Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 
  • Users Online: 483
  • Home
  • Print this page
  • Email this page


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2021  |  Volume : 16  |  Issue : 5  |  Page : 528-546

Quinazoline analogues as cytotoxic agents; QSAR, docking, and in silico studies


1 Department of Medicinal Chemistry, Faculty of Pharmacy; Department of Medicinal Chemistry, Pharmaceutical Sciences Research Center, Shiraz University of Medical Sciences, Shiraz, I.R. Iran
2 Department of Medicinal Chemistry, Faculty of Pharmacy, Shiraz University of Medical Sciences, Shiraz, I.R. Iran
3 Department of Medicinal Chemistry, Pharmaceutical Sciences Research Center, Shiraz University of Medical Sciences, Shiraz, I.R. Iran

Date of Submission05-Dec-2020
Date of Decision13-Feb-2021
Date of Acceptance25-Jul-2021
Date of Web Publication18-Aug-2021

Correspondence Address:
Razieh Sabet
Department of Medicinal Chemistry, Faculty of Pharmacy, Shiraz University of Medical Sciences, Shiraz
I.R. Iran
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/1735-5362.323919

Rights and Permissions
  Abstract 


Background and purpose: Synthesis and investigation of pharmacological activity of novel compounds are time and money-consuming. However, computational techniques, docking, and in silico studies have facilitated drug discovery research to design pharmacologically effective compounds.
Experimental approach: In this study, a series of quinazoline derivatives were applied to quantitative structure-activity relationship (QSAR) analysis. A collection of chemometric methods were conducted to provide relations between structural features and cytotoxic activity of a variety of quinazoline derivatives against breast cancer cell line. An in silico-screening was accomplished and new impressive lead compounds were designed to target the epidermal growth factor receptor (EGFR)-active site based on a new structural pattern. Molecular docking was performed to delve into the interactions, free binding energy, and molecular binding mode of the compounds against the EGFR target.
Findings/Results: A comparison between different methods significantly indicated that genetic algorithm-partial least-squares were selected as the best model for quinazoline derivatives. In the current study, constitutional, functional, chemical, resource description framework, 2D autocorrelation, and charge descriptors were considered as significant parameters for the prediction of anticancer activity of quinazoline derivatives. In silico screening was employed to discover new compounds with good potential as anticancer agents and suggested to be synthesized. Also, the binding energy of docking simulation showed desired correlation with QSAR and experimental data.
Conclusion and implications: The results showed good accordance between binding energy and QSAR results. Compounds Q1-Q30 are desired to be synthesized and applied to in vitro evaluation.

Keywords: Cytotoxic; Molecular docking; QSAR; Quinazoline.


How to cite this article:
Emami L, Sabet R, Khabnadideh S, Faghih Z, Thayori P. Quinazoline analogues as cytotoxic agents; QSAR, docking, and in silico studies. Res Pharma Sci 2021;16:528-46

How to cite this URL:
Emami L, Sabet R, Khabnadideh S, Faghih Z, Thayori P. Quinazoline analogues as cytotoxic agents; QSAR, docking, and in silico studies. Res Pharma Sci [serial online] 2021 [cited 2022 Aug 14];16:528-46. Available from: https://www.rpsjournal.net/text.asp?2021/16/5/528/323919




  Introduction Top


Among the nitrogen-containing compounds, the quinazoline ring was a very privileged and effective scaffold in pharmaceutical and medicinal chemistry; they have a broad spectrum of biological activities such as anticancer, diuretic, anti-inflammatory, anticonvulsant, antimicrobial, antiviral, antiplasmodial, and antihypertensive effects[1]. Also, among the different quinazoline scaffolds, 2-substituted-4(3H)-quinazolinone has been used as an attractive pharmacophore for drug design purposes[2],[3]. Quinazoline which is substituted at C4, C6, and C7 has been applied as one of the most significant classes of quinazoline-based epidermal growth factor receptor (EGFR) inhibitors[4]. The quinazoline skeleton is also a substructure of natural purine bases, plant alkaloids, and several FDA-approved drugs such as prazosin, alfuzosin, erlotinib, gefitinib [Figure 1][5],[6].
Figure 1: Previously reported epidermal growth factor receptor inhibitors bearing quinazoline scaffold.

Click here to view


Synthesis and investigation of pharmacological activity of novel compounds, usually take large amounts of expenditures and time. The use of computational techniques, docking, and in silico studies for designing pharmacologically effective compounds has opened a new approach to drug discovery research. Quantitative structure-activity relationships (QSAR) studies, as one of the important subjects in chemometrics, provide the ability to biological activities prediction for the novel or even non-synthesized compounds by medicinal chemists[7],[8],[9],[10],[11].

Linear QSAR models are mathematical equations that deliver good information for better explaining the mechanisms of action of the compounds, and proving the relationship between chemical structures and pharmacological activities.

The most important phase in constructing QSAR models is the suitable description of the structural and physicochemical properties of chemical structures[12],[13],[14],[15]. These properties called molecular descriptors have a high impact on the pharmacological properties of a compound[16],[17],[18],[19]. Molecular descriptors have been classified into different groups including physiochemical, constitutional, geometrical, topological, and quantum chemical descriptors. Dragon and Hyperchem as two famous computational packages are able to calculate more than 7000 of these parameters[20],[21].

