Development and validation of a robust QSAR model for prediction of carcinogenicity of drugs.

Carcinogenicity is one of the toxicological endpoints causing the highest concern. Also, the standard bioassays in rodents used to assess the carcinogenic potential of chemicals and drugs are extremely long, costly and require the sacrifice of large numbers of animals. For these reasons, we have attempted development of a global quantitative structure–activity relationship (QSAR) model using a data set of 1464 compounds (the Galvez data set available from http://www.uv.es/~galvez/tablevi.pdf), including many marketed drugs for their carcinogenesis potential. Though experimental toxicity testing using animal models is unavoidable for new drug candidates at an advanced stage of drug development, yet the developed global QSAR model can in silico predict the carcinogenicity of new drug compounds to provide a tool for initial screening of new drug candidate molecules with reduced number of animal testing, money and time. Considering large number of data points with diverse structural features used for model development (ntraining = 732) and model validation (ntest = 732), the model developed in this study has an encouraging statistical quality (leave-one-out Q2 = 0.731, R2pred = 0.716). Our developed model suggests that higher lipophilicity values and conjugated ring systems, thioketo and nitro groups contribute positively towards drug carcinogenicity. On the contrary, tertiary and secondary nitrogens, phenolic, enolic and carboxylic OH fragments and presence of three-membered rings reduce the carcinogenicity. Branching, size and shape are found to be crucial factors for drug-induced carcinogenicity. One may consider all these points to reduce carcinogenic potential of the molecules.

Comparative QSAR modeling of COX-2 inhibitor 1,2-diarylimidazoles using E-state and physicochemical parameters.

Considering importance of developing selective COX-2 inhibitors, COX-2 binding affinity data of 4-(2-aryl-1-imidazolyl)-phenyl methyl sulfones and sulfonamides (n = 83) have been modeled using electrotopological state (E-state) index as electronic parameter, hydrophobic substituent constant (pi) and molar refractivity (MR) of aryl ring substituents as lipophilic and steric parameters, respectively. Additionally, suitable dummy parameters have been used for the development of multiple regression equations in a stepwise manner. The study suggests that lipophilicity of ortho, meta and para substituents of the aryl ring increases the binding affinity, while molar refractivity (MR) of ortho and meta substituents of the aryl ring decreases the binding affinity. Again, electron-withdrawing substituents at meta and para positions of the aryl ring increase the binding affinity. Additionally, a 4-fluoro substituent on the aryl ring, a trifluoromethyl substituent at R position and simultaneous presence of 3-chloro and 4-methyl groups on the aryl ring are conducive to the binding affinity. Also, an amino substituent is preferred over a methyl group at R2 position suggesting preference of the sulfonamide moiety over the methyl sulfone moiety for the COX-2 binding affinity. Furthermore, importance of E-state values of different atoms in the generated relations suggests the influence of electron density distribution over the 1,2-diarylimidazole nucleus for the binding affinity. For this data set, E-state parameters perform better as electronic parameters in comparison to Hammett sigma parameters. When lipophilic whole molecular descriptor (ClogP) is used, instead of hydrophobic substituent constant (pi), the former performs better than the latter.

Exploring QSAR of peripheral benzodiazepine receptor binding affinity of N,N-dialkyl-2-phenylindol-3-yl-glyoxylamides using physico-chemical descriptors.

The present QSAR study has attempted to explore the structural and physicochemical requirements of ligands N,N-dialkyl-2-phenylindol-3-yl-glyoxylamides for binding with peripheral benzodiazepine receptor (PBR). The calculated partition coefficient values show parabolic relations with the PBR binding affinity, suggesting that the binding affinity increases with increase in the partition coefficient of the compounds until it reaches the critical value after which the affinity decreases. The critical value of logP is within range of 6.052-6.410. Furthermore, positive Wang-Ford.charge values of carbonyl oxygens of the glyoxamide moiety and negative Wang-Ford charge value of the glyoxamide nitrogen are conducive for the binding affinity. Again, the indole moiety should have favorable charge distribution. Higher values of the parameters dipole moment (Dipole) and moment of inertia (I_z) of the ligands are conducive for the binding affinity. The presence of hydrogen atom at R2 and cyclic moiety at R1 and R2 positions are detrimental to the binding affinity.

Exploring selectivity requirements for peripheral versus central benzodiazepine receptor binding affinity: QSAR modeling of 2-phenylimidazo[1,2-a]pyridine acetamides using topological and physicochemical descriptors.

Considering the potential of peripheral benzodiazepine receptor (PBR) ligands in therapeutic applications and clinical benefit in the management of a large spectrum of different indications, quantitative structure-activity relationship (QSAR) study has been attempted to explore the structural and physicochemical requirements for selectivity of 2-phenylimidazo[1,2-a]pyridineacetamides for binding with peripheral over central benzodiazepine receptors (CBRs). For PBR binding affinity, molar refractivity (MR) shows a parabolic relation with binding affinity suggesting that binding affinity increases with increase in volume of the compounds, until it reaches the critical value, after which the affinity decreases. The negative coefficients of S_aaN and S_ssNH indicate that binding affinity increases with decrease in E-state value of (N/) (aromatic nitrogen) and HN< (secondary amino group) fragments. The coefficient of 3XVC and JX term indicates the importance of shape and branching for binding affinity. For CBR binding affinity, lipophilicity of molecules is detrimental to the binding affinity, while presence of hydrogen at Y position is conducive to the activity. Selectivity pattern of these ligands for peripheral (cortex) over central receptors requires the presence and absence of methyl group at R2 and R3 positions respectively, and shows the importance of MR and shape parameter. Similarly, selectivity of these ligands for peripheral (ovary) over central receptors requires the presence and absence of methyl group at R2 and R3 positions respectively, presence of phenyl group at R1 and R2 positions and selectivity relation shows importance of MR, shape and branching.

Recognition schemes for protein-nucleic acid interactions.

The molecular forces involved in protein-nucleic acid interaction are electrostatic, stacking and hydrogen-bonding. These interactions have a certain amount of specificity due to the directional nature of such interactions and the spatial contributions of the steric effects of different substituent groups. Quantum chemical calculations on these interactions have been reported which clearly bring out such features. While the binding energies for electrostatic interactions are an order of magnitude higher, the differences in interaction energies for structures stabilised by hydrogen-bonding and stacking are relatively small. Thus, the molecular interactions alone cannot explain the highly specific nature of binding observed in certain segments of proteins and nucleic acids. It is therefore logical to assume that the sequence dependent three dimensional structures of these molecules help to place the functional groups in the correct geometry for a favourable interaction between the two molecules. We have carried out 2D-FT nuclear magnetic resonance studies on the oligonucleotide d- GGATCCGGATCC. This oligonucleotide sequence has two binding sites for the restriction enzyme Bam H1. Our studies indicate that the conformation of this DNA fragment is predominantly B-type except near the binding sites where the ribose ring prefers a 3E conformation. This interesting finding raises the general question about the presence of specificity in the inherent backbone structures of proteins and nucleic acids as opposed to specific intermolecular interactions which may induce conformational changes to facilitate such binding.