- Email: [email protected]
New aminocoumarin antibiotics as gyrase inhibitors
International Journal of Medical Microbiology 304 (2014) 31–36
Contents lists available at ScienceDirect
International Journal of Medical Microbiology journal homepage: www.elsevier.com/locate/ijmm
New aminocoumarin antibiotics as gyrase inhibitors Lutz Heide ∗ Pharmaceutical Institute, Eberhard Karls-Universität Tübingen, Auf der Morgenstelle 8, 72076 Tübingen, Germany
a r t i c l e
i n f o
Keywords: Aminocoumarins Novobiocin Simocyclinone Gyrase Staphylococcus aureus
a b s t r a c t The aminocoumarins novobiocin, clorobiocin and coumermycin A1 are structurally related antibiotics produced by different Streptomyces strains. They are potent inhibitors of bacterial gyrase. Their binding sites and their mode of action differ from those of fluoroquinolones such as ciprofloxacin. Novobiocin has been introduced into clinical use against Staphylococcus aureus infections, and S. aureus gyrase is particularly sensitive to inhibition by aminocoumarins, while topoisomerase IV is much less sensitive. Modern genetic techniques have allowed the engineering of the producer strains, resulting in a diverse range of new aminocoumarins, including compounds which are more active than the natural antibiotics as well as a compound which is actively imported across the cell envelope of Gram-negative bacteria. A further group of aminocoumarins are the simocyclinones which bind simultaneously to two different sites of gyrase and show a completely new mode of inhibition. Both the simocyclinones and the “classical” aminocoumarins strongly inhibit the fluoroquinolone-induced activation of RecA and thereby the SOS response in S. aureus. Therefore, a combination of aminocoumarins and fluoroquinolones strongly reduced the risk of resistance development and may offer new prospects in anti-infective therapy. © 2013 Elsevier GmbH. All rights reserved.
Introduction DNA gyrase belongs to the best-validated targets in antibacterial drug therapy (Chopra et al., 2012; Collin et al., 2011; Pommier et al., 2010; Sanyal and Doig, 2012). Gyrase was discovered in E. coli in 1976 (Gellert et al., 1976), but the two most important classes of gyrase inhibitors were already discovered before that date. The quinolone nalidixic acid was found accidentally as a side product of the synthesis of the antimalarial chloroquine (Lesher et al., 1962). Based on nalidixic acid, the fluoroquinolones were developed and became one of the principal clinical weapons in the fight against bacterial infections. However, their usefulness is now endangered by the rapid emergence of resistance. The only other class of gyrase inhibitors which has been introduced into clinical use are the aminocoumarins. Novobiocin (Fig. 1) was discovered in the 1950s in screening programs for antibiotics produced by microorganisms. It was introduced into human anti-infective therapy in 1964 under the name Albamycin® by the Upjohn company and used for the treatment of Staphylococcus aureus infections, including multiresistant MRSA strains. Novobiocin is also active against Borrelia burgdorferi, the causative agent of Lyme disease (Samuels and Garon, 1993). Shortly after novobiocin, the even more potent compounds clorobiocin and coumermycin A1 were discovered. These three
∗ Tel.: +49 7071 2972460; fax: +49 7071 295250. E-mail address: [email protected] 1438-4221/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.ijmm.2013.08.013
“classical” aminocoumarins show structural similarity (Fig. 1). An aminocoumarin moiety is linked via an amide bond to an aromatic acid, and via a glycosidic bond to the unusual deoxysugar L-noviose, which carries an acyl moiety attached to its 3-hydroxy group. This acyl moiety, the deoxysugar and the aminocoumarin moiety form the principal site of interaction with gyrase. Beyond these three “classical” compounds (and close analogs thereof), only two other natural aminocoumarins have been discovered: rubradirin which lacks the deoxysugar moiety at the 7-OH group of the aminocoumarin and which therefore is not a gyrase inhibitor (Kim et al., 2008), and the simocyclinones which will be discussed below. All these compounds are produced by soil bacteria of the genus Streptomyces. Coumermycin has also been found in strains of the genus Actinoallomurus, distantly related to Streptomyces (Pozzi et al., 2011). Further, a new isolate of a Streptomyces strain producing novobiocin and analogs thereof has been described (Cheenpracha et al., 2010). In biochemical assays, aminocoumarins inhibit DNA gyrase with IC50 values which are lower than that of modern fluoroquinolones (Alt et al., 2011b). However, the advantage of a higher target affinity is offset by a more attractive mechanism of action exerted by the fluoroquinolones. While the aminocoumarins act as competitive inhibitors of gyrase, the fluoroquinolones act as “poisons” of this enzyme. They stabilize the covalent gyrase-DNA complex, thereby leading to protein-stabilized breaks of DNA and ultimately to cell death, even at a relatively low occupancy of the inhibitor (Collin et al., 2011). Therefore, minimal inhibitory concentrations
32
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Fig. 1. (A) Structures of the three “classical” aminocoumarin antibiotics novobiocin, clorobiocin and coumermycin A1 . (B) Structure of simocyclinone D8.
(MICs) of the fluoroquinolones against pathogenic bacteria are in a similar range as those of aminocoumarin antibiotics (Schröder et al., 2012). Despite the fact that fluoroquinolones are usually referred to a “gyrase inhibitors”, their primary target in many pathogenic bacteria is in fact not gyrase but topoisomerase IV (= topo IV). Topo IV is a similar heterotetrameric enzyme as gyrase. E. coli gyrase consists of two GyrA and two GyrB subunits, and topo IV of two ParC and two Par E subunits. Gyrase and topo IV share 40% sequence identity. Both belong to the type II topoisomerases which catalyze changes in the topology of DNA involving the transient break of both strands of DNA. Both gyrase and topo IV can relax supercoiled DNA, but gyrase is unique by its capability to introduce negative supercoils. The energy for this reaction is derived from the hydrolysis of ATP, catalyzed by the GyrB subunit. In Gram-negative bacteria, usually gyrase is the primary target of the fluoroquinolones, while in Gram-positive bacteria often topo IV is the primary target (Collin et al., 2011). Also aminocoumarins can interact both with gyrase and with topo IV, but gyrase is the primary target (see below). Detailed structures of the ternary complexes of fluoroquinolones with gyrase and DNA have been published only recently (Laponogov et al., 2010; Wohlkonig et al., 2010). In contrast, the structural basis of the interaction of aminocoumarins with gyrase has been extremely well characterized from several high resolution crystal structures over the last 15 years (Collin et al., 2011; Lewis et al., 1996; Maxwell and Lawson, 2003). The binding site of the aminocoumarins (especially of their substituted deoxysugar moiety) overlaps with the binding site of ATP on the GyrB subunit, and this explains the competitive inhibition of gyrase-catalyzed ATP hydrolysis by aminocoumarins.
