JBC Biosymposia, Inc.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schweppe, R. E.
Right arrow Articles by Bradford, A. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schweppe, R. E.
Right arrow Articles by Bradford, A. P.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Volume 272, Number 49, Issue of December 5, 1997 pp. 30852-30859

Functional Components of Fibroblast Growth Factor (FGF) Signal Transduction in Pituitary Cells
IDENTIFICATION OF FGF RESPONSE ELEMENTS IN THE PROLACTIN GENE*

(Received for publication, July 30, 1997, and in revised form, September 16, 1997)

Rebecca E. Schweppe Dagger , Ashley A. Frazer-Abel Dagger , Arthur Gutierrez-Hartmann Dagger § and Andrew P. Bradford §

From the Departments of § Medicine and of Dagger  Biochemistry and Molecular Genetics, Program in Molecular Biology, and the Colorado Cancer Center, University of Colorado Health Sciences Center, Denver, Colorado 80262

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES

ABSTRACT

Fibroblast growth factors (FGFs) have been implicated in pituitary lactotroph tumorigenesis; however, little is known about the molecular mechanisms of FGF signal transduction. We used a transient transfection approach, in GH4 cells, to identify components of the FGF signaling pathway leading to activation of the rat prolactin (rPRL) promoter. Using dominant-negative constructs of p21Ras, Raf-1 kinase, and mitogen-activated protein (MAP) kinase, we show that FGF activation of the rPRL promoter is independent of Ras and Raf-1 but requires MAP kinase. Furthermore, MAP kinase but not Raf-1 kinase catalytic activity is stimulated by FGFs. The rPRL promoter FGF response maps to two Ets binding sites, centered at -212 (FRE1) and -96 (FRE2), and co-transfection of dominant-negative Ets inhibits FGF activation. FRE1 co-localizes with a composite, Ets/GHF-1, Ras response element. However, overexpression of Ets-1 and GHF-1, which potentiate the Ras response, inhibits FGF stimulation of the rPRL promoter, implying that Ras and FGF signaling pathways target distinct factors to elicit their effects. These data suggest that Ets factors serve to sort and integrate MAP kinase-dependent growth factor signals, allowing highly specific transcriptional responses to be mediated via the interaction of distinct Ets proteins and cofactors at common response elements.

INTRODUCTION

Fibroblast growth factor-2 (FGF-2 or basic FGF)1 and FGF-4 (hst-1) are members of a family comprised of 14 heparin-binding polypeptides (1-3). These proteins play important roles in growth and differentiation and have been implicated in the formation and progression of tumors in a variety of tissues, including the pituitary (4-7). FGF-2 is most abundant in the pituitary gland, where it was first isolated (8), and has been found in human pituitary adenomas (4). In addition, FGF-2 has been shown to stimulate prolactin (PRL) secretion from normal rat pituitary cells (9) and cultured human pituitary adenomas (5). Furthermore, elevated levels of immunoreactive FGF-2 are present in patients with multiple endocrine neoplasia type-1 and pituitary tumors (6).

FGF-4 was first isolated as a transforming gene in human stomach cancer (10). Subsequently, FGF-4 gene sequences have been isolated from human prolactinomas and shown to exhibit transforming activity in NIH 3T3 assays (11). In addition, FGF-4 has been shown to induce PRL secretion and gene transcription in rat pituitary cells (7), and rats injected with GH4 cells stably transfected with FGF-4 expression vectors developed highly aggressive, prolactin-secreting tumors (7). Thus, FGF-4 may be directly involved in the development, progression, and metastasis of pituitary tumors. In addition, it has recently been shown that pituitary lactotroph adenomas express different FGF receptor isoforms and subtypes compared with the normal pituitary (12).

Despite these advances in our understanding of FGF-2 and FGF-4 expression in the pituitary, the precise mechanisms of FGF signal transduction in pituitary cells have not been elucidated. The initial event in the FGF signal transduction pathway is activation of FGF receptor tyrosine kinase activity. Both Ras/Raf/MAP kinase-dependent and -independent pathways have been implicated as downstream components of the FGF signaling pathway in various systems. The Ras/Raf/MAP kinase pathway is required for mesoderm induction by FGF in Xenopus (13-15), neurite outgrowth in PC12 cells (16), and migration of tracheal cells in Drosophila (17). FGFs have also been suggested to signal via Ras/Raf-independent pathways including pertussis toxin-sensitive G proteins (18, 19), protein kinase C (PKC), (19, 20) and protein kinase B (21), and recently, FGF receptor-3 has been shown to be associated with Stat-1 (22).

Little is known about the nuclear components of FGF signaling pathways. FGF signaling in Xenopus requires AP-1/Jun for normal embryonic development and mesoderm induction (23). Similarly, few FGF- responsive cis-acting DNA elements (FREs) have been identified. FGF activation of the proenkephalin gene requires ATF-3 and c-Jun, which bind to a sequence previously identified as a cAMP response element (24). In addition, an FRE in the rat osteocalcin promoter confers a synergistic induction by FGF-2 and cAMP (25), and a similar element has been found in the human osteocalcin promoter (26). Finally, the FGF response of the urokinase-type plasminogen activator (uPA) promoter has been mapped to a PEA3-like/AP1 composite element (27). However, the precise identity of the factors mediating these FGF responses have not been identified.

Identification of the nuclear and cytoplasmic components of the FGF signal transduction pathway has been hampered by the lack of a convenient, rapid assay, since FGF responses characterized to date have typically involved induction of cellular proliferation, morphological changes, or differentiation. Regulation of the rat (r) prolactin (PRL) gene in GH4 pituitary tumor cell line provides an excellent, physiologically relevant model system to study the components of FGF signaling. GH4 cells are highly differentiated, retain cell-specific functions and hormonal responses, and express the phenotypic marker, PRL, under the control of the pituitary-specific POU homeodomain transcription factor, GHF-1/Pit-1 (28). Moreover, this model system has been successfully employed to study the mechanisms of the epidermal growth factor and Ras signal transduction pathways as they impinge on the rPRL promoter (29-32). GH4 cells are FGF-responsive, and FGF-4 has been shown to specifically activate rPRL gene transcription (7).

To investigate the molecular mechanisms of FGF-2 and FGF-4 signaling to the rPRL promoter in GH4 cells, a transient transfection system was utilized. In this study we show that FGF-2 and FGF-4 induction of the rPRL promoter is independent of Ras and Raf-1 kinase but requires MAP kinase. We have identified two FREs, both containing Ets binding sites, and provide evidence that a member of the Ets family of transcription factors is a functional nuclear target of the FGF response. Of note, the rPRL FREs co-localize with 1) a previously identified Ras response element (32) and 2) a basal transcription element implicated in insulin activation of the rPRL promoter (33, 34). However, the FGF, Ras, and insulin responses appear to be mediated by distinct Ets transcription factors. Thus, we have identified critical functional components of the FGF signaling pathway in pituitary cells, and we present evidence that different growth factors, acting via MAP kinase, act to target distinct Ets members that bind common cis-acting response elements, thus serving to sort and integrate multiple signaling pathways resulting in specific changes in gene expression.

EXPERIMENTAL PROCEDURES

Cell Culture and Transfections

GH4T2 rat pituitary tumor cells were grown in Dulbecco's modified Eagle's medium (DMEM, Life Technologies, Inc.) supplemented with 10% fetal calf serum (FCS) (Hyclone, Logan UT) and 50 µg/ml penicillin and streptomycin (Gemini, Calabasas, CA). Cells were maintained at 37 °C in 5% CO2.

