INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WE RECENTLY MADE the observation (23) that bradykinin (BK) has a direct hypertrophic effect on rat ventricular cardiomyocytes (VCM) as defined by an increase in protein synthesis and an increase in atrial natriuretic peptide mRNA and secretion that is similar to that of angiotensin II. The requisite kinases located in the BK signaling pathway relevant to the increase in protein synthesis in VCM are not known. A candidate kinase in regulating protein synthesis is 70-kDa S6 kinase (p70^S6K). This kinase phosphorylates a 40S ribosomal protein S6 that plays an important role in translation initiation of a class of mRNA that contain an oligopyrimidine tract at their translational start site (10, 12, 14, 18). A second kinase, 90-kDa ribosomal S6 kinase (p90^rsk), has also been observed to phosphorylate S6 in vitro (30). However, the ability of p90^rsk to phosphorylate S6 in vivo is unclear, because its primary function has been attributed to phosphorylation of transcription factors (3, 5). Consequently, the goal of this study was to determine the role that these kinases play in BK-stimulated protein synthesis.

In general, with a few noted exceptions (11, 24), p90^rsk and p70^S6K are independently activated by two distinct signaling pathways (4, 19). p90^rsk is a distal kinase in the mitogen-activated protein (MAP) kinase pathway. It is activated by phosphorylation by MAP kinase (ERK) (29). p70^S6K is a downstream kinase in the phosphatidylinositol (PI) 3-kinase signaling pathway (8, 32). However, the mechanism by which p70^S6K is activated is less clear. One of the activating kinases of p70^S6K is mammalian target of rapamycin (mTOR; also called FRAP or RAFT1), which has recently been shown to phosphorylate p70^S6K in its autoinhibitory domain, thereby allowing it to be activated by phosphorylation in its catalytic domain by constitutively active phosphoinositide-dependent protein kinase, PDK1 (2, 6, 7, 21, 22). In the current study, we determined the extent to which BK activates p90^rsk and p70^S6K.

To determine the relative contribution of each kinase to BK-stimulated protein synthesis, we employed specific inhibitors to block their activation. Activation of p90^rsk was blocked with PD-098059, which inhibits Raf activation of MAP kinase kinase (MEK) (1, 9). Activation of p70^S6K was blocked with rapamycin (12). Rapamycin is a bacterial macrolide that forms an inhibitory complex with the immunophilin FKBP12. This complex binds to mTOR, which blocks its ability to phosphorylate p70^S6K.

In this study, we found that BK activated p90^rsk and p70^S6K to a similar degree. Activation of p70^S6K but not p90^rsk was associated with an increase in protein synthesis. The distinctive finding in this study was the observation that the MAP kinase pathway inhibitor PD-098059 blocked activation of p70^S6K and protein synthesis. In general, the MAP kinase and PI 3-kinase signaling pathways have been considered to be parallel pathways with minimal cross talk between them (4, 19). However, the results of this study suggest that, in VCM, BK-stimulated protein synthesis and p70^S6K activation are dependent on signaling through the MAP kinase pathway.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of rat ventricular cardiomyocytes. VCM from adult male Sprague-Dawley rats were isolated as previously described (23). In brief, myocytes were freshly dissociated and plated onto laminin-coated (Collaborative Biomedical Products, Bedford, MA) six-well or 60-mm tissue-culture plates (Falcon/Becton Dickinson, Franklin Lakes, NJ) in serum-free medium 199 (Sigma, St. Louis, MO). VCM were incubated at 37°C until required (2-24 h), with >98% myocyte content as determined by examination of multiple fields by microscopy.

[³H]phenylalanine incorporation. Protein synthesis measurements were performed as previously described (23). Briefly, adult VCM, plated on six-well plates, were incubated with L-[2,3,4,5,6-³H]phenylalanine (1.5 µCi/ml; sp. act. 132 Ci/mmol; Amersham, Arlington Heights, IL) and with study drugs at 37°C for 2 h. The medium 199 contained 303 µM unlabeled D,L-phenylalanine. Experiments were terminated by washing myocytes with ice-cold PBS, pH 7.4. The protein and DNA were precipitated with 10% trichloroacetic acid at 4°C. The precipitates were washed with 95% ethanol. Precipitates were dissolved in 0.15 M sodium hydroxide. [³H]phenylalanine incorporation was determined by scintillation counting of an aliquot of each sample. With the use of a double-stranded DNA (dsDNA) fluorescent quantification reagent (PicoGreen, Molecular Probes, Eugene, OR), results were standardized to the DNA content of each well. Neonatal VCM were incubated at 37°C with [³H]phenylalanine (0.44 µCi/ml) in DMEM containing 400 µM unlabeled L-phenylalanine. At the end of the 48-h incubation period, myocytes were harvested as described above for the adult VCM. Each individual experiment was conducted with six replicates. Results are expressed as a percentage of control. Solutions containing BK were supplemented with lisinopril (1 µM) to limit BK degradation.