Gyrase belongs to the GHKL family of enzymes, named after its founding members gyrase, heat-shock protein 90 (Hsp90), certain protein kinases and the DNA mismatch repair protein MutL (Dutta and Inouye, 2000). Proteins of this family share a common type of ATP-binding fold, and therefore the aminocoumarins interact not only with bacterial gyrase but also with eukaryotic Hsp90 which is an emerging target in anticancer therapy (Hong et al., 2012). A number of novobiocin analogs have been synthesized and tested as Hsp90 inhibitors (Burlison et al., 2006; Donnelly and Blagg, 2008). Following the discovery and the functional analysis of the biosynthetic gene clusters of the three classical aminocoumarins (Pojer et al., 2002; Steffensky et al., 2000; Wang et al., 2000), it has become possible to generate a multitude of new aminocoumarin analogs by modern genetic techniques, such as combinatorial biosynthesis, mutasynthesis and synthetic biology (Heide, 2009b). This, together with the ever-increasing problem of antibacterial drug resistance, has renewed the interest in aminocoumarins as antibacterial agents. Measurement of gyrase and topoisomerase IV inhibition by aminocoumarin antibiotics Determination of the inhibitory activity of aminocoumarins and fluoroquinolones on gyrase and topo IV by biochemical assays in vitro is not trivial, and the IC50 values reported in the literature vary considerably according to the methods used. Different assay types are available, e.g. supercoiling, relaxation, DNA cleavage and decatenation assays (Morgan-Linnell et al., 2007; Pan and Fisher, 1999). The supercoiling assay is considered most suitable for the assessment of the inhibition of gyrase by aminocoumarins, and the decatenation assay for the topo IV inhibition. The former
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
assay measures the conversion of relaxed to supercoiled plasmid DNA and the latter the conversion of concatenated to decatenated kinetoplast DNA. Since the topoisomerases from E. coli are the best studied enzymes, most literature data on gyrase and topo IV inhibition have been obtained with these enzymes. However, Staphylococcus aureus is the most important pathogen addressed by aminocoumarin therapy, and the sensitivity of S. aureus topoisomerases to aminocoumarins is somewhat different from that of the E. coli enzymes. In vitro investigation of S. aureus gyrase is complicated by the fact that this enzyme requires high concentrations of potassium glutamate (K-Glu) for its activity (e.g. 700 mM) (Alt et al., 2011b). In contrast, E. coli gyrase can be, and usually is, assayed in the absence of K-Glu. However, also this enzyme shows higher activity in the presence of K-Glu. A recent study (Alt et al., 2011b) showed that the presence of K-Glu increased the sensitivity of E. coli gyrase toward aminocoumarin antibiotics approximately 10-fold, and in case of one specific compound even 150-fold. Therefore, the assay conditions are a very important factor when IC50 values of aminocoumarins for gyrase and topo IV inhibition are compared. Table 1 shows the inhibitory activity of different aminocoumarin antibiotics on gyrase and topo IV from E. coli and S. aureus, determined using identical K-Glu concentrations for the enzymes from both organisms (Alt et al., 2011b). These data confirm the excellent activity of aminocoumarins on S. aureus gyrase: 50% inhibition is achieved in the low nanomolar range. The S. aureus gyrase is more sensitive to most aminocoumarins than the E. coli enzyme. In both organisms, the sensitivity of topo IV is lower by several orders of magnitude than that of gyrase, clearly establishing gyrase as the primary target of the aminocoumarins. Despite these very clear in vitro results, the biological significance of topo IV inhibition by aminocoumarins may not be completely dismissed: cultivation of S. aureus in the presence of novobiocin results not only in the selection of mutants with altered gyrase, but as a second step of resistance development also in mutations of topo IV (Fujimoto-Nakamura et al., 2005). Furthermore, the biosynthetic gene clusters of the most potent natural aminocoumarins, clorobiocin and coumermycin, contain not only a gene coding for an aminocoumarin-resistant gyrase B subunit for self-resistance of the producer strain, but also a gene coding for a resistant ParE subunit of topo IV (Schmutz et al., 2004; Schmutz et al., 2003). This indicates that the toxic effect of aminocoumarins on topo IV is a relevant problem for the producer strain. New aminocoumarin antibiotics from combinatorial biosynthesis and mutasynthesis Since the discovery of the sulfonamides and the penicillins, there have been two principal routes for the discovery of new anti-infectives: chemical synthesis and screening of natural products (mostly produced by microorganisms). Most of the antibiotics which are presently in clinical use are natural products or derivatives thereof, with the penicillins as the best-known example. In contrast, only few classes of synthetic compounds have achieved an important role in antibacterial therapy, most prominently the sulfonamides and the fluoroquinolones. However, the discovery of new antibiotics by screening of microbial natural products has become increasingly difficult, as many microbial strains which are easily accessible and which can be cultivated under laboratory conditions have already been investigated. With the advent of modern genetic and genomic techniques, however, new and promising routes have been opened for antibiotic drug discovery from microbial strains. These routes include e.g. combinatorial biosynthesis, mutasynthesis and synthetic biology, which have been very successfully exploited for the generation of new aminocoumarins.
33
Fig. 2. Structures of selected aminocoumarin antibiotics obtained from combinatorial biosynthesis, mutasynthesis and synthetic biology.
