The ADMA/DDAH Pathway Regulates VEGF-Mediated Angiogenesis
Objectives— Asymmetrical dimethylarginine (ADMA) is a nitric oxide synthase (NOS) inhibitor and cardiovascular risk factor associated with angiogenic disorders. Enzymes metabolising ADMA, dimethylarginine dimethylaminohydrolases (DDAH) promote angiogenesis, but the mechanisms are not clear. We hypothesized that ADMA/DDAH modifies endothelial responses to vascular endothelial growth factor (VEGF) by affecting activity of Rho GTPases, regulators of actin polymerization, and focal adhesion dynamics.
Methods and Results— The effects of ADMA on VEGF-induced endothelial cell motility, focal adhesion turnover, and angiogenesis were studied in human umbilical vein endothelial cells (HUVECs) and DDAH I heterozygous knockout mice. ADMA inhibited VEGF-induced chemotaxis in vitro and angiogenesis in vitro and in vivo in an NO-dependent way. ADMA effects were prevented by overexpression of DDAH but were not associated with decreased proliferation, increased apoptosis, or changes in VEGFR-2 activity or expression. ADMA inhibited endothelial cell polarization, protrusion formation, and decreased focal adhesion dynamics, resulting from Rac1 inhibition after decrease in phosphorylation of vasodilator stimulated phosphoprotein (VASP). Constitutively active Rac1, and to a lesser extent dominant negative RhoA, abrogated ADMA effects in vitro and in vivo.
Conclusion— The ADMA/DDAH pathway regulates VEGF-induced angiogenesis in an NO- and Rac1-dependent manner.
High plasma levels of ADMA are associated with cardiovascular disorders including atherosclerosis, hypertension, homocysteinemia, diabetes mellitus, insulin resistance, and hypercholesterolemia.1,2 ADMA and the related methylarginine analogue monomethylarginine (L-NMMA) inhibit NO production by competing with L-arginine for binding to NOS.2 Levels of methylarginines are regulated by the enzymes DDAH I and II, which metabolize these factors.2 DDAH I heterozygote (HT) knockout mice have raised ADMA levels and a hypertensive phenotype as a consequence of reduced NO signaling.3 Further, dysregulation of the ADMA/DDAH pathway is associated with impaired angiogenesis, which contributes to disease pathology. In a mouse model of hypercholesterolemia, DDAH I activity is inhibited by lipid-induced oxidative stress, leading to elevated ADMA levels, reduced NO production, and impaired angiogenesis.4 ADMA metabolism through DDAH overexpression promotes angiogenesis in vitro,5,6 promotes sprouting from aortic rings,7,8 and enhances angiogenesis in tumors and after ischemia,6,9 but the mechanisms are not fully understood.
ADMA/DDAH pathway may influence angiogenesis by affecting expression of vascular endothelial growth factor (VEGF). DDAH overexpression enhances VEGF expression in C6 glioma cells6 and in human and murine endothelial cell lines.5 ADMA/DDAH may also influence VEGF responses by affecting NO bioavailability. NO signaling is critically involved in VEGF-mediated chemotaxis and angiogenesis.10,11 NO production is increased through a mechanism requiring VEGFR-2 activation, which may also involve upregulation of eNOS, increased cytosolic calcium, or phosphatidyl inositol 3 kinase (PI3K) activation.10,11 Changes in NO levels affect the activity of Rho GTPases, key regulators of actin dynamics.12 Coordinated activation of Rho GTPases Rac1, RhoA, and Cdc42 is required for cell motility.12 Rac1 and Cdc42 mediate focal contact/complex formation and polymerization of the actin cytoskeleton at the leading edge of the cell, whereas RhoA controls stability of focal adhesions, stress fiber formation in the main body of the cell, and retraction at the rear.12 ADMA activates RhoA and inhibits Rac1 and Cdc42 in pulmonary endothelial cells, leading to inhibition of spontaneous endothelial cell motility and vessel sprouting from aortic rings.7 Rho GTPases also play an essential role in regulating motility and angiogenesis in response to growth factors including VEGF, although the requirement for individual GTPases appears to depend on the experimental model of angiogenesis.13–16 RhoA is required for VEGF-induced angiogenesis in Matrigel plugs,13 whereas Rac1 and Cdc42 but not RhoA are important in phorbol ester– and growth factor–induced angiogenesis in collagen.14 VEGF-mediated motility requires Rac1 activity,15 but the role of RhoA is controversial.13,15,16
We hypothesized that ADMA metabolism alters VEGF signaling via its effects on Rho GTPases. We show for the first time that ADMA inhibits VEGF-induced endothelial chemotaxis and angiogenesis in vitro and in vivo. These effects result from reduced NO signaling and inhibition of pSer239VASP and Rac1, important in cell polarization and focal adhesion turnover.
