INTRODUCTION

Angiotensin II (AII)¹ is a peptide hormone that exerts a variety of effects on the cardiovascular system. These include vasoconstriction, regulation of salt and fluid homeostasis, induction of gene expression, and promotion of growth in cardiac and vascular smooth muscle cells (1, 2, 3, 4, 5). Acting through a seven-transmembrane, G-protein coupled receptor, this octapeptide elicits numerous signal transduction pathways, including the elevation of intracellular Ca²⁺, stimulation of phospholipase C, protein kinase C, phospholipase A₂, phospholipase D, Raf-1 kinase, and MAP kinase (6, 7, 8, 9, 10, 11). More recently, AII has been shown to activate components of the JAK-STAT pathway, which include members of the JAK kinases and STAT family of transcription factors (12, 13, 14). Activation of these signal transduction pathways may collectively contribute to the ability of AII to cause hypertrophic and hyperplastic effects in cardiovascular cells (15).

The JAK-STAT pathway is initiated by the binding of cytokines and growth factors to their cognate cell surface receptors. Receptors that lack intrinsic tyrosine kinase activity (e.g. IL-6 and IFN-) cause tyrosine phosphorylation of STAT proteins in the cytoplasm through the activation of JAK kinases (16, 17), whereas receptors with intrinsic tyrosine kinase activity (e.g. epidermal growth factor) tyrosine phosphorylate STAT members through a JAK-independent pathway (18, 19). Tyrosine-phosphorylated STAT proteins form homo- or heterodimers and translocate to the nucleus to induce gene transcription. Recent evidence suggests that, in addition to tyrosine phosphorylation, some STAT proteins (Stat1 and Stat3) are also phosphorylated at serine residues, possibly through the action of MAP kinases (20, 21, 22, 23). Serine phosphorylation appears to enhance the transactivating potential of tyrosine-phosphorylated Stat1 and Stat3 (20, 21). Thus, two of the major signal transduction pathways, i.e. the MAP kinase pathway and the JAK-STAT pathway, appear to modulate the capacity of STAT proteins to induce gene transcription through phosphorylation.

Using CHO-K1 cells expressing AT_1A receptors (T3CHO/AT_1A), we recently reported the ability of AII to transiently inhibit IL-6-induced Stat3 tyrosine phosphorylation and subsequent formation of the DNA-protein complex, SIF-A (13). We also observed that the phorbol ester PMA, which activates PKC, mimics the actions of AII in inhibiting IL-6-induced SIF-A formation. However, in PKC down-regulated cells, PMA failed to inhibit the IL-6-induced SIF-A response, but the inhibitory actions of AII were unaffected (13). These data suggested that AII-dependent inhibitory actions were not mediated by PMA-sensitive isoforms of PKC. We hypothesized that differential activation of the MAP kinase cascade may be responsible for inhibition of the IL-6-induced Stat3/SIF-A response by AII and PMA. Therefore, inhibiting the MAP kinase pathway would prevent the inhibitory effect. In the present study, we demonstrate that AII and PMA induce high levels of MAP kinase activity, in contrast to the weak MAP kinase activation by IL-6. Inhibition of the MAP kinase pathway by treatment with the MAPKK1 inhibitor PD98059 significantly attenuated the AII- and PMA-mediated inhibition of IL-6-induced Stat3 tyrosine phosphorylation and SIF-A formation. These data suggest, for the first time, involvement of the MAP kinase pathway in the negative regulation of IL-6-induced Stat3 signaling by AII.

EXPERIMENTAL PROCEDURES

Cell culture media, fetal bovine serum, antibiotics, Geneticin, tissue culture flasks, and IL-6 were purchased from Life Technologies, Inc; AII was purchased from U. S. Biochemical Corp.; nitrocellulose membranes were purchased from Amersham Corp; polyvinylidene difluoride membrane and [-³²P] were purchased from DuPont NEN; polyclonal antibodies to Stat3 (C-20) and protein A/G-agarose were purchased from Santa Cruz Biotechnology; anti-phosphotyrosine antibodies were purchased from Upstate Biotechnology, Inc.; goat antirabbit IgG and rabbit antimouse IgG were purchased from Bio-Rad; and other chemicals were purchased from Sigma.

