added as well as 0.1 mCi/ml Trans 35 S label and 10% dialyzed calf serum. The cells were harvested after 2 h by washing with ice-cold saline solution. Nuclear extract was prepared as previously described (23) and stored at 280 C in 20 mm HEPES (pH 7.9), 20% glycerol, 100 mm KCl, 0.2 mm EDTA, 0.2 mm EGTA, 2 mm dithiothreitol, and 1 mm phenylmethylsulfonylfluoride.

Assay of DNA-protein binding by gel electrophoresis An oligonucleotide to the PRL promoter sequences 101/264 was prepared, purified on polyacrylamide gels, and end labeled with [32P]deoxy-CTP. The sequence of this oligonucleotide is 59-AATGACGGAAATAGATTGGGAGGGGAAGAGGATGC-39. Labeled PRL 59flanking DNA was then used in mobility shift experiments with unlabeled nuclear extracts performed as previously described (21). Two micrograms of nuclear extract (prepared as described above) were incubated at 25 C for 30 min with 10,000 cpm (10 –20 fm) 32P-labeled PRL 2106/287. The protein-DNA complexes were then analyzed by electrophoresis on a 6% polyacrylamide gel in Tris-acetate-EDTA buffer.

Analysis of DNA-bound proteins by gel shift/ Western immunoblot GH4 cell nuclear extracts were incubated with 32P-labeled PRL promoter DNA, 2106/287, and the protein-DNA complexes were resolved using PAGE as described above. The gel was then electroblotted in Towbin’s buffer through nitro-cellulose to DE81 paper as previously described (24). The DE81 paper captures the DNA components of the complex, whereas the protein components are captured on the nitrocellulose. The radioactive image captured on the DE81 paper was developed using the Molecular Dynamics, Inc. PhosphorImager (Sunnyvale, CA). The proteins were visualized using Western blotting with specific antibody to C/EBPa (Santa Cruz Biotechnology, Inc.) and CREB and phospho-CREB (Dr. Michael Greenberg). Enhanced chemiluminescence was performed with reagents from Pierce Chemical Co. as described above.

Analysis of PRL promoter responsiveness using transient transfection Electroporation experiments and CAT assays were performed as previously described (21). GH4 cells were harvested with an EDTA solution, and 20 – 40 3 106 cells were used for each electroporation. Trypan blue exclusion before electroporation ranged from 95–99%. The voltage of the electroporation was 1550 V. This gives trypan blue exclusion of 70 – 80% after electroporation. The transformed cells were then plated in multiwell dishes (Falcon Plastics, Oxnard, CA) at 5 3 106 cells/9-cm2 tissue culture well in DMEM with 10% hormone-depleted serum [prepared with ion exchange resin and charcoal as previously described (25)]. Cells were refed at 24 h with DMEM with 10% hormonedepleted serum with or without insulin (1 mg/ml bovine insulin; Calbiochem, San Diego, CA), EGF (40 ng/ml recombinant human EGF; R&D Systems, Minneapolis, MN), or cAMP (0.1 mm, 8-(4-chlorophenylthio)-cAMP; Sigma Chemcial Co.). After 48 h, the flasks were washed three times with normal saline and frozen. The cells were harvested, and CAT activity was assayed as described previously (26), except that in later experiments [14C]chloramphenicol was replaced with BODIPY chloramphenicol (Molecular Probes, Inc., Eugene, OR), and fluorescence

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

4544

C/EBP REGULATION OF THE PRL GENE

FIG. 1. Analysis of proteins in specifically retarded complexes of 32Plabeled GH4 cell nuclear extract and PRL promoter DNA. GH4 cells were metabolically labeled with 35S, and nuclear extract was prepared as described in Materials and Methods. These proteins (20 mg) were then incubated with 200,000 cpm 32P-labeled PRL 2106/287 (200 – 400 fmol). The protein/DNA complexes were resolved by electrophoresis on 6% polyacrylamide gels in Tris-acetate-EDTA buffer, pH 7.6. The wet gel was exposed to X-Omat AR5 film to determine the location of the shifted bands. The specifically shifted band was then cut out from the gel, and the proteins were eluted into SDS sample buffer. The proteins were also eluted from an analogous position in a lane that had protein but no DNA. The proteins were resolved by SDS-PAGE on a 10% gel. The left panel shows the gel mobility shift pattern (the free DNA was run off the gel to improve resolution of the shifted complex). The right panel shows the resolution of the proteins on SDS-PAGE. The 35S-labeled proteins eluted from the shifted complex were electrophoresed in lane 1, whereas the 35S-labeled proteins eluted from the area of the gel shift without DNA were loaded in lane 2. The arrows on the left show the positions and sizes of the proteins in the complex, whereas the arrows on the right show the migration of mol wt markers (Rainbow markers, Amersham Pharmacia Biotech). intensity was measured using a FluorImager 575 (Molecular Dynamics, Inc.) with ImageQuant software. Control of transfection efficiency was performed using a Rous sarcoma virus (RSV)-b-galactosidase expression plasmid. Briefly, 2 mg RSVb-galactosidase expression plasmid were included in all electroporations. The b-galactosidase activity in the cell lysates was determined using o-nitrophenyl-b-d-galactopyranoside. Transfection efficiency did not vary significantly among transfections performed at the same time. In a typical experiment the mean OD420 for seven electroporations was 0.724, with a sd of 0.06. The percent acetylation was then corrected for minor variations in b-galactosidase activity by converting the percent acetylation to percent acetylation per OD430 b-galactosidase activity/mg protein. The fold stimulation or inhibition was then determined. Significance was determined using Student’s t test and is reported where appropriate.

