Activation of Go-coupled Dopamine D2 Receptors Inhibits ERK1/ERK2 in Pituitary Cells
2002; Elsevier BV; Volume: 277; Issue: 39 Linguagem: Inglês
10.1074/jbc.m202920200
ISSN1083-351X
AutoresJeffrey Liu, Ross E. Baker, Clement Sun, V.C. Sundmark, Harry P. Elsholtz,
Tópico(s)Pituitary Gland Disorders and Treatments
ResumoIn pituitary lactotrophs the prolactin gene is stimulated by neuropeptides and estrogen and is suppressed by dopamine via D2-type receptors. Stimulatory signals converge on activation of the mitogen-activated protein kinases ERK1/2, but dopamine regulation of this pathway is not well defined. Paradoxically, D2 agonists activate ERK1/2 in many cell types. Here we show that in prolactin-secreting GH4ZR7 cells and primary pituitary cells, dopamine treatment leads to a rapid, pronounced, and specific decrease in activated ERK1/2. The response is blocked by D2-specific antagonists and pertussis toxin. Interestingly, in stable lines expressing specific pertussis toxin-resistant Gα subunits, toxin treatment blocks dopamine suppression of MAPK in Gαi2- but not Gαo-expressing cells, demonstrating that Go-dependent pathways can effect the inhibitory MAPK response. At the nuclear level, the MEK1 inhibitor U0126 mimics the D2-agonist bromocryptine in suppressing levels of endogenous prolactin transcripts. Moreover, a good correlation is seen between the IC50 values for inhibition of MEK1 and suppression of prolactin promoter function (PD184352 > U0126 > U0125). Both dopamine and U0126 enhance the nuclear localization of ERF, a MAPK-sensitive ETS repressor that inhibits prolactin promoter activity. In addition, U0126 suppression is transferred by tandem copies of the Pit-1-binding site, consistent with mapping experiments for dopamine responsiveness. Our data suggest that ERK1/2 suppression is an obligatory step in the dopaminergic control of prolactin gene transcription and that bidirectional control of ERK1/2 function in the pituitary may provide a key mechanism for endocrine gene control. In pituitary lactotrophs the prolactin gene is stimulated by neuropeptides and estrogen and is suppressed by dopamine via D2-type receptors. Stimulatory signals converge on activation of the mitogen-activated protein kinases ERK1/2, but dopamine regulation of this pathway is not well defined. Paradoxically, D2 agonists activate ERK1/2 in many cell types. Here we show that in prolactin-secreting GH4ZR7 cells and primary pituitary cells, dopamine treatment leads to a rapid, pronounced, and specific decrease in activated ERK1/2. The response is blocked by D2-specific antagonists and pertussis toxin. Interestingly, in stable lines expressing specific pertussis toxin-resistant Gα subunits, toxin treatment blocks dopamine suppression of MAPK in Gαi2- but not Gαo-expressing cells, demonstrating that Go-dependent pathways can effect the inhibitory MAPK response. At the nuclear level, the MEK1 inhibitor U0126 mimics the D2-agonist bromocryptine in suppressing levels of endogenous prolactin transcripts. Moreover, a good correlation is seen between the IC50 values for inhibition of MEK1 and suppression of prolactin promoter function (PD184352 > U0126 > U0125). Both dopamine and U0126 enhance the nuclear localization of ERF, a MAPK-sensitive ETS repressor that inhibits prolactin promoter activity. In addition, U0126 suppression is transferred by tandem copies of the Pit-1-binding site, consistent with mapping experiments for dopamine responsiveness. Our data suggest that ERK1/2 suppression is an obligatory step in the dopaminergic control of prolactin gene transcription and that bidirectional control of ERK1/2 function in the pituitary may provide a key mechanism for endocrine gene control. D2-type receptors extracellular signal-regulated kinase mitogen-activated protein kinase mitogen-activated protein kinase kinase thyrotropin-releasing hormone green fluorescence protein prolactin growth hormone Tris-buffered saline dopamine ETS-2 repressor factor beta-adrenergic receptor kinase rous sarcoma virus Dopaminergic activation of G-protein-coupled D2-type receptors (D2R)1 regulates a range of behavioral and locomotor functions in the brain and leads to tonic inhibition of prolactin synthesis and release from the anterior pituitary. Hyperprolactinemia is observed in mice with a targeted disruption of the D2R gene along with the hypertrophic expansion of the pituitary lactotroph population and formation of pituitary adenomas in older animals (1Kelly M.A. Rubinstein M. Asa S.L. Zhang G. Saez C. Bunzow J.R. Allen R.G. Hnasko R. Ben-Jonathan N. Grandy D.K. Low M.J. Neuron. 