An important approach for the researchers is establishing a complete SAR of quinazoline derivatives of cytotoxic agents to modify the quinazoline moiety[22],[23]. In this paper, we investigated QSAR studies of some quinazoline derivatives which have been recently reported to exhibit cytotoxic activity against the MCF-7 cell line. Different QSAR models including multiple linear regressions were used to establish the relationship between descriptors and anti-breast cancer activity of compounds[16],[24],[25],[26]. Our QSAR models would be established mathematical equations between pharmacological activities and calculable parameters such as topological, quantum, physicochemical, stereochemical, or electronic parameters. In addition, molecular docking simulation was also done to reach the most favorable conformation and binding modes of all compounds as well as newly designed compounds towards EGFR as possible targets for their anticancer effect.

The molecular docking simulation help us to understand the possible interactions between the ligands and enzyme’s active sites in detail and also helps to design novel potent inhibitors. In 2017, Tu et al. reported docking studies of new synthesized quinazoline bearing aryl semi carbazole ([Figure 1], structure I) and showed that tetrahydrofuran substituent was necessary for the anti-cancer activity of these compounds[27]. The in vitro cytotoxic activities of new quinazoline compounds were determined against three human cancer cell lines (A549, MCF-7, and PC-3), and showing that the introduction of quinazoline derivatives bearing 2,3 dihydro-indole or 1, 2, 3, 4-tetrahydroquinoline moiety ([Figure 1], structure II] could act appropriately to interact with the active site of the EGFR target [Figure 1][28].


  Material and Methods Top


Data set and descriptor generation

In this study, the experimental biological activities of anticancer activity against MCF-7 cell line were used (in terms of -log IC50), of a set of 81 quinazoline derivatives[29],[30],[31],[32],[33]. The structural feature and antiproliferative values of these compounds were shown in [Table 1].
Table 1: Chemical structure of quinazoline derivatives used in this study and their docking binding energy.

Click here to view


Hyperchem 8.0 software (Hypercube Inc. version 8.0.3) was used to draw the structures and optimization by semi-empirical AM1 calculations and Polak-Ribiere algorithm until the root mean square gradient of 0.01 kcal/mol[20]. Molecular volume, molecular surface area, hydrophobicity (Log P), hydration energy, and molecular polarizability were achieved by Hyperchem software. The other descriptors such as topological, constitutional, and functional descriptors were achieved by the Dragon package (version 5.5), developed by the Milano chemometrics and QSAR group[21]. The calculated descriptors are briefly introduced in [Table 2]. Finally, kenardston algorithm was used for the classification of the data set into calibration and test sets. This method is not accompanied by error and randomness.
Table 2: Brief description of some descriptors used in this study.

Click here to view


Data screening and model building

Stepwise regression SPSS (version 22.0) software was used to analyze the experimental data to achieve effective descriptors. Then, the effective descriptors were hoarded in a data matrix that the number of molecules and descriptors were collected in rows and columns, respectively. Multiple linear regressions (MLR), partial least squares (PLS), MLR with factor analysis (FA-MLR), and principal component regression analysis (PCRA) methods were utilized to derive the QSAR equations.

A cross-validation and external test set were used to evaluate the model’s stability and prediction ability. For this purpose leave-one-out, leave-group-out were used[34]. Also, the PLS regression with NIPALS-based algorithm existed in the chemometrics toolbox of MATLAB software (version 21 Math work Inc.) was used to obtain the optimum number of factors based on the Harland and Thomas F-ratio criterion[35].

Molecular modeling procedure

3D X-ray crystal structures of EGFR in complex with erlotinib (PDB ID: 1M17) were selected from Protein Data Bank (http://www.rcsb.org) based on the similarity of its co-crystal ligand to studied compounds[36],[37]. The structure of studied compounds was generated, minimized, and converted to pdbqt format. The pdbqt formats of the ligands were prepared by adding Gasteiger charges and setting the degree of torsions by the AutoDock tools package (1.5.6). For the preparation of pdbqt format of protein, we removed cognate ligand, water molecules, added missing hydrogen atoms, and finally merged non-polar hydrogens according to their corresponding carbons[38],[39]. All preparation was performed by the AutoDock tools package (1.5.6). Docking procedure was done at the grid box with a size of 70 × 70 ×70 and the centre of x = 20.14, y = 0.3, z = 52.2 Å by AutoDock Vina (1.1.2)[40]. Using an in-house batch script (DOCKFACE)[41]. The exhaustiveness was set to 100, and other docking parameters were set as default. The binding interactions of the docked compounds and the receptor were visualized by a fully automated protein-ligand interaction profiler[42].


  Results Top


MLR analysis

The resulted QSAR models from different types of descriptors for the compounds (65 molecules as calibration and 16 molecules as prediction sets) are listed in [Table 3]. The stepwise selection was used to achieve MLR analyses.
Table 3: The results of multiple linear regressions analysis with different types of descriptors.