Results A 43-kDa protein binds to the PRL promoter multihormone response element

GH4 cells were metabolically labeled with [35S]methionine, and gel mobility shift experiments were performed with the 35S-labeled nuclear extracts to identify transcription factors that bind to the PRL promoter. The 35S-labeled extract was incubated with 32P-labeled PRL promoter DNA 2106/ 287, and the DNA/protein complexes were resolved by

Endo • 1999 Vol 140 • No 10

FIG. 2. Analysis of transcription factors in GH4 cell nuclear extracts by immunoblotting. The proteins in a cytosolic extract of GH4 cells were resolved by SDS-PAGE. The proteins were then blotted to nitrocellulose, and different primary antibodies were applied to this membrane using a Mini-PROTEAN II multiscreen apparatus (Bio-Rad Laboratories, Inc., Richmond, CA). Lane 1, Antibody to CREB (Upstate Biotechnology, Inc., Lake Placid, CA); lane 2, antibody to C/EBPa (Santa Cruz Biotechnology, Inc.); lane 3, antibody to C/EBPb (D. Ron); lane 4, antibody against C/EBPd (Santa Cruz Biotechnology, Inc.). The arrow on the left indicates the position of CREB/C/EBPa. The arrows on the right indicate the migration of mol wt markers (Rainbow markers, Amersham Pharmacia Biotech).

electrophoresis in a polyacrylamide gel under nondenaturing conditions. The major band (Fig. 1, left), previously determined to be specific by inhibition with excess unlabeled oligonucleotide (26), was then cut out, and the proteins were eluted into SDS sample buffer and resolved on SDS-PAGE (Fig. 1, right, lane 1). Several proteins are contained in this complex. The most prominent has a molecular size of 43 kDa, whereas minor bands are seen at 52 and 72 kDa. There are no bands seen in lane 2 (Fig. 1, right). This lane resolved proteins that were eluted from a region of the gel shift parallel to the specific complex (Fig. 1, left) that contained 35Slabeled nuclear extract without DNA (the migration of uncomplexed proteins into the gel). A number of transcription factors have molecular masses 610% of 43 kDa including CREB, C/EBPa, and c-Jun. Several factors suggested that this 43-kDa protein might be either C/EBPa or CREB. First, this sequence is a cAMP response element, and both CREB and C/EBP bind to such sequences (27). Second, it has been reported that C/EBP associates with Ets-related transcription factors (28) that are known to bind to this sequence (29). Finally, C/EBPa was found to be a major regulator of the related GH gene (30). CREB and C/EBPa bind to the PRL promoter multihormone response element

Immunoblot analysis of GH4 cell nuclear extracts was performed to determine whether GH4 cells express CREB and/or C/EBP, as CREB and/or C/EBP must be present in GH4 cells to be physiologically relevant to PRL gene expression. Figure 2 shows an immunoblot of GH4 cell cytoplasmic proteins resolved on SDS-PAGE and blotted with specific antibodies to CREB, C/EBPa, C/EBPb, and C/EBPd. A distinct band is seen at 42 kDa in the lanes that were blotted with antibody against CREB or C/EBPa. This is the reported size of C/EBPa. The C/EBPa band comigrates with CREB, and the sizes of the CREB and C/EBPa also agree with the sizes of CREB and C/EBPa from in vitro translated and 35S-labeled CREB and C/EBPa (data not shown). The antibodies to

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

C/EBP REGULATION OF THE PRL GENE

C/EBPb and C/EBPd appear to be significantly less specific than the antibodies for CREB and C/EBPa, but the major bands in these lanes do not correspond to the expected molecular size for the C/EBPb isoforms (31.5 and 20 kDa) or for C/EBPd (35 kDa). Thus, both CREB and C/EBPa, but not C/EBPb and C/EBPd, are components of GH4 cell nuclear extract. C/EBP family members can bind to diverse sequences (28). The multifactor response element at 2101/292 of the PRL promoter has the sequence ATGACGGAAA, which is a partial match (8 of 10) with the consensus C/EBP response element AT(T/G)NNG(A/C/T)AA(T/G). It was possible that this is also a C/EBP response element as well as a cAMP response element. Therefore, several approaches were used to determine whether C/EBPa or CREB or both proteins can bind to this PRL promoter sequence. First, competition experiments were performed. Nuclear extract was incubated with 32P-labeled PRL promoter DNA with or without an oligonucleotide for the consensus C/EBP response element (Fig. 3, lanes 1– 4) or cAMP response element (Fig. 3, lanes 5– 8). Competition with the unlabeled PRL DNA was included as a control (Fig. 3, lane 10). In the absence of competitor, mobility shifts of GH4 cell nuclear extracts with PRL promoter DNA show a number of specific DNA-protein complexes (compare Fig. 3, lanes 9 and 10). One of these complexes becomes less prominent when an unlabeled oligonucleotide to the C/EBP response element is added to the

FIG. 3. Competition analysis of binding activities present in GH4 cell nuclear extracts. PRL 2101/264 labeled with 32P was incubated with 2 mg nuclear extract from GH4 cells, and the protein/DNA complexes were resolved on a 6% polyacrylamide gel under nondenaturing conditions (lanes 1, 5, and 9). Unlabeled oligonucleotide for a consensus C/REB response element (TTGCGCAAT) was included in the incubations resolved in lane 2 (2-fold molar excess), lane 3 (10-fold molar excess), and lane 4 (100-fold molar excess). Unlabeled oligonucleotide to a consensus cAMP response element (TGACGTCA) was added to the incubations resolved in lane 6 (2-fold molar excess), lane 7 (10-fold Molar excess), and lane 8 (100-fold molar excess). Lane 10 shows the competition with a 100-fold molar excess of unlabeled PRL 2101/264. The location of the DNA protein complex that is lost in incubations with unlabeled C/EBP-RE and CRE DNA is indicated on the left.

