目的：构建可缓释骨形态发生蛋白-2（bone morphogenetic protein-2, BMP-2）的仿生磷酸钙（biomimetic calcium phosphate，BioCaP）共沉淀三维支架（BMP-2-coprecipitated biomimetic calcium phosphate，BMP-2-BioCaP），检测其理化特性，探究其对人脂肪间充质干细胞（human adiposederived stem cells, hASCs）体内外成骨分化的影响，最终构建以hASCs和BMP-2-BioCaP为基础的新型组织工程化骨。方法：构建BMP-2-BioCaP三维缓释支架，扫描电子显微镜观察表面形貌，体外检测其缓释能力。将BMP-2-BioCaP颗粒分别浸泡于增殖培养基（proliferation medium, PM）与成骨诱导培养基（osteogenic medium，OM）中，每2天提取上清液用于hASCs的体外培养。CCK-8实验检测各组hASCs的体外增殖能力，诱导7 d及14 d后进行碱性磷酸酶（alkaline phosphatase, ALP）染色及活性定量检测，14 d及21 d进行茜素红染色及矿化沉积定量分析，4 d及14 d检测成骨相关基因的表达情况。体内实验使用6只裸鼠，于其背部正中做皮肤切口，向两侧共分离出4个皮下植入腔，分别植入：（1）单纯BioCaP支架，（2）BioCaP支架+hASCs，（3）BMP-2-BioCaP缓释支架，（4）BMP-2-BioCaP缓释支架+hASCs（实验组）。植入4周后取材，标本制成组织学切片进行HE染色观察。结果：BioCaP表面由不规则晶体组成三维立体多孔结构，孔直径约为5~10 μm，加入BMP-2后，不影响BioCaP原有的立体结构。缓释曲线结果显示，蛋白质在前2天释放速度较快，随后释放速度放缓并于5 d后趋于平稳，之后每天释放量较稳定，至第21天仍有少量释放，累积释放量达20%。CCK-8结果显示，BMP-2-BioCaP缓释支架不会影响hASCs的早期增殖。诱导7 d与14 d后，OM+BMP-2-BioCaP组ALP染色及活性定量检测均显著高于其他组（P<0.01）。诱导21 d后，OM+BMP-2-BioCaP组矿化结节染色及钙沉积定量检测均高于其他组（P<0.01）。诱导4 d时OM+BMP-2-BioCaP组的Runt相关转录因子2基因（Runt-related transcription factor 2，RUNX2）与ALP基因表达水平较对照组显著升高（P<0.01），诱导14 d时RUNX2、ALP、骨桥蛋白（osteopontin，OPN）和骨钙素（osteocalcin, OC）基因表达量均显著高于其他组（P<0.01）。HE染色分析可见，实验组和BMP-2-BioCaP缓释支架组中，细胞外基质呈强嗜酸性，出现类似骨陷窝的结构，并可见包含其中的类骨细胞。与BMP-2-BioCaP缓释支架组相比，实验组的细胞外基质嗜酸性更强，类骨组织的面积更大，结构更加典型，其他组未见矿化基质和类骨组织形成。结论：BMP-2-BioCaP支架能够实现BMP-2的良好缓释，并能显著促进hASCs的体内外成骨分化，以hASCs和BMP-2-BioCaP为基础构建的新型组织工程化骨具有潜在的应用前景。

Objective： To construct a novel biomimetic calcium phosphate (BioCaP) scaffold loaded with bone morphogenetic protein-2 (BMP-2), and to investigate its role in the osteogenesis of human adipose-derived stem cells (hASCs) in vitro and in vivo. Methods: The BioCaP scaffold coprecipitated with BMP-2 (BMP-2-BioCaP) was constructed in this study. Field emission scanning electron microscopy (SEM) was used to analyze the morphology of the surfaces. The release kinetics was measured to evaluate the slow-release characteristics in vitro. BMP-2-BioCaP was immersed in proliferation medium (PM) or osteogenic medium (OM), respectively. The supernatants were collected and used to culture hASCs in vitro. Cell numbers were determined using the cell-counting kit-8 (CCK-8) to assess the cell proliferation. After 7 and 14 days, alkaline phosphatase (ALP) staining and quantification were performed to test the activity of ALP. After 14 and 21 days, the calcification deposition was determined by alizarin red S (ARS) staining and quantification. The expressions of the osteoblast-related genes were tested on day 4 and day 14. In the in vivo study, 6 nude mice were used and implanted subcutaneously into the back of the nude mice for 4 groups: (1) BioCaP scaffold only, (2) BioCaP scaffold+hASCs, (3) BMP-2-BioCaP scaffold, (4) BMP-2-BioCaP scaffold+hASCs (test group). After 4 weeks of implantation, hematoxylin-eosin (HE) staining was performed to evaluate the in vivo osteogenesis of hASCs. Results: SEM observations showed that BioCaP and BMP-2-BioCaP scaffold were entirely composed of straight, plate-like and sharp-edged crystal units, and the length of the crystal units varied between 5 and 10 μm. Release kinetics analysis demonstrated that BMP-2 incorporated with BioCaP could be released at certain concentration and last for more than 21 days, and the accumulative protein release could reach 20%. CCK-8 assays showed that cell proliferation was not significantly affected by BMP-2-BioCaP. ALP activity was higher by the induction of OM+BMP-2-BioCaP than of the other groups (P<0.01). More mineralization deposition and more expressions of osteoblast-related genes such as Runt-related transcription factor 2 (RUNX2), ALP, osteopontin (OPN) and osteocalcin (OC) were determined in the OM+BMP-2-BioCaP group at different time points (P<0.01). HE staining showed that, in the test group and BMP-2-BioCaP scaffold group, the extracellular matrix (ECM) with eosinophilic staining were observed around hASCs, and newly-formed bone-like tissues could be found in ECM around the scaffold materials. Moreover, compared with the BMP-2-BioCaP scaffold group, more bone-like tissues could be observed in ECM with typical structure of bone tissue in the test groups. No obvious positive results were found in the other groups. Conclusion: BMP-2-BioCaP scaffold could achieve slow-release of BMP-2 and promote the osteogenic differentiation of hASCs in vitro and in vivo. The novel tissue-engineered bone composed of hASCs and BMP-2-BioCaPis promising for the repair of bone defect.