Click here to view


[Table 4] includes the statistical parameters of prediction that indicate the suitability of the proposed QSAR model based on MLR analysis of molecular parameters. The correlation coefficient of prediction is 0.97, which means that the resulted QSAR model could predict 97% of variances in the anti-breast cancer activity data. It has a root mean square error (RMSE) of 0.21.
Table 4: Statistical parameters for testing prediction ability of the MLR, GA-PLS, PCR, and FA-MLR models

Click here to view


Genetic algorithm-PLS analysis

Genetic algorithm (GA) method was used to select the variable to find more effective descriptors to improve the performance of PLS analysis. The data set (n = 81) was disported in calibration (n = 65) and prediction set (n = 16) groups. A cross-validation procedure was used to achieve the optimum number of latent variables for each PLS model. In this work, many different GA-PLS runs were conducted using the different initial set of populations (50-250), therefore, the most convenient GA-PLS model that resulted in the best fitness contained 15 descriptors including six constitutional descriptors (nI, nR10, nBnz, nArX, nO, nX), five functional descriptors (n ROR, nC=S, npyridine, nHAcc, nArNO2), one chemical (SA), one resource description framework (RDF) parameter (RDF115v), one 2D autocorrelation parameter (MATS3m), and one charge (Qneg) parameter. The majority of these descriptors are constitutional indices. All of them being those obtained by different MLR-based QSAR models. The PLS estimate of the regression coefficients is shown in [Figure 2].
Figure 2: Partial least squares regression coefficients for the variables used in the genetic algorithm-partial least squares model.

Click here to view


In order to investigate the relative importance of the variable that appeared in the final model, the variable important in projection (VIP) was employed[43]. VIP values reflect the importance of terms in the PLS model. According to Erikson et al. X-variables (predictor variables) could be classified according to their relevance in explaining y (predicted variable), so that VIP > 1.0 and VIP < 0.8 mean highly or less influential, respectively, and 0.8 < VIP< 1.0 means moderately influential [Figure 3].
Figure 3: Plot of VIP for the descriptors used in genetic algorithm-partial least squares model. VIP, variables important in projection.

Click here to view


FA-MLR and PCRA

In FA-MLR analysis, Factor Analysis was used to decrease the number of variables and also to identify the effective predictor variables and to decrease collinearities between them[44]. PCRA was tried for the dataset along with FA-MLR[45]. In this method, factor scores, as obtained from FA, were used as the predictor variables[44].

Based on the process explained in the experimental part, the following five-parametric equation is shown in [Table 5].
Table 5: The results of multiple linear regressions analysis with factor analysis with different types of descriptors

Click here to view




This equation could explain about 97% of the variance and predict 96% of the variance in pIC50 data. It has an RSME of 0.15. This equation describes the effect of functional, constitutional and so on descriptors (nI, n = CHR, nO-O, nCONR2 and nN-N) on anti-breast activity of the studied molecules. Multiple regression equation using stepwise selection method (PCRA), the following equation was gained [Table 6].
Table 6: The results of principal component regression analysis.

Click here to view




This equation could explain and predict 93% and 91% of the variances in pIC50 data, respectively. The RSME of PCRA analysis was 0.19. PCRA equation is better than those derived from FA-MLR. While the data of this analysis showed acceptable prediction, we observed that the predicted values of some molecules are near to each other.

As it is shown in [Table 5], in the case of each factor, the loading values for some descriptors are much higher than those of the others. These high values for each factor indicate that this factor delivers higher information about each index. It should be noted that all factors have information from all descriptors but the contribution of the descriptor in different factors are not equal. For example, factors 1 and 2 have higher loadings for the chemical, atom-center, constitutional, functional, BCUT (modified burden eigenvalues) information, geometrical, 2D autocorrelations, and Walk and path counts indices whereas information about the weighted holistic invariant molecular (WHIM descriptor), connectivity indices, 3D MoRSE descriptors and functional descriptors are highly combined in factors 3 and 4.

To assess the robustness and applicability domain of all gained models, leverage as standard procedures was used [Table 7]. Warning leverage (h*) is another criterion to explain results. Leverage greater than warning leverage h* means that the predicted response is the result of substantial extrapolation of the model and therefore may not be reliable[46].
Table 7: Leverage (h) of the external test set molecules for different models. The last row (h*) is the warning leverage.

Click here to view


Docking studies

Molecular docking simulation is the computational method of ligand binding to a receptor that indicates the affinity of the binding ligand to the active site of the target. Quinazoline derivatives are well-known as EGFR inhibitors[47]. So, the docking study was done for EGFR kinase to predict the binding modes, affinities, and orientation of ligands at the active sites of the enzyme. In the present research, molecular docking simulations were done on 81 molecules of data set as well as 30 designs of new compounds. The docking method was verified by docking of co-crystallized ligand (erlotinib) in EGFR active site and comparison of the results between docked and crystallized erlotinib conformations. RMSD of docking for erlotinib in comparison with its coordination in the crystal structure was 1.07. The free energy binding of docked compounds is summarized in [Table 1].

Docking study on EGFR

All investigated complexes showed better docking binding energies than the reference drug (erlotinib) with a binding energy value of -5.8 kcal/mol which showed a good correlation with QSAR results, it means almost that the compounds with the lowest IC50 have the highest binding energy. The docked conformation of protein crystal structure (1M17) with erlotinib and three other quinazoline derivatives (33, 55, and 79), which possess strong binding energy, are shown in [Figure 2]. As depicted in [Figure 4], the nitrogen atom of the quinazoline moiety in compound 33 binds to the Thr 766 via hydrogen bond interactions, and some hydrophobic interactions with the residues Lys 721, Ala 719, Leu 764, Met 769, and Leu 820 were observed.
Figure 4: The main interaction between the active site of epidermal growth factor receptor (PDB ID: 1M17) with compounds 33, 55, 79, and erlotinib.

