目的 构建基于人脑微血管内皮细胞（HCMEC）、星形胶质细胞（SVGp12）和周细胞（HBVP）的三细胞体 外血脑屏障模型，以更准确地模拟中枢神经系统的复杂微环境，并观察氧糖剥夺再灌注（OGD/R）损伤对其的影 响。方法 通过 Transwell 共培养技术，将 HCMEC、SVGp12 和 HBVP 共同引入体外血脑屏障模型，约培养 7 d 后，待跨内皮电阻（TEER）值稳定在＞110 Ω·cm2即可开始实验。氧糖剥夺培养 4 h、复氧复糖 2 h，将模型组 置于充满 5%CO2及 95%N2的缺氧盒中密闭培养，构建体外血脑屏障 OGD/R 诱导损伤模型。观察体外血脑屏障 模型中各类细胞的形态变化；采用电阻计测量 TEER 值；CCK-8 法测定细胞活力；检测 D-乳酸脱氢酶（LDH） 渗出率；通过 4 h 渗透实验、荧光素钠（SF）渗透实验评价血脑屏障的渗透性；重复 3 次实验考察 TEER 值、 LDH 渗出率、4 h 渗透实验、SF 渗透实验结果，以评估模型的重复性及稳定性；Western Blot 法检测细胞 ZO-1、occludin、claudin-5、MMP9 蛋白表达水平。结果 （1）三细胞体外血脑屏障模型在培养 7 d 后，TEER 值稳定在 110 Ω·cm2 ，模型成功建立。与对照组比较，模型组在 OGD/R 诱导损伤后 TEER 值显著下降（P＜ 0.01），体外血脑屏障模型的屏障作用被破坏；CCK-8 实验细胞吸光度（A）值显著降低（P＜0.01），表明细胞活 力显著下降；LDH 渗出率显著增高（P＜0.01）；A 值（4 h 渗透实验）显著降低（P＜0.01），平均荧光强度（SF 渗 透实验）显著提高（P＜0.01），表明体外血脑屏障模型被破坏，渗透性显著增加；细胞的紧密连接蛋白 ZO-1、 occludin、claudin-5 表达显著下调（P＜0.01），MMP9 蛋白表达显著上调（P＜0.01）。（2）与 HCEMC-SVGp12 对照 组比较，HCEMC-HBVP-SVGp12 对照组的 A 值（4 h 渗透实验）明显升高（P＜0.05），平均荧光强度（SF 渗透实 验）显著降低（P＜0.01）；与 HCEMC-SVGp12 模型组比较，HCEMC-HBVP-SVGp12 模型组的 TEER 值、A 值 （4 h 渗透实验）显著升高（P＜ 0.05，P＜0.01），平均荧光强度（SF 渗透实验）显著降低（P＜0.01）。表明周细胞 HBVP 能明显增强体外血脑屏障模型的屏障功能。（3）重复性实验结果表明体外血脑屏障模型及 OGD/R 诱导损 伤模型的稳定性较好，具有可重复性。结论 三细胞体外血脑屏障模型在正常条件下表现出良好的功能稳定 性，并能够有效模拟体内血脑屏障的生理状态和病理变化。

Objective To construct a three-cell in vitro blood-brain barrier （BBB） model based on human cerebral microvascular endothelial cells （HCMEC），astrocytes （SVGp12），and pericytes （HBVP），aiming to more accurately simulate the complex microenvironment of the central nervous system， and to observe the effects of oxygen-glucose deprivation/reperfusion （OGD/R） injury on this model. Methods Using the Transwell co-culture technique，HCMEC， SVGp12，and HBVP were incorporated into the in vitro BBB model. Experiments commenced after approximately 7 days of culture when the transendothelial electrical resistance （TEER） value stabilized above 110 Ω·cm2 . To induce OGD/R injury，the model group was subjected to 4 hours of oxygen-glucose deprivation followed by 2 hours of reperfusion， achieved by culturing in a sealed hypoxia chamber filled with 5%CO2 and 95%N2. Morphological changes of the various cell types within the model were observed. TEER values were measured using a volt-ohm meter； cell viability was assessed via the CCK-8 assay；D-lactate dehydrogenase （LDH） leakage rate was determined；BBB permeability was evaluated using the 4-hour permeability assay and the sodium fluorescein （SF） permeability assay. Experiments measuring TEER，LDH leakage rate，and the results of the 4-hour and SF permeability assays were repeated three times to assess the model's reproducibility and stability. Protein expression levels of ZO-1，occludin，claudin-5，and MMP9 were detected by Western Blot analysis. Results （1） The three-cell in vitro BBB model was successfully established，with TEER values stabilizing at 110 Ω·cm2 after 7 days of culture. Compared to the control group，the model group subjected to OGD/R injury showed a significant decrease in TEER value （P＜0.01），indicating disruption of the barrier function； the CCK-8 assay revealed a significant reduction in absorbance （A） value （P＜0.01）， indicating markedly decreased cell viability；the LDH leakage rate was significantly increased （P＜0.01）；the A value from the 4-hour permeability assay was significantly decreased （P＜0.01），while the mean fluorescence intensity from the SF permeability assay was significantly increased （P＜0.01），collectively demonstrating that the in vitro BBB model was compromised， with significantly increased permeability. The expression levels of tight junction proteins ZO-1， occludin， and claudin-5 were significantly downregulated （P＜0.01）， whereas MMP9 protein expression was significantly upregulated （P＜0.01）.（2） Compared to the HCEMC-SVGp12 control group， the HCEMC-HBVPSVGp12 control group showed a significantly higher A value in the 4-hour permeability assay （P＜0.05） and a significantly lower mean fluorescence intensity in the SF permeability assay （P＜0.01）. Compared to the HCEMCSVGp12 model group， the HCEMC-HBVP-SVGp12 model group exhibited significantly higher TEER values and A values from the 4-hour permeability assay （P＜0.05， P＜0.01， respectively）， and a significantly lower mean fluorescence intensity from the SF permeability assay （P＜0.01）. These findings indicate that the pericyte HBVP significantly enhanced the barrier function of the in vitro BBB model.（3） Reproducibility experiments confirmed that both the in vitro BBB model and the OGD/R-induced injury model exhibited good stability and were reproducible. Conclusion The three-cell in vitro BBB model demonstrates favorable functional stability under normal conditions and effectively simulates both the physiological state and pathological alterations of the in vitro BBB.

R285.5；R965.1

国家自然科学基金重点项目（82030124）；中国中医科学院科技创新工程项目（CI2021A04614）。