【作者】 李夏林；

【导师】 靳安民；

【作者基本信息】 南方医科大学 ， 外科学， 2014， 博士

【摘要】 临床上治疗粉碎骨折,肿瘤切除,或骨髓炎病灶清除后大段骨缺损仍然是骨科医生面临的一大挑战。自体骨移植一直作为骨缺损治疗的金标准,然而存在着诸如取骨区疼痛,出血,来源有限等不足。同种异体骨,作为自体骨替代品,近些年在临床上应用也较广泛,但其也存在成本较高,病毒传播潜在风险及不能完全避免的免疫排斥反应等问题,限制了其在临床大规模使用。作为自体骨及同种异体骨的替代品,组织工程骨的研究有望彻底解决这一难题。具备骨诱导活性的复合生物材料是组织工程骨的一个研究热点。一些有着较强的诱导成骨活性的生长因子,如重组人骨形态发生蛋白-2(rhBMP-2)及重组人骨形态发生蛋白-2(rhBMP-7),已结合不同的生物材料作为载体,被美国FDA批准用于临床上促进骨折愈合及脊柱融合。但BMP制剂广泛应用于临床后发现由于使用剂量远高于生理水平,活性因子的释放动力学不够科学等原因,出现了诸如局部软组织水肿,脊髓脊神经根炎,异位骨化,种植体周围骨吸收及引发癌症等一系列的并发症。理想的,可以广泛地应用于临床的有成骨诱导能力的人工骨填充材料仍需要进一步研究不断完善。在骨折的自然愈合过程中,有多种细胞因子共同参与,如BMP家族,血管内皮生长因子(VEGF),胰岛素样生长因子(IGF)等,他们的成骨机制包括：诱导间充质干细胞(MSCs)成骨分化,促进局部新生血管的生成,作为细胞激活分化的调控因子等,促进骨组织愈合。BMP家族中BMP-2和BMP-7是被研究得最多,有较强的成骨诱导活性的两种因子。之前的研究表明,BMP-2是骨折初始阶段各种级联反应的启动因子,在骨折后的第一天即大量表达至峰值,而后持续低浓度表达。而BMP-7则在骨折2周后才出现。最近一些研究表明,BMP-2和BMP-7的复合使用比单因子更能促进骨组织再生和骨折愈合。所以,如果研发一种骨修复材料可以仿生控释BMP-2和BMP-7,则可以在安全浓度范围内,以较低的浓度,较少的制备成本最大限度地促进使骨缺损修复。微型包囊技术系利用天然的或合成的高分子材料(囊材)作为囊膜壁壳,将固态药物或液态药物(囊心物)包裹而成药库型微型胶囊。药物在囊内通过扩散和渗透等形式在特定的位置以适当的速度和持续的时间释放出来,以达到更大限度的发挥药效的作用。到目前为止已有200多种药物采用了微囊化技术,尤其是近年来生物大分子作为药物在临床上广泛应用,包括蛋白质,多肽类,基因,生长因子,抗原等都能成功地被包封制成缓释微囊,从而最大限度地保护生物大分子的结构和活性,并能达到靶向给药,控释给药目的,使得药物释放更可控,更安全,更有效,从而带动了制药工业的飞速发展。近年来已有不少学者尝试将BMPs以不同的高分子聚合物包封成微囊制成缓释制剂,或与其他无机材料共同合成能缓释活性因子的人工骨支架,细胞实验及动物实验结果令人满意,如何让缓释系统释放出的BMPs最大限度地保留其诱导成骨活性,又不至于剂量过高引起副作用,是现在BMPs缓释载体研究的新的趋势。一直以来,载药微囊作为控释载体大都作为支架或水凝胶组成成分之一。2007年Ana Jaklenec等试验了全新的支架合成方法,将包封小牛血清蛋白(BSA)的PLGA微囊通过二氯甲烷密闭熏蒸法直接融合成支架并获得成功。次年,这个研究团队继续用该方法制备了由包封胰岛素样生长因子-1及转化生长因子β1的微囊融合成的支架用于软骨组织工程的研究,这种支架能成功地缓释这两种生长因子并保留其生物活性。该方法制备过程简单,可通过控制微囊大小及熏蒸时间来控制支架的孔隙率,体外细胞培养可让纤维母细胞顺利附着为其扩增提供支持。这为组织工程骨及软骨支架的制备提供了一条崭新的思路。然而,其进一步的生物学实验成果并未见报道。将包封活性因子的高分子聚合物微囊直接熏蒸融合制成微球支架,它最大的优点在于制备过程简单,一步到位,避免了以往复合支架制备过程中反复添加不同有机溶剂,化学稳定剂及交联剂等,而且不需要旋蒸,冻干,高温,高压等步骤,这样最大限度地避免了制备过程中对功能蛋白质的结构和活性产生的负面影响。同时,合成的支架具有一定的粘弹性,可以根据骨缺损的不同形态予以塑性,并且电子显微镜证实其有一定的孔隙率,适合细胞贴附及营养液的渗入。本研究首先利用复乳溶剂挥发法制备了包封rhBMP-2的微囊,聚乳酸-聚乙二醇-聚乳酸三嵌段共聚物(PELA)被用来作为微囊的囊材。我们期望通过响应面法优化实验参数以得到最高包封率的微囊。然后应用并改良Ana Jaklenec等的方法制备微囊支架用于促进骨愈合。制备的微囊其外粘附rhBMP-2,其内包封rhBMP-7。用这种微囊融合成的支架能序贯释放rhBMP-2及rhBMP-7,并进一步研究这种支架材料释放BMPs的控释曲线是否能模拟自然骨愈合过程中的BMPs的表达特点。最后通过大鼠的骨髓间充质干细胞体外实验及大鼠活体内骨缺损修复实验来进一步验证材料的成骨能力,降解能力及细胞毒性。目的：研发出制备过程简单,具有仿生控释BMP-2和BMP-7双因子的新型微球支架,其组成为BMP-2/PELA/BMP-7。研究这种生物活性支架的粘弹性,表征,生长因子的缓释曲线,缓释出因子的生物活性,生物相容性,诱导骨髓干细胞成骨转化能力。并进一步在活体骨缺损动物模型中评价其成骨能力,为今后骨缺损治疗的应用提供一种新方法和新思路。方法：1负载rhBMP-2的PELA微囊制备及包封率优化研究将改良复乳溶剂挥发法制备包封rhBMP-2的PELA微囊,通过单因素实验研究对微囊包封率影响最大的几个实验参数,包括PELA中PEG嵌段的分子量,投入的PELA的质量及rhBMP-2的质量,内水相乳化剂浓度,外水相中聚乙烯醇(PVA)浓度以及微囊固化的搅拌时间。并在单因素实验结果的基础上选取三个对包封率影响最大的实验参数,应用响应面法(RSM)设计三因素三水平的实验方案,优化实验参数,以得到最佳rhBMP-2包封率的PELA微囊。每个样本均设三个重复样以减少实验误差。采用DPS13.0软件行实验设计及其数据分析,包括回归分析及回归方程系数的确定。用STATISTICA7.0软件绘制的三维响应面图来描述自变量对于应变量的单个效应和交互效应。用方差分析检验该模型的统计学差异,P<0.05被认为差异有统计学意义。最后用扫描电镜观察制备的微囊的形态,大小,并计算平均直径。2BMPs/PELA微囊支架材料制备及体外缓释特点通过改良Ana Jaklenec等的微囊支架制备方法,用PLA-PEG-PLA三嵌段共聚物(PELA)作为微囊的壁材,制备出其内包封rhBMP-7,其外粘附rhBMP-2的微胶囊,再通过二氯甲烷熏蒸将微囊聚合成支架材料(BMP-2/PELA/BMP-7,A组),并同时制备仅包封rhBMP-7的微囊支架(PELA/BMP-7,B组),仅表面粘附rhBMP-2的微囊支架(BMP-2/PELA,C组)以及单纯PELA微囊支架(D组)作为对照组。本部分首先研究这种支架材料的制备方法,将熏蒸不同时间的微囊支架在扫描电镜下观察其表面形态。再将微囊支架浸泡在37℃的PBS(pH7.4)缓冲液中,在固定的时间点称重,在46天内记录其干重及膨胀率的变化,以了解支架材料体外降解的性能。然后以同样的实验条件,在不同时间点检测缓释液中BMP-2及BMP-7的浓度,以获得微囊支架材料BMPs的释放曲线。每个样品均设六个重复样以减少误差。3BMPs/PELA微囊支架材料诱导大鼠骨髓间充质干细胞成骨分化实验研究采用体外培养的大鼠骨髓间充质干细胞(MSCs)来检验BMPs/PELA微囊支架材料的成骨能力及细胞毒性。先用全骨髓分离法分离培养大鼠骨髓间充质干细胞,通过观察细胞形态及流式细胞仪检测细胞表面标志物进一步鉴定；将微囊支架与MSCs共培养,在第3,7,14天时通过MTT实验来检验支架材料是否会对MSCs的活性产生影响；将支架的体外缓释液加入细胞培养液中诱导MSCs成骨分化,在第7天及14天时通过细胞碱性磷酸酶(AKP)活性测定来检验各组微囊支架缓释的BMP的生物活性及各种支架材料诱导MSCs成骨分化的能力。每组实验均设6个重复细胞样以减少误差,用SPSS19.0软件进行分析实验结果,组间数据比较用单因素方差分析法,两两比较用SNK法,P值小于0.05被认为差异有统计学意义。