Journal of Receptors and Signal Transduction

ISSN: 1079-9893 (Print) 1532-4281 (Online) Journal homepage: http://www.tandfonline.com/loi/irst20

QSAR and molecular docking studies on oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors Cong-Min Kang, Dong-Qing Liu, Xu-Hao Zhao, Ying-Jie Dai, Jia-Gao Cheng & Ying-Tao Lv To cite this article: Cong-Min Kang, Dong-Qing Liu, Xu-Hao Zhao, Ying-Jie Dai, Jia-Gao Cheng & Ying-Tao Lv (2015): QSAR and molecular docking studies on oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors, Journal of Receptors and Signal Transduction, DOI: 10.3109/10799893.2015.1049364 To link to this article: http://dx.doi.org/10.3109/10799893.2015.1049364

Published online: 29 Sep 2015.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=irst20 Download by: [University of Tasmania]

Date: 29 September 2015, At: 22:45

http://informahealthcare.com/rst ISSN: 1079-9893 (print), 1532-4281 (electronic) J Recept Signal Transduct Res, Early Online: 1–7 ! 2015 Taylor & Francis. DOI: 10.3109/10799893.2015.1049364

RESEARCH ARTICLE

QSAR and molecular docking studies on oxindole derivatives as VEGFR2 tyrosine kinase inhibitors Cong-Min Kang1, Dong-Qing Liu1, Xu-Hao Zhao1, Ying-Jie Dai1, Jia-Gao Cheng2, and Ying-Tao Lv1 College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao, P.R. China and 2Shanghai Key Laboratory of Chemical Biology, School of Pharmacy, East China University of Science and Technology, Shanghai, P.R. China

Downloaded by [University of Tasmania] at 22:45 29 September 2015

1

Abstract

Keywords

The three-dimensional quantitative structure–activity relationships (3D-QSAR) were established for 30 oxindole derivatives as vascular endothelial growth factor receptor-2 (VEGFR-2) tyrosine kinase inhibitors by using comparative molecular field analysis (CoMFA) and comparative similarity indices analysis comparative molecular similarity indices analysis (CoMSIA) techniques. With the CoMFA model, the cross-validated value (q2) was 0.777, the non-cross-validated value (R2) was 0.987, and the external cross-validated value (Q2ext ) was 0.72. And with the CoMSIA model, the corresponding q2, R2 and Q2ext values were 0.710, 0.988 and 0.78, respectively. Docking studies were employed to bind the inhibitors into the active site to determine the probable binding conformation. The binding mode obtained by molecular docking was in good agreement with the 3D-QSAR results. Based on the QSAR models and the docking binding mode, a set of new VEGFR-2 tyrosine kinase inhibitors were designed, which showed excellent predicting inhibiting potencies. The result revealed that both QSAR models have good predictive capability to guide the design and structural modification of homologic compounds. It is also helpful for further research and development of new VEGFR-2 tyrosine kinase inhibitors.

3D-QSAR, CoMFA, CoMSIA, molecular docking, VEGFR-2 tyrosine kinase inhibitors

Introduction Angiogenesis is a complex and highly regulated process that is crucial for tumor growth and metastasis, and a substantial number of growth factors and cytokines have been identified in recent years that activate and maintain angiogenesis throughout tumorigenesis, of which vascular endothelial growth factor receptor-2 (VEGFR-2) is the most dominant player (1–3). Inhibition of VEGFR-2 tyrosine kinase could suppress both angiogenesis and tumor growth in vivo (4,5). Consequently, a number of small molecules with VEGFR tyrosine kinase inhibitory properties have been developed (6). Many of these have been evaluated as potent inhibitors and some are currently in clinical trials for various angiogenic related disorders (7). The principal compound of oxindole derivatives reaching the clinic was sunitinib, which has been approved and marketed for the treatment of renal cell carcinoma (8). Most oxindole derivatives have a low molecular weight and bind to the ATP binding site of protein kinases, competing with ATP. Moreover, the structures and activities of oxindole derivatives as kinase inhibitors were summarized by Prakash et al. (9).

