A Study On The Mouse Model Of Plaque Rupture And Atherothrombosis And The Relevant Mechanisms

BackgroundAtherosclerotic plaque rupture and subsequent luminal thrombus formation (atherothrombosis) are responsible for most acute coronary syndromes (ACS). An animal model that can consistently reproduce the process of atherothrombosis is essential for mechanistic studies and drug development for coronary heart disease. Although some features of plaque instability are observed in advanced plaques in mice, atherothrombosis cannot be consistently reproduced in atherosclerotic mice even after further manipulation of the plaques, such as overexpressing p53, matrix metalloproteinase (MMP)-9or urokinase.The mechanisms of this'atherothrombosis-resistance'phenomenon in mice are poorly understood. Nonetheless, previous studies have suggested that targeting a single factor or pathway is not sufficient to effectively reproduce atherothrombosis in mice. Mental stress has been recognized as a potential trigger of ACS in humans, which may involve diverse mechanisms. In line with this notion, we have demonstrated that exposure to mental stress induced prominent plaque disruptions in high fat-fed ApoE-/-mice, as evidence by the presence of buried fibrous caps. However, luminal thrombi formation was still rare.In humans, ACS are associated with prothrombotic conditions of the blood, caused by increased platelet activation, dysregulated coagulation factors, and/or decreased anticoagulation factors and endogenous fibrinolysis activity. In ApoE-/- mice, Kuijpers et al. demonstrated that ultrasound treatment of advanced plaque led rapid platelet deposition on plaques, however, followed by a single and small thrombus. This indicates the coagulation dysfunction might be the potential mechanism of atherothrombosis-resistence in ApoE-/-mice. On the other hand, Carrie et al. reported that atherothrombosis was observed in～20%of Apoe/Npc1double knockout mice, which might be related to increased thrombin generation and platelet activation resulted from Npc1deficiency. In light of these findings, we attempt to create a procoagulant status by overexpressing prothrombin, the last protease in the coagulation cascade, in mental stressed ApoE-/-mice to induce atherothrombosis.ObjectivesWe sought to induce atherothrombosis in advanced plaques by overexpressing prothrombin in mental stressed ApoE-/-mice, and elucidate the potential mechanisms involved in plaque rupture and atherothrombosis in this model.MethodsConstruction of a recombinant adenovirus expressing prothrombinRecombinant adenoviruses carrying the murine prothrombin (Ad-ProT) were prepared, and the adenovirus expressing EGFP alone (Ad-EGFP) was used as a control.Animal experimental protocolOne hundred and seventy ApoE-/-mice (male,12-week old) were fed on a high-fat diet and atherosclerotic lesions were induced by placement of a perivascular collar around the right carotid artery. Mice were assigned to four parts of the in vivo study:In the first part (n=20), to examine the temporal changes of the adenoviral transfection efficiency, ApoE-/-mice were euthanized at0day,4days,8days,15days and22days after Ad-ProT transfection (n=4for each time point).In the second part, to examine the effects of prothrombin and stress on advanced carotid plaques, one hundred and four ApoE-/-mice were randomly divided into four groups at the end of week11:unstress, stress, Ad-ProT and Ad-ProT+stress groups (n=26for each). Mice in the unstress and stress groups received an intravenous injection of Ad-EGFP, while others received Ad-ProT transfection. At the end of week12, mice in stress and Ad-ProT+stress groups were exposed to mental stress stimulation exerted by restraint in a well ventilated syringe (50ml) for6hrs per day and5days per week. All mice were euthanized at the end of week14.In the third part, to examine the reproducibility of the model of atherothrombosis and the effects of anti-platelet drugs, thirty-two ApoE-/-mice were randomly divided into two groups (n=16for each) at the end of week11: Ad-ProT+stress and Ad-Pro T+stress with anti-platelet treatment (oral administration of aspirin at5mg/kg/day and clopidogrel at25mg/kg/day during the last3weeks).In the fourth part, to determine whether atherothrombosis was able to be induced in less severe lesions, fourteen mice were treated with Ad-Prot transfection and stress stimulation