Efiects And Underlying Mechallisms Of Tumor Necrosis Factor Alpha Converting Enzyme On Atherosclerosis

BackgroudAn abundant increase in infl ammatory factors leads to acute inflammatory diseases, such as acute coronary syndrome (ACS). C-reactive protein (CRP), the prototypic marker of inflammation, is one of the strongest predictors of cardiovascular events. Recent studies have demonstrated that CRP is present in atherosclerotic plaques and plays a pivotal role in promoting atherogenesis by regulating the expression and release of inflammatory cytokines.Lectin-like oxidized low density lipoprotein receptor-1(LOX-1), a type Ⅱ membrane glycoprotein acting as a receptor for oxidized low-density lipoprotein, mediates vascular dysfunction. LOX-1is expressed on the cell surface and can be proteolytically cleaved at its extracellular domain and released as a soluble form (sLOX-1).sLOX-1level reflects increased oxidative stress of vascular walls and has been identified as a novel marker for early diagnosis of ACS. However, the exact mechanisms of sLOX-1release from the cell membrane are poorly understood.Tumor necrosis factor-a converting enzyme (TACE), a disintegrin and metalloproteinase, mediates the release of growth factors, receptors, and adhesion molecules. TACE is synthesized in a latent form and activated by reactive oxygen species (ROS) before reaching the cell membrane. Studies report that CRP can upregulate ROS production by activating NAPDH oxidase in macrophages. However, it is still unknown whether CRP could stimulate sLOX-1release by activating TACE.Objective:(1) To determine wehther CRP could stimulate sLOX-1release from activated macrophages(2) To determine wehther CRP could stimulate sLOX-1release from macrophages via Fc g receptor gamma Ⅱ(Fc g RII)-mediated p47phox phosphorylation, ROS production, and subsequent TACE activation.Materials and Methods1. ReagentsTrypsin, penicillin/streptomycin solution (P/S), fetal bovine serum (FBS), RPMI1640, and Opti-MEM were from Life Technologies. LOX-1ELISA kit was from R&D Systems. Recombinant human CRP, also from R&D Systems, is endotoxin-free (endotoxin level<0.1endotoxin units/1μg protein) and azide-free. The protein extraction kits were from Millipore or Biovision.2. Cell culture and treatmentTHP-1cells purchased from the American Type Culture Collection were grown in RPMI1640medium with10%FBS and1%P/S. For monocyte differentiation, cells were seeded in culture plates at2×106cells/1ml per well and allowed to adhere and differentiate overnight at37℃in the presence of100nM phorbol myristate acetate (PMA, Sigma). Macrophages derived from THP-1cells were treated with TNF-a (5ng/ml) for12h in serum free medium and were then divided into control and CRP treatment groups. In addition, macrophages in CRP treatment group were pretreated with siRNA oligonucleotides48h before CRP treatment while inhibitors, including N-acetylcysteine (NAC,10mM, Sigma), apocynin (Apo,10mM, Sigma) and tumor necrosis factor-a protease inhibitor1(TAP1-1,100μM, Peptides International Inc.), phenylmethyl sulfonylfl uoride (PMSF,3mM, Sigma), were added1h before CRP treatment. To identify the effect of the CRP to be specifi c, polymixin B sulfate (10μg/ml, Amerson) and boiled CRP (25μg/ml) were used. Cell viability was>95%in all experiments.Peripheral blood samples were obtained from six normal subjects and six patients with ACS after their informed consent. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll density gradient centrifugation and cultured in RPMI1640medium with20%autologous serum at37℃in5%CO2for seven days to induce differentiation into macrophages. Then macrophages from PBMCs were stimulated similarly to the macrophages from THP-1cells but without TNF-a treatment.Cell-free supernatants were harvested and kept at-80℃for cytokine measurements. The experiment was approved by the Committee of Shandong University.3. In vivo experimentTwenty New Zealand white rabbits,8weeks old, underwent balloon-induced endothelial injury in the abdominal