Endothelin-1 Activates Endothelial Cell Nitric-oxide Synthase via Heterotrimeric G-protein βγ Subunit Signaling to Protein Kinase B/Akt*

Endothelin-1 has dual vasoactive effects, mediating vasoconstriction via ETA receptor activation of vascular smooth muscle cells and vasorelaxation via ETB receptor activation of endothelial cells. Although it is commonly accepted that endothelin-1 binding to endothelial cell ETB receptors stimulates nitric oxide (NO) synthesis and subsequent smooth muscle relaxation, the signaling pathways downstream of ETB receptor activation are unknown. Here, using a model in which we have utilized isolated primary endothelial cells, we demonstrate that ET-1 binding to sinusoidal endothelial cell ETB receptors led to increased protein kinase B/Akt phosphorylation, endothelial cell nitric-oxide synthase (eNOS) phosphorylation, and NO synthesis. Furthermore, eNOS activation was not dependent on tyrosine phosphorylation, and pretreatment of endothelial cells with pertussis toxin as well as overexpression of a dominant negative G-protein-coupled receptor kinase construct that sequesters βγ subunits inhibited Akt phosphorylation and NO synthesis. Taken together, the data elucidate a G-protein-coupled receptor signaling pathway for ETB receptor-mediated NO production and call attention to the absolute requirement for heterotrimeric G-protein βγ subunits in this cascade.

ET-1 binds to one of two endothelin receptor subtypes, known as ET A or ET B . It is commonly accepted that endothelin stimulation of ET A receptors on smooth muscle cells leads to cellular contraction, whereas activation of ET B receptors in endothelial cells leads to NO production (11)(12)(13)(14). Furthermore, the physiologic relevance of this receptor specificity has been emphasized in in vivo studies examining vascular responsiveness after stimulation of endothelin receptors (12,(15)(16)(17).
Each of the endothelin receptors is a classic G-protein-coupled receptor (GPCR). In the canonical signaling pathway, receptor activation triggers the exchange of GTP for GDP by the G␣ subunit of the heterotrimeric G-protein, causing a conformational change in the G␣ subunits. GTP-bound G␣ and free G␤␥ subunits then bind and activate downstream effectors (18). GPCR pathways that have been shown to be stimulated downstream of ET-1 appear to be complicated and diverse (19 -25). For example, stimulation of ET A or ET B receptors stimulates a classic GPCR-linked phosphatidylinositol pathway (26). However, endothelin has the capability to signal to a multitude of other pathways, including MAPK pathways (27), cyclo-oxygenase (23), and others (24,25,28). Additionally, ET A activation appears to stimulate Src-family tyrosine kinases (29,30) presumably via GPCR cross-talk pathways, making signaling pathways potentially even more complex (31).
Phosphoinositide (PI) 3-kinase phosphorylates the 3Ј-hydroxyl group of phosphoinositides, leading to the formation of phosphatidylinositol 3-phosphate, phosphatidylinositol 3,4bisphosphate, and phosphatidylinositol 3,4, 5-trisphosphate. Lipid products of PI 3-kinase have been implicated in diverse cellular processes such as membrane ruffling and cell growth (32). Downstream of PI 3-kinase is the serine-threonine kinase, protein kinase B/Akt, which like PI 3-kinase has been implicated in a wide array of cellular processes (33)(34)(35). Moreover, recent data suggest that the Akt pathway may be involved in endothelin signaling pathways that control cell survival (36,37), although the mechanism of this connection has not been established.
The relationship between NO and endothelins in the control of vascular tone, although physiologically important (38), is incompletely understood. In particular, the link between ET B receptor activation and NO synthesis and the signaling cascade that leads to NO genesis remains poorly characterized. Given data suggesting that GPCRs can signal to PI 3-kinase, and that ET-1 is capable of signaling to Akt, we have hypothesized that one potential link between ET-1, ET B receptor, and NO synthesis is the Akt signaling pathway. We have examined primary isolates of sinusoidal endothelial cells, which, as for vascular endothelial cells, are known to possess only ET B receptors and to express eNOS (39 -42) and show that ET-1 binding to the ET B receptor causes eNOS activation via Akt phosphorylation, with resultant downstream eNOS phosphorylation and NO synthesis. Furthermore, we find a crucial role for G-protein ␤␥ subunits in the endothelin/NO signaling cascade.
