胶质瘤血管巢：肿瘤源性内皮细胞的作用及机制

Increasing evidence demonstrates that, similar to neural stem cells (NSCs), glioma stem cells (GSCs) are dependent on a perivascular niche. Although the cellular components of this perivascular niche are not fully understood, more and more studies point to the functional contribution of the vascular endothelial cells. It is hypothesized that cross talk through P13K/eNOS/Notch signal pathway between those stem cells and regional vascular endothelial cells plays a main role in maintaining the stem cells in an undifferentiated state. Traditionally, it is believed that the vascular endothelial cells surrounding the tumor stem cells are initiated from nonmalignant blood vessels or recruited from blood cells. However,our recent study in combination with the reports by others, revealed that glioma stem cells unexpectedly transdifferentiated into vascular endothelial like cells and form functional microvasculature. This finding raises an important question about whehter this transdifferentiated endothelial cells help to build tumor niche in soide tummor tissues and whether this abnormal niche tends to maintain the tumor stem cells in an undifferentiated state preferentially. By employing fluorescent protein tracing animal models, transwell co-culture system, RT-PCR, RNAi,and other methods, we are to certify:1) tumor-derived endothelial cells lie adjacent to glioma stem cells through reconstruction in three dimension of capillaries in solid tumor tissues; 2) compared with normal endothelial cells, tumor-derived endothelial cells, have stronger potential in maintaining stem cell proliferation and self-renewal; 3)intervening the signaling cross talk between glioma stem cells and malignant endothelial cells by manipulating P13K/eNOS/Notch signal pathway is a potential strategy to break glioma stem cell self-renewal and promote their differentiation.

和神经干细胞一样，胶质瘤干细胞（glioma stem cells, GSCs）也存在血管巢，但其内皮细胞的来源并不清楚。课题组前期研究发现，GSCs除常规分化为胶质瘤细胞外，还经旁路机制转分化为具有肿瘤特征的血管内皮细胞。我们推测，这些转分化的内皮细胞很可能参与并主导了胶质瘤血管巢的构建。本申请拟利用双色荧光示踪的GSCs裸小鼠移植模型、Transwell共培养系统，通过激光共聚焦、ELISA、RT-PCR和RNA干扰等证明：1）肿瘤源性血管内皮细胞在空间结构上和GSCs关系紧密；2）在功能上比正常内皮细胞更能促进GSCs增殖和自我更新；3）其作用受到GSCs与内皮细胞间信号传递（P13K/ eNOS /Notch）的调控。本研究具有重要意义，一旦假设获证实，将是首次证明GSCs通过转分化自行构建血管巢，并为选择性破坏胶质瘤血管巢提供细胞和分子靶点。

本研究的目的是探索胶质瘤干细胞能否转分化为内皮细胞并参与胶质瘤血管巢的构建。本课题在获取胶质瘤干细胞,成功建立起裸小鼠移植模型后，拟按照原定计划进一步实施课题时，Cheng L等(Cell. 2013) 等通过荧光示踪技术等证明，胶质瘤干细胞主要通过转分化为周细胞而不是内皮细胞参与血管巢的构建。鉴于此，我们改变了研究方向，探索通过基因转入，促使胶质瘤干细胞向良性方向分化，结果发现：1）和神经干细胞相比，胶质瘤干细胞中神经元分化基因NeuroD1, Sox1处于低表达或不表达状态；2）通过慢病毒将NeuroD1, Sox1转入胶质瘤干细胞后，胶质瘤干细胞在一周后出现大量死亡，而存活的细胞在形态上呈现神经元的改变；3)免疫组化染色显示细胞β－tubulin，MAP2阳性，而CD133, Nestin, Sox2等干细胞标志物的表达显著降低，甚至消失。另外，我们在和同济大学医学院章小清教授的合作研究中偶然发现温度、胞外pH和胞外渗透压的变化都会改变胶质瘤干细胞胞内pH的变化，而胞内pH的变化最终影响Smad5的核浆分布，也就是说Smad5是胶质瘤干细胞微环境的重要感受蛋白，这是一个全新的发现。.胶质瘤干细胞在转入神经元分化基因后出现的成熟分化具有重要意义，至少说明一点，胶质瘤干细胞可以通过改造由恶转良，从而为胶质瘤的治疗开辟新的思路。另外，针对Smad5的发现，我们设想脑肿瘤干细胞存在于微血管和周围其它细胞共同维持的特定微环境中，因而其胞内pH是相对特异和稳定的，如果小分子药物能特异改变脑肿瘤干细胞的胞内pH，就有可能对脑肿瘤干细胞分化产生影响，进而对胶质瘤具有治疗意义。