Address for correspondence: Ying-Tao Lv, College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao 266042, P.R. China. Tel: +86 53284023290. E-mail: [email protected]

History Received 1 April 2015 Accepted 4 May 2015 Published online 28 September 2015

In this paper, two different three-dimensional quantitative structural activity relationship (3D-QSAR) studies, comparative molecular field analysis (CoMFA) (10) and comparative molecular similarity indices analysis (CoMSIA) (11) were carried out on a set of active oxindole derivatives that exhibit activity against VEGFR-2 tyrosine kinase. Molecular docking (12) was applied to study the interactions between oxindole derivatives and VEGFR-2 tyrosine kinase. The developed models not only can help understand the SAR of the oxindole derivatives, but also guide us to design novel excellent inhibitors.

Materials and methods Data set selection A total set of 30 oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors were collected from literatures (13,14). All molecules contain the same common template, i.e. oxindole ring. The biological activity of each molecule in data set was assayed using the same experimental protocol. The reported biological activities (IC50) of all molecules were converted into corresponding pIC50 values (log10 IC50) and utilized as dependent variables for the development of QSAR model. The chemical structures of all molecules along with their corresponding experimental biological activity (pIC50) values are displayed in Table 1, in which 25 compounds were used as training set and 5 compounds labeled with ‘‘*’’ were

2

C.-M. Kang et al.

J Recept Signal Transduct Res, Early Online: 1–7

Table 1. Structures and activities of VEGFR-2 tyrosine kinase inhibitors. Predicted Activity (pIC50) Compound series

No.

1

n

O

NH O

Downloaded by [University of Tasmania] at 22:45 29 September 2015

NH

R2

CoMFA

CoMSIA

2

R (CH2)2N(C2H5)2

R n 5-F 1

8.30

8.10

8.26

2

(CH2)2N(C2H5)2

5-Cl 1

8.00

8.14

8.11

3*

(CH2)2N(C2H5)2

5-Br 1

7.80

8.04

8.01

4

(CH2)2N(CH3)2

5-F 1

8.70

8.52

8.45

5

(CH2)2N(CH2)2

5-F 1

8.10

7.96

8.22

6

(CH2)2N(C2H4)O

5-F 1

7.60

7.79

7.65

7

(R)-CH2CH(OH)CH2N(C2H4)2O

5-F 1

9.00

9.07

9.06

8

(CH2)2N(C2H5)2

5-F 2

7.85

7.69

7.73

9

(CH2)2N(C2H4)2O

5-F 2

9.00

9.00

8.95

10

(R)-CH2CH(OH)CH2N(C2H4)2O

5-F 2

7.57

7.59

7.56

11

(CH2)2N(C2H5)2

5-F 3

6.95

7.02

7.04

R

2

N

O HN

NH

R

12

NO2

7.32

7.41

7.36

13

COOMe

7.44

7.21

7.15

14

COOEt

6.96

7.14

7.15

15

Cl

6.89

6.97

7.07

16*

NH2

6.88

6.95

6.83

17

CN

6.61

6.65

6.56

18*

CONH2

6.12

6.08

6.17

19

Pyrrol-1-yl

6.10

6.06

5.92

20

CONHipr

5.91

5.88

6.05

21

CONMe2

5.88

5.81

5.86

22

COCH3

5.76

5.80

5.72

23

CONHCH3

5.68

5.66

5.72

24

H

6.21

6.34

6.28

25 4-(NCH3)COCH2-(4-methyl-piperazin-1-yl)

8.30

8.18

8.28

26

4-(NCH3)COCH2-(imidazol-1-yl)

8.22

8.12

8.22

27

4-(NCOCH3)CH2CONMe2

8.10

7.94

8.16

28*

4-(NCOCH3)(CH2)3NMe2

8.05

7.98

7.95

29

4-(NCOCH3)(CH2)2NMe2

7.92

7.80

7.94

30

4-(NSO2CH3)(CH2)2NMe2

7.64

7.79

7.56

R

3

R2

NH O

NH

O

Observed activity (pIC50)

1*

1 R1

O

Groups

QSAR and molecular docking studies on oxindole derivatives

DOI: 10.3109/10799893.2015.1049364

randomly selected as a test set. Training set molecules were utilized for the development of QSAR model, whereas test set molecules were used for the validation of the generated model.