Cell Isolation and Culture-Preparations of sinusoidal endothelial cells were from male Sprague-Dawley rat (450 -500 g) (Harlan Sprague-Dawley, Indianapolis, IN) as described previously (43). In brief, after in situ perfusion of the liver with 20 mg% Pronase (Roche Molecular Biochemicals, Indianapolis, IN), followed by collagenase (Crescent Chemical, Hauppauge, NY), dispersed cell suspensions were layered on a discontinuous density gradient of 8.2% and 15.6% Accudenz (Accurate Chemical and Scientific, Westbury, NY). Endothelial cells, present in the lower layer, were further purified by centrifugal elutriation (18 ml/min flow) and were grown in medium containing 20% serum (10% horse/calf). To verify the purity of endothelial cells, we routinely document their uptake of fluorescently labeled di-I-acetoacetylated low density lipoprotein as described (43). Additionally, endothelial isolates were probed with anti-CD31 (BD Biosciences, San Diego, CA). Contamination with stellate cells and/or Kupffer cells was detected by immunolabeling with anti-desmin (Dako, Carpenteria, CA) (44) and a specific antibody (45) as described (46). These methods demonstrated that the purity of primary endothelial cell isolates was greater than 95%; all experiments were performed with primary cells.
Adenoviral Gene Transfer-Recombinant adenovirus particles were purified from infected 293 cells by lysis in virus storage buffer followed by two sequential rounds of ultracentrifugation in CsCl gradients. Viral titers were measured by standard plaque assay using 293 cells. Sinusoidal endothelial cells were infected with a replication defective adenovirus that expresses myristoylated constitutively active Akt (Admyr.Akt), a dominant negative Akt construct (Ad-dn.Akt) containing mutations at amino acid 308 (T 3 A) and amino acid 473 (S 3 A) (provided by Dr. Ken Walsh, Boston University Medical Center) (47) or an identical adenovirus without vector DNA (Ad-EV) at a multiplicity of infection (m.o.i.) of 250. Recombinant adenovirus containing a construct encoding a cDNA corresponding to the carboxyl terminus fragment of bovine GRK2 (48) (Ad-GRK2CT) was used to infect cells (also m.o.i. of 250). Recombinant adenovirus containing a construct encoding a cDNA corresponding to PTEN, and an identical adenovirus without cDNA was as described (49) (each at an m.o.i. of 100). For cellular transduction of adenovirus, sinusoidal endothelial cell were plated at a density of 1 ϫ 10 6 cells/ml in 35-or 60-mm collagen-coated culture dishes and exposed to adenovirus in 2% serum for 16 h at 37°C. Subsequently, adenoviruscontaining medium was exchanged and cells were harvested at the indicated time points.
NO Measurement-Medium was collected from sinusoidal endothelial cells after indicated treatment. To assess NO production, we analyzed the release of nitrite, the stable breakdown product of NO, using a Sievers Chemiluminescence NO Analyzer (Sievers Instruments, Inc., Boulder, CO). Measurements of known concentrations of nitrite were used to generate a standard curve between 25 and 500 pmol of nitrite.
Immunoblotting-Sinusoidal endothelial cells were washed twice with ice-cold phosphate-buffered saline, and total cell lysates were prepared by scraping cells in lysis buffer (137 mM NaCl, 2 mM EDTA, 10% glycerol, 1% Triton X-100, 20 mM Tris-HCl, pH 8.0) containing protease inhibitors (1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 1 mM NaF, 1 g/ml leupeptin, 1 g/ml pepstatin). Lysates were subjected to SDS-PAGE, proteins were transferred to nitrocellulose, and specific epitopes were detected with specific primary antibody. Bound primary antibody was detected using a chemiluminescent detection kit (Tropix, Inc., Bedford, MA) over a linear range. Specific bands were scanned and quantitated by densitometry (Quantity One, Bio-Rad).