1997; 19: 103-113Abstract Full Text Full Text PDF PubMed Scopus (367) Google Scholar, 2Saiardi A. Bozzi Y. Baik J.H. Borrelli E. Neuron. 1997; 19: 115-126Abstract Full Text Full Text PDF PubMed Scopus (169) Google Scholar, 3Asa S.L. Kelly M.A. Grandy D.K. Low M.J. Endocrinology. 1999; 140: 5348-5355Crossref PubMed Google Scholar). Inhibition of prolactin synthesis by dopamine occurs at the transcriptional level (4Maurer R.A. Nature. 1981; 294: 94-97Crossref PubMed Scopus (194) Google Scholar) and is dependent on the proximal promoter region of the prolactin gene (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar, 6McChesney R. Sealfon S.C. Tsutsumi M. Dong K.W. Roberts J.L. Bancroft C. Mol. Cell. Endocrinol. 1991; 79: R1-R7Crossref PubMed Scopus (24) Google Scholar). This region also confers transactivation by multiple stimulatory pathways, including those involving cAMP/protein kinase A, calcium, phospholipases, protein kinase C, and MAPKs. It is generally held that by antagonizing the elevation of intracellular cAMP or calcium, D2R signaling may inhibit the transactivation functions of factors like Pit-1, ETS-domain proteins, or specific transcription co-activators. Although activation of MAPK cascades are known to have an important role in mediating stimulatory responses of the prolactin gene to growth factors (7Schweppe R.E. Frazer-Abel A.A. Gutierrez-Hartmann A. Bradford A.P. J. Biol. Chem. 1997; 272: 30852-30859Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar, 8Castillo A.I. Tolon R.M. Aranda A. Oncogene. 1998; 16: 1981-1991Crossref PubMed Scopus (43) Google Scholar), thyrotropin-releasing hormone (TRH) (9Wang Y.H. Maurer R.A. Mol Endocrinol. 1999; 13: 1094-1104Crossref PubMed Scopus (53) Google Scholar), and even estrogen (10Watters J.J. Chun T.Y. Kim Y.N. Bertics P.J. Gorski J. Mol Endocrinol. 2000; 14: 1872-1881Crossref PubMed Scopus (108) Google Scholar), the role of MAPK regulation in the dopaminergic suppression of prolactin has not been defined. Indeed, D2R stimulation activates MAPKs in a wide range of cultured cells, including COS (11Faure M. Voyno-Yasenetskaya T.A. Bourne H.R. J. Biol. Chem. 1994; 269: 7851-7854Abstract Full Text PDF PubMed Google Scholar), Balb-c/3T3 (12Ghahremani M.H. Forget C. Albert P.R. Mol. Cell. Biol. 2000; 20: 1497-1506Crossref PubMed Scopus (41) Google Scholar), Chinese hamster ovary (13Oak J.N. Lavine N. van Tol H.H. Mol. Pharmacol. 2001; 60: 92-103Crossref PubMed Scopus (116) Google Scholar), C6 glioma (14Luo Y. Kokkonen G.C. Wang X. Neve K.A. Roth G.S. J. Neurochem. 1998; 71: 980-990Crossref PubMed Scopus (94) Google Scholar), and tissues (e.g. brain slices (15Yan Z. Feng J. Fienberg A.A. Greengard P. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 11607-11612Crossref PubMed Scopus (189) Google Scholar, 16Calabresi P. Gubellini P. Picconi B. Centonze D. Pisani A. Bonsi P. Greengard P. Hipskind R.A. Borrelli E. Bernardi G. J. Neurosci. 2001; 21: 5110-5120Crossref PubMed Google Scholar) and lung epithelium (17Guerrero C. Lecuona E. Pesce L. Ridge K.M. Sznajder J.I. Am. J. Physiol. Lung Cell Mol. Physiol. 2001; 281: L79-L85Crossref PubMed Google Scholar)). This activation is generally blocked by the ADP-ribosylating agent pertussis toxin (12Ghahremani M.H. Forget C. Albert P.R. Mol. Cell. Biol. 2000; 20: 1497-1506Crossref PubMed Scopus (41) Google Scholar, 13Oak J.N. Lavine N. van Tol H.H. Mol. Pharmacol. 2001; 60: 92-103Crossref PubMed Scopus (116) Google Scholar, 14Luo Y. Kokkonen G.C. Wang X. Neve K.A. Roth G.S. J. Neurochem. 1998; 71: 980-990Crossref PubMed Scopus (94) Google Scholar), indicating a requirement for heterotrimeric Gi/o-type proteins, and in some cases by the C-terminal sequence of βARK kinase (12Ghahremani M.H. Forget C. Albert P.R. Mol. Cell. Biol. 2000; 20: 1497-1506Crossref PubMed Scopus (41) Google Scholar) or Gα subunit of retinal transducin (11Faure M. Voyno-Yasenetskaya T.A. Bourne H.R. J. Biol. Chem. 1994; 269: 7851-7854Abstract Full Text PDF PubMed Google Scholar), consistent with a role for Gβ/γ subunit dimers in stimulatory D2R signaling. Because a stimulatory effect of D2R agonists on MAPKs appears inconsistent with their inhibitory actions on prolactin gene transcription, we examined how D2R activation alters MAPK function in prolactin-secreting cells. We show here that in the pituitary cell line