4545

incubation at 2-fold the concentration of labeled oligonucleotide (Fig. 3, compare lanes 1 and 2). This band is not seen when a 10- or 100-fold molar excess of unlabeled oligonucleotide is added to the incubation (lanes 3 and 4). However, this band is also competed by the related sequence for the CREB response element (Fig. 3, lanes 6 – 8). Therefore, another gel shift was performed comparing the binding of in vitro translated C/EBPa and CREB with the pattern of protein/DNA complexes in GH4 cell nuclear extracts. C/EBPa that was made in reticulocyte lysates produces a specific shifted complex (Fig. 4, lane 5) that is eliminated by including the unlabeled oligonucleotide for either the C/EBP (Fig. 4, lane 6) or CREB response element (Fig. 4, lane 7). This band runs with approximately the same mobility as the shifted band in the GH4 cell nuclear extract (Fig. 4, lane 1) that is inhibited by the C/EBP (Fig. 4, lane 3) or CREB response element DNA (Fig. 4, lane 4). This suggests that this complex seen in nuclear extract might be C/EBPa. However, in vitro translated CREB also produced a specifically DNA-protein complex of approximately the same mobility (Fig. 4, lane 8) that is likewise competed by incubation with unlabeled C/EBP (Fig. 4, lane 9) or CREB response element DNA (Fig. 4, lane 10). C/EBPa in GH4 cell nuclear extracts binds to the PRL promoter multihormone response element

These experiments demonstrated that both CREB and C/EBPa can bind to PRL promoter DNA and that nuclear extracts contain a binding activity producing a similar shift.

FIG. 4. Binding of C/EBP and CREB to the PRL promoter. The 32Plabeled PRL 2101/264 was incubated with 2 mg nuclear extract (lanes 1– 4), 10 ml C/EBPa programmed reticulocyte lysate (lanes 5–7), or 2 ml CREB programmed reticulocyte lysate (lanes 8 –10). Specific complexes were identified by incubation with a 100-fold molar excess of unlabeled DNA to PRL 2101/264 (lane 2), the C/EBP response element (lanes 3, 6, and 9), or CRE (lanes 4, 7, and 10). The arrow on the left indicates the position of the CREB/C/EBP band seen in lanes using reticulocyte lysates of CREB and C/EBPa. Lanes 1– 4 and lanes 5–10 are separate exposures of the same gel.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

4546

C/EBP REGULATION OF THE PRL GENE

However, they do not demonstrate the presence of these factors in the shifted complex. Gel mobility shift/Western blot experiments (24) confirmed that the C/EBPa in nuclear extracts produced this shift. This technique uses a sandwich of papers with different affinities for DNA and protein. The protein/DNA complexes resolved on a gel mobility shift are electroblotted first to nitro-cellulose. The proteins in the DNA/protein complexes are captured in this layer. The labeled DNA passes through the nitro-cellulose and is bound by the DE-81 paper. The radiolabeled DNA on the DE-81 paper was then visualized using the PhosphorImager, whereas the nitro-cellulose was probed with antibodies to C/EBPa and CREB. Nuclear extract incubated with 32P-labeled PRL promoter DNA (Fig. 5A, lane 2) resulted in two specific complexes with approximately the same mobility as the C/EBPa-DNA and CREB-DNA complexes seen in the previous experiment (Fig. 4, lanes 5 and 8). When the matching nitro-cellulose blot was probed with C/EBPa antibody, two strong C/EBPa signals were seen in this lane (Fig. 5A, lane 4). These bands correspond exactly with the two uppershifted bands seen on the left of this panel (Fig. 5A, lane 2). Some C/EBPa-immunoreactive material is also seen in the lane that was without PRL DNA (Fig. 5A, lane 3). Although this C/EBPa also migrates into the gel to the same position as the upper-shifted band, its intensity is much less than that in lane with PRL DNA (compare Fig. 5A, lanes 3 and 4). No C/EBPa was seen at the position corresponding to the middle-shifted band in the lane that contained no DNA even with longer exposures of the Western blot. This experiment indicates that C/EBPa is one component of specific complexes formed by the PRL promoter and GH4 cell nuclear extract. A control experiment (Fig. 5B) demonstrates the binding of C/EBPa in GH4 cell nuclear extract to the consensus C/EBP response element. The gel mobility shift shows two major and one minor shifted bands (Fig. 5B, lane 2). C/EBPaimmunoreactive material was identified that corresponded to several of these bands (Fig. 5B, lane 4), although not to the major shifted band (the middle band of Fig. 5B, lane 2). Other experiments (data not shown) demonstrated that this major complex was not specific. Over exposure of the Western blot shows that a small amount of C/EBPa-immunoreactive material migrates into the gel without DNA to approximately the same position as the major upper band of C/EBPa immunoreactive material seen in 32P-labeled DNA-containing lane. However, this was approximately 10- to 100-fold less than that in the shifted band. An identical experiment using antibodies to CREB and phospho-CREB failed to convincingly identify CREB binding to the PRL promoter DNA (Fig. 5C). Two distinct bands of CREB-immunoreactive material migrated into the gel. However, these bands were identical in the lanes with and without DNA (Fig. 5C, lanes 3 and 4). Moreover, neither of these bands corresponded to the shifted band seen in the analysis of [32P]DNA migration into the gel (Fig. 5C, lane 2). This suggests that little or no CREB binds to PRL (2101/264). However, it cannot be ruled out that the CREB antibody did not have a high enough affinity to identify CREB in these complexes. The control experiment seen in Fig. 5D rules out this possibility. This was an identical gel shift/Western blot exper-

