Adenovirus E4 gene promotes selective endothelial cell survival and angiogenesis via activation of the vascular endothelial-cadherin/Akt signaling pathway.

The early 4 region (E4) of the adenoviral vectors (AdE4(+)) prolongs human endothelial cell (EC) survival and alters the angiogenic response, although the mechanisms for the EC-specific, AdE4(+)-mediated effects remain unknown. We hypothesized that AdE4(+) modulates EC survival through activation of the vascular endothelial (VE)-cadherin/Akt pathway. Here, we showed that AdE4(+), but not AdE4(-) vectors, selectively stimulated phosphorylation of both Akt at Ser(473) and Src kinase in ECs. The phosphatidylinositol 3-kinase (PI3K) inhibitors LY294002 and wortmannin abrogated AdE4(+) induction of both phospho-Akt expression and prolonged EC survival. Regulation of phospho-Akt was found to be under the control of various factors, namely VE-cadherin activation, Src kinase, tyrosine kinase, extracellular signal-regulated kinase (ERK), and c-Jun N-terminal kinase (JNK). Downstream targets of Akt signaling resulted in glycogen synthase kinase-3alpha/beta phosphorylation, beta-catenin up-regulation, and caspase-3 suppression, all of which led to AdE4(+)-mediated EC survival. Furthermore, infection with AdE4(+) vectors increased the angiogenic potential of ECs by promoting EC migration and capillary tube formation in Matrigel plugs. This selective AdE4(+)-mediated enhanced motility of ECs was also blocked by PI3K inhibitors. Taken together, these results suggest that activation of the VE-cadherin/Akt pathway is critical for AdE4(+)-mediated survival of ECs and angiogenic potential.

Adenoviral (Ad) 1 vectors are efficient vectors for gene delivery in vitro and in vivo, especially for local gene delivery to the endothelial cells (EC) within the vascular system. ECs are important target cells for gene therapy because they are readily accessible to Ad vectors via the circulation and play a critical role in the progression of physiological and pathophysiological processes, including wound healing, apoptosis, inflammation, tissue revascularization, and tumor angiogenesis. We and other groups demonstrated previously that infection of ECs with first-generation AdE4 ϩ vectors, but not AdE4 Ϫ vectors, selectively modulates the angiogenic potential of ECs by altering apoptosis and the inflammatory status of ECs (1)(2)(3)(4). Remarkably, infection with AdE4 ϩ vectors resulted in the generation of a unique "suspended animation" state where the cells remain viable without requiring stimulation with serum or growth factors for several months. However, the mechanisms by which AdE4 ϩ promotes long-term survival of ECs have yet to be clarified.
The phosphatidylinositol 3-kinase (PI3K)/Akt signal cascade is involved in the regulation of apoptosis, survival, and proliferation of a wide variety of cell types (5), including ECs (6). PI3K activates the Akt serine/threonine kinase by generating specific inositol phospholipids that recruit Akt to the cell membrane and enable its activation. PI3K/Akt signaling is activated by various growth factors, including VEGF-A, FGF-2, and platelet-derived growth factor (7,8) as well as the vascular endothelial cadherin (VE-cadherin)/␤-catenin complex (9). PI3K is activated by binding to phosphorylated receptors, and the resulting phosphoinositides recruit the kinases Akt and 3-phosphoinositide-dependent protein kinase-1 via pleckstrin homology domains (10).
VE-cadherin is a member of the cadherin superfamily that is specifically expressed at EC junctions (11). Its extracellular adhesive domain interacts via its cytoplasmic tail with the armadillo family proteins ␤-catenin, plakoglobin, and ␣-catenin (12), which couple the cadherin-catenin complex to the actin cytoskeleton. This complex is known to be involved in controlling endothelial survival and motility.
