Central Role of Calcium-Dependent Tyrosine Kinase PYK2 in Endothelial Nitric Oxide Synthase–Mediated Angiogenic Response and Vascular Function
Background— The involvement of Ca2+-dependent tyrosine kinase PYK2 in the Akt/endothelial NO synthase pathway remains to be determined.
Methods and Results— Blood flow recovery and neovessel formation after hind-limb ischemia were impaired in PYK2−/− mice with reduced mobilization of endothelial progenitors. Vascular endothelial growth factor (VEGF)–mediated cytoplasmic Ca2+ mobilization and Ca2+-independent Akt activation were markedly decreased in the PYK2-deficient aortic endothelial cells, whereas the Ca2+-independent AMP-activated protein kinase/protein kinase-A pathway that phosphorylates endothelial NO synthase was not impaired. Acetylcholine-mediated aortic vasorelaxation and cGMP production were significantly decreased. Vascular endothelial growth factor–dependent migration, tube formation, and actin cytoskeletal reorganization associated with Rac1 activation were inhibited in PYK2-deficient endothelial cells. PI3-kinase is associated with vascular endothelial growth factor–induced PYK2/Src complex, and inhibition of Src blocked Akt activation. The vascular endothelial growth factor–mediated Src association with PLCγ1 and phosphorylation of 783Tyr-PLCγ1 also were abolished by PYK2 deficiency.
Conclusion— These findings demonstrate that PYK2 is closely involved in receptor- or ischemia-activated signaling events via Src/PLCγ1 and Src/PI3-kinase/Akt pathways, leading to endothelial NO synthase phosphorylation, and thus modulates endothelial NO synthase–mediated vasoactive function and angiogenic response.
Received August 9, 2005; accepted June 12, 2007.
Nitric oxide (NO) has multiple functions in NO-mediated vascular action and angiogenic response. This was confirmed by endothelial NO synthase (eNOS)−/− mice exhibiting hypertension1 or impaired vascular endothelial growth factor (VEGF) –induced angiogenesis.2 VEGF phosphorylates eNOS, which is directly activated on the phosphorylation of 1177serine in human (1176serine in mouse) by Akt,3,4 whereas the upstream molecules that activate Akt-eNOS system have not been fully clarified.
Clinical Perspective p 1051
Tyrosine kinases activate the PI3-kinase/Akt or Ca2+ signaling pathways, suggesting that tyrosine kinase is upstream of eNOS. Indeed, Src and VEGF receptor-2 activate eNOS through activation of the PI3-kinase/Akt pathway.5 PYK2 (proline-rich tyrosine kinase), also known as RAFTK, CAK, and CADTK,6,7 is the cytoplasmic tyrosine kinase and exhibits 45% amino acid sequence identity to focal adhesion kinase. Tyrosine phosphoryla-tion of PYK2 and focal adhesion kinase was triggered by integrin-mediated adhesion.8 PYK2 was stimulated by a broad range of physiological stimuli such as stimuli for G-protein–coupled receptors that elevate intracellular Ca2+,6,7 phorbol ester, inflammatory cytokines, and stress signals, including ischemia.9 PYK2 acts in concert with Src, which links Gi- or Gq-coupled receptors, leading to the MAP kinase pathway.10 Furthermore, PYK2 binds to proteins that interact with the cytoskeleton, suggesting a role in the regulation of cellular morphology. The phenotype of PYK2-deficient mice was recently described as having macrophages that exhibit impaired migration as a result of cytoskeleton abnormality induced by diminished Ca2+ mobilization and reduced activation of PI3-kinase.11 In this study, we newly generated PYK2−/− mice and investigated the molecular mechanism for the effects of PYK2 on NO-mediated vascular function and angiogenic response.
Results are expressed as mean±SEM. All data were transformed by the natural logarithm before ANOVA corresponding to each experiment. Repeated-measures ANOVA was used to analyze the time course experiment. The Scheffé test was used as the multiple comparison test. For comparisons between 2 groups, a 2-sample t test was performed. A value of P< 0.05 (2 tailed) was considered statistically significant.