Downloaded by [University of Tasmania] at 22:45 29 September 2015

Molecular alignment All calculations were carried out using SYBYL 7.0 from Tripos (St. Louis, MO, USA). Structures of all the compounds were sketched with Sketch module in SYBYL and minimized with minimize module by utilizing Tripos force field with the Gasteiger–Huckel charges (15) and conjugated gradient method, and gradient convergence criteria of 0.05 kcal/mol (16). Finally, the optimized structures of all compounds were got, and then were employed for molecular alignment and 3DQSAR analysis. The training set was aligned using the align database module. The most active compound (compound 9) was used as an alignment template molecule and the rest molecules of the training set were aligned to it by utilizing the common substructure as shown in Figure 1. Superimposition of all the training set molecules is shown in Figure 2. CoMFA Models of steric and electrostatic fields for CoMFA were based on both Lennard-Jones and Coulomb potentials (10). The steric and electrostatic fields were calculated using the Tripos force field at each lattice intersection for the aligned ˚ . A sp3 carbon atom molecules with a grid spacing of 2.0 A ˚ was used as a having a charge of +1 and a radius of 1.52 A probe to calculate the steric and electrostatic fields. The energy cut-off value was set to 30 kcal/mol and the column filtering value was set to 2.0 kcal/mol to improve the signal to noise ratio by omitting those lattice points whose energy

O

4'

5'

N

1' *

*

2' * * 3'

*

*

N H

*

N* H

*

variation was below this threshold. With standard options for scaling of variables, the regression analysis was carried out by using the full cross-validated partial least squares (PLS) method (17,18). The final model, non-cross-validated conventional analysis, was developed with the optimum number of components to yield a non-cross-validated R2 value. CoMSIA In CoMSIA interaction energy calculation, the steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor potential fields were calculated at each lattice intersection of the same lattice box used for CoMFA calculations (19). In this study, standard settings of CoMSIA were utilized to calculate the steric, electrostatic, hydrophobic, donor and acceptor fields. A sp3 carbon probe ˚ , hydrophobicity atom with a charge of +1.0, a radius of 1.0 A +1.0, and H-bond donor and acceptor property +1 were used to calculate the respective fields. PLS analysis The relationship between the structural parameters and the biological activities has been quantified by the PLS algorithm. PLS method was used to correlate oxindole derivatives inhibitory activity of the CoMFA and CoMSIA fields to derive 3D-QSAR models. Cross validation analysis was performed internally using leave one out (LOO) method (20) in which one compound is removed from the data set and its activity is predicted using the model derived from the rest of the data set (21). The cross validation q2 that resulted in optimum numbers of components and the lowest standard error of prediction was considered for further analysis. Final analysis was performed to calculate non-cross-validated R2 using the optimum number of components. The cross validation q2, Fischer’s statistic (F-test), standard error of estimate (SEE) and predicted R2 were calculated.

N

External validation

6'

F

3

O

Figure 1. The structure of template molecule.

O

The internal validations of QSAR models were used the LOO method, and the external validations were through the Q2ext (22). External validations of various models were performed by using a test set of five molecules. Pn ðyi  ^yi Þ2 Q2ext ¼ 1  Pni¼1 , ytr Þ2 i¼1 ðyi   where yi is the observed activity (pIC50) of the test set, ^yi is the predicted activity of the test set and ytr is the mean value of observed activity of training set. Molecular docking

Figure 2. Image of superimposed all training set molecules.

