Escalating problems with drug resistance continue to compromise the effectiveness of commercial antibiotics, necessitating the search for novel classes of antimicrobial agents. To circumvent problems with resistance, a multi-target single-pharmacophore approach has been employed to discover inhibitors that maintain a balanced activity against multiple target enzymes. In this chapter we examine the application of computational techniques, in particular, structure-based drug design approaches, to

more »

... design new dual-targeting antibacterial agents against bacterial topoisomerases. 2 Keywords: Structure-based drug design, Structure-activity relationships, multi-target single pharmacophore, virtual screening, docking, DNA gyrase. Introduction Almost seventy years ago, penicillin was used to save the life of an infected patient suffering from streptococcal sepsis (1). Since then, many new classes of antibiotics have been discovered and developed. However, bacteria, the champions of evolution, have adapted and developed resistance against frontline antibiotics, such as vancomycin, methicillin, fluoroquinolones and macrolides (2-7). A growing clinical concern is that bacteria are becoming increasingly multidrug-resistant where even new antibiotics, such as linezolid (8) and daptomycin (9), are encountering significant resistance. Consequently, the unmet medical need caused by prevailing bacterial drug resistance has renewed interest in the discovery and development of new classes of antibiotics (10) with novel mechanisms of action (11-13). Pharmaceutical researchers are pursuing a multi-target single-pharmacophore approach (14) to fight antibiotic resistance. This new strategy focuses on the design of inhibitors that maintain well-balanced activity against multiple target enzymes from multiple pathways, like bacterial topoisomerase IV and DNA gyrase, dramatically reducing the probability of resistance incidence (15). The utilization of microbial genomics (16) for target identification, advances in x-ray crystallography for 3D target-ligand interactions, and exploitation of computational approaches for designing and developing (17, 18) novel small molecules should accelerate the discovery of new antibacterial agents with the desired broad spectrum efficacy and the low potential for resistance development. This chapter discusses the application of computational structure-based drug design methods in the discovery of novel small molecule inhibitors against well-validated targets involved in bacterial DNA replication (19, 20) and cell division; the type IIA topoisomerases, gyrase B 4 (GyrB) and topoisomerase IV (ParE). The focus will be on specific ligand-protein interactions observed in publically available crystallographic structures of novobiocin and/or adenylylimidodiphosphate (ADPNP) bound to the E. coli GyrB and ParE subunits as the initial guide for inhibitor discovery and optimization. 2. The design of antibacterial agents that inhibit the function of type IIA topoisomerases -GyrB and/or ParE Background There are many classes of antibiotics that target different aspects of bacterial function. Two important classes of antibiotics, including the coumarins (e.g. novobiocin) and fluoroquinolones (e.g. ciprofloxacin), target proteins involved in DNA replication, specifically the type IIA family of topoisomerases, DNA gyrase and topoisomerase IV (19, 20). These enzymes are structurally homologous proteins and exist as heterotetramers in vivo, comprising 2 GyrA and 2 GyrB subunits for DNA gyrase and 2 ParC and 2 ParE subunits for topoisomerase IV. Type II topoisomerases possess both DNA cleavage and ligation functionalities, and an ATP-dependent clamp that work in a concerted fashion to resolve DNA catenates and supercoils during replication. The GyrA and ParC subunits are catalyze DNA cleavage and re-ligation via an intermediate where the 5'-ends of the DNA chain are covalently bound to conserved tyrosine residues. The GyrB and ParE subunits are responsible for passage of a separate DNA strand through the cut DNA strands in an ATP-dependent manner. At the cleavage site, DNA becomes single stranded due to a four base-pair staggered break within the binding cavity, allowing the fluoroquinolone inhibitors to bind to the DNA-topoisomerase complex. The drug-DNA-enzyme 5 ternary complex locks the tetrameric complex in a catalytically non-competent state (21), arresting DNA replication. The accumulation of single-and double-stranded breaks irreversibly damage the bacterial chromosome (22, 23). Although dual-targeting in principle, fluoroquinolones appear to kill via single targets in different bacterial classes. In E. coli, a gram-negative bacteria, fluoroquinolones kill mainly via inhibition of DNA gyrase, while in gram-positive organisms like S. aureus, fluoroquinolones appear to operate primarily via inhibition Topoisomerase IV (24). Coumarin antibiotics, on the other hand, which target the ATP-binding domains of the GyrB and ParE subunits, are poor dual targeting agents that operate mainly via the inhibition of GyrB. However, the high degree of similarity at the sequence and structural level between GyrB and ParE suggests that the design of potent dual targeting agents against the GyrB and ParE domains should be possible (25, 26). Structural features of the type IIA topoisomerases A number of X-ray structures (27) of the ATPase domain of DNA-gyrase complexed with phosphoaminophosphonic acid-adenylate ester (ADPNP) and novobiocin are available ( Figure 1 ). The detailed analysis of these structures demonstrates that the ligands bind similarly to a key aspartic acid side-chain and a conserved water molecule in the binding cavity. In both cases, each ligand donates a hydrogen bond to the aspartate side chain and accepts a hydrogen bond from the conserved water molecule. Figure 2 illustrates the hydrogen-bond network observed for the adenine of the ADPNP and the carbamate moiety of novobiocin. Novobiocin is an ATPcompetitive inhibitor that overlaps partially with the adenine moiety of ATP (27). This overlapping competitive binding area is the focus of novel inhibitor design. [Insert Figures 1 & 2 here] 6 Recently, Bellon et. al from Vertex (27) reported the structure of the ATP-binding domain of E. coli ParE complexed with adenylyl-imidodiphosphate (ADPNP) at 2.0 Å resolution as well as a ParE novobiocin complex at 2.1 Å resolution (Figure 3) . A comparison of the GyrB and ParE structures demonstrate strikingly similarity within the ATP-binding sites. Superposition of the E. coli ParE and GyrB structures shows the amino acids in the active site overlap closely. The ParE and the corresponding E. coli GyrB residues (in parenthesis) are as follows when ADPNP is bound ( Figure 4 ): Y5