Designing Type II Topoisomerase Inhibitors: a Molecular Modeling Approach

Home , Aminocoumarin, Clinafloxacin, Clorobiocin, Coumermycin A1, Novobiocin

Send Orders for Reprints to [email protected] 40 Current Topics in Medicinal Chemistry, 2014, 14, 40-50 Designing Type II Topoisomerase Inhibitors: A Molecular Modeling Approach

Juan J. Perez*, Cecylia S. Lupala and Patricia Gomez-Gutierrez

Department of Chemical Engineering. Universitat Politecnica de Catalunya, ETSEIB. Av. Diagonal, 647; 08028 Barcelona, Spain

Abstract: Nosocomial infections are produced by pathogens with the ability to persist in hospital environments and with the propensity to develop resistance to diverse antimicrobials. In order to tackle resistance, it has been pointed as good strategy to select resilient drug targets that are evolutionally constrained to design drugs less susceptible to develop resistance. Molecular modeling can help to fulfill this goal by providing a rationalization of the observed resistance at the molecular level and, suggesting modifications on existing drugs or in the design of new ones to overcome the problem. The present report focus on type II topoisomerases, a clinical validated target for antibacterials and describe diverse modes of intervention including, inhibition of their ATPase function, stabilization of the cleavage complex or prevention of DNA strand hydrolysis. Moreover, the origin of resistance is also rationalized on the base of ligand-target interactions. Finally, efforts are described to circumvent the effect of non-susceptible strains by the design of new drugs based on existing ones, like the case of diones that act through the same mechanism as quinolones or the newly released quinole-carbonitrile derivatives that inhibit type II topoisomerases through a new mechanism. Keywords: Topoisomerases inhibitors, antibiotics design, nosocomial infections, quinolones.

INTRODUCTION cephalosporins prevent the formation of the bacterial wall, while others like polymixins disrupt the structure of the Nosocomial infections affect about 5-10% of hospitalized membrane by intercalation [5]. Finally, other compounds patients and are among the major causes of death and in- inhibit protein synthesis by binding to the ribosome, like creased morbidity in this population. Furthermore, these in- macrolides that prevent the transfer of the peptidyl-tRNA fections hamper the success of advanced surgical procedures, complex from the A to the P site [6], lincosamides that cause including organ and prosthetic transplants. In addition to a premature separation of peptidyl-tRNA complex from the their morbidity and mortality burdens, nosocomial infections ribosomes [7] or tetracyclines preventing the attachment of are associated to higher healthcare costs due to the increased new amino acids to the nascent polypeptide chain [8]. length of stay in hospitals for infected patients. These infections are caused by pathogens -commonly bacteria- with the Antibacterial resistance is defined as the point at which ability to persist in hospital environments and with propen- the administration of the drug can no longer safely treat the sity to develop resistance to diverse antimicrobials. Accord- infection due to an induced change in the drug target or an ingly, there is a great deal of interest to establish new proto- inability of the drug to reach the target. Moreover, microor- cols that include stricter measures in infection control and ganisms may also exhibit resistance towards more than one antibiotic use, as well as to find new antibiotics [1]. As it drug, in this case, depending on the number of the antibacte- will be discussed later, molecular modeling can help in un- rial classes towards they show resistance, strains are referred derstanding the causes of antibacterial resistance at the mo- to exhibit multidrug, extended drug, and pandrug resistance lecular level and to direct efforts to design new compounds. [9]. Antibacterial resistance is nothing new, microorganisms have survived for thousands of years due to their ability to Antibacterials are drugs that arrest bacterial growth or adapt against these compounds. Drug resistance is the result kill them. This goal can be achieved acting at different levels of the evolutionary pressure on bacteria and is typically ac- on the microorganism. Some drugs target key enzymes. For quired by gene mutation or lateral gene transfer within or example, rifamycins and lipiarmycins inhibit prokaryotic between species. Genetic diversity within populations com- RNA-polymerase function [2]; quinolones kill bacteria by bined with rapid microbial generation time gives microbes a increasing levels of DNA strand breaks generated by type II remarkable adaptability in response to selective pressure topoisomerases [3]; or sulphamides kill bacteria by inhibit- from antimicrobials. ing dihydropteroate synthetase, an enzyme involved in folate synthesis [4]. Other group of drugs, like penicillins and The observed antibiotic resistance experienced nowadays is a consequence of a complex interaction among natural selection, environment, and patterns of drug use and misuse *Address correspondence to this author at the Department of Chemical [10]. Thus, considering that most of the antibacterials are Engineering. Universitat Politecnica de Catalunya, ETSEIB. Av. Diagonal, natural or semi-synthetic compounds and, the robust antibac- 647; 08028 Barcelona, Spain; Tel: +34934016679; Fax: +34934017150; Email: [email protected] terial machinery of these microorganisms together with their

