萜类化合物结构与其经皮促渗活性间的定量构效关系研究

发布时间：2018-06-18 05:42

本文选题：结构参数 + 热力学参数　；参考：《山西医科大学》2014年硕士论文

【摘要】：目的：使用促渗剂是提高药物经皮吸收的最有效和方便的方法，但其种类繁多，结构各异，逐一筛选非常困难，建立定量构效关系模型（QSAR）可对未知化合物的活性进行预测，从而指导促渗剂的筛选和结构优化。本文将密度泛函和从头算的量子化学计算方法用于41种萜类化合物结构参数及热力学参数的确定，以建立一种预测精度高的萜类化合物经皮促渗活性QSAR模型。方法：选择41种萜类化合物作为模型促渗剂，以氟哌啶醇（HP）为模型药物。采用量子化学法来确定化合物的结构参数及热力学参数。首先，利用内坐标做出萜类化合物的分子结构，在GaussianView软件中对以上各单体进行构造，其次，在HF/3-21G、B3LYP/6-311++G(2d,p)水平下进行各个单体的结构优化，得到最优结构，，最后，从out文件中采集或计算药物的结构参数及热力学参数：包括自然电荷（CN）、APT电荷（CAPT）、Mulliken电荷（CMu）、表面静电势（包含：最大值（VS,max）、最小值（VS,min）、总面积（VS,ta）、正面积（VS,pa）、负面积（VS,na））、双重偶极距（D）、分子量（M）、分子体积（Vm）、最高占据分子轨道（EHOMO）、最低未占据分子轨道（ELUMO）、前线轨道能极差（EH-U）、硬度（η）、化学势（λ）、焓（H）、吉布斯自由能（G）、熵（S）、摩尔热容（Cp）。分别考察41种萜类促渗剂对氟哌啶醇药物的促渗活性，用Kp来表示。本文的Kp值均来自L. Kang等人的文献，L. Kang等首先建立氟哌啶醇的高效液相色谱体外分析方法（HPLC），然后选择离体皮肤，用流式细胞扩散池测定用以表示促渗剂促渗活性指标的变量渗透系数（Kp）。采用多元线性回归建立萜类化合物的结构参数及热力学参数之间的QSAR模型。将所选择的萜类化合物分为三类：含-OH的化合物、含-C=O的化合物、烃类或烯萜类化合物，分别建立这三类化合物的QSAR模型，即Ya、Yb、Yc，最后建立41种萜类化合物总的QSAR模型Yd，根据得到的R值来评价数据间的拟合程度，采用留一法对模型进行内部检验，通过计算RMSE值来进行模型外部检验。结果：采用后退法回归建立的模型分别Ya=-6.107+10.873CAPT+14.485CMu-0.002VS,max-0.013VS,min+0.091VS,ta-1.496D+0.099M-0.068Vm-64.572Cp（n=22）、Yb=-10.675-17.105CMu-2.701D+0.067M（n=12）、Yc=0.491-0.014M+40.222η（n=7）、Yd=-15.823-0.008VS,max-0.009VS,min+0.02VS,ta-0.319D-51.952λ-11.15Cp（n=41）。模型的R值分别为0.933、0.860、0.985、及0.809，S分别为0.849、0.796、0.176、及1.000，经留一法交互验证得到RCV2=0.871、0.739、0.970、0.6540.5，对模型的准确性能检验得RMSE分别为0.627、0.650、0.133、0.910。根据RCV2和RMSE值可知模型a、b、c、d预测准确程度较高；且模型的稳定性和预测精度较好。由模型a、b、c、d表明，促渗剂促渗效果主要与APT电荷、Mulliken电荷、表面静电势（最大值、最小值、总面积）、双重偶极距、分子量、分子体积、硬度、化学势、摩尔热容等因素有关。在大多数情况下，最大静电势越小、最小静电势越小、静电势的总面积越大、双重偶极矩越小、摩尔热容越小的萜类化合物的促渗能力越强。结论:本课题所建立的模型均具有统计学意义，与L. Kang文中所建立模型相比，本文所建立的QSAR模型的预测精度相对较高，稳定性较好，从而进一步更好的指导萜类化合物的活性预测和结构优化，提高了促渗剂筛选的预测准确性。
[Abstract]:Objective: the use of the promoter is the most effective and convenient method to improve the drug transdermal absorption, but it has a wide variety and different structures. It is very difficult to screen one by one. A quantitative structure-activity relationship model (QSAR) can be used to predict the activity of the unknown compounds, so as to guide the screening and structural optimization of the enhancers. The quantum chemical calculation method is used to determine the structural parameters and thermodynamic parameters of 41 terpenoids, so as to establish a QSAR model for the percolation activity of terpenoids with high prediction accuracy.
Methods: 41 terpenoids were selected as model penetration enhancers and haloperidol (HP) was used as model drug. Quantum chemical method was used to determine the structural parameters and thermodynamic parameters of the compounds. First, the molecular structures of terpenoids were made using the internal coordinates, and the above monomers were constructed in the GaussianView software, followed by HF/3-21G, B At the level of 3LYP/6-311++G (2D, P), the structure of each monomer is optimized and the optimal structure is obtained. Finally, the structural parameters and thermodynamic parameters of the drug are collected or calculated from the out file, including the natural charge (CN), the APT charge (CAPT), the Mulliken charge (CMu), the surface electrostatic potential (including the maximum value (VS, max), the minimum value of the VS, the total area, and the total area. VS, PA, VS, Na), double polar distance (D), molecular weight (M), molecular volume (Vm), the highest occupying the molecular orbital (EHOMO), the lowest unoccupied molecular orbital (ELUMO), the front orbital energy difference (EH-U), the hardness (NA), the enthalpy (H), the Gibbs free energy (G), the entropy, and the molar heat capacity. The percolation activity of the acetamol drug is expressed in Kp. The Kp values of this article are all from the literature of L. Kang et al., L. Kang and so on first establish the high performance liquid chromatography (HPLC) method of haloperidol (HPLC), and then choose the isolated skin and determine the variable osmotic coefficient (Kp) using the flow cell diffusion pool to indicate the index of the promoter activity of the promoter (Kp). The QSAR model between the structural parameters and the thermodynamic parameters of terpenoids is established by the meta linear regression. The selected terpenoids are divided into three categories: -OH containing compounds, -C=O containing compounds, hydrocarbons or terpenoids, and the QSAR models of these compounds, namely, Ya, Yb, Yc, and the final establishment of the total QSAR of 41 terpenoids. Model Yd, according to the obtained R value to evaluate the degree of data fitting, using one method to carry out the internal test of the model, through the calculation of the RMSE value to carry out the model external test.
Results: the models established by regression method are Ya=-6.107+10.873CAPT+14.485CMu-0.002VS, max-0.013VS, min+0.091VS, ta-1.496D+0.099M-0.068Vm-64.572Cp (n=22), Yb=-10.675-17.105CMu-2.701D+0.067M (n=12), Yc=0.491-0.014M+40.222 (n=7), Yd= -15.823-0.008VS, max-0.009VS, ta-1.496D+0.099M-0.068Vm-64.572Cp. The R values of the model are 0.933,0.860,0.985, and 0.809, S are 0.849,0.796,0.176, and 1 respectively, and RCV2=0.871,0.739,0.970,0.6540.5 is obtained by one method. The accuracy of the model is verified to be 0.627,0.650,0.133,0.910. based on RCV2 and RMSE values for a, B, C, and the model is more accurate. The stability and prediction accuracy are good.
The model a, B, C, and d show that the penetration enhancers are mainly related to APT charge, Mulliken charge, surface static potential (maximum, minimum, total area), double even distance, molecular weight, molecular volume, hardness, chemical potential, mole heat capacity and so on. In most cases, the smaller the maximum electrostatic potential is, the smaller the minimum electrostatic potential is, the more the total area of the electrostatic potential is, the more the total area is. The smaller the double dipole moment and the smaller the molar heat capacity, the stronger the penetration of terpenes.
Conclusion: the model established in this project is of statistical significance. Compared with the model established in L. Kang, the prediction accuracy of the QSAR model is relatively high and the stability is better, thus further better guiding the activity prediction and structural optimization of terpenoids, and improving the accuracy of the prediction of the selection of the permeating agent.
【学位授予单位】：山西医科大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R943

【参考文献】