Available online through ISSN

Transcription

1 International Journal of Pharmaceutical Archive-1(2), 2012, Available online through I Research Article QAR tudy of Benzimidazole Derivatives Against MetAPs Using MLR Approach Ajeet* Department of Medicinal Chemistry,. D. College of Pharmacy and Vocational tudies, Bhopa Road, Muzaffarnagar, U.P., India, PI (Received on: ; Revised & Accepted on: ) ABTRACT ere Benzimidazole analogues have been used to correlate the inhibition activity with the topohape and geomhape descriptors for studying the Quantitative tructure Activity Relationship (QAR). Correlation may be an adequate predictive model which can help to provide guidance in designing and subsequently yielding greatly specific compounds that may have reduced side effects and improved pharmacological activities. We have used Multiple Linear Regression (MLR) for developing QAR model. For the validation of the developed QAR model, statistical analysis such as cross validation test, quality factor, fischers test, root mean square deviation (RMD) etc.; have been performed and all the tests validated this QAR model with fraction of variance r 2 = ubject classification-bio-tatistics Key words- QAR; Multiple Linear Regression; Benzimidazole analogues; Methionine aminopeptidases (MetAPs). ITRODUCTIO The enzyme methionine aminopeptidase (MetAP) is a dinuclear metallo-protease that removes the -terminal methionine from proteins (nascent proteins) (Giglione et al., 2003; Lowther and Matthews, 2000). MetAPs are conserved in all life forms ranging from bacteria to humans. MetAP1 and MetAP2 are two classes of MetAPs, which differ in the presence of an internal polypeptide insertion present within the catalytic domain of MetAP2 (Arfin et al., 1995) In the present study, we developed a QAR model on a series of benzimidazole analogues with respect to their IC 50. The QAR studies are perfect tool for understanding the drug design process in terms of their chemicalpharmacological activity interaction, along with it is also used in toxicology and pesticide research. QAR studies can focus on mechanism of action of ligands with uman, bacteria, virus, membranes, enzymes etc. It can also be used for the evaluation of the metabolism, absorption, distribution and excretion phenomena. The QAR methodology comprises of computationally derived descriptors to correlate with pharmacological activities. These descriptors are principally of four types such as electronic, stearic, hydrophobic and topological indices (Verma et. al. 2010). The descriptors used for developing the QAR model are topohape and geomhape (Yap 2011), ( Petitjean 1992). Rational Drug Design helps to facilitate and fasten the drug designing process, which involves various methods to identify novel compound (Ajeet et. al a), (Ajeet 2012), (Ajeet et. al b). MATERIAL AD METOD All the bioactivity values and information about 2D structure of benzimidazole analogues were taken from literature (Zahra Garkani-ejad et. al. 2010). IC 50 is a variable that comprises the bioactivity parameter for the QAR model. In order to calculate the 2D molecular descriptors, PaDEL descriptor software (Yap 2011), which incorporate CDK library for descriptor calculation has been used after optimitizing the benzimidazole analogues. For the development of QAR model, Multiple Linear Regression (Verma et. al. 2010) has been employed and all were validated through statistics. MODELIG PARAMETER AD TRUCTURE OPTIMIZATIO The 2D structure construction, energy minimization and geometry optimization of the selected benzimidazole derivatives were carried out by using ChemDraw Ultra 7.0 and Chem3D Pro 7.0 (Cambridgeoft Corporation, 100 CambridgePark Drive, Cambridge MA, UA) on an Intel(R) Core(TM)2 Duo Central Processing Unit 2.20 Gz and 4.00 GB of RAM, running the Windows 7 ome Basic, 64-bit compatible operating system. The energy minimization was carried out to minimum RM Gradient of 0.100, with step interval of 2.0 Fs and frame interval of 10 Fs. Descriptors calculated for the training set are given in Table 1. Corresponding author: Ajeet*, Department of Medicinal Chemistry,. D. College of Pharmacy and Vocational tudies, Bhopa Road, Muzaffarnagar, U.P., India, PI International Journal of Pharmaceutical Archive- 1 (2), ov

2 Table 1. Descriptors calculated for the training set. ALOGP APol Aromatic atoms count Aromatic bonds count Atom count Autocorrelation (arge) Autocorrelation (Mass) Autocorrelation (Polarizability) BCUT Bond count BPol Carbon types Chi chain Chi cluster Chi path cluster Chi path Eccentric connectivity index Atom type electrotopological state Fragment complexity bond acceptor count bond donor count Kappa shape indices Largest chain Molecular linear free energy relation Petitjean number Rotatable bonds count Rule of five TPA VadjMa Weight Weighted path Wiener numbers XlogP Zagreb index CPA Gravitational index Length over breadth Moment of inertia Petitjean shape index (topohape, geomhape) WIM Largest Pi system Longest aliphatic chain Mannhold LogP McGowan volume MDE DECRIPTOR ELECTIO The selection of descriptors among the calculated descriptors for the multiple linear regression analysis is based on the correlation matrix. This matrix is prepared and analyzed for the least correalated descriptors. The correlation matrix is given in Table 2. Table 2. Correlation matrix topohape geomhape topohape geomhape TATITICAL PARAMETER Fraction of variance (r 2 ): The value of fraction of variance may vary between 0 (means model without explanatory power) and 1 (means perfect model). QAR model having r 2 > 0.6 will only be considered for validation (Verma et. al. 2010). Cross-Validation Test (q 2 ): A QAR model must have q 2 > 0.5 for the predictive ability (Verma et. al. 2010). tandard deviation (s): The smaller s value is always required for the predictive QAR model. r 2 -q 2 < 0.3: The difference between r 2 and q 2 should never be exceeding by 0.3. A large difference suggests the following: presence of outliers, over-fitted model, and presence of irrelevant variables in data (Verma et. al. 2010). Quality Factor (Q): Overfitting and chance correlation, due to excess number of descriptors, can be detected by Q value. Positive value for this QAR model suggests its high predictive power and lack of overfitting. Fischer tatistics (F): The F value of QAR model was compared with their literature value at 95% level. MODEL VALIDATIO The QAR model validation was carried with statistical analysis. 2012, IJPA Online, All Rights Reserved 43

