【作者】 宋海霞；

【导师】 王全师；

【作者基本信息】 南方医科大学 ， 影像医学与核医学（专业学位）， 2014， 博士

【摘要】 [研究背景]类风湿关节炎(rheumatoid arthritis, RA)是一种以累及周围关节为主,渐进性、多系统性自身免疫疾病。RA在我国发病率约为0.4%,在全世界约为0.5%-1.0%。目前RA的发生学机制尚未完全明了,研究显示,RA和环境、患者的遗传因素、免疫水平、激素水平有一定关系；同时一些特殊人群如嗜烟人群,糖尿病患者,肥胖者的RA发病率也较高。此外,不同人种,不同性别发病也有较大差异。因此,RA可能为易感个体被感染因子直接或间接激发,同时B细胞、T细胞、及多种细胞因子共同参与的关节及全身性病变。RA以滑膜炎为首发改变,逐步累及关节软骨、韧带、肌腱,造成软骨破坏及骨侵蚀,最终导致关节畸形和功能丧失。如未获得及早有效治疗,RA三年内关节破坏率达70%,不仅严重影响患者的生活质量,还给社会带来了巨大的经济负担。因临床目前尚无有效药物可逆转RA关节骨质破坏,因此,早期诊断RA,尽早开始有效治疗,对控制和延缓RA病情发展,降低RA致残率有重要意义。RA早期特征性病理改变是炎症细胞浸润、滑膜细胞增生及新生血管形成。以滑膜为首发病变,关节滑膜细胞增生及炎性细胞浸润,形成富含毛细血管的肉芽组织----血管翳；后者侵蚀破坏软骨和骨,最终导致关节畸形、功能丧失。显然,新生血管形成是产生和维持RA血管翳的重要标志,它增强了血管翳的侵袭性,在RA的侵蚀和破坏过程中发挥了重要的作用。研究显示新生血管形成在RA早期即开始作用并贯穿整个RA病程。因此,观察RA滑膜病理改变,尤其新生血管的形成对于RA的早期诊断、活动性判断、疗效观察和预后判断均有重要意义。2010版美国风湿病学会/欧洲抗风湿联合会(ACR/EULAR)修订的类风湿性关节炎诊断标准指出滑膜炎是类风湿关节炎早期诊断指标之一。同时,侵入滑膜和腱鞘的炎性反应细胞和新生血管能产生包括血管内皮细胞生长因子(vascular endothelial growth factor, VEGF),肿瘤坏死因子-α (tumor necrosis factor-α, TNF-α)等在内的等多种细胞因子。这些细胞因子以网络形式相互作用,不但进一步促进新血管的形成,同时也是关节结构破坏的基础。VEGF是一种特异作用于血管内皮细胞的多功能细胞因子,促进血管内皮细胞有丝分裂、增生、迁移,最终导致新生血管生成增加。此外,VEGF还可增加血管通透性,促进血管内物质渗出,导致慢性炎症形成和发展。因此,目前认为VEGF是参与RA最重要的细胞因子之一,在整个新生血管的形成过程中处于核心地位[10]TNF-α是RA中另一种重要免疫调节因子,由单核巨噬细胞产生,通过调节血管新生和滑膜炎性改变,降解软骨基质等过程参与RA关节病变。TNF-α产生于免疫反应早期,通过自分泌和旁分泌的方式产生大量炎性介质,刺激纤维样细胞增殖并增加IL-6、IL-8等细胞因子的释放；同时又反过来活化环境中的巨噬细胞,促进自身生成,从而在滑膜细胞和巨噬细胞之间形成了一个正反馈,使炎症持续发展,激发炎性连锁反应。有动物实验研究显示,在血管内皮细胞中TNF-α对血管新生的作用是通过整合素αvβ-3来完成的[14]。整合素为细胞黏附分子家族的成员之一,正常生理条件下在血管内皮低表达或不表达,而在肿瘤诱导的新生血管组织高表达,说明其对血管的生成起重要作用。研究显示整合素αvβ-3可能在多环节影响类风湿关节炎病变滑膜血管新生,在血管生成中起着重要的调节作用。动物模型在类风湿性关节炎的发病机制,早期诊断,药物治疗等研究中具有非常重要的价值。胶原诱导性关节炎(collagen induced arthritis, CIA)以多发性关节炎性改变并伴有关节损害为特征,是国内外较为理想的类风湿关节炎动物模型之一。大鼠则因饲养容易,实验成本低廉而成为首选。采用何种敏感而有效的手段或指标对RA模型进行实时观察和监测,即活体观察RA模型关节病理变化过程,尤其滑膜血管翳的形成和发展,对RA实验研究至关重要。组织病理检查毫无疑问可以准确反映关节滑膜病变,但需要处死动物,实验无法连续性进行。血液生化指标多波动较大,且特异性不理想。普通X线因无法显示滑膜及软骨,对RA早期滑膜改变诊断价值有限。微型CT图像分辨率约为90μm,微型核磁为500μm,对于大鼠关节细微结构的辨别仍显不足。且CT、MRI均适合瞬时图像的采集,难以进行实时动态观察。缺乏理想的专门针对小动物(如大鼠)的骨关节影像学方法是目前RA实验研究中面临的问题。超声对人体软组织,特别是含液体的组织细微结构具有较高的分辨力。RA关节滑膜囊内多有液体渗出,与增生的滑膜之间形成良好的反射界面,使滑膜增生的形态学改变得以直观显示。目前,高频超声对RA的诊断价值已得到临床认可,彩色多普勒、频谱多普勒、能量多普勒等超声技术的应用增加了超声早期诊断RA的准确性及敏感性,尤其能量多普勒成像(power Doppler imaging PDI),克服了彩色多普勒显像对血流角度的依赖性,特别适用于显示滑膜低速血流,且不易被干扰,可靠性较高。但超声检查应用于类风湿动物实验研究鲜见报道,尤其在RA小动物模型的应用,至今国内未见报道。原因在于鼠体格小,肢体关节纤细,临床常用高频超声探头频率多为15-20MHz,轴向辨率约3mm,对大鼠关节细微结构显示欠佳。超声生物显微镜(ultrasoundbiomicroscopy, UBM)由高频率换能器与高分辨率超声仪结合而成,探头频率20-100MHz,轴向分辨力可达100-20μm[27]。UBM是近年新兴的活体成像技术,逐步用于观察小动物模型及人类浅表组织。该技术具有非侵入性、高分辨率、实时性等独特优势,目前UBM在小动物心血管疾病、胚胎发育等领域的研究应用逐渐成为热点,而在肌骨关节的应用尚处于起步阶段。本实验在传统胶原诱导性关节炎基础上加以改进,通过二次免疫加强法对SD大鼠尾根部及腹股沟皮内多点注射Ⅱ型胶原乳剂,建立类风湿关节炎大鼠模型---胶原诱导性关节炎(CIA)。尝试使用UBM动态连续在体观察CIA大鼠关节变化,并采用能量多普勒成像对滑膜血流进行半定量分析,目前这一研究尚鲜见报道。本研究试图从临床症状,声像学特点,病理学改变等多角度,结合VEGF、TNF-α、αvβ-3表达水平,动态观察RA发生发展过程,分析滑膜细胞增生,炎性活动、血管生成与类风湿关节炎发病机制的关系；探讨UBM对RA小动物模型早期诊断的可能性及应用价值。此外,实验通过UBM观察不同抗血管生成药物对RA的治疗作用,为RA抗血管生成靶向治疗提供实验依据,并进一步评价UBM在RA小动物模型药物疗效观察中应用价值。本实验主要包括以下三个部分：第一章超声生物显微镜在建立大鼠CIA模型中的应用[目的]1.探讨UBM能否应用于大鼠关节检查,初步总结大鼠正常关节及CIA声像图特点。2.通过滑膜厚度及PDI血流分级半定量分析探讨CIA大鼠关节声像图表现与滑膜增生,血管新生的关系,评价UBM对CIA早期小动物模型的应用价值。[方法]1.110只雌性SD大鼠根据随机数字表分成2组：空白对照组,10只；胶原诱导性关节炎(CIA)组100只。2.CIA组采用牛Ⅱ型胶原乳剂二次免疫诱导制备RA大鼠模型。II型胶原乳剂分初次免疫与加强免疫两次注射给药。将大鼠用10%水合氯醛(注射剂量0.3ml/100g体重)腹腔注射麻醉后,初次免疫将浓度为lmg/mL的Ⅱ型胶原乳剂以0.2ml/只的剂量在CIA组大鼠尾根部进行多点皮内注射。2周后加强免疫一次,使用不完全弗氏佐剂配制的浓度为lmg/mL的Ⅱ型胶原乳液于大鼠腹股沟皮下注射,剂量0.2ml/只。对照组同期同部位0.2ml/只生理盐水注射。3.试验当天注射造模药物前及注射造模药物后每5天对实验大鼠进行一般状态及局部关节肿胀度观察,测量体重、关节炎指数。4.实验第0天、15天、35天UBM检查实验大鼠四肢关节,记录检出滑膜增厚关节数目(检出关节数),PDI显示关节腔内出现血流信号的关节数目(血流关节数),测量膝、踝、足爪关节滑膜厚度,增生范围,PDI观察滑膜血流,根据血流信号丰富程度半定量评分。5.第35天将造模成功的部分CIA大鼠及未成功CIA大鼠、空白组大鼠全部处死,取四肢关节进行病理检查,对滑膜病理损害半定量评分。滑膜细胞增生评分标准：滑膜细胞排列整齐,无增生,少于3层,为0分；滑膜增生5-6层,无明显淋巴细胞浸润,1分；滑膜增生6层以上,出现淋巴细胞浸润,2分；滑膜增生明显伴血管翳形成,为3分；出现软骨,骨侵蚀为4分。单只大鼠滑膜病理评分以诸病变关节中病理评分最高者计入。病理关节数为单只大鼠滑膜病理评分≥1分的关节个数。6.应用SPSS17.0统计软件包进行统计分析。计量资料统计描述采用χ±s表示,两组间比较采用独立样本T检验、配对样本T检验,空白组全部是0的情况下根据其余各组的95%置信区间是否包含0来判断结果。多组间比较采用单因素方差分析,首先进行方差齐性检验,方差齐性,则组间多重比较采用LSD法；若方差不齐则采用近似方差分析Welch法；组间多重比较采用Dunnett’sT3法。