We hypothesized that the Akt signaling pathway, through activation of the VE-cadherin, plays an important role in AdE4 ϩ promotion of EC survival and that the VE-cadherin/Akt signaling pathway can regulate the AdE4 ϩ pro-survival effects on EC. Here, we demonstrate that AdE4 ϩ significantly protected ECs against apoptosis through the recruitment of the VE-cadherin/Akt-signaling cascade. Moreover, this AdE4 ϩ effect is associated with alterations of EC migration and vessellike tube formation, suggesting that AdE4 ϩ -mediated activa-tion of this pathway also regulates the angiogenic potential of ECs.

MATERIALS AND METHODS
Cell Culture-Human umbilical vein endothelial cells (HUVEC) and human umbilical smooth muscle cells were isolated as described (13) and cultured in EC medium (M199 medium containing 10% (v/v) fetal bovine serum, 20 g/ml EC growth factor, 20 units/ml heparin, 100 g/ml penicillin, and 100 g/ml streptomycin) in a humidified incubator at 37°C with 95% air/5% CO 2 . HUVEC monolayers between passages 2 to 4 were used in these studies. A549 (human lung carcinoma cells), HL60 (leukemic cell lines), HeLa (human cervix carcinoma cells), and primary human foreskin fibroblast cells (HFF) were purchased from the American Type Culture Collection (ATCC, Manassas, VA); human pulmonary microvascular endothelial cells, human liver endothelial cells, and human vein endothelial cells were obtained from Clonetics. Cell viability was assayed by the trypan blue exclusion method, which indicated that Ͻ5% of the cells took up the dye both before and after the infection of Ad vectors.
Terminal Deoxynucleotidyltransferase-mediated dUTP Nick-end Labeling (TUNEL) Assay-The TUNEL assay was performed by using the in situ cell detection kit (fluorescein isothiocyanate) following the manufacturer's instructions (Roche Molecular Biochemicals). In brief, cells grown on gelatin-pretreated glass coverslips were fixed by a freshly prepared 4% paraformaldehyde solution (in PBS, pH 7.4) for 1 h at room temperature. Coverslips were then washed with PBS and incubated in permeabilization solution (0.1% Triton X-100 and 0.1% sodium citrate) for 2 min on ice. Then, 50 l of TUNEL reaction mixture was added on coverslips and incubated in a humidified chamber for 1 h at 37°C in the dark. Finally, cells were mounted and analyzed by fluorescence microscopy.
Determination of the Caspase-3 Activity-The activity of caspase-3 in EC lysates (2 ϫ 10 5 cells/sample) was determined using the caspase-3 colorimetric assay kit according to the manufacturer's instructions (R&D Systems). 150 g of protein were used per reaction and incubated for 2 h at 37°C. The colorimetric reaction products were determined at 505 nm.
Migration Assays-Migration assays were performed by a modification of the procedure described previously using a Boyden chamber (17). HUVECs were infected with or without Ad vectors overnight in serumfree media (X-vivo medium). Transduced cells were suspended with 1% collagenase and washed with phosphate-buffered saline. Cells were resuspended in serum-free medium and put into the coated fibronectin chamber (3-m pores size; BD Biosciences). VEGF-A (10 ng/ml) and FGF-2 (10 ng/ml) in 5% fetal bovine serum X-vivo medium were put into the lower chamber. The chamber was incubated for 22 h at 37°C. The filter was carefully removed, and cells attached on the upper side were wiped off. HUVECs migrating through the filter and appearing on the lower side were fixed by careful immersion of the filter into 70% ethanol FIG. 3. Protective effect of AdE4 ؉ infection in preventing apoptosis and reducing caspase-3 activity in ECs. A, HUVEC were exposed to the AdNull (100 m.o.i.) or PBS (control) in growth factor-free medium for 3 days. Apoptosis was revealed by fluorescent phase-contrast microscopic using the TUNEL method. Positive staining for apoptotic cells was detected in ECs cultured in growth factor-free medium. No staining was observed in ECs infected with AdE4 ϩ . B, the level of caspase-3 activity was measured by caspase-3 assay. The cells were infected with AdE4 Ϫ , AdE4 ϩ , or PBS (control) for 48 h. The cell lysates were then tested for caspase-3 activity. The data points represent the mean Ϯ S.D. of three separate experiments, **, p Ͻ 0.01 compared with control; ##, p Ͻ 0.01 compared with AdE4 Ϫ . C, AdE4 ϩ prevented the A23187-induced apoptosis in HUVECs. Cells were infected with or without the 100 m.o.i. of AdE4 ϩ for 2 h and then treated with or without A23187 (100 nM) or vehicle (control) in EC medium containing serum and EC growth factors. Following treatment for the indicated time points, the number of ECs was quantified by the trypan blue exclusion method.