GH4ZR7, dopamine treatment lowers constitutive and hormone-stimulated levels of activated MAPKs, ERK1 and ERK2. The inhibitory response is rapid and dependent on specific heterotrimeric G-proteins and specific MAPK types in that p38 MAPKs are not regulated in a similar manner to ERKs. The effects of MAPKK (MEK1) inhibitors on prolactin transcription parallel those of dopamine and are dependent in part on Pit-1 and ETS-type transcription factors. Finally, dopaminergic inhibition of ERK function is not restricted to transformed pituitary cell lines but is observed also in normal primary pituicytes, suggesting a physiological role for this regulatory mechanism. MEK1 inhibitors PD98059, PD184352, U0126, U0125, and pertussis toxin were purchased fromCalbiochem. Dopamine, bromocryptine, sulpiride, spiperone, and sorbitol were from Sigma. TRH was from Roche Molecular Biochemicals. The luciferase reporter plasmid −422 rPRL-Luc, rGH-Luc, RSV-Luc, 3x1P-Luc, and 3xSp1-Luc constructs were described previously (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar, 18Lew A.M. Yao H. Elsholtz H.P. J. Biol. Chem. 1994; 269: 12007-12013Abstract Full Text PDF PubMed Google Scholar). GFP-ERF fusion protein expression vector was prepared by in-frame insertion of the ERF cDNA sequence into pEGFP-C1 (CLONTECH). GH4ZR7 cells were maintained in Ham's F-10 with 12.5% horse serum and 2.5% fetal calf serum. Transfections were done as previously described (18Lew A.M. Yao H. Elsholtz H.P. J. Biol. Chem. 1994; 269: 12007-12013Abstract Full Text PDF PubMed Google Scholar). For primary culture, the pituitaries were isolated from 3-month-old Sprague-Dawley rats, washed with ice-cold phosphate-buffered saline and Dulbecco's modified Eagle's medium, and resuspended in defined medium (Dulbecco's modified Eagle's medium, penicillin/streptomycin, 30 μg/ml putrescine, 1 μm hydrocortisone, 5 μg/ml insulin, 5 μg/ml transferrin, 0.375% bovine serum albumin, and 10 pm T3). The cells were separated mechanically by passing progressively through a Pasteur pipette, 18- and 23-gauge needles. Dispersed cells were plated onto poly-l-lysine-coated culture plates and incubated in defined media for 48 h before treatments. Pertussis toxin-insensitive Gαi/o mutants containing C-terminal Cys to Ser substitutions and cloned into expression vector pcDNA3 (Invitrogen) were kindly provided by Dr. Paul Albert, University of Ottawa) (12Ghahremani M.H. Forget C. Albert P.R. Mol. Cell. Biol. 2000; 20: 1497-1506Crossref PubMed Scopus (41) Google Scholar). GH4ZR7 cells were co-transfected with the mutant Gαi/o subunit constructs and pcDNA3.1/hygromycin vector using electroporation (500 μfarad capacitance, 280 volts) and cultured in Ham's F-10 medium (12.5% horse serum, 2.5% fetal bovine serum) containing 300 μg/ml hygromycin-B for 3–4 weeks. Antibiotic-resistant clones were picked (25 clones/transfection) and tested for expression of recombinant Gαi/o RNA transcripts using 32P-labeled probes that recognized 3′ non-coding sequences specific to the vector. Transcript-positive clones were assessed by Western blot for the presence of corresponding Gαi/o proteins. mRNA from GH4ZR7 cells was prepared using oligo-dT cellulose (Collaborative Biomedical Tech.). Blots were probed with random primer labeled ([32P]dATP) cDNAs for Gαi2, Gαo, PRL, GH, or tubulin as previously described (26Gutkind J.S. J. Biol. Chem. 1998; 273: 1839-1842Abstract Full Text Full Text PDF PubMed Scopus (692) Google Scholar). Cells from 6-cm dishes were harvested in 0.2 ml of radioimmune precipitation assay buffer, extract protein was quantified by BCA protein assay (Pierce, Rockford, IL), samples were resolved on SDS 12% polyacrylamide gels at 100 V, and proteins were transferred to nitrocellulose. Blots were incubated for 2 h in 5% nonfat dry milk in 1× TBS. The blots were then incubated overnight with primary antibody in fresh 5% nonfat dry milk in 1× TBS followed by a 1-h incubation with horseradish peroxidase-conjugated secondary antibody at room temperature. The peroxidase product was developed and exposed to Kodak Blue X-Omat film. ETS mutations of the prolactin promoter were generated by using the PCR method of Kammanet al. (19Kammann M. Laufs J. Schell J. Gronenborn B. Nucleic Acids Res. 1989; 17: 5404Crossref PubMed Scopus (181) Google Scholar). First PCR used a wild-type primer specific for either the 5′-end (ccggctcgagcttttaatttaccca) or 3′-end (ggccaagcttgaccacacttccc) of the prolactin promoter and an