Endo • 1999 Vol 140 • No 10

iment performed with [32P]cAMP response element DNA. One major shifted band is seen using labeled cAMP response element DNA and GH4 cell nuclear extract (Fig. 5D, lane 2). The Western blot of this experiment demonstrates that this shifted band contained CREB (compare Fig. 5D, lanes 2 ad 4). In the lane without labeled DNA, the same two bands of CREB immunoreactivity are seen as in the experiment with the PRL promoter DNA (compare lane 3 of Fig. 5C with lane 3 of Fig. 5D). Thus, the CREB that the Western blots (Fig. 2) identified in GH4 cell nuclear extract appropriately binds its consensus response element, and this complex can be identified using the gel shift/Western blot technique. However, it does not bind the PRL promoter in sufficient abundance to be detected. C/EBPa increases basal expression of the PRL promoter

Plasmids containing the complementary DNA for CREB and C/EBPa were then used in transient transfection experiments to determine whether overexpression of CREB or C/EBPa could affect basal or hormone-increased PRL gene expression. Expression of CREB did not result in a significant change from levels of PRL-CAT expression in control cells. However, expression of C/EBPa results in a 5.5-fold increase in basal PRL-CAT expression (P , 0.01, basal-control vs. basal-C/EBP) from 2 6 0.8% acetylation/mg proteinzh in control cells to 11 6 1.1% acetylation/mg proteinzh in C/EBPa-expressing cells (Fig. 6). PRL-CAT expression in the presence of both insulin and exogenous C/EBPa was increased in an additive manner (P , 0.025, insulin-control vs. insulin-C/EBP) from 20 6 1.3% to 32 6 2.1% acetylation/mg proteinzh with expression of C/EBPa. EGF- and cAMP-increased PRL gene expression was not increased significantly beyond the increase seen due to C/EBPa expression (P . 0.05). These studies indicate that C/EBPa might play a role to increase both basal and hormone-increased PRL gene expression. To determine whether C/EBP might play a role in physiological regulation of PRL gene transcription, the protein Chop10 was used as a dominant negative inhibitor of C/EBP. Chop10 is a bZIP protein similar to C/EBP that has two proline substitutions in its DNA-binding domain that make it unable to bind DNA (16). However, it heterodimerizes with C/EBP and prevents its binding to DNA. This has been shown to block C/EBP-increased transcription of the angiotensin gene that has a well defined response element for C/EBP. Expression of Chop10 in GH4 cells reduces basal and insulin- or EGF-increased PRL gene expression (Fig. 6). The cAMP-increased expression is not affected under the same conditions. Basal transcription is reduced from 2 6 0.8% acetylation/mg proteinzh in control cells to 0.7 6 0.2% acetylation/mg proteinzh in Chop10 electroporated cells (P , 0.05). Insulin-increased PRL gene expression is reduced from 20 6 1.3% acetylation/mg proteinzh in control cells treated with insulin to 10 6 1.3% acetylation/mg proteinzh in Chop10 electroporated cells treated with insulin (P , 0.025). Similar changes were noted for EGF-increased PRL-CAT expression. EGF-increased PRL-CAT expression was 12 61.3% acetylation/mg proteinzh in control cells and 5 6 0.3% acetylation/mg proteinzh in Chop10-treated cells (P , 0.025). The