Cell Culture and Treatments
Primary human umbilical vein endothelial cells (HUVECs, PromoCell, Heidelberg, Germany) were cultured in endothelial growth medium containing 2% fetal calf serum (FCS) and growth supplements (C-22010, PromoCell). For experiments, the cells were incubated in starvation media containing 1% FCS and no growth supplements, or where stated, in L-arginine-free media (PromoCell). ADMA (100 μmol/L, Calbiochem, Merck Biosciences) or NG-Nitro-l-arginine Methyl Ester (L-NAME, 1 mmol/L, Calbiochem) were added to the medium for 16 hours before measurement of cell responses. S-nitroso-N-acetylpenicillamine (SNAP, 10 μmol/L, Calbiochem) and human recombinant VEGF-165 (25 ng/mL, R&D Systems, Abingdon, UK) were added for the duration of experiments. cGMP in cell lysates was measured after treatment with VEGF for 5 minutes, using a commercial kit (Biotrak EIA system, RPN226, GE Healthcare, for details please see the supplemental materials, available online at http://atvb.ahajournals.org).
Adenovirus Construction and Transfection
Adenoviral gene transfer was used to overexpress DDAH I, RhoAN19, Rac1V12, or Cdc42V12 or GFP (transfection control) as described at a multiplicity of infection (MOI) of 200.7
Analysis of VEGF, VEGFR-2, and Cleaved Caspase-3
Soluble human VEGF secretion was quantified using an ELISA kit (QIA51, Calbiochem). VEGFR-2, phospho(Tyr1175)-VEGFR-2, and cleaved caspase-3 (Asp175) were measured using sandwich ELISA kits (Cell Signaling, New England Biolabs, UK; for details please see the supplemental materials).
RhoA, Rac1, and Cdc42 GTP-Binding
Active RhoA, Rac1, and Cdc42 was measured by affinity precipitation and detected by Western blotting as previously described.7
Cell Proliferation and Migration
Migration assays were performed in Transwell dishes, and the optical density of 6 random fields per filter (triplicate samples) was analyzed in Adobe Photoshop. Proliferation was evaluated in HUVECs cultured for 2 days in serum-reduced media (1% FCS) by colorimetric analysis using CellTiter 96 AQueous One Solution Cell Proliferation Assay (MTS; Promega). Both assays were validated by manual counting of cells (for details please see the supplemental materials).
Focal Adhesion Analysis
Focal adhesions were visualized in HUVECs by immunofluorescence using antivinculin antibodies. Alternatively, HUVECs were transfected with pDsRed2-N1-Zyxin and seeded in optical bottom 96-well plates (for details please see the supplemental materials). Images were recorded using confocal microscopy and processed using LaserPix software. Cell adhesions in the first frame (t=0) were labeled red, and in the last were labeled green (pseudocolours, t=16 minutes). Protrusions and cell elongation (circularity of a fitted ellipse) were analyzed using Image J software (National Institute of Health). Focal adhesion turnover was calculated as the proportion of remodelling adhesions (green and red) to constant adhesions (yellow).
Collagen I In Vitro Angiogenesis Assay
Cells were seeded onto 24-well plates (5×104 per well) precoated with PureCol Collagen I solution (Nutacon BV, 2 mg/mL, 300 μL/well). After 24 hours at 37°C, the cells were fixed (4% formaldehyde/PBS) and images from 4 random fields were analyzed. Total tubule length was calculated using Image J software.
