AP-1 (Fra-1/c-Jun)-mediated Induction of Expression of Matrix Metalloproteinase-2 Is Required for 15(S)-Hydroxyeicosatetraenoic Acid-induced Angiogenesis*

To understand the involvement of matrix metalloproteinases (MMPs) in 15(S)-hydroxyeicosatetraenoic acid (15(S)-HETE)-induced angiogenesis, we have studied the role of MMP-2. 15(S)-HETE induced MMP-2 expression and activity in a time-dependent manner in human dermal microvascular endothelial cells (HDMVECs). Inhibition of MMP-2 activity or depletion of its levels attenuated 15(S)-HETE-induced HDMVEC migration, tube formation, and Matrigel plug angiogenesis. 15(S)-HETE also induced Fra-1 and c-Jun expression in a Rac1-MEK1-JNK1-dependent manner. In addition, 15(S)-HETE-induced MMP-2 expression and activity were mediated by Rac1-MEK1-JNK1-dependent activation of AP-1 (Fra-1/c-Jun). Cloning and site-directed mutagenesis of MMP-2 promoter revealed that AP-1 site proximal to the transcriptional start site is required for 15(S)-HETE-induced MMP-2 expression, and Fra-1 and c-Jun are the essential components of AP-1 that bind to MMP-2 promoter in response to 15(S)-HETE. Hind limb ischemia led to an increase in MEK1 and JNK1 activation and Fra-1, c-Jun, and MMP-2 expression resulting in enhanced neovascularization and recovery of blood perfusion in wild-type mice as compared with 12/15-Lox−/− mice. Together, these results provide the first direct evidence for a role of 12/15-Lox-12/15(S)-HETE axis in the regulation of ischemia-induced angiogenesis.

Angiogenesis has been implicated in the pathogenesis of cancer, diabetic retinopathy, and atherosclerosis (1). On the other hand, a therapeutic role for angiogenesis in the treatment of coronary artery disease and wound healing has been identified (2). In the initiation of new capillaries, endothelial cells of existing blood vessels degrade the underlying basement membrane and invade into the stroma of the neighboring tissue (3,4). The endothelial cell invasion and migration require the cooperative activity of the plasminogen activator system and the matrix metalloproteinases (MMPs) 2 (3). MMPs were initially discov-ered in 1962 as collagenolytic activity degrading extracellular matrix protein during tadpole tail resorption (5). Since then, more than 23 members of the MMP family of matrix-degrading enzymes have been identified, characterized, and sub-classified based on their ability to degrade a specific extracellular matrix protein (6). Besides their role in extracellular matrix protein degradation, MMPs have been reported to play a role in the activation of cell surface receptors (7). MMPs not only have been shown to play a role in a number of physiological processes such as embryogenesis (8) and angiogenesis (9,10), but also have been reported to contribute to the pathogenesis of various diseases, including tumor metastasis (11) and inflammation (12). Among many MMPs, MMP-2 and -9 (also known as gelatinase A and gelatinase B, respectively) have been identified as important players in several cardiovascular diseases, including atherosclerosis, ischemic heart disease, heart failure, aneurysm, and stroke (13)(14)(15)(16)(17). MMP-2 is expressed ubiquitously in various cell types, including cardiomyocytes, endothelial cells, fibroblasts, and vascular smooth muscle cells, and plays a crucial role in the regulation of angiogenesis (18 -20). Despite the importance of MMPs in cardiovascular development and diseases (21)(22)(23), relatively very little is known in regard to the role of these enzymes in eicosanoid-mediated vascular wall remodeling.
