多氯联苯生物降解速率常数的电性拓扑模型

唐自强1,2,冯惠1,*,幺冰1,冯长君1,#

1. 徐州工程学院材料与化学工程学院,徐州 221018 2. 徐州技师学院,徐州 221151

摘要: 基于拓扑化学理论,用原子类型的电性拓扑状态指数(EA)描述了66个多氯联苯分子的化学微环境。基于EA和最佳变量子集回归,建立上述化合物生物降解速率常数(lnK)的定量结构-生物降解性关系(QSBR)模型。其最优三元(EC2EC3ECl) QSBR模型的判定系数(R2)和逐一剔除法交叉验证系数分别为0.848和0.824。经函数(FIT)、Akaike信息判据(AIC)检验,QSBR模型具有良好的估计稳定性和预测能力。结果显示影响多氯联苯生物降解速率常数的主要因素是分子内所含氯原子的数目及其所处位置。

关键词: 多氯联苯;生物降解速率常数;电性拓扑指数;定量结构-生物降解性关系

多氯联苯(polychlorinated biphenyls, PCBs)具有优良的阻燃性、热稳定性、惰性和介电能力,而被广泛用作与人类生产生活紧密相关的阻燃剂、塑化剂油漆、复印纸和变压器油等。然而,1968年日本“米糠油事件”(即“油症”,20世纪世界八大公害事件之一)首次将人们的关注点从PCBs的商业应用转向对人类健康的影响上。因PCBs良好的化学稳定性,使其在自然环境中很难降解,而通过水、土、气和生物等载体转移,在人或动物体内富集,进而对生物体产生致癌、致畸、致突变,以及内分泌干扰等作用,对人类生存繁衍和可持续发展构成严重威胁[1-3]。近几十年来,对PCBs的修复一直是环境科学技术领域的研究热点。以微生物为核心的降解技术是公认的成本低、无二次污染、环境友好的修复技术,且特别适合土壤和河底底泥中低浓度持久性有机污染物的原位修复。不过,现已获得的PCBs降解菌却存在着降解高氯代PCBs能力差、抗毒能力低下、降解谱窄等缺陷。贾凌云[4]从变压器油污染物中筛选出一株降解PCBs的新菌株,经分子生物学鉴定和生理生化指标鉴定,确认该菌株为肠杆菌属(Enterobacter sp.),命名为LY402。实验证明该菌株不仅对PCBs具有良好的降解能力,而且呈现很强的抗污染物毒性的能力,她在无碳源液相中测定了LY402对PCBs生物降解的速率常数(K),单位为h-1

由于进入环境中的有机污染物种类繁多,数量庞大,难以通过实验测定所有污染物的生物降解性,并且不同学者所采用的实验方法、条件也不尽相同,导致已公布的降解性数据之间的可比性较差。因此,最早用于药物的定量构效关系(quantitative structure activity relationships, QSAR)[5-15]研究方法便被移植来估算与预测有机污染物的生物降解性,由此形成物质的定量结构-生物降解性相关性(quantitative structure-biodegradability relationship, QSBR)分支学科[16-19]。已成为环境科学、医药和生物等学科的热门领域。因此,本文基于Hall与Kier等提出的原子类型电性拓扑指数(electrotopological state index of atom type,用“EA”表示)[20-22],建立PCBs生物降解速率常数(lnK)[4]的QSBR模型,以预测PCBs的lnK,并探讨影响lnK的主要结构因素。

1 材料与方法(Materials and methods)

1.1 研究对象

PCBs的基本结构如图1所示,分子通式为“C12Hn+mCl10-n-m”,其中n≤5,m≤5。按照联苯环上氯原子取代数目和取代位置的不同,理论上有209种化合物,而在环境中出现频率较高的仅有90余种。

图1 多氯联苯(PCBs)的基本结构
Fig. 1 Basic structure of polychlorinated biphenyls (PCBs)

