QSAR and Docking Studies of Thiazolidine-4-carboxylic Acid Derivatives as Neuram

TONG Jian-Bo WANG Tian-Hao WU Ying-Ji FENG Yi

QSAR and Docking Studies of Thiazolidine-4-carboxylic Acid Derivatives as Neuraminidase Inhibitors①

TONG Jian-Boa, b②WANG Tian-Haoa, bWU Ying-Jia, bFENG Yia, b

a(710021)b(710021)

In order to understand the chemical-biological interactions governing their activi- ties toward neuraminidase (NA), QSAR models of 28 thiazolidine-4-carboxylic acid derivatives with inhibitory influenzavirus were developed. Here a quantitative structure activity relationship (QSAR) model was built by three-dimensional holographic atomic vector field (3-HoVAIF) and multiple linear regression (MLR). The estimation stability and prediction ability of the model were strictly analyzed by both internal and external validations. The correlation coefficient (2) of established MLR model was 0.984, and the cross-validated correlation coefficient (2) of MLR model was 0.947. Furthermore, the cross-validated correlation coefficient for the test set (Q2) was 0.967. The bindingmode pattern of the compounds to the binding site of integrase enzyme was confirmed by docking studies. The results of present study indicated that this model can aid in designing more potent neuraminidase inhibitors.

3-QSAR, thiazolidine-4-carboxylic acid derivatives, 3-HoVAIF, molecular designing, molecular docking;

1 INTRODUCTION

Neuraminidase (NA) inhibitors are the only class of antiviral drugscurrently approved for use against the influenza virus[1]. Influenza often occurs in winter seasons and occasionallycauses pandemics due to the emergence of virus mutants and unex- pected transmission to humans. The influenza virus, avian influenza(H7N9) virus in particular,is a great threat to human health potentially causing serious publichealth and economic problems[2]. Due to the key role in replication, spread, infection and pathogenesis of influenza virus, the application of neuraminidase inhibitors (NAIs) has been considered as a dominant approach for the treatment of influenza infection[3], because the active site of NA is highly conserved acrossthe influenzaandviral strains[4].This inspired us to explore novel pyrimidine deriva- tives asNA inhibitors. Moreover, QSAR[5]is widely used in drug design.

The availability of computational techniques on quantitative structure activity relationship (QSAR) might provide a potential direction for accelerating the drug design process. In fact, QSAR can be viewed as a technique attempting to summarize chemical and biological information in a form that allows one to generate relationships between che- mical structure and biological activity[6]. As is well known, the success of a QSAR study depends also on the selection of variables (molecular descriptors) and on the representation of the information. Variables should give the maximum of information in active variables. 3-QSAR model would better reflect the interaction between the ligand and receptor com- pared to 2-QSAR. Three-dimensional holographic atomic vector field (3-HoVAIF) is a widely used method for 3-QSAR. In this paper, QSAR models of 28 thiazolidine-4-carboxylic acid derivatives as novel influenza NAIs were constructed to understand the chemical-biological interactions governing their activities towardinfluenza NA[7]. This article by means of 3-HoVAIF and MLR studied the 3-QSAR models of molecular drugs, and then using molecular docking studied the combination of the active site between drugs and neuraminidase.

2 PRINCIPLE AND METHODOLOGY

In 3-QSAR, 3molecular fields were used as molecular descriptors. Assuming that the compounds considered in some applications bind to the target in marginally same way, the value of a 3molecular field at every point in space was used as a descriptor. If then one assumes that the variation in biological activity can be related to the change in these 3molecular field values between the molecules, a 3-QSAR model can be obtained.

3-QSAR models are relatively simply interpre- table through visualization where the important regions that correspond to descriptors retained in the QSAR are located. Once it is known in what regions these resides and what effect they have on QSAR, the ligand candidates to maximally exploit these regions and hence increase the (predicted) activity of the compounds were manipulated by drug designers. The first available 3-QSAR analysis was carried out by 3-HoVAIF which was proposed as a three-dimensional molecular structure description and multiple drug system has been applied[8-11].

3-HoVAIF was generated considering three common non-bonding interactions of the biological activities,, electrostatic, steric,and hydrophobic interaction related with atomic relative distance and atomic self-properties. Thesedescriptors neither resort to any experimental parameters nor consider configuration overlap of samples. Ordinary atoms of organic molecules includingH, C, N, P, O, S, F, Cl, Br and I are classified intofive types in the Periodic Table ofElements. According to the hybridization state of atoms, theatoms are furthermore subdivided into ten types. Thus, there are 55 interactions in a molecule[12]. In this paper, electrostatic, steric and hydrophobic potential energies takepart in the representation of different interactions, producing 3 × 55 = 165 interaction items for organic molecules.

