Endothelin 1 Induces (cid:1) 1 Pix Translocation and Cdc42 Activation via Protein Kinase A-dependent Pathway*

p21-activated kinase (Pak)-interacting exchange factor (Pix), a Rho family guanine nucleotide exchange factor (GEF), has been shown to co-localize with Pak and form activated Cdc42- and Rac1-driven focal complexes. In this study we have presented evidence that treatment of human mesangial cells (HMC) with endothelin 1 (ET-1) and stimulation of adenylate cyclase with either forskolin or with the cAMP analog 8-Br-cAMP activated the GTP loading of Cdc42. Transient expression of constitutively active G (cid:2) s also stimulated Cdc42. In addition, overexpression of (cid:1) 1 Pix enhanced ET-1-in-duced Cdc42 activation, whereas the expression of (cid:1) 1 Pix SH3m(W43K), which lacks the ability to bind Pak, and (cid:1) 1 PixDHm(L238R/L239S), which lacks GEF activity, de- creased ET-1-induced Cdc42 activation. Furthermore, ET-1 stimulation induced (cid:1) 1 Pix translocation to focal complexes. Interestingly, pretreatment of HMC with protein kinase A (PKA) inhibitors blocked both Cdc42 activation and (cid:1) 1 Pix translocation induced by ET-1, in- dicating the involvement of the PKA pathway. Through site-directed mutagenesis studies of consensus PKA phosphorylation sites and in vitro PKA kinase assay, we have shown that (cid:1) 1 Pix

p21-activated kinase (Pak)-interacting exchange factor (Pix), a Rho family guanine nucleotide exchange factor (GEF), has been shown to co-localize with Pak and form activated Cdc42-and Rac1-driven focal complexes. In this study we have presented evidence that treatment of human mesangial cells (HMC) with endothelin 1 (ET-1) and stimulation of adenylate cyclase with either forskolin or with the cAMP analog 8-Br-cAMP activated the GTP loading of Cdc42. Transient expression of constitutively active G␣ s also stimulated Cdc42. In addition, overexpression of ␤ 1 Pix enhanced ET-1-induced Cdc42 activation, whereas the expression of ␤ 1 Pix SH3m(W43K), which lacks the ability to bind Pak, and ␤ 1 PixDHm(L238R/L239S), which lacks GEF activity, decreased ET-1-induced Cdc42 activation. Furthermore, ET-1 stimulation induced ␤ 1 Pix translocation to focal complexes. Interestingly, pretreatment of HMC with protein kinase A (PKA) inhibitors blocked both Cdc42 activation and ␤ 1 Pix translocation induced by ET-1, indicating the involvement of the PKA pathway. Through site-directed mutagenesis studies of consensus PKA phosphorylation sites and in vitro PKA kinase assay, we have shown that ␤ 1 Pix is phosphorylated by PKA. Using purified recombinant ␤ 1 Pix(wt) and ␤ 1 Pix mutants, we have identified Ser-516 and Thr-526 as the major phosphorylation sites by PKA. ␤ 1 Pix(S516A/T526A), in which both phosphorylation sites are replaced by alanine, blocks ␤ 1 Pix translocation and Cdc42 activation. Our results have provided evidence that stimulation of PKA pathway by ET-1 or cAMP analog results in ␤ 1 Pix phosphorylation, which in turn controls ␤ 1 Pix translocation to focal complexes and Cdc42 activation.
Mesangial cells are smooth muscle-like cells situated within the renal glomerulus that play an important role in regulating glomerular filtration and function, both by contraction and release of proinflammatory substances. Endothelin (ET), 1 a potent vasoconstrictor peptide implicated in chronic renal diseases, plays a crucial role in the physiology and pathology of glomerular cells. ET release is increased in response to inflammatory cytokines, suggesting that ET-1 synthesis might increase in glomerulonephritis by intrinsic glomerular cells such as glomerular endothelial, mesangial, and epithelial cells (1). Two receptors for ET isopeptides, ET A and ET B , are G proteincoupled receptors with seven transmembrane domains (2,3). ET not only stimulates mesangial cell proliferation (4) but also increases the expression of extracellular matrix proteins such as collagen and fibronectin (5) and induces active cytoskeletal rearrangement. This process is governed largely by the precise temporal and spatial modulation of small GTPase proteins of the Rho family, Cdc42, Rac, and RhoA.