at the end of week6and7after collar treatment respectively and euthanized at the end of week9.Prothrombotic status evaluationPlasma prothrombin concentration and activity were examined by western blot and Automatic Coagulation Analyzer respectively. Thrombin generation test was performed in vitro to evaluate the plasma coagulation function. The status of platelet activation was assessed with flow cytometry using CD62P as a marker.Histological and morphological analysisSections of the carotid plaques were stained with hematoxylin and eosin at a100μm interval for the assessment of plaque morphology and the idenfication of atherothrombosis, intraplaque hemorrhage. Other sections were stained with oil red O for lipids, picrosirius red for collagen, Verhoeff s stain for elastic fiber, and Perl's stain for ferric iron, and immuno-staining for macrophage, smooth muscle cells and inflammatory factors (IL-1β,1L-6, MCP-1). Apoptosis was assessed with a terminal deoxynucleotidyl transferase end-labeling (TUNEL) kit from Millipore. Arterial cross-sectional lumen narrowing was quantified with the Image-Pro Plus v6.0software (Media Cybernetics, Bethesda, Md).Molecular biological analysisMouse atherosclerosis PCR array experiments were performed to screen the potential pathways involved in the atherothrombosis. The expression of caspase-3, cleaved capase-3and (pro-) MMP-2,-9were detected by western blot with specific antibodies.ResultsProthrombin overexpression increased blood thrombogenicity in miceMost of the Ad-ProT was enriched in the liver and also detectable in the carotid plaques. The transgene expression was detectable4days after injection and reached a peak around15days. Plasma prothrombin activity was significantly enhanced. TGT revealed that the lag time and the time-to-peak were significantly shortened and the peak thrombin was increased by Ad-ProT transfection. Prothrombin overexpression precipitated acute atherothrombotic eventsIn Ad-ProT+stress mice, we identified acute ruptures associated with luminal thrombi (atherothrombosis) in71%(10of14) of the carotid plaques. Of note, we found that5of the10acute atherothrombotic lesions also colocalized with old, partially organized thrombi buried in the same plaque, indicating that these plaques had ruptured for more than once, an important feature of culprit lesions in human coronary arteries. Moreover, we observed severe intra-plaque hemorrhage in4carotid plaques in Ad-ProT+Stress group, which could be identified macroscopically as hemotoma-like lesions in the vessel wall. Three of the four cases of intra-plaque hemorrhage were associated with plaque rupture and atherothrombosis.In contrast to the Ad-ProT+Stress group, mice in the other three groups displayed rare atherothrombosis. Pathological examination revealed2(14%) cases of fresh disruption and8(57%) buried fibrous caps (indicating old plaque disruptions) in the stress group (n=14). However, only2(14%) typical atherothrombosis was observed in these mice. Three (21%) mice in the Ad-ProT group exhibited plaque rupture and atherothrombosis in the carotid plaques, while only2(14%) buried fibrous caps were found in the unstress group (n=14for each group).Plaque destabilization was associated with increased inflammation, apoptosis and metalloproteinasesPCR array revealed that regardless of the presence of prothrombin overexpression, stress upregulated expression of9genes involved in inflammation,1involved in lipid uptake, and downregulated2anti-apoptotic genes. We further demonstrated that the number of apoptotic cells in the plaque and cleaved caspase-3were significantly increased in stressed mice. In addition, we found that both MMP-2and MMP-9were increased in vessels from stressed mice. However, prothrombin overexpression alone only had modest effects on inflammation and apoptosis as compared to the effects of stress, although prothrombin increased MMP expression.Ad-Pro T+Stress mice displayed high reproducibility and relevance to atherothrombosis in humansTo confirm that systemic administration of adenovirus expressing prothrombin could produce consistent results, we tested a different preparation of the viral vector using the same dosage in another Ad-ProT+Stress group. The rates of atherothrombosis and intra-plaque hemorrhage in the carotid plaques were69%(11out of16) and25%(4out of16) respectively, similar with the results of the previous