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

C/EBP REGULATION OF THE PRL GENE

4547

FIG. 5. Identification of C/EBPa binding to the PRL promoter by gel mobility shift/Western blotting of GH4 cell nuclear proteins. A, Analysis of C/EBPa association with the PRL promoter. Twenty micrograms of GH4 cell nuclear extract were incubated without any DNA or with 100,000 cpm of an oligonucleotide to PRL 2101/264 (32P-PRL-101/264) and without any DNA and resolved on a 6% polyacrylamide gel as described in Materials and Methods. The gel was then electroblotted through nitro-cellulose (binding proteins) to DE81 paper (binding DNA). The labeled DNA in the gel was visualized using a PhosphorImager (Molecular Dynamics, Inc.), and the migration of 32P labeled DNA into the gel is shown on the left of each panel (PhosphorImager). The arrows on the left indicate the migration of the specifically retarded DNA protein complexes. The proteins from the gel that were captured on nitro-cellulose were Western blotted against an antibody to C/EBPa (Santa Cruz Biotechnology, Inc.). The pattern of C/EBPa migration in the gel shift is seen on the right of the panel (Western blot). The arrows on the right indicate the location of C/EBPa immunoreactivity. B, Analysis of C/EBPa binding to a C/EBP response element. Twenty micrograms of GH4 cell nuclear extract were incubated without any DNA or with 100,000 cpm of a 32P-labeled oligonucleotide to the C/EBP response element. The protein-DNA complexes were resolved, blotted to nitro-cellulose and DE81 paper, and analyzed as described above. The arrows on the left indicate the migration of the specifically retarded DNA protein complexes in the gel shift (lanes 1 and 2; PhosphorImager). The arrows on the left indicate the migration of the specifically retarded DNA protein complexes. The proteins from the gel that were captured on nitro-cellulose were Western blotted against an antibody to C/EBPa (Santa Cruz Biotechnology, Inc.). The pattern of C/EBPa migration in the gel shift is seen on the right of the panel (lanes 3 and 4; Western blot). The arrows on the right indicate the location of C/EBPa immunoreactivity. C, Analysis of CREB binding to the PRL promoter. Twenty micrograms of GH4 cell nuclear extract were incubated without any DNA or with 100,000 cpm of a 32P-labeled oligonucleotide to PRL 2101/264 (left panel; 32P-PRL-101/264) and without any DNA. The protein-DNA complexes were resolved, blotted to nitro-cellulose and DE81 paper, and analyzed as described above. The arrows on the left indicate the migration of the specifically retarded DNA protein complexes in the gel shift (left two lanes; PhosphorImager). The proteins from this gel mobility shift that were captured on nitro-cellulose were Western blotted against an antibody to CREB and phospho-CREB (Dr. Michael Greenberg, Harvard University). The pattern of CREB migration in the gel shift is seen in the right two lanes of the panel (Western blot). The arrows on the right of the panel indicate the location of CREB immunoreactivity. D, Analysis of CREB binding to the CRE. Twenty micrograms of GH4 cell nuclear extract were incubated without any DNA or with 100,000 cpm of a 32P-labeled oligonucleotide to the CREB response element (32P-CRE). The arrows on the left indicate the migration of the specifically retarded DNA protein complexes in the gel mobility shift (PhosphorImager). The proteins from this gel that were captured on nitro-cellulose were Western blotted against an antibody to CREB and phospho-CREB (Dr. Michael Greenberg, Harvard University). The pattern of CREB migration in the gel shift is seen on the right of the panel (Western blot). The arrows on the right indicate the location of CREB immunoreactivity.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

4548

Endo • 1999 Vol 140 • No 10

C/EBP REGULATION OF THE PRL GENE

FIG. 6. Effect of expression of C/EBP on insulin-increased PRL-CAT transcription. In control transfections, GH4 cells were cotransfected with 10 mg PRL(2173/175)CAT (34), 2 mg RSV-bGal, and 5 mg of an expression vector for the human insulin receptor, pRT3HIR2 (J. Whittaker, Stony Brook, NY) alone or with 10 mg of a cytomegalovirus expression vector pRK5. Vectors expressing CREB (M. Montminy, The Salk Institute), C/EBPa (S. L. McKnight, Tularik), or Chop10 (D. Ron, New York University School of Medicine) under control of the cytomegalovirus promoter were included at 10 mg in the designated experiments. After 24 h, the medium was exchanged, and 1 mg/ml insulin, 40 ng/nl EGF, or 0.1 mM 8-(chlorophenylthio)-cAMP were added to the appropriate cultures. The plates were harvested 48 h after electroporation by washing three times with normal saline and freezing. The average percent acetylation per 10 mg protein in control and insulin-, EGF-, or cAMP-treated cultures was determined and adjusted for b-galactosidase expression. The results are expressed as normalized percent acetylation. That is, the percent acetylation per 10 mg protein/h adjusted to the b-galactosidase activity per 10 mg protein. The results are from three separate experiments performed in duplicate.

effect of cAMP was not significantly affected by Chop10 expression and was 16 6 1.8% acetylation/mg proteinzh in control cells and 12 6 1.6% acetylation/mg proteinzh in the Chop10-transfected cells. C/EBPa requires the sequences 2101/292 to activate PRL gene expression

The effect of C/EBPa on basal PRL-CAT expression was determined using deletion and linker-scanning mutants of the PRL promoter to determine the promoter element affected by C/EBPa. The activation of the PRL promoter was eliminated using the linker-scanning mutant LS(2101/ 292)CAT (Fig. 7), and basal expression from this plasmid was reduced from 3 6 0.5% acetylation/10 mg proteinzh in control cells with the wild-type reporter, PRL(2173/ 175)CAT to 1 6 0.3% acetylation/10 mg proteinzh in cells expressing LS(2101/292)CAT. Reduced basal PRL-CAT transcription (1.1 6 0.5% acetylation/10 mg proteinzh) and no C/EBPa stimulation (1.0 6 0.6% acetylation/10 mg proteinzh) was also seen with the reporter PRL(‚2106/267)CAT, which has a 40-bp deletion that includes the sequence 2101/ 292. The plasmid LS(296/287,276/267)CAT is a double linker-scanning mutant that has been shown to eliminate

FIG. 7. Deletion analysis of C/EBPa activation of basal PRL gene expression. GH4 cells were electroporated with 2 mg RSV-bGal and 10 mg of the PRL-CAT reporter plasmid indicated with or without 5 g C/EBPa expression plasmid. The cultures were refed at 24 h, and the plates were harvested 48 h after electroporation by washing three times with normal saline and freezing. The average percent acetylation per 10 mg protein/h was determined and adjusted for b-galactosidase expression. The normalized percent acetylation is shown. The results are from three separate experiments performed in duplicate.