Immunoprecipitation of p90^rsk and p70^S6K. Adult VCM, plated on 60-mm tissue-culture plates, were incubated for 30 min with the kinase inhibitors PD-098059 (25 µM) or rapamycin (50 nM) before the addition of 10 µM BK. After 10 min, the cells were washed with ice-cold PBS before lysis in 0.5 ml ice-cold myocyte lysis buffer (MLB). The MLB contained the following (in mM): 2 dithiothreitol (DTT), 1 EDTA disodium salt, 5 EGTA, 50 -glycerophosphate, 0.1 leupeptin, 5 magnesium chloride, 1.15 phenylmethylsulfonyl fluoride (PMSF), 10 phosphate buffer (pH 7.4), and 1 sodium orthovanadate, in addition to 10 µg/ml aprotinin, 10 nM okadaic acid, and 0.5% Triton X-100 as previously described (26). After the protein content of the cell lysates was determined (Lowry method), 750 µg of protein were incubated with 1 µg of polyclonal antibody recognizing either p90^rsk (sc-231 G, Santa Cruz Biotechnology, Santa Cruz, CA) or p70^S6K (sc-230, Santa Cruz Biotechnology) with gentle agitation for 2 h at 4°C. The antigen-antibody complex was then precipitated with protein A-Sepharose beads with gentle agitation for 1 h at 4°C. The immunoprecipitates were washed three times with ice-cold PBS before resuspension in 45 µl of ice-cold MLB.

Phosphotransferase activity of p90^rsk and p70^S6K. Kinase activity was determined in immunoprecipitates using a kit (17-136) from Upstate Biotechnology (Lake Placid, NY). Briefly, substrate peptide ([AKRRRLSSLRA]) and inhibitor cocktail targeted to protein kinase C, protein kinase A, and calmodulin-dependent protein kinases were added to 10 µl of immunoprecipitate. ATP (10 µl) stock solution (75 mM magnesium chloride and 500 µM ATP) containing [-³²P]ATP (6,000 Ci/mmol, Du Pont-New England Nuclear, Boston, MA) was added for a final assay volume of 40 µl. Samples were incubated at 30°C for 20 min (p90^rsk) or 40 min (p70^S6K). The samples were placed on ice, and 30 µl were spotted onto P81 phosphocellulose filter papers. The papers were allowed to dry for ~5 min before being rinsed seven times in 0.75% phosphoric acid followed by two acetone washes. S6 kinase activity was expressed as picomoles of phosphate incorporated per minute per milligram of protein. Final results were expressed as a ratio of activity in control vs. experimental samples.

Immunoblot analysis of p70^S6K. Neonatal VCM, isolated as described above, were plated onto 12-well tissue-culture plates. After culture in serum-free medium for 24 h, VCM were incubated for 15 min with the kinase inhibitor PD-098059 (25 µM) before the addition of BK (10 µM) and lisinopril (1 µM). After 20 min, cells were washed with ice-cold PBS, and 0.2 ml SDS-containing buffer was added to each well. Proteins were resolved by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Millipore, Marlborough, MA). Immunoblotting was performed for analysis of p70^S6K phosphorylation at Ser-411 and at Thr-421/Ser-424 using phospho-specific antibodies (catalog nos. 9201 and 9204; Upstate Biotechnology, Lake Placid, NY). Detection was by enhanced chemiluminescence (ECL) using SuperSignal ULTRA (Pierce, Rockford, IL). All blots were stained with amido black to verify that protein-loading conditions and transfer efficiency were similar for all lanes. This approach was verified by immunoblotting with a non-phospho-specific antibody that recognizes p70^S6K (C-18, Santa Cruz Biotechnology).