3 REULT AD DICUIO From the data in Table 3, QAR equation have been developed, here 95% confidence intervals are given in parantheses. Their structures are given in Table 4. IC 50 = ( ) ( ) ( ) (topohape) ( ) (geomhape) Table 3. Descriptors used to derive QAR equation along with bioactivities Training set IC 50 Descriptors used.no. Obs. Pred. Diff. topohape geomhape 1 Thiabendazole Methyl-2-thiazol-4-yl-1(3)-benzoimidazole itro-2-thiazol-4-yl-1(3)-benzoimidazole Pyridin-2-yl-3-benzoimidazol-5-ylamine Fluoro-2-pyridin-2-yl-1-benzoimidazole Chlor-2-thiazol-4-yl-1-benzimidazol Chloro-2-pyridin-2-yl-1(3)-benzoimidazole Thiazol-4-yl-1-imidazo[4,5-b] pyridine Pyrazin-2-yl-1-benzoimidazol Amino-2-thiazol-4-yl-1- benzimidazole Fluor-2-thiazol-4-yl-1-benzimidazol Phenyl-(2-pyridin-2-yl-3-benzoimidazol-5-yl) - methanone Benzimidazol-2-yl-thiourea Table 4. tructures used for model development O + O F 2 5. Cl 6. Cl O 11. F , IJPA Online, All Rights Reserved 44

4 Validation of QAR model A quantitative assessment of model robustness has been performed through model validation. All the statistical results of model validation have been given in Table 5. n/p (>=4) Table 5. Results of statistical validation r 2 q 2 r 2 - q 2 < 0.3 Q RMD F n= no. of molecules taken for modeling, p= no. of descriptors used According to the developed QAR model, the benzimidazole analogues must have positive topohape for enhanced activity. Moving towards the effects of the geomhape on the bioactivity of derivatives of benzimidazole analogues, the developed QAR model suggest that a negative geomhape will definitely be favourable to the activity, as discussed by Verma and ansch (2010), Ajeet et. al. (2012 a), Ajeet (2012), Ajeet et. al. (2012 b). A comparison (multiple linear regression plots) of observed values and predicted values of IC 50 for benzimidazole analogues used for development of QAR equation is shown in Figure 1 and multiple linear graph is shown in Figure 2. In Figure 2 the calculated IC 50 values have been shown with the mean error with the crossed sign. Calculated IC R² = Predicted IC50 Fig. 1. Multiple linear regression plot for QAR study IC umber of Data Points Fig. 2. Multiple linear graph between umber of data points and bioactivities COCLUIO With deluged data of QAR studies of benzimidazole analogues, we could draw a number of conclusions. On the basis of discussion given earlier we could conclude that benzimidazole analogues must have positive topohape for enhanced activity. Moving towards the effects of the geomhape on the bioactivity of derivatives of benzimidazole analogues, the developed QAR model suggests that a negative geomhape will definitely be favourable to the activity. o for developing the novel benzimidazoles on the basis of topohape and geomhape we have to select such groups or substituents which increase the topohape of the novel molecules and simultaneously decrease the geomhape of the molecule and we can evaluate the novel molecules for their IC 50 values on the basis of the derived QAR equation. 2012, IJPA Online, All Rights Reserved 45

5 REFERECE [1] Ajeet, Kumar Vipul, ingh Brajpal. QAR modeling of gelatinase inhibitors: MLR approach. Int. Res. Jour. of Pharm. 2012; 3(3), [2] Ajeet, ingh Brajpal, Kumar Vipul. QAR modeling of 2-[C(O)X]-5,8-(OY) 2-1,4-naphthoquinines against L1210 cells using multiple linear regression. Indonesian J. Pharm. 2012; 23(3), [3] Ajeet. QAR modeling of recombinant human stromelysin inhibitors MLR approach. Int. J. Adv. Pharm. Biol. ci. 2012; 2(2), [4] Arfin. M., Kendall R. L., all L., Weaver L.., tewart A. E., Matthews, B. W., Bradshaw, R.A. Eukaryotic methionyl aminopeptidases: twoclasses of cobalt-dependent enzymes. Proc. atl. Acad. ci. UA. 1995; 92, [5] Giglione C., Vallon, O., Meinnel, T. Control of protein life-span by -terminal methionine excision. EMBO J. 2003; 22, [6] Lowther W.T., Matthews B.W. tructure and function of the methionine aminopeptidases. Biochim. Biophys. Acta. 2000;1477, [7] Michel Petitjean. Applications of the radius-diameter diagram to the classification of topological and geometrical shapes of chemical compounds. J. Chem. Inf. Comput. ci. 1992; 32 (4), [8] Verma RP, ansch C. QAR modeling of taxane analogues against colon cancer. Eur. J. Med. Chem.2010; 45: [9] Yap C.W. PaDEL-descriptor: An open source software to calculate molecular descriptors and fingerprints. J. Comput. Chem. 2011; 32(7), [10] Zahra Garkani-ejad, Fereshteh aneie. QAR study of benzimidazole derivatives inhibition on escherichia coli methionine aminopeptidase. Bull. Chem. oc. Ethiop. 2010; 24(3): ource of support: il, Conflict of interest: one Declared 2012, IJPA Online, All Rights Reserved 46