1873-5294/14 $58.00+.00 © 2014 Bentham Science Publishers Designing Type II Topoisomerase Inhibitors Current Topics in Medicinal Chemistry, 2014, Vol. 14, No. 1 41 rapid generation time, it is understandable the ability of these ligands, putatively binders of the same target, permits to microorganisms to escape new antibacterials after treatment identify the stereospecific features of the ligand-receptor for a short period of time. For example, it took only four interaction, that are described through a pharmacophore that years after penicillin was commercialized in the early 1940s, can be used to design new compounds [19]. Finally, a broad to identify penicillin-resistant strains of S. aureus [11]. In a class of studies can be classified as quantitative structure- recent study taking advantage of the advances in sequencing, activity studies, where the activity of a series of compounds the time evolution of the genome of a S. aureus multidrug is related with structural or physicochemical parameters of a resistance strain was tracked from the blood samples of a set of ligands [20, 21]. patient undergoing chemotherapy with vancomycin and The present contribution is aimed at reviewing available other antibiotics. Interestingly, it was found that from the structure-activity studies on the inhibition of type II topoiso- sequence of the first vancomycin susceptible isolate to the imerases, to exemplify how structural information helps to last vancomycin non-susceptible isolate, 35 point mutations understand how specific mutations are responsible for resis- could be identified in three months of treatment [12]. tance and how new ligands can be designed to avoid these In regard to the mechanisms bacteria use to raise resis- limitations [17]. tance to antibacterial drugs, the most direct mechanism pro- ceeds via modification of the binding site features in the tar- BACTERIAL TYPE IIA TOPOISOMERASES get macromolecule. In fact, slight modifications of the bind- DNA is organized in large loops both in eukaryotic and ing site can reduce dramatically the effectiveness of a drug, prokaryotic cells, which essentially renders it a closed circu- being numerous the reports describing point mutations in a lar system. This structure is mechanically constrained, and therapeutic target that provoke resistance to a group of anti- imposes that DNA adopts diverse topological isoforms, in- bacterials [13]. In addition, there are other resistance mecha- cluding supercoiled configurations (positively or negatively) nisms that use as strategy the reduction of antibacterial up- as compared to the relaxed reference form, knots or even take. One of them consists of enzymatic antibacterial degra- catenanes. Interestingly, DNA topology plays an active role dation. Indeed, many bacteria contain specific versions of - in regulating basic biological processes, including control of lactamases, capable to hydrolyze antibiotics like penicillins, gene transcription, DNA replication and segregation, ge- cephalosporins, monobactams or carbapenems [14]. Bacte- nome maintenance and cellular differentiation [22]. Accord- rial resistance may also rise by adaptation of these enzymes ingly, control of DNA topology is essential for cellular func- to a specific antibacterial after a period of treatment, turning tion. This control is exercised in all organisms by a family of the strain non-susceptible to the drug. Other mechanisms enzymes called topoisomerases, found throughout all cellular concern drug transport systems in and out the cell. Thus, domains of life [23]. Topoisomerases alter DNA topology by bacteria may raise resistance through drug influx reduction repeated cycles of DNA strand breaking and religation at the by alteration of porin expression [15]. Moreover, resistance incredible frequency of 250-6000 cycles per minute, depend- may also be raised through efflux enhancement. Efflux is a ing of enzyme type. Although the double-stranded DNA mechanism to pump out unwanted toxic substances out of breaks generated by topoisomerases are essential for cell the cell. Active efflux is an active transport mechanism thus, viability, it is a dangerous process in which an aberrant op- it requires binding of the drugs to a transporter, resistance eration can damage chromosomal integrity. Due to their im- may rise by adaptation of the transporter to a specific drug portant role in cell viability, topoisomerases have been target after a period of treatment [16]. for pharmaceutical intervention for designing antibacterial and antitumor agents for the treatment of cancer in humans MOLECULAR MODELING [24, 25]. Molecular modeling can be used to design ligands Topoisomerases fall into two categories: type I that pro- against resistant bacteria, through the analysis of the stereo- duce DNA single strand breaks and, type II that produce chemical modifications caused by mutations appeared in the double strand breaks [26]. The process catalyzed by the for- binding pocket of specific targets. In addition, considering mer does not require energy input, whereas the latter requires other mechanisms of resistance, molecules can be designed ATP hydrolysis. Every organism needs at least one type I in such a way they are not substrates of bacterial hydrolyses and one type II enzymes to fully deal with maintaining ap- or that can escape the active efflux mechanism. propriate levels of DNA supercoiling and removing chromo- Molecular modeling comprise a diverse set of computa- somal entanglements. Type I topoisomerases can be divided tional techniques aimed at designing new drugs. The choice further into three subclasses [23]: type IA, that share many of one of them to tackle a specific problem, depends on the structural and mechanistic features with the type II topoi- structural information available of the target. In the case the somerases, like topoisomerase I and III in bacteria or 3 and three dimensional structure of the macromolecule-ligand 3 in humans; type IB, found in eukaryotic cells and in some complex is known, the key features of the ligand- viruses and are characterized for utilizing a controlled rotary macromolecule interaction, as well as the effect of mutations mechanism; and type IC, bifunctional enzymes that carry out on ligand binding can be easily assessed [17]. In some cases, both topoisomerase and DNA repair activities within the even if this information is not available, a model of the three same protein, being topoisomerase V the only member de- dimensional structure of the target can be constructed by scribed so far. On the other hand, type II topoisomerases can homology modeling, using another protein with a high se- be divided into two subclasses.Type IIA is the most widely quence identity whose structure is known as template [18]. distributed class and it is found throughout eukaryotes, bac- Moreover, in the case the three dimensional model of the teria, and some archaea. The other class is referred as type target cannot be constructed, the comparison of a set of IIB, like topoisomerase VI that is restricted to archaea and 42 Current Topics in Medicinal Chemistry, 2014, Vo l. 14, No. 1 Perez et al. higher plants. Members of the same subfamily are structur- Indeed, the latter is responsible for the differential function- ally and mechanistically similar, whereas those of different ality between gyrase and topoisomerase IV. The A2B2 dimers subfamilies are distinct. In the present report we focus on exhibit three major subunit interfaces or gates between the type IIA topoisomerase inhibitors as antibacterial agents. two AB/CE subunits that play an important role for its func- Bacteria express two distinct type IIA topoisomerases, tion, including the ATP gate formed by the GHKL domains; the DNA gate formed by the TOPRIM domains; and the C known as DNA gyrase and topoisomerase IV, although a few gate formed between coiled-coil domains (see Fig. 1) [31]. species like Mycobacterium tuberculosis express only gyrase [26]. Gyrase is involved primarily in regulating the super- Type II topoisomerases bind to DNA and form a complex helical density of chromosomal DNA and alleviating tor- in which around 120 base pairs wrap around the protein core. sional stress that accumulates ahead of DNA tracking sys- In a further step, the enzyme transiently cleaves a pair of tems. In contrast, topoisomerase IV efficiently decatenates opposing phosphodiester bonds situated four base pairs DNA and is the major enzyme that is responsible for unknot- apart, generating a topoisomarase-DNA cleavage complex. ting and untangling the bacterial genome. They are large Passage of a second DNA segment through this enzyme- proteins consisting of two units, encoded by two different bridged DNA gate and its resealing complete the topological gens and, organized in the form of a tetramer A2B2. In gyrase change of the DNA [32]. In order to fulfill its function, the these units are the 97 kDa protein GyrA and the 90 kDa pro- enzyme has two active sites: one involved in ATP hydroly- tein GyrB, whereas in the case of topoisomerase IV, they are sis, located at each of the GHKL domains, and another one the proteins ParC and ParE, with 75 kDa and 70 kDa mo- involved in DNA cleavage and ligation, consisting of a large lecular mass, respectively. gorge on the upper part of the heart shaped topoisomerase In spite of size differences and low sequence identity, and defined by the TOPRIM domain of GyrB/ParE and the both enzymes share a common quaternary structure [27]. The winged helix domain, the tower and the coiled-coil domains structure can be described as a heart-shaped protein with a of GyrA/ParCof each of the two AB/CE subunits. Structural large central hole, exhibiting a two-fold symmetry axis de- and kinetic studies suggest that binding of ATP to fined between the two GyrA-GyrB units in gyrase or ParC- GyrB/ParE promotes dimerization and gate closure. This ParE units topoisomerase IV. GyrB and ParE subunits con- structural change in turn, allows entry and capture of a dou- sist of three domains: a GHKL domain (Gyrase, Hsp90, ble stranded DNA molecule, called the transported-segment Histidine Kinase, MutL), found in several ATP-binding pro- or T-segment. Concomitantly, on the site where DNA cleav- teins [28];a second domain that acts as transducer; and, a age and ligation are produced, another 20 base pair long third domain, called TOPRIM (topoisomerase-primase), DNA duplex called the gate-segment or G-segment, is ac- found in a wide variety of enzymes involved in nucleic acid commodated and subjected to a strand cut. Breakage of DNA manipulation [29]. On the other hand, GyrA and ParC is produced by means of topoisomerases double functionality subunits consist of four domains: a winged helix domain as nucleases and ligases by transphosphorilation. Actually, (WHD), a widespread nucleic-acid-binding protein structural the conserved tyrosine residue Tyr122 in GyrA and Tyr120 in element [30]; a small domain called the tower; a third long ParC (E.Coli numbering, that will be used throughout this domain classified as a coiled-coil and a variable C-terminus. report), participates in a nucleophilic attack on the 5’ side of

