2D-and 3D-QSBR Studies on the Relationship between Structure and Biodegradability of Phthalates

HAN Xing-Yun SHI Ji-Qi CHEN Tin-Ming

?

2-and 3-QSBR Studies on the Relationship between Structure and Biodegradability of Phthalates

HAN Xiang-Yuna①SHI Jia-QibCHEN Tian-Minga

a(224051) JGHX201408b(210023)

Nine phthalates were calculated at the B3LYP/6-311G** level using DFT method. The corresponding linear relationship equations (2were 0.853 and 0.936 respectively) for the biodegradation rate (b) and half-life time (1/2) of biodegradation were obtained with the structural parameters as theoretical descriptors. Furthermore, CoMFA method was also applied to establish 3models which revealed the fields influencing these properties. The relationship between the properties and the structure was obtained. The correlation coefficients of the models were 0.992 and 0.999, respectively. Analyses of 2and 3models demonstrated that the molecular volume was an important factor affecting the biodegradability of these compounds.

phthalates, biodegradability, density functional theory, comparative molecular field analysis

1 INTRODUCTION

Phthalates (referred to as PAEs) are ones of the synthetic organic compounds[1]that are produced in great quantity and used commonly as pesticide carriers and plastic reinforcer and modifier widely in the whole world[2, 3]. There are about 14 phthalates that are used commercially. As worldwide pollutants, they exist widely in air, water, soil and organisms and have caused serious harm to the environment where the human being lives and the health of human body.

Biodegradation is the chief way by which envi- ronmental pollutants are degraded and eliminated. Quantitative structure-activity/biodegradation rela- tionship (QSAR/QSBR) study is one of the indispen- sable methods used to assess the ecological risk of organic pollutants[4]. A great number of biodegrading properties of compounds may be acquired through constructing the effective models that possess similar structure of the targeted compound, which may help to guide the practical control of pollution.calculation is a calculating method based on non-relativistic approximation, Born-Oppen- heimer approximation and Hartree approximation and does not use any empirical parameter, and it may solve Schrodinger equation through calculating the integration of all molecules in a system; in contrast with the multiple semi-empirical calculating methods used to calculate molecular orbits,method may give the most precise result, and it is the most strict theoretically, but its calculated amount is much greater than that of semi-empirical methods.method includes Hartree-Fock method, density functional theory (DFT), electron correlation method,. In recent years, with the steady increase of calculating speed of computers,method has become the mainstream for quantum chemistry calculation in the world, and it has been applied in the study on the structure activity relationship for a series of different com- pounds[5-12]because the models given by it may correlate better to the targeted compound as com- pared with models given by semi-empirical methods. Comparative molecular field analysis (CoMFA) as a common 3-QSAR method uses micro molecular 3structure as descriptor and it broke through the limitation of 2-QSAR method in the characteri- zation of molecular structure and configuration. Combining 2-QSAR with 3-QSAR may explain the quantitative structure-activity/biodegradation relationship from different aspects such as molecular structure and molecular force field[13-17].

In this paper, authors used DFT in Gaussian 09 program[18]and 6-311G** basis set to calculate 9 phthalates, and then introduced calculated structural and thermodynamic parameters as theoretical descri- ptors to QSBR method and adopted GQSARF2.0 program[19]to get the quantitative relationship be- tween the biodegradation rate constant and para- meters of half-life time of compounds and their structural and thermodynamic parameters through fitting, and then adopted CoMFA[20]to study the relationship between biodegradation rate constant and half-life time of phthalates and their respective 3structures.

2 CALCULATING METHODS

The basic structures of 9 phthalates are shown in Fig. 1, and their specific names are listed in Table 1. Biodegradation rate constantb(d-1) and half-life time1/2(d)[21]of these 9 phthalates are shown in Table 1.

Table 1. Names and Experimental and Predicted Biodegradation Rate Constant(b) and Half-life Time (1/2) of Degradation for 9 Phthalates

9 phthalates were calculated with DFT in Gaus- sian 09 program at the B3LYP/6-311G** level to obtain their structural and thermodynamic parame- ters. The structural parameters include dipole mo- ment (), energy of the highest occupied molecular orbital (HOMO), energy of the lowest unoccupied orbital (LUMO), the most negative net atomic charge of molecule (-), net charge of the most positive hydrogen atom (H+), molecular volume (i) and mean molecular polarization (). The thermody- namic parameters include total energy (), zero- point vibrational energy (), enthalpy (), free energy (), corrected thermal energyth(i.e., the total of vibrational energy, rotational energy and translational energy of a molecule), heat capacity at constant volume (C°) and entropy (°). The calcu- lation of vibrational frequency demonstrated that all compounds obtained were with the least energy and no imaginary frequency existed. While calculating the molecular volumei(?3), keyword “Volume” was adopted to indicate how to calculate it by defining it in the range of 0.001 electron/bohr3, and option “Tight” was chosen to enhance the precision of integration. Table 2 shows the structural para- meters calculated at the B3LYP/6-311G** level which affected the biodegradability of phthalates remarkably.

