Activation of the Luteinizing Hormone b Promoter by Gonadotropin-releasing Hormone Requires c-Jun NH 2 -terminal Protein Kinase*

Regulation of the mitogen-activated protein kinase (MAPK) family by gonadotropin-releasing hormone (GnRH) in the gonadotrope cell line L b T2 was investi-gated. Treatment with gonadotropin-releasing hormone agonist (GnRHa) activates extracellular signal-regu-lated kinase (ERK) and c-Jun NH 2 -terminal kinase (JNK). Activation of ERK by GnRHa occurred within 5 min, and declined thereafter, whereas activation of JNK by GnRHa occurred with a different time frame, i.e. it was detectable at 5 min, reached a plateau at 30 min, and declined thereafter. GnRHa-induced ERK activation was dependent on protein kinase C or extracellular and intracellular Ca 2 1 , whereas GnRHa-induced JNK activation was not dependent on protein kinase C or on extracellular or intracellular Ca 2 1 . To determine whether a mitogen-activated protein kinase family cascade regulates rat luteinizing hormone b (LH b ) promoter activity, we transfected the rat LH b ( 2 156 to 1 7)-luciferase construct into L b T2 cells. GnRH activated the rat LH b promoter activity in a time-dependent manner. Neither treatment with a mitogen-activated protein ki-nase/ERK kinase (MEK) inhibitor, PD98059, nor cotransfection

GnRH, 1 a hypothalamic decapeptide, serves as a key regu-lator of the reproductive system. GnRH acts on anterior pituitary gonadotropes to stimulate the synthesis and release of the pituitary gonadotropins LH and FSH. The gonadotropins are subunit hormones, each containing noncovalently linked ␣and ␤-subunits (1,2). Within a species, the ␣-subunits are identical, while the ␤-subunits differ and confer the physiological specificity of the heterodimeric hormone. Each ␤-subunit as well as the common ␣-subunit is encoded by different genes on separate chromosomes. When GnRH binds to its seven-transmembrane receptor (3), it induces interaction of the receptor with the heterotrimeric G q protein, which leads to activation of phospholipase C and formation of inositol 1,4,5-triphosphate and diacylglycerol, leading to elevation of intracellular Ca 2ϩ and activation of protein kinase C (PKC) (4 -6).
Intracellular transmission of extracellular signals is mediated in large part by several groups of sequentially activated protein kinases, which are collectively known as the mitogenactivated protein kinase (MAPK) cascades. In growth factor signaling, the key elucidated MAPK cascade is the extracellular signal-regulated kinase (ERK). Recent evidence indicates that many G protein-coupled receptors can activate the ERK cascade (7)(8)(9)(10)(11). The signals transmitted through the ERK cascade lead to activation of a set of regulatory molecules that eventually initiates cellular responses such as growth and differentiation (12)(13)(14). Recently, it has been shown that GnRHa is capable of activating ERK in pituitary organ culture (15) and the ␣T3-1 gonadotrope cell line (16,17). However, the ERK cascade is not the only link between membrane receptors and their intracellular targets, and in the past few years several other ERK-like cascades have been identified (13). One of the most studied of these cascades is the Jun NH 2 -terminal kinase (JNK; also known as stress-activated protein kinase (SAPK); Refs. 18 and 19) cascade, which is known to be activated in response to cellular stresses such as apoptosis (18,20). ERK, JNK, and p38 (21) constitute the MAPK family. Recent data suggest that GnRH is capable of activating JNK (22) and p38 (23) in the ␣T3-1 gonadotrope cell line.
It was reported that GnRH induction and basal control of the ␣-subunit gene seem to occur through the PKC/ERK pathway, while induction of the LH␤ gene is dependent on calcium influx in the ␣T3-1 gonadotrope cell line, suggesting the differential stimulation of transcription of rat LH subunit genes by GnRH (24). However, the ␣T3-1 gonadotrope cell line does not express the LH␤ gene. Mellon and co-workers (25), using targeted oncogenesis in transgenic mice, have recently generated an immortal gonadotrope cell line (L␤T2). The cells of this line express the mRNA of GnRH receptor and of both the ␣and ␤-subunits of LH (26,27).
Taken together, these facts led us to examine whether GnRH stimulates the activity of ERK and/or JNK, and whether the respective cascades play a role in the transcriptional activation of the rat LH␤ gene in L␤T2 cells.