ETS core mutagenic primer: −212, gattaattacagcaaaaatcgatgagagaaatgctg; −180, tagtggccagaaagtctagattttgattaattacag; and −160, ttctggccactatgagatcttgaatatgaataagaaat. The 150–200 base pair product was used in a second reaction to amplify a full-length promoter. The PCR product was restriction digested usingXhoI and HindIII and ligated into the luciferase-containing vector. Antibodies to ERK1/2, phospho-ERK1/2, p38, and phospho-p38 (Santa Cruz) were used to measure MAPK phosphorylation by Western analysis. ERK1/2 activity was measured using the p44/42 MAP Kinase Assay kit from Cell Signaling Technology. Immunoprecipitation was done with GH4ZR7 cell extract using immobilized phospho-p44/42 antibody. The precipitate was washed and used in kinase reactions with Elk-1 protein as substrate. The level of ERK activity was determined by Elk-1 phosphorylation in Western blot using anti-phospho-Elk-1 antibody. GH4ZR7 cells were transiently transfected with GFP-ERF expression vector and plated on cover slides coated with poly-l-lysine. After treatment, the cells were washed and fixed in 4% paraformaldehyde. Slides were prepared by coating the cells with 90% glycerol in phosphate-buffered saline and examined by confocal microscope (Fluoview BX50/PC system). GFP-ERF proteins were visualized using an argon ion laser at 488 nm. Regulation of MAPKs in D2R-expressing GH4ZR7 cells was determined by quantifying the activated (i.e. MAPKK-phosphorylated) form of the enzyme and by measuring the ability of immunoprecipitated MAPKs to phosphorylate the substrate ETS protein, Elk1. Initial studies showed that activated ERK1 and ERK2 are readily detected, and at surprisingly comparable levels, in GH4ZR7 cells cultured in serum-containing or serum-free medium for 24 h, even after removal of the weak estrogenic dye, Phenol Red. 2J. Liu, unpublished data. This serum-independent "basal" level of activated ERKs may derive from stimulatory factors released from (or expressed on) pituitary cells, as suggested by a biphasic pattern of phospho-ERK regulation. As shown in Fig. 1, phospho-ERK levels rapidly decline by 4–5-fold following serum withdrawal but recover to 70% control by 6 h post-withdrawal. Control cells (e.g.NIH3T3) treated in a similar manner showed minimal recovery in phospho-ERK levels over the same time course (Fig. 1). Dopamine regulation of phospho-ERK was examined under basal (serum-free) conditions and in the presence of ERK activators such as the hypothalamic peptide thyrotrophin-releasing hormone. In either case, phospho-ERK1/2 were suppressed 2–3-fold by brief exposure of cells to dopamine (Fig. 2, Aand B). This suppression was observed at dopamine concentrations previously shown to inhibit prolactin gene transcription (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar, 18Lew A.M. Yao H. Elsholtz H.P. J. Biol. Chem. 1994; 269: 12007-12013Abstract Full Text PDF PubMed Google Scholar), and was blocked completely by D2R-specific antagonists, sulpiride and spiperone (Fig. 2 A and data not shown). In contrast to ERK1/2, the stress-inducible MAPK, p38, was not inhibited by dopamine (Fig. 2 C), demonstrating selectivity in the MAPK response to D2R activation in GH4ZR7 cells. To further examine the cell context for dopaminergic suppression of ERK1/2, we measured the dopamine response in normal rat pituitary cells. Dispersed primary cultures were prepared in serum-free defined medium and treated with D2 agonists under time and dose conditions found effective using GH4ZR7 cells. In three separate experiments of similar design, dopamine or bromocryptine reduced phospho-ERK levels by 15–30% (Fig. 3), demonstrating that D2R-dependent regulation of MAPK in normal pituicytes parallels the response in the GH4ZR7 model. To assess whether suppression of ERKs by dopamine involves specific G-protein subtypes we examined the sensitivity of this response to pertussis toxin. Pretreatment of GH4ZR7 cells with pertussis toxin had no effect on the basal level of phospho-ERK1/2 or on stimulation of ERK1/2 by TRH, which signals predominantly via Gq-coupled receptors, but prevented dopamine-dependent suppression of ERK1/2 (Fig.4 A), indicative of a Gi/o-coupled response. Mutation of the terminal cysteine residue of Gαi/o-subunits renders them insensitive to ADP-ribosylation by pertussis toxin, providing a strategy to identify which Gi/o proteins are critical for coupling to specific signaling pathways. We established stable GH4ZR7 clones that express