Statistics-All results were expressed as the mean Ϯ S.E. All experiments were performed in replicates utilizing cell isolates from different rats. Statistical analysis was performed using the two-tailed Student's t test, and p Ͻ 0.05 was considered statistically significant.

Endothelin Receptor Activation Leads to Akt
Phosphorylation-We initially investigated whether endothelin receptor activation led to Akt phosphorylation in sinusoidal endothelial cells. Exposure of sinusoidal endothelial cells to ET-1 (which has equal affinity for either the ET A or ET B receptor) or sarafotoxin S6C (which binds only to the ET B receptor) (50), led to FIG. 1. Endothelin receptor activation leads to phosphorylation of Akt. In A, sinusoidal endothelial cells were serum-starved and exposed to ET-1 (10 nM), sarafotoxin S6C (10 nM), or PDGF (0.25 ng/ml) for 30 min. Cellular lysates (50 g of total protein) were subjected to immunoblotting with specific antibody directed against phospho Akt; these immunoblots were stripped and reprobed with polyclonal anti-total Akt antibody. Below the immunoblot, specific phospho-Akt bands were quantitated and normalized to total Akt levels, and the data are presented graphically (control was arbitrarily set to "1," n ϭ 3; *, p Ͻ 0.05 compared with control). In B, serum-starved sinusoidal endothelial cells were pretreated with the ET B receptor inhibitor, BQ-788 (10 M), for 2 h, exposed to ET-1 (10 nM) for 30 min, and subjected to immunoblotting with anti-phospho-Akt antibody. In C, sinusoidal endothelial cells were serum-starved and exposed to ET-1 (10 nM) for the indicated times. Cells were subjected to immunoblotting with anti-phospho-Akt, after which blots were stripped and probed with anti-total Akt antibody. The data shown are representative of three to six different experiments, each performed with cells from a different isolation. significant increases in Akt phosphorylation at serine 473 ( Fig.  1, A and B). PDGF leads to phosphorylation of tyrosine residues (Tyr-740 and Tyr-751) on the PDGF B-receptor and activates the lipid kinase activity of PI 3-kinases, formation of PI (3,4,5)-P3 and activation of Akt. Importantly, PDGF also stimulated Akt phosphorylation, consistent with previous data (9,10). As expected, blockade of the ET B receptor with the specific ET B receptor antagonist BQ-788 inhibited ET-1-mediated Akt phosphorylation (Fig. 1B), emphasizing that in sinusoidal endothelial cells, ET-1-mediated Akt activation proceeds via ET B activation rather than ET A receptor signaling.
A unique property of the endothelin and endothelin receptor interaction is tight binding and a slow rate of dissociation (51). This property results in prolonged physiologic activity of ET-1, despite a relatively short half-life (52). Furthermore, dissociation rates and signaling activity vary among cells from different tissues (53). In sinusoidal endothelial cells, Akt phosphorylation occurred within minutes of endothelin receptor binding, peaked at 30 min, and subsequently declined (Fig. 1C), consistent with rapid signaling as well as rapid induction of inhibitory pathways.
ET B Activation Induces eNOS Phosphorylation and NO Production-A number of studies have indicated that endothelial nitric-oxide synthase (eNOS) is an important Akt substrate and that eNOS is activated by phosphorylated Akt (9, 10). Therefore, we examined whether ET-1-induced Akt activation led to phosphorylation of eNOS as well as NO release in sinusoidal endothelial cells. We found that ET B receptor activation led to Akt phosphorylation and caused eNOS to be phosphorylated at serine 1179 ( Fig. 2A). ET-1 induced not only Akt phosphorylation, but it also led to production of nitrite, indicative of NO synthesis (Fig. 2B). Likewise, stimulation of ET B receptors with sarafotoxin S6C led to eNOS phosphorylation (not shown) and NO synthesis (Fig. 2B). Interestingly, PDGF did not induce NO production in sinusoidal endothelial cells (Fig. 2B). Furthermore, exposure to endothelins and PDGF did not alter the level of total Akt protein expression or total eNOS protein expression (Fig. 1A). These results further substantiate the finding that ET-1 causes NO production in sinusoidal endothelial cells as a result of Akt activation and eNOS phosphorylation.
