治疗骨髓瘤肾化合物的筛选暨CypA在小细胞肺癌发生中的作用分析

【作者】 杨慧蓉；

【导师】 余龙；

【作者基本信息】 复旦大学 ， 遗传学， 2007， 博士

【摘要】 本论文分为3个部分,第一部分是基于免疫球蛋白轻链结构筛选可用以治疗骨髓瘤肾的先导化合物;第二部分是D8C结构域的鉴定和结构模拟;第三部分是研究Cyclophilin A在小细胞肺癌发生中的作用分析。在第一部分中,通过计算机辅助药物设计来筛选潜在的可用以治疗骨髓瘤肾的先导化合物。居于血液恶性肿瘤第二位的多发性骨髓瘤是由于浆细胞过度增生所致,过度增殖的浆细胞大量分泌免疫球蛋白和游离的免疫球蛋白轻链,过量的轻链不能被重吸收,就会与位于肾髓袢升支粗段的uromodulin蛋白的225-233位的9个氨基酸结合并堵塞肾小管,导致骨髓瘤肾,进而导致多发性骨髓瘤病人的死亡。我们采用同源模建和分子动力学模拟优化轻链结构,指标检测结构比较可信后,利用AutoDock程序,基于结构匹配和能量匹配的分子对接计算,从化合物的三维结构数据库Maybridge中筛选能够与轻链进行较好结合的化合物,从理论上讲它们能作用于Iglc从而抑制其与uromodulin结合。通过实验验证,得到可以抑制两者结合的小化合物三个,为开发可用于治疗骨髓瘤肾的药物提供了有价值的先导化合物。在第二部分中,分析uromodulin蛋白序列,我们通过生物信息学的方法鉴定了一个新的位于199-289位氨基酸的结构域D8C,并预测了它的结构。D8C含有非常保守的芳香氨基酸和8个半胱氨酸,存在于uromodulin、LZP、GP-2蛋白和一些斑马鱼和海葵中的细胞表面糖蛋白等膜蛋白的胞外部分,可能介导蛋白之间的相互作用。我们通过同源模建和分子动力学优化预测D8C的结构,与实验得到的CD光谱结果比较,有一定的可信度。此结构可以用来筛选治疗骨髓瘤肾的小分子化合物。另外,编码uromodulin蛋白的基因UMOD突变导致家族性早幼年高尿酸血症和2型髓质囊性肾病。从2002年将疾病致病基因定位到UMOD到今为止,已有报道50多个来源于不同种族的家族中发现这种疾病并测出了UMOD基因的38个突变位点,其中DSC中包含11个突变位点,相对来说此结构域突变率比较高。我们以野生型的D8C结构为模板,模拟了突变体的结构。分析动力学模拟轨迹,发现野生型与突变体的单个氨基酸波动相比较,在201-205aa(GWYFR)处野生型有个锐利的峰,波动比较大,在此期间突变体基本都是下降的;野生型的总疏水面积和回旋半径都大于突变体,这些结构的变化导致蛋白不能通过内质网的“质控系统”,可能与伴侣蛋白不能解离进行继续转运而滞留在内质网。这种推断与文献已经报导的实验结果相一致,uromodulin突变蛋白与野生型蛋白转入细胞,观察荧光定位发现突变体多集中在内质网中,高尔基体很少;观察病人的肾组织,发病后到细胞质膜上的uromodulin蛋白减少,尿液中蛋白也减少。另一方面,我们推测D8C起着介导蛋白之间相互作用的关系,而其中非常保守的芳香氨基酸WYFR区可能对于蛋白结合是非常重要的。在第三部分中,我们首先运用实时荧光定量RT-PCR分析了CypA和它位于膜上的受体CD147在正常肺细胞和肺癌细胞株中的表达情况。结果表明CypA和CD147在小细胞肺癌中表达量大大提高,两者在各种细胞中的表达趋势相同。外源的CypA蛋白能促进小细胞肺癌细胞的生长,这一作用可能与其激活了有丝分裂原激活的蛋白激酶(MAPK)中的ERK1/2信号通路有关,我们推测CypA蛋白可能通过CD147激活了ERK1/2,促进细胞生长。这些研究为深入了解CypA在小细胞肺癌中的作用提供了有重要价值的线索。更多还原

【Abstract】 The dissertation consists of three parts: the first part describes virtual screening of lead compounds for myeloma kidney based on Iglc structure; the second part describes the identification and structure simulation of D8C, a novel protein domain; the third part describes the expression of Cyclophilin A in lung cancer cells and normal lung cell. The exogeneous CypA protein can stimulate ERK1/2 signal in small cell lung cancer cell, which results in cell’s proliferation.In the first part, we performed virtual screening based on the Iglc structure. It is known that binding between Iglc and uromodulin leads to myeloma kidney, the main cause of death in patients with multiple myeloma. We used homology modeling and molecular dynamics to optimize the structure of Iglc, then structure-based virtual screening was used to search for small compounds that could inhibit its binding with uromodulin. According to the binding energy and the characterization of compounds, we have chosen some compounds and tested their effects by pull down method. Three compounds could inhibit the binding. Then we analyzed the binding motif between Iglc and the useful compounds. This may offer guidelines for the rational design of novel drugs that inhibit binding between Iglc and uromodulin for treating myeloma kidney.In the second part, through analyzing uromodulin sequence, we identified a novel domain, D8C (199-289aa), which is present in uromodulin, LZP, GP-2 and cell surface glycoproteins in zebra fish and actinia, etc. These proteins are membrane proteins with D8C located in the extracellular part. We did homologue modeling and molecular dynamics to simulate the structure of D8C and its mutants. The second structures from DSSP accord with CD results. So this structure could be used to screen small compounds for myeloma kidney. Mutation of UMOD gene, which code uromodulin, is responsible for familial juvenile hyperuricaemic nephropathy (FJHN) and Medullary Cystic Kidney Disease type 2 (MCKD2). So far 38 mutants were identified and 11 of them located in D8C domain. D8C consists of fourα-helices and twoβ-sheets. Through analyzing the simulation trajectories, the most obvious change is that wild type D8C shows a sharp peak located in GWFRY (201-205aa) compared with mutants. The radius of gyration and solvent accessible area are bigger in wild type than in mutants. We predict that the mutant protein can not pass ER "quality control", maybe can not dissociate from chaperon protein and delay in ER for these structure changes. This is in agreement with fluorescence experiment that mutant protein is delayed in ER and little trafficks to Golgi compartment and plasma membrane, which have been reported. Also we infer that D8C is involved in protein-protein interaction, among which the conserved aromatic amino acids GWFRY may be important for the process.In the third part, we analyzed the expression of CypA and its membrane receptor CD147 in several kinds of lung cancer cells as well as normal lung cell and found that in H446 cell, a small cell lung cancer cell, the expressions of both CypA and CD147 are the highest. The exogeneous CypA protein can substantially stimulate H446 cell growth in dependence on its PPIase activity. We also showed that CypA protein can stimulate ERK1/2 signal in a dose and time dependent manner; but have no stimulation effect on p38 and JNK signals. We predict that exogeneous CypA may function through its receptor CD147. Elucidation of the precise role of CypA in these pathways may lead to new targeted therapies for small cell lung cancer.更多还原