insulin- and EGF-increased PRL-CAT expression (10, 13). Basal PRL-CAT expression with this plasmid (2.9 6 0.8% acetylation/10 mg proteinzh) is not significantly different from the control value. However, C/EBPa produces only a 2-fold increase over basal levels (5.6 6 1.0% acetylation/10 mg proteinzh) using this reporter plasmid. Finally, PRL(‚296/267)CAT has significantly reduced (P , 0.05) basal PRL-CAT expression (1.2 6 0.3% acetylation/10 mg proteinzh) as well as reduced stimulation by C/EBPa (3.9 6 0.4% acetylation/10 mg proteinzh). These data suggest that C/EBPa acts at the sequence 2101/292 to increase basal PRL-CAT gene expression, as mutations that completely eliminate this sequence (LS(2101/292)CAT and PRL(‚2106/267)CAT) are completely inactive, whereas those where this sequence is only partly mutated (LS(296/ 287,276/267)CAT and PRL(‚296/267)CAT) are only partially inactive. Discussion

These studies demonstrate that C/EBPa binds the PRL promoter and that C/EBPa expression may participate in the physiological regulation of PRL gene transcription. First, four experiments demonstrate that C/EBPa is present in GH cell nuclear extracts and binds to the PRL promoter. 1) A number of DNA-protein complexes were observed when 32 P-labeled PRL DNA was incubated with GH4 cell nuclear extract and resolved on polyacrylamide gels. One of those complexes was lost when unlabeled C/EBP response element DNA was included in this incubation. 2) C/EBPa translated in vitro binds to PRL promoter DNA and produces a shifted complex with the same mobility as that in GH4 cell

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

C/EBP REGULATION OF THE PRL GENE

nuclear extract. 3) Gel shift/Western blots identify C/EBPa in a complex with PRL promoter DNA. Thus, C/EBPa can be considered a PRL promoter-interacting factor. Second, functional studies also suggest a role for C/EBPa in PRL gene transcription. Over expression of C/EBPa increases basal PRL-CAT expression. More convincingly, however, Chop10 inhibits basal and hormone-stimulated PRLCAT expression. Chop10 is known to act by forming nonproductive complexes with other bZIP proteins to inhibit their function (16). Thus, this experiment suggests that the Chop10 expressed from the plasmid binds to endogenous C/EBPa and blocks its normal function to maintain physiological levels of PRL gene expression. The C/EBPa-mediated increase is dependent on the sequence 2101/292, as shown by the deletion analysis presented in Fig. 7. This region of the PRL promoter was previously shown to be important for both basal and EGF-, insulin-, and cAMPincreased PRL-CAT expression (5, 11, 12, 31, 32). It is possible that C/EBPa is the factor binding to this sequence that maintains high basal levels of PRL gene expression. This would agree with studies of the GH promoter that have demonstrated the binding of C/EBPa to the GHF3 site of the GH promoter and cooperation between C/EBPa and Pit-1 (30). C/EBP or Pit-1 alone was able to increase GH gene transcription, and the interaction of C/EBPa and Pit1 led to higher levels of basal GH transcription in pituitary progenitor, GHFT1–5 cells, which do not normally express C/EBPa. The interaction of Pit1 and C/EBPa on the GH promoter is similar to C/EBP function on the albumin promoter in liver, where C/EBP interacts with hepatocyte nuclear factors to achieve maximal stimulation (33). Finally, the maximal stimulation of the GH promoter required hormonal inputs as well as expression of C/EBPa. This is analogous to the PRL promoter, where insulin and C/EBPa induce maximal PRL-CAT expression. The bZIP proteins interact with each other to form heterodimers. Therefore, it is possible that C/EBPa acts alone (as a homodimer) or in combination with other bZIP proteins. The data suggest that C/EBPa is the only bZIP factor acting in these experiments. Some of the bZIP proteins, such as Chop10 and Lip, an alternately initiated C/EBPb, are inhibitors of transcription, and thus, C/EBPa is probably not complexed with these factors in these experiments. C/EBPa could complex with C/EBPb or C/EBPd. No evidence of these proteins was seen on Western blots. However, other bZIP proteins are being isolated (34). Thus, it is possible that C/EBPa interacts with a family member for which we do not have antibody or which has not been isolated. This could be addressed in glutathione-S-transferase-C/EBP pull-down experiments to identify factors interacting with the C/EBP leucine zipper. It has been shown that C/EBPb activates the somatostatin promoter through the CRE and that CREB competes with C/EBPb for occupancy of the CRE. In the presence of a heat-stable inhibitor of protein kinase A, CREB inhibits somatostatin gene expression due to competition between the inactive, unphosphorylated CREB and the constitutively active C/EBPb (35). However, other studies have indicated no functional interference between CREB and C/EBP. Thus, the binding of C/EBPa to the CRE of the phosphoenolpyruvate