相关性比较采用Pearson、Spearman相关分析；计数资料的统计描述用率来表示,两组间比较采用χ2检验；P<0.05为差异有统计学意义。[结果]1.CIA组共72只大鼠造模成功,造模成功率72%。CIA成功组大鼠临床表现,UBM声像学特征及病理改变与空白组、CIA未成功组有显著性差异(P<0.05),空白组和CIA未成功组无显著性差异(P>0.05)。2. UBM、AI对CIA大鼠阳性检出率分别为88%和44%,二者具有显著性差异(P<0.01)。3.CIA成功组大鼠UBM声像学指标(检出关节数,检出血流关节数、关节滑膜厚度,滑膜UBM分级,血流PDI分级)随病程发展而增加,注射造模药物后第35与第15天相比具有显著性差异(P<0.01)。4.CIA成功组大鼠病理关节数、滑膜病理评分均高于未成功组,具有显著性差异(P<0.01)。5.CIA成功组滑膜厚度与滑膜病理评分、PDI分级、UBM分级呈正相关(r分别为0.649、0.528、0.619,P<0.01)；血流PDI分级与滑膜病理评分、UBM分级呈正相关,r分别为0.610、0.690,P<0.01,具有统计学意义。[结论]1.本研究成功建立了RA早期大鼠模型,其临床特点、关节变化及病理改变符合RA病理特征。2.UBM能够清晰观察CIA大鼠滑膜增厚,血管新生,关节积液,是RA早期诊断、病情评估等实验研究的理想影像学方法。3.滑膜厚度和血流PDI分级可做为早期诊断RA,评估滑膜增生及血管新生程度的客观指标。第二章RA早期UBM声像特点与VEGF、TNF-α、αvβ-3相关性实验研究[目的]1.观察VEGF, TNF-α, αvβ-3在CIA大鼠血清及滑膜组织中的动态表达,探讨其在滑膜炎症及血管新生中的作用,为RA血管抑制靶向治疗提供实验依据。2.分析CIA大鼠UBM声像学指标与血清、滑膜组织VEGF, TNF-α, αvβ-3表达相关性,探讨滑膜厚度、PDI血流分级等声像学指标无创性评估RA早期滑膜增生和血管翳形成的价值。[方法]1.110只实验大鼠根据随机数字表分为两组,空白对照组10只,CIA组100只。采用牛Ⅱ型胶原乳剂二次免疫诱导制备CIA大鼠模型,对照组同期给与生理盐水注射。2.注射造模药物前(第0天)及注射药物后第15、35天大鼠心脏取血2ml,分离血清,采用酶联免疫吸附分析法测定VEGF, TNF-α, αvβ-3水平。3.注射造模药物前(0天)及注射药物后第15、35天使用UBM观察大鼠四肢关节滑膜厚度,血流及关节积液等声像图变化。4.注射药物后第35天处死对照组大鼠(10只)、部分造模成功CIA大鼠(22只)和造模不成功大鼠(18只),取四肢关节进行病理切片检查,半定量评估关节滑膜病理损伤,评分标准同前。5.滑膜病理评分≥1者,使用鼠αvβ-3、VEGF、TNF-α免疫组织化学试剂盒,严格按照操作说明进行滑膜VEGF, TNF-α, αvβ-3表达水平测量,结果采用半定量计分法判定：①按阳性着色程度评分：0分为无着色,与背景一致；1分为浅黄色,略高于背景色；2分为棕黄色,明显高于背景色。3分为棕褐色。②按阳性细胞所占比例评分：阴性为0分,1分为10%以下；11-50%为2分；51-75%为3分；75%以上为4分。①②两项得分乘积,0-1分为阴性(一),2-4分为弱阳性(+),5-8分为中度阳性(++),9-12分为强阳性(+++)。6.应用SPSS17.0统计软件包进行统计分析,计量资料统计描述采用Z+s表示,多组间比较采用单因素方差分析,首先进行方差齐性检验,若检测结果方差齐性,则组间多重比较采用LSD法；若方差不齐则采用近似方差分析Welch法,组间多重比较采用Dunnett’s T3法；相关性比较采用Pearson、Spearman相关分析；P<0.05为差异有统计学意义。[结果]1.CIA成功组大鼠注射造模药物后血清VEGF、TNF-α、αvβ-3水平随时间升高,第35天与第15天、第0天相比具有显著性差异(P<0.05)。2.CIA成功组大鼠在注射造模药物后第35天血清VEGF、TNF-α、αvβ-3水平分别为98.29±13.86pg/ml,27.37±5.10pg/ml,2.15±0.58pg/ml,高于CIA未成功组和对照组,具有显著性差异(P<0.05)。3.CIA未成功组血清VEGF、TNF-α、αvβ-3水平和对照组无显著性差异(P>0.05)。4.CIA成功组大鼠滑膜VEGF、TNF-α、αvβ-3免疫组化表达水平分别为2.23±0.92、1.91±0.97、1.81±0.91,与未成功组和对照组有统计学差异(P<0.05)；CIA未成功组和对照组滑膜VEGF、TNF-α、αvβ3表达水平无显著性差异(P>0.05)。5.CIA成功组大鼠滑膜厚度与滑膜VEGF、TNF-α、αvβ-3表达水平呈正相关,r分别为0.713、0.749、0.548,P<0.01,具有统计学意义。6.CIA成功组大鼠滑膜PDI分级与滑膜VEGF、TNF-α、αvβ-3表达水平呈正相关,r分别为0.576、0.635、0.789,P<0.01,具有统计学意义。[结论]1.CIA大鼠血清VEGF和TNF-α, αvβ-3含量随病程进展而增加,与临床关节评分及病理评分相一致,血清VEGF和TNF-α, αvβ-3升高与类风湿活动有关。2.CIA大鼠关节滑膜VEGF, TNF-α, αvβ-3的高表达与RA关节滑膜增生及血管生成关系密切。3.滑膜厚度、血流PDI分级分别与滑膜VEGF, TNF-α, αvβ-3表达呈正相关,UBM是无创性评价滑膜增生和血管生成的有效工具。第三章UBM在RA早期靶向治疗实验研究中的应用[目的]1.观察贝伐单抗和益普赛治疗前后CIA大鼠UBM关节声像图、临床症状、血清学指标等变化情况,评估抗血管生成对早期RA治疗效果,为RA靶向治疗提供实验依据。2.评价UBM在CIA靶向治疗过程中评估炎性反应和治疗效果的作用,探讨声像学指标能否作为观察RA药物疗效,指导用药的客观指标。[方法]1.将前期实验造模成功的50只CIA大鼠,根据侵犯关节数和程度相匹配原则分为3组：贝伐单抗治疗组,20只；益赛普治疗组20只；对照组,10只,不做药物治疗,同期给予生理盐水注射。根据体重换算相应剂量分别于第0天,7天两次给予,疗程共14天。2.第0、7、14天观察3组CIA大鼠一般状况,记录体重,关节炎指数。3.UBM观察治疗前后CIA大鼠膝,踝,关节,测量滑膜厚度,软骨、骨侵蚀及血流变化。检查方法、内容、图像判读标准同前。4.酶联免疫吸附分析测量治疗前后各组CIA大鼠血清VEGF, TNF-α, αvβ-3表达水平,方法同前。5.治疗结束后CIA大鼠全部处死,取膝,踝,足爪关节制作病理切片,常规染色,进行关节炎病理损伤评分。病理评分≥1者,进行免疫组化检测滑膜组织中VEGF、TNF-α、αvβ-3表达水平,方法及评分标准同前。6.应用SPSS17.0统计软件包进行统计分析,计量资料统计描述采用χ±s表示,同一组治疗前后比较采用配对样本T检验,多组间比较采用单因素方差分析,首先进行方差齐性检验,若检测结果方差齐性,则组间多重比较采用LSD法,若方差不齐则采用近似方差分析Welch法,组间多重比较采用Dunnett’s T3法；相关性比较采用Pearson Spearman相关分析；P<0.05为差异有统计学意义。[结果]1.药物干预前,三组CIA大鼠UBM声像学指标(检出病变关节数,检出显示血流关节数、病变关节滑膜厚度、UBM分级、PDI分级)组间无显著性差异(P>0.05)；药物干预后检出血流关节数及PDI分级三组间有极显著性差异(P<0.01),贝伐单抗组和益赛普组分别高于空白对照组,具有极显著性差异(P<0.01)；贝伐单抗组和益赛普组二者之间无组间差异(P>0.05)。2.药物干预前三组CIA大鼠血清VEGF、TNF、αvβ-3水平组间无差异(P>0.05),药物干预后三组间出现极显著性差异(P<0.01)；贝伐单抗和益赛普组均低于对照组,具有极显著性差异(P<0.01),二者无组间差异(P>0.05)。3.贝伐单抗和益赛普组CIA大鼠治疗后,AI, UBM声像学各指标与治疗前比较下降,有显著性差异(P<0.05)。4.对照组给予生理盐水两周后滑膜厚度、PDI分级较治疗前增加,差异具有统计学意义(P<0.05)；治疗前后AI、UBM检出病变关节数、显示血流关节数,UBM分级比较,无显著性差异(P>0.05)。5.药物干预后,三组间病理关节数无显著性差异(P>0.05)；滑膜VEGF、TNF-α、avβ-3表达水平及病理评分三组间有显著性差异(P<0.05)；空白对照组上述指标均高于贝伐单抗组和益赛普组,有显著性差异(P<0.05)；贝伐单抗和益赛普组间无显著性差异(P>0.05)。6.药物干预后,滑膜厚度,PDI分级的下降分别与滑膜病理评分及VEGF、TNF-α、αLvp-3表达水平的下降正呈相关(P<0.05)。[结论]1.贝伐单抗和益普赛抑制RA早期滑膜增生,减少滑膜血管新生,降低滑膜病理评分,对早期RA治疗作用肯定。2.滑膜厚度、PDI分级可作为RA治疗过程中评估炎性反应和治疗效果的客观指标,与滑膜厚度相比,PDI血流变化对靶向治疗更为敏感。3.UBM对RA药物试验,疗效观察有推广应用价值。更多还原

【Abstract】 BackgroundRheumatoid arthritis (RA) is a chronic, progressive and multi-systemic, inflammatory autoimmune disease, usually involving polyarthritis. The morbidity of RA is about0.4%in nation and the number in the world is about0.5%to1.0%. At present the mechanism of RA is not fully understood. Studies have shown the pathogenesis of RA is multiple, including the environmental, genetic factors, the immune and hormone levels. Maybe RA has some relationships with special populations, such as smoking crowd, diabetics, obese people who have a higher RA incidence. In addition, the morbility of RA is quite different in different races and genders[2’3]. Thus, RA may be stimulated by a susceptible individual factors directly or indirectly, while B cells, T cells, and a variety of cytokines participate the disease process of joints and systemic organse. Synovitis is the primer pathyologhy change of RA, gradually the cartilage, ligaments, tendons and whole tissue is affected, causing joints pain followed by cartilage damage and joint space becoming narrow. Without effective suppression treatment in early course, the joint destruction rate of RA is70%within three years, leading to joint deformity and disablity. Now, there is no effective drugs can reverse the destruction of bone joints in clinical, and the joint destructions has begun in the early course of RA (within four to twelve months). So it is important to diagnose RA and begin active treatment as soon as possible, which can control and hold back the progression of the disease with a significant reduction in morbidity.Inflammatory cell infiltration, synovial hypertrophy and angiogenesis are characteristics of RA primer pathology. Synovitis is the first step, then, Synovium hyperplasia and inflammatory cell infiltration appear in the joints. Synovial granulation tissue riches in capillaries which is named as pannus. Pannus destructs the cartilage, causing bone erosion and leads to joint deformity and loss of function at last. Obviously, Angiogenesis is an important symbol for generation and maintenance of RA pannus. Pannus enhances the invasiveness and plays an important role in RA erosion and destruction process. Angiogenesis start early in the course of RA and throughout the whole course. Thus, synovial pathology of RA, particularly, the vascularization of synovium is vital for the early diagnosis of RA, activity judgment, observation of drug efficacy and prognosis.2010edition of the American College of Rheumatology/European League Against Rheumatism Association (ACR/EULAR) points out that synovitis is one of the diagnostic criteria for early diagnosis of rheumatoid arthritis (RA).Meanwhile, the inflammatory cells and vascularization can produce a variety of cytokines, including vascular endothelial growth factor(VEGF), tumor necrosis factor-a (tumor necrosis factor-a, TNF-a) and the likes。These cytokines interact in form of net. Not only further promote the formation of new blood vessels, also lead to joints structural damage. VEGF is a multifunctional cytokine having vascular endothelial cell-specific effects,which can promote endothelial cell mitosis, proliferation, migration and result in more angiogenesis [9]. VEGF also increases vascular permeability, promoting substances leaking, which leads to the formation and development of chronic inflammation. So, VEGF is currently considered one of the most important factors in RA and the core of the whole angiogenesis[10].TNF-a is another important immunoregulatory factor in RA, released by mononuclear macrophages. TNF-a takes part in RA by regulating angiogenesis, synovitis and degradation of cartilage matrix. Released early in the immune response, TNF-a produces a large number of inflammatory mediators through autocrine and paracrine activation, stimulating fibroblast proliferation and increase of the release of cytokines, such as IL-6, IL-8[11]. Meanwhile, TNF-a can activate macrophages to promote cytokine production. So, a positive feedback forms between synovial cells and macrophages, which keeps the sustainable development of inflammation, stimulates inflammatory chain reaction[12’13]. Experimental studies in animals show TNF-α is affected by integrin αvβ-3in angiogenesis[14]. Integrins αvβ-3is a member of cell adhesion molecules family, which has low or no expression in the vascular endothelium under normal physiological conditions and high expression in tumor-induced angiogenesis, indicating an important role in the generation of blood vessels[15,16]. Studies have shown integrin may affect many aspects of synovial lesions angiogenesis in rheumatoid arthritis, playing an important regulatory role in angiogenesis.The application of animal models plays an important vole in the study of rheumatoid arthritis. Collagen-induced arthritis (CIA) is one of ideal animal models for rheumatoid arthritis, which is characterized by polyarthritis associated with joint damage[17].Rats become the best choice due to easy feeding and lower experiment costs. A sensitive and effective tool for vivo animal model is crucial, which can provide real-time observation and monitoring RA joint pathological changes, particularly synovial pannus formation and development in the experimental study of RA. Obviously, histopathological examination can reflect joint disease accurately, but the animals have to be killed and continuity of the experiment can not be kept.Specificity of laboratory biochemical parameters is not ideal, for appearing fluctuationssometimes[18]. X-ray has little help for synovium and cartilage in the early RA. Micro-CT image resolution is about100um, and the finger of MRI is500um, which is insufficient to distinguish the structure of rat joints[19’20]. Besides, CT and MRI are suitable for instantaneous image acquisition and can not carry out real-time observation. Lack of ideal imaging methods of bones and joints specifically for small animals (such as rats) is the problem of RA experimental study at present.Ultrasound has high resolution for the soft tissue, especially comprising with liquid. RA synovial capsule has more liquid oozing as synovial hyperplasia and provides a good reflection interface. Synovial morphological changes can be visualized by ultrasonography. Currently, the diagnostic value of high frequency ultrasound for RA has been accepted in clinical[21,23]. Color Doppler, spectral Doppler, and power Doppler ultrasound increase the accuracy and sensitivity of diagnosis in early RA[24,25].Especially, power Doppler imaging (PDI) overcomes angle dependence of color Doppler flow imaging, having high reliability for the detection of synovium low blood flow, and not susceptible to be interferenced [26].However, the report of ultrasound applied in animal studies in rheumatoid is rare. Especially, the application for small animal models of RA, such as rats, has not been reported because clinical US devices is not adapted for rat joints given the small sizeUBM (ultrasound biomicroscopy, UBM) has high resolution provided by the high frequency transducers (up to100MHz), the axial resolution is up to100-20um[27]. As a kind of vivo imaging technique emerging in recent years, UBM is used to observe small animal models and human superficial tissue. The technology has the unique advantage of non-invasive, high-resolution and real-time. Resently, Application of UBM in cardiovascular disease in small animals, embryos and other areas study is becoming a hot[28,29]. However, the application in musculoskeletal joints is still in primary stage[30,31].In this experiment, collagen-induced arthritis (CIA)was induced in SD rat through injection subcutaneously at the base of the tail with100mg of CII emulsified in Freund’s adjuvant[32]. Ultrasonography and Doppler analysis were performed on all rats by ultrasoundbiomicroscope (UBM). The changes of CIA rats were observed in vivo and the flow of synovium were semi-quantitative analyzed. At present, the similar research is rarely reported. The objective of our study was to determine whether UBM analysis could accurately detect arthritic lesions in RA model, named CIA. We thus aim at validating this non-invasive method by comparing it with clinical and histological joint examination, which are the classical methods used until now. Moreover, the change of cytokine expression level, including VEGF, TNF-α, αvβ-3, were analysised, trying to expore the correlation between ultrasonographic changes and RA immune mechanisms. Neovascularization is strongly linked with joint inflammation during arthritis, and is therefore an important parameter of disease evolution. We also evaluated the possibility to estimating vascularization in arthritic joints of CIA rats using power Doppler. By observing the therapeutic effect of different antiangiogenic drugs for RA, we evaluated UBM for efficacy to treatments in RA experimental study.This experiment included the following three parts:Chapter1Ultrasound and Doppler micro-imaging in a model of rheumatoid arthritis in ratsObjective1. To determine whether ultrasonography analysis could accurately detect arthritis lesions in a rat model of rheumatoid arthritis, summarize sonographic features of CIA and normal rats by UBM.2. To study the relationship of sonographic feature and pathological changes in joint of CIA rats by quantitative analysis of synovial thickness, PDI blood and synovial angiogenesis, evaluate the role of UBM in study synovial hyperplasia and angiogenesis in CIA.Materials and Methods1.110female SD rats were randomly divided into two groups according to the random number table:the blank control group,10and CIA group,100.2. CIA group rats were induced by immunized with native bovine collagen type Ⅱ (CII). Every rat was injected subcutaneously at the base of the tail with0.2ml of CⅡ emulsified in Freund’s adjuvant. On the15th day, rats were boosted with a subcutaneous injection of CII in incomplete Freund’s adjuvant,wjile the control group rat was given0.2ml saline injections subcutaneously at the base of the tail at the same time.3. Before and after injection the modeling drug, the clinical features of the experimental rats were observed. The weight of the rats and the paw volume were observed, and arthritis index(AI) was calculated the swelling of joint and every five days.4. Joints examinations were performed using UBM (VEVO770, Visualsonics) both in B mode and power Doppler imaging(PDI) on the day of0,15and35. Synovial fluid and synovial hypertrophy intra-articular were recessed, and Synovial flow at each joint was evaluated by PDI. The number of joint synovial thickening detected and the number of intraarticular joint appearing flow signal in knee, ankle, and feet joint were recorded.5. Parts of CIA rats were killed, their legs were dissected and processed for histological studies. The lesions were evaluated in each joint as previously described using five point scale (0-4, where0is normal and4severe) reflecting synovitis:synovial cells arranged in neat rows, no proliferation, less than three layers.0points;synovial hyperplasia to5-6layers no significant infiltration of lymphocytes, score1; synovial hyperplasia more than six layers, lymphatic cell infiltration, score2;synovial hyperplasia was significantly associated with pannus formation, score3; cartilage and bone erosion, score4. Single rats articular synovial pathological score is the highest score in various pathological lesions. Number of pathological articular is the summary of joint, which synovium pathology score≥1point.6. SPSS17.0statistical package was appliced for statistical analysis. The experimental data and measurement data were indicated by means士standard. The data between the two groups were compared using independent samples T-test and paired samples T-test. The data comparison among several groups was made by means of one-way ANOVA. LSD method was used when test for homogeneity of variance. when test of homogeneity of variance, multiple comparisons between the groups using; If heterogeneity of variance is adopted, Welch approximate variance analysis and multiple comparisons were used between groups. Pearson and Sperman correlation analysis were used for relevance comparison. Enumeration data was test by X2. A value of P<0.05was considered statistically significant.Results1.72rats were successful set up in CIA group, modeling success rate was72%. The clinical and pathology performance of72successful modeling rat of the CIA group was significantly. There was a significant difference between unsuccessful CIA group and control group (P<0.05).2. Positive detection rate of CIA rats used by UBM was88%, the finger of AI was 44%, the difference is significant (P<0.01).3. Counts of joints with synovial hypertrophy signal, joints with PDI signal, Synovial thickness, synovial UBM grade, PDI grade increase with the progression in CIA successful group. The difference of rats diseased joints was statistically significant on the day of15and35after injection modeling drug(P<0.01).4. Count of pathology joint and synovial pathology scores in successful CIA rat were higher than the unsuccessful group, with significant difference(P<0.01).5. Synovial thickness was positively correlated respectively with synovial pathology score, PDI classification and UBM grade in CIA successful group (r=0.649,0.528,0.619, P<0.01). PDI grade was positively correlated with pathology score and UBM grade respectively (r,0.610,0.690, P<0.01), with significant difference.Conclusions1. RA rat model of CIA was successfully established。The clinical and pathology performance of CIA rats were in accordance with RA.2. UBM can display Synovial hyperplasia, angiogenesis and joint effusion. So UBM maybe an ideal imaging methods for early diagnosis and severity assessment of small animal models of RA.3. Synovial thickness and PDI grade can be used as an objective indicators for diagnose of early RA and the extent of synovial proliferation and angiogenesis. Chapter2Correlation of ultrasonographic features and level of VEGF, TNF-α, αvβ-3expression in early RA experimental studyObjective1. To explore the role of VEGF, TNF-α, αvβ-3in synovial inflammation and angiogenesis by observing the expression level in serum and synovial tissue of CIA rats, which can provide experimental basis for RA vascular targeting therapy2. To study the correlation between ultrasonographic findings and level of VEGF, TNF-α, αvβ-3in serum and synovial tissue and explore the possibility of whether synovial thickness or PDI grade can be used as sonographic markers to assess synovial hyperplasia and pannus formation in early RA.Materials and Methods1.110female SD rats were randomly divided into two groups, the blank control group,10and CIA group,100.2. CIA group rats were immune-induced by bovine type Ⅱ of collagen to make RA model, named CIA rat.3. Before and on the15th and35th day after injection modeling drug, the VEGF, TNF-α and αvβ-3level in serum of the experimental rats were detected by By enzyme-linked immunosorbent assay (ELISA).4. Joints examinations were performed using UBM both in B mode and PDI on the day of before and after injection modeling drug15and35days respectively. Joint effusion and synovial proliferation within intra-articular were recessed. The flow of synovium at each joint was evaluated by PDI. Recorded the number of joint synovial thickening detected and the number of intraarticular joint appearing flow signal in knee, ankle, and feet joint.5. All of the blank group and parts of CIA rats were killed on the35th day. Take Pathology and immunohistochemical examination for experimental rats joint. Synovial pathological damage was assessed by with the former Semi-quantitative scoring criteria.6. VEGF, TNF-α, αvβ-3expression level in synovium was measured using mouse αvβ-3, VEGF, TNF-a immunohistochemical kit in strict accordance with the operating instructions for the joints which synovial pathology score≥1. The expression of αvβ-3, VEGF, TNF-a was determined by semi-quantitative integral method:positive cells<5%, score0;6%-25%, score1;26%-50%, score2;51%-75%, score3;>75%, score4. The score of stain:no specific staining, score0; yellow staining, score1; light brown, score2; brown, score3. To multiply the score marks, the scores were defined as flows:0-1, negative(-),2-4, weekly positive(+);5-8, middle positive(++);and9-12, strong positive(+++).7. Statistical methods were same as the former.Results1. The levels of serum VEGF, TNF-α, αvβ-3in CIA successful group rats increased with time after modeling injection, the levels of the35th day compared with the15th day and the0day with a significant difference (P<0.05).2. After injection modeling drug35days, the VEGF, TNF-α, αyβ-3level in serum of CIA successful group rats were:98.29±13.86pg/ml,27.37±5.10pg/ml,2.15±0.58pg/ml, and the difference compared with the blank control group and the unsuccessful CIA group was statistically significant (P<0.05).3. There had no significant difference between the VEGF, TNF-α, αvβ-3level in serum of CIA unsuccessful group between control group (P>0.05).4. On the35th day after injection modeling drug, the levels of VEGF, TNF-α, αvβ-3in synovium of CIA successful group rats were2.23±0.92,1.91±0.97,1.81±0.91respectively, and the difference compared with the blank control group and CIA unsuccessful group was statistically significant (P<0.05).5. In CIA successful group, synovium thickness and VEGF, TNF-α, αvβ-3expression levels were positively correlated respectively, r was0.713,0.749and0.548, P<0.01, with statistical significance.6. In CIA successful group, synovial PDI grade were positively correlated with VEGF, TNF-α, αvβ-3expression level respectively, r was0.576,0.635and0.789, P<0.01, with statistical significance.Conclusions1. The expression of VEGF, TNF-α and αvβ-3in serum of CIA rats increase with the progression and consistent with the clinical and pathological joint score. The expression level of VEGF, TNF-α, and αvβ-3in serum may reflect activity in rheumatoid progression.2. VEGF, TNF-α and αvβ-3in synovium showed high expression level in CIA successful group, which were closely related with synovial lesions and angiogenesis in RA.3. Synovium thickness and PDI grade were positively correlated with synovial VEGF, TNF-α, αvβ-3expression respectively. UBM is an effective tool for noninvasive evaluation of synovial proliferation and angiogenesis. Chapter3Evaluation of UBM in anti-angiogenic therapy for early RA in experimental studyObjective1. To observe effect of Bevacizumab and etanercept treatment in rheumatoid arthritis model rats and provide anti-angiogenic therapy experimental basis. 2. To evaluate the use of UBM, including quantitative Doppler analysis of synovial vascularisation, before and after anti-angiogenic treatment with Bevacizumab and etanercept in rheumatoid arthritis model rats.Materials and Methods1.50CIA rats were divided into three groups:bevacizumab group,20; etanercept treatment group according to the number and extent of arthritic joints,20; control group,10. Drug according to the weight was given once a week in14days.2. Before and after drug intervention, the clinical features of the experimental rats were observed. The weight of the rats and the arthritis index scores were measured.3. Joints examinations were performed by UBM just as previously described.4. Before and after drug intervention, the VEGF, TNF-α and αvβ-3level in serum of the experimental rats were detected by ELISA.5. Synovial VEGF, TNF-α, αvβ-3expression level were measured using mouse VEGF, TNF-a, avP-3immunohistochemistry kit in strict accordance with the operating instructions for the joints which synovial pathology score≥1. The method was same as previously described.6. SPSS17.0statistical package was appliced for statistical analysis. The experimental data and measurement data were indicated by means±standard. The data comparison among several groups was made by means of one-way ANOVA. LSD method was used when test for homogeneity of variance. If heterogeneity of variance was adopted, Welch approximate variance analysis and multiple comparisons were used between groups. Pearson and Sperman correlation analysis were used for relevance comparison. Enumeration data was test by X2. A value of P<0.05was considered statistically significant.Results1. Before drug intervention, there was no significant difference for Sonographic Ultrasonographic features in three groups (P>0.05). After treatment, comparied with control group, the count of joint appearing flow and flow PDI grade reduced in both bevacizumab group and etanercept group, and there were significant differences (P <0.05); Meanwhile, the count of joint appearing flow and flow PDI grade had no difference between bevacizumab group and etanercept group(P>0.05).2. Before drug intervention, serum VEGF, TNF, αvβ-3levels had no differences in the three groups(P>0.05); while a significant difference appeared after treatment(P <0.01). Comparied with control group, the level of bevacizumab group and etanercept group declined, there was significant differences (P<0.01). There was no significant difference the between bevacizumab group and etanercept group(P>0.05).3. After bevacizumab and etanercept treatment, AI and Ultrasonographic features decreased with significant differences (P<0.05).4. In control group, compared with before treatment, synovial thickness and PDI grade increased with statistically significant difference(P<0.05). Before and after treatment, AI, atthritis joint count detected by UBM, the number of joints with blood and UBM grade didn’t change significantly(P>0.05).5. After drug intervention, there was no significant difference for pathological joint count among the three groups(P>0.05); Synovial VEGF, TNF-a, αvβ-3expression and pathological scores were significantly different (P<0.05) among the three groups; Blank control group were higher than the indicators of bevacizumab group and the etanercept group, there was a significant difference(P<0.05);There was no significant difference between bevacizumab and etanercept group (P>0.05).6. Synovial thickness decreased after drug intervention, PDI and synovial pathology grading score showed a positive correlation VEGF, TNF-α, αvβ-3expression levels respectively(P<0.05). Conclusions1. Bevacizumab and etanercept can suppress early RA synovial hyperplasia, decrease synovial angiogenesis and lower synovial pathology score. There was no difference between the treatment of the two drugs.2. Synovial thickness, PDI grade can be used as sonographic markers to assess synovial hyperplasia and pannus formation in early RA, and value response to treatment. compared with synovial thickness, PDI is more sensitive in anti-angiogenic therapy.3. UBM has application value for RA drug trials and efficacy.更多还原