Exploring Structural Requirement for Design and Development of compounds with Antimalarial Activity via CoMFA, CoMSIA and HQSAR

 

Mr. Nilesh Mandloi¹*, Dr. Rajesh Sharma², Dr. Jitendra Sainy², Dr. Swaraj Patil²

¹GRY Institute of Pharmacy, Borawan-451228, Dist.–Khargone (M.P.) India

²School of Pharmacy, Devi Ahilya Vishwavidyalaya, Takshashila Campus,

Khandwa Road, Indore-452001 M.P., India

 

ABSTRACT:

Malaria is lethal infectious diseases in the world caused by the protozoal species Plasmodium claiming more lives than any other parasitic infections. Novel therapies are needed against malaria because of emergence of multidrug resistant plasmodium falciparum to existing drugs. Molecular modeling studies were performed by using three different QSAR methods, Comparative Molecular Field Analysis (CoMFA), Comparative similarity indices analysis (CoMSIA) and Hologram QSAR (HQSAR), to determine the factors required for the activity of these compounds. The developed models on one hundred and twelve 7-substituted 4-aminoquinoline derivatives showed good statistical significance in internal cross validation (q²) and non-cross validation (r²) values of 0.509, 0.992 by the CoMFA model and 0.358 0.838 by the CoMSIA model respectively for antimalarial activity. Structural features were correlated in terms of their several properties as steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor interactions. Furthermore, the bioactive conformation was explored and explained by docking of compounds the active binding site of lactate dehydrogenase of Plasmodium falciparum.

 

 

 

INTRODUCTION:

Malaria originated more than 30 million years ago is caused by mosquito-borne eukaryotic protists of the genus Plasmodium five species (P. ovale, P. malariae, P. vivax, P. falciparum, and P. Knowlesi) that affects humans. of these, P. falciparum is the most problematic, mainly due to its high prevalence, virulence and drug resistance.^{[1, 2]} According to world malaria report 2013 there were an estimated 207 million cases of malaria in 2012 (uncertainty range: 135–287 million) and an estimated 627 000 deaths (uncertainty range: 473000–789000), malaria killed an estimated 483000 children under five years of age. That is 1300 children every day or one child almost every minute^[3].

 

Historically 4-aminoquinoline based entities, particularly chloroquine (CQ), have the first choice in the malaria chemotherapy. Chloroquine (CQ), antimalarial evolved as substitute for quinine during World War II, was approved as a safe, efficacious, easily available, inexpensive and economic treatment against malaria. It is cleared that this class of compounds enter the food vacuole and inhibit the parasite growth by forming a complex with hematin thereby inhibiting the hemozoin formation^[4].

 

The Plasmodium species are strongly linked with human malaria, having been reported to be frequently over expressed in many human organ systems, including severe pain and also loss of blood contents. Out of four destructive parasite species, Plasmodium falciparum is amenable for the majority of malaria cases, which repeatedly confirm mortal from past centuries to mankind^[5].

 

In present study we used software Sybyl X2.0 to manifest the molecular modeling analysis in search of new 4-amino quinoline derivatives as antimalarial. The CoMFA and CoMSIA studies gave an insight into the quantitative role of chemical features in fluctuating in terms of favorable and unfavorable contours. Thus CoMFA and CoMSIA studies will not only interpret the conformation or spatial orientation but also provide useful information for further design of new drugs. HQSAR uses counts of key molecular substructures and PLS to generate fragment-based structure-activity relationships identify the sub-structure requirement of antimalarial activity. We also performed the molecular docking to explore the drug-receptor interaction. In our docking study we used plasmodium falciparum lactate dehydrogenase (PfLDH), which is an tetrameric enzyme contains 316 amino acids and found in p.viavax, P. malariae and p. ovale all exhibit ~90% identity to LDH. It is an important enzyme of glycolysis and necessary for energy production in plasmodium, besides plasmodium lack functional kreb cycle during some erythrocytic stage. As an essential enzyme it catalyzes the interconversion of pyruvate and lactate with concurrent conversion of NADH and NAD.^[6]

 

 

 

 

The QSAR contour maps and docking binding affinity analysis of these structures enabled us to investigate and suggested structure-activity relationship of dataset which guided us to design a series of new inhibitors with accepted predicted activities (pEC₅₀) and docking score.

 

MATERIAL AND METHODS:

3D-QSAR Data set:

In present studies one hundred and twelve 7-substituted 4-aminoquinoline derivatives and their antimalarial activities were taken from previous literatures^[7]. In extend to examine the predictive power of the QSAR models, the dataset divided into the training set of 91 (2, 4, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 19, 20, 21, 22, 23, 26, 28, 30, 31, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 54, 55, 56, 58, 59, 60, 61, 62, 63, 64, 66, 68, 69, 70, 72, 73, 74, 75, 76, 77, 78, 80, 81, 82,  83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 111) molecules and remaining in test set were selected by diversity method in such a way that the structural diversity and wide range of biological activity in the data set were added. The antimalarial activity EC₅₀ values were transformed in corresponding pEC₅₀ (-logEC₅₀) and used as dependent variable. The structures of the compounds and their biological data are listed in Table 1. ^[1,8]

 

Table 1:-Actual and predicted activity of training and test sets

Sr. No.	R1	R2	EC50 (µM)	PEC50	CoMFA predicted	CoMSIA predicted	HQSAR Predicted
1	4-MeO-PhO	Furfuryl	0.004	8.39	7.34	6.63	7.82
2	3-Me2N-PhO	3-F-6-MeO-Bn	0.007	8.15	8.11	8.12	7.11
3	PhO	Piperonyl	0.018	7.74	7.76	7.78	7.13
4	2-MeO-Pho	Furfuryl	0.001	9	9.02	8.57	8.12
5	4-F-PhO	Furfuryl	0.026	7.58	7.28	7.75	7.43
6	4-Cl-PhO	3-F-6-MeO-Bn	0.031	7.50	6.91	7.08	7.25
7	4-F-PhO	3-F-6-MeO-Bn	0.297	6.52	6.56	6.69	6.97
8	PhO	3-F-6-MeO-Bn	0.162	6.79	6.79	6.62	6.93
9	3-MeO-PhO	Furfuryl	0.009	8.04	7.98	7.48	7.21
10	3-Me2N-PhO	Furfuryl	0.027	7.56	7.57	7.53	7.42
11	4-tertBu-PhO	2-HO-3-MeO-Bn	0.054	7.26	7.20	7.09	6.95
12	3-MeO-PhO	Piperonyl	2.36	5.62	7.58	7.80	6.90
13	4-Cl-PhO	Piperonyl	0.038	7.42	7.38	7.48	7.57
14	4-MeO-PhO	2-HO-3-MeO-Bn	0.004	8.39	8.38	8.38	8.26
15	2-MeO-PhO	Piperonyl	0.014	7.85	7.90	7.81	8.01
16	3-MeO-PhO	Piperonyl	0.012	7.92	7.97	7.88	7.10
17	4-MeO-PhO	Piperonyl	0.054	7.26	7.33	7.20	7.52
18	4-F-PhO	Piperonyl	0.015	7.82	7.02	7.40	7.47
19	4-Cl-PhO	2-HO-3-MeO-Bn	0.011	7.95	7.99	7.88	7.96
20	4-Cl-PhO	Furfuryl	0.04	7.39	7.38	7.75	7.53
21	2-MeO-PhO	Piperonyl	0.022	7.65	7.64	7.56	7.82
22	4-MeO-PhO	Piperonyl	0.046	7.33	7.32	7.31	7.71
23	PhO	2-HO-3-MeO-Bn	0.015	7.82	7.83	7.76	7.68
24	3-Me2N-PhO	2-HO-3-MeO-Bn	0.01	8	7.67	7.80	7.86
25	4-tertBu-PhO	Piperonyl	0.013	7.86	7.29	7.05	6.71
26	4-tertBu-PhO	Piperonyl	0.106	6.97	6.98	7.05	6.91
27	3-MeO-PhO	2-HO-3-MeO-Bn	0.072	7.14	7.97	8.00	7.65
28	4-F-PhO	2-HO-3-MeO-Bn	0.011	7.95	7.97	8.00	7.65
29	3-Me2N-PhO	Piperonyl	1.363	5.86	7.63	7.97	7.31
30	4-tertBu-PhO	Furfuryl	1.11	5.95	5.95	6.22	6.51
31	2-MeO-PhO	2-HO-3-MeO-Bn	0.009	8.04	8.06	7.77	8.56
32	PhO	Furfuryl	0.314	6.50	7.75	7.87	7.24
33	4-MeO-Ph	Furfuryl	0.011	7.95	7.44	7.53	7.62
34	Piperonyl	3-F-6-MeO-Bn	0.031	7.50	7.44	7.68	7.42
35	4-CF3-Ph	2-HO-3-MeO-Bn	0.013	7.88	7.87	8.01	7.89
36	4-MeO-Ph	Piperonyl	0.029	7.53	7.55	7.61	7.51
37	Piperonyl	Piperonyl	0.026	7.58	7.50	7.42	7.61
38	4-MeO-Ph	3-F-6-MeO-Bn	0.036	7.44	7.01	7.02	7.32
39	Ph	Piperonyl	0.017	7.76	7.74	7.68	7.69
40	4-F-Ph	Piperonyl	0.018	7.74	7.80	7.59	7.77
41	Ph	3-F-6-MeO-Bn	0.011	7.95	8.00	7.91	7.50
42	Piperonyl	Furfuryl	0.019	7.72	7.75	7.83	7.73
43	Piperonyl	2-HO-3-MeO-Bn	0.008	8.09	8.06	7.73	8.17
44	4-F-Ph	Furfuryl	0.032	7.49	7.51	7.75	7.72
45	4-MeO-Ph	2-HO-3-MeO-Bn	0.013	7.88	7.85	7.89	8.06
46	4-CF3-Ph	3-F-6-MeO-Bn	0.054	7.26	7.21	6.94	7.31
47	4-F-Ph	3-F-6-MeO-Bn	0.018	7.74	7.69	7.73	7.78
48	4-CF3-Ph	Piperonyl	0.025	7.60	7.59	7.42	7.50
49	Ph	Furfuryl	0.02	7.69	7.68	7.86	7.80
50	3,5-CF3-Ph	2-HO-3-MeO-Bn	0.029	7.53	7.58	7.69	7.29
51	4-CF3-Ph	Furfuryl	0.02	7.69	7.67	7.62	7.45
52	4-tertBu-Ph	Furfuryl	0.704	6.15	7.34	7.26	6.31
53	4-tertBu-Ph	2-HO-3-MeO-Bn	0.251	6.60	7.52	7.38	6.75
54	4-tertBu-Ph	Piperonyl	0.223	6.65	6.64	6.71	6.56
55	1-Naphtyl	Piperonyl	0.598	6.22	6.21	6.41	6.68
56	3,5-CF3-Ph	Piperonyl	0.284	6.54	6.53	6.73	6.88
57	3,5-CF3-Ph	3-F-6-MeO-Bn	0.092	7.03	7.87	7.88	6.69
58	4-F-Ph	2-HO-3-MeO-Bn	0.009	8.04	8.00	8.08	8.10
59	3,5-CF3-Ph	Furfuryl	0.285	6.54	6.54	6.51	6.83
60	1-Naphtyl	3-F-6-MeO-Bn	1.154	5.93	5.94	5.81	6.48
61	1-Naphtyl	2-HO-3-MeO-Bn	0.033	7.48	7.42	7.54	7.22
62	1-Naphtyl	Furfuryl	0.031	7.50	7.46	7.69	6.78
63	4-tertBu-Ph	3-F-6-MeO-Bn	0.144	6.84	6.88	6.79	6.40
64	Ph	2-HO-3-MeO-Bn	0.009	8.04	8.06	8.14	8.26
65	1-Et-Pr	2-HO-3-MeO-Bn	0.004	8.39	7.57	6.97	7.97
66	3,5-Me-Bn	2-HO-3-MeO-Bn	0.009	8.04	8.03	8.09	8.18
67	3,5-Me-Bn	Furfuryl	0.009	8.04	7.55	7.48	7.74
68	4-CN-Bn	2-HO-3-MeO-Bn	0.011	7.95	7.95	7.86	8.07
69	Bn	3-F-6-MeO-Bn	0.011	7.95	7.85	7.66	7.21
70	2-Cl-4-F-Bn	2-HO-3-MeO-Bn	0.013	7.88	7.91	7.84	7.70
71	PhEt	Piperonyl	0.013	7.88	7.71	7.49	7.35
72	4-F-Bn	2-HO-3-MeO-Bn	0.015	7.82	7.76	8.01	7.59
73	PhEt	2-HO-3-MeO-Bn	0.017	7.76	7.88	7.81	7.89
74	Bn	2-HO-3-MeO-Bn	0.017	7.76	7.81	7.92	7.95
75	cHex	2-HO-3-MeO-Bn	0.019	7.72	7.72	7.71	7.50
76	2-Cl-4-F-Bn	Furfuryl	0.02	7.69	7.68	7.92	7.46
77	iso-butyl	2-HO-3-MeO-Bn	0.02	7.69	7.69	7.70	7.68
78	iso-pentyl	2-HO-3-MeO-Bn	0.022	7.65	7.70	7.61	7.60
79	4-CN-Bn	Piperonyl	0.023	7.63	7.06	7.33	7.53
80	3-CF3-Bn	2-HO-3-MeO-Bn	0.025	7.60	7.55	7.61	7.73
81	2-Cl-4-F-Bn	Piperonyl	0.027	7.56	7.57	7.68	7.31
82	1-Et-Pr	Piperonyl	0.027	7.56	7.57	7.52	7.43
83	iso-pentyl	Piperonyl	0.027	7.56	7.57	7.46	7.06
84	4-CN-Bn	3-F-6-MeO-Bn	0.028	7.55	7.54	7.62	7.33
85	3,5-Me-Bn	3-F-6-MeO-Bn	0.034	7.46	7.50	7.80	7.44
86	4-CN-Bn	Furfuryl	0.039	7.40	7.39	7.48	7.63
87	3-CF3-Bn	Piperonyl	0.039	7.40	7.39	7.26	7.34
88	Bn	Furfuryl	0.04	7.39	7.39	7.27	7.51
89	3-CF3-Bn	3-F-6-MeO-Bn	0.04	7.39	7.41	7.25	7.15
90	1-Et-Pr	3-F-6-MeO-Bn	0.04	7.39	7.35	7.43	7.23
91	cHexmethyl	Piperonyl	0.041	7.38	7.37	7.58	7.21
92	3,5-Me-Bn	Piperonyl	0.041	7.38	7.34	7.27	7.64
93	PhEt	Furfuryl	0.046	7.33	7.30	7.31	7.45
94	cHexmethy	2-HO-3-MeO-Bn	0.05	7.30	7.32	7.27	7.39
95	cHexmethy	Furfuryl	0.055	7.25	7.29	7.17	6.96
96	3-CF3-Bn	Furfuryl	0.056	7.25	7.19	7.22	7.29
97	3-F-Bn	Furfuryl	0.057	7.24	7.18	7.18	7.15
98	iso-butyl	Piperonyl	0.066	7.18	7.24	7.11	6.50
99	4-F-Bn	3-F-6-MeO-Bn	0.075	7.12	7.10	7.27	7.27
100	1-Et-Pr	Furfuryl	0.086	7.06	7.15	7.07	7.53
101	iso-butyl	3-F-6-MeO-Bn	0.089	7.05	7.13	6.92	6.94
102	cHex	3-F-6-MeO-Bn	0.09	7.04	7.05	6.96	7.11
103	iso-pentyl	3-F-6-MeO-Bn	0.116	6.93	6.96	6.82	6.86
104	cHex	Furfuryl	0.118	6.92	6.95	7.05	7.06
105	3-F-Bn	Piperonyl	0.127	6.89	6.95	7.19	7.2
106	cHex	Piperonyl	0.128	6.89	6.87	6.77	7.31
107	PhEt	3-F-6-MeO-Bn	0.132	6.87	6.87	7.10	7.15
108	cHexmethy	3-F-6-MeO-Bn	0.138	6.86	6.81	6.81	7.01
109	iso-pentyl	Furfuryl	0.176	6.75	6.74	6.87	7.14
110	2-Cl-4-F-Bn	3-F-6-MeO-Bn	0.206	6.68	7.34	7.46	7.13
111	iso-butyl	Furfuryl	0.581	6.23	6.22	6.33	7.30
112	Bn	Piperonyl	0.72	6.14	7.37	7.49	6.73

 

Molecular modeling and Structural alignment:

In QSAR analysis structural superposition plays a crucial role, because the relative interaction energies depend strongly on relative molecular positions. Flexible groups and conformation diversity present dilemmas to molecular alignment.^[9] All the molecular modeling studies were performed using SYBYL X2.0 software running on a core-2 duo Intel processor workstation.^[10] The molecules should be aligned together based on a certain rule, because the molecules in our investigation have a common substructure. The molecules to be analyzed were aligned on an appropriate 4-aminoquinoline template, which is considered to be common substructure. For the multifit alignment the most active compound (4) was used as alignment template (Figure 1). The alignment, automatically executed in SYBYL program, is made based on this common substructure using “database alignment”.

 

Molecular superposition plays a decisive role in CoMFA analysis, since the relative interaction energies depend strongly on relative molecular positions. Flexible groups and conformation diversity present dilemmas to molecular alignment.^[11]

 

 

Figure 1:-Image of Alignment of Structures

 

CoMFA:

In CoMFA biological properties of molecules are correlated with steric and electrostatic potential energies in terms of Lennard-Jones and Coulombic potentials, respectively.^[11] For the CoMFA analysis, the sp3 hybridised carbon atom were used with +1charge served as a probe, steric (S) and electrostatic (E) fields energies  between the probe and molecules at each lattice intersection were calculated with a grid step of 2A˚, which are automatically generated by the CoMFA-STD method with cut off value of 30 kcal mol^-1. The Gasteiger, Gasteiger–Huckel, MMFF, Del-Re, and Pullman charges were used to generate the partial charges on the studied molecules and explored models^[12].

 

CoMSIA:

The CoMSIA analysis has the advantage in exploring more similarity descriptors, viz. steric (S), electrostatic (E), hydrophobic (H), hydrogen bond donor (D), and hydrogen bond acceptor (A)^[13]. All these descriptors were generated using a sp³ hybridized carbon atom with+1 charge; the attenuation factor was set to 0.3 and a van der Waals radius of 1.4Å, CoMSIA similarity indices (A_F,K) between a molecule j and atoms i at a grid point were calculated.

 

HQSAR studies:

Hologram quantitative structure activity relationship (HQSAR) technique is based on the 2-dimensional descriptors which eliminate the need for generation of 3D structures, putative binding conformations and molecular alignments. It employs specialized fragment fingerprints as predictive variables of biological activity. HQSAR is sensitive to three parameters concerning hologram generation, including hologram length, fragment size, and fragment distinction. The fragments distinct are atoms (A), bonds (B), connections (C), hydrogen atoms (H), chirality (Ch), and donor (D). Initially, various models were developed by using the default fragment size of 4 to 7 and different component, then based on the different fragment distinctions determined by the first step, the models were developed using different fragment sizes. The models with better result were applied to different fragment size and component number. The better statistical results were obtained in fragment size 5 to 8, A/B/Ch/D distinct and number of component 17^[15].

 

Partial least square (PLS) and predictive r² analysis:

Partial least square has been broadly used in QSAR studies. It is a popular linear modeling technique for relating dependent variables and independent variables by a linear multivariate model. PLS regression studies were carried out by using anti-malarial activity as dependent variable and structural properties of CoMFA, CoMSIA and HQSAR as independent variables. Leave one out (LOO) validation option was used to determine the optimum number of component and no validation was utilized for determining the predictability of the developed model. The full PLS analysis was carried out with a column filtering of 2.0kcal/mol. The predictive r² value is based on only the test set molecules, which may be defined as shown in Equation 1

 

r²pred=SD-PRESS/SD                                                  (1)

 

Where SD are the sum of squared deviation between the biological activities of the test set molecules to the mean activity of the training set molecules, while PRESS are the sum of squared deviations between the observed and the predicted activities of the test molecules^{[14- 17]}.

 

Molecular Docking:

Docking studies yielded crucial information about the orientation of the inhibitors in the binding pocket of the enzyme and the interaction between the target (enzyme) and the small molecules (ligands) at the molecular level.

 

Molecular docking studies were carried out using the Surflex Dock in SYBYL X2.0 to explore possible binding affinity conformations and offer more insight into the understanding of the relations of lactate dehydrogenase receptor with inhibitors. The protein structure of Plasmodium falciparum L-lactate dehydrogenase along with its inhibitor was retrieved from RCSB Protein Data Bank (PDB entry code: 1LDG). The protein structures were subjected to energy minimization and charge calculation (AMBER7FF99). All ligands and water molecules were removed. The bloat value was set as 1 and the threshold value as 0.5 for generation of protomol and position was considered to be the active sites for potential receptor’s binding sites^[18].

 

RESULTS AND DISCUSSION:

CoMFA and CoMSIA results:

From a training set of 91 molecules of 7-Substituded 4-Aminoquinoline analogues with pEC₅₀ values ranging from 9 to 5.937794 µM, CoMFA and CoMSIA models were derived. Test set of 21 molecules was used to evaluate the predictive power of the resulting models. The CoMFA model using both steric and electrostatic fields contributions were 0.320 and 0.680, respectively which gave a cross-validated correlation coefficient (q²) =0.509 with an optimized component 2. A high non-cross-validated correlation coefficient (r²)=0.992 with a low standard error estimate (SEE)=0.048, obtained F value=157.58.In CoMSIA analysis there is advantage to unfold more field in comparison to CoMFA. It consist Steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor field contribution were 0.059, 0.175, 0.129, 0.104 and 0.99 respectively, which gave q²=0.358 with an optimized component 9 and r²=0.838 with a low standard error 0.221, obtained F value of 157.58.The statistical parameters associated in CoMFA and CoMSIA models are listed in the table 2.

 

Table 2-Summary of the CoMFA and CoMSIA statistical results

Statistical parameters	CoMFA	CoMSIA	Field contribution (%)	CoMFA	CoMSIA
q2(cross validated)	0.509	0.358	Steric	0.320	0.059
r2	0.992	0.838	Electrostatic	0.680	0.175
F values	157.58	44.79	Hydrophobic	-	0.129
SEE	0.048	0.221	HB donor	-	0.104
No. of Components	9	9	HB acceptor	-	0.99

q²=cross-validation correlation coefficient; r²=conventional correlation; SEE=standard error of estimate; F=Fischer test; N=optimal number of component; HB=Hydrogen bond

 

CoMFA and CoMSIA contour map analysis:

Contour maps were obtained by using descriptors from the CoMFA and CoMSIA models, which establish the significant physicochemical properties responsible for antimalarial activity and also investigate the importance of various substituents. The contours obtained revealed information about the physicochemical properties of quinoline based molecules, essential in defining the antimalarial activity. It also gives an indication of that portion of the molecule that requires particular property to improve the binding affinity. The developed contour maps from the CoMFA and CoMSIA analyses are shown in Figures.

 

CoMFA contour map interpretation

The PLS CoMFA contours maps in terms of common steric and electrostatic potential were generated taking molecule no 4 having greatest antimalarial activity of the dataset indicating that

 

 

a.    COMFA STERIC:

In CoMFA steric counter map (fig.A1) two green and yellow color shows. The region comes under green color favors bulky group and the region comes under yellow color disfavor the bulky group. Hence if we substitute bulky group at green region and remove at yellow region it will contribute in enhancing the antimalarial activity. In fig (A1) C-OMe at R₁ substitution, quinoline ring, amino group at carbon chain and C=O at R₂ substitution favor, while some region of furan ring at R₂ substitution disfavors steric/bulky group.

 

 

Figure A1:-CoMFA STERIC

 

b.    COMFA ELECTROSTATIC:

CoMFA electrostatic counter map (fig.A2) shows red and blue color. The region comes under blue color favors electropositive substitution and the region comes under red color favor electronegative substitution. Hence if we substitute electropositive group at blue region and electronegative at red region it will contribute in enhancing the antimalarial activity. In fig (A2) nitrogen in quinoline ring and furan-2-carbaldehyde at R₂ position Favors Electropositive substitution. While favors Electronegative Substitution at R₁ position.

 

 

Figure A2:-CoMFA ELECTROSTATIC

 

CoMSIA contour map interpretation

The PLS contours from CoMSIA analysis were carried out taking the most active compound number 4.

 

a.    CoMSIA STERIC:

CoMSIA steric (fig.B1) interactions are represented by favored green within which the occupation of bulky group is likely to increase activity and disfavored yellow counters within which occupation of bulky group is likely to decrease activity. In fig (B1) the quinoline ring and region of R₁ substitution strongly favor bulky group while the region near furan ring at R₂ substitution slightly favor bulky substitution.

 

 

Figure B1:-CoMSIA STERIC

 

b.    CoMSIA ELECTROSTATIC:

CoMSIA electrostatic (fig.B2) interactions are represented by negative charge favored red and positive charge favored blue counter. It is revealed that increase in positive charge will enhance activity in blue regions and increase negative charges will enhance activity in red regions. In fig (B2) substitution R₂ and some areas of quinoline ring favor positive charge in blue color, while some areas near R₁ substitution and carbon linker favor negative charge in red color.

 

 

Figure B2:-CoMSIA ELECTROSTATIC

 

c.     CoMSIA HYDROPHOBIC

In hydrophobic contour map (fig.B3) yellow color signifies hydrophobic favored regions while the white color indicates hydrophobic unfavored regions, which indicates that presence of hydrophilic groups in these areas may increase the activity. In fig (B3) carbon chain between two amino groups, C=O and lower region of furan ring at R2 substitution and some region of R1 substitution favor, while OMe at R1 and some region of furan ring at R2 substitution slightly disfavor hydrophobic.

 

 

Figure B3:- CoMSIA HYDROPHOBIC

d.    CoMSIA DONAR

Hydrogen bond donors (fig.B4) are shown by favored cyan and disfavored purple counters.

In fig (B4) cyan counters shows near the 4-aminoquinoline ring and substitution R₂ which implies that substitution with hydrogen bond donating groups may increase the activity.

 

 

Figure B4:- CoMSIA DONOR

 

e.     CoMSIA ACCEPTOR

Hydrogen bond acceptors (fig.B5) are shown by favored magenta and disfavored red counters.

In fig (B5) there are two small magenta counters shows near the quinoline ring and furan ring, which prefer hydrogen bonding acceptor atoms.

 

 

Figure B5:-CoMSIA ACCEPTOR

 

HQSAR study:

For HQSAR analyses we screened the 6 default series of hologram length values ranging from 53 to 401 bins, initially using the fragment size default (4–7) on 60 different fragment distinct taking Atom(A), Bond(B), Connection(C), Hydrogen atom(H), Chirality(Ch) Donor/acceptor(D). The patterns of fragment counts from the training set inhibitors were related to the experimental biological activity using the PLS analysis. The best statistical parameter was obtained from PLS analyses with A/C/D, A/B/Ch/D, A/B/D, A/B/Ch/D and A/C/Ch/D. The influence of fragment size is of fundamental importance in the generation of HQSAR models, as this parameter controls the minimum and maximum lengths of fragments to be encoded in the hologram fingerprint. Hence, all 5 sets of distinct with fragment size combinations 2-5, 3-6, 4-7, 5-8, 6-9 and 7-10 were investigated for the best model. Best result from which was investigated further by changing number of components (default 20). Final two models were selected on the basic of best statistical parameter are shown in Table 3.

 

Table 3-Statistical results for the two best HQSAR models obtained using different combinations of fragment distinction and different fragment size

Model	Fragment distinctions	Statistics parameters
		R2 best cv	R2 best full	R2 Ensemble	SEE	N	Fragments size	HL
1	A/B/Ch/D	0.640	0.640	0.634	0.389	17	5-8	151
2	A/C/D	0.638	0.638	0.632	0.364	15	6-9	353

SEE=standard error of estimate; HL=Best hologram length (bin); N=optimal number of components; Fragment distinction: A=atoms; B=bonds; C=connections; Ch=chirality; D=donor/acceptor.

 

Graphical interpretation of the HQSAR model:

Besides predicting the activities of untested molecules an important role of HQSAR model is to provide hints about what molecular fragments may be important contributors to activity through the use of different color coding which reflects the individual atomic contributions. The potent compounds contribution map obtained from the HQSAR module implemented in Sybyl X2.0 uses a color scheme is depicted in the figure2. Taking compound number 4 (best actual activity) HQSAR fingerprint was generated. In contribution map green and yellow color (represents positive contribution), white color (indicates intermediate or moderate contribution), red and orange (negative contribution) and cyan color (represents maximal common structure) suggest the structure fragment requirement for enhancing the binding affinity.

 

 

Figure2:-molecular fragments of best active compound 4

 

The contours coding depends on the biological activity contribution of the individual fingerprint of atoms of the molecules. The contribution map obtained from the HQSAR module implemented in Sybyl X2.0 uses a fingerprint scheme to discriminate individual atomic contributions to activity. The fingerprints at the negative end of the spectrum reflect poor contributions, whereas fingerprints at the positive end reflect favorable contributions. Atoms with intermediate contributions are white. Atoms corresponding to the maximal common structure (MCS) are neutral, since it is common to all compounds and contributes in the same manner for all molecules in the training set. The molecular fragments of compound 4 are shown in Figure 2. According to the contribution maps, quinoline ring comes under green color strongly related to positive contribution toward biological affinity.

 

Docking:

For Docking Analysis all compounds were selected to evaluate antimalarial activity as series under consideration was reported with only antimalarial activity. Docking was done using PDB code 1LDG, using software Sybyl-X-2.0.All compounds were showing interaction with lactate dehydrogenase enzyme whose data mentioned in table 4.

 

Docking analysis:

7-substituted 4-aminoquinoline derivatives were docked at active site of crystal structure of 1ldg.pdb. The docking result present in table shows that 27 molecules were found to exhibit good antimalarial activity in which compound no. 85 has the highest total score 11.01, crash-2.59 and polar 2.39. The ligand-receptor interaction analysis of this structure shows that the nitrogen atom present in quinoline forms hydrogen bond with GLY 99, amino group of 4-aminoqionoline ring forms hydrogen bond with ASP53, OCh₃ and C=O  of R₂ substitution forms hydrogen bond with MET30 and ILE31 respectively. In lipophilic potential pose view (figure C) aminoquinoline rings situated near medium lipophilic region but the nitrogen at 7 position showing slightly higher lipophilicity. In electrophilic potential (figure D) Aminoquinoline rings and carbon linker situated near medium electrostatic region.  In Cavity depth pose (figure E) Analysis reveals that the compounds are in good Cavity depth which supports activity.

 

Table 4:- Results of Docking Analysis

Sr. No.	Name	Total Score	Crash	Polar
1	85	11.01	-2.59	2.39
2	21	10.13	-1.59	3.91
3	89	9.75	-1.74	2.19
4	101	9.75	-1.72	0.67
5	17	9.63	-3.09	2.30
6	57	9.14	-1.58	2.23
7	25	9.02	-3.03	2.40
8	105	8.97	-2.22	2.58
9	13	8.94	-1.64	3.13
10	73	8.79	-2.77	2.58
11	93	8.72	-4.58	2.93
12	45	8.72	-1.62	3.78
13	53	8.70	-1.33	3.00
14	29	8.54	-2.80	1.18
15	61	8.26	-1.80	3.64
16	81	8.13	-2.19	0.82
17	65	8.09	-1.69	0.84
18	37	7.90	-2.48	0.03
19	77	7.86	-1.61	2.59
20	9	7.79	-2.17	0.59
21	69	7.73	-1.76	3.87
22	49	7.63	-2.25	2.025
23	109	7.17	-1.03	0.74
24	41	6.90	-1.86	1.01
25	97	6.81	-1.72	3.22
26	33	6.59	-2.93	0.83
27	1	6.55	-1.05	2.95

 

 

Figure C :- Lipophilic potential

 

 

Figure D :- Electrophilic potential

 

 

Figure E :- Cavity depth

 

 

Figure F:- Interaction with hydrogen

 

Figures:-Docking poses visualisation of compound 85

 

 

CONCLUSION:

The present study we successfully established Quantitative structure activity relationship (QSAR) of a series of 7-substituted 4-aminoquinoline derivatives with corresponding antimalarial activity were investigated by using 3D CoMFA, CoMSIA and HQSAR methods. We also performed molecular docking study to characterize a set of 7-substituted 4-aminoquinoline based derivatives and to identify essential structural requirements in 3D chemical space for the modulation of Antimalarial activity. The CoMFA, CoMSIA and HQSAR models showed meaningful statistical significance results in internal validation (q²), external validation (r²) and predicted r². The CoMFA and CoMSIA models were further used to generate the counter maps which highlight the key structural features that affected bioactivity and provided explicit indication for lead compound optimization. HQSAR provides information about positive, negative and intermediate contribution of sub-structural fingerprint requirements for imparting the biological activity. Docking studies suggested that Aminoquinoline ring is necessary for antimalarial activity. The CoMFA, CoMSIA, HQSAR contour maps, along with the docking binding affinity molecules, revealed sufficient information to understand the structure-activity relationship (SAR) and to recognize structural features influencing activity.

 

ACKNOWLEDGEMENT:

The authors are thankful to the Head, School of Pharmacy for providing facilities.

 

CONFLICT OF INTEREST:

The authors declare no conflict of interest.

 

REFERENCES:

1        Li J, Li S, Bai C, liu H, Gramatica P. Structural Requirements of 3-carboxyl-4 (1H)-quinolones as Potential Antimalarials from 2D and 3D QSAR Analysis. Journal of Molecular Graphic and Modeling. 2013; 44: 266-277.

2         Andayi WA, Egan WA, Gut J, Rosenthal PJ, Chibale. Synthesis Antiplasmodium Activity and β-Hematin Inhibition of Hydroxypyridone−Chloroquine Hybrids. Journal of Medicinal Chemistry. 2013; 4: 642-646.

3        http://www.who.int/malaria/media/world_malaria_report_2013/en/index.html#

4        Sunduru N, Srivastava K, Rajakumar S, Puri SK, Saxena JK, Chauhan PMS. Synthesis of Novel Thiourea, Thiazolidinedione and Thioparabanic Acid Derivatives of 4-aminoquinoline as Potent Antimalarials. Journal of Bioorganic and Medicinal Chemistry. 2017; 19: 2570-2573.

5        Saxena S, Chaudhary SS, Varshney K, Saxena AK. Pharmacophore-based Virtual Screening and Docking Studies on Hsp90 Inhibitors. SAR QSAR Environmental Research. 2016; 21: 445-462.

6        Cortopassi WA, Oliveira AA, Guimaraes AP, Renno MN, Krettli AU, Franca TC. Docking Studies on the Binding of Quinoline Derivatives and Hematin to Plasmodium Falciparum Lactate Dehydrogenase. Journal of Biomolecular Structure and Dynamics. 2017; 29: 0739-1102.

7        Hwang JY, Kawasuji T, Lowes DJ, Clark JA, Connelly MC, Zhu F, Guiguemde WA, Sigal MS, Wilson EB, DeRisi JL, Guy RK. Synthesis and Evaluation of 7-Substituted 4-Aminoquinoline Analogues for Antimalarial Activity. Journal of Medicinal Chemistry. 2011; 54: 7084-7093.

8        Adhikari N, Halder AK, Mondal C, Jha T. Exploring Structural Requirements of Aurone Derivatives as Antimalarials by Validated DFT-based QSAR, HQSAR, and COMFA–COMSIA approach.  Medicinal Chemistry Research. 2013; 22: 6029-6045.

9        Xie A, Sivaprakasama P, Doerksen RJ. 3D-QSAR Analysis of Antimalarial Farnesyltransferase Inhibitors Based on a 2, 5-diaminobenzophenone Scaffold. Journal of Bioorganic and Medicinal Chemistry. 2015; 14: 7311-7323.

10      SYBYL6.9, Tripos Inc., St. Louis, MO, USA.

11      Zhang L, Tsai KC, Du L, Fang H, Li M. How to Generate Reliable and Predictive CoMFA Models. Current Medicinal Chemistry. 2014; 18: 923-930

12      Xue CX, Cui SY, Liu MC, Hu ZD, Fan BT. 3D QSAR Studies on Antimalarial Alkoxylated and Hydroxylated Chalcones by CoMFA and CoMSIA. European Journal of Medicinal Chemistry. 2014; 39: 745-753.

13      Cheng F, Shen J, Luo X, Zhu W, Gu J, Ji R,  Jiang H, Chen K. Molecular Docking and 3-D-QSAR Studies on the Possible Antimalarial Mechanism of Artemisinin Analogues. Journal of Bioorganic and Medicinal Chemistry. 2014; 10: 2883-2891.

14      Villalobos TPJ, Ibarra RG, Acosta JJM. 2D, 3D-QSAR and Molecular Docking of 4(1H)-Quinolones Analogues with Antimalarial Activities. Jornal of Molecular Graphics and Modelling. 2015; 26: 105-124.

15      Saxena AK, Prathipati P. Comparison of MLR, PLS and GA-MLR in QSAR Analysis, SAR and QSAR in Environmental Research. 2015; 14: 433-446.

16      Neves BJ, Bueno RV, Braga RC, Andrade CH. Discovery of New Potential hits of Plasmodium falciparum enoyl-ACP reductase through ligand- and Structure-based Drug Design Approaches, Journal of Bioorganic and Medicinal Chemistry. 2016; 23: 2436-2441.

17      Ojha PK, RoyK. Comparative QSAR for Antimalarial Endochins: Importance of descriptor-thinning and noise reduction prior to feature selection, Chemometrics and Intelligent Laboratory Systems. 2011; 109: 146-161.

18      Bhat HR, Singh UP, Gahtori P, Ghosh SK, Prakash A, Singh RK. 4-Aminoquinoline-1,3,5-triazine: Design, Synthesis, in vitro Antimalarial activity and Docking Studies. New Journal of Chemistry. 2013; 37: 2654-2662.

 

Accepted on 25.05.2018        © RJPT All right reserved

Research J. Pharm. and Tech 2018; 11(8): 3341-3349.