T3CHO/AT_1A cells (24) were grown in -minimal essential medium containing 10% fetal bovine serum and 200 µg/ml Geneticin antibiotic for 12-24 h, serum starved for 12 h in F-12 medium, and treated with one or more agents as indicated.

Preparation of nuclear extracts and electrophoretic mobility shift assays were performed as described previously (12, 13). The sequence of sis-inducing element (SIE) was described previously (13, 25).

Serum-starved cells were treated with various agents for the indicated times; then the cells were scraped and washed in phosphate-buffered saline and lysed in buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1 mM Na₃VO₄, 1 mM NaF, 1% aprotinin, 0.1% leupeptin, 0.5 mM phenylmethylsulfonyl fluoride, and 15% glycerol) for 15 min on ice. Protein concentrations were determined, and samples were stored at 80 °C. To 100 µg of total cell extract (in 100 µl), 2 volumes of lysis buffer (200 µl) were added and further diluted with 2 volumes of buffer A (200 µl) (10 mM Tris, pH 7.4, 50 mM KCl, 20% glycerol, 0.5 mM EDTA, 0.5 M dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 1 mM Na₃VO₄). To the samples, 500 ng each of two different anti-Stat3 antibodies (Santa Cruz Biotechnology and Upstate Biotechnology) were added and incubated at 4 °C on a rocker for 8 h. Protein A/G-agarose was added to the samples, and incubation continued for 4 h at 4 °C. Immunocomplexes were collected by centrifugation, washed with 1 ml of buffer A, and dissolved in SDS-sample buffer. The samples were analyzed by SDS-polyacrylamide gel (8%) electrophoresis, transferred to nitrocellulose, and probed with antiphosphotyrosine antibody as described previously (13). The blots were stripped and reprobed with monoclonal anti-Stat3 antibody (Transduction Laboratories) as described previously (13).

In-gel MAP kinase assays were performed as described previously (26). Ten µg of the lysate in SDS-loading buffer were loaded on a SDS-polyacrylamide mini gel (80 × 50 × 0.75 mm) copolymerized with 1 mg/ml myelin basic protein. After electrophoresis, the gel was sequentially incubated for 1 h each with 50 mM Tris, pH 8.0, containing 20% 2-propanol, followed by 50 mM Tris, pH 8.0, containing 5 mM dithiothreitol (buffer A). Denaturation and renaturation of the proteins were performed by incubating the gel, respectively, in buffer A containing 6 M guanidine HCl for 1 h and buffer A containing 0.04% Tween 40 for 16 h at 4 °C. The gel was pre-incubated with kinase assay buffer (40 mM HEPES, pH 8.0, 2 mM dithiothreitol, 0.1 mM EGTA, and 5 mM magnesium acetate) at room temperature for 30 min, followed by another incubation at the same temperature, in kinase assay buffer containing 10 µCi/ml [-³²P]ATP for 90 min. The gel was washed six times (20 min each) with 5% trichloroacetic acid and 1% sodium pyrophosphate and then dried and exposed to x-ray film or to a phosphor screen. Quantitations were made on an SF PhosphoImager (Molecular Dynamics) using the ImageQuant Program.