Fig. (1). Three dimensional structure of Saccharomyces cerevisiae DNA gyrase with the G-segment of DNA bound on it: a) front view; b) front view rotated 90º in regard to a); c) view from the top. Designing Type II Topoisomerase Inhibitors Current Topics in Medicinal Chemistry, 2014, Vol. 14, No. 1 43 a nucleotide phosphate, generating a transient covalent bond of bacterial DNA gyrase [38], exhibiting lower performance between the protein and the DNA. These two processes re- in topoisomerase IV inhibition [39]. However, these com- quire the presence of a divalent ion, preferably Mg2+, al- pounds have not been successful as drugs due to their low though Mn2+, Ca2+ or Co2+ also work [33], presumably to activity against Gram-negative bacteria, toxicity in eukaryo- promote the leaving of the ribose 3’-OH. Moreover, in a step tes and low solubility in water [40]. forward the symmetric tyrosine executes the same move- ment, producing a transient covalent bond and break on the other DNA strand, four base pairs apart. The whole process generates a transient DNA gate through which the T-segment goes through. Following T- segment transport, DNA strand breaks are resealed, and the T-segment released. So, we can picture the enzyme as an ATP-modulated clamp with two sets of jaws at opposite ends, connected by multiple joints. The enzyme with a bound DNA segment can admit a second DNA duplex through one set of jaws, transport it through the cleaved first duplex, and expel it through the other set of jaws [34].

INHIBITION OF BACTERIAL TYPE II TOPOI- SOMERASES Type II topoisomerases are good targets to develop antibacterial agents for several reasons: (1) they are essential in all bacteria for replication and cell division; (2) an accumula- tion of cleavage complexes makes cells not viable and eventually destroyed (bactericidal effect); (3) targeting bacterial type II topoisomerases is not poisonous for human enzymes; and (4) due to the high degree of homology between gyrase and Topo IV, type II topoisomerase inhibitors tend to target both enzymes. Fig. (2). Chemical structures of classical aminocoumarins: novo- Taking into account that the enzyme exhibits two active biocin (1), clorobiocin (2) and coumermycin A1 (3). sites, two different mechanisms can be thought to pursuit inhibition: prevention of ATP binding and alternatively, pre- Classical aminocoumarins bind to the gyrB subunit of vention of substrate binding to the cleavage and relegation DNA gyrase competing with ATP, acting pharmacologically site. The former approach has already been exploited and as an ATP competitive inhibitors [41]. The analysis of the involves the design of competitive inhibitors of ATP, called crystallographic structure of the complex gyrB subunit- catalytic inhibitors. Compounds like novobiocin use this novobiocin shows that although the ligand does not bind in mechanism. The latter approach has not been used yet. It the same spot as ATP, the binding mode of the two mole- would involve discovering molecules that interfere binding cules exhibit some overlap with common interactions with of DNA to the gorge, which is not necessarily simple. Alter- diverse residues of the binding pocket. Fig. 3 shows the ATP natively, two approaches have been used. One consist of binding pocket of the E.Coli DNA gyrase with novobiocin stabilizing the covalent enzyme–DNA cleavage complex, and the ATP inhibitor adenylyl---imidodiphosphate preventing DNA relegation and inducing cell death [35]. (ADPNP) bound. As it can be seen, the ligands show partial Quinolones are a family of antibacterials that use this overlap. Actually, both molecules form a hydrogen bond mechanism to execute their action. The other, consist of pre- with residues Ala47 and Asp73: in the case of novobiocin this venting the enzyme to hydrolyze DNA. This mechanism is is mediated through the 3'-carbamoyl substituent of the sugar used by a class of recently designed antibacterials, like moiety and in the case of ADPNP through its adenine amine. GSK299423 [36]. In addition, novobiocin also forms hydrogen bonds with the backbone carbonyl of Gly77 and with the side chain of Thr165 CATALYTIC INHIBITORS mediated through a water molecule. Moreover, the 2'- hydroxyl of novobiocin forms a hydrogen bond to the car- One of the best characterized catalytic inhibitors of type 46 II topoisomerase are the classical aminocoumarins. They bonyl oxygen of Asn , and the 4'-methoxyl oxygen forms a hydrogen bond to the side chain of Asn46. Finally, the cou- represent a subset of the aminocoumarins, a heterogeneous 136 group of natural antibacterials isolated from different Strep- marin ring forms two hydrogen bonds, both with Arg . In tomyces strains [37] and characterized for having a 3-amino- addition, there are a number of hydrophobic contacts involv- 4, 7-dihydroxycoumarin scaffold. Classical aminocoumarins ing the 5',-5'-dimethyl group of the sugar or the 4'-methoxy include: novobiocin (1), clorobiocin (2) and coumermycin methyl group. Further analysis of the structure suggests that A1 (3) shown in Fig. 2. The antibacterial activity of these the 3'-iso-pentenyl-4'-hydroxybenzoate group of novobiocin compounds was discovered in the early 1950s, demonstrat- is not important for DNA gyrase interactions, although it is ing them to be effective agents against Gram-positive bacte- thought to influence uptake of the compound into the bacteria. Later it was demonstrated that they are potent inhibitors rial cells. Since the overlap between the aminocoumarins and 44 Current Topics in Medicinal Chemistry, 2014, Vo l. 14, No. 1 Perez et al.