pertussis toxin-resistant forms of Gαi2 and Gαo and examined whether either Gα subtype was required for D2R-dependent inhibition of ERK1/2. RNA analysis using probes specific for the 3′-end of recombinant Gα transcripts, 3R. Baker, unpublished data. together with immunoblot data in Fig. 4 B (inset) identified several cloned GH4ZR7-PTXr lines that express the mutant Gα subtypes. Following pertussis toxin pretreatment and dopamine addition, Gαi2-PTXr-expressing cells behaved similarly to parental controls where suppression of ERK1/2 function by dopamine was completely blocked. In contrast, pertussis toxin was ineffective in blocking dopamine regulation of phospho-ERK levels in Gαo-PTXr-expressing cells (Fig. 4 B), indicating that this Gα subtype may be critical in coupling D2R activation to MAPK regulation. D2R activation in lactotrophs triggers several signaling events that may reduce prolactin synthesis, including a reduction in cAMP levels, inhibition of calcium channels, and a decrease in phosphatidylinositol turnover. Because dopamine potently suppresses basal ERK1/2 function in GH4ZR7 cells and primary pituitary cells, we investigated the impact of this regulatory mechanism on expression of the endogenous prolactin gene. Fig.5 shows that similar to bromocryptine, MEK1 inhibitors U0126 and U0125 cause a 2–3-fold reduction in prolactin RNA transcripts over a 48-h period. Expression of the prolactin-related growth hormone gene and tubulin control were unchanged in response to the treatments. The aminated phenylthiobutadiene U0126 is a more potent inhibitor of MEK1 than the closely related analog U0125 and also a better inhibitor of the endogenous prolactin gene (Fig. 5, at 48 h). To establish a more quantitative relationship between ERK1/2 suppression and decreases in prolactin gene transcription, we compared the ability of MEK1 inhibitors having a wide range of IC50 values for the suppression of ERK1/2 activation (Fig.6 A) to repress prolactin promoter function. Fig. 6 B shows there is a 100-fold range in the potency of PD184352, U0126, and U0125 to inhibit the prolactin promoter, in good agreement with the range and hierarchy of these compounds to block ERK1/2 activation (i.e. PD184352 > U0126 > U0125). Moreover, the selectivity of MEK1 inhibitors for the prolactin promoter is demonstrated by the dopamine-insensitiveRSV promoter (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar), which was unaffected by even the most potent MEK1 inhibitors (i.e. PD184352 and U0126) at concentrations exceeding 100 μm (Fig. 6 B). We have previously shown that ERF, a ubiquitous transcriptional repressor of the ETS-domain family can selectively inhibit the prolactin gene promoter by interacting at composite ETS/Pit-1-binding sites and potentially other Pit-1 sites (21Day R.N. Liu J. Sundmark V. Kawecki M. Berry D. Elsholtz H.P. J. Biol. Chem. 1998; 273: 31909-31915Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). In fibroblasts, ERF is a direct target for MAPK phosphorylation (22Sgouras D.N. Athanasiou M.A. Beal Jr., G.J. Fisher R.J. Blair D.G. Mavrothalassitis G.J. EMBO J. 1995; 14: 4781-4793Crossref PubMed Scopus (146) Google Scholar), and MAPK activation triggers export of the repressor from nuclei (23Le Gallic L. Sgouras D. Beal Jr., G. Mavrothalassitis G. Mol. Cell. Biol. 1999; 19: 4121-4133Crossref PubMed Scopus (90) Google Scholar), providing an attractive mechanism for de-repression of gene transcription. We examined by confocal microscopy whether a reduction in basal ERK1/2 function in dopamine- or U0126-treated GH4ZR7 cells could alter the subcellular location of ERF. In untreated cells, a GFP-ERF fusion protein was largely excluded from nuclei, contrasting with a uniformly distributed GFP control (Fig.7 A). Brief exposure of cells to either dopamine or U0126 caused a redistribution of GFP-ERF to nuclei within 10–30 min (Fig. 7 B). These agents had no effect on the distribution of GFP in control cultures,2indicating that ERF sequences were critical for regulating nuclear translocation. Quantification of localization data (Fig. 7 C) demonstrated that nuclear GFP-ERF was detected in 40% in dopamine-treated cultures. Nearly all cells showed nuclear localization of GFP-ERF following exposure to U0126. The dopamine-responsive promoter region of the prolactin gene includes ras-, TRH-, and MAPK-inducible elements that have been mapped to ETS-binding sites centered at −160 and −212 (24Bradford A.P. Conrad K.E. Tran