aorta after being anesthetized with3%pentobarbital sodium (30mg/kg). They were fed a high cholesterol diet for12weeks to establish an animal model of atherosclerosis as described previously. At the end of week12, rabbits were fasted for24h and randomly divided into group A (n=10) and group B (n=10). Group A received intravenous injection of human recombinant CRP (3mg/kg), and group B received intravenous injection of saline. Blood samples were obtained6h after injection, and levels of human CRP in the serum were measured by a highly sensitive nephelometric assay. The experiment complied with the Animal Management Rule of the Ministry of Public Health, People's Republic of China (documentation55,2001) and was approved by the Animal Care Committee of Shandong University.4. ELISAsLOX-1in the culture supernatants or serum was measured by ELISA assay according to the manufacturer's instruction. The level of sLOX-1in the culture supernatants was measured from2×106cells/l ml per well. 5. RNA interferenceMacrophages were washed with Opti-MEM and incubated for2h with Opti-MEM medium before being transfected with siRNA oligonucleotides (Santa Cruz Biotechnology). The cells were incubated with siRNA of TACE (200nM), p47phox (100nM), CD32(100nM), CD64(100nM) or control (200/100nM) using Lipofectamine2000(5μL/ml, Invitrogen) or Endo-Porter (6μL/ml) as the transfection reagent. After12hr transfection. the Opti-MEM medium was re placed with fresh RPMI1640(containing10%FBS/20%autologous serum), and the samples were incubated for another36h.6. Determination of intracellular ROS productionIntracellular ROS generation was measured by use of2',7'-dichlorodihydrofl uorescein diacetate (DCF-DA, Sigma), a cellpermeable indicator for ROS. After cells were treated with TNF-a (5ng/ml) for12h, they were incubated with antibody against CD32(3μg/ml) or with NAC (10mM), Apo (10mM) for1h, or with siRNA oligonucleotides for48h, followed by CRP (25μg/ml) challenge for the indicated times (10min to1h), then loaded with DCF-DA (10μM) for10min at37℃. ROS-mediated fluorescence was detected by use of a fluorescence microplate reader (Varioskan Flash, Thermo Scientific) with excitation498nm and emission520nm.7. Measurement of TACE activityTACE activity of cell lysates was measured by use of the Sensolyte520TACE activity assay kit (Anaspec) according to the manufacturer's instruction. Briefly, after macrophages were collected in assay buffer, incubated for10min at4℃and centrifuged at2500g for10min at4℃, the supernatants were collected. Equal amounts of cell lysates were incubated with50μl TACE substrate for30min at37℃, and changes in fluorescence were monitored by the fluorescence microplate reader with excitation490nm and emission520nm. Fluorescence quenching was used to calculate percentage activity with the appropriate control values.8. Western blot analysis Extracts containing cytoplasmic, membrane, or total proteins were separately prepared according to the manufacturer's instructions. Equal amounts of cytosolic, membrane, or total protein extracts were separately subjected to Western analysis with antibodies against LOX-1(1:250, R&D Systems), TACE (1:200, Abeam), phosphorylated p47phox (1:200, Syd Labs), CD32(1:200, Santa Cruz Biotechnology), CD64(1:200, Santa Cruz Biotechnology), Gas(1:400,Santa Cruz Biotechnology), and b-actin (1:1000, Santa Cruz Biotechnology). The antigen-antibody complexes were detected by enhanced chemiluminescence. All blots were probed with G a s or β-actin as a loading control, and densitometric analysis was performed with an image analyzer (Alphalmager2200, Alpha).9. Real-time PCRTotal RNA was isolated from macrophages by use of TRIzol Reagent (Invitrogen) and treated with DNase (Ambion) to remove contaminating genomic DNA. cDNA was prepared from500ng RNA by use of PrimeScriptTM Reverse Transcriptase (Takara Bio Inc.) according to the manufacturer's instructions. Real-time PCR reactions involved the SYBR Green method for45cycles with a LightCycler (Roche), and a melt curve analysis