The crystal structure of VEGFR-2 (PDB code: 4AGD (23)) was downloaded from the Protein Data Bank (24). Before docking, the receptor was prepared based on PDB 4AGD (complex with sunitinib), with all ligands and water molecules removed and with hydrogens and charges added (using Sybyl). The Autodock Vina (25) was used to get the affinity energy and the binding mode of VEGFR-2 tyrosine kinase with oxindole derivatives.

4

C.-M. Kang et al.

J Recept Signal Transduct Res, Early Online: 1–7

Table 2. The statistical parameters of the best CoMFA and CoMSIA models. Model

N

q2

R2

F

SEE

Steric

Electrostatic

Hydrophobic

CoMFA CoMSIA

6 6

0.777 0.710

0.987 0.988

53.484 72.390

0.213 0.225

0.531 0.220

0.469 0.476

– 0.304

N is the number of components from PLS analysis, q2 is the correlation coefficient of the LOO crossvalidation, R2 is the non-cross-validation coefficient, F is the F-test value and SEE is the standard deviation of the regression, respectively.

Results and discussion

Downloaded by [University of Tasmania] at 22:45 29 September 2015

CoMFA analysis The CoMFA model showed LOO cross-validation q2 ¼ 0.777 with six optimum components, non-cross-validation R2 ¼ 0.987, F (F-test) value ¼ 53.484, SEE ¼ 0.213; the steric and electrostatic contributions were 53.1% and 46.9%, respectively (Table 2). Often, a high q2 value (q240.5) is considered as a proof of high predictive ability of the model (26). The developed model was found to be statistically significant due to its high cross-validation q2. CoMSIA analysis The CoMSIA analysis was performed by using steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor fields. Various models were developed using a combination of different fields and the statistically significant model is reported here as in Table 2. The best CoMSIA model showed cross-validation q2 ¼ 0.710 with six optimum components, non-cross-validation R2 ¼ 0.988, F value ¼ 72.390, SEE ¼ 0.225; the steric, electrostatic and hydrophobic contributions were 22.0%, 47.6% and 30.4%, respectively. Validation of QSAR models 3D-QSAR models were evaluated for their predictive abilities by the test set molecules. External validation was considered to be the most acceptable validation method for a predictive QSAR. To test the stability and predictive ability of the 3D-QSAR model, five compounds were selected as the test set for validation. The predicted pIC50 for this test set is listed in Table 1. With the CoMFA model, the external cross-validated value (Q2ext ) was 0.72. And with the CoMSIA model, the Q2ext value was 0.78. The correlation between experimental and predicted pIC50 for training set is shown in Figure 3. It was shown that the CoMFA and CoMSIA model were all stable and provided robust predictive ability. CoMFA contour plots The CoMFA and CoMSIA contour plots were generated to visualize the information content of the derived 3D-QSAR models. These contour plots showed regions in 3D space around the molecules, where variation in specific molecular properties increase or decrease the activity. CoMFA steric and electrostatic contour plots are shown in Figure 4 using compound 9 as a reference molecular. In Figure 4(a), the green contours indicate regions, where bulky groups will

Figure 3. Plots of experimental and calculated pIC50 for CoMFA and CoMSIA analysis.

increase the activity, while the yellow contours indicate regions of steric hindrance to activity. In Figure 4(b), blue contours indicate regions, where electron-donating groups will increase activity, and red contours indicate regions, where electron-withdrawing groups will increase the activity. As shown in Figure 4(a), the 50 site of the compound 9 was oriented toward green contours indicating that a bulky substituent would be favored. This may explain why compounds 1, 3 and 28, which possess bulky substituents at 50 position have significantly better activity than compounds 16 and 18, which have minor groups at the 50 position. The yellow contours at the 20 site of the compound 9 indicated that bulky groups at this position might decrease the activity.