P.H. Ostrowski M.C. Gutierrez-Hartmann A. J. Biol. Chem. 1996; 271: 24639-24648Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar, 25Howard P.W. Maurer R.A. J. Biol. Chem. 1995; 270: 20930-20936Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar). We mutated the core GGA(A/T) motifs of these ETS sites in a 450-base pair prolactin promoter. A third potential ETS motif positioned at −180, 3′ to the Pit-1-binding site 3P was also mutated (Fig. 8 A). As shown in Fig. 8 B, dopamine and U0126 inhibited activity of the wild-type prolactin promoter by 40 and 62%, respectively. However, a loss in responsiveness to dopamine and the MEK1 inhibitor was not seen following mutation of Ras/TRH-regulated ETS sites of the prolactin promoter. We have shown that a multimerized 1P Pit-1-binding site (coordinates −62 to −38) is sufficient to confer dopamine inhibition to a minimal TATA box promoter, whereas other binding sites (e.g. Sp1) are not regulated by dopamine (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar). Interestingly, as shown in Fig. 8 B, U0126 also inhibits activity of a 3x1P-TATA promoter but not a 3xSp1-TATA promoter, further demonstrating that dopamine signaling at the nuclear level in GH4ZR7 cells may involve targeting of the ERK1/2 pathway. Consistent with a role for Pit-1 sites in conferring inhibition by U0126, we found that the growth hormone promoter is also suppressed, albeit with lesser efficiency than the prolactin promoter. However, the inability of U0126 to lower steady state levels of growth hormone mRNA (see Fig. 3) argues that in a chromatin context transcriptional responses to MAPK suppression may depend on cooperative interactions that occur in the prolactin gene but not the growth hormone gene. This study demonstrates that dopamine D2R signaling in normal pituitary cells and prolactin-secreting cell lines leads to a reduction in ERK1/2 function. Dopamine not only antagonizes stimulatory effects of exogenous hormones (e.g. TRH) on ERK1/2, but it also suppresses basal levels of activated ERK1/2 observed in serum-free cultures of GH4ZR7 cells and primary pituitary cells. The biphasic pattern of phospho-ERK1/2 regulation, in which an initial sharp decline in phospho-ERK levels under serum-free conditions is followed by a recovery phase, suggests that secreted or membrane-associated autocrine factors may contribute to the elevated basal levels of activated ERK in pituitary cells. Possible candidates may include one or more members of the fibroblast growth factor family that are expressed in GH4 cells and primary pituicytes as these can be potent activators of ERK1/2 in pituitary cultures (7Schweppe R.E. Frazer-Abel A.A. Gutierrez-Hartmann A. Bradford A.P. J. Biol. Chem. 1997; 272: 30852-30859Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). 4S. Ezzat, personal communication. Efforts to address this issue using immunoneutralization strategies are currently in progress. Although activation of ERK1/2 by Gi/o-protein-coupled receptors, including D2Rs, is now a well established paradigm in several cell-types and tissues (26Gutkind J.S. J. Biol. Chem. 1998; 273: 1839-1842Abstract Full Text Full Text PDF PubMed Scopus (692) Google Scholar, 27Luttrell L.M. van Biesen T. Hawes B.E. Koch W.J. Krueger K.M. Touhara K. Lefkowitz R.J. Adv. Second Messenger Phosphoprotein Res. 1997; 31: 263-277Crossref PubMed Scopus (97) Google Scholar), the mechanism of rapid inhibition of ERK1/2 by this group of receptors is less well understood. In GH4ZR7 cells, dopamine inhibition of ERK1/2 could involve inhibitory effects on calcium channels or adenylate cyclase (18Lew A.M. Yao H. Elsholtz H.P. J. Biol. Chem. 1994; 269: 12007-12013Abstract Full Text PDF PubMed Google Scholar, 28Albert P.R. Neve K.A. Bunzow J.R. Civelli O. J. Biol. Chem. 1990; 265: 2098-2104Abstract Full Text PDF PubMed Google Scholar, 29Vallar L. Muca C. Magni M. Albert P. Bunzow J. Meldolesi J. Civelli O. J. Biol. Chem. 1990; 265: 10320-10326Abstract Full Text PDF PubMed Google Scholar), as agents that stimulate either calcium influx/PKC or adenylate cyclase/PKA can also stimulate ERK1/2 phosphorylation. Interestingly, two earlier studies using PTX-resistant Gα proteins in GH4 cells (20Senogles S.E. J. Biol. Chem. 1994; 269: 23120-23127Abstract Full Text PDF PubMed Google Scholar) and fibroblasts (12Ghahremani M.H. Forget C. Albert P.R. Mol. Cell. Biol. 2000; 20: 1497-1506Crossref PubMed Scopus (41) Google Scholar) demonstrate a requirement for