was performed after each reaction to verify that primer dimers were absent. Data analysis was performed with LightCycler Software4.0(Roche) and the2-ΔΔCT method was used to assess the relative mRNA expression level normalized to that of GAPDH.10. Statistical analysisAll experiments were repeated at least three times. Data are presented as mean±SD. Data analysis was performed with oneway ANOVA followed by Newman-Keuls posthoc test or unpaired Student's t-test. P<0.05was considered statistically significant.Results1. CRP stimulated sLOX-1release from macrophages activated by TNF-aThe LOX-1protein expression of macrophages was low at baseline and was significantly upregulated after TNF-α(5ng/ml) treatment for12hr. Incubating macrophages with CRP (2.5～25μg/ml) for a further5hr after TNF-a treatment resulted in dose-dependent increase in sLOX-1levels, with a stepwise and significant increase from the dose of10μg/mL. However, exposure of macrophages to CRP (25μg/mL) without TNF-a stimulation did not affect the sLOX-1level. Moreover, boiled CRP and polymixin B sulphate produced no effects on the sLOX-1levels induced by CRP. Furthermore, CRP (25μg/ml) treatment for5hr caused a time-dependent decrease in membrane-bound LOX-1(mLOX-1) levels and increase in sLOX-1levels, but the cytoplasmic LOX-1(cLOX-1) protein levels were unaffected. Pretreating activated macrophages with PMSF (3mM), an inhibitor of serine protease, could attenuate the sLOX-1increase induced by CRP. Furthermore, CRP (25μg/ml) treatment for6hr significantly upregulated LOX-1mRNA expression but had no effect on the LOX-1protein expression, possibly due to the fact that6hrs of CRP treatment were too short for LOX-1protein to enhance expression. These results indicated that CRP specifically induced sLOX-1release from activated macrophages but this effect could be blocked by protease inhibitor.2. TACE regulated sLOX-1release from activated macrophages induced by CRPTo determine whether TACE was involved in sLOX-1release, we designed loss-of function experiments with the TACE inhibitor TAPI-1(100μM) and TACE siRNA (200nM). TACE siRNA successfully knocked down the expression of TACE at both mRNA and protein levels. Pretreatment with TAPI-1and TACE siRNA effectively prevented the increase of sLOX-1levels induced by CRP. However, CRP (25μg/ml) treatment for5hr had no effect on the TACE protein expression but enhanced TACE activity in a time-dependent manner with the peak TACE activity by47.5%after5-hr treatment. Furthermore, the activity of TACE induced by CRP was attenuated by pretreatment with TAPI-1(100μM). Thus, TACE might be involved in sLOX-1release from activated macrophages induced by CRP. On the other hand, TACE siRNA had no effect on LOX-1 mRNA expression levels.3. CRP stimulating sLOX-1release from activated macrophages was mediated by ROS generated from NADPH oxidaseWe observed that exposure of macrophages to CRP (25μg/ml) led to a time-dependent increase of intracellular ROS generation, with a maximal response at40min. Moreover, pretreatment with ROS scavenger NAC (10mM) significantly decreased sLOX-1levels induced by CRP. Similarly, pretreatment with the NADPH oxidase inhibitor Apo (10mM) markedly reduced ROS production and sLOX-1levels induced by CRP. In contrast, exposure of unactivated macrophages to Apo or NAC did not affect basal DCF fluorescence (P>0.05).NADPH oxidase is a multi-component protein whose activation depends on phosphorylation of the cytosolic subunit p47phox and its translocation to the cell membrane. CRP (25μg/ml) led to a time-dependent increase in the phosphorylation and translocation of p47phox to the cell membrane, with a maximal response at20min. We also found that p47phox siRNA successfully knocked down the expression of p47phox protein and that pretreatment with p47phox siRNA (100nM) successfully decreased sLOX-1levels induced by CRP. Furthermore, pretreatment with NAC (10mM), Apo (10mM) and p47phox siRNA (100nM) effectively attenuated TACE activity induced by CRP. Thus, sLOX-1release