Basement Membrane Plug Angiogenesis Assay
All experiments were carried out under a Home Office License and conducted according to the Animal Scientific Procedures Act 1986. Eight-week-old wild-type (WT) or DDAHI heterozygous (HT) mice3 were injected subcutaneously in both flanks with Cultrex Matrigel basement membrane extract (R&D Systems) and 500 ng/mL VEGF,17 and with adenoviruses where indicated (4×108 plaque forming units/ml Matrigel). After 7 days, mice (n=5) were euthanized and plugs recovered. Histological sections were stained with toluidine blue or antibodies against PECAM-1, with DAPI nuclear counterstain. Angiogenesis was quantified by measuring hemoglobin (Hb) content in plugs (for details please see the supplemental materials).18
Retrovirus Construction and Transfection of VASP Mutants
Retroviral constructs encoding human VASP mutants (S239A-VASP and S239D-VASP) were inserted into the retroviral vector LXSN. Retroviral stocks were generated by cotransfection with pVSVG (Clontech) and pGag-pol (Clontech) into HEK 293T cells and used to infect HUVECs. HUVECs stably expressing transgenes were selected with G418 (0.6 mg/mL, Sigma). Expression of vesicular stomatitis virus glycoprotein (VSVG)–tagged VASP were assessed by Western blotting and distribution of pSer239-VASP in HUVECs by immunostaining with mouse monoclonal antipSer239VASP antibody (1:100, Upstate Biotechnology; for details please see the supplemental materials).
Data are expressed as means±SEM. Differences among groups were assessed by ANOVA followed by posthoc tests. A probability value <0.05 was taken as statistically significant, n=3, unless stated otherwise.
ADMA Inhibits VEGF-Induced Chemotaxis and Capillary Network Formation In Vitro
HUVEC migration across Transwell filters in basal conditions (no VEGF) was 4.5±0.5 cells/200 μm2 in 4-hour assay and increased 2-fold on the addition of VEGF (P<0.001, Figure 1A and 1B). Exogenous ADMA (100 μmol/L) did not influence basal cell migration but reduced VEGF response to control levels (P<0.001, Figure 1A and 1⇓B). Lower dose of ADMA (10 μmol/L) had no effect on VEGF-mediated migration in basal culture medium, but had a significant inhibitory effect in arginine-free culture medium (supplemental Figure I). Overexpression of DDAH I restored the chemotactic response toward VEGF in the presence of ADMA (Figure 1A and 1⇓B).
ADMA Impairs VEGF-Induced Angiogenesis In Vitro and In Vivo
To study the effects of ADMA on early stages of VEGF-induced angiogenesis, we used a collagen I tube formation assay. Within 24 hours after plating, endothelial cells began to align and form a network of tubular structures. Whereas untreated cells showed little evidence of tube formation, VEGF induced formation of extensive cell networks, increasing tubulogenesis by 150%, compared to controls (P<0.001; Figure 2A and 2B). ADMA (100 μmol/L) had no effect in basal conditions but reduced VEGF effects by 40% (P<0.01, Figure 2A and 2⇓B).
To investigate the relevance of these observations in vivo, we analyzed blood vessel formation in Matrigel plugs coinjected with VEGF in DDAH I heterozygous (HT) knockout mice. These mice have raised ADMA levels and show abnormal cardiovascular responses as a result of decreased NO signaling.3 VEGF-mediated angiogenesis was severely impaired in DDAH I HT mice compared to their WT littermates (Figure 2C through 2E). The presence of blood vessels was confirmed by immunofluorescent staining of platelet endothelial cell adhesion molecule-1 (PECAM-1; supplemental Figure II). In the absence of VEGF, angiogenesis was minimal (Figure 2E). The analysis of hemoglobin levels confirmed that VEGF increased blood supply into the plug, and that this was significantly impaired in DDAH I HT mice compared to WT (83% decrease, P<0.05, Figure 2E and supplemental Figure II). When VEGF was coinjected with DDAH I adenoviruses into HT mice, angiogenesis was enhanced by 197% (P<0.05, HT compared to HT+DDAH I, Figure 2E). Sustained adenovirus-mediated overexpression of DDAH I was confirmed by Western blotting (supplemental Figure II).
