QSAR and molecular docking studies on 4-quinoline carboxylic acid derivatives as inhibition of vesicular stomatitis virus replication European Journal of Chemistry 10 (1) (2019) 45-51 European Journal of Chemistry View Journal Online View Article Online QSAR and molecular docking studies on 4-quinoline carboxylic acid derivatives as inhibition of vesicular stomatitis virus replication Tawassl Tajelsir Hassan Hajalsiddig * and Ahmed Elsadig Mohammed Saeed Department of Chemistry, Collage of Science, Sudan University of Science and Technology, Khartoum, 407, Sudan toooota7@hotmail.com (T.T.H.H.), aemsaeed@gmail.com (A.E.M.S.) * Corresponding author at: Department of Chemistry, Collage of Science, Sudan University of Science and Technology, Khartoum, 407, Sudan. Tel: +249.183.769363 Fax: +249.183.774559 e-mail: toooota7@hotmail.com (T.T.H. Hajalsiddig). 10.5155/eurjchem.10.1.45-51.1795 Received: 04 October 2018 Received in revised form: 10 November 2018 Accepted: 12 November 2018 Published online: 31 March 2019 Printed: 31 March 2019 The current study describes the development of in silico models based on quantitative structure-activity relationship (QSAR) analysis has been performed on 4-quinoline carboxylic acid derivatives as inhibition capacity of vesicular stomatitis virus replication in Madin Darby canine kidney epithelial cells. A highly descriptive and predictive QSAR model was obtained through the calculation of alignment-independent descriptors using MOE 2009.10 software. For a training set of 20 compounds, the partial least squares analyses result in a model which displays a squared correlation coefficient (r2) of 0.913. Validation of this model was performed using leave-one-out (q2) of 0.842. This model gives (r2pre) of 0.889 for a test set of five compounds. Docking studies were performed for 25 compounds to investigate the mode of interaction between 4-quinoline carboxylic acid derivatives and the active site of the human dihydroorotate dehydrogenase. PLS QSAR Docking study Vesicular stomatitis virus 4-Quinoline carboxylic acid Dihydroorotate dehydrogenase Cite this: Eur. J. Chem. 2019, 10(1), 45-51 Journal website: www.eurjchem.com 1. Introduction Vesicular stomatitis virus (VSV) is an enveloped, non- segmented, negative-stranded RNA virus and the prototype member of the family Rhabdoviridae, genus Vesiculovirus [1,2]. Clinical disease presents as severe vesiculation and/or ulceration of the tongue, oral tissues, feet, and teats, and results in substantial loss of productivity. Except for its appearance in horses, it is clinically indistinguishable from foot-and-mouth disease. Unlike foot-and-mouth disease, it is very infectious for man and can cause a temporarily debili- tating disease [3]. The VSV and its membrane glycoprotein G (VSVG) are often used as models to study endocytosis and secretory traffic. For the same reasons, VSVG is often used for pseudo typing of retroviral vectors for gene delivery [4]. VSV is an oncolytic virus currently being investigated as a promising tool to treat cancer because of its ability to selectively replicate in cancer cells [5]. In the following, we have focused on dihydroorotate dehydrogenase (DHODH), the fourth enzyme in the de novo pyrimidine nucleosides biosynthetic pathway [6-8]. Pyrimi- dine nucleotides play a critical role in cellular metabolism serving as activated precursors of RNA and DNA, CDP diacylglycerol phosphoglyceride for the assembly of cell membranes and UDP-sugars for protein glycosylation and glycogen synthesis [9]. Therefore, DHODH considered as a key enzyme