Molecular Mechanisms Underlying the Proangiogenic Effect of Factor XIII
http://www.100md.com
动脉硬化血栓血管生物学 2005年第3期
From the Institute of Thrombosis and Hemostasis, Sheba Medical Center, Tel Hashomer and Sackler Faculty of Medicine (R.D., R.E., A.I.), Tel Aviv University, Israel; and the Whitaker Cardiovascular Institute and Evans Department of Medicine (J.L.), Boston University School of Medicine, Mass.
Correspondence to Aida Inbal, MD, Institute of Thrombosis and Hemostasis, Sheba Medical Center, Tel Hashomer, Israel 52621. E-mail aida.inbal@sheba.health.gov.il
Abstract
Objective— Coagulation Factor XIII (FXIII) was previously shown by us to induce angiogenesis. The aim of this study was to elucidate the molecular events underlying the proangiogenic effects of activated FXIII (FXIIIa) on human umbilical vein endothelial cells (HUVECs).
Methods and Results— As shown by coimmunoprecipitation studies, FXIIIa crosslinked v?3 with vascular endothelial growth factor receptor 2 (VEGFR-2) and enhanced the noncovalent interaction between the 2 receptors. In addition, FXIIIa induced tyrosine phosphorylation of VEGFR-2 in both the crosslinked high-molecular-weight and the noncovalent VEGFR-2/v?3 complexes. These effects as well as FXIIIa-induced proliferation and migration of HUVECs were abolished by iodoacetamide treatment of FXIIIa (I-FXIII) or by PTKI, an inhibitor of VEGFR-2. FXIIIa induced upregulation of c-Jun and Egr-1 as revealed by quantitative RT-PCR. Electrophoretic mobility-shift assay experiments showed that FXIIIa treatment of HUVECs enhanced binding of Wilm’s tumor-1 (WT-1) but not of early growth response (Egr)-1 to the thrombospondin-1 (TSP-1) promoter sequence, suggesting that WT-1 but not Egr-1 is involved in downregulation of TSP-1 expression.
Conclusions— The proangiogenic effect of FXIIIa is mediated by (1) enhancement of crosslinked and noncovalent v?3/VEGFR-2 complex formation; (2) tyrosine phosphorylation and activation of VEGFR-2; (3) upregulation of c-Jun and Egr-1; and (4) downregulation of TSP-1 induced indirectly by c-Jun through WT-1. These processes may clarify FXIII role in vascular remodeling and tissue repair.
Factor XIIIa (FXIIIa) was previously shown by us to induce angiogenesis. In the present work we show that the proangiogenic effect of FXIIIa is mediated by (1) enhancement of crosslinked and noncovalent v?3/vascular endothelial growth factor 2 (VEGFR-2) complex formation; (2) activation of VEGFR-2; (3) upregulation of cyclin D-1, Egr-1, and c-Jun; and (4) downregulation of thrombospondin-1 by WT-1.
Key Words: Factor XIII ? angiogenesis ? vascular endothelial growth factor receptor 2 ? cJun ? early growth response-1
Introduction
Angiogenesis is a complex process involving both proangiogenic and antiangiogenic factors. Vascular endothelial growth factor A (VEGF-A) is a major angiogenic growth factor regulating the key steps of angiogenesis, including endothelial cell (EC) proliferation, migration, and survival.1 VEGF-A exerts its proangiogenic effects by binding to the EC-specific tyrosine-kinase receptor VEGFR-2.1,2
After binding of VEGF-A to VEGFR-2, the receptor is autophosphorylated on tyrosine residues, followed by activation of the phosphatidyl-inositol kinase/protein kinase B (PI3K/PKB) and phospholipase C/protein kinase C (PLC/PKC) pathways, which play an important role in cell survival.2,3 In addition, binding of VEGF-A to VEGFR-2 has been shown to activate the mitogen-activated protein kinase kinase/extracellular signal regulated kinase (MEK/ERK) element of the mitogen-activated protein (MAP) kinase pathway, leading to upregulation of the transcription factor early growth response-1 (Egr-1), which is associated with cell growth and differentiation.4
v?3 is one of the most important integrins involved in angiogenesis and vasculogenesis. This integrin is highly expressed by angiogenic ECs in tumors and wounds where it functions as a regulator of the balance between the proliferative and apoptotic signals.5,6 In addition to a broad spectrum of adhesive glycoproteins, v?3 is capable of interacting with several growth factor receptors, such as the insulin receptor,7 the platelet-derived growth factor ? (PDGF-?) receptor,8 and VEGFR-2.8–10 The association of v?3 with VEGFR-2 was found to be essential for VEGFR-2 activation and phosphorylation in response to VEGF-A.9,10 Interestingly, VEGFR-2 levels are significantly elevated in mice lacking v?3, suggesting that the v?3/VEGFR-2 interaction might also regulate VEGFR-2 expression in ECs.11
Factor FXIII (FXIII) is a plasma transglutaminase that participates at the final step of the coagulation cascade. Thrombin-activated FXIII (FXIIIa) catalyzes formation of covalent crosslinks between -glutamyl and -lysyl residues of adjacent fibrin chains in polymerized fibrin to yield the mature clot.12
Besides its role in hemostasis, FXIII participates in tissue remodeling, wound healing, and embryo implantation as can be inferred from defects in these processes in patients with inherited FXIII deficiency.12 Wound healing and embryo implantation are complex processes that involve cell proliferation and angiogenesis. Until recently, the precise role of FXIII in these processes has not been sufficiently studied. In our previous work we showed for the first time that FXIIIa supports angiogenesis by enhancing EC migration, proliferation, and survival.13 In addition, FXIIIa induced new vessel formation in a rabbit cornea model, and this proangiogenic effect was associated with downregulation of thrombospondin-1 (TSP-1).13 In this study, we explore the molecular mechanism(s) underlying the proangiogenic activity of FXIII.
Materials and Methods
Cells
Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical cords and maintained in EGM-2 medium (Clonetics) as described previously.14 Dermal microvascular ECs were purchased from Clonetics and grown according to the manufacturer’s instructions. Subconfluent cells from passages 1 to 6 plated on fibronectin (unless stated otherwise) were used in all the experiments.
FXIII Activation
FXIII was activated by immobilized thrombin as described previously.13 FXIIIa was inactivated by treatment with 2 mmol/L iodoacetamide (I-FXIII), an inhibitor of transglutamination (Sigma) as described previously.13 (Detailed procedure appears in online Methods, available at http://atvb.ahajournals.org.)
Western Blot Analysis of HUVEC Proteins
Detailed procedure appears in online Methods.
Coimmunoprecipitation Experiments
Triton-X-100 extracts of untreated HUVECs or HUVECs treated with either FXIIIa or I-FXIII for 16 hours at 37°C were incubated with the appropriate primary antibody (goat anti-human VEGFR-2 [R&D Systems, Minneapolis, Minn] or sheep anti-human ?3 [Affinity Biological Inc]), or anti-v?3 complex LM609 (Chemicon, Temecula, Calif), followed by addition of protein A-Sepharose beads (Pharmacia). The beads were washed 3x with 1% Triton X-100 lysis buffer and boiled in SDS-PAGE sample buffer for 3 minutes. The precipitated proteins were resolved by 4% to 12% SDS-PAGE and transferred to a nylon membrane as described above. In some experiments, HUVECs were preincubated with 500 nM VEGFR-2 kinase inhibitor (Z)-5-bromo-3 (4,5,7-tetrahydro-1H-indol-2-ylmethylene)1,3-dihydroindol-2-one (Calbiochem; Cat. No. 676485) for 30 minutes at 37°C before the addition of FXIIIa. This compound is a membrane-permeable tyrosine kinase inhibitor with IC50=70 nM for VEGFR-2.15 Blots containing the proteins immunoprecipitated with anti-VEGFR-2 antibody were probed with mouse anti-?3 (AP3, GTI) or mouse anti-PY (4G10; Upstate Biotechnology, Lake Placid, NY) antibodies. Blots containing the proteins immunoprecipitated with the anti-?3 antibody were probed with either anti-VEGFR-2, anti-PY, anti--glutamyl--lysine isopeptide bond (clone 814-MAM, CovalAb, France), or anti-?3 (either sheep polyclonal or mouse monoclonal) antibodies. The signals were detected using the appropriate horseradish peroxidase-conjugated antibodies followed by enhanced chemiluminescence. The expression of VEGFR-2 on HUVECs was tested and found to be the same throughout the 1 to 6 passages used in these experiments.
HUVEC Migration and Proliferation
HUVEC migration was analyzed by a modified Boyden chamber assay13; proliferation was assessed by measuring 3[H]-Thymidine incorporation.13 Detailed migration and proliferation procedures are provided in online Methods.
Real-Time PCR
Total RNA extracted from one 75-cm2 flask of either untreated HUVECs or HUVECs treated with 50 μg/mL FXIIIa for 16 hours at 37°C was isolated using the Qiagen RNA isolation kit (Qiagen). RNA was then reverse transcribed into cDNA by MMLV reverse transcriptase (Promega). cDNA samples were amplified using specific primers for either c-Jun (forward: 5'-agtccagcaacgggcacatc; reverse: 5'-ctcctgctcatctgtcacgttct; accession number: JO4111.1), Egr-1 (forward: 5'-cgtcctcagcaccatctc; reverse: 5'-ggctcagggaaaatgtcag;accession number: NM_001964), or GAPDH (forward: cagcctcaagatcatcagca; reverse: gtcttctgggtggcagtgat; accession number: M33197) and the Quanti Test Sybr Green RT-PCR kit (Qiagen). Amplification was monitored by real-time PCR analysis using the ABI/Prism 7700 Sequence Detector System (Applied Biosystems).