4BMPs/PELA微囊支架材料修复大鼠骨缺损实验研究建立大鼠股骨髁大块骨缺损的实验模型,于活体内检验各组微囊支架材料诱导成骨的能力。将24只大鼠随机分为4组,在全麻下于股骨髁部钻孔形成5×5mm圆形单层皮质的骨洞,分别用A, B, C, D四组支架材料填入检验其体内成骨效果。于术后第4周及第8周取材,观察大鼠的一般状况及股骨缺损部位软组织及骨组织的大体情况；将取下的股骨通过微型CT (mCT)扫描进行影像学分析以及新生骨各参数的测量；再通过HE染色及Masson三色染色法进一步行组织学观察评估新生骨生成及支架材料体内降解吸收情况。实验结果以SPSS19.0软件进行分析,组间数据比较用单因素方差分析,两两比较用SNK检验,P值<0.05被认为差异有统计学意义。结果：1负载rhBMP-2的PELA微囊制备及包封率优化研究本研究成功地以复乳溶剂蒸发法制备了包封rhBMP-2的PELA微胶囊。并应用了响应面法设计优化了制备过程中的关键参数以得到最高包封率的微囊。制备过程中一些参数,包括聚合物中PEG嵌段的分子量,投入原料的量(PELA和rhBMP-2),复乳液的搅拌固化时间,乳化剂的浓度及外水相中PVA浓度等均可影响最终微囊的rhBMP-2包封率,其最优值分别为：PEG嵌段分子量4000Da, PELA330mg, BMP3ìg,司盘-20浓度0.5%,聚乙烯醇浓度0.5%,在800rpm的转速下固化搅拌时间为30分钟。在单因素实验结果的基础上,我们选取PELA的质量(X1),PVA的浓度(X2)和rhBMP-2质量(X3)这三个对包封率影响最大的三个因素作为自变量,考察它们对包封率(Y)的影响。通过Box-Behnken法设计三因素三水平的响应面实验方案,得到的二次多项式回归方程模型：Y=27.33602+0.29086X1+86.69629X2-30.05764X3-0.00055X12-48.92593X22+2.79563X32+0.02088X1X3-7.36111X2X3方程回归的F值是47.8993,相应的P值为0.0001(<0.01),回归方程有统计学意义,其R2(决定系数)为0.9846。三维响应面图显示出PVA的浓度与rhBMP-2质量,及PELA的质量与rhBMP-2质量均对最后的包封率产生交互效应。以此模型推导出最佳的实验参数为：PELA的量为282.3mg,rhBMP-2量为1.0μg,PVA浓度为0.8%。在此条件下推算出微囊的包封率理论最大值为76.5%,而以此参数制备的微囊实际的rhBMP-2包封率为75.0%,仅存在微量误差,这也进一步证实该模型能很好地代表各变量值与响应值(包封率)之间的关系。电子显微镜下可以观察到大部分微囊都呈现出完整的圆球形,其直径分布基本均匀一致,微囊平均直径为40μm。2BMPs/PELA微囊支架材料制备及体外缓释特点参考第一部分得到的优化的实验参数,用改良的复乳溶剂挥发法成功制备出内包封BMP-7,其外粘附BMP-2的微胶囊。然后采用二氯甲烷熏蒸法能将其成功融合成多孔的三维支架材料(BMP-2/PELA/BMP-7)。并同时制备单因子材料组PELA/BMP-7和BMP-2/PELA,以及单纯PELA材料组微囊支架作为对照组。这种支架制备方法优点在于制备过程简单,一步到位,很大程度上避免了添加各种化学试剂及使用高温,紫外线照射交联等物理方法对功能蛋白质的结构和活性产生的负面影响,同时也节约了材料的制备成本。这种骨修复材料还具有一定的粘弹性,可以根据骨缺损的不同形态予以塑性。扫描电镜显示融合后的骨填充材料,其球状微囊的基本结构仍清晰可见。材料表面及剖面均可见一定的孔隙率,孔隙大小范围为50-200μm。体外降解实验表明：在磷酸盐缓冲液(PBS)中开始的两个星期,材料的膨胀率增加,于14日达到了峰值后开始快速。而材料干重在前两周无明显变化,2周后开始迅速下降,表明支架材料开始降解。缓释实验表明：BMP-2/PELA/BMP-7微囊支架材料能同时缓释BMP-2及BMP-7两大成骨活性因子,但释放曲线明显不同。rhBMP-2呈现出明显的初始突释现象,随后是一个较长时间的缓慢释放；而rhBMP-7初始突释也存在,但明显被抑制,而在缓慢释放2周后,其释放速度呈现出一个小幅加速。这两种活性因子的控释时间均长于42天。这些缓释特点能简单模拟自然骨折修复过程BMPs表达的特点。3BMPs/PELA微囊支架材料诱导大鼠骨髓间充质干细胞成骨分化实验研究用MTT法检测与各组支架材料共培养的大鼠MSCs的活性,结果表明,在共培养的第3天(F=2.08,P=0.135),第7天(F=O.12,P=0.954)及第14天(F=43.98,P=0.732)时,BMP-2/PELA/BMP-7组,PELA/BMP-7组和BMP-2/PELA组支架材料与纯细胞对照组的OD值差异无统计学意义。这结果表明BMP/PELA支架材料在短期内对大鼠MSC细胞活力没有负面影响,短期内未发现明显细胞毒性。再进一步通过检测细胞AKP活性来评价MSCs成骨分化的程度。于MSCs的培养液中加入各组支架材料缓释液,在培养第7天及14天的时候检测各组细胞的AKP活性,其结果表明：第7天时各组细胞AKP表达量有统计学差异(F=28.57,P=0.000),与纯PELA组相比,BMP-2/PELA/BMP-7组,PELA/BMP-7组和BMP-2/PELA组均能增强细胞的AKP活性,差异有统计学意义(P值均=0.000)而这三组之间AKP值差异无统计学意义。到第14天,各组细胞AKP值的总体差异也具有统计学意义(F=43.98,P=0.000)。BMP-2/PELA/BMP-7组中细胞的AKP活性高于PELA/BMP-7组和BMP-2/PELA组,差异有统计学意义(P=0.007及P=0.014)。PELA/BMP-7组和BMP-2/PELA组间差异无统计学意义。而纯PELA组的AKP值低于PELA/BMP-7组(P=0.000)和BMP-2/PELA组(P=0.007),其差异有统计学意义。这一结果提示：微囊支架缓释的的BMP仍保留很强的诱导干细胞成骨分化的生物学活性,而BMP-2和BMP-7这两大成骨活性因子对诱导MSCs成骨分化存在着协同效应。4BMPs/PELA微囊支架材料修复大鼠骨缺损实验研究大鼠手术后均成活,均未出现病理性骨折,伤口流脓,窦道等情况。取股骨标本,见所有标本周围软组织均未见脓肿,纤维化,骨组织未见死骨,大部分骨缺损生物材料填充处均可见新生组织覆盖良好。股骨mCT扫描可见：术后第4周时BMP-2/PELA/BMP-7组股骨缺损部位的新生骨较其他组多；PELA/BMP-7组和BMP-2/PELA组形成的新生骨多于纯PELA组。术后第8周各组骨缺损部的新生骨均多于第4周；BMP-2/PELA/BMP-7组的新生骨组织完全覆盖骨缺损部,并且可见新生骨皮质；PELA/BMP-7组和BMP-2/PELA组可见更多的新生骨；纯PELA组8周可在缺损部位发现少部分的新生骨组织.再通过mCT的几个重要参数来分析大鼠骨缺损部的新生骨,包括骨密度(BMD),新生骨骨体积分数(BV/TV),骨小梁数量(Tb.N)等,其结果表明：各组的骨密度值(BMD)术后第4周时各组差异有统计学意义(F=58.86,P=0.000),BMP-2/PELA/BMP-7组新生骨BMD均高于其他各组,差异有统计学意义(P值均=0.000),PELA/BMP-7组和BMP-2/PELA组骨密度值差异无统计学意义(P=0.54),但均高于PELA组(P=0.004；P=0.001)。而至术后第8周时,各组的BMD值总体差异也有统计学意义(F=22.57,P=0.000),BMP-2/PELA/BMP-7组新生骨BMD值仍高于PELA/BMP-7组(P=0.000),BMP-2/PELA组(P=0.039)和PELA组(P=0.000),差异有统计学意义。BMP-2/PELA组新生骨BMD均值要高于PELA/BMP-7组,差异则有统计学意义(P=0.013)。各组的BV-TV值在术后第4周时总体差异有统计学意义(F=171.34,P=0.000)。BMP-2/PELA/BMP-7组的BV/TV值高于其他各组,差异有统计学意义(P值均=0.000)。