贾凌云[4]在无碳源液相中测定了LY402对66种PCBs生物降解的K,具体数值如表1所示。由于K值相差太大,取其对数用于建模,即lnK,其数值如表1所示。

表1 PCBs的分子结构与生物降解速率常数(lnK)
Table 1 The molecular structures and rate constants for biodegradation (lnK) of PCB congeners

No.PCB同系物PCB congenersEC2EC3EClK/h-1[4]lnKExp.[4]Cal.Err.13,4,2’-PCB13.0953.71417.8590.2650-1.328-2.2260.89722,3,4’-PCB13.0563.76017.8510.2470-1.398-2.1680.77032,6,3’-PCB12.8963.68018.0910.2340-1.452-1.354-0.09942,3,2’,3’-PCB10.8023.51024.1320.2250-1.492-2.3790.88752,4,3’-PCB12.9673.87717.8230.2210-1.510-2.0670.55862,2’4,5-PCB10.7623.67024.0120.2160-1.532-2.5681.03574,4’-PCB15.4883.79411.6060.2050-1.585-2.1990.61482,3,2’-PCB13.0453.49118.1310.2000-1.609-1.536-0.07492,6,2’-PCB12.8953.50018.2710.1720-1.760-0.953-0.807102,3,2’,4’-PCB10.6873.76123.9960.1670-1.790-2.4660.676112,5,2’,5’-PCB10.4523.95524.0380.1620-1.820-1.9500.129122,5,3’,4’-PCB10.6394.00223.8040.1350-2.002-2.8010.799132,3,4,3’-PCB10.9363.62923.8800.1210-2.112-3.1801.068142,4,2’,4’-PCB10.5774.00723.8610.1190-2.129-2.5610.432152,6,4’-PCB12.8863.79617.9850.1050-2.254-1.564-0.690162,4,5,3’-PCB10.7613.85223.8320.0995-2.308-2.9620.655172,3,6,2’-PCB10.7123.44724.2850.0887-2.422-1.880-0.542182,5,3’-PCB12.8833.87717.9070.0589-2.832-1.727-1.105192,3,5,4’-PCB10.7223.89823.8240.0579-2.849-2.9060.056202,3,3’,4’-PCB10.8143.77923.8520.0575-2.856-3.0190.163212,3,5,2’,3’-PCB8.5743.52130.1270.0559-2.884-3.2660.382222,3,5,2’,5’-PCB8.3993.74330.0800.0545-2.910-3.0500.140232,2’,3,4’,5-PCB8.4443.79029.9880.0489-3.018-3.3320.314242,3,4,2’,5’-PCB8.5723.62730.0240.0479-3.039-3.4910.452252,3,6,2’,3’-PCB8.5603.35730.3050.0478-3.041-2.856-0.185262,4,5,2’,5’-PCB8.3973.84929.9760.0443-3.117-3.2710.154272,4,5,2’,6’-PCB8.4613.59030.1720.0438-3.128-2.964-0.164282,3,6,3’,4’-PCB8.5123.69930.0110.0413-3.187-3.4100.223292,4,5,2’.3’-PCB8.5723.62730.0240.0347-3.361-3.4910.130302,3,6-PCB13.1443.36818.1550.0329-3.414-1.667-1.747312,3,5,6,2’,5’-PCB6.4603.21436.3260.0325-3.427-3.9150.488322,3,6,2’,4’-PCB8.4233.63630.1640.0322-3.436-2.912-0.524332,4,5,4’-PCB10.7483.97123.7260.0307-3.483-3.170-0.314342,3,6,2’,3’,6’-PCB6.4463.05136.5030.0304-3.493-3.5020.009352,3,5,6,2’,3’-PCB6.6352.99236.3740.0283-3.565-4.1340.569362,4,6,2’,4’-PCB8.3753.80430.0430.0261-3.646-3.084-0.561373,4,5,2’,5’-PCB8.5863.79029.8470.0238-3.738-3.9030.165382,3,4,3’,5’-PCB8.6333.73129.8580.0183-4.001-3.961-0.040392,3,4,2’,4’-PCB8.6223.66829.9330.0178-4.029-3.779-0.249402,3,5,2’,3’,6’-PCB6.4233.25936.3180.0173-4.057-3.865-0.192412,4,5,2’,4’,5’-PCB6.3833.69335.9230.0161-4.129-4.6550.526422,3,4,3’,4’-PCB8.7333.70529.7850.0161-4.129-4.3060.177432,3,4,2’,3’,5’-PCB6.5723.35236.0770.0135-4.305-4.6670.362442,3,4,6,2’,3’-PCB6.6953.02036.2850.0134-4.313-4.4370.125452,3,6,2’,4’,5’-PCB6.3983.39336.2100.0118-4.440-4.058-0.381462,4,6,2’,4’-PCB8.3753.80430.0430.0117-4.448-3.084-1.364472,3,4,6,2’,4’,5’-PCB4.5743.00542.1980.0109-4.519-5.6981.179483,4,5,2’,3’-PCB8.7613.56729.8940.0099-4.616-4.119-0.497493,4,3’,4’-PCB10.8594.00723.5780.0095-4.653-3.694-0.959502,3,5,6,2’,4’,5’-PCB4.5252.96442.2890.0084-4.785-5.4080.622512,3,4,5,2’,4’-PCB6.6813.32135.9990.0071-4.952-5.0370.085522,3,4,6,2’,3’,5’-PCB4.6102.85942.3090.0066-5.028-5.5220.493532,3,4,5,6,2’,3’-PCB4.7142.90542.1600.0042-5.468-6.0400.572542,3,4,2’,3’,4’-PCB6.7343.24836.0180.0041-5.504-5.091-0.413