RESULTS

We demonstrated previously that in CHO-K1 cells stably transfected with the angiotensin AT_1A receptor (T3CHO/AT_1A), IL-6 induced nuclear Stat3 tyrosine phosphorylation and SIF-A formation. Supershift assays with antibodies to Stat3 demonstrated that Stat3 was a component of IL-6-induced SIF-A complex in these cells (13). In addition, pretreatment of T3CHO/AT_1A cells with AII resulted in an inhibition of the IL-6-induced nuclear Stat3 tyrosine phosphorylation and SIF-A formation (13). In the present study, we performed experiments to determine if this AII-mediated inhibition occurs in a concentration-dependent manner in total cell lysates. Serum-starved T3CHO/AT_1A cells were treated with different concentrations of AII for 25 min and then were stimulated with IL-6 for 15 min. Total cell lysates were prepared, immunoprecipitated with anti-Stat3 antibody, and immunoblotted with antiphosphotyrosine antibody. Fig. 1A demonstrates the concentration-dependent inhibition of the IL-6-induced Stat3 tyrosine phosphorylation by AII. Complete inhibition of the IL-6-induced Stat3 tyrosine phosphorylation by AII was observed at 5 nM or greater (Fig. 1A, lanes 3-6), and inhibition was 50% at 1 nM AII (Fig. 1A, lane 7). When the blot corresponding to Fig. 1A was stripped and reprobed with anti-Stat3 antibody, an equal amount of Stat3 was present in all the lanes (Fig. 1B), suggesting that AII causes the inhibition of Stat3 tyrosine phosphorylation without affecting the total Stat3 protein content.

We next determined if the inhibition of IL-6-induced Stat3 tyrosine phosphorylation in the total cell extract in Fig. 1 was reflected in the concentration-dependent inhibition of nuclear SIF-A activity. Nuclear extracts were prepared from cells treated with IL-6 alone or treated sequentially with AII and IL-6. The samples were analyzed in an electrophoretic mobility shift assay using c-fos regulatory element SIE. Fig. 1C demonstrates the concentration-dependent inhibition of IL-6-induced nuclear SIF-A response by AII. Densitometric analysis of the bands of Fig. 1C indicated that the inhibition of IL-6-induced SIF-A activity by AII was one-half maximal at a concentration of 1.3 nM (data not shown). This is similar to the K_d of AII-binding to the AT_1A receptor in T3CHO/AT_1A cells (24). The results obtained in Fig. 1 collectively demonstrate that: 1) AII potently inhibits the IL-6-induced Stat3 tyrosine phosphorylation and subsequent formation of SIF complex A; and 2) the inhibitory actions of AII do not involve the degradation of Stat3 protein.

Numerous studies indicate that the rate, magnitude, and duration of signals transmitted through the MAP kinase cascade may significantly contribute to the ability of different ligands to cause different end responses (27, 28, 29, 30, 31). It was shown that epidermal growth factor, by a transient MAP kinase activation, caused PC12 cells to proliferate, whereas in the same cells, NGF, through a sustained activation of MAP kinase, caused the cells to differentiate (reviewed in Ref. 27). To explain the inhibitory effects of AII and PMA on IL-6-induced Stat3 tyrosine phosphorylation/SIF-A formation, we hypothesized that: 1) the AII- and PMA-induced MAP kinase activation profile may be significantly different from that of IL-6 and that such differences in the MAP kinase activation profile may contribute to the inhibitory effects of AII and PMA; and 2) signals elicited by AII and PMA may converge at the level of MAPKK and, therefore, blocking the MAP kinase pathway at the level of MAPKK would attenuate the inhibitory effects. Using an in-gel kinase assay, we first determined the MAP kinase activation profile of AII and PMA and compared this to the MAP kinase activity stimulated by IL-6. Both AII (Fig. 2A, lanes 1-9) and PMA (Fig. 2B, lanes 1-6) rapidly induced high levels of MAP kinase activity (8-fold at 2 min), which was sustained throughout the time periods examined (2 h for AII and 30 min for PMA). In contrast, the IL-6-elicited response was weak (1.7-fold at 10 min) and transient, returning to basal levels within 30 min (Fig. 2A, lanes 10-15).

