Stimulation of cardiac β-adrenergic receptors (β-AR) activates both the Gs- and Gi-coupled signaling cascades, including the phosphoinositide 3 kinase (PI3K) pathway, that have important physiological implications. Multiple isoforms of PI3K exist in the heart. The goals of this study were to examine the intracellular signaling pathways linking β-AR to PI3K and to identify the PI3K isoform mediating this transactivation in a cardiac context. Acute β-AR stimulation with isoproterenol resulted in increased tyrosine kinase-associated PI3K activity and phosphorylation of Akt and p70S6K in H9c2 cardiomyocytes. Cotreatment with ICI-118,551, but not CGP-20712, abolished the increase in PI3K activity, suggesting a β2-AR-mediated event. PI3K activation was also abrogated by cotreatment with pertussis toxin, 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolol[3,4-d]pyrimidine (PP2, a selective Src-family tyrosine kinases inhibitor), or AG-1296 [selective platelet-derived growth factor receptor (PDGFR) inhibitor] but not with an inhibitor for protein kinase A, protein kinase C, Ras, adenylyl cyclase, epidermal growth factor receptor, or insulin-like growth factor-1 receptor. β-AR stimulation induced an increase in tyrosine phosphorylation of PDGFR, which was abolished by inhibition of Src either by PP2 or small interfering RNA. Moreover, H9c2 cardiomyocytes stably transfected with a vector expressing a Gβγ sequestrant peptide derived from the COOH-terminus of β-AR kinase-1 failed to activate PI3K after β-AR stimulation, suggesting Gβγ is required for the transactivation. Furthermore, acute β-AR stimulation in vivo resulted in increases in PDGFR-associated PI3K and PI3Kα isoform activities but not the activities of other isoforms (PI3Kβ, -δ, -γ) in adult mouse heart. Taken together, these data provide in vitro and in vivo evidence for a novel mechanism of β-AR-mediated transactivation of cardiac PI3Kα via sequential involvement of Gαi/Gβγ, Src, and PDGFR.
- platelet-derived growth factor receptor
the β-adrenergic receptors (β-AR) are members of the seven transmembrane, G protein-coupled receptor (GPCR) family (4). These receptors mediate the effects of catecholamines, including activation of the cardiovascular system and regulation of energy metabolism in a variety of physiological and pathological conditions. Three β-AR subtypes (β1-AR, β2-AR, β3-AR) have been characterized at the gene level (10, 11, 20). In terms of receptor number, the β1-AR is the major subtype of the β-AR in the heart (26). The physiological responses and signal transduction mechanisms induced by β1-AR and β2-AR stimulation are different (48, 49). For example, the β1-AR is coupled only to Gαs and induces cAMP production, whereas the β2-AR can interact with both Gαs and Gαi (37). Although there are many downstream effectors of β-AR, there recently has been an increase in interest in the phosphoinositide 3-kinase (PI3K) signaling pathway. The mammalian PI3K can be divided into three major classes (class I, II, and III) based on their structure and substrate specificity (47). The class I PI3K can be further divided into two subclasses. The mammalian class IA PI3Ks are heterodimers of a 110-kDa catalytic subunit (p110α, p110β, or p110δ) and a regulatory subunit of 85 or 55 kDa (p85/p55), whereas the class IB PI3K (PI3Kγ) is composed of a p110γ catalytic subunit and a p101 regulatory subunit. The class I PI3Ks phosphorylate the 3′-OH of the inositol ring of phosphoinositide. Phosphoinositide 4,5-bisphosphate (PIP2) is the main substrate in vivo and is converted to phosphoinositide 3,4,5,-trisphosphate (PIP3). In most systems, the basal level of PIP3 in cells is low and only rises sharply upon cellular stimulation, which in turn controls a wide variety of intracellular signaling pathways (3). Several genetic models with alterations of PI3K isoforms and other signaling molecules, including phosphatase and tensin homologue deleted from chromosome 10 (PTEN) and Akt, have demonstrated altered cardiac phenotypes (review Refs. 1 and 31). Past results suggest the class IA PI3Ks are activated by receptor tyrosine kinase pathways, whereas the class IB PI3K (PI3Kγ) is coupled to GPCR (47). This has been studied most extensively in leukocytes (13, 38). In cardiomyocytes prepared from mutant mice with deletion of cardiac PI3Kγ, there is an increase in basal cAMP level (8). Although all class I PI3K isoforms are expressed in the heart (8), it is unclear which isoform of PI3K is functionally linked to β-AR in vivo. Experiments using purified recombinant proteins show that all class IA PI3K (-α, -β, -δ) respond to tyrosine kinase, but both PI3Kβ and PI3Kγ are sensitive to Gβγ (25).