induced by CRP was mediated by ROS generated from NADPH oxidase. However, incubation with p47p ox siRNA partially suppressed LOX-1mRNA levels but had no effect on LOX-1protein expression after6-hr CRP treatment, suggesting that the pathway by which ROS and p47phox enhance LOX-1mRNA expression is probably different from those by which ROS and p47phox increase sLOX-1release.4. sLOX-1release from activated macrophages induced by CRP was mediated by FcγRⅡCRP has been reported to bind to the family of FcyRs in human macrophages, including FcyRI (CD64) and FcyR II (CD32). Our results demonstrated that FcyR siRNA effectively abolished FcyR protein expression. Pretreatment of activated macrophages with antibody against CD32(3μg/ml) or with CD32siRNA (100nM) effectively prevented sLOX-1release induced by CRP. However, pretreatment of activated macrophages with antibody against CD64(3μg/ml) or IgG (3μg/ml) or with CD64siRNA (100nM) did not affect sLOX-1release induced by CRP. Moreover, ROS production, TACE activation and p47phox protein phosphorylation and translocation were all suppressed by pretreatment with antibody against CD32(3μg/ml) or with CD32siRNA (100nM). In addition, antibodies against both CD32and CD64showed significant suppression of LOX-1mRNA expression but exerted no effects on LOX-1protein expression within6-hr of CRP treatment.5. CRP regulated sLOX-1release from macrophages derived from PBMCs of ACS patientsHaving established the role of CRP in activated macrophages derived from THP-1cells, we went on further to examine the effects of CRP on macrophages derived from PBMCs of normal subjects and ACS patients. We observed that CRP enhanced sLOX-1release from macrophages derived from PBMCs of ACS patients; in addition, pretreatment with antibody against CD32(3μg/ml), or with NAC (10mM), Apo (10mM), p47phox siRNA (100nM), TAPI-1(100μM) and TACE siRNA (200nM) attenuated sLOX-1release induced by CRP. However, CRP had no effects on sLOX-1release from the macrophages of normal subjects. Therefore, a signaling pathway of FcyRII-NADPH oxidase (p47phox)-ROS-TACE was involved in sLOX-1release from activated macrophages induced by CRP.6. CRP elevated sLOX-1levels in vivo in a rabbit model of atherosclerosisWe further examined whether CRP enhanced sLOX-1levels in vivo by use of a rabbit model of atherosclerosis. Serum levels of human CRP were higher in group A than in group B (16±4μg/ml vs. undetected, P<0.001). Likewise, circulating sLOX-1levels were significantly higher in group A than in group B; median circulating sLOX-1level increased1.68-fold:64.3pg/mL (interquartile range59.9to70.7pg/mL) in group A versus39.1pg/mL (34.0to52.0pg/mL) in group B (P<0.01).Conclusions:1. CRP stimulated sLOX-1release from macrophages activated by TNF-a.2. TNF-a converting enzyme mediates soluble LOX-1release from activated macrophages.3. The signaling pathway of FcγRⅡ-NADPH oxidase (p47phox)-ROS-TACE might be involved in sLOX-1release from activated macrophages induced by CRP. BackgroundPathological features of atherosclerosis can briefly be characterized as chronic inflammation, increased neoangiogenesis, enhanced oxidative stress, and disruption of the extracellular matrix (ECM). In terms of inflammation, atherosclerosis begets from the initial phases of leukocyte recruitment to eventual rupture of vulnerable atherosclerotic plaque, while inflammatory mediators play key roles in the pathogenesis process of atherosclerosis. Moreover, matrix metalloprotease, especially MMP2and MMP9, play crucial roles in plaque instability and subsequent rupture by promoting matrix degradation.Tumor necrosis factor alpha converting enzyme (TACE) which was initially discovered as the protease that cleaves the26-kDa precursor of TNF-a to yield the TNF soluble form. It was later recognized that TACE activity is also responsible for proteolytic cleavage of numerous other transmembrane proteins such as growth factors, growth factor receptors, cytokines. cytokine receptors and adhesion molecules. Some studies provided