Table 2. Structural Descriptors Calculated by B3LYP/6-311G** Affecting the Biodegradability of Phthalates Remarkably

In this paper, authors adopted GQSARF2.0 pro- gram and structural and thermodynamic parameters as arguments of 9 phthalates to make regressive analyses so as to obtain equations associated with their biodegradation rate constants and half-life time, and their qualities were measured with correlation coefficient (2), standard error (), cross corre- lation coefficient (2) andvalue.

CoMFA analysis was completed by SYBYL 7.3 software package, Tripos standard field was adopted, the threshold of electrostatic field energy and steric field energy was set to 30 kcal·mol–1(1 cal = 4.18 J), and3-hybridized C+ was used as probe to cal- culate values and distributions of the energy of electrostatic field and steric field on the peripheral grids of aligned molecules, detecting points being set at intervals of 2.0 ? and all other values being defaults. The structures of all molecules were optimized based on their energies using Tripos standard molecular force field, the criteria for energy convergence was 0.05 kcal·mol–1·?–1, and the net charge of atoms in a molecule was calculated using Gasteriger-Hückel method to get the molecular configuration of the lowest energy. Dimethyl phtha- late with the highest biodegradation rate was used as template for all phthalates. The skeleton used for molecular alignment is shown in Fig. 2 (hydrogen atoms were removed) and the aligned molecules are shown in Fig. 3.

Fig. 1. Basic structure for 9 phthalates

Fig. 2. Skeleton used for alignments

Fig. 3. Aligned molecules

Partial least-squares regression (PLS) was used to perform statistical analysis, the best principal com- ponent () and cross-validation correlation coef- ficient (2) were determined through cross validation by leave-one out (LOO) method, and then the result was verified by non-cross validation to construct CoMFA model. The model quality is measured by cross-validation correlation coefficient (2), routine correlation coefficient (2), standard error () and-test values[9].

3 RESULTS AND DISCUSSIONS

3. 1 2D-QSBR study on biodegradation rate and half-life time

3.1.1 2-QSBR model

The best correlation equations and model parame- ters for biodegradation rate and half-life time ob- tained by GQSARF2.0 program are shown in Table 3.

Table 3. QSBR Equations for Biodegradation Rate and Half-life Time with Structural and Thermodynamic Parameters as Descriptorsa)

a)2,,, and2represent respectively the correlation coefficient, standard error,statistical value and cross-validation correlation coefficient of those equations.

From Table 3 it may be seen thati/100 will appear in the equation firstly when only a variable is taken, demonstrating that a certain linear relation- ship does exist betweenbandi/100 (2= 0.722), and the biodegradation rate of compounds will decrease along with the increase ofi/100;i/100andH+will appear in equation when 2 variables are taken. Meanwhile, the correlation coefficient of the equation will be increased remarkably and the standard error will be decreased noticeably. The biodegradation rate decreases along with the in- crease ofi/100 of compounds, showing that the increase of the spatial volume of a compound will hinder its entering organism, which is disadvanta- geous for biodegradation; the biodegradation rate will increase along with the increase ofH+. Fur- thermore, equation 3 shows that the half-life time (1/2) of biodegradation will increase along with the increase ofi/100 and decrease along with the decrease ofH+.

3.1.2 Validation of the models

Stability and predicting ability of constructed QSBR models should be verified to avoid the colinearity among variables in a model. Often, the degree of correlation among all variables in a model is assessed with the variance inflation factor ()[22]defined as= 1/(1–2), whererepresents the multiple regression correlation coefficient between a variable and the others in the equation.= 1.0 means no autocorrelation existing among the variables;= 1.0~5.0 shows that the correlation equation may be acceptable; if> 10, this regres- sion equation is unstable and therefore needs repea- ted validation. For models at the B3LYP/6-311G** level, the correlation coefficients2amongindepen- dent variables in equation (2) are 0.691 and 0.691 respectively, andare 3.236 and 3.236 (as shown in Table 4), indicating equation (2) has great statistical significance.

Table 4. Correlation Coefficient (r2), Variance Inflation Factors (VIF), Standard Regression Coefficients (SR) and T-scores for Eq. (2)

3.1.3 Validation of the predicting ability

Cross-validation correlation coefficient (2) cha- racterizes the predicting ability and stability of the constructed models. Generally speaking,2> 0.5 indicates a higher predicting ability of a model[19], and higher2means higher predicting ability. The cross-validation correlation coefficients for the equation containing 2 variables listed in Table 3 are 0.6958 and 0.8490 respectively, indicating that equations (2) and (3) possess higher predicting ability. Data about the biodegradation of various compounds predicted by equations (2) and (3) are shown in Table 1. Meanwhile, the errors of the predicted values are shown. From Table 1 we can see that the prediction error for dimethyl phthalate is less than 1%, and that for diundecyl phthalate is relatively remarkable.