Cell Cultures-L␤T2 cells (26) were generously provided by P. Mellon (La Jolla, CA). Cells were cultured at 37°C in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum in a water-saturated atmosphere of 95% O 2 and 5% CO 2 .
Clone Selection-The dominant negative c-Jun (dnJun) expression plasmid pLHCc-JUN (S63A, S73A) was constructed as described previously (32). L␤T2 cells were transfected for 12 h in six-well tissue culture plates with 2 g of pLHCdnc-JUN (S63A, S73A) using Lipo-fectAMINE Plus (Life Technologies, Inc.) (36). Clone selection was performed by adding hygromycin to the medium at 200 g/ml final concentration 2 days after the transfection. After 3 weeks, several clones were isolated using cloning rings. Selected clones were then maintained in medium supplemented with hygromycin (100 g/ml), and only low passage cells (p Ͻ 10) were used for the experiments described here.
Assay of ERK Activity-Cells were incubated overnight in the absence of serum and then treated with various substances. They were then washed twice with phosphate-buffered saline and lysed in ice-cold HNTG buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl 2 , 1 mM EDTA, 10 mM sodium pyrophosphate, 100 M sodium orthovanadate, 100 mM NaF, 10 g/ml aprotinin, 10 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride) (37). The extracts were centrifuged to remove cellular debris, and the protein content of the supernatants was determined using the Bio-Rad protein assay reagent. Erk1 rabbit polyclonal antibody was bound to protein A-Sepharose beads, and 300 g of protein from the lysate samples was immunoprecipitated at 4°C for 2 h. The immunoprecipitated products were washed once in HNTG buffer, twice in 0.5 M LiCl, 0.1 M Tris, pH 8.0, and once in kinase assay buffer (25 mM HEPES, pH 7.2-7.4, 10 mM MgCl 2 , 10 mM MnCl 2 , and 1 mM dithiothreitol), and samples were resuspended in 30 l of kinase assay buffer containing 10 g of myelin basic protein and 40 M [␥-32 P]ATP (1 Ci) as described previously (16). The kinase reaction was allowed to proceed at room temperature for 5 min and stopped by the addition of Laemmli SDS sample buffer (38). Reaction products were resolved by 15% SDS-PAGE.
Assay of 42-kDa ERK Activity Using a Transient Expression System-This activity was assayed as described (9,10,39). Briefly, L␤T2 cells cultured in 100-mm dishes were transfected with Myc-tagged p42 MAPK expression plasmid (1 g of pEXV-Erk2-tag) in combination with 9 g of pLNCX, pLNCX-MAPK (K3 M), pcDL-SR␣, or pcDL-SR␣-SAPK-VPF using LipofectAMINE Plus. At 72 h after transfection, serum-deprived cells were incubated with 1 M GnRHa for 5 min, and the expressed Myc-tagged p42 MAPK was immunoprecipitated with 1 g of antibody 9E10. The ERK activity in the immunoprecipitate was measured as described above.
Assay of JNK Activity-JNK activity was precipitated from 250 g of whole cell lysates by incubation with 2 g of GST-c-Jun-(1-89) fusion protein/GSH-Sepharose beads for 18 h at 4°C (New England Biolabs; Ref. 19). c-Jun-(1-89) contains a high affinity binding site for JNK close to the NH 2 terminus; this site contains two phosphorylation sites at Ser 63 and Ser 73 . The beads were washed and resuspended in 50 l of kinase buffer containing 100 M ATP for 30 min at 30°C as described (39). The solid-phase kinase reaction was terminated by addition of Laemmli sample buffer, and phosphorylation of GST-c-Jun on Ser 63 was examined after SDS-PAGE and immunoblotting with anti-phospho(Ser 63 ) c-Jun antibody.
Assay of JNK Activity Using a Transient Expression System-L␤T2 cells cultured in 100 mm dishes were transfected with HA-tagged wild type SAPK/JNK expression plasmid (1 g of pcDL-SR␣-wt-SAPK) or HA-tagged dominant negative SAPK/JNK expression plasmid (1 g of pcDL-SR␣-SAPK-VPF) using LipofectAMINE Plus. At 72 h after transfection, serum-deprived cells were incubated with 1 M GnRHa for 30 min, and cell lysates were immunoprecipitated with anti-HA antibody. The expressed HA-tagged wild type SAPK/JNK or dominant negative SAPK/JNK was eluted with 1% SDS, and the JNK activity was measured as described above.