Electrophoretic Mobility-Shift Assay Experiments
Nuclear proteins were prepared as described.16 Protein concentrations were measured using Bradford analysis. The oligonucleotides used for electrophoretic mobility-shift assay (EMSA) experiments were: probe 1 forward: 5' -gctgcgtgggcgggcaccgact-3'; probe 1 reverse: 5'-agtcggtgcccgcccacgcagc-3'; probe 2 forward: 5'-ccgacccccgcccccttc-3'; probe 2 reverse: 5'- gaagggggcgggggtcgg-3'. Each oligonucleotide (10 pmol) was end-labeled with 32 [P]-ATP using polynucleotide kinase (New England Biolabs Inc). The radiolabeled oligonucleotides were purified from unincorporated 32 [P]-ATP using G-50 Spin columns (New England Biolabs Inc), mixed, and incubated for 15 minutes at room temperature to allow annealing. Ten μg of HUVEC nuclear proteins prepared as described above were mixed with 1 μg of poly-(dI-dC) and 1 μL (50 000 cpm) of the double-stranded oligonucleotides in binding buffer (20 mmol/L HEPES, pH 7.5, 1 mmol/L EDTA, 5 mmol/L MgCl2, 50 mmol/L KCl, 1 mmol/L DTT, and 10% glycerol). For supershift experiments, the nuclear extracts were preincubated with either anti-Egr-1 (rabbit anti-human Egr-1, Santa Cruz Biotechnology Inc, Santa Cruz, Calif), or anti-WT-1 antibodies (mouse anti-human WT-1, Santa Cruz Biotechnology Inc, Santa Cruz, Calif) at a ratio of 10 μg IgG/10 μg of total nuclear protein for 30 minutes at room temperature before the addition of the radiolabeled probe. To test the specificity of the formed complexes, competition assays using a 100-fold excess of cold double-stranded oligonucleotides were performed. The DNA-protein complexes were resolved on 6% nondenaturing polyacrylamide gels and detected by autoradiography.
Analysis of v?3 Activation
v?3 activation was analyzed by flow cytometry using WOW-1 antibody (kindly provided by Dr S. Shattil, Scripps Research Institute), which recognizes the v?3 integrin in its active conformation. Detailed procedure appears in online Methods.
Statistical Analysis
Statistical analysis was performed by unpaired Student t test. Results are expressed as mean±SD. P<0.05 was considered significant.
Results
Tyrosine Phosphorylation of HUVEC Proteins Induced by FXIIIa
Triton X-100 soluble lysates of control HUVECs or HUVECs incubated in the presence of either FXIIIa or I-FXIII were tested for tyrosine phosphorylation by Western blot (WB) analysis. As shown in Figure 1A, FXIIIa, but not I-FXIII, stimulated tyrosine phosphorylation of 2 proteins with of 250 and 230 kDa. Another phosphorylated protein with of 120 kDa was present in untreated cells and was not affected by FXIIIa or I-FXIII treatment.
Figure 1. Tyrosine phosphorylation of HUVEC proteins after FXIIIa treatment. The representative picture of 3 separate experiments is shown. A, WB analysis with anti-PY antibody of Triton X-100 soluble extracts of either untreated HUVEC or HUVEC treated with FXIIIa or I-FXIII. The numbers above each lane indicate FXIIIa concentration in μg/mL. B, Triton X-100 soluble extracts of either control HUVECs (C) or HUVECs treated with 10 ng/mL VEGF-A, 50 μg/mL FXIIIa, 50 μg/mL FXIIIa after immunodepletion (D), and VEGF-A after preincubation with 500 nM protein tyrosine kinase inhibitor (VEGF-A+PTKI). The samples were immunoprecipitated (IP) with anti-VEGFR-2, followed by WB analysis with either anti-VEGFR-2 or anti-PY antibodies. The 230-kDa band denotes VEGFR-2.
To identify the candidate protein undergoing tyrosine-phosphorylation in response to FXIIIa, the blot shown in Figure 1A was stripped and incubated with VEGFR-2 antibody, which bound to the 230-kDa tyrosine-phosphorylated band (not shown).
Tyrosine phosphorylation of VEGFR-2 was further verified by immunoprecipitation of VEGFR-2 followed by WB analysis with anti-PY antibody (Figure 1B). VEGFR-2 was tyrosine phosphorylated in FXIIIa-treated and in VEGF-A-treated samples. VEGF-A- or FXIIIa-induced VEGFR-2 phosphorylation was abolished by a specific VEGFR-2 tyrosine kinase inhibitor PTKI (an inhibitor of VEGFR-2) or depletion of the FXIIIa preparation by passage over an immobilized anti-FXIII antibody (D), respectively (Figure 1B).
The Role of FXIIIa Crosslinkng in v?3–VEGFR-2 Interaction
We tested whether FXIIIa affects the v?3 and VEGFR-2 interaction, because this interaction was previously shown to play an important role in VEGFR-2–induced signaling.9,10 Triton X-100 lysates of control and FXIIIa or I-FXIII-treated HUVECs were immunoprecipitated with anti-?3 antibody followed by WB analysis using anti-VEGFR-2 antibody (Figure 2A). A high-molecular-weight complex recognized by the 2 antibodies was observed at the interface between stacking and running gels in the lanes containing FXIIIa-treated HUVECs. This complex was absent in the cells treated with I-FXIII. In addition, a 230-kDa band corresponding to the mature form of VEGFR-2 was observed in all the lanes, with an increase in intensity in a FXIIIa dose-dependent manner indicating VEGFR-2/?3 complex formation without crosslinking. To determine whether the VEGFR-2/?3 complex undergoes phosphorylation, this blot was stripped and reprobed with anti-PY antibody. WB with anti-PY revealed that VEGFR-2 in both the high-molecular-weight and the noncovalent VEGFR-2/?3 complexes underwent phosphorylation (Figure 2A). To prove that VEGFR-2 and ?3 within the high-molecular-weight complex are crosslinked by FXIIIa, the blot containing immunoprecipitated VEGFR-2 and ?3 proteins was subjected to WB analysis using an antibody recognizing the -glutamyl--lysine isopeptide bonds. As shown in Figure 2A, this antibody reacted with VEGFR-2/?3 high-molecular-weight complex present only in FXIIIa-treated samples. Immunoprecipitation of HUVECs with anti-VEGFR-2 and WB with anti-?3 detected the high-molecular-weight complex only in the lane corresponding to FXIIIa-treated HUVECs (Figure 2B), suggesting that VEGFR-2 was crosslinked to ?3. Non-crosslinked ?3 that coimmunoprecipitated with VEGFR-2 was observed in control as well as in FXIIIa or I-FXIII samples (Figure 2B). Assessment of the extent of crosslinking (see online Methods) revealed that 12.5% of total VEGFR-2 and 8% of total v?3 extracted from HUVECs underwent crosslinking.
Figure 2. Coimmunoprecipitation of HUVEC proteins with anti-VEGFR-2 and anti-?3 antibodies. The representative picture of 3 separate experiments is shown. A through C, Triton X-100 soluble proteins of either untreated or HUVECs treated with FXIIIa or I-FXIII. The numbers above each lane indicate concentration in μg/mL. The samples were subjected to immunoprecipitation (IP) with either anti-?3 (A and D), anti-VEGFR-2 (B), or LM609 (C) antibodies. The immunoprecipitated proteins were analyzed by WB using anti-VEGFR-2, anti-PY, anti--glutamyl- -lysine isopeptide bond antibodies, or anti-?3 (A), anti-?3, or anti-VEGFR-2 (B through D).
To determine whether ?3 is in complex with v chain while interacting with VEGFR-2, HUVEC proteins were immunoprecipitated with LM609 antibody recognizing the v?3 integrin complex, followed by WB analysis with anti-VEGFR-2 (Figure 2C). A high-molecular-weight crosslinked VEGFR-2/v?3 complex was observed only in FXIIIa-treated cells. The v?3 integrin within the complex was not activated as demonstrated by lack of WOW-1 antibody binding to FXIIIa-treated cells (Figure I, available online at http://atvb.ahajournals.org).
To evaluate the role of the tyrosine kinase activity of VEGFR-2 in the VEGFR-2/?3 interaction, we performed a coimmunoprecipitation experiment using lysates of control and FXIIIa-treated HUVECs preincubated with a specific VEGFR-2 tyrosine kinase inhibitor (PTKI; Figure 2D). This inhibitor blocks VEGF-A-induced VEGFR-2 phosphorylation as shown by Sun et al15 and by us in Figure 1B. As demonstrated in Figure 2D, this inhibitor markedly decreased VEGFR-2/?3 complex formation in nontreated or FXIIIa-treated cells and completely abolished crosslinking of the VEGFR-2/?3 complex induced by FXIIIa. FXIIIa mediated VEGFR-2/v?3 crosslinking, and VEGFR-2 phosphorylation was also observed in dermal microvascular ECs (data not shown).