PELA/BMP-7组和BMP-2/PELA组的BV/TV值无统计学差异(P=0.134),而均高于PELA组,差异有统计学意义(P=0.044,P=0.001)。术后第8周时,各组总体差异有统计学意义(F=67.95,P=0.000),BMP-2/PELA/BMP-7的BV/TV值仍高于其他各组,差异有统计学意义(P值均=0.000)。BMP-2/PELA组BV/TV值高于PELA/BMP-7组,其差异有统计学意义(P=0.002)。而PELA组的BV/TV值在术后第8周仍低于BMP-2/PELA/BMP-7组(P=0.000),PELA/BMP-7组(P=0.035)和BMP-2/PELA(P=0.000),差异有统计学意义。各组新生骨的骨小梁数量(Tb.N)也体现出类似的趋势。术后第4周时方各组总体差异有统计学意义(F=20.35,P=0.000)。BMP-2/PELA/BMP-7组新生骨Tb.N值高于其他各组,差异有统计学意义(P值均=0.000)。PELA/BMP-7组和BMP-2/PELA组Tb.N值差异无统计学意义,而均大于PELA组(P=0.005及P=0.010),差异有统计学意义。术后第8周时各组总体差异也有统计学意义(F=12.01,P=0.000),BMP-2/PELA/BMP-7组Tb.N值仍高于PELA/BMP-7组(P=0.000)和PELA组(P=0.000),差异有统计学意义,而与BMP-2/PELA组比较无统计学差异。BMP-2/PELA组Tb.N值高于PELA/BMP-7组,差异有统计学意义(P=0.018)。这些数据结果表明：与BMP-2或BMP-7单因子缓释材料相比,BMP-2/PELA/BMP-7支架材料有更强大的活体内成骨诱导能力；而BMP-2/PELA支架材料中远期的成骨诱导能力强于PELA/BMP-7材料；所有复合BMPs的PELA微囊支架材料均比单纯的PELA聚合物显示出更好的成骨效果。大鼠股骨的组织学观察可见：BMP-2/PELA/BMP-7组在术后4周时可见新骨形成,其周围有残留的部分降解的材料和大量的血细胞,其切片上可见到共聚物材料降解并被新生骨替代的动态过程。到术后第8周,新生骨组织体积更大,而且成熟度更高,可以见到较多类似于成熟骨皮质的板层骨,在骨基质中可以发现带有血管的完整的骨单位。PELA/BMP-7组和BMP-2/PELA组在术后第4周时均能观察到部分新生的编织骨,骨基质中能发现支架材料的残留。到术后第8周,可以见到更多的新生骨组织,但没有发现成熟的板层骨及骨单位。而单纯的PELA组术后第4周无新骨形成,到术后第8周才能见少量新骨组织出现并开始取代支架材料。这些组织学结果与mCT图像结果相一致。结论：用改良的复乳溶剂挥发法成功制备出包封rhBMP-2的PELA微囊,并通过响应面法优化实验参数,微囊包封率达75%。然后用二氯甲烷熏蒸法将负载BMPs的微囊融合成多孔支架材料,可以简单模拟自然骨折修复过程BMPs表达的特点缓释BMP-2和BMP-7,其缓释时间超过42天。该材料具备较强的诱导大鼠间充质干细胞成骨分化的能力,并且生物相容性好,可完全降解,短期内未检测到细胞毒性。活体动物实验证实其能很好地修复大鼠股骨大块骨缺损,这为其今后广泛应用于临床治疗骨缺损提供了理论依据。更多还原

【Abstract】 The treatment of large bone defect caused by comminuted fracture, tumor excision, or osteomyelitis still remains a challenge in clinic. As the gold standard for bone defects’ treatment, autogenous bone grafts bring problems such as donor site pain and limited supply. Allogeneic bone are alternatives but the issue is high cost, risk of virus transmission and adverse host immune reaction which restrain their application. Osteoinductive biomaterials are a promising method to solve this problem and provide an alternative to autogenous bone or allografts. Some growth factors for promoting bone healing, such as bone morphogenetic protein-2(BMP-2) and bone morphogenetic protein-7(BMP-7), have been combined with biomaterials as carriers and used in clinic for fracture healing or spinal fusion. However, the doses of BMP used in clinical situation was far above physiological levels which causes many side effects such as local soft tissue edema, spinal radiculitis, ectopic ossification, bone resorption around implant and cancer risk. So, the ideal synthetic bone graft substitutes which can be widely used in clinical situation needs further research.In the natural process of bone healing after fracture a variety of growth factors, such as BMPs, vascular endothelial growth factor (VEGF), and insulin-like growth factor (IGF) can regulate cellular activities and induce mesenchymal stem cells (MCS) to osteogenic differentiation and promote bone formation. BMP-2and BMP-7have been proved to have strong osteoinductive activity. Previous study have indicated that BMP-2reaches the peak on the first day after the fracture, and BMP-7emerge after2weeks in the natural healing process of fracture. Recently some studies have shown that the combined delivery of BMP-2and BMP-7enhances bone regeneration and healing of fracture compared with single factor delivery. So if scaffolds can release these growth factors sequentially and slowly in a biomimetic manner, it would optimize the effect of bone defect repair within the safe concentration range.Microcapsule technology just uses natural or synthetic polymer material as the capsule wall material to encapsulate the solid or liquid drugs for microcapsule. By diffusion and permeation mode wrapped drug is released in an appropriate rate in the setted position to be more effective. So far, more than200kinds of drugs have been encapsulated as microcapsule form. In recent years microcapsule technology have been successfully used in biological macromolecules, including proteins, polypeptides, genes, growth factors and antigens, which can protect of the structure and activity of biological macromolecules and achieve its control release style to make it more controllable, effective and safer. Recently, many scholars have tried to encapsulate BMPs with different polymer to prepare microcapsules for sustained release or fabrication of artificial bone scaffold. The in vitro and in vivo experiments got satisfactory results, so how to optimize the BMPs delivery system to avoid side effects caused by high doses and keep its osteoinductive activity as soon as possible remain the trend of present research.For a long time, microcapsule has been only used as one controlled release component of scaffolds or hydrogel for tissue engineering. In2007, Ana Jaklenec et al invented a simple novel method to fabricate scaffolds fused only by protein-loaded microspheres using dichloromethane vapor for control release. In the following year, he continued to prepare a scaffold fused by IGF-I and TGF-β1loaded microcapsules for cartilage tissue engineering. The final scaffold could succeed to deliver bioactive growth factors in a sequential manner and have shown its promising applications in cartilage or bone tissue engineering field. However no further relative studies were reported in this area. The advantage of this microcapsule-based scaffold is its simple progress of fabrication. The one-step preparation process can avoid adding various organic solvents, chemical stabilizers or cross-linking agent, and do not require freeze-dried, high temperature, high pressure and other physical processing, so it can protect the structure and bioactivity of proteins. Meanwhile, the synthetic viscoelastic character is helpful to repair bone defect with different forms, and the interlinked pores are suitable for nutrient solution infiltration.In the present study, the rhBMP-2loaded microcapsules were prepared using double emulsion solvent evaporation method firstly. PLA-PEG-PLA triblock copolymer (PELA) was used as wall material of microcapsule. We wanted to find optimal entrapping conditions for high encapsulation efficiency by response surface methodology. Then we employed and modified the Ana Jaklenec et al. method to prepare the scaffolds for bone healing. The scaffolds fused directly by BMPs loaded microscapsules was fabricated to promote impaired bone healing. The microcapsules encapsulating BMP-7were covered by BMP-2. The scaffold fused by these microcapsules could deliver BMP-2and BMP-7sequentially. Further, we investigated whether the BMPs release style of the scaffold can mimic the natural bone regeneration cascade. The release kinetics of growth factors and degradation character were studied. The effect on bone regeneration was investigated in vitro using rat’s MSCs and in vivo through rat’s femoral bone defect models.Object:To fabricate the microcapsule-based scaffold (BMP-2/PELA/BMP-7) which can be prepared simply and release BMP-2and BMP-7sequently. We will investigate its viscoelasticity, morphology, biocompatibility, release kenitics of growth factors, osteogenic induction ability and the bioactivity of released BMPs. Further we will employ rat’s bone defect model to evaluate the scaffold osteogenic capability in vivo and we hope to provide a new approach for future therapy of bone defects in clinic.Method:1Preparation of rhBMP-2loaded PELA microcapsules and parameters optimization for high encapsulation efficiency (EE) The rhBMP-2loaded PELA microcapsules were prepared using modified double emulsion solvent evaporation method. This work was conducted to optimize entrapping conditions of rhBMP-2loaded PELA copolymer. The effect of different molecular weight of PEG in copolymer, PELA amount, rhBMP-2amount, span-20concentration, polyvinyl alcohol (PVA) concentration and stirring time on EE tested. On the basis of single-factor experiments, the optimum parameters for highest EE were achieved by response surface methodology (RSM). The effects of three independent variables at three levels were investigated using Box-Behnken experimental design with software DPS13.0and software STATISTICA7.0was employed to draw the response surface plot for EE of rhBMP-2. The statistical significance of the model was determined by the application of F-test and P<0.05was considered statistically significant difference. Morphology of these microcapsules is observed by scanning electron microscope (SEM).2Preparation of BMPs/PELA microcapsule-based scaffold and its release keniticsIn this study, we employed and modified the Ana Jaklenec et al. method to prepare the microcapsules scaffolds. PLA-PEG-PLA triblock copolymer (PELA) was used as wall material of microcapsule. The microcapsules encapsulating BMP-7were covered by BMP-2. The scaffold was fused directly with these microscapsules by dichloromethane vapor (BMP-2/PELA/BMP-7, group A). Meanwhile PELA/BMP-7(group B), BMP-2/PELA (group C) and PELA (group D) scaffolds were fused as control group. In this chapter preparation method of scaffold was studied and the morphology of was imaged by SEM. Scaffold’s swelling and degradation tests, were performed in10ml PBS (pH7.4) at37℃. The weight loss and swelling ratio were recorded within46days. The BMPs’ release kenitics of the scaffold were studied in the same condition. Each sample was run in six copies to minimize the error.3Effect of BMPs/PELA scaffold on rat’s mesenchymal stem cells’(MSCs) osteogenic differentiation in vitroRat’s MSCs were employed to investigate osteogenic ability and cytotoxicity of BMPs/PELA scaffold. First bone marrow MSCs were isolated from male wistar rats and identified through the observation of cell morphology and detection of cell surface markers with flow cytometry. MTT assay was used to evaluate the activity of MSCs seeded on scaffold at day3,7and14. To estimate the bioactivity of released BMPs from scaffolds and its ability to induce osteogenic differentiation of MSCs, cells’alkline phosphatase (AKP) activity in different group were test. Six copies for each sample were carried out to reduce the error. All data are presented as mean+standard deviation. One-way ANOVA followed by SNK test was performed to compare multiple groups of data with SPSS19.0software. The difference in groups was considered significant if P<0.05.4BMPs/PELA scaffold repair the rat’s bone defect in vivoWe investigated the in vivo effect of BMPs loaded PELA scaffolds in bone regeneration through rat’s femoral bone defect models. Twenty-four wistar rats were randomly assigned to four groups. Under general anesthesia a5mm×5mm unicortical hole in condyle was filled with50mg biomaterials. At4and8weeks after implantation, the rats were sacrificed and the femurs removed. The general situation of femur and soft tissue around was observed. Quantitative3D analysis of bone ingrowth in the scaffolds was performed by micro-computed tomographic (mCT). The newly formed bone in each group was analyzed with the parameters the ratio of bone volume and total defected volume (BV/TV), bone mineral density (BMD) and trabecular number (Tb.N). To obtain more detailed information about new bone tissue and degradation progress of scaffold in vivo, histological analysis was performed with hematoxylin and eosin (HE) and Masson Trichrome methods. The results were analysed by One-way ANOVA method with software SPSS19.0, followed by SNK test for the determination of the significance of difference among groups. Statistically difference was defined as P<0.05.Result:1Preparation of rhBMP-2loaded PELA microcapsules and parameter optimization for high EEIn this study, it was successful to employ response surface methodology to optimize the preparing parameters of double emulsion solvent evaporation technique for high EE of rhBMP-2loaded PELA microcapsules. In preparing process, many factors such as the molecular weight of PEG block in polymer, amount of raw material (PELA and rhBMP-2), stirring time of double emulsion, concentration of emulsifier in oil phase and PVA concentration in outer aqueous phase could influence the final EE of rhBMP-2apparently. The result showed that the highest EE of BMP-2was achieved at PEG molecule weight4000Da, PELA330mg, BMP3μg, span-20concentration0.5%, PVA concentration0.5%, stirring time30min in800rpm/min.Then the effects of three independent variables X1(PELA), X2(PVA concentration) and X3(rhBMP-2) at three levels on EE were investigated using Box-Behnken experimental design. The average EE was taken as the response Y. An second order polynomial model be found and the R square was98.46%. Y=27.33602+0.29086X1+86.69629X2-30.05764X3-0.00055X12-48.92593X22+2.79563X32+0.02088X1X3-7.36111X2X3. The3D response surface plot showed that rhBMP-2amount had significant interaction with amount of PELA and PVA concentration on EE. The optimal preparing parameters could be determined from this model:the amount of PELA282.3mg, rhBMP-21μg and PVA concentration0.79%. Under these conditions, the actual EE value (75%) was in agreement with the predicted highest EE (76.5%). The data proved that the developed model could adequately represent the real relationship between the chosen parameters and EE. Most of microcapsules were rounded and the mean diameter of microcapsule was40μn by SEM.2Preparation of BMPs/PELA microcapsule-based scaffold and its release keniticsMicrocapsules encapsulating BMP-7and covered by BMP-2were prepared using the improved double emulsion solvent evaporation technique, as our previous method. Porous scaffolds (BMP-2/PELA/BMP-7) were fused by the dichloromethane vapor. Even after fusion the spheres’ underlying structure was still visible with SEM. Its inherent pores could be found with the size ranged from50μm to200μm. Scaffolds’ swelling and degradation tests showed that in the first two weeks, the swelling ratio increased obviously and attained a peak in day14in PBS solution. There was no significant mass loss until day12, then the weight began to decease quickly. In vitro BMPs release study showed that the BMP-2/PELA/BMP-7porous scaffolds could control release rhBMP-2and rhBMP-7sequently. The release profile of two growth factors were different. RhBMP-2had a burst release on day2, followed by a long sustained release until the end of experiment. To rhBMP-7the initial burst release was relatively low. After a period of slow release an accelerated release was found at2weeks later. Controlled-release interval was more than42days which can partially mimic the BMPs’ release profile in natural fracture healing process.3Effect of BMPs/PELA scaffold on rat’s MSCs’ osteogenic differentiation in vitroMTT assay was used to evaluate the rat’s MSCs activity which cultured with BMPs/PELA scaffold. After statistical analysis cells’ activity had no significant different between PELA scaffolds and positive control at day3(F=2.08, P=0.135), day7(F=0.12, P=0.954) and day14(F=43.98, P=0.732). The result indicates the BMPs loaded PELA scaffold has no negative impact on cell activity in the short term.Then MSCs were used to cultured with the releasates of4groups of scaffolds and the AKP activity was test on day7and14. Results showed that the difference at day7was statistically significant (F=28.57, P=0.000). The releasates from group A, B and C could markedly enhance cells’AKP activity at day7compared to the PELA group (P=0.000) and there was no statistical difference between these three groups. At day14the difference of AKP activity was also statistically significant (F=43.98, P=0.000). The group A was significantly higher than the group B(P=0.007) and group C (P=0.014). There was no statistical difference between group B and C. The PELA group’s AKP activity was lower than B (P=0.000) and C group (P=0.007). Thus our observations suggest that:BMPs/PELA scaffold has not been found with obvious cytotoxicity in a short term; the BMPs released from BMPs/PELA scaffold can keep their bioactivity and incorporation of BMP-2and BMP-7has the synergistic effect on promotion to rat’s MSCs’osteogenic differentiation.4BMPs/PELA scaffold repair the rat’s bone defect in vivoBMPs/PELA scaffold could repair the rat’s femoral defect so well. The femur’s mCT analysis showed that at week4after implantation the group A enhanced more new bone ingrowth in femoral defect than other groups, and new bone formation in the group B and C were more than the group PELA, which implied BMP-2and BMP-7released from scaffolds could keep bioactivity and promote new bone formation compared to control group. At week8all groups showed more new bone formation than week4. The femoral hole filled with the group A was covered completely by new bone tissue and full cortical bone could be found. There was more new bone formation in the group B and C than group D, the group C seemed to have stronger ability to promote bone healing than the group B. And for the group D, there was partial new bone could be found in the defect site until8weeks.The regenerated bone mineral density (BMD), the volume of new bone formation per total scaffold volume (BV/TV) and the trabecular number of defect area (Tb.N) were calculated for more accurately analysis. All parameters of group A were higher than other groups at week4and week8. The BMD’s difference of each group was statistically significant at week4(F=58.86, P=58.86). The group A was higher than other groups significantly (P=0.000). The difference between group B and C was not significant (P=0.54), but both were higher than group D (P=0.004; P=0.001). At week8the BMD’s difference was also statistically significant (F=22.57, P=22.57). The group A was still higher than group B (P=0.000), group C (P=0.039) and group D (P=0.000). The mean BMD of group C was high than group C (P=0.013).The difference of BV/TV value between four groups was statistically significant at week4(F=171.34, P=0.000). The group A was higher than other groups significantly (P=0.000); The difference between group B and C was not significant (P=0.134), but both were higher than group D (P=0.044; P=0.001). At week8the difference of all groups was also statistically significant (F=67.95, P=0.000). The group A was still higher than other groups(P=0.000). The group C had higher BV/TV value than group B (P=0.002). The Tb.N value of all groups showed similar trend. At week4, the difference of Tb.N between four groups was statistically significant (F=20.35, P=0.000). The group A was higher than other groups significantly (P=0.000); Both group B and C were higher than group D (P=0.005; P=0.010). At week8the difference of all groups was also statistically significant (F=12.01, P=0.000). The group A was still higher than group B (P=0.000) and group D (P=0.000). The group C was high than group B (P=0.018). These results indicated in vivo BMP-2/PELA/BMP-7material have the stronger osteogenetic ability than other groups; The long term osteoinductive ability of BMP-2/PELA scaffold was stronger than PELA/BMP-7scaffold; All BMPs combined PELA scaffold material shows better osteogenetic effect than simple PELA polymer.Histological analysis of rats’ femur showed that in the group A new bone was observed to surround the residual materials and plenty of blood cells were around at4weeks. The copolymers seemed to degrade along with the new bone formation. At week8, much more and thicker new bone was observed and tended to become lamellar bone, Some osteon with vessels can be seen in the bone matrix. In both the group B and C, limited new woven bone combined with materials was observed in the bone matrix at week4, and material’s degradation product was found to be surrounded by new bone tissue. At week8, more new bone was observed, but no matured lamellar bone was found in both groups. In the group PELA, no new bone formation was found at week4. Until week8new bone began to appear and replace the biomaterials. The results of histological observation consisted with the image of mCT.Conclusion:The rhBMP-2loaded PELA microcapsules have been successfully prepared by double emulsion solvent evaporation technique and the optimum parameters for high EE (75.0%) were achieved by response surface methodology. Then the microcapsules encapsulating rhBMP-7and covered by rhBMP-2were prepared and the porous scaffolds (BMP-2/PELA/BMP-7) were fused with these microcapsules by the dichloromethane vapor. The scaffold could control release rhBMP-2and rhBMP-7for more than42days which can partially mimic the BMPs’release profile in natural fracture healing process. Except superior biocompatibility and biodegradation, the scaffold could enhance rat’s MSCs osteogenic differentiation in vitro and repair the rat’s bone defect in vivo so well. Together, the results indicate promising potential for the use of these microcapsule-based PELA scaffolds in the treatment of bone defect in the future.更多还原