functional evidence for TACE as a candidate gene of atherosclerosis susceptibility, other studies showed that polymorphisms of the TACE genes had relation to cardiovascular mortality. TACE expression increased in aortic lesions together with their development in atherosclerotic leisions of ApoE-/-mouse. In human atherosclerotic samples, there was no TACE expression in nomal arteries, but local levels of TACE at the site of ruptured plaque were much higher. All those implicated that TACE might play important roles in the plaque rupture, but the mechanisms are still unknown. We would discuss whether TACE took part in atherosclerotic rupture and the probable mechanism.Objective1To determine whether TACE was involved in the atherosclerotic plaques unstability.2To determine the probable molecular mechanism of TACE regulating atherosclerotic plaques unstability.Materials1Small hairpin RNAs interference and lentivirus constructWe used the pGLV-U6-EGFP (pGLV1-1), which contained the U6expression cassette, an RNA polymerase Ⅲ-dependent transcription of shRNA transcript. This vector also expressed green fluorescent protein from a cytomegalovirus promoter, which allowed for monitoring of the efficiency of transfection. Small hairpin RNAs were designed to contain21-nucleotide sense sequences identical to the target molecule(s), followed by a short (7-nucleotide) nonspecific loop sequence and an antisense sequence, followed by two thymidines, which serve as a stop signal for RNA polymerase III. The oligonucleotides were annealed and cloned into pGLV1-1between the HpaI and Xhol sites. One small hairpin RNA construct that resulted in nearly complete inhibition of TACE mRNA expression in rabbit abdominal smooth muscle cells (rSMCs) was chosen for use in our studies. This construct targets the sequence5'-ggatttaaaggttatggaata-3'in TACE splice variant. As a control, we used21-nucleotide scrambled small hairpin RNA, which did not give more than a18-nucleotide match against any rabbit genomic sequence.2Rabbit abdominal SMCs culture and Lentivirus transfection8-week old rabbit abdominal aorta were dissected and washed twice in fresh phosphate buffered saline (PBS) to remove any blood. The abdominal aorta were dissected into small pieces, suspended in0.25%trypsin-EDTA (Invitrogen), and incubated for30min at37℃and then were incubated in lmg/ml collagenase of type I,0.5mg/ml elastase and1.25mg/ml trypsinfor45min to1h, after that, they were centrifuged at3000rpm for5min, finally, the supernants were incubated in Dulbecco modified Eagle medium (DMEM) containing4,500mg of glucose/liter,0.1mM nonessential amino acids,2mM L-glutamine, and10%fetal bovine serum (FBS, Gibico) and incubated at37℃with5%CO2. Rabbit abdominal SMCs were passaged after85-95%confluence. SMCs were randomly divided into four groups:control group (2X106cells receiving2X106TU/ml lentivirus-scramble shRNA), TACE shRNA group (2×106cells receiving2×106TU/ml lentivirus-TACE shRNA), TNF-a group (2×106cells receiving10ng/ml TNF-a for24hrs followed by2×106TU/ml lentivirus-scramble shRNA), TACE shRNA+TNF-a group (2×106cells receiving10ng/ml TNF-a for24hrs followed by2×106TU/ml lentivirus-TACE shRNA).3Atherosclerotic animal model and Lentivirus-mediated TACE shRNA transfection in vivoFifty-seven adult male New Zealand White (NZW) rabbits (～1.9-2.1kg) received balloon induced abdominal aorta endothelium injury (3.5mm; three times,14-16atm inflations) under general anaesthesia with3%pentobarbital sodium (30mg/kg) and then were fed with1%cholesterol diet for16weeks. These rabbits were randomly divided into3groups:TACE shRNA group (n=19, receiving0.2mL of1×109TU/mL recombinant Lentivirus-TACE shRNA), mock group (n=19, receiving0.2mL of2×109TU/mL recombinant Lentivirus-scramble shRNA), and control group (receiving0.2mL PBS). Lentivirus and PBS were transfected using intraplaque injection method as before reported in our laboratory at week8. GFP gene expression of atherosclerotic plaques was detected1week,2weeks and8weeks after transfection. Blood