It is well established that lipoxygenases, particularly 12/15lipoxygenase , play an important role in the pathogenesis of various vascular diseases and cancers (24 -28). One of the initial mechanisms by which 12/15-Lox is implicated in atherosclerosis is its capacity to oxidize low density lipoprotein particles (24). However, many studies also provided clues for additional mechanisms of 12/15-Lox involvement in the pathogenesis of atherosclerosis, restenosis, and cancer. Of these, its role in the regulation of cell growth, migration, and apoptosis have received much attention (29 -32). Besides, one of the compelling evidences for the role of 12/15-Lox in atherosclerosis comes from the findings that atherosclerotic arteries produce 15-HETE as a major eicosanoid (33), and its levels are increased in rabbit atherosclerotic arteries (34). Furthermore, the work from our laboratory provided clues for the role of 15(S)-HETE in both vascular diseases and cancers. Of note are the observa-tions that 15(S)-HETE induces angiogenesis and that this phenomenon requires the autocrine production of angiogenic factors such as vascular endothelial growth factor and interleukin-8 (35,36). In addition, 15(S)-HETE was found to stimulate various signaling molecules, including protein kinase B (Akt), Src, activating transcription factor-2, signal transducer and activator of transcription-3/5, and all of these molecules appear to be involved in the mediation of angiogenesis in response to this stimulus (35)(36)(37)(38). During the course of these studies, we also observed that 15(S)-HETE activates Rac1, MEK1, and JNK1 in human retinal microvascular endothelial cells facilitating their migration and tube-like structure formation (37,39). Because MMPs via their capacity to degrade extracellular matrix proteins play a pivotal role in cell migration, one of the major cellular events underlying angiogenesis, we asked the question whether 15(S)-HETE stimulates MMPs in microvascular endothelial cells and thereby mediates its angiogenic response. Our results indicate that 15(S)-HETE induces the expression of MMP-2 via involving AP-1 (Fra-1/c-Jun) and that MMP-2 activity is required for 15(S)-HETE-induced angiogenesis. In support of this conclusion, we also observed that 12/15-Lox Ϫ/Ϫ mice exhibit decreased MEK1 and JNK1 activation and c-Jun, Fra-1, and MMP-2 expression, thereby reduced angiogenic response following hind limb ischemia as compared with WT mice.
Cell Culture-HDMVECs were bought from Cascade Biologics (Portland, OR) and were grown in medium 131 containing microvascular growth supplements, 10 g/ml gentamycin, and 0.25 g/ml amphotericin B. Cultures were maintained at 37°C in a humidified 95% air and 5% CO 2 atmosphere. Cells were quiesced by incubating in medium 131 without microvascular growth supplements for 24 h and used to perform the experiments unless otherwise indicated.
Transfections and Transductions-HDMVECs were transfected with specific siRNA molecules at a final concentration of 100 nM using DharmaFECT 2 transfection reagent according to the manufacturer's instructions. In the case of adenoviral vectors, cells were transduced with adenovirus harboring GFP or target molecule at 40 m.o.i. overnight in complete medium. After transfections or transductions, cells were quiesced for 24 h and used as required.
Gelatin Zymography-Activity of MMP-2 was determined using a gelatin zymography assay. Conditioned medium or tissue extracts containing an equal amount of protein in an equal volume were mixed with an equal volume of non-reducing sample buffer (2% SDS, 125 mM Tris-HCl, pH 6.8, 10% glycerol, and 0.001% bromphenol blue) and electrophoresed onto 10% polyacrylamide gels containing 0.1% gelatin. After electrophoresis, the gels were incubated for 1 h at 37°C in 2.5% Triton X-100 solution, washed in 50 mM Tris-HCl buffer (pH 7.5) for 30 min, and incubated at 37°C for overnight in 50 mM Tris-HCl buffer (pH 7.5) containing 200 mM NaCl, 10 mM CaCl 2 , and 0.05% Brij-35. The gels were stained with 0.05% Coomassie Brilliant Blue G-250 (Bio-Rad) solution containing 2% acetic acid and 45% methanol and then destained with 40% methanol and 10% acetic acid. Gelatinolytic activity was detected as unstained band, and the band intensities were quantified using National Institutes of Health ImageJ.
Quantitative Real-time Reverse Transcriptase-PCR-After appropriate treatments, total cellular RNA was isolated from HDMVECs or tissues by using a RiboPure TM kit (Ambion, Austin, TX) or an RNeasy Fibrous Tissue Mini Kit (Qiagen, Valencia, CA), respectively, as per the manufacturers' guidelines. Reverse transcription was carried out with a High Capacity cDNA reverse transcription kit for reverse transcription (RT)-PCR based on the supplier's protocol (Applied Biosystems, Foster City, CA). The cDNA was then used as a template for PCR amplification using TaqMan Gene Expression Assays for human MMP-2 (Hs01548724_m1), mouse MMP-2 (Mm00439506_m1), human ␤-actin (Hs99999903_m1), and mouse ␤-actin (Mm02619580_g1). The amplification was carried out on 7300 Real-Time PCR Systems (Applied Biosystems) using the following conditions: 95°C for 10 min followed by 40 cycles at 95°C for 15 s with extension at 60°C for 1 min for both MMP-2 and ␤-actin. The PCR amplification run was examined, using the 7300 real-time PCR system-operated SDS version 1.4 program. The program uses the Delta Rn analysis method (Applied Biosystems).
