人類骨髓、臍帶血、羊水與羊膜複能性間葉系幹細胞特性之研究

Abstract

中文摘要 複能性間葉系幹細胞在人體中扮演著重要的角色，它們可分化為多種間葉系細胞，如：骨骼細胞、脂肪細胞、軟骨細胞、肌肉細胞等，並主要負責體內組織的建立與修復。本論文以不同來源的人類複能性間葉系幹細胞，分為兩大部分進行相關研究。 第一部份研究為比較骨髓間葉系幹細胞（bmMSCs）與臍帶血間葉系幹細胞（cbMSCs）對於骨骼與脂肪的分化能力。藉由基因晶片分析與體外分化研究，我們發現cbMSCs對骨骼分化潛力較bmMSCs強，而bmMSCs則在脂肪分化上的潛力較cbMSCs強。除此之外我們首次證明了leptin可直接影響間葉系幹細胞的分化能力，leptin可促進間葉系幹細胞的骨骼分化能力並抑制其脂肪分化能力，經由real-time PCR的分析也證明了，在脂肪分化的過程中，bmMSCs較cbMSCs中可表現出較高量脂肪分化專一基因PPARγ2的 mRNA，而在骨骼分化的過程中，cbMSCs則較bmMSCs可表現出較高量的Cbfa1 mRNA。 當我們進一步建立cbMSCs 中的單一細胞群落時，發現臍帶血中存有兩種不同型態的細胞群落，由佔有大多數的flattened fibroblastic clones與少部分的spindle-shaped fibroblastic clones所構成，這樣的分佈比例恰與bmMSCs的MSCs單一細胞群落相反。兩種群落之間除CD90外擁有相同的表面抗原表現，且二者在骨骼與軟骨分化能力並無太大差異。最大的不同在於cbMSCs中佔極大多數的CD90－ flattened fibroblastic clones其脂肪分化能力遠低於CD90+ spindle-shaped fibroblastic clones的群落。雖然目前對於體內間葉系幹細胞形成脂肪細胞的詳細機制仍尚未明朗，但臍帶血與成人骨髓中spindle-shaped fibroblastic clones與flattened fibroblastic clones所佔比例不同，可反應在人類發育不同時期脂肪細胞所代表的生理意義。 第二部分研究為由羊水與羊膜中分別建立間葉系幹細胞（AFMSCs 與AMMSCs），並就其特性加以分析。AFMSCs與AMMSCs皆具有一般間葉系幹細胞的特性。而AFMSCs可表現胚胎幹細胞的調控因子（Oct-4, Nanog），但無法表現胚胎幹細胞標記SSEA-1, SSEA-3, SSEA-4, TRA-1-60與TRA-1-81。AFMSCs其端粒長度亦較臍帶血與骨髓之間葉系幹細胞長，我們推測其可能屬於較早期之間葉系幹細胞。而AMMSCs 除了間葉系分化能力外，可經由insulin與isobutyl-methylxanthine 誘導而分化為神經細胞，並表現神經細胞專一標記β-tubulin III、tyrosine hydroxylase 與 NeuN。

Abstract

During human entire lifespan, from fetus to adult, mesenchymal stem cells (MSCs) proliferate and differentiate into mesenchyme-lineage tissues. Bone marrow and umbilical cord blood are reported to be the main sources of MSCs and have been proposed for possible clinical applications. This study evaluates the tendency in bone marrow-derived MSCs (bmMSCs) and cord blood-derived MSCs (cbMSCs) by in vitro induction and quantification of their characteristics. Results indicated that cbMSCs had a significantly stronger osteogenic potential but less capacity in adipogenic differentiation than bmMSCs. Leptin, an important regulator of mesenchymal differentiation, also acted in bmMSCs and cbMSCs. In both types of MSCs, leptin was found to support osteogenesis, and inhibited adipogenesis. However, Cbfa1 mRNA expression in bmMSCs and cbMSCs was affected at different degree by leptin during osteogenesis. In contrast, leptin reduced PPARγ2 mRNA expression at same level during adipogenesis in both types of MSCs. These results demonstrate the diverse capacity of MSCs from bone marrow and cord blood and suggest that they be used differently in experimental and therapeutic studies. We also isolated the clonogenic MSCs from cord blood by limiting dilution method. These cells exhibited two different morphologic phenotypes, including flattened fibroblastic clones (majority) and spindle-shaped fibroblastic clones (minority). Both types of MSCs shared similar cell surface markers except CD90 and had similar osteogenic and chondrogenic potentials. However, the spindle-shaped clones possessed the positive CD90 expression and a greater tendency in adipogenesis than the flattened clones. The high number of flattened MSCs might actually be linked to the less sensitivity of cbMSCs in adipogenic differentiation. Amniotic fluid and amniotic membrane were also the good sources of mesenchymal stem cells. We successfully isolated MSCs from second-trimester amniotic fluid (AFMSCs) and term amniotic membrane (AMMSCs). AFMSCs and AMMSCs were very similar with MSCs from other sources in the phenotypic morphology, marker profiles (positive for SH2, SH3, SH4, CD29, CD44, CD90 and HLA-I; negative for CD26, CD31, CD34, CD45 and HLA-II). All AFMSCs and AMMSCs samples have mesenchymal-lineage differentiated potentials to differentiate into osteoblasts and adipocytes. The telomerase activity was not detected in AFMSCs, cbMSCs and bmMSCs, and the telomere length of AFMSCs was longer than cbMSCs and bmMSCs. AFMSCs could express embryonic regulators, Oct-4 and Nanog, but they could not express the embryonic markers SSEA-1, SSEA-3, SSEA-4, TRA-1-60 and TRA-1-81. AMMSCs could be induced into neuronal cells, which expressed neuronal markers β-tubulin III, tyrosine hydroxylase and NeuN, by insulin and isobutyl-methylxanthine.