in biosynthesis pathway in most prokaryotic and eukaryotic cells [10]. The therapeutic potential of inhibiting de novo pyrimidine biosynthesis at the dihydroorotate dehydro- genase catalyzed step was revealed by the antiproliferative agents’ leflunomide and brequinar (6-fluoro-2-(2'-fluoro-[1,1'- biphenyl]-4-yl)-3-methylquinoline-4-carboxylic acid) [11]. Recently, Das et al. reported the 4-quinoline carboxylic acid derivatives as antiviral activity and also tested the ability of these compounds to inhibit in vitro VSV replication in Madin Darby canine kidney (MDCK) epithelial cell [12]. In this work, we collected brief group of 4-quinoline carboxylic acid derivatives with biological activity to QSAR study to obtain model, which was used to predict the biological against VSV replication as human DHODH inhibitor and disclosed to docking these derivatives in the target enzyme for the antiviral activity to be human DHODH. 2. Experimental 2.1. QSAR study ABSTRACT RESEARCH ARTICLE KEYWORDS European Journal of Chemistry ISSN 2153-2249 (Print) / ISSN 2153-2257 (Online) – Copyright © 2019 The Authors – Atlanta Publishing House LLC – Printed in the USA. This work is published and licensed by Atlanta Publishing House LLC – CC BY NC – Some Rights Reserved. http://dx.doi.org/10.5155/eurjchem.10.1.45-51.1795 http://dx.doi.org/10.5155/eurjchem.10.1.45-51.1795 https://crossmark.crossref.org/dialog/?doi=10.5155/eurjchem.10.1.45-51.1795&domain=pdf&date_stamp=2019-03-31 http://www.eurjchem.com/ http://dx.doi.org/10.5155/eurjchem.10.1.45-51.1795 mailto:toooota7@hotmail.com mailto:aemsaeed@gmail.com mailto:toooota7@hotmail.com http://www.eurjchem.com/ https://crossmark.crossref.org/dialog/?doi=10.5155/eurjchem.10.1.45-51.1795&domain=pdf&date_stamp=2019-03-31� 46 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 Table 1. Experimental EC50, experimental pEC50, predicted pEC50 and residual values of quinoline derivatives of 25 compounds used in training and test sets for inhibit in vitro VSV replication in MDCK epithelial cells *. N O OH R OR1 R2 R3 R4 No Compound R R1 R2 R3 R4 EC50 (μM) (Exp.) pEC50 (Exp.) pEC50, (Predicted) Residual 1 Brequinar F See Figure 1 0.3 6.52 6.61 -0.09 2 C1 Cl (CH2)2CH3 H H H 4.7 5.33 5.52 -0.19 3 C2 Cl CH3 H H H 7.1 5.15 4.99 0.16 4 C3 Cl CH2CH3 H H H 5.7 5.24 4.97 0.27 5 C4 T Cl (CH2)3CH3 H H H 6.3 5.20 5.49 -0.29 6 C11 T Cl Ph H H H 1.3 5.89 7.18 -1.29 7 C12 F Ph H H H 0.1 7.00 6.55 0.45 8 C16 NO2 Ph H H H 2.0 5.70 5.41 0.29 9 C18 H Ph H H H 1.0 6.00 6.50 -0.50 10 C22 F CH3 H H H 6.4 5.19 5.05 0.14 11 C24 F (CH2)2CH3 H H H 19.0 4.72 5.25 -0.53 12 C26 T F CH2(C2H5) H H H 6.8 5.16 5.17 0.01 13 C29 F Ph-4-NO2 H H H 4.9 5.31 5.53 -0.22 14 C30 F Ph-3,4-(OCH3O) H H H 0.9 6.04 6.23 0.19 15 C31 F Ph-2-F H H H 0.3 6.52 5.94 0.58 16 C32 F Ph-3-F H H H 0.1 7.00 7.04 -0.04 17 C33 T F Ph-4-F H H H 1.0 6.00 6.81 -0.81 18 C34 F Ph-2-pyridyl H H H 22.9 4.64 4.64 0.00 19 C35 F Ph-3-pyridyl H H H 2.5 5.60 5.73 -0.13 20 C36 F Ph-2-thiazolyl H H H 14.6 4.84 4.91 -0.07 21 C39 F CH2-Ph H H H 0.232 6.69 6.56 0.13 22 C40 T F 3,5-Dimethylphenyl H H H 0.522 6.28 7.06 -0.78 23 C42 F Ph-3-C(CH)3 H H H 0.062 7.18 7.66 -0.48 24 C43 F Ph CH3 CH3 H 0.023 7.63 7.73 -0.10 25 C44 F Ph CH(CH3)2 H CH3 0.002 8.69 8.18 0.51 * T = Test set. N O OH R OR1 R2 R3 R4 Brequinar N O OH F CH3 F Figure 1. Basic structure 4-quinoline carboxylic acids derivatives and brequinar. 