Cell Migration Assay-Cell migration was performed using a modified Boyden chamber method (38). The cell culture inserts containing membranes with 10 mm diameter and 8.0-m pore size (Nalgene Nunc International, Rochester, NY) were placed in a 24-well tissue culture plate (Costar, Corning Inc., Corning, NY). The lower surface of the porous membrane was coated with 70% Matrigel at 4°C overnight and then blocked with 0.1% heat-inactivated bovine serum albumin at 37°C for 1 h. HDMVECs were quiesced for 24 h in medium 131, trypsinized, and neutralized with trypsin-neutralizing solution. Cells were seeded into the upper chamber at 1 ϫ 10 5 cells/well. Vehicle or 15(S)-HETE was added to the lower chamber. Both the upper and lower chambers contained medium 131. When the effect of a pharmacological inhibitor was tested on 15(S)-HETE-induced HDMVEC migration, cells were incubated with the inhibitor for 30 min first and then added to the upper chamber. In the siRNA approach, cells were transfected with scrambled or specific siRNA and quiesced before they were subjected to migration assay. After 8 h of incubation at 37°C, non-migrated cells were removed from the upper side of the membrane with cotton swabs, and the cells on the lower surface of the membrane were fixed in methanol for 15 min. The membrane was then stained with 4Ј,6-diamidino-2-phenylindole in Vectashield mounting medium (Vector Laboratories Inc.) and observed under a diaphot fluorescence microscope with a photometrics CH250 charge-coupled device camera (Nikon, Garden City, NY). Cells were counted in five randomly selected fields per well and presented as the number of migrated cells/field.
Tube Formation Assay-Tube formation assay was performed as described previously (38). Twenty-four well culture plates (Costar, Corning Inc.) were coated with growth factorreduced Matrigel (BD Biosciences) in a total volume of 280 l/well and allowed to solidify for 30 min at 37°C. HDMVECs were trypsinized, neutralized with trypsin-neutralizing solution, and resuspended in medium 131 at 5 ϫ 10 5 cells/ml, and 200 l of this cell suspension was added into each well. Vehicle or 15(S)-HETE, at the indicated concentration, was added to the appropriate well, and the cells were incubated at 37°C for 6 h. When the effect of a pharmacological inhibitor was tested on 15(S)-HETE-induced HDMVEC tube formation, cells were incubated with the inhibitor for 30 min first and then plated. In the siRNA approach, cells were transfected with scrambled or specific siRNA and quiesced before they were subjected to tube formation assay. Tube formation was observed under an inverted microscope (Eclipse TS100, Nikon). Images were captured with a charge-coupled device color camera (KP-D20AU, Hitachi, Japan) attached to the microscope, and tube length was measured using ImageJ.
Western Blot Analysis-After appropriate treatments and rinsing with cold phosphate-buffered saline, HDMVECs were lysed in 500 l of lysis buffer (phosphate-buffered saline, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 100 g/ml phenylmethylsulfonyl fluoride, 100 g/ml aprotinin, 1 g/ml leupeptin, and 1 mM sodium orthovanadate) and scraped into 1.5-ml Eppendorf tubes. Tissues extracts were prepared by homogenization using Tissue Tearor (Biospec Products Inc.) in the above mentioned lysis buffer. After standing on ice for 20 min, the cell and tissue extracts were cleared by centrifugation at 12,000 rpm for 20 min at 4°C. The cell and tissue extracts containing equal amounts of protein were resolved by electrophoresis on 0.1% SDS and 10% polyacrylamide gels. The proteins were transferred electrophoretically to a nitrocellulose membrane (Hybond, Amersham Biosciences). After blocking in 10 mM Tris-HCl buffer, pH 8.0, containing 150 mM NaCl, 0.1% Tween 20, and 5% (w/v) nonfat dry milk, the membrane was treated with appropriate primary antibodies followed by incubation with horseradish peroxidase-conjugated secondary antibodies. The antigen-antibody complexes were detected using chemiluminescence reagent kit (Amersham Biosciences).
Pull-down Assay-An equal amount of protein from control and each treatment was incubated with glutathione S-transferase-p21-activated kinase (Cdc42 and Rac1 interactive binding domain)-conjugated Sepharose CL4B beads for 45 min at 4°C. The beads were collected by centrifugation, washed in lysis buffer and heated in Laemmli sample buffer for 5 min, and the released proteins were resolved on 0.1% SDS-12% PAGE and analyzed by immunoblotting for Rac1 levels using anti-Rac1 antibodies.