ET-1 Induces NO Synthesis via a PI 3-Kinase/Akt Pathway-To further characterize the role of the PI 3-kinase/Akt pathway in endothelin-mediated NO synthesis, we examined ET B -induced Akt and eNOS phosphorylation and NO synthesis after PI 3-kinase inhibition. LY29004, a specific inhibitor of PI 3-kinase, prevented ET-1-mediated Akt phosphorylation (Fig.  3A), including at all time points after exposure to ET-1 (Fig.  3B). Additionally, LY29004 efficiently inhibited NO production (Fig. 3C); of note, NO appeared to be produced rapidly after ET B receptor activation (nitrite levels were measured in conditioned medium so that total nitrite levels increased incrementally over time). As expected, LY29004 inhibited PDGFinduced Akt phosphorylation (Fig. 3A) but had no effect on total Akt expression.
Modulation of PI 3-Kinase Activation by PTEN-The inositol 3Ј-phosphatase, PTEN, which hydrolyzes PI 3-kinase lipid products phosphatidylinositol 3,4-bisphosphate and phosphatidylinositol 3,4,5-trisphosphate, functions in opposition to PI 3-kinase (54,55). Therefore, to further assess the role of ET-1 FIG. 2. ET B receptor activation, Akt and eNOS phosphorylation and subsequent NO production. In A, serum-starved sinusoidal endothelial cells were exposed to ET-1 (10 nM) and subjected to immunoblotting with anti-phospho-eNOS antibody; blots were stripped and reprobed with anti-phospho-Akt, anti-total Akt, or anti-eNOS antibody. The immunoblot shown is representative of three different experiments, each performed with cells from a different isolation. In B, NO production in sinusoidal endothelial cells was detected by chemiluminescence detection of nitrites in conditioned medium (n ϭ 3; *, p Ͻ 0.005 compared with control, nitrite levels were normalized to total protein content in cell monolayers).

FIG. 3. ET-1 induces NO synthesis via a PI 3-kinase-Akt pathway.
In A, serum-starved sinusoidal endothelial cells were pretreated with the PI 3-kinase inhibitor LY294002 (20 M) for 4 h, exposed to ET-1 (10 nM) or PDGF (25 ng/ml) for 30 min, harvested, and subjected to immunoblotting as in Fig. 1. In B, sinusoidal endothelial cells were serum-starved and pretreated with LY-294002 for 4 h and exposed to ET-1 (10 nM) for the indicated times and subjected to immunoblotting as in Fig. 1. The immunoblot is representative of three different experiments, each performed with cells from a different isolation. In C, nitrite production in conditioned sinusoidal endothelial cell medium was detected by chemiluminescence (n ϭ 3; *, p Ͻ 0.005 compared with control or ET/Ly, nitrite levels were normalized to total protein content in cell monolayers). ET signifies ET-1, and LY signifies LY294002. in activation of PI 3-kinase, we examined the effect of overexpression of a wild-type PTEN construct in sinusoidal endothelial cells stimulated with ET-1. ET-1-mediated phosphorylation of Akt in sinusoidal endothelial cells was inhibited by overexpression of wild-type PTEN (Fig. 4), further confirming that ET-1-mediated activation of Akt is PI 3-kinase-dependent.
A Direct Effect of Akt on NO Production-To assess the specific role of Akt activation in ET B receptor-mediated signaling to NO, we examined the effect of dominant negative and dominant active Akt constructs (47,56). Expression of a dominant negative Akt construct inhibited ET-1-mediated Akt phosphorylation (Fig. 5A) and NO production (Fig. 5B). As expected, overexpression of dominant active Akt led to enhanced Akt phosphorylation; this subsequently led to an in-crease in NO synthesis (Fig. 5, A and B). A negative control (an identical adenovirus vector without insert DNA) failed to induce Akt phosphorylation or NO synthesis. Interestingly, the additional effect of ET-1 on both Akt phosphorylation and NO synthesis was minimal when active Akt was expressed (Fig. 5,  A and B).