To find out whether VEGFR-2/?3 crosslinking follows kinetics similar to that of downstream signaling, time-course experiments were performed and are presented in Figure 3. Weak bands corresponding to phosphorylated VEGFR-2 (Figure 3A) and VEGFR-2/?3 crosslinked complex (Figure 3B) were observed only after 1 hour of incubation with FXIIIa. Both bands increased in intensity in a time-dependent manner, reaching a plateau after 8 hours of exposure to FXIIIa. Similar kinetics was observed for downstream activation of Akt (Figure 3C). In contrast, maximal activation of ERK was observed already after 2 hours of incubation (Figure 3C). It is possible that partial activation of VEGFR-2 is sufficient to trigger a full activation of the MEK/ERK system. Treatment of HUVECs with VEGF resulted in more rapid maximal activation of both ERK and Akt (after only 10 minutes of incubation; Figure 3C). Densitometric assessment of the time course experiments is shown in Figure II (available online at http://atvb.ahajournals.org).
Figure 3. Time course of FXIIIa-induced events. The representative picture of 2 separate experiments is shown. In A and B, Triton X-100 extracts of HUVECs treated with 50 μg/mL FXIIIa for various time periods were immunoprecipitated (IP) and analyzed by WB. A, Time course of VEGFR-2 phosphorylation: IP with anti-VEGFR-2; WB analysis using either anti-PY or anti-VEGFR-2 antibodies. B, Time course of crosslinked VEGFR-2/?3 complex formation: IP with anti-?3; WB analysis using either anti--glutamyl--lysine or anti-?3 antibodies. C, Total Triton X-100 extracts of HUVECs treated with either 50 μg/mL FXIIIa or 10 ng/mL VEGF-A for various time periods were analyzed with either anti-phospho-ERK, anti-ERK, anti-phospho-Akt, or anti-Akt antibodies. The numbers above the bars indicate incubation periods in minutes and hours for VEGF-A and FXIIIa, respectively.
To provide functional evidence that the pathways studied are responsible for the angiogenic action of F-XIII, we performed migration and proliferation assays in the presence of FXIIIa, VEGF, or both. As shown in Figure 4, FXIIIa significantly increased EC migration and proliferation, and both processes were inhibited by VEGFR-2 tyrosine kinase inhibitor PTKI. Similar results were observed with VEGF. In addition, these experiments showed that FXIIIa had no additional effect on VEGF-A-induced migration and proliferation.
Figure 4. Effect of FXIIIa/VEGF on HUVEC proliferation and migration. [3H]-thymidine incorporation and migration of HUVECs induced by either 10 ng/mL VEGF-A, or 50 μg/mL FXIIIa, or a combination of both in the absence or presence of 500 nM PTKI. Data for proliferation and migration are presented as percent of control and number of migrating cells per field, respectively (mean±SD of 3 separate experiments performed in quadruplicates). *P<0.05.
Effect of FXIIIa on Gene Expression of Transcription Factors and Signaling Molecules
After treatment of HUVECs with FXIIIa, suspected upregulation of cyclin D1 and Egr-1 was observed by microarray analysis (data not shown). However, validation of the upregulation by real-time PCR and WB disclosed that only Egr-1 gene and protein were upregulated. A 7.5±2.2-fold increase in mRNA (P<0.05) and 3.7±1.5-fold increase in protein level (by densitometry; P<0.05) was observed (Figure III, available online at http://atvb.ahajournals.org).
Because we have previously shown that FXIIIa proangiogenic effect is associated with downregulation of TSP-1,13 and Dejong et al have demonstrated that downregulation of TSP-1 was associated with upregulation of c-Jun,16 we examined by real-time PCR whether FXIIIa affects c-Jun expression. FXIIIa induced a 1.65±0.15-fold increase in c-Jun mRNA as shown by real-time PCR (Figure III, P<0.05). Accordingly, a 2.1±0.3-fold increase (by densitometry; P<0.05) in the amount of c-Jun protein was observed by WB analysis (representative WB in Figure IIIB). Thus, FXIIIa effect is associated with upregulation of c-Jun.
Effect of FXIIIa on Binding of Nuclear Proteins to the TSP-1 Promoter
c-Jun was previously shown to downregulate TSP-1 by inducing WT-1 binding to the TSP-1 promoter.16 We examined whether WT-1 binding to TSP-1 promoter is affected by FXIIIa by an EMSA experiment using radiolabeled oligonucleotide probes containing a GC-rich TSP-1 promoter sequence (probe 1) as outlined in Methods. As shown in Figure 5A, the treatment of HUVECs with FXIIIa increased complex formation between the probe and one of the nuclear proteins, which was significantly reduced by addition of an anti-WT-1 antibody. Because the amount of the WT-1 protein in the nuclear extracts of HUVECs was not affected by FXIIIa (Figure 5B), the downregulation of TSP-1 associated with FXIIIa may be caused by modulation of WT-1 binding to TSP-1 rather than WT-1 protein synthesis. Similarly, an EMSA experiment using radiolabeled oligonucleotide probes 1 and 2 (Methods) was undertaken to analyze binding of Egr-1 to TSP-1 promoter after FXIIIa treatment. Figure 5A shows that anti-Egr-1 antibody had no effect on FXIIIa-induced complex formation between probe 1 and nuclear proteins. Binding of probe 2 to nuclear proteins was not affected by FXIIIa although it was inhibited by anti-Egr-1 antibody. WB analysis of FXIIIa-treated HUVECs with anti-Egr-1 antibody revealed that Egr-1 was markedly upregulated (Figure 5B). Taken together, these data suggest that although Egr-1 protein is upregulated after FXIIIa treatment, its binding to the TSP-1 gene is not visibly regulated by FXIIIa.
Figure 5. EMSA of FXIIIa-induced binding of TSP-1 promoter to WT-1. The representative picture of 4 separate experiments is shown. A, Ten μg of nuclear proteins extracted from untreated HUVECs (–) or HUVECs treated with 50 μg/mL of FXIIIa (+) were subjected to EMSA analysis using radiolabeled probe 1 or 2 (50 000 cpm) from the TSP-1 promoter as described in Materials and Methods. Anti-WT-1 or anti-Egr-1 antibodies were used to identify the DNA-protein complexes (denoted by arrows). NSB indicates nonspecific binding. B, Fifty μg of nuclear proteins extracted from untreated HUVECs (–) or HUVECs treated with 50 μg/mL of FXIIIa (+) were subjected to WB analysis using either anti-WT-1 or anti-Egr-1 antibody.
Discussion
In this study, we have explored the molecular mechanisms that underlie the proangiogenic activity of FXIII reported previously by us.13 Our results show that FXIIIa crosslinks ?3 integrin to VEGFR-2 as well as enhances the noncovalent complex formation between the ?3 integrin and VEGFR-2. Moreover, FXIIIa induces tyrosine phosphorylation of VEGFR-2 in both the crosslinked and noncovalent ?3-VEGFR-2 complexes. The formation of the complexes appears to be dependent on the intact catalytic site of VEGFR-2, because it is inhibited by a specific VEGFR-2 tyrosine kinase inhibitor (PTKI). It is, however, unclear why VEGFR-2/?3 noncovalent and crosslinked complexes depend on the intact catalytic site of VEGFR-2. One of the possibilities is that the inhibitor binding induces a conformational change in the receptor, thereby interfering with VEGFR-2/v ?3 interaction.
The effects of FXIIIa on both the v?3 integrin-VEGFR-2 association and tyrosine phosphorylation of VEGFR-2 were found to be absolutely dependent on its transglutaminase activity, as indicated by the lack of effect of iodoacetamide-inhibited FXIIIa. Taken together, these data suggest that the proangiogenic effect of FXIIIa, in part, involves crosslinking of the 2 receptors v?3 integrin and VEGFR-2 followed by activation of VEGFR-2 similarly to that elicited by VEGF-A.2 The faster kinetics of VEGF-induced as compared with FXIIIa-induced phosphorylation of signaling proteins may suggest that VEGF is the primary proangiogenic factor, the signals of which are immediately/rapidly transduced, whereas FXIII may be essential for maintaining the angiogenic process and therefore acts at a much slower pace. This is further supported by the incapability of FXIIIa to enhance maximal VEGF-A-induced migration and proliferation of HUVECs.
The tyrosine-phosphorylated VEGFR-2 was shown to mediate activation of MAP kinase, PI-3 kinase, c-GMP-dependent kinase, Jun-N-terminal kinase, and focal adhesion kinase, all of which are involved in regulation of the angiogenic response of ECs to VEGF-A.17 Our data indicate that, similar to VEGF-A, FXIIIa induces tyrosine-phosphorylation of VEGFR-2 in HUVECs; however, it does not increase the amounts of VEGF-A or VEGFR-2 as shown by us previously.13 It seems that VEGFR-2 phosphorylation is the major and vital step in the initiation of FXIIIa-induced angiogenesis. Inhibition of FXIII-mediated proliferation and migration by PTKI further confirm that VEGFR-2 phosphorylation is the event responsible for the angiogenic action of FXIII.