Luciferase Assay-HDMVECs were transfected with MMP-2 promoter-luciferase constructs or pGL3 basic vector using Lipofectamine 2000 reagent. After growth synchronization in growth factor free-medium for 24 h, cells were treated with and without 15(S)-HETE (0.1 M) for 8 h. Cells were then washed once with ice-cold phosphate-buffered saline and lysed with 200 l of lysis buffer. The cell extracts were collected into microcentrifuge tubes and centrifuged for 2 min at 12,000 ϫ g at 4°C. The supernatants were collected and assayed for luciferase activity using Luciferase Assay System (Promega) and a single tube luminometer (TD20/20 Turner Designs, Sunnyvale, CA), and the values are expressed as relative luciferase units.
Chromatin Immunoprecipitation Assay-Chromatin immunoprecipitation (ChIP) assay was performed on HDMVECs using ChIP assay kit following the supplier's protocol (Upstate Biotechnology). AP-1 DNA complexes were immunoprecipitated using anti-Fra-1 antibodies or anti-c-Jun antibodies. Preimmune rabbit serum was used as a negative control. The immunoprecipitated DNA was uncross-linked, subjected to proteinase K digestion, and purified using QIAquick columns (Qiagen). The purified DNA was used as a template for PCR amplification using primers (forward, 5Ј-CTTCTCAAACTG-TTCCCTGC-3Ј; reverse, 5Ј-GGAACGCCTGACTTCAGC-3Ј) flanking the putative AP-1 located at Ϫ1263 and Ϫ1270 in human MMP-2 promoter region (accession no. AJ298926). The PCR products were resolved on 1.5% agarose gels and visualized by ethidium bromide staining.
Matrigel Plug Angiogenesis Assay-A Matrigel plug assay was performed essentially as described previously (36). The C57BL/6 and 12/15-LOX Ϫ/Ϫ mice were obtained from the Jackson Laboratory (Bar Harbor, ME). C57BL/6 (WT) mice or 12/15-LOX Ϫ/Ϫ mice (8 weeks old) were lightly anesthetized with ketamine (100 mg/kg) and xylazine (8 mg/kg) and were injected subcutaneously with 0.5 ml of Matrigel, which was premixed with vehicle or 5 M of 15(S)-HETE along the abdominal midline. The injections were made rapidly with a 30-gauge 1/2 needle to ensure the entire content was delivered as a single plug. Wherever the effect of a pharmacological inhibitor was tested on 15(S)-HETE-induced angiogenesis, it was added to the Matrigel prior to injecting into mice. The mice were allowed to recover, and 7 days later, unless otherwise stated, the animals were sacrificed by inhalation of CO 2 ; the Matrigel plugs were harvested from underneath the skin. The plugs were homogenized in 1 ml of deionized H 2 O on ice and cleared by centrifugation at 10,000 rpm for 6 min at 4°C. The supernatant was collected and used in duplicate to measure hemoglobin content with Drabkin's reagent along with hemoglobin standard essentially according to the manufacturer's protocol (Sigma). The absorbance was read at 540 nm in a Microplate reader (Spectra Max 190, Molecular Devices, Sunnyvale, CA). These experi-ments were repeated at least three times with six mice for each group, and the values are expressed as grams hemoglobin/ dl/mg of plug.
Hind Limb Ischemia-Hind limb ischemia was induced in wild-type and 12-LOX Ϫ/Ϫ mice by ligating the left common femoral artery proximal to the origin of the profunda femoris artery (41). Mice were anesthetized with intraperitoneal injection of ketamine (100 mg/kg) and xylazine (8 mg/kg). An incision was made in the left groin of mice, and blunt dissection was performed to identify the left common femoral artery by its pale pink color and pulsatile nature. The common femoral vein and femoral nerve were separated from the artery. Two ligations were performed in the common femoral artery proximal to the origin of the profunda femoris artery using 6-0 nylon sutures. The common femoral artery was then transected between the ligation sites. Immediate blanching was noted in the distal left hind limb after ligation. The incision was sutured in a single layer using the same suture material as for ligation. Mice were noted to be limping and dragging the left hind limb after recovery from anesthesia. After 7 days, mice were anesthetized and placed on a heating pad (ϳ37°C), and the limb blood perfusion was measured by Laser Doppler Imager System (Laser

12/15-Lox-15(S)-HETE-induced Angiogenesis Requires MMP-2
Doppler Perfusion Imager System, moorLDI-2HR, Moor Instruments, Wilmington, DE). Perfusion was expressed as the ratio of ischemic to non-ischemic hind limb perfusion. In parallel experiments, mice were sacrificed, adductor muscles from ischemic and non-ischemic limbs of wild type and 12/15-Lox Ϫ/Ϫ mice were dissected out, and either protein or RNA was isolated or embedded in OCT compound. Tissue extracts were analyzed for MEK1 and JNK1 phosphorylation and Fra-1, c-Jun, and MMP-2 levels. Total RNA was used to measure MMP-2 expression. Embedded tissues were sectioned and stained for CD31, vWF, and Hu177.