Rat LH␤ Promoter Assay-L␤T2 cells cultured in 24-well plates were transfected with the rat Ϫ156 to ϩ7 LH␤-luciferase construct and CMV-␤-galactosidase plasmid (to normalize for cell viability and transfection efficiency) in combination with the indicated plasmids using LipofectAMINE Plus. At 48 h after transfection, serum-deprived cells were incubated with 100 nM GnRH for the indicated times. In some of the experiments, cells were treated with 20 M PD98059 for 15 min before the addition of 100 nM GnRH. Cell extracts were prepared by lysing the cells with three sequential freeze-thaw cycles in a buffer containing 100 mM potassium phosphate, pH 7.8, and 10 mM dithiothreitol. Vigorous vortexing was used to enhance cell lysis. Unlysed cells and insoluble material were pelleted at 10,000 rpm for 10 min at 4°C. The supernatant volume was measured, and aliquots of the supernatant were used in the subsequent luciferase and ␤-galactosidase assays.
Luciferase was assayed as described previously (40). Briefly, the luciferase assay mixture contained 100 mM KPO 4 , pH 7.8, 1 mM dithiothreitol, 3.7 mM MgSO 4 , 530 M ATP, and 470 M luciferin plus 20 l of cell extract in a final volume of 100 l. Luciferin was added just before measuring light units, which were measured in duplicate during the first 40 s of the reaction at 25°C in a luminometer (41).
␤-Galactosidase was assayed as described previously (40). The ␤-galactosidase buffer contained 60 mM sodium phosphate, pH 7.5, 1 mM MgCl 2 , 0.80 mg/ml O-nitrophenyl-␤-␦Ϫgalactopyranoside, and 40 mM ␤-mercaptoethanol. A standard curve for 100 microunits to 2 milliunits of ␤-galactosidase was made with each assay. A 30-l aliquot of cell extract was incubated with assay buffer until color developed (30 -120 min), and the reaction was then stopped by adding Na 2 CO 3 to a final concentration of 625 mM. Absorbance was then read at 405 nm.
Luciferase light units were normalized relative to the activity of ␤-galactosidase. The control value was set at 1 and the data expressed as -fold stimulation relative to control. Data are expressed as the mean Ϯ S.E.
Statistics-Statistical analysis was performed by Student's t test, and p Ͻ 0.01 was considered significant. Data are expressed as the mean Ϯ S.E.

RESULTS
Activation of ERK and JNK by GnRH-To evaluate whether ERK is activated by GnRH in L␤T2 cells, cultured cells were exposed to GnRHa for the indicated times (Fig. 1A). Cell lysates were immunoprecipitated with anti-ERK antibody and examined for ERK activity by assaying the incorporation of 32 P into MBP. The GnRHa-dependent increase in ERK activity reached a plateau from 5 min through 10 min and rapidly declined thereafter. We next examined the effect of GnRH on the activation of JNK, which is a member of the MAP kinase family. Cultured cells were exposed to GnRHa for the indicated times and cell lysates were incubated with GST-c-Jun fusion protein, followed by precipitation and Western analysis using antiphospho-c-Jun antibody (Fig. 1B). The activation of JNK by GnRHa in L␤T2 cells was detectable at 5 min, reached a broad plateau from 30 min through 3 h, and declined thereafter. These results indicate that JNK activation by GnRHa was slower than ERK activation. We also found that GnRH activated both ERK and JNK, and that the time courses for ERK and JNK activation by GnRH were similar to that of GnRHa (data not shown).
Effect of Pertussis Toxin on GnRH-induced ERK and JNK Activation-We compared the mechanisms of ERK and JNK activation induced by GnRH. It has been shown that the receptor for GnRH (3-6) is a member of the superfamily of G proteincoupled receptors. To determine what type of G protein is coupled to the GnRH receptor, we pretreated L␤T2 cells with 100 ng/ml pertussis toxin (PTX) for 4 h in order to inactivate G i and G o proteins, and then treated the cells with 1 M GnRHa for 5 min ( Fig. 2A) or 30 min (Fig. 2B, upper panel). Whereas PTX clearly caused a decrease in GnRHa-induced ERK activation ( Fig. 2A, lane 7), PTX did not have a detectable effect on GnRHa-induced JNK (Fig. 2B, upper panel, lane 7) activation. Thus, although PTX-sensitive G proteins are partly involved in the effect of GnRHa on ERK activity, as previously reported (42), PTX-sensitive G proteins are not involved in the effect of GnRHa on JNK activity, as was also previously reported (22).