Cdc42 and Rac1 function as molecular switches (6,7). They are converted from the GDP-bound inactive form to a GTPbound active state by a reaction catalyzed by guanine nucleotide exchange factors (GEFs) (8). Since their identification, GEFs have become increasingly involved in mediating the effects of G protein-coupled receptor agonists. Recently, a Cdc42/ Rac-GEF termed Pix (Pak-interacting exchange factor) was identified (9). Pix has a diffuse B cell lymphoma homology (DH) domain and a flanking pleckstrin homology domain, which are conserved in all of the GEFs for Rho GTPases. Pix family proteins consist of two isoforms, ␣Pix and ␤Pix, and recently a new splice variant of ␤Pix designated ␤ 2 Pix has been identified (10). The human Pix family bind tightly through an N-terminal SH3 domain to a conserved proline-rich Pak sequence located at the C terminus and are colocalized with Pak to form activated Cdc42-and Rac1-driven focal complexes (9). Recently, Pix has been shown to form a trimolecular complex with Pak1 and p95PKL (also known as G protein-coupled receptor kinaseinteracting target, GIT1) (11). Furthermore, tyrosine-phosphorylated p95PKL can also bind paxillin (12,13) and therefore provides the link between Pix/Pak and focal complexes through this interaction. The presence of several domains allows Pix to interact with a variety of signaling proteins and suggests that Pix might have an important role in mediating the effects of extracellular signals (2, 14 -17).
The physiological and pathological responses elicited in mesangial cells ultimately require changes in the cytoskeletal organization whose architecture is modulated by the Rho family proteins. Therefore, understanding the signal transduction pathways associated with the activation of Rho family GTPase by ET and its regulation by ␤ 1 Pix is of pivotal importance in delineating the effects of ET on cytoskeletal organization in mesangial cells. was performed using Lipofectamine™ 2000 according to the manufacturer's instructions.
RT-PCR Analysis-Total RNA isolated from rat mesangial cells was reverse transcribed using Superscript reverse transcriptase (Invitrogen), oligo(dT) primers (Invitrogen), and deoxynucleotide triphosphate as specified by the manufacturer. ␤ 1 Pix was amplified by PCR with TITANIUM™ Taq polymerase (Clontech Laboratories, Inc.) in the presence of deoxynucleotide triphosphate, the forward primer 5Ј-GGAAT-TCCATGACTGATAACGCCAACAGCCAA-3Ј and the reverse primer 5Ј-GCTCTAGAGCTAGATTGGTCTCATCCCAAGCAGG-3Ј. The PCR products were subjected to electrophoresis in a 1% acrylamide gel, and the results were visualized with a Bio Imaging analyzer. The ␤ 1 Pix cDNA was cut with EcoRI and XbaI and inserted into the EcoRI-XbaI site of pcDNA3.1/Myc-His vector.
Pulldown Assays of Rho Family GTPases-Cells were transfected with empty vector, Myc-tagged ␤ 1 Pix, or its mutants for 24 h. After stimulation with ET-1, cells were lysed in lysis/wash buffer (25 mM HEPES, pH 7.5, 150 mM NaCl, 1% Igepal CA-630, 10 mM MgCl 2 , 1 mM EDTA, 1% glycerol, 10 g/ml leupeptin, and 10 g/ml aprotinin). To measure the active GTP-bound form of Rho family GTPases in the cell lysates, we performed pulldown assays (Cytoskeleton) using recombinant GST-tagged Pak-1⅐PBD (for Cdc42 and Rac1) and Rhetokin-RBD (for RhoA). Aliquots (500 g) of the supernatants mixed with glutathione-agarose with 10 g of GST⅐Pak-1⅐PBD or 10 g of GST⅐RBD were precipitated by centrifugation. Complexes were boiled in a Laemmli sample buffer and then separated on 15% SDS-polyacrylamide gels. The separated proteins were immunoblotted using each antibody against Rho family small GTPases.