2.1.1. Data set The set of selected compounds reported by Das et al. [12] was used to QSAR study. Only 25 compounds have were selected from three combine set according to which that compounds have F, Br, and H atom in position R, and also phenyl, alkyl and heterocycle group in position R1. Structure of compound with substitution of R and R1 position and their biological activity as inhibit in vitro VSV replication in MDCK epithelial cells were reported (Figure 1). These compounds were evaluated for their ability to inhibit VSV replication in MDCK epithelial cells in terms of half maximal effective concentration (EC50) values. For the purpose of modeling study, all 25 derivatives have been divided into training and test sets. Out of the 25 derivatives fifth compounds have been placed in the test set for the validation of derived models. The biological activities of 25 compounds transformation to pEC50, see Table 1. 2.1.2. Theoretical molecular descriptors The compounds of training and test set were first drawn using ACD/Lab (Copyright 1994-2010, ACD/Labs 12.00, product Version 12.01, Build 36726, 26Feb 2010) freeware and saved in “mol” file format. Then, the saved file was opened by MOE 2009.10 were minimized energy, the different molecular 25 descriptors were calculated and decrease the redundancy existing in the descriptors data matrix, the correlations of descriptors with each other and with pEC50 of the molecules are examined, and collinear descriptors (r < 0.9) are detected. Those descriptors which have the pairwise correlation coefficient above 0.9 and having the lower correlation with pEC50 values are removed from the data matrix [13]. Eight descriptors were left in clouding Log octanol/water partition coefficient (Log Po/w), molar refractivity (mr), heat of formation (AM1-HF), dipol moment (AM1-Dipol), total polar surface area (ASA-P), atomic connectivity index order zero (Chi0), mass density (Density) and ionization potential (AM1- IP). The values of descriptors were listed in Table 2. 2.1.3. Model development The QSAR models were constructed based on the partial least square method using to descriptors in MOE 2009.10 software. QSAR was built using the descriptor as an independent variable and EC50 as a dependent variable by forward stepwise regression analyses. QSAR equations were acquired according to different combinations of various descriptors. The data matrix was analyzed using the partial least squares (PLS) method. 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 47 Table 2. The values of molecular descriptor of training and test sets compounds *. No Compound AM1-Dipole AM1-HF AM1-IP ASA-P Chi0 mr Log P(o/w) Density 1 Brequinar 4.82 -86.57 8.95 76.12 19.84 10.58 7.40 0.76 2 C1 3.39 -63.08 8.90 113.38 17.10 9.50 5.28 0.78 3 C2 2.96 -50.81 8.93 142.51 15.69 8.55 4.33 0.80 4 C3 3.25 -56.28 8.89 115.68 16.40 9.02 4.67 0.79 5 C4 T 2.30 -10.72 8.97 104.69 18.80 10.57 5.98 0.78 6 C11 T 2.18 -48.26 8.94 102.78 18.80 10.14 5.54 0.77 7 C12 5.37 1.97 9.26 187.70 20.38 10.59 5.32 0.79 8 C16 2.45 -3.20 8.93 104.83 17.93 10.07 5.35 0.73 9 C18 1.56 -88.06 8.93 142.71 15.69 8.12 3.89 0.78 10 C22 2.59 -99.27 8.85 113.75 17.10 9.07 4.84 0.76 11 C24 3.90 -72.52 8.89 159.81 17.97 9.45 5.01 0.76 12 C26 T 4.79 -43.76 9.35 188.86 21.25 10.66 5.48 0.82 13 C29 2.14 -103.4 8.90 156.98 20.66 10.72 5.25 0.80 14 C30 3.00 -91.77 9.03 104.41 19.67 10.18 5.69 0.80 15 C31 1.43 -93.74 9.08 102.43 19.67 10.17 5.73 0.80 16 C32 1.37 -90.91 9.02 104.58 19.67 10.17 5.69 0.80 17 C33 T 3.79 -26.18 9.20 125.27 18.80 9.98 4.71 0.78 18 C34 1.54 -38.35 9.10 123.41 18.80 9.98 4.31 0.78 19 C35 2.80 21.34 9.17 135.70 18.10 9.84 4.52 0.83 20 C36 2.43 -59.56 8.87 117.75 19.51 10.62 5.68 0.76 21 C39 2.59 -63.06 8.93 102.59 20.54 11.04 6.25 0.75 22 