Src-dependent STAT-3-mediated Expression of Monocyte Chemoattractant Protein-1 Is Required for 15(S)-Hydroxyeicosatetraenoic Acid-induced Vascular Smooth Muscle Cell Migration*

To understand the role of human 15-lipoxygenase 1 (15-LOX1) in vascular wall remodeling, we have studied the effect of the major 15-LOX1 metabolite of arachidonic acid, 15(S)-hydroxyeicosatetraenoic acid (15(S)-HETE), on vascular smooth muscle cell (VSMC) migration both in vitro and in vivo. Among 5(S)-HETE, 12(S)-HETE, and 15(S)-HETE, 15(S)-HETE potentially stimulated more vascular smooth muscle cell (VSMC) migration. In addition, 15(S)-HETE-induced VSMC migration was dependent on Src-mediated activation of signal transducer and activator of transcription-3 (STAT-3). 15(S)-HETE also induced monocyte chemoattractant protein-1 (MCP-1) expression via Src-STAT-3 signaling, and neutralizing anti-MCP-1 antibodies completely negated 15(S)-HETE-induced VSMC migration. Cloning and characterization of a 2.6-kb MCP-1 promoter revealed the presence of four putative STAT-binding sites, and the site that is proximal to the transcription start site was found to be essential for 15(S)-HETE-induced Src-STAT-3-mediated MCP-1 expression. Rat carotid arteries that were subjected to balloon injury and transduced with Ad-15-LOX1 upon exposure to [3H]arachidonic acid ex vivo produced 15-HETE as a major eicosanoid and enhanced balloon injury-induced expression of MCP-1 in smooth muscle cells in Src and STAT-3-dependent manner in vivo. Adenovirus-mediated delivery of 15-LOX1 into rat carotid artery also led to recruitment and homing of macrophages to medial region in response to injury. In addition, transduction of Ad-15-LOX1 into arteries enhanced balloon injury-induced smooth muscle cell migration from media to intima and neointima formation. These results show for the first time that 15-LOX1–15(S)-HETE axis plays a major role in vascular wall remodeling after balloon angioplasty.

VSMC 3 migration from media to intima plays a determinant role in atherosclerosis and restenosis (1)(2)(3). Arachidonic acid (AA) and its oxygenated metabolites, collectively known as eicosanoids, are involved in the maintenance of vascular tone (4,5). Lipoxygenases (LOXs) are non-heme iron dioxygenases that stereospecifically introduce molecular oxygen into polyunsaturated fatty acids, resulting in the formation of hydroperoxyeicosatetraenoic acids, which are subsequently converted to hydroxyeicosatetraenoic acids (HETEs) (6). Two 15-LOXs, namely, 15-LOX1 and 15-LOX2, have been shown to be expressed in humans (7,8). Both enzymes metabolize linoleic acid to 13(S)-hydroperoxyoctadecadienoic acid and AA to 15(S)hydroperoxyeicosatetraenoic acid preferentially (9,10). In regard to their tissue distribution, whereas 15-LOX1 shows a narrow cell-specific expression, including human reticulocytes and airway epithelial cells, 15-LOX2 appears to be expressed in epithelial cell types in cornea, lung, prostate, and skin (11). The studies from our laboratory showed that both 15-LOX1 and 15-LOX2 are expressed in human retinal microvascular endothelial cells (12). Although the presence of 15-LOX2 in VSMC has yet to be reported, these cells express 15-LOX1 (also known as 12/15-LOX in murines) and, when exposed to AA, produce both 15(S)-HETE and 12(S)-HETE (13,14). It is also known that 15-LOX1 and 12/15-LOX are involved in the oxidation of low density lipoprotein, a contributing factor in the pathogenesis of atherosclerosis (15,16). Furthermore, using either transgenic or knock-out mice models, a number of studies have demonstrated that 15-LOX1 and its murine ortholog 12/15-LOX play a role in atherosclerosis and restenosis (17)(18)(19). In addition, human atheroma homogenates upon incubation with AA converted it mainly to .
In recent years LOX products of polyunsaturated fatty acids have also been shown to be potent chemoattractants for resi-dential and invading immune cells recruited to vascular lesion areas (21). Although the association of LOX products of polyunsaturated fatty acids with the pathophysiology of vessel wall diseases is well documented, the precise mechanisms by which these lipid molecules, particularly 15(S)-HETE, the preferentially produced AA product of both 15-LOX1 and 15-LOX2, act on VSMC is not well understood. Here, we report for the first time that 15(S)-HETE induces the expression of MCP-1, a potent chemoattractant for inflammatory and smooth muscle cells (22,23), via a mechanism involving Src and STAT-3 and mediates VSMC migration. In addition, we present evidence that adenovirus-mediated expression of 15-LOX1 in the vessel wall led to increased production of 15-HETE and enhanced expression of MCP-1 in smooth muscle cells (SMC) and the migration of these cells from media to intima resulting into neointima formation. In addition, adenovirus-mediated expression of 15-LOX1 in the vessel wall enhanced the recruitment and homing of macrophages in the medial region in response to injury.
