8/10/2019 140210130063 Abdul Falahitawan Tugas 5

1/2

Computational analysis and predictive modeling of polymorphdescriptors

Yungyung Lee 1, Sourav Jana 1, Gayathri Acharya 2 and Chi H Lee 2*

1 School of Computing and Endineering, University of Missouri-Kansas City, Missouri, MO 64 !, US"2 #ivision of $harmaceutical Sciences, College of $harmacy, University of Missouri-Kansas City, Missouri, MO 64 !%, US"

Keywords: Binding affinity, QSAR, BCR , o!y"or#h, $itoa%ntrone &

The computational tools intended for quantitativeassessment of protein-ligand interactions are basedon several factors including protein-ligand docking,molecular dynamis simulation and free energycalculations. To better define the role of bindingaffinity in forming a protein-ligand complex, astructural characterization for putative himan off-targets was recently performed on Nelfinavir, a

potent !"-protase inhibitor with pleiotropic effectsin cancer cells. !n this experiment, they have adaptednumerous computational models that integratedmolecular dynamic simulation, free energycalculations with ligand binding site comparison and

biological network analysis.There are two integral screening approaches that

could help identify and characterize the substratesand inhibitors of the efflux proteins and#or transporter system$ the measurement of bindingaffinity and toxicity analysis of substratecompounds. There was a report that drug residenttime and uptake amount are better correlated withdrug efficacy than binding affinity, suggesting thatlead optimization could be efficiently accomplishedwith analyzing the drug uptake profiles. %lthoughnumerous methodologies have been proposed for drug-target screening strategies based on bindingaffinity, there are no efficient computational toolsavailable for the accurate estimation of the druguptake profiles from the point of the molecular structures. !n this study, the uptake rates of &itoaxtrone in the presence of various substratecompounds were examined as an in vitro screeningindex that could help to characterize the binding

properties of chemotherapeutic drugs to tumor cellsor efflux proteins.

'reast cancer resistant protein (')*+ alsoknown as %')+ or & * or %') / is a member of transporter super family %T+ binding cassette (%')

proteins. ')*+ is known to affect thetherapeutically available concentrations of variousclinical agents. 0ince the ')*+ effluxes a widerange of structurally diverse xenobiotic compounds

from cells, the board distribution of ')*+ not onlyrenders less complete distribution of drugs but alsocauses a poor response of cells tochemotherapeutics. ')*+ in con1unction with +-gpexpression at target sites affected the

pharmacokinetic profiles of substrates andinhibitors. 0ubsequently, the therapeuticallyavailable concentrations of certain agents increasedin ')*+ knock-out animal models that were highly

prone to &itoxantrone induced toxicity.The in vitro studies on the ')*+ efflux system

have demonstrated that some cell lines displayederratic efflux profiles of doxorubicin and rhodamine2/3, and these observations were attributable to45/ nd position in the amino acid sequence consistingof arginine, glycine or threonine residues which aresusceptible to numerous posttranslationalmodifications. Three polymorphs, namely 45/* (wild type and two mutants (45/ and 45/T of ')*+, have been identified, and alterations in their expressions and functions were reported 6/27. 8ild-type ')*+ and its variants were markedly expressedin human embryonic kidney ( 9: cells.

The present study was intended to establish therelationships between chemical properties involvedwith the uptake rates of structurally diversesubstrates and ')*+ polymorphs. To achieve thisgoal, we have designed the computational modelconsisting of numerous molecular descriptors. Theuptake rates of &itoxantrone by ')*+ wereexamined in the presence of various

pharmacological classes of %') transporter inhibitors, such as anticiral (i.e. 9rythromysin,;oscarnet , antibiotic (i.e. )iprofloxacin,;ebendazole, Novobiocin,

8/10/2019 140210130063 Abdul Falahitawan Tugas 5

2/2

$an&ano et a!'

&itoxantrone are due to competitive binding of these substrates to ')*+.

!n the development of computational model prediction of structural specificity of substratecompouns to ')*+, three dimensional structures of substrates were built using %&+%) with raphical=ser !nterface (0emichem, 0hawnee &ission, :0 .%&+%) used %ustin &odel 2 (%&2 for thequantum mechanical semi-empirical calculations of interactive energy. )>?900% can generate thenumerical values for molecular descriptors, whereasthe heuristic method ( & preselects appropriatemolecular descriptors. The multiple linear