3.2 Analysis of the result of CoMFA (3D-QSBR models)

As Fig. 3 shows, all molecules may be aligned well despite the long chain of phthalates. Statistical parameters and contribution values of various fields for CoMFA models are listed in Table 5.2, SE andvalues demonstrate higher stability of those models.2> 0.5 shows stronger predicting ability of the models. Biodegradation rate constants and data about the half-life time predicted by CoMFA for those compounds, as well as the prediction errors are listed in Table 1. From Table 1 it may be seen that the error of predicted biodegradation rate constant is greater (>25%) for dioctyl phthalate, dinonyl phthalate and diundecyl phthalate, showing that those models are more suitable for analysis and prediction of phthalates with shorter chain. CoMFA shows a strong ability in the prediction of half-life time, as proven by that the error of the predicted half-life time for all phthalates is less than 1%. Statistical parameters demonstrate that stability and predicting ability of CoMFA models are higher remarkably than those of 2models. The con- tribution values of each field demonstrate that the steric field has stronger effect as compared with the electrostatic field.

Fig. 4 is the 3contour map of each field for CoMFA models, and this contour map reflects clearly the effect of various fields around a molecule on the biodegradability. Dimethyl phthalate with the highest biodegradation rate was taken as reference, and different color represents the effect of energy of different fields. Fig. 4(a) shows the distribution of steric field for the biodegradation rate constant, and the green area in Fig. 4(a) indicates that the volume of substituent group is smaller around here. The biodegradation rate constant of the compound is bigger.Fig. 4(b) shows the distribution of electros- tatic field for biodegradation rate constant, and the blue area in Fig. 4(b) indicates that if the group introduced here has stronger electropositivity, the biodegradation rate constant will be bigger; and the red area shows that the stronger the electronegativity of substituent groups introduced here is, the bigger the biodegradation rate constant will be. As the aliphatic group is electronegative and the electro- negativity decreases with longer aliphaticchain, the fact shown in Fig. 4(b) accords with the experi- mental data. Electronegative groups attract electron and combine with degeneration enzyme in the biological body to make the chemical degrade more easily. Fig. 4(c) shows the distribution of steric field for the half-life time of biodegradation, and, simi- larly, the green area in this figure suggests that the greater volume of substituent groups around here, the longer half-life time of biodegradation; and the yellow area indicates that the smaller the volume of substituent groups introduced around here is, the longer the half-life time of biodegradation will be. The experimental data also reflect this phenomenon. The reason can be that it is more difficult for great molecules to pass through the biological membrane and degrade. Fig. 4(d) shows the distribution of electrostatic field for half-life time of biodegradation, and the blue area in Fig. 4(d) indicates that the stronger electropositivity of substituent groups introduced here, the longer half-life time of biodegradation; and the red area shows that the stronger the electronegativity of substituent groups introduced here is, the longer the half-life time of biodegradation of the compound will be. As steric field plays a more important role in the half-life time (shown in Table 5), Fig. 4(c) reflects the factors affecting the biodegradation to a greater extent. Green near position R1 or R2 (Fig. 1) means small volume of substituent groups around here benefits the biodegradation here. The fact and interpretation are in accordance with that on the biodegradation rate constant.

Fig. 4. Contour map of CoMFA models

(a) Steric field forb; (b) electrostatic field forb; (c) steric field for1/2; (d) electrostatic field for1/2

Table 5. Statistical Parameters of CoMFA Models

Note:

4 CONCLUSION

Parameters of quantum chemistry for 9 phthalates were calculated using DFT at the B3LYP/6-311G** level for constructing 2-QSBR models to predict the biodegradation rate constant (b) and half-life time (1/2) of biodegradation of those compounds.iandH+appeared in the regression equations, showing that they were highly correlated to the biodegradability with less standard error. Further- more, CoMFA analysis on the biodegradation rate constant and half-life time of biodegradation demonstrated that the characters of steric field of substituent groups affect remarkably the biode- gradability of phthalates. All models constructed using the above two methods possess high stability and predicting ability. In the above two kinds of models, the common factor that affects the biode- gradability of phthalates was the molecular volume. The 3-QSBR models were superior remarkably to the 2-QSBR models in terms of the stability and predicting ability.

(1) Yang, C. F.; Feng, C. J. Investigation into the biodegradability of phthalic acid ester compounds with molecular connectivity indices and electrotopological states indices.. 2007, 29, 36–39.

