ERK-dependent and -independent pathways in bovine luteal cells

Molecular and Cellular Endocrinology 200 (2003) 141 /154 www.elsevier.com/locate/mce

Epidermal growth factor induces c-fos and c-jun mRNA via Raf-1/ MEK1/ERK-dependent and -independent pathways in bovine luteal cells Dong-bao Chen a,*, John S. Davis a,b,1 a

The Women’s Research Institute, Departments of Obstetrics and Gynecology, and Internal Medicine, University of Kansas School of Medicine-Wichita, 1010 North Kansas, Wichita, KS 67214, USA b The Research Service of the Department of Veterans Affairs Hospital, Wichita, KS 67218, USA Received 20 June 2002; accepted 30 September 2002

Abstract Epidermal growth factor (EGF) modulates the actions of gonadotropins in the corpus luteum. The membrane-associated EGF receptors undergo rapid tyrosine phosphorylation and internalization upon ligand binding in ovarian cells, including luteal cells. However, little is known about the post-receptor signaling events induced by EGF that lead to the transcriptional regulation of EGF-responsive genes in the ovary. The present study was designed to examine in bovine luteal cells (1) activation of the extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase (MAPK) signaling cascade (Raf/MEK/ERK) by EGF; (2) mRNA expression of AP-1 transcription factors, i.e. c-fos and c-jun, in response to EGF; and (3) the role of ERK in EGFinduced expression of c-fos and c-jun mRNA. Raf-1 and B-Raf, but not A-Raf, were activated by EGF (10 ng/ml) and the pharmacological protein kinase C (PKC) activator phorbol myristate acetate (PMA, 20 nM). Activation of Raf resulted in the phosphorylation and activation of MAPK kinase (MEK1) which subsequently activated ERKs. Treatment with EGF-induced the phosphorylation of both ERK2 and ERK1 in a time and concentration dependent manner. Additionally, activated ERK was found in the nucleus of the cells following treatment with EGF (10 ng/ml) and PMA (PMA, 20 nM) for 5 min. Depletion of PKC by chronic PMA treatment (2.5 mM, 24 h) only partially inhibited the stimulatory effects of EGF on Raf-1, ERK2 and ERK1. These data demonstrate that PKC-dependent and independent-mechanisms are involved in EGF activation of the Raf/MEK/ERK signaling cascade in bovine luteal cells. EGF rapidly and transiently stimulated the expression of c-fos and c-jun mRNA in bovine luteal cells. Maximal induction of c-fos and c-jun mRNA by EGF occurred within 30 min of treatment with 10 ng/ml EGF. Treatment with the MEK1 inhibitor PD098059 (50 mM) abolished EGF-induced ERK activation. However, blocking EGF-induced ERK activation by pretreatment with PD098059 only partially attenuated EGF-induced c-fos and c-jun mRNA expression. Thus, additional pathways are implicated in the regulation of c-fos and c-jun mRNA expression by EGF in bovine luteal cells. # 2002 Published by Elsevier Science Ireland Ltd. Keywords: Epidermal growth factor; Raf; Mitogen-activated protein kinase; Protein kinase C; c-fos/c-jun; Corpus luteum

1. Introduction

* Corresponding author. Address: Division of Maternal-Fetal Medicine, Department of Reproductive Medicine, School of Medicine, University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0802, USA. Tel.:
/1-619-5437279; fax:
/1-6195432919. E-mail addresses: [email protected] (D.-b. Chen), [email protected] (J.S. Davis). 1 Present address: Olson Center for Women’s Health, Departments of Obstetrics and Gynecology, 983255 Nebraska Medical Center, Omaha, NE 68198-3225, USA. Tel.:
/1-402-559-9079.

Epidermal growth factor (EGF) and the EGF family of growth factors as well as their receptors (EGFR) have been demonstrated to be important intraovarian modulators (Ackland et al., 1992). EGF, like other growth factors, can regulate diverse cellular responses including proliferation, differentiation and many metabolic processes (Carpenter and Cohen, 1990). The follicular granulosa cells are well known to possess EGFR (StArnaud et al., 1983; Ayyagari and Khan-Dawood, 1987;

0303-7207/02/$ - see front matter # 2002 Published by Elsevier Science Ireland Ltd. PII: S 0 3 0 3 - 7 2 0 7 ( 0 2 ) 0 0 3 7 9 - 9

142

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

Maruo et al., 1993; Scurry et al., 1994). Upon exposure to EGF, granulosa cells can be stimulated in vitro to proliferate (May et al., 1992), and increase the binding affinity of follicle-stimulating hormone (May et al., 1987). In granulosa cells, EGF stimulates aromatase activity and estrogen production (Gangrade et al., 1991; Misajon et al., 1999) in some species while suppressing estrogen production in others (Murray et al., 1993). Following ovulation, follicular cells differentiate into luteal cells. Luteal steroidogenic cells express EGF (Tekpetey et al., 1995; Singh et al., 1995) and EGFR (Tekpetey et al., 1995; Singh et al., 1995; Huang et al., 1995). EGFR are present in both large and small porcine luteal cells (Singh et al., 1995). Additionally, bovine luteal cell membranes possess specific, high affinity, and saturable EGFR (Budnik and Mukhopadhyay, 1991, 1996). The presence of EGF and EGFR in luteal cells implies that EGF may be important for autocrine and paracrine regulation of luteal function (Knecht and Catt, 1983; Davis et al., 1996; Pescador et al., 1999). Interestingly, luteal cells, compared to their progenitors, lose the ability to proliferate in response to EGF (Vlodavsky et al., 1978). The EGFR is a member of the ErbB family of receptors that consist of a cystine-rich extracellular domain, a single membrane-spanning domain, and a large intracellular domain containing a tyrosine kinase domain. Upon ligand binding, EGFR dimerizes and autophosphorylates on several tyrosine residues in the intracellular domain, which serve as docking sites for several physiological substrates (Carpenter and Cohen, 1990). A number of these effector proteins contain sequence motifs termed src homology 2 domains that couple to different intracellular signaling systems, including the adaptor proteins Shc, Grb2 and the Ras guanine nucleotide exchange factor Sos (Sasaoka et al., 1994), phospholipase Cg (Margolis et al., 1990; Rotin et al., 1992), some phosphotyrosine phosphatases (Liu and Chernoff, 1997), and phosphatidylinositol 3-kinase (Habib et al., 1998). Among these effector proteins, the adaptor proteins have been demonstrated to activate the p21Ras-directed mitogenic signal transduction pathway, which transmits growth and differentiation signals of EGFR into the nucleus via activation of mitogenactivated protein kinase (MAPK) (Liebmann, 2001). MAPK cascades are evolutionary conserved signal transduction modules that are used in a wide variety of biological responses. In vertebrates, multiple isoforms of MAPK have been identified and categorized into three subfamilies, i.e. the extracellular signal-regulated kinases (ERKs), p38mapk, and the Jun N-terminal kinases (JNKs) or stress-activated protein kinases. The ‘classical’ ERKs, ERK2 or p42mapk and ERK1 or p44mapk, are positioned downstream of Raf-1 and MEK1, and together comprise an orderly signaling cascade in response to a variety of extracellular stimuli including

growth factors (Davis, 1993). The MAPK signaling pathways have been implicated in control of different and even opposite cellular responses including proliferation, differentiation and cell death. Such actions are elicited at least in part through translocation of activated MAPK into the nucleus (Chen et al., 1992), where it phosphorylates and thereby activates nuclear transcription factors, including Elk-1 (Gille et al., 1992), Sap1 (Dalton and Treisman, 1992), c-Jun (Rozek and Pfizer, 1993) and ATF2 (Gille et al., 1992; Rozek and Pfizer, 1993). These transcription factors stimulate the expression of the immediate-early response oncogenes, i.e. c-fos (Gille et al., 1992) and c-jun (Rozek and Pfizer, 1993). Fos and Jun proteins are constituents of the activator protein-1 (AP-1) transcription factors that in turn regulate the transcription of numerous genes possessing promoter AP-1 binding sites (Karin et al., 1997; Shaulian and Karin, 2002). Upon ligand binding, the EGFR on isolated corpus luteum cell membranes undergoes rapid tyrosine phosphorylation (Budnik and Mukhopadhyay, 1996; Lamm et al., 1999) and internalization (Budnik and Mukhopadhyay, 1996). Activated EGFR may associate with various docking molecules and subsequently activate their corresponding downstream signaling events in luteal cells. It has been previously demonstrated that EGF activates the ERK MAPKs in porcine granulosa cells (Keel et al., 1995; Cameron et al., 1996). However, the signaling pathways that stimulate gene expression by EGF in ovarian cells have not been defined. In the present study we tested the hypothesis that EGF stimulation of the Raf/MEK1/ERK MAPK signaling cascade mediates the expression of the early response genes c-fos and c-jun in bovine luteal cells. We also tested whether the response to EGF requires the involvement of protein kinase C (PKC). Our results demonstrate that EGF activates a Raf-1/MEK1/ERK MAPK cascade via PKC-dependent and independentmechanisms. We also report that EGF stimulates the expression of c-fos and c-jun oncogenes by ERKdependent and ERK-independent mechanisms in bovine luteal cells.

