Neovasculogenic effect of 11,12-epoxyeicosatrienoic acid involves the Akt/eNOS signaling pathways in human endothelial progenitor cells

The 11,12-epoxy-eicosatrienoic acid (11,12-EET) is formed from arachidonic acid (AA) by cytochrome P450 2J2 (CYP 2J2) epoxygenase and function as an effector in blood vessels. Human endothelial progenitor cells (hEPCs), a preceding cell source for endothelial cells (ECs), involve in the vascular tissue repairing by postnatal neovasculogenesis. However, the effect of 11, 12-EET on hEPCs and neovasculogenesis is not well known. In the current study, we examined the function of 11, 12-EET in hEPCs-mediated neovasculogenesis by using tubular formation analysis, Western Blotting assay, immunofluorescence staining, flow cytometry analysis and zymogram analysis. The results suggest that 11, 12-EET significantly induces neovasculogenesis through the phosphorylation of phosphoinositide 3-kinase (PI3–K)/Akt, endothelial-nitric oxide synthase (e-NOS) and extracellular signal-regulated kinase 1/2 (ERK 1/2) signaling pathways. 11, 12-EET up-regulates the expression of cyclin D1, cyclin–dependent kinase 4 (CDK4) and nuclear factor kappa B (NF-κB) proteins. Moreover, 11, 12-EET augments the expression of VE-cadherin and CD31 proteins in hEPCs. 11, 12-EET also augmented Rac1/Rho A signaling cascades, cell migration and an up-regulation of matrix metalloproteinase (MMP) −2 and −9 proteins. These results demonstrate that 11, 12-EET exerts a significant function in the neovasculogenesis of hEPCs.


Introduction
B one marrow (BM) provides stem cells and endothelial progenitor cells (hEPCs) for tissue repairment and the prevention of ischemic injuries [1]. Study indicated that human BM-derived hemangioblast, a preceding progenitor for hematopoietic cells, are evolved consequently into hEPCs/endothelial colony forming cells (ECFCs) and endothelial cells (ECs) during embryogenesis and provides the sources for neovascularization [2]. Many studies also suggest that hEPCs are implicated in postnatal neovasculogenesis/angiogenesis process in the adulthood [3,4]. Signaling cascades lead to the neovascularization are involved in the proliferation and outgrowth of hEPCs [5]. Recent studies suggested several important roles of BM -derived circulating hEPCs in neovascularization during different conditions such as wound healing and ischemic injuries [6e8]. Previous studies suggested that the phosphorylation of Akt and extracellular signal-regulated kinase 1/2 (ERK 1/2) proteins induces the proliferation of hEPCs through an increment of nuclear factor kappa B (NF-kB), cyclin D1 and cyclin edependent kinase 4 (CDK4) proteins [9,10]. The activation of Akt further augments the phosphorylation of downstream targets nuclear factor kappa B inhibitor (IkBa) and endothelial-nitric oxide synthase (e-NOS) proteins to enhance cell migration capability [11]. Moreover, small GTPases such as Rho A and Rac1 proteins are involved in cell movement and angiogenesis through the changes of cytoskeleton, assembly of contractile stress fibers and filopodial/lamellipodial extension [12]. These evidences suggest that RhoA and Rac1 proteins play important roles in cell migration through a modulation of actin cytoskeleton and the stability of microtubules [13,14]. A study also indicated that matrix metalloproteinase (MMP) proteins are involved in the cell recruitment and migration of hEPCs during the neovascularization process [15]. MMP-9 is involved in the augmentation of cell motility of BM-derived progenitor cells [15,16]. Recent study suggests that ECFCs, a late stage of hEPC population, possess the characteristics of a true endothelial progenitor and has a potential to differentiate into ECFCs [17]. A significant up-regulation of CD31 and VE-cadherin was observed during the differentiation of hEPCs into mature ECs [18e20].
A study indicate that CYP2J2 epoxygenase is expressed in vascular ECs and responsible for the conversion of arachidonic acid to the eicosanoid metabolites including 11,12-epoxyeicosatrienoic acids (EETs) [21]. An earlier study suggested that EETs involved in the VEGF-mediated angiogenesis and cancer development [22]. Therefore, the current study would examine the molecular mechanisms of 11, 12-EET in hEPCs-mediated neovasculogenesis.