We reasoned that, either alone or in combination, the faster kinetics, higher magnitude, and sustained activation of MAP kinase by AII and PMA observed in Fig. 2 may play a role in the inhibition of IL-6-induced Stat3 tyrosine phosphorylation and subsequent SIF-A complex formation. To address this question, we used PD98059, a recently developed MAPKK1 inhibitor (32, 33, 34, 35), to block the AII- and PMA-induced MAP kinase pathway. The use of PD98059 to block activation of the MAP kinase pathway induced by a variety of agents, including platelet-derived growth factor, NGF, insulin, and PMA, has been reported recently (32, 33, 34, 35). This compound exhibits a high degree of selectivity for inhibiting MAPKK1 activation both in vivo and in vitro, without affecting the function of other serine/threonine and tyrosine kinases (32). Fig. 2C demonstrates the concentration-dependent activation of MAP kinase by AII (lanes 2-5) and its inhibition by 20 µM PD98059 (lanes 6-9). PD98059 completely inhibited the MAP kinase activity elicited by AII at 1 nM (Fig. 2C, lane 9). As shown previously for other agonists, at higher concentrations of AII (5 nM or greater), inhibition by PD98059 was partial (Fig. 2C, lanes 7 and 6), possibly due to the ability of AII to induce MAP kinase through multiple routes, which may involve MAPKK2 (15). The same concentration of PD98059 (20 µM) completely inhibited the MAP kinase activity elicited by IL-6 (Fig. 2C, lanes 10 and 11) and caused a 90% reduction in the PMA (100 nM)-induced MAP kinase activity (Fig. 2D, lanes 2 and 3).

We next determined if treatment of cells with PD98059 would attenuate the AII-mediated inhibition of IL-6-induced Stat3 tyrosine phosphorylation and nuclear SIF-A response. For these experiments, AII was used at 1 nM to inhibit the IL-6 response, because activation of MAP kinase at this concentration of AII was completely inhibitable by 20 µM PD98059. Total lysates were prepared from cells treated with IL-6, AII + IL-6, or PD98059 + AII + IL-6. Following immunoprecipitation with anti-Stat3 antibodies, samples were separated electrophoretically, transferred to nitrocellulose, and immunoblotted with antiphosphotyrosine antibody. Fig. 3A demonstrates that treatment of cells with 1 nM AII for 25 min resulted in a significant inhibition of Stat3 tyrosine phosphorylation, in particular the M_r 89,000 form (Fig. 3A, lane 3). Cells that were exposed to the MAPKK1 inhibitor PD98059, before the addition of AII and IL-6, showed no loss of tyrosine phosphorylation of either forms (M_r 89,000 and M_r 92,000) of Stat3 (Fig. 3A, lane 4). To determine if Stat3 protein was present in equal amounts in each lane, the blot in Fig. 3A was stripped and reprobed with anti-Stat3 antibody. Fig. 3B demonstrates that Stat3 is present in similar amounts in the immunoprecipitated samples of Fig. 3A. These results suggest that inhibition of MAPKK1 by PD98059 attenuates AII-mediated inhibition of IL-6-induced Stat3 tyrosine phosphorylation.

We substantiated the above results by determining the nuclear SIF-A response, using electrophoretic mobility shift assays. Nuclear extracts were prepared from cells treated with IL-6, AII + IL-6, and PD98059 + AII + IL-6 and subjected to electrophoretic mobility shift assays, using ³²P-labeled SIE. Fig. 4A demonstrates that AII pretreatment (1 nM) caused a 50% inhibition of the IL-6-induced SIF-A response (lanes 2 and 3). PD98059 significantly attenuated this inhibitory action of AII (Fig. 4A, lane 4). To determine if the intensities of SIF-A are reflected in the degree of tyrosine phosphorylation of Stat3, the nuclear extracts were immunoprecipitated with anti-Stat3 antibodies and immunoblotted with antiphosphotyrosine antibodies. As shown in Fig. 4B, a higher amount of tyrosine-phosphorylated Stat3 was present in cells pretreated with PD98059, before addition of AII and IL-6 (lane 4), suggesting that PD98059 attenuates the inhibitory actions of AII. To determine if the degree of tyrosine phosphorylation of Stat3 in these nuclear extracts was related to the amount of Stat3 protein, samples used in Fig. 4A were directly analyzed in a Western blot and probed with anti-Stat3 antibodies. Fig. 4C demonstrates that the Stat3 tyrosine phosphorylation observed in Fig. 4B correlates with the amount of Stat3 protein, consistent with tyrosine phosphorylation being required for nuclear translocation (16).