Gi2 in the D2R-mediated inhibition of cAMP. From our study the ability of PTX-resistant Gαo, but not PTX-resistant Gαi2, to rescue dopamine suppression of ERK1/2 suggests that pathways independent of cAMP inhibition may play a significant role. This finding may be of particular interest in understanding transcriptional inhibition of the prolactin gene, as we have previously demonstrated that a GTPase-deficient Gαomutant inhibits prolactin promoter function without causing a decrease in intracellular cAMP (18Lew A.M. Yao H. Elsholtz H.P. J. Biol. Chem. 1994; 269: 12007-12013Abstract Full Text PDF PubMed Google Scholar). Inhibitory control of ERK1/2 by dopamine may involve the regulation of specific phosphatases, either dual specificity enzymes that directly target MAPKs or phospho-Ser/Thr or phospho-Tyr phosphatases that might act earlier in the signaling cascade. Florio et al. (30Florio T. Pan M.G. Newman B. Hershberger R.E. Civelli O. Stork P.J. J. Biol. Chem. 1992; 267: 24169-24172Abstract Full Text PDF PubMed Google Scholar) reported that dopamine can rapidly stimulate a phospho-Tyr phosphatase activity in GH4ZR7 cell membranes, an effect blocked by the antagonist haloperidol and sensitive to pertussis toxin. Activation of somatostatin receptors was unable to stimulate the PTPase activity (30Florio T. Pan M.G. Newman B. Hershberger R.E. Civelli O. Stork P.J. J. Biol. Chem. 1992; 267: 24169-24172Abstract Full Text PDF PubMed Google Scholar), suggesting functional differences in the signaling pathways evoked by these Gi/o-coupled receptors in GH4ZR7 cells. Among phospho-Ser/Thr phosphatases the ubiquitous PP1 is a possible effector in D2R-mediated ERK1/2 suppression. Inhibition of PP1 can stimulate ERK1/2 signaling in certain prolactin-secreting cell lines (31Manfroid I. Martial J.A. Muller M. Mol. Endocrinol. 2001; 15: 625-637Crossref PubMed Scopus (19) Google Scholar) most likely by preventing dephosphorylation of an upstream component in the kinase cascade. A reversal of PP1 inhibition by dopamine could lead to a reduction in activated ERK1/2. Moreover, a PP1- and actin-binding protein, spinophilin/neurabin II (32Allen P.B. Ouimet C.C. Greengard P. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9956-9961Crossref PubMed Scopus (394) Google Scholar, 33Satoh A. Nakanishi H. Obaishi H. Wada M. Takahashi K. Satoh K. Hirao K. Nishioka H. Hata Y. Mizoguchi A. Takai Y. J. Biol. Chem. 1998; 273: 3470-3475Abstract Full Text Full Text PDF PubMed Scopus (197) Google Scholar), has recently been identified in yeast two-hybrid screens as a target for the D2R third intracellular loop (34Smith F.D. Oxford G.S. Milgram S.L. J. Biol. Chem. 1999; 274: 19894-19900Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar), providing a further link between the D2R and PP1. However, although spinophilin is expressed at low levels in various cell types, including those in which D2Rs activate ERK1/2, its role in dopaminergic inhibition of ERK1/2 in lactotrophs remains to be tested. Promoter activity of the prolactin gene is strongly suppressed by MEK1 inhibitors having unique chemical structures with a ranking for suppression of promoter function that corresponds well to the IC50 values for MEK1 inhibition. Although interpretation of kinase inhibitor data is limited by the specificity of such compounds, it is noteworthy from a recent cross-analysis of multiple kinase inhibitors (35Davies S.P. Reddy H. Caivano M. Cohen P. Biochem. J. 2000; 351: 95-105Crossref PubMed Scopus (3957) Google Scholar) that MEK1 inhibitors (particularly PD184352) demonstrate remarkable target specificity relative to many other kinase inhibitors. In addition, given the prolactin promoter-specific effects of all MEK1 inhibitors tested in our study, it is unlikely that these compounds suppress transcription by a MEK1/ERK-independent mechanism. Hence, together with the RNA blot analysis, these data argue that ERK1/2 activity, whether stimulated by neuroendocrine hormones or maintained at elevated basal levels by autocrine/paracrine pituitary factors, may be required for prolactin gene expression, and thereby provides an effective target for inhibitory control by D2R signaling pathways. The mechanisms involved in transcriptional inhibition by dopamine and MEK1 inhibitors appear to include translocation of the ERF repressor to nuclei and regulation at Pit-1 sites of the prolactin promoter. Although in some fibroblast lines the ERK-sensitive repressor is localized to nuclei following serum withdrawal, requiring the addition of mitogens for nuclear export (23Le Gallic L. Sgouras D. Beal Jr., G. Mavrothalassitis G. Mol. Cell. Biol. 1999; 19: 4121-4133Crossref PubMed Scopus (90) Google Scholar), ERF in GH4ZR7 cells is restricted to the cytoplasm even after prolonged serum withdrawal. A similar distribution is seen in pituitary GHFT-1 cells. 