Transient transfections were carried out by electroporation, as described previously (31, 32). Briefly, media were changed 4-12 h prior to each transfection, and cells were harvested at 50-70% confluency and electroporated in 10% FCS/DMEM, as described (32). After electroporation, cells were plated in DMEM without serum for a final concentration of 0.66% FCS. Similar results were obtained when cells were plated in 10% FCS/DMEM. Cells were incubated for a total of 24 h and treated with FGF-2 (kindly provided by Brad Olwin, University of Colorado, Boulder), FGF-4 (R & D Systems, Minneapolis, MN), or diluent (0.1% bovine serum albumin in phosphate-buffered saline) at a final concentration of 1 ng/ml to the existing media. Minimal FGF responses were detected as early as 2 h post-treatment; however, optimal responses occurred after 6 h of incubation. Similar results were obtained when cells were treated with FGFs ± 1 µg/ml heparin (Sigma). Electroporations were performed in triplicate or duplicate where indicated for each condition within a single experiment, and experiments were repeated using different plasmid preparations of each construct. Luciferase and beta -galactosidase assays were measured as described previously (31, 32).

Plasmid Constructs

The promoter constructs pA3rPRLluc-425, pA3rGHluc (35), the -255, -212, -189, -138, -125, -54, -36 rPRL promoter deletions, the site-specific mutations in FPII (pA3(mFPII)rPRLluc), the RRE (pA3(mEBS)rPRLluc), and in FPIV (pA3(mFPIV)rPRLluc), and the FPI deletion (pA3(FPIdel)rPRLluc) have been described previously (36). The plasmids pSVras, Fp6V Fp6V c-src (527F), pZCR17N, RSV-Raf-C4, pLNCXiMAPK(K-M), pAPr-EtsZ, pSG5c-Ets-1, pRSVGHF-1, pRSVGHF-2, and pCMVbeta (CLONTECH, Palo Alto, CA) have been described elsewhere (31, 36). Plasmid DNAs were purified by either the alkaline sodium dodecyl sulfate extraction and cesium chloride density gradient centrifugation or by the Qiagen Mega protocol (Qiagen, Chatsworth, CA). Plasmids were quantitated by both absorbance at 260 nm and verified by agarose electrophoresis with DNA standards.

Western Blot Analysis

GH4T2 cells were serum-starved overnight and treated with 1 ng/ml FGF-2 or FGF-4, 25 nM EGF, or the equivalent volume of diluent for the indicated times. Cells (107) were washed in cold phosphate-buffered saline and harvested in 500 µl of extraction buffer (EB) (1% Triton X-100, 10 mM Tris, pH 7.4, 5 mM EDTA, 50 mM NaCl, 50 mM NaF, 0.1% bovine serum albumin, 20 µg/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, and 2 mM Na3VO4). Equal amounts of protein (50 µg), as determined by the Bio-Rad DC protein assay, were resolved by electrophoresis on 10% SDS-polyacrylamide gels and transferred to Immobilon-P membrane (Millipore, Bedford, MA) according to the manufacturer's protocol. Membranes were probed with primary antibodies to p42/p44 MAP kinase and to the tyrosine-phosphorylated form of MAP kinase (New England Biolabs, Beverly, MA) according to the manufacturer's protocol, except membranes were incubated with a goat anti-rabbit horseradish peroxidase antibody (Life Technologies, Inc.) diluted 1:10,000 in blocking buffer. Protein was detected using the chemiluminescence assay, ECL (Amersham Life Sciences) according to the manufacturer's protocols. Where indicated, membranes were stripped for 30 min at 50 °C according to the ECL protocol and re-probed as described above.

Raf-1 Kinase Assays

Raf-1 kinase assays were performed essentially as described (37, 38) using the syntide II peptide (PLARTLSVAGLPGKK) as a substrate (generously provided by Dr. Roger Colbran, Vanderbilt University). GH4T2 cell extracts were prepared as described above for the MAPK Western blots. Equal protein was immunoprecipitated with a polyclonal anti-Raf-1 (C12, Santa Cruz Biotechnology, Santa Cruz, CA) antibody (1:500) or a rabbit IgG monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA), as a negative control, at 4 °C for 1 h. The immunocomplexes were then bound to 25 µl of protein A-CL-4B Sepharose slurry (Zymed, San Francisco, CA) for 1 h and washed one time with 1 ml of EB and one time with 1 ml of 0.5 M LiCl, 100 mM Tris, pH 7.6. The immunocomplexes were resuspended in 36 µl of kinase buffer (25 mM Tris-HCl, pH 7.5, 10 mM MnCl2, 10 µM ATP, 1 mM dithiothreitol, 25 mM beta -glycerophosphate), 1 µl of 4 mM syntide II peptide, and 0.1 µCi of [gamma -32P]ATP. The reaction was allowed to proceed for 20 min at room temperature. The reaction was terminated with 2 µl of 10% trichloroacetic acid, spotted in duplicate on P81 Whatman paper, and air-dried. The filters were washed extensively in 0.85% phosphoric acid and counted in scintillation fluid. Radioactive counts from the preimmune reactions were subtracted from counts from the anti-Raf-1 immunoprecipitations.

Site-directed Mutagenesis

The pA3(mBTE)rPRLluc construct and the double mutant, pA3(mBTE/mEBS)rPRLluc, were created by nested polymerase chain reaction site-directed mutagenesis in pG7rPRL or pG7(mEBS)rPRL, respectively, using specific mutagenic oligonucleotides in combination with commercially available SP6 and T7 primers (Promega, Madison, WI), as described previously (36). pG7(mEBS)rPRL contains a site-specific mutation in the Ets binding site (EBS) within the RRE (-214 to -209) (36). The core -97CGGAAA-92 EBS within the BTE was changed to CTCGAG, generating a unique XhoI restriction site for the mBTE construct and a second XhoI site for the mBTE/mEBS construct. Thus, the double mutant has the EBS at the RRE at position -214 to -209 mutated and the EBS at -99 to -94 also mutated to XhoI sites. The resulting constructs were cloned into the HindIII site of pA3luc and sequenced to confirm the presence of the mutant EBSs and verify that the resulting promoters are otherwise identical to pA3rPRLluc and pA3(mEBS)rPRLluc, respectively.

RESULTS

FGF Activation of the rPRL Promoter Is Independent of Ras

FGFs have been shown to selectively activate the rat (r) prolactin (PRL) promoter but not the ancestrally related rat growth hormone (GH) promoter, in GH4 pituitary cells (7). This selective activation of the rPRL promoter is reminiscent of the Ras response of these related promoters in that V-12 Ras activates the rPRL promoter but not the rGH promoter in GH4 cells (31, 32). Based on these results, and other observations implicating the Ras/Raf/MAP kinase pathway in FGF signal transduction (13-17, 37, 39), we investigated whether FGF-2 or FGF-4 signal via the Ras pathway in this system. To determine the functional role of Ras in FGF-2 and FGF-4 activation of the rPRL promoter, a dominant-negative N17Ras construct (40) was transiently transfected into GH4 cells along with the rPRL promoter fused to the luciferase gene (pA3rPRLluc). As shown in Fig. 1A, in the absence of N17Ras, FGF-2 and FGF-4 activate the rPRL promoter approximately 3-fold. Co-transfection of N17Ras does not significantly block either FGF-2 or FGF-4 activation of the rPRL promoter. To demonstrate that N17Ras is functionally expressed in this system, we show that N17Ras blocks the 43-fold induction mediated by mutationally activated c-src (Fp6V c-src 527F) (41), which acts via the Ras pathway, to about 15-fold (Fig. 1B). Thus, although FGF-2 and FGF-4, like Ras, selectively activate the rPRL promoter, these results indicate that FGF-2 and FGF-4 induction of rPRL promoter activity in GH4 cells is not mediated by Ras.