Materials. ANG II, aprotinin, DTT, EGTA, -glycerophosphate, leupeptin, lisinopril, okadaic acid, PMSF, protein A, sodium hydroxide, sodium orthovanadate, and Triton X-100 were purchased from Sigma. Acetone, disodium EDTA, magnesium chloride, phosphoric acid, potassium phosphate (mono- and dibasic), and sodium chloride were obtained from Fisher Scientific (Fair Lawn, NJ). BK and rapamycin were obtained from Research Biochemicals International (Natick, MA), and ethanol was from Aaper Alcohol and Chemical (Shelbyville, KY). PD-098059 was kindly donated by Parke-Davis (Ann Arbor, MI).

Data analysis. Results are expressed as means ± SE. Statistical comparisons with control were by the Wilcoxon signed-rank test. The null hypothesis was rejected at the P < 0.05 level. The Bonferroni correction for multiple comparisons was applied where appropriate.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BK stimulates p90^rsk activity through the MAP kinase pathway. BK (10 µM) increased p90^rsk activity in adult VCM 2.5 ± 0.3-fold (Fig. 1). Pretreatment of myocytes with the MEK inhibitor PD-098059 (25 µM) diminished BK-stimulated p90^rsk activity 1.4 ± 0.4-fold. In contrast, pretreatment with the p70^S6K inhibitor rapamycin (50 nM) failed to block BK-stimulated activation of this kinase. Neither inhibitor significantly affected basal p90^rsk activity (data not shown).

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Bradykinin (BK) stimulates p90rsk activity in adult cardiomyocytes. BK (10 µM) significantly increased p90rsk activity (n = 6 experiments). PD-098059 (PD; 25 µM), but not rapamycin (Rapa; 50 nM), attenuated the response. * P < 0.05 vs. control.

BK stimulates p70^S6K activity through the MAP kinase pathway. In addition to increasing p90^rsk activity, 10 µM BK increased p70^S6K activity 2.0 ± 0.3-fold in adult VCM (Fig. 2). Pretreatment with the p70^S6K inhibitor rapamycin (50 nM) completely blocked BK-stimulated p70^S6K activity. Unexpectedly, treatment of myocytes with 25 µM PD-098059 attenuated BK-stimulated activity 1.3 ± 0.5-fold. To further verify this observation, we examined the effect of PD-098059 on BK-stimulated phosphorylation of p70^S6K. It has recently been shown that phosphorylation of p70^S6K parallels changes in its level of activity (33). Two phospho-specific antibodies were used that target phosphorylation of p70^S6K on Ser-411 and Thr-421/Ser-424. These phosphorylation sites reside in the autoinhibitory domain of p70^S6K (21). Phosphorylation at these and a fourth site, Ser-418, provides access to Thr-389 and Thr-229, two additional phosphorylation sites critical in the full activation of the kinase (21). We used neonatal VCM for these experiments because the ECL signal generated from the phospho-specific p70^S6K antibodies was too low in adult VCM. BK increased phosphorylation of p70^S6K on Ser-411 and Thr-421/Ser-424 (Fig. 3). Preincubation with 25 µM PD-098059 blocked BK-stimulated phosphorylation at these sites. Similar results were found when 50 nM rapamycin was used in these studies (data not shown).

View larger version (34K): [in this window] [in a new window]   Fig. 2.   BK stimulates p70S6K activity in adult cardiomyocytes. BK (10 µM) significantly increased p70S6K activity (n = 7 experiments). Both rapamycin (Rapa; 50 nM) and PD-098059 (PD; 25 µM) attenuated the response. * P < 0.05 vs. control.

View larger version (45K): [in this window] [in a new window]   Fig. 3.   PD-098059 inhibits BK-stimulated phosphorylation of p70S6K. Representative immunoblots probed with phospho-specific antibodies for p70S6K phosphorylated at Ser-411 or Thr-421/Ser-424 are shown. Preincubation with 25 µM PD-098059 blocked BK (10 µM)/lisinopril (1 µM)-stimulated phosphorylation. This experiment was repeated 3 times with similar results. Lane 1, control; lanes 2 and 3, BK/lisinopril; lanes 4 and 5, BK/lisinopril + PD-098059; lanes 6 and 7, PD-098059; lane 8, lisinopril. Intensity of bands is reflective of degree to which p70S6K is phosphorylated at Thr-421/Ser-424 when blot is probed with anti-Thr-421/Ser-424 antibody. Appearance of a slower migrating, highly phosphorylated form of p70S6K is indicative of activated p70S6K when blot is probed with anti-Ser-411 antibody. Samples from this experiment were again resolved by SDS-PAGE and probed with non-phospho-specific anti-p70S6K. These results are presented in lane 3. This experiment was repeated 3 times with similar results.