5T. Voss and R. N. Day, personal communication. The ability of dopamine and the MEK inhibitor U0126 to trigger nuclear translocation of ERF, supports the view that D2R-dependent inhibition of ERK1/2 is a requirement for regulation of this transcription repressor. Although our previous data showed that ERF repression can be conferred by the 3P Pit-1/ETS composite site of the prolactin promoter (21Day R.N. Liu J. Sundmark V. Kawecki M. Berry D. Elsholtz H.P. J. Biol. Chem. 1998; 273: 31909-31915Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar), mutations of this site or a second Pit-1/ETS site (4P) were surprisingly unable to diminish the transcriptional response to dopamine or U0126. Other more proximal ETS elements may therefore be required, or alternatively, ERF may inhibit at non-composite Pit-1 sites as suggested by binding analysis of the prolactin 1P element (21Day R.N. Liu J. Sundmark V. Kawecki M. Berry D. Elsholtz H.P. J. Biol. Chem. 1998; 273: 31909-31915Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). Although the 1P element confers dopamine responsiveness (5Elsholtz H.P. Lew A.M. Albert P.R. Sundmark V.C. J. Biol. Chem. 1991; 266: 22919-22925Abstract Full Text PDF PubMed Google Scholar), it has not previously been considered a target for MAPK regulation based on studies of stimulatory signaling pathways in GH4 cells. Besides serving as a potential site for ERF-dependent repression, the 1P element likely plays a key role in the Pit-1-dependent recruitment of transcriptional coactivators. Dopamine inhibition of ERK1/2 activity may lead to impaired Pit-1/coactivator interactions with a consequent decrease in transactivation. In conclusion, we show that suppression of ERK1/2 activity by dopamine may play a key role in the negative regulation of the prolactin gene. This finding complements studies on the stimulatory control of prolactin, showing that ERK1/2 serves as an integrative node for diverse upstream signals including Gs- (36Le Pechon-Vallee C. Magalon K. Rasolonjanahary R. Enjalbert A. Gerard C. Neuroendocrinology. 2000; 72: 46-560Crossref PubMed Scopus (36) Google Scholar) and Gq-coupled receptors (9Wang Y.H. Maurer R.A. Mol Endocrinol. 1999; 13: 1094-1104Crossref PubMed Scopus (53) Google Scholar), receptor tyrosine kinases that activate ras-dependent (24Bradford A.P. Conrad K.E. Tran P.H. Ostrowski M.C. Gutierrez-Hartmann A. J. Biol. Chem. 1996; 271: 24639-24648Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar) or -independent (7Schweppe R.E. Frazer-Abel A.A. Gutierrez-Hartmann A. Bradford A.P. J. Biol. Chem. 1997; 272: 30852-30859Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar) pathways, and even steroid hormones (10Watters J.J. Chun T.Y. Kim Y.N. Bertics P.J. Gorski J. Mol Endocrinol. 2000; 14: 1872-1881Crossref PubMed Scopus (108) Google Scholar). Consistent with its inhibitory role in vivo, hypothalamic dopamine may reduce levels of activated ERKs to antagonize this stimulation or suppress basal prolactin gene transcription, which may be maintained in part by local pituitary-derived signals. Finally, although inhibition of ERK1/2 is generally viewed as a restorative mechanism that follows an acute or protracted stimulatory phase, our experiments support a model in which some Gi/o-coupled receptors, or as shown recently some receptor tyrosine kinases (37Miao H. Wei B.R. Peehl D.M., Li, Q. Alexandrou T. Schelling J.R. Rhim J.S. Sedor J.R. Burnett E. Wang B. Nat. Cell Biol. 2001; 3: 527-530Crossref PubMed Scopus (283) Google Scholar), cause a dynamic suppression of ERK1/2 with resultant changes in gene transcription or cell growth. We thank Drs. Richard Day and Ty Voss (University of Virginia) for GFP expression vectors, discussion, and communication of unpublished data, Drs. Paul Albert and Mohammad Ghahremani (University of Ottawa) for mutant Gα expression vectors, Drs. Sylvia Asa and George Fantus (University of Toronto) for providing rat pituitaries and primary culture reagents, and Drs. Peter Backx and Myron Cybulsky (University of Toronto) for use of confocal microscope systems.
Referência(s)