Fig. 1. FGF-2 and FGF-4 activation of the prolactin promoter is not inhibited by dominant-negative Ras. GH4 cells were co-transfected with 3 µg of pA3rPRL-425luc and 0.3 µg of pCMVbeta gal with or without 10 µg of pZCRN17Ras, where indicated. Cells were treated with 1 ng/ml FGF-2, FGF-4, or diluent, as indicated, for 6 h prior to harvest. Cells were harvested 24 h after transfection. Luciferase activity was normalized to beta -galactosidase activity, and the basal activity of pA3rPRL-425luc was set to 1. Data are expressed as mean fold activation ± S.E. of 30 transfections. Statistical analysis (Student's t test) indicates that there is not a significant difference in the FGF response ± N17Ras. B, N17Ras blocks src induction of the rPRL promoter. GH4 cells were transfected with pA3rPRL-425luc, pCMVbeta gal, and N17Ras as in A along with 10 µg of Fp6v c-Src (527F) where indicated. Cells were treated with FGFs, harvested, and assayed as above. Results are a representative experiment of three transfections.

[View Larger Version of this Image (23K GIF file)]

Activation of the rPRL Promoter Is Not Mediated via Raf-1

We next sought to determine whether Raf-1 kinase, a typical downstream component in the Ras pathway, is involved in FGF activation of the rPRL promoter in GH4 cells, since Raf-1 kinase has been shown to be activated independently of Ras, e.g. by protein kinase C (42). To test the functional role of Raf-1 in the FGF response, GH4 cells were co-transfected with a dominant-negative Raf (C4 Raf) construct (43). Fig. 2A shows that FGF-2 and FGF-4 activate the rPRL promoter 3- to 4-fold. Co-transfection of dominant-negative C4 Raf does not significantly block activation by FGF-2 or FGF-4. Consistent with previous results (31), C4 Raf blocked the 5-fold activation by Ras of the rPRL promoter by 60%, showing that the dominant-negative construct is functional in this system. These data indicate that FGF-2 and FGF-4 activation of the rPRL promoter in GH4 cells is independent of Raf-1 kinase.

Fig. 2. The role of Raf-1 kinase in the FGF response. A, dominant-negative Raf does not inhibit FGF-2 or FGF-4 activation of the prolactin promoter. GH4 cells were co-transfected with 3 µg of pA3rPRL-425luc and 0.3 µg of pCMVbeta gal in the presence or absence of 10 µg of pRSV-Raf-C4B and 2 µg of V-12 Ras, where indicated. Cells were treated, harvested, and assayed as described in Fig. 1. Data are expressed as mean fold ± S.E. of six transfections as in Fig. 1. Statistical analysis (Student's t test) indicates that FGF activation of the rPRL promoter is not significantly different ± C4 Raf. B, FGF-2 and FGF-4 do not activate Raf-1 kinase catalytic activity. GH4 cells were serum-starved overnight and treated with 1 ng/ml FGF-2, FGF-4, diluent, or 25 nM EGF, as indicated. Raf-1 kinase activity was measured using an immunocomplex kinase assay with syntide II as substrate as described under "Experimental Procedures."

[View Larger Version of this Image (26K GIF file)]

To investigate further the role of Raf-1 in FGF signaling, we examined the effects of FGF-2 and FGF-4 on Raf-1 kinase catalytic activity. Fig. 2B shows the results of a Raf-1 immunocomplex kinase assay using Syntide II as a substrate (37, 38). EGF activates Raf-1 kinase activity approximately 2-21/2-fold after 5 min of treatment. However, treatment with FGF-2 or FGF-4 for as long as 30 min does not activate Raf-1 kinase activity (Fig. 2B). Thus, both biochemical and functional analyses indicate that Raf-1 kinase is not a component of the FGF signaling pathway leading to activation of the rPRL promoter.

FGF-2 and FGF-4 Signal via MAP Kinase to Activate the rPRL Promoter

MAP kinase has been shown to be a critical point of convergence of many hormone and growth factor signaling pathways, including some that are Ras- and/or Raf-independent (44). Thus, we tested the role of MAP kinase in FGF signaling and activation of the rPRL promoter. GH4 cells were transfected with a p42 MAP kinase construct (iMAPK), mutated in its ATP-binding site, that has been shown to specifically inhibit MAP kinase-dependent signaling (31). Fig. 3A shows that transfection of iMAPK completely abolished the 2-5-fold activation of the rPRL promoter by FGF-2 and FGF-4, respectively. Consistent with previous observations (31), the iMAPK construct attenuated the 6-fold Ras activation of the rPRL promoter to about 3-fold (Fig. 3A). Expression of iMAPK has no effect on forskolin activation of the rPRL promoter mediated by protein kinase A (data not shown and Ref. 31). These results implicate p42 MAP kinase as a functional component of the FGF signaling pathway leading to activation of the rPRL promoter.

Fig. 3. The role of p42 MAP kinase in the FGF response. A, the effect of an inhibitory MAP kinase (iMAPK) expression vector on FGF-2- and FGF-4-stimulated rPRL promoter activity. GH4 cells were co-transfected with 3 µg of pA3rPRL-425luc and 0.3 µg of pCMVbeta gal with or without 25 µg of pLNCXiMAPK (31) and 3 µg of V-12 Ras, where indicated. Cells were treated, harvested, and assayed as described in Fig. 1. Fold activation was determined relative to the basal activity of the rPRL promoter. Results show a representative experiment performed in triplicate. Similar results were obtained in six other transfections. B, FGF-2 and FGF-4 activate MAP kinase catalytic activity. GH4 cell extracts were prepared as in Fig. 2B and resolved on 10% SDS-polyacrylamide gels along with control proteins for p42/p44 MAP kinase (lane 11) and a phosphorylated p42/p44 MAP kinase (lane 12). After transfer to Immobilon-P, membranes were probed with a phospho-specific antibody to p42/p44 MAP kinase (top panel) and detected using ECL. The membrane was stripped and reprobed with an antibody to p42/p44 MAP kinase that is not phospho-specific, and protein was detected using ECL.

[View Larger Version of this Image (36K GIF file)]

To determine the effect of FGF-2 and FGF-4 on MAP kinase catalytic activity, extracts derived from GH4 cells treated with either FGF-2, FGF-4, EGF, or diluent were probed with an antibody specific for the phosphorylated and activated form of MAP kinase (45). Fig. 3B shows that MAP kinase is not phosphorylated in control/untreated cells (lane 1). Treatment with EGF for 5 min results in phosphorylation and activation of MAP kinase (lane 2). FGF-2 and FGF-4 treatment induced MAP kinase phosphorylation and activation for up to 30 min (lanes 3-10). Specificity of the antibody for the phosphorylated MAP kinase protein is illustrated in lanes 11 and 12. To demonstrate equal loading of protein, the blot was stripped and re-probed with a second MAP kinase antibody, which recognizes both phosphorylated and nonphosphorylated forms (Fig. 3B). Taken together, these data indicate that FGF-2 and FGF-4 signaling to the rPRL promoter is independent of Ras and Raf-1 but requires MAP kinase.