4549

carboxykinase promoter has no effect on cAMP stimulation of that gene (36). Likewise, the C/EBP-related factor, CLEF, binds to the CRE of the substance P promoter, but does not interfere with cAMP stimulation of this gene (27). As no CREB binding to the PRL promoter was observed in our studies, it is possible that cAMP activates PRL gene expression through C/EBPa. Our data demonstrate that exogenous C/EBPa increases PRL-CAT expression, and that neither cAMP nor EGF has any further effect to increase PRL gene expression in the presence of exogenously expressed C/EBPa. Further, the fold stimulation of PRL-CAT by insulin is reduced in the presence of C/EBPa. This might result if C/EBPa competed with the EGF and the cAMP effector molecules for binding to this common response element. Alternately, the phosphorylation or increased production of C/EBPa due to EGF and/or cAMP might mediate EGF and/or cAMP activation of PRL gene expression. The lack of any additive effect of EGF and cAMP in the presence of exogenous C/EBPa supports this explanation. The reduced fold stimulation by insulin could result from inhibition of the binding of the insulin-responsive transcription factor or the effect of both agents might be the maximal stimulation that this site is capable of maintaining. This explanation is supported by previous studies showing that insulin, EGF, and cAMP, added in any combination of two factors, produced an additive activation of the PRL promoter. However, the addition of all three agents together did not increase PRL-CAT expression more than any two of these agents (10). This suggests that a 30- to 40-fold stimulation is the maximal achievable through this site. Insulin influences both the phosphorylation (17) and production of C/EBP family members (37). Therefore, insulin might act at least partially by altering the abundance or activity of C/EBPa. However, we do not regard this as likely. First, dephosphorylation of C/EBPa by insulin has not been shown to be responsible for the insulin-mediated inhibition of Glut4 expression in 3T3-L1 adipocytes (17). Second, insulin has been shown to decrease the production of C/EBPa in 3T3-L1 cells (17) and to decrease the production of C/EBPb in liver of mice and rats (37), whereas insulin- and EGFincreased PRL-CAT expression require increased amounts of C/EBPa. Finally, the sequences through which insulin and EGF are active are different from the C/EBP-responsive sequences defined here. It seems more likely that the effects of C/EBPa on insulin- and EGF-increased PRL gene expression are a consequence of the effects of C/EBPa on basal PRLCAT expression. These studies have identified C/EBPa as a transcription factor that binds to the PRL promoter multihormonal response element (2101/292) to increase basal and insulinincreased PRL gene expression. Further, Chop10, which blocks C/EBPa binding to DNA, reduces basal and hormone-increased PRL-CAT expression. Thus, it is possible that physiological interaction of C/EBPa with this sequence is responsible for maintaining high levels of PRL gene transcription and explains the decrease in basal PRL-CAT expression when the 2101/292 sequence is mutated (5).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

4550

C/EBP REGULATION OF THE PRL GENE Acknowledgments

We thank S. L. McKnight, D. Ron, R. Treisman, and J. Whittaker for plasmids and antibodies used in these studies.

References 1. Cao Z, Baron EA, Carrillo AJ, Sharp ZD 1987 Reconstitution of cell-typespecific transcription of the rat prolactin gene in vitro. Mol Cell Biol 7:3402–3408 2. Gutierrez-Hartmann A, Siddiqui S, Loukin S 1987 Selective transcription and DNase I protection of the rat prolactin gene by GH3 pituitary cell-free extracts. Proc Natl Acad Sci USA 84:5211–5215 3. Schuster WA, Treacy MN, Martin F 1988 Tissue specific trans-acting factor interaction with proximal rat prolactin gene promoter sequences. EMBO 7:1721–1733 4. Day R 1998 Visualization of Pit-1 transcription factor interaction in the living cell nucleus by fluorescence resonance energy transfer microscopy. Mol Endocrinol 12:1410 –1419 5. Jacob KK, Stanley FM 1994 The insulin and cAMP response elements of the prolactin gene are overlapping sequences. J Biol Chem 269:25:515–525, 520 6. Peers B, Nalda AM, Monget P, Voz ML, Belayew A, Martial JA 1992 Binding of a 100 kDa ubiquitous factor to the human prolactin promoter is required for its basal and hormone-regulated activity. Eur J Biochem 210:53–58 7. Maurer RA, Notides AC 1987 Identification of an estrogen-responsive element from the 59-flanking region of the rat prolactin gene. Mol Cell Biol 7:4247– 4254 8. Day RN, Koike S, Sakai M, Muramatsu M, Maurer RA 1990 Both Pit-1 and the estrogen receptor are required for estrogen responsiveness of the rat prolactin gene. Mol Endocrinol 4:1964 –1971 9. Howard PW, Maurer RA 1995 A composite Ets/Pit-1 binding site in the prolactin gene can mediate transcriptional responses to multiple signal transduction pathways. J Biol Chem 270:20930 –20936 9a.Bradford AP, Conrad KE, Tran PH, Ostrowski MC, Gutierrez-Hartmann A 1996 GHF-1/Pit-1 functions as a cell-specific integrator of Ras signaling by targeting the Ras pathway to a composite Ets-1/GHF-1 response element. J Biol Chem 271:24639 –24648 10. Jacob KK, Wininger E, DiMini K, Stanley FM 1999 The EGF response element of the prolactin promoter. Mol Cell Endocrinol 152:137–145 11. Liang J, Kim KE, Schoderbek WE, Maurer RA 1992 Characterization of a nontissue-specific, 39,59-cyclic adenosine monophosphate-responsive element in the proximal region of the rat prolactin gene. Mol Endocrinol 6:885– 892 12. Peers B, Monget P, Nalda MA, Voz ML, Berwaer M, Belayew A, Martial JA 1991 Transcriptional induction of the human prolactin gene by cAMP requires two cis-acting elements and at least the pituitary-specific factor Pit-1. J Biol Chem 266:18127–18134 13. Jacob KK, Ouyang L, Stanley FM 1995 A consensus insulin response element of the prolactin, somatostatin, and thymidine kinase genes is ets-factor activated. J Biol Chem 270:27773–27779 14. Cao Z, Umek R, McKnight S 1991 Regulated expression of three C/EBP isoforms during adipose conversion of 3T3–L1 cells. Genes Dev 5:1538 –1552 15. Park EA, Gurney AL, Nizielski SE, Hakimi P, Cao Z, Moorman A, Hanson RW 1993 Relative roles of CCAAT/enhancer-binding protein b and cAMP regulatory element-binding protein in controlling transcription of the gene for phosphoenolpyruvate carboxykinase (GTP). J Biol Chem 268:613– 619 16. Ron D, Habener JF 1992 CHOP, a novel developmentally regulated nuclear protein that dimerizes with transcription factors C/EBP and LAP and functions as a dominant negative inhibitor of gene transcription. Genes Dev 6:439 – 453 17. Hemati N, Ross SE, Erickson RL, Groblewski GE, MacDougald OA 1997 Signaling pathways through which insulin regulates CAAT/enhancer binding protein a (C/EBPa) phosphorylation and gene expression in 3T3–L1 adipocytes. J Biol Chem 272:25913–25919