β-AR stimulation has been shown to affect insulin signaling via PI3K/Akt in cardiomyocytes (27). In cultured neonatal cardiomyocytes, β-AR stimulation activates Akt via PI3K (28). Both the β1-AR and β2-AR have been reported to transactivate PI3K in vitro (15, 22). These studies, however, do not provide information regarding the PI3K isoform involved since the antibody used for immunoprecipitation (IP) in the in vitro lipid kinase assay recognizes all class I PI3K (p110α, -β, -δ, and -γ). We have demonstrated that the β-AR are involved in regulation of neonatal cardiomyocyte proliferation, which also involves the PI3K signaling pathway (45). The goal of our study was to examine the intracellular signaling cascade linking β-AR stimulation to PI3K activation. We used both in vitro and in vivo approaches to delineate the β-AR subtype and PI3K isoform critical for the transactivation. For in vitro study, we used the H9c2 cardiomyocytes, an embryonic ventricular myocyte cell line that has proved to be a useful model for in vitro study of cardiomyocytes (2, 12). The results suggest a novel transactivation mechanism between the β-AR and PI3K signaling pathway.
MATERIALS AND METHODS
The cAMP-dependent protein kinase (PKA) inhibitor H-89, the Gαs activator cholera toxin A (CTX), the Gαi inhibitor pertussis toxin (PTX), and the epidermal growth factor receptor (EGFR) inhibitor AG1478 were purchased from Sigma. The highly selective β1-AR (CGP-20712) and β2-AR (ICI-118,551) antagonists, the Src family-selective tyrosine kinases inhibitor PP2, the protein kinase C (PKC) inhibitor GF-109203X, and the activator of adenylyl cyclase forskolin were purchased from Tocris. The insulin-like growth factor-1 receptor (IGF-1R) inhibitor tyrphostin AG1024, the platelet-derived growth factor (PDGFR) inhibitor AG1296, and PDGF-AB were obtained from Calbiochem. The Ras inhibitor farnesylthioacetic acid (FTA) was purchased from Biomol. A monoclonal anti-α-sarcomeric actin antibody was purchased from Sigma. Antibodies against phospho-Akt (Ser-473 and Thr-308) and phospho-p70S6K (Thr-389) were obtained from Cell Signaling, and the antibody specific for phosphotyrosine (anti-pY) was purchased from Upstate Biotechnology. Antibodies against cSrc, PDGFR, as well as antibodies against each of the catalytic subunit isoform of class I PI3K: p110α (SC-7174), p110β (SC-7175), p110δ (SC-7176), and p110γ (SC-7177) were purchased from Santa Cruz Biotechnology.
H9c2 rat fetal cardiomyoblasts (ATCC no. CRL-1446) were grown in Dulbecco's modified Eagle-high glucose medium supplemented with 10% (vol/vol) fetal bovine serum containing 50 U/ml penicillin G and 50 μg/ml streptomycin in a humidified atmosphere containing 5% CO2 at 37°C. Cells were first grown to ∼50% confluency and synchronized overnight in serum-free medium before treatment. No cells were allowed to grow to confluency before treatment. For IP and PI3K assay, cells were treated with saline or isoproterenol (Iso, 10 μM) for 2 min. In some experiments selective inhibitors were added 30 min before Iso treatment unless specified otherwise.
All animal handling and procedures, adhering to American Physiological Society's “Guiding Principles in the Care and Use of Animals,” were approved by the Institutional Animal Care and Use Committee. Male C57BL/6 mice (6 wk old) were used for the in vivo studies. Animals were fed standard laboratory chow ad libitum. Mice were injected with either saline or Iso (1.25 mg/kg sc) for 1 h before being euthanized as previously described (46). Whole hearts were harvested and left ventricles were identified, flash frozen in liquid nitrogen, and stored at −70°C. Ventricular tissue lysates were prepared as described (46).
H9c2 cell and cardiac tissue lysates were prepared as described, and protein concentrations were determined with the bicinchoninic acid assay (41, 46). PI3K activity was determined with in vitro immunoprecipitation lipid kinase assay as described previously (46). Cell lysates (0.5 mg) were immunoprecipitated with anti-pY antibody, and l-α- phosphoinositide (Avanti Polar Lipids) was used as the lipid substrate (2 μg/reaction). Antibodies against individual class I PI3K isoform (p110α, -β, -δ, or -γ) and PDGFR were also used for immunoprecipitation in experiments involving mouse heart. After incubation, the final extracted reaction mixtures were spotted onto TLC plates, and the results were analyzed by phosphoimaging (Bio-Rad Laboratories).
A β-AR kinase-1 COOH-terminus (B-ARKct) minigene was inserted into an eukaryotic expression vector, pUSEamp(+) (pUSE), at the EcoRI and XbaI sites. Stable transfection of pUSE (empty vector) or pUSE-B-ARKct in H9c2 cells was performed by using lipofectamine according to the manufacture's instructions (Invitrogen). Individual single cells were isolated and screened for neomycin resistance. The final transfected cells expressing either pUSE or pUSE-B-ARKct were treated similarly as described above, and PI3K activity was measured.
To knockdown Src, two complementary hairpin small interfering RNA (siRNA) template oligonucleotides containing 21 nt target sequences of the mouse cSrc tyrosine kinase (5′-AAGTACAACTTCCATGGCACT-3′, GenBank no. BC052006) were synthesized, annealed, and ligated into pScilencer 5.1-H1 Retro vector (Ambion). The vector was then transfected to H9c2 cells by using Lipofectamine 2000 transfection reagent (Invitrogen). After transfection, individual single cells were isolated and screened for puromycin (1,000 ng/ml) resistance. As a control of the experiment, H9c2 cells were also transfected with scrambled oligos of the same length with minimum homology to mammalian genes.
The knockdown of cSrc in H9c2 cells was confirmed at the gene and protein levels by RT-PCR and Western blot analysis, respectively. The following primer pair was used in RT-PCR: sense: 5′-GCTACAGACGATAGGAAAGG-3′ and antisense 5′-CTCCACACATCAGACTTG-3′. As an internal control, RT-PCR was carried out simultaneously using a primer pair specific for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene sense 5′-TGGCAAAGTGGAGATTGTTG-3′ and antisense 5′-CTTCTGGGT GGCAGTGATG-3′. Western blot analysis was carried out by using antibodies specific for cSrc and α-sarcomeric actin.
Immunoprecipitation and Western blot analysis.
For IP studies, H9c2 cardiomyocytes were treated with Iso (10 μM) for 1 to 15 min. Protein lysates (0.5 mg) were incubated with protein G sepharose beads for 2 h at 4°C. After centrifugation, the supernatant was transferred to a fresh tube and incubated with anti-PDGFR at 4°C for 4 h with continuous rotation. A 20-μl packed volume of protein G sepharose was added, and the mixture was incubated overnight at 4°C with continuous rotation. After the mixture was washed, 45 μl of 2× Laemmli sample buffer were added. The sample was heated in boiling water for 5 min and quenched on ice for 2 min. After vortex and centrifuge, 20 μl of the supernatant were resolved on an SDS PAGE gel and immunoblotted with anti-pY or ant-PDGFR. Protein phosphorylation of Akt and p70S6K were measured in cells treated with Iso (0–100 μM, 0–60 min). Western blot analysis of α-sarcomeric actin, phospho-Akt, and phospho-p70S6K were performed as previously described (46). The results were visualized by chemiluminescence.
All results are shown as means ± SE. Statistical significance of the difference among groups was analyzed by one-way analysis of variance followed by Newman-Keuls test. In studies involving only control and the Iso-treated groups, statistical significance of the difference between the groups was analyzed by the paired Student's t-test. A probability of P < 0.05 was considered to represent a significant difference.
β-AR stimulation induces increases in phosphotyrosine-associated PI3K activity and phosphorylation of p70S6K and Akt.
In H9c2 cardiomyocytes, acute β-AR stimulation (10 μM Iso, 2 min) induced a significant increase in phosphotyrosine-associated PI3K activity (Fig. 1). This effect was mediated by the β2-subtype of β-AR because cotreatment with ICI-118,551 abrogated the effect of Iso, whereas cotreatment with CGP-20712 did not affect the outcome. In H9c2 cells, PDGF is a proven mitogen, an effect that requires PI3K (2). We therefore used PDGF as a positive control in our study. As expected, a low dose of acute PDGF-AB (5 ng/ml, 2 min) treatment significantly increased PI3K activity in H9c2 cardiomyocytes (Fig. 1). To assess whether the signaling molecules within the PI3K signaling pathway were activated by β-AR stimulation, H9c2 cells were treated with various doses of Iso (0, 10, 50, and 100 μM) for 5 min. β-AR stimulation induced an increase in Akt phosphorylation (Ser-473 and Thr-308) at all doses tested (Fig. 2A). The levels of total Akt were not changed. We next examined the effect of β-AR stimulation on phosphorylation of p70S6K. H9c2 cells were treated with Iso (10 μM) for 5 to 60 min. Again, Iso induced a significant increase in p70S6K phosphorylation (Thr-389) at all time points tested (Fig. 2B). These results demonstrate that acute β-AR stimulation induces activation of the PI3K signaling pathway in H9c2 cardiomyocytes.
β-AR-mediated increase in PI3K activity is Gαi-dependent but Gαs-independent and requires the involvement of Src family tyrosine kinases and PDGFR.
Src family tyrosine kinases and G proteins (both Gαs and Gαi) have been shown to contribute to β-AR-mediated extracellular signal-regulated kinase 1 and 2 (Erk1,2) phosphorylation and hypertrophy in cardiomyocyte (50). We therefore investigated the intracellular signaling event responsible for β-AR-mediated activation of PI3K activity. H9c2 cardiomyocytes were treated with Iso alone or in combination with activators or inhibitors of several signaling pathways. CTX induces accumulation of cellular cAMP through the GDP-ribosylation of Gαs, whereas H-89 is a selective potent inhibitor of PKA (5, 6). Treatment with CTX alone or forskolin (FO), an adenylyl cyclase activator, alone did not increase PI3K activity in H9c2 cardiomyocytes (Fig. 3A, lanes 9 and 10). Neither did cotreatment with CTX potentiate the effect of Iso on PI3K activity (Fig. 3A, lane 3). Cotreatment with H-89 also failed to abolish the effect of Iso on PI3K activity (Fig. 3A, lane 4). These results suggest that the Gαs-PKA-cAMP pathway is not involved in the β-AR-mediated increase in phosphotyrosine-associated PI3K activity in H9c2 cardiomyocytes.
Cotreatment with PTX abolished the effect of Iso, suggesting the Gαi subunit of G protein is involved in β-AR-mediated increase in PI3K activity (Fig. 3B). Cotreatment with an inhibitor for PKC (GF-109203X) or Ras (FTA) did not significantly alter the effect of Iso, whereas cotreatment with PP2, an inhibitor for the Src-family tyrosine kinases, completely abolished the effect of Iso (Fig. 3A, lanes 5–7). Cotreatment with an inhibitor for EGFR (AG1478) or IGF-1R (AG1024) did not alter the effect of Iso, whereas cotreatment with an inhibitor for PDGFR (AG1296) fully abrogated the effect of Iso (Fig. 3A, lanes 13–15). Treatment with individual inhibitor alone did not affect the basal PI3K activity in H9c2 cells (data not shown). These results suggest that β-AR-mediated increase in phosphotyrosine-associated PI3K activity in H9c2 cardiomyocytes is dependent on Gαi and requires the involvement of Src and PDGFR.
We next examined the effect of β-AR stimulation on activation of PDGFR and to further investigate the roles of Src in β-AR-mediated activation of PDGFR. First, tyrosine phosphorylation of PDGFR following β-AR stimulation was examined. H9c2 cardiomyocytes were treated with saline or Iso for one to 15 min. Cell lysates were subjected to IP with an antibody against PDGFR, followed by immunoblotting using anti-PDGFR and an antibody specific for phosphotyrosine. The basal levels of PDGFR phosphorylation (in nonstimulated cells) were very low. Acute β-AR stimulation significantly increased tyrosine phosphorylation of PDGFR (Fig. 4A). This effect can be observed as early as 1 min and peaked at 5 min after Iso treatment. The critical role of Src in β-AR-mediated activation of PDGFR was further established by experiments using siRNA against Src. The levels of Src were knocked down in H9c2 cells by stable transfection of a mouse cSrc siRNA. H9c2 cells transfected with scrambled oligos were used as the control. We confirmed the knockdown of Src at the gene and protein levels. As revealed by RT-PCR, H9c2 cells stably transfected with cSrc siRNA had a significantly reduced level of cSrc gene compared with that of scrambled oligonucleotide-transfected cells (Fig. 4B, top). GAPDH was used as an internal control in RT-PCR, and there was no difference in GAPDH expression between the two transfected cells. Western blot analysis revealed that H9c2 cells stably transfected with Src siRNA had a significant reduction in the level of Src protein compared with that of scrambled oligonucleotide-transfected cells (Fig. 4B, bottom). α-Sarcomeric actin was used as an internal control in Western blot analysis, and there was no difference in α-sarcomeric actin protein levels between the two transfected cells. These transfected cells were treated with saline or Iso for 2 to 15 min and processed as described in Fig. 4A. As similarly seen in nontransfected cells, cells transfected with scrambled oligos had a significant increase in tyrosine phosphorylation of PDGFR after acute β-AR stimulation (Fig. 4C). The peak phosphorylation was seen at 2 min after Iso treatment, a slight difference from that of normal H9c2 cells (5 min, Fig. 4A). These do not indicate a change in the time course of PDGFR phosphorylation following β-AR stimulation in scrambled oligos-transfected cells as we routinely observed peak phosphorylation between 2 and 5 min following Iso treatment in both the normal and transfected cells. In contrast, Src knockdown by siRNA completely abrogated the effect of β-AR stimulation on tyrosine phosphorylation of PDGFR. Since we have shown β-AR-mediated activation of PI3K requires Src (Fig. 3A), we next examined whether Src is also required for PDGF activation. H9c2 cardiomyocytes were treated with saline, PP2 alone, Iso alone, or Iso in combination with PP2 for 2 min. Cell lysates were IP/IB as described above to assess tyrosine phosphorylation of PDGFR. As expected, β-AR stimulation induced a significant increase in tyrosine phosphorylation of PDGFR, which was abolished by cotreatment with PP2 (Fig. 4D). This suggests that Src tyrosine kinase functions upstream of PDGFR following β-AR stimulation in H9c2 cardiomyocytes. Taken together, these data demonstrate that β-AR-mediated increase in PI3K activity requires the Src family tyrosine kinases and PDGFR and that in this signaling cascade Src functions upstream of PDGFR.
β-AR-mediated increase in PI3K activity requires the involvement of Gβγ.
Since we have shown that β-AR-mediated increases in PI3K activity is PTX sensitive (i.e., a Gαi-dependent signaling) and not involved Gαs, we next examined whether the Gβγ subunits of G protein are required. H9c2 cells were stably transfected with an empty vector pUSE or the vector expressing a B-ARKct minigene to inhibit Gβγ (21). In H9c2 cardiomyocytes stably transfected with the empty vector, β-AR stimulation (10 μM Iso, 2 min) induced a significant increase in PI3K activity (Fig. 5). In contrast, cells overexpressing B-ARKct failed to respond to β-AR stimulation. These results suggest that the Gβγ subunits of G protein are required for β-AR stimulation-induced increase in PI3K activity in H9c2 cardiomyocytes.
β-AR-mediated increase in PI3K activity in vivo is PDGFR dependent and PI3Kα isoform specific.
We have shown that β-AR stimulation in vitro induces an increase in tyrosine kinase-associated PI3K activity. This suggests that the class IA PI3K (PI3Kα, -β, and -δ) are involved in this transactivation. These results, however, did not rule out the involvement of the class IB PI3K (PI3Kγ). Although PI3Kγ can be activated by GPCR in vitro (13, 25, 38), there is very little available knowledge regarding the specific PI3K isoform(s) involved in β-AR-induced transactivation in vivo. We therefore investigated the specific PI3K isoform responsible for β-AR-mediated activation of PI3K in the mouse heart. Adult male C57BL/6 mice were treated with saline (10 μl/g body wt) or Iso (1.25 mg/kg sc) for 1 h, and left ventricle lysates were subjected to in vitro lipid kinase assay by using antibodies specific for phosphotyrosine and each of the catalytic subunit isoform of PI3K (p110α, -β, -γ, and -δ). Upon receptor activation in vivo, the catalytic subunit of PI3K is recruited to the plasma membrane by the regulatory subunit where the lipid substrates are located (47). Since mouse hearts were harvested after acute Iso stimulation, the use of individual p110 should represent actual individual PI3K isoform activity. Consistent with the in vitro data from H9c2 cardiomyocytes, β-AR stimulation in vivo increased phosphotyrosine-associated PI3K activity (Fig. 6A, lanes 1 and 2). PI3Kα activity was readily detectable and was significantly increased following acute β-AR stimulation (Fig. 6A, lanes 3 and 4). PI3Kδ activity was not affected by β-AR stimulation (Fig. 6A, lanes 9 and 10). The basal cardiac PI3Kγ activity is usually low in the heart without chronic β-AR stimulation (30, 32, and personal communication, Drs. Sathyamangla Naga Prasad and Howard A. Rockman). We found that both the activities of PI3Kβ and PI3Kγ in the heart, although detectable, were substantially lower than those of PI3Kα and PI3Kδ and were not affected by acute β-AR stimulation (Fig. 6A, lanes 5–8). The relatively lower cardiac PI3Kβ and PI3Kγ activities do not necessarily imply that their expression levels are lower than that of PI3Kα because these results could be dictated by the differences in the affinity and efficiency of the antibodies used in the assay. PI3Kβ and PI3Kγ proteins, however, are readily detectable in the mouse heart, neutrophils/macrophages, spleen, and bone marrow in Western blot analysis by using the same antibodies as we used in the immunoprecipitation step of in vitro lipid kinase assay (8, 13). One possibility to account for such discrepancy is the lower intrinsic kinase activities for cardiac PI3Kβ and PI3Kγ. As a loading control, the levels of α-sarcomeric actin were not changed. To further investigate whether PDGFR and PI3K are associated in vivo following β-AR stimulation, animals were treated with saline and Iso as described above. Cardiac tissue lysates were IP with an antibody specific for PDGFR, and the immunoprecipitates were subjected to in vitro lipid kinase assay. There was a fivefold increase (509 ± 213%, P < 0.05 vs. control) in PDGFR-associated PI3K activity by the acute Iso treatment. These results demonstrate that the activation of PI3K following acute β-AR stimulation in vivo is PI3Kα isoform specific and PDGFR dependent.
β-AR-mediated activation of the PI3K signaling pathway is of great interest and has significant biological implications. In the cardiac context, β2-AR stimulation induces an anti-apoptotic response via a Gαi-dependent cross talk to PI3K (48, 49). β-AR stimulation in vitro leads to recruitment of PI3Kγ to B-ARK, which results in receptor internalization (30). Disruption of these interactions in vivo is beneficial because it 1) prevents β-AR downregulation induced by chronic agonist administration; and 2) preserves cardiac function under chronic pressure overload (32). Animals with disruption of PI3Kγ gene are protected from chronic β-AR stimulation-induced heart failure (33). Our data provide in vitro evidence for the involvement of β-AR in the activation of PI3K signaling pathway. We demonstrate that this transactivation is mediated via a Gαi pathway and requires the Gβγ-subunits of G protein, the Src-family tyrosine kinases, and PDGFR. We further showed that β-AR-mediated increases in PI3K activity in vivo is PDGFR dependent. More importantly, we demonstrated β-AR stimulation in vivo induces activation of cardiac PI3Kα but not other isoforms of PI3K. The levels and activity of cardiac PI3K signaling pathway in rats are tightly controlled during development, with the highest activity found in late gestation and early postnatal life but reduces dramatically in older animals (46). In animals over 2 wk old, acute β-AR stimulation induces increases in p70S6K level and activity (46). Thus the current finding that acute β-AR stimulation in vivo induces increases in cardiac PI3Kα activity complements and extends our earlier studies in rats. It has been reported that GPCR can activate PI3Kγ via binding of Gβγ to the regulatory subunit of PI3Kγ (25, 43). On the other hand, stimulation of GPCR/Gβγ also leads to activation of PI3Kβ and Akt (25, 29). In human neutrophils, GPCR stimulation induces an increase in PTX-sensitive, phosphotyrosine-associated PI3K activity (42). Of note, these studies were conducted either in cells or with purified recombinant proteins. In contrast, our studies were performed on adult mice in an acute regimen of β-AR stimulation. Thus our finding represents a novel in vivo observation that stimulation of GPCR induces activation of PI3Kα isoform in the heart.