Cell Culture-Rat VSMCs were isolated and subcultured as described previously (24). VSMCs were used between 6 and 12 passages.
Cell Migration-Cell migration was measured using a modified Boyden chamber method as described previously (29). Wherever adenovirus was used, cells were first transduced with the respective adenovirus at a m.o.i. of 40 and growth-arrested before they were subjected to stimulus-induced migration. Cell motility was presented as number of migrated cells per field.
RT-PCR and Quantitative Real Time-PCR-After appropriate treatments, total cellular RNA was isolated from VSMCs using Trizol reagent as per manufacturer's instructions (Invitrogen). Reverse transcription was carried out with the Superscript III First-Strand Synthesis System for RT-PCR based on the supplier's protocol (Invitrogen). The cDNA was then used as a template for PCR using specific primers. The primers used are as follows: rat MCP-1 (accession no. AF058786), 5Ј-CAGAAACCAGCCAACTCTCA-3Ј (forward) and 5Ј-GCTTGAGGTGGTTGTGGAAA-3Ј (reverse); rat ␤-actin (accession no. EF156276), 5Ј-CGTTGACATCCGTA-AAGACC-3Ј (forward) and 5Ј-GATAGAGCCACCAATC-CACA-3Ј (reverse). The amplification was carried out on Gene Amp PCR System 2400 (Applied Biosystems, Foster City, CA). The amplified RT-PCR products were separated on 1.2% (w/v) agarose gels and stained with ethidium bromide, pictures were captured using AlphaEase Digital Imaging System (Alpha Innotech Corporation, San Leandro, CA), and the band intensities were quantified using NIH Image J. In the case of quantitative real time-PCR, the cDNA was used as a template for PCR using Taqman gene expression assay kit for rat MCP-1 (catalog no. Rn00580555_m1) and ␤-actin (catalog no. Rn00667869_m1).
The amplification was carried out on an Applied Biosystems 7300 real time PCR system (Applied Biosystems) using the following amplification conditions for the above-mentioned genes as follows: 95°C for 10 min followed by 40 cycles at 95°C for 15 s with a final extension at 60°C for 1 min. The amplification was examined using the 7300 real time PCR system-SDS Version 1.4 program.
ELISA-MCP-1 release into medium was measured by using its specific ELISA kit following the manufacturer's instructions.
Luciferase Reporter-gene Assay-VSMC were transfected with MCP-1 promoter-luciferase constructs using Lipofectamine 2000 reagent. After growth synchronization in serum-free medium for 36 h, cells were treated with and without 15(S)-HETE (0.5 M) for 2 h. Cells were then washed once with ice-cold PBS and lysed with 200 l of lysis buffer, and the cell extracts were collected into microcentrifuge tubes and centrifuged for 10 min at 12,000 ϫ g at 4°C. The supernatants were assayed for luciferase activity using the luciferase assay system (Promega) and a single tube luminometer (TD20/20 Turner Designs, Sunnyvale, CA) and expressed as relative luciferase units/mg of protein.
Electrophoresis Mobility Shift Assay-After appropriate treatments, VSMC nuclear extracts were prepared and analyzed for STAT-3 DNA binding activity as described previously (26). 32 P-Labeled double-stranded oligonucleotides from the Ϫ168 to Ϫ136 (5Ј-CTCCCTTCTACTTCCTGGAAACATCC-AAGGG-3Ј) region of rat MCP-1 promoter (accession no. AF079313) were used as a probe to measure STAT-3 DNA binding activity.
Chromatin Immunoprecipitation Assay-Chromatin immunoprecipitation assay was done on VSMCs by using the chromatin immunoprecipitation assay kit following the supplier's protocol (Upstate Biotechnology). STAT-3-DNA complexes were immunoprecipitated using anti-STAT-3 antibody (SC-482). Preimmune serum (SC-2338) was used as a negative control. After heating at 65°C and treatment with proteinase K to reverse DNA-protein cross-links, the immunoprecipitated DNA was extracted using QIAquick columns (Qiagen). The purified DNA was used as a template for PCR amplification using primers (5Ј-CCAATCCGCGGTTTCTCC-3Ј (forward) and 5Ј-GGCTTCAGTGAGAGTTGG-3Ј (reverse)) flanking the putative STAT-binding site located at Ϫ149 in the rat MCP-1 promoter region. The PCR products were resolved on 2% agarose gel and stained with ethidium bromide.