Activated tyrosine-phosphorylated VEGFR-2 binds and activates PLC, which in turn activates PKC.2 Activated PKC induces activation of MAPK,17 which is in accordance with our results showing that FXIIIa-mediated VEGFR-2/v?3 crosslinking with subsequent VEGFR-2 phosphorylation is associated with MAPK activation. Activation of VEGFR-2 by VEGF-A induces upregulation of Egr-1 in a PKC- and MAPK-dependent manner,4,18 which is also in agreement with our data showing that the FXIIIa-induced proangiogenic effect is associated with upregulation of Egr-1.
Egr-1 is a zinc-finger transcription factor involved in cell proliferation and differentiation and regulation of multiple genes implicated in vasculogenesis and angiogenesis. Upregulation of Egr-1 demonstrated in this study indicates that Egr-1 may serve as a direct mediator of FXIIIa-induced EC proliferation rather than mediating TSP-1 suppression, because its binding to the TSP-1 promoter is not affected by FXIIIa. That Egr-1 participates in angiogenesis and in wound healing18,19 supports its involvement in FXIII-induced wound healing12 owing to the ability of FXIIIa to upregulate Egr-1.
We have recently shown that the proangiogenic activity of FXIIIa is associated with downregulation of TSP-1 in in vitro and in vivo models.13 Several lines of evidence indicate that downregulation of TSP-1 is linked to overexpression of the transcription factor c-Jun, a positive regulator of cell proliferation.16,20 In rat fibroblasts, the negative regulation of TSP-1 transcription by c-Jun was shown to be mediated by transcription factor WT-1 rather than by direct repression by c-Jun.16 In this study, c-Jun was found to be upregulated by FXIIIa at both mRNA and protein levels. In addition, FXIIIa induced an increase in binding of WT-1 but not Egr-1 to the TSP-1 promoter. Based on these results and the work of Dejong and colleagues,16 it seems most likely that the downregulation of TSP-1 by FXIIIa stems from indirect suppression of TSP-1 transcription by c-Jun through WT-1.
We and others have previously reported that FXIIIa binds to the v?3 integrin expressed on the surface of ECs.21,22 In this study, FXIIIa-v?3 interaction on HUVECs was not associated with v?3 activation. Because FXIIIa does not contain the conventional integrin recognition and activation sequence RGD, its binding to v?3 may be mediated by the LDV sequence without activating the integrin. LDV was previously shown to mediate the interaction of fibronectin with 4?1.23
Hemostasis and angiogenesis are interrelated. Proteins generated by the hemostatic system coordinate the spatial localization and stabilization of ECs followed by growth and repair of a damaged blood vessel.24 After clot stabilization, angiogenesis is regulated through proteins secreted by platelets and cryptic fragments generated from coagulation and fibrinolytic system.24 The well-characterized proangiogenic coagulation proteins are thrombin, FVII, and tissue factor. The antiangiogenic proteins are antithrombin, domain 5 of the high-molecular-weight kininogen, prothrombin fragments 1 and 2, u-PA, plasminogen activator inhibitor-1, and angiostatin.24 In this study, we focused on the regulation of angiogenesis mediated by clot stabilization protein-FXIII. Based on our results from this and previous studies, we propose the following model for the proangiogenic effect of FXIIIa (Figure IV, available online at http://atvb.ahajournals.org): FXIIIa binding to the EC v?3 integrin is followed by enhancement of the v?3–VEGFR-2 interaction resulting in tyrosine phosphorylation and activation of VEGFR-2. Activated VEGFR-2 then initiates a signaling cascade leading to upregulation of Egr-1 and c-Jun, which may then be responsible for the enhanced cell proliferation and survival induced by FXIIIa. In addition, downregulation of TSP-1 by c-Jun through WT-1 may further contribute to the proangiogenic activities exhibited by FXIIIa. This FXIIIa-mediated proangiogenic effect may finally clarify the role of FXIII in vascular remodeling and tissue repair.
Received July 27, 2004; accepted December 3, 2004.
References
Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med. 2003; 9: 669–676.
Takahashi T, Shibuya M. The 230 kDa mature form of KDR/Flk-1 (VEGF receptor-2) activates the PLC-gamma pathway and partially induces mitotic signals in NIH3T3 fibroblasts. Oncogene. 1997; 14: 2079–2089.
Matsumoto T, Claesson-Welsh L. VEGF receptor signal transduction. Sci STKE. 2001; 2001: RE21. Review.
Mechtcheriakova D, Schabbauer G, Lucerna M, Clauss M, De Martin R, Binder BR, Hofer E. Specificity, diversity, and convergence in VEGF and TNF-alpha signaling events leading to tissue factor up-regulation via EGR-1 in endothelial cells. FASEB J. 2001; 15: 230–242.
Varner JA, Brooks PC, Cheresh DA. REVIEW: the integrin alpha V beta 3: angiogenesis and apoptosis. Cell Adhes Commun. 1995; 3: 367–374.
Eliceiri BP, Cheresh DA. Role of alpha v integrins during angiogenesis. Cancer J. 2000; 6 (suppl 3): S245–S249.
Vuori K, Ruoslahti E. Association of insulin receptor substrate-1 with integrins. Science. 1994; 266: 1576–1578.
Borges E, Jan Y, Ruoslahti E. Platelet-derived growth factor receptor beta and vascular endothelial growth factor receptor 2 bind to the beta 3 integrin through its extracellular domain. J Biol Chem. 2000; 275: 39867–39873.
Soldi R, Mitola S, Strasly M, Defilippi P, Tarone G, Bussolino F. Role of alphavbeta3 integrin in the activation of vascular endothelial growth factor receptor-2. EMBO J. 1999; 18: 882–892.
Masson-Gadais B, Houle F, Laferriere J, Huot J. Integrin alphavbeta3, requirement for VEGFR2-mediated activation of SAPK2/p38 and for Hsp90-dependent phosphorylation of focal adhesion kinase in endothelial cells activated by VEGF. Cell Stress Chaperones. 2003; 8: 37–52.
Reynolds LE, Wyder L, Lively JC, Taverna D, Robinson SD, Huang X, Sheppard D, Hynes RO, Hodivala-Dilke KM. Enhanced pathological angiogenesis in mice lacking beta3 integrin or beta3 and beta5 integrins. Nat Med. 2002; 8: 27–34.
Muszbek L, Adany R, Mikkola H. Novel aspects of blood coagulation factor XIII. I. Structure, distribution, activation, and function. Crit Rev Clin Lab Sci. 1996; 33: 357–421.
Dardik R, Solomon A, Loscalzo J, Eskaraev R, Bialik A, Goldberg I, Schiby G, Inbal A. Novel proangiogenic effect of factor XIII associated with suppression of thrombospondin 1 expression. Arterioscler Thromb Vasc Biol. 2003; 23: 1472–1477.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest. 1973; 52: 2745–2756.
Sun L, Tran N, Liang C, Hubbard S, Tang F, Lipson K, Schreck R, Zhou Y, McMahon G, Tang C. Identification of substituted 3--1,3-dihydroindol-2-ones as growth factor receptor inhibitors for VEGF-R2 (Flk-1/KDR), FGF-R1, and PDGF-beta tyrosine kinases. J Med Chem. 2000; 43: 2655–2663.
Dejong V, Degeorges A, Filleur S, Ait-Si-Ali S, Mettouchi A, Bornstein P, Binetruy B, Cabon F. The Wilms’ tumor gene product represses the transcription of thrombospondin 1 in response to overexpression of c-Jun. Oncogene. 1999; 18: 3143–3151.
Petrova TV, Makinen T, Alitalo K. Signaling via vascular endothelial growth factor receptors. Exp Cell Res. 1999; 253: 117–130.
Lucerna M, Mechtcheriakova D, Kadl A, Schabbauer G, Schafer R, Gruber F, Koshelnick Y, Muller HD, Issbrucker K, Clauss M, Binder BR, Hofer E. NAB2, a corepressor of EGR-1, inhibits vascular endothelial growth factor-mediated gene induction and angiogenic responses of endothelial cells. J Biol Chem. 2003; 278: 11433–11440.
Bryant M, Drew GM, Houston P, Hissey P, Campbell CJ, Braddock M. Tissue repair with a therapeutic transcription factor. Hum Gene Ther. 2000; 11: 2143–2158.
Kim SA, Um SJ, Kang JH, Hong KJ. Expression of thrombospondin-1 in human hepatocarcinoma cell lines and its regulation by transcription factor Jun/AP-1. Mol Cell Biochem. 2001; 216: 21–29.
Dardik R, Shenkman B, Tamarin I, Eskaraev R, Harsfalvi J, Varon D, Inbal A. Factor XIII mediates adhesion of platelets to endothelial cells through alpha(v)beta(3) and glycoprotein IIb/IIIa integrins. Thromb Res. 2002; 105: 317–323.
Dallabrida SM, Falls LA, Farrell DH. Factor XIIIa supports microvascular endothelial cell adhesion and inhibits capillary tube formation in fibrin. Blood. 2000; 95: 2586–2592.
Mould AP, Wheldon LA, Komoriya A, Wayner EA, Yamada KM, Humphries MJ. Affinity chromatographic isolation of the melanoma adhesion receptor for the IIICS region of fibronectin and its identification as the integrin alpha 4 beta 1. J Biol Chem. 1990; 265: 4020–4024.