2. Materials and methods 2.1. Chemicals and reagents Recombinant mouse EGF and Protein-A conjugated agarose beads were obtained from Upstate Biotechnology, Inc. (Lake Placid, CA). Rabbit polyclonal anti-Raf1, conjugated agarose beads, rabbit polyclonal antiERK2 and ERK1 conjugated agarose beads, and rabbit polyclonal anti-Raf-1, A-Raf and B-Raf antibody were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Monoclonal anti-PanERK antibodies were

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

from Transduction Laboratories (Lexington, KY). Active MAPK antibody was from Promega (Madison, WI). Recombinant [His]6-tagged-MEK1 (MEK1) and recombinant [His]6-tagged-K52R (K52R, a full length but catalytically inactive ERK2) were from Dr Dent (Dent et al., 1995). Recombinant GST-Elk-1 was a gift from Dr Roberson (Roberson et al., 1995). The rabbit polyclonal anti-3b-hydroxysteroid dehydrogenase (3bHSD) antibody was provided by Dr Mason. [g-32P]ATP (37 MBq, 10 mCi/ml) and [a-32P]-dCTP (37 MBq, 3000 Ci/mM) were from DuPont (Boston, MA). Tissue culture plasticware was from Corning (Corning, NY). Fetal calf serum (FCS), 0.24 /9.5 kb RNA ladder, medium-199 (M-199) were from Life Technologies Inc. (Grand Island, NY). Bio-Rad protein reagents (500 / 0006), electrophoresis reagents and Zeta probe blotting membrane were from Bio-Rad Laboratories (Richmond, CA). Enhanced chemiluminesence (ECL) kits were from Armersham (Arlington Heights, IL). Immobilon-P (PVDF) membrane was from Millipore (Beldford, MA). Random primed labeling kit was from Boehringer Mannheim Co. (Indianapolis, IN). 1.8 kb human b-actin cDNA was purchased from Clontech (Palo Alto, CA). The MEK-1 inhibitor PD098059 was purchased from Calbiochem (La Jolla, CA). Prestained protein molecular marker, and all other chemicals and reagents were from Sigma (St. Louis, MO).

143

7.4, 100 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton-100, 0.5% NP-40, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF, 10 mg/ml of each leupeptin, aprotinin and pepstatin-A) on ice with continuous shaking for 15 min. After clarification by centrifugation (5 min, 15 000/g), the protein content of the samples was measured by a Bio-Rad procedure using BSA as the standard. Aliquots of the extracts were subjected to immunoprecipitation or were added in Laemmli buffer and frozen at /80 8C until western blot analysis could be performed. 2.3. Western blot analysis After SDS-PAGE size-fractionation, proteins were electrically (100 V, 1.5 h) transferred to a PVDF membrane. Non-specific binding was blocked with 5% fat-free milk in TBST (50 mM Tris /HCl, pH 7.5, 0.15 M NaCl, 0.05% Tween-20) overnight at 4 8C, after which the membrane was incubated with appropriate amounts of primary antibodies in TBST-3% BSA (for monoclonal antibodies) or in 5% fat-free milk in TBST (for polyclonal antibodies) at room temperature for 1 h. Following four washes (5 min) with TBST, the membrane was incubated with anti-mouse or anti-rabbit peroxidase conjugated IgG for 1 h, respectively. The membrane was washed four times with TBST, and bound antibodies were detected with the ECL reagents according to the manufacturer’s instruction.

2.2. Cell isolation, culture, experimental conditions and total cell extracts

2.4. Immunoprecipitation

Bovine ovaries were collected during early pregnancy (fetal crown rump length B/15 cm) from a local slaughterhouse and transported to the laboratory in cold M-199. The luteal tissue was dissociated with collagenase as described previously (Chakravorty et al., 1993; Chen et al., 1998, 2001). The cell viability was determined by trypan blue test. The luteal cell preparations with /90% viability were used and plated in standard tissue culture plasticware in M-199 with 5% FCS for 18 h at 37 8C in an humidified atmosphere of 5% CO2 in air. The cells were serum-starved for 48 h prior to experiments by replacing medium with serumfree M-199 supplemented with 0.1% bovine serum albumin (BSA, Fraction V) and insulin (5 mg/ml). Following incubation for 24 h, some groups of luteal cells were treated for an additional 24 h with 2.5 mM PMA to obtain the PKC-deficient cell model as described previously (Chen et al., 2001). Prior to treatment, the media were changed and cultures were used following a 1/3 h equilibration with fresh M-199 plus 0.1% BSA. After cell stimulation as stated in the figure legends, the cells were rapidly rinsed twice with cold phosphate-buffered saline (PBS). Total cell extracts were prepared by lysing the cells with a non-denaturing lysis buffer A (10 mM Tris /HCl, pH

Cell extracts (200 mg total protein) were cleared by coincubation with 1 ml normal goat serum and 20 ml Protein-A conjugated agarose beads for 1 h at 4 8C. After centrifugation (15 000 /g , 4 8C, 2 min), the supernantants were transferred to new tubes. Immunoprecipitations were then performed in the cleared cell extracts at 4 8C as described below. For Raf-1 immunoprecipitation, cell extracts were incubated with PBS-washed rabbit polyclonal anti-Raf-1 conjugated agarose beads (5 mg) in lysis buffer A with continuous rotation for 2 h. The beads (immuno-complex) were collected by centrifugation. Following 3 washes with buffer A, 3 washes with LiCl solution (0.5 M LiCl, 100 mM Tris /HCl, pH 8.0) and 2 washes with kinase assay buffer B (50 mM b-glycerophosphate, pH 7.3, 1.5 mM EGTA, 1 mM DTT and 0.09% Brig35), the beads were resuspended in 40 ml of kinase buffer B. For A-Raf and B-Raf immunoprecipitation, cell extracts were incubated with rabbit polyclonal anti-A-Raf and rabbit polyclonal anti-B-Raf antibodies (5 mg) in buffer A overnight with continuous rotation. Protein-A conjugated agarose beads (20 ml) were then added and the samples were incubated for an additional 1 h with continuous rotation. The immuno-complexes were collected and washed as described for Raf-1 immunopre-

144

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

cipitation, and finally resuspended in 40 ml of kinase buffer B. For ERK2 and ERK1 immunoprecipitation, cell extracts were incubated with anti-ERK2 or ERK1 antibody conjugated agarose beads in lysis buffer A with continuous rotation for 2 h. The immuno-complex was collected, washed and resuspended as described for Raf1 immunoprecipitation. Aliquots of Raf-1, A-Raf, BRaf, ERK2 and ERK1 immunoprecipitates were prepared in Laemmli buffer for subsequent western blot analysis to verify that the same amounts of immunoprecipitates were used in the kinase assay (data not shown). 2.5. Immuno-complex kinase assay Raf kinase activity was measured by a two-step immuno-complex kinase assay by its ability to incorporate [g-32P]-ATP into recombinant MEK1, and phosphorylated MEK1 to subsequently phosphorylate K52R as described (Dent et al., 1995; Chen et al., 1998). Briefly, 15 ml of Raf-1, A-Raf or B-Raf immunoprecipitates was incubated with 10 mCi [g-32P]-ATP, 30 mM ATP in buffer C (20 mM Tris /HCl, pH 7.4, 0.5 mM MnCl2, 5 mM MgCl2) in the presence of 1 mg MEK1 (final volume 20 ml) at 30 8C for 20 min. The reaction mixtures were chilled on ice for 5 min and then centrifuged (15 000 /g , 5 min). A portion of the supernatant (15 ml) was transferred to a new reaction tube containing 0.1 mg K52R, 5 mCi [g-32P]-ATP and 30 mM ATP in buffer C (final volume 20 ml), and incubated at 30 8C for 10 min. The reaction was stopped by the addition of Laemmli buffer and proteins were sizefractionated on 10% SDS-PAGE. Following commassie blue staining, the gels were dried. Autoradiography was performed on dried gels to visualize phosphorylated MEK1 and K52R bands. The MEK1 and K52R bands were cut out and quantified by liquid scintillation counting. ERK MAPK activity was measured by in vitro phosphorylation of GST-Elk-1 as described (Chen et al., 1998). Briefly, ERK2 or ERK1 immunoprecipitates (20 ml) were incubated with 10 mCi [g-32P]-ATP, 30 mM ATP, and 0.5 mg GST-Elk-1 in buffer C for 30 min at 30 8C. The reaction was stopped by addition of Laemmli buffer, and analyzed by 10% SDS-PAGE and autoradiography to visualize phosphorylated GST-Elk1. Elk-1 bands were cut out and 32P incorporation was quantified by liquid scintillation counting. 2.6. Phosphorylation status of ERK2 and ERK1 The phosphorylation status of ERK2 and ERK1 was assessed by a gel-shift assay based upon the slower migration rate of the phosphorylated active forms of ERK2 and ERK1 than the non-phosphorylated inactive forms on SDS-PAGE gel (Chen et al., 1998). The