Carotid Artery Balloon Injury-All the animal protocols were performed in accordance with the relevant guidelines and regulation approved by the Institutional Animal Care and Use Committee of the University of Tennessee Health Science Center. Balloon injury (BI) was performed essentially as described previously (30). At 3 days or 2 weeks after BI, the animals were sacrificed with an overdose of pentobarbital (200 mg/kg), and the carotid arteries were collected and processed for either RNA isolation or morphometric analysis. For morphometric analysis, carotid arteries were fixed in 10% formalin and embedded in OCT (Sakura Finetek U. S. A. Inc., Torrance, CA), and 5-m-thick sections were made at equally spaced intervals in the middle of injured and uninjured common carotid artery segments and stained with hematoxylin and eosin. The intimal (I), medial (M), and luminal areas were measured using NIH Image J and the I/M ratios were calculated.
Reverse Phase HPLC Analysis of Arachidonic Acid Metabolites-At 3 days after balloon injury and transduction with Ad-GFP or Ad-15-LOX1, the arteries were isolated, minced, and incubated with [ 3 H]arachidonic acid (0.5 Ci/ml) for 2 h. The eicosanoids released into the medium were extracted in methanol at a final concentration of 10% (v/v). The carotid artery minces were rinsed with methanol, and the rinse was added to the above mixture. The pH of the mixture was adjusted to 2.5 with 10% (v/v) glacial acetic acid, and it was then passed through a Pre-Sep C18 column. The arachidonic acid metabolites were eluted with 100% methanol followed by evaporation to dryness. The metabolites were reconstituted in 50 l of 100% methanol and analyzed by reverse-phase high performance liquid chromatography using a C18 column (5-m particle size, 4.6 ϫ 250 mm, Beckman Instruments, Berkeley, CA) as described previously using methanol:water:acetic acid with 55:45:0.01 and 75:25:0.01 ratios and 100% methanol as the mobile phases at a flow rate of 1.0 ml/min for 40, 30, and 20 min (31).
Immunohistochemistry-Serial carotid artery cross-sections, after rinsing in PBS for 5 min, were incubated with 0.3% H 2 O 2 for 30 min to block endogenous peroxidase activity. After blocking in 5% goat serum in PBS for 30 min, sections were incubated with mouse anti-rat CD68 antibodies (1:200 dilution) or rabbit anti-rat MCP-1 antibodies (1:100 dilution) in PBS containing 5% goat serum at room temperature for 60 min followed by incubation in biotinylated goat anti-mouse IgG or goat anti-rabbit IgG for 30 min. Peroxidase labeling was carried out using the ABC kit (Vector Laboratories), and the signals were visualized by using a DAB kit (Vector Laboratories) to display the reaction product with a brown color. The sections were then counterstained with hematoxylin and mounted. After each step, the slides were washed 3 times for 5 min each in PBS. As a negative control the sections were incubated with PBS instead of the primary antibody. Six different fields at ϫ400 magnification were evaluated for the semiquantitative analysis of macrophage recruitment or MCP-1 expression. The recruited macrophage cell number was presented as the number of CD68-positive cells/0.04-mm 2 area including intima, media, and adventitia regions, whereas MCP-1 expression was evalu-ated as the ratio of positively stained medial area to total medial area in the observed field.
Double Immunofluorescence Staining-Soon after isolation, the carotid arteries were snap-frozen in OCT compound. Cryosections (5 m) were made using Leica Kryostat (model CM3050S, Leica, Wetzlar, Germany). After blocking in goat serum, the cryosections were incubated with rabbit anti-rat MCP-1 antibodies and mouse anti-rat SMC␣-actin antibodies for 1 h. After washing in PBS, all slides were incubated with TRITC-conjugated goat anti-rabbit and fluorescein-conjugated goat anti-mouse secondary antibodies. Fluorescence was observed under Zeiss Axio Observer Z1 Motorized Inverted microscope. Negative controls were processed exactly as described above except that they were not incubated with primary antibodies.
In Vivo SMC Migration Assay-The in vivo SMC migration was determined as described by Bendeck et al. (32). Briefly, 3 days after balloon injury, the carotid arteries were fixed in vivo with 10% buffered formalin at physiological pressure. The middle 1 cm of the denuded (injured) common carotid artery was cut and fixed in cold acetone for 10 min. The artery was then opened longitudinally and pinned down onto an agar plate with the luminal surface facing upward. The arteries were rinsed in PBS and then placed in 0.3% H 2 O 2 for 30 min to block endogenous peroxidase activity. Nonspecific protein binding was blocked by incubating the arteries in 5% normal goat serum in PBS for 30 min. The arteries were incubated with anti-SMC␣actin antibodies (A2547 Clone 1A4, Sigma) diluted 1:300 in PBS for 1 h followed by incubation in biotinylated goat anti-mouse IgG for 30 min. Peroxidase labeling was carried out by using ABC kit (Vector Laboratories), and the signals were visualized by using the DAB kit (Vector Laboratories). After each step, the slides were rinsed 3 times for 5 min each in PBS. Finally, the opened arteries were placed intimal side up on glass slides with coverslips. As a negative control, samples of the same specimens without the primary antibody incubation were used. The intimal surface of the vessel was examined under a light microscope at ϫ200 magnification, and the total number of positively stained cells per 0.1 mm 2 of the luminal surface area was counted.
