QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lo, R. K. H. Articles by Wong, Y. H. Search for Related Content PubMed PubMed Citation Articles by Lo, R. K. H. Articles by Wong, Y. H.

Signal Transducer and Activator of Transcription 3 Activation by the -Opioid Receptor via G₁₄ Involves Multiple Intermediates

Department of Biochemistry, Molecular Neuroscience Center, and Biotechnology Research Institute, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China

Received November 3, 2003; accepted February 26, 2004

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The hematopoietic-specific G₁₄ links a variety of G protein-coupled receptors to phospholipase C (PLC) stimulation. Recent studies reveal that several G subunits are capable of activating signal transducer and activator of transcription (STAT) proteins. In the present study, we investigated the mechanism by which G₁₄ mediates receptor-induced stimulation of STAT3. In human embryonic kidney 293 cells, coexpression of G₁₄ with -opioid receptor supported [D-Pen², D-Pen⁵]enkephalin (DPDPE)-induced STAT3 phosphorylations at both Tyr⁷⁰⁵ and Ser⁷²⁷ in a pertussis toxin-insensitive manner. The constitutively active G₁₄QL mutant also induced STAT3 phosphorylations at these sites and promoted STAT3-dependent luciferase activity. Requirements for PLC, protein kinase C (PKC), and calmodulin-dependent kinase II (CaMKII) in G₁₄QL-induced STAT3 activation were demonstrated by their respective inhibitors as well as by coexpression of their dominant-negative mutants. Inhibition of c-Src and Janus kinase 2 and 3 activities abolished STAT3 activation induced by G₁₄QL, but no physical association between G₁₄QL and c-Src could be detected by coimmunoprecipitation. Various intermediates along the extracellular signal-regulated kinase signaling cascade were apparently required for G₁₄QL-induced STAT3 activation; they included Ras/Rac1, Raf-1, and mitogen-activated protein kinase kinase-1/2. In contrast, functional blockade of c-Jun N-terminal kinase, p38 mitogen-activated protein kinase, and phosphatidylinositol-3 kinase had no effect on G₁₄QL-induced responses. PLC, PKC, and CaMKII were shown to be involved in G₁₄QL-mediated c-Src phosphorylation. Similar results were obtained with human erythro-leukemia cells upon DPDPE treatment. These results demonstrate for the first time that G₁₄ activation can lead to STAT3 stimulation via a complex signaling network involving multiple intermediates.

It has become widely accepted that G subunits play critical roles in cell growth, proliferation, and differentiation by activating the MAPK cascades. Apart from MAPKs, signal transducers and activators of transcription (STATs), a family of latent cytoplasmic transcription factor, have also been shown to play crucial roles in these cellular events (Hirano et al., 2000). The regulatory mechanisms for STAT3 activation are rather complex. In response to cytokines, growth factors, or interferons, activated Janus kinases (JAKs) phosphorylate and recruit cytoplasmic STATs. Phosphorylated STATs dimerize and translocate to the nucleus where they bind to specific gene-promoter sequences and induce DNA transcription. In addition to JAKs, the nonreceptor Src tyrosine kinase has been reported to induce the activation of STATs. v-Src can tyrosine-phosphorylate and constitutively activate STAT3, resulting in the induction of STAT3-mediated gene expression (Turkson et al., 1998). Members of MAPKs such as the extracellular signal-regulated kinase (ERK) are also capable of activating STATs. For instance, erythropoietin-induced STAT3 Ser⁷²⁷ phosphorylation is mediated by ERK signaling cascades (Haq et al., 2002). Taken together, STAT3 activation is modulated by complex mechanisms involving multiple signaling intermediates such as JAKs, Src kinase, and ERK.

Although STATs are considered to be primarily regulated by cytokines and growth factors, their activities can also be modulated by GPCRs. Angiotensin II type I receptor has been shown to activate the JAK2/STAT3 signaling pathway in neonatal rat cardiac myocytes (McWhinney et al., 1997). The precise mechanism by which GPCRs tap into the JAK/STAT pathway is unclear, but there is ample evidence to support the involvement of G subunits. In NIH-3T3 cells expressing a dominant-negative mutant of G_i2, Src kinase activity and STAT3 tyrosine phosphorylation are inhibited, resulting in a reduction of v-fms-induced proliferation (Corre et al., 1999). In contrast, constitutively active G_o and G_i mutants induce neoplastic transformation of NIH-3T3 cells via the activation of Src and STAT3 pathways (Ram et al., 2000). These studies imply that G_i subfamily members may possess the ability to activate STAT3, possibly via Src. Our recent demonstration that G₁₆ can mediate the stimulatory actions on STAT3 by the ORL₁, N-formyl-methionyl-leucyl-phenylalanine, and C5a receptors (Lo et al., 2003; Wu et al., 2003) further suggested that G_q subfamily members may similarly regulate the activities of STATs. In the present study, we sought to investigate whether activated G₁₄ can stimulate STAT3 activity through intermediate signaling molecules.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Reagents. The human G₁₄ and G₁₄QL cDNAs were obtained from Guthrie Research Institute (Sayre, PA). The cDNAs of STAT3 and its dominant-negative mutants STAT3Y705F and STAT3S727A were gifts from Dr. Nancy C. Reich (State University of New York, NY). The origins of plasmids encoding the DOR, dominant-negative mutants of PI3K (PI3K-DN), and c-Src (c-Src-DN), Ras and Rac1 (wild-type, dominant-negative mutants, and constitutively activated mutants) were as described previously (Kam et al., 2003). The PKC and PKC dominant-negative mutant cDNAs, PKC-KR and PKC-KR, were kindly obtained from Dr. Bernard Weinstein (Columbia University, New York, NY). The luciferase reporter gene pSTAT3-TA-luc was purchased from BD Biosciences Clontech (Palo Alto, CA). Human embryonic kidney 293 cells were obtained from the American Type Culture Collection (Manassas, VA). Human erythro-leukemia (HEL) cells were purchased from German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). Cell culture reagents, including LipofectAMINE Plus, were from Invitrogen (Carlsbad, CA). Antisera against G₁₄ and c-Src were purchased from Gramsch Laboratories (Schwabhausen, Germany) and Santa Cruz Biochemicals (Santa Cruz, CA), respectively, and other antibodies were supplied from Cell Signaling Technology Inc. (Beverly, MA). All kinase inhibitors and their negative analogs were ordered from Calbiochem (Darmstadt, Germany). Epidermal growth factor (EGF) and [D-Pen²,D-Pen⁵]enkephalin (DPDPE) were purchased from Sigma Chemical (St. Louis, MO). Pertussis toxin (PTX) was from List Biological Laboratories Inc.(Campbell, CA). Protein G-agarose and dithiobis[succinimidylpropionate] (DSP) cross-linker were obtained from Pierce Chemical (Rockford, IL). The luciferase substrate and its lysis buffer were purchased from Roche Diagnostics (Mannheim, Germany).