The H9c2 cardiomyocytes have characteristics of electrophysiological elements and signaling pathways of adult cardiomyocytes (12). This cell line provides a useful model for studying the interaction between β-AR and PI3K because in these cells, population of β-AR subtypes can be readily detected and PI3K has been shown to regulate their differentiation (9, 19). We showed that β-AR stimulation-induced activation of PI3K is β2-AR dependent. In the neonatal rat heart, acute β-AR blockade-induced reduction in p70S6K activity is β1-AR specific (unpublished data). This discrepancy could be explained by their differences in the subtype ratio of β-AR. H9c2 cardiomyocytes have a subtype ratio of 71% β2-AR versus 29% β1-AR (9). This is in contrast to that of the heart where β1-AR is the majority (26).
Src, the first identified protein tyrosine kinase, is important for GPCR-mediated signal transduction (24, 40). Both Gαs and Gαi, but not Gβγ, can directly interact with and activate Src (24). Overexpression of Gβγ in COS-7 cells, however, increases c-Src autophosphorylation (23) suggesting although there is no direct interaction, Gβγ is still able to regulate Src activity. Src plays essential roles in β2-AR internalization (14). Lyn, a Src-related tyrosine kinase, has been shown to temporally associate with and activate class IA PI3K (34, 36). Consistent with these studies, we show that inhibition of Src is sufficient to abrogate β-AR-induced increase in tyrosine kinase-associated PI3K activity. We used PP2 to inhibit Src. PP1, structurally and biologically similar to PP2, is known to block PDGFR-mediated downstream events (e.g., activation of mitogen-activated protein kinase and Akt/PI3K, increases in DNA synthesis). These effects, however, are mainly mediated via the effect of PP1 on Src family tyrosine kinase and not via direct interaction with PDGFR (7). The effect of PP2 on PDGFR has also been reported (16). Activation of PDGFR involves receptor autophosphorylation that results in activation of other signaling systems including Src and PI3K (44). This is achieved by either relocalization of signaling molecules from the cytosol to plasma membrane (PI3K) or by relaxing the intrinsic suppression of kinase activity (Src) (17, 39). We demonstrated that tyrosine phosphorylation of PDGFR is increased shortly following β-AR stimulation. Because inhibition of Src, either by pharmacological inhibition with PP2 or by siRNA targeting Src, abolishes β-AR-mediated increase in tyrosine phosphorylation of PDGFR (Fig. 4, C and D), it is likely that Src acts upstream of PDGFR in H9c2 cardiomyocytes. Moreover, since Gβγ is required for β-AR-mediated increase in PI3K activity but does not interact with Src (24), it is likely there are intermediate signaling molecule(s) between Gβγ and Src.
It has been shown that GPCR can induce shedding of heparin-binding epidermal growth factor through activation of metalloproteinases, which results in transactivation of EGFR (35). Downstream targets activated by this transactivation include the PI3K signaling pathway, Erk1,2, and p38 mitogen-activated protein kinase. In rat cardiac fibroblast, β2-AR stimulation induces mitogenesis via EGFR transactivation (18). This transactivation, however, was not observed in H9c2 cardiomyocytes. We found that cotreatment with AG1478, a selective inhibitor of EGFR, did not abolish the effect of β-AR stimulation on PI3K activation. Immunoprecipitation of H9c2 cell lysates with an antibody specific for EGFR did not result in a detectable level of protein (not shown). Thus it is likely that the intermediate signaling molecules for transactivation of PI3K and the downstream targets activated after β-AR stimulation in the heart is dependent on cellular context (e.g., fibroblast vs. cardiomyocyte).
In summary, our data suggest a novel mechanism in H9c2 cardiomyocyte for the cross talk between activation of β-AR and PI3K. Our data support a model depicted in Fig. 7 where stimulation of the β2-AR activates PI3Kα via the sequential involvement of Gαi/Gβγ, Src, and PDGFR.
This work was supported by National Institutes of Health Grant 1 P20 RR018728.
We are grateful to Drs. Lewis Cantley and Ji Luo of Harvard Medical School for critical input in PI3K assay. Discussion of assaying cardiac PI3Kγ activity with Drs. Sathyamangla Naga Prasad (Cleveland Clinic Foundation) and Howard A. Rockman (Duke University) was very helpful.
↵* N. Yano and V. Ianus contributed equally to this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 by the American Physiological Society