Role of PKC in Activation of ERK and JNK-Many G protein-linked receptors can mediate stimulation of ERK activity via the phospholipase C-dependent activation of PKC (43)(44)(45)(46). Activation of ERK (16,17) or JNK (22) by GnRH requires PKC in ␣T3-1 cells. Therefore, the role of PKC in GnRH-induced ERK ( Fig. 2A) or JNK (Fig. 2B) activation in L␤T2 cells was examined. Exposure of L␤T2 cells to PMA caused stimulation of ERK activity (Fig. 2A, lane 1). However, the ability of PMA to induce the activation of ERK does not necessarily mean that the PKC pathway is involved in GnRHa-induced ERK activa-tion, as has been shown in the case of norepinephrine-induced ERK activation in both adipocytes (47) and GT-1 GnRH neuronal cell lines (10). Whether PKC is indeed involved in GnRH signaling was determined using PKC depletion. Role of Extracellular and Intracellular Ca 2ϩ in ERK and JNK Activation-It has been reported that elevated Ca 2ϩ is necessary for GnRH-induced ERK activation in ␣T3-1 cells (16,17). We therefore evaluated the role of extracellular and intracellular Ca 2ϩ in the GnRH-induced ERK (Fig. 3A) and JNK (Fig. 3B) activation in L␤T2 cells. Elimination of extracellular Ca 2ϩ by treatment with 3 mM EGTA for 1 min or with 1 M nifedipine for 10 min clearly attenuated GnRHa-induced ERK activation (Fig. 3A, lanes 3 and 5), indicating that Ca 2ϩ influx is required for GnRHa-induced ERK activation. Moreover, treatment with either 50 M 1,2-bis(o-aminophenoxy)ethane-N,N,NЈ-tetraacetic acid-acetoxymethyl ester (BAPTA-AM) for 20 min to eliminate intracellular Ca 2ϩ (Fig. 3A, lane 6) or 3 mM EGTA for 15 min to eliminate extracellular and intracellular Ca 2ϩ (49) (Fig. 3A, lane 4) clearly attenuated GnRHa-induced ERK activation, indicating that intracellular Ca 2ϩ is also required for GnRHa-induced ERK activation. On the other hand, elimination of extracellular Ca 2ϩ by either treatment with 3 mM EGTA for 1 min or with 1 M nifedipine for 10 min or elimination of intracellular Ca 2ϩ by treatment with 50 M BAPTA-AM had no effect on GnRHa-induced JNK activation (Fig. 3B). These results suggest that GnRH-induced ERK acti- vation was dependent on extracellular and intracellular Ca 2ϩ , whereas GnRH-induced JNK activation was independent of extracellular and intracellular Ca 2ϩ in L␤T2 cells.
Stimulation of LH␤ Promoter Activity by GnRH-We sought to determine whether the ERK and/or JNK cascades are involved in the regulation of LH␤ synthesis induced by GnRH. A rat LH␤ promoter (Ϫ156 to ϩ7 bp)-luciferase reporter construct was transiently transfected into L␤T2 cells. As shown in Fig. 4, addition of 100 nM GnRH enhanced the luciferase activity in a time-dependent fashion. To examine whether the stimulation of the LH␤ promoter by GnRH is the result of activation of the ERK cascade, either PD98059, an inhibitor of MEK, or an expression vector, pLNCX-MAPK (K3 M), encoding a catalytically inactive form of MAPK (iMAPK), was used. PD98059 is relatively specific for MEK, with no inhibitory activity against a number of other serine/threonine and tyrosine kinases (50 -52). Although pretreatment with 20 M PD98059 for 15 min completely abolished the GnRHa-induced ERK activation (Fig. 5B), pretreatment with 20 M PD98059 had no effect on GnRH-induced LH␤ promoter activation (Fig. 5A). In addition, cotransfection with pLNCX-MAPK (K3 M) at doses up to 2.4 g had no effect on GnRH-induced LH␤ promoter activation (Fig. 5A), whereas cotransfection with pLNCX-MAPK (K3 M) completely abolished the GnRHa-induced ERK activation (Fig. 5C). These results suggest that the ERK cascade is not involved in the GnRH-induced LH␤ promoter activation. We next examined the involvement of the JNK cascade in the stimulation of the LH␤ promoter by GnRH. An expression plasmid that encodes a dominant negative SAPK/JNK (pcDL-SR␣-SAPK-VPF) was used to inhibit the JNK cascade (53). GnRHa-induced JNK activation in cells transfected with pcDL-SR␣-SAPK-VPF was clearly attenuated compared with that in cells transfected with pcDL-SR␣-wt-SAPK (Fig. 6B), confirming the negative effects of the expression of a dominant negative SAPK/JNK on the endogenous kinase in L␤T2 cells. Moreover, cotransfection with pcDL-SR␣-SAPK-VPF did not interfere with GnRHa-induced ERK activation (Fig. 6C), suggesting the specificity of action of the SAPK-VPF vector in L␤T2 cells. Cotransfection with pcDL-SR␣-SAPK-VPF significantly attenuated the GnRH-induced LH␤ promoter activation dose-dependently, whereas cotransfection with pcDL-SR␣ had no effect (Fig. 6A), suggesting that the JNK cascade is involved in the GnRH-induced LH␤ promoter activation.
