In Silico Studies on Biaryloxytriazoles as Antifungal Agents

 

Maryam Iman1, Yassamin EbrahimiNassimi2, Asghar Davood*2

 

1Chemical Injuries Research Center, Baqiyatallah University of Medical Sciences, Tehran, Iran.

2Department of Medicinal Chemistry, Pharmaceutical Sciences Branch, Islamic Azad University, Tehran, Iran

 

Received: 2014/05/5 �������������� Accepted: 2014/09/14

 

Abstract

Biaryloxy-substituted triazoles have been recognized as lanosterol 14α- demethylase inhibitor. In this research a group of newly synthesized biaryoxytriazoles with CYP51 inhibitor activities that possessing a variety of substituents at the different positions of the phenyl ring, were subjected to docking modeling and Quantitative Structure-Activity Relationship (QSAR) analysis.

All the desired substituted triazoles were built using Hyper Chem. The conformational studies and optimization were performed through MM+ and PM3 methods. Docking studies were performed using Autodock4.2. QSAR studies were done using SPSS and Matlab softwares.

The developed QSAR model indicates that the importance of RDF040m, nRNHR, Mor24u, H052 and Mor11p indices on antifungal activity. The sums of the RDF040m, Mor24u and nRNHR were identified as the most significant descriptors.

The docking results revealed that all of the compounds can interact with the 14α-demethylasemainly byazole-heme coordination.

The obtained QSAR equation can be used to design and estimate the lanosterol 14α-demethylase inhibitory activity of new derivatives of this series of compounds.

 

Key words: Triazoles, Antifungal, Molecular modeling, Lanosterol 14 α-demethylase, Docking, QSAR.


Introduction

Lanosterol 14α-Demethylase (14α-DM) is a cytochrome P450 enzyme that converts lanosterol to cholesterol. It is a target for azole antifungal drugs, inhibiting the production of ergosterol. Over-expression of CYP51A1 can lead to resistance to these antifungals (1).

Azoles bind to heme in CYP51 via the un-protonated nitrogen atom. The side chains in many triazole antifungals bear one or two 1, 4-disubstituted aryl substituent to improved antifungal activity.� Current researches mainly focused on the optimization of the side chain attached to the azole�s pharmacophore. Optimization led to new compounds with better biological or pharmacological properties (2-4).

Docking protocols aid in elucidation of the most energetically favorable binding pose of a ligand to its receptor. QSAR models are mathematical equations relating chemical structure to their biological activity. A hybrid QSAR/docking approach can be used to accelerating in silico high throughput screening and to explore the probable binding conformations of compounds at the active sites of the protein target (5-17).

This research investigated the combined molecular docking and QSAR approach to model the antifungal activity of a series of triazole derivatives bearing a variety of biaryloxy side chains (Figure 1). These compounds were designed by insertiona variety of hydrophilic substituents into the biphenyloxy side chain to improve their solubility and antifungal activity (18).

Figure 1. structure of biaryloxytriazole derivatives

Materials and Methods

Molecular Modeling

The chemical structure of desired triazole derivatives (Table 1) was constructed using Hyper Chem software (Version 7, hypercube Inc.). Molecular mechanic (MM+) and Semi-empirical molecular orbital calculations (PM3) of the structure were performed using the Hyper Chem program and among all energy minimal conformers the conformer with the lowest energy was considered in docking and QSAR studies.

Molecular docking

Docking studies were performed using AutoDock software (Version 4.2). The crystal structure of lanosterol 14α-Demethylase (14α-DM) was downloaded from the PDB bank server (PDB entry 1EA1) (19).


Table 1. Chemical structure of the desired triazole derivatives.

Compound

R

A

Compound

R

A

1

C

16

N

2

C

17

N

3

C

18

N

4

C

19

N

5

C

20

N

6

C

21

N

7

C

22

N

8

C

23

N

9

C

24

N

10

C

25

N

11

C

26

N

12

C

27

N

13

N

28

N

14

N

29

N

15

N

30

N


For the macromolecule, crystal structure of lanosterol 14α- demethylase, polar hydrogens were added and then kollman united atom charges were assigned. The grid maps of docking studies were computed using the AutoGrid included in to the AutoDock distribution. Finally docking was performed using the Lamarckian Genetic Algorithm method and the default parameters based on method that described previously (5-7, 16).