FIG. 4. AdE4 ؉ induced activation of the PI3K/Akt pathway in ECs.
A, effects of PI3K inhibitors on AdE4 ϩ -modulated EC viability. The viability of confluent ECs cultured in growth factor-free medium was measured over time after infection with AdE4 ϩ at 100 m.o.i. with or without 500 nM wortmannin (WT) or 10 M LY-294002 (LY). B, AdE4 ϩ induced phospho-Akt and phospho-Ser in ECs. ECs were infected with AdE4 ϩ , AdE4 Ϫ , or PBS (control) for the indicated time points. The cell lysates were then analyzed by immunoblot using polyclonal anti-phospho-Src antibody (pSrc), polyclonal anti-phospho-Ser-473 Akt/PKB antibody (pAkt), and total Akt antibody (Akt). C, non-EC cells lines, including the A549 cell line or the HFF cell line, were infected by AdE4 ϩ , AdE4 Ϫ , or PBS (control) for the indicated time. The cell lysates were then analyzed by immunoblot using the polyclonal anti-phospho-Ser-473 Akt/PKB antibody (pAkt). for 15 min followed by crystal violet staining at 25°C for 1 h, and the HUVECs were counted in 10 random fields per side of chamber. Each experiment was repeated three times to achieve statistical significance.
Tube Formation Assay-The formation of vascular-like structures by HUVECs on Matrigel (Becton Dickinson) was semi-quantified by phasecontrast microscopy. Twenty-four-well culture plates were coated with Matrigel according to the manufacturer's instructions. HUVECs were infected with or without adenovirus and then seeded on coated plates at 5 ϫ 10 4 cell/well in serum-free medium and incubated at 37°C for 8 h. HUVECs were seeded on coated plate in serum-free medium containing VEGF-A, 10 ng/ml, and FGF-2, 10 ng/ml, as positive control.

AdE4 ϩ Vectors Selectively Prolong Survival of ECs by Pre-
venting Apoptosis-Infection of HUVEC with E4 ϩ , but not E4 Ϫ , adenoviral vectors supports survival of these cells in serumand growth factor-free culture conditions. Remarkably, HUVEC monolayers infected with AdE4 ϩ vectors show no decrease in cell number for at least 10 days (Fig. 1, A and B). In contrast, HUVEC monolayers infected with either control or E4-deficient Ad (AdE4 Ϫ ) vectors, including AdE4 Ϫ LacZ and AdE4 ORF 6, do not survive in serum-or growth factor-free conditions by day 10. The infection of HUVEC with adenovirus vectors is highly efficient, and virtually all of the HUVEC express ␤-gal after infection with AdE4 ϩ LacZ or AdE4 Ϫ LacZ (Fig. 1C).
Second generation Ad vectors containing E4 with deletions of E1, E3, and, additionally, E2b genes are associated with decreased risk of adenovirus-derived gene expression and, thus, are less likely to induce immune responses or to be cleared via cell-mediated immune responses (15,18). HUVECs infected with a second generation AdE4 ϩ vector survive at rates similar to those treated with first generation AdE4 ϩ (Fig. 2A).
Microvascular ECs from each organ have unique physiological attributes. Therefore, we investigated the effects of AdE4 ϩ infection on survival in various organ-specific ECs, namely human pulmonary endothelial cells, human liver endothelial cells, and human vein endothelial cells. AdE4 ϩ infection of these various ECs also resulted in survival effects similar to those observed in HUVECs under serum-and growth factorfree conditions (Fig. 2B).