phosphorylated active forms and non-phosphorylated inactive forms of ERK2 and ERK1 were detected by western blot analysis with the PanERK antibodies. Phosphorylation of ERK2 and ERK1 was also measured by western blot analysis with the active MAPK antibody which recognizes phosphorylated ERK2 and ERK1 on both tyrosine and threonine residues. 2.7. Fluorescence immunocytochemistry Luteal cells were cultured in 8-well Lab-Tek glass chamber slides as described above. Following treatment with EGF (10 ng/ml) and PMA (20 nM) for 5 min, the cells were washed with PBS, and fixed with cold methanol for 10 min at /20 8C. The cells were air dried, washed with 0.1% BSA /TBS (10 mM Tris /HCl, pH 7.4, 150 mM NaCl), and permeabilized with 0.02% Triton-100 in 0.1% BSA /TBS for 10 min. Non-specific binding was blocked by incubation with 3% BSA /TBS for 30 min at room temperature, after which the active MAPK antibody (5 mg/ml) or rabbit polyclonal anti-3bHSD antibody (1:100) diluted in 0.1% BSA /TBS was applied and incubated overnight at 4 8C. Following three washes (10 min each) with TBS, the samples were incubated with Cy3-conjugated anti-rabbit IgG (1:100) for 1 h at room temperature. Following four washes (5 min each) with 0.1% BSA /TBS, the samples were examined by confocal microscopy Bio-Rad (Richmond, CA) MRC-1024 Laser Scanning Confocal Imaging System using an argon/krypton lamp. Digital imaging was captured by the Bio-Rad laser-shop software. For negative controls, the same procedure as described above was performed either without primary antibody or without secondary antibody. 2.8. RNA extraction and northern blot analysis Following cell stimulation, the medium was discarded. Total RNA was isolated by the single-step guanidine /thiocyanate /phenol /chloroform procedure (Chomczynski and Sacchi, 1987), and quantified based upon the absorbance value at 260 nM by UV spectrophotometer. RNA samples (15 mg/lane) were electrophoresed on a denaturing gel of 1% agarose and 1.5% formaldehyde, transferred and UV cross-linked to a Zeta probe blotting membrane. The membrane was prehybridized in 500 mM NaHPO4, pH 7.2, 7% SDS and 1 mM EDTA, at 60 8C for 4 h. Hybridization was performed in the same buffer containing denatured [32P]-labeled bovine c-fos (GeneBank accession number #AF069515) and c-jun (GeneBank accession number #AF069514) cDNA probes (/2.5 /106 cpm/ml) overnight at 60 8C. The cDNAs were random-primed radiolabeled with [a-32P]-dCTP. Following four washes (20 min each) in prewarmed (60 8C) buffer (40 mM NaHPO4, pH 7.2, 1% SDS, 1 mM EDTA, pH 8.0) at

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

145

room temperature, hybridization signals were detected by autoradiography. The membranes were stripped and reprobed with the human b-actin cDNA probes as a control for sample loading. 2.9. Statistical analysis Experiments were replicated using cell preparations collected from three or more corpora lutea, each from separate animals. Data are presented as mean9/S.E.M and were analyzed by the InStat program. One way analysis of variance was followed by Dunn’s t-test. P B/ 0.05 were considered statistically significant.

3. Results 3.1. Activation of Raf oncoproteins in response to EGF Raf activation was measured in a two-step immunocomplex assay. The first step of the assay employs endogenous Raf immuno-complexes and catalytically active recombinant MEK1. The second step involves the MEK1 derived from the first step and a full length but catalytically inactive ERK2 (K52R) to assess MEK1 activation upon exposure to Raf (Chen et al., 1998). Phosphorylation of K52R by Raf was mediated by MEK1, because Raf immunoprecipitates were incapable of phosphorylating K52R in the absence of exogenous MEK1. This data suggests an orderly sequence in the activation of Raf, MEK1 and ERK2, as shown in Fig. 1a and previously (Dent et al., 1995; Chen et al., 1998). We have previously reported that all three Raf proteins (A-Raf, B-Raf and Raf-1) are present in bovine luteal cells (Chen et al., 1998). Treatment with EGF did not significantly alter the kinase activity of A-Raf in ARaf immunoprecipitates prepared from bovine luteal cells (Fig. 1). Conversely, compared to untreated control cells, we observed that treatment with EGF for 5 min provoked 6-fold increase (P B/0.01) in the phosphorylation of MEK1 (Raf kinase activity) and 7-fold increase in the phosphorylation of K52R (MEK1 kinase activity) in B-Raf immunoprecipitates (Fig. 1b). Furthermore, compared to untreated control cells, treatment with EGF for 5 min provoked a 12-fold increase (P B/0.01) in the phosphorylation of MEK1 (Raf kinase activity) and a 13-fold increase in the phosphorylation of K52R (MEK1 kinase activity) in Raf-1 immunoprecipitates (Fig. 1b). 3.2. Phosphorylation of ERK2 and ERK1 in response to EGF We have previously demonstrated that phosphorylation of ERK2 and ERK1 in luteal cells can be measured by Western blot analysis with a PanERK antibody

Fig. 1. Activation of A-Raf, B-Raf and Raf-1 by EGF in bovine luteal cells. Bovine luteal cells were washed and equilibrated with fresh M199-0.1% BSA for 1 h, then treated with EGF (10 ng/ml) for 5 min. Total cell lysates were made with a cold non-denaturing buffer. The enzymatic activities of immunoprecipitated A-Raf, B-Raf and Raf-1 from cell lysates (200 mg protein) were then measured by [32P]-ATP incorporation into [His]6-tagged-MEK1 (1 mg) and its subsequent phosphorylation of [His]6-tagged-K52R (0.1 mg) by the 2-step immuno-complex kinase assay as described in Section 2. Panel (a) typical autoradiograms representing one of three independent experiments with each of A-Raf, B-Raf and Raf-1 immunoprecipitates are shown. Lane 1: Raf immunoprecipitates cannot phosphorylate K52R in the absence of MEK1. Lane 2: MEK1 phosphorylates K52R in the absence of cellular protein. Lanes 3 /4: Raf-initiated phosphorylation of MEK1 and K52R in Raf immuno-complexes isolated from cells treated with control media (lane 3), EGF (10 ng/ml) (lane 4). Panel (b) Data from three independent experiments are expressed as means9/ SEM. In each immuno-complex assay the levels of 32P-ATP incorporation into MEK1 and K52R were obtained after subtraction of their corresponding levels of phosphorylation in the absence of Raf immuno-complexes. Autophosphorylation of MEK1 was 1399/25 cpm and phosphorylation of K52R in the presence of MEK1 was 2219/37 cpm. Significant differences in the levels of K52R phosphorylation in Raf immuno-complexes from control and EGF treated cells are shown by horizontal lines. Letters on the top of the bars (a vs. b) represent significant differences (P B/0.05) in the levels of MEK1 phosphorylation in Raf immuno-complexes.

based on the reduced migration rate (mobility shift) of the phosphorylated active forms compared to the nonphosphorylated inactive forms in SDS-PAGE (Chen et al., 1998). Upon EGF stimulation, a mobility shift of ERK2 and ERK1 rapidly occurred in luteal cells (Figs. 2 and 3). We obtained optimal separation of the phosphorylated and non-phosphorylated forms by using 10% SDS-PAGE with an acrylamide:bisacrylamide ratio of 29.8:0.2. Western blot analysis revealed that ERK2 and ERK1 were the major EGF-responsive isoforms of the ERK subfamily of MAPK in bovine luteal cells.

146

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

Fig. 2. Concentration-dependence of EGF-induced phosphorylation of ERK2 and ERK1 in bovine luteal cells. Bovine luteal cells were treated with increasing concentrations (0 /100 ng/ml) for 5 min and lysed in a non-denaturing buffer. Total cell lysates (20 mg/lane) were size-fractionated by 10% SDS-PAGE, electrically transferred on to a PVDF membrane, and analyzed by Western blotting with the PanERK antibody (1:500, panel a) or phospho-MAPK antibody (1:1000, panel b). A ‘gel-shift’ on SDS-PAGE demonstrating ERK1 and ERK2 activation was shown in upper panel with longer exposure (45 s) and lower panel with shorter exposure (15 s) in panel a, respectively. Results from one of three independent experiments are shown.

Additionally, the levels of ERK2 were greater than the levels of ERK1, as shown by western blot analysis with both the PanERK antibody (Fig. 2aFig. 3a) and the active MAPK antibody (Fig. 2bFig. 3b). Additionally, two unknown proteins with molecular weights of approximate 85 and 56 kDa were also observed in PanERK immunoblots. These two proteins were not altered in response to EGF and other ligands as we have showed previously (Chen et al., 1998). Kinetic studies showed that EGF stimulated the mobility shift of ERK2 (upper panel) and ERK1 (lower panel) in both concentration- and time-dependent patterns. When luteal cells were treated with increasing concentrations (0.01 /100 ng/ml) of EGF for 5 min, maximal effects of EGF on the mobility shift of ERK2 (upper panel) and ERK1 (lower panel) were observed with 1/10 ng/ml of EGF (Fig. 2a). The stimulatory effects of EGF (10 ng/ml) on the mobility shift of ERK2 and ERK1 in bovine luteal cells peaked at 5 /10 min following treatment, remained high up to 60 min, and then returned to basal levels after 120 min (Fig. 3a).