Click here to view


In compound 55, the nitrogen atom formed hydrogen bond interaction with Thr 766 and the phenyl ring formed pi-stacking interaction with Phe 699 and also there is some hydrophobic interaction with Val 702, Leu 764, and Leu 694. Compound 79 can be summarized by hydrogen bond interactions with Glu 738 and Thr 830 and the pi-cation interaction between quinazoline ring and Lys 721. Also, interaction with Leu 699, 764, 766, 820, Val702, and Ala 719. To explain the binding mode of erlotinib, in brief, it should be announced that the oxygen atom of the erlotinib chain formed a hydrogen bond via Cys 733. In addition, pi-stacking interaction between the phenyl ring and Phe 699 was observed.

In silico screening

In silico methods can help to discover newly designed drugs before synthesized and offers a lot of benefits such as cost savings, time-consuming, and increase the speed of predict and achieve new pharmaceutical compounds. Virtual screening is a computational technique to search active compounds from very large libraries by the effect of structural modification on binding energy and biological activity at the parent molecule. A new compound designed according to the best QSAR model and binding energy from docking simulation. We used the important descriptors selected in the GA-PLS model to design new active compounds because the GA-PLS model has the greatest statistical parameters. The in silico screen was applied by substituting different groups in different places. We also analyzed the robustness and applicability domain of the model by calculation of leverage value on the whole data set. The warning leverage (h*) was found to be 0.66 and it is cleared that all of the compounds have a smaller leverage value (h) than warning leverage (h*). The results of this investigation are summarized in [Table 8]. We designed 30 compounds (Q1-Q30) and biological activities based on the GA-PLS equation were achieved. And also, docking binding energy was obtained. From the results, the Q14 has the best activity and have a good candidate for anticancer agents.
Table 8: Structural modification of studied compounds and their predicted activities and docking binding energies for epidermal growth factor receptor inhibitory based on genetic algorithm-partial least squares equation.

Click here to view


ADME and drug-likeness properties of designed compounds

The physicochemical properties of some potent designed compounds were calculated by http://www.swissadme.ch/, and the results are shown in [Table 9]. The molecular weight of compounds (Q7, Q14, and Q17) was in the accepted range. The log P of hit compound Q14 values is lower than 5 and displayed that it has desire lipophilicity. Furthermore, the hydrogen bond properties and total polar surface area, and rotatable bond number of compound Q14 is within the acceptable limit. Generally, the data indicated that compound Q14 followed Lipinski’s rule of five and Veber rule. The drug-likeness and absorption, distribution, metabolism, and excretion (ADME) properties of some potent compounds were derived from the preADMET online server (http://preadmet.bmdrc. org/). ADME properties including human intestinal absorption, in vitro caucasian colon adenocarcinoma cell permeability (Caco-2), skin permeability, in vitro plasma protein binding, and in vivo, blood-brain barrier penetration are shown in [Table 10].
Table 9: Physicochemical properties of some potent designed compounds.

Click here to view
Table 10: In silico absorbance distribution metabolism and excretion of some potent designed compounds.

Click here to view


Human intestinal absorption analysis indicated that hit compound Q14 indicated well human intestinal absorption, which causes quick absorption from the intestine to the bloodstream. The Caco-2 permeability parameters suggest that compound Q14 has moderate permeability for penetration to biological membranes. Moreover, compound Q14 binds moderately to plasma proteins, so it may be used as an oral dosage form. It has low blood-brain barrier penetration; so, it has a low neurotoxicity effect.


  Discussion Top


In MLR analysis the equation 1 [Table 2] explained the effect of constitutional descriptors on the cytotoxicity activity of these compounds. It showed that increasing the numbers of substitution halogen atoms (nF, nCl, nI) on the compounds results in an activity enhancement, such as molecular series 6-7, 9-14. It also showed that the iodine substitution on the 6th position of quinazoline ring would have better activity. It also explains the positive effect nR10 and nR06 (number of 6-membered rings) such as phenyl ring on activity (such as molecule series 19-24, 25-30, and 33-37 have good activity). This equation also showed the effect of O numbers on the activity. The Keto group on the quinazoline ring has better activity than thio keto group on the ring (molecules number 50 has low activity). It also explains the positive effect of nDB (number of double bonds) on the activity. Quinazoline and phenyl ring have multiple double bonds.

The result of the topological group counts parameter on cytotoxicity activity of the studied molecules has been introduced by equation 2 of [Table 2]. It showed that among the topological descriptors T (O. O), PCR, and JhetV have positive effects and T (O..I), JhetZ, SEigv, piID, and T(F..I) have negative effects on cytotoxic activity of the compounds.