Soil bacteria and fungi which produce antibiotics possess specific genes for the biosynthesis of these compounds. Usually, all genes required for the biosynthesis of a given antibiotic are grouped together in one contiguous stretch of DNA, the so-called biosynthetic gene cluster. E.g. the biosynthetic gene clusters of the three “classical” aminocoumarin antibiotics novobiocin, clorobiocin and coumermycin A1 span between 23 and 38 kb and comprise between 20 and 33 genes, including biosynthetic, regulatory and resistance genes and genes for transporters (Heide, 2009a). The function of nearly all these has been experimentally identified, partly revealing completely new enzymes and enzyme classes which were undiscovered previously (Bonitz et al., 2011; Heide, 2009a). Knowledge of the function of biosynthetic genes readily allows the modification of these genes and gene clusters in order to generate modified antibiotic structures. E.g. the chlorine atom of clorobiocin (Fig. 1) is introduced by the halogenase enzyme Clohal. Inactivation of the clo-hal gene in the producer strain led to the accumulation of a clorobiocin analog lacking the chlorine atom (novclobiocin 101, Fig. 2) (Eustáquio et al., 2003). Novobiocin contains not chlorine but a methyl group in the corresponding position of the aminocoumarin moiety (Fig. 1), and this methyl group is attached under catalysis of the methyl transferase NovO. Expression of the gene novO in the above mentioned mutant of the clorobiocin producer led to the unnatural “hybrid” antibiotic novclobiocin 102 (Fig. 2), carrying the methyl group typical of novobiocin in an otherwise unmodified skeleton of clorobiocin. Such experiments have allowed the investigation of structurefunction relationships in the class of aminocoumarin antibiotics. As shown in Table 1, removal of the chlorine group from the clorobiocin structure (resulting in novclobiocin 101) clearly reduces the activity. Activity is largely restored by introduction of the methyl group in this position (resulting in novclobiocin 102). Removal of the acyl group attached to the 3-OH group of the deoxysugar of clorobiocin (resulting in novclobiocin 103) strongly reduces the activity, especially against S. aureus gyrase. Many more modifications of the aminocoumarins have been generated in similar experiments, defining the structural elements of aminocoumarin antibiotics which are required for optimal interaction with the target (Alt et al., 2011b; Flatman et al., 2006), and the aminocoumarins
34
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Table 1 Activity of selected aminocoumarin antibiotics against DNA gyrase and topoisomerase IV. Gyrase activity was determined in a supercoiling assay in the presence of 700 mM potassium glutamate (K-Glu), and topo IV activity in a decatenation assay in the presence of 100 mM K-Glu (Alt et al., 2011a; Alt et al., 2011b). IC50 [nM] against DNA gyrase
novobiocin clorobiocin coumermycin A1 simocyclinone D8 novclobiocin 101 novclobiocin 102 novclobiocin 217 novclobiocin 225 novclobiocin 401 novclobiocin 103
IC50 [nM] against topoisomerase IV
from E. coli
from S. aureus
from E. coli
from S. aureus
80 30 30 100 300 30 6 6 30 100
10 6 6 2000 50 10 1 1 6 1000
10 000 3000 5000 >10 000 3000 300 8000 8000 >50 000 >10 000
20 000 10 000 100 000 17 000 35 000 5000 >50 000 >50 000 35 000 >50 000
have provided an example of the power of such “combinatorial biosynthesis” experiments for the generation of new compounds (Heide, 2009c). Yet, even larger structural diversity can be generated when the power of genetic recombination experiments is combined with the possibilities of organic synthesis. This is e.g. achieved in mutasynthesis experiments, in which the biosynthesis of a genuine structural moiety of the antibiotic is blocked by an appropriate gene inactivation. Subsequently, analogs of this structural moiety are chemically synthesized and fed to the mutant strain. These analogs can then be incorporated into the antibiotic, leading to new variants of the drug. This technique has been further refined (Anderle et al., 2007), leading e.g. to novclobiocins 217 and 225 (Fig. 2). As shown in Table 1, these compounds are more potent inhibitors of gyrase than the most potent natural aminocoumarins. They both show IC50 values of 1 nM against S. aureus gyrase, and this translates into excellent antibacterial activity against S. aureus in vitro, with MIC values < 0.06 g/ml (Anderle et al., 2008). Therefore, even though the affinity of natural aminocoumarins for their target has been optimized over millions of years in evolution, this affinity can still be improved by suitable structural modifications.
cluster for clorobiocin biosynthesis, deficient in a gene required for the biosynthesis of the genuine substituted benzoic acid moiety of clorobiocin, was introduced into this strain. This synthetic biology approach resulted in the efficient formation of the desired catecholsubstituted aminocoumarin antibiotic novclobiocin 401 (Fig. 2) (Alt et al., 2011a). The structural modification did not reduce the activity of this compound against gyrase, compared to the genuine antibiotic clorobiocin (Table 1). However, it clearly increased the activity against an E. coli test strain, as shown by the larger inhibition zone in a disk diffusion assay (Fig. 3). In contrast, an E. coli test strain deficient in tonB (required for energizing the catechol transporters) was much less inhibited by novclobiocin 401 (Fig. 3), indicating that this compound was actively imported across the Gram-negative cell envelope by the TonB-dependent catechol siderophore transporters. Novclobiocin 401 provides an example for the power of modern synthetic biology techniques for the generation of new antibiotics, and for the possibility to improve the activity of aminocoumarin antibiotics against Gram-negative organisms by suitable structural modification.
A new aminocoumarin antibiotic is actively imported across the cell envelope of Gram-negative bacteria A principal shortcoming of the aminocoumarin antibiotics is their poor activity against Gram-negative bacteria. One important reason for this is their poor penetration across the outer membrane of the Gram-negative cell envelope. However, the outer membrane contains transporters for the active import of certain compounds, e.g. of so-called catechol siderophores, compounds which complex iron through their ortho-diphenol (= “catechol”) structures and thereby facilitate uptake of this vital mineral ion into the bacterial cell. The recently generated aminocoumarin antibiotic novclobiocin 401 (Fig. 2) contains the catechol structural moiety 3,4-dihydroxybenzoic acid (3,4-DHBA). Since it mimics the structure of the catechol siderophores, it is actively imported across the bacterial cell envelope (Alt et al., 2011a). In contrast to the successful mutasynthesis experiments described above, novclobiocin 401 could not be obtained by feeding of 3,4-DHBA to an appropriate mutant strain, due to quick degradation of this benzoic acid derivative during external feeding. However, 3,4-DHBA was generated successfully in the cell by a synthetic biology strategy. An artificial gene operon was chemically synthesized, encoding a twostep pathway from chorismic acid (an intermediate of aromatic amino acid biosynthesis) to 3,4-DHBA. This pathway combines two enzymes of completely unrelated organisms and does not exist in nature. The two genes were synthesized with an optimized sequence (i.e. optimized codon usage) for efficient translation in a Streptomyces host strain, and with a strong constitutive promoter for efficient transcription. This construct was introduced into a specific, genetically engineered host strain. In addition, a modified gene
Simocyclinone D8, a bifunctional aminocoumarin antibiotic simultaneously binding to two different sites of gyrase Simocyclinone D8 (Fig. 1) is an aminocoumarin antibiotic discovered by Fiedler and coworkers (Schimana et al., 2000). It is produced, together with several structural analogs, by a Streptomyces strain. Like the classical aminocoumarins, it contains a 3-amino-4,7-dihydroxy-coumarin moiety, but it lacks the substituted deoxysugar moiety bound to the 7-hydroxy group. Therefore, it cannot bind to the typical aminocoumarin binding site on the gyrase B subunit, and it does not inhibit gyrase B-catalyzed ATP hydrolysis. Unexpectedly, simocyclinone D8 was nevertheless found to be a very efficient inhibitor of gyrase (Flatman et al., 2005). X-ray crystallographic studies (Edwards et al., 2009) as well as mass spectrometric studies (Edwards et al., 2011) showed that simocyclinone D8 simultaneously binds to two different site of the gyrase A subunit. One binding pocket interacts with the aminocoumarin moiety, and the other binding pocket with the angucyclinone moiety of the antibiotic. Both pockets lie in the predicted DNA binding saddle of gyrase, and therefore occupancy of either site by the inhibitor would prevent DNA binding and thereby the very first step in the gyrase reaction. The mode of action of simocyclinone D8 is therefore different from that of the fluoroquinolones and from that of the classical aminocoumarins, and represents a completely new mode of action. The two binding sites for simocyclinone D8 are adjacent to but not overlapping with the binding site for fluoroquinolones. Therefore, the discovery of simocyclinone D8 raises the prospect of developing inhibitors acting on new sites of the gyrase
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
35
Fig. 3. Determination of the antibacterial activity of clorobiocin, novobiocin and the catechol-compound novclobiocin 401 against E. coli mutants with and without active TonB. Living cells were stained with 2,3,5-triphenyltetrazolium chloride to improve visualization. C = solvent control. Adapted from Alt et al. (2011a).