C40 T 1.74 -28.34 8.74 104.74 21.37 11.75 6.76 0.76 23 C42 2.99 -67.37 8.88 106.26 22.17 11.94 7.04 0.73 24 C43 1.35 -60.40 9.11 104.60 20.54 11.05 6.21 0.75 25 C44 1.94 -70.03 8.92 101.77 22.12 11.97 7.05 0.73 * T = Test set. The quality of each regression model was evaluated using a squared correlation coefficient (r2 > 0.7) and root mean square error (RMSE) [14]. Finally, to exclude false or artificial correlations, the external consistency of the variables of the model have been addressed in terms of cross-validated r2 or q2 (>0.5) criteria from the leave-one-out (LOO) cross-validation procedure as default option. The coefficient r2 indicated how well the equation fits the data. The q2 was considered as an indicator of the predictive performance and stability of a QSAR mode [15]. About ten QSAR models were generated by using partial least square regression method coupled with stepwise forward backward method. Among the various models two significant QSAR models were finally selected. Models summary of best models are given below. 2.1.4. Validation model Validation is a crucial and important aspect for the determination of the reliability of models [16]. In this study, the data set is divided into training set for model development and test set for external prediction. Goodness of fit of the models was assessed by examining the multiple correlation coefficient r2, the standard deviation (s), the F-ratio between the variances of calculated and observed activities (F) [15,17]. 2.2. Docking study Twenty five ligands of 4-quinoline carboxylic acid derivatives “mol” file format were opened in Molecular Operating Environment (MOE 2009.10). The 3-D protonated structures of compound were energy minimized. Then, 25 compounds have been compute conformational and saved in a molecular database (mdb) file for further studies. The crystal structure of the complex of human dihydro- orotate dehydrogenase protein (PDB code: 1D3G) was retrieved from a protein data bank. The pdb file was imported to MOE suite where receptor preparation module was used to prepare the protein. All the bound water molecules were removed from the complex. 3D protonation and the active site identification were done. MOE Docking Simulation Program was used to perform the total of ten independent docking runs. The docked poses were inspected and the top scored pose for each compound was reserved for further studies of interaction evaluation. The ligand-protein interactions were visualized in 2-dimensional space by making use the MOE Ligand Interactions Program [17]. 3. Results and discussion 3.1. QSAR results As the number of 4-quinoline carboxylic acid derivatives in the training set was 20, it was important to reduce the number of descriptors until the ratio was ≦ 4. After eight descriptors (Log Po/w, mr, AM1-HF, AM1-Dipol, ASA-P, Chi0, Density and AM1-IP) were selected. The correlation between the selected descriptors and pEC50 was established. The value of the correlation coefficient for each pair of selected descriptors was examined. The greatest value of the correlation coefficient (0.859) is that belonging to the pair of descriptors Log Po/w and AM1-Dipol. The models obtained for the prediction of inhibitory concentration of 4-quinoline carboxylic acid derivatives, using 25 compounds, with highest significant models in four descriptors are given below: pEC50 = 0.11131 - 0.56687×AM1-Dipole + 1.12247×Log Po/w + 0.00702×AM1-HF + 0.00732×ASA.P (1) pEC50 = 1.38478 - 0.49189×AM1-Dipole + 0.85383×Log Po/w + 0.00642×AM1-HF + 0.09659×Chi0 (2) pEC50 = 1.40898 - 