Pin‐Chih Su

To validate a method for predicting the binding affinities of FabI inhibitors, three implicit solvent methods, MM-PBSA, MM-GBSA, and QM/MM-GBSA were carefully compared using 16 benzimidazole inhibitors in complex with Francisella tularensis FabI. The data suggests that the prediction results are sensitive to radii sets, GB methods, QM Hamiltonians, sampling protocols, and simulation length, if only one simulation trajectory is used for each ligand. In this case, QM/MM-GBSA using 6 ns MD simulation trajectories together with GB(neck2) , PM3, and the mbondi2 radii set, generate the closest agreement with experimental values (r(2) = 0.88). However, if the three implicit solvent methods are averaged from six 1 ns MD simulations for each ligand (called "multiple independent sampling"), the prediction results are relatively insensitive to all the tested parameters. Moreover, MM/GBSA together with GB(HCT) and mbondi, using 600 frames extracted evenly from six 0.25 ns MD simulations, can also provide accurate prediction to experimental values (r(2) = 0.84). Therefore, the multiple independent sampling method can be more efficient than a single, long simulation method. Since future scaffold expansions may significantly change the benzimidazole's physiochemical properties (charges, etc.) and possibly binding modes, which may affect the sensitivities of various parameters, the relatively insensitive "multiple independent sampling method" may avoid the need of an entirely new validation study. Moreover, due to large fluctuating entropy values, (QM/)MM-P(G)BSA were limited to inhibitors' relative affinity prediction, but not the absolute affinity. The developed protocol will support an ongoing benzimidazole lead optimization program.

In Escherichia coli, ClpYQ (HslUV) is a two-component ATP-dependent protease in which ClpQ is the peptidase subunit and ClpY is the ATPase and the substrate-binding subunit. The ATP-dependent proteolysis is mediated by substrate recognition in the ClpYQ complex. ClpY has three domains, N, I, and C, and these domains are discrete and exhibit different binding preferences. In vivo, ClpYQ targets SulA, RcsA, RpoH, and TraJ molecules. In this study, ClpY was analyzed to identify the molecular determinants required for the binding of its natural protein substrates. Using yeast two-hybrid analysis, we showed that domain I of ClpY contains the residues responsible for recognition of its natural substrates, while domain C is necessary to engage ClpQ. Moreover, the specific residues that lie in the amino acid (aa) 137 to 150 (loop 1) and aa 175 to 209 (loop 2) double loops in domain I of ClpY were shown to be necessary for natural substrate interaction. Additionally, the two-hybrid system, together with random PCR mutagenesis, allowed the isolation of ClpY mutants that displayed a range of binding activities with SulA, including a mutant with no SulA binding trait. Subsequently, via methyl methanesulfonate tests and cpsB::lacZ assays with, e.g., SulA and RcsA as targets, we concluded that aa 175 to 209 of loop 2 are involved in the tethering of natural substrates, and it is likely that both loops, aa 137 to 150 and aa 175 to 209, of ClpY domain I may assist in the delivery of substrates into the inner core for ultimate degradation by ClpQ.

The enzymes of the bacterial fatty acid biosynthetic pathway (FAS II), represent attractive targets for antimicrobial drug design, because their mammalian counterpart (FAS I) uses a single, multifunctional enzyme with low sequence and structural similarity. This provides an opportunity to selectively target this essential bacterial pathway without interfering with mammalian enzymes that could result in off-target effects. The enoyl-[acyl-carrier-protein] reductase enzyme, FabI, catalyzes the reduction of a double bond in enoyl-ACP to acyl-ACP as a key step in the bacterial production of fatty acids. The FabI essentiality in the two target bacteria in this dissertation, F. tularensis and S. aureus, has been strongly proven in vivo.\nThis thesis project is divided into three parts. The first part of the thesis project is to perform computer aided drug design techniques (virtual screening and data mining algorithms) to identify novel scaffolds FabI inhibitors, as highlighted in Chapter 2. In this hit discovery part, three novel FabI inhibitor scaffolds showed FabI enzymatic inhibitory activities. Among these three inhibitors, the initial SAR extension suggests that the benzimidazole scaffold inhibitors have tractable SAR in enzymatic activities and show Gram-positive and –negative anti-bacterial activities. Therefore, the benzimidazole scaffold was selected for further optimization. \nThe second part is to build structure based computational models (both implicit and explicit solvent methods) to predict FabI benzimidazole inhibitors’ activities and to prioritize inhibitors to synthesize, as detailed in Chapter 3 & 4. The last part (Chapter 5) is to use bioinformatics analysis to identify more bacterial pathogens, of which FabI would be their only enoyl acyl reductase, in the hope to expand the inhibitory spectrum and clinical values of the developed FabI inhibitors. \nIn summary, it is anticipated that the above computer aided lead optimization cycle will help generate benzimidazole inhibitors in sub-nM range for FabI enzymatic activities, and single digit or lower MICs (µg/mL) against the tested Gram-positive and –negative bacteria. Promising benzimidazole inhibitors will be selected for animal challenge and pharmacokinetic tests. We believe that FabI inhibitors will be developed into a FDA-approved antibiotic, and ease the biodefense and public health threats that our society is facing.