The equations 3-12 of [Table 3] demonstrated the positive and negative effects of BCUT, Galvz topological, 2D autocorrelations, geometrical, RDF, 3D MoRSE, WHIM, GETAWAY, functional, and atom-centered fragments on the anti-breast cancer activity of these compounds. The MLR equation of [Table 2] obtained from the pool of functional groups descriptors, E11, explained the positive impact of the nCb (presence of benzene ring, especially with halogen atoms such as F and, Cl, is good for example compounds 33-37) n = CHR, nArX (number of X on the aromatic ring for example phenyl ring), nArNO2 (number of nitro groups (aromatic)), nN-N(number of N hydrazines), nROR (number of ethers), nSO2 (number of sulfone compounds 19-21), nHAcc(number of acceptor atoms for H-bonds (N, O, F)) on the anticancer activity. The nCONR2 (number of tertiary amides compound 2), nC=S (number of thioketo compound 42), npyridine (number of pyridines compound 21) had negative compacts on the anti-breast cancer activity. The negative sign of this group proposed that a decrease in the number of these descriptors resulted in an activity enhancement. This equation, which has a high statistical quality (R2 = 0.97, Q2 = 0.96).

The model extracted from PLS model has a high cross-validation statistic, and it also represents a high ability for modelling external test samples. It could explain and predict about 93% of variances in the cytotoxic activity of the studied molecules. There is a close agreement between the experimental and predicted values of anticancer activity data to evaluate the significance of the 15 selected PLS descriptors in the protein tyrosine kinase inhibitory activity. The VIP analysis of PLS equation is shown in [Figure 2]. As it is observed, nArX, nI, Qneg, nROR, and nArNO2 indices represent the most significant contribution in the resulted QSAR model. In addition, functional group parameters such as nO, n BnZ, and MATS3m are moderately influential parameters. The calculated leverage values of the prediction set samples for all models and the warning leverage, as the threshold value for accepted prediction, are shown in [Table 6]. As seen, the leverages of all test samples were lower than h* for all models. This means that all the predicted values were acceptable.

Finally, the FA-MLR and PCRA results shown in [Table 11], demonstrate the four-factor loadings of the variables (after VARIMAX rotation) for the compounds tested for cytotoxic activity. As it is observed, about 62% of variances in the original data matrix could be explained by the selected four factors.
Table 11: Numerical values of factor loading numbers 1–4 for descriptors after varimax rotation

Click here to view


The results of the docking studies indicated that the important amino acid in the active site of EGFR enzyme is Thr 766, Val 719 and 702. The majority interaction of studied compounds are hydrogen bond between substituent at quinazoline moiety with Thr 766 and the pi-hydrogen bond between receptor and phenyl group of ligands. Molecular docking simulations were also applied to the newly designed compounds to obtain binding mode and energy with EGFR target. The physicochemical and ADME properties of the designed compounds were also predicted. It was indicated that these compounds were compatible with the Lipinski and Veber rules and have desired pharmacokinetic characteristics. In future studies, 30 newly designed quinazoline compounds will be synthesized and subjected to cytotoxic laboratory evaluation.


  Conclusion Top


The quantitative relationship between molecular descriptor and cytotoxic activity of different quinazoline derivatives was discovered by different chemometrics methods, such as MLR, FA-MLR, PCR, and GA-PLS. The reliability, accuracy, and predictability of the proposed models were evaluated by RSME of cross-validation, cross-validation, and the RMSE of prediction. A static comparison between different QSAR methods indicated that GA-PLS represented a good relationship between the chemical structure of quinazoline and anticancer activity for MCF-7 cell line. This study suggested that constitutional descriptors, functional descriptors, chemical, RDF parameter, 2D autocorrelation parameter, and charge parameter descriptors of molecules were important for the prediction of anticancer activity of quinazoline derivatives. In silico screening was employed to discover new compounds such as Q14, with good potential inhibitory as anticancer agents and suggested to be synthesized, furthermore, ADME and drug-likeness properties of designed compounds indicated that compound Q14 is compatible with Lipinski and Veber rules and has desired pharmacokinetic properties. Also, the binding energy of docking simulation showed a desire correlation with QSAR and experimental data.

Acknowledgments

This work was financially supported by Shiraz University of Medical Sciences, Shiraz, I.R. Iran, under Grant No. 97-01-36-16805.

Conflict of interest statement

The authors declared no conflict of interest in this study.

Authors’ contribution

L. Emami and S. Khabnadideh analyzed the data and co-wrote the paper. R. Sabet designed the experiment, analyzed the data, and co-wrote the paper. Z. Faghih and P. Thayori analyzed the data



 
  References Top

1.
Haghighijoo Z, Rezaei Z, Taheri S, Jani M, Khabnadideh S. A rapid and convenient method for synthesis of anilinoquinazoline: an improved synthesis of erlotinib derivatives. Trends Pharmacol Sci. 2015;1(3):173-178.  Back to cited text no. 1
    
2.
Rahmannejadi N, Khabnadideh S, Yavari I. Synthesis, docking, and cytotoxic activities of novel 2-aryl-4-(arylamino) quinazolines. Monatsh Chem. 2018;149(11):2085-2092. DOI: 10.1007/s00706-018-2270-3.  Back to cited text no. 2
    
3.
Rahmannejadi N, Zomorodian K, Faghih Z, Faghih Z, Khabnadideh S, Yavari I. Preparation, docking, antimicrobial and cytotoxic activities of 2-arylquinazolinones. Br J Pharm Res. 2017;20(3):1-11. DOI: 10.9734/JPRI/2017/38244.  Back to cited text no. 3
    