Browder T, Folkman J, Pirie-Shephered S. The hemostatic system as a regulator of angiogenesis. J Biol Chem. 2000; 275: 1521–1524.(Rima Dardik; Joseph Losca)
Correspondence to Aida Inbal, MD, Institute of Thrombosis and Hemostasis, Sheba Medical Center, Tel Hashomer, Israel 52621. E-mail aida.inbal@sheba.health.gov.il
Abstract
Objective— Coagulation Factor XIII (FXIII) was previously shown by us to induce angiogenesis. The aim of this study was to elucidate the molecular events underlying the proangiogenic effects of activated FXIII (FXIIIa) on human umbilical vein endothelial cells (HUVECs).
Methods and Results— As shown by coimmunoprecipitation studies, FXIIIa crosslinked v?3 with vascular endothelial growth factor receptor 2 (VEGFR-2) and enhanced the noncovalent interaction between the 2 receptors. In addition, FXIIIa induced tyrosine phosphorylation of VEGFR-2 in both the crosslinked high-molecular-weight and the noncovalent VEGFR-2/v?3 complexes. These effects as well as FXIIIa-induced proliferation and migration of HUVECs were abolished by iodoacetamide treatment of FXIIIa (I-FXIII) or by PTKI, an inhibitor of VEGFR-2. FXIIIa induced upregulation of c-Jun and Egr-1 as revealed by quantitative RT-PCR. Electrophoretic mobility-shift assay experiments showed that FXIIIa treatment of HUVECs enhanced binding of Wilm’s tumor-1 (WT-1) but not of early growth response (Egr)-1 to the thrombospondin-1 (TSP-1) promoter sequence, suggesting that WT-1 but not Egr-1 is involved in downregulation of TSP-1 expression.
Conclusions— The proangiogenic effect of FXIIIa is mediated by (1) enhancement of crosslinked and noncovalent v?3/VEGFR-2 complex formation; (2) tyrosine phosphorylation and activation of VEGFR-2; (3) upregulation of c-Jun and Egr-1; and (4) downregulation of TSP-1 induced indirectly by c-Jun through WT-1. These processes may clarify FXIII role in vascular remodeling and tissue repair.
Factor XIIIa (FXIIIa) was previously shown by us to induce angiogenesis. In the present work we show that the proangiogenic effect of FXIIIa is mediated by (1) enhancement of crosslinked and noncovalent v?3/vascular endothelial growth factor 2 (VEGFR-2) complex formation; (2) activation of VEGFR-2; (3) upregulation of cyclin D-1, Egr-1, and c-Jun; and (4) downregulation of thrombospondin-1 by WT-1.
Key Words: Factor XIII ? angiogenesis ? vascular endothelial growth factor receptor 2 ? cJun ? early growth response-1
Introduction
Angiogenesis is a complex process involving both proangiogenic and antiangiogenic factors. Vascular endothelial growth factor A (VEGF-A) is a major angiogenic growth factor regulating the key steps of angiogenesis, including endothelial cell (EC) proliferation, migration, and survival.1 VEGF-A exerts its proangiogenic effects by binding to the EC-specific tyrosine-kinase receptor VEGFR-2.1,2
After binding of VEGF-A to VEGFR-2, the receptor is autophosphorylated on tyrosine residues, followed by activation of the phosphatidyl-inositol kinase/protein kinase B (PI3K/PKB) and phospholipase C/protein kinase C (PLC/PKC) pathways, which play an important role in cell survival.2,3 In addition, binding of VEGF-A to VEGFR-2 has been shown to activate the mitogen-activated protein kinase kinase/extracellular signal regulated kinase (MEK/ERK) element of the mitogen-activated protein (MAP) kinase pathway, leading to upregulation of the transcription factor early growth response-1 (Egr-1), which is associated with cell growth and differentiation.4
v?3 is one of the most important integrins involved in angiogenesis and vasculogenesis. This integrin is highly expressed by angiogenic ECs in tumors and wounds where it functions as a regulator of the balance between the proliferative and apoptotic signals.5,6 In addition to a broad spectrum of adhesive glycoproteins, v?3 is capable of interacting with several growth factor receptors, such as the insulin receptor,7 the platelet-derived growth factor ? (PDGF-?) receptor,8 and VEGFR-2.8–10 The association of v?3 with VEGFR-2 was found to be essential for VEGFR-2 activation and phosphorylation in response to VEGF-A.9,10 Interestingly, VEGFR-2 levels are significantly elevated in mice lacking v?3, suggesting that the v?3/VEGFR-2 interaction might also regulate VEGFR-2 expression in ECs.11
Factor FXIII (FXIII) is a plasma transglutaminase that participates at the final step of the coagulation cascade. Thrombin-activated FXIII (FXIIIa) catalyzes formation of covalent crosslinks between -glutamyl and -lysyl residues of adjacent fibrin chains in polymerized fibrin to yield the mature clot.12
Besides its role in hemostasis, FXIII participates in tissue remodeling, wound healing, and embryo implantation as can be inferred from defects in these processes in patients with inherited FXIII deficiency.12 Wound healing and embryo implantation are complex processes that involve cell proliferation and angiogenesis. Until recently, the precise role of FXIII in these processes has not been sufficiently studied. In our previous work we showed for the first time that FXIIIa supports angiogenesis by enhancing EC migration, proliferation, and survival.13 In addition, FXIIIa induced new vessel formation in a rabbit cornea model, and this proangiogenic effect was associated with downregulation of thrombospondin-1 (TSP-1).13 In this study, we explore the molecular mechanism(s) underlying the proangiogenic activity of FXIII.
Materials and Methods
Cells
Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical cords and maintained in EGM-2 medium (Clonetics) as described previously.14 Dermal microvascular ECs were purchased from Clonetics and grown according to the manufacturer’s instructions. Subconfluent cells from passages 1 to 6 plated on fibronectin (unless stated otherwise) were used in all the experiments.
FXIII Activation
FXIII was activated by immobilized thrombin as described previously.13 FXIIIa was inactivated by treatment with 2 mmol/L iodoacetamide (I-FXIII), an inhibitor of transglutamination (Sigma) as described previously.13 (Detailed procedure appears in online Methods, available at http://atvb.ahajournals.org.)
Western Blot Analysis of HUVEC Proteins
Detailed procedure appears in online Methods.
Coimmunoprecipitation Experiments
Triton-X-100 extracts of untreated HUVECs or HUVECs treated with either FXIIIa or I-FXIII for 16 hours at 37°C were incubated with the appropriate primary antibody (goat anti-human VEGFR-2 [R&D Systems, Minneapolis, Minn] or sheep anti-human ?3 [Affinity Biological Inc]), or anti-v?3 complex LM609 (Chemicon, Temecula, Calif), followed by addition of protein A-Sepharose beads (Pharmacia). The beads were washed 3x with 1% Triton X-100 lysis buffer and boiled in SDS-PAGE sample buffer for 3 minutes. The precipitated proteins were resolved by 4% to 12% SDS-PAGE and transferred to a nylon membrane as described above. In some experiments, HUVECs were preincubated with 500 nM VEGFR-2 kinase inhibitor (Z)-5-bromo-3 (4,5,7-tetrahydro-1H-indol-2-ylmethylene)1,3-dihydroindol-2-one (Calbiochem; Cat. No. 676485) for 30 minutes at 37°C before the addition of FXIIIa. This compound is a membrane-permeable tyrosine kinase inhibitor with IC50=70 nM for VEGFR-2.15 Blots containing the proteins immunoprecipitated with anti-VEGFR-2 antibody were probed with mouse anti-?3 (AP3, GTI) or mouse anti-PY (4G10; Upstate Biotechnology, Lake Placid, NY) antibodies. Blots containing the proteins immunoprecipitated with the anti-?3 antibody were probed with either anti-VEGFR-2, anti-PY, anti--glutamyl--lysine isopeptide bond (clone 814-MAM, CovalAb, France), or anti-?3 (either sheep polyclonal or mouse monoclonal) antibodies. The signals were detected using the appropriate horseradish peroxidase-conjugated antibodies followed by enhanced chemiluminescence. The expression of VEGFR-2 on HUVECs was tested and found to be the same throughout the 1 to 6 passages used in these experiments.
HUVEC Migration and Proliferation
HUVEC migration was analyzed by a modified Boyden chamber assay13; proliferation was assessed by measuring 3[H]-Thymidine incorporation.13 Detailed migration and proliferation procedures are provided in online Methods.
Real-Time PCR
Total RNA extracted from one 75-cm2 flask of either untreated HUVECs or HUVECs treated with 50 μg/mL FXIIIa for 16 hours at 37°C was isolated using the Qiagen RNA isolation kit (Qiagen). RNA was then reverse transcribed into cDNA by MMLV reverse transcriptase (Promega). cDNA samples were amplified using specific primers for either c-Jun (forward: 5'-agtccagcaacgggcacatc; reverse: 5'-ctcctgctcatctgtcacgttct; accession number: JO4111.1), Egr-1 (forward: 5'-cgtcctcagcaccatctc; reverse: 5'-ggctcagggaaaatgtcag;accession number: NM_001964), or GAPDH (forward: cagcctcaagatcatcagca; reverse: gtcttctgggtggcagtgat; accession number: M33197) and the Quanti Test Sybr Green RT-PCR kit (Qiagen). Amplification was monitored by real-time PCR analysis using the ABI/Prism 7700 Sequence Detector System (Applied Biosystems).