Ad-ProT+Stress group. In order to confirm that our mouse model is relevant to human disease, we tested whether anti-platelet drugs used in clinical practice could prevent the occurrence of atherothrombosis. We demonstrated that co-treatment with aspirin and clopidogrel by oral administration for the last3weeks repressed platelet activation and significantly reduced the incidence of atherothrombosis to19%(3out of16, P<0.05). Aspirin and clopidogrel treatment had no significant effects on the pathological features of the lesion.Onset of atherothrombosis was plaque burden-dependentAt9weeks after collar placement, most of the collared carotid displayed plaques with an obvious necrotic core. However, in14Ad-ProT+Stress mice, only2(14%) exhibited atherothrombosis, including1fresh rupture and1old thrombus. By analyzing the relationship between atherothrombosis and plaque burden in all mice involved in this study, we demonstrated that there was a trend that the rates of plaque rupture and atherothrombosis increased with the level of percent cross-sectional lumen narrowing.ConclusionsCombination of prothrombin expression and mental stress could recapitulate high incidence of plaque rupture and atherothrombosis in carotid plaques of ApoE-/-mice. The episodes of atherothrombosis were associated with plaque vulnerability, plaque burden and blood vulnerability. Increased inflammation, apoptosis and upregulation of metalloproteinases contributed to the development of plaque destabilization and rupture. This approach may provide a novel model for studying mechanisms of acute plaque events and the development of new anti-atherothrombotic therapies. BackgroundTissue factor (TF) is the main initiator of blood coagulation. In contrast to other clotting factors such as prothrombin, TF is abundantly expressed in vascular cells, and serves as the determinant of tissue thrombogenicity. In atherosclerotic lesions, TF is highly expressed in the necrotic core, macrophages, smooth muscle cells and around cholesterol crystal. TF contents and procoagulant activity plays a pivotal role in determining plaque thrombogenicity. Clinical evidence demonstrated that the content of TF in coronary plaques of patients with unstable angina pectoris was markedly higher than that in patients with stable angina pectoris. When plaque ruptured, TF was exposed to the flowing blood and formed a TF/FVIIa complex with clotting factor Ⅷa (FVIIa), which then catalyzed the activation of factor X and IX, finally leading to the formation of a fibrin clot. Recent studies confirmed the existence of blood-borne TF, which mainly originated from the ruptured plaque, and might play an important role in the maintenance of the local thrombus. We have demonstrated the importance of blood-plaque interaction in atherothrombosis in Paper I. Based on these finding, we sought to observe whether atherothrombosis would be induced when plaque thrombogenicity is increased in the blood-plaque interaction.Furthermore, accumulating evidences demonstrated that in addition to coagulation cascade, TF/FⅧa complex elicits multiple effects on a variety of cells via protease-activated receptor-2(PAR2). TF/FⅧa promotes metastasis, progression and angiogenesis in tumors. In cultured smooth muscle cells and macrophage, TF/FVIIa complex upregulates the expression of a number of inflammatory factors. Angiogenesis and inflammation have been confirmed to play pivotal roles in plaque vulnerability and complications. These indicate that TF in plaques might lead to plaque destabilization.In this program, we attempt to test the hypothesis that TF high expression in advanced plaque would not only lead to atherothrombosis, but also promote plaque instability via PAR2.Objectives(1) To induce plaque rupture and atherothrombosis in carotid plaques by overexpressing TF in mental stressed ApoE-/-mice;(2) To elucidate the effects of TF on plaque instability and the potential mechanisms.MethodsTwo in vivo studies and two in vitro studies were involved in this program.In vivo study I. To induce plaque rupture and atherothrombosis in carotid plaques by TF overexpressing and mental stress in ApoE-/-mice1. Construction of a recombinant adenovirus expressing prothrombinRecombinant adenoviruses carrying the murine TF (Ad-TF) were prepared, and the adenovirus expressing EGFP alone (Ad-EGFP) was used as a control.2. Animal experimental protocolForty-eight ApoE-/-mice (male,12-week old) were fed on a high-fat diet and atherosclerotic lesions were induced by placement of a perivascular collar around the right carotid artery. Mice were divided into two groups (n=24for each) at the end of week11: Ad-TF group transfected with Ad-TF, and Ad-EGFP group transfected with Ad-EGFP, and then exposed to restraint stress during week12to week14. All mice were euthanized at the end of week14.3. Histological and morphological analysisSections of the carotid plaques were stained with hematoxylin and eosin at a100μm interval for the assessment of plaque morphology and the idenfication of plaque rupture, atherothrombosis and intraplaque hemorrhage. Other sections were stained with oil red O for lipids, picrosirius red for collagen, Verhoeff s stain for elastic fiber, and Perl's stain for ferric iron, and immuno-staining for macrophage, smooth muscle cells, and inflammatory factors (IL-1β, IL-6, MCP-1). Arterial cross-sectional lumen narrowing was quantified with the Image-Pro Plus v6.0software (Media Cybernetics, Bethesda, Md).4. Molecular biological analysisMouse atherosclerosis PCR array experiments were performed to screen the potential pathways involved in the atherothrombosis. The expression of TF and PAR2were detected by western blot with specific antibodies. In vitro study I. The potential effects of TF on the expression of inflammatory factors in human aorta smooth muscle cells (HASMC) and macrophage via TF/FVJIa/PAR2(1) To observe the effects of TF/FVIIa on the expression of gene selected by PCR array in in vivo study I,e.g. Ccl2, Csf2,114, Lif, Itga2, Selplg, FⅧa was added into the medium at0,10,100and200nM respecitively. After incubation for another12hrs, all cells were harvested for qRT-PCR analysis.(2) To find out whether the potential effects of TF/FⅧa on gene expression were related to the coagulation, medium was added with thrombin inhibitor PPACK at0,1,10and50uM and then added with FVIIa2hrs later. Cells were incubated for another12hrs, and harvested for qRT-PCR analysis. Cells incubated with normal medium were used as controls.(3) To elucidate whether the effects of TF/FⅧa on gene expression were via PAR1or PAR2, medium was added with PAR1antagonist SCH7979(10,100,200nM, respectively) or PAR2antagonist ENMD-1068(1,5,10uM respectively) followed by the addition of FVIIa100nM, or PAR1agonist TFLLR-NH2(10,100,200uM repectively), or PAR2agnoist SLIGKV-NH2(10,100,200uM repectively). Cells were incubated for another12hrs, and harvested for qRT-PCR analysis. Cells incubated with normal medium were used as controls.In vitro study Ⅱ. The effect of inflammatory factors on TF expression in HASMC and macrophage(1) To observe whether LPS would increase the TF expression in HASMC and macrophage, medium was added with LPS at0,1,10,100,200and500ng/ml, respectively. After incubation for another12hrs, all cells were harvested for qRT-PCR analysis.(2) To elucidate the potential mechanism of the effects of LPS on TF expression in HASMC and macrophage, cells were incubated with medium added with the specific inhibitors of Erkl/2(U0126), p38(SB202190) and JNK (JNK inhibitor Ⅱ) respecitively followed by LPS100ng/ml. Cells incubated with normal medium were used as blank controls. After incubation for another12hrs, all cells were harvested for qRT-PCR analysis.(3) To observe whether the enhaced expression of inflammatory factors induced by TF/FⅧa via PAR2would reciprocally increase TF expression, cells were incubated with medium collected from cells incubation with FⅧa (100nM,24hrs). Cells incubated with medium collected from cells incubation with normal medium were used as controls. After incubation for another12hrs, all cells were harvested for qRT-PCR analysis.In vivo study Ⅱ. The effects of TF/FⅧa/PAR2pathway on plaque instability 1. Seventy-two ApoE-/-mice (male,12-week old) were fed on a high-fat diet and atherosclerotic lesions were induced by placement of a perivascular collar around the right carotid artery. Mice were divided into three groups (n=24for each) at the end of week11: Ad-TF, Ad-TF+PAR2antagonist, and Ad-EGFP groups. Mice in the Ad-TF and Ad-TF+PAR2antagonist groups were transfected with Ad-TF, while mice in Ad-EGFP group with Ad-EGFP. Mice in Ad-TF+PAR2antagonist group were adiministrated with PAR2antagnoist ENMD-1068during week11to week14. At week12, all mice were exposed to restraint stress for two weeks as described above and then euthanized at the end of week14.2. Histological and morphological analysis(1) Sections of the carotid plaques were stained with hematoxylin and eosin at a100μm interval for the assessment of plaque morphology and the idenfication of atherothrombosis, intraplaque hemorrhage.(2) RNA was extracted from carotid plaques, and qRT-PCR analysis was done for the expression of gene selected by PCR array in in vivo study I, e.g. Ccl2, Csf2,114, Lif, Itga2, Selplg, Lamal, Serpinel, Serpinb2.ResultsTF overexpression combined with restraint stress induced atherothrombosis and intraplaque hemorrhageTF expression in the carotid plaques of Ad-TF group was much higher than that in Ad-EGFP group. We identified carotid plaque rupture and atherothrombosis in67%(8out of12) of mice in Ad-TF group. Moreover, we observed intra-plaque hemorrhage in8carotid plaques, including2related to plaque rupture,3related to intra-plaque neovascularization and3related to both factors. In contrast, although pathological examination revealed plaque disruption in9(75%) carotid plaques in mice of Ad-EGFP group (n=12), only2(14%) typical thrombi were observed upon these disrupted plaques. Intra-plaque hemorrhage was observed in2carotid plaques and both were related to plaque disruption. We identified1carotid plaque with intra-plaque neovascularization; however, it was not related to intra-plaque hemorrhage.TF overexpression was associated with plaque destabilizationAlthought there were no significant differences in the contents of lipid, collagen, smooth muscle cells and macrophage of carotid plaques between Ad-TF and Ad-EGFP groups, we found that MCP-1expression in carotid plaques in Ad-TF group was higher than that in Ad-EGFP group. PCR array further revealed that compared with Ad-EGFP group, the expression of6genes related to inflammatory (Ccl2, Csf2,114, Lif, Itga2and Selplg),1gene related to extracellular medium (Lamal) and another2anti-fibrinolysis gene (Serpinel and Serpinb2) were significantly upregulated in Ad-TF group.TF/FⅧa complex upregulated inflammation-related gene expression via PAR2, not PAR1TF/FⅧa significantly increased the mRNA expression of inflammatory factors selected by PCR array including Ccl2, Csf2,114, Lif, Itga2and Selplg, and the alteration show dose-dependence to FVIIa. Thrombin specific inhibitor PPACK did not show interference with the effects of TF/FⅧa on gene expression, which indicated that TF/FⅧa/inflammation pathway was not associated with coagulant cascade. PAR2antagonist ENMD-10685uM andlOuM could significantly inhibited TF/FⅧa effects on the expression of inflammatory genes, while PAR2agonist could increase these genes expression significantly. In contrast, PAR1antagnonist SCH79797exhibited none effects on the upregulation of the related gene expression induced by TF/FVIIa. These indicated that TF/FⅧa upregulated inflammation-related gene expression via PAR2, but not PAR1.Inflammatory factors increased TF expression reciprocally in HASMC and macrophageLPS could upregulated TF expression dose-dependently in cultured HASMC and macrophage. Specific inhibitors of Erkl/2, p38and JNK could inhibit LPS-induced TF expression. What's more, cells incubated with TF/FVIIa-induced inflammatory medium exhibited enhanced TF expression. These indicated that TF expression could be increased reciprocally by TF/FVIIa induced inflammatory factors.TF increased plaque instability via PAR2Compared with Ad-TF group, the mRNA expression of PCR array seleceted gene was downregulated in mice of Ad-TF+PAR2antagonist group, although there were no significant differences in the incidence of plaque rupture and atherothrombosis between the Ad-TF and Ad-TF+PAR2antagonist groups, the incidence of intra-plaque neovasularization (4out of12,33%) and intra-plaque hemorrhage (3out of12,25%) were significantly lower in Ad-TF+PAR2antagonist group, than that in Ad-TF group.Conclusions1. Enhanced plaque thrombogenicity by TF overexpression could induce high incidence of plaque rupture and atherothrombosis in carotid plaques of ApoE-/-mice, combined with mental stress.2. Beyond contribution to plaque thrombogenicity, TF might induce plaque instability via PAR2. The potential mechanisms involved increased inflammation which also could upregulated TF expression and might served as a vicious cycle, and neovasuclarization in the carotid plaque.