4.
Emami L, Faghih Z, Khabnadideh S, Rezaei Z, Sabet R, Harigh E, et al. 2-(Chloromethyl)-3- phenylquinazolin-4(3H)-ones as potent anticancer agents; cytotoxicity, molecular docking and in silico studies. J Iran Chem Soc. 2021;18:1877-1889. DOI: 10.1007/s13738-021-02168-1.  Back to cited text no. 4
    
5.
Divar M, Zomorodian K, Bastan S, Yazdanpanah S, Khabnadideh S. Synthesis of some quinazolinone derivatives using magnetic nanoparticles-supported tungstic acid as antimicrobial agents. J Iran Chem Soc. 2018;15(7):1457-1466. DOI: 10.1007/s13738-018-1337-8.  Back to cited text no. 5
    
6.
Faghih Z, Rahmannejadi N, Sabet R, Zomorodian K, Asad M, Khabnadideh S. Synthesis of some novel dibromo-2-arylquinazolinone derivatives as cytotoxic agents. Res Pharm Sci. 2019;14(2):115-121. DOI: 10.4103/1735-5362.253358.  Back to cited text no. 6
    
7.
Ferreira MMC. Multivariate QSAR. In: Wang Z, editor. Encyclopedia of Physical Organic Chemistry. John Wiley & Sons; Vol. 3. 2016. pp. 1-38. DOI: 10.1002/9781118468586.epoc3013.  Back to cited text no. 7
    
8.
Nasab RR, Mansourian M, Hassanzadeh F. Synthesis, antimicrobial evaluation and docking studies of some novel quinazolinone Schiff base derivatives. Res Pharm Sci. 2018;13(3):213-220. DOI: 10.4103/1735-5362.228942.  Back to cited text no. 8
    
9.
Sabet R, Fassihi A, Hemmateenejad B, Saghaei L, Miri R, Gholami M. Computer-aided design of novel antibacterial 3-hydroxypyridine-4-ones: application of QSAR methods based on the MOLMAP approach. J Comput Aided Mol Des. 2012;26(3):349-361. DOI: 10.1007/s10822-012-9561-2.  Back to cited text no. 9
    
10.
Sabet R, Fassihi A, Moeinifard B. QSAR study of PETT derivatives as potent HIV-1 reverse transcriptase inhibitors. J Mol Graph Model. 2009;28(2):146-155. DOI: 10.1016/j.jmgm.2009.05.002.  Back to cited text no. 10
    
11.
Sadeghian-Rizi S, Sakhteman A, Hassanzadeh F. A quantitative structure-activity relationship (QSAR) study of some diaryl urea derivatives of B-RAF inhibitors. Res Pharm Sci. 2016;11(6):445-453. DOI: 10.4103/1735-5362.194869.  Back to cited text no. 11
    
12.
Jain SK, Yadav AK, Nayak P. 2D QSAR analysis on oxadiazole derivatives as anticancer agents. Int J Curr Pharm Res. 2011;3(4):27-33.  Back to cited text no. 12
    
13.
Sainy J, Sharma R. QSAR analysis of thiolactone derivatives using HQSAR, CoMFA and CoMSIA. SAR QSAR Environ Res. 2015;26(10):873-892. DOI: 10.1080/1062936X.2015.1095238.  Back to cited text no. 13
    
14.
Todeschini R, Consonni V. Handbook of Molecular Descriptors. Weinheim: WileyVCH; 2008. pp. 516-520. DOI: 10.1002/9783527613106.  Back to cited text no. 14
    
15.
Wan J, Zhang L, Yang G, Zhan CG. Quantitative structure-activity relationship for cyclic imide derivatives of protoporphyrinogen oxidase inhibitors: a study of quantum chemical descriptors from density functional theory. J Chem Inf Comput Sci. 2004;44(6):2099-2105. DOI: 10.1021/ci049793p.  Back to cited text no. 15
    
16.
Hamann M, Alonso D, Martín-Aparicio E, Fuertes A, Pérez-Puerto MJ, Castro A, et al. Glycogen synthase kinase-3 (GSK-3) inhibitory activity and structure-activity relationship (SAR) studies of the manzamine alkaloids. Potential for Alzheimer’s disease. J Nat Prod. 2007;70(9):1397-1405. DOI: 10.1021/np060092r.  Back to cited text no. 16
    
17.
Akhtar MJ, Siddiqui AA, Khan AA, Ali Z, Dewangan RP, Pasha S, et al. Design, synthesis, docking and QSAR study of substituted benzimidazole linked oxadiazole as cytotoxic agents, EGFR and erbB2 receptor inhibitors. Eur J Med Chem. 2017;126:853-869. DOI: 10.1016/j.ejmech.2016.12.014.  Back to cited text no. 17
    
18.
Ray S, De K, Sengupta C, Roy K. QSAR study of lipid peroxidation-inhibition potential of some phenolic antioxidants. Indian J Biochem Biophys. 2008;45:198-205.  Back to cited text no. 18
    
19.
Hemmateenejad B, Emami L, Sharghi H. Multi-wavelength spectrophotometric determination of acidity constant of some newly synthesized Schiff bases and their QSPR study. Spectrochim Acta A Mol Biomol Spectrosc. 2010;75(1):340-346. DOI: 10.1016/j.saa.2009.10.037.  Back to cited text no. 19
    