To assess whether AdE4 ϩ pro-survival effect is restricted to human ECs, we evaluated the effect of AdE4 ϩ in non-ECs, including A549 (lung carcinoma), HL60 (leukemic cell lines), HeLa (cervical carcinoma), HFF cells, and human umbilical smooth muscle cells. Remarkably, in contrast to ECs, there were no alterations in the apoptotic or survival state of these other non-EC cell types after infection with AdE4 ϩ vectors (data not shown) as assessed by TUNEL assay or cell number. These findings indicate that the pro-survival effects of AdE4 ϩ are primarily specific for EC type.
To examine the effect of AdE4 ϩ on the apoptosis of ECs, both TUNEL assay and caspase-3 activity were employed to evaluate the extent to which the AdE4 ϩ gene suppressed EC apoptosis. After 48 h of serum and growth factor starvation, HUVEC monolayers began to apoptose, as evidenced by the TUNEL assay (Fig. 3A). In contrast, AdE4 ϩ -treated HUVECs remained robust and showed no indication of apoptosis. Caspase-3 activation was also measured as an indicator of apoptosis induction, because different upstream pathways that lead to apoptosis depend on caspase-3 induction for final apoptotic execution. Fig. 3B shows that AdE4 ϩ , but not AdE4 Ϫ , markedly suppressed caspase-3 activity in ECs, even under serum-and growth factor-free conditions. In addition, EC monolayers cultured in the presence of serum and growth factors were also treated with calcium ionophore A23187, which directly increases the concentration of intracellular Ca 2ϩ and induces cell apoptosis. Remarkably, AdE4 ϩ effectively blocked Quantitation of pAkt level after normalization to total Akt from Western analyses using a densitometer is also presented. the onset of apoptosis induced by A23187 (Fig. 3C). These data suggest that AdE4 ϩ infection can also protect ECs from Ca 2ϩ influx-induced cell apoptosis.
AdE4 ϩ Inhibition of EC Apoptosis Is Mediated via Activation of the PI3K/Akt Pathway-Because the protein kinase Akt is an important survival factor that suppresses EC apoptosis (7), we examined whether AdE4 ϩ regulates the PI3K/Akt signaling cascade. Treatment of EC cultures with two selective and structurally unrelated PI3K inhibitors, LY294002 (10 M) or wortmannin (500 nM), abolished the ability of AdE4 ϩ to rescue EC from apoptosis following growth factor withdrawal (Fig. 4A). LY294002 and wortmannin also reverted the inhibition of caspase-3 activation by AdE4 ϩ (data not shown). Because Akt activation is a critical downstream effector of PI3K, we initially tested whether AdE4 ϩ infection of EC is associated with Akt phosphorylation. Akt phosphorylation was assessed by immu-noblot analysis in EC using a phospho-specific anti-Akt antibody. Immunoblot analysis of EC infected with AdE4 ϩ vector in serum-and growth factor-free conditions revealed significant up-regulation of the expression of Akt phosphorylated at Ser 473 at a maximum of 48 h by AdE4 ϩ , but not AdE4 Ϫ infection, and remained constant until at least 72 h (Fig. 4B). Total Akt levels remained unchanged, and Akt phosphorylated at Thr 308 expression was not detectable (data not shown). In addition, AdE4 ϩ induced phosphorylation of Src kinase, which is an upstream activator of PI3K/Akt (Fig. 4B). However, AdE4 ϩ did not significantly alter the expression of phospho-Akt in non-EC lines, including the A549 cell line or primary HFF cells (Fig.  4C). These data suggest that AdE4 ϩ promotes survival specifically in EC via the Akt phosphorylation pathway.