Its electrostatic interactions can be used in the classical Coulomb theorem (type (1)) to describe;

Stereoscopic effect of the Lennard-Jones equation to describe the effect of the type (type (2));

Whereεrepresents the potential well[13, 14],is 0.01[14], representing the calibration constant of interatomic interaction deduced byexperience; R* =(C·R*+ C·R*)/2 is Van derwaals radius with its calibration factor of 1.00 in thecase of3, 0.95 in2and 0.90 inhybridization[15].

Hydrophobic action uses Hint of Kellogg and method to define the hydrophobic effect between the two atoms (type (3));

Whereis the sign function, indicating the entropychange resulting from different types of ato- micinteractions[16-20];representsthe solvent accessible surface area of atom (SASA),., surface area formed by a hydrateprobe spherecally rolling at the surface of this atom[21];is the hydrophobic constants[22].

Molecular docking was performed by AutoDock4.2 software package[23]. Using simulated annealing and genetic algorithm to find the best junction of close position of the receptor and the ligand, by computing free energy of the semi-empirical method, we evaluatethe matching in receptor and the ligand, and using functional form (type (4))[24]to calculate.

Where ΔG, ΔG, ΔG, ΔGand ΔGare the semi-empirical parameters obtained by fitting.

3 RESULTS AND DISCUSSION

3.1 Selection and partition of the data

28 Thiazolidine-4-carboxylic acid derivatives withinhibitory influenzavirus were obtained from refe- rences[25].50values were measured spectrofluoro- metrically using2΄-(4-methylumbelliferyl)-a-D-ace- tylneuraminic acid (MUNANA) assubstratefor neuraminidase to yield a fluorescent productwhich was quantified[25]. From structurally diversemolecules possessing activities of a wide range, 28NIs were divided into the training set with 22 samples and the test set containing 6 samples. The common structure is shown in Fig. 1, and the structures and predicted activity values of 28 thiazolidine-4-carboxylic acid derivatives are listed in Table 1.

Fig. 1. Skeleton of 28 thiazolidine-4-carboxylic acid derivatives

Table 1. Structures and Predicted Activities of 28 Thiazolidine-4-carboxylic Acid Derivatives

*Samples in the test set

Molecular steric structures were firstly constructed byChemoffice 8.0, and then optimized at the AM1level by MOPAC half-experience quantum chemistry software in Chem3. Then net electric charge ofatoms was calculated in single-point form byMulliken methods. After the above two items wereinput respectively into forms of Descartes coor- dinates and net electric charge amounts, 3-HoVAIFdescriptors were produced by applying 3-HoVAIF.EXE is an applied program written inlanguage. The ultimate vectors for 28 thiazolidine-4-carboxylic acidderivatives involve 105 non-zeroitems. The interpreted unification of the 3-HoVAIFmethod may lead to some information overlapamong these different descriptors. To address theabovementioned problems, stepwise multiple regression (SMR) method was employed which used the SPSS 13.0 software to selectvariables; Multiple linear regression (MLR) and partial least squares (PLS)ways were applied to construct the modelaccording to the values of Fisher prominent test bySMR. Theobtained original variable matrix by SMR was thensubjected to a PLS regression modeling or MLR regression modelingand theoptimal model was determined when cross-validation correlative coefficients (CV2) in leave-one-outcross-validation (LOO-CV) achieved the maximumvalue, and the relative statistics to these modelswere presented. QSAR models were validated by leave-one-outcross-validation (LOO-CV) and external predictionby test setR2andQ2[26]. Meanwhile, the modelswere validated by different modeling methods.

3.2 Model building

In this study, two models, PLS and MLR, were obtained. PLS was widely used in analytical, physical, andmedicinal chemistry as a data analyzing method. Ithas many advantages over ordinary MLR. For instance, it can avoid harmful effects in modeling dueto multi-collinearity, and it particularly fits forregressing when the number of variables is lessthan that of samples,. In QSAR studies, variables, as few as possible, should be included in amodel. Not all the structural descriptors relevant tobiological activities will be easily interpreted, sothose redundant descriptors should be eliminated inorder to promote its robustness and predictivecapability especially when the number of variables isvery large. The correlation coefficients (2) ofestablished PLS model was 0.964, and the cross-validated correlation coefficients (2) of PLS model was 0.935. Furthermore, the correlation coefficient for the testset (R2) was 0.911 and the cross-validated correlation coefficient for the test set (Q2) was 0.644, respectively.