Fig. 3. Time-course of EGF-induced phosphorylation of ERK2 and ERK1 in bovine luteal cells. Bovine luteal cells were treated with 10 ng/ ml of EGF for various times and lysed in a non-denaturing buffer. Total cell lysates (20 mg/lane) were size-fractionated by 10% SDSPAGE, electrically transferred on to a PVDF membrane, and analyzed by Western blotting with the PanERK antibody (1:500, panel a) or phospho-MAPK antibody (1:1000, panel b). A ‘gel-shift’ on SDSPAGE demonstrating ERK1 and ERK2 activation was shown in upper panel with longer exposure (45 s) and lower panel with shorter exposure (15 s) in panel a, respectively. Results from one of three independent experiments are shown.

Activation of ERK MAPK results from the phosphorylation of threonine/tyrosine residues in the conserved ‘TEY’ motif (Davis, 1993). When analyzed with the phospho-specific MAPK antibody, we observed similar profiles of time-courses and concentration-responses of EGF-induced phosphorylation of ERK2 and ERK1 as compared to those of EGF-induced mobility shift of ERK2 and ERK1. As shown in Fig. 2b, dose/ response studies revealed that EGF-induced phosphorylation of ERK2 and ERK1 was maximal when the cells were treated with 1 /10 ng/ml of EGF for 5 min. Timecourse studies (Fig. 3b) revealed that EGF-induced phosphorylation of ERK2 and ERK1 was maximal at 2 /30 min, declined at 60 min and returned to basal levels at 120 min after treatment with EGF (10 ng/ml). A similar time- and concentration-dependent changes in mobility shift (Fig. 2aFig. 3a) and phosphorylation (Fig. 2bFig. 3b) of ERK2 and ERK1 indicates that the mobility shift of ERK2 and ERK1 indeed correlates to the phosphorylation of ERK2 and ERK1.

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

3.3. PKC is required for EGF-induced activation of Raf1, ERK2 and ERK1 PKC plays an important role in the regulation of corpus luteum function as exemplified by the decrease in progesterone secretion following infusion of PMA (McGuire et al., 1994). We have recently, demonstrated that activation of PKC by acute treatment with PMA and PGF2a stimulates Raf-1 (Chen et al., 1998); and PKC mediates the activation of ERK MAPK in response to PGF2a (Chen et al., 2001). Accordingly, we examined the role of PKC in EGF activation of the Raf-1-MEK1-ERK signaling cascade using the PKCdeficient luteal cell model established with chronic treatment with 2.5 mM PMA for 24 h (Chen et al., 2001). Compared to untreated control cells, treatment with EGF (10 ng/ml) for 5 min significantly (P B/0.01) stimulated the activities of Raf-1 (phosphorylation of MEK1) and MEK1 (phosphorylation of K52R) by 9.9and 11.8-fold, respectively. Depletion of PKC by chronic treatment with PMA (2.5 mM, 24 h) did not significantly alter basal Raf-1 or MEK1 activity. Compared to untreated control cells, in PKC-deficient cells EGF only provoked a 4.3- and 3.9-fold increase in MEK1 phosphorylation (Raf-1 activity) and K52 phosphorylation (MEK1 activity), respectively (Fig. 4a).

147

These results indicate that the activation of Raf-1 by EGF is partially inhibited in the absence of PKC. Using the PKC-deficient luteal cell model, experiments were also performed to determine the role of PKC in ERK2 and ERK1 activation in response to EGF. Western blot analysis with the active MAPK antibody showed that treatment with EGF and PMA both stimulated the phosphorylation of ERK2 and ERK1. In keeping with our previous report (Chen et al., 2001), Fig. 4b showed that PKC was required for PMAinduced phosphorylation of ERK2 and ERK1 because it was abolished in the PKC-deficient cells. ERK activation in response to EGF is much less dependent upon PKC than is ERK activation in response to PGF2a and PMA in bovine luteal cells (Chen et al., 1998, 2001) because EGF-induced ERK phosphorylation was only partially attenuated in the PKC-deficient cells (Fig. 4b). 3.4. Activation of ERK2 and ERK1 in response to EGF Activated MAPK translocates into the nucleus to regulate gene expression, in part, by the ability of MAPK to phosphorylate and activate nuclear substrates including the Ets family transcription factor such as Elk1 (Davis, 1993). We employed the transcription factor Elk-1 as a specific substrate to determine the activities of ERK2 and ERK1 by immuno-complex kinase assays. EGF significantly (P B/0.01) stimulated Elk-1 phosphorylation in both ERK2 and ERK1 immunoprecipitates. Treatment with EGF (10 ng/ml) for 5 min increased the activities of ERK2 and ERK1 by 11.2and 9.9-fold, respectively. Activation of PKC by PMA also increased the activities of ERK2 and ERK1. Treatment with PMA (20 nM) for 5 min stimulated Elk-1 phosphorylation by 4.3- and 5.5-fold in ERK2 and ERK1 immunoprecipitates, respectively (Fig. 5). 3.5. ERK translocates into the nucleus upon EGF stimulation

Fig. 4. PKC is involved in activation of Raf-1, MEK1 and ERK2/1 in response to EGF in bovine luteal cells. Bovine luteal cells were incubated for 24 h, then 2.5 mM PMA was added for an additional 24 h to deplete phorbol ester-responsive isoforms of PKC, then challenged with EGF (10 ng/ml) or PMA (20 nM) for 5 min. Total cell lysates were prepared with a non-denaturing buffer. Panel (a) Raf-1 immunocomplex kinase assay as described in Fig. 1. The fold induction of MEK1 and K52R phosphorylation was included in the corresponding lanes of each assay. Panel (b) Western blot analysis with the active MAPK antibody as described in Fig. 3. The results represent one of three similar independent experiments.

Upon stimulation, activated MAPK rapidly translocates into the nucleus (Chen et al., 1992) where it phosphorylates and thereby activates nuclear substrates to regulate gene transcription (Davis, 1993). To test whether this phenomenon is also associated with ERK activation upon EGF stimulation in bovine luteal cells, we localized activated ERK by indirect fluorescence immunocytochemistry using active MAPK antibody. The active MAPK signal observed either in the cytoplasm or the nucleus of untreated control cells was weak. Following treatment with EGF (10 ng/ml) for 5 min, a very strong nuclear labeling of activated MAPK was induced while the cytoplasmic signals of activated MAPK labeling remained weak (Fig. 6D). As expected, treatment with PMA (20 nM) for 5 min to activate PKC

148

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

Fig. 5. Activation of ERK2 and ERK1 in response to EGF and PMA in bovine luteal cells. Bovine luteal cells were treated with EGF (10 ng/ ml) and PMA (20 nM) for 5 min, lysed in a non-denaturing buffer. Total cell lysates (200 mg protein) were subjected to ERK2 and ERK1 immunoprecipitation as described in Section 2. The ERK2 and ERK1 immuno-complexes were incubated with GST-Elk-1 (0.5 mg) and [g-32P]-ATP for 20 min at 30 8C. The reaction mixtures were then analyzed with 10% SDS-PAGE followed by autoradiography. 32P incorporation into GST-Elk-1 was quantified by liquid scintillation counting of the corresponding bands. (a) Typical autoradiograms represent one of three independent experiments with each of ERK2 and ERK1 immunoprecipitates are shown. Lane 1: GST-Elk-1 does not have auto kinase activity. Lane 2 /4: ERK immunoprecipitates initiate GST-Elk-1 phosphorylation. (b) Data from three independent experiments are expressed as means9/SEM. Differences in the levels of Elk-1 phosphorylation in ERK2 immuno-complexes from control and treated cells are shown by horizontal lines. Letters on the top of the bars (a vs. b) represent significant differences (P B/0.05) in the levels of Elk-1 phosphorylation in ERK1 immuno-complexes from control and treated cells.

also induced a strong nuclear signal of active MAPK (Fig. 6C, Chen et al., 2001). In parallel, 3b-HSD immunolabeling was primarily present in the cytoplasm of untreated control cells (Fig. 6B). The majority of the cells were 3b-HSD positive cells, demonstrating their steroidogenic nature. 3.6. Expression of c-fos and c-jun mRNA in response to EGF The AP-1 family of transcription factors, e. g. the gene products of the c-fos and c-jun families of protooncogenes, plays important roles in regulating gene expression (Karin et al., 1997). In order to explore the role of AP-1 transcription factors in the ovary, we examined the effects of EGF on induction of fos and jun

Fig. 6. Subcellular localization of active ERK in response to EGF in bovine luteal cells. Bovine luteal cells were cultured on 8-well Lab-Tek glass chamber slides. Following serum-starvation, the cells were treated with EGF (10 ng/ml) and PMA (20 nM) for 5 min, washed twice with cold-PBS, and fixed with methanol (/20 8C) for 10 min. Following incubation with 3% BSA /TBS for 30 min, the cells were incubated with active MAPK antibody (5 mg/ml) or polyclonal rabbit anti-3b-HSD antibody (1:100) in 0.1% BSA /TBS overnight at 4 8C. After four washes with 0.1% BSA /TBS, the cells were incubated with Cy3-conjugated anti-rabbit IgG (1:100). The cells were washed and examined by confocal microscopy (400 /). The labeling patterns of active MAPK in untreated cells, PMA- and EGF-treated cells are shown in Panel A, C, and D, respectively. Panel B shows cells labeled with anti-3b-HSD antibodies. The results represent one of three similar independent experiments.