molecule, including bifunctional inhibitors which bind simultaneously to two different sites of their target. Initial studies reported that simocyclinone D8 was primarily active against Gram-positive bacteria, but a recent publication reported that it is also active against clinical isolates of certain Gram-negative strains (Richter et al., 2010). Furthermore, evidence was reported for the interaction of simocyclinone D8 with an additional binding site on the gyrase B subunit (Sissi et al., 2010), though the significance of this binding site is yet unclear. Combinations of fluoroquinolones and aminocoumarins reduce resistance development Fluoroquinolones like ciprofloxacin have become one of the most important drug classes in the fight against bacterial infections. However, their therapeutic usefulness in endangered by the alarming spread of fluoroquinolone-resistant pathogen strains. Resistance against fluoroquinolones arises mostly from point mutations close to the active site tyrosine in the gyrase A subunit, and may be rapidly spread in between bacteria by horizontal gene transfer, a process which frequently involves phage transduction and DNA recombination events. One important reason that especially fluoroquinolones suffer from rapid emergence of resistance is that fluoroquinolones lead to a strong activation of RecA, for example in S. aureus (Mesak et al., 2008). RecA is a key enzyme involved in DNA repair, DNA recombination and induction of the SOS response, a process which also involves expression of an error-prone DNA polymerase and thereby leads to an increased frequency of point mutations (Schröder et al., 2012). The activation of RecA is therefore central to the development of resistance against fluoroquinolones. Vickers et al. (2007) reported that resistance development against fluoroquinolones was strongly reduced when S. aureus was exposed to a combination of fluoroquinolones and aminocoumarins. Under these conditions, the frequency of resistance development was so low that, under a combination therapy with both classes of gyrase inhibitors, resistance may no longer be selected for (Vickers et al., 2007). The biological basis of this phenomenon has recently been unraveled in a study by Schröder et al. (2012). The authors first confirmed the strong (i.e. 13-fold) activation of recA expression by the fluoroquinolone ciprofloxacin in S. aureus. Subsequently, the authors proved that the aminocoumarin novobiocin led to a 17-fold repression of recA expression. Notably, the inhibitory effect of novobiocin was clearly dominant over the activating effect of ciprofloxacin: simultaneous application of novobiocin and ciprofloxacin repressed rather
than induced recA expression. Thereby, novobiocin inhibited the ciprofloxacin-induced SOS response in S. aureus. Novobiocin also reduced the spontaneous mutation frequency, as evidenced by the reduced emergence of non-hemolytic variants of S. aureus. Furthermore, novobiocin reduced the rate of phage transduction of chromosomal markers, an indicator of recombination frequency. As expected from these observations, resistance development against ciprofloxacin was reduced when S. aureus was exposed to a combination of ciprofloxacin and novobiocin, as compared to ciprofloxacin alone. The aminocoumarin antibiotics clorobiocin and simocyclinone D8 showed similar effects on recA expression as novobiocin (Schröder et al., 2012). Also for coumermycin A1 , advantages of its combination with fluoroquinolones have been described (van der Auwera et al., 1987). Therefore, the combination of aminocoumarins with fluoroquinolones may offer a strategy to provide effective antibacterial therapy with a reduced risk of resistance development. Conclusions and perspectives Aminocoumarin antibiotics act on gyrase, one of the bestestablished targets in antibacterial drug therapy. They show extremely high affinity to this target, and strong antibacterial activity against multiresistant Gram-positive strains including MRSA. Their binding sites to gyrase are different from those of the fluoroquinolones, and therefore they are also active against fluoroquinolone-resistant strains. The recent finding that aminocoumarins efficiently repress the fluoroquinolone-induced activation of RecA and thereby reduce resistance development may open new prospects for combination therapies using both fluoroquinolones and aminocoumarins. Furthermore, the discovery that the new aminocoumarin simocyclinone D8 binds at different sites of gyrase than the classical aminocoumarins, and acts by a different mechanism, opens new routes for the development of gyrase inhibitors. Besides the fluoroquinolones, the aminocoumarins are the only class of gyrase inhibitors which have been introduced into clinical anti-infective therapy. However, their clinical success has been limited for several reasons. One of these is their eukaryotic toxicity, which at least in part results from their action on eukaryotic Hsp90. As both bacterial gyrase and eukaryotic Hsp90 are well-characterized targets and readily accessible to X-ray crystallographic studies, it appears quite possible to rationally develop inhibitors with selective activity against either gyrase or Hsp90, for use as either antibacterial of anticancer drugs, respectively.