Pretreatment of T3CHO/AT_1A cells with PMA interferes with the IL-6-induced Stat3/SIF-A response (13); therefore, we determined the ability of PD98059 to reverse this inhibitory effect. Analogous experiments as described for AII (Fig. 4) were performed with PMA as the inhibitory agent. Nuclear extracts were prepared from cells treated with IL-6, PMA + IL-6, or PD98059 + PMA + IL-6 and analyzed by electrophoretic mobility shift assays and immunoblotting. As demonstrated in Fig. 5, PD98059 significantly attenuated the PMA-mediated inhibition of IL-6-induced SIF-A (Fig. 5A, lanes 2-4) and Stat3 tyrosine phosphorylation (Fig. 5B, lanes 2-4). The partial reversal of the inhibitory effect of PMA by PD98059, observed in Fig. 5, A and B, possibly reflects the incomplete inhibition of PMA-induced MAP kinase activation by PD98059 shown in Fig. 1D. PD98059 alone had no significant effect on the IL-6-induced SIF-A response (Fig. 5A, lane 5). To determine if PKC activation was required for the PMA-mediated inhibitory action, cells were treated with PMA for 24 h to down-regulate PKC activity, and the effect of PMA to inhibit the IL-6-induced Stat3/SIF-A response was determined. In PKC down-regulated cells, PMA failed to inhibit the IL-6-induced SIF-A response (Fig. 5A, lanes 7 and 8) and Stat3 tyrosine phosphorylation (Fig. 5B, lanes 6 and 7), suggesting that PKC activation was required for PMA actions (13). As shown above for AII, the degree of Stat3 tyrosine phosphorylation in these nuclear extracts was proportional to the amount of Stat3 protein (Fig. 5C). Thus, the results of Fig. 5 suggest that: 1) inhibition of MAPKK1 by treatment with PD98059 attenuates the PMA-mediated inhibitory actions on IL-6-induced nuclear Stat3 tyrosine phosphorylation and SIF-A complex formation; and 2) the PD98059 sensitive inhibition by PMA requires PKC activation.

DISCUSSION

The studies described in this manuscript are an extension of our previous report on the ability of AII and PMA to inhibit IL-6-induced Stat3 tyrosine phosphorylation and SIF-A formation (13). The major finding in the present study is that exposure of CHO-K1 cells (expressing the angiotensin AT_1A receptor) to the specific MAPKK1 inhibitor PD98059 attenuated both the AII- and PMA-mediated interference of IL-6-induced Stat3 tyrosine phosphorylation and SIF-A formation. These results suggest that MAPKK1 or a downstream effector(s) is involved in the inhibitory actions of AII and PMA, indicating a role for the AII- and PMA-induced MAP kinase pathway in the inhibition of IL-6-induced Stat3 activation.

Comparison of the MAP kinase activation profile indicated that the AII- and PMA-induced activation profile was significantly different from that of IL-6 with respect to rate, magnitude, and duration. Angiotensin II and PMA, both of which induced MAP kinase activity with greater magnitude (8-fold) and duration (2 h and 30 min, respectively), potently inhibited IL-6-induced Stat3 tyrosine phosphorylation. In contrast, the weak MAP kinase activation by IL-6 (1.7-fold) was not associated with inhibition of Stat3 tyrosine phosphorylation. How the AII-induced MAP kinase pathway modulates the inhibition of IL-6-induced Stat3 tyrosine phosphorylation is not known. Several reports suggest that differences in the strength and duration of MAP kinase activation can cause distinct end responses (reviewed in Ref. 27). For example, in PC12 cells, sustained MAP kinase activation (1 h) by NGF was shown to cause cell differentiation, whereas transient activation (less than 20 min) by epidermal growth factor caused cell proliferation (27, 29). Sustained activation of MAP kinase by NGF resulted in translocation of MAP kinase to the nucleus (29, 31, 36), whereas transient activation by epidermal growth factor did not cause nuclear translocation. Transient activation, therefore, has different consequences for gene expression, compared to sustained activation, since nuclear accumulation of active MAP kinases would result in phosphorylation of distinct substrates, which may subsequently modulate the activity of transcription factors. Thus, the quantitative difference in MAP kinase activation is translated into a qualitative difference in end responses. It remains to be determined if stimulation of cells with AII causes differential translocation or compartmentalization of MAP kinase activity to bring about the inhibitory effect on IL-6-induced Stat3 tyrosine phosphorylation.

