Chemotype-based Designing of Mycobacterium tuberculosis BioA inhibitors: Docking studies and ADME analysis Aparna Bahuguna, Shiv Shyam Maurya, Deepak Kumar, Prija Ponnan, Diwan S. Rawat* Department of Chemistry, University of Delhi, Delhi-110007, India E-mail: [email protected] Adenosylmethionine-8-amino-7-oxononanoate aminotransferase (BioA),a PLP-dependent transaminase of M. tuberculosis (Mtb), catalyzes the reversible transfer of the α-amino group from SAM to KAPA to form DAPA. 7-Keto-8-aminopelargonic acid (KAPA) 7,8-Diaminopelargonic acid (DAPA) S-Adenosyl-(2-oxo-4-thiobutyrate) S-Adenosylmethionine (SAM) BioA Crystal structure bound to KAPA (PDB ID: 4CXQ) showing H-bond interactions residues Tyr25, Trp64, Arg400, Tyr157, Gly316 and covalent interaction With co-factor PLP . BioA has recently gained much importance as anti-TB drug target for its specificity in Mtb and absence of the enzyme in higher eukaryotic organisms such as plants and mammals. Hence,t here has been considerable effort in identifying novel lead molecule as BioA inhibitors. Active Site of Mtb BioA Bioactive conformation of KAPA ABA-604 ABA-601 ABA-603 ABA-600 SSM-7 ABA-3 ABA-30 ABA-100 ABA-101 ABA-602 mol_MW :334.43,QPlogPo/w: 1.13, DonorHB:3, AccptHB:9 PSA: 92.41 #rotor:7 QPlogHERG: -6.382 QPlogKhsa: -0.366 PercentHumanOralAbsorption:68.708 QPPCaco: 91.842 mol_MW: 441.435,QlogPo/w: 2.775,DonorHB: 1 AccptHB: 6 #Rotor: 6 QPlogHERG: -6.71 QPPCaco: 17.56 QPlogKhsa: 0.42 HumanOralAbsorption: 3 mol_MW : 290.318,QPlogPo/w:5, DonorHB:0, AccptHB:4.75 PSA: 76.445 #rotor:5 QPlogHERG: 76.445 QPlogKhsa: -0.13 PercentHumanOralAbsorption: 91.974 QPPCaco: 656.433 mol_MW :311.383,QPlogPo/w:1.009 DonorHB:1, AccptHB:1.75 PSA:76.464 #rotor:2 QPlogHERG:-4.551 QPlogKhsa:-0.398 PercentHumanOralAbsorption: QPPCaco:131.385 Mol_MW: 360.52, QlogPo/w: 2.58, DonorHB: 1 AccptHB: 8 #rotor: 7 QPlogHERG: -7.27 QPPCaco: 164.29 QPlogKhsa: 0.09 HumanOralAbsorption: 3.00 mol_MW: 304.39, QlogPo/w: 3.42,DonorHB: 1 AccptHB: 5 #rotor: 3 QPlogHERG: -5.22 QPPCaco: 779.82 QPlogKhsa: 0.49 HumanOralAbsorption: 1.00 mol_MW 347.30, QlogPo/w: 2.93,DonorHB: 2 AccptHB: 6 # rotor: 3 QPlogHERG: -5.96 QPPCaco: 530.44 QPlogKhsa: 0.08 HumanOralAbsorption: 3.00 mol_MW: 270.33, QlogPo/w: 0.23,DonorHB: 2 AccptHB:7 # Rotor: 3 QPlogHERG: -4.22 QPPCaco: 35.18 QPlogKhsa: -0.53 HumanOralAbsorption: 2.00 mol_MW: 341.46, QlogPo/w: 2.58,DonorHB: 1 AccptHB: 7 #rotor: 5 QPlogHERG: -7.85 QPPCaco: 85.63 QPlogKhsa: 0.33 HumanOralAbsorption: 3.00 Mol_MW: 355.48, QlogPo/w: 2.64,DonorHB: 1 AccptHB: 7 # rotor: 5 QPlogHERG: -7.42 QPPCaco: 89.29 QPlogKhsa: 0.36 HumanOralAbsorption: 3.00 KAPA was taken as the reference molecule, XED force field was used to generate field points of KAPA Fragment structure (colored in black), suggested by Dai et al., 2014 was aligned to KAPA Cresset’s Spark program was used to grow new ligand by Bioisosteric substitution Final structures of 3D molecules were selected based on field and shape similarity relative to the reference template molecule KAPA. 4CXQ active site was used as the excluded volume to penalize the steric clashes. Alignment scores ranged from 0.54-0.62. Positive Negative Hydrophobic Shape References 1. R. Dai, D. J. Wilson, T. W. Geders. et al. ChemBioChem. 2014, 15, 575. 2. S. Dey, J. M. Lane, R. E. Lee, E. J. Rubin, J.C. Sacchettini . Biochemistry. 2010, 49, 6746. 3. R. Dai, T. W. Geders, F. Liu. et al. J. Med. Chem. 2015, 58, 5208. 4. S. W. Park, D. E. Casalena, D. J. Wilson. et al. Chem. Biol. 2015, 22, 76. Pharmacokinetics properties prediction of lead molecules by QikProp (Schrödinger, Inc., New York, NY, 2012) 2D-Diagram showing interaction of lead molecules in the active site of Mtb BioA predicted by Glide docking program (Schrödinger, Inc., New York, NY, 2012) Glide energy (-59.70 kcal mol-1 to -42.35 kcal mol-1) and high GlideXP score (-11.32 kcal mol-1 and -9.38 kcal mol-1) Cresset’s Forge program was used to Field-based alignment of structurally diverse hit molecule obtained from Spark 3D molecular electrostatic potential (MEP) Field Points Radial plots help to quickly visualize the physical properties data of molecules by converting numerical information into plots. It provides trends, and sort on overall fit to the project profile or find outliers enabling to focus on the results. ABA-3 ABA-30 ABA-100 ABA-101 ABA-601 SSM-7 ABA-603 ABA-602 ABA-600 ABA-604 Conclusion 1. Spark and Forge program (Cresset Bio Molecular Discovery Ltd., Cambridgeshire, UK, 2015) were used for linking chemistry. scaffold replacement and Field-based alignment. 2. Final structures of 3D molecules were selected based on field and shape similarity relative to the reference template molecule KAPA. 3. Protein active site was used as the excluded volume to penalize the steric clashes. 4. A series of 300 molecules were designed and QikProp program (Schrödinger, Inc., New York, NY, 2012) was used to screen the molecular dataset according to ADMET parameters. 5. Compounds with ADMET parameters in permissible ranges were docked in the active site of Mtb BioA using Glide program (Schrödinger, Inc., New York, NY, 2012). 6. Compounds ABA series and SSM-7, having lowest Glide energy (-59.70 kcal mol-1 to -42.35 kcal mol -1 ) and high GlideXP score (-11.32 kcal mol-1 and -9.38 kcal mol -1 ) showed good interactions with the active site of Mtb BioA. 7. These novel lead molecules designed by in silico approach could be viewed as potential Mtb BioA inhibitors.