36
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Another shortcoming of the aminocoumarins is their poor activity against Gram-negative bacteria, partly due to their poor penetration across the outer membrane. As described, the development of novclobiocin 401 which is actively imported across the Gram-negative cell envelope may open prospects to overcome this problem. A pharmaceutical problem of the aminocoumarins (especially of coumermycin A1 ) is their low solubility. However, with the recent possibility to generate a multitude of derivatives of the aminocoumarins by genetic engineering techniques, as well as with the recent advances in chemical synthesis of aminocoumarin analogs, the generation of compounds with better solubility as well as improved target selectivity should be possible. Thereby, new prospects arise for the development of new and improved aminocoumarins which may be used alone, or in combination with fluoroquinolones, as gyrase inhibitors in antibacterial therapy. Acknowledgement The author gratefully acknowledges financial support from the Deutsche Forschungsgemeinschaft (SFB 766). References Alt, S., Burkard, N., Kulik, A., Grond, S., Heide, L., 2011a. An artificial pathway to 3,4-dihydroxybenzoic acid allows generation of new aminocoumarin antibiotic recognized by catechol transporters of E. coli. Chem. Biol. 18, 304–313. Alt, S., Mitchenall, L.A., Maxwell, A., Heide, L., 2011b. Inhibition of DNA gyrase and DNA topoisomerase IV of Staphylococcus aureus and Escherichia coli by aminocoumarin antibiotics. J. Antimicrob. Chemother. 66, 2061–2069. Anderle, C., Hennig, S., Kammerer, B., Li, S.M., Wessjohann, L., Gust, B., Heide, L., 2007. Improved mutasynthetic approaches for the production of modified aminocoumarin antibiotics. Chem. Biol. 14, 955–967. Anderle, C., Stieger, M., Burrell, M., Reinelt, S., Maxwell, A., Page, M., Heide, L., 2008. Biological activities of novel gyrase inhibitors of the aminocoumarin class. Antimicrob. Agents Chemother. 52, 1982–1990. Bonitz, T., Alva, V., Saleh, O., Lupas, A.N., Heide, L., 2011. Evolutionary relationships of microbial aromatic prenyltransferases. PLoS ONE 6, e27336. Burlison, J.A., Neckers, L., Smith, A.B., Maxwell, A., Blagg, B.S., 2006. Novobiocin: redesigning a DNA gyrase inhibitor for selective inhibition of Hsp90. J. Am. Chem. Soc. 128, 15529–15536. Cheenpracha, S., Vidor, N.B., Yoshida, W.Y., Davies, J., Chang, L.C., 2010. Coumabiocins A-F, aminocoumarins from an organic extract of Streptomyces sp. L-4-4. J. Nat. Prod. 73, 880–884. Chopra, S., Matsuyama, K., Tran, T., Malerich, J.P., Wan, B., Franzblau, S.G., Lun, S., Guo, H., Maiga, M.C., Bishai, W.R., Madrid, P.B., 2012. Evaluation of gyrase B as a drug target in Mycobacterium tuberculosis. J. Antimicrob. Chemother. 67, 415–421. Collin, F., Karkare, S., Maxwell, A., 2011. Exploiting bacterial DNA gyrase as a drug target: current state and perspectives. Appl. Microbiol. Biotechnol. 92, 479–497. Donnelly, A., Blagg, B.S., 2008. Novobiocin and additional inhibitors of the Hsp90 C-terminal nucleotide-binding pocket. Curr. Med. Chem. 15, 2702–2717. Dutta, R., Inouye, M., 2000. GHKL, an emergent ATPase/kinase superfamily. Trends Biochem. Sci. 25, 24–28. Edwards, M.J., Flatman, R.H., Mitchenall, L.A., Stevenson, C.E., Le, T.B., Clarke, T.A., McKay, A.R., Fiedler, H.P., Buttner, M.J., Lawson, D.M., Maxwell, A., 2009. A crystal structure of the bifunctional antibiotic simocyclinone D8, bound to DNA gyrase. Science 326, 1415–1418. Edwards, M.J., Williams, M.A., Maxwell, A., McKay, A.R., 2011. Mass spectrometry reveals that the antibiotic simocyclinone D8 binds to DNA gyrase in a “bentover” conformation: evidence of positive cooperativity in binding. Biochemistry 50, 3432–3440. Eustáquio, A.S., Gust, B., Luft, T., Li, S.M., Chater, K.F., Heide, L., 2003. Clorobiocin biosynthesis in Streptomyces, Identification of the halogenase and generation of structural analogs. Chem. Biol. 10, 279–288. Flatman, R.H., Eustáquio, A., Li, S.M., Heide, L., Maxwell, A., 2006. Structure-activity relationships of aminocoumarin-type gyrase and topoisomerase IV inhibitors obtained by combinatorial biosynthesis. Antimicrob. Agents Chemother. 50, 1136–1142. Flatman, R.H., Howells, A.J., Heide, L., Fiedler, H.P., Maxwell, A., 2005. Simocyclinone D8, an inhibitor of DNA gyrase with a novel mode of action. Antimicrob. Agents Chemother. 49, 1093–1100. Fujimoto-Nakamura, M., Ito, H., Oyamada, Y., Nishino, T., Yamagishi, J., 2005. Accumulation of mutations in both gyrB and parE genes is associated with high-level resistance to novobiocin in Staphylococcus aureus. Antimicrob. Agents Chemother. 49, 3810–3815. Gellert, M., Mizuuchi, K., O’Dea, M.H., Nash, H.A., 1976. DNA gyrase: an enzyme that introduces superhelical turns into DNA. Proc. Natl. Acad. Sci. U.S.A. 73, 3872–3876.
Heide, L., 2009a. The aminocoumarins: biosynthesis and biology. Nat. Prod. Rep. 26, 1241–1250. Heide, L., 2009b. Aminocoumarins: mutasynthesis, chemoenzymatic synthesis, and metabolic engineering. Methods Enzymol. 459, 437–455. Heide, L., 2009c. Genetic engineering of antibiotic biosynthesis for the generation of new aminocoumarins. Biotechnol. Adv. 27, 1006–1014. Hong, D.S., Banerji, U., Tavana, B., George, G.C., Aaron, J., Kurzrock, R., 2012. Targeting the molecular chaperone heat shock protein 90 (Hsp90): Lessons learned and future directions. Cancer Treat. Rev.. Kim, C.G., Lamichhane, J., Song, K.I., Nguyen, V.D., Kim, D.H., Jeong, T.S., Kang, S.H., Kim, K.W., Maharjan, J., Hong, Y.S., Kang, J.S., Yoo, J.C., Lee, J.J., Oh, T.J., Liou, K., Sohng, J.K., 2008. Biosynthesis of rubradirin as an ansamycin antibiotic from Streptomyces achromogenes var. rubradiris NRRL3061. Arch. Microbiol. 189, 463–473. Laponogov, I., Pan, X.S., Veselkov, D.A., McAuley, K.E., Fisher, L.M., Sanderson, M.R., 2010. Structural basis of gate-DNA breakage and resealing by type II topoisomerases. PLoS ONE 5, e11338. Lesher, G.Y., Froelich, E.J., Gruett, M.D., Bailey, J.H., Brundage, R.P., 1962. 