FGF-2- and FGF-4-responsive Elements Map to EBSs on the rPRL Promoter

To date, few functionally relevant FGF-responsive cis-acting DNA elements have been characterized. FGF response elements (FREs) have been identified in the proenkephalin (24), the uPA (27), and the osteocalcin genes. Likewise, no transcription factor nuclear targets of FGF signaling have been clearly elucidated, but ATF-3 and c-Jun have been implicated for the proenkephalin promoter (24), and AP1 for the uPA promoter (27). The factors that bind to the osteocalcin FRE have not been identified. However, AP1/Jun has been shown to be required for mesoderm induction by FGFs in Xenopus (23).

To facilitate the identification of trans-acting factors involved in the rPRL promoter FGF response, and to further characterize FGF's mode of action in the pituitary, we used a series of 5' exonuclease deletions (32) and site-specific mutations (34, 36) to map the FGF-responsive regions of the rPRL promoter. Fig. 4B shows the effects of the 5' exonuclease deletions on activation of the rPRL promoter by FGF-2 and FGF-4. The intact -425 promoter exhibits a 3-7-fold activation by FGF-2 and FGF-4 (100%), and the -255 construct showed equivalent FGF responses. However, deletion to -212 of the promoter significantly reduces induction by both FGF-2 and FGF-4 to approximately 20-30% of wild-type (Fig. 4B). The rPRL promoter constructs with end points at -189, -138 (data not shown), and -125 show similar attenuated FGF responses of 20-40%. This residual FGF response is eliminated upon further deletion to position -54, since the -54 and -36 constructs do not exhibit significant induction by FGF-2 or FGF-4 compared with the promoterless pA3luc vector (Fig. 4B). These results suggest that the FGF-2 and FGF-4 response maps to two elements of the rPRL promoter, the first between -255 and -212, and the second between -125 and -54. Interestingly, both of these regions contain binding sites for the Ets family of transcription factors. The Ets binding site (EBS) centered at -212, along with the adjacent GHF-1 binding site (FPIV), have previously been shown to function as a composite Ras response element (RRE) (32). The EBS centered at -96 lies within a basal transcription element (BTE) (34) that has been shown to be important in the insulin response of the rPRL promoter (33). Additionally, the most proximal FRE is immediately adjacent to a repressor binding site (FPII), which is thought to interact with the BTE to maintain pituitary-specific expression of the rPRL gene (34).

Fig. 4. Mapping of the FGF response elements of the proximal rPRL promoter. A, structural organization of the proximal rPRL promoter. The region between nucleotides -425 to +73 is depicted. The end points of 5' exonuclease deletions constructed in pA3luc and verified by dideoxy sequencing are indicated by the numbers in boldface. GHF-1 footprints (FPI, FPIII, and FPIV), as determined by DNase protection, are indicated by shaded rectangles. Putative Ets binding sites are shown by solid rectangles. The FPII repressor site and the basal transcription element (BTE) (34) are denoted by the circle and triangle, respectively. B, the effect of FGF-2 and FGF-4 on rPRL promoter deletions. GH4 cells were co-transfected with 3 µg of the indicated series of rPRL promoter deletions or the empty vector pA3luc along with 0.3 µg of pCMVbeta gal. Cells were treated with FGF-2, FGF-4, or diluent and assayed as described in Fig. 1. FGF activation of each rPRL promoter construct was determined relative to the basal activity of that promoter construct. FGF-2 and FGF-4 fold activation of the -425 promoter (3-7-fold) was set to 100%, and basal promoter activity (1-fold) was set to 0%. Results are the mean ± S.E. of seven transfections. Statistical analysis (Student's t test) indicates that the -212, -189, and -125 rPRL constructs gave a significantly lower FGF responses (p < .01)(*) compared with the -425 promoter. The responses of the -212, -189, and -125 are statistically different from the -54 and -36 rPRL constructs (p < .01) (**), which did not exhibit FGF responses relative to the promoterless vector (pA3luc).

[View Larger Version of this Image (27K GIF file)]

The functional role of each of these putative FREs was evaluated using site-specific mutations. First, we tested rPRL promoter constructs harboring site-specific mutations of the Ets binding site (mEBS) and the GHF-1 binding site (mFPIV) that comprise the RRE (36) (Fig. 5A). As shown in Fig. 5B, site-directed mutagenesis of the EBS and FPIV reduces the FGF response to 35-50% of the intact -425 promoter. These results imply that the EBS and the GHF-1 binding site (FPIV) within the RRE are important contributors to the FGF response. The 3'-FRE lies between -125 and -54 which spans the end point of the FPII repressor site and all of the BTE. Mutation of FPII (mFPII) reduces the FGF response by approximately 60% (Fig. 5B). Similar attenuated FGF-2 and FGF-4 responses are also observed upon mutation of the EBS within the BTE (mBTE). In addition, deletion of the BTE in the context of an FPII mutation markedly inhibited the FGF response, reducing it to approximately 20% of wild type (data not shown). Of note, the rPRL promoters bearing mutations in the BTE retain significant basal activity and are fully responsive to Ras (data not shown). To analyze further the functional role of the Ets binding sites within the BTE and RRE, a double mutant (mEBS/mBTE) was created using polymerase chain reaction site-directed mutagenesis (Fig. 5A). Fig. 5B shows that the rPRL promoter containing the double mutant Ets sites in both FREs exhibits a minimal FGF response (20% of wild-type). These two FREs may be acting as composite and interacting elements, consisting of an EBS and GHF-1 binding site (FPIV) within the RRE and the EBS within the BTE and the adjacent FPII. This is consistent with the mechanism of Ets transcription factors, as they typically act with other factors at composite elements (46). To examine the role of FPI on the FGF response, an rPRL promoter construct containing an FPI deletion (34) was tested, and it retains almost complete FGF-2 and FGF-4 responses (Fig. 4B). Taken together, these results suggest the Ets binding sites within the composite FREs are critical for FGF activation of the rPRL promoter and that the FPIV and FPII regions also contribute.

Fig. 5. Site-specific mutations of putative FGF response elements. A, putative FGF response elements of the rPRL promoter. GHF-1 binding sites are indicated by the shaded rectangles; putative Ets binding sites are indicated by the solid rectangles, and the FPII repressor site and the BTE are indicated by a circle and triangle, respectively. Site-specific mutations of the GHF-1 binding site (mFPIV), the Ets binding sites within the RRE (mEBS) and BTE (mBTE), and site-directed mutagenesis of FPII (mFPII) are indicated below the wild type sequences. B, GH4 cells were co-transfected with 0.3 µg of pCMVbeta gal and 3 µg of the indicated rPRL promoter constructs and treated with FGF-2, FGF-4, or diluent and assayed as described in Fig. 1. FGF activation is expressed as percent of the intact -425 promoter (21/2- to 6-fold). Results are the mean ± S.E. of 8-12 transfections.