Western Blot Analysis-After appropriate treatments and rinsing with cold PBS, VSMCs were lysed in 500 l of lysis buffer (1ϫ PBS containing 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 100 g/ml aprotinin, 1 g/ml leupeptin, and 1 mM sodium orthovanadate) and collected into 1.5-ml microcentrifuge tubes. After 20 min incubation on ice, the cell lysates were cleared by centrifugation at 12,000 rpm for 20 min at 4°C. Cell lysates containing equal amounts of protein were resolved by electrophoresis on 0.1% SDS and 10% polyacrylamide gels. The proteins were electroblotted to a nitrocellulose membrane (Hybond, GE Healthcare). After blocking in 10 mM Tris-HCl buffer, pH 8.0, containing 150 mM sodium chloride, 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 (GE Healthcare).
Statistics-All the experiments were repeated three times with similar results. Data are presented as the means Ϯ S.D. The treatment effect was analyzed by Student's t test. p values Ͻ0.05 were considered to be statistically significant. In the case of Western blotting, electrophoretic mobility shift assay, histochemistry, immunohistochemistry, and RT-PCR, one of the representative set of data is shown.

Hydroxyeicosatetraenoic Acids Stimulate VSMC Migration-
Previously we have reported that 15(S)-HETE stimulates VSMC migration in a dose-dependent manner with maximum effect at 0.5 M, which is a pathophysiological concentration (29). To understand the role of eicosanoids in the pathogenesis of atherosclerosis and restenosis, we first studied the comparative effect of various LOX products of AA, namely, 5(S)-HETE, 12(S)-HETE, and 15(S)-HETE, on VSMC migration using a modified Boyden chamber method. We found that 5(S)-HETE, 12(S)-HETE, and 15(S)-HETE at 0.5 M concentration stimulated VSMC migration more than 2-fold as compared with control (Fig. 1). Among these HETEs, 15(S)-HETE was found to be more potent. To see whether chirality has any role in the chemotactic effects of 15(S)-HETE, we next tested the effects of its enantiomer, 15(R)-HETE, in the stimulation of VSMC migration. Although both enantiomers stimulated VSMC migration robustly, 15(S)-HETE was observed to be more potent than  Fig. 1). Hence, we focused further studies on understanding the mechanisms of 15(S)-HETE-induced VSMC migration.

STAT-3 Mediates 15(S)-HETE-induced MCP-1 Expression
Activation of STAT-3 Is Required for 15(S)-HETE-induced VSMC Migration-STATs play an important role in cellular processes, including cell migration and proliferation (33)(34)(35). We have demonstrated a role for STAT-3 and STAT-5B in thrombin and platelet-derived growth factor-BB-induced VSMC migration and proliferation (36 -38). Phosphorylation of STAT-3 at Tyr-705 is required for its dimerization, nuclear translocation, and binding to DNA (39). To test whether STAT-3 is activated by 15(S)-HETE, we analyzed cell extracts prepared from control and various time periods of 15(S)-HETE (0.5 M)-treated VSMCs by Western blotting using its phosphospecific antibodies. We found that 15(S)-HETE stimulates phosphorylation of STAT-3 at Tyr-705 in a sustained manner with a maximum increase of 3-fold at 1 h ( Fig. 2A) (Fig. 3A). We next applied a dominant negative mutant approach to test the role of Src in 15(S)-HETE-induced STAT-3 phosphorylation and VSMC migration. Transduction of VSMCs with adenovirus harboring dnSrc at 40 m.o.i. attenuated 15(S)-HETE-induced Src and STAT-3 tyrosine phosphorylation in VSMCs and the migration of these cells (Fig. 3, B-D). 15(S)-HETE Induces MCP-1 Expression in VSMCs-Several cytokines including interleukin-6 (IL-6) and MCP-1 have been shown to play an important role in the vessel wall diseases (41)(42)(43). MCP-1 has been reported to be one of the most potent cytokines involved in the progression of atherosclerosis and observed at 2 and 4 h, and these levels were declined at 6 h (Fig. 4, A and  B). To confirm the effect of 15(S)-HETE on MCP-1 expression, we also measured its release into the medium by ELISA. Consistent with its effects on MCP-1 mRNA expression, 15(S)-HETE (0.5 M) induced its secretion in a time-dependent manner as well (Fig. 4C). To demonstrate the role of MCP-1 in 15(S)-HETE-induced VSMC migration, we used neutralizing anti-MCP-1 antibodies. Neutralizing anti-MCP-1 antibodies (3 g/ml) completely abrogated 15(S)-HETE (0.5 M)-induced VSMC migration (Fig. 4D).