ATP is partial, resistance arises from mutations on key residues involved in the interaction including, Ile78, Ile94, Met95 dues that are involved in the interaction between the ligand and Val120 [41]. and the receptor and that do not affect ATP binding. Interest- A number of flavonoids have been reported to exhibit an- ingly, these mutations may act by destroying a binding inter- tibacterial activity. However, only few have been demon- action or by adding steric hindrance. For example, mutation strated to inhibit DNA gyrase [45]. These include catechins of Arg136 or Gly164 (to Valine) provokes resistance without (6) in Fig. 4, extracted from the green tea leafs and, the fla- affecting significantly topoisomerase function [24]. vonolquercetin (7) in Fig. 4, present in many fruits and vege- tables [46, 47]. Their activity is exercised by binding to the gyrB subunit, interfering the ATPase activity of the enzyme, consequently being competitive inhibitors of ATP. Modeling studies suggest that these compounds bind in a different way in regard to aminocoumarins and cyclothialidines, but there is not enough experimental evidence to substantiate this hypothesis [46, 47]. In order to discover new antibiotics, there have been published numerous molecular modeling studies describing analogs synthesized based on the structure-activity information available on natural catalytic inhibitors. For example, an in silico screening of a data base biased for compounds with a resorcinol moiety, based on the features of GR122222X bound to the gyrB subunit, as well as similar mono- or poly- substituted phenol analogues was carried out. The study Fig. (3). Superimposition of the bound conformations of novobio- identified a few thiazole analogs, like compound (8) in Fig. cin (yellow) and the ATP inhibitor adenylyl---imidodiphosphate 4, as promising catalytic inhibitors of DNA gyrase [48]. Pyr- (ADPNP) (cyan) inside the ATP binding site. rolamides is another group of catalytic inhibitors developed using fragment-based NMR screening [49]. From a lead Aminocoumarins exhibit a differential binding to DNA compound, and using the techniques of medicinal chemistry, gyrase and topoisomerase IV. Being in general, poorer in- analog (9), shown in Fig. 4 was identified. The compound hibitors of the latter, although it depends on the molecule. was shown to exhibit good inhibition profile against different Whereas clorobiocin still retains reasonable potency against Gram-positive pathogens, including Enterococcus faecium topoisomerase IV, novobiocin exhibits a significant reduced and Streptococcus spp. More recently, using a combined potency [42]. The differential behavior can be explained at approach involving in silico screening and fragment-base the molecular level to be due to a specific residue. Compari- crystallographic screening, potent pyrrolopyrimidine inhibitors were identified as catalytic inhibitors of gyrase and son of the bound conformation of novobiocin to the two re- topoisomerase IV, exhibiting a broad spectrum of activity ceptors permits to identify a few structural differences. against Gram-negative bacteria [50]. These were tested by means of site directed mutagenesis concluding that the differential profile observed is due to residue Ile78 in gyrB that is a Met74 in parE that produces a STABILIZERS OF THE ENZYME–DNA CLEAVAGE COMPLEX differential interaction [43]. Cyclothialidines isolated from Streptomyces spp, consti- Quinolones are the prototype of binders to the transient tute another group of natural catalytic inhibitors of type II type II topoisomerase-DNA cleavage complex, preventing topoisomerases [44]. Cyclothialidine (4) in Fig. 4, contains a DNA relegation. They represent a large group of molecules unique 12-membered lactone ring with an integrated pen- with a 4-oxo-1,4-dihydroquinoline skeleton (10). Like many tapeptide chain (Ser-Hyp-Ser-Met-Ala) attached to a resorci- other drugs, quinolones have been used as antibacterials for nol moiety. Cyclothialidine potently inhibits DNA gyrases decades and it has only recently been shown how these from several bacterial species in vitro, including E. Coli and molecules interact with their receptors at the atomic level. Staphylococcus aureus, with a high degree of selectivity. The first active compound of this group was identified after Despite its excellent in vitro activity cyclothialidines possess screening a subproduct obtained in the production of the no antibacterial activity in vivo, due to its insufficient pene- antimalarial chloroquine [51]. This compound served as a tration of the bacterial cell wall. GR122222X (5) in Fig. 4, is basis to synthesize new analogs following a traditional me- an analog of cyclothialidine where the first amino acid of the dicinal chemistry approach, guided by a pharmacophoric peptide chain is alanine instead of serine. Analysis of the hypothesis raised from the known structure-activity data crystallographic structure of gyrB-GR122222X complex [52]. In a parallel effort, the pharmacological target of qui- reveals that, like in the case of novobiocin, both molecules nolones was identified about ten years later the release of the overlap partially in the ATP binding pocket. In this case, first quinolone antibacterial [53] and, only very recently, the most of the interactions occur through the resorcinol ring. crystallographic structure of quinolone-DNA-topoisomerase Key residues involved in hydrogen bonding interactions with II complex has been published [36, 54-56]. All this informa- the ligand either direct or water mediated are Asn46, Ala47, tion now permits to rationalize all the efforts carried out dur- Asp73 and Thr165. There are also several hydrophobic resi- ing years in this area and eventually provides ground knowl- edge to design new antibacterial drugs [57]. Designing Type II Topoisomerase Inhibitors Current Topics in Medicinal Chemistry, 2014, Vol. 14, No. 1 45

Fig. (4). Chemical structures of diverse catalytic inhibitors: Cyclothialidine (4); GR122222X (5); catechin (6); quercetin (7);thiazole analog (8); pyrrolamide analog (9).

In contrast to most antimicrobial agents which were first Structure-activity studies of the more than 10,000 ana- found as products of bacteria or yeasts, quinolones are syn- logs synthesized have provided evidence of the effect of dif- thetic products. Nalidixic acid (11), in Fig. 5, a naphthyri- ferent substitutions on the 4-quinolone scaffold [59]. So, it is dine derivative with a moderate antibacterial activity and known that position 2 does not tolerate bulky substituents narrow Gram-negative spectrum, can be considered the first due to steric hindrance. Moreover, small substituents at posi- quinolone analog used as antibacterial drug. The compound tion 5 such as an amino, hydroxyl, or methyl groups can was introduced in the early 1960s for the treatment of un- markedly increase in vitro activity against Gram-positive complicated urinary tract infections caused by enteric bacte- bacteria as well as enhance potency. Substitution of a fluo- ria and it still in the market. Nalidixic acid is the result of an rine atom at position 6 controls bacterial potency and a bulky optimizing process of a lead that showed weak antibacterial group at position 7 controls potency spectrum and pharma- profile that had been identified as an impurity produced in cokinetics. The heterocycle scaffold can also be modified, the chemical manufacture of a batch of the antimalarial agent preserving activity. Thus, the carbon in position 8 can be chloroquine [51]. However, the real success of quinolones as replaced by a nitrogen giving rise to napthylridines, like in antibacterial agents came a decade later with the develop- nalixic acid. Other successful scaffolds used include quina- ment of fluoroquinolones. Actually, a substitution of a zolines (15), quinazolinediones (16) or isothiazoquinolones piperazine ring at position 7 and the addition of a fluorine (17), shown in Fig. 5. atom at position 6, resulted in an increase of potency, low As mentioned above, it took about ten years after the re- toxicological profile and improved pharmacokinetics com- lease of nalidixic acid to identify gyrase as the therapeutic pared to first generation quinolones, giving rise to a second target of quinolones [53]. However, for years it was not clear generation of quinolones. Ciprofloxacin (12) and levoflox- if the ligands bound directly to DNA or to gyrase. After acin (13) in Fig. 5, are model compounds of this series which some speculation about quinolone action mechanism [60], is still in widespread clinical use today. These are antibacte- fifty years later their mechanism of action begun to be eluci- rials of broad spectrum both Gram-negative and Gram- dated thanks to the resolution of the crystal structure of sev- positive bacteria. Ciprofloxacin is one of the most potent eral target-quinolone complexes [36, 54-56]. Actually, qui- fluoroquinolones, being effective against susceptible strains nolones stabilize the transient cleavage complex produced on of Pseudomonas aeruginosa and Actinobacter baumannii. the G-segment, inhibiting DNA strand resealing. Superimpo- Despite its wide spectrum of activity, they exhibit moderate sition of the crystallographic structures of the cleaved topoi- activity against pneumococcus. This burden was solved a somerase IV-DNA-quinolone complexes with levofloxacin few years after with the addition of a cyclopropyl group at (14) (pdb entry 3RAE) [54], clinafloxacin(15) (pdb entry position 1 together with substitution of methoxy or chlorine 3RAD) [54], and moxifloxacinn (13) (pdb entry 2XKK) [55] moieties at position 8, given rise to a third generation of qui- together with the gyrase-DNA-ciprofloxacin (12), (pdb entry nolones. These compounds also called respiratory qui- 2XCT) [36], shows that the ligand sits in the same position nolones improve the activity against Gram-positive cocci as can be seen in Fig. 6. The fused rings sit intercalated be- and the activity against anaerobes, whilst retaining the activ- tween the nucleotides of the DNA strand stabilized by means ity against Gram-negative pathogens. A model compound of of stacking interactions. However, as previously shown [17], this generation is moxifloxacin (14), shown in Fig. 5 [58]. although this interaction is strong enough to guarantee 46 Current Topics in Medicinal Chemistry, 2014, Vo l. 14, No. 1 Perez et al.