续表1No.PCB同系物PCB congenersEC2EC3EClK/h-1[4]lnKExp.[4]Cal.Err.552,3,5,6,2’,3’,4’-PCB4.7002.74242.3370.0036-5.627-5.6280.001562,3,4,5,2’,3’-PCB6.8223.03836.1400.0026-5.972-4.984-0.987572,3,4,5,6,2’,3’-PCB5.0722.27842.4280.0024-6.049-6.1080.059582,3,4,5,2’,3’,6’-PCB4.7382.69542.3450.0022-6.133-5.680-0.453592,3,4,5,2’,3’,5’,6’-PCB2.9332.18548.4380.0019-6.255-7.1250.869602,3,5,6,3’,4’,5’-PCB4.6542.97742.1460.0019-6.271-5.957-0.315612,3,4,5,3’,4’-PCB6.7693.38535.8460.0017-6.395-5.531-0.864622,3,4,6,2’,3’,4’-PCB4.7502.78342.2460.0015-6.502-5.917-0.585632,3,4,5,6,2’,3’,6’-PCB3.1161.78148.6590.0009-6.968-6.9760.009642,3,4,6,3’,4’,5’-PCB4.7123.00942.0570.0008-7.170-6.257-0.913652,3,4,3’,4’,5’-PCB6.7323.43035.8380.0007-7.240-5.482-1.758662,3,4,5,6,2’,3’,5’-PCB3.0552.03548.4660.0006-7.349-7.287-0.062

1.2 原子类型电性拓扑指数(EA)

将化合物的抽象结构予以参数化是建立定量构效关系(QSAR/QSPR)模型的最重要步骤。拓扑指数为实现分子结构的数值化表征提供了简便方法。原子类型电性拓扑状态指数(E-state indices, En)是表征分子中每种非氢原子类型的结构参数,由两部分构成:其一是非氢原子类型本身的原子结构及局部拓扑环境,由此形成非氢原子的固有状态值(即本征值);其二是反映该原子受分子中其他非氢原子扰动程度的本征值增量。对于PCBs,只存在3种原子类型:C—、C⟨、—Cl,对应3种指数,依次为:EC2EC3ECl,具体计算过程参见文献[20-22]。