Data set and descriptor�s generation

The biological data used in this study are the anti-Candida albicans activities as minimum inhibition concentrations (MIC80) from a set of 30 biaryloxy-substituted triazoles derivatives (18) that was used for subsequent QSAR analysis as dependent variables. Using Kennard Stone algorithm, the data set was divided into two groups, training set (n = 24) and test set (n = 6) in which compounds 5, 15, 18, 22, 24 and 28 were considered as test set. A large number of molecular descriptors were calculated using Hyper Chem, AutoDock and Dragon software that were used as independent variables in QSAR model building. The Dragon program was used to calculate different functional groups, topological, geometrical, and constitutional descriptors for each molecule. Chemical parameters including molar volume (V), molecular surface area (SA), hydrophobicity (logP), hydration energy (HE) and molecular polarizability (MP) were calculated using HyperChem software (Table 2).

QSAR equations

QSAR ��techniques ��were ��used ��to� �develop correlations between� �biological activity and physicochemical
properties of a set of molecules (5-11,17). For this study, many descriptors were calculated, after� standardization of variables, in order to select the set of descriptors that are most relevant to the MIC80 of these agents, the multiple linear regression (MLR)with Step-wise variable selection method were used. QSAR equations were established using SPSS and Matlab softwares.

 

Results and discussion

Molecular modeling and docking

In current study we have tried to dock the inhibitors azoles 1-30 insight to 14α-DM. AutoDock 4.2, was used for this study. This program utilizes Lamarckian Genetic Algorithm for configurationally search and evaluates the energy using grid-based molecular affinity potential. Ten best configurations of the protein-inhibitors complexes were recovered on binding energy. The hydrogen bond and non-bonded contacts for the complexes were calculated using the AutoDock tools. Azole-Heme coordination, lowest energy and maximum number of conformations per cluster were set as the criteria to predict the binding modes of the compounds.

The results show that all of compounds interact with the 14α-demethylase and azole-heme coordination, π-π and π-cation interactions are involved in drug-receptor interaction (Figures2-5). The obtained results demonstrate that the coordination between nitrogen (N-3) of triazole and iron atom of heme is the most important interaction to inhibitory activity.


 

Table 2. Calculated physicochemical parameters of azoles 1-30 using the HyperChem software.

Comp.

V

Surface A (approx.)

Surface A (grid)

