HIV-1整合酶抑制剂筛选及其活性检测方法研究

发布时间：2018-10-18 09:59

【摘要】：自从1981年首次发现艾滋病患者以来，艾滋病在全球范围迅速传播，已经严重威胁人类的健康，，对社会造成巨大的破坏。艾滋病是由人类免疫缺陷病毒（HIV-1）引起，病毒复制周期中特异性的活性位点和关键酶为抗病毒化学治疗提供了潜在的靶标。其中，HIV-1编码的整合酶（IN）将HIV-1反转录后的产物cDNA整合入宿主基因组，它是病毒复制过程中不可缺少的酶。由于人类宿主的正常细胞中并不存在IN的功能类似物，使得特异性作用于IN的抑制剂对于正常细胞毒性较小；同时，IN采用单一的活性位点去适应两种不同的DNA底物，这在一定程度上限制了HIV-1产生对IN抑制剂的耐药株，因此是研发抗HIV-1药物的一个非常有意义的靶点。在体内，整合作用是一个多步骤的反应过程，主要包括：病毒cDNA与IN结合；IN发挥内切酶作用对病毒cDNA进行3'-加工；IN通过链转移反应将病毒cDNA整合到宿主细胞DNA。在体外，使用重组IN蛋白、寡聚核苷酸DNA底物以及二价金属离子可以实现IN在体内所发挥的整合功能，据此建立高效的IN抑制剂的筛选方法，是当前IN抑制剂体外筛选的主要手段。本研究表达纯化了重组IN蛋白，建立了高通量的HIV-1整合酶抑制剂筛选方法。与原有方法相比，本研究新建立的筛选方法灵敏可靠、步骤简便、成本低廉。药物筛选不仅包括生物实验方法，计算机辅助虚拟筛选方法在药物筛选中也占据了越来越重要的地位。本论文不仅在生物学实验基础上进行了整合酶抑制剂筛选方法的探索，还从计算机模拟和机器学习的角度对整合酶抑制剂虚拟筛选进行了研究。利用CoMFA分析以及支持向量机（SVM）建模等方法考察HIV-1整合酶抑制剂活性，建立了基于定量构效关系（QSAR）和SVM的预测模型，通过预测结果来判断模型的外部预测能力，结果显示SVM表现出了较强的建模和预测能力。论文工作主要包括以下三个方面：（1）表达纯化整合酶蛋白并进行链转移反应活性鉴定首先将HIV-1整合酶原核表达载体pET-28a（+）/IN（F185K/C280S）转化入大肠杆菌感受态细胞BL21（DE3）进行原核表达，该表达载体含有IN基因并引入了F185K/C280S突变以提高蛋白溶解性。用镍琼脂糖凝胶进行亲和层析纯化，获得了纯度和活性均较高的整合酶重组蛋白。运用高通量的酶联免疫吸附方法（ELISA），对重组蛋白进行链转移活性鉴定，表明重组蛋白具有较好的催化活性，该工作为建立整合酶链转移反应抑制剂筛选方法奠定了基础。（2）建立便捷高效的整合酶链转移反应抑制剂筛选方法研究的目标是建立高效便捷的HIV-1整合酶链转移反应抑制剂筛选方法，设计了生物素标记的供体DNA和荧光基团FITC标记的靶DNA，用链霉亲和素磁珠捕获反应体系中的DNA产物；最后用荧光分析仪检测DNA产物的荧光信号，并计算待测样品的抑制率。用已知整合酶抑制剂S-1360和MK-0518对筛选方法进行了验证，测定结果与已有实验数据相符合，表明本筛选方法能够有效并可靠地应用于HIV-1整合酶链转移反应抑制剂的筛选。与原有的整合酶链转移反应抑制剂筛选方法相比，本筛选方法还具有显著的改进：(i)原有的HIV-1整合酶链转移反应抑制剂体外筛选方法是用碱性磷酸酶标记的抗地高辛抗体检测地高辛基团，并利用碱性磷酸酶的特异性显色反应来读取信号，而本方法是利用荧光基团标记靶DNA，直接检测荧光基团的荧光信号即可对整合酶的链转移反应活性或待测样品的抑制活性进行评价，故本方法步骤更为简单，省力省时；(ii)原有的方法需要使用碱性磷酸酶标记的抗地高辛抗体，而本方法无需使用抗地高辛抗体，成本更为低廉。该方法还能应用于去整合反应和抑制剂作用机理的研究，具有重要的学术和应用价值。（3）应用CoMFA和SVM构建嘧啶酮类HIV-1整合酶抑制剂筛选模型 SVM应用于抑制剂筛选，通过对抑制剂进行定量构效关系（QSAR）建模，将化合物的活性与化合物的各种结构参数建立起相关联系，确定回归方程以预测新化合物的活性。应用CoMFA软件计算68个嘧啶酮类（pyrimidones）化合物的拓扑、分子极化、亲水性等结构参数，用所选的结构参数作为SVM的输入，建立起非线性的支持向量回归模型。模型获得训练集的活性预测值与实验值相关系数为0.938，获得测试集的活性预测值与实验值相关系数为0.804，表明模型所获得的活性预测值与实验结果相吻合，说明模型能够有效地进行HIV-1IN抑制剂筛选。本研究利用SVM算法与CoMFA方法相结合，为建立抑制剂筛选模型提供了一些有益的探索和有价值的信息，不仅可以进行药物虚拟筛选，并且可以为抗HIV-1药物设计提供有效的生物学信息。
[Abstract]:Since the discovery of AIDS patients for the first time in 1981, the rapid spread of AIDS worldwide has seriously threatened the health of mankind and has caused enormous damage to society. AIDS is caused by human immunodeficiency virus (HIV-1), and specific active sites and critical enzymes in viral replication cycles provide potential targets for antiviral chemotherapy. in which HIV-1-encoded integrase (IN) binds HIV-1 back-transcribed product cDNA into the host genome, which is an indispensable enzyme in the viral replication process. Since the functional analogue of IN is not present in the normal cells of the human host, the inhibitor specifically acting on IN is less toxic to normal cells, and in the meantime, the IN employs a single active site to accommodate two different DNA substrates, This, to some extent, limits HIV-1-induced resistance to the IN inhibitor and is therefore a very meaningful target for the development of anti-HIV-1 drugs. In vivo, the whole action is a multi-step reaction process, which mainly includes the combination of virus cDNA and IN, 3 '-processing; IN binds the virus cDNA to the host cell D by chain transfer reaction NA. In vitro, using recombinant IN protein, oligo-DNA substrate and divalent metal ion can realize the whole function of IN in vivo, thereby establishing a screening method of an effective IN inhibitor, which is the main screening method of the IN inhibitor in vitro. The recombinant IN protein was purified by this study, and the high-throughput screening of HIV-1 integrase inhibitor was established. Compared with the original method, the new screening method is sensitive and reliable, Low cost. The drug screening not only includes the biological experiment method, but also the computer-aided virtual screening method also occupies more and more in drug screening. This paper not only carries on the exploration of the screening method of the whole enzyme inhibitor on the basis of biology experiment, but also the virtual screening of the whole enzyme inhibitor from the angle of computer simulation and machine learning. The activity of HIV-1 integrase inhibitor was investigated by means of CoMFA analysis and support vector machine (SVM) modeling, and a prediction model based on quantitative structure effect relation (QSAR) and SVM was established. The results show that SVM shows strong modeling and prediction capability. The work of the thesis mainly comprises the following three aspects: (1) expression and purification of the whole enzyme protein and carrying out the activity identification of chain transfer reaction, firstly, transforming the prokaryotic expression vector of HIV-1 into the prokaryotic expression vector of the whole enzyme of HIV-1 (+)/ IN (F185K/ C280S) into E. coli JM109 cell B, L21 (DE3) was expressed in the prokaryotic expression vector, which contained IN gene and introduced F185K/ C. 280S mutation to improve protein solubility. Purification by affinity chromatography with nickel agarose gel to obtain purity and carrying out chain transfer activity identification on the recombinant protein by using a high-flux enzyme-linked immunosorbent assay (ELISA), The method lays a foundation for the screening method of the reaction inhibitor. The aim of this study is to establish an efficient and convenient screening method for the enzyme transfer inhibitor of HIV-1 integrase chain transfer. The target DNA labeled with FITC-labeled body DNA and fluorescent group is used to capture the DNA product in the reaction system by streptavidin magnetic beads; and finally, the DN is detected by a fluorescence analyzer. The fluorescence signal of A product is calculated and the inhibition rate of the sample to be measured is calculated. The screening method is verified by using the known whole enzyme inhibitor S-1360 and MK-0518, and the measurement result is consistent with the existing experimental data, so that the screening method can be effectively and reliably applied. Compared with the screening method of the original whole enzyme chain transfer reaction inhibitor, the screening method has the remarkable improvement: (i) the original HIV-1 integrase chain transfer reaction inhibitor in vitro screening method The method is characterized by that a fluorescent group is used to mark the target DNA, the fluorescent signal of the fluorescent group can be directly detected, The chain transfer reaction activity of the enzyme or the inhibition activity of the sample to be measured is evaluated, so that the method is simpler and labor-saving, The method does not need to use the anti-goxin antibody, and the cost is lower. The method can also be applied to the de-integration. The research on the mechanism of reaction and inhibitor action has important academic and application values. CoMFA and SVM are used to construct a screening model SVM for the anti-HIV-1 integrase inhibitor, which is applied to inhibitor screening, and the quantitative structure effect relationship (Q) is carried out on the inhibitor. SAR modeling to combine the activity of the compounds A regression equation is established to predict the activity of a new compound by establishing a regression equation to predict the activity of the new compound, and the structure parameters such as the topology, molecular polarization, hydrophilicity and the like of 68 oxycodone compounds are calculated by the CoMFA software, Using the selected structural parameters as the input of the SVM, a nonlinear support vector regression model was established. The correlation coefficient between the activity predicted value and the experimental value of the training set was 0.9938, and the correlation coefficient of the activity predicted value and the experimental value of the test set was 0. 804, indicating the predicted activity predicted value of the model. This study uses SVM and CoMFA to provide some useful exploration and valuable information to establish the inhibitor screening model, which not only can
【学位授予单位】：北京工业大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：R512.91;R446.1

【相似文献】