samples of the three groups were collected at week16and plasma was stored at-80℃. In addition, ten adult male NZW rabbits (～1.9-2.1kg) received balloon induced abdominal aorta endothelium injury and were fed with1%cholesterol diet for16weeks. Sections from these10rabbits were used for plaques differentiation. The experiment complied with the Animal Management Rule of the Ministry of Public Health, People's Republic of China (documentation55,2001) and obeying guidelines approved by the Animal Care Committee of Shandong University.4Intravascular ultrasound (IVUS) examinationIVUS study was performed at week16by use of a3.2F catheter containing a single rotating element transducer of40MHz connected to an IVUS system (Galaxy, Boston Scientific Corp, Boston, MA, USA). IVUS study was performed according to the standard procedure. The resolution of the image frame was150mm and300mm respectively in axial and lateral direction. After the catheter reached the left renal artery, it was withdrawn by use of a motorized pullback device at a constant speed of0.5mm·s-1until to8cm below the left renal artery. External elastic lamina (EEL) area and lumen area were measured from cross-sectional images and remodeling index (RI) was calculated as lesion segment EEL area/reference segment EEL area. Plaque burden was calculated as (EEL area-lumen area)/EEL area×100%. The segments near the plaque with the least lesion within a10-mm distance from the plaque were regarded as the reference segments, and the mean EEL area derived from five different reference segments or lesion segment was calculated. Plaques with RI>1.05was regarded as positive remodeling, RI=0.95-1.05as no remodeling, and RI<0.95as negative remodeling. The IVUS images were reviewed by two independent observers and the averaged values were used for data analyses.5Histopathology examinationAfter IVUS examination at week16, animals were administered an overdose of sodium pentobarbital followed by intravenous injection with heparin (500U/kg), after that, the abdominal segments from the left renal aorta to8cm below it was excised, rinsed with PBS and cut in cross section at3to6mm intervals. Some nonadjacent intervals of per abdominal segments were post-fixed in4%buffered paraformaldehyde and embedded in paraffin or OCT (frozen tissue matrix) for frozen sections, other segments were stored at-80℃. Structural changes of the atherosclerotic vessel wall were seen in sections stained with HE staining, Picrosirius staning, oil red O staining and immuno-histochemistry. Plaque vulnerable index was calculated as the ratio between the two histological areas in atherosclerotic plaques:the lipid core area (macrophages+extracellular lipid deposits)/the fibromuscular area (smooth muscle cells+collagen).6Definition of atherosclerotic plaquesThe atherosclerotic plaques were classified into two groups based on fibromuscular cap thickness (cut-off value,100μm):stable plaques (>100μm) and unstable plaques (≤100μm). Cap thickness was measured at the point of the thickest lipid core with a clear demarcation of the fibromuscular cap-lipid core interface. The lipid core stained with oil red O staining was defined as a layer more than100μm thick consisting of macrophage foam cells and extracellular lipid deposits. Fibromuscular caps stained with Masson's trichrome were defined as areas consisting of smooth muscle cells and collagen fibers covering the lipid core. Lipid cores and fibromusclar caps were measured with an image analysis system (Image-Pro Plus5.0, Media Cybernetics, USA) attached to a color CCD video camera.7ImmunohistochemistrySerial sections were subjected to immuno-staining for Mouse anti-rabbit RAM11monoclonal antibody (Lab Vision Neomarkers, dilution in1:200), mouse anti-human smooth muscle cell a-actin1A4monoclonal antibody (Boster,1:100). goat anti-human PECAM-1(CD31) polycolonal antibody (Santa Cruz,1:200), goat anti-human TACE polycolonal antibody (Santa Cruz,1:100), goat anti-human P65polycolonal antibody (Santa Cruz,1:200), goat anti-human