However, PLS model has so many advantages but MLR model has also many traits. Linear Regression Multiple (MLR) is a more classic modeling method, which can optimize the activity of the lead compounds. The assumption of the MLR method is that when the molecular structure changes, the biological activity of the molecules will change, and the fundamental cause of this change is closely related to the physical parameters. In addition, another obvious advantage of it is that it can obtain a causal model, and has a clear physical meaning. In this paper, the results of MLR were better than those of PLS. The results obtained by MLR were as follows:

p50= 2.8450.000×3-5-0.841×1-8-0.364×

1-3-0.002×2–5+0.000×1–5+0.000×

5–5–200.684×2–10

= 21,cum= 0.992,= 0.107,= 119.322,cv2= 0.947,cv= 0.199,cv= 33.138,

1= –7.903,2= –9.618,3= –3.235,4= 3.397,

5= 1.960,6= 1.618,7= –0.653

for the training samples,cumas the correlation coefficient,is the standard deviation,istest values,CV2,CV,CVandifor the cross-validated correlation coefficients, standard deviation,test value andtest value of each variable.

In the model,3-5is hydrophobic function between C of2hybridization and N of3hybridization,1–8is electrostatic function between H and O of3hybridization,1-3is electrostatic function between H and C of2hybridization,2–5is steric function between C of3hybridization and N of3hybridization,1–5is hydrophobic function between H and N of3hybridization,5–5is hydro- phobic function of N of3hybridization,2-10is steric function between C of3hybridization and Cl.

3.3 Model analysis

All the parameters in the model of Table 2 were taken from the Ref. [7], indicating that the obtained model of 3-HoVAIF is superior to other models, so it is robust with good exterior predictive capabilities. Although the method is old and troublesome, the result is good. Hence, for the model, the method of 3-HoVAIF was chosen at last.

Table 2. Statistical Results and Relative Contributions of the Models

2: squared multiple correlation coefficients;: standard error of estimate;2: squared cross-validated correlation coefficient;cv: cross-validated standard error of estimate;: components;: Fisher statistic;2testand2extare

2and2of the test set, respectively.

From the above results, it can be seen that the resultsof PLS areworse than those of MLR, so the way of MLR was obtained topredict the test set. At the same time, the predicted activities of training and test sets compounds were achieved. And then the scatter plot between calculatedand predicted activities of the training and test sets compounds was given in Fig. 2. As shown in Fig. 2, calculatedactivities are in a linear relationship with predicted activities of the samples. It told that all samples were almost uniformly distributedaround the diagonal, and no obviously exceptional pointwas selected.Q2was 0.967, so the results of internal and external inspection showed the model was accurate and stable. Thereby, theobtained QSAR model with goodexterior predictive capabilitywas robust. The adopted QSAR model can be used to designand screen new NIs.

Fig. 2. Scatter plot between the observed and predictedactivity of training and test sets of 28 NAIs

3.4 Drug molecular design

From the above regression equation, the p50increases with the decrease of1–3,1-8,2–5and2-10values, which are to enhance the activities of drug molecules. So the highest activity compound 27 of 28 thiazolidine-4-carboxylic acid derivatives which have been synthesized and evaluated as neurami- nidase inhibitors was chosen for optimization, and then it would get higher activity compounds. In accord with compound 27 as the template for the design of new drug molecules, newly designed molecules and predictive activity are shown in Table 3. By comparing the newly designed compound 1 with compound 27 of the original template molecule, the substitution with -COOH group of the former showed better activity than the substitution with -OH group of the template molecule. On account of=conjugate, the biological activity of compounds increases. By comparing the newly designed com- pound 6 with compound 27 of the original template molecule, the (CH3)2CH- of compound 6 replaced CH3O- of the template molecule, which can increase the biological activity by decreasing the electrostatic interaction.

Table 3. Newly Designed Molecular Structures and Predictive Activity

3.5 Docking study

Docking was performed using AutoDock software packageversion 4.2. 3co-crystallized structure of neuraminidase was taken from the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (PDB ID: 4B7R).Processing of the protein was done by adding hydrogen atomsto assign appropriate ionization states to both acidic and basicamino acid residues and removing the remaining water moleculesfrom protein and adding charges. In this experiment,the No. 27 compound withthe highest activity of template and No. 6 compound with the highestactivity of newly designed compounds were adopted by molecular docking, respectively, docking lattice was40 × 40 × 40, and grid point spacing was 0.375 Angstroms, coordinates of the central grid point of maps for No. 27 template compound was (23.282, 21.870, –42.954),while coordinates of the central grid point of maps for newly designed compound 6 was (40.696, –1.808, –5.945), Lamarckian Genetic AlGotithm (LGA) was used as ligand conformation search process, and the otherparameters were by default.

The docked conformations showed that all ligands bind to theactive residues in the predefined hydro- phobic binding pocket. As can be seen from Fig. 3, the location of the docked ligands agreed well with that ofthe docked reference ligand. For Fig. 3a, the reference ligand formed hydrogen bonding (H···bond) with ASP151, GLU278, TYR402, ARG293 and ARG118 in the binding sites. The docked reference ligand was found to have H···bond interactions: NH of3hybridization of the ligand and Ogroup of ASP151 (hydrogen bond distance 2.1 nm), OH of3hybridization of ligand and O group of GLU278 (hydrogen bond distance 2.1 nm), C=O of2hybridization of ligand and NH groups of ARG118 and ARG368 (hydrogen bond distances2.1 and 1.7 nm, respectively), OHof3hybridization of the ligandand OH group of TYR402 (hydrogen bond distance 1.9 nm) and OHof3hybridization of the ligandand NH group of ARG293 (hydrogen bond distance 2.1 nm).

However, for Fig. 3b, the reference ligand formed hydrogen bonding (H···bond) with GLU277, ARG368 and ARG293 in the binding sites. The docked reference ligand was found to have H···bond interaction between NH of3hybridization of the ligand and Ogroup of GLU277 (hydrogen bond distance 2.3 nm), C=O of2hybridization of ligand and NH2group of ARG368 (hydrogen bond distance 2.3 nm) and OH of3hybridization of ligand and NH2group of ARG293 (hydrogen bond distance 2.3 nm).

Fig. 3. Docking interaction pattern of 4B7R active residues with ligands (a) Compound 27 and 4B7R active residues, (b) New compound 6 and 4B7R active residues

From molecular docking model, in the docking process of ligand and receptor, the formation of hydrogen bonds between them determined the ligands in live. The positions of the cavity have an important role. Hydrogen bonding between drug molecules and biological macromoleculereceptors is a common way of bonding, which can increase the drugactivity and make the combination between drug molecules more stable. The docking results agreed well with the observedbiological activity data, which showed that these docking conformations are desirable to analyze the drug models.

4 CONCLUSION

28 Thiazolidine-4-carboxylic acid derivatives are a kind of anti-neuraminidase drug having great deve- lopmentprospects and good exterior predictivecapability by using 3-HoVAIF way. In this paper,3-QSAR model was set up by means of 3- HoVAIF and MLR, finally achieving good results. It is illustrated that the obtained 3-QSAR is reliable for describing chemical structure and biological activity of the drug molecules, and then molecular docking approach was employed to study the rela- tionship between drug ligand and macromolecular receptor. Ultimately, the docking results indicate that the ligands would form hydrogen bonding interact- tions with GLU, TYR, ARG and ASP of the protein receptor, generally. So, the adopted QSAR model can help to design and screen new compounds to obtain newNAIs with high activities.

(1) Chen, Y. Q.; Wohlbold, T. J.; Zheng, N. Y. Influenza infection in humans induces broadly cross-reactive and protective neuraminidase-reactive antibodies.2018, 173, 417-429.

(2) Wang, X. L.; Jiang, H.; Wu, P. Epidemiology of avian influenzaH7N9 virus in human beings across five epidemics in mainland China, 2013-17: an epidemiological study of laboratory-confirmed case series.2017, 17, 822-832.

(3) Zhao, H. Y.; Chen, Z. L. Screening of neuraminidase inhibitors from traditional Chinese medicines by integrating capillary electrophoresis with immobilized enzyme microreactor.2014, 1340, 139-145.

(4) Anne Moscona, M. D. Neuraminidase inhibitors for influenza.2005, 353, 1363-1373.

