本文選題：QSAR + 急性毒性　； 參考：《山東大學(xué)》2016年博士論文

【摘要】：隨著化學(xué)工業(yè)的發(fā)展,越來越多新型化學(xué)品進(jìn)入了人類的生活。根據(jù)美國化學(xué)文摘(Chemical Abstracts Service, CAS)的最新統(tǒng)計數(shù)據(jù),目前注冊的已發(fā)現(xiàn)和合成的化學(xué)物質(zhì)總數(shù)已超過1億種,而其中很多物質(zhì)對環(huán)境和人類健康是有害的。面對數(shù)目如此龐大的化學(xué)物質(zhì),獲得其毒性信息從而進(jìn)行合理的監(jiān)管和使用,是保護(hù)人類和環(huán)境免受危害的根本方法。然而根據(jù)統(tǒng)計,目前處于監(jiān)管狀態(tài)下的化學(xué)品僅有30多萬種,絕大多數(shù)化學(xué)品缺少毒性信息,存在著巨大的數(shù)據(jù)缺口。而傳統(tǒng)生物實驗方法耗時長、花費大,無法滿足大量化學(xué)品急需測試的需要,且與目前提倡的動物保護(hù)“3R(替代(Replacement)、減少(Reduction)和優(yōu)化(Refinement))原則”相悖,因此能快速高效地獲取化學(xué)品生物毒性數(shù)據(jù)的計算機(jī)模擬預(yù)測方法一定量構(gòu)效關(guān)系(Quantitative structure-activity relationship, QSAR)方法已被越來越多的研究者所接受和使用。豐富的實驗數(shù)據(jù)是進(jìn)行QSAR研究的基礎(chǔ),只有在數(shù)據(jù)充足的前提下才能夠借助計算機(jī)、應(yīng)用統(tǒng)計學(xué)方法構(gòu)建出合理的構(gòu)效關(guān)系方程,然后結(jié)合生物化學(xué)及物理化學(xué)等相關(guān)知識,總結(jié)出化合物性質(zhì)變化和分子結(jié)構(gòu)特征變化之間的聯(lián)系及規(guī)律,進(jìn)而通過已知的分子結(jié)構(gòu)特征對化合物的未知性質(zhì)進(jìn)行預(yù)測。因此,QSAR的研究對象應(yīng)該是具有豐富實驗數(shù)據(jù)的性質(zhì)變量。目前在環(huán)境及生態(tài)毒理學(xué)領(lǐng)域常見的性質(zhì)變量有急性毒性、慢性毒性、致癌性和皮膚敏感性等。本論文選擇其中的急性毒性和致癌性進(jìn)行了QSAR研究。QSAR研究成功的一個重要因素是獲取合理準(zhǔn)確的分子結(jié)構(gòu)及相應(yīng)的結(jié)構(gòu)特征變量,為此需要對分子結(jié)構(gòu)進(jìn)行合理的幾何優(yōu)化。本論文中的分子結(jié)構(gòu)優(yōu)化都是在密度泛函理論(Density Functional Theory, DFT)方法的MPWB1K/6-31+G(d,p)水平下進(jìn)行的,并且通過研究OH自由基引發(fā)的滅多威在大氣中的降解機(jī)制對該水平的優(yōu)化方法進(jìn)行了實際效用的驗證,在此基礎(chǔ)上開展了有機(jī)物對嚙齒動物致癌性和有機(jī)物對水生生物急性毒性的研究。本論文的主要研究內(nèi)容及成果總結(jié)如下：(1)基于DFT方法研究OH自由基引發(fā)的滅多威在大氣中降解機(jī)理滅多威是一種氨基甲酸酯類農(nóng)藥,主要用作廣譜殺蟲劑,但其對動物和人類也會產(chǎn)生毒害,因此研究其在大氣中的降解機(jī)制具有重要的實際意義。本文研究了滅多威兩種同分異構(gòu)體在大氣中O2、NO和H20存在的情況下由OH自由基引發(fā)的降解反應(yīng)。在DFT方法的MPWB1K/6-31+G(d,p)水平下對反應(yīng)物、中間體和產(chǎn)物進(jìn)行了幾何結(jié)構(gòu)優(yōu)化和頻率計算,構(gòu)建了反應(yīng)的勢能剖面圖并討論了所有可能的降解路徑。采用小曲率隧道效應(yīng)(small-curvature tunneling,SCT)校正的正則變分過渡態(tài)理論(Canonical variational theory, CVT)計算了每條反應(yīng)路徑在200-400K溫度范圍內(nèi)基元反應(yīng)的速率常數(shù)。研究發(fā)現(xiàn),滅多威兩種同分異構(gòu)體的反應(yīng)狀況非常相似,既可以反生OH自由基加成反應(yīng),也可以發(fā)生H原子抽提反應(yīng)。根據(jù)反應(yīng)速率常數(shù)的分支比可以看出,對于Z式異構(gòu)體,常溫下抽提反應(yīng)占較大比重,隨著溫度上升,加成反應(yīng)比重上升。而對于E式異構(gòu)體,常溫下加成反應(yīng)占較大比重,并隨著溫度上升比重不斷增大。除此之外,還計算了兩種同分異構(gòu)體的大氣壽命,目前還未見其他有關(guān)報道。(2)基于DFT方法研究有機(jī)物對嚙齒動物致癌性的QSAR預(yù)測模型本論文基于CPDB腫瘤實驗數(shù)據(jù)庫,選擇數(shù)據(jù)最多的肝臟腫瘤實驗數(shù)據(jù)作為分析研究對象。通過對比雌雄兩種性別嚙齒生物的腫瘤實驗數(shù)據(jù),發(fā)現(xiàn)多種化合物存在明顯差異,而且總體而言,雌性實驗動物對致癌物表現(xiàn)出比雄性更強(qiáng)的耐受性。因此,綜合考慮分子的結(jié)構(gòu)特征,構(gòu)建了有環(huán)化合物對雄性大鼠、無環(huán)化合物對雄性大鼠、有環(huán)化合物對雌性大鼠、無環(huán)化合物對雌性大鼠、有環(huán)化合物對雄性小鼠、無環(huán)化合物對雄性小鼠、有環(huán)化合物對雌性小鼠、無環(huán)化合物對雌性小鼠等八個定量預(yù)測模型。然后根據(jù)模型、結(jié)合已有知識對致癌機(jī)理進(jìn)行了討論。研究所構(gòu)建的預(yù)測模型可以為研究性別對致癌作用的影響以及危險物質(zhì)的管理提供重要的理論支持。(3)基于DFT方法研究有機(jī)物對水生生物急性毒性QSAR預(yù)測模型本章選擇有機(jī)物對不同營養(yǎng)級典型水生生物(包括魚類和水蚤)的急性毒性作為研究對象,基于DFT方法優(yōu)化的分子結(jié)構(gòu)分別構(gòu)建了QSAR預(yù)測模型并進(jìn)行了對比分析。在魚類的急性毒性實驗中,黑頭呆魚是最常用的實驗對象之一,因此具有豐富的實驗數(shù)據(jù),為利用QSAR方法研究有機(jī)物的急性毒性預(yù)測模型提供了重要的數(shù)據(jù)支撐。本論文搜集整理了963種有機(jī)物對黑頭呆魚的96小時急性毒性實驗數(shù)據(jù),在DFT方法的MPWB1K/6-31+G(d,p)水平上對所有化合物的分子進(jìn)行了幾何結(jié)構(gòu)優(yōu)化,選擇Dragon描述符和辛醇水分配系數(shù)logKow作為建模的備選變量,借助遺傳算法和多元線性回歸構(gòu)建了目前數(shù)據(jù)集最大的黑頭呆魚急性毒性預(yù)測通用模型。除此之外,根據(jù)毒性比率及有機(jī)物毒性作用方式的劃分,分別構(gòu)建了麻醉性毒物和過量毒性毒物的急性毒性局域預(yù)測模型。大型蚤屬于世界種,是毒性實驗標(biāo)準(zhǔn)測試生物,具有豐富的實驗數(shù)據(jù)。本章搜集整理了479種有機(jī)物對大型蚤的48小時急性毒性實驗數(shù)據(jù),同樣在DFT方法的MPWB1K/6-31+G(d,p)水平上進(jìn)行了化合物的幾何結(jié)構(gòu)優(yōu)化,選擇Dragon描述符及l(fā)ogKow作為建模的備選變量,借助遺傳算法和多元線性回歸構(gòu)建了通用的急性毒性預(yù)測模型。除此之外，參考Verhaar區(qū)分框架,利用毒性比率將所研究化合物進(jìn)行了毒性作用機(jī)制的分類并分別構(gòu)建了QSAR模型,借助模型討論了每種機(jī)制化合物的作用機(jī)理。最后,將有機(jī)物對黑頭呆魚和大型蚤呈現(xiàn)出的不同毒性作用機(jī)制進(jìn)行了對比。研究發(fā)現(xiàn),大型蚤對毒物的敏感性要高于黑頭呆魚；除了苯胺類物質(zhì),硫代磷酸酯類物質(zhì)對黑頭呆魚和大型蚤的毒性也存在明顯差異,其對黑頭呆魚僅表現(xiàn)為麻醉毒性,對大型蚤卻呈現(xiàn)出過量毒性。
[Abstract]:With the development of the chemical industry, more and more new chemicals have entered human life. According to the latest statistics of Chemical Abstracts Service (CAS), more than 100 million kinds of chemicals have been found and synthesized, and many of them are harmful to the environment and human health. It is a fundamental way to protect human and environment from harm by obtaining such huge chemicals and obtaining its toxic information, which is the fundamental way to protect human and environment from harm. However, according to statistics, there are only about 300000 kinds of chemicals under supervision. Most chemicals are lacking in toxic information and there are huge data gaps. Biological experiments are time-consuming and costly to meet the needs of a large number of chemicals needed to test, and are contrary to the proposed animal protection "3R (Replacement), Reduction and Refinement"), so a computer simulation prediction method for obtaining chemical biological toxicity data quickly and efficiently The Quantitative structure-activity relationship (QSAR) method has been accepted and used by more and more researchers. The rich experimental data is the basis of QSAR research. Only with the sufficient data, a reasonable structure effect relation equation can be constructed with the help of the computer. The related knowledge of Biochemistry and physicochemistry, summarizes the relation and law between the properties of compounds and the changes of molecular structure characteristics, and then predicts the unknown properties of the compounds by the known molecular structure characteristics. Therefore, the research object of QSAR should be the property variable with rich and rich experimental data. The common variables in the field of ecotoxicology are acute toxicity, chronic toxicity, carcinogenicity, and skin sensitivity. In this paper, the acute toxicity and carcinogenicity of the QSAR are chosen. An important factor in the success of.QSAR research is to obtain a reasonable and accurate molecular structure and structural characteristic variables. The structure optimization is reasonable. The optimization of molecular structure in this paper is carried out under the MPWB1K/6-31+G (D, P) level of the Density Functional Theory (DFT) method, and the actual utility of the optimization method of the degradation of the Do-win in the atmosphere is studied by studying the degradation machine of the OH free radical. On this basis, the study on the carcinogenicity of organic substances to rodents and the acute toxicity of organic substances to aquatic organisms is carried out. The main contents and results of this paper are summarized as follows: (1) the study of the mechanism of the degradation of Do-win induced by OH radical based on the degradation mechanism of Do-win in the atmosphere is a kind of carbamate pesticide, which is mainly used as a wide range. Insecticides, however, are also toxic to animals and humans, so it is of great practical significance to study their degradation mechanism in the atmosphere. In this paper, the degradation reaction caused by two isomers in the atmosphere under the presence of O2, NO and H20 in the atmosphere was investigated by the OH free radical. The reaction at the MPWB1K/6-31+G (D, P) level of the DFT method was in response to the reaction. The materials, intermediates and products were optimized for geometric structure and frequency calculation. The potential energy profile of the reaction was constructed and all possible degradation paths were discussed. The regular variational transition state theory (Canonical variational theory, CVT) corrected by the small-curvature tunneling (SCT) was used to calculate each reaction path in 20 The rate constant of the base element reaction in the temperature range of 0-400K has been found. It is found that the reaction condition of two isomers of the Do-win is very similar, which can both reverse the OH radical addition reaction and the H atom extraction reaction. According to the branch ratio of the reaction rate constant, the extraction reaction of the Z isomer at normal temperature is a large proportion. As the temperature rises, the proportion of the addition reaction increases. For E isomers, the addition reaction at normal temperature is a large proportion and increases with the increase of temperature. In addition, the atmospheric life of two isomers has been calculated, and there are no other related reports. (2) the carcinogenicity of organic compounds to rodents is studied based on the DFT method. The QSAR prediction model is based on the CPDB tumor experiment database to select the most data of the liver tumor experimental data as the research object. By comparing the tumor experimental data of two sex rodents of male and female, it is found that there are obvious differences in various compounds, and in general, the female experimental animal shows the carcinogen to the male. Therefore, in consideration of the structural characteristics of the molecular structure, a cyclic compound was constructed for male rats, an acyclic compound against male rats, a ring compound against female rats, an acyclic compound to a female rat, a ring compound against male mice, an acyclic compound against male mice, and a cyclic compound against female mice. Eight quantitative prediction models of female mice, such as compound, were discussed in accordance with the model, combined with the existing knowledge. The prediction model constructed by the study can provide important theoretical support for the study of the effects of sex on carcinogenesis and the management of hazardous substances. (3) the study of organic matter to aquatic organisms is based on the DFT method. The acute toxicity of organic substances to the typical aquatic organisms (including fish and water fleas) of different trophic levels was selected in this chapter, and the QSAR prediction model was constructed based on the molecular structure optimized by the DFT method and compared. In the acute toxicity test of fish, the black head fish was the most commonly used experiment. One of the objects, therefore, has abundant experimental data and provides important data support for the acute toxicity prediction model of organic matter using the QSAR method. This paper collects and collects the 96 hour acute toxicity test data of 963 kinds of organic compounds to black head fish, and the molecules of all compounds on the level of MPWB1K/6-31+G (D, P) of the DFT method. The geometric structure optimization, the selection of the Dragon descriptor and the octanol water distribution coefficient logKow are selected as the alternative variables of the modeling. By means of genetic algorithm and multiple linear regression, a general model for predicting acute toxicity of black head fish is constructed with the help of the genetic algorithm and multiple linear regression. A local prediction model of acute toxicity of narcotic toxicants and toxic toxic substances was constructed. The large fleas belong to the world species, are the standard test organisms for toxicity test and have abundant experimental data. This chapter collects and collects the 48 hour acute toxicity test data of 479 kinds of organic compounds to the fleas, as well as on the MPWB1K/6-31+G (D, P) level of the DFT method. The geometric structure of the compound is optimized, the Dragon descriptor and logKow are selected as alternative variables of the modeling, and a general acute toxicity prediction model is constructed with the aid of genetic algorithm and multiple linear regression. In addition, the Verhaar classification framework is used as a reference for the classification of the toxic mechanism of the compounds using the toxicity ratio. The QSAR model was constructed, and the mechanism of each mechanism was discussed with the help of the model. Finally, the different toxicity mechanism of the organic matter to the black head fish and the large fleas was compared. The study found that the large fleas were more sensitive to the poison than the black head fish; besides aniline, thiophosphate was the same. Toxicities of the black headed fish and Daphnia magna are also different. They only show narcotic toxicity to the black headed fish, but show excessive toxicity to Daphnia magna.

【學(xué)位授予單位】：山東大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：X592
，

本文編號：1868805