0.47655×AM1-Dipole + 0.85460×Log Po/w + 0.00565×AM1-HF + 0.170832×mr (3) The four relevant descriptors (variables) in Equations (1, 2 and 3) of 25 compound (n training = 20 and n test set = 5) could explain 91.3, 90.4 and 89.7% of the variance (adjusted coefficient of variation) of the inhibitory concentration. The difference between r2 and q2 of three models were be < 0.1. These differences were less than 0.3, signifying the robustness of the models. The values of all the statistical parameters being within the acceptable limit reflect the internal and external predictive potential of the developed models [18]. Root mean square error (RMSE) and standard error of estimate (SEE) were lower value is better for both to good model. These two values are lower and acceptable F (F-test) or p-value showed value higher, so this is batter for models. 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 48 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 Table 3. The statistical parameter for fives equations. Equation ntraining ntest set r2 q2 r2Pred RMSA F-value p-value SEE 1 20 5 0.913 0.842 0.889 0.311 39.283 <0.0001 0.359 2 20 5 0.904 0.821 0.902 0.328 35.064 <0.0001 0.379 3 20 5 0.897 0.805 0.885 0.337 32.776 <0.0001 0.390 4 20 5 0.891 0.804 0.855 0.348 43.641 <0.0001 0.389 5 20 5 0.859 0.782 0.894 0.397 51.702 <0.0001 0.429 y = 0.9132x + 0.5252 R² = 0.9126 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 4.00 5.00 6.00 7.00 8.00 9.00 pE C 50 (P re d. ) pEC50 (Exp.) Figure 2. Plot of predicted training set versus experimental pEC50 values. y = 0.8885x + 0.6904 R² = 0.8418 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 4.00 5.00 6.00 7.00 8.00 9.00 X- pE C 50 (P re d. ) pEC50 (Exp.) Figure 3. Plot of cross validation prediction (X-pEC50) versus experimental pEC50 values. The highest significant models in three and two descriptors are given also below: pEC50 = 2.43728 - 0.49571×AM1-Dipole + 1.01033×Log Po/w + 0.00731×AM1-HF (4) pEC50 = 2.00731 - 0.42239×AM1-Dipole + 0.95099×Log Po/w (5) The three and two relevant descriptors (variables) in Equations (4 and 5) showed Criteria Model signifying the robustness of the two models. Summary of the all statistical parameter for fives equations reported in Table 3. The plot showing goodness of fit between observed and calculated activities for the training and test set compounds is given in Figure 2-4 for the best model. The QSAR model of Equation (1) was developed using 25 compounds as training set and test set molecules in Table 1. 3.2. Docking results To develop a deeper insight into the molecular mechanism of 4-quinoline carboxylic acid derivatives as human dihydro- orotate dehydrogenase comprising, the compounds brequinar (Reference), C12, C32, C42, C43 and C44 were simulated computationally to the active sites of human DHODH protein (PDB code: 1D3G). Human DHODH protein consisted of active site as shown in Figure 5. In silico molecular docking results, produced the different docking conformations based on binding energy. The variants with the minimal energy of the enzyme-inhibitor complex were selected for studies of binding mode. Preferred docked conformations of most of the ligands formed one cluster inside the active [19]. All the docked conformations for each compound were analyzed and it was found that the most favorable docking poses with maximum number of interactions were those which were ranked the highest based on the minimal binding energy, which was computed as a negative value by the software. The most favorable docking poses of the 25 docked conformations for each