Role of PKC and Ca 2ϩ in GnRH-induced LH␤ Promoter Activation-Since PKC and Ca 2ϩ are not involved in GnRHainduced JNK activation (Figs. 2 and 3), we next examined the effect of PKC and Ca 2ϩ on GnRH-induced LH␤ promoter activation. Pretreatment with 10 M GF 109203X for 10 min, 3 mM EGTA for 15 min, 1 M nifedipine for 10 min, or 50 M BAPTA-AM for 20 min had no effect on GnRHa-induced LH␤ promoter activation (Fig. 7). Thus, neither PKC nor Ca 2ϩ is involved in GnRHa-induced LH␤ promoter activation, just as they are not involved in GnRHa-induced JNK activation.
A c-Jun Transcription Factor Is a Nuclear Acceptor of the JNK Signaling Cascade-It has been demonstrated that JNK phosphorylates c-Jun and ATF-2 at the putative regulatory amino-terminal serine residues and thereby increases their transcriptional activities (18,19). Moreover, JNK has been reported to activate Elk-1, resulting in an increase in c-fos gene expression (54). The Ets domain transcription factor Elk-1 is a substrate for three distinct classes of MAP kinase family members (55)(56)(57)(58). In addition, Ets family binding sites have been identified in the rat LH␤ promoter between Ϫ156 and ϩ7. Therefore, we first examined whether Ets transcription factors are the nuclear acceptors for GnRH signaling. Members of the Ets transcription factor family contain a transactivation domain at the amino terminus and a highly conserved DNAbinding domain at the carboxyl terminus, and this latter domain defines the Ets family of transcription factors since it lacks homology to other DNA-binding motifs (59,60). To examine the functional role of Ets transcription factors in GnRHinduced activation of the LH␤ promoter, the effect of an expression plasmid that encodes a dominant negative Ets construct (pAPr-etsZ) was examined. The pAPr-etsZ construct encodes the highly conserved DNA-binding domain of c-Ets-2 protein devoid of the transactivation domain, and inhibits both Ets-1- and Ets-2-mediated responses (31) since Ets-1 and Ets-2 recognize the same DNA sequence motif (31,59). Cotransfection with pAPr-etsZ at doses up to 2.4 g had no effect on GnRHinduced transcriptional stimulation (Fig. 8A). We also examined the effect of an expression plasmid that encoded a truncated Ets-2 with dominant-negative activity (pRK5-ets-2⌬1-328). Cotransfection with pRK5-ets-2⌬1-328 also had no effect on GnRH-induced transcriptional stimulation (Fig. 8B). These results suggest that the nuclear acceptor for the stimulation of LH␤ promoter activity by GnRH is not a member of the Ets transcription factor family. Since the LH␤ promoter does not contain a consensus ATF/CREB site, we next examined whether c-Jun is involved as a nuclear acceptor of a JNK signal. A dominant negative inhibitor (61,62) of the JNK cascade, dnJun, was used to inhibit the phosphorylation of c-Jun. The dnJun mutant cannot be phosphorylated at the NH 2 -terminal serine residues owing to substitution of serines 63 and 73 by alanine, and the mutant consequently blocks the enhanced transactivation promoted by JNK-dependent phosphorylation of these sites (61,62). Thus, dnJun blocks c-Jun phosphorylation-dependent events of the JNK cascade (32,61,62). Cotransfection with a dnJun expression vector significantly attenuated the GnRH-induced LH␤ promoter activation (Fig. 9A). In addition, cotransfection of TAM-67, a well characterized transdominant negative inhibitor of AP-1 owing to a deletion of residues 2-122 (34), significantly attenuated the GnRH-induced LH␤ promoter activation in a dose-dependent manner (Fig. 9B). We further examined whether the LH␤ promoter was activated by GnRH in clonal lines of L␤T2 cells, which stably expressed a dominant negative inhibitor (61,62) of the JNK cascade, dnJun (dnJun-L␤T2). GnRH-induced ERK activation was not attenuated in dnJun-L␤T2 cells, suggesting that there is