In Vitro Kinase Assay-The ␤ 1 Pix-bound beads were resuspended in 80 l of kinase buffer (50 mM HEPES, pH 7.5, 10 mM MgCl 2 , 1 mM EGTA, 1 mM dithiothreitol, 0.015% Tween 20) and incubated with 30 units of protein kinase A catalytic subunit (Sigma), 20 Ci of [␥-32 P]ATP at 30°C for 30 min. The beads were pelleted by centrifugation and washed three times. The beads were then suspended in Laemmli sample buffer. Proteins were subjected to SDS-PAGE, Coomassie Blue staining, and autoradiography. For recombinant ␤ 1 Pix and its mutants, 5 g were used in the kinase reaction as described above.
Measurement of cAMP-A Biotrack cAMP enzyme immunoassay kit (Amersham Biosciences) was used to measure cyclic AMP levels in HMC according to the manufacturer's instructions.
PKA Assay-SignaTECT®cAMP-dependent PKA assay (Promega) was used to measure PKA enzyme activity according to the manufacturer's instructions. In brief, 20 l of substrate mixture containing 100 M kemptide peptide, 5 M cAMP, and 5 g of cell lysate were added in order. Then the reaction was started by adding 5 l of the mixture containing 0.5 mM ATP and 10 Ci/l of [␥-32 P]ATP (3,000 Ci/mmol). After incubation, 10 l of the mixture was spotted onto streptavidincoated membrane, washed repeatedly, dried, and placed in scintillation vials for radioactive counting.
Immunocytochemistry-100-mm dish-cultured human glomerular mesangial cells were transfected with 10 g of plasmid containing either wild type or mutant ␤ 1 Pix. The following day, transfected cells were seeded on glass coverslips housed in a 12-well plate, where all subsequent incubations were performed. After overnight serum starvation, cells were stimulated with ET-1, washed in PBS, and fixed in 4% paraformaldehyde in PBS for 15 min. Cells were then washed, permeabilized with 0.2% Triton X-100 in PBS for 5 min, and incubated in 10% goat serum in PBS for 30 min. Primary antibodies against c-Myc and paxillin (Santa Cruz) were diluted 1:500 and incubated with cells for 1 h. Cells were washed several times with PBS, followed by a 1-h incubation with Alexa Fluor 488 and Alexa Fluor 546 secondary antibodies (Molecular Probes) diluted 1:1000. Cells were equilibrated and mounted in ProLong anti-fade reagent (Molecular Probes). Cells were visualized under a fluorescent microscope (E600; Nikon) and photographed using the SPOT system (Diagnostic Instruments). Cells of roughly equal and average fluorescent intensity were chosen as comparative examples.

ET-1 Stimulates Cdc42 in Human Mesangial
Cells-We first sought to determine whether Cdc42 is activated by ET-1 in HMC. Using a pulldown assay to separate the activated GTPbound form of Cdc42, we found that treatment of HMC with ET-1 induced the formation of the GTP-bound form of Cdc42 (Fig. 1A). ET-1 stimulates Cdc42 in a dose-and time-dependent manner, peaking at 2 min, and this activation decreases 10 min after stimulation (Fig. 1B). The dependence of Cdc42 stimulation on ET-1 concentration was determined at 5 min. Maximal stimulation was observed at a concentration of 100 nM ET-1. To confirm that ET-1 specifically stimulates Cdc42, HMC were pretreated with 10 ng/ml of Clostridium difficile toxin B, which glycosylates and inactivates Rho family proteins (19), and then stimulated with ET-1. As shown in Fig. 1C, toxin B completely inhibited ET-1-induced Cdc42 stimulation.
We also found that ET-1 induced RhoA stimulation (Fig. 1D). However, ET-1-induced stimulation of Rac1 was not observed (data not shown).