ADMA Effects Are Mediated by Inhibition of NO Signaling and Are Independent of Proliferation and Apoptosis
Because VEGF-mediated motility and angiogenesis require NO, we hypothesized that ADMA effects on VEGF-induced angiogenesis may result from inhibition of NO signaling. L-NAME, a NOS inhibitor structurally unrelated to methylarginines, inhibited VEGF-mediated chemotaxis to the same extent as ADMA (P<0.001 compared to VEGF), whereas the NO donor SNAP significantly reduced the effects of ADMA, confirming previously reported role of NO in ADMA signaling (supplemental Figure III).7 Intracellular levels of cGMP, a downstream effector of NO, were reduced by ADMA in control and VEGF-treated group. DDAH did not significantly change cGMP levels in basal conditions but prevented ADMA effects in VEGF-treated group (supplemental Figure III). Changes in cGMP were consistent with previously reported changes in nitrite levels in culture media induced by ADMA/DDAH.7,19
Endothelial cell proliferation and apoptosis are important in the regulation of angiogenic responses. VEGF enhanced endothelial cell proliferation by 34% (P<0.001, compared to no treatment), but ADMA had no effect on VEGF-mediated proliferation within 48 hours of treatment (supplemental Figures III and IV). It has recently been reported that ADMA is proapoptotic.20 However, whereas VEGF inhibited cell apoptosis by 27% (P<0.05 compared to no treatment), ADMA had no significant effect on apoptosis regardless of the presence of serum, growth factor supplements, or VEGF (supplemental Figure III).
ADMA Prevents VEGF-Induced Protrusion Formation and Focal Adhesion Turnover and Reduces Cell Polarity
Inhibition of basal NO production by endothelial cells is associated with decreased motility and enlarged focal adhesions, which may result from imbalance in the assembly and disassembly of cell attachments.7,21 We hypothesized that ADMA would affect motility by altering focal adhesion turnover and distribution. Within 1 hour after seeding, VEGF promoted formation of lamellipodia and peripheral localization of the focal adhesion protein, vinculin (Figure 3A). ADMA reduced peripheral vinculin localization, whereas overexpression of DDAH I restored the VEGF-induced phenotype (Figure 3A). At later stages of cell spreading, ADMA-treated cells showed accumulation of enlarged central focal adhesions, the effect prevented by DDAH (Figure 3B). To study focal adhesion turnover and protrusion formation in live endothelial cells, the cells were transfected to express fluorescently-labeled zyxin, a marker of intermediate/late focal adhesions.22 Image analysis of time-lapse confocal images was used to quantify the numbers of remodeled focal adhesions and extension/retraction of the cell periphery (supplemental Figure IV). VEGF promoted protrusion formation and focal contact remodeling, and these effects were significantly inhibited by ADMA and rescued by overexpression of DDAH I (Figure 3C). Migratory endothelial cells exhibit an elongated polarized morphology. ADMA prevented cell elongation in both control and VEGF-treated groups (Figure 3D).
ADMA Alters Rho GTPase Activation Independently of VEGFR-2 Activation
DDAH II was reported to increase VEGF mRNA in endothelial cell lines.5 However, we did not observe any measurable changes in VEGF protein expression in HUVECs overexpressing DDAH I (data not shown). Because VEGFR-2–mediated signaling is involved in NO dependent endothelial cell responses to VEGF,10 we tested whether ADMA could affect VEGFR-2 expression or activation. However, neither activity nor expression of VEGFR-2 was affected by either ADMA or DDAH I (supplemental Figure V).