no cross-talk between the ERK and JNK cascades (Fig. 9C). Expression of dnJun significantly attenuated the GnRH-induced LH␤ promoter activation (Fig. 9D). These results suggest that c-Jun is involved in the GnRH-induced LH␤ transcriptional activation. DISCUSSION Both the biosynthesis and the secretion of the gonadotropins are under the regulation of GnRH. Previous studies indicated that the GnRH receptor couples the G proteins of the G q/11 family with phosphoinositide turnover and a resultant increase in intracellular calcium concentration and PKC activation, to stimulate secretion of LH and FSH (4 -6). However, the molecular mechanisms by which GnRH mediates its transcriptional effects remain largely unknown. It was reported that GnRHinduced activation and basal control of the ␣-subunit gene seem to occur through the PKC/ERK pathway, while induction of the LH␤ gene is dependent on calcium influx using ␣T3-1 cells, which do not express the LH␤ gene (24). The ␤-subunits confer the physiological specificity of the heterodimeric hormone. Thus, a systematic approach to identifying mechanisms of hormonal regulation of gonadotropin subunit gene expression has been hampered by the lack of an available cell line that expresses the ␣, LH␤, and FSH␤ genes in a regulated manner. An immortal gonadotrope cell line (L␤T2) which expresses mRNA for GnRH receptor and for both the ␣and ␤-subunits of LH, has recently been generated (26,27). Therefore, we examined the mechanism by which GnRH induces the biosynthesis of LH␤ using L␤T2 cells in this study. As in ␣T3-1 cells (17,22), GnRHa caused both ERK and JNK activation in L␤T2 cells.
Although it was reported that activation of ERK by GnRH contributes to stimulation of the ␣-subunit promoter (16,24,63), the role of JNK activation by GnRH has not been clarified hitherto. We present here the first evidence that a JNK cascade is necessary to elicit LH␤ promoter activity in a c-Jun-dependent mechanism.  2 and 3). Lysates of cells were assayed for ERK activity as described in the legend for Fig.  1A. C, cells were transfected with pLNCX (lanes 1 and 2) or pLNCX-MAPK (K3 M) (lanes 3 and 4) together with Myc-tagged p42 MAPK expression plasmid (pEXV-Erk2-tag) and, after 72 h, were stimulated with 1 M GnRHa for 5 min (lanes 2 and 4). Cell lysates were immunoprecipitated with anti-Myc antibody (A-Myc), and the immunoprecipitates were incubated with [␥-32 P]ATP in the presence of MBP, as described under "Experimental Procedures." After the reactions were stopped with Laemmli sample buffer, samples were subjected to SDS-PAGE and autoradiography.
Activation of ERK is induced by phosphorylation of both threonine and tyrosine residues of the enzyme as a result of successive stimulation of Ras, ERK kinase kinase which may be Raf-1, MEK kinase, or an alternative kinase, and MEK (12)(13)(14). Protein kinase C␣ activates Raf-1 by direct phosphorylation (64). Distinct pathways of G i -and G q -mediated ERK activation have been reported (65). The activation of G i -coupled receptors, such as oxytocin (8), prostaglandin F2␣ (9), endothelin-1 (11), and prolactin-releasing peptide (40), appears to be PTX-sensitive and PKC-independent. However, in the case of receptors coupled to G q , such as bombesin (43) and thyrotropin-releasing hormone (7), activation is thought to be secondary to stimulation of phosphatidylinositol 4,5-bisphosphate-phospholipase C, leading to production of inositol phosphate and diacylglycerol, with subsequent PKC-mediated stimulation of ERK. In the present study, pretreatment with PTX detectably blocked the GnRHa-induced ERK activation (Fig. 2) and apparent down-regulation of PKC by prolonged incubation with PMA attenuated the stimulation of ERK activity by Gn-RHa (Fig. 2), suggesting the involvement of both G i and G q protein in GnRHa-induced ERK activation.