1,8Naphthyridine derivatives. A new class of chemotherapeutic agents. J. Med. Pharm. Chem. 91, 1063–1065. Lewis, R.J., Singh, O.M.P., Smith, C.V., Skarzynski, T., Maxwell, A., Wonacott, A.J., Wigley, D.B., 1996. The nature of inhibition of DNA gyrase by the coumarins and the cyclothialidines revealed by X-ray crystallography. EMBO J. 15, 1412–1420. Maxwell, A., Lawson, D.M., 2003. The ATP-binding site of type II topoisomerases as a target for antibacterial drugs. Curr. Top. Med. Chem. 3, 283–303. Mesak, L.R., Miao, V., Davies, J., 2008. Effects of subinhibitory concentrations of antibiotics on SOS and DNA repair gene expression in Staphylococcus aureus. Antimicrob. Agents Chemother. 52, 3394–3397. Morgan-Linnell, S.K., Hiasa, H., Zechiedrich, L., Nitiss, J.L., 2007. Assessing sensitivity to antibacterial topoisomerase II inhibitors. Curr. Protoc. Pharmacol. 39, 3.13.1113.13.26. Pan, X.S., Fisher, L.M., 1999. Streptococcus pneumoniae DNA gyrase and topoisomerase IV: overexpression, purification, and differential inhibition by fluoroquinolones. Antimicrob. Agents Chemother. 43, 1129–1136. Pojer, F., Li, S.M., Heide, L., 2002. Molecular cloning and sequence analysis of the clorobiocin biosynthetic gene cluster: new insights into the biosynthesis of aminocoumarin antibiotics. Microbiology 148, 3901–3911. Pommier, Y., Leo, E., Zhang, H., Marchand, C., 2010. DNA topoisomerases and their poisoning by anticancer and antibacterial drugs. Chem. Biol. 17, 421–433. Pozzi, R., Simone, M., Mazzetti, C., Maffioli, S., Monciardini, P., Cavaletti, L., Bamonte, R., Sosio, M., Donadio, S., 2011. The genus Actinoallomurus and some of its metabolites. J. Antibiot. (Tokyo) 64, 133–139. Richter, S.N., Frasson, I., Palumbo, M., Sissi, C., Palu, G., 2010. Simocyclinone D8 turns on against Gram-negative bacteria in a clinical setting. Bioorg. Med. Chem. Lett. 20, 1202–1204. Samuels, D.S., Garon, C.F., 1993. Coumermycin A1 inhibits growth and induces relaxation of supercoiled plasmids in Borrelia burgdorferi, the Lyme disease agent. Antimicrob. Agents Chemother. 37, 46–50. Sanyal, G., Doig, P., 2012. Bacterial DNA replication enzymes as targets for antibacterial drug discovery. Expert Opin. Drug Discov. 7, 327–339. Schimana, J., Fiedler, H.P., Groth, I., Süssmuth, R., Beil, W., Walker, M., Zeeck, A., 2000. Simocyclinones, novel cytostatic angucyclinone antibiotics produced by Streptomyces antibioticus Tü 6040. I. Taxonomy, fermentation, isolation and biological activities. J. Antibiot. (Tokyo) 53, 779–787. Schmutz, E., Hennig, S., Li, S.M., Heide, L., 2004. Identification of a topoisomerase IV in actinobacteria: purification and characterization of ParYR and GyrBR from the coumermycin A1 producer Streptomyces rishiriensis DSM 40489. Microbiology 150, 641–647. Schmutz, E., Mühlenweg, A., Li, S.M., Heide, L., 2003. Resistance genes of aminocoumarin producers: Two type II topoisomerase genes confer resistance against coumermycin A1 and clorobiocin. Antimicrob. Agents Chemother. 47, 869–877. Schröder, W., Goerke, C., Wolz, C., 2012. Opposing effects of aminocoumarins and fluoroquinolones on the SOS response and adaptability in Staphylococcus aureus. J. Antimicrob. Chemother., doi: 10.1093/jac/dks456. Sissi, C., Vazquez, E., Chemello, A., Mitchenall, L.A., Maxwell, A., Palumbo, M., 2010. Mapping simocyclinone D8 interaction with DNA gyrase: evidence for a new binding site on GyrB. Antimicrob. Agents Chemother. 54, 213–220. Steffensky, M., Li, S.M., Heide, L., 2000. Cloning, overexpression, and purification of novobiocic acid synthetase from Streptomyces spheroides NCIMB 11891. J. Biol. Chem. 275, 21754–21760. van der Auwera, P., Vandermies, A., Grenier, P., Klastersky, J., 1987. Comparative in vitro activity of CI934, a new fluoroquinolone, alone and in combination with coumermycin, against gram-positive bacteria. Drugs Exp. Clin. Res. 13, 125–132. Vickers, A.A., O’Neill, A.J., Chopra, I., 2007. Emergence and maintenance of resistance to fluoroquinolones and coumarins in Staphylococcus aureus: predictions from in vitro studies. J. Antimicrob. Chemother. 60, 269–273. Wang, Z.X., Li, S.M., Heide, L., 2000. Identification of the coumermycin A1 biosynthetic gene cluster of Streptomyces rishiriensis DSM 40489. Antimicrob. Agents Chemother. 44, 3040–3048. Wohlkonig, A., Chan, P.F., Fosberry, A.P., Homes, P., Huang, J., Kranz, M., Leydon, V.R., Miles, T.J., Pearson, N.D., Perera, R.L., Shillings, A.J., Gwynn, M.N., Bax, B.D., 2010. Structural basis of quinolone inhibition of type IIA topoisomerases and target-mediated resistance. Nat. Struct. Mol. Biol. 17, 1152–1153.
Contents lists available at ScienceDirect
International Journal of Medical Microbiology journal homepage: www.elsevier.com/locate/ijmm
New aminocoumarin antibiotics as gyrase inhibitors Lutz Heide ∗ Pharmaceutical Institute, Eberhard Karls-Universität Tübingen, Auf der Morgenstelle 8, 72076 Tübingen, Germany
a r t i c l e
i n f o
Keywords: Aminocoumarins Novobiocin Simocyclinone Gyrase Staphylococcus aureus
a b s t r a c t The aminocoumarins novobiocin, clorobiocin and coumermycin A1 are structurally related antibiotics produced by different Streptomyces strains. They are potent inhibitors of bacterial gyrase. Their binding sites and their mode of action differ from those of fluoroquinolones such as ciprofloxacin. Novobiocin has been introduced into clinical use against Staphylococcus aureus infections, and S. aureus gyrase is particularly sensitive to inhibition by aminocoumarins, while topoisomerase IV is much less sensitive. Modern genetic techniques have allowed the engineering of the producer strains, resulting in a diverse range of new aminocoumarins, including compounds which are more active than the natural antibiotics as well as a compound which is actively imported across the cell envelope of Gram-negative bacteria. A further group of aminocoumarins are the simocyclinones which bind simultaneously to two different sites of gyrase and show a completely new mode of inhibition. Both the simocyclinones and the “classical” aminocoumarins strongly inhibit the fluoroquinolone-induced activation of RecA and thereby the SOS response in S. aureus. Therefore, a combination of aminocoumarins and fluoroquinolones strongly reduced the risk of resistance development and may offer new prospects in anti-infective therapy. © 2013 Elsevier GmbH. All rights reserved.