Fig. (5). Chemical structure of diverse quinolones: quinolone skeleton (10); nalidixic acid (11); ciprofloxacin (12); levofloaxacin (13); moxifloxacin (14); quinazoline skeleton (15); quinazolinedione skeleton (16); isothiazoquinolone skeleton (17).

Fig. (6). Superimposition of the bound conformations of levofloxacin (yellow), clinafloxacin (magenta) and moxifloxacin (blue) in topoisomerase IV together with ciprofloxacin (green) bound to DNA gyrase. (The color version of the figure is available in the electronic copy of the article). binding, additional interactions are required to orient the erably magnesium, has been shown to be necessary for molecules properly in the binding pocket. These are pro- topoisomerase function and furthermore for quinolone bind- vided with diverse residues of gyrase or topoisomerase IV. ing. On the one hand, a catalytic magnesium that is required Thus, analysis of the crystallographic structures permits to for the transphosphorilation reaction [61] and on the other, a identify key structural elements that help to explain the bind- non-catalytic magnesium that is necessary to stabilize the ing of quinolones to the transient cleavage complex. One of quinolone-topoisomerase-DNA complex [62]. Interestingly, them are residues Ser79 and Asp83 located on the WHD do- two of these water molecules act as a bridges to interact with main of the parC subunit of topoisomerase IV or the corre- Ser79 and Asp83 in topoisomerase IV. Thus, these residues sponding Ser83 and Glu87 in gyrA subunit of DNA gyrase. can be considered to be anchoring points of quinolones and The other is an octahedrally coordinated Mg2+ ion bound to in fact, as discussed below, mutations of these two residues the C3/C4 keto-acid moiety of quinolones and to four addi- are responsible of most of the cases of quinolone resistance. tional water molecules. The presence of divalent ions, pref- Another important anchoring point is provided by Arg121 Designing Type II Topoisomerase Inhibitors Current Topics in Medicinal Chemistry, 2014, Vol. 14, No. 1 47 located in ParC that interacts with the hydroxyl group in C3. the atomic level. For example, in the case of the diones the On the opposite region of the ligand, the fluorine atom in hydrogen interaction with Arg117 is mediated through the position 6 interacts with polar atoms of DNA bases. Close to carbonyl moiety in position 2. Moreover, unlike quinolones, it, the bulky moiety attached to position 7 projects into a PD0305970 does not require Mg2+ for binding [64], which is large solvent-accessible volume. This region in topoi- consistent with the amine in position 2 that prevents its coor- somerase IV contains several polar interconnected residues, dination binding. Moreover, residues Ser79 and Asp83 are far including Asp435, Glu474, Glu475 and Arg456 that create an from the quinazolinedione and are not involved in ligand environment adequate to host a polar/charged moiety. Qui- binding. Finally, the substituent in position 7 has an amine nolones are expected to act on gyrase and topoisomerase IV group that interacts with Glu474 and Arg456. The differential with similar activity, since the key residues involved in structural binding requirements of PD0305970 compared to ligand binding are conserved. Moreover, they are conserved fluroquinolones explains the activity of this class of com- throughout diverse species. pounds against quinolone resistant strains [65]. Moreover, it also explains that dione resistance strains exhibit mutations Early after the beginning of use of fluoroquinolones the in Glu474 and Arg456 [64]. species Staphylococcus aureus and Pseudomonas aeruginosa showed resistance, however from the mid-1990s quinolone resistance started to increase in almost all Gram- INHIBITORS OF DNA CLEAVAGE positive and Gram-negative species [63]. Resistance mecha- These compounds bind to a site close to that of the qui- nism corresponds basically to mutations on the binding site, nolones, but are not associated with the stabilization of the 79 83 specifically, mutations of residues Ser and Asp in parE of cleaved complex [36, 66, 67]. Among them, GSK299423 topoisomerase IV or the corresponding residues of gyrA in (18), a quinoline-carbonitrile found by unbiased antibacterial 83 87 DNA gyrase: Ser and Glu that cause the loss of the coor- screening, or the related compound NXL101 (19), shown in dination bridge between the quinolone and the enzyme [62]. Fig. 8. GSK299423 is a potent antibacterial agent with broad It has recently been suggested that resistance could be spectrum of Gram-negative and Gram-positive pathogens, tackled by selecting resilient drug targets that are evolution- and active against quinolone-resistant strains. This com- ally constrained and the development of robust drugs that are pound stabilizes a pre-cleavage enzyme-DNA complex, in- less susceptible to the development of resistance. Accord- hibiting strand cleavage. Analysis of the crystallographic ingly, taking into account that type II topoisomerases are structure of the GSK299423-gyrase-DNA complex (pdb en- clinical validated targets, one strategy to fight resistance is to try 2XCS) permits to understand the binding mode of this design new structures that bind to the same target. In this compound. The compound binds midway of the two active direction, structures with alternative skeletons, like quina- sites, with the quinolone-carbonitrile group intercalated be- zolines (15), quinazolinediones (16) or isothiazoquinolones tween two base pairs and with the oxathiolo-pyridine group (17), shown in Fig. 5, have been described to bind in the sitting on a large hydrophobic pocket on top of the DNA- same way as quinolones. As a proof of concept the crystallo- gate [36]. graphic structure of the cleaved complex of a dione (PD0305970)-DNA-topoisomerase IV gives support to this MOLECULAR MODELING AND NOSOCOMIAL IN- idea. Fig. 7 shows the superimposition of this compound FECTIONS with clinafloxacin (pdb entry 3RAD) [54]. As can be seen Acinetobacter baumannii, a Gram-negative bacteria, is the modification from a 3-carboxylic acid and a 4-ketone to a an increasingly important nosocomial pathogen due to the 2,4-dione, does not alter the binding mode of the ligand to increasing number of strains that become resistant to the the target protein, however there are differences in binding at

Fig. (7). Superimposition of the crystal structures of ciprofloxacin (green) bound to DNA gyrase and PD0305970 (pink) bound to topoisomerase IV. (The color version of the figure is available in the electronic copy of the article). 48 Current Topics in Medicinal Chemistry, 2014, Vo l. 14, No. 1 Perez et al.