Cell Culture and Transfection. HEL cells were cultured in RPMI 1640 at 5% CO₂ and 37°C with 10% fetal bovine serum, 50 units/ml penicillin, and 50 µl/ml streptomycin. HEK 293 cells were grown in Eagle's minimum essential medium (growth medium) at 5% CO₂ and 37°C with 10% fetal bovine serum, 50 units/ml penicillin, and 50 µl/ml streptomycin. HEK 293 cells were seeded onto 96-well microtiter plates at 15,000 cells/well and were maintained in the growth medium for 24 h before transfection. Cells were transiently cotransfected with various cDNAs using LipofectAMINE Plus reagents. The transfection cocktails in serum-free Opti-MEM medium (100 µl/well; Invitrogen) contained 0.2 µl of both PLUS and LipofectAMINE reagents, 10 ng of G proteins, small GTPases or the control vector cDNAs, and 0.1 µg of pSTAT3-TA-luc. Three hours after transfection, 50 µl of Opti-MEM medium containing 30% fetal bovine serum was added into the wells and cultured for another 30 h.

Luciferase Assay. After 30 h of transfection, HEK 293 cells were serum-starved for 24 h or treated with a different kinase inhibitor overnight. Growth medium was then removed. 25 µl of lysis buffer from the luciferase assay kit (Roche Diagnostics) was added to the wells and then gently shaken on ice for 30 min. A further 25 µl of lysis buffer and 25 µl of luciferase substrate were added to each well to initiate the reaction. Luciferase activity was measured in a Micro-LumatPlus LB96V luminometer (PerkinElmer Life and Analytical Sciences, Boston, MA).

Immunoblotting, Cross-Linking, and Immunoprecipitation Assays. HEK 293 cells were seeded on six-well plates at a density of 5 x 10⁵ cells/well and maintained in growth medium the day before transfection. Cells were transiently cotransfected with various cDNAs by means of LipofectAMINE Plus reagents. Forty-eight hours later, the transfectants were serum-starved or treated with different kinase inhibitors overnight. Cells were lysed in 150 µl of lysis buffer (50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 100 mM NaCl, 1% Triton X-100, 40 mM NaP₂O₇, 1 mM dithiothreitol, 200 µM Na₃VO₄, 0.7 µg/ml pepstatin, 4 µg/ml aprotinin, 100 µM phenylmethylsulfonyl fluoride, and 2 µg/ml leupeptin) and gently shaken on ice for 15 min. Supernatants were collected by centrifugation at 16,000g for 2 min. HEL cells were seeded at a density of 1 x 10⁶ cells using serum-free medium with or without PTX (100 ng/ml) treatment overnight. Cells were incubated with different kinase inhibitors for 30 min and then treated with 100 nM DPDPE for another 20 min at 37°C. Subsequently, cells were lysed by the same lysis buffer. Clarified lysates (40 µg) were resolved on 12% SDS-polyacrylamide gels and transferred to a nitrocellulose membrane (GE Osmonics, Westborough, MA). STAT3, phospho-STAT3-Tyr⁷⁰⁵, phospho-STAT3-Ser⁷²⁷, ERK, phospho-ERK, JNK, phospho-JNK, p38 MAPK, phospho-p38 MAPK, c-Src, phospho-c-Src, Akt, and phospho-Akt proteins were detected by specific primary antisera and horseradish peroxidase-conjugated secondary antisera. The immunoblots were visualized by chemiluminescence with the enhanced chemiluminescence kit from Amersham Biosciences Inc. (Piscataway, NJ), and the images detected in X-ray films were quantified by densitometric scanning using the Eagle Eye II still video system (Stratagene, La Jolla, CA). For cross-linking, transfectants were rinsed with phosphate-buffered saline twice and then incubated in the same buffer plus 0.5 mM DSP for 15 min at room temperature. After that, cells were rinsed as above, incubated 5 min in quenching solution (50 mM glycine in phosphate-buffered saline, pH 7.4), and lysed in RIPA buffer (25 mM HEPES, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 1 mM dithiothreitol, 200 µM Na₃VO₄, 0.7 µg/ml pepstatin, 4 µg/ml aprotinin, 100 µM phenylmethylsulfonyl fluoride, and 2 µg/ml leupeptin). For immunoprecipitation, cell lysates were incubated with an anti-Src antiserum (4 µg/sample) at 4°C overnight and then followed by incubation with 30 µl of protein G Agarose (50% slurry) at 4°C for 2 h. Immunoprecipitates were washed with 400 µl of RIPA buffer and then resuspended in 50 µl of RIPA buffer and 10 µl of 6x sample buffer and boiled for 5 min. The c-Src and G₁₄ in the immunoprecipitates were detected by specific primary antisera and horseradish peroxidase-conjugated secondary antisera.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Stimulation of STAT3 Induced by DPDPE-Activated DOR via G₁₄ in a Time-Dependent Manner. We have demonstrated previously that the G_i-linked DOR can efficiently stimulate PLC via G₁₄ in COS-7 cells (Ho et al., 2001). Hence, we began our investigation by examining the effects of DOR-mediated activation of G₁₄ on STAT3 activity. HEK 293 cells were transiently transfected with pcDNA1 (vector control), DOR, or DOR together with G₁₄. Transfectants were challenged with 100 nM DPDPE, a specific agonist of DOR, for different time periods up to 30 min. The phosphorylation status of STAT3 was then assessed by Western blot analysis using anti-phospho-STAT3 antisera, whereas total STAT3 expression was monitored by anti-STAT3 antiserum. As shown in Fig. 1A, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations were unaffected by DPEPE in transfectants expressing the DOR alone compared with the controls (vector and basal). Yet a 30-min treatment of these cells with DPDPE significantly suppressed cAMP accumulation but had no effect on STAT3 phosphorylation (data not shown), implying that DOR could not mediate STAT3 phosphorylation via endogenous G_i/o proteins. In cells coexpressing DOR and G₁₄, phosphorylation of STAT3 at Tyr⁷⁰⁵ and Ser⁷²⁷ was significantly stimulated upon treatment with DPDPE (Fig. 1A). These phosphorylations began at 5 min and were sustained up to 15 min, which then decreased to the basal level by 30 min (Fig. 1A). To confirm that DOR-stimulated STAT3 phosphorylations were indeed mediated by G₁₄, transfectants were preincubated with 100 ng/ml PTX to abolish signaling via the endogenous PTX-sensitive G proteins. As illustrated in Fig. 1A, the DPDPE-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations were insensitive to PTX treatment. The effectiveness of PTX was confirmed by its ability to abolish DPDPE-induced inhibition of forskolin-stimulated cAMP accumulation (data not shown). Taken together, Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations of STAT3 were induced by DPDPE-activated DOR via G₁₄ in a time-dependent manner.