ET-1 Stimulated Akt Phosphorylation Is Pertussis Toxinsensitive-Sequence homology and hydropathic analysis of endothelin receptor primary structure suggests that the known endothelin receptors belong to the rhodopsin class of GPCRs (41,56,57). Pathways downstream of GPCRs have been classified as either pertussis toxin (PTX)-sensitive (i.e. coupled to G␣ i/o ) or PTX-insensitive (i.e. coupled to G␣ q/11 , G␣ s , or G␣ 12/13 ) (56,58,59). PTX catalyzes the ADP-ribosylation of the alpha subunits of heterotrimeric guanine regulatory proteins G i , G o , and G t . This prevents these specific G-proteins from interacting with their activating receptors and thus prevents propagation of downstream receptor signals. We found that PTX pretreatment inhibited ET-1-induced Akt phosphorylation (Fig. 6), suggesting that Akt phosphorylation after ET B receptor stimulation is linked to G i/o . Notably, PTX did not influence basal Akt or phospho-Akt levels (Fig. 6).  4. Modulation of PI 3-kinase activation by PTEN. Sinusoidal endothelial cells were transduced with adenovirus encoding WT PTEN or with adenovirus containing the identical adenovirus backbone without a cDNA insert (Ad-EV) for 18 h. The cells were subsequently serum-starved for 6 h and then exposed to ET-1 (10 nM) for 30 min. Cellular lysates were subjected to immunoblotting with antibodies directed against phospho-Akt; blots were stripped and probed with antitotal Akt antibody or anti-PTEN antibody. The data shown are representative of three different experiments, each performed with cells from a different isolation.

FIG. 5.
A direct effect of Akt on NO production. In A, sinusoidal endothelial cells were transduced with adenovirus containing a dominant-negative form of Akt (Ad-dn.Akt), adenovirus containing a constitutively active form of Akt (Ad-myr.Akt), or adenovirus containing the identical adenovirus backbone without a cDNA insert (Ad-EV) overnight. Cells were washed and incubated in serum-free medium for 6 h, followed by exposure to ET-1 (10 nM) for 30 min. Cellular lysates (50 g of total protein) were subjected to immunoblotting with anti-phospho-Akt, after which blots were stripped and reprobed with anti-total Akt antibody. The data shown are representative of three different experiments, each performed with cells from a different isolation. In B, conditioned medium was collected, and nitrite levels were measured by chemiluminescence (n ϭ 3; *, p Ͻ 0.005 for control or Ad-dn.Akt cells compared with ET-1 treated cells or Ad-myr.Akt cells).

FIG. 6. ET-1 activated Akt phosphorylation via a pertussis toxin-sensitive manner.
Sinusoidal endothelial cells were pretreated with 100 ng/ml pertussis toxin (PTX) for 6 h prior to stimulation with ET-1 for 30 min. Cellular lysates (50 g of total protein) were subjected to immunoblotting with anti-phospho-Akt antibody, after which blots were stripped and reprobed with anti-total Akt antibody. The data shown are representative of three different experiments, each performed with cells from a different isolation.
FIG. 7. G-proteins are critically linked to NO signaling. In A, sinusoidal endothelial cells were transduced with adenovirus encoding the GRK2 carboxyl terminus (GRK2CT) or a control adenovirus containing an identical adenovirus backbone without a cDNA insert (Ad-EV) for 18 h. Cells were washed and serum-starved overnight and exposed to ET-1 (10 nM) for 30 min as indicated. Cells were lysed and subjected to immunoblotting (50 g of total protein) with antibodies directed against phospho-Akt; blots were stripped and probed with anti-total Akt antibody or anti-GRK2CT antibody. The data shown are representative of three different experiments, each performed with cells from a different isolation. In B, nitrite levels were measured in conditioned medium by chemiluminescence (n ϭ 3; *, p Ͻ 0.05 for control or Ad-GRK2CT-exposed cells compared with ET-1-treated cells).