The intermediate in the MAP kinase pathway responsible for AII- and PMA-mediated inhibitory actions may be either MAPKK1, MAP kinase, or a downstream kinase activated by MAP kinase. This in turn may activate a kinase or a phosphatase, which could directly alter the signal transducing components of IL-6, through protein modifications. The interference may occur at the level of the IL-6 receptor (gp80) or the dimerization of the signal transducer protein gp130 during IL-6 signaling. Inhibition is unlikely to occur at the level of the JAK kinases, since AII treatment alone was shown to cause rapid activation of two of the JAK kinase family members, JAK2 and tyk2 (14). As shown in Fig. 1, inhibition does not appear to be due to Stat3 protein degradation. An interesting possibility would involve a tyrosine phosphatase that would dephosphorylate tyrosine-phosphorylated Stat3. In support of this, we have observed that inhibition of IL-6-induced Stat3 tyrosine phosphorylation occurs even if AII is added for 25 min after stimulation with IL-6,² suggesting an active dephosphorylation event. Although this possibility requires further elucidation, our data point to a potent cross-regulation of cytokine responses by the angiotensin AT_1A receptor, through the activation of the MAP kinase pathway.

Angiotensin II dose-dependently inhibited the IL-6-induced Stat3/SIF-A response with 50% inhibition occurring at 1 nM and complete inhibition at concentrations greater than 5 nM. These data indicate that AII mediates activation of a potent inhibitory pathway. In experiments where the ability of PD98059 to attenuate the inhibitory effects of AII was determined (Fig. 3), we used 1 nM AII to inhibit the IL-6-induced responses, since PD98059 (20 µM) completely blocked AII-induced MAP kinase activation at this concentration (Fig. 1C). The inability of PD98059 to completely inhibit the MAP kinase activation caused by AII at a concentration of 5 nM or greater, is consistent with a previous report (32), demonstrating that this inhibitor did not completely block the in vivo activation of MAP kinase, when cells were stimulated with high concentrations of agonists that were potent activators of MAPKK. Therefore, an effect of PD98059 on some biological processes could only be demonstrated at low concentrations of agonist (32).

It is important to reiterate the high degree of specificity of the MAPKK1 inhibitor PD98059 used in the present investigation. PD98059 binds to the dephosphorylated form (inactive form) of MAPKK1 and thus prevents the activation and phosphorylation of MAPKK1 (IC₅₀, 2 µM) (32). Only at higher concentrations does it exhibit a partial inhibitory effect on MAPKK2 (IC₅₀, 50 µM) (32). It has been used effectively to block the in vivo activation of MAP kinase activity, induced by a variety of agents including platelet-derived growth factor (33), NGF (34), and insulin (35). The high degree of specificity of PD98059 in vitro and in vivo is supported by its failure to inhibit a variety of protein kinases (18 protein Ser/Thr kinases and 4 protein Tyr kinases) that have so far been studied (32, 33). Moreover, it does not prevent the in vivo activation of Raf or the activation of other MAPKK and MAP kinase homologues, such as c-JUN kinase or p38 (32). Kinetic analysis has indicated that PD98059 does not compete for ATP binding or MAP kinase binding to MAPKK1, suggesting that it most likely inhibits MAPKK1 through an allosteric mechanism (33).³ This unusual mode of action may account for its high degree of selectivity for MAPKK1 inhibition (32, 33).

The cross-talk between AII and IL-6 or related cytokines may be important in acute phase and inflammatory responses and other pathophysiological processes such as cardiac hypertrophy. IL-6 has been shown to have a significant role in the acute phase response (37, 38), and additionally, several IL-6 family members (leukemia inhibitory factor, IL-11, oncostatin M, and cardiotrophin-1) contribute to cardiomyocyte hypertrophy (39). It is possible that through differential activation of the MAP kinase pathway, AII may influence the status of Stat3 tyrosine phosphorylation, thus modulating the cytokine signaling pathway and gene expression.