Src and STAT-3 Activation Are Required for 15(S)-HETE-induced MCP-1 Expression in VSMCs-To test the role of Src and STAT-3 in 15(S)-HETE-induced expression of MCP-1 in VSMCs, cells were transduced with Ad-GFP, Ad-dnSrc, or
Ad-dnSTAT-3 at 40 m.o.i., quiesced, and treated with and without 0.5 M 15(S)-HETE for 2 h, and either total cellular RNA was isolated and analyzed for MCP-1 mRNA levels by RT-PCR or medium was collected and tested for MCP-1 release by ELISA. Adenovirus-mediated expression of dnSrc or dnSTAT-3 substantially attenuated 15(S)-HETE (0.5 M)induced MCP-1 expression and secretion (Fig. 5, A and B). Although a role for Jak-STAT signaling in the regulation of MCP-1 by plasmin has been reported (45), the underlying mechanisms were not clear. To understand the mechanism(s) by which STAT-3 mediates 15(S)-HETE-induced MCP-1 expression, we cloned a 2.6-kb rat MCP-1 promoter and analyzed it for STATbinding sites by Transfac search (46). From Transfac search results, we identified four putative STATbinding sites and other cis-acting elements in the cloned MCP-1 promoter (supplemental Fig. 2). To determine the functional role of these sites in MCP-1 expression, we performed promoter deletion analysis. MCP-1 promoter with serial 5Ј deletions starting at Ϫ2478, Ϫ1881, Ϫ1249, Ϫ1192, Ϫ763, Ϫ532, and Ϫ182 nucleotides were generated by polymerase chain reaction and cloned into pGL3 basic vector. VSMCs were transfected with these MCP-1 pro-  NOVEMBER 6, 2009 • VOLUME 284 • NUMBER 45 moter-luciferase constructs, quiesced, and treated with and without 0.5 M 15(S)-HETE for 2 h, and cell extracts were prepared and analyzed for luciferase activity. No significant changes in luciferase activity were observed with all MCP-1 promoter-luciferase constructs except pGL3-MCP1(0.2 kb)-Luc in response to 15(S)-HETE treatment as compared with vector control (Fig. 6A). However, a 4-fold increase in MCP-1 promoter-luciferase activity was observed in response to 15(S)-HETE with pGL3-MCP1(0.2 kb)-Luc construct that contains one putative STAT-binding site. In addition, mutational disruption of this STAT-binding site completely abrogated both basal and 15(S)-HETE-induced pGL3-MCP1(0.2 kb)-Luc activity (Fig. 6A).

STAT-3 Mediates 15(S)-HETE-induced MCP-1 Expression
To further confirm the role of Src and STAT-3 in 15(S)-HETE-induced MCP-1 promoter activity in VSMCs, we performed an electrophoretic mobility shift assay using a 32 P-labeled oligonucleotide probe that contained the putative STAT binding sequence proximal to the transcription start site (Ϫ149 to Ϫ157 nucleotides; 5Ј-TTCCTG-GAA-3Ј). 15(S)-HETE induced STAT DNA binding activity in a timedependent manner, and this effect was completely inhibited by adenovirus-mediated expression of dnSrc or dnSTAT-3 (supplemental Fig. 3,  A and B). To confirm STAT-3 binding to MCP-1 promoter in vivo, we performed a chromatin immunoprecipitation assay. Time-dependent binding of STAT-3 to MCP-1 promoter was observed by chromatin immunoprecipitation assay (Fig. 6B). In addition, blockade of Src and STAT-3 via adenovirus-mediated expression of their dominant negative mutants suppressed the binding of STAT-3 to MCP-1 promoter in vivo (Fig. 6C).