View larger version (39K): [in this window] [in a new window]   Fig. 1. STAT3 stimulation induced by DPDPE-activated DOR via G14 in HEK 293 cells and HEL cells. A, HEK 293 cells were transiently transfected with pcDNA1, DOR, or DOR together with G14. Two days after transfection, the transfectants were serum-starved with or without PTX treatment (100 ng/ml) overnight and then incubated with 100 nM DPDPE at indicated times. B, HEL cells were seeded at a density of 1 x 106 cells using serum-free medium with or without PTX (100 ng/ml) treatment overnight. Cells were incubated with 100 nM DPDPE for the indicated times at 37°C. For Western blot analysis, cell lysates were immunoblotted with anti-phospho-STAT3-Tyr705, anti-phospho-STAT3-Ser727, or anti-STAT3 antiserum. The immunoblots shown represent one of three sets; two other sets yielded similar results. Results of densitometric analyses are shown below the immunoblots; data shown represent the mean ± S.E.M. from three separate experiments. , p < 0.05, significantly higher than the control (zero time point) by Dunnett's t test.

Because G₁₄ is mainly expressed in hematopoietic cells, we sought to investigate whether STAT3 phosphorylation can be stimulated by DOR via G₁₄ in HEL cells. The presence of G₁₄ in HEL cells was confirmed by immunodetection with a G₁₄-specific antiserum (data not shown). We treated HEL cells with 100 nM DPDPE for various durations and then determined STAT3 phosphorylation status with anti-phospho-STAT3 antisera. As indicated in Fig. 1B, DPDPE significantly activated the phosphorylation of STAT3 at both Tyr⁷⁰⁵ and Ser⁷²⁷ residues compared with the untreated cells, whereas total STAT3 expression was unaffected. The optimal phosphorylation time of STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ induced by DPDPE was between 20 and 30 min (Fig. 1B). Because DOR can transduce signals through both PTX-sensitive and -insensitive G proteins, the potential involvement of G_i proteins in regulating DPDPE-induced STAT3 phosphorylation was eliminated by pretreating the HEL cells with PTX (100 ng/ml) overnight. As shown in Fig. 1B, DPDPE remained fully capable of mediating STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations in PTX-treated cells. These observations implied that DPDPE-induced STAT3 phosphorylation was induced by PTX-insensitive G proteins, such as G₁₄ and G₁₆. The identity of the PTX-insensitive G protein is unlikely to be G_q because DOR is incapable of interacting with G_q (Liu et al., 2003).

Stimulation of STAT3 Induced by Constitutively Active G₁₄QL. Next we examined the mechanism by which G₁₄ activates STAT3. The glutamine 205 to leucine (Q205L) mutation in G₁₄ results in a reduction of its intrinsic GTPase activity and produces a constitutively active phenotype (Chan and Wong, 2000). It mimics the ligand stimulation of GPCRs and is useful for mapping out the downstream signaling mechanism. To verify whether STAT3 could be stimulated by G₁₄QL in the absence of GPCR activation, HEK 293 cells were transiently transfected with pcDNA1 (vector control), G₁₄, or G₁₄QL cDNAs. As shown in Fig. 2A, the expression level of total STAT3 was unaffected in cells transfected with G₁₄ or G₁₄QL compared with the vector control. In contrast to the vector control or G₁₄ transfectants, G₁₄QL significantly induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations (Fig. 2A). Because phosphorylated STAT3 is capable of stimulating gene transcription, we sought to determine whether G₁₄QL could enhance STAT3-dependent transcriptional activity. We performed reporter gene assays by cotransfecting the cells with a STAT3-dependent luciferase construct (pSTAT3-TA-luc). As expected, G₁₄QL enhanced STAT3-dependent transcriptional activity significantly compared with both vector control and G₁₄ (Fig. 2A). The observed luciferase activity was positively correlated with increasing amounts of G₁₄QL cDNA used in the transfection (Fig. 2B). In contrast, the STAT3-dependent transcriptional activity of wild-type G₁₄ remained around the basal level even at a cDNA concentration of 100 ng/ml (Fig. 2B). These results indicated that G₁₄QL enhances STAT3-dependent transcriptional activity in a cDNA dose-dependent manner.

View larger version (33K): [in this window] [in a new window]   Fig. 2. STAT3 stimulation induced by constitutively active G14QL. A, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL. B, 1 µg/ml pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells with varying concentrations of G protein cDNAs (up to 100 ng/ml cDNA and the total amount was corrected with pcDNA1). C, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL in combination with STAT3, STAT3Y705F, or STAT3S727A. Cell lysates were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle) or anti-STAT3 (bottom) antiserum. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results. For reporter gene assays, cell lysates were used to measure luciferase activities. The luminescence emitted was quantified as the relative luminescent unit (RLU). RLUs were expressed as a percentage of the pcDNA1 control. Data shown represent the mean ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than the corresponding pcDNA1 control by Dunnett's t test. #, p < 0.05, significantly attenuated compared with those obtained with STAT3-overexpressing cells by Dunnett's t test.

G₁₄QL-Induced STAT3 Stimulation Was Inhibited by STAT3 Dominant-Negative Mutants. To further confirm the phosphorylation and activation of STAT3 by G₁₄QL, we assessed the ability of tyrosine and serine dominant-negative mutants of STAT3 (STAT3Y705F and STAT3S727A) to suppress G₁₄QL-induced responses. HEK 293 cells were cotransfected with the cDNAs of pcDNA1, G₁₄, or G₁₄QL in combination with wild-type STAT3 or its dominant-negative mutants. Western blot analysis with anti-STAT3 antiserum revealed increased STAT3 protein expression in cells transfected with wild-type or dominant-negative mutants of STAT3 (Fig. 2C). Basal and G₁₄QL-induced Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations of STAT3 were elevated in STAT3-expressing cells compared with the corresponding controls (Fig. 2C). G₁₄QL-induced STAT3 Tyr⁷⁰⁵ phosphorylation, but not STAT3 Ser⁷²⁷ phosphorylation, was attenuated in STAT3Y705F-expressing cells (Fig. 2C). On the other hand, STAT3S727A reduced G₁₄QL-induced STAT3 Ser⁷²⁷ phosphorylation only and had no influence on STAT3 Tyr⁷⁰⁵ phosphorylation induced by G₁₄QL (Fig. 2C). In either case, the magnitudes of the suppressed responses were similar to those observed with the control cells, in which only endogenous STAT3 proteins were present. These results implied that G₁₄QL independently induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations. Despite the increased expression of STAT3 in wild-type STAT3 transfectants, G₁₄QL-induced luciferase activity was no different from that of the corresponding vector control and G₁₄ transfectants (Fig. 2C), presumably because other factors such as substrate availability became limiting. For cells coexpressing G₁₄QL and either STAT3Y705F or STAT3S727A, the STAT3-dependent luciferase activities were significantly lower than that of wild-type STAT3-coexpressing cells (Fig. 2C). This might be explained by the fact that simultaneous phosphorylations at Tyr⁷⁰⁵ and Ser⁷²⁷ are required for maximal stimulation of STAT3.