iNOS polycolonal antibody (Novus Biologicals,1:200), goat anti-human ICAM-1polycolonal antibody (Santa Cruz,1:200), goat anti-human TGF β1polycolonal antibody (Santa Cruz,1:200), and goat anti-human P47polycolonal antibody (Santa Cruz,1:200). Briefly, sections were treated with3%hydrogen peroxide for10minutes, blocked with5%normal BSA at37℃for20minutes, and incubated with the primary antibody at4℃overnight. Incubation with PBS instead of the primary antibody was served as a negative control. The color was developed with the DAB chromogen and counterstained with hematoxylin. For hispathology immunofluorescence, the secondary antibody was marked with fluorescence.8Oil red O stainingLipid content in atherosclerotic lesions were measured by lipid deposition stained with oil red O in frozen sections. To quantify lipid depositions of the abdominal aorta.5μm sequential sections were cut from aorta segments embedded in OCT. We calculated for each aortic ring an average of lipid deposition from nonadjacent20separate sections each animal.9Neovascularization quantitative analysisNeovessels were identified as channels surrounded by a layer of endothelial cells highlighted by immuno-staining with anti-CD31antibody. We calculated the adventitia neovessels in the area encircling the EEL within one×200opitical field. All neovessels in the section were counted by two observers by use of fields at×200magnification by use of Image Pro-plus (Media Cybernetics, USA).10Western Blot AnalysisProtein was extracted from tissues or cell lystes as instructions and protein content was measured by a Bio-Rad protein assay. Equal amounts of protein (30μg/lane) were separated by10%SDS-PAGE and electro-transferred to nitrocellulose membranes. Nonspecific antibody binding is blocked with5%skim milk in TBS. Membrane is incubated with a primary antibody for TACE (1:1000, Santa Cruz), MMP2(1:1500, Santa Cruz), MMP9(1:1500, Santa Cruz), VEGF (1:1000, Abcam), Flt-1(1:1000, R&D), β-actin (1:1500, Santa Cruz) and GFP (1:2000, Santa Cruz) overnight and then with HRP-conju gated secondary antibody for one hour in TBS. After each antibody incubation, the blot is washed (three times,5min each) with TBS containing0.1%Tween20. The antigen-antibody complexes were exposed to chemi luminescent substrate (Millipore) and scanned using an image analyzer (Alphalmager2200, Alpha). All blots were probed with β-actin as a loading control, and densitometric analysis was performed with an image analyzer (Alphalmager2200, Alpha).11Real-time quantitative reverse transcription polymerase chain reaction (RT-PCR)Rabbit mRNA from the abdominal aorta (from the left renal aorta to8cm below it) or from rSMCs was prepared with TRIzol Reagent (Invitrogen, USA) and PrimeScript(?) RT reagent Kit (Takara Bio Inc, Japan) according to the manufacturer's instructions. RT-PCR was used to determine the mRNA levels of TACE, MMP2, ICAM-1, VCAM-1. VEGFA, Fit-1and P47and glyceraldehyde phosphate dehydrogenase (GAPDH). The sequences of primers were shown in Table.2. Melt curve analysis was performed after each reaction to verify that primer dimers were absent. Data analysis was performed with Bio-Rad MyiQTM Single Color Real-Time PCR Detection System (USA) and the2-ΔΔCT method was used to assess the relative mRNA expression level normalized in multiplex reactions using GAPDH control.12Fluorescence enzymatic activity measurementTACE activity of rSMCs or abdominal tissues was measured by use of the Sensolyte520(?) TACE activity assay kit (Anaspec) according to the manufacturer's instruction. Briefly, after rSMCs or tissues were lysed in assay buffer, incubated for10min at4℃and centrifuged at2500g for10min at4℃, the supernatants were collected. Equal amounts of proteins were incubated with50μl TACE substrate for30min at37℃and changes in fluorescence were monitored by Varioskan Flash software (USA) with excitation490nm and emission520nm.13ELISA detection and biochemical assaySoluble inflammatory factors of the plasma were measured by ELISA assay according to