Overexpression of 15-LOX1 Enhances Neointima Formation in Rat
Carotid Artery in Response to Balloon Injury-Migration of smooth muscle cells (SMCs) from media to intima is a hallmark both in atherosclerosis and restenosis after angioplasty (1)(2)(3)47). To understand the role of 15(S)-HETE in MCP-1 expression in SMCs and their migration from media to intima in vivo, we used an adenoviral vector for expression of 15-LOX1. Soon after balloon injury, Ad-GFP or Ad-15-LOX1 (10 10 pfu) was transduced into arteries, and 3 days later these arteries were dissected out, minced, and incubated with [ 3 H]AA for 2 h, and AA metabolites released into the medium were extracted and measured by reverse-phase-HPLC. As shown in Fig. 7A, balloon-injured arteries that were transduced with Ad-15-LOX1 produced 15-HETE as a major eicosanoid, and it was at least 3-fold higher than that produced by balloon-injured Ad-GFPtransduced arteries. In parallel experiments, 3 days after balloon injury and transduction with Ad-GFP or Ad-15-LOX1, the arteries were isolated, and RNA was extracted and analyzed for MCP-1 mRNA levels by RT-PCR. Although no MCP-1 mRNA expression was observed in uninjured arteries, balloon injury resulted in its induction by severalfold (Fig. 7B). In addition, balloon-injured arteries that were transduced with Ad-15-LOX1 expressed MCP-1 mRNA levels at least by 2-fold higher than that in balloon-injured Ad-GFP-transduced arteries (Fig.  7B). Furthermore, immunostaining of the cross-sections of carotid artery with anti-MCP-1 antibodies revealed the presence of more MCP-1-positive staining in the medial region of balloon-injured and Ad-GFP-transduced arteries as compared with uninjured arteries (Fig. 7C). MCP-1-positive staining was found to be even much higher in Ad-15-LOX1-transduced arteries as compared with Ad-GFP group in response to injury (Fig. 7C). To find whether the enhanced expression of MCP-1 in balloon-injured Ad-15-LOX1-transduced arteries occurs in SMC and whether this effect is mediated by Src-STAT-3 signaling in vivo, soon after injury arteries were transduced with Ad-GFP or Ad-15-LOX1 in combination with and without Ad-dnSrc or Ad-dnSTAT-3 (10 10 pfu/ml); 3 days later arteries were dissected out, cross-sections were made, and double immunofluorescence staining for MCP-1 and SMC␣-actin was performed. Double immunofluorescence studies revealed that balloon injury-induced and 15-LOX1-enhanced MCP-1 expression occurs in SMC of injured arteries (Fig. 8). In addition, transduction of either Ad-dnSrc or Ad-dnSTAT-3 substantially reduced the expression of MCP-1 in balloon-injured Ad-15-LOX1-transduced arteries, suggesting that Src and STAT-3 mediate injury-induced 15-LOX1-enhanced MCP-1 expression in the artery (Fig. 8). To find whether the 15-HETE-MCP-1 axis causes the recruitment of macrophages, these carotid artery sections were also stained for CD68, a marker for macrophages. Surprisingly, compared with uninjured artery where only a very few CD68-positive cells are present in the adventitia, balloon injury induced the infiltration of more of these cells into both adventitia and media in Ad-GFP-transduced arteries. In addition, the presence of CD68-positive cells was even further enhanced in Ad-15-LOX1-transduced arteries in response to injury (Fig. 9). To test the role of 15-LOX1 on SMC migration in vivo, arteries that were isolated 3 days after balloon injury and transduction with Ad-GFP or Ad-15-LOX1, were fixed, opened longitudinally, and stained for SMC␣-actin.
The SMC␣-actin-positive cells were counted as a measure of SMC migration from media to intima. Compared with Ad-GFP group, adenovirus-mediated transduction of 15-LOX1 led to a 1.6-fold increase in the migration of SMC from media to intima (Fig. 10A). To test the effect of 15-LOX1 on neointima formation, soon after balloon injury arteries were transduced with Ad-GFP or Ad-15-LOX1 (10 10 pfu); two weeks later the arteries were isolated and fixed, cross-sections were made and stained with hematoxylin and eosin, and morphometric analysis was preformed. Adenovirusmediated expression of 15-LOX1 enhanced neointima formation by 35% as compared with Ad-GFP group (Fig. 10B).