1.3 统计回归分析

将PCBs分子的EC2EC3ECl作为自变量集,其生物降解速率常数(lnK)[4]作为因变量,采用最佳变量子集回归的方法(leaps-and-bounds regression)确定lnK模型的最佳变量组合。一般按照“三性原则”确定QSAR模型:一是统计性,样本容量(f)与自变量数(b)之比,称为容变比(SV),即SV=f/b>5的模型才具有统计意义[23],此是建立模型的基础。二是相关性,要求模型的可决系数(R2)>0.8,为高度相关[24]。三是预测性,要求模型的交叉验证相关系数显示良好的预测能力[25]。另外,Akaike信息判据(Akaike’s information criterion, AIC)、Kubinyi函数(Kubinyi function, FIT)[26-27]也用于衡量模型的质量,其计算公式如下:

(1)

(2)

导致AIC增大,FIT减小的自变量,不宜引入模型;式中,RSS表示估计标准误差(standard error of estimate)。

2 结果(Results)

2.1 lnK的线性回归模型

将66种PCBs的lnK[4]和上述3种拓扑指数输入MINITAB统计分析软件,运用其中的最佳子集回归方法选择最佳变量组合,建立的最佳QSAR模型如表2所示。其中SD分别为削减误差比例、交叉验证判定系数、校正判定系数、Fisher统计值、容变比和估计标准误差。

由表2可知,FIT等随自变量数增加而递增,AICSD均递降,都指向三元模型质量最优。故选定PCBs的lnK的QSBR模型:

lnK=148.522(±22.988)-2.597(±0.378)ECl-

6.614(±1.023)EC2-4.781(±0.826)EC3

(3)

115.076

表2 生物降解速率常数(lnK)和电性拓扑指数(EA)的最佳子集回归结果
Table 2 The results between electrotopological state index (EA) and rate constants for biodegradation (lnK) with leaps-and-bounds regression

No.R2R2cvR2adjAICFITFSVSD变量Variables10.7440.7280.7400.7852.776185.781660.854ECl20.7650.7480.7580.7632.930102.821330.824ECl, EC230.8480.8240.8400.5584.612115.076220.669ECl, EC2, EC3

2.2 lnK模型的质量检验

首先,模型(3)的SV=22,远大于5,显示模型具有很好的统计意义,随机性低,具备建模的基本要求。二是R2=0.848>0.8,呈现良好的拟合性。R2又称为削减误差比例,因此,模型(3)中隐含影响多氯联苯lnK的84.8%因素,仅有不足15.2%属于未知因素。三是远大于0.5,呈现良好的预测准确性。由表1可知,预测值(lnKcal.)与相应实验值(lnKexp.)较好吻合。为防止模型存在过拟合及离域点,要求本模型远小于0.3,可见没有拟入噪音及奇异值。

3 讨论(Discussion)

EC2EC3ECl与氯原子的数目(d)的判定系数依次为0.995、0.653和1.000,显示高度相关。另外,这3种指数对66种PCBs的分子结构实现唯一性表征,不存在相同的数值。由此可见,模型(3)揭示了在生物降解环境相同情况下,PCBs生物lnK呈现的规律:一是与其分子内所含氯原子的数目(d)负相关;二是对于同分异构体,与氯原子所处位置有关。此与文献[4]给出的结论是一致的。

在模型(3)中,EC2EC3ECl前的系数均<0,说明PCBs分子中所含氯原子越多,其降解速率常数越小。其原因是联苯环上被氯取代的位点越多,其性质越稳定,降解菌破坏稳定的碳-氯键进而实现开环降解需要的能量也越高,因此,导致K降低。

综上所述:(1)由表1可知,EC2EC3ECl等指数对PCBs分子结构呈现良好的选择性,实现唯一性表征,即对不同的化合物,不存在相同的数值。(2) 经“三性”原则等检验,所建lnK线性模型不仅具有良好的相关性、稳定性,而且呈现良好的预测能力。(3) 根据进入模型的EC2EC3ECl可知,影响PCBslnK的主要因素是分子内所含氯原子的数目及其所处位置。

参考文献(References):

[1] 唐文雅, 竺美, 黄冬梅, 等. 危险废物鉴定中痕量多氯联苯的前处理优化分析[J]. 环境监测管理与技术, 2021, 33(3): 49-52

Tang W Y, Zhu M, Huang D M, et al. Study on optimization of sample preparation for trace PCBs in hazardous waste identification [J]. The Administration and Technique of Environmental Monitoring, 2021, 33(3): 49-52 (in Chinese)