mRNA expression in bovine luteal cells. As measured by Northern analysis, the basal levels of c-fos mRNA was almost undetectable (Lane 1 in Fig. 7aFig. 8a). In contrast, the basal levels of c-jun mRNA were higher than that of c-fos in cultured primary bovine luteal cells (Lane 1 in Fig. 7aFig. 8a). A single c-fos transcript about 2.37 kb (Fig. 7aFig. 8a) and two c-jun transcripts with a major band about 3.10 kb and a minor band about 3.25 kb (Fig. 7bFig. 8b) were found in bovine luteal cells. The sizes for c-fos and c-jun transcripts were the same as described in previous reports by others in a number of cells and tissues (Li et al., 1997). EGF is a potent inducer of c-fos and c-jun mRNA expression in many cell systems (Stass and Mixson, 1997). We observed that EGF dramatically stimulated cfos and c-jun mRNA expression in bovine luteal cells. Treatment with increasing concentrations of EGF (0.01/100 ng/ml) for 30 min provoked concentrationdependent increases in the levels of c-fos (Fig. 7a) and cjun (Fig. 7b) mRNA. As little as 0.01 ng/ml EGF elevated the levels of c-fos mRNA, and following treatment with 1 ng/ml EGF c-fos mRNA levels reached nearly maximal levels. As little as 0.1 ng/ml EGF elevated the levels of c-jun mRNA, and treatment with

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

Fig. 7. Concentration-dependent c-fos and c-jun mRNA expression in response to EGF in bovine luteal cells. Bovine luteal cells were treated with increasing concentrations of EGF (0.01 /100 ng/ml). Total RNA was extracted. The RNA samples (15 mg/lane) were size-fractionated on formaldehyde */1% agarose gel and Northern blot analysis was performed with [a-32P] dCTP labeled bovine specific c-fos (panel a) and c-jun (panel b) cDNA probes as described in Section 2. The membranes were stripped and reprobed with human b-actin cDNA probe (panel c) as an internal control. The results represent one of three similar independent experiments.

Fig. 8. Time-course of c-fos and c-jun mRNA expression in response to EGF in bovine luteal cells. Bovine luteal cells were treated with EGF (10 ng/ml) up to 240 min. Total RNA was extracted, sizefractionated on formaldehyde-1% agarose gel, transferred, and crosslinked on a Zeta-probe membrane. Northern blot analysis was performed to determine mRNA levels of c-fos (panel a) and c-jun (panel b) as described in Fig. 7. The membranes were stripped and reprobed with human b-actin cDNA probe as an internal control (panel c). The results represent one of three similar independent experiments.

10 ng/ml EGF elevated c-jun mRNA levels to maximal levels. Treatment with EGF did not significantly alter bactin mRNA, which were served as the internal control for sample loading of Northern analysis of c-fos and cjun mRNA (see lower panels of Figs. 7/9c). The genes encoding the Fos and Jun families of transcription factors are ‘immediate-early’ response

149

Fig. 9. ERK is required for increased c-fos and c-jun mRNA expression in response to EGF in bovine luteal cells. Panel (a) PD098059 blocks ERK phosphorylation in response to EGF. Bovine luteal cells were pretreated with 20 or 50 mM PD098059 for 1 h as indicated in the figure, then treated with EGF (10 ng/ml) for 5 min. Total cell lysates were made with a non-denaturing buffer. Phosphorylation of ERK2 and ERK1 was determined by immunoblotting with the active MAPK antibody as described in Fig. 3. The results represent one of three similar independent experiments. Panel (b) Effects of PD098059 on increased c-fos and c-jun mRNA expression by EGF and PMA. Bovine luteal cells were pretreated with 50 mM PD098059 for 1 h, followed by treatment with PMA (20 nM) or EGF (10 ng/ml) for 30 min. Total RNA was isolated. The RNA samples (15 mg/lane) were size-fractionated on formaldehyde-1% agarose gel and Northern blot analysis was performed with [a-32P] dCTP labeled bovine specific c-fos and c-jun cDNA probes as described in Fig. 7. The membranes were stripped and reprobed with human b-actin cDNA probe as an internal control. The results represent one of three similar independent experiments.

genes whose transcription is rapidly induced and independent of protein synthesis (Karin et al., 1997). Time-course studies revealed that EGF rapidly increased the mRNA levels of c-fos and c-jun in bovine luteal cells. When the cells were treated with EGF (10 ng/ml) up to 240 min, the levels of c-fos mRNA dramatically were increased after 15 min, reached maximal levels at 30 min, and maintained at high levels up to 60 min of treatment. Thereafter, EGF-induced cfos mRNA levels rapidly returned to basal levels at 90/ 120 min of treatment (Fig. 8a). The early time-course of c-jun gene expression was similar to c-fos in response to EGF. However, EGF-induced c-jun mRNA expression was maintained at high levels up to 90 min, declined at 120 min, and then diminished at 240 min of treatment (Fig. 8b). Notably, the profiles of the temporal and concentration responses of EGF-induced c-fos c-jun mRNA expression were very consistent with those of EGF-induced activation (mobility shift and phosphorylation) of ERK2 and ERK1 (Figs. 2 and 3).

150

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

3.7. Role of ERK MAPK in EGF-induced c-fos and c-jun mRNA expression Elk-1 is a substrate for activated MAPKs including members of all three MAPK subfamilies (Gille et al., 1992). Phosphorylation of Elk-1 by MAPKs stimulates the transcription of immediate-early response gene c-fos by facilitating the formation of a ternary complex with the serum response factor (SRF) and c-fos promoter serum response element (SRE) (Davis, 1993). Therefore, the question that arises is whether ERK activation is sufficient for the induction of c-fos and c-jun mRNA expression by EGF in bovine luteal cells. To test this, we first examined the effects of the specific MEK1 inhibitor PD098059 on phosphorylation of ERK2 and ERK1 by EGF in bovine luteal cells. Western blot analysis with the active MAPK antibody showed that following pretreatment with 20 or 50 mM of PD098059 for 60 min, EGF-induced ERK2 and ERK1 phosphorylation was completely attenuated (Fig. 9a). In keeping with the kinetic studies of c-fos and c-jun mRNA expression in response to EGF as shown in Figs. 7 and 8, experiments were performed to determine whether blocking ERK activation attenuates the maximal induction of c-fos and c-jun mRNA expression in response to EGF. As shown in Fig. 9b, treatment with EGF (10 ng/ml) and PMA (20 nM) for 30 min significantly stimulated both c-fos (top panel) and c-jun (middle panel) mRNA expression. Blocking ERK activation with PD098059 completely inhibited PMA-stimulated c-fos and c-jun mRNA expression. This result suggests that activation of the ERK MAPK plays an obligatory role in induction of cfos and c-jun oncogenes by PMA in bovine luteal cells. In contrast, the expression of c-fos and c-jun mRNA induced by EGF inhibited nearly 50% in cells pretreated with PD098059. This result indicates that ERK activation contributes toward maximal but is not required for EGF-induced transcription of c-fos and c-jun oncogenes.

4. Discussion This report provides evidence that EGF activates the MAPK signaling cascade in luteal cells. Mammalian cells contain three Raf genes encoding the serine/ threonine protein kinases, i. e. A-Raf, B-Raf, and Raf1, that function as MAPK kinase kinase to initiate the ERK2 and ERK1 signaling cascade (Dent et al., 1995). Chen et al. (1998) previously reported that Raf proteins are present in bovine luteal cells. In this report, we show that EGF phosphorylates and activates Raf-1 and BRaf. These data are similar to our previous report demonstrating the stimulatory effects of PGF2a and PMA on these Raf isoforms (Chen et al., 1998). However, the stimulatory effects of EGF on the