Endo • 1999 Vol 140 • No 10

18. Huang JH, Liao WSL 1994 Induction of the mouse serum amyloid A3 gene by cytokines requires both C/EBP family proteins and a novel constitutive nuclear factor. Mol Cell Biol 14:4475– 4484 19. Tae HJ, Luo X, Kim KH 1994 Roles of CCAAT/enhancer binding protein and its binding site on repression and derepression of acetyl-CoA carboxylase gene. J Biol Chem 269:10475–10484 20. Stanley FM 1988 Stimulation of prolactin gene expression by insulin. J Biol Chem 263:13:444 – 448 21. Stanley FM 1989 Transcriptional regulation of prolactin gene expression by thyroid hormone: alternate suppression and stimulation in different GH cell lines. Mol Endocrinol 3:1627–1633 22. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680 – 685 23. Gorski K, Carneiro M, Schibler U 1986 Tissue-specific in vitro transcription from the mouse albumin promoter. Cell 47:767–776 24. Demczuk S, Harbers M, Vennstrom B 1993 Identification and analysis of all components of a gel retardation assay by combination with immunoblotting. Proc Natl Acad Sci USA 90:2574 –2578 25. Samuels H, Stanley F, Casanova J 1979 Depletion of l-3,5,39-triiodothyronine and l-thyroxine in euthyroid calf serum for use in cell culture studies of the action of thyroid hormone. Endocrinology 105:80 – 85 26. Stanley F. M 1992 An element in the prolactin promoter mediates the stimulation of insulin on transcription of the prolactin gene. J Biol Chem 267:16719 –16726 27. Kageyama R, Sasai Y, Nakanishi S 1991 Molecular characterization of transcription factors that bind to the cAMP responsive region of the substance P precursor gene. cDNA cloning of a novel C/EBP related factor. J Biol Chem 266:15525–15531 28. Grove M, Plumb M 1993 C/EBP, NF-kB, and c-Ets family members and transcriptional regulation of the cell-specific and inducible macrophage inflammatory protein 1a immediate-early gene. Mol Cell Biol 13:5276 –5289 29. Ouyang L, Jacob KK, Stanley FM 1996 GABP mediates insulin-increased prolactin gene transcription. J Biol Chem 271:10425–10428 30. Schaufele F 1996 CCAAT/Enhancer-binding protein a activation of the rat growth hormone promoter in pituitary progenitor GHFT1–5 cells. J Biol Chem 271:21484 –21489 31. Keech C, Jackson S, Siddiqui S, Ocran K, Gutierrez-Hartmann A 1992 Cyclic Adenosine 39,59-monophosphate activation of the rat prolactin promoter is restricted to the pituitary-specific cell type. Mol Endocrinol 6:2059 –2070 32. Rajnarayan S, Chiono M, Alexander L, Gutierrez-Hartmann A 1995 Reconstitution of protein kinase A regulation of the rat prolactin promoter in HeLa nonpituitary cells: identification of both GHF-1/Pit-1-dependent and -independent mechanisms. Mol Endocrinol 9:502–512 33. Friedman AD, Landschulz WH, McKnight SL 1989 CCAAT/enhancer binding protein activates the promoter of the serum albumin gene in cultured hepatoma cells. Genes Dev 3:1314 –1322 34. Haas NB, Cantwell CA, Johnson PF, Burch JBE 1995 DNA-binding specificity of the PAR basic leucine zipper protein VBP partially overlaps those of the C/EBP and CREB/ATF families and is influenced by domains that flank the core basic region. Mol Cell Biol 15:1923–1932 35. Vallejo M, Gosse ME, Beckman W, Habener JF 1995 Impaired cyclic AMPdependent phosphorylation renders CREB a repressor of C/EBP-induced transcription of the somatostatin gene in an insulinoma cell line. Mol Cell Biol 15:415– 424 36. Park EA, Roesler WJ, Liu J, Klemm DJ, Gurney AL, Thatcher JD, Shuman J, Friedman A, Hanson RW 1990 The roles of the CCAAT/enhancer-binding protein in the transcriptional regulation of the gene for phosphoenolpyruvate carboxykinase (GTP). Mol Cell Biol 10:6264 – 6272 37. Bosch F, Sabater J, Valera A 1995 Insulin inhibits liver expression of the CCAAT/enhancer-binding protein b. Diabetes 44:267–271

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.