Double Immunofluorescence Staining-After retrieving the Matrigel plugs and adductor muscles from mice, they were snap-frozen in OCT compound. Cryosections (5 m) were made using Leica Kryostat (Model CM3050S, Leica, Wetzlar, Germany). After blocking in normal goat serum, the cryosections were incubated with rabbit anti-mouse von Willebrand Factor (vWF) antibodies and rat anti-mouse CD31 antibodies or monoclonal Hu177 antibodies (42) and rat anti-mouse CD31 for 1 h (1:500). After washing in phosphate-buffered saline, all slides were incubated with goat anti-rabbit secondary antibodies conjugated with Alexa Fluor 568 or goat anti-rat secondary antibodies conjugated with Alexa Fluor 488 or goat anti-rat secondary antibodies conjugated with Alexa Fluor 568 or goat anti-mouse secondary antibodies conjugated with Alexa Fluor 488. Fluorescence was observed under a Zeiss inverted microscope (Model AxioVision AX10).
Statistics-All the experiments were repeated three times, and data are presented as mean Ϯ S.D. The treatment effects were analyzed by Student t test, and the p values Ͻ0.01 were considered statistically significant. In the case of double immunofluorescence staining, ChIP analysis, EMSA, gelatin zymography, and Western blotting, one representative set of data is shown.

15(S)-HETE Induces MMP-2 Expression and Activation in
HDMVECs-To understand the mechanisms of 15(S)-HETEinduced angiogenesis, we studied the role of MMP-2. 15(S)-HETE (0.1 M)-induced MMP-2 mRNA levels in a time-dependent manner with a 2.5-fold increase at 8 h compared with control (Fig. 1A). To confirm these results, we also studied the time course effect of 15(S)-HETE on MMP-2 activity. As measured by gelatin zymography, we found that treatment with 15(S)-HETE (0.1 M) led to a time-dependent release of MMP-2 into the conditioned medium with 2.5-fold increase at 8 h as compared with untreated cells (Fig. 1B).
To test whether AP-1 binds to MMP-2 promoter in response to 15(S)-HETE, we cloned a 1.7-kb human MMP-2 promoter region and identified three putative AP-1 binding sites located each at Ϫ1263, Ϫ1615, and Ϫ1673 nt relative to the transcription start site by Transfac analysis (Fig. 5A). Promoter-luciferase reporter gene assays along with site-directed mutagenesis showed that the AP-1 site proximal to the transcriptional start site is important for 15(S)-HETE-induced MMP-2 expression (Fig. 5B). To identify the AP-1 components in 15(S)-HETEinduced MMP-2 promoter-luciferase activity, we next tested the role of Fra-1 and c-Jun. Down-regulation of either Fra-1 or c-Jun levels by their respective siRNAs significantly attenuated  15(S)-HETE-induced MMP-2 promoter-luciferase activity (Fig. 6A). To confirm these observations, we also studied a time course effect of 15(S)-HETE on AP-1 DNA-binding activity using an AP-1 element at Ϫ1263 nt as a 32 P-labeled probe. 15(S)-HETE induced AP-1 DNA-binding activity in a time-dependent manner with ϳ3-fold increase at 1 h (Fig. 6B). Blockade of Rac1-MEK1 signaling by adenovirus-mediated expression of their dominant negative mutants completely inhibited 15(S)-HETEinduced AP-1 DNA-binding activity (Fig. 6C). To obtain additional evidence for the role of Fra-1 and c-Jun in the regulation of MMP-2 promoter activity, we performed a ChIP assay. ChIP analysis revealed a timedependent binding of Fra-1 and c-Jun to MMP-2 promoter in vivo (Fig. 6D). In addition blockade of Rac1 or MEK1 via adenovirus-mediated expression of their dominant negative mutants suppressed the binding of Fra-1 and c-Jun to MMP-2 promoter in vivo (Fig. 6E).