One important downstream biochemical event that occurs after ligand binding to many growth-promoting receptors is the activation of members of the MAP kinase family, including ERK and JNK (21). The ERK cascade is strongly activated by growth and differentiation factors, and sustained activation is thought to be an important signal for promoting cell proliferation and differentiation (12)(13)(14). The JNK cascade is also activated by cellular stresses (18, 19). These observations suggest the existence of parallel cascades leading to activation of either ERK or JNK. Is the mechanism of GnRHa-induced ERK acti-vation different from that of GnRHa-induced JNK activation? In most cases (7, 10), PKC and Ca 2ϩ have been shown to stimulate ERK activity. However, in endothelin-1-stimulated Rat-1 cells, JNK but not ERK activation was inhibited by chelation of Ca 2ϩ and by down-regulation of PKC (67). Similarly, in cardiac myocytes, activation of JNK by angiotensin II was strongly suppressed by down-regulation of PKC or by chelation of intracellular Ca 2ϩ (68). On the other hand, in GN4 rat liver epithelial cells, angiotensin II activated JNK in a Ca 2ϩ -dependent, PKC-independent manner (69). In the present study, GnRHa-induced JNK but not ERK activation was independent of both extracellular Ca 2ϩ and intracellular Ca 2ϩ (Fig. 3). Moreover, GnRHa-induced ERK activation was PKCdependent, whereas JNK activation was PKC-independent (Fig. 2). The time course of JNK activation (Fig. 1B) in response to GnRHa was slower than that of ERK activation (Fig. 1A). Thus, the regulation of the JNK activation by GnRHa appeared to be different from that of the ERK activation.
Ca 2ϩ is a critical mediator of the induction of gonadotropin secretion by GnRH (5,70,71). Studies have shown that calcium ionophores and calcium channel agonists can stimulate gonadotropin secretion. The stimulatory actions of GnRH on LH and FSH secretion can be inhibited by calcium channel antagonists and culturing the secretory cells in calcium-free medium. Elimination of extracellular Ca 2ϩ by treatment with 3 mM EGTA for 1 min or 1 M nifedipine for 10 min or elimination of intracellular Ca 2ϩ by treatment with 50 M BAPTA-AM for 20 min did not abolish the GnRHa-induced activation of JNK (Fig. 3B). These results suggest that GnRHa-induced activation of JNK is independent of extracellular and intracellular Ca 2ϩ and does not seem to involve gonadotropin secretion. Although little is known regarding the role of the activation of the MAP kinase family in the biosynthesis of hormones, we recently showed that both ERK and JNK are necessary to elicit rat prolactin promoter activity by prolactin-releasing peptide in an Ets-dependent mechanism (40). GnRH-induced activation of the LH␤ promoter was not attenuated by either pretreatment with MEK inhibitor PD98059 or by cotransfection with an iMAPK construct (Fig. 5A). On the other hand, GnRHinduced activation of the LH␤ promoter was attenuated by cotransfection with a dominant negative SAPK/JNK construct (Fig. 6A). Moreover, the lack of involvement of PKC and Ca 2ϩ in GnRHa-induced LH␤ promoter activation (Fig. 7) is similar to that in GnRHa-induced JNK activation (Figs. 2 and 3). Thus, although GnRH induces the activation of both ERK and JNK, the JNK cascade seems to be required for the GnRH-induced LH␤ promoter activation, as in the case of tumor necrosis factor-␣ production in mast cells (72). In addition, since it was reported that stimulation of JNK by GnRH was mediated by c-Src in ␣T3-1 cells (22), we examined the effect of the Srcselective tyrosine kinase inhibitor herbimycin A on GnRHainduced JNK activation and LH␤ promoter activation. Pretreatment with 5 M herbimycin A for 18 h had no effect on GnRHa-induced JNK activation or LH␤ promoter activation (data not shown), suggesting that the mechanism of JNK acti- No transcription factors that bind to a GnRH-responsive region of the LH␤ promoter have been identified yet. JNKs phosphorylate two sites of the NH 2 -terminal transactivating domain of c-Jun (Ser 63 and Ser 73 ), ATF-2/CREB, and Elk-1, thereby increasing their transcriptional activities (18,19,54). Although Ets family binding sites have been