20.
Mojaddami A, Sakhteman A, Fereidoonnezhad M, Faghih Z, Najdian A, Khabnadideh S, et al. Binding mode of triazole derivatives as aromatase inhibitors based on docking, protein ligand interaction fingerprinting, and molecular dynamics simulation studies. Res Pharm Sci. 2017;12(1):21-30. DOI: 10.4103/1735-5362.199043.  Back to cited text no. 20
    
21.
Todeschini R. Milano Chemometrics and QSAR Research Group. Available from: https://michem.unimib.it/.  Back to cited text no. 21
    
22.
Emami L, Faghih Z, Sakhteman A, Rezaei Z, Faghih Z, Salehi F, et al. Design, synthesis, molecular simulation, and biological activities of novel quinazolinone-pyrimidine hybrid derivatives as dipeptidyl peptidase-4 inhibitors and anticancer agents. New J Chem. 2020;44(45):19515-19531. DOI: 10.1039/D0NJ03774E.  Back to cited text no. 22
    
23.
Ashok UP, Kollur SP, Arun BP, Sanjay C, Suresh KS, Anil N, et al. In vitro anticancer activity of 4(3H)-quinazolinone derived Schiff base and its Cu (II), Zn (II) and Cd (II) complexes: preparation, X-ray structural, spectral characterization and theoretical investigations. Inorganica Chim Acta. 2020;511:119846. DOI: 10.1016/j.ica.2020.119846.  Back to cited text no. 23
    
24.
Fassihi A, Abedi D, Saghaie L, Sabet R, Fazeli H, Bostaki G, et al. Synthesis, antimicrobial evaluation and QSAR study of some 3-hydroxypyridine-4-one and 3-hydroxypyran-4-one derivatives. Eur J Med Chem. 2009;44(5):2145-2157. DOI: 10.1016/j.ejmech.2008.10.022.  Back to cited text no. 24
    
25.
Fassihi A, Sabet R. QSAR study of p56lck protein tyrosine kinase inhibitory activity of flavonoid derivatives using MLR and GA-PLS. Int J Mol Sci. 2008;9(9):1876-1892. DOI: 10.3390/ijms9091876.  Back to cited text no. 25
    
26.
Sabet R, Fassihi A. QSAR study of antimicrobial 3-hydroxypyridine-4-one and 3-hydroxypyran-4-one derivatives using different chemometric tools. Int J Mol Sci. 2008;9(12):2407-2423. DOI: 10.3390/ijms9122407.  Back to cited text no. 26
    
27.
Tu Y, OuYang Y, Xu S, Zhu Y, Li G, Sun C, et al. Design, synthesis, and docking studies of afatinib analogs bearing cinnamamide moiety as potent EGFR inhibitors. Bioorg Med Chem. 2016;24(7):1495-1503. DOI: 10.1016/j.bmc.2016.02.017.  Back to cited text no. 27
    
28.
OuYang Y, Zou W, Peng L, Yang Z, Tang Q, Chen M, et al. Design, synthesis, antiproliferative activity and docking studies of quinazoline derivatives bearing 2,3-dihydro-indole or 1,2,3,4-tetrahydroquinoline as potential EGFR inhibitors. Eur J Med Chem. 2018;154:29-43. DOI: 10.1016/j.ejmech.2018.05.006.  Back to cited text no. 28
    
29.
Abbas SE, Barsoum FF, Georgey HH, Mohammed ER. Synthesis and antitumor activity of certain 2,3,6-trisubstituted quinazolin-4 (3H)-one derivatives. Bull Fac Pharm Cairo Univ. 2013;51(2):273-282. DOI: 10.1016/j.bfopcu.2013.08.003.  Back to cited text no. 29
    
30.
Abdel-Fattah HA, El-Etrawy AS, Gabr NRM. Synthesis and biological evaluation of some new 1,3,4-oxa, thiadiazole and 1,2,4-triazole derivatives attached to benzimidazole. Int J Pharm Chem. 2014;04(03):112-118. DOI:10.7439/IJPC.V4I3.91.  Back to cited text no. 30
    
31.
El-Azab AS, Al-Omar MA, Alaa AM, Abdel-Aziz NI, Magda AA, Aleisa AM, et al. Design, synthesis and biological evaluation of novel quinazoline derivatives as potential antitumor agents: molecular docking study. Eur J Med Chem. 2010;45(9):4188-4198. DOI: 10.1016/j.ejmech.2010.06.013.  Back to cited text no. 31
    
32.
Ghorab MM, Alsaid MS. Novel 4-aminoquinazoline derivatives as new leads for anticancer drug discovery. Acta Pharm. 2015;65(3):299-309. DOI: 10.1515/acph-2015-0021.  Back to cited text no. 32
    
33.
Madhavi S, Sreenivasulu R, Yazala JP, Raju RR. Synthesis of chalcone incorporated quinazoline derivatives as anticancer agents. Saudi Pharm J. 2017;25(2):275-279. DOI: 10.1016/j.jsps.2016.06.005.  Back to cited text no. 33
    
34.
Koljonen J, Nordling TE, Alander JT. A review of genetic algorithms in near infrared spectroscopy and chemometrics: past and future. J Near Infrared Spec. 2008;16(3):189-197. DOI: 10.1255/jnirs.778.  Back to cited text no. 34
    