Introduction DNA gyrase belongs to the best-validated targets in antibacterial drug therapy (Chopra et al., 2012; Collin et al., 2011; Pommier et al., 2010; Sanyal and Doig, 2012). Gyrase was discovered in E. coli in 1976 (Gellert et al., 1976), but the two most important classes of gyrase inhibitors were already discovered before that date. The quinolone nalidixic acid was found accidentally as a side product of the synthesis of the antimalarial chloroquine (Lesher et al., 1962). Based on nalidixic acid, the fluoroquinolones were developed and became one of the principal clinical weapons in the fight against bacterial infections. However, their usefulness is now endangered by the rapid emergence of resistance. The only other class of gyrase inhibitors which has been introduced into clinical use are the aminocoumarins. Novobiocin (Fig. 1) was discovered in the 1950s in screening programs for antibiotics produced by microorganisms. It was introduced into human anti-infective therapy in 1964 under the name Albamycin® by the Upjohn company and used for the treatment of Staphylococcus aureus infections, including multiresistant MRSA strains. Novobiocin is also active against Borrelia burgdorferi, the causative agent of Lyme disease (Samuels and Garon, 1993). Shortly after novobiocin, the even more potent compounds clorobiocin and coumermycin A1 were discovered. These three
∗ Tel.: +49 7071 2972460; fax: +49 7071 295250. E-mail address: [email protected] 1438-4221/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.ijmm.2013.08.013
“classical” aminocoumarins show structural similarity (Fig. 1). An aminocoumarin moiety is linked via an amide bond to an aromatic acid, and via a glycosidic bond to the unusual deoxysugar L-noviose, which carries an acyl moiety attached to its 3-hydroxy group. This acyl moiety, the deoxysugar and the aminocoumarin moiety form the principal site of interaction with gyrase. Beyond these three “classical” compounds (and close analogs thereof), only two other natural aminocoumarins have been discovered: rubradirin which lacks the deoxysugar moiety at the 7-OH group of the aminocoumarin and which therefore is not a gyrase inhibitor (Kim et al., 2008), and the simocyclinones which will be discussed below. All these compounds are produced by soil bacteria of the genus Streptomyces. Coumermycin has also been found in strains of the genus Actinoallomurus, distantly related to Streptomyces (Pozzi et al., 2011). Further, a new isolate of a Streptomyces strain producing novobiocin and analogs thereof has been described (Cheenpracha et al., 2010). In biochemical assays, aminocoumarins inhibit DNA gyrase with IC50 values which are lower than that of modern fluoroquinolones (Alt et al., 2011b). However, the advantage of a higher target affinity is offset by a more attractive mechanism of action exerted by the fluoroquinolones. While the aminocoumarins act as competitive inhibitors of gyrase, the fluoroquinolones act as “poisons” of this enzyme. They stabilize the covalent gyrase-DNA complex, thereby leading to protein-stabilized breaks of DNA and ultimately to cell death, even at a relatively low occupancy of the inhibitor (Collin et al., 2011). Therefore, minimal inhibitory concentrations
32
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Fig. 1. (A) Structures of the three “classical” aminocoumarin antibiotics novobiocin, clorobiocin and coumermycin A1 . (B) Structure of simocyclinone D8.
(MICs) of the fluoroquinolones against pathogenic bacteria are in a similar range as those of aminocoumarin antibiotics (Schröder et al., 2012). Despite the fact that fluoroquinolones are usually referred to a “gyrase inhibitors”, their primary target in many pathogenic bacteria is in fact not gyrase but topoisomerase IV (= topo IV). Topo IV is a similar heterotetrameric enzyme as gyrase. E. coli gyrase consists of two GyrA and two GyrB subunits, and topo IV of two ParC and two Par E subunits. Gyrase and topo IV share 40% sequence identity. Both belong to the type II topoisomerases which catalyze changes in the topology of DNA involving the transient break of both strands of DNA. Both gyrase and topo IV can relax supercoiled DNA, but gyrase is unique by its capability to introduce negative supercoils. The energy for this reaction is derived from the hydrolysis of ATP, catalyzed by the GyrB subunit. In Gram-negative bacteria, usually gyrase is the primary target of the fluoroquinolones, while in Gram-positive bacteria often topo IV is the primary target (Collin et al., 2011). Also aminocoumarins can interact both with gyrase and with topo IV, but gyrase is the primary target (see below). Detailed structures of the ternary complexes of fluoroquinolones with gyrase and DNA have been published only recently (Laponogov et al., 2010; Wohlkonig et al., 2010). In contrast, the structural basis of the interaction of aminocoumarins with gyrase has been extremely well characterized from several high resolution crystal structures over the last 15 years (Collin et al., 2011; Lewis et al., 1996; Maxwell and Lawson, 2003). The binding site of the aminocoumarins (especially of their substituted deoxysugar moiety) overlaps with the binding site of ATP on the GyrB subunit, and this explains the competitive inhibition of gyrase-catalyzed ATP hydrolysis by aminocoumarins.