Cell culture procedure
hEPCs were seeded onto culture dish and cultured in MCDB-131 medium with EGM-2 growth kit and 10% fetal bovine serum (FBS). Culture media were changed every 2 days. For these in vitro experiments, 11,12-EET was dissolved in dimethyl sulfoxide (DMSO) at a stock solution of 100 mM. hEPCs were treated with 11,12-EET (for 8 h) for further analysis of neovascularization, protein expression, cell migration, proliferation or zymogram analysis.

Assessment of cell survival
hEPCs (2 Â 10 4 cells) were seeded in 24-well plates with MCDB-131 medium containing 11,12-EET and different inhibitors for 8h. MTT assay was used to measure the cell proliferation. At the end of the experiment, the optical absorbance was analysed at wave length of 570 nm with a microplate reader.

Extraction of cellular proteins
Protein extractions were executed by NE-PER kit with inhibitors of phosphatase and protease. To remove the cell debris, cellular proteins were centrifuged for 10 min at 12,000Âg. The remaining H.-C. HUNG ET AL NEOVASCULOGENIC EFFECTS OF 11,12-EET supernatants were obtained as a cytoplasmic fraction.

Western Blotting analysis
Cellular proteins (70 mg) were separated by running in 10% sodium-dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The resulting SAS-PAGE gel was electroblotted to polyvinylidene difluoride (PVDF) membrane and covered with primary antibody solution. The blots were stripped and reprobed with internal control antibody.
Detection of other proteins including p-ERK 1/2, p-Akt, p-eNOS, p-IkBa, p-Rac 1 and RhoA was performed by using similar procedure described above. The t-ERK 1/2, t-Akt and t-eNOS antibodies were used as internal controls for p-ERK 1/2, p-Akt, p-eNOS proteins, respectively. Expression of nuclear proteins including cyclin D1, p65/NF-kB (RelA) and CDK4 was measured by also using similar procedure. The Lamin A protein was used as internal control.

Vascular tube formation assay
For the vascular tube formation assay, aliquots of 50 mL matrigel (4 mg/mL) were transferred to each well of a 96-well plate and incubated at 37 C until gelatinization occurred. For the neovasculogenesis assay, hEPCs (1 Â 10 4 cells) were cultured in the matrigel-coated 96 well plate with 10% FBS MCDB-131 medium. After cell seeding for 8 h, photos of neovasculogenesis was documented under inverted phase-contrast microscope. The final results were collected for data analysis by using the accessory software (Olympus imaging system) (Tokyo, Japan).

F-actin assembly detection
hEPCs growing on culture Tek-chamber slides with 11,12-EET (0 and 50 nM) at different time points (0, 0.5 and 2 h) were fixed with phosphate buffer saline (PBS) containing 3.7% formaldehyde and labeled with 5 units/mL of Alexa 488 phalloidin (Invitrogen Inc.). Cells were rinsed three times with PBS and were monitored using Confocal Microscope Detection System (Leica, Wetzlar, Germany) to perform image documentation and analysis.

Flow cytometric analysis of biomarker proteins
hEPCs for biomarker analysis were stained with either anti-human c-kit-fluorescein isothiocyanate (FITC), anti-human CD31-phycoerythrin (PE) or anti-human VE-cadherin-PE solution. Cells stayed on ice for 30 min until analysis by BD FACS Canto flow cytometry (BD Biosciences, Franklin Lakes, NJ). Results from staining cells were analyzed using the accessory software.

Gelatin zymography
Supernatant protein (20 mg) from conditioned medium of cultured hEPCs was transferred into an 8% polyacrylamide gel containing gelatin. At the end of electrophoresis, the polyacrylamide gel was washed with 2.5% Triton X-100 at room temperature and subsequently incubated in a reaction buffer (10 mM CaCl 2 , 0.15 M NaCl and 50 mM Tris) at 37 C overnight. To detect the enzymatic reaction of MMPs, gel was covered with 0.25% Coomassie blue solution and photograph documented on a light box. Proteolysis within polyacrylamide gel was observed as a white range in a dark field.