AdE4 ϩ Directly Induces Akt Phosphorylation via Tyrosine Kinase, ERK, and JNK Pathways-To determine whether AdE4 ϩ activates signals leading to phosphorylation of Akt or inhibits the dephosphorylation of Akt, HUVECs were infected with AdE4 ϩ in the presence of PI3K inhibitors LY294002, wortmannin, or Src kinase inhibitor PP2. As shown in Fig. 5A, AdE4 ϩ -mediated induction of phospho-Akt expression was blocked by LY294002 or wortmannin. The Src kinase inhibitor PP2, but not the inactive tyrosine kinase inhibitor PP3 (data not shown), also partly suppressed AdE4 ϩ -mediated induction of phospho-Akt expression (Fig. 5A); however, PP2 did not significantly alter the AdE4 ϩ -prolonged EC survival (Fig. 5C). These data suggest that the AdE4 ϩ effect is mediated by phosphorylation of Akt in ECs.
Both receptor tyrosine kinase and nonreceptor tyrosine kinase are involved upstream of PI3K/Akt (19,20). To dissect the pathway mediating phospho-Akt activation, we examined the effects of genistein, a widely used inhibitor of tyrosine kinase activity, on AdE4 ϩ -promoted EC survival. Fig. 5C shows that survival of EC by AdE4 ϩ was significantly inhibited by treatment with 50 M genistein. Also, pretreatment with genistein abolished AdE4 ϩ -induced phoshpo-Akt expression (Fig. 5B).
As MAPKs have been implicated in the regulation of phosphorylation of Akt, we also assessed the role of MAPKs in the regulation of phospho-Akt in EC after AdE4 ϩ infection. As shown in Fig. 5D, AdE4 ϩ -induced activation of phospho-Akt was inhibited by pretreatment with PD98059 (a selective inhibitor of ERK) or SP600125 (a selective inhibitor of JNK), but not by SB203580 (a selective inhibitor of p38-MAPK). These data indicate that ERK and JNK are the upstream activators of PI3K in the signaling pathways activated by AdE4 ϩ .
Selective Activation of EC-specific VE-cadherin by AdE4 ϩ Vectors-VE-cadherin and the intracellular ␤-catenin binding region of VE-cadherin have been shown to play an important role in Akt activation and EC survival (9). To assess the effect of AdE4 ϩ on VE-cadherin and ␤-catenin levels, EC lysates from control and AdE4 Ϫ and AdE4 ϩ cell cultures were analyzed by Western blot using specific anti-VE-cadherin and anti-␤-catenin antibodies. As shown in Fig. 6A, VE-cadherin and ␤-catenin protein levels were increased after infection with AdE4 ϩ . Because the anti-VE-cadherin antibody can directly inhibit VEcadherin function in ECs (21), we then used neutralizing monoclonal antibody (BV9) to VE-cadherin in cell cultures infected with AdE4 ϩ . BV9 significantly suppressed AdE4 ϩ -promoted EC survival (Fig. 6B) and blocked AdE4 ϩ -induced phospho-Akt in ECs (Fig. 6C). These data suggest that VE-cadherin and ␤-catenin participate as activators of the PI3K signaling pathway in AdE4 ϩ -mediated survival of ECs. As VE-cadherin is only expressed on the ECs but not on the other cell types, this may explain why AdE4 ϩ -mediated induction of EC survival is restricted to ECs. It is conceivable that AdE4 ϩ gene products specifically recruit VE-cadherin.
FIG. 6. VE-cadherin and ␤-catenin play an important role in AdE4 ؉ -induced EC survival. A, AdE4 ϩ increased VE-cadherin and ␤-catenin protein expression in HUVECs. HUVECs were infected with AdE4 ϩ , AdE4 Ϫ , or PBS (control) for 48 h, and cell extracts were analyzed by Western blotting with antibodies against VE-cadherin, ␤-catenin, and ␤-actin. B, effect of neutralizing monoclonal antibody to VEcadherin on AdE4 ϩ -promoted EC survival. Phase-contrast micrographs of representative monolayers of HUVECs that were infected with AdE4 ϩ with or without VE-cadherin antibody (BV9 10 g/ml) for 48 h are shown. C, the effect of neutralizing monoclonal antibodies to VEcadherin on AdE4 ϩ -induced, phosphorylated Akt (pAkt). HUVECs were infected with AdE4 ϩ with or without VE-cadherin Ab (BV9) for 48 h, and cell extracts were analyzed by Western blotting with antibodies against pAkt and Akt.