ADMA Impairs Focal Adhesion Remodeling and Angiogenesis via Rac1
The effects of ADMA on VEGF-induced cell motility are likely to be mediated by Rho GTPases, important regulators of actin dynamics. ADMA reduced the activity of Rac1 and Cdc42 and increased the activity of RhoA in both control and VEGF-treated cells, consistent with its effects on pulmonary endothelial cells under basal conditions (Figure 4A and 4B).7 To further investigate the role of Rho GTPases in the effects of ADMA, mutant Rho GTPases were overexpressed by adenoviral gene transfer. Constitutively active Rac1 (Rac1V12) prevented inhibition of focal adhesion remodelling by ADMA, restoring it to similar levels as VEGF, whereas constitutively active Cdc42 (Cdc42V12) or dominant negative RhoA (RhoAN19) had no effect (Figure 4C). To investigate how these observations might relate to angiogenesis in vivo, Matrigel plugs implanted in DDAHI HT mice were coinjected with VEGF and Rho GTPase mutants. Rac1V12 significantly enhanced angiogenesis in DDAH I HT mice, increasing hemoglobin (Hb) content in plugs from 32±26 to 559±97 (arbitrary units, P<0.01, Figure 4D), comparable to Hb levels in WT mice either without (513±138) or with Rac1V12 (554±110, Figure 4D). Overexpression of RhoAN19 in WT mice did not affect VEGF-induced angiogenesis (Figure 4D), whereas in DDAHI HT knockout mice, some noticeable though statistically not significant increase was observed (P=0.081, Figure 4D). Overexpression of Cdc42V12 did not influence VEGF-induced angiogenesis in either WT or DDAH I HT knockout mice (Figure 4D).
ADMA reduces Rac1 activity in pulmonary endothelial cells by decreasing NO/PKG-mediated phosphorylation of VASP on Ser239.19 Consistently, ADMA decreased VASP phosphorylation on Ser239 in HUVECs treated with VEGF (Figure 5A through 5D). The upper and lower band of the VASP doublet, which is detected by antipSer239 antibody is Ser157-phosphorylated and -nonphosphorylated protein, respectively (Figure 5C).23 Overexpression of mutant of VASP (S239A-VASP), in which the preferred PKG site is blocked from phosphorylation, inhibited tube formation to the same extent as ADMA (Figure 5E). Conversely, a phosphomimetic VASP mutant (S239D-VASP) increased tube formation in control cells and cells treated with VEGF and ADMA, whereas nonphosphorylatable VASP mutant (S239A-VASP) decreased tube formation, mimicking the effects of ADMA (Figure 5E). The small reduction in tube formation in VEGF-treated cells by S239D-VASP compared to untreated vector control may indicate that dynamic changes in VASP phosphorylation are important in this response. Expression of VSVG-VASP mutants was confirmed by Western blotting (supplemental Figure V). Overexpression of S239D-VASP increased Rac1 activity in HUVECs (supplemental Figure V). Rac1 can act as an upstream regulator of RhoA activity,24 and therefore we verified its role in ADMA-induced effects on RhoA. Overexpression of constitutively activated Rac1 (V12Rac1) inhibited RhoA activity in control (untreated) HUVECs but did not abolish RhoA activation by ADMA (supplemental Figure VI).
We show for the first time that a functional DDAH I pathway is required for the angiogenic response to VEGF by regulating Rho GTPase activity. Our in vitro studies indicate that ADMA, a substrate for DDAH, impairs endothelial cell responses to VEGF by reducing NO dependent chemotaxis and focal adhesion turnover, and abrogating protrusion formation, cell polarization, and alignment into capillary networks. ADMA does not affect VEGFR-2 activation or expression but inhibits VEGF-mediated phosphorylation of VASP and activation of Rac1. Focal adhesion turnover and angiogenesis in vivo are restored by increasing Rac1 activity, indicating a key role for this GTPase as a target in ADMA-induced angiogenic pathologies.