35.
Sabet R, Fassihi A, Saghaie L. Octanol-water partition coefficients determination and QSPR study of some 3-hydroxy pyridine-4-one derivatives. J Pharm Res Int. 2018;22(4):1-15. DOI: 10.9734/JPRI/2018/41142.  Back to cited text no. 35
    
36.
Salentin S, Schreiber S, Haupt VJ, Adasme MF, Schroeder M. PLIP: fully automated protein-ligand interaction profiler. Nucleic Acids Res. 2015;43(Web Server issue):W443-W447. DOI: 10.1093/nar/gkv315.  Back to cited text no. 36
    
37.
Fereidoonnezhad M, Faghih Z, Mojaddami A, Rezaei Z, Sakhteman A. A comparative QSAR analysis, molecular docking and PLIF studies of some N-arylphenyl-2,2-dichloroacetamide analogues as anticancer agents. Iran J Pharm Res. 2017;16(3):981-998. PMID: 29535790.  Back to cited text no. 37
    
38.
Emami L, Sabet R, Sakhteman A, Zade MK. Quantitative structure-activity relationship and molecular docking studies of imidazolopyrimidine amides as potent dipeptidyl peptidase-4 (DPP4) inhibitors. J Pharm Res Int. 2019;27(6):1-15. DOI: 10.9734/jpri/2019/v27i630186.  Back to cited text no. 38
    
39.
Sabet R, Khabnadideh S, Fathi D, Emami L. QSAR and docking study of isatin analogues as cytotoxic agents. J Pharm Res Int. 2019;27(5):1-22. DOI: 10.9734/jpri/2019/v27i530184.  Back to cited text no. 39
    
40.
Morris GM, Huey R, Olson AJ. Using autodock for ligand-receptor docking. Curr Protoc Bioinformatics. 2008;8:Unit 8.14. DOI: 10.1002/0471250953.bi0814s24.  Back to cited text no. 40
    
41.
Zarenezhad E, Esmaielzadeh S, Behrouz S, Emami L. Copper Schiff base complex as a new immobilized heterogeneous catalyst: experimental, theoretical, biological and docking study. J Iran Chem Soc. 2021;18:283-295. DOI: 10.1007/s13738-020-02024-8.  Back to cited text no. 41
    
42.
Ferreira LG, Dos Santos RN, Oliva G, Andricopulo AD. Molecular docking and structure-based drug design strategies. Molecules. 2015;20(7):13384-13421. DOI: 10.3390/molecules200713384.  Back to cited text no. 42
    
43.
Lo YC, Rensi SE, Torng W, Altman RB. Machine learning in chemoinformatics and drug discovery. Drug Discov Today. 2018;23(8):1538-1546. DOI: 10.1016/j.drudis.2018.05.010.  Back to cited text no. 43
    
44.
van de Waterbeemd H, Timmerman H, Mannhold R, Krogsgaard-Larsen P. Chemometric Methods in Molecular Design. John Wiley & Sons; 2008. pp. 1-379. DOI: 10.1002/9783527615452.  Back to cited text no. 44
    
45.
Mahani NM, Zarandi AM, Horzadeh A. QSAR studies of novel iminochromene derivatives as carbonyl reductase 1 (CBR1) inhibitors. Marmara Pharm J. 2018;22(2):227-236. DOI: 10.12991/mpj.2018.60.  Back to cited text no. 45
    
46.
Qian Y, Wang Y, Sa R, Yan H, Pan X, Yang Y, et al. Metabolic fingerprinting of Angelica sinensis during growth using UPLC-TOFMS and chemometrics data analysis. Chem Cent J. 2013;7(1):42-51. DOI: 10.1186/1752-153X-7-42.  Back to cited text no. 46
    
47.
Pasha FA, Muddassar M, Srivastava AK, Cho SJ. In silico QSAR studies of anilinoquinolines as EGFR inhibitors. J Mol Model. 2010;16(2):263-277. DOI: 10.1007/s00894-009-0534-x.  Back to cited text no. 47
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8], [Table 9], [Table 10], [Table 11]


This article has been cited by
1 Fluconazole-Like Compounds as Potential Antifungal Agents: QSAR, Molecular Docking, and Molecular Dynamics Simulation
Farnaz Salehi, Leila Emami, Zahra Rezaei, Soghra Khabnadideh, Behnaz Tajik, Razieh Sabet, Josefina Pons
Journal of Chemistry. 2022; 2022: 1
[Pubmed] | [DOI]
2 Efficient synthesis of 1,3-naphtoxazine derivatives using reusable magnetic catalyst (GO-Fe3O4–Ti(IV)): anticonvulsant evaluation and computational studies
Soghra Khabnadideh, Aida solhjoo, Reza Heidari, Leila Amiri Zirtol, Amirhossein Sakhteman, Zahra Rezaei, Elaheh Babaei, Samaneh Rahimi, Leila Emami
BMC Chemistry. 2022; 16(1)
[Pubmed] | [DOI]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
   Abstract
    Introduction
    Material and Methods
    Results
    Discussion
    Conclusion
   References
   Article Figures
   Article Tables

 Article Access Statistics
    Viewed1076    
    Printed26    
    Emailed0    
    PDF Downloaded179    
    Comments [Add]    
    Cited by others 2    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]