12/15-Lox-15(S)-HETE-induced Angiogenesis Requires MMP-2
Lack of a Neovascularization Response in 12/15-Lox Ϫ/Ϫ Mice following Hind Limb Ischemia-To demonstrate the in vivo relevance of the in vitro findings, we used a murine model of hind limb ischemia. It was observed that, as compared with WT mice, 12/15-Lox Ϫ/Ϫ mice had impaired recovery of blood flow 7 days after hind limb ischemia (Fig. 7A). To confirm these results further, immunofluorescence was performed on the adductor muscles of both WT and 12/15-Lox Ϫ/Ϫ mice for the presence of endothelial cells. Double immunofluorescence staining demonstrated the presence of increased CD31 and vWF in the ischemic muscles of WT mice as compared with 12/15-Lox Ϫ/Ϫ mice (Fig. 7B). To understand the signaling events of neovascularization in vivo, we have also measured the activation of MEK1 and JNK1 in the adductor muscles of WT and 12/15-Lox Ϫ/Ϫ mice with and without ischemia. As compared with non-ischemia control, ischemia induced both MEK1 and JNK1 phosphorylation at least by 2-fold in WT mice (Fig. 8A). Similarly, ischemia induced Fra-1, c-Jun, and MMP-2 expression as well as MMP-2 activity by 2-to 5-fold in the adductor muscles of WT mice (Fig. 8, B-E). Surprisingly, as compared with WT mice, ischemia failed to stimulate the acti- MAY 28, 2010 • VOLUME 285 • NUMBER 22

12/15-Lox-15(S)-HETE-induced Angiogenesis Requires MMP-2
vation of MEK1 and JNK1 as well the induction of expression of Fra-1, c-Jun, and MMP-2 in the adductor muscles of 12/15-Lox Ϫ/Ϫ mice. Consistent with a lack of effect on MMP-2 levels, ischemia also had no significant effect on MMP-2 activity in 12/15-Lox Ϫ/Ϫ mice. HU177 was selectively exposed in the interstitial matrix of tumors as well as in the extracellular matrix of angiogenic blood vessels as a result of extracellular matrix degradation by MMPs (42). To confirm the role of MMP-2 in ischemia-induced neovascularization, we also performed double immunofluorescence staining for Hu177 and CD31 in ischemic adductor muscles of both WT and 12/15-Lox Ϫ/Ϫ mice. It was observed that ischemia increased Hu177 cryptic collagen-positive staining in WT mice as compared with 12/15-Lox Ϫ/Ϫ mice (Fig. 8E).

DISCUSSION
Humans express 15-Lox1 and 15-Lox2, and both convert arachidonic acid to 15(S)-HETE as a predominant eicosanoid (50 -53). Furthermore, the atherosclerotic arteries upon incubation with arachidonic acid produced 15(S)-HETE as a major eicosanoid (33,34). Despite these observations, many studies have focused on the role of 12/15-Lox, a murine ortholog of 15-Lox1, in the pathogenesis of atherosclerosis and restenosis (24 -26, 54), thus underscoring the importance of 15-Lox/15-HETE in these diseases. To understand the role of 15-Lox/15-HETE in vascular diseases, we have previously shown that it stimulates angiogenesis as well as the migration of smooth muscle cells (35-39, 52, 53). A role for angiogenesis in the progression of both atherosclerosis and restenosis has also been reported (55,56). We have shown that 15(S)-HETE-induced angiogenesis requires induction of expression of angiogenic factors such as fibroblast growth factor, vascular endothelial growth factor, and interleukin-8 (35)(36)(37). MMPs have been reported to be involved in the progression of atherosclerosis and restenosis (13,19,57). Similarly, MMPs, particularly MMP-2, have been shown to be critical in the onset of tumor angiogenesis (58). The findings that atherosclerotic arteries produce 15-HETE as a major eicosanoid and it possesses the capacity to induce angiogenesis suggest that one of the mechanisms by which 15(S)-HETE could promote vascular diseases is via its involvement in the regulation of angiogenesis. Similarly, 15(S)-HETE via its capacity to induce MMP-2 expression and angiogenesis may be involved in tumor growth and metastasis as well.  