Materials, construction of targeting vector, generation of PYK2−/− mice (Figure I of the online Data Supplement), Western blotting, immunohistochemistry, transfection of DNA plasmid, measurement of GTP-Rho and GTP-Rac, migration, tubular formation, in vivo angiogenesis, preparation of endothelial progenitor cell (EPC) –like cells, hind-limb ischemia, laser Doppler perfusion image, cGMP assay, measurement of NO metabolites and NO levels, acetylcholine (Ach)- and nitroprusside-mediated vasodilatation, measurement of cytoplasmic Ca2+ concentration, fluorescence-activated cell sorting, and isolation of endothelial cells (ECs) from aorta and primary culture are described in the Methods section of the online Data Supplement. C57B1/6 strain mice (SHIMIZU Laboratory Supplies, Kyoto, Japan) were used. The animal experiments were approved by our institutional review board.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Impaired eNOS/Akt Activation and Ca2+ Mobilization by PYK2 Deficiency
eNOS was reported to be activated by VEGF, Ach, or ischemic stress.12 Incubation of the aorta with VEGF (100 ng/mL) phosphorylates PYK2 time dependently with a peak level around 5 minutes (Figure 1A). Ach (1 μmol/L) also caused PYK2 phosphorylation with a similar peak level, the extent of which was comparable to that in VEGF stimulation (Figure 1A). Ischemic stress time dependently increased the phosphorylation of PYK2 in the hind-limb muscle (Figure 1B). Such PYK2 phosphorylation was observed in the primary cultured aortic von Willebrand factor–positive ECs after stimulation with VEGF (100 ng/mL) and Ach (1 μmol/L) and exposure to 1% hypoxia (Figure 1C). To clarify the cell types expressing PYK2, an immunohistochemical analysis was performed. Figure 1D showed that PYK2 was present mainly in the endothelial (CD31+ ECs) and medial layers (vascular smooth muscle cells) in the aorta and in the CD31+ vessels in the skeletal muscle, in which a few cells in the interstitial region (CD31−) also expressed PYK2.
We next examined whether the stimuli that induce PYK2 activation led to the phosphorylation of eNOS. Hind-limb ischemia causes eNOS phosphorylation in the skeletal muscle in wild-type (+/+) mice, whereas eNOS activation in the PYK2−/− mice was markedly inhibited (Figure 2A). Exposure of the aorta to VEGF or Ach also induced eNOS phosphorylation, whereas their activated levels were significantly diminished in the PYK2−/− mice (Figure 2B). Induction of eNOS phosphorylation by VEGF or Ach or exposure to 1% hypoxia also was observed (2.3- to 2.5-fold, respectively; P<0.005) in the wild-type aortic ECs but strongly inhibited in the PYK2-deficient ECs (Figure 2C).
The ischemia-induced Akt phosphorylation in the skeletal muscle was significantly lower in the PYK2−/− mice (46±3% at 2 hours, 34±3% decrease 1 day after ischemia; P<0.01) than the wild-type mice (Figure 2A), whereas the tyrosine phosphorylation level in VEGF receptor-2 (Flk-1) did not significantly differ between the wild-type and PYK2-deficient muscle (n=6 each; data not shown).
Akt phosphorylation in the aorta (Figure 2B) and ECs (Figure 2C) from the wild-type mice was significantly increased by VEGF stimulation (1.7±0.3-fold, P<0.05; and 3.0±0.7-fold, P<0.005, respectively), whereas Akt activation in the PYK2-deficient aorta and ECs was markedly attenuated (43±3% and 42±6% inhibition, respectively; P<0.05). Significant inhibition of Akt phosphorylation in ECs by PYK2 deficiency also was observed after exposure to 1% hypoxia (72±5% inhibition; P<0.01; Figure 2C).
We examined the involvement of AMP-activated protein kinase (AMPK)13 and cAMP-dependent protein kinase-A (PKA),14 known as the Ca2+-independent kinase for phosphorylation of 1176Ser-eNOS. The phosphorylation levels of AMPK and PKA after hind-limb ischemia did not differ significantly between PYK2−/− and wild-type mice (Figure 2A). Furthermore, the phosphorylation of AMPK and PKA in PYK2-deficient ECs exposed to 1% hypoxia for 18 hours also was similar to the wild-type ECs (data not shown). Akt, AMPK, and PKA showed peak phosphorylation on day 1 and at 2 hours, respectively; PKA and AMPK then reversed to baseline levels on day 7, whereas moderate activation of Akt was observed on day 7 (220±30% increase compared with basal level; Figure 2A). Neither AMPK nor PKA was activated 5 minutes after VEGF (100 ng/mL) treatment in both the wild-type and PYK2-deficient cells (data not shown), whereas dibutylic cAMP (1 mmol/L) and 5-aminoimidazole-4-carboxamide-1-β-d-riboside AICAR (1 mmol/L) apparently phosphorylated PKA and AMPK in the wild-type ECs (positive controls; data not shown). These findings suggest that Akt, rather than Ca2+-independent AMPK or PKA, is involved in VEGF-mediated eNOS phosphorylation.