Fig. (8). Chemical structures of GSK299423 (18), NXL101 (19), DS-8587 (20) and PD0305970 (21). most common clinically utilized antibacterials including qui- work by means of the first mechanism include the natural nolones, aminoglycosides, broad-spectrum cephalosporins, inhibitors aminocoumarins, cyclothialidines and some fla- carbapenems, tigecycline and colistin, which are the last re- vonoids. In addition, different families of compounds have sort antibiotics [2]. Resistance to quinolones comes from been recently described, including thiazoles, pyrrolamides or specific mutations on Ser86 (Ser83 in E.Coli) in the gyrA pyrrolopyrimidines, among others. Compounds in the second subunit of DNA gyrase and Ser80 (Ser79 in E.Coli) in the group include quinolones and compounds with similar scaf- parC subunit of topoisomerase IV. In addition, efflux- folds, like quinazolines, quinazolinediones or isothiazoqui- mediated resistance to quinolones has also been described. nolones. Finally, within the third group of compounds quinolone-carbonitriles, pyrazoles or quinolines have described. In the quest for new antibacterials to fight resistant strains, a fine tune of the affinity of the ligand can be a good A detailed study of the molecular interactions that ex- strategy to design new antibacterials, since with this process plain the available structure-activity relationships, permits to the safety range of use is also improved. In this direction, find new directions for improvement of the compounds and DS-8587 (20) in Fig. 8, was designed. It represents a novel furthermore, explain the rationale of resistance and how it broad-spectrum quinolone with an order of magnitude higher can be avoided. Molecular modeling studies also permits to affinity for type II topoisomerases than fluoroquinolones understand the effect of resistant-strain mutations, and to [68]. From the chemical point of view, the compound keeps foresee ways to improve presently available compounds or the scaffold of quinolones, and as novelty it incorporates even design new ones, necessary to tackle with resistance moieties inspired on PD0305970 (21) in Fig. 8, on positions strains. 474 456 7 and 8 which provide access to residues Glu and Arg . In summary, we have seen how a combined approach of In addition, the compound exhibits an improved per- molecular modeling, medicinal chemistry and a robust bioas- formance against the efflux-mediated mechanism of resis- say permits to develop new drugs directed to a specific target tance. As result, the compound is active against A.baumannii and to rationalize the effect of specific mutations on the po- resistant strains and a promising antibacterial agent to treat tency of available compounds. infections caused by this pathogen. CONFLICT OF INTEREST CONCLUSION The author(s) confirm that this article content has no con- Fighting resistant bacteria require the use of new drugs. flicts of interest. Design of these compounds, either from modification of the ACKNOWLEDGEMENTS existing ones or from the novo, can highly benefit from molecular modeling studies. In the present report we have fo- CSL wants to acknowledge partial support from the Uni- cused on type II topoisomerase, a clinical validated target versitat Politecnica de Catalunya. and analyzed possible mechanisms of inhibition. Basically, inhibition can be performed using three different mecha- REFERENCES nisms: preventing ATP hydrolysis, stabilizing the cleavage [1] Davies, J.; Davies, D. Origins and Evolution of Antibiotic Resis- complex and, preventing DNA cleavage. Compounds that tance. Microbiol. Mol. Biol. Rev., 2010, 10, 417-433 Designing Type II Topoisomerase Inhibitors Current Topics in Medicinal Chemistry, 2014, Vol. 14, No. 1 49