Log P

HE

MP

1

1165.69

565.16

680.3

4.19

3.36

44.39

2

1237.93

614.04

723.02

4.04

-10.58

47.49

3

1293.52

627.22

745.28

4.4

-8.06

49.32

4

1293.14

617.41

749.7

4.39

-9.76

49.32

5

1314.47

612.29

743.04

4.8

-8.81

51.16

6

1325.5

626.48

767.33

3.6

-15.95

49.96

7

1445.3

681.37

814.14

4.89

-9.08

56.53

8

1462.5

661.76

831.42

4.31

-9.22

57.15

9

1450.05

692.18

836.26

6.39

-12.65

57.24

10

1495.2

703.38

864.42

5.57

-17.64

57.87

11

1167.63

562.55

680.04

3.98

-15.63

44.94

12

1208.82

580.78

709.58

4.85

-21.93

45.87

13

1124.11

488.42

662.07

3.59

-15.44

43.11

14

1383.13

613.75

791.25

0.05

-13.44

54.60

15

1413.21

634.32

810.99

5.44

-19.87

55.24

16

1422.25

661.56

821.07

5.44

-20.32

55.24

17

1442.2

678.07

830.34

6.24

-13.73

56.53

18

1247.08

597.06

730.62

4.29

-11.73

46.78

19

1493.25

775.02

866.93

6.34

-10.3

55.95

20

1502.94

713.28

867.08

5.74

-13.49

58.45

21

1284.08

624.97

743.41

3.81

-13.2

48.79

22

1456.56

692.96

836.56

5.6

-13.4

56.52

23

1453.03

670.01

838.18

4.52

-13.56

56.52

24

1481.58

708.7

858.96

4.88

-16.36

57.16

25

1685.9

822.67

966.19

5.7

-12.09

65.22

26

1703.71

776.09

969.4

5.74

-12.19

66.28

27

1768.48

817.54

998.4

5.5

-14.01

68.75

28

1313

622.36

755.46

2.72

-13.96

50.05

29

1482.28

679.87

846.6

3.81

-12.24

56.62

30

1457.95

646.94

818.6

2.52

-13.99

56.13


Some of the calculated docking results like as the predicted binding energy, docked energy and inhibition-constant (Ki) of these inhibitors into the active site are listed in Table 3.

These results reveal all of the compounds possess good and acceptable binding energy. our docking results reveal that based on the predicted binding energy, compounds 1, 2, 3, 4, 5, 11, 12, 13, 14, 18, 21, 22 are more potent when compared to fluconazole with -6.67 kcal/mol binding energy, and more tightly bind to the active site and compounds 5a and 14 c with -5.38 and -3.9 kcal/mol are less active than fluconazole. Based on the Ki, compounds 1, 6, 11, 12, 13, 14, 15, 16, 17, 18, 24, 26, 27 respectively can inhibit the enzyme more efficiently than fluconazole with 12.99 �m inhibition constant.


 

 

Figure 2.Docked structure of compound 1 in the model of 14-DM; Heme coordination with triazole, hydrogen binding between ILE322 and aldehyde are shown.

 

 

Figure 3.Docked structure of compound 15 in the model of 14-DM; Heme coordination with triazole, hydrogen binding between heme and imidazole ring (distance 2.929 �), pi-cation interaction of phenyl ring and ARG96, pi-pi interaction between phenyl ring of ligand and phenyl ring of TYR 76 are shown.

 

Figure 4.Docked structure of compound 16 in the model of 14-DM; Heme coordination with nitrogen atom of triazole, hydrogen binding with ALA73 (distance 2.06 �) and pi- cation interaction with PHE 83 are shown.

 

Figure 5.Docked structure of compound 26 in the model of 14-DM; Heme coordination with nitrogen atom of triazole, hydrogen binding with ALA73 (distance 2.94 �), and pi- cation interaction with PHE 83 are shown.

 


QSAR equation

Based on the procedure was explained in the material and methods section, a stepwise multi linear regression (MLR) method was performed to find out the best predictive QSAR model (equation 1). To avoid the over fitted models and chance correlations, the developed model validated using internal and external validation methods. In the present study, using the Kennard Stone algorithm based on biological activity, the data set was divided into the training set and test set� in which the test set was used for external validation of prepared QSAR model. Leave-one-out cross validation (Q2LOOCV) was used as internal validation of prepared model.

Reliability of the equation 1 was judged by examining the squared correlation coefficient (r2), Fisher�s value (F), and standard deviation. Large F, small S, very small p-value, as well as R2and q2 values close to one indicate a good QSAR model.

Using a stepwise multiple linear regression method, the following 5-parametric equation was derived for triazoles 1�30.

PMIC80 = (3.779� 0.061)- (0.436 � 0.067) RDF040m-(0.289436 � 0.071) nRNHR-(0.381436 � 0.066) mor24u - (0.251� 0.068)Mor11p - (0.242� 0.72) H052.

n= 30, R2= 0.905, S= 0.294, F= 34.310, q2= 0.905, p≤0.0001

Equation 1 explains 90% of the variance in pMIC80 data, in which describes the effect of RDF040m, nRNHR, Mor24u , H052 and Mor11p indices on antifungal activity. RDF040m is Radial Distribution Function that weighted by atomic masses and nRNHR is number of aliphatic secondary amines.� Mor24u and Mor11p corresponds to 3D-MORSE descriptors in which the Mor11p is weighted by atomic polarizabilities. H052 is the Ha attached to C0 (SP3) with 1X attached to next C Comparison of descriptors signs reveals that H052, RDF040m, nRNHR, Mor24u and Mor11p have the negative contribution on the equation. Finally considering the coefficients, concept and value of descriptors reveal that RDF040m, Mor24u and nRNHR have the most effect on activity..


Table 3.Docking results of biaryloxytriazoles 1-30 using AutoDock 4.2 software.

Comp

Intermolecular

energya

Ligand

efficiencya

Electrostatic

energya

Total

Internal

energya

Torsional energya

Unbound energya

Ki (�m)

Docked energyb

Binding energyc

1

-14.22

-0.37

-0.07

-1.19

2.39

-1.19

2.11

-15.41

-11.84

2

-13.12

-0.32

0.06

-0.77

2.68

-0.77

22.59

-13.89

-10.43

3

-13.02

-0.3

0.08

-1.04

2.68

-1.04

26.63

-14.06

-10.33

4

-13.37

-0.31

0.08

-1.15

2.98

-1.15

24.41

-14.52

-10.39

5

-14.08

-0.32

0.1

-1.19

2.98

-1.19

7.35

-15.27

-11.1

6

-10.65

-0.21

-0.16

-1.37

3.28

-1.37

3.94

-12.02

-7.37

7

-9.4

-0.17

-0.17

-1.37

2.68

-1.37

11.99

-10.77

-6.71

8

-8.36

-0.14

0.11

-1.26

2.98

-1.26

114.1

-9.62

-5.38

9

-11.43

-0.22

0.09

-1.81

2.98

-1.81

640.5

-13.24

-8.45

10

-9.46

-0.15

0.37

-1.69

3.28

-1.69

29.42

-11.15

-6.18

11

-13.99

-0.36

-0.03

-1.28

2.39

-1.28

3.13

-15.27

-11.6

12

-14.21

-0.36

-0.02

-1.15

2.39

-1.15

2.15

-15.36

-11.82

13

-14.17

-0.39

-0.01

-1

2.09

-1

1.4

-15.17

-12.08

14

-15.12

-0.32

-0.05

-1.22

2.98

-1.22

1.27

-16.34

-12.14

15

-11.05

-0.21

0.06

-1.59

2.98

-1.59

1.23

-12.64

-8.06

16

-10.39

-0.19

0.02

-0.99

2.98

-0.99

3.72

-11.38

-7.41

17

-10.58

-0.19

-0.04

-1.22

2.98

-1.22

2.68

-11.8

-7.6

18

-14.52

-0.36

-0.05

-1.27

2.68

-1.27

2.11

-15.79

-11.84

19

-10.73

-0.17

0.03

-1.84

4.18

-1.84

15.68

-12.57

-6.56

20

-8.96

-0.15

0.09

-0.92

2.98

-0.92

41.58

-9.88

-5.98

21

-13.69

-0.31

0.02

-0.46

2.68

-0.46

8.6

-14.15

-11

22

-16.1

-0.34

-0.02

-1.01

2.68

-1.01

147.7

-17.11

-13.41

23

-11.58

-0.22

0.02

-1.17

2.68

-1.17

301.7

-12.75

-8.9

24

-10.35

-0.18

0.05

-0.79

2.98

-0.79

3.99

-11.14

-7.37

25

-10.35

-0.14

-0.27

-0.2

3.88

-0.2

18.03

-10.55

-6.47

26

-10.83

-0.16

-0.14

0.63

3.28

0.63

2.92

-10.2

-7.55

27

-8.38

-0.11

-0.12

15.7

3.28

15.7

182.6

-7.32

-5.11

28

-10.39

-0.21

-0.1

-1.11

2.98

-1.11

3.69

-11.5

-7.41

29

-9.94

-0.17

-0.01

1.02

2.98

1.02

7.94

-8.92

-6.96

30

-6.89

-0.1

-0.38

-1.51

2.98

-1.51

1.38

-8.4

-3.9

Fluco

nazole

-8.16

-0.3

0.02

0.89

1.49

-0.84

12.99

-9.05

-6.67

a Kcal/mol�� b� Docking energy is the sum of intermolecular energy and ligand�s internal energy

c� binding energy is the sum of intermolecular energy and torsional energy


Table 4. The experimental and calculated antifungal activity of biaryloxytriazole 1-30.

Comp.

PMIC80Exp.a

PMIC80Calc.b

|REP|c

1

4.20

4.15

0.011

2

3.60

3.26

0.103

3

3.60

3.83

0.060

4

3

2.92

0.025

5*

3

2.97

0.007

6

2.39

2.39

0.0001

7

4.20

4.55

0.075

8

3

3.001

0.0004

9

3.60

3.43

0.048

10

4.80

5.02

0.042

11

4.80

4.87

0.0127

12

5.41

4.91

0.102

13

4.20

4.52

0.069

14

3.60

3.88

0.073

15*

4.20

4.007

0.049

16

4.20

4.33

0.030

17

2.39

2.59

0.076

18*

4.20

4.49

0.065

19

3.60

3.60

0.019

20

4.80

4.80

0.058

21

3.60

3.60

0.144

22*

4.20

4.45

0.056

23

3.60

3.60

0.040

24*

4.20

3.60

0.165

25

4.20

4.20

0.001

26

4.20

4.20

0.050

27

4.20

4.20

0.109

28*

4.20

2.55

0.646

29

4.20

4.20

0.021

30

1.79

1.79

0.187

a PMIC80 in Candida albicans������ b The calculated PMIC80 using QSAR equation����� cAbsolute� relative error of prediction

*Compounds used as prediction set

Figure 6. Plot of cross-validated calculated antifungal activity obtained by QSAR equation.

 


The experimental and calculated activity and absolute relative error of prediction (REP) of biaryloxytriazole derivatives 1-30 are presented in table 4.

The graphical representation of cross validated of the experimental and calculated antifungal activity of biaryloxytriazole derivatives values was shown in Figure 6.

Conclusion

Docking studies showed that the coordination between nitrogen of triazole and iron of heme is the most important interaction to inhibitory activity. The developed QSAR model indicated the importance of RDF040m, nRNHR, Mor24u, H052 and Mor11p indices on antifungal activity that RDF040m, Mor24u and nRNHR were identified as the most significant descriptors.

 

Acknowledgment

We thank for Professor Arthur J. Olson for his kindness in offering us the AutoDock 4.2 program. Technical assistance of the Medicinal Chemistry Department of Azad University in performing the computational studies is gratefully acknowledged.��������

 

References

1. VandenBossche H, Dromer F, Improvisi I, Lozano-Chiu M, Rex JH, Sanglard D. Antifungal drug resistance in pathogenic fungi. Med. Mycol. 1998; 36 (1): 119�28.

2. Sheng C, Zhang W, Ji H, Zhang M, Song Y, Xu H, Zhu J, Miao Z, Jiang Q, Yao J, Zhou Y, Zhu J, Lu JJ.Structure-based optimization of azole antifungal agents by CoMFA, CoMSIA, and molecular docking. Med. Chem. 2006; 49: 2512.

3. Park JS, Yu KA, Kang TH, Kim S, Suh YG.Discovery of novel indazole-linked triazoles as antifungal agents.Bioorg. Med. Chem. Lett. 2007; 17: 3486-3490.Unprecedented Rearrangement of 2-(2-Aminoethyl)-1-aryl-3,4-dihydropyrazino[1,2-b]indazole-2-ium 6-oxides to 2,3-Dihydro-1H-imidazo[1,2-b]indazoles

Unprecedented Rearrangement of 2-(2-Aminoethyl)-1-aryl-3,4-dihydropyrazino[1,2-b]indazole-2-ium 6-oxides to 2,3-Dihydro-1H-imidazo[1,2-b]indazoles

4. Lebouvier N, Pagniez F, Duflos M, Le Pape P, Na YM, Le Baut G, Le Borgne M.2-[2-(2,6-Di�chloro�benz�yloxy)-2-phenyl�eth�yl]-2H-indazole,Bioorg. Med. Chem. Lett. 2007; 17: 3686.

5. Davood A, Nematollahi A, Iman M and Shafiee A. Computational studies of new 1,4-dihydropyridines containing 4-(5)-chloro-2-ethyl-5-(4)-imidazolyl substituent: QSAR and docking. Med Chem Res2010; 19(1): 58-70.

6. Davood A and Iman M. Docking and QSAR studies of β-phenylethylidenehydrazine derivatives as a Gamma-aminobutyric acid aminotransferase inhibitor. Med Che Res2011; 20: 955-961.

7. Nematollahi A and Davood A. Docking and QSAR studies of novel (E)-3-(4-methanesulfonylphenyl)-2-(aryl) acrylic acids as dual inhibitors of cyclooxygenases and lipoxygenases. Int J Chem Tech Res 2010;2(3): 1808-1815.

8. Iman M, Davood A, Nematollahi A, Dehpoor AR and Shafiee A. Design and synthesis of new 1,4-dihydropyridines containing 4(5)-chloro-5(4)-imidazolyl substituent as a novel calcium channel blocker. Arch Pharm Res 2011; 34(9), 1417-1426.

9. Iman M and Davood A. QSAR and QSTR study of selenocyanate derivatives to improve their therapeutic index as antileishmanial agents. Med Che Res 2014; 23:818�826.

10. Iman M and Davood A. QSAR and QSTR study of pyrimidine derivatives to improve their therapeutic index as antileishmanial agents. Med Che Res 2013;22(10):� 5029-5035.

11. Iman M,Davood AandBanarouei N. QSAR study of chalcone derivatives as anti-Leishmania agents. Turk J Chem2014; 1-9.

12. Iman M and Davood A. Homology modeling of lanosterol 14α-demethylase of Candida albicans and insights into azole binding. Med Che Res 2014; 23: 2890-2899.

13. Iman M, Saadabadi A and Davood A.Docking Studies of PhthalimidePharmacophore as a Sodium Channel Blocker.Iran J Basic Med Sci. Sep 2013; 16(9): 1016�1021.

14. Masand VH, Mahajan DT,Patil KN, Padole N, Hadda TB, Alafeefy AA, ShibiIG.Molecular docking and 3D QSAR studies of quinoxaline derivatives as potential influenza NS1A protein inhibitors. J. Comput. Method. Mol. Design 2011; 1(3), 49-56.

15. Wang Q, Mach RH, Reichert DE. Docking and 3D-QSAR studies on isatin sulfonamide analogues as caspase-3 inhibitors.J. Chem. Inf. Model. 2009; 49 1963-1973.

16. Cherkasov A, Ban F, Li Y, Fallahi M, Hammond GL.Progressive docking: a hybrid QSAR/docking approach for accelerating in silico high throughput screening. J. Med. Chem. 2006; 49, 7466-7478.

17. Davood A and Iman M. Molecular docking and QSAR study on imidazole derivatives as 14-demethylase inhibitors. Turk J Chem2013; 37: 119-133.

18. PingL, Zhu S,Li P, Xie W, Jin Y, Sun Q, Wu Q, SunP, ZhangY, YangX, JiangY, ZhangD. Synthesis and SAR studies of biaryloxy-substituted triazoles as antifungal agents. Bioorg Med ChemLett2008;18, 3261�3265.

19. Podust LM, Poulos TL, Waterman MR.� Crystal structure of cytochrome P450 14alpha -sterol demethylase (CYP51) from Mycobacterium tuberculosis in complex with azole inhibitors.Proc.Natl.Acad.Sci.USA 2001;98: 3068.



 

 

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

Creative Commons License
Journal of Health Policy and Sustainable Health is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.