The Involvement of PLC Signaling Pathway in G₁₄QL-Induced STAT3 Stimulation. As a member of the G_q subfamily, G₁₄ activates PLC, which hydrolyzes phosphatidylinositol-4,5-bisphosphate into diacylglycerol and IP₃. In A10 cells expressing PTH/PTHrP receptors, transient overexpression of G₁₄ conferred the ability of PTHrP to increase intracellular Ca²⁺ and IP₃ formation (Maeda et al., 1996). These studies led us to examine whether PLC and its downstream effectors were required for G₁₄QL-induced STAT3 phosphorylation and activation. HEK 293 cells were thus treated with an inhibitor for either PLC (U73122 [GenBank] ) or CaMKII (KN62) at 10 µM. As illustrated in Fig. 3A, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations induced by G₁₄QL were significantly attenuated by U73122 [GenBank] and KN62 compared with the corresponding controls. Likewise, for the reporter gene assays, U73122 [GenBank] or KN62 inhibited G₁₄QL-induced STAT3 activation (Fig. 3A). To further confirm these data, HEK 293 cells were incubated with 10 µM U73343 [GenBank] or KN92, the inactive analog of U73122 [GenBank] or CaMKII, respectively. As shown in Fig. 3B, G₁₄QL-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations as well as STAT3 transcriptional activation were not suppressed by these inactive analogs, indicating that both PLC and CaMKII participated in the mediation of G₁₄QL-induced STAT3 phosphorylation and activation.

View larger version (19K): [in this window] [in a new window]   Fig. 3. The involvement of PLC-, CaMKII, and JAK2/3 in G14QL-induced STAT3 stimulation. pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL. A, transfectants were treated with 0.1% DMSO (v/v), 10 µM U73122 [GenBank] , 10 µM KN62, 100 µM AG490, or 100 µg/ml JAK3 inhibitor (WHI-P131) using serum-free growth medium overnight. B, transfectants were treated with 0.1% DMSO (v/v), 10 µM U73343 [GenBank] , 10 µM KN92, or 100 µg/ml JAK3 inhibitor negative control (WHI-P258). For reporter gene assays, STAT3-dependent luciferase activity was determined as described in Fig. 2. Data shown represent the mean ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly increased compared with pcDNA1 control by Dunnett's t test. For Western blot analysis, cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results.

Next, we explored the role of PKC in G₁₄QL-induced STAT3 phosphorylation and activation. HEK 293 cells were treated with a PKC-selective inhibitor, calphostin C (100 nM). Inhibition of PKC by calphostin C suppressed the ability of G₁₄QL to induce STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations as well as the induction of STAT3-dependent luciferase activity (Fig. 4A). In addition to calphostin C, two PKC dominant-negative mutants, PKC-KR and PKC-KR, were tested. We chose these isoforms because PKC and PKC can be taken as representative members of conventional Ca²⁺-dependent PKC and novel Ca²⁺-independent PKC, respectively. As demonstrated in Fig. 4B, both PKC-KR and PKC-KR abolished STAT3 activation induced by G₁₄QL compared with the corresponding controls. Taken together, G₁₄QL-induced STAT3 phosphorylation and activation seemed to be mediated via a signaling pathway involving PLC, CaMKII, and PKC.

View larger version (19K): [in this window] [in a new window]   Fig. 4. The involvement of PKC in G14QL-induced STAT3 stimulation. A, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL. Transfectants were treated with 0.1% DMSO (v/v) or 100 nM calphostin C (Cal C) using serum-free growth medium overnight. B, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL in combination with the dominant-negative mutants of PKC (PKC-KR) or PKC (PKC-KR). For reporter gene assays, STAT3-dependent luciferase activity was determined as described in Fig. 2. Data shown represent the mean ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than the corresponding pcDNA1 control by Dunnett's t test. #, p < 0.05, significantly attenuated by the dominant-negative mutants of PKC by Dunnett's t test. For Western blot analysis, cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results.

The Involvement of JAK2/3 and c-Src in G₁₄QL-Induced STAT3 Stimulation. There is ample evidence to support the notion that STAT activation by cytokines or growth factors is mediated via JAKs and Src family members. It is interesting that G_s and G_i can directly bind to the catalytic domain of c-Src and stimulate its kinase activity (Ma et al., 2000), thereby providing an avenue for GPCRs to regulate the JAK/STAT pathway. Indeed, G_i-coupled chemokine receptors have been shown to integrate G protein and JAK/STAT signals during chemotaxis (Soriano et al., 2003). These observations raised the likelihood that G₁₄QL-induced STAT3 phosphorylation and that JAKs and c-Src might mediate activation. To explore this possibility, transfectants were treated with selective inhibitors against JAK2 (AG490; 100 µM) or JAK3 (WHI-P131; 100 µg/ml). As shown in Fig. 3A, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations induced by G₁₄QL were almost completely inhibited by AG490 and WHI-P131. Likewise, both AG490 and WHI-P131 attenuated the G₁₄QL-induced STAT3-dependent luciferase activity (Fig. 3A), suggesting that JAK2/3 were involved in the activation of STAT3 by G₁₄QL. In contrast, 100 µg/ml WHI-P258 (the inactive analog of WHI-P131) failed to affect the G₁₄QL-induced responses (Fig. 3B).

Next, we investigated the role of c-Src in G₁₄QL-induced STAT3 phosphorylation and activation by treating the transfectants with a specific c-Src inhibitor, PP2. As indicated in Fig. 5A, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations in G₁₄QL-expressing cells were inhibited in the presence of PP2 (from 5 to 50 µM). PP2 had no effect on the endogenous STAT3 expression level (Fig. 5A, top) but reduced the amount of phosphorylated STAT3 at or greater than 10 µM, with the attenuation of Ser⁷²⁷ phosphorylation being more prominent. Furthermore, 25 µM PP2 attenuated STAT3-dependent luciferase activity in G₁₄QL-expressing cells compared with the corresponding controls (Fig. 5B). The specificity of PP2 was confirmed with the use of its inactive analog, PP3. G₁₄QL-induced STAT3 activation was not inhibited by 25 µM PP3 (Fig. 5B). To gain further insight into the involvement of c-Src in G₁₄QL-induced activation of STAT3, the effects of a c-Src dominant-negative mutant (c-Src-DN) on STAT3 phosphorylation and activation were examined. As shown in Fig. 5C, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations were attenuated in cells coexpressing c-Src-DN and G₁₄QL compared with those expressing G₁₄QL alone. Likewise, coexpression of c-Src-DN significantly inhibited the G₁₄QL-induced STAT3-dependent luciferase activity (Fig. 5C), further supporting the notion that G₁₄QL-induced STAT3 phosphorylation and activation were c-Src-dependent.

View larger version (34K): [in this window] [in a new window]   Fig. 5. The involvement of c-Src in G14QL-induced STAT3 stimulation. HEK 293 cells were transiently transfected with pcDNA1, G14 or G14QL cDNAs. A, transfectants were treated with indicated concentration of PP2. B, transfectants were treated with 0.1% DMSO (v/v), 25 µM PP2, or 25 µM PP3 overnight. C, HEK 293 cells were further cotransfected pcDNA1, G14, or G14QL in combination with c-Src dominant-negative mutant c-Src-K295R/Y527F (c-SrcDN). For Western blot analysis, cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results. For reporter gene assays, STAT3-dependent luciferase activity was determined as described in Fig. 2. Data shown represent the means ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than the corresponding pcDNA1 control by Dunnett's t test. D, after cross-linking, cell lysates from the transfectants were immunoprecipitated (IP) by c-Src antiserum (top). The immunoprecipitates were immunoblotted with G14 or c-Src antiserum. Aliquots of crude lysates were used to detect G14 and c-Src expression levels by Western blot analysis (bottom). E, transfectants were treated with 0.1% DMSO (v/v), 10 µM U73122 [GenBank] , 100 nM calphostin C (Cal C), or 10 µM KN62 using serum-free growth medium overnight. Cell lysates from the transfectants were immunoblotted with anti-phospho-c-Src (top) or anti-c-Src (bottom). Numerical values shown above the immunoreactive bands represent the relative intensities of c-Src phosphorylation compared with the basal.

Because it has been demonstrated that activated G_s and G_i (but not G_q, G₁₂, and G) can directly bind to and activate c-Src (Ma et al., 2000), perhaps G₁₄QL can also associate with c-Src directly. To test this idea, we performed cross-linking and coimmunoprecipitation experiments using cells transfected with pcDNA1, G₁₄, or G₁₄QL. Transfectants were treated with the cross-linker DSP before immunoprecipitation with a c-Src antiserum. As shown in Fig. 5D (top), neither G₁₄ nor G₁₄QL could be coimmunoprecipitated with c-Src despite clear evidence of c-Src being pulled down. The absence of G₁₄ immunoreactivity in the c-Src immunoprecipitates was not caused by a lack of expression of G₁₄ and G₁₄QL in the transfectants, because both subunits were clearly detectable in the crude lysates by a G₁₄-specific antiserum (Fig. 5D, bottom). Equivalent levels of endogenous c-Src were detected in the immunoprecipitates and the crude lysates (Fig. 5D). These findings indicated that, like G_q, no direct physical association between G₁₄QL and c-Src could be detected.

We next examined whether c-Src was downstream of PLC in response to stimulation by G₁₄QL. As shown in Fig. 5E, G₁₄QL significantly induced the phosphorylation of c-Src compared with the vector- and G₁₄-transfected cells. G₁₄QL-transfected cells were incubated with 10 µM U73122 [GenBank] , 100 nM calphostin C, or 10 µM KN62. G₁₄QL-induced c-Src phosphorylation was partially inhibited by U73122 [GenBank] , calphostin C, and KN62, whereas the total expression level of c-Src was unaffected (Fig. 5E). These observations suggested that G₁₄QL-induced c-Src phosphorylation required PLC, PKC, and CaMKII.

The Involvement of ERK in G₁₄QL-Induced STAT3 Stimulation. G proteins are important upstream regulators of numerous signal transduction pathways, including the MAPK cascades. The ERK, JNK, and p38 MAPK have all been shown to exert regulatory influence on STAT3 activity. Thus, we sought to determine whether MAPKs were required for G₁₄QL-induced STAT3 activation. We first established whether G₁₄QL can activate ERK, JNK, and p38 MAPK by using anti-phospho-MAPK antisera. All three subfamilies of MAPK became significantly phosphorylated in cells expressing G₁₄QL, in contrast to the corresponding control cells (Fig. 6A). Total amounts of endogenous ERK, JNK, and p38 MAPK were not affected by overexpressing G₁₄ or G₁₄QL (Fig. 6A). These observations indicated that G₁₄QL was capable of inducing the phosphorylation of ERK, JNK, and p38 MAPK.

View larger version (37K): [in this window] [in a new window]   Fig. 6. The involvement of ERK signaling cascades in G14QL-induced STAT3 stimulation. HEK 293 cells were transiently transfected with pcDNA1, G14, or G14QL. A, phosphorylations of ERK, JNK, and p38 MAPK (p38) in the cell lysates were determined by immunoblotting with anti-phospho-specific antisera against each of the three MAPKs (top). The result of the densitometry analysis is shown above the immunoblots; data shown represent the mean ± S.E.M. from three separate experiments. , p < 0.05, significantly higher than pcDNA1 control by Dunnett's t test. The bottom of each set of immunoblots represents total MAPKs present in the cell lysates. B, cells were treated with 0.1% DMSO (v/v), 10 µM SB202190, 10 µM SB203580, 10 µM Raf-1 kinase inhibitor, 10 µM U0126, or 10 µM U0124 using serum-free growth medium overnight. Cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. C, cells were incubated with 0.1% DMSO (v/v), 10 µM SB202190, or 10 µM SB203580 using serum-free growth medium overnight. Phosphorylations of JNK and p38 MAPK (p38) in the cell lysates were determined by immunoblotting with anti-phospho-specific antisera against each of the two MAPKs. For reporter gene assays, STAT3-dependent luciferase activity was determined as described in Fig. 2. Data shown represent the mean ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than pcDNA1 control by Dunnett's t test. For Western blot analysis, data shown represent one of three sets of immunoblots; two other sets yielded similar results.

To test whether one or more MAPKs were involved in G₁₄QL-induced STAT3 phosphorylation and activation, specific kinase inhibitors were used. Before determining the activation status of STAT3, cells transfected with pcDNA1, G₁₄, or G₁₄QL were pretreated with selective inhibitors (10 µM each) against Raf-1 and MEK1/2 (U0126), two upstream kinases in the ERK cascade. As shown in Fig. 6B, G₁₄QL-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations were attenuated in the presence of either inhibitor, whereas total STAT3 expression was unaffected. Moreover, both inhibitors significantly suppressed the STAT3-dependent luciferase activity in G₁₄QL-expressing cells (Fig. 6B). The specificity of U0126 was verified with the use of its inactive analog, U0124. G₁₄QL-induced STAT3 responses were not inhibited by 10 µM U0124 (Fig. 6B). Our findings demonstrated that the ERK signaling cascade participated in G₁₄QL-induced STAT3 phosphorylation and activation. We then used specific kinase inhibitors to determine whether JNK and p38 MAPK were also involved. Our findings demonstrated that neither SB202190 nor SB203580 at 10 µM inhibited Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations of STAT3 or STAT3-dependent luciferase activity in G₁₄QL-expressing cells (Fig. 6B), implying that both JNK and p38 MAPK were not required for G₁₄QL-induced STAT3 activation. In Fig. 6C, the control experiment indicated that G₁₄QL-induced JNK and p38 MAPK activations were completely suppressed by SB202190, whereas SB203580 only inhibited the activation of p38 MAPK. The lack of effect of JNK/p38 MAPK inhibitors further highlighted the importance of the ERK pathway for G₁₄QL-induced STAT3 activation.

The Involvement of Ras and Rac1 in G₁₄QL-Induced STAT3 Stimulation. A number of G subunits have recently been reported to regulate MAPKs via the monomeric GTPases such as Ras and Rac1 (Woo and Kim, 2002). It has previously been demonstrated that Rac is necessary for GPCR-induced activation of JAK/STAT signaling (Pelletier et al., 2003). It was therefore of particular interest to study the role of Ras and Rac1 in G₁₄QL-induced STAT3 phosphorylation and activation. HEK 293 cells were overexpressed with the cDNAs of pcDNA1, G₁₄, or G₁₄QL in combination with various mutants of either Ras or Rac1. Transfectants were then assayed for STAT3 activity. If the monomeric GTPases were required for G₁₄QL-induced STAT3 activity, their constitutively active mutants should facilitate the response, whereas their dominant-negative mutants should suppress it. As depicted in Fig. 7B, STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations were increased in cells transfected with the constitutively activated mutant of Ras (RasG12V) in combination with either G₁₄ or G₁₄QL. RasG12V alone was capable of inducing the phosphorylations of STAT3 (data not shown). On the contrary, a dominant-negative mutant of Ras (RasS17N) significantly reduced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations in G₁₄QL-expressing cells (Fig. 7B). Participation of Ras was further demonstrated with the reporter gene assay. RasG12V significantly stimulated STAT3-dependent luciferase activity in G₁₄-expressing cells compared with the vector or wild-type Ras transfectants (Fig. 7A). The dominant-negative RasS17N effectively abolished the G₁₄QL-induced STAT3 luciferase activity (Fig. 7A). Likewise, the constitutively activated mutant of Rac1 (Rac1G12V) stimulated STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations as well as STAT3-dependent luciferase activity when expressed alone (data not shown) or with G₁₄ (Fig. 7, A and B). The dominant-negative mutant of Rac1 (Rac1T17N) inhibited STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations as well as the STAT3-dependent luciferase activity in cells coexpressing G₁₄QL (Fig. 7, A and B). These results implied that both Ras and Rac1 were involved in G₁₄QL-induced STAT3 phosphorylation and activation.

View larger version (39K): [in this window] [in a new window]   Fig. 7. The involvement of Ras and Rac1 in G14QL-induced STAT3 stimulation. pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL in combination with the wild-type or mutant of Ras or Rac1. A, for reporter gene analysis, STAT3-dependent luciferase activity was determined as in the legend to Fig. 2. Data shown represent the mean ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than the pcDNA1 control by Dunnett's t test. B, for Western blot analysis, cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results.

Phosphoinositide-3 Kinase Is Not Required for G₁₄QL-Induced STAT3 Stimulation. Phosphoinositide-3 kinase (PI3K) family members control several cellular responses, including cell growth and survival. It has been demonstrated that the erythropoietin receptor activates MAPK via Ras/PI3K signaling in Chinese hamster ovary cells, whereas c-Src is capable of stimulating PI3K phosphorylation (Kumar et al., 2003). In addition, PI3K has been shown to play a critical role in the regulation of STAT3 Ser⁷²⁷ phosphorylation triggered by interleukin-2 (Fung et al., 2003). These established linkages between PI3K and the various signaling molecules (c-Src, Ras, and ERK) prompted us to examine the role of PI3K in G₁₄QL-induced STAT3 activation. We treated G₁₄QL-expressing cells with PI3K-specific inhibitors, 100 nM wortmannin or 10 µM LY294002. Neither wortmannin nor LY294002 had any inhibitory effect on G₁₄QL-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations in G₁₄QL-expressing cells (Fig. 8A). Similar expression level of endogenous STAT3 was detected by STAT3 antiserum in these cells (Fig. 8A). In addition, G₁₄QL-induced STAT3-dependent luciferase activity was not inhibited by the PI3K inhibitors compared with the corresponding control (Fig. 8A). The role of PI3K in G₁₄QL-induced STAT3 phosphorylation and activation was further characterized by using a dominant-negative mutant of PI3K (PI3KDN). As illustrated in Fig. 8B, PI3KDN did not affect STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations induced by G₁₄QL, and the dominant-negative mutant had no effect on the total amount of endogenous STAT3. Expression of PI3KDN did not suppress the STAT3-dependent luciferase activity in G₁₄QL-expressing cells (Fig. 8B). Because Akt is a downstream effector of PI3K, we were able to demonstrate that wortmannin or LY294002 treatment completely inhibited EGF-induced Akt phosphorylation but not ERK phosphorylation, confirming that the activity of PI3K was indeed inhibited under these conditions (Fig. 8C). These studies collectively indicated that PI3K was not involved in G₁₄QL-induced STAT3 phosphorylation and activation in HEK 293 cells.