Statistical analysis
The biostatistic analysis was performed to investigate the difference in the vascular formation between 11,12-EET groups and control group of hEPCs by using SYSTAT software (Chicago, IL, USA). Confirmation of difference in neovasculogenic index was performed by using the one way ANOVA model and Tukey's post hoc test at the P ¼ 0.05 level. Confirmation of difference in protein expression was performed by using student t-test at the P ¼ 0.05 level.
It seems probable, therefore, that 11,12-EET effectively functioned as neovasculogenic agents through increased expression of VE-cadherin proteins as well as the phosphorylation levels of PI3eK/Akt, eNOS, NF-kB, and MAPK/ERK signaling molecules.
To further examine the neovasculogenic effects of 11,12-EET in hEPCs, we measured the expression of nuclear proteins. As shown in Fig. 3B and 11,12-EET enhanced nuclear level of NF-kB (p-p65; RelA) protein in hEPCs. Moreover, 11,12-EET significantly augmented the expression of cyclin D1 and CDK4 proteins in hEPCs. These results suggested that 11,12-EET mediated neovasculogenesis in hEPCs.

11, 12-EET significantly induced the expression of VE-cadherin and CD31 proteins in hEPCs
To further investigate the vasculogenic effects of 11,12-EET in hEPCs, we measured the expression of VE-cadherin, CD31 and c-kit biomarker proteins by using flow cytometry analysis. In Fig. 4, treatment of 11,12-EET significantly increased the VE-cadherinþ and CD-31þ cell subpopulations of hEPCs. These results suggested that EET treatment might augment the differentiation of hEPCs toward to VE-cadherinþ/ CD31þ late EPCs (ECFCs).

11,12-EET augmented Rac1/Rho A cascade, cell migration and upregulation of MMP-2,-9 proteins in hEPCs
To verify actions of 11, 12-EET on cell migration, we further elucidated the probable effects of 11,12-EET on Rac 1 and Rho A proteins. As demonstrated in Fig. 5A and 11,12-EET increased phosphorylated levels (i.e. activation) of the RAC 1 and an upregulation of Rho A proteins. The results suggested that 11,12-EET could modulate cell migration and enhance the expression of phosphorylated-Rac 1 and Rho A proteins in hEPCs.
To confirm these findings, we further examined whether 11,12-EET altered the distribution of F-actin proteins in hEPCs. As shown in Fig. 5B, F-actin stress fibers is randomly distributed in hEPCs at starting time point (Fig. 5B, a-b). Treatment of 11, 12-EET (50 nM) significantly induced the redistribution of F-actin stress fibers into a well-aligned pattern in hEPCs (Fig. 5B, c-f). 11,12-EET also significantly induces cell migration of hEPCs (P < 0.05) (Fig. 5C). Our results demonstrated that 11,12-EET increased cell proliferation levels in hEPCs (Fig. 5D). Previous study suggested that neovasculogenesis and angiogenesis are correlated with an increasing enzymatic activity of MMP-2 and MMP-9 proteins. Therefore, we further investigated whether 11,12-EET modulated MMP-2 and MMP -9 expression in hEPCs. In Fig. 5E, 11