activities of B-Raf and Raf-1 are much greater (2 /3fold) than the effects of PMA and PGF2a in bovine luteal cells. Although A-Raf exhibits substantial homology to Raf-1 within the kinase domain of the two molecules but less homology elsewhere (Huleihel et al., 1986), A-Raf was not activated in bovine luteal cells by EGF (in the present study) and PGF2a and PMA (Chen et al., 1998). The mechanism by which these physiological ligands differentially activate these Raf isoforms in luteal cells is currently unclear and requires further investigation. Multiple mechanisms exist for the regulation of Raf activation. At present, at least three models of Raf-1 activation have been described. First, Raf-1 specifically associates with the active form of GTP-bound p21Ras, both in vitro and in vivo, via its regulatory domain. This pathway transmits and modulates extracellular mitogenic signals with receptor intrinsic tyrosine kinase activity. Ras binding does not activate Raf-1 directly but rather serves to recruit Raf-1 from the cytosol to the plasma membrane in response to growth factors. Second, some of the diacylglycerol-dependent isoforms of PKC, i.e. a (Kolch et al., 1993; Cai et al., 1997), b1 and o(Cai et al., 1997), can activate Raf-1 by direct phosphorylation of serine residues in the regulatory as well as the kinase domain (Kolch et al., 1993). Third, Raf-1 can be activated by tyrosine phosphorylation via membrane-associated Src family tyrosine kinases such as Lck (Chao et al., 1997) and PYK2 (Dikic et al., 1996). Although Ras-dependent activation of Raf-1 is the predominant signaling event that leads to ERK activation by tyrosine kinase receptors (Davis, 1993; Pearson et al., 2001), our present data suggest that PKC appears to be necessary for the full-activation of Raf-1, ERK2 and ERK1 by EGF in luteal cells. We showed that EGF-induced Raf-1 and ERK activation was reduced ( :/60%) in luteal cells rendered PKC-deficient by chronic PMA treatment. This result suggests that PKC is involved in EGF-induced Raf activation in luteal cells. Partial inhibition of EGF-induced Raf-1 and ERK activation in the absence of PKC indicates that mechanisms other than PKC also account for Raf-1 and ERK activation in response to EGF in bovine luteal cells. Further studies to determine the signaling events directly associated with EGFR and the role of p21Ras and nonreceptor tyrosine kinases in EGF-induced Raf activation in luteal cells are necessary to identify their involvement in the signaling event(s) which lead to Raf-1 activation. It is noteworthy that the stimulatory effects of EGF on the ERK MAPK signaling cascade were transient, and returned to basal levels within 60 min of treatment with EGF. These data, together with our previous observations that a similar and even more transient activation phenomenon occurred in the activation of Raf-1, MEK1, ERK2 and ERK1 in PGF2a-treated bovine luteal cells (Chen et al., 1998), demonstrate that

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

activation of the Raf-MEK/ERK MAPK cascade rapidly desensitizes in the presence of EGF and PGF2a in bovine luteal cells. The molecular mechanism that mediates agonist-induced desensitization of ERK MAPK signaling in luteal cells is currently unclear. In keeping with our previous observations that the activities of Raf, MEK1, ERK2 and ERK1 are coordinately regulated by PGF2a (Chen et al., 1998), the desensitization of ERK MAPK cascade most likely occurs at or precedes the activation of Raf-1. Phosphorylation of Raf reduces the ability of Raf to associate with the plasma membrane and contributes to desensitization of the ERK MAPK signaling cascade (Wartmann et al., 1996). Interestingly, in this report Raf desensitization appeared to require activation of ERK2. Thus, ERK2 may directly or indirectly modulate the activity of Raf. Additionally, desensitization of agonist-induced activation of ERK MAPK signaling cascade may result from down-regulation of the activity of tyrosine kinase receptors such as EGFR (Langlois et al., 1995; Budnik and Mukhopadhyay, 1996). The present series of experiments demonstrate that EGF rapidly stimulates the expression of the prototypic immediate-early response genes, i.e. c-fos and c-jun . This family of oncogenes includes the members of the Fos family (c-fos , fos-B , fra-1, and fra-2 ) and Jun family (c-jun , jun-B , and jun-D ) whose products are the constituents of the AP-1 transcription factors that form either Jun homodimers or Jun /Fos heterodimers (Karin et al., 1997). Except the Jun homodimers, AP-1 is predominantly induced rapidly at the transcriptional level by novel synthesis of its subunits with limited posttranslational modifications (Treisman, 1994). The transcription of c-fos and c-jun is controlled by the cis acting elements in their promoter regions, including the SRE and the TPA response element. The best-characterized SRE is that of the c-fos gene (Gille et al., 1992). Full-activation of the c-fos SRE requires association with the ubiquitous transcription factor SRF and formation of a ternary complex with ternary complex factors, including the Ets family of transcription factors Elk-1 and Sap1 (Treisman et al., 1994). The DNAbinding domain at the N-terminus of Elk-1 and Sap facilitates ternary complex formation, while the trans activation domain at the C-terminus contains several conserved MAPK phosphorylation sites. Upon cell stimulation, MAPK translocates into the nucleus where it phosphorylates the Elk-1 C-terminus (Chen et al., 1992), which then cooperates with the C-terminus of SRF activation domain to initiate c-fos transcription (Whitmarch et al., 1995). Our data suggest that phosphorylation of Elk-1 by the ERK pathway is involved in the activation of the c-fos gene by EGF in bovine luteal cells. The following evidence supports this notion. First, ERK2 and ERK1 were phosphorylated and activated upon EGF stimulation. Second, EGF-activated ERK2

151

and ERK1 were able to phosphorylate Elk-1 in vitro. Third, preincubation of cells with the specific MEK1 inhibitor PD098059 attenuated the phosphorylation of ERK2 and ERK1 by EGF. Fourth, active ERK was primarily observed in the nucleus of cells treated with EGF. Lastly, preincubation of cells with PD098059 inhibited EGF-induced c-fos mRNA expression. Unlike PMA-induced c-fos mRNA expression, which was completely attenuated by blocking the ERK pathway, our data showed that EGF-induced c-fos mRNA expression in bovine luteal cells was only partially attenuated by blocking the ERK pathway. This suggests that mechanisms in addition to ERK activation are also involved in EGF-induction of c-fos gene in bovine luteal cells. In other experiments, we have found that EGF also stimulated the activity of JNK1 in bovine luteal cells (Chen et al., 1997). Current evidence suggests that activated JNK translocates into the nucleus (Mizukami et al., 1997) and phosphorylates transcription factors other than Elk-1 to regulate c-fos expression (Whitmarch et al., 1995). Therefore, EGF-activated JNK pathway may contribute to ERK-independent activation of the c-fos gene in bovine luteal cells. The regulatory mechanisms for the induction of the cjun gene appear more complicated and much less understood. Early studies suggest that a putative AP-1 element in the c-jun promoter mediates the positive auto-induction of c-jun by mitogens (Angel et al., 1988). However, more recent studies show that several consensus binding sites for different transcription factors, including AP-1, MEF2, Sp-1, and CTF, are present in the c-jun promoter (Rozek and Pfizer, 1993). The MEF2 site positioned at /59 and, to a lesser extent, the AP-1 like site at /72 are absolutely required for induction of the c-jun promoter in response to EGF receptor activation in PC-12 cells (Leppa et al., 1998). Under similar conditions, deletion of the Sp-1 and CTF sites did not affect the activity of the c-jun promoter (Coso et al., 1997; Leppa et al., 1998). Therefore, the AP-1 transcription factors appear to be important regulators of c-jun expression and may be involved in EGF regulation of c-jun mRNA expression in luteal cells. Current evidence suggests that c-Jun activity can be regulated by post-translational modifications of preexisting c-Jun protein. c-Jun can be directly modulated by regulatory phosphorylations on several serine and threonine residues (Ser63, Ser73, Thr91 and Thr93) within the N-terminal trans -activation domain (Papavassiliou et al., 1995). Phosphorylation of these residues results in enhanced trans -activation, DNA-binding, and stability of c-Jun protein (Musti et al., 1997). The direct up-stream kinases that phosphorylate the residues within c-Jun N-terminus include members of the two MAPK subfamilies, i.e. the ERKs including ERK2 and ERK1 (Pulverer et al., 1991; Musti et al., 1997), and JNK (Kyriakis et al., 1994; Leppa et al.,

152

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

1998). Similar to the ERK pathway, JNK activation involves a kinase cascade comprising MEKK1, SEK and JNKs (Kyriakis et al., 1994). In addition, upstream activators of MEKK1 have been identified, including the small G-proteins Rac1, cdc42 and Ras (Minden et al., 1995). Analogous to the mechanisms associated with ERK activation (Sasaoka et al., 1994), JNK translocates into the nucleus (Mizukami et al., 1997) and binds directly to and phosphorylates c-Jun (Hibi et al., 1993). In Hela cells, transfection of active forms Ras, RacI, cdc42Hs, and MEKK1 increased, while transfection of dominant-negative mutants of Ras, RacI and MEKK1 inhibited the expression of a c-jun promoter-driven luciferase reporter gene in response to EGF (Leppa et al., 1998). This result suggested that the Rac /MEKK and JNK pathway was involved in the activation of the c-jun promoter by EGF. Our present data showed that EGF-induced enhanced c-jun mRNA levels was partially inhibited by blocking the ERK pathway using the MEK1 inhibitor PD098059. This suggests that additional mechanisms other than ERK MAPK are involved in EGF-induced c-jun mRNA expression. Again, our preliminary data suggested that EGF-activated JNK in bovine luteal cells (Chen et al., 1997), which provides an alternative pathway for EGF actions in bovine luteal cells. However, induction of c-jun gene through the MEF-2 sites may be a more important pathway in response to EGF. A recent report demonstrates that EGF activates the big MAPK (Bmk1/ERK5) which is required for cell cycle regulation (Kato et al., 1998). BMK1/ERK5, a newly identified MAPK family member, shares the same activation motif (TEY) with ERK1/ 2 but is activated by MEK5. The functions and regulatory mechanisms of Bmk/ERK5 are different from other MAPKs because of its long COOH-terminal tail (Lee et al., 1995). It has been recently, shown that BMK1/ERK5 phosphorylates and activates MEF2C possibly other MEF2 transcription factors that subsequently modulate the expression of c-jun (Kato et al., 1997). This may provide an alternative pathway for EGF-induced c-jun expression in bovine luteal cells. Taken together, we have demonstrated that EGF activates B-Raf and Raf-1, which subsequently activate the classical ERK MAPKs, i.e. ERK2 and ERK1, in bovine luteal cells. Activation of the Raf/MEK/ERK MAPK signaling cascade by EGF in bovine luteal cells is mediated by PKC-dependent and PKC-independent pathways. Upon EGF exposure, activated ERK MAPKs are present in the nucleus. We have also demonstrated that EGF rapidly stimulates the mRNA expression of the c-fos and c-jun oncogenes in bovine luteal cells. Blocking the ERK MAPK pathway by a specific MEK1 inhibitor only partially inhibited of EGF-induced mRNA levels of c-fos and c-jun oncogenes. These data demonstrate that induction of c-fos and c-jun oncogenes by EGF is mediated by both ERK-

dependent and ERK-independent pathways in bovine luteal cells.