(2) Staples, C. A.; Peterson, D. R.; Parkerton, T. F. The environmental fate of phthalate esters – a literature review.1997, 35, 667–749.

(3) Xia, F. Y.; Zheng, P.; Zhou, Q. Aerobic die-away degradation of nine phthalic acid esters (M).. Beijing, China, Oct. 2001, 8–12.

(4) Chen, Y. J.; Wang, Z. Y.; Mao, L.; Gao, S. X. QSBR study on the biodegradation rate constant of chloro-phenol compounds.2010, 29, 895–899.

(5) Han, X. Y.; Wang, Z. Y.; Yang, C. S. Quantitative correlation of the acute toxicity of phenylthio-carboxylates with their structural and thermodynamic parameters by DFT calculation.. 2005, 24, 145–150.

(6) Zheng, Q.; Wang, L. S. Studies on the quantitative structure-activity relationship of toxicity of chlorophenol serial compounds in themethods and substitutive position of chlorine atom (NPCS).2007, 8, 933–938.

(7) Eroglu, E.; Türkmen, H. A DFT-based quantum theoretic QSAR study of aromatic and heterocyclic sulfonamides as carbonic anhydrase inhibitors against isozyme CA-II.2007, 26, 701–708.

(8) Mo, L. Y.; Liu, H. Y.; Yi, Z. S.; Li, Y. H.; Dou, R. N. QSTR study on the freshwater photobacteria toxicity of substituted benzenes.. 2011, 10, 1403–1411.

(9) Li, X. J.; Shan, G.; Liu, H.; Wang, Z. Y. Determination of lgKow and QSPR study on some fluorobenzene derivatives.. 2009, 10, 1236–1241.

(10) Yang, D. S.; Dai, Y. Z.; Li, J. H.; Zhu, F. QSBR study on the anaerobic biodegradation of chlorophenols.. 2006, 10, 1183–1188.

(11) Pasha, F. A.; Srivastava, H. K.; Singh, P. P. Comparative QSAR study of phenol derivatives with the help of density functional theory.2005, 13, 6823–6829.

(12) Xu, S.; Liu, H.; Liu, H. X.; Yang, G. Y.; Wang, Z. Y. DFT study on the thermodynamic properties of poly-bromine carbazole.. 2010, 4, 535–542.

(13) Han, X. Y.; Shi, J. Q.; Zheng, Q.; Zhang, Y. G.; Zhang, L. 2D- and 3D-QSBR study on the relationship between the structure and biodegradation property of sulfonylurea herbicides.. 2012, 12, 1693–1700.

(14) Wang, C. H.; Jiang, M.; Li, X. L.; Shen, X. Q.; Yu, H. X.; Wu, Y. 2D-QSAR using MLR and 3D-QSAR using CoMSIA studies on the toxicity of aromatic hydrocarbons on Larval Sinonvaculina onstricta.. 2012, 31, 420–428.

(15) Zhao, H. M.; Zhang, C. Y.; Ge, Z. G.; Wang, Z. Y. Toxicity (–lg50) measurement of the mluorobenzene derivants against vibrio qinghaiensis (Q67) and their 2D, 3D-QSAR study.. 2010, 29, 1467–1476.

(16) Waller, C. L. A comparative QSAR study using CoMFA, HQSAR, and FRED/SKEYS paradigms for estrogen receptor binding affinities of structurally diverse compounds.. 2004, 44, 758–765.

(17) Ge, Z. G.; Sun, P.; Liu, H.; Tan, J.; Liu, H. X. Toxicity measurement of halogeno-benzene against vibrio qinghaiensis (Q67) and their 2D,3D-QSAR study.. 2011, 30, 630–637.

(18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A. Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A.09,A.9, Gaussian, Inc.: Wallingford CT 2009.

(19) Liu, S. S.; Liu, H. Y.; Yin, C. S. VSMP: a novel variable selection and modeling method based on the prediction.. 2003, 43, 964–969.

(20) Cramer, R. D.; Patterson, D. E.; Bunce, J. D. Comparative molecular field analysis (CoMFA). 1. Effects of shape on binding of steroids to carrier proteins.. 1988, 110, 5959–5967.

(21) Xia, F. Y.; Zheng, P.; Zhou, Q.; Feng, X. S. Relationship between quantitative structure and biodegradability for phthalic acid ester compounds.(Agric. & Life Sci.) 2004, 30, 141–146.

(22) Famini, G. R.; Carl, A. P.; Wilson, L. Y. Using theoretical descriptors in quantitative structure-activity relationships: some physicochemical properties.1992, 5, 395–408.

16 January 2014;

17 April 2014

. Female, associate professor. E-mail: hxy16_2000@163.com