Effect of ␤ 1 Pix Overexpression on ET-1-induced Cdc42 Activation-Pix family proteins are GEFs for small GTPase proteins Cdc42/Rac (9) and have been shown to signal via these proteins. Therefore, we studied the effect of ␤ 1 Pix and its inactive mutants on ET-1-induced Cdc42 activation in HMC. In our experiments, Cdc42 activation was measured after ET-1 treatment of HMC-overexpressing wild type rat ␤ 1 Pix or its mutants, ␤ 1 Pix SH3m(W43K) and ␤ 1 Pix DHm(L238R/L239S) (9). As shown in Fig. 2A, ET-1 induced Cdc42 activation in cells expressing empty vector, and this activation was enhanced by ␤ 1 Pix overexpression. By contrast, mutated ␤ 1 Pix DHm-(L238R/L239S), which lacks GEF activity, and SH3 domainmutated ␤ 1 Pix SH3m(W43K), which lacks the ability to bind to Pak, decreased ET-1-induced Cdc42 activation ( Fig. 2A). The expression of ␤ 1 Pix alone in the absence of ET-1 stimulation had no effect on Cdc42 activity (data not shown). Fig. 2B shows quantitative analysis of the results obtained by densitometry. ␤ 1 Pix overexpression did not enhance ET-1-induced RhoA activation (data not shown), indicating that ␤ 1 Pix specifically regulates the activation of Cdc42 by ET-1 in HMC.
Role of PKA in Stimulation of Cdc42 by ET-1-To determine the enzyme(s) involved in the regulation of Cdc42 activation by ET-1, we utilized different pharmacological inhibitors such as PKC inhibitor Ro31-7549, phosphatidylinositol-3-kinase inhibitor wortmannin, PKA inhibitor H-89, and PP2, a Src kinase inhibitor. We found that only H-89, a selective PKA inhibitor, blocked Cdc42 activation induced by ET-1 (Fig. 3A). The inhibitory effect of H-89 was confirmed using a myristoylated, cellpermeable peptide derivative of the naturally occurring PKA inhibitor, PKI (Fig. 3B). PKA is activated in response to cAMP generated by adenylate cyclase. To investigate the role of PKA in Cdc42 activation, HMC were stimulated either by ET-1, cAMP analog 8-Br-cAMP, or by the adenylate cyclase activator, forskolin. Cells transiently expressing the constitutively active ␣-subunit of G s served as positive control to mimic receptormediated activation of adenylate cyclase. Fig. 4A shows that ET-1 stimulated Cdc42 to a level similar to those produced by stimulating PKA, either by 8-Br-cAMP or forskolin or by expressing the constitutively active G␣ sQL . Treatment of the cells with H-89 inhibited Cdc42 activation induced by ET-1 and all other agonists mentioned above. These results indicate that Cdc42 activation in response to ET-1 requires PKA activity. We next wondered whether ET-1 is able to stimulate cAMP increase, which in turn activates PKA enzyme in the cells. In the presence of 3-isobutyl-1-methylxanthine, ET-1 induced a maximum of 2.5-fold cAMP accumulation over the basal level at 50 nM and higher concentrations (Fig. 4B). Next, we sought to determine whether ET-1 would stimulate PKA activity. Results show that stimulation by ET-1 (100 nM) increases PKA activity by 2-fold over non-stimulated cells (Fig. 4C). Cells expressing constitutively active G␣ sQL also showed an increase in PKA activity. In both cases the stimulated PKA activity was inhibited by H-89 to the basal level. Taken together, these results demonstrate that (i) activation of G s and increase in intracellular cAMP lead to the stimulation of Cdc42, and (ii) activation of PKA plays an important role in the regulation of ET-1-mediated stimulation of Cdc42.
Identification of PKA Phosphorylation Sites on ␤ 1 Pix-The functional activity of Cdc42 is known to be regulated by ␤ 1 Pix, which controls its GDP/GTP-bound state. The fact that PKA inhibitors H-89 and PKI both blocked ET-1-induced Cdc42 activation prompted us to examine whether ␤ 1 Pix was phosphorylated by PKA. Upon examination of the ␤ 1 Pix amino acid sequence, we found two potential phosphorylation sites based on the PKA consensus sequence represented by RKXS/T. Interestingly, both potential PKA phosphorylation sites, Ser-516 and Thr-526, are highly conserved in ␣Pix, ␤ 1 Pix, and ␤ 2 Pix, suggesting that these residues may be of functional significance.
To further confirm that ␤ 1 Pix is a direct target for PKA phosphorylation in vitro, we performed an in vitro kinase assay using purified recombinant ␤ 1 Pix and its mutants as substrate. Autoradiography of reaction products showed that recombi-nant ␤ 1 Pix(wt) was phosphorylated by PKA (Fig. 5B, lane 1). Substitution of Ser-516, Thr-526, or both to Ala strongly reduced phosphorylation as compared with ␤ 1 Pix(wt). Slight residual phosphorylation suggested that an additional residue could be targeted by PKA under these optimal conditions. Coomassie staining of gels demonstrated comparable loading of proteins (Fig. 4, lower panels). The apparent discrepancy between the ability of PKA to phosphorylate immunoprecipitated ␤ 1 Pix(T526A) (Fig. 5A, lane 4) and not recombinant ␤ 1 Pix(T526A) (Fig. 5B, lane 3) suggests the presence of other residues phosphorylated by PKA or the presence of endogenous Pak in the immunoprecipitate. Indeed, ␤ 1 Pix is a well known target for Pak phosphorylation (10). In conclusion, our results show for the first time that Ser-516 and Thr-526 are the major in vitro phosphorylation sites by PKA on ␤ 1 Pix. ␤ 1 Pix Phosphorylation by PKA Mediates ET-1-and cAMPinduced Cdc42 Activation-To determine whether ␤ 1 Pix phosphorylation at Ser-516 and Thr-526 has a functional significance, we investigated the effect of ␤ 1 Pix double mutant ␤ 1 Pix(S516A/T526A) on Cdc42 activation by ET-1. We hypothesized that ␤ 1 Pix(S516A/T526A) is a biologically inactive mutant and therefore will inhibit ET-1-induced Cdc42 activation. To specifically measure the effect of ␤ 1 Pix and its mutant on Cdc42 activity, we co-transfected HMC with HA-tagged Cdc42 and either Myc-tagged ␤ 1 Pix or Myc-tagged ␤ 1 Pix(S516A/ T526A). The co-transfection enabled us to increase the sensitivity of measuring activated Cdc42 only in the cells expressing both ␤ 1 Pix and Cdc42, given the transient nature of transfection used. In this assay we detected GST⅐PBD bound to active HA-Cdc42 by immunoblotting with anti-HA tag antibody instead of using anti-Cdc42. Consistent with our previous results, ET-1 treatment induced Cdc42 activation (Fig. 6A). The Cdc42 activation was greatly increased in cells co-expressing HA-Cdc42 and ␤ 1 Pix. However, in cells co-expressing HA-Cdc42 and ␤ 1 Pix(S516A/T526A), activation of Cdc42 was completely inhibited in response to ET-1 (Fig. 6A).
