【作者】 赵鹏；

【导师】 朱樑； 姚定康；

【作者基本信息】 第二军医大学 ， 内科学， 2009， 博士

【摘要】 【研究背景及目的】原发性胆汁性肝硬化(Primary biliary cirrhosis, PBC)是一种主要以肝内中小胆管的非化脓性进行性炎性损伤为特征的自身免疫性疾病。PBC起病隐匿,进展缓慢,最终可导致肝硬化等终末期肝病,其确切的发病机制至今尚不清楚。目前,PBC的诊断主要依据临床生化及抗线粒体抗体(AMA)的检测,但是临床上仍有部分AMA阴性的PBC患者容易漏诊。熊去氧胆酸(UDCA)是其主要治疗药物,但是其对某些PBC的远期治疗效果并不理想。因而如何早期诊断PBC,如何进行有效的干预处理是PBC研究的一个方向。由于人体诸多生命功能最终是通过蛋白质来完成的,在疾病的发生过程中,人体内蛋白质的种类和含量也会发生相应的改变,这就为我们对PBC的研究提供了一个思路。而蛋白质组学(Proteomics)正是以蛋白质组为研究对象,分析细胞内、外动态变化的蛋白质组成、表达水平与修饰状态,了解蛋白质分子之间的相互作用与联系。迄今国内外有关蛋白质组学技术研究PBC的相关报道较少。我们从蛋白质组学的角度出发,利用同位素标记相对和绝对定量(iTRAQ)联合液相色谱-质谱(LC-MS/MS)技术通过比较PBC、HBV肝纤维化、HBV肝硬化患者和正常对照人群的血清蛋白质,筛选、鉴定PBC患者血清差异表达蛋白,并进行验证,旨在对PBC发病机制和诊治研究提供某些理论依据。与此同时,疾病的动物模型也是研究疾病的一个重要方法。目前PBC动物模型大致分为三类:抗原免疫、化学物质诱导和基因缺陷。但是这些动物模型在肝脏病理变化和/或血清学生化以及自身抗体的表达等方面与人类PBC总是存在一定差异。国外有文献报道部分肝炎患者在利用IFN-α治疗的过程中会继发PBC,而我们在临床工作中也有类似发现,这给了我们一个建立PBC动物模型的启示——通过Poly I:C腹腔注射,诱导小鼠体内IFN-α水平升高,进而建立PBC动物模型。同时,利用iTRAQ和LC-MS/MS技术对不同阶段PBC动物模型及其对照组血清及肝脏组织中的蛋白质进行比较,筛选、鉴定PBC动物模型血清及肝组织差异表达蛋白,并进行验证,以期通过PBC动物模型的研究对人PBC可能的发病机制和诊断有所启示。【实验方法】一、PBC动物模型建立1、6-8周龄C57BL/6雌性小鼠80只,体重约20g,随机分为PBS对照组和Poly I:C模型组,SPF级饲养。PBS液稀释Poly I:C至1mg/ml。模型组给予Poly I:C 5mg/kg腹腔内注射2次/周,对照组给予PBS 5mg/kg腹腔内注射2次/周;每组分别与第4周、8周、12周和16周各处死10只。2、血清样品制备眼球摘除取血,4℃1400g离心5分钟,取血清,部分血清行生化及自身抗体检测,部分血清立即置于-80℃保存备用。3、肝脏组织样品制备断颈处死后,取肝脏于冰生理盐水中清洗,部分肝脏置入4%福尔马林液固定,石蜡包埋,行HE染色和CK19免疫组化染色,光镜下观察肝脏病理变化及肝内胆管变化;部分肝脏立即置于-80℃保存备用。二、PBC动物模型血清及肝脏组织差异表达蛋白筛选、鉴定1、血清及肝脏组织样品制备将每组血清各自混合,4℃1400g离心5分钟,过柱去除高丰度蛋白;肝脏组织于PBS中剪碎洗去血液;液氮研磨混合;按照裂解液体积:组织重量(v:w)=5:1的关系加入裂解液(7M尿素,2M硫脲,0.1% PMSF,0.5% DTT),混悬,冰上裂解半小时。2、样品蛋白质定量通过Bradford法测定各组样品中的总蛋白含量。3、蛋白质的消化和标记加入还原试剂,60℃反应1小时;加入半胱氨酸封闭试剂,室温处理10分钟;每管各加入预冷的丙酮(丙酮:样品体积比=5:1),-20℃沉淀1小时后,12000 rpm,4℃,离心20分钟,取沉淀;加入溶解液20μl,混悬,充分溶解样品;按照酶:蛋白质=1: 20的比例加入胰蛋白酶,37℃酶解过夜;iTRAQTM4、差异表达蛋白质谱分析、筛选及鉴定不同标记的iTRAQ标记蛋白样品混合,经第一维强阳离子柱(SCX)分离,共收取14个梯度进行第二维分析;第二维反相色谱-质谱联用(RPLC-MS),色谱分离70分钟后行质谱鉴定,MS扫描范围m/z 400-1800,MS/MS扫描范围m/z 100-2000。经生物信息学处理,筛选出统计学上差异明显的蛋白质峰进行搜库、鉴定。Reagent114,115,116,117分别标记正常对照、4周、12周和16周模型组,共4管,各管标记试剂中加入70μl乙醇,混匀,分别加入各管样品中,室温反应一小时,之后各加入三倍体积水,使标记试剂分解;合并各管标记好的样品,真空冷冻干燥。三、人PBC血清差异表达蛋白筛选、鉴定及验证1、血清样品制备正常人群对照、PBC、HBV肝纤维化和HBV肝硬化患者各20例,清晨空腹取血,4℃静置1小时,4℃1400g离心5分钟,取上清,分装后立即置于-80℃保存备用;将每组血清各自混合成1管,4℃1400g离心5分钟,过柱去除高丰度蛋白。2、样品蛋白质定量参见动物模型样品蛋白质定量。3、蛋白质的消化和标记参见动物模型的蛋白质消化及标记。4、差异表达蛋白质谱分析、筛选及鉴定参见动物模型质谱分析方法。5、利用Western免疫印迹技术验证淋巴管内皮透明质酸受体1(LYVE-1)和锌-α2糖蛋白(AZGP1)在PBC患者与正常人群血清中的表达差异;利用ELISA法验证视黄醇结合蛋白4(RBP4)在PBC患者与正常人群血清中的表达差异。【实验结果】一、利用Poly I:C腹腔注射建立的PBC动物模型与人PBC类似1、血清ALP检测模型组血清ALP水平随时间的延长逐渐升高,16周时,模型组血清ALP显著高于对照组。2、血清AMA检测模型组自第4周始即有小鼠血清中出现AMA阳性,并随时间延长,其阳性率逐渐升高,至16周模型组血清AMA阳性率达80%,而对照组血清一直未检测到AMA。3、血清ANA检测模型组和对照组自第4周始均有小鼠血清可检出ANA阳性,但是其滴度不高;随着建模时间的延长,模型组从第8周至第16周其血清ANA阳性率均为100%,并且滴度均高于相应时间点的对照组,但是从第8周开始模型组血清ANA滴度呈下降趋势。4、肝组织病理检测肝组织HE染色显示,光镜下对照组肝内胆管周围及汇管区无或极少量炎性细胞浸润,而模型组第4周可见汇管区周围有少量淋巴细胞浸润,随着建模时间的延长,模型组肝组织炎性细胞浸润率逐渐升高,第12周可见肝内小胆管周围有淋巴细胞浸润,第16周汇管区和肝内小胆管周围淋巴细胞浸润现象更为明显;模型组与对照组肝组织病理检测均未发现肝纤维化特征。