Inhibition of p70^S6K blocks protein synthesis. To determine if p70^S6K or p90^rsk plays a role in BK-stimulated protein synthesis, we examined the effect of rapamycin and PD-098059 on BK-induced [³H]phenylalanine incorporation into protein. BK (10 µM) increased protein synthesis by 27 ± 3% in adult VCM (Fig. 4). Rapamycin (50 nM) completely abolished the increase in [³H]phenylalanine incorporation induced by BK. Similar results were observed in neonatal VCM. In these studies, 10 µM BK increased protein synthesis by 39 ± 6%, and again, rapamycin negated this effect. For comparison, we also determined the effect of rapamycin on ANG II-stimulated protein synthesis. ANG II has previously been recognized to increase the activity of p70^S6K and p90^rsk in VCM (25, 26). ANG II (1 µM) increased [³H]phenylalanine incorporation by 29 ± 3% in adult and by 26 ± 2% in neonatal VCM. Rapamycin also blocked the increase in protein synthesis stimulated by ANG II. These results suggest that p90^rsk plays little to no role in BK or ANG II-stimulated protein synthesis, because rapamycin does not block p90^rsk activation (25; Fig. 1) yet blocks protein synthesis by BK or ANG II.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Rapamycin blocks increases in protein synthesis induced by ANG II and BK. Both ANG II (1 µM)- and BK (10 µM)-induced increases in [3H]phenylalanine incorporation in adult rat ventricular cardiomyocytes (A; n = 13 experiments) and neonatal rat ventricular cardiomyocytes (B; n = 4 experiments) were abolished by p70S6K inhibitor rapamycin (50 nM). * P < 0.05 vs. control.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   PD-098059 blocks increases in protein synthesis induced by ANG II and BK. Both ANG II (1 µM)- and BK (10 µM)-induced increases in [3H]phenylalanine incorporation in adult rat ventricular cardiomyocytes (A; n = 14 experiments) and neonatal rat ventricular cardiomyocytes (B; n = 7 experiments) were abolished by mitogen-activated protein kinase kinase inhibitor PD-098059 (25 µM). * P < 0.05 vs. control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Activation of p70^S6K is a requisite event in BK-stimulated protein synthesis. This conclusion is supported by the observations that rapamycin, which blocks activation of p70^S6K but not p90^rsk, inhibits BK-stimulated protein synthesis and PD-098059, which also blocks p70^S6K activation/phosphorylation, inhibits BK-stimulated protein synthesis. The essential role for p70^S6K in the regulation of protein synthesis is not unique to VCM. It plays a key role in other cells as well, because p70^S6K is recognized as the kinase that regulates the phosphorylation of 40S ribosomal protein S6 in vivo, distinct from MAP kinase-stimulated p90^rsk (4, 19).

An unexpected finding is the mechanism by which p70^S6K is activated by BK. The PI 3-kinase pathway is a well-recognized signaling pathway leading to the activation of p70^S6K (8, 32), whereas activation of p70^S6K through the MAP kinase pathway is not as well recognized or occurs less frequently. A few reports have emerged indicating a necessary role for activation of the MAP kinase pathway in stimulating protein synthesis. Increased protein synthesis stimulated by ANG II, insulin, or insulin-like growth factor I has been found to be dependent on activation of the MAP kinase pathway in vascular smooth cells, myoblasts, Chinese hamster ovary (CHO) cells, and VCM (15, 16, 27, 28, 34). However, in vascular smooth cells and myoblasts, activation of p70^S6K by ANG II or insulin was not inhibited by PD-098059 (15, 28). In contrast, in CHO cells, PD-098059 inhibited activation of p70^S6K at low concentrations of insulin (27). There is little information regarding the effect of PD-098059 on stimulated activation of p70^S6K in VCM. Thus the current study is one of the first to report that stimulation of p70^S6K is dependent on MAP kinase pathway activation in VCM. PD-098059 also blocked ANG II-stimulated protein synthesis in our studies, suggesting that ANG II-stimulated protein synthesis may also be mediated through a similar signaling pathway. However, we did not determine the effect of PD-098059 on p70^S6K activity with ANG II stimulation. Also of note, we found that inhibition of the MAP kinase pathway blocked stimulated protein synthesis in both adult, fully differentiated cardiac myocytes and neonatal VCM.