compound were analyzed to further investigate the interactions of the docked conformations within the active sites. The detailed docking results are tabulated in Table 4. The active site consisted of hydrophilic amino acids (His56, Tyr63, Tyr356, Tyr380, Arg136 and Gly 97) and the hydrophobic portions were constructed (Ala55, Ala59, Val134, Val62, Leu46, Leu67, Leu68, Leu359, Met43, Phe62, Pro52 and Pro364). Mostly ligands showed strong three polar interact- tions between two oxygen atoms in carboxylic group with hydrogen atom of amino group in Arg136 and Gln47 as hydrophilic interaction Figure 6. 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 http://banglajol.info/index.php/BJP/article/downloadSuppFile/31428/7544 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 49 Table 4. Free binding energy (kJ/mol), bond interaction and interacted amino acid of the investigated compounds. No Compound Free binding energy, S Type of bond interacted Interaction group Amino acid interacted Length (Å) 1 Brequinar analog * -42.43 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.32 1.62 1.27 2 Brequinar ** -35.03 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.68 1.87 1.89 3 C1 -28.91 1 Polar bonds C=O of carboxylic acid Arg 136 2.10 4 C2 -15.54 π-Interaction Phenyl Phe62 - 5 C3 -38.19 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.90 1.92 1.85 6 C4 -39.67 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.85 1.96 1.85 7 C11 -41.81 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.85 1.94 1.81 8 C12 -42.31 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.75 1.97 1.81 9 C16 -42.51 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.78 1.92 1.83 10 C18 -42.02 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.98 1.74 1.83 11 C22 -36.34 3 polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.98 1.79 1.81 12 C24 -38.67 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.82 1.92 1.77 13 C26 -39.10 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.73 2.07 1.75 14 C29 -17.56 1 Polar bonds C=O of carboxylic acid Tyr265 3.69 15 C30 -43.85 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.82 1.97 1.77 16 C31 -43.28 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.80 1.97 1.75 17 C32 -43.06 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.75 1.96 1.81 18 C33 -47.29 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.95 1.77 1.89 19 C34 -41.81 3 Polar bonds H-O of carboxylic acid H-O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.75 1.96 1.82 20 C35 -41.41 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.75 1.96 1.82 21 C36 -41.58 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.82 1.96 1.76 22 C39 -42.23 3 Polar bonds π-Interaction H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Phenyl Arg136 Gln47 Arg136 Phe62 1.75 1.99 1.81 - 23 C40 -37.76 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid Arg136 Arg136 1.83 1.67 24 C42 -45.40 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.83 1.97 1.77 25 C43 -39.33 3 Polar bonds H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Arg136 Gln47 Arg136 1.73 2.02 1.73 26 C44 -29. 71 3 Polar bonds π-Interaction H-O of carboxylic acid C=O of carboxylic acid C=O of carboxylic acid Phenyl Arg136 Gln47 Arg136 Phe62 2.19 2.05 1.95 - * 2-Biphenyl-4-yl-6-fluoro-3-methyl-quinoline-4-carboxylic acid. ** 6-Fluoro-2-(2'-fluoro-[1,1'-biphenyl]-4-yl)-3-methylquinoline-4-carboxylic acid. 