【作者】 李强；

【导师】 韩英； 时永全；

【作者基本信息】 第四军医大学 ， 内科学， 2013， 博士

【摘要】 肝硬化对人类健康仍然是一个巨大的威胁。对于失代偿期肝硬化，原位肝移植是唯一能够延长患者生存期的治疗方法，但供肝缺乏、费用高昂等因素严重限制了其临床应用。近年来，干细胞研究领域取得了一系列重要进展。大量研究表明，成体干细胞具有可塑性、能够被诱导分化为多种组织细胞。临床探索性研究发现，自体干细胞移植能够有效地促进肝脏疾病的恢复。目前用于肝脏疾病治疗的干细胞主要包括骨髓来源的间充质干细胞（mesenchymal stem cells，MSCs）、造血干细胞（hematopoietic stem cells，HSCs）及未分类的混合细胞。基础与临床研究表明，MSCs和HSCs都能改善肝损伤、促进肝功能恢复，但目前尚不清楚哪一种干细胞效果更好。此外，干细胞治疗骨损伤、糖尿病和心脏病的研究提示，MSCs和HSCs可能具有一定的协同作用。但在干细胞促进肝损伤恢复的过程中，MSCs和HSCs是否发挥协同效应仍未见报道。本研究拟采用绿色荧光蛋白（Green fluorescent protein，GFP）转基因雄性小鼠分离纯化MSCs和HSCs，将这些细胞单一和两者组合之后移植给同品系肝损伤雌性小鼠（CCl₄肝损伤模型，不表达GFP），进而对比观察两种干细胞在小鼠体内的示踪及其促进肝损伤修复的作用与机制。实验目的1、观察骨髓来源的MSCs和HSCs移植后在肝损伤小鼠体内的分布，比较两种干细胞向损伤肝脏归巢的能力；2、对比分析骨髓来源的MSCs和HSCs促进小鼠肝损伤恢复的效果，观察MSCs和HSCs有无协同作用；3、初步明确MSCs和HSCs促进小鼠肝损伤恢复的机制。材料和方法1、取GFP转基因小鼠（雄性小鼠）的骨髓，利用全骨髓细胞贴壁培养法分离MSCs，利用流式分选技术分离纯化HSCs；2、借助细胞形态学、流式细胞术和成骨成脂诱导等方法对分离的干细胞进行表型和功能鉴定；3、用CCl₄制备肝损伤小鼠模型（雌性小鼠）。将正常小鼠或肝损伤小鼠随机分为五组，一组接受MSCs回输，一组接受HSCs回输，一组接受MSCs和HSCs混合细胞回输（按1:1比例），一组作为空白对照，一组作为模型对照组，每只小鼠接受的细胞数量为1x106；4、通过小动物荧光成像技术、荧光细胞示踪和PCR检测Y染色体序列等技术，比较移植的干细胞在小鼠各个主要脏器的分布，观察干细胞向损伤肝脏的归巢能力；5、通过生存分析、肝组织HE染色及血清学检查等方法比较MSCs和HSCs促进肝损伤修复的效果；6、通过苦味酸天狼星红染色、免疫组织化学染色和Real-time PCR等方法，检测干细胞移植对小鼠肝纤维化及纤维化相关分子表达的影响；7、借助激光共聚焦显微镜观察干细胞在损伤肝脏中的定植；8、通过免疫组化和免疫荧光染色等方法，观察干细胞移植对肝细胞再生能力的影响；9、通过ELISA方法检测干细胞移植对小鼠体内多种生长因子及免疫因子的影响。结果1、分离鉴定GFP小鼠骨髓来源的MSCs及HSCs用全骨髓贴壁培养法成功分离出MSCs。第一代的MSCs在显微镜下表现为形状不规则的细胞，在荧光显微镜下可见细胞荧光强度不均匀。当体外培养至第三代时，MSC细胞大小及形状较规则，荧光强度也较均匀。绝大多数细胞表达CD90（95.2±6.1%）、CD29（85.6±3.5%）、CD105（96.5±4.3%）、CD45（6.3±2.2%）、CD34（7.2±1.8%）和CD80（3.8±1.2%），符合MSCs表型。功能试验显示，MSCs可向成骨细胞及脂肪细胞分化。提示我们成功获得了MSCs。同时，我们采用流式细胞术成功分选获得了HSCs。2、MSCs和HSCs的体内示踪当把MSCs、HSCs及两者等比例混合的细胞（MSCs+HSCs）通过尾静脉注入肝损伤小鼠或正常小鼠，GFP信号首先在小鼠肺内蓄积，2h后GFP信号在肺内开始逐渐减弱，之后开始在肝脏和脾脏中蓄积。在接下来的观察时间内GFP信号强度在肝脏中一直保持增长趋势，然而脾中的信号强度从7d开始减弱。肾脏中的GFP信号强度几乎检测不到。三组中GFP强度变化趋势一致。肝脏中的GFP信号在MSCs组较HSCs组及MSCs+HSCs组显著增强，提示MSCs较HSCs向损伤肝脏的归巢能力更强。此外，无论是MSCs还是HSCs，在4w时在肝脏中的蓄积都比2w时要多。Y染色体示踪的结果与GFP示踪结果相似。与接受干细胞的正常小鼠相比，在所有的观察点上，肝损伤小鼠肝脏内的GFP信号均显著增强，提示肝损伤是诱导干细胞向肝脏归巢的重要因素。流式细胞术分析及Real-Time PCR发现，MSC（培养至第三代时）上的CXCR4（干细胞的归巢受体）表达量显著高于HSC。3、MSCs和HSCs促进肝损伤修复效果的对比制备肝损伤小鼠模型，并使用MSCs、HSCs或两者等比例混合细胞进行尾静脉回输治疗。生存分析结果显示，三组接受不同干细胞移植的小鼠生存率较肝损伤模型组显著提高，其中MSCs组的生存率显著高于HSCs和MSCs+HSCs组。肝组织切片的HE染色显示，肝损伤模型小鼠的肝组织内可见大量的炎性细胞浸润和大量的肝细胞坏死。天狼星红染色显示，肝损伤模型小鼠肝组织内可见大量的胶原沉积。与未接受干细胞治疗的对照小鼠相比，干细胞治疗显著减轻了小鼠的肝纤维化和炎症，其中以MSCs的效果最强。研究还表明，MSCs和HSCs在促进肝损伤修复中未显示协同效应。4、MSCs和HSCs促进肝损伤修复的机制分离接受干细胞治疗的肝损伤小鼠的肝脏。对肝组织进行免疫荧光染色结果显示，HSCs组GFP+/AFP+和GFP+/ALB+双阳性细胞数量（分别为4.3±0.6%和3.5±0.7%）较MSCs组（1.4±0.5%和2.1±0.3%）及MSCs+HSCs组（2.8±0.4%和2.4±0.6%）显著提高。干细胞治疗组α-SMA+细胞数量较对照组均显著降低，其中MSCs组α-SMA+细胞数量最低。未发现GFP+/α-SMA+双阳性细胞，提示移植的干细胞未转分化为肝星状细胞。干细胞治疗组肝组织中Ki-67和PCNA的表达量显著升高，以MSCs组最高。接受干细胞治疗4w后，用ELISA法检测各组小鼠血清中的生长因子及免疫调节相关因子。结果显示，MSCs组神经生长因子（NGF）水平较其它干细胞组显著提高，而肝细胞生长因子（HGF）及血管内皮生长因子（VEGF）水平在各干细胞移植组之间未见显著差异。MSCs组IL-10水平较其它干细胞移植组显著升高而IL-6、TNF-α水平较其它干细胞移植组显著降低。结论1、骨髓来源的MSCs和HSCs均能向损伤肝脏归巢，并且MSCs的肝脏归巢能力更强。2、对于CCl₄诱导的小鼠肝损伤，MSCs和HSCs都能有效促进损伤修复，其中MSCs的作用更强。3、干细胞可能通过调节肝细胞再生相关因子和免疫因子的表达促进肝损伤的恢复。4、在促进CCl₄诱导小鼠肝损伤修复过程中，MSCs和HSCs未表现出协同效用。更多还原

【Abstract】 Cirrhosis and its related morbidity place a significant burden on health careworldwide. Liver transplantation remains the definitive treatment option for end-stageliver disease. But the mismatch between the number of patients requiring transplantationand the amount of available organs is set to grow, highlighting the need to develop newstrategies to reduce liver scarring and stimulate liver regeneration.More recently, reportsof unexpected plasticity in adult bone marrow have raised hopes that stem cell therapymay offer exciting therapeutic possibilities for patients with chronic liver disease.To date,there are several published human clinical studies investigating the effects of stem celltherapy in patients with liver disease and most of the studies yielded positive results.The cells mostly used to transplant were derived from bone marrow including MSCs,HSCs and unsorted mononuclear cells.Among them, MSCs and HSCs can be obtained in a great quantity. Both stem cells have anti-fibrotic and proregenetative effects in the injuredliver. However, which cell type is more effective in treating the injured liver remains to bedetermined. Moreover, several recent studies have suggested that MSCs and HSCsfunction synergistically for the therapy of diabetes and heart failure and for vascularizingbioengineered tissues. Whether these cells can work synergistically in the injured liver isunclear.In this experiment, the MSCs and HSCs were isolated from the male green fluorescenceprotein transgenic mice. A total of1×106isolated stem cells were resuspended in PBS andslowly infused into the female liver injured mice via the tail vein to evaluate thebiodistribution after the peripheral and the anti-fibrotic activities of these two stem cells.Objectives1. Our aim was to evaluate the biodistribution of the stem cells after the peripheralinfusion of MSCs or HSCs into liver injured mice and compared their homing capacity tothe injured liver.2. To evaluated the anti-inflammatory and anti-fibrotic activities of these two stem cells inthe injured liver and to study whether MSCs and HSCs exhibit synergistic effects intreating liver injury.3. To investigate the underlying molecular mechanisms through which MSCs and HSCsparticipate in injuried liver repair. We hope that these findings contribute to betterunderstanding of the interactions between stem cells and the environment that leads tohoming and integration into livers.Materials and methods1. MSCs were isolated using whole bone marrow culture method and purified usingattachment method. HSCs from bone marrow were sorted by magnetic nanoparticles andflow cytometry using multiple antibody panels. 2. The adhered cells in the first passage and the third passage were observed byfluorescence microscope．Stem cell surface markers were identified by flowcytometer．The cells were also induced to difierentiate into bone and fat cells in vitro.3.After final CCl4injection at3months, the mice except the control group were randomlydivided into different groups to ensure that the relatively level of liver disfunction wasconstant. Mice from the same cohort were randomly allocated to receive different cellstem cells （1x106cells/mice）via injections of the tail vein.4. Distribution of transplanted cells in injuried liver was detected by bio-imaging system,fluorescence assay and Y chromosome sequence.5. Therapeutic potential of transplanted cell for liver cirrhosis was determined by serumassay, survival curve for the liver injured mice and representative photomicrographs ofH&E-stained mouse livers from the different groups.6. Liver fibrosis was quantified with sirius red staining, real-time PCR andimmunofluorescence analysis.7. The characterization of transplanted stem cells in injuried liver was detected byconfocal microscopy assay.8. Hepatocyte regeneration was determined by immunohistochemistry andimmunofluorescence analyses.9. Quantification of the mouse serum levels of growth factors and cytokine weredetermined using enzyme-linked immunosorbent assays kits per the manufacturer’sinstructions.Results1. Isolation and characterization of MSCs and HSCs from GFP transgenic miceThe MSCs were isolated from the bone marrow of GFP-transgenic mice. In the firstpassage, the cells derived from the donors emitted heterogeneous levels of greenfluorescence when observed under the fluorescence microscope and were of various sizes,as observed in bright field. In the third passage, the GFP signal intensity was uniformamong the cells, and the cells exhibited a homogeneous morphology. Flow cytometryanalyses were used to characterize the surface markers of the cultured cells. Most of the cells expressed the standard MSC surface markers, CD90(95.2±6.1%), CD29(85.6±3.5%) and CD105(96.5±4.3%), whereas they were negative for CD45(6.3±2.2%),CD34(7.2±1.8%) and CD80(3.8±1.2%). MSCs could differentiate into osteoblasts andfat cells respectively. HSCs were also successfully isolated by flow cytometry for use inthe following experiments.2. In vivo tracking of MSCs and HSCs.After intravenous infusion, the GFP signals first accumulated in the lung, but by2h,those signals began to decrease, whereas they started to accumulate in the liver and spleenat2h after infusion. During the following hours to days, the GFP signal intensitygradually increased in the liver. From24h to7d, the GFP signal intensity graduallyincreased in the spleen and then decreased. The GFP signal was barely detectable in thekidney. These trends were similar in the MSCs, HSCs and MSCs+HSCs groups. The GFPintensity of the livers in the MSCs group was significantly higher than in the HSCs orMSCs+HSCs group. The number of homing stem cells to the CCl4-induced cirrhotic liverwas significantly higher than that in the normal group. The donor-derived signals werestronger in the livers4w after the cell injection than at2w. These results were confirmedby the percentage of GFP-positive cells detected after nuclear staining with DAPI. From2w on, most of the transplanted cells were located infiltrating the areas around the liver’sportal tracts and interlobular connective tissue. Only a few of the cells migrated toward thecentral region of the hepatic lobes and can be detected in the sinusoids. FACS analyses ofCXCR4expression on MSCs and HSCs shows that CXCR4expressed on MSCs (in thethird passage)(33.2±8.1%) was significantly higher than HSCs(24.5±6.8%, P<0.05).Real-time PCR revealed that CXCR4mRNA expression was higher in MSCs (in the thirdpassage) than in HSCs.3. Comparing the anti-inflammatory and anti-fibrotic activities of transplantatedstem cells in the injured liver.The mice in the normal group survived the observation period. The survival of themice in the three groups that underwent stem cell transplantation was significantly higherthan in the group treated with CCl4. The survival percentage in the MSCs group (68.2%) was significantly higher than in the HSCs (36.4%) or MSCs+HSCs (45.5%) group, whilethe survival percentage in the MSCs+HSCs group was significantly higher than in theHSCs group. The histological sections of the CCl4-injured mice demonstrated that whencompared with normal mice, a large number of inflammatory cells had infiltrated thesinusoids and centrilobular regions, and the coagulation necrosis of hepatocytes wasobserved (H&E staining). In addition, liver fibrosis had increased significantly,characterized by fibrotic septum formation starting in the portal areas (Sirius red staining).After transplantation of MSCs, the injured livers showed maximal restoration with thinnerfibrotic areas and decreased collagen depositions (3.6±0.7%). However, the fibrotic areashad decreased to a lesser extent in the mice transplanted with HSCs (7.6±0.8%) orMSCs+HSCs (6.2±1.7%, P<0.05). These results were confirmed by the expression oftype I collagen (4.0±1.1%,6.3±3.1%,5.7±1.5%, respectively, P<0.05). Moreover, whencompared with the mice in the HSCs group and the MSCs+HSCs group, the micereceiving MSCs showed the best improvement of liver function, however, liver function inMSCs group was still inferior than in the normal cohort, as demonstrated by the ALB,ALT and AST levels of the peripheral blood.4. The mechanism of transplanted MSCs and HSCs promoting liver function andreversing liver fibrosis.The immunofluorescence staining in the HSCs group revealed a higher percentage ofdouble-labeled GFP+/AFP+and GFP+/ALB+cells in the host livers (4.3±0.6%and3.5±0.7%, respectively), compared with the MSCs (1.4±0.5%and2.1±0.3%, respectively)or the MSCs+HSCs (2.8±0.4%and2.4±0.6%, respectively,P<0.05) groups. Theexpression of α-SMA was significantly decreased in the livers of mice transplanted withany cell type. The α-SMA expression in the MSCs group was significantly lower than inthe other two groups. Double-labeled GFP+/α-SMA+cells were not found in any of thegroups. To evaluate whether stem cell transplantation enhances the proliferation ofhepatocytes in cirrhotic livers, the PCNA and Ki-67expression levels were assessed byimmunofluorescence and immunohistochemistry. In the MSCs-transplanted livers, thepercentage of PCNA+(11.5±3.4%) and Ki-67+(8.2±2.7%) cells was increased significantly when compared with those of the HSCs (6.9±1.8%,6.0±1.1%, respectively) andMSCs+HSCs (8.5±3.1%,7.2±1.9%, respectively, P<0.05) groups.Four weeks after cell transplantation, the serum showed a significant increase in NGFin the MSCs (101±12pg/ml) group when compared with the HSCs and MSCs+HSCsgroups (53±5pg/ml and69±7pg/ml, respectively, P<0.05). The levels of HGF and VEGFwere not significantly different among the cell transplantation groups. The expression ofIL-10in the MSCs group (98±22pg/ml) was higher than in the other cell transplantationgroups (HSCs and MSCs+HSCs groups:42±14pg/ml,66±21pg/ml, respectively) or theCCl4group (21±4pg/ml, P<0.05), whereas the concentration of IL-6(45±15pg/ml) andTNF-α (140±75pg/ml) in the MSCs group was lower than that of HSCs (IL-6:77±13pg/ml, TNF-α:227±87pg/ml) and MSCs+HSCs (IL-6:60±9pg/ml, TNF-α:183±15pg/ml,P<0.05) groups.Conclusions1. When compared with HSCs alone and a combination of HSCs and MSCs, MSCs hadthe greatest homing capability to the injured liver.2. MSCs showed the greatest capability to restore the injured liver in vivo and facilitatemice survival when compared with other two groups.3. MSCs exhibited more remarkable paracrine effects and immunomodulatory propertieson hepatic stellate cells and native hepatocytes in the treatment of the liver pathology.4. Synergistic actions of MSCs and HSCs were most likely not observed because the stemcells in liver were detected mostly as single cells, and single MSCs are insufficient toprovide a beneficial niche for HSCs.更多还原