Angiogenic morphogenesis requires tightly controlled regulation of apoptotic and proliferative pathways. In our studies, ADMA did not modulate the effects of VEGF on cell proliferation or apoptosis, in contrast to other reports showing antiproliferative and proapoptotic effects.20,25 Although these discrepancies may result from the use of different cell types and culture conditions, it is unlikely that proapoptotic effects of ADMA play a role in our system; we observed significant short-term effects of ADMA on chemotaxis in vitro, and in vivo substantial numbers of endothelial cells accumulated in Matrigel plugs in DDAH I HT mice in spite of high tissue and plasma levels of ADMA. DDAH I overexpression also did not induce secretion of soluble VEGF, nor did it affect VEGFR-2 activation or expression. These differences are likely to be isoform-specific as DDAH II but not DDAH I overexpression induces VEGF mRNA and secretion in endothelial cells.5,26
The dose of ADMA used in our in vitro studies was relatively high (100 μmol/L), whereas lower doses were only effective in media deprived of NOS substrate L-arginine, confirming the involvement of NO signaling. In vivo, heterozygous DDAH I gene deletion in mice increases plasma concentration of ADMA 0.5 to 0.7 μmol/L similar to the levels reported in patients with multiple cardiovascular risk factors.3 We found that these mice showed decreased angiogenic responses to VEGF. It is likely that local changes in ADMA concentrations in cells and tissues are higher than estimated. Plasma levels of ADMA vary from 0.2 to 1 μmol/L but tissue levels may reach 15 μmol/L.27 Another likely mechanism to explain the effect of low concentrations of ADMA in vivo may involve endothelial progenitor cells (EPCs). EPCs are important in neoangiogenesis and their mobilization, differentiation, and function are inhibited by plasma concentrations of ADMA.25
Under basal conditions, ADMA has been shown to activate RhoA and inhibit Rac1 in an NO/protein kinase G (PKG)–dependent manner,7,19 and we observed similar effects in the presence of VEGF. Inhibition of Rac1 by ADMA appears to be particularly deleterious to VEGF-induced angiogenic responses. This may reflect a more critical role played by Rac1 in angiogenesis. Endothelial-specific deletion of Rac1 is lethal at embryonic day 9.5 because of aberrant angiogenesis, and Rac1-deficient endothelial cells fail to form lamellipodia, migrate, and undergo angiogenic morphogenesis in response to VEGF.28 Further, and reminiscent of the DDAH I HT phenotype,3 endothelial-specific downregulation of Rac1 results in mild hypertension and impaired angiogenesis as a result of decreased NO signaling.29 In support of the role of NO/Rac1 in ADMA effects, VEGF-mediated chemotaxis was restored by NO donors, and overexpression of constitutively activated Rac1 restored focal adhesion remodeling and angiogenesis in DDAH I HT knockout mice. The mechanism by which Rac1 regulates angiogenesis is likely to involve control of both cell–cell and cell–matrix interactions. Cell–matrix interaction induces dynamic cellular protrusions followed by the assembly of endothelial cell aggregates and rudimentary cords, requiring Rac1-dependent integration of both actin remodeling and microtubule dynamics.30 These dynamic protrusions are thought to be critical in supporting proper cell–cell contact and alignment in capillary morphogenesis.30 In line with a Rac1-dependent mechanism, therefore, we observed that ADMA abrogated lamellipodia formation, cell polarization, and capillary morphogenesis in vitro, whereas in vivo, endothelial cells were able to invade Matrigel plugs in DDAH I HT mice but failed to align and coalesce into capillary vessels.
Although Rac1 appears more important in control of VEGF-induced angiogenesis, a role of RhoA cannot be completely excluded. RhoA inhibition increased remodeling of focal adhesions in basal conditions, possibly as an effect of decreased adhesion to the substratum7 and increased blood supply in Matrigel plugs in DDAH I HT knockout mice. RhoA may contribute to inhibition of VEGF-mediated angiogenesis by regulating mechanical remodeling of extracellular matrix (ECM) and matrix invasion.13 Rac1 reduced RhoA activity in basal conditions but did not abolish RhoA activation by ADMA, which suggests that RhoA activity is controlled by two different mechanisms. Rac1 is thought to inhibit RhoA by activation of p190 Rho GTPase activating protein,31 whereas NO/PKG inhibit RhoA by increasing phosphorylation on Ser188, which enhances its interaction with Rho guanine-dissociation inhibitor.32
VEGFR-2 mediates the effects of VEGF on endothelial motility.11 Interestingly, although ADMA did not prevent VEGFR-2 autophosphorylation, it may affect downstream signaling pathways in a spatially-dependent manner by preventing VEGFR-2 localization to lamellipodia.33,34 Rac1 regulates lamellipodia formation,12,35,36 and a decrease in its activity may prevent proper localization of activated VEGFR-2. In support of this, we observed reduced localization of VEGFR-2 in lamellipodia in HUVECs treated with ADMA after monolayer wounding (supplemental Figure VII).