G-proteins Are Critically Linked to NO Signaling-We demonstrated that ET-1 stimulates Akt and eNOS through a G i/ocoupled mechanism. Because activation of G i/o -coupled receptors leads to dissociation of activated G␣ i/o and G-protein ␤␥ subunits (56,59,60), we hypothesized that Akt phosphorylation was dependent on G-protein ␤␥ subunit signaling in this system. To test this postulate, we utilized a construct coding for the carboxyl terminus of G-protein-coupled receptor kinase 2 (GRK2CT), which binds to G-protein ␤␥ subunits and prevents their downstream signaling (48). Transduction of sinusoidal endothelial cells with an adenovirus expressing this construct abrogated ET-1-induced Akt phosphorylation and NO production (Fig. 7, A and B). These findings suggest that G-protein ␤␥ subunits rather than G␣ i/o subunits contribute to the ET-1-dependent activation of Akt.
ET-1 Stimulation of eNOS Is Protein-tyrosine Kinase-independent-Non-receptor tyrosine protein kinases have been shown to exhibit cross-talk with G-protein coupled receptorsignaling cascades (31). Therefore, we postulated that nonreceptor tyrosine protein kinases might have a role in ET-1-mediated signaling. However, we found that, although the tyrosine kinase inhibitor, genistein, inhibited PDGF (a well known inducer of Akt)-mediated Akt phosphorylation, it had no effect on ET-1-mediated activation of Akt (Fig. 8A). Additionally, inhibition of protein-tyrosine kinases with genistein had no effect on NO production by sinusoidal endothelial cells (Fig.  8B). These data support the concept that ET-1-mediated Akt phosphorylation and NO production is G-protein-dependent and does not involve a tyrosine kinase pathway. DISCUSSION The endothelins comprise a family of 21-amino acid peptides that activate specific G-protein-coupled receptors in a variety of cell types to regulate blood flow and multiple other physiologic effects (61). The physiologic effects of ET-1 are mediated by at least two types of endothelin receptors, designated ET A and ET B . It is commonly accepted that stimulation of ET B receptors leads to NO production in endothelial cells. However, the signaling pathways linking the ET B receptor to NO have been poorly characterized. In this study, we have elucidated a potential mechanism linking endothelin-mediated activation of ET B receptors and eNOS (and thus, NO). We show that Akt plays a central role in this process, consistent with previous data demonstrating that Akt phosphorylation leads to eNOS phosphorylation and subsequent activation of eNOS enzymatic activity (10,62). Moreover, we reveal that heterotrimeric Gprotein ␤␥ subunits released from receptors-activated G i/o -coupled receptors represent the critical connection to Akt, and thus eNOS.
Previous data have shown that ET-1 activates several intracellular signal pathways, including adenylyl cyclase, phospholipase C, protein kinase Cs, and the mitogen-activated protein kinase (MAPK) cascades (20,21,63). Here, we have extended the ET-1 signaling paradigm to include Akt, eNOS, and NO. We utilized several different approaches to elucidate the critical role of Akt in endothelin-mediated activation of eNOS. First, we demonstrated that Akt phosphorylation paralleled closely eNOS phosphorylation after stimulation with ET-1. Second, we showed that a dominant negative Akt construct abrogated ET-1 phosphorylation of Akt and eNOS. A central role for Akt in endothelin-mediated NO production was further supported by pharmacologic inhibition of PI 3-kinase with LY294002, because Akt activation is known to be PI 3-kinasedependent. Finally, we demonstrated that ET-1-mediated activation of Akt was inhibited by the potent lipid phosphatase, PTEN (54,64), which dephosphorylates the 3Ј-phosphoinositide products of PI 3-kinase, phosphatidylinositol 3,4-biphosphate, and phosphatidylinositol 3,4,5-triphosphate (55) and has been shown to regulate vascular epidermal growth factor-FIG. 8. Tyrosine-protein kinase activity was not required for ET-1-mediated Akt activation. A, sinusoidal endothelial cells were serum-starved overnight and exposed to genistein for 2 h, then ET-1 (10 nM) or PDGF (25 ng/ml) for 30 min. Cellular lysates (50 g of total protein) were subjected to immunoblot with anti-phospho Akt antibody; blots were stripped and probed with anti-total Akt antibody. The data shown are representative of three different experiments, each performed with cells from a different isolation. In B, nitrite levels were measured in conditioned medium by chemiluminescence (n ϭ 3; *, p Ͻ 0.05 for ET-1 or ET-1/genistein cells compared with other cells).