DISCUSSION
Vascular wall remodeling is a slow progressing disease process in atherosclerosis and restenosis (1,2). Although the risk factors for these diseases have been identified by epidemiological studies, the disease initiation and progression mechanisms still remain unclear (3,48). Since the discovery of increased colocalization of 15-LOX mRNA and protein with epitopes of oxidized low density lipoprotein particles, growing numbers of experimental evidence showed an important role for lipoxygenases in these diseases (15, 19, 49 -51). However, most of these studies have focused on 12/15-LOX, a murine ortholog of human 15-LOX1 that converts AA preferentially to 12(S)-HETE. In contrast, human 15-LOX1 converts AA mainly to 15(S)-HETE (9, 10). In fact, several studies have now shown that 15-HETE is produced preferentially by, and accumulated in human and rabbit atherosclerotic arteries (20,52,53). Thus, the importance of 15-LOX1-15(S)-HETE in human vascular disease has been underscored. Toward this end, the present study examines the potential role of 15(S)-HETE in the regulation of VSMC migration, a hallmark event associated with the pathogenesis of restenosis after angioplasty and fibrous plaque formation at the early stages of atherosclerosis (1,3,47). Our results show that 15(S)-HETE is a potent chemoattractant for VSMCs both in vitro and in vivo. In addition, the major mechanism underlying 15(S)-HETE chemotactic activity was found to be Src-dependent STAT-3-mediated MCP-1 expression. Cloning and characterization of MCP-1 promoter revealed the presence of several STAT-binding sites spanning a 2.6-kb region. However, by comparison of luciferase activity of MCP-1 promoter constructs with serial 5Ј deletions and site-directed mutagenesis, we found that the STAT-binding site proximal to transcriptional start site is essential for 15(S)-HETE-induced Src-dependent STAT-3-mediated MCP-1 expression. This conclusion is supported further by the findings that STAT-3 binds directly to MCP-1 promoter in vivo in

STAT-3 Mediates 15(S)-HETE-induced MCP-1 Expression
response to 15(S)-HETE treatment and that interference with Src-STAT-3 activation prevents its binding. It is interesting to note that the basal promoter activity observed with 2.6-kb promoter fragment was decreased with deletions between the 1.9and 0.5-kb region, suggesting the presence of a possible repressor(s) in this region of the promoter. The recovery of the basal promoter activity and its enhancement by 15(S)-HETE with the 0.2-kb promoter fragment compared with the 0.5-kb fragment may suggest inactivation and/or displacement of a possible repressor in the Ϫ1.9to Ϫ0.5-kb region by this eicosanoid, leading to STAT-3-mediated induction of the promoter activity. This observation may also indicate that in addition to STAT-3, other mechanisms are involved in the regulation of MCP-1 promoter activity, at least to some level by 15(S)-HETE. It is also noteworthy to point out that although MCP-1 mRNA levels were prevalent even at 6 h of treatment with 15(S)-HETE, its protein levels were decreased to almost basal levels. This observation leads to a speculation of transcript-initiated translational silencing of MCP-1 expression during the termination phase of its induction by 15(S)-HETE. Such a transcript-mediated translational silencing of gene regulation was also observed in the case of ceruloplasmin by ␥-interferon in macrophages (54).
To understand the pathophysiological relevance of our in vitro findings in vivo, we overexpressed 15-LOX1 in ballooninjured rat carotid arteries and studied its capacity on AA con-  These observations suggest that 15-LOX1 and its AA metabolite, 15(S)-HETE, play an important role in vascular wall remodeling in response to injury. VSMC migration from media to intima is also a critical event in vascular wall remodeling in atherosclerosis (1-3, 23, 47). The enhanced capacity of human atherosclerotic arteries to convert AA mainly to  and the detection of increased 15-LOX1 expression and 15-HETE levels in rabbit atherosclerotic arteries (52,53) suggest that this eicosanoid may also be an important lipid mediator of VSMC migration in atherosclerosis. Based on these observations and the present findings, it is quite likely that in addition to its earlier identified role in the oxidation of low density lipoprotein (15), 15-LOX1 via its capacity to produce 15-HETE may play an important role in the stimulation of SMC migration in this vascular disease process. Several cytokines and chemokines were reported to be produced by various cell types, including SMCs and macrophages, during atherogenesis (55). Among these, MCP-1 has been observed to play a contributing role in atherogenesis (23). It is also interesting to note that increased numbers of CD68-positive cells were observed in injured arteries that were transduced with Ad-15-LOX1 as compared with Ad-GFP group. It appears that 15-LOX1-15-HETEinduced MCP-1 expression in the medial region may influence the recruitment and homing of macrophages. The capacity of 15-LOX1-15-HETE axis to induce the expression of MCP-1 in vivo in intact arteries in response to injury further lends a strong support for the role of 15-HETE in atherosclerosis. It was reported that MCP-1 and its receptor CCR2 play an important role in neointima formation (56). In addition to its role as a potent chemotactic factor for monocytes and macrophages, MCP-1 also exerts both chemotactic and mitogenic effects on VSMCs (56). The MCP-1 produced by SMCs of the medial region in Ad-15-LOX1-transduced arteries via its chemotactic activity could account for the recruitment and homing of macrophages into this region. In addition, upon recruitment and activation, or Ad-15-LOX1 (10 10 pfu) in combination with and without Ad-dnSrc (10 10 pfu) or Ad-dnSTAT-3 (10 10 pfu), injured and uninjured common carotid arteries were isolated and fixed, and cross-sections were made and immunostained for colocalization of MCP-1 and SMC␣-actin using anti-MCP-1 and anti-SMC␣-actin antibodies followed by fluorescent-labeled secondary antibodies. In the case of negative controls, incubation with primary antibodies is omitted. DAPI, 4,6-diamidino-2-phenylindole. these macrophages could produce various cytokines, which in turn enhance SMC migration from media to intima and their proliferation in intima forming neointima.