The Rho group of small GTPases plays a key role in cell motility (44,45). In addition, Rac1 mediates MMP-2 activation by membrane type 1-MMP, and MMP-2 activity is required for Rac1-facilitated cell invasion across the type I collagen barrier (47). Because 15(S)-HETE stimulates Rac1 in the induction of expression of MMP-2, whose activation is needed for cell migration, it is logical to assume that the Rac-1-MMP-2 axis constitutes a motile signaling for endothelial cell invasion and migration in 15(S)-HETE-induced angiogenesis. Rac1 plays a role in mediating receptor tyrosine kinase and cytokine receptor signaling to activate JNK1 via recruiting MAPK kinase kinases, MEKK1-4, and MAPK kinases, MKK4/7 (59 -61). Furthermore, activation of JNK1 is required for stimulus-induced expression of the Fos and Jun family of AP-1 transcriptional factors (62). Previously, we have shown that 15(S)-HETE activates JNK1 via Rac1-dependent stimulation of MEK1 (38). In the present study, we observed that activation of The promoter regions of many MMPs, including MMP-1, -3, -7, -9, -10, -12, and -13, are highly conserved with respect to the ϳ7-bp AP-1 binding site (63). The functional significance of the AP-1 site in the mediation of MMP-2 promoter activity in response to hormones and cytokines such as angiotensin II,

12/15-Lox-15(S)-HETE-induced Angiogenesis Requires MMP-2
endothelin 1, and interleukin 1␤ was confirmed by EMSA studies and site-directed mutagenesis in cardiac cells (49). In addition, it was found that both Fra-1/Jun-B and Fos-B/Jun-B occupy the AP-1 site in the MMP-2 promoter in response to these stimulants. Although 15(S)-HETE induced c-Fos, Fra-1, Jun-B, and c-Jun levels very robustly, blockade of Rac1-MEK1 signaling suppressed predominantly Fra-1 and c-Jun levels indicating a possible role for these AP-1 components in the induction of MMP-2 expression. This assumption can be further corroborated by the findings that interference with JNK1 activation blocks 15(S)-HETE-induced Fra-1 and c-Jun levels. In addition, siRNA-mediated inhibition of Fra-1 or c-Jun levels led to a lack of effect of 15(S)-HETE on MMP-2 expression and its promoter activity. Interference with Rac1-MEK1 signaling negated the binding of these AP-1 transcriptional factors to MMP-2 promoter in response to 15(S)-HETE. Furthermore, data obtained by site-directed mutagenesis, EMSA, and ChIP assay suggest that, among the three AP-1 sites present in the 1.7-kb promoter, the AP-1 site proximal to the transcriptional start site is essential for 15(S)-HETE-induced MMP-2 promoter activity. Previous studies have demonstrated that MMPs by proteolytic degradation releases soluble ectodomain of fibroblast growth factor receptor-1 that possesses the capacity to bind to FGF (64). We observed that 15(S)-HETE induces the expression of FGF-2 (37, 38). These findings appear to be very intriguing and lead to the speculation that 15(S)-HETE while inducing the expression of FGF-2 may also via activation of MMPs release the soluble active ectodomain of its receptor so that the growth factor can bind and activate its receptor leading to angiogenesis. In this context, it is also noteworthy that MMPs and ADAMs (a disintegrin and metalloprotease) via releasing growth factors such as heparinbound-EGF and heparin-bound-FGF from degradation of proteoglycans facilitates the activation of receptor tyrosine kinases by G-protein-coupled receptor agonists (65).