[2] 王涛, 吴启美. 植物根际降解土壤多氯联苯机理研究进展[J]. 环境科学与技术, 2021, 44(4): 36-44

Wang T, Wu Q M. Research progress on the mechanism of plant rhizosphere degradation of soil polychlorinated biphenyls [J]. Environmental Science& Technology, 2021, 44(4): 36-44 (in Chinese)

[3] 马丽莎, 谢文平, 田斐, 等. 广东沿海养殖牡蛎中多氯联苯残留水平及人体饮食暴露风险评估[J]. 南方水产科学, 2021, 17(2): 11-19

Ma L S, Xie W P, Tian F, et al. Bioaccumulation and human health risk assessment of polychlorinated biphenyls (PCBs) in farmed oysters along Guangdong coast [J]. South China Fisheries Science, 2021, 17(2): 11-19 (in Chinese)

[4] 贾凌云. 多氯联苯降解菌的筛选及其降解性能研究[D]. 大连: 大连理工大学, 2008: 65-70

Jia L Y. Isolation and characterization of a novel polychlorinated biphenyl-degrading bacterium [D]. Dalian: Dalian University of Technology, 2008: 65-70 (in Chinese)

[5] Feng H, Du X H, Chen Y, et al. 3D-QSAR models of anti-tumor activity for histone deacetylase inhibitors containing dihydropyridin-2-one [J]. Chinese Journal of Structural Chemistry, 2020, 39(5): 855-860

[6] Zhu L L, Qin Z L, Feng C J. CoMFA Model and Molecular Design of Anti -excitatory Activity for Benzodiazepinooxazole Derivatives against Mice. Chin. J. Struct. Chem., 2021, 40(8): 1075-1081

[7] Feng H, Feng C J. 3D-QSAR studies on the anti-tumor activity of N-aryl-salicylamide derivatives [J]. Chinese Journal of Structural Chemistry, 2019, 38(11): 1874-1880

[8] 廖立敏, 李建凤, 雷光东. 含氯苯酚类化合物结构表征与毒性预测[J]. 生态毒理学报, 2017, 12(6): 266-272

Liao L M, Li J F, Lei G D. Structural characterization and toxicity prediction of chlorinated phenolic compounds [J]. Asian Journal of Ecotoxicology, 2017, 12(6): 266-272 (in Chinese)

[9] 张文灏, 陈景文, 徐童, 等. 外源化合物在鱼体内生物半减期的QSAR模型[J]. 生态毒理学报, 2019, 14(3): 90-98

Zhang W H, Chen J W, Xu T, et al. QSAR models for predicting biological half-life of xenobiotics in fish [J]. Asian Journal of Ecotoxicology, 2019, 14(3): 90-98 (in Chinese)

[10] 冯长君. 基于CoMFA研究氟喹诺酮C-3噻唑酮衍生物抗胰腺癌活性[J]. 徐州工程学院学报:自然科学版, 2021, 36(3): 8-12

Feng C J. Study on anti-pancreatic cancer activity of fluoroquinolone C-3 thiazolone derivatives based on CoMFA [J]. Journal of Xuzhou Institute of Technology: Natural Sciences Edition, 2021, 36(3): 8-12 (in Chinese)

[11] 郑玉婷, 乔显亮, 于洋, 等. 有机化学品生物富集因子定量结构-活性关系模型[J]. 生态毒理学报, 2019, 14(2): 214-221

Zheng Y T, Qiao X L, Yu Y, et al. Quantitative structure-activity relationship model for bioconcentration factors of organic chemicals [J]. Asian Journal of Ecotoxicology, 2019, 14(2): 214-221 (in Chinese)

[12] 唐自强, 冯长君. 取代苯酚类化合物抑藻活性的CoMFA模型[J]. 生态毒理学报, 2019, 14(4): 192-196

Tang Z Q, Feng C J. CoMFA model for inhibitory activity of chlorinated phenolic compounds to Dunaliella salina [J]. Asian Journal of Ecotoxicology, 2019, 14(4): 192-196 (in Chinese)