To prove that the reduced phosphorylation of eNOS is due directly to the loss of PYK2, we transfected GFP-tagged PYK2 plasmid to the PYK2-deficient ECs and studied by immunostaining with anti–phospho-1176Ser-eNOS antibody whether VEGF-mediated phosphorylation of eNOS can be restored (Figure 2D). eNOS phosphorylation was observed in VEGF-exposed cells in which GFP-tagged PYK2 plasmid was transfected, whereas phospho-eNOS–positive cells were barely detected in the control GFP-transfected cells.
We next studied whether intracellular Ca2+ mobilization was influenced by PYK2 deficiency. Figure 3A shows that VEGF-mediated elevation of cytoplasmic Ca2+ levels was markedly inhibited in the PYK2-deficient ECs (Ca2+ concentrations in PYK2−/−, 95±24 nmol/L versus wild type, 432±69 nmol/L; P<0.001), whereas Ca2+ mobilization with ATP that directly opens the Ca2+ channel on the plasma membrane15 was comparable to the wild type (Figure 3A), suggesting that the pathway for the receptor-independent Ca2+ mobilization is not impaired. To study the effect of PYK2 deficiency on the other Ca2+ signaling pathway not involving NO formation, we studied the activation of the Ca2+-dependent transcription factor nuclear factor of activated T cells 2 (NFATc2), which was shown to be crucial for VEGF-mediated angiogenesis.16 The results showed that 68±8% of the wild-type cells showed nuclear translocation of NFATc2 from cytoplasm 30 minutes after VEGF stimulation, whereas in the PYK2-deficient cells, the translocation was markedly reduced (22±6%; P<0.01), indicating that PYK2 deficiency inhibits Ca2+ signaling pathway not involving NO formation (Figure 3B). To prove that the functional effects are due directly to the loss of PYK2, we transfected the GFP-tagged PYK2 plasmid to the PYK2-deficient cells and observed VEGF-mediated nuclear translocation of NFATc2. In the PYK2-deficient cells transfected with GFP-tagged PYK2 plasmid, the number of NFATc2-translocated cells increased significantly (from 22±6% to 48±9%; P<0.05; Figure 3B), whereas in the control GFP-transfected cells, the translocated cells did not increase significantly (data not shown). Pretreatment with a chelator of intracellular Ca2+ store, AM-BAPTA (5 μmol/L), in the Ca2+-free medium for 45 minutes did not affect VEGF-induced Akt activation in ECs (Figure 3C), indicating that Akt activation is Ca2+ independent in our system.
Reduced Response by PYK2 Deficiency in NO Production and Ach-Mediated Vasodilatation
The intracellular NO level was measured with 4-amino-5-methyl-amino-2′,7′-difluorofluorescein diacetate (DAF-FM DA). NO was visualized as green under laser microscopy (Figure 4A). In the wild-type ECs, VEGF increased NO levels by 3.2-fold (P<0.005 versus untreated cells), whereas this increase was severely impaired in the PYK2-deficient ECs. NO metabolites in the 24-hour urine sample were significantly lower in the PYK2−/− mice than the wild-type mice (4084±820 versus 9326±1163 nmol; P<0.005; Figure 4B). Oral administration of NG-nitro-l-arginine methyl ester (L-NAME; 3 mmol/L in drinking water) to the wild-type mice reduced the NO metabolite production to a level comparable to the PYK2−/− mice (Figure 4B).
Ach-mediated vasodilatation is induced by the eNOS-NO system. We next examined whether Ach-mediated relaxation of the aorta is influenced by PYK2 deficiency. The dose-dependent relaxation of the isolated aorta constricted by norepinephrine was evaluated (Figure 4C). Ach (10 μmol/L) -mediated relaxation response was much weaker in the PYK2-deficient aorta than in the wild-type aorta (32±8% versus 59±5% relaxation; P<0.01), whereas norepinephrine (300 nmol/L) -mediated vasoconstriction in the PYK2-deficient aorta did not significantly differ from the wild-type aorta (190±30% versus 165±6% relative to 50 mmol/L KCl–mediated constriction; n=5). This result suggests that Ca2+ signaling transmitted mainly through voltage-dependent Ca2+ channel leading to the constriction of vascular smooth muscle cells is not impaired in the PYK2-deficient aorta.
To evaluate the NO dependency on Ach-mediated vasorelaxation, the effect of L-NAME was studied. Addition of L-NAME (10 μmol/L) markedly suppressed the Ach-mediated maximum relaxation of the wild-type aorta (from 59±5% to 17±2%; P<0.001; n=4), whereas in the PYK2-deficient aorta, the reduced relaxation response was suppressed further to a level comparable to L-NAME–treated wild-type aorta (from 32±8 to 18±3%; P<0.005; n=4). These findings indicate that Ach-mediated vasorelaxation depends mainly on NO production, in which an involvement of PYK2-mediated NO signaling was estimated to be ≈64% [(59−32/59−17)×100] of the total NO-mediated vasodilation activated downstream of Ach. NO donor (nitroprusside) –induced vasorelaxation of the aorta was similar in both groups (82±4 versus 83±4%; n=5), suggesting that NO-mediated signal transduction for vasodilatation is not impaired in the PYK2-deficient aorta (Figure 4C).