PKA-dependent Translocation of ␤ 1 Pix to Focal Complexes-It has been shown that ␤ 1 Pix⅐Pak complex co-localizes to focal complexes where p95PKL plays a role in mediating the association of ␤ 1 Pix with paxillin (9, 13). Expression and localization of Myc-tagged ␤ 1 Pix were analyzed under a fluorescent microscope using anti-c-Myc antibody. Having found that H-89 inhibited Cdc42 activation induced by ET-1, we examined its effect on ET-1-induced ␤ 1 Pix translocation to focal complexes. The control cells show a diffuse distribution of ␤ 1 Pix throughout the cytosol (Fig. 7a). ET-1 treatment induced ␤ 1 Pix translocation to focal complexes (Fig. 7, b and c) that co-localized with paxillin (Fig. 7, lower panels). Pretreatment of HMC with 10 M H-89 (Fig. 7d) or with 1 M PKI (Fig. 7f) for 30 min completely inhibits ET-1-induced ␤ 1 Pix translocation to focal complexes. PKI alone has no effect on ␤ 1 Pix distribution (Fig.  7e). Altogether, these results indicate that ET-1-induced ␤ 1 Pix translocation is dependent on PKA activity. The next step was to determine whether the phosphorylation of Ser-516 and Thr-526 residues is important for ␤ 1 Pix translocation. To this end we compared the targeting of ␤ 1 Pix wild type with that of double mutant ␤ 1 Pix(S516A/T526A). The unstimulated cells expressing ␤ 1 Pix(S516A/T526A) show a comparable distribution (Fig. 7g) as unstimulated cells expressing ␤ 1 Pix wild type (Fig. 7a). However, in ET-1-stimulated cells the translocation of ␤ 1 Pix(S516A/T526A) was strongly inhibited (Fig. 7h), indicating that the phosphorylation of these two residues is functionally important to ␤ 1 Pix complex translocation to the focal complexes. The mutation of either Ser-516 or Thr-526 alone did not inhibit ␤ 1 Pix translocation after ET-1 stimulation (data not shown).

DISCUSSION
In the present study, we have reported for the first time that ET-1 stimulates the GTPase Cdc42 by a ␤ 1 Pix-mediated, PKAdependent pathway. Moreover, we have demonstrated that ET-1-induced ␤ 1 Pix translocation also requires PKA activity.
Many G protein-coupled receptor agonists activate small GT-Pases that, in turn, regulate a variety of biological responses such as cell differentiation and growth (2). Our results have shown that ET-1 stimulates Cdc42 and RhoA but fails to stimulate Rac1. A critical role for Cdc42 in ET-1 signaling is further supported by recent findings (20) showing that ET-1 stimulates Cdc42, but not Rac1, in human kidney epithelial cells.
PKA is involved in ET-1-induced Cdc42 activation because PKA inhibitors blocked Cdc42 stimulation. Our results demonstrating ET-1-induced cAMP increase and activation of PKA suggest that ET-1 activates Cdc42 via cAMP production by a G␣ s -dependent pathway. Indeed, we found that stimulation of adenylate cyclase by forskolin, or loading of HMC with the cAMP analog 8-Br-cAMP, induced Cdc42 activation. Further- Control reactions were carried out in the absence of PKA catalytic subunit (5) or in the absence of ␤ 1 Pix (6). The phosphorylated products were analyzed on SDS-PAGE gel and exposed to x-ray film. Equal loading is shown by Coomassie Blue-stained gels (lower panels). Similar results were obtained in four independent experiments. more, the expression of constitutively active G␣ s induced Cdc42 activation. The data presented here show cross-talk between cAMP and Cdc42. It is unknown how widespread this phenomenon is, but previous study showed that Cdc42 activation by cAMP occurs in human mast cells, Chinese hamster ovary-K1 cells, and COS-7 cells (21). Our finding that cAMP can activate Cdc42 extends the spectrum of possible pathways involved in transducing a signal from G protein-coupled receptor to Cdc42.
The activity of small GTPase proteins is regulated by GEFs that modulate their GDP/GTP-bound state. It is well established now that upon activation Cdc42 and Rac 1 bind to Pak and form an active complex that regulates several signaling pathways, including cytoskeletal rearrangement (22)(23)(24)(25). To further examine the mechanism involved in the regulation of Cdc42 stimulation, we investigated the role of the Cdc42/Rac1-GEF, ␤ 1 Pix, in ET-1-induced Cdc42 activation in HMC. Using ␤ 1 Pix and its double mutant, ␤ 1 Pix(L238R/L239S), we showed that ␤ 1 Pix acts upstream of Cdc42 and provides a key link between ET-1 receptor and Cdc42. In addition, we confirmed a direct role of PKA by establishing ␤ 1 Pix as an in vitro substrate of PKA at Ser-516 and Thr-526. The mutation of both residues inhibited Cdc42 activation and ␤ 1 Pix translocation. However, the mutation of a single residue does not have functional importance. Thr-526 is also phosphorylated by Pak (9), which makes this residue a molecular convergent point for PKA and Pak. Furthermore, inhibition of PKA blocks ET-1-induced ␤ 1 Pix translocation to focal complexes (Fig. 7d). This has prompted us to speculate that phosphorylation of ␤ 1 Pix by PKA is the mechanism by which ␤ 1 Pix complex is recruited to focal complexes where it binds p95PKL (13). PKA phosphorylation of ␤ 1 Pix could also accomplish Cdc42 activation by bringing ␤ 1 Pix within the vicinity of its target Cdc42, as seen in the case of the Ras exchange factor Sos, which is brought to the membrane by binding to Grb2, thereby enabling Ras activation (26,27). The inhibition of PKA by H-89 or the expression of ␤ 1 Pix double mutant ␤ 1 Pix(S516A/T526A) induced a complete inhibition of Cdc42 activation by ET-1 or cAMP analog (Fig. 6), which confirms the functional role of ␤ 1 Pix phosphorylation by PKA. Moreover, the expression of ␤ 1 Pix(S516A/T526A) results in a very strong inhibition of ␤ 1 Pix translocation induced by ET-1 (Fig. 7f). This result indicates that phosphorylation by PKA may induce conformational changes in ␤ 1 Pix that facilitate or stabilize ␤ 1 Pix complex formation that allow its proper targeting. Previous studies have shown that the stimulation of extracellular signal-related kinase pathway in PC12 cells induces Pak2⅐␤ 1 Pix complex translocation through phosphorylation of ␤ 1 Pix on Ser-525 and Thr-526 (28,29). However, in these studies Pix translocation was found to occur independently of Rac1 activation. Considering that Pak acts downstream of PKA, the phosphorylation of ␤ 1 Pix by PKA might provide the first step before Pak phosphorylation provides the signal for ␤ 1 Pix complex to be recruited to focal complexes. ␤ 1 Pix is usually associated with a multicomplex of proteins binding to paxillin and Pak, an effector of Cdc42 function associated with filopodia formation (30,31). This complex is linked to paxillin through an ADP-ribosylation factor GAP protein, p95PKL (13). Although ADP-ribosylation factor GTPases play a critical role in vesicle transport, they have been involved in regulating the actin cytoskeleton (32) and the translocation to the plasma membrane of focal adhesion proteins such as paxillin (33). Alignments between different Pix isoforms show that Ser-516 and Thr-526 are highly conserved, indicating that these residues may have a functional importance in the formation of these signaling complexes. The physiological significance of ␤ 1 Pix phosphorylation remains unclear. Increasing evidence indicates that phosphorylation is an important mechanism of regulation of GEFs of the diffuse B cell lymphoma family. Although it has been shown that the phosphorylation status of ␤ 1 Pix has no effect on its GEF activity (9), recent findings show that the activation of Rac1 is dependent on ␤ 1 Pix phosphorylation (29).
We can speculate that PKA regulation of Cdc42 activation through phosphorylation of ␤ 1 Pix may have an important consequence for regulating mesangial cell migration. Indeed, it has been shown that increased Cdc42 activity results in increased cell motility (34). During initial stages of mesangial cell migration new protrusions are formed at the leading edge of the cell. It can be hypothesized that PKA amplifies Cdc42 activation through ␤ 1 Pix phosphorylation, resulting in the initiation of microspike formation and cell protrusion. This scenario implies that activated Cdc42 is localized at the leading edge of migrating cells. In summary, our findings describe a new ET-1 receptor-initiated pathway that involves the activation of cAMP/PKA pathway and results in the activation of ␤ 1 Pix translocation and Cdc42; the findings identify important venues for future work regarding the role of this pathway in mesangial cell migration.