DOI: 10.3109/10799893.2015.1049364

Downloaded by [University of Tasmania] at 22:45 29 September 2015

Figure 4. Contour plots of CoMFA steric field (a) and electrostatic field (b) analysis in combination with compound 9. Compound 9 is depicted in stick representation, colored by atom type (white C, blue N, red O, cyan H). Steric fields (a): green contours regions where bulky groups increase activity, yellow contours regions where bulky groups decrease activity. Electrostatic fields (b): blue contours regions where electron-donating groups increase activity, red contours regions where electron-withdrawing groups increase activity.

Figure 5. Contour plots of CoMSIA analysis in combination with compound 9. Compound 9 is depicted in stick representation, colored by atom type (white C, blue N, red O, cyan H). Steric fields (a): green contours regions where bulky groups increase activity, yellow contours regions where bulky groups decrease activity. Electrostatic fields (b): blue contours regions where electron-donating groups increase activity, red contours regions where electron-withdrawing groups increase activity. Hydrophobic fields (c): yellow contours regions where hydrophobic groups increase activity, white contours regions where hydrophobic groups decrease activity.

QSAR and molecular docking studies on oxindole derivatives

5

6

C.-M. Kang et al.

J Recept Signal Transduct Res, Early Online: 1–7

well compared with that of the electrostatic contour plot of CoMFA. The similar pattern of electrostatic contour plot in CoMFA and CoMSIA indicated that the electrostatic interactions between the active site of VEGFR-2 and substituents present in these regions were crucial for the inhibitory activity. The hydrophobic field contour plot is shown in Figure 5(c) using compound 9 as a reference structure, white and yellow contours indicate regions, where hydrophilic and hydrophobic properties were favored. The yellow contours at the 20 and 50 site of the compound 9 indicated that hydrophobic group at this region could increase the activity. This shows why compounds 16 and 18, which possess hydrophilic groups at 20 site, showed significant decreased activities, and compounds 1, 3 and 28, which possess hydrophobic groups at 50 site, displayed much better potency.

Downloaded by [University of Tasmania] at 22:45 29 September 2015

Docking analysis Compound 9, the most active compound in the data set, was used to investigate the binding mode of the oxindole inhibitors. As shown in Figure 6, the oxindole ring inserted into the end of the binding pocket, which suggested that bulky substituents will have steric hindrance with amino acid residues and then decrease the potency, such as compounds 19 and 21 have a low activity. The 50 site of the compound is at the start of the binding pocket, which exposes to solvents and should be bulky and hydrophobic, such as compounds 4, 7 and 9, which have bulky substituents at 50 site, showed significant activities. The observations gained from molecular docking were in agreement with those of QSAR models. Figure 6. Docked binding modes of compound 9 in the binding site of receptor (PDB code 4AGD). Compound 9 is shown in stick representation.

Design for novel inhibitors based on 3D-QSAR and docking studies

This may be the reason why the compounds 1 and 3 have significantly superior activity. The CoMFA electrostatic field contour plot is shown in Figure 4(b). A blue contour near the 10 site of the compound 9 revealed that the increase in the positive charge will result in an increase in the activity. Comparing compound 16, which possesses electron donating group at 10 site with compound 18, which possesses electron withdrawing group, it was found that their activity discrepancies can be explained easily by this blue contour.

Based on the QSAR and molecular docking results, 12 new compounds were designed. When the 50 site of designed compound was substituted by bulky groups, which also were hydrophobic, the compound will have good activity. If the 20 site of designed compound was substituted with electron donating substituents, the activity of designed compounds will be increased. But, if the substituents were too large, the activity will be decreased. The activities of the new designed compounds were evaluated by the established QSAR models. The structures and predicted activities of designed compounds are listed in Table 3. All compounds showed good predicted activities.