[2] Chopra, I. Bacterial RNA polymerase: a promising target for the [27] Sissi, C.;Palumbo, M. In front of and behind the replication fork: discovery of new antimicrobial agents. Curr. Opin, Investig. Drugs, bacterial type IIA Topoisomerases. Cell. Mol. Life Sci., 2010, 67, 2007, 8, 600-607 2001-2024 [3] Heeb, S.; Fletcher, M. P.; Chhabra, S. R.; Diggle, S. P.; Williams, [28] Dutta, R.; Inouye, M. GHKL, an emergent ATPase/kinase super- P.; Camara, M. Quinolones: from antibiotics to autoinducers. family. Trends Biochem. Sci., 2000, 25, 24-28 FEMS Microbiol. Rev., 2011, 35, 247-274 [29] Aravind, L.; Leipe, D.D.; Koonin, E.V. Toprim--a conserved cata- [4] Sköld, O. Sulfonamide resistance: mechanisms and trends. Drug lytic domain in type IA and II topoisomerases, DNAG-type pri- Resist. Updat., 2000, 3, 155-160 mases, OLD family nucleases and RecR proteins. Nucleic Acids [5] Dixon, R.A.; Chopra, I. Polymyxin B and polymyxin B nonapep- Res., 1998, 26, 4205-4213 tide alter cytoplasmic membrane permeability in Escherichia coli. [30] Harami, G. M.; Gyimesi, M.; Kovács,M. From keys to bulldozers: J. Antimicrob. Chemother., 1986, 18, 557-563. expanding roles for winged helix domains in nucleic-acid-binding pro- [6] Tenson, T.; Mankin, A. Antibiotics and the ribosome. Mol. Micro- teins. Trends Biochem. Sci., 2013, 38, 364-371 biol., 2006, 59, 1664-1677 [31] Schmidt, B. H.; Osheroff, N.; Berger, J. M. Structure of a topoi- [7] Noli, C.; Boothe, D. Microlides and lincosamides. Veter. Derma- somerase II-DNA-nucleotide complex reveals a new control tol., 1999, 10, 217-223 mechanism for ATPase activity. Nat. Struct. Mol. Biol., 2012, 19, [8] Chopra, I.; Roberts, M. Tetracycline Antibiotics: Mode of Action, 1147-1154 Applications, Molecular Biology, and Epidemiology of Bacterial [32] Schoeffler, A. J.; Berger, J. M. DNA topoisomerases : harnessing Resistance. Microbiol. Mol. Biol. Rev., 2001, 65, 232-260 and constraining energy to govern chromosome topology. Quart. [9] J. Vilaand J. Pachon. Therapeutic options for Acinetobacter bau- Rev. Biophys., 2008, 41, 41-101 mannii infections: an update. Expert Opin. Pharmacother., 2012, [33] Deweeseab, J. E.; Osheroff, N. The use of divalent metal ions by 13, 2319-2336 type II topoisomerases. Metallomics, 2010, 2, 450-459 [10] Davies, J.; Davies, D. Origins and Evolution of Antibiotic Resis- [34] Roca. J.; Berger, J.M.; Harrison, S.C.; Wang, J.C. DNA transport tance. Microbiol. Mol. Biol. Rev., 2010, 74, 417-433 by a type II topoisomerase: direct evidence for a two-gate mecha- [11] Rammelkamp, C.H.; Maxon, T. Resistance of Staphylococcus nism. Proc. Natl. Acad. Sci. USA, 1996, 93, 4057-4062 aureus to the Action of Penicillin. Proc. Soc. Exp. Biol. Med., [35] Drlica, K.; Malik, M.; Kerns, R. J.; Zhao, X. Quinolone-Mediated 1942, 51, 386-389 Bacterial Death. Antimicrob. Agents Chemother. , 2008, 52, 385- [12] Mwangi, M.M.; Wu, S. W.; Zhou, Y.; Sieradzki, K.; Lencastre, H. 392 de; Richardson, P.; Bruce, D.; Rubin, E.; Myers, E.; Siggia, E. D.; [36] Bax,B.D.; Chan,P.F.; Eggleston,D.S.; Fosberry,A.; Gentry,D.R.; Tomasz, A. Tracking the in vivo evolution of multidrug resistance Gorrec, F.; Giordano, I.; Hann, M.M. Hennessy, A.; Hibbs,M.; in Staphylococcus aureus by whole-genome sequencing. Proc. Huang, J.; Jones, E.; Jones, J.; Brown, K. K.; Lewis, C. J.; May, E. Natl. Acad. Sci. USA, 2007, 104, 9451-9456 W.; Saunders, M. R.; Singh, O.; Spitzfaden, C. E.; Shen, C.; Shil- [13] Davies, J.; Davies, D. Origins and Evolution of Antibiotic Resis- lings, A.; Theobald, A. J.; Wohlkonig, A.; Pearson, N. D.; Gwynn, tance. Microbiol. Mol. Biol. Rev., 2010 , 74, 417-433 M. N. Type IIA topoisomerase inhibition by a new class of antibac- [14] Biondi, S.; Long, S.; Panunzio, M.; Qin, W. L. Current Trends in terial agents. Nature, 2010, 466, 935-940. beta-Lactam Based beta-Lactamases Inhibitors. Curr. Med. Chem., [37] Oblak, M.; Kotnik, M.; Solmajet, T. Discovery and development of 2011, 18, 4223-4236 ATPase inhibitors of DNA gyrase as antibacterial agents. Curr [15] Pagès, J.-M.; James, C. E.; Winterhalter, M. The porin and the Med. Chem., 2007, 14, 2033 permeating antibiotic: a selective diffusion barrier in Gram- [38] Gellert, M.; O'Dea, M. H.; Itoh, T.; Tomizawa. J.-I. Novobiocin negative bacteria. Nat. Rev. Microbiol., 2008, 6, 893-903 and coumermycin inhibit DNA supercoiling catalyzed by DNA gy- [16] Coyne, S. Courvalin, P.; Perichon, B. Efflux-Mediated Antibiotic rase. Proc. Natl. Acad. Sci. USA, 1976, 73, 4474-4478 Resistance in Acinetobacter spp. Antimicrob. Agents Chemother., [39] Alt, S.; Mitchenall, L. A.; Maxwell, A.; Heide, L. Inhibition of 2011, 55, 947-953 DNA gyrase and DNA topoisomerase IV of Staphylococcus aureus [17] Lupala, C.L.; Gomez-Gutierrez, P.; Perez, J.J. Molecular determi- and Escherichia coli by aminocoumarin antibiotics. J. Antimicrob. nants of the bacterial resistance to fluoroquinolones: A computa- Chemother., 2011, 66, 2061-2069 tional study. Curr. Comp.-Aided Drug Des., 2013, 9, 281-288 [40] Heide, L. The aminocoumarins: biosynthesis and biology. Nat. [18] Yadav, P. K.; Singh, G.; Gautam, B.; Singh, S.; Yadav, M.; Srivas- Prod. Rep., 2009, 26, 1241-1250 tav, U.; Singh, B. Molecular modeling, dynamics studies and vir- [41] Lewis, R. J.; Singh, O. M. P.; Smith, C. V.; Skarzynski, T.; Max- tual screening of Fructose 1, 6 biphosphate aldolase-II in commu- well, A.; Wonacott, A. J.; Wigley, D. B. The nature of inhibition of nity acquired- methicillin resistant Staphylococcus aureus (CA- DNA gyrase by the coumarins and the cyclothialidines revealed by MRSA). Bioinformation, 2013, 9, 158-164. X-ray crystallography. EMBO J., 1996, 15, 1412-1420 [19] Gampe, C.M.; Tsukamoto, H.; Doud, E. H.; Walker, S.; Kahne, D. [42] Flatman, R. H.; Eustaquio, A.; Li, S.-M.; Heide, L.; Maxwell, A. Tuning the moenomycin pharmacophore to enable discovery of Structure-activity relationships of aminocoumarin-type gyrase and bacterial cell wall synthesis inhibitors. J. Am. Chem. Soc., 2013, topoisomerase IV inhibitors obtained by combinatorial biosynthe- 135, 37763779 sis. Antimicrob. Agents Chemother., 2006, 50, 1136-1142 [20] Prado-Prado, F.J.; Gonzalez-Diaz, H.; Santana, L.; Uriarte, E. [43] Bellon, S.; Parsons, J. D.; Wei, Y.; Hayakawa, K.; Swenson, L. L.; Unified QSAR approach to antimicrobials. Part 2: predicting activ- Charifson, P. S.; Lippke, J. A.; Aldape, R.; Gross, C. H. Crystal ity against more than 90 different species in order to halt antibacte- structures of Escherichia coli topoisomerase IV ParE subunit (24 rial resistance. Bioorg. Med. Chem., 2007, 15, 897-902 and 43 kilodaltons): a single residue dictates differences in novo- [21] Speck-Planche, A.; Kleandrova, V.V.; Cordeiro, M.N.D.S. biocin potency against topoisomerase IV and DNA gyrase. Antimi- Chemioinformatics for rational discovery of safe antibacterial crob. Agents Chemother., 2004, 48, 1856-1864 drugs: Simultaneous predictions of biological activity against strep- [44] Nakada, N.; Shimada, H.; Hirata, T.; Aoki, Y.; Kamiyama, T.; tococci and toxicological profiles in laboratory animals. Biorg. Watanabe, J.; Arisawa, M. Biological characterization of cyclothi- Med. Chem., 2013, 21, 2727-2732 alidine, a new DNA Gyrase Inhibitor. Antimicrob. Agents Che- [22] Liu, Z.; Deibler, R. W.; Chan, H. S.; Zechiedrich, L. The why and mother., 1993, 37, 2656-2661 how of DNA unlinking. Nucleic Acids Res., 2009, 37, 661-671 [45] Cushnie, T. P. T.; Lamb, A. J. Antimicrobial activity of flavonoids. [23] Nitiss, J. L. DNA topoisomerase II and its growing repertoire of Int. J. Antimicrob. Ag., 2005, 26 343-356 biological functions. Nat Rev Cancer. , 2009, 9, 327-337 [46] Plaper A, Golob M, Hafner I, Oblak M, Solmajer T, Jerala R. [24] Collin, F.; Karkare, S.; Maxwell, A. Exploiting bacterial DNA Characterization of quercetin binding site on DNA gyrase. Bio- gyrase as a drug target: current state and perspectives. Appl. Micro- chem. Biophys. Res. Commun., 2003, 306, 530-536 biol. Biotechnol., 2011, 92, 479-497 [47] Gradisar, H.; Pristovsek, P.; Plaper, A.; Jerala, R. Green tea cate- [25] Kathiravan, M. K., Khilare, M. M., Nikoomanesh, K., Chothe, A. chins inhibit bacterial DNA gyrase by interaction with its ATP S.; Jain, K.S. Topoisomerase as target for antibacterial and antican- binding site. J. Med. Chem., 2007, 50, 264-271 cer drug discovery. J. Enzym. Inhibit. Med. Chem., 2013, 28, 419- [48] Brvar, M.; Perdih, A.; Oblak, M.; Masic, L. P.; Solmajer, T. In 435 silico discovery of 2-amino-4-(2,4-dihydroxyphenyl)thiazoles as [26] Ghilarov,D. A.; Shkundina, I. S. DNA Topoisomerases and Their novel inhibitors of DNA gyrase B. Bioorg. Med. Chem. Lett., 2010, Functions in a Cell. Mol.Biol., 2012, 46, 47-57. 20, 958-962 50 Current Topics in Medicinal Chemistry, 2014, Vo l. 14, No. 1 Perez et al.

