Akt/Protein Kinase B and Endothelial Nitric Oxide Synthase Mediate Muscular Neovascularization Induced by Tissue Kallikrein Gene Transfer
Background— Angiogenesis gene therapy with human tissue kallikrein (hTK) has shown promise for ischemic disease. The present study was undertaken to (1) assess an optimal gene transfer modality, (2) clarify hTK angiogenic pathways, and (3) discount possible side effects.
Methods and Results— The hTK gene was transferred to murine adductors by increasing doses of an adenovirus (Ad.hTK). Heterologous protein production was evaluated by ELISA and immunohistochemistry. Structural and functional characteristics of hTK-induced neovascularization were assessed. Muscular endothelial nitric oxide synthase (eNOS) and vascular endothelial growth factor (VEGF)-A mRNA and protein content were evaluated by real-time polymerase chain reaction and Western blotting. The ability of hTK to phosphorylate-activate Akt/protein kinase B (Akt-B) and VEGF receptor 2 (VEGF-R2) was also determined. Implication of the aforementioned mechanisms in Ad.hTK-induced neovascularization was challenged by blocking Akt-B with a dominant-negative Akt construct; NOS with NG-nitro-l-arginine methyl ester; and VEGF-A with neutralizing antibody, VEGF-R2 antagonist, or Ad carrying soluble VEGF-R1 gene. We found that 107 PFU Ad.hTK led to peak increases in capillary and arteriole density. Newly developed arterioles persisted for up to 8 weeks. Ad.hTK did not change microvascular permeability. Ad.hTK upregulated eNOS mRNA and protein and activated Akt-B through Ser-473 phosphorylation. Inhibitory studies documented that these biochemical events were instrumental to Ad.hTK-induced neovascularization. In contrast, Ad.hTK neither affected VEGF-A and VEGF-R2 levels nor increased VEGF-R2 phosphorylation. Consistently, Ad.hTK-induced neovascularization was not disturbed by any of the different approaches used to block VEGF-A.
Conclusions— Our findings provide new information on the pathway involved in hTK-induced neoangiogenesis and represent an advancement toward clinical applications with Ad.hTK.
Received July 14, 2003; de novo received December 29, 2003; revision received April 22, 2004; accepted April 26, 2004.
Neovascularization gene therapy has been proposed as a rescue for ischemic disease.1,2 After successful application in animal models, the strategy has been transferred from the bench to the bedside. However, clinical results have not matched the level of efficacy for which researchers had hoped.3–7 The serious side effects derived from the use of robust infecting doses of viral vectors and unregulated expression of angiogenic genes7,8 point to the necessity that clinical application must be preceded by in-depth clarification of the involved pathways and improvements in the therapeutic-risk index. Furthermore, simple initiators of angiogenesis may not represent the best option to construct a durable and well-tempered neovascularization. In this area of high medical need, substances capable of successfully completing the angiogenic process appear more suitable to the task.
The serine protease human tissue kallikrein (hTK), a recent entry in the angiogenic factor family, has shown promise for the treatment of ischemic disease. Kallikrein-kinin system components are upregulated in animal models of and in patients with peripheral vascular disease,9,10 and local hTK gene delivery enhances native neovascularization and accelerates tissue healing.9 It is noteworthy that hTK does not require an ischemic environment for its curative properties.11
At present, information about molecular effector(s) of adenovirus (Ad) hTK-induced neovascularization remains largely incomplete, being mainly limited to the basic notion that kinins generated by hTK play a fundamental role in these vascular effects.9 In addition, from in vitro studies we know that bradykinin (BK) activates endothelial NO synthase (eNOS) through either the phosphoinositide 3-kinase–Akt/protein kinase B (Akt-B) pathway or calcineurin-mediated mechanisms.12 Furthermore, vascular endothelial growth factor (VEGF)-receptor 2 (R2) trans-activation by the BK-B2 receptor seemingly induces endothelial cell (EC) tube formation on Matrigel through eNOS mediation.13 In general, VEGF-A and Akt-B cooperate in the control of endothelial growth and viability,14 but recent evidence supports the possibility that Akt-B may be activated by alternative pathways independent of VEGF-A.15 Therefore, we considered it worthwhile to clarify whether hTK stimulates angiogenesis in vivo through Akt-B, eNOS, and VEGF-A. In view of future clinical use, we also evaluated an optimal gene transfer modality, the composition and persistence of neovascularization, and the microvascular permeability of Ad.hTK-infected muscles.
hTK Gene Delivery and Histological Assessment of Muscular Neovascularization
Procedures complied with the standards stated in the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Academy of Sciences, Bethesda, Md, 1996).