CoMSIA contour plots

Conclusion

The CoMSIA contours plots derived using steric, electrostatic and hydrophobic fields, which are presented in Figure 5. CoMSIA steric and electrostatic contours are more or less similar to those of CoMFA. The steric field contour plot of the CoMSIA is shown in Figure 5(a). The green contours regions of the CoMSIA steric contour plot could be well compared with steric contour plot of the CoMFA, which indicated that bulky groups were essentially required for the steric interactions with receptor active sites. The electrostatic field contour of the CoMSIA is displayed in Figure 5(b). Its electropositive favorable blue contours regions can also be

In this study, 3D-QSAR and molecular docking have been successfully applied to a set of oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors. The predictive abilities of QSAR have been shown by the external validation, which also been confirmed by molecular docking. For the oxindole derivatives, the 50 site should have bulky and hydrophobic groups and the 10 , 20 site should have small donating substituents. A number of novel derivatives were designed by using the QSAR and molecular docking results. The predicted activities of these newly designed molecules may be reliable. The results obtained from 3D-QSAR and molecular

QSAR and molecular docking studies on oxindole derivatives

DOI: 10.3109/10799893.2015.1049364

7

Table 3. Structures and predicted activities of designed compounds. Predicted activity (pIC50) Designed compounds R1 O

NH O

Downloaded by [University of Tasmania] at 22:45 29 September 2015

NH

R2

R2

CoMFA

CoMSIA

9.26

9.15

No.

R1

D1

–CH2CH(OH)CH2C6H5

D2

–CH2CH(OH)CH2CH2C6H5

5-F

9.31

9.17

D3

–CH2CH(OH)CH2CH2C6H5

6-OH

9.12

8.95

D4

–CH2CH(OH)CH2CH2C6H5

6-NH2

9.32

9.27

D5

–CH2CH(OH)CH2C6H5

6-OH

8.85

8.83

D6

–CH2CH(OH)CH2C6H5

6-NH2

9.07

9.01

D7

–CONHC6H5

6-NH2

9.10

9.06

D8

–CONHC6H5

6-OH

8.98

9.04

5-F

D9

–CONHC6H5

5-F

9.14

9.03

D10

–CONH(-6-benzimidazole)

5-F

9.32

9.13

D11

–CONH(-6-benzimidazole)

6-OH

9.34

9.27

D12

–CONH(-6-benzimidazole)

6-NH2

9.35

9.31

docking studies can be served as a useful guideline for further modification of oxindole derivatives, which function as VEGFR-2 tyrosine kinase inhibitors.

Declaration of interest The project was supported by the National Natural Science Foundation of China (21072111, 21172070, 21272131) and Shandong Provincial Natural Science Foundation, China (ZR2011BM015).

References 1. Cai Z-W, Wei D, Borzilleri RM, et al. Synthesis, SAR, and evaluation of 4-[2,4-difluoro-5-(cyclopropylcarbamoyl)phenylamino] pyrrolo [2,1-f][1,2,4] triazine-based VEGFR-2 kinase inhibitors. Bioorg Med Chem Lett 2008;18:1354–8. 2. Ramsden J. Angiogenesis in the thyroid gland. J Endocrinol 2000; 166:475–80. 3. Li D, Xie K, Ding G, et al. Tumor resistance to anti-VEGF therapy through up-regulation of VEGF-C expression. Cancer Lett 2014; 346:45–52. 4. Cao J, Zhang J, Wang Z, et al. Hypothyroidism as a potential biomarker of efficacy of famitinib, a novel VEGFR-2 inhibitor in metastatic breast cancer. Cancer Chemother Pharmacol 2014;74: 389–98. 5. Patel A, Fenton C, Ramirez R, et al. Tyrosine kinase expression is increased in papillary thyroid carcinoma of children and young adults. Front Biosci 2000;5:A1–9. 6. Shibuya M. Vascular endothelial growth factor and its receptor system: physiological functions in angiogenesis and pathological roles in various diseases. J Biochem 2013;153:13–9. 7. Sharma PS, Sharma R, Tyagi T. VEGF/VEGFR pathway inhibitors as anti-angiogenic agents: present and future. Curr Cancer Drug Targets 2011;11:624–53. 8. Chow LQM, Eckhardt SG. Sunitinib: from rational design to clinical efficacy. J Clin Oncol 2007;25:884–96. 9. Prakash CR, Theivendren P, Raja S. Indolin-2-ones in clinical trials as potential kinase inhibitors: a review. Pharmacol Pharm 2012;3: 62–71. 10. Cramer RD, Patterson DE, Bunce JD. Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. J Am Chem Soc 1988;110:5959–67. 11. Klebe G, Abraham U, Mietzner T. Molecular similarity indices in a comparative analysis (CoMSIA) of drug molecules to correlate and predict their biological activity. J Med Chem 1994;37:4130–46.

12. Park H, Jeon TJ, Chien PN, et al. Discovery of novel DUSP4 inhibitors through the virtual screening with docking simulations. Bull Korean Chem Soc 2014;35:2655–9. 13. Tang PC, Su YD, Feng J, et al. Novel potent orally active multitargeted receptor tyrosine kinase inhibitors: synthesis, structure–activity relationships, and antitumor activities of 2-indolinone derivatives. J Med Chem 2010;53:8140–9. 14. Roth GJ, Heckel A, Colbatzky F, et al. Design, synthesis, and evaluation of indolinones as triple angiokinase inhibitors and the discovery of a highly specific 6-methoxycarbonyl-substituted indolinone (BIBF 1120). J Med Chem 2009;52:4466–80. 15. Gasteiger J, Marsili M. Iterative partial equalization of orbital electronegativity: a rapid access to atomic charges. Tetrahedron 1980;36:3219–22. 16. Sivan SK, Manga V. Molecular docking and 3D-QSAR studies on triazolinone and pyridazinone, non-nucleoside inhibitor of HIV-1 reverse transcriptase. J Mol Model 2010;16:1169–78. 17. McIntosh A, Bookstein F, Haxby JV, et al. Spatial pattern analysis of functional brain images using partial least squares. Neuroimage 1996;3:143–57. 18. Wold S, Sjostrom M, Eriksson L. PLS-regression: a basic tool of chemometrics. Chemom Intell Lab Syst 2001;58:109–30. 19. Wu H-Q, Yao J, He Q-Q, et al. Docking-based CoMFA and CoMSIA studies on naphthyl-substituted diarylpyrimidines as NNRTIs. SAR QSAR Environ Res 2014;25:761–75. 20. Bush BL, Nachbar Jr RB. Sample-distance partial least squares: PLS optimized for many variables, with application to CoMFA. J Comput Aided Mol Des 1993;7:587–619. 21. Kumar R, Kumar A, Jain S, et al. Synthesis, antibacterial evaluation and QSAR studies of 7-[4-(5-aryl-1,3,4-oxadiazole-2-yl)piperazinyl] quinolone derivatives. Eur J Med Chem 2011;46:3543–50. 22. Alexander G, Alexander T. Beware of Q2. J Mol Graph Model 2002;20:269–76. 23. McTigue M, Murray BW, Chen JH, et al. Molecular conformations, interactions, and properties associated with drug efficiency and clinical performance among VEGFR TK inhibitors. Proc Natl Acad Sci USA 2012;109:18281–9. 24. Berman HM, Battistuz T, Bhat T, et al. The protein data bank. Acta Crystallogr D Biol Crystallogr 2002;58:899–907. 25. Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 2010;31:455–61. 26. Bo¨hm M, Stu¨rzebecher J, Klebe G. Three-dimensional quantitative structure–activity relationship analyses using comparative molecular field analysis and comparative molecular similarity indices analysis to elucidate selectivity differences of inhibitors binding to trypsin, thrombin, and factor Xa. J Med Chem 1999;42:458–77.

QSAR and molecular docking studies on oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors.

The three-dimensional quantitative structure-activity relationships (3D-QSAR) were established for 30 oxindole derivatives as vascular endothelial gro...
1KB Sizes 4 Downloads 16 Views