[View Larger Version of this Image (27K GIF file)]

Dominant-negative Ets Inhibits the FGF Response

The above promoter mapping data indicate that the FGF response maps to Ets binding sites within the RRE and BTE. To examine the functional role of Ets transcription factors in FGF activation of the rPRL promoter, a dominant-negative Ets construct, pAPrEts-Z, was transiently transfected into GH4 cells. This dominant-negative Ets encodes the highly conserved DNA binding domain of Ets-2 and inhibits the effects of Ets family members since it is able to recognize and bind the same DNA sequences but lacks a transactivation domain (47). Fig. 6A shows that transfection of pAPrEts-Z reduces the FGF-2 and -4 response by 60 and 70%, respectively. Consistent with our previous observations (31), pAPrEts-Z inhibits V-12 Ras activation of the rPRL promoter to a similar extent (Fig. 6A). These results are in agreement with the mapping data (Fig. 5) and further corroborate that an Ets factor is a critical nuclear component of the FGF response leading to activation of the rPRL promoter.

Fig. 6. Dominant-negative Ets but not GHF-2 inhibits FGF-2 and FGF-4 activation of the rPRL promoter. A, pAPrEts-Z (47) inhibits FGF activation of the rPRL promoter. GH4 cells were co-transfected with 3 µg of pA3PRL-425luc, 0.3 µg of pCMVbeta gal with or without 10 µg of pAPrEts-Z, and 3 µg of V-12 Ras, where indicated. GH4 cells were treated with FGF-2, FGF-4, or diluent and assayed as described in Fig. 1. FGF activation is expressed as percent of the maximum response (3-6-fold). The results are a representative experiment of six transfections done in triplicate. B, effect of GHF-2 on FGF and Ras activation of the rPRL promoter. GH4 cells were co-transfected with 3 µg of pA3rPRL-425luc, 0.3 µg of pCMVbeta gal with or without 20 µg pRSVGHF-2, and 3 µg of V-12 Ras where indicated. Cells were treated with FGF-2, FGF-4, or diluent and assayed as described in Fig. 1. FGF activation is expressed as percent maximum response (3-5-fold). The results are a representative experiment performed in triplicate. FGF activation of the rPRL promoter is not statistically different ± GHF-2 (Student's t test). Similar results were observed in six other transfections. Error bars indicate standard deviation.

[View Larger Version of this Image (26K GIF file)]

GHF-2 Fails to Block FGF Activation of the rPRL Promoter

Site-specific mutation and deletion analysis of the rPRL promoter indicate that a composite Ets/GHF-1 binding site, previously defined as an RRE (32), may function as an FRE. Specifically, mutation of FPIV, a GHF-1 binding site, inhibits the FGF response by approximately 50%, suggesting a role for GHF-1 in FGF signaling. To address further the functional role of GHF-1, GH4 cells were transfected with the alternatively spliced isoform, GHF-2, which functions as a dominant-negative in this system and serves as a probe for certain GHF-1-mediated events (32, 48). As shown previously (32, 48), co-transfection of GHF-2 reduces activation of the rPRL promoter by V-12 Ras (Fig. 6B). However, GHF-2 has no significant effect on the FGF-2 or FGF-4 responses (Fig. 6B). These results indicate that unlike the Ras response (32, 48), GHF-1 is not required for the FGF response. These data also show that GHF-2 does not mediate the FGF response, as it does the PKA response (48).

The Effect of Ets-1 and GHF-1 on the FGF Response

The above results show that the FGF and Ras pathways target a common composite element, the RRE, yet they also suggest that FGFs and Ras are utilizing distinct transcription factors to mediate their responses. The Ras response is mediated by a functional interaction between an Ets-1-like factor and GHF-1 (32), whereas the FGF response also requires an Ets family member but appears to be independent of GHF-1 (Fig. 6). To examine directly the role of Ets-1 and GHF-1 in the FGF response, these factors were expressed in GH4 cells with or without FGFs (Fig. 7). Overexpression of GHF-1 or Ets-1 has minimal effect on the basal activity of the rPRL promoter (data not shown). Consistent with our previous findings (32), the 4-fold Ras response of the rPRL promoter is enhanced to almost 6-fold in the presence of Ets-1, to 10-fold in the presence of GHF-1, and to an optimum 16-fold in the presence of both Ets-1 and GHF-1 (Fig. 7A). As shown in Fig. 7B, the 6-fold induction of the rPRL promoter by FGF-4 is not enhanced by Ets-1 but is in fact inhibited to approximately 4-fold. Similarly, co-transfection of GHF-1 reduced the FGF-4 response to the same extent. Finally, in the presence of both GHF-1 and Ets-1, the FGF response is further inhibited to only 3-fold (Fig. 7B). Similar results were obtained with FGF-2 (data not shown). These results show that the specific transcription factors Ets-1 and GHF-1 are not functional components of FGF activation of the rPRL promoter and provide further evidence that Ras and FGFs are targeting distinct nuclear factors.

Fig. 7. FGFs and Ras target distinct nuclear factors. A, Ets-1 and GHF-1 enhance the Ras response. GH4 cells were co-transfected with 3 µg of pA3PRL-425luc, 0.3 µg of pCMVbeta gal with or without 10 µg of pSG5c-Ets-1, 10 µg of pRSVGHF-1, or 3 µg of V-12 Ras, where indicated, harvested, and assayed as in Fig. 1. B, Ets-1 and GHF-1 inhibit the FGF response. GH4 cells were transfected with pA3rPRL-425luc, pCMVbeta gal, c-Ets-1, and GHF-1 as in A. Cells were treated with FGF-4 or diluent and assayed as described in Fig. 1. Fold activation was determined relative to rPRL activity in the presence of c-Ets-1 or GHF-1 alone or c-Ets-1 and GHF-1 together. Results are a representative experiment of six transfections done in duplicate.

[View Larger Version of this Image (28K GIF file)]

DISCUSSION

FGFs and their receptors play a critical role in development and in the regulation of cell growth and differentiation (2) and have been implicated in tumorigenesis in a variety of tissues including the pituitary (4-7). In the studies shown here we have utilized the rPRL gene in GH4 pituitary cells to identify both cytoplasmic and nuclear functional components of the FGF signaling pathway regulating expression of an endogenous gene in a differentiated cell line. Both FGF-2 and FGF-4 selectively activate the rPRL promoter; however, in contrast to other systems, the FGF response is independent of p21 Ras and is not mediated via Raf-1 kinase. Moreover, FGFs fail to stimulate Raf-1 kinase catalytic activity in this system. FGF induction of the rPRL promoter is, however, dependent on MAP kinase, whose catalytic activity is stimulated by FGFs in GH4 cells. Using a series of site-specific mutations and 5' deletions, we have identified two interacting rPRL promoter FGF response elements, each containing binding sites for members of the Ets family of transcription factors, and we present evidence that an Ets factor is a critical nuclear component of the FGF response.

FGFs mediate their biological effects via a family of transmembrane tyrosine kinase receptors that undergo dimerization and autophosphorylation in response to FGF binding. FGF receptors are thought to be linked to the Ras pathway via a novel lipid binding adaptor protein, FRS2 (49), and current evidence indicates that the Ras/Raf-1 pathway acting via MAP kinase is critical for several FGF responses, including induction of mesoderm in Xenopus (13-15), neurite outgrowth in PC12 cells (16), and migration of tracheal cells in Drosophila (17). Other FGF responses may be mediated via Ras-independent pathways, involving pertussis toxin-sensitive G-proteins (18, 19) or protein kinase C (PKC) (18, 19). We have shown that FGF activation of the rPRL promoter is not mediated via Ras or Raf-1 (Figs. 1 and 2) but is dependent on MAP kinase (Fig. 3). Ras-independent signal transduction via tyrosine kinase receptors is not without precedence in the pituitary, since EGF activation of the rPRL promoter also does not require the Ras/Raf-1 pathway. Indeed the Ras and EGF pathways are mutually antagonistic (29, 30). The EGF response is blocked by inhibitors of PKC that also attenuates activation of the rPRL promoter by FGFs.2 Of note, PKC-dependent activation of MAP kinase has been recently documented in two pituitary cell systems, thyrotropin releasing hormone signaling in GH3 lactotrophs (50) and gonadotropin-releasing hormone signaling in the gonadotrope-alpha T3-1 cell line (51). However, in both of these studies, only biochemical data were presented, and the physiological relevance of the PKC-mediated activation of MAP kinase remains to be determined. Nevertheless, the possibility remains that the Ras-independent activation of MAP kinase by FGFs in GH4 pituitary cells may be mediated via PKC.

Recently, Ras-independent activation of MAP kinase leading to activation of the Ets factor, Elk-1, and differentiation of neuroendocrine PC12 cells has been described (52). In this case, activation of the MAP kinase cascade occurs via B-Raf kinase, which is selectively activated by the small GTPase protein, Rap1. B-Raf is expressed predominantly in neuronal and endocrine cells (53) and is present in GH4 pituitary cells.2 Immunocomplex kinase assays used in this study were specific for Raf-1 kinase and would not detect B-Raf activity. Thus, it will be interesting to determine the role of B-Raf and Rap1 in the FGF activation of the pituitary rPRL promoter.

Growth factor tyrosine kinase signaling via Ras-independent pathways has been reported in several highly differentiated cell lines. For example, insulin activates MAP kinase via a Ras-dependent pathway in 3T3-L1 fibroblasts, but upon differentiation to adipocytes, insulin stimulation of MAP kinase is mediated by a predominantly Ras-independent mechanism (54). Furthermore, EGF activation of the rPRL promoter in GH4 pituitary cells is independent of Ras (29). Thus, regulation of tissue-specific gene expression, e.g. prolactin, and/or establishment of a highly differentiated phenotype, may be mediated in part by growth factor tyrosine kinase receptors selectively utilizing Ras-independent signaling pathways.

Nuclear components of the FGF signal transduction pathway have not been well characterized, but a role for AP1 proteins has been suggested. FGF response elements have been identified in the uPA and proenkephalin genes containing AP-1 binding sites (24), and AP-1 is required for mesoderm induction by FGFs in Xenopus (23). The rat and human osteocalcin genes also contain FREs; however, the factors that bind to these elements have not been identified (25, 26). In contrast, the rPRL promoter used in these studies lacks a consensus AP1 site, and both basal (42) and FGF-stimulated2 transcription is inhibited by expression of c-Jun. We have mapped the FGF response of the rPRL promoter to two elements (centered at -212 and -96), each of which contain binding sites for members of the Ets family of transcription factors. These two FREs appear to contribute equally, and both elements are required for optimal FGF activation (Figs. 4 and 5). Furthermore, expression of a dominant-negative Ets factor inhibited FGF induction of the rPRL promoter (Fig. 6). In addition, we and others (36, 55) have shown specific binding of Ets factors to both of these elements in vitro.2 However, the possibility that other as yet unknown transcription factors bind to these Ets cognate sequences has not been excluded. Thus, we have identified one of the first examples of functionally relevant FREs containing Ets binding sites. Recently, FGF-2 activation of the human interstitial collagenase promoter (matrix metalloproteinase, MMP1) has been shown to be mediated in a Ras-dependent manner via a bipartite Ets-AP1 element (56). FGF treatment up-regulates an AP1 complex containing Fra1 and c-Jun. Binding to the Ets element is not FGF-inducible, and the Ets factor remains to be identified.

We have previously defined the FRE centered at -212 as a composite Ras response element (RRE) that is comprised of juxtaposed Ets and GHF-1 binding sites (FPIV) (36). Ras activation of the rPRL promoter is mediated via a functional interaction between Ets-1 and GHF-1 at this element and is enhanced by overexpression of either or both factors. Mutation of either the EBS or GHF-1 site reduced the FGF response of the rPRL promoter (Fig. 5). However, in contrast to Ras activation of rPRL promoter activity, the FGF response is inhibited by overexpression of Ets-1 or GHF-1 (Fig. 7). Thus, an Ets factor(s) and a POU homeodomain protein(s) distinct from Ets-1 and GHF-1, respectively, may mediate the FGF response. The inhibition of FGF activation of the rPRL promoter by Ets-1 and GHF-1 may reflect the formation of nonproductive complexes that block access to the FRE. Moreover, expression of the alternatively spliced isoform GHF-2, which blocks Ras activation of the rPRL promoter (Fig. 6B and Ref. 36), had no effect on the FGF response. Furthermore, in contrast to FGF stimulation, the Ras response is mediated solely by the composite RRE, whereas the BTE and FPII are not required (Fig. 8). Taken together, these results imply that, despite targeting a common cis element (-207 to -190) in the rPRL promoter, the Ras and FGF pathways utilize distinct nuclear factors to transduce their effects.

Fig. 8. FGF activation of the rPRL promoter. FGF-2 and FGF-4 activate the rPRL promoter independently of Ras and Raf-1 but require MAP kinase. The FGF response maps to two composite elements, both containing Ets binding sites. These sites have previously been identified as a Ras (RRE) and an insulin response element (IRE), as indicated. Even though FGFs, Ras, and insulin converge at MAP kinase, these pathways target distinct Ets factors; the Ras response requires Ets-1 and GHF-1, the insulin response requires the Ets factor GABP, and the FGF response is mediated by a yet to be identified member of the Ets family of transcription factors. Thus, different Ets factors may serve to sort different signaling pathways to elicit a specific response.

[View Larger Version of this Image (22K GIF file)]

The second rPRL promoter FRE (-96) lies within a basal transcription element (BTE) (34, 57) that has also been implicated in the cAMP, thyrotropin releasing hormone, phorbol ester, and EGF responses (57-60). This element is also required for the insulin activation of the rPRL promoter via the Ets factor GABP (33, 55). Of note, rPRL promoter constructs used to study the insulin response lacked the RRE, and thus its contribution has not been evaluated. The BTE is immediately adjacent to FPII, which binds an unknown factor and exerts a modulatory effect on the BTE to repress PRL transcription in nonpituitary cells (34). FPII has also been implicated in EGF activation of the rPRL promoter (30). Here we show that mutation of either FPII or the EBS within the BTE reduces the FGF response (Fig. 5), suggesting that this region, like the RRE, may also function as a composite response element.

Recent evidence has suggested a central role for members of the Ets family of transcription factors in the regulation of basal lactotroph-specific rPRL gene expression (61) and in rPRL promoter regulation in response to Ras (32, 36), insulin (55), and in this study FGFs. Several Ets factors are nuclear targets of growth factor signaling pathways acting via MAP kinase (44, 46, 62) and can both positively and negatively regulate transcription (32, 63-66). Thus, based upon our data, we propose a model by which growth factors can elicit distinct responses, despite acting via common cis-acting response elements, by targeting different members of the Ets family (Fig. 8). Moreover, since Ets factors typically act in concert with other transcription factors at composite elements (44, 46, 62), further specificity can be conferred by interactions with different coactivators. Thus, the rPRL promoter Ras response is mediated via interaction of Ets-1 and the pituitary-specific POU homeodomain protein GHF-1 at the RRE (32, 36), whereas FGF induction of promoter activity is mediated by distinct Ets members and other cofactors, perhaps homeodomain proteins, which bind to this same composite element (FRE1). Similarly, insulin activation of rPRL promoter activity may be mediated via the Ets factor GABP binding to the BTE (55). The FGF response also utilizes this element but may require interactions with coactivators binding at FPII which remain to be identified (Fig. 8). Thus, Ets factors may serve to sort, integrate, and coordinate transcriptional responses to different growth factor signaling pathways, resulting in highly selective regulation of tissue-specific gene expression.

FOOTNOTES

*   This work was supported by National Institutes of Health Grant DK37667 (to A. G. H.) and in part by Sigma Xi Grant-in-Aid of Research (to R. E. S.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
   To whom correspondence should be addressed: Dept. of Medicine, Program in Molecular Biology, and the Colorado Cancer Center, University of Colorado Health Sciences Center, 4200 East Ninth Ave., Box B-151, Denver, CO 80262. Tel.: 303-315-8443; Fax: 303-315-4525; E-mail: Andy.Bradford{at}UCHSC.edu.
1   The abbreviations used are: FGF, fibroblast growth factors; rPRL, rat prolactin; PRL, prolactin; PKC, protein kinase C; FRE, FGF response elements; uPA, urokinase-type plasminogen activator; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; ECL, enhanced chemiluminescence; EBS, Ets binding site; MAPK, mitogen-activated protein kinase; EGF, epidermal growth factor; GH, growth hormone; RRE, Ras response element; BTE, basal transcription element.
2   R. E. Schweppe and A. P. Bradford, unpublished observations.

ACKNOWLEDGEMENTS

We thank Drs. Natalie Ahn, Carol Carter, Jeff Dunkelberg, David Gordon, Lynn Heasley, and Cheryl Pickett for discussion and reading of the manuscript. We thank Kelley Fantle, Nicole Manning, and Dr. Scott Diamond for technical assistance. We thank Drs. Roger Colbran for the gift of Syntide II and Brad Olwin for FGF-2.

REFERENCES

  1. Yamasaki, M., Miyake, A., Tagashira, S., and Itoh, N. (1996) J. Biol. Chem. 271, 15918-15921 [Abstract/Free Full Text]
  2. Mason, I. (1994) Cell 78, 547-552 [CrossRef][Medline] [Order article via Infotrieve]
  3. Smallwood, P., Munoz-Sanjuan, I., Tong, P., Macke, J., Hendry, S., Gilbert, D., Copeland, N., Jenkins, N., and Nathans, J. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 9850-9857 [Abstract/Free Full Text]
  4. Li, Y., Koga, M., Kasayama, S., Matsumotot, K., Arita, N., Hayakawa, T., and Sato, B. (1992) J. Clin. Endocrinol. Metab. 75, 1436-1441 [Abstract]
  5. Atkin, S., Landolt, A., Jeffreys, R., Diver, M., Radcliffe, J., and White, M. (1993) J. Clin. Endocrinol. & Metab. 77, 831-837 [Abstract]
  6. Zimering, M., Katsumata, N., Sato, Y., Brandi, M., Aurbach, G., Marx, S., and Friesen, H. (1993) J. Clin. Endocrinol. & Metab. 76, 1182-1187 [Abstract]
  7. Shimon, I., Huttner, A., Said, J., Spirina, O., and Melmed, S. (1996) J. Clin. Invest. 97, 187-195 [Medline] [Order article via Infotrieve]
  8. Gospodarowicz, D. (1975) J. Biol. Chem. 250, 2515-2520 [Abstract/Free Full Text]
  9. Baird, A., Mormede, P., Ying, Y., Wehrenberg, B., Ueno, N., Ling, N., and Guillemin, R. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 5545-5549 [Abstract/Free Full Text]
  10. Sakamoto, H., Mori, M., Taira, M., Yoshida, S., Matsukawa, K., Shimizu, M., Sekiguchi, M., Terada, M., and Sugimura, T. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 3997-4001 [Abstract/Free Full Text]
  11. Gonsky, R., Herman, V., Melmed, S., and Fagin, J. (1991) Mol. Endocrinol. 5, 1687-1695 [Abstract]
  12. Ali Asghar Abbass, S., Asa, S. L., and Ezzat, S. (1997) J. Clin. Endocrinol. & Metab. 82, 1160-1166 [Abstract/Free Full Text]
  13. LaBonne, C., Burke, B., and Whitman, M. (1995) Development 121, 1475-1486 [Abstract]
  14. MacNicol, A., Muslin, A., and Williams, L. (1993) Cell 73, 571-583 [CrossRef][Medline] [Order article via Infotrieve]
  15. Umbhauer, M., Marshall, C., Mason, C., Old, R., and Smith, J. (1995) Nature 376, 58-62 [CrossRef][Medline] [Order article via Infotrieve]
  16. Kremer, N., D'Arcangelo, G., Thomas, S., De Marco, M., Brugge, J., and Halegoua, S. (1991) J. Cell Biol. 115, 809-819 [Abstract/Free Full Text]
  17. Reichman-Fried, M., Dickson, B., Hafer, E., and Shilo, B. (1994) Genes Dev. 8, 428-439 [Abstract/Free Full Text]
  18. Estival, A., Robberecht, P., Fanjul, M., Rouots, B., Hollande, E., Vaysse, N., and Clemente, F. (1996) Biochem. J. 315, 619-624
  19. Logan, A., and Logan, S. (1991) Cell Signalling 3, 215-223 [CrossRef][Medline] [Order article via Infotrieve]
  20. Hurley, M., Marcello, K., Abreu, C., and Kessler, M. (1996) J. Bone Miner. Res. 11, 1256-1263 [Medline] [Order article via Infotrieve]
  21. Burgering, B., and Coffer, P. (1995) Nature 376, 599-602 [CrossRef][Medline] [Order article via Infotrieve]
  22. Su, W.-C. S., Kitagawa, M., Xue, N., Xie, B., Garofalo, S., Cho, J., Deng, C., Horton, W., and Fu, X.-Y. (1997) Nature 386, 288-292 [CrossRef][Medline] [Order article via Infotrieve]
  23. Dong, Z., Xu, R.-H., Kim, J., Zhan, S.-N., Ma, W.-Y., Colburn, N. H., and King, H.-F. (1996) J. Biol. Chem. 271, 9942-9946 [Abstract/Free Full Text]
  24. Tan, Y., Low, K., Boccia, C., Grossman, J., and Comb, M. (1994) Mol. Cell. Biol. 14, 7546-7556 [Abstract/Free Full Text]
  25. Boudreaux, J. M., and Towler, D. A. (1996) J. Biol. Chem. 271, 7508-7515 [Abstract/Free Full Text]
  26. Schedlich, L., Flanagan, J., Crofts, L., Gillies, S., Goldberg, D., Morrison, N., and Eisman, J. (1994) J. Bone Miner. Res. 9, 143-152 [Medline] [Order article via Infotrieve]
  27. Besser, D., Presta, M., and Nagamine, Y. (1995) Cell Growth Differ. 6, 1009-1017 [Abstract]
  28. Castrillo, J. L., Theill, L. E., and Karin, M. (1991) Science 253, 197-199 [Abstract/Free Full Text]
  29. Pickett, C. A., and Gutierrez-Hartmann, A. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 8612-8616 [Abstract/Free Full Text]
  30. Pickett, C., and Gutierrez-Hartmann, A. (1995) Mol. Cell. Biol. 15, 6777-6784 [Abstract]
  31. Conrad, K. E., Oberwetter, J. M., Vallaincourt, R., Johnson, G. L., and Gutierrez-Hartmann, A. (1994) Mol. Cell. Biol. 14, 1553-1565 [Abstract/Free Full Text]
  32. Bradford, A. P., Conrad, K. E., Wasylyk, C., Wasylyk, B., and Gutierrez-Hartmann, A. (1995) Mol. Cell. Biol. 15, 2849-2857 [Abstract]
  33. Jacob, K. K., Ouyang, L., and Stanley, F. M. (1995) J. Biol. Chem. 270, 27773-27779 [Abstract/Free Full Text]
  34. Jackson, S. M., Keech, C. A., Williamson, D. J., and Gutierrez-Hartmann, A. (1992) Mol. Cell. Biol. 12, 2708-2719 [Abstract/Free Full Text]
  35. Conrad, K. E., and Gutierrez-Hartmann, A. (1992) Oncogene 7, 1279-1286 [Medline] [Order article via Infotrieve]
  36. Bradford, A. P., Conrad, K. E., Tran, P. H., Ostrowski, M. C., and Gutierrrez-Hartmann, A. (1996) J. Biol. Chem. 271, 24639-24648 [Abstract/Free Full Text]
  37. Huang, J., Mohammadi, M., Rodrigues, G. A., and Schlessinger, J. (1995) J. Biol. Chem. 270, 5065-5072 [Abstract/Free Full Text]
  38. Kovacina, K. S., Yonezawa, K., Brautigan, D. L., Tonks, N. K., Rapp, U. R., and Roth, R. A. (1990) J. Biol. Chem. 265, 12115-12118 [Abstract/Free Full Text]
  39. Whitman, M., and Melton, D. (1992) Nature 357, 252-254 [CrossRef][Medline] [Order article via Infotrieve]
  40. Feig, L., and Cooper, G. M. (1988) Mol. Cell. Biol. 8, 3235-3243 [Abstract/Free Full Text]
  41. Gupta, S. K., Gallego, C., Johnson, G. L., and Heasley, L. E. (1992) J. Biol. Chem. 267, 7987-7990 [Abstract/Free Full Text]
  42. Oberwetter, J. M., Conrad, K. E., and Gutierrez-Hartmann, A. (1993) Mol. Endocrinol. 7, 915-923 [Abstract]
  43. Bruder, J. T., Heidecker, G., and Rapp, U. R. (1992) Genes Dev. 6, 545-556 [Abstract/Free Full Text]
  44. Hill, C., and Treisman, R. (1995) Cell 80, 199-211 [CrossRef][Medline] [Order article via Infotrieve]
  45. Payne, D. M., Rossomando, A. J., Martino, P., Erickson, A. K., Her, J. H., Shabanowitz, J., Hunt, D. F., Weber, M. J., and Sturgill, T. W. (1991) EMBO J. 10, 885-892 [Medline] [Order article via Infotrieve]
  46. Janknecht, R., and Nordheim, A. (1993) Biochim. Biophys. Acta 1155, 346-356 [Medline] [Order article via Infotrieve]
  47. Langer, S. J., Bortner, D. M., Roussel, M. F., Sherr, C. J., and Ostrowski, M. C. (1992) Mol. Cell. Biol. 12, 5355-5362 [Abstract/Free Full Text]
  48. Diamond, S. E., and Gutierrez-Hartmann, A. (1996) J. Biol. Chem. 271, 28925-28932 [Abstract/Free Full Text]
  49. Kouhara, H., Hadari, Y., Spivak-Kroizman, T., Schilling, J., Bar-Sagi, D., Lax, I., and Schlessinger, J. (1997) Cell 89, 693-702 [CrossRef][Medline] [Order article via Infotrieve]
  50. Ohmichi, M., Sawada, T., Kanda, Y., Koike, K., Hirota, K., Miyake, A., and Saltiel, A. R. (1994) J. Biol. Chem. 269, 3783-3788 [Abstract/Free Full Text]
  51. Reiss, N., Llevi, L., Shacham, S., Harris, D., Seger, R., and Naor, Z. (1997) Endocrinology 138, 1673-1682 [Abstract/Free Full Text]
  52. Vossler, M., Yao, H., York, R., Pan, M.-G., Rim, C., and Stork, J. (1997) Cell 89, 73-82 [CrossRef][Medline] [Order article via Infotrieve]
  53. Barnier, J. V., Papin, C., Eychene, A., Lecoq, O., and Calothy, G. (1995) J. Biol. Chem. 270, 23381-23389 [Abstract/Free Full Text]
  54. Carel, K., Kummer, J. L., Schubert, C., Leitner, W., Heidenreich, K. A., and Draznin, B. (1996) J. Biol. Chem. 271, 30625-30630 [Abstract/Free Full Text]
  55. Newberry, E., Willis, D., Latifi, T., Boudreaux, J., and Towler, D. (1997) Mol. Endocrinol. 11, 1129-1144 [Abstract/Free Full Text]
  56. Iverson, R. A., Day, K. H., d'Emden, M., Day, R. N., and Maurer, R. A. (1990) Mol. Endocrinol. 4, 1564-1571 [Abstract]
  57. Liang, J., Kim, K. E., Schoderbeck, W. E., and Maurer, R. A. (1992) Mol. Endocrinol. 6, 885-892 [Abstract]
  58. Peers, B., Monget, P., Nalda, M. A., Voz, M. L., Berwaer, M., Belayew, A., and Martial, J. A. (1991) J. Biol. Chem. 266, 18127-18134 [Abstract/Free Full Text]
  59. Peers, B., Nalda, A. M., Monget, P., Voz, M. L., Belayew, A., and Martial, J. A. (1992) Eur. J. Biochem. 210, 53-58 [Medline] [Order article via Infotrieve]
  60. Ouyang, L., Jacob, K. K., and Stanley, F. M. (1996) J. Biol. Chem. 271, 10425-10428 [Abstract/Free Full Text]
  61. Bradford, A., Wasylyk, C., Wasylyk, B., and Gutierrez-Hartmann, A. (1997) Mol. Cell. Biol. 17, 1065-1074 [Abstract]
  62. Wasylyk, B., Hahn, S. H., and Giovane, A. (1993) Eur. J. Biochem. 211, 7-18 [Medline] [Order article via Infotrieve]
  63. Wasylyk, C., Bradford, A., Gutierrez-Hartmann, A., and Wasylyk, B. (1997) Oncogene 14, 899-913 [CrossRef][Medline] [Order article via Infotrieve]
  64. Yang, B.-S., Hauser, C., Henkel, G., Colman, M., Van Beveren, C., Stacey, K., Hume, D., Maki, R., and Ostrowski, M. (1996) Mol. Cell. Biol. 16, 538-547 [Abstract]
  65. Giovane, A., Pintzas, A., Maira, S.-M., Sobieszcuk, P., and Wasylyk, B. (1994) Genes Dev. 8, 1502-1513 [Abstract/Free Full Text]
  66. Sgouras, D., Athanasiou, M., Beal, G., Fisher, R., Blair, D., and Mavrothalassitis, G. (1995) EMBO J. 14, 4781-4793 [Medline] [Order article via Infotrieve]
Volume 272, Number 49, Issue of December 5, 1997 pp. 30852-30859
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles: