抗生素等有机物的水解速率常数及产物研究

发布时间：2018-05-12 21:49

本文选题：抗生素 + 水解速率常数　；参考：《大连理工大学》2016年硕士论文

【摘要】：水解是有机化合物重要的降解途径之一,水解速率常数是对其进行生态风险评价的基础参数,可以通过实验测定获得。但是仅凭实验方法获取水解数据存在耗时费力、依赖设备和标准品等问题,非常有必要基于实验数据发展预测方法。定量结构-活性关系(QSAR)是目前应用较为广泛的预测技术,能够基于化合物的分子结构对其理化性质、环境行为参数和毒理学参数进行预测。抗生素类化合物在水环境中频繁被检出,其环境行为和生态风险引起广泛关注,对抗生素水解行为的实验研究很少,水解预测模型未见报道。依据经济合作与发展组织(OECD)提出的QSAR发展和使用导则,本研究采用多元线性回归(MLR)算法构建了酯类、酰胺类和抗生素类的水解速率预测模型。考虑到抗生素类化合物的水解数据较为缺失,本文依照美国环境保护署于2008年制定的水解实验导则,测定了七个抗生素的水解速率,并对三个抗生素的水解产物进行分析。主要研究发现如下：(1)基于从数据库和文献中收集的70个碱催化水解数据,构建了酯类模型,模型具有良好的拟合优度、稳健性和预测能力(R2adj=0.898,Q2LOO=0.851,Q2ext=0.651)。基于从数据库和文献中收集的94个水解数据,构建了酰胺类模型,模型具有一定的拟合优度、稳健性(R2adj=0.681,Q2LOO=0.647)和预测能力。(2)基于20种抗生素的水解速率常数,构建了抗生素的QSAR模型,模型具有良好的拟合能力、稳健性(R2adj=0.885,RMSE=0.139,Q2LOO=0.848)。从模型描述符可以看出,分子大小、形状、对称性和原子分布等对水解速率影响较大。采用实验测试的阿莫西林、盐酸强力霉素的水解数据作为抗生素类预测模型的验证集,模型预测能力良好。(3)实验测定了七种抗生素的水解速率常数,其中阿莫西林、盐酸强力霉素、氯霉素、罗红霉素易水解；林可霉素、盐酸环丙沙星、洛克沙砷不易水解。通过高分辨质谱对阿莫西林、盐酸强力霉素、青霉素G的水解产物进行了表征,并对可能的水解路径进行了分析。以阿莫西林为例,水解先是亲核试剂(H2O或OH-)对β-内酰胺键中C-N键进攻,使其断裂,化合物开环,然后是脱羧基和脱水反应。
[Abstract]:Hydrolysis is one of the most important degradation methods for organic compounds. The hydrolysis rate constant is the basic parameter for ecological risk assessment and can be obtained by experiment. However, it is necessary to develop a prediction method based on experimental data because it is time-consuming and laborious to obtain hydrolytic data only, relying on equipment and standard materials. Quantitative Structure-Activity relationship (QSAR) is a widely used prediction technique, which can predict the physical and chemical properties, environmental behavior parameters and toxicological parameters of compounds based on their molecular structure. Antibiotic compounds are frequently detected in the water environment, their environmental behavior and ecological risks have attracted wide attention. There are few experimental studies on the hydrolysis behavior of antibiotics, and the hydrolysis prediction model has not been reported. Based on the QSAR development and application guidelines proposed by the Organization for Economic Cooperation and Development (OECD), a prediction model of hydrolysis rate of esters, amides and antibiotics was constructed by using the multivariate linear regression (MLR) algorithm. In view of the lack of hydrolytic data of antibiotic compounds, the hydrolysis rate of seven antibiotics was determined according to the hydrolytic experimental guidelines formulated by the United States Environmental Protection Agency in 2008, and the hydrolysis products of three antibiotics were analyzed. The main results are as follows: (1) based on 70 alkali catalytic hydrolysis data collected from database and literature, the ester model is constructed. The model has good fit degree, robustness and predictive ability. Based on 94 hydrolytic data collected from database and literature, an QSAR model of antibiotics was constructed based on hydrolysis rate constants of 20 antibiotics. The model has good fitting ability, and its robustness is 0.885RMSE 0.139Q _ 2LOO _ (0.848). It can be seen from the model descriptors that the molecular size, shape, symmetry and atomic distribution have great influence on the hydrolysis rate. The hydrolysate data of amoxicillin and doxycycline hydrochloride were used as the verification set of antibiotic prediction model. Doxycycline hydrochloride, chloramphenicol, roxithromycin are easy to hydrolyze; lincomycin, ciprofloxacin hydrochloride, roxacin are not easy to hydrolyze. The hydrolysates of amoxicillin, doxycycline hydrochloride and penicillin G were characterized by high resolution mass spectrometry. For example, amoxicillin was hydrolyzed by nucleophilic reagents (H 2O or OH), which attacked the C-N bond in 尾 -lactam bond, broke it, opened the ring of the compound, then decarboxyl group and dehydration reaction.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：X131.2

【相似文献】