Neovascularization is an adaptive biological response of ischemia. Therefore, we have used a murine hind limb ischemia model to understand the physiological relevance of our in vitro findings on the role of 15(S)-HETE in the regulation of angiogenesis in vivo (41). It is interesting to note that in 12/15-Lox Ϫ/Ϫ mice the blood flow recovery is significantly lower than that in WT mice after hind limb ischemia suggesting a crucial role for 12(S)-HETE/15(S)-HETE in the stimulation of angiogenesis. MMPmediated remodeling of extracellular matrix creates a permissive microenvironment for endothelial cell invasion and migration and thereby new blood vessel formation. Recently, it was shown that cryptic collagen epitope (Hu177) is present in both interstitial collagen type 1 and basement membrane collagen type IV and that selective exposure of these epitopes within ischemic muscle correlates with the enhanced MMP activity (42). Our findings reveal that 7 days after the initiation of ischemia, increased exposure of Hu177-positive staining was observed in the adductor muscles of WT mice suggesting increased MMP activity. In contrast, the adductor muscles of 12/15-Lox Ϫ/Ϫ mice showed decreased Hu177-positive staining as compared with WT mice, and this correlates with a lack of angiogenic response. These findings infer that 12/15-Lox and its metabolites of arachidonic acid, 12(S)-HETE/15(S)-HETE, are crucial in the induction of expression and activation of MMPs, most likely MMP-2, which is important for initiation of angiogenesis. Because exposure of Hu177-positive epitopes is a marker for increased MMP activities, including MMP-2 and MMP-9, our results may not exclude a possible role for other MMPs besides MMP-2 in 12(S)-HETE/15(S)-HETE-induced angiogenesis following ischemia. However, because ischemia mimicked 15(S)-HETE in the stimulation of MEK1 and JNK1 phosphorylation and induction of Fra-1, c-Jun, and MMP-2 expression, it is most likely that MMP-2 does play a role in ischemia-induced angiogene-FIGURE 7. Lack of 12/15-Lox gene impairs blood flow recovery after ischemia. WT and 12/15-Lox Ϫ/Ϫ mice were subjected to hind limb ischemia by left femoral artery excision. A, on day 7 after hind limb ischemia, blood flow was measured by using laser Doppler perfusion imaging. Perfusion is expressed as the ratio of the ischemic to the non-ischemic hind limb. B, blood vessels in the ischemic adductor muscles of WT and 12/15-Lox Ϫ/Ϫ mice were analyzed by double immunofluorescence staining for CD31 (green) and vWF (red). *, p Ͻ 0.01 versus WT mice.

12/15-Lox-15(S)-HETE-induced Angiogenesis Requires MMP-2
sis. Previous studies have reported that chronic hypoxia induces the expression of 15-Lox1 in aortic endothelial cells (66). We have shown that hypoxia induces the expression of 15-Lox1 in human microvascular endothelial cells and thereby increases the production of 15(S)-HETE (37). The hypoxia-induced expression of 15-Lox1 and the production of its arachidonic acid metabolite, 15(S)-HETE, may be attributed to its adaptive role in restoring the blood flow to ischemic tissues as observed in the present study. The lack of angiogenic response following mild ischemia in 12/15-Lox Ϫ/Ϫ mice is consistent with this assumption. In contrast to these observations, some recent reports have shown that 15-Lox1 inhibits vascular endothelial growth factor-A and placental growth factor-induced angiogenesis in rabbit skeletal muscle (67). These differential effects of 15-Lox1-15-HETE axis in the modulation of angiogenesis may be attributed to species variations.
Taken together, our studies provide a mechanistic evidence for the role of 12/15(S)-HETE in the regulation of angiogenesis through Rac1, MEK1, JNK1, and AP-1 (Fra-1/ c-Jun)-dependent expression and secretion of MMP-2. We have previously reported that 15(S)-HETEinduced angiogenesis requires the production of FGF-2, vascular endothelial growth factor, and interleukin-8 (35,36,68). Because angiogenesis is an orchestrated cellular process involving the actions of many molecules, it is quite expected that 15(S)-HETE-induced angiogenesis may also need the requirement of several molecules, including cytokines and MMPs. Based on the capacity of 15(S)-HETE to stimulate angiogenesis and its production in atherosclerotic arteries as a major eicosanoid, it is likely that this oxidized lipid molecule contributes to the pathological angiogenesis that occurs in atherosclerotic plaques and restenotic lesions. A-D, adductor muscles were isolated from ischemic and non-ischemic WT and 12/15-Lox Ϫ/Ϫ mice 7 days post-operation, and either tissue extracts were prepared or RNA was isolated. Tissue extracts were analyzed by Western blotting for MEK1 and JNK1 phosphorylation and Fra-1 and c-Jun levels using the respective antibodies (A). MMP-2 mRNA levels were measured by QRT-PCR (B), and its protein levels and activity were measured by Western blotting and gelatin zymography, respectively (C and D). E, cryptic collagen epitopes were evaluated in the ischemic adductor muscles of WT and 12/15-Lox Ϫ/Ϫ mice by immunofluorescence staining with anti-Hu177 antibodies (green) and anti-CD31 antibodies (red). The bar graphs in A, B, and D represent the mean Ϯ S.D. values for six animals. *, p Ͻ 0.01 versus WT mice.