NO-mediated vasodilatation requires cGMP as a second messenger. We therefore measured the amount of cGMP in the aorta. Basal cGMP production in the PYK2−/− mice was 41% lower than that in the wild-type mice. Ach increased aortic cGMP production 4.8-fold in the wild-type mice, whereas the increase in the PYK2−/− mice was 2.0-fold, significantly (P<0.01) lower than in the wild-type mice (Figure 4D).
Decrease in Neovessel Formation by PYK2 Deficiency
Angiogenesis in the ischemic tissue requires eNOS activation.17 We analyzed the blood flow recovery and neovessel formation after hind-limb ischemia. The ratio of blood flow recovery assessed by laser Doppler imaging was significantly lower in the PYK2−/− mice than in the wild-type mice (50% versus 76% recovery at 3 weeks after ischemia; P<0.01; Figure 5A). Oral administration of L-NAME (3 mmol/L in drinking water) significantly reduced the recovery ratio in the wild-type (from 76% to 62%; P<0.05) and PYK2−/− (from 50% to 37%; P<0.05) mice (Figure 5A). Considering that the recovery ratio of the L-NAME–treated wild-type mice (62%) is close to that of the PYK2−/− mice (50%), the blood flow recovery after hind-limb ischemia is considered to be regulated mainly by PYK2-mediated NO signaling. We also counted the number of CD31+ vessels in the ischemic muscle. There was no significant difference in the basal vessel numbers surrounding the muscle fibers. The vessel number (per muscle fiber) in the wild-type mice increased 2.3-fold (P<0.005) after hind-limb ischemia, whereas the PYK2−/− mice showed no significant increase (Figure 5B).
We next examined whether PYK2 deficiency affects the mobilization or differentiation of EPCs. We found that 3 days after limb ischemia, the number of circulating CD45−/Flk-1+ EPCs was significantly lower (36%; P<0.05) in the PYK2−/− mice than the wild-type mice (0.28±0.06% and 0.18±0.03% relative to total peripheral blood mononuclear cells, respectively; n=10 each; Figure 5C). Considering that the mobilization of EPCs was reportedly regulated by the eNOS function of EPCs in a VEGF-dependent manner,18 the present study suggests that PYK2 deficiency attenuates VEGF-mediated EPC mobilization by impairing Ca2+/eNOS signaling.
Impaired cGMP-Dependent Protein Kinase– and eNOS-Mediated Migration of PYK2-Deficient ECs
NO-mediated angiogenesis depends on the migration of ECs, in which cGMP-dependent protein kinase (GK) or eNOS plays a crucial role.19 Because it was reported that the migration of macrophages was severely impaired in PYK2−/− mice,11 the migration activity of aortic ECs was evaluated by the Boyden chamber assay. VEGF-mediated migration was 41% lower (P<0.01) in the PYK2-deficient ECs than wild-type ECs (Figure 6A). Pretreatment with L-NAME (3 mmol/L) or Rp-8-Br-cGMPS (1 μmol/L, a GK inhibitor) markedly inhibited the VEGF-mediated migration activities of wild-type ECs (50% and 45% inhibition, respectively; Figure 6A). l-Arginine (1 mmol/L) or GK activator 8-Bromo-cGMP (1 μmol/L) treatment restored the reduced migration of PYK2-deficient ECs to the wild-type level, whereas L-NAME or GK inhibitor (1 μmol/L) did not significantly affect their migration activities (Figure 6A).
The tubular formation activity and 3-dimensional angiogenesis in the Matrigel plug were significantly decreased in the PYK2-deficient ECs (59% and 38% decrease, respectively; Figure 6B and 6C). L-NAME diminished the tubular formation of wild-type ECs (39%; P<0.05), and L-arginine treatment improved the decreased activity of PYK2-deficient ECs (52%; P<0.01) (Figure 6B). These findings suggested that PYK2-mediated migration of ECs and angiogenic response are regulated mainly by NO and GK activation.