ADMA inhibits Rac1 activity in pulmonary endothelial cells by decreasing NO/cGMP-dependent phosphorylation of vasodilator activated phosphoprotein (VASP) on Ser239.19 VASP regulates actin dynamics, and its localization and activity is associated with Rac1 activation.23,37 In control and VEGF-treated HUVECs, ADMA decreased VASP Ser239 phosphorylation, whereas nonphosphorylatable mutant of VASP mimicked the effects of ADMA on endothelial tube formation in vitro, supporting a role upstream of Rac1 in ADMA-induced effects on angiogenesis. In further support of our data, VASP phosphorylation by PKG has recently been shown to enhance angiogenesis38 and increase vascular smooth muscle cell matrix invasion (Thomas Renné, unpublished observation 2009). Full characterization of the effects of Rac1/RhoA activity imbalance induced by ADMA on angiogenesis will require further studies.
In summary, our observations indicate that the DDAH I/ADMA pathway plays a critical role in regulating VEGF-mediated endothelial cell motility and angiogenesis and that altered NO signaling and Rho GTPases activity are central to angiogenic regulation by DDAH I/ADMA.
We thank Dr Anna Huttenlocher (University of Wisconsin-Madison, Madison, Wisconsin) for the gift of pDsRed2-N1-Zyxin and Dr Paul Frankel (Rayne Institute UCL) for the gift of pVSVG and pGag-pol.
Sources of Funding
This work was supported by British Heart Foundation grant PG/07/004.
L.R.F. and T.B. contributed equally to this study.
Received January 16, 2009; revision accepted September 11, 2009.
Miyazaki H, Matsuoka H, Cooke JP, Usui M, Ueda S, Okuda S, Imaizumi T. Endogenous nitric oxide synthase inhibitor: a novel marker of atherosclerosis. Circulation. 1999; 99: 1141–1146.
Vallance P, Leiper J. Cardiovascular biology of the asymmetric dimethylarginine: dimethylarginine dimethylaminohydrolase pathway. Arterioscler Thromb Vasc Biol. 2004; 24: 1023–1030.
Leiper J, Nandi M, Torondel B, Murray-Rust J, Malaki M, O'Hara B, Rossiter S, Anthony S, Madhani M, Selwood D, Smith C, Wojciak-Stothard B, Rudiger A, Stidwill R, McDonald NQ, Vallance P. Disruption of methylarginine metabolism impairs vascular homeostasis. Nat Med. 2007; 13: 198–120.
Wojciak-Stothard B, Torondel B, Tsang LY, Fleming I, Fisslthaler B, Leiper JM, Vallance P. The ADMA/DDAH pathway is a critical regulator of endothelial cell motility. J Cell Sci. 2007; 120: 929–942.
Jacobi J, Sydow K, von Degenfeld G, Zhang Y, Dayoub H, Wang B, Patterson AJ, Kimoto M, Blau HM, Cooke JP. Overexpression of dimethylarginine dimethylaminohydrolase reduces tissue asymmetric dimethylarginine levels and enhances angiogenesis. Circulation. 2005; 111: 1431–1438.
Ridley AJ. Rho GTPases and cell migration. J Cell Sci. 2001; 114: 2713–2722.
Hoang MV, Whelan MC, Senger DR. Rho activity critically and selectively regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci U S A. 2004; 101: 1874–1879.