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 50 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 y = 1.595x - 2.6944 R² = 0.8888 5.00 5.50 6.00 6.50 7.00 7.50 8.00 5.00 5.20 5.40 5.60 5.80 6.00 6.20 6.40 pE C 50 (P re d. ) pEC50 (Exp.) Figure 4. Plot of predicted test set versus experimental pEC50 values. (a) N O O F - Met 43 Leu 46 Gln 47 Pro 52 Ala 55 His 56 Ala 59 Thr 63 Leu 67 Leu 68 Val 134 Arg 136 Tyr 356 Leu 359 Thr 360 Pro 364 (b) Figure 5. (a) 3D model of active sites of human DHODH protein 1D3G with brequinar analog ligand, (b) 2D model of ligand interaction with protein. N O O F O - Tyr 38 Leu 42 Met 43 Leu 46 Gln 47 Pro 52 Ala 55 His 56 Ala 59 Phe 62 Thr 63 Leu 67 Leu 68 Val 134 Arg 136 Tyr 356 Leu 359 Thr 360 Pro 364 Figure 6. 2D model of compound C44 interactions by π-interaction with Phe62 and three hydrogen bonding interactions with Arg136 and Gln47. Strong hydrogen bonding interactions of ligands with Arg136 and Gln47 showed bond distances in range (1.32 and 2.10 Å). It is quite interesting to note that the binding free energies “S” in Escore-1 (London dG) for all ligands are in the range difference between them from -35.031 to -45.400 kcal/mol, except compounds C1m, C2 and C29 showed lower binding free energy. Compounds C42, C30, C31 and C32 showed binding free energies (-45.400, -43.835, -43.278 and - 43.059 kcal/mol, respectively) higher than reference analogues Brequinar (-42.4287 kcal/mol). This may explain the higher activity of them compared to analogues Brequinar. Although compound C44 showed higher biological activity than the others, it has lower binding free energy (-29.708 kcal/mol). The π-interaction between phenyl ring in compound C44 and phenyl ring in Phe62 beside to three hydrogen bonding interactions mentioned above, maybe that explains the higher activity of compound C44 than other compounds (Figure 6). 4. Conclusion The derived QSAR models have provided rationales to explain of 4-quinoline carboxylic acid derivatives inhibitory activity to VSV replication, according to descriptors; Log Po/w, mr, AM1-HF, AM1-Dipol, ASA-P, Chi0, density and AM1-P; by highlighted the role of these descriptors in biological activity. PLS analysis has also confirmed the suggested models have acceptable predictability. All the compounds are within the applicability domain of the proposed models and were evaluated correctly. These models can be used to design new 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 Hajalsiddig and Saeed / European Journal of Chemistry 10 (1) (2019) 45-51 51 4-quinoline carboxylic acid derivatives as inhibitors of human dihydroorotate dehydrogenase. The binding energies derived from docking simulations indicated that almost all the 25 compounds are exhibited stronger binding affinity for 1D3G. It is noteworthy, the higher biological activity of diaryl ether substituted 4-quionline carboxylic acid as inhibitors human as dihydroorotate dehydrogenase reported by Das et al. be through with result of docking [12]. In the view of this study, the further research can be carried on the newly designed compound, which has diaryl ether substituted 4-quinoline carboxylic acid, as inhibitors. Disclosure statement Conflict of interests: The authors declare that they have no conflict of interest. Author contributions: All authors contributed equally to this work. Ethical approval: All ethical guidelines have been adhered. ORCID Tawassl Tajelsir Hassan Hajalsiddig http://orcid.org/0000-0001-8053-6339 Ahmed Elsadig Mohammed Saeed http://orcid.org/0000-0002-7317-8040 References [1]. Hong, S.; Jung, Y.; Park, S.; Paik, S. Virus Genes 2005, 31, 195-201. [2]. Novella, I.; Ball, L.; Wertz, G. J. Virol. 