[49] Sherer, B. A.; Hull, K.; Green, O.; Basarab, G.; Hauck, S.; Hill, P.; [59] 59. Domagala, J.M. Structure-activity and structure-side-effect Loch III, J. T.; Mullen, G.; Bist, S.; Bryant, J.; Boriack-Sjodin, A.; relationships for the quinolone antibacterials. J. Antimicrob. Che- Read, J.; DeGrace, N.; Uria-Nickelsen, M.; Illingworth, R. N.; mother., 1994, 34, 851-851 Eakin, A. E. Pyrrolamide DNA gyrase inhibitors: Optimization of [60] Madurga, S.; Sanchez-Cespedes, J.; Belda, I.; Vila, J.; Giralt, E. antibacterial activity and efficacy. Bioorg. Med. Chem. Lett., 2011, Mechanism of binding of fluoroquinolones to the quinolone resis- 21, 7416-7420 tance-determining region of DNA gyrase: towards an understand- [50] Tari, L. W.; Trzoss, M.; Bensen, D. C.; Li, X.; Chen, Z.; Lam, T.; ing of the molecular basis of quinolone resistance. Chembiochem., Zhang, J.; Creighton, C. J.; Cunningham, M. L.; Kwan, B.; Stid- 2008, 9, 2081-2086. ham, M.; Shaw, K. J.; Lightstone, F. C.; Wong, S. E.; Nguyen, T. [61] Pitts, S. L.; Liou, G. F.; Mitchenall, L. A.; Burgin, A. B.; Maxwell, B.; Nix, J.; Finn, J. Pyrrolopyrimidine inhibitors of DNA gyrase B A.; Neuman, K. C.; Osheroff, N. Use of divalent metal ions in the (GyrB) and topoisomerase IV (ParE). Part I: Structure guided dis- DNA cleavage reaction of topoisomerase IV. Nucleic Acids Res., covery and optimization of dual targeting agents with potent, 2011, 39, 4808-4817 broad-spectrum enzymatic activity. Bioorg. Med. Chem. Lett., [62] Aldred, K. J.; McPherson, S. A.; TurnboughJr, C. L.; Kerns, R. J.; 2013, 23, 1529-1536 Osheroff, N. Topoisomerase IV-quinolone interactions are medi- [51] Mitscher, L. A. Bacterial Topoisomerase Inhibitors: Quinolone and ated through a water-metal ion bridge: mechanistic basis of qui- Pyridone Antibacterial. Chem. Rev., 2005, 105, 559-592 nolone resistance. Nucleic Acids Res., 2013, 41, 4628-4639 [52] Emmerson, A. M. ; Jones, A. M.. The quinolones: decades of de- [63] Dalhoff,A. Resistance surveillance studies: a multifaceted prob- velopment and use. J. Antimicrob. Chemother., 2003, 51, S1, 13-20 lem—the fluoroquinolone example. Infection, 2012, 40, 239-262 [53] Sugino, A.; Peeblest, C. l.; Kreuzer, K. N.; Cozzarelli, N.R. [64] Xiao-Su Pan, Katherine A. Gould, and L. Mark Fisher. Probing the Mechanism of action of nalidixic acid: Purification of Escherichia differential iInteractions of quinazolinedione PD 0305970 and qui- coli nalA gene product and its relationship to DNA gyrase and a nolones with gyrase and topoisomerase IV. Antimicrob. Agents novel nicking-closing enzyme. Proc. Natl. Acad. Sci. USA, 1977, Chemother., 2009, 53, 3822-3831 74, 4767-4771 [65] Aldred, K.J.; McPherson, S.A.; Wang, P.; Kerns, R.J.; Graves, [54] Laponogov, I.; Sohi, M. K.; Veselkov, D. A.; Pan, X.-S.; Sawhney, D.E.; Turnbough, Jr, C.L.; Osheroff, N. Drug interactions with Ba- R.; Thompson, A. W.; McAuley, K. E.; Fisher, L. M.; Sanderson, cillus anthracis topoisomerase IV: biochemical basis for quinolone M. R. Structural insight into the quinolone-DNA cleavage complex action and resistance. Biochemistry, 2012, 51, 370-381 of type IIA topoisomerases. Nat. Struct. Mol. Biol., 2009, 16, 667- [66] Gomez,L.; Hack, M. D.; Wu, J.; Wiener, J. J. M.; Venkatesan, H.; 669 Santillan, Jr., A.; Pippel, D. J.; Mani, N.; Morrow, B. J.; Motley, S. [55] Laponogov, I., Pan, X.S., Veselkov, D.A., McAuley, K.E., Fisher, T.; Shaw, K. J.; Wolin, R.; Grice, C. A.; Jones, T.K. Novel pyra- L.M. and Sanderson, M.R. Structural basis of gate-DNA breakage zole derivatives as potent inhibitors of type II topoisomerases. Part and resealing by type II topoisomerases. PLoS One, 2010, 5, 1: Synthesis and preliminary SAR analysis. Bioorg. Med. Chem. e11338. Lett., 2007, 17, 2723-2727 [56] Wohlkonig, A., Chan, P.F., Fosberry, A.P., Homes, P., Huang, J., [67] Black, M. T.; Stachyra, T.; Platel, D.; Girard, A.-M.; Claudon, M.; Kranz, M., Leydon, V.R., Miles, T.J., Pearson, N.D., Perera, R.L.; Bruneau, J.-M.; Miossec, C. Mechanism of Action of the Antibiotic Shillings, A. J.; Gwynn, M. N.; Bax, B. D. Structural basis of qui- NXL101, a novel nonfluoroquinolone Inhibitor of bacterial Type II nolone inhibition of type IIA topoisomerases and target-mediated Topoisomerases. Antimicrob. Agents Chemother., 2008, 52, 3339- resistance. Nat. Struct. Mol. Biol., 2010, 17, 1152-1153 3349 [57] Wiles, J. A.; Bradbury, B. J.; Pucci, M. J. New quinolone antibiot- [68] Higuchi, S.; Onodera,Y.; Chiba, M.; Hoshino, K.; Gotoh, N. Potent ics: a survey of the literature from 2005 to 2010. Expert Opin. in vitro antibacterial activity of DS-8587, a novel broad-spectrum Ther. Patents, 2010, 20.1295-1319 quinolone, against Acinetobacterbaumannii. Antimicrob. Agents [58] Andersson, M.I.; MacGowan, A.P. Development of the quinolones. Chemother., 2013, 57, 1978-1981 J. Antimicrob. Chemother., 2003, 51, Suppl. S1, 1-11

Received: July 22, 2013 Revised: August 17, 2013 Accepted: September 01, 2013