[13] 唐自强, 冯长君. 硝基芳烃对梨形四膜虫急性毒性的CoMFA模型[J]. 生态毒理学报, 2020, 15(5): 327-332

Tang Z Q, Feng C J. CoMFA model for acute toxicity of nitroaromatic compounds to Tetrahymena pyriformis [J]. Asian Journal of Ecotoxicology, 2020, 15(5): 327-332 (in Chinese)

[14] 冯长君. 苯磺酰脲类化合物除草活性的CoFMA模型[J]. 徐州工程学院学报: 自然科学版, 2019, 34(2): 21-25

Feng C J. CoFMA model of herbicidal activity of phenyl-sulfonylurea derivatives [J]. Journal of Xuzhou Institute of Technology: Natural Sciences Edition, 2019, 34(2): 21-25 (in Chinese)

[15] 冯长君. 硝基苯衍生物对发光菌抑制毒性的CoMFA模型[J]. 徐州工程学院学报: 自然科学版, 2020, 35(1): 28-31

Feng C J. CoMFA model of inhibitory activity for nitrobenzene derivatives to photobacteria [J]. Journal of Xuzhou Institute of Technology: Natural Sciences Edition, 2020, 35(1): 28-31 (in Chinese)

[16] 贺美, 向廷生, 谢瑶, 等. 影响有机化学品快速生物降解性的分子结构参数研究进展[J]. 生态科学, 2015, 34(3): 181-188

He M, Xiang T S, Xie Y, et al. A review on molecular structural descriptors affecting ready biodegradation of organic chemicals [J]. Ecological Science, 2015, 34(3): 181-188 (in Chinese)

[17] 樊健, 尚少鹏. 黄河包头段水体中取代酚的生物降解性研究[J]. 灌溉排水学报, 2017, 36(1): 85-90

Fan J, Shang S P. Biodegradability of substituted benzenes in Baotou section of the Yellow River [J]. Journal of Irrigation and Drainage, 2017, 36(1): 85-90 (in Chinese)

[18] Chen Y, Zeng Y X, Ji Z J, et al. Identification of stable quantitative trait loci for sheath blight resistance using recombinant inbred line [J]. Rice Science, 2019, 26(5): 331-338

[19] 张丹, 晁聪, 李玉坤, 等. 定量构效关系应用于水中有机污染物降解过程的研究进展[J]. 化工环保, 2021, 41(4): 418-426

Zhang D, Chao C, Li Y K, et al. Research progresses on application of quantitative structure-activity relationship to degradation process of organic pollutants in water [J]. Environmental Protection of Chemical Industry, 2021, 41(4): 418-426 (in Chinese)

[20] 堵锡华, 王超. 碳纳米管吸附能力的神经网络理论模型[J]. 环境化学, 2016, 35(7): 1445-1451

Du X H, Wang C. Research on the adsorption of carbon nanotubes to aromatic compounds by neural network [J]. Environmental Chemistry, 2016, 35(7): 1445-1451 (in Chinese)

[21] Kier L B, Hall L H. An electrotopological-state index for atoms in molecules [J]. Pharmaceutical Research, 1990, 7(8): 801-807

[22] Hall L H, Kier L B. Electrotopological state indices for atom types: A novel combination of electronic, topological, and valence state information [J]. Journal of Chemical Information and Computer Sciences, 1995, 35(6): 1039-1045

[23] 刘东, 章文军, 许禄. 手性羟酸和氨基酸类化合物的构效关系研究[J]. 化学学报, 2009, 67(2): 145-150

Liu D, Zhang W J, Xu L. Quantitative structure-activity/property relationships for chiral hydroxy acids and amino acids [J]. Acta Chimica Sinica, 2009, 67(2): 145-150 (in Chinese)

[24] 张骥, 申鹏, 陆涛, 等. 黄酮类化合物抑制MMP-9的定量结构-活性关系和结构修饰的理论研究[J]. 化学学报, 2011, 69(4): 383-392

Zhang J, Shen P, Lu T, et al. Theoretical studies on quantitative structure-activity relationship and structural modification for the inhibition of MMP-9 by flavonoids [J]. Acta Chimica Sinica, 2011, 69(4): 383-392 (in Chinese)

[25] 胡松青, 米思奇, 贾晓林, 等. 苯并咪唑类缓蚀剂的3D-QSAR研究及分子设计[J]. 高等学校化学学报, 2011, 32(10): 2402-2409

Hu S Q, Mi S Q, Jia X L, et al. 3D-QSAR study and molecular design of benzimidazole derivatives as corrosion inhibitors [J]. Chemical Journal of Chinese Universities, 2011, 32(10): 2402-2409 (in Chinese)

[26] Saíz-Urra L, Pérez González M, Teijeira M. 2D-autocorrelation descriptors for predicting cytotoxicity of naphthoquinone ester derivatives against oral human epidermoid carcinoma [J]. Bioorganic & Medicinal Chemistry, 2007, 15(10): 3565-3571

[27] Saíz-Urra L, González M P, Teijeira M. QSAR studies about cytotoxicity of benzophenazines with dual inhibition toward both topoisomerases Ⅰ and Ⅱ: 3D-MoRSE descriptors and statistical considerations about variable selection [J]. Bioorganic & Medicinal Chemistry, 2006, 14(21): 7347-7358

Electrotopological Model of Biodegradation Rate Constant for Polychlorinated Biphenyls

Tang Ziqiang1,2, Feng Hui1,*, Yao Bing1, Feng Changjun1,#

1. School of Materials and Chemical Engineering, Xuzhou Institute of Technology, Xuzhou 221018, China 2. Xuzhou Technical Institute, Xuzhou 221151, China

Abstract: Based on topological chemical theory, electrotopological state index of atom type (EA) were used to describe the chemical microenvionment of 66 polychlorinated biphenyls molecules (PCBs). The quantitative structure-biodegradability relationship (QSBR) models for estimating the rate constants (lnK) for biodegradation of above compounds was developed based on the EA and leaps-and-bounds regression. The coefficient of multiple determination (R2) and cross-validated coefficient of multiple determination of leave-one-out (LOO) of the optimal three variable (EC2, EC3, ECl) QSBR model were 0.848, 0.824, respectively. The QSBR model has both favorable estimation stability and good prediction capability by R2, Kubinyi function (FIT), Akaike’s information criterion (AIC) tests. The results showed that the main factor affecting the biodegradation rate constant (lnK) of PCBs was the number and location of chlorine atoms in the molecule.

Keywords: polychlorinated biphenyls; rate constant for biodegradation; electrotopological state index of atom type; quantitative structure-biodegradability relationship (QSBR)

收稿日期2021-10-12

录用日期2022-01-10

基金项目结构化学国家重点实验室开放基金(2016028)

第一作者唐自强(1963—),男,学士,教授,研究方向为物质构效学,E-mail: tzq63@163.com

*通讯作者(

Corresponding author), E-mail: fengh@xzit.edu.cn

# 共同通讯作者(Co-corresponding author), E-mail: fengcj@xzit.edu.cn

DOI: 10.7524/AJE.1673-5897.20211012002

唐自强, 冯惠, 幺冰, 等. 多氯联苯生物降解速率常数的电性拓扑模型[J]. 生态毒理学报,2022, 17(5): 319-324

Tang Z Q, Feng H, Yao B, et al. Electrotopological model of biodegradation rate constant for polychlorinated biphenyls [J]. Asian Journal of Ecotoxicology, 2022, 17(5): 319-324 (in Chinese)

文章编号: 1673-5897(2022)5-319-06

中图分类号: X171.5

文献标识码: A

Received 12 October 2021

accepted 10 January 2022

通讯作者简介:冯惠(1985—),女,博士研究生,讲师,主要研究方向为药物构效关系。

共同通讯作者简介:冯长君(1954—),男,学士,教授,主要研究方向为物质构效关系。