Acknowledgements The present study was supported by the Lalor Foundation and NIH R01 HL70562 (DB Chen), and the Research Service of the Department of Veterans Affairs, USDA and NIH RO1 HD38813 (JS Davis). We are grateful for the excellent technical assistance of HonW Fong and Suzanne W. Westfall. We also thank Dr Mark S. Roberson (Cornell University) for providing the glutathione-S -transferase Elk-1, Dr Ian Mason (University of Texas Southwestern Medical Center) for providing the polyclonal rabbit anti-human 3b-HSD antibody, Drs Paul Dent and Thomas W. Sturgill (University of Virginia Health Sciences Center) for providing the [His]6-tagged-MEK1 and K52R, and Dr John W. Schmidt, Wichita State University, for his assistance in confocal microscopy.

References Ackland, J.F., Schwartz, N.B., Mayo, K.E., Dodson, R.E., 1992. Nonsteroidal signals originating in the gonads. Physiol. Rev. 72, 731 /787. Angel, P., Hattori, K., Smeal, T., Karin, M., 1988. The jun protooncogene is positively autoregulated by its product, Jun/AP-1. Cell 55, 875 /885. Ayyagari, R.R., Khan-Dawood, F.S., 1987. Human corpus luteum: presence of epidermal growth factor receptors and binding characteristics. Am. J. Obstet. Gynecol. 156, 942 /946. Budnik, L.T., Mukhopadhyay, A.K., 1991. Epidermal growth factor, a modulator of luteal adenylate cyclase. Characterization of epidermal growth factor receptors and its interaction with adenylate cyclase system in bovine luteal cell membrane. J. Biol. Chem. 266, 13908 /13913. Budnik, L.T., Mukhopadhyay, A.K., 1996. Modulatory action of epidermal growth factor on differentiated human granulosa lutein cells: cross-talk between ligand activated receptors for EGF and gonadotropin. Mol. Cell Endocrinol. 124, 141 /150. Cai, H., Smola, U., Wixler, V., Eisenmann-Tappe, I., Diaz-Meco, M.T., Moscat, J., Rapp, L.I., Cooper, G.M., 1997. Role of diacylglycerol-regulated protein kinase C isotypes in growth factor activation of the Raf-1 protein kinase. Mol. Cell Biol. 17, 732 /741. Cameron, M.R., Foster, J.S., Bukovsky, A., Wimalasena, J., 1996. Activation of mitogen-activated protein kinases by gonadotropins and cyclic adenosine 5?-monophosphates in porcine granulosa cells. Biol. Reprod. 55, 111 /119. Carpenter, G., Cohen, S., 1990. Epidermal growth factor. J. Biol. Chem. 265, 7709 /7712. Chakravorty, A., Joslyn, M.I., Davis, J.S., 1993. Characterization of insulin and insulin-like growth factor-I actions in the bovine luteal cell: regulation of receptor tyroskine kinase activity, phosphatidylinositol-3-kinase, and deoxyribonucleic acid synthesis. Endocrinology 3, 1331 /1340. Chao, T.S., Abe, M., Hershenson, M.B., Gomes, I., Roser, M.R., 1997. Src tyrosine kinase mediates stimulation Raf-1 kinase and mitogen-activated protein kinase by the tumor promoter thapsigargin. Cancer Res. 57, 3168 /3173.

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154 Chen, D.B., Fong, H.W., Davis, J.S., 2001. Induction of c-fos and cjun mRNA expression is mediated by a protein kinase C-dependent ERK MAP kinase. Endocrinology 142, 887 /895. Chen, D.B., Fong, H.W., Westfall, S.W., Rueda, B.R., Davis, J.S., 1997. Activation of the mitogen-activated protein kinases (ERK, JNK and p38mapk) in bovine luteal cells. Biol. Reprod. 56 (Suppl. 1), 229. Chen, D.B., Westfall, S.D., Fong, H.W., Roberson, M.A., Davis, J.S., 1998. PGF2a stimulates the Raf/MEK1/MAPK signaling cascade in bovine luteal cells. Endocrinology 139, 3876 /3885. Chen, R.H., Sarnecki, C., Blenis, J., 1992. Nuclear localization and regulation of erk -and rsk -encoded protein kinases. Mol. Cell Biol. 12, 915 /927. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate /phenol /chloroform extraction. Anal. Biochem. 162, 156 /159. Davis, J.S., May, J.V., Keel, B.A., 1996. Mechanisms of hormone and growth factor action in the bovine corpus luteum. Theriogenology 45, 1351 /1380. Coso, O.A., Montauer, S., Fromm, C., Lacal, J.C., Pryues, R., Teramoto, H., Gutkind, J.S., 1997. Signaling from G-proteincoupled receptors to the c-jun promoter involves the MEF-2 transcription factor: evidence for a novel c-Jun amino-terminal kinase-independent pathway. J. Biol. Chem. 272, 20691 /20697. Dalton, S., Treisman, R., 1992. Characterization of Sap-1, a protein recruited by serum response factor to the c-fos serum response element. Cell 68, 597 /612. Davis, R.J., 1993. The mitogen-activated protein kinase signal transduction pathway. J. Biol. Chem. 268, 14553 /14556. Dent, P., Beardon, D.B., Morrison, D.K., Jove, R., Sturgill, T.W., 1995. Regulation of Raf-1 and Raf-1 mutants by Ras-dependent and Ras-independent mechanism in vitro. Mol. Cell Biol. 15, 4125 /4135. Dikic, I., Tokiwa, G., Lev, S., Courtneidge, S.A., Schlessinger, J., 1996. A role for Pyk2 and Src in linking G-protein-coupled receptors with MAP kinase activation. Nature 383, 547 /550. Gangrade, B.K., Davis, J.S., May, J.V., 1991. A novel mechanism for the induction of aromatase in ovarian cells in vitro: role of transforming growth factor alpha-induced protein tyrosine kinase. Endocrinology 129, 2790 /2792. Gille, H., Sharrocks, A.D., Shaw, P.E., 1992. Phosphorylation of transcription factor p62TCF by MAP kinase stimulates ternary complex formation at the c-fos promoter. Nature 358, 414 /417. Habib, A.A., Hognason, T., Ren, J., Stefansson, K., Ratan, R.R., 1998. The epidermal growth factor receptor associates with and recruits phosphatidylinositol 3-kinase to the platelet-derived growth factor beta receptor. J. Biol. Chem. 273, 6885 /6891. Hibi, M., Lin, A., Smeal, T., Minden, A., Karin, M., 1993. Identification of an oncoprotein- and UV-responsive protein kinase that binds and potentiates the c-Jun activation domain. Genes Dev. 7, 2135 /2148. Huang, J.C., Khan-Dawood, F.S., Dawood, M.Y., Yeh, J., 1995. Baboon corpus luteum: epidermal growth factor receptor messenger ribonucleic acid expression during early, midluteal, and late luteal phases. Fertil. Steril. 63, 1318 /1321. Huleihel, M., Goldsborough, M.D., Cleveland, J.L., Gunnell, M., Bonner, T., Rapp, U., 1986. Characterization of murine A-Raf, a new oncogene related to the v-Raf oncogene. Mol. Cell Biol. 6, 2655 /2662. Karin, M., Liu, Z.G., Zandi, E., 1997. AP-1 function and regulation. Curr. Opin. Cell Biol. 9, 240 /246. Kato, Y., Kravchenko, V.V., Tapping, R.I., Han, J.H., Ulevitch, R.J., Lee, D.W., 1997. BMK1/ERK5 regulates serum-induced early gene expression through transcription factor MEF2C. EMBO J. 16, 7054 /7066.

