Development of Efficient Drug Analogs for Dutasteride through Insilico Modeling

摘要

Action of drug depends on the quantity of drug that reaches to the receptor and the degree of affinity between drug and receptor. Once drug bound to its receptor, it shows its desired effect (intrinsic activity) along with major and minor side-effects. More than 500 drugs with significant side-effects are available in market. In the present study the efforts have been made to identify new candidate compounds for dutasteride through modeled structure of two existing drug targets to improve efficacy by optimizing various parameters. Three dimensional structures of both drug targets of dutasteride were generated by MODELLER, validated by PROCHECK and ERRAT programs (90.04% of model-I and 84.647% of model-II). Energy minimization and molecular dynamics calculations were done through GROMACS using OPLS force field. RMSD calculated for both simulated model attain a constant deviation within 10,000 cycles. For analog generation, mono-substitution is preferred instead of multi-substitution. Objective and subjective both approaches were used for identification and calculation of chemical descriptors. A comparison of the calculated binding affinities for structurally similar inhibitors of dutasteride gave suitable analogs. Total 169 analogs were generated and eight are selected for comparative study. Our finding reveals that three dutasteride drug analogs (-CH2NH2, -CH2OH, -CF2OH) were most suitable analogs, showing theoretically more superior results. These findings need to be further evaluations in laboratory. Introduction Soft-computing techniques are concerned with approximate calculation, imprecision, analysis where human capability of making decisions is in dilemma. The recent developments in soft-computing techniques are expanding its scope in pharmaceutical industry. At present thousands of drugs are in market but many of them needed to improve efficacy (minimization of biological toxicity effects and side effects) [1, 2]. In the present study dutasteride (AVODART) drug for hair loss treatment is studied [3]. Dutasteride inhibits both isoforms of 5-alpha reductase, which are considered as drug targets. The most common adverse reactions of Dutasteride are impotence, decreased libido, breast disorders, and ejaculation disorders. Dutasteride belongs to: Drug type - small molecule, Drug category - Approved, Investigational, Anti-baldness agent and Antihyperplasia agents (http://www.drugbank.ca/). The PDB structures of both the drug targets are not available and additionally the molecular modeling of targets was a tougher task due to very low sequence similarity with known PDB templates. In this paper an attempt has been made to model the tertiary structure of drug targets through multiple templates with incorporation of fold assignment and secondary structure information [4, 5]. After molecular dynamics calculations and model assessment, active site characterization, analogs generation, analog selection and descriptors calculations of analogs has been done through various offline and online softwares and tools. Furthermore to explore the efficacy of dutasteride, a comparative analysis of generated analogs has been performed with dutasteride. Material & Methods Dutasteride and its drug target information were obtained from drug bank. Dutasteride drug targets are existing in two isoform i.e. 3-oxo-5-alpha-steroid4-dehydrogenase 1(Drug target-I) and 3-oxo -5-alpha-steroid 4-dehydrogenase 2(Drug target-II) as reported in drug bank database. For these targets 290 and 292 SNPs are reported [6] at location chromosome-5 locus 5p15 and chromosome-2, locus-2p23 respectively.Modeling of drug targets The PDB structures of both drug targets are not available. So, modeling of both the target proteins were performed using MODELLER. A template search has been performed through BLAST and PSI-BLAST programs [7]. Global alignment method was used for comparison between the target-template sequences [4]. Gaps with variable gap penalty function are inclu