Electrophoretic Mobility-Shift Assay Experiments
Nuclear proteins were prepared as described.16 Protein concentrations were measured using Bradford analysis. The oligonucleotides used for electrophoretic mobility-shift assay (EMSA) experiments were: probe 1 forward: 5' -gctgcgtgggcgggcaccgact-3'; probe 1 reverse: 5'-agtcggtgcccgcccacgcagc-3'; probe 2 forward: 5'-ccgacccccgcccccttc-3'; probe 2 reverse: 5'- gaagggggcgggggtcgg-3'. Each oligonucleotide (10 pmol) was end-labeled with 32 [P]-ATP using polynucleotide kinase (New England Biolabs Inc). The radiolabeled oligonucleotides were purified from unincorporated 32 [P]-ATP using G-50 Spin columns (New England Biolabs Inc), mixed, and incubated for 15 minutes at room temperature to allow annealing. Ten μg of HUVEC nuclear proteins prepared as described above were mixed with 1 μg of poly-(dI-dC) and 1 μL (50 000 cpm) of the double-stranded oligonucleotides in binding buffer (20 mmol/L HEPES, pH 7.5, 1 mmol/L EDTA, 5 mmol/L MgCl2, 50 mmol/L KCl, 1 mmol/L DTT, and 10% glycerol). For supershift experiments, the nuclear extracts were preincubated with either anti-Egr-1 (rabbit anti-human Egr-1, Santa Cruz Biotechnology Inc, Santa Cruz, Calif), or anti-WT-1 antibodies (mouse anti-human WT-1, Santa Cruz Biotechnology Inc, Santa Cruz, Calif) at a ratio of 10 μg IgG/10 μg of total nuclear protein for 30 minutes at room temperature before the addition of the radiolabeled probe. To test the specificity of the formed complexes, competition assays using a 100-fold excess of cold double-stranded oligonucleotides were performed. The DNA-protein complexes were resolved on 6% nondenaturing polyacrylamide gels and detected by autoradiography.
Analysis of v?3 Activation
v?3 activation was analyzed by flow cytometry using WOW-1 antibody (kindly provided by Dr S. Shattil, Scripps Research Institute), which recognizes the v?3 integrin in its active conformation. Detailed procedure appears in online Methods.
Statistical Analysis
Statistical analysis was performed by unpaired Student t test. Results are expressed as mean±SD. P<0.05 was considered significant.
Results
Tyrosine Phosphorylation of HUVEC Proteins Induced by FXIIIa
Triton X-100 soluble lysates of control HUVECs or HUVECs incubated in the presence of either FXIIIa or I-FXIII were tested for tyrosine phosphorylation by Western blot (WB) analysis. As shown in Figure 1A, FXIIIa, but not I-FXIII, stimulated tyrosine phosphorylation of 2 proteins with
Figure 1. Tyrosine phosphorylation of HUVEC proteins after FXIIIa treatment. The representative picture of 3 separate experiments is shown. A, WB analysis with anti-PY antibody of Triton X-100 soluble extracts of either untreated HUVEC or HUVEC treated with FXIIIa or I-FXIII. The numbers above each lane indicate FXIIIa concentration in μg/mL. B, Triton X-100 soluble extracts of either control HUVECs (C) or HUVECs treated with 10 ng/mL VEGF-A, 50 μg/mL FXIIIa, 50 μg/mL FXIIIa after immunodepletion (D), and VEGF-A after preincubation with 500 nM protein tyrosine kinase inhibitor (VEGF-A+PTKI). The samples were immunoprecipitated (IP) with anti-VEGFR-2, followed by WB analysis with either anti-VEGFR-2 or anti-PY antibodies. The 230-kDa band denotes VEGFR-2.
To identify the candidate protein undergoing tyrosine-phosphorylation in response to FXIIIa, the blot shown in Figure 1A was stripped and incubated with VEGFR-2 antibody, which bound to the 230-kDa tyrosine-phosphorylated band (not shown).
Tyrosine phosphorylation of VEGFR-2 was further verified by immunoprecipitation of VEGFR-2 followed by WB analysis with anti-PY antibody (Figure 1B). VEGFR-2 was tyrosine phosphorylated in FXIIIa-treated and in VEGF-A-treated samples. VEGF-A- or FXIIIa-induced VEGFR-2 phosphorylation was abolished by a specific VEGFR-2 tyrosine kinase inhibitor PTKI (an inhibitor of VEGFR-2) or depletion of the FXIIIa preparation by passage over an immobilized anti-FXIII antibody (D), respectively (Figure 1B).
The Role of FXIIIa Crosslinkng in v?3–VEGFR-2 Interaction
We tested whether FXIIIa affects the v?3 and VEGFR-2 interaction, because this interaction was previously shown to play an important role in VEGFR-2–induced signaling.9,10 Triton X-100 lysates of control and FXIIIa or I-FXIII-treated HUVECs were immunoprecipitated with anti-?3 antibody followed by WB analysis using anti-VEGFR-2 antibody (Figure 2A). A high-molecular-weight complex recognized by the 2 antibodies was observed at the interface between stacking and running gels in the lanes containing FXIIIa-treated HUVECs. This complex was absent in the cells treated with I-FXIII. In addition, a 230-kDa band corresponding to the mature form of VEGFR-2 was observed in all the lanes, with an increase in intensity in a FXIIIa dose-dependent manner indicating VEGFR-2/?3 complex formation without crosslinking. To determine whether the VEGFR-2/?3 complex undergoes phosphorylation, this blot was stripped and reprobed with anti-PY antibody. WB with anti-PY revealed that VEGFR-2 in both the high-molecular-weight and the noncovalent VEGFR-2/?3 complexes underwent phosphorylation (Figure 2A). To prove that VEGFR-2 and ?3 within the high-molecular-weight complex are crosslinked by FXIIIa, the blot containing immunoprecipitated VEGFR-2 and ?3 proteins was subjected to WB analysis using an antibody recognizing the -glutamyl--lysine isopeptide bonds. As shown in Figure 2A, this antibody reacted with VEGFR-2/?3 high-molecular-weight complex present only in FXIIIa-treated samples. Immunoprecipitation of HUVECs with anti-VEGFR-2 and WB with anti-?3 detected the high-molecular-weight complex only in the lane corresponding to FXIIIa-treated HUVECs (Figure 2B), suggesting that VEGFR-2 was crosslinked to ?3. Non-crosslinked ?3 that coimmunoprecipitated with VEGFR-2 was observed in control as well as in FXIIIa or I-FXIII samples (Figure 2B). Assessment of the extent of crosslinking (see online Methods) revealed that 12.5% of total VEGFR-2 and 8% of total v?3 extracted from HUVECs underwent crosslinking.
Figure 2. Coimmunoprecipitation of HUVEC proteins with anti-VEGFR-2 and anti-?3 antibodies. The representative picture of 3 separate experiments is shown. A through C, Triton X-100 soluble proteins of either untreated or HUVECs treated with FXIIIa or I-FXIII. The numbers above each lane indicate concentration in μg/mL. The samples were subjected to immunoprecipitation (IP) with either anti-?3 (A and D), anti-VEGFR-2 (B), or LM609 (C) antibodies. The immunoprecipitated proteins were analyzed by WB using anti-VEGFR-2, anti-PY, anti--glutamyl- -lysine isopeptide bond antibodies, or anti-?3 (A), anti-?3, or anti-VEGFR-2 (B through D).
To determine whether ?3 is in complex with v chain while interacting with VEGFR-2, HUVEC proteins were immunoprecipitated with LM609 antibody recognizing the v?3 integrin complex, followed by WB analysis with anti-VEGFR-2 (Figure 2C). A high-molecular-weight crosslinked VEGFR-2/v?3 complex was observed only in FXIIIa-treated cells. The v?3 integrin within the complex was not activated as demonstrated by lack of WOW-1 antibody binding to FXIIIa-treated cells (Figure I, available online at http://atvb.ahajournals.org).
To evaluate the role of the tyrosine kinase activity of VEGFR-2 in the VEGFR-2/?3 interaction, we performed a coimmunoprecipitation experiment using lysates of control and FXIIIa-treated HUVECs preincubated with a specific VEGFR-2 tyrosine kinase inhibitor (PTKI; Figure 2D). This inhibitor blocks VEGF-A-induced VEGFR-2 phosphorylation as shown by Sun et al15 and by us in Figure 1B. As demonstrated in Figure 2D, this inhibitor markedly decreased VEGFR-2/?3 complex formation in nontreated or FXIIIa-treated cells and completely abolished crosslinking of the VEGFR-2/?3 complex induced by FXIIIa. FXIIIa mediated VEGFR-2/v?3 crosslinking, and VEGFR-2 phosphorylation was also observed in dermal microvascular ECs (data not shown).
To find out whether VEGFR-2/?3 crosslinking follows kinetics similar to that of downstream signaling, time-course experiments were performed and are presented in Figure 3. Weak bands corresponding to phosphorylated VEGFR-2 (Figure 3A) and VEGFR-2/?3 crosslinked complex (Figure 3B) were observed only after 1 hour of incubation with FXIIIa. Both bands increased in intensity in a time-dependent manner, reaching a plateau after 8 hours of exposure to FXIIIa. Similar kinetics was observed for downstream activation of Akt (Figure 3C). In contrast, maximal activation of ERK was observed already after 2 hours of incubation (Figure 3C). It is possible that partial activation of VEGFR-2 is sufficient to trigger a full activation of the MEK/ERK system. Treatment of HUVECs with VEGF resulted in more rapid maximal activation of both ERK and Akt (after only 10 minutes of incubation; Figure 3C). Densitometric assessment of the time course experiments is shown in Figure II (available online at http://atvb.ahajournals.org).
Figure 3. Time course of FXIIIa-induced events. The representative picture of 2 separate experiments is shown. In A and B, Triton X-100 extracts of HUVECs treated with 50 μg/mL FXIIIa for various time periods were immunoprecipitated (IP) and analyzed by WB. A, Time course of VEGFR-2 phosphorylation: IP with anti-VEGFR-2; WB analysis using either anti-PY or anti-VEGFR-2 antibodies. B, Time course of crosslinked VEGFR-2/?3 complex formation: IP with anti-?3; WB analysis using either anti--glutamyl--lysine or anti-?3 antibodies. C, Total Triton X-100 extracts of HUVECs treated with either 50 μg/mL FXIIIa or 10 ng/mL VEGF-A for various time periods were analyzed with either anti-phospho-ERK, anti-ERK, anti-phospho-Akt, or anti-Akt antibodies. The numbers above the bars indicate incubation periods in minutes and hours for VEGF-A and FXIIIa, respectively.
To provide functional evidence that the pathways studied are responsible for the angiogenic action of F-XIII, we performed migration and proliferation assays in the presence of FXIIIa, VEGF, or both. As shown in Figure 4, FXIIIa significantly increased EC migration and proliferation, and both processes were inhibited by VEGFR-2 tyrosine kinase inhibitor PTKI. Similar results were observed with VEGF. In addition, these experiments showed that FXIIIa had no additional effect on VEGF-A-induced migration and proliferation.
Figure 4. Effect of FXIIIa/VEGF on HUVEC proliferation and migration. [3H]-thymidine incorporation and migration of HUVECs induced by either 10 ng/mL VEGF-A, or 50 μg/mL FXIIIa, or a combination of both in the absence or presence of 500 nM PTKI. Data for proliferation and migration are presented as percent of control and number of migrating cells per field, respectively (mean±SD of 3 separate experiments performed in quadruplicates). *P<0.05.
Effect of FXIIIa on Gene Expression of Transcription Factors and Signaling Molecules
After treatment of HUVECs with FXIIIa, suspected upregulation of cyclin D1 and Egr-1 was observed by microarray analysis (data not shown). However, validation of the upregulation by real-time PCR and WB disclosed that only Egr-1 gene and protein were upregulated. A 7.5±2.2-fold increase in mRNA (P<0.05) and 3.7±1.5-fold increase in protein level (by densitometry; P<0.05) was observed (Figure III, available online at http://atvb.ahajournals.org).
Because we have previously shown that FXIIIa proangiogenic effect is associated with downregulation of TSP-1,13 and Dejong et al have demonstrated that downregulation of TSP-1 was associated with upregulation of c-Jun,16 we examined by real-time PCR whether FXIIIa affects c-Jun expression. FXIIIa induced a 1.65±0.15-fold increase in c-Jun mRNA as shown by real-time PCR (Figure III, P<0.05). Accordingly, a 2.1±0.3-fold increase (by densitometry; P<0.05) in the amount of c-Jun protein was observed by WB analysis (representative WB in Figure IIIB). Thus, FXIIIa effect is associated with upregulation of c-Jun.
Effect of FXIIIa on Binding of Nuclear Proteins to the TSP-1 Promoter
c-Jun was previously shown to downregulate TSP-1 by inducing WT-1 binding to the TSP-1 promoter.16 We examined whether WT-1 binding to TSP-1 promoter is affected by FXIIIa by an EMSA experiment using radiolabeled oligonucleotide probes containing a GC-rich TSP-1 promoter sequence (probe 1) as outlined in Methods. As shown in Figure 5A, the treatment of HUVECs with FXIIIa increased complex formation between the probe and one of the nuclear proteins, which was significantly reduced by addition of an anti-WT-1 antibody. Because the amount of the WT-1 protein in the nuclear extracts of HUVECs was not affected by FXIIIa (Figure 5B), the downregulation of TSP-1 associated with FXIIIa may be caused by modulation of WT-1 binding to TSP-1 rather than WT-1 protein synthesis. Similarly, an EMSA experiment using radiolabeled oligonucleotide probes 1 and 2 (Methods) was undertaken to analyze binding of Egr-1 to TSP-1 promoter after FXIIIa treatment. Figure 5A shows that anti-Egr-1 antibody had no effect on FXIIIa-induced complex formation between probe 1 and nuclear proteins. Binding of probe 2 to nuclear proteins was not affected by FXIIIa although it was inhibited by anti-Egr-1 antibody. WB analysis of FXIIIa-treated HUVECs with anti-Egr-1 antibody revealed that Egr-1 was markedly upregulated (Figure 5B). Taken together, these data suggest that although Egr-1 protein is upregulated after FXIIIa treatment, its binding to the TSP-1 gene is not visibly regulated by FXIIIa.
Figure 5. EMSA of FXIIIa-induced binding of TSP-1 promoter to WT-1. The representative picture of 4 separate experiments is shown. A, Ten μg of nuclear proteins extracted from untreated HUVECs (–) or HUVECs treated with 50 μg/mL of FXIIIa (+) were subjected to EMSA analysis using radiolabeled probe 1 or 2 (50 000 cpm) from the TSP-1 promoter as described in Materials and Methods. Anti-WT-1 or anti-Egr-1 antibodies were used to identify the DNA-protein complexes (denoted by arrows). NSB indicates nonspecific binding. B, Fifty μg of nuclear proteins extracted from untreated HUVECs (–) or HUVECs treated with 50 μg/mL of FXIIIa (+) were subjected to WB analysis using either anti-WT-1 or anti-Egr-1 antibody.
Discussion
In this study, we have explored the molecular mechanisms that underlie the proangiogenic activity of FXIII reported previously by us.13 Our results show that FXIIIa crosslinks ?3 integrin to VEGFR-2 as well as enhances the noncovalent complex formation between the ?3 integrin and VEGFR-2. Moreover, FXIIIa induces tyrosine phosphorylation of VEGFR-2 in both the crosslinked and noncovalent ?3-VEGFR-2 complexes. The formation of the complexes appears to be dependent on the intact catalytic site of VEGFR-2, because it is inhibited by a specific VEGFR-2 tyrosine kinase inhibitor (PTKI). It is, however, unclear why VEGFR-2/?3 noncovalent and crosslinked complexes depend on the intact catalytic site of VEGFR-2. One of the possibilities is that the inhibitor binding induces a conformational change in the receptor, thereby interfering with VEGFR-2/v ?3 interaction.
The effects of FXIIIa on both the v?3 integrin-VEGFR-2 association and tyrosine phosphorylation of VEGFR-2 were found to be absolutely dependent on its transglutaminase activity, as indicated by the lack of effect of iodoacetamide-inhibited FXIIIa. Taken together, these data suggest that the proangiogenic effect of FXIIIa, in part, involves crosslinking of the 2 receptors v?3 integrin and VEGFR-2 followed by activation of VEGFR-2 similarly to that elicited by VEGF-A.2 The faster kinetics of VEGF-induced as compared with FXIIIa-induced phosphorylation of signaling proteins may suggest that VEGF is the primary proangiogenic factor, the signals of which are immediately/rapidly transduced, whereas FXIII may be essential for maintaining the angiogenic process and therefore acts at a much slower pace. This is further supported by the incapability of FXIIIa to enhance maximal VEGF-A-induced migration and proliferation of HUVECs.
The tyrosine-phosphorylated VEGFR-2 was shown to mediate activation of MAP kinase, PI-3 kinase, c-GMP-dependent kinase, Jun-N-terminal kinase, and focal adhesion kinase, all of which are involved in regulation of the angiogenic response of ECs to VEGF-A.17 Our data indicate that, similar to VEGF-A, FXIIIa induces tyrosine-phosphorylation of VEGFR-2 in HUVECs; however, it does not increase the amounts of VEGF-A or VEGFR-2 as shown by us previously.13 It seems that VEGFR-2 phosphorylation is the major and vital step in the initiation of FXIIIa-induced angiogenesis. Inhibition of FXIII-mediated proliferation and migration by PTKI further confirm that VEGFR-2 phosphorylation is the event responsible for the angiogenic action of FXIII.
Activated tyrosine-phosphorylated VEGFR-2 binds and activates PLC, which in turn activates PKC.2 Activated PKC induces activation of MAPK,17 which is in accordance with our results showing that FXIIIa-mediated VEGFR-2/v?3 crosslinking with subsequent VEGFR-2 phosphorylation is associated with MAPK activation. Activation of VEGFR-2 by VEGF-A induces upregulation of Egr-1 in a PKC- and MAPK-dependent manner,4,18 which is also in agreement with our data showing that the FXIIIa-induced proangiogenic effect is associated with upregulation of Egr-1.
Egr-1 is a zinc-finger transcription factor involved in cell proliferation and differentiation and regulation of multiple genes implicated in vasculogenesis and angiogenesis. Upregulation of Egr-1 demonstrated in this study indicates that Egr-1 may serve as a direct mediator of FXIIIa-induced EC proliferation rather than mediating TSP-1 suppression, because its binding to the TSP-1 promoter is not affected by FXIIIa. That Egr-1 participates in angiogenesis and in wound healing18,19 supports its involvement in FXIII-induced wound healing12 owing to the ability of FXIIIa to upregulate Egr-1.
We have recently shown that the proangiogenic activity of FXIIIa is associated with downregulation of TSP-1 in in vitro and in vivo models.13 Several lines of evidence indicate that downregulation of TSP-1 is linked to overexpression of the transcription factor c-Jun, a positive regulator of cell proliferation.16,20 In rat fibroblasts, the negative regulation of TSP-1 transcription by c-Jun was shown to be mediated by transcription factor WT-1 rather than by direct repression by c-Jun.16 In this study, c-Jun was found to be upregulated by FXIIIa at both mRNA and protein levels. In addition, FXIIIa induced an increase in binding of WT-1 but not Egr-1 to the TSP-1 promoter. Based on these results and the work of Dejong and colleagues,16 it seems most likely that the downregulation of TSP-1 by FXIIIa stems from indirect suppression of TSP-1 transcription by c-Jun through WT-1.
We and others have previously reported that FXIIIa binds to the v?3 integrin expressed on the surface of ECs.21,22 In this study, FXIIIa-v?3 interaction on HUVECs was not associated with v?3 activation. Because FXIIIa does not contain the conventional integrin recognition and activation sequence RGD, its binding to v?3 may be mediated by the LDV sequence without activating the integrin. LDV was previously shown to mediate the interaction of fibronectin with 4?1.23
Hemostasis and angiogenesis are interrelated. Proteins generated by the hemostatic system coordinate the spatial localization and stabilization of ECs followed by growth and repair of a damaged blood vessel.24 After clot stabilization, angiogenesis is regulated through proteins secreted by platelets and cryptic fragments generated from coagulation and fibrinolytic system.24 The well-characterized proangiogenic coagulation proteins are thrombin, FVII, and tissue factor. The antiangiogenic proteins are antithrombin, domain 5 of the high-molecular-weight kininogen, prothrombin fragments 1 and 2, u-PA, plasminogen activator inhibitor-1, and angiostatin.24 In this study, we focused on the regulation of angiogenesis mediated by clot stabilization protein-FXIII. Based on our results from this and previous studies, we propose the following model for the proangiogenic effect of FXIIIa (Figure IV, available online at http://atvb.ahajournals.org): FXIIIa binding to the EC v?3 integrin is followed by enhancement of the v?3–VEGFR-2 interaction resulting in tyrosine phosphorylation and activation of VEGFR-2. Activated VEGFR-2 then initiates a signaling cascade leading to upregulation of Egr-1 and c-Jun, which may then be responsible for the enhanced cell proliferation and survival induced by FXIIIa. In addition, downregulation of TSP-1 by c-Jun through WT-1 may further contribute to the proangiogenic activities exhibited by FXIIIa. This FXIIIa-mediated proangiogenic effect may finally clarify the role of FXIII in vascular remodeling and tissue repair.
Received July 27, 2004; accepted December 3, 2004.
References
Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med. 2003; 9: 669–676.
Takahashi T, Shibuya M. The 230 kDa mature form of KDR/Flk-1 (VEGF receptor-2) activates the PLC-gamma pathway and partially induces mitotic signals in NIH3T3 fibroblasts. Oncogene. 1997; 14: 2079–2089.
Matsumoto T, Claesson-Welsh L. VEGF receptor signal transduction. Sci STKE. 2001; 2001: RE21. Review.
Mechtcheriakova D, Schabbauer G, Lucerna M, Clauss M, De Martin R, Binder BR, Hofer E. Specificity, diversity, and convergence in VEGF and TNF-alpha signaling events leading to tissue factor up-regulation via EGR-1 in endothelial cells. FASEB J. 2001; 15: 230–242.
Varner JA, Brooks PC, Cheresh DA. REVIEW: the integrin alpha V beta 3: angiogenesis and apoptosis. Cell Adhes Commun. 1995; 3: 367–374.
Eliceiri BP, Cheresh DA. Role of alpha v integrins during angiogenesis. Cancer J. 2000; 6 (suppl 3): S245–S249.
Vuori K, Ruoslahti E. Association of insulin receptor substrate-1 with integrins. Science. 1994; 266: 1576–1578.
Borges E, Jan Y, Ruoslahti E. Platelet-derived growth factor receptor beta and vascular endothelial growth factor receptor 2 bind to the beta 3 integrin through its extracellular domain. J Biol Chem. 2000; 275: 39867–39873.
Soldi R, Mitola S, Strasly M, Defilippi P, Tarone G, Bussolino F. Role of alphavbeta3 integrin in the activation of vascular endothelial growth factor receptor-2. EMBO J. 1999; 18: 882–892.
Masson-Gadais B, Houle F, Laferriere J, Huot J. Integrin alphavbeta3, requirement for VEGFR2-mediated activation of SAPK2/p38 and for Hsp90-dependent phosphorylation of focal adhesion kinase in endothelial cells activated by VEGF. Cell Stress Chaperones. 2003; 8: 37–52.
Reynolds LE, Wyder L, Lively JC, Taverna D, Robinson SD, Huang X, Sheppard D, Hynes RO, Hodivala-Dilke KM. Enhanced pathological angiogenesis in mice lacking beta3 integrin or beta3 and beta5 integrins. Nat Med. 2002; 8: 27–34.
Muszbek L, Adany R, Mikkola H. Novel aspects of blood coagulation factor XIII. I. Structure, distribution, activation, and function. Crit Rev Clin Lab Sci. 1996; 33: 357–421.
Dardik R, Solomon A, Loscalzo J, Eskaraev R, Bialik A, Goldberg I, Schiby G, Inbal A. Novel proangiogenic effect of factor XIII associated with suppression of thrombospondin 1 expression. Arterioscler Thromb Vasc Biol. 2003; 23: 1472–1477.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest. 1973; 52: 2745–2756.
Sun L, Tran N, Liang C, Hubbard S, Tang F, Lipson K, Schreck R, Zhou Y, McMahon G, Tang C. Identification of substituted 3--1,3-dihydroindol-2-ones as growth factor receptor inhibitors for VEGF-R2 (Flk-1/KDR), FGF-R1, and PDGF-beta tyrosine kinases. J Med Chem. 2000; 43: 2655–2663.
Dejong V, Degeorges A, Filleur S, Ait-Si-Ali S, Mettouchi A, Bornstein P, Binetruy B, Cabon F. The Wilms’ tumor gene product represses the transcription of thrombospondin 1 in response to overexpression of c-Jun. Oncogene. 1999; 18: 3143–3151.
Petrova TV, Makinen T, Alitalo K. Signaling via vascular endothelial growth factor receptors. Exp Cell Res. 1999; 253: 117–130.
Lucerna M, Mechtcheriakova D, Kadl A, Schabbauer G, Schafer R, Gruber F, Koshelnick Y, Muller HD, Issbrucker K, Clauss M, Binder BR, Hofer E. NAB2, a corepressor of EGR-1, inhibits vascular endothelial growth factor-mediated gene induction and angiogenic responses of endothelial cells. J Biol Chem. 2003; 278: 11433–11440.
Bryant M, Drew GM, Houston P, Hissey P, Campbell CJ, Braddock M. Tissue repair with a therapeutic transcription factor. Hum Gene Ther. 2000; 11: 2143–2158.
Kim SA, Um SJ, Kang JH, Hong KJ. Expression of thrombospondin-1 in human hepatocarcinoma cell lines and its regulation by transcription factor Jun/AP-1. Mol Cell Biochem. 2001; 216: 21–29.
Dardik R, Shenkman B, Tamarin I, Eskaraev R, Harsfalvi J, Varon D, Inbal A. Factor XIII mediates adhesion of platelets to endothelial cells through alpha(v)beta(3) and glycoprotein IIb/IIIa integrins. Thromb Res. 2002; 105: 317–323.
Dallabrida SM, Falls LA, Farrell DH. Factor XIIIa supports microvascular endothelial cell adhesion and inhibits capillary tube formation in fibrin. Blood. 2000; 95: 2586–2592.
Mould AP, Wheldon LA, Komoriya A, Wayner EA, Yamada KM, Humphries MJ. Affinity chromatographic isolation of the melanoma adhesion receptor for the IIICS region of fibronectin and its identification as the integrin alpha 4 beta 1. J Biol Chem. 1990; 265: 4020–4024.
Browder T, Folkman J, Pirie-Shephered S. The hemostatic system as a regulator of angiogenesis. J Biol Chem. 2000; 275: 1521–1524.(Rima Dardik; Joseph Losca)