identified in the rat LH␤ promoter between Ϫ156 and ϩ7, cotransfection with either pAPr-etsZ or pRK5-ets-2⌬1-328 had no effect on the GnRH-induced LH␤ promoter activation (Fig. 8). The proximal LH␤ promoter does not contain ATF/CREB sites. Since the proximal LH␤ promoter between Ϫ95 and Ϫ86 contains 75.3% homology with the consensus AP-1 sites (TGA(C/G)TCA) (73), it is conceivable that c-Jun could be involved as a nuclear acceptor of a JNK signal. Therefore, we used dnJun to examine whether c-Jun might be involved as a nuclear acceptor of the JNK signal. dnJun has been characterized and successfully used in a number of studies (32,61,62). In addition, overexpression of dnJun did not alter the enzyme activity of JNK, showing that the derivative acts at a point distal to JNK in the JNK signal transduction cascade, consistent with the inhibition of the AP-1 transactivation function as previously shown (61,62). Moreover, independent studies using well characterized antisense reagents complementary to the JNK-1 and JNK-2 isoforms confirm that dnJun specifically blocks the JNK cascade (74). Cotransfection of a dnJun expression vector significantly attenuated the ability of GnRH to induce LH␤ promoter activation (Fig. 9A), suggesting that c-Jun is a substrate for JNK in the GnRH-induced LH␤ promotor activation. In addition, cotransfection of TAM-67, a well characterized transdominant negative inhibitor of AP-1 owing to a deletion of residues 2-122 (34), significantly attenuated the GnRH-induced LH␤ promoter activation in a dose-dependent manner (Fig. 9B). Moreover, GnRH did not induce either the rat LH␤ promoter activity (Fig. 9D) in dnJun-L␤T2 cells. These results suggest that c-Jun seems to be a substrate for JNK, as in the case of the ceramide-induced cyclooxygenase-2 gene expression (75), and to be involved in the GnRH-induced LH␤ transcriptional activation.
The signaling cascades that couple the activation of GnRH receptor to transcription are not yet fully defined. Although the JNK cascade appeared to be necessary for the GnRH-induced LH␤ promoter activation, it is not clear whether the activation of the JNK cascade is sufficient to induce the activation. It remains to be determined whether other MAP kinase family members such as p38 or the newly described SAPK3 (21) are also involved in GnRH-induced LH␤ promoter activation. Although it was reported that activation of ERK by GnRH contributes to stimulation of the ␣-subunit promoter (16,24,63), the ERK cascade did not appear to be involved in the GnRHinduced LH␤ transcriptional activation in this study. What is the role of the ERK cascade in the biological response to GnRH in L␤T2 cells? Thus, the complete role of the MAP kinase family in the action of GnRH in gonadotrope remains to be explored. It was reported that GnRH induces pituitary cell differentiation (66). Therefore, apart from a contribution to mediating transcriptional responses to GnRH, either ERK or JNK activation may be associated with other yet unknown FIG. 9. dnJun inhibits GnRH activation of the rat LH␤ promoter. L␤T2 cells were transiently cotransfected with 0.4 g of the rat LH␤ (Ϫ156 to ϩ7)-luciferase construct and 0.04 g of an internal control, pCMV␤gal, with or without 1.2 g of pLHCX or pLHCdnJun (A) or 0.4, 0.8, or 1.2 g of TAM-67 (B), as indicated. In D, L␤T2 cells or dnJun-expressing L␤T2 (dnJun-L␤T2) cells were transiently cotransfected with 0.4 g of the rat LH␤ (Ϫ156 to ϩ7)-luciferase construct and 0.04 g of an internal control. After transfection, cells were treated with 100 nM GnRH for 24 h prior to harvesting. Luciferase activity was normalized relative to ␤-galactosidase activity, and the basal activity was set at 1.0. Data are expressed as the mean -fold activation Ϯ S.E. of six transfections. ** indicates p Ͻ 0.01 as compared with the respective control. In C, L␤T2 and dnJun-L␤T2 cells were treated with (lanes 2 and 4, respectively) or without 1 M GnRHa for 5 min (lanes 1 and 3, respectively). Lysates of cells were assayed for ERK activity as described in the legend for Fig. 1A. Autoradiograms of 32 P-labeled MBP are shown. cellular responses to GnRH, such as effects on long term maintenance of the gonadotrope phenotype.