153

Kato, Y., Tapping, R.I., Hwang, H., Watson, M.H., Ulevitch, R.J., Lee, J.D., 1998. Bmk1/Erk5 is required for cell proliferation induced by epidermal growth factor. Science 395, 713 /716. Keel, B.A., Hildebrandt, J.M., May, J.V., Davis, J.S., 1995. Effects of epidermal growth factor on the tyrosine phosphorylation of mitogen-activated protein kinases in monolayer cultures of porcine granulosa cells. Endocrinology 136, 1197 /1204. Knecht, M., Catt, K., 1983. Epidermal growth factor and gonadotropin-releasing hormone inhibit cyclic AMP-dependent luteinizing hormone receptor formation in ovarian granulosa cells. J. Cell Biochem. 21, 209 /217. Kolch, W., Heidecker, G., Kochs, G., Hummel, R., Vahidi, H., Finkenzeller, G., Marme, D., Rapp, U.R., 1993. Protein kinase C-a activates Raf-1 by direct phosphorylation. Nature 364, 249 /252. Kyriakis, J.M., Banerjee, P., Nikolakaki, E., Dai, T., Rubie, E.A., Ahmad, M.F., Avruch, J., Woodgett, J.R., 1994. The stressactivated protein kinase subfamily and c-Jun kinases. Nature 369, 156 /160. Lamm, M.L., Rajagopalan-Gupta, R.M., Hunzicker-Dunn, M., 1999. Epidermal growth factor-induced heterologous desensitization of the luteinizing hormone/choriogonadotropin receptor in a cell-free membrane preparation is associated with the tyrosine phosphorylation of the epidermal growth factor receptor. Endocrinology 140, 29 /36. Langlois, W.J., Sasaoka, T., Saltiel, A.R., Olefsky, J.M., 1995. Negative feedback regulation and desensitization of insulin- and epidermal growth factor-stimulated p21ras activation. J. Biol. Chem. 270, 25320 /25323. Lee, J.D., Ulevitch, R.J., Han, J., 1995. Primary structure of BMK1, a new mammalian MAP kinase. Biochem. Biophy. Res. Commun. 213, 715 /724. Leppa, S., Saftriech, R., Ansonge, W., Bohmann, D., 1998. Differnetial regulation of c-Jun by ERK and JNK during PC12 cell differentiation. EMBO J. 17, 4404 /4413. Li, X., Yu, H., Graves, L.M., Earp, H.S., 1997. Protein kinase C and protein kinase A inhibit calcium-dependent but not stress-dependent c-Jun N-terminal kinase activation in rat liver epithelial cells. J. Biol. Chem. 272, 14996 /15002. Liebmann, C., 2001. Regulation of MAP kinase activity by peptide receptor signaling pathway: paradigms of multiplicity. Cell. Signaling 13, 777 /785. Liu, F., Chernoff, J., 1997. Protein tyrosine phosphatase 1B interacts with and is tyrosine phosphorylated by the epidermal growth factor receptor. Biochem. J. 327, 139 /145. Margolis, B., Li, N., Koch, A., Mohammadi, M., Hurwitz, D.R., Zilberstein, A., Ullrich, A., Pawson, T., Schlessinger, J., 1990. The tyrosine phosphorylated carboxyterminus of the EGF receptor is a binding site for GAP and PLC-/g˜/. EMBO J. 9, 4375 /4380. Maruo, T., Ladines-Llave, C.A., Samoto, T., Matsuo, H., Manalo, A.S., Ito, H., Mochizuki, M., 1993. Expression of epidermal growth factor and its receptor in the human ovary during follicular growth and regression. Endocrinology 132, 924 /931. May, J.V., Bridge, A.J., Gotcher, E.D., Gangrade, B.K., 1992. The regulation of porcine theca cell proliferation in vitro: synergistic actions of epidermal growth factor and platelet-derived growth factor. Endocrinology 131, 689 /697. May, J.V., Buck, P.A., Schomberg, D.W., 1987. Epidermal growth factor enhances [125I]-follicle-stimulating hormone binding by cultured porcine granulosa cells. Endocrinology 120, 2413 /2420. McGuire, W.J., Juengel, J.I., Niswender, G.D., 1994. Protein kinase C second messenger system mediates the antisteroidogenic effects of prostaglandin F2a in the ovine corpus luteum in vivo. Biol. Reprod. 51, 800 /806. Minden, A., Lin, A., Claret, F.X., Abo, A., Karin, M., 1995. Selective activation of JNK signaling cascade and c-Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell 81, 1147 / 1157.

154

D.-b. Chen, J.S. Davis / Molecular and Cellular Endocrinology 200 (2003) 141 /154

Misajon, A., Hutchinson, P., Lolatgis, N., Trounson, A.O., Almahbobi, G., 1999. The mechanism of action of epidermal growth factor and transforming growth factor-a on aromatase activity in granulosa cells from polycystic ovaries. Mol. Hum. Reprod. 5, 96 / 103. Mizukami, Y., Yoshioka, K., Morimoto, S., Yoshida, K.I., 1997. A novel mechanism of JNK1 activation. Nuclear translocation and activation of JNK1 during ischemia and reperfusion. J. Biol. Chem. 272, 16657 /16662. Murray, J.F., Downing, J.A., Evans, G., Findlay, J.K., Scaramuzzi, R.J., 1993. Epidermal growth factor acts directly on the sheep ovary in vivo to inhibit oestradiol-17 beta and inhibit secretion and enhance progesterone secretion. J. Endocrinol. 137, 253 /264. Musti, A.M., Treier, M., Bohmann, D., 1997. Reduced ubiquitindependent degradation of c-Jun after phosphorylation by MAP kinases. Science 275, 400 /402. Papavassiliou, A.G., Treiver, M., Bohmann, D., 1995. Intramolecular signal transduction in c-Jun. EMBO J. 14, 2014 /2019. Pearson, G., Robinson, F., Gibson, T.B., Xu, B., Karandikar, M., Berman, K., Cobb, M.H., 2001. Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr. Rev. 22, 153 /183. Pescador, N., Stocco, D.M., Murphy, B.D., 1999. Growth factor modulation of steroidogenic acute regulatory protein and luteinization in the pig ovary. Biol. Reprod. 60, 1453 /1461. Pulverer, B.J., Kyriakis, J.M., Auruch, J., Nikolakaki, E., Woodgett, J.R., 1991. Phosphorylation of c-jun mediated by MAP kinases. Nature 353, 670 /674. Roberson, M.S., Misra-Press, A., Laurance, M.E., Stork, P.J.S., Maurer, R.A., 1995. A role for mitogen-activated protein kinase in mediating activation of the glycoprotein hormone alpha-subunit promoter by gonadotropin-releasing hormone. Mol. Cell Biol. 15, 3531 /3539. Rotin, D., Honegger, A.M., Margolis, B.L., Ullrich, A., Schlessinger, J., 1992. Presence of SH2 domains of phospholipase C gamma 1 enhances substrate phosphorylation by increasing the affinity toward the epidermal growth factor receptor. J. Biol. Chem. 267, 9678 /9683.

Rozek, D., Pfizer, G.P., 1993. In vitro protein /DNA interactions at the c-jun promoter: preformed complexes mediate the UV response. Mol. Cell Biol. 13, 5490 /5499. Sasaoka, T., Langlois, W.J., Leitner, J.W., Draznin, B., Olefsky, J.M., 1994. The signaling pathway coupling epidermal growth factor receptors to activation of p21ras. J. Biol. Chem. 269, 32621 /32625. Scurry, J.P., Hamand, K.A., Astley, S.B., Leake, R.E., Wells, I., 1994. Immunoreactivity of antibodies to epidermal growth factor, transforming growth factors alpha and beta, and epidermal growth factor receptor in the premenopausal ovary. Pathology 26, 130 / 133. Shaulian, E., Karin, M., 2002. AP-1 as a regulator of cell life and death. Nat. Cell Biol. 4, E131 /E136. Singh, B., Kennedy, T.G., Tekpetey, F.R., Armstrong, D.T., 1995. Gene expression and peptide localization for epidermal growth factor receptor and its ligands in porcine luteal cells. Mol. Cell Endocrinol. 113, 137 /143. St-Arnaud, R., Walker, P., Kelly, P.A., Labrie, F., 1983. Rat ovarian epidermal growth factor receptors: characterization and hormonal regulation. Mol. Cell Endocrinol. 31, 43 /52. Stass, S.A., Mixson, J., 1997. Oncogenes and tumor suppressor genes: therapeutic implications. Clin. Cancer Res. 3, 2687 /2695. Tekpetey, F.R., Singh, B., Barbe, G., Armstrong, D.T., 1995. Localization of epidermal growth factor (EGF) receptor in the rat corpus luteum, and EGF and transforming growth factor-alpha stimulation of luteal cell steroidogenesis in vitro. Mol. Cell Endocrinol. 110, 95 /102. Treisman, R., 1994. Ternary complex factors: growth factor regulated transcriptional activators. Curr. Opin. Genet. Dev. 4, 96 /101. Vlodavsky, I., Brown, K.D., Gospodarowicz, D., 1978. A comparison of the binding of epidermal growth factor to cultured granulosa and luteal cells. J. Biol. Chem. 253, 3744 /3750. Wartmann, M., Hofer, P., Turowski, P., Satiel, A.R., Hynes, N.E., 1996. Negative modulation of membrane localization of the Raf-1 protein in kinase by hyperphosphorylation. J. Biol. Chem. 272, 3915 /3923. Whitmarch, A.J., Shore, P., Sharrocks, A.D., Davis, R.J., 1995. Integration of MAP kinase signal transduction pathways of the serum response element. Science 264, 403 /440.