Considering that GK affects the cytoskeleton structure,20 PYK2 may modulate the cytoskeleton structure, leading to the regulation of cell migration and eventually angiogenesis. We therefore examined the effect of PYK2 on the F-actin structure of ECs. In the basal condition, we observed the stress fibers in 55±5% of the total wild-type ECs attaching on a fibronectin-coated dish and in 76±5% of PYK2-deficient ECs (P<0.05) (Figure 7A). VEGF stimulation markedly decreased the number of stress fiber–positive wild-type cells (from 55±5% to 15±4%; P<0.01) and 64±6% of the total attaching wild-type cells exhibited the accumulation of F-actin at the plasma membrane, whereas in the PYK2-deficient ECs, 72±8% of cells still had the apparent stress fibers and the cells showing the F-actin accumulation at the plasma membrane was only 23±3%. Pretreatment with L-NAME inhibited VEGF-mediated F-actin accumulation at the plasma membrane (P<0.01) and increased the stress fiber formation (P<0.05) in the wild-type cells, whereas the addition of L-arginine to PYK2-deficient ECs attenuated the stress fiber formation (P<0.01) and enhanced the VEGF-mediated F-actin accumulation at the plasma membrane (P<0.05).
Because F-actin structure was regulated by Rho-family small GTPases, we evaluated the activities of RhoA and Rac1 by pull-down assay using GST-RBD21 and GST-PBD,22 respectively (Figure 7B). The amount of GTP-bound RhoA was decreased significantly in the wild-type ECs 5 minutes after VEGF treatment, whereas in the PYK2-deficient ECs, they were reduced more extensively than in the wild-type ECs. In contrast, the amount of GTP-bound Rac1 was increased markedly in the wild-type ECs, whereas its increase was abolished by the PYK2 deficiency. Considering that Rac1 plays a pivotal role in F-actin reorganization23 and PYK2 promotes Rac-mediated JNK activation,24 lack of VEGF-mediated Rac1 activation may be associated with the altered F-actin structure in the PYK2-deficient cells. Furthermore, treatment with L-NAME attenuated VEGF-mediated Rac1 activation in the wild-type ECs, whereas addition of L-arginine to the PYK2-deficient ECs restored the response, indicating that PYK2-mediated Rac1 activation is NO dependent (Figure 7B).
Taniyama et al25 showed that in the angiotensin II–stimulated cells (vascular smooth muscle cells), Ca2+-activated PYK2 acts as a scaffold for Src-dependent phosphorylation of 3-phosphoinositide-dependent protein kinase, the activator of Akt, whereas VEGF-induced Akt activation appears to be Ca2+ independent (Figure 3C). We therefore evaluated the target of PYK2 in the VEGF-mediated signaling pathways leading to Akt activation or Ca2+ mobilization. We found that VEGF stimulation causes PYK2 association with Src in the wild-type ECs, leading to the phosphorylation of Src (Figure 7C) and Akt (Figures 2B and 7⇑C). PYK2 deficiency significantly inhibited VEGF-mediated Src and Akt activation, and inhibition of Src activity by PP1 blocked Akt and PYK2 phosphorylation (Figures 2B and 7⇑C). Immunoprecipitation experiments indicated that Src, but not PYK2, is closely associated with the p85 subunit of PI3K and that the Src/PI3K complex binds to PYK2 in response to VEGF (Figure 7C, bottom).
To study PYK2-mediated Ca2+ signaling after stimulation with VEGF, we studied PLCγ1 activation, known to cause an increase in Ca2+ level.26 We found that VEGF-mediated Src association with PLCγ1 and phosphorylation of 783Tyr-PLCγ1 (both basal and VEGF induced) were significantly decreased in the PYK2-deficient cells and that the treatment with the Src inhibitor PP1 abolished VEGF-induced PLCγ1 phosphorylation (Figure 7D).
Taken together, it is likely that the direct target of PYK2 is Src and that Src-bound PI3-kinase and Src-bound PLCγ are involved in activation of Akt and Ca2+ mobilization, respectively.
Tyrosine kinases have been assumed to be the upstream molecule for PI3K-Akt-eNOS or Ca2+-eNOS pathways. eNOS is activated by Akt, and intracellular Ca2+ upregulates eNOS activity, raising the possibility that Ca2+-dependent tyrosine kinase PYK2 is a possible eNOS activator. This study provides the first evidence that (1) PYK2 deficiency attenuates VEGF-mediated eNOS phosphorylation associated with decreased Akt activation and intracellular Ca2+ mobilization, (2) PYK2 is associated with the Src/PI3K complex and inhibition of Src blocked Akt phosphorylation, (3) Ach-mediated vasodilation of the aorta was diminished by decreased cGMP production in PYK2−/− mice, and (4) PYK2 plays a central role in VEGF- or ischemia-mediated eNOS activation followed by angiogenic response in which VEGF-induced EPC mobilization, VEGF-dependent migration, actin cytoskeletal reorganization associated with reduced Rac1 activation were markedly inhibited in the PYK2-deficient ECs. Cell migration reportedly requires the recycled mobilization of actin from the old focal adhesion toward the plasma membrane of the leading edge to form the lamellipodia and new focal contact,27 suggesting that impaired reorganization of the F-actin at the plasma membrane plausibly causes attenuated migration of the PYK2-deficient ECs.
eNOS activation is dependent on an increase in [Ca2+]i and the binding of Ca2+/calmodulin to the enzyme, leading to conformational change to displace the autoinhibitory loop.28 If Ca2+ mobilization is totally abolished, administration of the eNOS substrate arginine could not restore eNOS activity. However, VEGF-induced increase in [Ca2+]i and subsequent translocation of NFATc2 in PYK2-deficient ECs remain ≈23% and ≈32%, respectively, of the wild-type cells (Figure 3). This increase in [Ca2+]i could cause the conformational change in eNOS, resulting in the recovery in eNOS activation after arginine administration.
The present study showed that PYK2 deficiency attenuates VEGF-mediated association of Src with PLCγ1 and phosphorylation of PLCγ1. VEGF was shown to stimulate the association of VEGF receptor-2 with Src, and subsequent Src activation was a requisite for VEGF-mediated PLCγ1 activation.29 Taken together, it is conceivable that the lack of intracellular Ca2+ mobilization in VEGF-stimulated PYK2-deficient cells is due to the inhibition of Src-associated PLCγ activation.
Src and PYK2 are mutually activated in a stepwise manner. Association of Src-SH2-domain with 402Tyr-phosphorylated-PYK2 leads to conformational change in Src to release the internal autoinhibition, resulting in the upregulation of its activity. Conversely, activated Src phosphorylates 881Tyr-PYK2, leading to the downstream Grb2/Ras/MAPK pathway.10 In response to VEGF receptor-2 stimulation, Src binds toward the phosphorylated 1212Tyr of VEGF receptor-2 with its SH2 domain, leading to Src activation.30 In addition, VEGF promotes association of VEGF receptor-2 with integrin (αVβ3) and transmits integrin-dependent cell biological responses.31 Activation of integrin (αVβ3) induces phosphorylation of 402Tyr-PYK2 and its association with integrin β3.32 Integrin-activated PYK2 is associated with Src33 and involved in VEGF-mediated cell migration.34 Furthermore, integrin (αVβ5) plays a crucial role in angiogenesis,35 and inhibition of integrin (αVβ5) disrupted VEGF-mediated and Src-dependent angiogenesis.36 Thus, PYK2/Src complex is likely to integrate integrin with the VEGF receptor-2 signaling system.
Fluid shear stress–mediated activation of eNOS is dependent on Ca2+ mobilized through a mechanical stress–activated Ca2+ channel on the plasma membrane, whereas a Ca2+-independent system has been reported recently. Fleming et al37 demonstrated that shear stress elicits Src-mediated phosphorylation of platelet EC adhesion molecule-1 (PECAM-1) at the cell-to-cell contact, which is crucial for subsequent activation of Akt and eNOS. Furthermore, Tzima et al38 showed that PECAM-1 forms a mechanosensory complex with VEGF receptor-2, leading to activation of PI3-kinase. These findings suggest that VEGF receptor-2 is involved in Src/PECAM-1–mediated Akt/eNOS activation in Ca2+-independent manner. Unlike PECAM-1, PYK2 is localized at the cell-to–extracellular-matrix contact region.8,33 VEGF promotes the association of VEGF receptor-2 with integrin and integrin-dependent cell biological responses.31 Thus, the PYK2/Src complex transmits the VEGF signals in association with extracellular matrix/integrins, suggesting that the Ca2+-independent Src/PECAM-1 system is unlikely to be involved in PYK2-mediated Src activation.
Because the blood flow recovery ratio in the ischemic limbs was increased ≈5-fold on day 7 compared with the day 0 control level (Figure 5A), hemodynamic shear stress also could be proportionally elevated in the newly formed vessels. Shear stress was shown to elicit the phosphorylation of 1177Ser-eNOS by activating both Akt and PKA.28 As shown in Figure 2A, Akt, AMPK, and PKA showed peak phosphorylation on day 1 and at 2 hours, respectively, and then PKA and AMPK reversed to the baseline level on day 7, whereas moderate activation of Akt was observed on day 7 (210% increase compared with the basal level), suggesting that Akt, rather than PKA and AMPK, is involved in the eNOS activation 7 days after limb ischemia. However, further studies are required to define the involvement of other kinases associated with flow shear stress.
This analysis of PYK2−/− mice demonstrates the critical role of PYK2 in Akt/NO signals activated by vasoactive substances or ischemic stress that modulates the vascular tonus or angiogenesis. These findings indicate that PYK2 can operate as a modulator for extracellular versatile stimuli, leading to eNOS activation, and is closely involved in the receptor- or ischemia-activated NO signaling events and thus regulates the cytoskeleton structure, vasoreactive function, or angiogenic response.
The authors profoundly appreciate Dr Nobuo Shirahashi for his advice and help with the statistical analysis.
Source of Funding
This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (grants 13670763 and 15590778 to Dr Okigaki).
Shesely EG, Maeda N, Kim HS, Desai KM, Krege JH, Laubach VE, Sherman PA, Sessa WC, Smithies O. Elevated blood pressures in mice lacking endothelial nitric oxide synthase. Proc Natl Acad Sci U S A. 1996; 93: 13176–13181.
Fukumura D, Gohongi T, Kadambi A, Izumi Y, Ang J, Yun CO, Buerk DG, Huang PL, Jain RK. Predominant role of endothelial nitric oxide synthase in vascular endothelial growth factor-induced angiogenesis and vascular permeability. Proc Natl Acad Sci U S A. 2001; 98: 2604–2609.
Haynes MP, Li L, Sinha D, Russell KS, Hisamoto K, Baron R, Collinge M, Sessa WC, Bender Jr. Src kinase mediates phosphatidylinositol 3-kinase/Akt-dependent rapid endothelial nitric-oxide synthase activation by estrogen. J Biol Chem. 2003; 278: 2118–2123.
Yu H, Marchetto GS, Dy R, Hunter D, Calvo B, Dawson TL, Wilm M, Anderegg RJ, Graves LM, Earp HS. Activation of a novel calcium-dependent protein-tyrosine kinase: correlation with c-Jun N-terminal kinase but not mitogen-activated protein kinase activation. J Biol Chem. 1996; 271: 29993–29998.
Astier A, Avraham H, Manie SN, Groopman J, Canty T, Avraham S, Freedman AS. The related adhesion focal tyrosine kinase is tyrosine-phosphorylated after beta1-integrin stimulation in B cells and binds to p130cas. J Biol Chem. 1997; 272: 228–232.
Tian D, Litvak V, Lev S. Cerebral ischemia and seizures induce tyrosine phosphorylation of PYK2 in neurons and microglial cells. J Neurosci. 2000; 20: 6478–6487.
Okigaki M, Davis C, Falasca M, Harroch S, Felsenfeld D, Sheetz MP, Schlessinger J. Pyk2 regulates multiple signaling events crucial for macrophage morphology and migration. Proc Natl Acad Sci U S A. 2003; 100: 10740–10745.
Tamarat R, Silvestre JS, Kubis N, Benessiano J, Duriez M, deGasparo M, Henrion D, Levy BI. Endothelial nitric oxide synthase lies downstream from angiotensin II-induced angiogenesis in ischemic hindlimb. Hypertension. 2002; 39: 830–835.
Michell BJ, Chen Z, Tiganis T, Stapleton D, Katsis F, Power DA, Sim AT, Kemp BE. Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase. J Biol Chem. 2001; 276: 17625–17628.
Boo YC, Sorescu G, Boyd N, Shiojima I, Walsh K, Du J, Jo H. Shear stress stimulates phosphorylation of endothelial nitric-oxide synthase at Ser1179 by Akt-independent mechanisms: role of protein kinase A. J Biol Chem. 2002; 277: 3388–3396.
Yamamoto K, Sokabe T, Matsumoto T, Yoshimura K, Shibata M, Ohura N, Fukuda T, Sato T, Sekine K, Kato S, Isshiki M, Fujita T, Kobayashi M, Kawamura K, Masuda H, Kamiya A, Ando J. Impaired flow-dependent control of vascular tone and remodeling in P2X4-deficient mice. Nat Med. 2006; 12: 133–137.
Hernandez GL, Volpert OV, Iniguez MA, Lorenzo E, Martinez-Martinez S, Grau R, Fresno M, Redondo JM. Selective inhibition of vascular endothelial growth factor-mediated angiogenesis by cyclosporin A: roles of the nuclear factor of activated T cells and cyclooxygenase 2. J Exp Med. 2001; 193: 607–620.
Smolenski A, Poller W, Walter U, Lohmann SM. Regulation of human endothelial cell focal adhesion sites and migration by cGMP-dependent protein kinase I. J Biol Chem. 2000; 275: 25723–25732.
Surks HK, Mochizuki N, Kasai Y, Georgescu SP, Tang KM, Ito M, Lincoln TM, Mendelsohn ME. Regulation of myosin phosphatase by a specific interaction with cGMP- dependent protein kinase Ialpha. Science. 1999; 286: 1583–1587.