Bayless KJ, Davis GE. The Cdc42 and Rac1 GTPases are required for capillary lumen formation in three-dimensional extracellular matrices. J Cell Sci. 2002; 115: 1123–1136.
van Nieuw Amerongen GP, Koolwijk P, Versteilen A, van Hinsbergh VWM. Involvement of RhoA/Rho Kinase Signaling in VEGF-Induced Endothelial Cell Migration and Angiogenesis In Vitro. Arterioscler Thromb Vasc Biol. 2003; 23: 211–217.
Wojciak-Stothard B, Torondel B, Zhao L, Renne T, Leiper JM. Modulation of Rac1 activity by ADMA/DDAH regulates pulmonary endothelial barrier function. Mol Biol Cell. 2009; 20: 33–42.
Zaidel-Bar R, Ballestrem C, Kam Z, Geiger B. Early molecular events in the assembly of matrix adhesions at the leading edge of migrating cells. J Cell Sci. 2003; 116: 4605–4613.
Thum T, Tsikas D, Stein S, Schultheiss M, Eigenthaler M, Anker SD, Poole-Wilson PA, Ertl G, Bauersachs J. Suppression of endothelial progenitor cells in human coronary artery disease by the endogenous nitric oxide synthase inhibitor asymmetric dimethylarginine. J Am Coll Cardiol. 2005; 46: 1693–1701.
Hasegawa K, Wakino S, Tanaka T, Kimoto M, Tatematsu S, Kanda T, Yoshioka K, Homma K, Sugano N, Kurabayashi M, Saruta T, Hayashi K. Dimethylarginine dimethylaminohydrolase 2 increases vascular endothelial growth factor expression through Sp1 transcription factor in endothelial cells. Arterioscler Thromb Vasc Biol. 2006; 26: 1488–1494.
Cardounel AJ, Xia Y, Zweier JL. Endogenous methylarginines modulate superoxide as well as nitric oxide generation from neuronal nitric-oxide synthase: differences in the effects of monomethyl- and dimethylarginines in the presence and absence of tetrahydrobiopterin. J Biol Chem. 2005; 280: 7540–7549.
Tan W, Palmby TR, Gavard J, Amornphimoltham P, Zheng Y, Gutkind JS. An essential role for Rac1 in endothelial cell function and vascular development. FASEB J. 2008; 22: 1829–1838.
Sawada N, Salomone S, Kim H-H, Kwiatkowski DJ, Liao JK. Regulation of endothelial nitric oxide synthase and postnatal angiogenesis by Rac1. Circ Res. 2008; 103: 360–368.
Connolly JO, Simpson N, Hewlett L, Hall A. Rac regulates endothelial morphogenesis and capillary assembly. Mol Biol Cell. 2002; 13: 2474–2485.
Ellerbroek SM, Wennerberg K, Burridge K. Serine phosphorylation negatively regulates RhoA in vivo. J Biol Chem. 2003; 278: 19023–19031.
Ikeda S, Ushio-Fukai M, Zuo L, Tojo T, Dikalov S, Patrushev NA, Alexander RW. Novel role of ARF6 in vascular endothelial growth factor-induced signaling and angiogenesis. Circ Res. 2005; 96: 467–475.
Labrecque L, Royal I, Surprenant DS, Patterson C, Gingras D, Béliveau R. Regulation of vascular endothelial growth factor receptor-2 activity by caveolin-1 and plasma membrane cholesterol. Mol Biol Cell. 2003; 14: 334–347.
Kraynov VS, Chamberlain C, Bokoch GM, Schwartz MA, Slabaugh S, Hahn KM. Localized Rac activation dynamics visualized in living cells. Science. 2000; 290: 333–337.
Schlegel N, Burger S, Golenhofen N, Walter U, Drenckhahn D, Waschke J. The role of VASP in regulation of cAMP- and Rac 1-mediated endothelial barrier stabilization. Am J Physiol Cell Physiol. 2008; 294: C178–C188.
Chen H, Levine YC, Golan DE, Michel T, Lin AJ. Atrial natriuretic peptide-initiated cGMP pathways regulate vasodilator-stimulated phosphoprotein phosphorylation and angiogenesis in vascular endothelium. J Biol Chem. 2008; 283: 4439–4447.