Discussion
Many studies already showed that neovasculogenesis and angiogenesis play key steps in the physiological and pathological conditions including ischemia prevention and tumor development [23]. Moreover, studies suggested important roles of EETs in cardioprotection, angiogenesis and tumor 11,12-EET untreated control group (P < 0.05). (B) Analysis of nuclear proteins were conducted using antibodies against cyclin D1, CDK4, p-p65 (RelA) and lamin A. The integrated densities (mean ± SD) of these proteins are adjusted with the loading control lamin A protein are shown in the bottom row. A single asterisk represented a statistical difference in comparison with 11,12 -EET-untreated control group (P < 0.05). metastasis [22,24,25]. Therefore, EETs seem to play important roles in many aspects of chronic diseases. Adult neovasculogenesis occur through the circulating hEPCs into vascular sites and the proliferation of ECs [26e28]. A recent study suggested that welldifferentiated ECs with a low proliferative potential have low repairing capability for damaged ECs. hEPCs from adult BM obtain an early-onset differentiated capacity, an excellent repairing capability and neovasculogenesis-prone characteristics.
Many studies suggested that EETs generated by cytochrome P450 epoxygenases are vasodilators eicosanoids [29,30]. These P450 epoxygenases CYP2J2 are highly expressed in vascular ECs and capable of metabolizing arachidonic acid into several types of EETs including 11,12-EET [21]. Previous studies showed that a risk of CAD is inversely correlated with the levels of the Cytochrome P450 epoxygenases CYP2J2 and 11,12 EET [31,32]. EETs are also involved in the VEGF-mediated angiogenesis processes [22]. These results suggested that 11,12-EET has favorable effects on increasing endothelial functions, neovasculogenesis and even angiogenesis. A recent study indicated that 11,12-EET could promote organ and tissue regeneration [33]. Moreover, 11,12-EET could rescue impaired wound   healing under ischemic condition [34]. These in vivo findings suggested that 11,12-EET could act as an effective agent to increase neovascularization and prevent ischemic injuries. For the first time, we demonstrated that 11,12-EET induces neovascularization/angiogenesis through an EPC-differentiation into EC-like characteristics. The results showed that 11,12-EET significantly augmented neovasculogenesis of hEPCs, in part, through an up-regulation of VE-cadherin protein in vitro. The molecular actions of 11,12-EET were mediated through the activation of signaling pathways including Akt, eNOS and ERK 1/2 signaling cascades. 11,12-EET also augmented the expression of RhoA and the phosphorylation of Rac 1 proteins. These novel findings suggested that 11,12-EET may function as a signaling effecter to induce neovasculogenesis.
The results further demonstrated that 11,12-EET significantly increased the nuclear levels of cyclin D1 and CDK4 proteins in hEPCs. There are several possible scenarios for 11,12-EET to induce neovasculogenesis and angiogenesis. One explanation might be predominant activation of Akt, eNOS and MAPK/ERK molecules in hEPCs. Neovascularization requires the activation of Akt and ERK 1/2 pathways and eNOS protein in hEPCs. The Akt-mediated phosphorylation of eNOS would lead to an increasing cell migration of endothelial cells [35]. Inhibition of the signaling cascades including PI3eK/Akt and eNOS pathways was also correlated with increased apoptosis level in ECs [35]. Our results suggested that 11,12-EET plays an important role by induction of NO bioavailability in vascular tissues. Although we have not examined the stability of 11,12-EET and the metabolic consequence, it is probable that 11,12-EET obtains its effect on neovascularization.
Alternatively, we identify a new role of 11,12-EET in the upregulation of cyclin D1 and CDK4 proteins in hEPCs. It is probable that a closely correlation between the expression of cyclin D1 and CDK4 proteins in 11, 12-EET mediated neovasculogenesis. Results from the current study further implicated that 11,12-EET probably induced the differentiation of hEPCs into ECFCs (late-EPCs) through increased levels of VE-cadherin and CD31 proteins. Based on our observation, it is probable that 11,12-EET may function as an agent to induce differentiation of EPC into ECFCs. After all, 11,12-EET induced the differentiation of EPC into ECFCs and the proliferation of these cells through augmented expression of cyclinD1 and CDK4 proteins as well as the neovasculogenesis in hEPCs. These results prove the neovasculogenic and angiogenic effects of 11,12-EET. It may explain the important roles of 11,12-EET in the differentiation of BM-derived hEPCs and the proliferation of ECFCs.
To validate these important findings, we also measured the expression of MMP-2 and MMP-9 proteins in these 11,12 EET-treated EPCs. As shown in Fig. 5, these results suggested that 11,12-EETmediated cell migration and neovascularization were associated with augmented expression of MMP-2 and MMP-9 proteins in hEPCs. Augmented expression of MMP proteins and the activation of Rac 1 and RhoA molecules are observed in the 11,12 -EET -mediated neovasculogenesis. Although we have not verified the crucial role of 11,12-EET in experimental animal model yet, the probable limitation is to measure the bioavailability of 11,12-EET in an in vivo study. Previous studies indicated that increased activities of endothelial progenitor cells could play important roles in increment of wound healing, tissue repairment and tissue regeneration [1]. One of probable utilization of 11,12-EET could be applied to promote wound healing or tissue repairment in future clinical application.

Conclusion
In conclusion, the specific novel aspects of this study include the neovasculogenic effects of 11,12-EET on (i) the activation of PI3eK/Akt, eNOS and ERK 1/2 signaling pathways; (ii) the differentiation of EPCs into ECFCs; (iii) the proliferation of EPCs; and (iv) the cell migration of EPCs. The proposed mechanism was described in Fig. 6. Our results might provide an important insight leading to the application of 11,12-EET in the future preclinical and clinical studies.