STATs play a major role in the regulation of cell migration and proliferation during development and disease process (34,35,57,58). In our earlier studies we have shown that STAT-3 and STAT-5B, by mediating cyclin D1 expression, play a role in thrombin and platelet-derived growth factor-BB-induced VSMC migration and/or proliferation (36,37). Krymskaya et al. (59) reported that Src, via influencing the phosphatidylinositol 3-kinase pathway, plays an essential role in the regulation of VSMC migration and proliferation in response to platelet-derived growth factor-BB and epidermal growth factor. Previous studies from our laboratory showed that Src mediates STAT-3 phosphorylation in microvascular endothelial cells, facilitating their migration and tube formation in response to 15(S)-HETE (25). However, as reported by other studies, Src can also mediate cell migration independent of STAT-3 (60). Vice versa, STATs can be targeted by other non-receptor tyrosine kinases such as the Jaks in the regulation of cell migration. In support of Jak-STAT signaling in the regulation of cell motility, we reported that 15(S)-HETE-induced human retinal microvascu- FIGURE 9. Transduction of 15-LOX1 into balloon-injured arteries enhances the migration and homing of macrophages in the injured arteries. Three days after balloon injury and transduction with Ad-GFP (10 10 pfu) or Ad-15-LOX1 (10 10 pfu), injured and uninjured common carotid arteries were isolated and fixed, and cross-sections made and stained with anti-CD68 antibodies. In the case of negative controls, incubation with primary antibodies is omitted. *, p Ͻ 0.05 versus uninjured artery (n ϭ 6); †, p Ͻ 0.05 versus Ad-GFP-BI (n ϭ 6). FIGURE 10. Transduction of 15-LOX1 in rat carotid arteries leads to enhanced migration of SMC from media to luminal surface leading to neointima formation. A, 3 days after balloon injury and transduction with Ad-GFP (10 10 pfu) or Ad-15-LOX1 (10 10 pfu), injured common carotid arteries were isolated, opened longitudinally, and stained with anti-SMC␣-actin antibodies. B, 2 weeks after balloon injury and transduction with Ad-GFP (10 10 pfu) or Ad-15-LOX1 (10 10 pfu), rats were sacrificed, arteries were isolated and fixed, and cross-sections were made and stained with hematoxylin and eosin. After morphometry analysis, intimal/medial (I/M) ratios and lumen sizes were calculated. The upper panel shows the representative pictures of ballooninjured Ad-GFP-or Ad-15-LOX1-transduced carotid artery cross-sections that were stained with hematoxylin and eosin. The bar graphs in the bottom panel show the quantitative analysis of the intimal/medial ratios and lumen sizes of the balloon-injured Ad-GFP-or Ad-15-LOX1-transduced rat carotid arteries. †, p Ͻ 0.05 versus Ad-GFP-BI (n ϭ 6). NOVEMBER 6, 2009 • VOLUME 284 • NUMBER 45 lar endothelial cell migration and tube formation require Jak2mediated STAT-5B-dependent IL-8 expression (61). In regard to other mechanisms of eicosanoid-induced VSMC motility, we have shown that cyclic AMP response element-binding protein-mediated IL-6 production is needed for 15(S)-HETE-induced VSMC migration (62). In addition, we demonstrated that adenovirus-mediated transduction of 15-LOX2 that converts AA exclusively to 15-(S)-HETE enhances injury-induced expression of IL-6 and, thereby, SMC migration from media to intimal region forming neointima (62). In the present study we found that Src-STAT-3-mediated MCP-1 expression is necessary for 15(S)-HETE-induced VSMC migration in vitro and transduction of 15-LOX1 exacerbates the injury effect on MCP-1 expression, SMC migration from media to intimal region, and neointima formation in vivo. Based on these observations, it appears that 15(S)-HETE influences the induction of expression of several cytokines, including MCP-1 and IL-6, via activation of various signaling pathways in the stimulation of VSMC migration. Because neutralization of either MCP-1 or IL-6 almost completely suppressed 15(S)-HETE-induced VSMC migration, it is possible that both molecules via either a cross-talk or stimulating parallel signaling events mediate migration of these cells in response to this pathophysiologically important oxidized lipid molecule of polyunsaturated fatty acids. It is also intriguing to note that neutralizing anti-MCP-1 antibodies, in addition to suppressing 15(S)-HETE effects on VSMC migration, blocked basal VSMC migration as well. This finding suggests that VSMCs produce MCP-1 constitutively at least to some level. Despite the capacity of 15(S)-HETE to stimulate various signaling pathways in exerting its chemotactic effects, at least in vascular wall cells, the far upstream mechanisms of its actions at the cell surface level are yet to be explored.

STAT-3 Mediates 15(S)-HETE-induced MCP-1 Expression
In conclusion, in the present study we report for the first time that the 15-LOX1-15(S)-HETE axis, via influencing the expression of MCP-1 by Src-STAT-3 signaling, plays a determinant role in the stimulation of SMC migration from media to intima, leading to neointima formation in response to injury.