5、肝内胆管上皮细胞CK19免疫组化检测光镜下可见模型组肝组织较对照组CK19表达显著升高,并且随着造模时间的延长,CK19表达呈明显上升趋势。二、PBC动物模型血清及肝脏组织差异表达蛋白筛选、鉴定1、利用iTRAQ技术对Poly I:C诱导的PBC动物模型不同时间点与对照组的血清及肝组织蛋白分别进行标记。2、利用LC-MS/MS技术对标记蛋白进行鉴定,结果血清中成功鉴定出86个蛋白,肝组织中成功鉴定出519个蛋白。3、PBC动物模型16周组较正常对照组血清明显差异表达蛋白10个,其中在模型16周组血清中上调的差异表达蛋白5个,下调的差异表达蛋白5个;按照蛋白质生物学功能分类主要包括:免疫球蛋白、载脂蛋白、膜蛋白和酶类蛋白。4、PBC动物模型16周组较正常对照组肝组织明显差异表达蛋白64个,其中在模型16周组肝组织上调的差异蛋白35个,下调的差异蛋白29个;按照蛋白质生物学功能分类,主要包括:免疫相关、能量代谢、酶类、细胞生长调控、运输及载体和细胞粘附及运动等。5、通过比较分析不同时间点的PBC模型组与PBS对照组的肝组织差异表达蛋白,我们发现有34个蛋白质与模型的建模时间呈一定的相关性。其中,有17个蛋白在模型组呈持续高表达,并且随着建模时间的延长,其表达水平逐渐升高;从蛋白质的生物学功能看,主要包括能量代谢、细胞生长及分化调控、运输及离子通道等;通过搜索UniProt蛋白质库,这些差异表达蛋白的亚细胞定位主要位于线粒体,其次为质膜、细胞核和胞质。其余17个蛋白则在模型组肝组织中呈持续低表达,并且随着建模时间的延长,其表达水平逐渐降低;从蛋白质的生物学功能看,主要包括酶类、细胞骨架、能量代谢相关、细胞生长及分化调控等;通过搜索UniProt蛋白质库,这些差异表达蛋白的亚细胞定位,主要位于胞质,其次是位于质膜、细胞外基质、线粒体和胞核等。6、从已经鉴定的差异表达蛋白中,我们发现:①模型组血清中上调的蛋白多与免疫功能密切相关;②模型组血清中下调的蛋白多与脂质转运和代谢功能相关;③模型组肝组织中上调的蛋白多位于线粒体,功能上多与能量代谢相关;④模型组肝组织中下调的蛋白多位于细胞质,功能上多与蛋白酶类相关。以上四点发现与我们对该模型进行的组织病理和血清生化的检测结果基本一致,与人PBC相关的表现类似,存在一定的相关性。三、人PBC血清差异表达蛋白筛选、鉴定及验证1、利用iTRAQ联合LC-MS/MS技术,对PBC、HBV肝纤维化、HBV肝硬化患者及正常人群血清蛋白质分别进行标记,并成功鉴定出93个蛋白。根据蛋白质生物学功能分类,这些蛋白主要包括免疫相关蛋白、运输及贮存蛋白、脂质代谢相关蛋白、细胞粘附及运动蛋白、凝血相关蛋白、酶抑制剂、酶类、细胞生长调控蛋白、细胞骨架蛋白、能量代谢相关蛋白及未知功能蛋白等11种蛋白。2、经统计分析,从PBC患者与正常对照人群血清中筛选出30个明显差异表达蛋白;其中LYVE1、IGHM(免疫球蛋白μ链恒定区)、RBP4、AZGP1等14个蛋白质在PBC患者血清中呈高表达,而ApoB(载脂蛋白B)、ApoA2(载脂蛋白A2)、ApoA1(载脂蛋白A1)、ApoC-III(载脂蛋白C-III)、SERPINF2(α2-抗纤维蛋白溶酶)及ApoM(载脂蛋白M)等16个蛋白质下调明显。按照蛋白质生物学功能分类,这些差异表达蛋白主要包括脂质转运及代谢蛋白、运输及贮存蛋白、细胞粘附及运动蛋白、免疫相关蛋白、凝血相关蛋白、酶类、酶抑制剂及未知功能蛋白等8类蛋白。3、分析比较PBC患者与HBV肝纤维化及HBV肝硬化患者血清,我们共筛选出9个明显表达差异蛋白;其中FN1(纤连蛋白1)、GC(维生素D结合蛋白)、LYVE1、A2M(α2-巨球蛋白)和AGT(血管紧张素原)等5个蛋白在PBC患者血清中表达水平显著下调,但其LYVE1较正常对照人群血清表达水平呈明显上调; A1BG(α1B-糖蛋白)、ApoC-III、ApoC-II和IGHM等4个蛋白在PBC患者血清中表达水平显著上调,但其A1BG和ApoC-III较正常对照人群血清表达水平明显下调。4、利用Western免疫印迹检测证实LYVE-1和AZGP1蛋白在PBC患者血清中表达均较正常人群明显升高;利用ELISA法检测证实RBP4蛋白在PBC患者血清中表达明显高于其在正常人群血清中的表达;验证结果均与iTRAQ联合质谱技术鉴定结果一致。【结论】1、Poly I:C腹腔注射C57BL/6小鼠能够成功构建与人PBC在组织病理和血清学表现类似的PBC动物模型。2、iTRAQ结合LC-MS/MS技术能够快速、有效地进行差异蛋白质组学研究,同时,我们利用Western免疫印迹技术和ELISA方法对部分鉴定蛋白进行验证的结果也证明了iTRAQ结合LC-MS/MS技术在蛋白定性及半定量的蛋白质组学研究方面的可靠性。3、PBC动物模型与对照组动物之间在血清及肝组织的蛋白质表达上均存在明显差异,在蛋白质生物学功能上分别涉及脂质转运和代谢、免疫应答、细胞粘附和运动、能量代谢等诸多方面;而且部分差异蛋白表达水平与PBC动物模型的进程呈现出明显的正相关或负相关性,这对于后续对该PBC模型的发生机制进行进一步的功能蛋白质组学研究打下了基础。4、PBC患者、HBV肝纤维化患者、HBV肝硬化患者及正常人群其相互之间在血清蛋白质表达上均存在显著差异,在蛋白质生物学功能上分别涉及脂质转运及代谢、载体类、细胞粘附及运动、免疫应答、凝血相关、蛋白水解酶类等诸多方面;特别是相对于正常人群、HBV肝纤维化及HBV肝硬化患者,我们发现FN1、GC、LYVE1、A2M、AGT、A1BG、ApoC-III、ApoC-II和IGHM等9个蛋白质在PBC患者血清中与上述三组均存在显著的表达差异,这些蛋白质可能与PBC的发病机制密切相关,有待于进一步研究。5、通过对PBC患者血清差异蛋白质组学与PBC动物模型血清和肝组织差异蛋白质组学的比较研究,我们发现,诸如ApoC等载脂蛋白家族蛋白在PBC患者及PBC动物模型中均呈现低表达状态,提示脂质代谢紊乱与人PBC和PBC动物模型的进展关系密切。6、本研究通过比较生理和不同病理条件下血清蛋白质组在表达水平上的改变情况,从而发现和鉴定出PBC疾病相关的蛋白质,为在整体水平发掘和寻求潜在药物靶标以及早期诊断、治疗的分子标志物提供了可能,同时也为PBC的病理机制研究提出了一个新的研究思路和方法。7、目前有关差异表达蛋白的验证及功能研究还在进一步进行中。更多还原

【Abstract】 【Background and Objective】Primary biliary cirrhosis （PBC） is a slowly progressive autoimmune disease of the liver that primarily affects women. Histopathologically, PBC is characterized by portal inflammation and immune-mediated destruction of the intrahepatic bile ducts. The loss of bile ducts leads to decreased bile secretion and the retention of toxic substances within the liver, resulting in further hepatic damage, fibrosis, cirrhosis, and eventually, liver failure. Serologically, PBC is characterized by the presence of antimitochondrial antibodies （AMA）, which are present in 90 to 95 percent of patients. But some of the AMA-negative PBC patients will be easily missed diagnosis. Currently, ursodeoxycholic acid （UDCA） is the only drug approved for the treatment of PBC, while a meta-analysis showed no difference between UDCA and placebo in the incidence of liver related death, liver transplantation, and in the development of complications of liver disease.Early diagnosis and early effective intervention of PBC are critical. Proteins are the most important executors to realize versatile functions of life activity, which will dynamically change in disease status. In recent years, proteomics has been expansively applied in many areas, ranging from basic research, various disease and malignant tumors diagnostic and biomarker discovery to therapeutic applications. However, the technology of proteomics has not been applied in the study of PBC so far. Therefore in our study we used the technology of isotope labeling relative and absolute quantification （iTRAQ） with LC/MS or LC-MS/MS to analyze the sera from PBC, HBV related liver fibrosis, HBV related liver cirrhosis and normal control groups, and to screen, identify and verify the differentially expressed proteins of PBC. It may help to illustrate the pathogenesis of PBC and to explore the effective treatments.Establishment of animal model is an important method to study diseases. Several animal models have been set up for PBC study recently, including antigen-induced, chemical-induced and gene knockout-induced models. There are some differences between the models and human PBC on the aspects of the appearance of autoantibody or the changes of serology or pathology in liver tissues. Clinically, we found several hepatitis B patients occurred PBC during the treatment with IFN-αand some studies reported that the level of IFN-αin sera increased in some other autoimmune diseases, such as Rheumatoid Arthritis and Graves’disease. It is supposed that IFN-αmay induce an animal model for PBC. So we applied Poly I:C （a strong inducer of IFN-α） by intraperitoneal injection to set up a PBC mouse model. At the same time, we also applied iTRAQ and LC/MS （LC-MS/MS） to analyze, screen and identify the differentially expressed proteins of sera and liver tissues of the model on different time points.【Methods】1. Establishment of the PBC model by Poly I:C1.1 Eighty adult 6-8 week-old female C57BL/6 mice were maintained separately, under controlled conditions （22℃, 55% humidity, and 12 h day/night）.1.2 Poly I: C injection was dissolved in sterilized phosphate-buffered saline （PBS） at a concentration of 1mg/ml and stored at -20℃until needed. Female C57BL/6 mice were injected with poly I: C 5mg/kg of body weight intraperitoneally twice a week for 16 consecutive weeks. As controls, a group of female C57BL/6 mice was injected with PBS according to the poly I: C injection protocol.1.3 Mice were killed by cervical dislocation at different times （4w, 8w, 12w, 16w） after administration of poly I:C or PBS. Liver tissues were collected. Part of liver tissue was fixed in buffered formalin （10%） and Paraffin-embedded, followed by HE staining and CK19 immunohistochemical staining for histological evaluation. Part of serum specimens were examined for biochemistry and autoantibody. The remaining serum and tissue specimens were stored at -80℃until used.2. Proteomics analysis of the serum and liver tissue specimens of the model2.1 Serum specimens were divided into 4 groups （4w, 12w, 16w and PBS control）, with the same of liver tissue specimens.2.2 Grinded liver tissues into powder form at liquid nitrogen and then added lysate （7M urea, 2M thiourea, 0.1% PMSF,0.5% DTT） with the concentration of lysate vol: tissue weight （v:w）=5:1 to the samples. Resuspended them fully followed with lysating on the ice for half an hour.2.3 Mix samples in each group respectively and quantitated total proteins in each group by means of Bradford.2.4 Added reducing reagent to 100μg samples and reacted 1 h at 60℃. Cysteine block at room temperature 10 min. Added precooling acetone (V_acetone: V_sample=5:1)and precipitated at -20℃1 h, then samples were centrifugated 12000 rpm 20 min at 4℃to get precipitation. Added dissolution buffer 20μl and resuspended. Trypsin digested overnight. Each group sample was labeled with a different iTRAQTM Reagent（Reagent114,115,116 and 117）.Added 70μl ethanol in each tube,incubated 1 h at room temperature and then added 3 times volume H2O to resolve label reagent.Mixed the labeled samples of all thegroups and to freeze drying in vacuum.2.5 The first dimension was to separate the labeled protein samples by SCX. RPLC-MS was the second dimension.The samples were separated by RPLC for 70 min and then scanned by MS（sweeping range was m/z 400-1800）or MS/MS（sweeping range was m/z 100-2000）.3.Proteomics analysis of the sera of human PBC3.1 Collect blood from fasting patients and healthy volunteers in the morning. Standing 1h at temperature,then centrifugate at 4℃1400g for 5min to derive the serum.3.2 Serum specimens were divided into 4 groups（control,PBC,HBV related liver fibrosis and HBV related liver cirrhosis）.3.3 Mix samples in each group respectively and quantitate total proteins in each group by means of Bradford.3.4 After reduction,cysteine block and digestion,each group sample was labeled with a different iTRAQTM Reagent.3.5 Combine the iTRAQ Reagent-labeled the 4 groups samples digests into one sample mixture.3.6 Clean up or（optionally）perform high resolution fractionation.3.7 Analyze the mixture by LC-MS/MS for protein identification and quantitation.3.8 Verify the differentially expressed proteins of human PBC in serum by ELISA or Western blot.【Results】1.The effects of poly I:C induced PBC model1.1 We examined ALP among the 4 groups and found that the level of ALP in modelgroups increased with time and at 16th week it had significant difference between the model and control group.1.2 We examined AMA among the 4 groups and found that AMA appeared in 4 mice in 4th week model group and its positive rate increased with time.In 16 th week model group there were 16 mice with AMA positive（positive rate=80%）in serum.While it last negativity in control group.1.3 The level of ANA in serum was examined in the model group and control group at different time point sequentially.At 4th week,ANA was detected positively in either model group or control group with lower titer.From 8th week to 16th week,the positive rate of ANA in model group maintained 100%and the titer at each time point was higher than the corresponding control group,respectively.We also found that the titer of ANA decreased after 8th week in model group.1.4 With light microscope detection of liver pathology,we found that lymphocytes infiltration in the liver tissues was low in 4th week model roup;however,considerablenumbers of infiltrating cells were detected around intrahepatic bile ducts in 8th week model group.After this time point,the extent of the infiltrating cells progressively increased.In 16th week model group,numerous of inflammatory cells were observed not only around intrahepatic bile ducts but also in the portal areas.Some infiltrating cells accumulated around the damaged bile ducts.Contrast to model groups,we found little inflammatorycells infiltration around intrahepatic bile ducts or in the portal areas.Meanwhile,we found no characteristic of liver fibrosis.1.5 With light microscope detection after CK19 immunohistochemical staining,wefound that CK19 positive expression rate in the liver tissues of model group was significant higher than it in the control group.And the trend of the expression of CK19 was increased with the time.2.Proteomics analysis of the serum and liver tissue specimens of the model2.1 We identified 86 proteins in the serum in the model of PBC.Contrast to the normal group,there were 5 proteins up-regulated significantly,while 5 proteins down-regulated significantly in the 16th week model group.The differentially expressed proteins are related to lipid metabolism,immune response,cytoskeleton and enzymes.2.2 519 proteins were identified in the tissues of the PBC model.Contrast to the normal group,there were 35 proteins up-regulated significantly,while 29 proteins down-regulated significantly in the 16th week model group.The differentially expressed proteins are related to carrier protein,immune response,energy etabolism,enzymes,cell growth,cell adhesion and movement.2.3 Through a comparative analysis of the differential expression proteins in the liver tissues of the different time points model groups with the control group,we have identified 34 proteins had a certain degree of correlation with the modeling time.Of these, 17 proteins in the model group were persistently high expression,and with the modeling time,and their expression level gradually increased.The functions included energy metabolism, cell growth and differentiation regulation, transportation and ion channels, etc. From the expression of proteins in the cell location of view, they were mainly located in mitochondria, followed by plasma membrane, nucleus and cytoplasm. 17 proteins in the model group were persistently low expression, and with the modeling time, and their expression level gradually decreased. The functions included enzymes, cytoskeleton, energy metabolism, cell growth and differentiation regulation, etc. From the expression of proteins in the cell location of view, they were mainly located in the cytoplasm, followed by plasma membrane, extracellular matrix, mitochondria and nuclei.2.4 We found that （i） most of the significant high expression proteins in the serum of model groups were related to the immune function closely; （ii） most of the significant low expression proteins in the serum of model groups were related to lipid transportation and metabolism; （iii） most of the significant high expression proteins in the liver tissues of model groups were located in the mitochondria and related to energy metabolism; （iv） most of the significant low expression proteins in the liver tissues of model groups were located in the cytoplasm and related to enzymes. These results were consistent with the histopathological and serum biochemical test results we carried out before, and were similar to the performance of human PBC.3. Proteomics analysis of the sera of human PBC3.1 We identified 93 proteins successfully in serum of people with the function of immune response, transportation, lipid metabolism, cell adhesion and movement, coagulation, enzyme inhibitors, enzymes, cell growth and differentiation regulation, cytoskeleton, energy metabolism. 3.2 30 of them were significantly different between PBC and healthy people. 14 proteins of them, such as LYVE1, IGHM, RBP4 and AZGP1, were up-regulated significantly in PBC patients, while the other 16 proteins, such as ApoB, ApoA2, ApoA1, ApoC-III, SERPINF2 and ApoM, were down-regulated significantly. These differential expression proteins were related to lipid metabolism, transportation and storage, cell adhesion and movement, immune response, coagulation, enzymes and enzyme inhibitors.3.3 Comparison of PBC patients and HBV related liver fibrosis and liver cirrhosis patients, we found a total of 9 marked differential expression proteins. Five of them, such as FN1, GC, LYVE1, A2M and AGT, were lower expressed in PBC patients, while LYVE1 was higher expressed than it was in healthy control. The remaining four proteins, including A1BG, ApoC-III, ApoC-II and IGHM, were higher expressed in PBC patients, while A1BG and ApoC-III were lower than they were in healthy control.3.4 By using the technology of western blot and ELISA, we validated the level of LYVE-1, AZGP-1 and RBP-4, and got the accordant results with the iTRAQ and LC-MS/MS.【Conclusion】1. By intraperitoneal injection of poly I:C, we established an animal model of PBC, which was similar to the human PBC on the characteristic of live pathology, autoantibody and biochemistry of serum.2. The technology of iTRAQ combining with LC-MS/MS is an efficient way to distinguish and identify the differential expression proteins in diseases. 3. There were significant differences on the aspect of protein expression in serum and liver tissues between PBC model group and control group. We identified that most of the differential expression proteins were involved with immune response, lipid metabolism, energy metabolism, cell adhesion and movement. The level of some differential expression proteins showed a significant positive correlation or negative correlation with the modeling time. It needs a further study of functional proteomics for these findings to explain the mechanism of this PBC model.4. Comparison with healthy people, patients of HBV related liver fibrosis or HBV related liver cirrhosis, we found that there were significant differentially expressed proteins in the serum of PBC, most of which were involved with lipid transportation, lipid metabolism, extracellular matrix, and so on. Some of the proteins we found might be related to the pathogenesis of PBC and should be explored further. 5. In the candidate proteins we found, the apolipoprotein family seems to be important to PBC, because of its decrease in both of PBC and its model.6. In this study, by comparing the physiological and different pathological conditions of different protein components of the change in expression level, we detected and identified some disease-related proteins （group）, especially Ig family and apolipoprotein family. It may be explored to the search for potential drug targets, as well as early diagnosis, treatment, molecular markers, but also for the study of the pathogenesis of PBC.更多还原