(5) Chen, B. Y.; Zhang, T.; Bond, T. Development of quantitative structure activity relationship (QSAR) model for disinfection byproduct (DBP) research: a review of methods and resources.2015, 299, 260-279.

(6) Bhole, R. P.; Bhusari, K. P. Synthesis, antihypertensive activity, and 3-QSAR studies of some new-hydroxybenzohydrazide derivatives.2011, 344, 119-134.

(7) Sun, J. Y.; Wang, J. C.; Mei, H. QSAR and pharmacophore studies of thiazolidine-4-carboxylic acid derivatives as novel influenza neuraminidase inhibitors using HQSAR, Topomer CoMFA and CoMSIA.2013, 32, 744-750.

(8) Van Damme, S.; Bultinck, P. 3QSAR based on conceptual DFT molecular fields: antituberculotic activity.2010, 943, 83-89.

(9) Tong, J. B.; Liu, S. L. Three-dimensional holographic vector of atomic interaction field applied in QSAR of anti-HIV HEPT analogues.2008, 27, 330-337.

(10) Tong, J. B.; Chen, Y.; Liu, S. L. QSAR studies of antituberculosis drug using three-dimensional structure descriptors.2013, 22, 4946-4952.

(11) Tong, J. B.; Che, T.; Li, Y. F. A descriptor of amino acids: SVRG and its application to peptide quantitative structure-activity relationship.2011, 22, 611-620.

(12) Levitt, M. Protein folding by restrained energy minimization and molecular dynamics.1983, 170, 723-764.

(13) Levitt, M.; Perutz, M. F. Aromatic rings act as hydrogen bond acceptors.1988, 201, 751-754.

(14) Hahn, M. Receptor surface models. 1. Definition and construction.1995, 38, 2080-2090.

(15) Kellogg, G. E.; Semus, S. F.; Abraham, D. J. HINT: a new method of empirical hydrophobic field calculation for CoMFA.1991, 5, 545-552.

(16) Wireko, F. C.; Kellogg, G. E.; Abraham, D. J. Allosteric modifiers of hemoglobin. 2. Crystallographically determined binding sites and hydrophobic binding/interaction analysis of novel hemoglobin oxygen effectors.1991, 34, 758-767.

(17) Tong, J. B.; Zhong, L.; Zhao, X. Quantitative structure-activity relationship studies of diarylpyrimidine derivatives as anti-HIV drugs using new three-dimensional structure descriptors.2014, 23, 1634-1642.

(18) Kellogg, G. E.; Abraham, D. J. Key, lock, and locksmith: complementary hydropathic map predictions of drug structure from a known receptor-receptor structure from known drugs.1992, 10, 212-217.

(19) Tong, J. B.; Chen, Y.; Liu, S. L. A novel 3molecular structural characterization method applied in QSAR study.2013, 64, 707-712.

(20) Li, Z. H.; Chen, G.; Chen, Z. T. Three-dimensional holographic vector of atomic interaction field (3-HoVAIF) for the QSPR/QSAR of polychlorinated naphthalenes.2012, 31, 345-352.

(21) Shu, M.; Zhang, Y. R.; Tian, F. F. Molecular docking and 3-QSAR research of biphenyl carboxylic acid MMP-3 inhibitors.2012, 31, 443-451.

(22) Liao, L. M.; Huang, X.; Li, J. F. Structural characterization and acute toxicity simulation for nitroaromatic compounds.2016, 35, 449-456.

(23) Huey, R.; Morris, G. M.; Olson, A. J. A semiempirical free energy force field with charge-based desolvation.2007, 28, 1145-1152.

(24) Liu, Y.; Jing, F. B.; Xu, Y. Y. Design, synthesis and biological activity of thiazolidine-4-carboxylic acid derivatives as novel influenza neuraminidase inhibitors.2011, 19, 2342-2348.

(25) Golbraikh, Alexander; Tropsha, Alexander. Beware of q2!2002, 20, 269-276.

(26) Tong, J. B.; Wang, Y.; Lei, S. 3-QSAR and docking studies of 1,3,4-thiazolidinone derivatives using R-group search and surflex-dock.2019, 38, 464-475.

21 June 2019;

20 November 2019

① This work was supported by the National Natural Science Funds of China (21475081), the Natural Science Foundation of Shaanxi Province (2019JM-237), and the Graduate Innovation Fund of Shaanxi University of Science and Technology

.Professor. E-mail: jianbotong@aliyun.com

10.14102/j.cnki.0254–5861.2011–2508