the manufacturer's instruction. Biochemical assay was assayed obeying the manufacturer's instruction.14Zymography MethodsMMP activities of abdominal atherosclerotic plaques were assessed with zymography method. Briefly, samples were minced and homogenized in ice-cold10mM PBS (pH7.2) containing150mmol/L NaCl,1%Triton X-100,0.1%SDS,0.5%sodium deoxycholate, and0.2%NaN3. Tissue homogenates were centrifuged at14000rpm for10minutes at4℃and the supernatant was collected. Protein content was measured by a Bio-Rad protein assay, and SDS-polyacrylamide gel electrophoresis zymography was performed. One part of homogenate containing30μg of protein mixed with1part of4×SDS sample buffer containing0.32%Tris-HCL,4%SDS (PH7.2),16%glycerol, bromophenol blue, and molecular weight markers were added. Each sample was loaded on either a10%polyacrylamide gel containing0.1%gelatin.15Statistical analysesValues were expressed as the mean±S.D. Statistical analysis was performed with by one-way ANOVA followed by the Students-Newman-Keuls (SNK) post hoc test. Correlation coefficients were assessed with Spearman rank correlation coefficients. Statistical significance was defined as P<0.05.Results1TACE expression in atherosclerotic plaquesTACE expression was mainly observed in RAM-11-positive areas of atherosclerotic lesions, in addition, intimal smooth muscle cells also expressed. TACE expression was not observed in aortic sections from normal diets rabbits.Levels of TACE expression density in unstable plaques were significantly higher than that in stable plaques (42.6±7.6%vs.25.2±6.5%, P<0.001). Moreover, TACE expression density was positively correlated with neovessels (R2=0.512, P<0.001) and macrophages content (R2=0.705, P<0.001). 2TACE shRNA effectively down-regulated TACE expression and inflammatory factors expression in rSMCs activated by TNF-aTo evaluate the efficiency of gene silencing in vitro, TACE expression in rSMCs were measured and shown to be substantially reduced by TACE shRNA in mRNA level and protein level both in basal condition and activated condition (P<0.05). Moreover, the mRNA expression levels of MMP2, ICAM-1, VCAM-1, VEGFA, Flt-1and P47in the TNF-a+TACE shRNA group were significantly lower than those in the TNF-a group (P<0.05).3TACE shRNA effectively down-regulated TACE expression in vivoA local intra-plaque injection method has been proven to be an efficient approach to transfer gene into atherosclerotic plaques. Since GFP expression provides a convenient monitor for checking the transfection efficiency of lentivirus, GFP protein expression in the abdominal atherosclerotic plaques was examined serially, which revealed a successful transfectionl week after transfection, strong expression in the abdominal aorta plaque2weeks after transfection and still expression8weeks after transfection. These results demonstrated an efficient in vivo transfection of lentivirus shRNA in atherosclerotic plaque.To examine the efficacy of lentivirus-mediated gene silencing in vivo, the levels of mRNA and protein expression of TACE in the abdominal atherosclerotic plaques were detected by RT-PCR or immuno-histochemistry. Compared with the control group, the TACE protein expression levels in the TACE shRNA group were lowered by53.5%70.4%and65.1%at1week,2weeks and8weeks after lentivirus transfection, respectively (all P<0.05) and the TACE mRNA expression level in the TACE shRNA group was reduced by61.3%,76.9%and70.1%at1week,2weeks and8weeks after transfection, respectively (all P<0.05). In contrast, there was no significant difference in the TACE mRNA or protein expression levels between the control group and the mock group (P>0.05). Moreover, activities of TACE in the TACE shRNA group were significantly higher than those of the control group and the mock group (P<0.05).4Down-regulating TACE improved atherosclerotic vessel remodeling in the balloon injuried rabbit abdominal aortaIVUS examination showed that external elastic lamina (EEL) area in the TACE shRNA group was significantly lower than t...