Drug Design And Molecular Modeling Studies Of Macrophage Migration Inhibitory Factor

Macrophage migration inhibitory factor (MIF) is a multi-function protein thatoperates as a cytokine and enzyme. As a cytokine, MIF is a major regulator ofinflammation and a central upstream mediator of innate immune response and functionsas a key mediator to counter-regulate the inhibitory effects of glucocorticoids within theimmune system. Furthermore, multiple studies have highlighted a role of MIF in tumorgrowth. MIF can directly affect normal cell division and malignant transformationinduced by tumor gene, or regulate immune response indirectly, such as control of cellproliferation and promotion of angiogenesis. Apart from its physiologic andpathophysiologic activities, the most unusual property of MIF is its ability to act as acatalyst with several unique biochemical activities, such as a D-dopachrometautomerase, a phenylpyruvate tautomerase and a thiol-protein oxidoreductase. Due tothe wide involvement of several aspects of immune, inflammation and tumors, such aspathogenic mechanism, diagnosis, prognosis and treatment, MIF has been regarded asan important therapeutic target in inflammation and tumor diseases. Current therapeuticstrategies for targeting MIF mainly focus on developing small inhibitors toward itstautomerase activities. Although these inhibitors provide the proof of concept for thetherapeutic benefit of targeting MIF, most of them show low activities, and thus none ofthem is in clinical use. In addition, the pathogenic mechanisms and biological functionsof MIF-associated diseases, the exact role of the enzymatic functions of MIF, thecomplicated signaling pathways implicated in its pleiotropic functions, the interactionsbetween MIF and its transmembrane receptor CD74or functional receptor CXCR,remain unclear.First, the interactions between MIF and a series of phenolic hydrazones wereinvestigated by molecular docking, molecular dynamics (MD) simulations, MM/GBSAbinding free energy calculations and binding energy decomposition analysis. Thedynamic binding process of the studied inhibitors, the impact of the essential residues ofMIF and the important substituents of the inhibitors on ligand binding were analyzed. Based on the molecular modeling studies obtained above, a series of derivatives weredesigned and higher inhibitory activities of three derivatives were confirmed bytheoretical predictions.Then, based on the crystal structure of MIF, Glide docking was utilized to identifypotential molecules that can interact with MIF from two chemical databases(ChemBridge and Specs) with~1.1million structures. After employing the REOS rulesand structural clustering basis on fingerprints,150compounds were purchased andsubmitted to the experimental assays. The in vitro enzyme-based assay shows that tenchemically diverse compounds have concentration-dependent inhibitory activity againstMIF in the micromolar regime, including three compounds with IC50values below10μM and one below1μM (0.55μM). The chemotaxis assay, enzyme-linkedimmunosorbent assay and cell viability assay illustrate that these three compounds canalso inhibit the biological properties of MIF in vitro, suggesting that these two activitiesare closely linked. These compounds are potential lead compounds to develop reagentsfor combating inflammatory diseases.Recently, CXCR2was reported to be a functional receptor of MIF. Severalmolecular modeling methods were employed to explore the binding mode of MIF withCXCR2. First, homology modeling was used to construct the3D structure of CXCR2.Then, ZDOCK was employed to generate the possible binding structures of MIFmonomer in complex with CXCR2. Three representative docked poses chosen fromZDOCK were submitted to20ns MD simulations. MM/GBSA binding free energycalculations and free energy decomposition were employed to determine the mostfavorable CXCR2-MIF binding pattern and predict the hot-spot residues in theprotein-protein interface. The predicted CXCR2-MIF binding pattern is in goodagreement with the experimental data, and these results contribute to the insights intothe interactions between MIF and CXCR. CXCR4is another functional receptor of MIF,and it is the only crystal structure among all chemokine receptors. The protein-proteininteraction and near-native structure between CXCR4and its natural ligand CXCL12,were predicted by an integrated protocol. MM/GBSA binding free energy calculationand free energy decomposition analysis are in good agreement with the experimentaldata. Based on the dynamic and energetic analyses, a two-site binding model wasproposed.In the last chapter of this thesis, we systematically explored the capability of theMM/GBSA and MM/PBSA approaches to rank the binding free energies of five sets of protein ligand systems by mixing five AMBER force fields (ff99, ff03, ff99SB,ff99SB-ILDN, and ff12SB) for proteins and four methods (RESP, ESP, AM1-BCC, andGasteiger) of obtaining partial charges for small molecules. The data set for theMM/PBSA and MM/GBSA calculations include five receptors: Avidin, Humanthrombin, Neuraminidase, Pim1and Spleen tyrosine kinase. The following conclusionscan be obtained:(1) for short time-scale MD simulations (1ns or less), the ff03forcefield gives the best predictions by both MM/GBSA and MM/PBSA;(2) for middletime-scale MDsimulations (24ns), MM/GBSA based on the ff99force field yields thebest predictions, while MM/PBSA based on the ff99SB force field does the best;(3) formost cases, MM/PBSA with the Tan’s parameters shows better ranking capability thanMM/GBSA (GBOBC1);(4) the RESP charges show the best performance for bothMM/PBSA and MM/GBSA, and the AM1-BCC and ESP charges can also give fairlysatisfactory predictions. Our results are useful for selecting proper force field and ligandcharges for binding free energy calculation based on relatively short MD simulations(≤10ns).