2004, 78, 9837-9841. [3]. Letchworth, G.; Rodriguez, L.; Del Cbarrera, J. Veterinary J. 1999, 157, 239-260. [4]. Johannsdottir, H.; Mancini, R.; Kartenbeck, J.; Amato, L.; Helenius, A. J. Virol 2008, 83, 440-453. [5]. Ausubel, L.; Meseck, M.; Derecho, I.; Lopez, P.; Knoblauch, C.; Mcmahon, R.; Anderson, J.; Dunphy, N.; Quezada, V.; Khan, R.; Huang, P.; Dang, W.; Luo, M.; Hsu, D.; Woo, S.; Couture, L. Hum. Gene Ther. 2011, 22, 489-497. [6]. Munier-Lehmann, H.; Vidalain, P.; Tangy, F.; Janin, Y. J. Med. Chem. 2013, 56, 3148-3167. [7]. Arakaki, T.; Buckner, F.; Gillespie, J.; Malmquist, N.; Phillips, M.; Kalyuzhniy, O.; Luft, J.; DeTitta, G.; Verlinde, C.; Van Voorhis, W.; Hol, W.; Merritt, E. Mol. Microbiol. 2008, 68, 37-50. [8]. Schroder, M.; Giermann, N.; Zrenner, R. Plant Physiology. 2005, 138, 1926-1938. [9]. Evans, D.; Guy, H. J. Biol. Chem. 2004, 279, 33035-33038 [10]. Ohishi, T.; Inaoka, D. K.; Kita, K.; Kawada, M. Chem. Pharm. Bull. 2018, 66, 239-242. [11]. Liu, S.; Neidhardt, E. A.; Grossman, T. H.; Ocain, T.; Clardy, J. Struc. 2000, 8(1), 25-33 [12]. Das, P.; Deng, X.; Zhang, L.; Roth, M.; Fontoura, B.; Phillips, M.; De Brabander, J. ACS Med. Chem. Lett. 2013, 4, 517-521. [13]. Pourbasheer, E.; Vahdani, S.; Aalizadeh, R.; Banaei, A.; Ganjali, M. R. J. Chem. Sci. 2015, 127, 1243-1251. [14]. Xu, X.; Wang, J.; Yao, Q. Bioorg. Med. Chem. Lett. 2015, 25, 241-244. [15]. Sharma, B. K.; Pilania, P.; Singh, P. Eur. J. Chem. 2010, 1(4), 325-334. [16]. Wang, R.; Jiang, J.; Pan, Y.; Cao, H.; Cui, Y. J. Hazard. Mater. 2009, 166, 155-186 [17]. Ahirwar, S.; Agarwal, P.; Patra, P. World J. Pharm. Pharmaceut. Sci. 2015, 3, 671-684. [18]. Mitra, I.; Saha, A.; Roy, K. Sci. Pharm. 2013, 81(1), 57-80. [19]. Arif, M.; Jabeen, F.; Saeed, A.; Qureshi, I. Z.; Mushtaq, N. Bangladesh J. Pharmac. 2017, 12, 319-332. Copyright © 2019 by Authors. This work is published and licensed by Atlanta Publishing House LLC, Atlanta, GA, USA. The full terms of this license are available at http://www.eurjchem.com/index.php/eurjchem/pages/view/terms and incorporate the Creative Commons Attribution-Non Commercial (CC BY NC) (International, v4.0) License (http://creativecommons.org/licenses/by-nc/4.0). By accessing the work, you hereby accept the Terms. This is an open access article distributed under the terms and conditions of the CC BY NC License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited without any further permission from Atlanta Publishing House LLC (European Journal of Chemistry). No use, distribution or reproduction is permitted which does not comply with these terms. Permissions for commercial use of this work beyond the scope of the License (http://www.eurjchem.com/index.php/eurjchem/pages/view/terms) are administered by Atlanta Publishing House LLC (European Journal of Chemistry). 2019 – European Journal of Chemistry – CC BY NC – DOI: 10.5155/eurjchem.10.1.45-51.1795 http://orcid.org/0000-0001-8053-6339 http://orcid.org/0000-0002-7317-8040 http://www.eurjchem.com/index.php/eurjchem/pages/view/terms http://creativecommons.org/licenses/by-nc/4.0 http://www.eurjchem.com/index.php/eurjchem/pages/view/terms 1. Introduction 2. Experimental 2.1. QSAR study 2.1.1. Data set 2.1.2. Theoretical molecular descriptors 2.1.3. Model development 2.1.4. Validation model 2.2. Docking study 3. Results and discussion 3.1. QSAR results 3.2. Docking results 4. Conclusion Disclosure statement ORCID References PrintField10: PrintField11: PrintField12: PrintField13: PrintField14: PrintField15: PrintField16: PrintField20: PrintField21: PrintField22: PrintField23: PrintField24: PrintField25: PrintField26: