Quantitative Structure Activity Relationship of Some Substituted 2-Aminopyridines and Fused Bicyclic Rings as inhibitors of Nitric Oxide Synthases

 

Amol S. Sherikar*

Department of Pharmaceutical Chemistry, Tatyasaheb Kore College of Pharmacy, Warananagar, Tal- Panhala, Dist- Kolhapur, Maharashtra, India. 416 113.

*Corresponding Author E-mail: amol.sherikar@rediffmail.com

ABSTRACT:

In the present work, QSAR study has been performed to explore physicochemical requirements of some substituted 2-aminopyridines and fused rings as inhibitors of Nitric Oxide Synthases. QSAR study has been carried out for two isoforms of the Nitric Oxide Synthases (eNOS and nNOS). QSAR models were derived by stepwise multiple regression analysis employing the method of least squares. The best QSAR model describing the Nitric Oxide Synthases inhibitory activity of 2- Aminopyridine and fused bicyclic ring system was selected on the basis of statistical significance and predictive ability as gauged by cross-validation procedure and external test-set method. The statistical quality of generated equations was evaluated considering parameters like correlation coefficient (r), standard error of estimation (SEE), Fischer ratio (F). The predictive ability of models was checked by external test set by calculating Q². The result of present study may be useful for designing more potent aminopyridine analogues as inhibitors of nitric oxide synthases.

 

 

 

INTRODUCTION:

Nitric oxide (NO) has emerged as one of the most interesting and seemingly ubiquitous mediators of normal and pathothysiological processes^1-4. Several researchers and scientist have been identified 2-Aminopyridines as Nitric oxide Synthase (NOS) inhibitors^5-6. Identifications of the potent inhibitors of nitric oxide synthases which produce more NO have been the subject of interest. In search of effective nitric oxide synthases inhibitors Hagmann et al,⁷ synthesized and evaluated a novel set of 2- Aminopyridines and fused bicyclic rings for their NOS inhibitory activity. We have this series of compounds for our QSAR studies. All the NOS inhibitory activity used in the present study were expressed as pIC₅₀, where IC₅₀ micro molar inhibitory concentration of compounds producing 50% inhibition of nitric oxide synthases. The compounds which were not showing NOS inhibitory activity in the above literature have not been taken for our study.

 

MATERIALS AND METHODS:

Hyperchem (8.0.5 Demo version, Hypercube, Inc),⁸ Molecular modeling pro 6.2.3 (Demo version Cambridge Software corp.),⁹ DRAGON-Software for the Calculation of Molecular Descriptors (release 5.5 for windows 2007)¹⁰ and STATISTICA version 6 (StatSoft, Inc., Tulsa, USA)¹¹. The three dimensional structures of all the molecules were drawn with the Molecular Modeling Pro. All molecules were transferred into Hyperchem software and pre-optimized using MM+ molecular mechanics force field. A more precise optimization is done with the semi-empirical AM1 method. The molecular structures were optimized until the root mean square gradient becomes smaller than 0.001 k/mol A⁰. The optimized molecules were saved as MOL file format. Most stable structure for each compound was generated and used for calculating various physicochemical descriptors like thermodynamic, steric and electronic values of descriptors.

 

The resulting structures were used to calculate constitutional, functional and topological descriptors by using DRAGON software. Meanwhile some electronic descriptors such as frontier molecular orbital (HOMO, LUMO), dipole moment and partial charges were calculated by using Hyperchem software. STATISTICA version 6 (StatSoft, Inc., Tulsa, USA) software was used to generate QSAR models by stepwise multiple linear regression analysis. Statistical measures used were n-number of compounds in regression, r-correlation coefficient, r²- squared correlation coefficient, F- test (Fischer’s value) for statistical significance, SEE- standard error of estimation, Q²- cross validated correlation coefficient and correlation matrix to show correlation among the parameters.

 

RESULTS AND DISCUSSION:

A data set of 26 molecules for inhibitory activity for isoforms eNOS and 17 molecules for nNOS was used for the present study. For eNOS the data set is divided into training set and test set. The chemical structures, physicochemical parameters and indicator variables of 2 – amino pyridines are given table 1. The QSAR study of these series resulted in several QSAR equations. The best equations are

For eNOS

pIC₅₀ = -0.045 (±0.431) MlogP -19.025 (±6.392) Mv -0.004 (±0.007) WhetZ + 18.004 (±3.833)                                   (1)

n = 22, r = 0.640, r² = 0.410, SEE= 0.756, F= 4.163, P= 0.021, Q² = 0.144, PRESS = 14.932, PE= 0.083

 

For nNOS

pIC₅₀ = 0.395 (±0.085) LUMO -9.349 (±3.994) Mp+ 10.362 (±2.740).                                                                             (2)

n= 17, r = 0.849, r² = 0.721, SEE = 0.437, F = 18.080, P = 0.000, Q² = 0.643, PRESS = 3.426, PE = 0.045

 

In model 1 MlogP is Moriguchi octanol-water partition coefficient; Mv is mean atomic van der Waals volume. Both MlogP and Mv contribute negatively for biological activity. It means that least lipophilic substituents should be on the pyridine ring for maximum eNOS inhibitory activity. The volume occupied by the molecule on the van der Waals surface should be least. WhetZ contributes negatively for biological activity which is Wiener-type index from Z weighted distance matrix (Barysz matrix). This model shows good correlation coefficient (r) of about 0.640 between descriptors (MlogP, Mv and WhetZ) and eNOS inhibitory activity. Squared correlation coefficient (r²) of 0.410 explains 41.00 % variance in biological activity. F represents Fischer ratio between the variances of calculated and observed activities which is 4.163. P is the probability value. The cross validated correlation coefficient of this model was 0.144 which shows poor internal prediction power of this model. PRESS is sum of squared deviation between predicted and actual activity values from each molecule in test set.

 

In model 2 LUMO (lowest unoccupied molecular orbital) is the lowest energy level in the molecule that contains no electrons. It is important in governing molecular reactivity and properties Molecules with low-lying LUMOs are more able to accept electrons than those with high LUMOs; thus the LUMO descriptor should measure the electrophilicity of a molecule. Mean atomic polarizability (Mp) which is a constitutional descriptor, and is measured by summation of the atomic contributions (∑a_iE_i, where E_i is the electric field at the corresponding atom and a_i the corresponding polarizability assumed to be isotropic). Model 2 shows good correlation coefficient (r) of 0.849 between descriptors (LUMO and Mp) and nNOS inhibitory activity. Square correlation coefficient (r²) of 0.721 explains 72.10% variance in biological activity. This model also indicates statistical significance > 99.99% with F value = 18.080. The cross correlation coefficient of this model was (Q²) 0.643 which shows good internal predictivity power. Further support in this regard was obtained by high PRESS. The observed, predicted and residuals values of substituted 2-amino pyridines and fused bicyclic rings used in QSAR analysis of eNOS (training set) and nNOS inhibitory activities are given in table 2. Table 3 contains observed, predicted and residuals values of substituted 2-amino pyridines and fused bicyclic rings used in QSAR analysis of eNOS (Test set).

 

Goodness of fit is calculated as PE = 2 (1 – r²)/3√n, if the value of correlation coefficient (r) is six times of PE then the expression is good and reliable. Both the model shows PE of 0.083 and 0.045 respectively.

 

Figure 1. Plot of observed pIC₅₀ against predicted pIC₅₀ of model 1. (Training Set)

 

Figure 2. Plot of observed pIC₅₀ against predicted pIC₅₀ of model 1. (Test set)

 

The test set was designed only for eNOS inhibitory activity and not for nNOS because of less number of molecules reported for nNOS in original series.  Plots of observed vs predicted pIC₅₀ of eNOS for training set is given in fig. 1 and Plot of observed vs predicted pIC₅₀ of eNOS for test set is given in fig. 2. Plots of observed vs predicted pIC₅₀ of nNOS is given in fig. 3.

 

Table 1 Structures, Physicochemical parameters and indicator variables of 2 – Aminopyridines.

Comp no.	R	MlogP	mv	WhetZ	LUMO	Mp
1.	H	0.667	0.63	27.699	2.625	0.66
2.	3-CH3	1.045	0.61	41.860	ND	ND
3.	4-CH3	1.045	0.61	42.914	3.961	0.64
4.	5-CH3	1.045	0.61	43	ND	ND
5.	6-CH3	1.045	0.61	42.143	ND	ND
6.	3,4-(CH3)2	1.034	0.60	59.619	ND	ND
7.	3,5-(CH3)2	1.376	0.60	60.571	3.783	0.63
8.	4,5-(CH3)2	1.784	0.60	60.762	3.96	0.66
9.	4,6-(CH3)2	0.527	0.54	60.571	ND	ND
10.	5,6-(CH3)2	1.875	0.60	60.190	4.111	0.70
11.	4-C2H5	1.763	0.60	65.81	ND	ND
12.	4-CF3	1.835	0.63	108	ND	ND
13.	3-C2H5, 4-CH3	1.732	0.59	85.524	ND	ND
14.	3-NH2, 4-CH3	0.477	0.60	58.476	ND	ND
15.	5-C2H5, 4-CH3	1.732	0.59	87.81	5.873	0.67
16.	4-CH3, 6-C2H5	1.864	0.59	87.429	6.823	0.63
17.	4-CH3, 6-n-C3H7	1.090	0.58	124.28	6.783	0.62
18.	4-CH3, 6-i-C3H7	2.451	0.58	116.28	5.536	0.70
19.	4-CH3, 6-n-C4H9	2.652	0.58	172.14	5.783	0.62
20.	4-CH3, 6-i-C4H9	2.352	0.58	147.14	5.982	0.62
21.	4-CH3, 6-i-C5H11	2.763	0.57	190	3.499	0.62
22.	4-CH3,6(CH2)3Ph	3.458	0.62	200.94	3	0.66
23.		2.056	0.67	92.952	3.873	0.7
24.		2.265	0.67	95.238	3.433	0.66
25.		2.073	0.62	85.857	3.44	0.64
26.		2.049	0.61	112.23	3.523	0.65

ND= not determined

 

Table 2. Observed, predicted and residuals values of substituted 2-amino pyridines and fused bicyclic rings used in QSAR analysis of eNOS and nNOS inhibitory activities.

Comp no.	Model 1. eNOS pIC50                                                                (Training set)				Model 2. nNOS pIC50
	IC50.	Obs. pIC50.	Pred. pIC50.	Res.	IC50.	Obs. pIC50.	Pred. pIC50.	Res.
1.	2.8	5.53	5.86	- 0.339	4.8	5.31	5.22	0.080
2.	1.20	5.92	6.17	- 0.251	ND	ND	ND	ND
3.	0.072	7.14	6.16	0.972	0.075	7.12	6.28	0.837
4.	3.1	5.50	6.16	- 0.666	ND	ND	ND	ND
5.a	0.82	6.08	Test set	Test set	ND	ND	ND	ND
6.	0.15	6.82	6.28	0.533	ND	ND	ND	ND
7.	3.6	5.44	6.26	-0.82	2.0	5.69	5.96	-0.277
8.	0.6	6.22	6.24	-0.027	0.34	6.46	5.69	0.768
9.	0.045	7.34	7.44	-0.106	ND	ND	ND	ND
10.	2.8	5.55	6.24	-0.695	2.2	5.65	5.44	0.207
11.	0.23	6.63	6.22	0.403	ND	ND	ND	ND
12.	72.4	4.14	5.47	-1.330	ND	ND	ND	ND
13.a	3.3	5.48	Test set	Test set	ND	ND	ND	ND
14.	0.081	7.09	6.31	0.773	ND	ND	ND	ND
15.	3.4	5.46	6.32	-0.863	0.61	6.21	6.41	-0.209
16.a	0.049	7.30	Test set	Test set	0.1	7.00	7.16	-0.169
17.	1.0	6.00	6.31	0.980	0.09	7.04	7.24	-0.206
18.	0.2	6.69	6.38	-0.386	1.2	5.92	6.00	-0.086
19.	0.1	7.00	6.35	0.330	0.1	7.00	6.85	0.148
20.	0.15	6.82	6.11	0.889	0.10	7.00	6.93	0.069
21.	1.9	5.72	6.23	0.588	0.51	6.29	5.94	0.341
22.	17.2	4.76	6.21	-0.498	3.3	5.48	5.37	0.102
23.	1.4	5.85	5.18	-0.429	6.6	5.18	5.34	-0.168
24.	24	4.61	4.76	1.084	11	4.95	5.54	-0.598
25.a	16.6	4.77	Test set	Test set	10.6	4.97	5.73	-0.768
26.	19.8	4.70	4.74	-0.13	2.4	5.61	5.67	-0.067

ND = not determined, Out = outlier, Obs = observed activity, Pred. = predicted activity, Res. = residual between observed activity and predicted activity, a = test set for eNOS inhibitory activity.

 

 

Figure 3. Plot of observed pIC₅₀ against predicted pIC₅₀ of model 2.

 

Table 3. Observed, predicted and residuals values of substituted 2-amino pyridines and fused bicyclic rings used in QSAR analysis of eNOS. (Test set)

Comp. no.	IC50	Observed pIC50	Predicted pIC50	Residual
1.	0.82	6.08	6.28	-0.2
2.	3.3	5.48	6.52	-1.04
3.	0.049	7.30	6.51	0.79
4.	16.6	4.77	6.28	-1.51

 

CONCLUSION:

The QSAR analysis for inhibitory activity of nitric oxide synthases by some 2-Aminopyridines and fused bicyclic rings reveals a wide range of correlation coefficient  r = 0.640 for eNOS and  r = 0.849 for nNOS. The squared correlation coefficient r² = 0.410 for eNOS and  r² = 0.721 for nNOS. Cross–validated squared correlation coefficient Q² = 0.144 for eNOS and Q² = 0.643 for nNOS. The predictive power of model 2 is very good than model 1. The study showed that for inhibitory activity of nNOS by 2-Aminopyridines and fused bicyclic rings mostly depend on mean atomic polarizability.

 

ACKNOWLEDGMENT:

Authors are thankful to Principal and Management, Tatyasaheb Kore College of Pharmacy, Warananagar for providing necessary facilities to complete this work.

 

REFERENCES:

1.       Chan K L. Role of Nitric Oxide in Ischemia and Reperfusion Injury. Curr. Med. Chem. – Anti-inflammatory and Anti-Allergic Agents. 2002; 1: 1-13.

2.       Walley M, Behofsits C H,  Simpson R  and Bjarnason I. Nitric oxide: potential role for reducing gastro-enteropathy.  Inflammopharmacology. 2003; 11: 429-436.

3.       Ignarro L J, Napolis C and Loscalzo J. Nitric Oxide Donors and Cardiovascular Agents Modulating the Bioactivity of Nitric Oxide: An Overview. J. Ame. Heart Asso. 2002; 11/25: 22-27.

4.       Wang P G, Xian M, Tang X, Wu X, Wen Z, Cai T  and Janczuk A J. Nitric Oxide Donor: Chemical Activities and Biological Applications. Chem. Rev.2002; 102: 1091-1134.

5.       W. S. Faraci, A. A. Nagel, K. A. Verdries, L. A. Vincent, H. Xu, L. E. Nicholes, J. M. Labase, E. D. Salter and E. R.  Pettipher. 2-Amino-4-methylpyridine as a potent inhibitor of inducible NO synthase activity in vitro and in vivo. Br. J. Pharmacol. 1996, 119, 1101.

6.       E. R.  Pettipher, T. A. Hibbs, M. A. Smith and R. Griffiths. Analgesic activity of 2-amino-4-methylpyridine, a novel NO synthase inhibitor.  J. Infamm. Res. 1997, 46, S135.

7.       Hagmann W K, et al. Substituted 2-Aminopyridines as Inhibitors of Nitric Oxide Synthases. Bioorg. Med. Chm. Lett. 2000; 10: 1975-1978.

8.       Hyperchem, 8.0.5 Demo version, Hypercube Inc. USA.

9.       Molecular modeling pro, 6.2.3 Demo version, Cambridge Software corp., Software publishers Association, 1730 M Street, suite 700, Washington D. C. 20036 (202) 452–1600, U.S.A.

10.     DRAGON, release 5.5 for windows, TALETE srl, Process and Product Optimization and Complex system analysis by chemometric methods via V. Pisani, 13- 20124, Milano, Italy.

11.     STATISTICA, version 6, Stat Soft, Inc., Tulsa, USA

 

 

 

Received on 24.08.2011        Modified on 30.08.2011

Accepted on 05.09.2011        © AJRC All right reserved