AdE4 ϩ Increased GSK3 ␣/␤ Phosphorylation and Decreased Caspase-3 Activation-GSK3␣/␤ and caspase-3 are downstream targets of Akt signaling; both phosphorylation of GSK3␣/␤ and suppression of activation of caspase-3 block the pathways leading to EC apoptosis. As shown by Western blot analysis in Fig. 7, AdE4 ϩ , but not AdE4 Ϫ , stimulated GSK3␣/␤ phosphorylation in a time-dependent manner with maximal expression occurring within 48 h and sustained phosphorylation lasting for at least 4 days. At the same time, activation of caspase-3 was suppressed by AdE4 ϩ , suggesting that phosphorylation of GSK3␣/␤ and reduction of caspase-3 activity are involved in the AdE4 ϩ survival effect.
AdE4 ϩ Promotes EC Migration and Tube Formation in ECs-Migration and tube formation are important angiogenic functions of ECs essential for the assembly of functional neo-vessels. To examine whether AdE4 ϩ also affects the angiogenic potential of EC, the migration of AdE4 ϩ -infected ECs was assessed in a modified Boyden chamber. AdE4 ϩ , but not AdE4 Ϫ , increased EC migration in response to VEGF-A by 3-fold, whereas AdE4 Ϫ infection resulted in a statistically insignificant rate of migration (Fig. 8). This response was blocked by prior administration of the PI3K inhibitor LY294002 (10 M).
A Matrigel tube formation assay was also employed to test the effect of AdE4 ϩ on the angiogenic potential of ECs. Infection of AdE4 ϩ vectors, but not AdE4 Ϫ vectors, induced the assembly of typical sprouting and tube-like structures that are reminiscent of vessels typically formed in vivo (Fig. 8B). This angiogenic effect is similar to that of VEGF-A treatment, which induces EC tube formation in Matrigel. These results indicate that AdE4 ϩ -infected ECs maintain their angiogenic potential in the absence of growth factors. Therefore, the overall effect of AdE4 ϩ gene products is not only to increase survival but also to maintain the pro-angiogenic properties of ECs. DISCUSSION Dissecting the mechanism by which E4 ϩ adenoviral vectors modulate angiogenesis is important to diminish the toxicities associated with adenoviral gene therapy. The importance of VE-cadherin/Akt signaling in the selective modulation of EC survival (9) offers a new perspective for discerning the mechanism whereby E4 gene products specifically support EC survival. In the present study, our results provide further evidence that AdE4 ϩ vectors selectively protect ECs from apoptosis and maintain their angiogenic potential through recruitment of the VE-cadherin/Akt signaling pathway.
Our data show that protein tyrosine kinases, such as Src family kinases, are involved in AdE4 ϩ -mediated PI3K/Akt activation and EC survival. AdE4 ϩ activation of Src or other receptor tyrosine kinase ligands is the key mediator of activation of Akt. Src kinase has been implicated in the control of cell division, the production of autocrine growth factors, and the cell's survival response (22). The inhibitory effect of the Src inhibitor PP2 and tyrosine kinase inhibitor genistein suggests that the Src family or some other tyrosine kinase is a crucial upstream activator of the PI3K/Akt pathway that is activated by the AdE4 ϩ vectors.
FIG. 8. Effect of AdE4 ؉ in VEGF-stimulated migration and tube formation of HUVEC. A, migration assay. AdE4 ϩ enhanced EC migration in Boyden chamber assay. ECs were plated on the upper chambers in 250 l of serum-free medium and then treated with control, AdE4 Ϫ , AdE4 ϩ , and AdE4 ϩ plus LY294002 (10 M). VEGF-A (VEGF165) was immediately added to the lower chambers. After 22 h at 37°C, the cells on the surface of chambers were wiped off, and cells that migrated to the bottom surface were stained with crystal violet and quantified. Results are shown as the mean Ϯ S.D. of cell numbers for 10 random fields in triplicate experiments. *, p Ͻ 0.01 compared with control. B, AdE4 ϩ -induced tube formation. ECs were infected with AdE4 Ϫ , AdE4 ϩ , or PBS and plated on Matrigel-coated culture plates for 8 h and analyzed for typical neo-angiogenic tube formation by phasecontrast microscopy.
ited by GSK-3␣/␤ activation in an in vivo Matrigel plug assay, whereas the inhibition of GSK-3␣/␤ signaling enhances capillary formation (26). In addition, it has been shown that FGF, insulin, and epidermal growth factor can inhibit GSK-3␣/␤ activation through increased Ser 9 phosphorylation to promote cell survival (27). Enhancement in Akt and GSK-3␣/␤ phosphorylation could be due to increase in glucose uptake, glycolysis, and glucose transporter expression, which have been shown to inhibit cytochrome c release and prevent apoptosis (28). We showed that the infection of ECs with AdE4 ϩ vectors resulted in GSK-3␣/␤ phosphorylation, which may contribute to AdE4 ϩinduced prolongation of EC survival.
The ability of EC to migrate in response to VEGF-A was significantly increased by the infection of AdE4 ϩ vectors in the Matrigel and migration assays. Inhibition of PI3K by LY294002 suppressed this effect, suggesting that PI3K is central to the angiogenic pathway. Previous studies have shown that the anti-VE-cadherin antibody is able to block tube formation in a fibrin gel (29,30). Taken together, these results indicate that the VE-cadherin/Akt pathway is a critical intracellular signaling step for angiogenesis in ECs. Remarkably, the effect of the AdE4 ϩ vector on protective apoptosis and maintaining migration in ECs is similar to that of classic angiogenic factors, including FGF-2, VEGF-A, or platelet-derived growth factor.
Inhibition of VE-cadherin did not completely block the prosurvival effect of AdE4 ϩ vectors, suggesting that VE-cadherinindependent signaling pathways may also be involved in AdE4 ϩ pro-survival effects. Alternatively, up-regulation of FGF-2, which activates additional survival pathways, may contribute to AdE4 ϩ -mediated survival of ECs. Both VEGF-A and FGF-2 have been shown to differentially activate Raf, resulting in protection from distinct pathways of EC apoptosis (31). In agreement with this finding, we showed that neutralizing monoclonal antibodies to VE-cadherin did not suppress FGF-2-mediated survival of ECs.
First generation E1 Ϫ E3 Ϫ Ad vectors are not completely replication defective, whereas the E2, E3, and E4 promoters may be still active and allow viral replication and late gene expression to proceed at high titers of Ad vectors (32). The resulting replication-competent adenoviral vector (RCA) may affect EC survival. However, new Ad vectors with deletions of the E1, E3, and E2b genes decrease the risk for RCA and late gene expression and have undetectable levels of RCA (15,18,33). Our results demonstrated that a recombinant adenovirus vector with E1, E2a, and E3 deletions has a similar effect with the first generation Ad vector to promote EC survival ( Fig. 2A), which further indicated that AdE4 ϩ gene products contribute to EC survival.
In summary, the present study demonstrated that AdE4 ϩ , but not AdE4 Ϫ , induced VE-cadherin/Akt activation in ECs. Phospho-Akt activation by AdE4 ϩ is mediated by VE-cadherin, tyrosine kinase, and the MAPK of ERK and JNK. Our results suggest that the AdE4 ϩ -mediated VE-cadherin/Akt pathway ultimately effects EC survival and angiogenesis. E4 region of Ad vectors contains at least six ORFs that confer a variety of regulatory functions. The precise identity of E4 ORF proteins that induce activation of Akt/VE-cadherin is not known and is the subject of ongoing studies.