View larger version (29K): [in this window] [in a new window]   Fig. 8. No involvement of PI3K in G14QL-induced STAT3 stimulation. A, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL. Transfectants were treated with 0.1% DMSO (v/v), 100 nM wortmannin (Wort), or 10 µM LY294002 (LY) using serum-free growth medium overnight. B, pSTAT3-TA-luc cDNA was transiently transfected into HEK 293 cells along with pcDNA1, G14, or G14QL in combination with PI3K dominant-negative mutant (PI3KDN). For reporter gene assays, STAT3-dependent luciferase activity was determined as described in Fig. 2. Data shown represent the means ± S.E.M. from four separate experiments performed in triplicate. , p < 0.05, significantly higher than the corresponding pcDNA1 control by Dunnett's t test. For Western blot analysis, cell lysates from the transfectants were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. C, HEK 293 cells were treated with 0.1% DMSO (v/v), 100 nM wortmannin, or 10 µM LY294002 and then incubated with EGF for 5 min. Phospho-Akt, Akt, phospho-ERK, and ERK proteins were detected by specific primary antisera. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results.

The Involvement of Various Signaling Intermediates in DPDPE-Induced STAT3 Phosphorylation in HEL Cells. Finally, we investigated the involvement of various signaling intermediates in DPDPE-induced STAT3 phosphorylation in HEL cells. HEL cells were pretreated with 100 ng/ml PTX overnight to eliminate the potential involvement of PTX-sensitive G_i proteins in regulating DPDPE-induced STAT3 phosphorylation. As already demonstrated in Fig. 1B, DPDPE was capable of inducing the phosphorylation of STAT3 at Tyr⁷⁰⁵ and Ser⁷²⁷ in PTX-treated HEL cells; however, the DPDPE-induced STAT3 phosphorylation was markedly inhibited by Raf-1 inhibitor, U73122 [GenBank] , AG490, calphostin C, U0126, and PP2 (Fig. 9). These findings implied that DPDPE-induced STAT3 phosphorylation in HEL cells required the participation of Raf-1, PLC, JAK2, PKC, MEK1/2, and c-Src. Because the same signaling intermediates apparently were required for DPDPE-induced STAT3 phosphorylation in both HEK 293 and HEL cells, it further supported the notion that G₁₄ may well be the PTX-insensitive G protein mediating DPDPE-induced STAT3 activation in HEL cells.

View larger version (28K): [in this window] [in a new window]   Fig. 9. The involvement of various signaling intermediates in DPDPE-induced STAT3 phosphorylation in HEL cells. HEL cells were seeded at a density of 1 x 106 cells using serum-free medium with 100 ng/ml PTX treatment overnight. Before 100 nM DPDPE treatment for 30 min, cells were preincubated with 0.1% DMSO (v/v), 10 µM Raf-1 inhibitor, 10 µM U73122 [GenBank] , 100 µM AG490, 100 nM calphostin C, 10 µM U0126, or 25 µM PP2 for 30 min. Cell lysates were immunoblotted with anti-phospho-STAT3-Tyr705 (top), anti-phospho-STAT3-Ser727 (middle), or anti-STAT3 (bottom) antiserum. Numerical values shown above the immunoreactive bands represent the relative intensities of STAT3 phosphorylation compared with the corresponding basal. The immunoblots shown represent one of three sets of immunoblots; two other sets yielded similar results.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The restrictive expression pattern of G₁₄ suggests that it may subserve specialized functions, such as regulation of hematopoietic cell differentiation. Because hematopoietic-specific G₁₆ can regulate gene transcription (Yang et al., 2001; Lo et al., 2003), it is plausible that G₁₄ may possess a similar ability. First, we demonstrated that DPDPE-stimulated DOR was capable of inducing PTX-insensitive STAT3 phosphorylation in the presence of G₁₄. Second, expression of G₁₄QL, but not G₁₄, induced STAT3 Tyr⁷⁰⁵/Ser⁷²⁷ phosphorylation and STAT3 transcriptional activation. Third, coexpression of STAT3 phosphorylation-resistant mutants revealed that G₁₄QL-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylation could occur independently. These observations illustrate that STAT3 activity can indeed be elicited by G₁₄ activation.

Many GPCRs have been shown to regulate STAT activities. Initial studies have implicated the involvement of JAK in angiotensin II AT1 receptor-induced STAT1/3 activation (McWhinney et al., 1997). Our results also demonstrate that JAK2/3 is required for G₁₄QL-induced STAT3 activation. Recent studies suggest a possible involvement of c-Src. STAT3 activation induced by G_i2 and G_o is Src-dependent (Corre et al., 1999; Ram et al., 2000), and JAKs can be directly tyrosine-phosphorylated by c-Src (Tsai et al., 2000). Our data indicate that G₁₄QL-induced STAT3 activation is inhibited by PP2 and c-Src-DN, implying that c-Src plays a role in G₁₄QL-induced STAT3 activation. Moreover, c-Src acts as a direct effector of G proteins because it can associate with activated G_s or G_i, but not G_q, G₁₂, and G (Ma et al., 2000). Like G_q, G₁₄QL was unable to associate with c-Src in immunoprecipitation assays, suggesting that G₁₄QL-induced c-Src activation relies on mechanisms other than direct physical association. Hence, some adapter proteins or protein kinases may be involved in G₁₄QL-induced c-Src activation. An adaptor protein, tetratricopeptide repeat 1 (TPR1), has indeed been demonstrated to interact with G₁₆ and G_q (Marty et al., 2003). Because extensive cross-talk between GPCRs and growth factor receptors is known to exist (Lowes et al., 2002), the Src/JAK/STAT3 pathway may represent yet another example of signaling cascades shared by two classes of receptors.

Our results also support a role for PLC cascade in G₁₄QL-induced STAT3 activation. PLC, CaMKII, and PKC-selective inhibitors, as well as coexpression of PKC and PKC dominant-negative mutants, significantly suppressed G₁₄QL-induced STAT3 activation. The fact that dominant-negative mutants of both PKC and PKC were effective might perhaps be explained by their binding to a common activator or effector. A previous study has shown that both PKC and PKC are required to mediate platelet-derived growth factor-induced ERK stimulation through the Grb2-associated binder-1 protein in vascular smooth muscle cells (Saito et al., 2002). It remains to be determined whether a similar mechanism applies to STAT3 activation. Although we did not investigate the contribution of atypical PKC in this signaling pathway, these PKC isoforms are unlikely to be required for G₁₄QL-induced STAT3 activation because their upstream activator, PI3K, did not participate in this signaling cascade (Fig. 8A). Apart from that, G₁₄QL was capable of inducing c-Src phosphorylation, which was partially suppressed by inhibitors against PLC, CaMKII, and PKC. Partial suppression of G₁₄QL-induced c-Src phosphorylation by these inhibitors implies that c-Src may play a secondary role in G₁₄QL-induced STAT3 activation. Nevertheless, these results provide clues as to how G₁₄QL can be indirectly linked to c-Src. In hypothalamic gonadotropin-releasing hormone neurons, gonadotropin-releasing hormone induces Src association with PKC and PKC, leading to Src-dependent transactivation of the EGF receptor (Shah et al., 2003). It has also been demonstrated in glomerular mesangial cells that c-Src can be activated via CaMKII (Wang et al., 2003). Hence, G₁₄QL can potentially stimulate the Src/JAK/STAT3 pathway through the PLC/PKC cascade (Fig. 10). If the PLC/PKC/CaMKII cascade constitutes a primary signal for G₁₄QL-induced STAT3 activation, other members of G_q subfamily may also activate STAT3. Indeed, we have demonstrated that several GPCRs can activate STAT3 in the presence of G₁₆ (Lo et al., 2003; Wu et al., 2003). It should be noted that the PLC/PKC/CaMKII cascade is not required for G_i2/o-induced STAT3 activation (Corre et al., 1999; Ram et al., 2000).

View larger version (44K): [in this window] [in a new window]   Fig. 10. Proposed signaling pathways leading to STAT3 activation by DPDPE-stimulated DOR via G14. The PLC and c-Src/JAK pathways stimulated by G14 converge at Ras and activate STAT3 via the Ras/Raf/MEK/ERK cascade. G14 may directly interact with TPR1 and in turn activate the ERK cascade. Full activation of STAT3 by the activated G14 is achieved by costimulation of c-Src/JAK- and ERK-dependent pathways. P in solid circle represents Tyr705 phosphorylation, whereas P in open circle represents Ser727 phosphorylation. Broken lines demonstrate that the protein regulations involve multiple steps. The experimental evidence identifying individual pathways and the interactions between their intermediates is described in the text.

Small GTPases seem to occupy a critical juncture in signaling network that leads to STAT3 activation. Ras and Rac1 act as molecular switches for numerous extracellular stimuli. In the present study, the inhibition of G₁₄QL-induced STAT3 activation by Ras and Rac1 dominant-negative mutants verified the importance of these GTPases. Because Ras is downstream of Src, JAK, and CaMKII (Abe and Berk, 1999; Odajima et al., 2000), it raises the possibility that G₁₄QL-stimulated c-Src/JAK and PLC pathways converge at Ras to activate STAT3 via the Ras/Raf/MEK/ERK cascade. Apart from Ras, Rac1 activation by c-Src has been demonstrated (Timokhina et al., 1998). Activated Rac1 induces JAK2 and STAT3 activation in COS-1 cells (Simon et al., 2000). ERK activation by Rac1 is not without precedence. It has been shown that ERK is situated downstream of Rac1 (Woo and Kim, 2002) and Rac1-induced ERK activation occurs in a PAK- and Raf-dependent manner (Tang et al., 1999). Because PAK is activated mainly by Rac1, it may represent a link between Rac1 and ERK cascade. Recent studies demonstrated that both Ras and Rac1 are involved in lysophosphatidic acid-induced ERK activation (Stahle et al., 2003). It is reasonable to suggest that G₁₄QL-induced stimulatory signals of Ras and Rac1 may converge at the level of ERK and ultimately activate STAT3.

There is a large body of evidence to illustrate the regulatory role of MAPKs in STAT3 activation. STAT3 activation can be induced by various components along the ERK, JNK, and p38 MAPK pathways (Haq et al., 2002). Although G₁₄QL can activate ERK, JNK, and p38 MAPK, only the ERK pathway is required for G₁₄QL-induced STAT3 activation (Fig. 6). Linkages to ERK from PLC cascades can be traced back to PKC and CaMKII, both of which can activate the Ras/Raf/MEK/ERK cascades (Yang et al., 2003). PKC is known to phosphorylate Raf-1 (Kolch et al., 1993), whereas CaMKII is required for ERK activation of vascular smooth muscle (Ginnan and Singer, 2002). The possibility of G₁₄QL activating Ras via TPR1 should also be considered. It is therefore plausible that ERK serves as a convergent point for multiple signals (e.g., Src, PLC, PKC, and CaMKII) that originate from G₁₄QL. In contrast, G_oQL-mediated STAT3 activation is ERK-independent (Ram et al., 2000). G_i and G_q subfamily members may used overlapping yet distinct mechanisms to stimulate STAT3, because they both require c-Src but differ in their dependencies on ERK. Another observation in the present study pertains to the lack of PI3K involvement. Recent studies have revealed that PI3K can phosphorylate STAT3 at Ser⁷²⁷ (Fung et al., 2003), and it acts downstream of Ras and c-Src (Kumar et al., 2003). It is surprising that PI3K inhibition in the present study could not abolish the ability of G₁₄QL to induce STAT3 activation, implying that PI3K does not participate in G₁₄QL-induced STAT3 activation.

The present study provides the first evidence that G₁₄ can link GPCR stimulation to STAT3 activation via complicated signaling pathways. Our data indicate that G₁₄QL-induced STAT3 activation involves multiple pathways ranging from PLC to c-Src/JAK to ERK. A rudimentary map for the signaling network can be constructed from the experiments described herein (Fig. 10). It is unclear why JNK and p38 MAPK do not participate in G₁₄QL-induced STAT3 activation even though they have been demonstrated to stimulate STAT3 (Turkson et al., 1999). Because the use of G₁₄QL eliminates any potential contributions by G, STAT3 activation by G₁₄-coupled receptors may exhibit different characteristics, especially when G can modulate a host of effectors. For GPCRs that activate both G_i/o and G₁₄ (e.g., DOR and ORL₁ receptors), their mechanisms for activating STAT3 will be more complicated because the two G protein subfamilies may use different pathways.

The biological implications of G₁₄-mediated STAT3 activation are currently unclear, but the present study might provide a framework for such explorations. Because STAT3 regulates gene transcription in cell proliferation and differentiation, it plays a prominent role in hematopoiesis (Coffer et al., 2000). Little is known regarding the capability of GPCRs to regulate hematopoiesis. Among GPCRs recognized by G₁₄, DOR is expressed in immune cells in which it modulates T-lymphocytes activation via PTX-sensitive G proteins (Sharp et al., 1998). However, the possibility that PTX-insensitive G proteins may also participate in the regulation of DOR-stimulated immune responses cannot be eliminated. In HEL cells, we illustrated that DPDPE-induced STAT3 Tyr⁷⁰⁵ and Ser⁷²⁷ phosphorylations are resistant to PTX, implying that these phosphorylations are G_i/o-independent. Because HEL cells express both G₁₄ and G₁₆ and that the latter can also interact productively with DOR, it remains to be demonstrated whether either or both G subunits are required for DPDPE-induced STAT3 phosphorylation. Nevertheless, DPDPE-mediated STAT3 phosphorylation required Raf-1, PLC