A dose-response curve (103 to 109 plaque-forming units [PFU]) to Ad.hTK was established in CD1 mice (Charles River, Calco, Italy). Anesthetized (2,2,2-tribromoethanol, 880 mmol/kg IP, Sigma) mice received 3 injections (3.5 μL/injection, 10.5 μL total volume) at different sites of the left adductor muscle along the projection of the femoral artery. This procedure enabled Ad vector diffusion along the entire mouse adductor. Ad carrying the luciferase gene (Ad.Luc) and saline served as controls. Each group consisted of at least 6 mice.
Two or 8 weeks after gene transfer, limb muscles were perfusion-fixed and processed for histological analyses of capillary and arteriole density, as described.11 Capillary density was expressed as capillary number per area (ncap/mm2) of transverse section or normalized to myofiber density (ncap/nfiber). Arteriole density was expressed as arteriole number per area (nart/mm2).11
The effect of Ad.hTK on EC proliferation in vivo was evaluated. To this aim, adductors (n=5 per group) that had received Ad.hTK or Ad.Luc (each at 108 PFU) 3 days in advance were processed for immunohistochemical identification of proliferating cell nuclear antigen (PCNA) by using a monoclonal antibody (Dako). Total capillaries and PCNA-positive ECs were counted in at least 25 fields at ×1000 magnification, and the number of PCNA-positive ECs per square millimeter and for every 1000 capillaries was calculated.
Transgene Expression After Ad.hTK Gene Transfer
We established the relation between Ad infecting dose and transgenic protein content. An ELISA (AngioProgen) selectively recognizing hTK16 was used to measure hTK content in plasma and adductors (n=4 per dose) at 5 days from gene delivery, ie, at the peak of transgene expression.16 Saline-injected muscles served as controls. We also measured hTK levels in the eyes (a tissue at risk for pathological angiogenesis) of mice receiving intramuscular Ad.hTK (109 PFU) or saline. Total protein concentration was measured by the Lowry method.
The cellular localization of transgenic protein in infected muscles was determined by immunohistochemistry with a polyclonal rabbit antibody that recognizes exclusively the hTK form.17 Specific binding was detected with the biotin-streptavidin LSAB+ system (Dako). Controls included preabsorption of anti-hTK with purified human urinary TK (Calbiochem) and replacement of anti-hTK antibody by nonimmune serum at the same dilution. Muscles injected with Ad.Luc and sterile saline served as negative controls.
Plasma protein extravasation (PPE) provoked by angiogenesis therapy may cause local edema.18 Because kinins may cause similar reactions,19,20 we tested whether this undesired effect occurs after hTK application. Muscular PPE at 3 or 14 days after 109 PFU of Ad.hTK or Ad.Luc administration or saline was determined by the Evans blue method20 and normalized to PPE in the contralateral, noninjected adductors. BK (10 μg/10 μL IM, given 5 minutes before Evans blue dye) served as positive control. Each group consisted of 6 mice.
Molecular Pathways Activated by Ad.hTK In Vivo
Western blot analysis of Akt-B was performed on muscles harvested 3 days after 108 PFU Ad.hTK or Ad.Luc (n=6 each group) had been administered. Twenty-six micrograms of protein for each sample was loaded. The reaction was carried out using primary antibodies raised against Ser-473–phosphorylated or total Akt (Cell Signaling Technology). Tubulin served as the loading control. Specific protein was detected by chemiluminescence reaction (Amersham), followed by analysis of immunoblot density by dedicated software (Scion Corp).
eNOS Expression: Real-Time PCR Primers
Real-time quantitative polymerase chain reaction (PCR; ABI PRISM 7000 sequence detection system software version 1.0, Perkin-Elmer) was used to determine eNOS mRNA content in muscles (n=6 per group) harvested 3 days after the injection of 108 PFU Ad.hTK or Ad.Luc. Total RNA was isolated with TRIzol reagent (Invitrogen) and treated with DNAse (Qiagen). RNA was reverse-transcribed with M-MLV reverse transcriptase (Invitrogen). eNOS and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) primer sequences were designed on GenBank database NM-008713 and NM-008084, respectively, and were as follows: 5′-CCTTCCGCTACCAGCCAGA-3′ (eNOS forward), 5′-CAGAGATCTTCACTGCATTGGCTA-3′ (eNOS reverse), 5′-CGT GGG GCT GCC CAG AAC AT-3′ (GAPDH forward), and 5′-TCT CCA GGC GGC ACG TCA GA-3′ (GAPDH reverse). Conventional PCR products of murine eNOS (105 bp) and GAPDH (156 bp) were obtained with the primers designed for the real-time PCR and were cloned into pGEM-T Easy vector (Promega) to be used as DNA standards. eNOS cDNA level was normalized to GAPDH cDNA level.
Muscular eNOS protein content at 5 days from gene transfer was analyzed by Western blotting (n=6 muscles per group) with a rabbit polyclonal antibody (1:1000, Santa Cruz). Blots were reprobed with anti–β-actin antibody (1:1000, Santa Cruz). Immunoblot density was analyzed by Image Gauge V3.41 (Fuji Film).
VEGF-A and VEGF-R2
VEGF-A mRNA content was determined on the same samples used for the eNOS experiment. VEGF-A primers (designed on GenBank M95200) generate a 111-bp fragment and were as follows: 5′-CCA GCG AAG CTA CTG CCG TCC A-3′ (forward) and 5′-ACA GCG CAT CAG CGG CAC AC-3′ (reverse).
Western blots for VEGF-A (rabbit polyclonal antibody at 1:2000, Santa Cruz), VEGF-R2 (goat polyclonal antibody at 1:2500, R&D Systems), and Tyr-951–phosphorylated VEGF-R2 (rabbit polyclonal antibody at 1:1000, Cell Signaling) were performed on the same samples (n=6 per group) used for eNOS analysis. After phospho-VEGF-R2 analysis, the membrane was stripped and reprobed with antibody for total VEGF-R2. In all cases, β-actin was used for normalization.
Functional Role of Akt-B, NOS, and VEGF-A in Ad.hTK-Induced Angiogenesis
Ad.hTK and Ad.DN–Akt-B Cotransfection
To elucidate the functional role of Akt-B in Ad.hTK-induced neovascularization, a cotransfection experiment with Ad.dominant negative-Akt308/547 (Ad.DN–Akt-B)21 was performed. Adductors received 109 PFU of Ad.hTK or Ad.Luc in combination with either Ad.DN–Akt-B or Ad.Luc (each at 5×106 PFU). Neovascularization was evaluated at 2 weeks in 6 mice per group.
Capillary and arteriole density was counted at 2 weeks from Ad.hTK or Ad.Luc administration (each at 109 PFU) in combination with the NOS inhibitor NG-nitro-l-arginine methyl ester (L-NAME, 1.4 mmol/kg body weight daily in drinking water) or the inactive enantiomer D-NAME. Each group consisted of 6 mice.
Inhibition of VEGF-A Action
The role of VEGF-A was addressed by 3 approaches: (1) A VEGF-A neutralizing antibody (2.5 μg IP twice a week, R&D Systems)22 or nonimmune IgG was given in combination with Ad.hTK or Ad.Luc (109 PFU IM). (2) The VEGF-R2 antagonist PTK 787 (kindly provided by Dr J. Wood, Novartis Pharma AG, Basel, Switzerland), that was previously shown to block VEGF-A–induced angiogenesis,23 was given in drinking water (25 mg/kg body weight per day for 15 days) starting 1 day before Ad.hTK or Ad.Luc (109 PFU IM). Control mice drank regular water. (3) An Ad carrying soluble VEGF-R1 gene (Ad.s-flt-1, 109 PFU, kindly provided by Drs S.A. Karumanchi, Beth Israel Deaconess Hospital and Harvard Medical School, Boston, Mass, and R. Mulligan, Harvard Medical School and Children’s Hospital, Boston, Mass) was cotransfected with Ad.hTK or Ad.Luc (each at 108 PFU). Soluble VEGF-R1 is able to entrap several VEGFs, including VEGF-A. Therefore, it inhibits the biological effects of VEGF-A.
The capacity of VEGF-A antibody, PTK 787, or Ad.s-flt-1 to block VEGF-A–induced neovascularization was confirmed by using them or their respective controls (nonimmune IgG, normal drinking water, or Ad.Luc) in mice whose muscles were infected with Ad.VEGF-A (107 PFU). Mice (n=6 per group) were humanely killed at 14 days from gene transfer for evaluation of neovascularization.
Akt-B Involvement in BK-Induced EC Proliferation In Vitro
We evaluated whether Akt-B blockade by wortmannin inhibits the BK-induced proliferation of human umbilical vein ECs (HUVECs). Proliferation was evaluated by MTS assay (Promega). HUVECs were seeded on 96-well plates. After 4 hours, the medium was changed to medium 199 containing reduced fetal bovine serum (2%) plus wortmannin (10 nmol/L) or vehicle (phosphate-buffered saline). To avoid BK degradation, the angiotensin-converting enzyme inhibitor captopril (10−5 mol/L) was added 2 hours before BK (10−12 to 10−8 mol/L) or its vehicle. Proliferation was assessed after 48 hours. Each experiment was repeated 8 times.
Results are expressed as mean±SEM. Multivariate repeated-measures ANOVA was performed to test for interaction between time and grouping factor. In multiple comparisons in which ANOVA indicated significant differences, the statistical value was determined according to the Bonferroni method. Differences within and between groups were determined by a paired or unpaired Student t test, respectively. A probability value <0.05 was interpreted to denote statistical significance.
hTK Protein Expression
As shown in Figure 1A, muscular hTK protein increased dose-dependently after Ad.hTK administration. hTK was not detected in plasma, contralateral noninjected muscles, or eyes. Skeletal myocytes expressed transgenic protein in variable amounts that in some cells appeared as small dots, whereas in others it was associated with the cell membrane (Figure 1B, i and ii). No immunoreactivity was observed in muscles injected with saline or Ad.Luc or when anti-hTK antibody was preabsorbed with the purified human enzyme (Figure 1B, iii and iv) or replaced by nonimmune serum (not shown).
Ad.hTK Titration and Microvascular Effects
At 14 days from gene transfer, Ad.hTK increased capillary and arteriole density in a dose-related manner (see Figure 2A and 2B). The plateau was reached with 107 PFU, indicating that moderate infecting doses are sufficient to elicit maximal biological effect. Results were confirmed after normalization by myofiber density (data not shown).
Previously, we reported that Ad.hTK-promoted capillarization is stable for up to 4 weeks, a time by which transgene expression has expired.9,16 Here, we examined the destiny of newly formed vessels over a longer period. At 8 weeks, capillary density of Ad.hTK-injected tissue had returned to the levels of Ad.Luc- or saline-injected muscles (Figure 2C). In contrast, arteriole density remained elevated (Figure 2D), with the increase encompassing vessels of any size of luminal diameter (data not shown).
As expected, Ad.hTK increased EC proliferation (9±3 versus 2±2 PCNA-positive ECs/mm2 in Ad.Luc and 12±3 versus 5±1 PCNA-positive ECs in every 1000 capillaries at 3 days after injection; P<0.05 for both comparisons). Data Supplement Figure I (available online only at http://www.circulationaha.org) shows representative images of PCNA-stained sections from muscles injected with Ad.hTK (B) or Ad.Luc (A).
Ad.hTK Gene Transfer Does Not Alter Microvascular Permeability
Ad.hTK did not increase PPE (data not shown), whereas the positive control BK increased PPE by 4.96-fold (P<0.01).
Ad.hTK Stimulates Akt-B Phosphorylation
As shown in Figure 3, Ad.hTK augmented Akt-B phosphorylation in Ser-473–treated preparations (P<0.01 versus Ad.Luc treated or untreated). Ad.hTK did not change the ratio between total Akt-B and tubulin (data not shown).
Ad.hTK Upregulates eNOS Expression
Ad.hTK increased eNOS mRNA by 2.84 times (P<0.05 versus Ad.Luc, Figure 4A). As shown in Figure 4B, Ad.hTK also augmented eNOS protein content (0.15±0.02 versus 0.09±0.01 eNOS/β-actin ratio in Ad.Luc, P<0.05).
Ad.hTK Does Not Affect VEGF-A Expression
Ad.hTK did not change VEGF-A expression at the mRNA (11±1 versus 9±2 [VEGF-A/GAPDH] × 104 in Ad.Luc, P=NS) or protein (0.17±0.02 versus 0.16±0.02 VEGF-A/β−actin in Ad.Luc, P=NS) level.
Ad.hTK Does Not Change VEGF-R2 Expression or Phosphorylation State
Ad.hTK did not change VEGF-R2 content (0.11±0.01 versus 0.13±0.04 VEGF-R2/β-actin in Ad.Luc, P=NS) or phosphorylation state (4.00±1.51 versus 5.63±0.71 phosphorylated/total VEGF-R2 in Ad.Luc, P=NS).
Functional Implication of Akt-B and eNOS in hTK-Induced Neovascularization
Ad.hTK-induced neovascularization involves both Akt-B and eNOS. In fact, hTK-induced angiogenesis and arteriogenesis were significantly inhibited by either Ad.DN-Akt-B (Figure 5A and 5B) or L-NAME (Figure 6A and 6B).
hTK-Induced Angiogenesis Is Independent of VEGF-A
As shown in Data Supplement Figure II (available online at http://www.circulationaha.org), Ad.hTK-induced neovascularization was not affected by VEGF-A neutralizing antibody (A and B), PTK 787 (C and D), or Ad.s-flt-1 (E and F). By contrast, each of the 3 compounds blocked angiogenesis and arteriogenesis caused by Ad.VEGF-A (data not shown). These results suggest that the angiogenic action of hTK is not mediated by either VEGF-A–dependent or -independent activation of VEGF-R1 or VEGF-R2.
Akt-B Mediates BK-Induced HUVEC Proliferation In Vitro
The Akt-B inhibitor wortmannin reduced BK-induced HUVEC proliferation (data not shown).
Our study documents that a low infecting dose (106 PFU) of Ad.hTK was sufficient to promote capillary and arteriole growth in limb muscle. With 107 PFU, microvascular density was further augmented. No additional effect occurred with escalating doses despite incrementally increased immunoreactive hTK. Our findings introduce the novel concept that low-dosage gene transfer could be effectively applied in vascular medicine with obvious reduction of side effects. This advantageous property could be conferred by the presence of a secretory signal in the hTK encoding sequence, allowing release of recombinant protein from infected cells into the extracellular environment.24 Thus, a few transduced cells are enough to release an optimal amount of hTK for the desired biological effect to be achieved. Moreover, the enzymatic nature of hTK allows a reduction in infection levels, inasmuch as single molecule of enzyme generates angiogenic kinin effectors continuously. It should be cautiously noted that gene transfer efficiency could be reduced in aged or diseased animals.25 However, low-dosage Ad.hTK showed the capacity of preventing muscular microangiopathy in type I diabetic mice.11
Ad.hTK encodes for a pre-pro form of tissue kallikrein. Activation to the mature enzyme was demonstrated in a previous study, where we found increased kinin levels in Ad.hTK-infected, normoperfused mouse adductors.16 Moreover, the ELISA used in the present study is able to recognize only active kallikrein.
A novel finding of the present study consists of the demonstration that hTK generates a stable and well-tempered vascularization. Muscularization of nascent capillaries and stimulation of growth and remodeling of preexisting collateral vessels may account for arteriogenesis promoted by hTK. Support for the latter possibility comes from clinical studies showing that hTK is involved in adaptive growth of collateral circulation.10 This result could have clinical implications, inasmuch as arterioles are principally responsible for providing tissue relief after the occurrence of vascular occlusion.
The present study provides novel insight into the mechanisms mediating Ad.hTK-induced neovascularization. Kinins activate the phosphoinositide 3-kinase–Akt-B pathway in cultured ECs,12 and here we report for the first time that Akt-B blockade inhibits the proliferative effect of BK on ECs. Moreover, our in vivo studies documented that Akt-B and eNOS are functionally involved in the neovascularization pathway of Ad.hTK, whereas VEGF-A or its receptors play no role. Ad.hTK-produced kinins share important features with VEGF-A, because they induce vasodilation, angiogenesis, and NO production. However, we found that Ad.hTK did not affect VEGF-A expression. Most important, VEGF-A blockade failed to inhibit Ad.hTK-induced neovascularization. Therefore, we conclude that VEGF-A does not participate in Ad.hTK-induced microvessel growth. Furthermore, relevant to a clinical perspective, Ad.hTK does not cause the increase in microvascular permeability observed with VEGF-based gene tranfer.18,26 Safety of Ad.hTK gene therapy is additionally supported by the observation that transgene expression was confined to the injection site, thereby not endangering distant organs, including the retina.
In conclusion, the present discoveries unravel novel mechanisms responsible for Ad.hTK-induced neovascularization and represent further progress toward its clinical application. The discovery that hTK induces a VEGF-A–independent activation of the Akt-B–eNOS pathway may have relevance for therapeutic strategies alternative to or combined with VEGF-A.
This study was supported by the Telethon-Onlus Foundation (grant No. GP0300Y01); the European Foundation for the Study of Diabetes (EASD)–Servier; and the Italian MIUR. INBB laboratories and INSERM U541 are partners of the European Community EGVN Network of Excellence and the contributions of Dr Silvestre and Michel Clerque were undertaken in the context of EVGN. The contribution of Dr Figueroa was partly realized during his visit at INBB, supported by a grant from the Sardinia Government to Dr Madeddu. Dr Linthout is a recipient of a European Community–founded Marie Curie Post-Doctorial Fellowship. The help of Dr Michael Latronico (ISS, Rome, Italy) in performing Western blots for Akt is acknowledged.
The online-only Data Supplement, which contains additional figures, is available with this article at http://www.circulationaha.org.
Leiden JM. Human gene therapy: the good, the bad, and the hugly. Circ Res. 2000; 86: 923–925.
Lee R, Springer M, Blanco W, et al. VEGF gene delivery to myocardium: deleterious effects of unregulated expression. Circulation. 2000; 102: 898–901.
Emanueli C, Minasi A, Zacheo A, et al. Local delivery of human tissue kallikrein gene accelerates spontaneous angiogenesis in mouse model of hindlimb ischemia. Circulation. 2001; 103: 125–132.
Porcu P, Emanueli C, Kapatsoris M, et al. Reversal of angiogenic growth factors upregulation by revascularization of lower limb ischemia. Circulation. 2002; 105: 67–72.
Emanueli C, Salis MB, Pinna A, et al. Prevention of diabetes-induced microangiopathy by human tissue kallikrein gene transfer. Circulation. 2002; 106: 993–999.
Harris MB, Ju H, Venema VJ, et al. Reciprocal phosphorylation and regulation of endothelial nitric-oxide synthase in response to bradykinin stimulation. J Biol Chem. 2001; 276: 16587–16591.
Miura S, Matsuo M, Saku S. Transactivation of KDR/Flk-1 by the B2 receptor induces tube formation in human coronary endothelial cells. Hypertension. 2003; 41: 1118–1123.
Jiang BH, Zheng JZ, Aoki M, et al. Phosphatidylinositol 3-kinase signaling mediates angiogenesis and expression of vascular endothelial growth factor in endothelial cells. Proc Natl Acad Sci U S A. 2000; 97: 1749–1753.
Skaletz-Rorowski A, Lutchman M, Kureishi Y, et al. HMG-CoA reductase inhibitors promote cholesterol-dependent Akt/PKB translocation to membrane domains in endothelial cells. Cardiovasc Res. 2003; 57: 253–264.
Emanueli C, Zacheo A, Minasi A, et al. Adenovirus-mediated human tissue kallikrein gene delivery induces angiogenesis in normoperfused skeletal muscle. Arterioscler Thromb Vasc Biol. 2000; 20: 2379–2385.
Gowdak LH, Poliakova L, Wang X, et al. Adenovirus-mediated VEGF (121) gene transfer stimulates angiogenesis in normoperfused skeletal muscles and preserves tissue perfusion after induction of ischemia. Circulation. 2000; 102: 565–571.
Emanueli C, Grady EG, Madeddu P, et al. Acute ACE inhibition causes plasma extravasation in mice that is mediated by bradykinin and substance P. Hypertension. 1998; 31: 1299–1304.
Condorelli G, Drusco A, Stassi G, et al. Akt induces enhanced myocardial contractility and cell size in vivo in transgenic mice. Proc Natl Acad Sci U S A. 2002; 99: 12333–12338.
Tamarat R, Silvestre JS, Kubis N, et al. Endothelial nitric oxide synthase lies downstream from angiotensin II–induced angiogenesis in ischemic hindlimb. Hypertension. 2002; 39: 830–835.
Bold G, Frei J. Furet P, et al. CGP 79787D (PTK787/ZK222584), CGP 84738, NVP-AAC789, NVP-AAD777 and related 1-anilino-(4-pyridylmethyl)phthalazines as inhibitors of VEGF- and bFGF-induced angiogenesis. Drugs Future. 2002; 27: 43–55.
Communal C, Huq F, Lebeche D, et al. Decreased efficiency of adenovirus-mediated gene transfer in aging cardiomyocytes. Circulation. 2003; 107: 1170–1175.