FIG. 9. Schematic representation of the putative ET B receptor-eNOS signaling pathway. In this pathway, ET B receptor stimulation leads to G-protein ␤␥ activation, and subsequent PI 3-kinase stimulation. PI 3-kinase activation leads to production of 3-phosphoinositides (e.g. PI(3,4,5)P 3 ), which results in recruitment and activation of Akt, which in turn phosphorylates and activates eNOS, leading to NO production. The pathway is interrupted by various pharmacologic and intrinsic signaling molecules as shown. Ly294002 selectively inhibits PI 3-kinase by blocking its active site, whereas PTEN hydrolyzes 3-phosphoinositides, preventing plasma membrane recruitment of Akt. mediated PI 3-kinase signaling in human umbilical vein endothelial cells (49). Inhibition of endothelin-mediated eNOS activation by PTEN indicates that this process is dependent on 3-phosphoinositide production by PI 3-kinase.
Extensive data from previous studies (56,61) have linked ET-1 to canonical GPCR signaling pathways. Here, we have extended this work to emphasize the central role not only of ␤␥ subunits, but also GRK2. Additionally, we demonstrated that PTX, which catalyzes the ADP-ribosylation of the G i family alpha subunits and thereby interrupts G i ␤␥ subunit function, effectively abrogates ET-1-mediated Akt phosphorylation and NO production. Taken together, these data point to a critical role for ␤␥ subunits in the signaling cascade. Although the data using the dominant negative GRK2CT construct, which selectively binds to and inhibits the function of ␤␥ subunits, provides convincing evidence for the involvement of ␤␥ subunits in the signaling cascade, it should be emphasized that our studies do not unequivocally exclude a role for G␣ subunits. Selective inhibition of G␣ i/o would be an attractive approach; however, it is not possible to inhibit G␣ i/o without also inhibiting ␤␥. Nonetheless, ␤␥ subunits appear to be necessary for Akt phosphorylation and eNOS activation after endothelin binding.
Our data as well as previous work emphasize a mechanism by which GPCR activation leads to Akt phosphorylation; for example, PI 3-kinase activation appears to proceed via heterotrimeric G-protein ␤␥ signaling (Fig. 9). Supporting the presence of this signaling pathway are previous data that demonstrate a direct interaction of ␤␥ subunits with the catalytic domain of the PI 3-kinase ␥ isoform (65). In addition, cardiac hypertrophy led to ␤␥ subunit-dependent activation of PI 3-kinase (66). Such data are highly consistent with our findings, which emphasize a specific and novel endothelin-mediated pathway involving ␤␥ subunit recruitment and activation of PI 3-kinase with downstream phosphorylation of Akt and eNOS.
Although GPCR activation has been shown to lead to Akt activation via PI 3-kinase ␥ (67), PI 3-kinase/Akt is well known to be linked to tyrosine kinase phosphorylation (55). For example, PDGF, a classic tyrosine kinase inducer, stimulates recruitment and activation of PI 3-kinase ␣ via its p85 regulatory subunit and subsequent Akt phosphorylation (68,69). We found that endothelin-mediated Akt activation and NO production were not affected by the tyrosine kinase inhibitor genistein, supporting the observation that ET B receptors signal through PI 3-kinase ␥. Furthermore, in our system, PDGF activated Akt but had no effect on eNOS.
To our knowledge, this is the first report documenting a mechanism linking ET B receptor activation directly to Akt and eNOS activation in an endothelial cell system. The work extends that of others in which it has been shown that ET-1 can stimulate NO synthesis in isolated endothelial cells (14). It is notable that we have failed to identify NO production in a variety of transformed endothelial cell lines. 2 Thus, a major advantage of the current study is that it utilized only primary cell isolates. In the context of the data presented, we further postulate that primary cells and cell lines possess dissimilar endothelin signaling pathways. Finally, the current data not only emphasize an important endothelin signaling pathway but also highlight a unique system with which to examine endothelin/NO-coupled signaling.