Ren XD, Kiosses WB, Schwartz MA. Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J. 1999; 18: 578–585.
Benard V, Bohl BP, Bokoch GM. Characterization of Rac and Cdc42 activation in chemoattractant-stimulated human neutrophils using a novel assay for active GTPases. J Biol Chem. 1999; 274: 13198–13204.
Tokiwa G, Dikic I, Lev S, Schlessinger J. Activation of Pyk2 by stress signals and coupling with JNK signaling pathway. Science. 1996; 273: 792–794.
Taniyama Y, Weber DS, Rocic P, Hilenski L, Akers ML, Park J, Hemmings BA, Alexander RW, Griendling KK. Pyk2- and Src-dependent tyrosine phosphorylation of PDK1 regulates focal adhesions. Mol Cell Biol. 2003; 23: 8019–8029.
Carroll DJ, Ramarao CS, Mehlmann LM, Roche S, Terasaki M, Jaffe LA. Calcium release at fertilization in starfish eggs is mediated by phospholipase Cgamma. J Cell Biol. 1997; 138: 1303–1311.
Fleming I, Busse R. Molecular mechanisms involved in the regulation of the endothelial nitric oxide synthase. Am J Physiol Regul Integr Comp Physiol. 2003; 284: R1–R12.
He H, Venema VJ, Gu X, Venema RC, Marrero MB, Caldwell RB. Vascular endothelial growth factor signals endothelial cell production of nitric oxide and prostacyclin through flk-1/KDR activation of c-Src. J Biol Chem. 1999; 274: 25130–25135.
Meyer RD, Dayanir V, Majnoun F, Rahimi N. The presence of a single tyrosine residue at the carboxyl domain of vascular endothelial growth factor receptor-2/FLK-1 regulates its autophosphorylation and activation of signaling molecules. J Biol Chem. 2002; 277: 27081–27087.
Soldi R, Mitola S, Strasly M, Defilippi P, Tarone G, Bussolino F. Role of alphavbeta3 integrin in the activation of vascular endothelial growth factor receptor-2. EMBO J. 1999; 18: 882–892.
Butler B, Blystone SD. Tyrosine phosphorylation of beta3 integrin provides a binding site for Pyk2. J Biol Chem. 2005; 280: 14556–14562.
Avraham HK, Lee TH, Koh Y, Kim TA, Jiang S, Sussman M, Samarel AM, Avraham S. Vascular endothelial growth factor regulates focal adhesion assembly in human brain microvascular endothelial cells through activation of the focal adhesion kinase and related adhesion focal tyrosine kinase. J Biol Chem. 2003; 278: 36661–36668.
Friedlander M, Brooks PC, Shaffer RW, Kincaid CM, Varner JA, Cheresh DA. Definition of two angiogenic pathways by distinct alpha v integrins. Science. 1995; 270: 1500–1502.
Hood JD, Frausto R, Kiosses WB, Schwartz MA, Cheresh DA. Differential alphav integrin-mediated Ras-ERK signaling during two pathways of angiogenesis. J Cell Biol. 2003; 162: 933–943.
Fleming I, Fisslthaler B, Dixit M, Busse R. Role of PECAM-1 in the shear-stress-induced activation of Akt and the endothelial nitric oxide synthase (eNOS) in endothelial cells. J Cell Sci. 2005; 118: 4103–4111.
Although endothelial dysfunction causes atherosclerosis or vascular aging, drugs to improve endothelial functions have not been developed. Nitric oxide (NO) plays pivotal roles in the maintenance of endothelial function and vascular homeostasis, including vasodilatation, antiinflammatory effect, anticoagulation, antiproliferative effect of vascular smooth muscle cells, angiogenesis, and vasculogenesis. NO is produced by endothelial NO synthase (eNOS); therefore, a drug that upregulates eNOS function may improve endothelial function. For this purpose, it is important to clarify the molecular mechanism to activate eNOS. In the present study, we showed that tyrosine kinase PYK2 plays a crucial role in eNOS activation. Both PYK2 and eNOS are activated by hemodynamic mechanical stress, ischemic stress, and stimulation with endothelial growth factors, and PYK2 transmits calcium and Akt signaling pathways, both of which activate eNOS, suggesting that the drug developed to activate PYK2 may be feasible therapy for maintaining vascular homeostasis in various stress conditions. However, PYK2 also is involved in angiotensin II–mediated signaling to induce atherosclerosis, including vasoconstriction, vascular inflammation, and proliferation of vascular smooth muscle cells, indicating that PYK2 has dual effects on vascular function. The role of PYK2 in the maintenance of vascular homeostasis should be investigated further.
The online Data Supplement, consisting of an expanded Methods section and a figure, can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.645416/DC1.