Our study suggests that activation of a kinase distal to Raf in the MAP kinase pathway plays a necessary role in activating p70^S6K with BK stimulation. A candidate kinase for this role is MEK, based on the current study and a study by Lenormand et al. (17). They developed a stable clone of Chinese hamster lung fibroblasts expressing Raf-1:ER, an estradiol-regulated form of oncogenic Raf-1. They found MAP kinase and p70^S6K activities to be increased in cells expressing Raf-1:ER with the administration of estradiol. However, the activation of p70^S6K was not mediated by MAP kinase because blocking MAP kinase activation by expression of the phosphatase MKP-1 did not prevent p70^S6K activation by Raf-1:ER. These results suggest that either Raf-1 or MEK mediates activation of p70^S6K. Additional studies by Lenormand et al. (17) indicate that MEK is the key kinase in the MAP kinase signaling pathway responsible for the activation of p70^S6K (17). In our study, PD-098059 blocked activation of p70^S6K by BK. PD-098059 binds to MEK and inhibits its activation by Raf (9). Therefore, Raf does not appear to be the proximal kinase responsible for the activation of p70^S6K. Our results are consistent with the observation that MEK plays a critical role in the activation of p70^S6K by BK in VCM.

The mechanism by which the MAP kinase pathway (MEK) activates p70^S6K is not known. In fact, the mechanism by which p70^S6K activity is regulated is still under investigation. p70^S6K has numerous phosphorylation sites (22). There is evidence that at least two kinases and one phosphatase play important roles in its phosphorylation and subsequently its level of activity; mTOR and PDK1 have been show to directly phosphorylate p70^S6K (2, 7, 21). Additionally, protein kinase B may also phosphorylate p70^S6K (2). mTOR also negatively regulates a phosphatase that recognizes a critical phosphorylation site necessary for p70^S6K activity (21). Activation of the MAP kinase pathway could affect any of these regulators of p70^S6K phosphorylation or directly phosphorylate p70^S6K.

mTOR/p70^S6K regulates translation of an essential class of mRNA that contains an oligopyrimidine tract at their transcriptional start site, termed 5'-TOP, which confers translational control on their expression (12, 14, 18). mRNA transcripts for all ribosomal proteins studied to date, as well as protein synthesis elongation factors, contain 5'-TOP at their translational start site (18). Failure to recruit these messages into polysomes suppresses the biogenesis of the translational machinery required for increased protein synthesis and cell cycle progression. Rapamycin selectively represses translation of 5'-TOP mRNA (13). The mechanism by which rapamycin blocks translation of 5'-TOP mRNA is well characterized. Rapamycin binds to FKBP12. This complex binds to and inhibits mTOR phosphorylation of p70^S6K (12). It has recently been shown that p70^S6K phosphorylates S6 in vivo and that it is required for induction of 5'-TOP mRNA translation (12, 31). These studies indicate that activation of p70^S6K is required for enhanced translation of 5'-TOP mRNA, which results in increased protein synthesis.

It is possible that the results of the experiments in which rapamycin and PD-098059 were used could be influenced by nonspecific effects of these drugs. Nonetheless, both have been well characterized and thus far have been found to be quite specific (1, 9, 12). There has been a greater amount of experience with rapamycin than PD-098059. However, PD-098059 has thus far been found to be a specific inhibitor of Raf activation of MEK (1, 9). Notably, PD-098059 does not inhibit PI 3-kinase (9). The specificity of PD-098059 may be conferred by it binding to an essential recognition site required for Raf activation of MEK (1). It does not interfere with the ATP binding site as do other kinase inhibitors that are not as specific as PD-098059 (1). Therefore, it is unlikely that PD-098059 blocks BK-induced activation of p70^S6K by a mechanism other than inhibition of Raf activation of MEK.

In conclusion, BK activates both p90^rsk and p70^S6K. Activation of p70^S6K but not p90^rsk is associated with an increase in protein synthesis in VCM. The distinctive finding in this study is the observation that signaling through the MAP kinase pathway activates p70^S6K in VCM when stimulated with BK. Activation of p70^S6K plays a critical role in BK-induced protein synthesis in VCM.