Gyrase belongs to the GHKL family of enzymes, named after its founding members gyrase, heat-shock protein 90 (Hsp90), certain protein kinases and the DNA mismatch repair protein MutL (Dutta and Inouye, 2000). Proteins of this family share a common type of ATP-binding fold, and therefore the aminocoumarins interact not only with bacterial gyrase but also with eukaryotic Hsp90 which is an emerging target in anticancer therapy (Hong et al., 2012). A number of novobiocin analogs have been synthesized and tested as Hsp90 inhibitors (Burlison et al., 2006; Donnelly and Blagg, 2008). Following the discovery and the functional analysis of the biosynthetic gene clusters of the three classical aminocoumarins (Pojer et al., 2002; Steffensky et al., 2000; Wang et al., 2000), it has become possible to generate a multitude of new aminocoumarin analogs by modern genetic techniques, such as combinatorial biosynthesis, mutasynthesis and synthetic biology (Heide, 2009b). This, together with the ever-increasing problem of antibacterial drug resistance, has renewed the interest in aminocoumarins as antibacterial agents. Measurement of gyrase and topoisomerase IV inhibition by aminocoumarin antibiotics Determination of the inhibitory activity of aminocoumarins and fluoroquinolones on gyrase and topo IV by biochemical assays in vitro is not trivial, and the IC50 values reported in the literature vary considerably according to the methods used. Different assay types are available, e.g. supercoiling, relaxation, DNA cleavage and decatenation assays (Morgan-Linnell et al., 2007; Pan and Fisher, 1999). The supercoiling assay is considered most suitable for the assessment of the inhibition of gyrase by aminocoumarins, and the decatenation assay for the topo IV inhibition. The former
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
assay measures the conversion of relaxed to supercoiled plasmid DNA and the latter the conversion of concatenated to decatenated kinetoplast DNA. Since the topoisomerases from E. coli are the best studied enzymes, most literature data on gyrase and topo IV inhibition have been obtained with these enzymes. However, Staphylococcus aureus is the most important pathogen addressed by aminocoumarin therapy, and the sensitivity of S. aureus topoisomerases to aminocoumarins is somewhat different from that of the E. coli enzymes. In vitro investigation of S. aureus gyrase is complicated by the fact that this enzyme requires high concentrations of potassium glutamate (K-Glu) for its activity (e.g. 700 mM) (Alt et al., 2011b). In contrast, E. coli gyrase can be, and usually is, assayed in the absence of K-Glu. However, also this enzyme shows higher activity in the presence of K-Glu. A recent study (Alt et al., 2011b) showed that the presence of K-Glu increased the sensitivity of E. coli gyrase toward aminocoumarin antibiotics approximately 10-fold, and in case of one specific compound even 150-fold. Therefore, the assay conditions are a very important factor when IC50 values of aminocoumarins for gyrase and topo IV inhibition are compared. Table 1 shows the inhibitory activity of different aminocoumarin antibiotics on gyrase and topo IV from E. coli and S. aureus, determined using identical K-Glu concentrations for the enzymes from both organisms (Alt et al., 2011b). These data confirm the excellent activity of aminocoumarins on S. aureus gyrase: 50% inhibition is achieved in the low nanomolar range. The S. aureus gyrase is more sensitive to most aminocoumarins than the E. coli enzyme. In both organisms, the sensitivity of topo IV is lower by several orders of magnitude than that of gyrase, clearly establishing gyrase as the primary target of the aminocoumarins. Despite these very clear in vitro results, the biological significance of topo IV inhibition by aminocoumarins may not be completely dismissed: cultivation of S. aureus in the presence of novobiocin results not only in the selection of mutants with altered gyrase, but as a second step of resistance development also in mutations of topo IV (Fujimoto-Nakamura et al., 2005). Furthermore, the biosynthetic gene clusters of the most potent natural aminocoumarins, clorobiocin and coumermycin, contain not only a gene coding for an aminocoumarin-resistant gyrase B subunit for self-resistance of the producer strain, but also a gene coding for a resistant ParE subunit of topo IV (Schmutz et al., 2004; Schmutz et al., 2003). This indicates that the toxic effect of aminocoumarins on topo IV is a relevant problem for the producer strain. New aminocoumarin antibiotics from combinatorial biosynthesis and mutasynthesis Since the discovery of the sulfonamides and the penicillins, there have been two principal routes for the discovery of new anti-infectives: chemical synthesis and screening of natural products (mostly produced by microorganisms). Most of the antibiotics which are presently in clinical use are natural products or derivatives thereof, with the penicillins as the best-known example. In contrast, only few classes of synthetic compounds have achieved an important role in antibacterial therapy, most prominently the sulfonamides and the fluoroquinolones. However, the discovery of new antibiotics by screening of microbial natural products has become increasingly difficult, as many microbial strains which are easily accessible and which can be cultivated under laboratory conditions have already been investigated. With the advent of modern genetic and genomic techniques, however, new and promising routes have been opened for antibiotic drug discovery from microbial strains. These routes include e.g. combinatorial biosynthesis, mutasynthesis and synthetic biology, which have been very successfully exploited for the generation of new aminocoumarins.
33
Fig. 2. Structures of selected aminocoumarin antibiotics obtained from combinatorial biosynthesis, mutasynthesis and synthetic biology.
Soil bacteria and fungi which produce antibiotics possess specific genes for the biosynthesis of these compounds. Usually, all genes required for the biosynthesis of a given antibiotic are grouped together in one contiguous stretch of DNA, the so-called biosynthetic gene cluster. E.g. the biosynthetic gene clusters of the three “classical” aminocoumarin antibiotics novobiocin, clorobiocin and coumermycin A1 span between 23 and 38 kb and comprise between 20 and 33 genes, including biosynthetic, regulatory and resistance genes and genes for transporters (Heide, 2009a). The function of nearly all these has been experimentally identified, partly revealing completely new enzymes and enzyme classes which were undiscovered previously (Bonitz et al., 2011; Heide, 2009a). Knowledge of the function of biosynthetic genes readily allows the modification of these genes and gene clusters in order to generate modified antibiotic structures. E.g. the chlorine atom of clorobiocin (Fig. 1) is introduced by the halogenase enzyme Clohal. Inactivation of the clo-hal gene in the producer strain led to the accumulation of a clorobiocin analog lacking the chlorine atom (novclobiocin 101, Fig. 2) (Eustáquio et al., 2003). Novobiocin contains not chlorine but a methyl group in the corresponding position of the aminocoumarin moiety (Fig. 1), and this methyl group is attached under catalysis of the methyl transferase NovO. Expression of the gene novO in the above mentioned mutant of the clorobiocin producer led to the unnatural “hybrid” antibiotic novclobiocin 102 (Fig. 2), carrying the methyl group typical of novobiocin in an otherwise unmodified skeleton of clorobiocin. Such experiments have allowed the investigation of structurefunction relationships in the class of aminocoumarin antibiotics. As shown in Table 1, removal of the chlorine group from the clorobiocin structure (resulting in novclobiocin 101) clearly reduces the activity. Activity is largely restored by introduction of the methyl group in this position (resulting in novclobiocin 102). Removal of the acyl group attached to the 3-OH group of the deoxysugar of clorobiocin (resulting in novclobiocin 103) strongly reduces the activity, especially against S. aureus gyrase. Many more modifications of the aminocoumarins have been generated in similar experiments, defining the structural elements of aminocoumarin antibiotics which are required for optimal interaction with the target (Alt et al., 2011b; Flatman et al., 2006), and the aminocoumarins
34
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Table 1 Activity of selected aminocoumarin antibiotics against DNA gyrase and topoisomerase IV. Gyrase activity was determined in a supercoiling assay in the presence of 700 mM potassium glutamate (K-Glu), and topo IV activity in a decatenation assay in the presence of 100 mM K-Glu (Alt et al., 2011a; Alt et al., 2011b). IC50 [nM] against DNA gyrase
novobiocin clorobiocin coumermycin A1 simocyclinone D8 novclobiocin 101 novclobiocin 102 novclobiocin 217 novclobiocin 225 novclobiocin 401 novclobiocin 103
IC50 [nM] against topoisomerase IV
from E. coli
from S. aureus
from E. coli
from S. aureus
80 30 30 100 300 30 6 6 30 100
10 6 6 2000 50 10 1 1 6 1000
10 000 3000 5000 >10 000 3000 300 8000 8000 >50 000 >10 000
20 000 10 000 100 000 17 000 35 000 5000 >50 000 >50 000 35 000 >50 000
have provided an example of the power of such “combinatorial biosynthesis” experiments for the generation of new compounds (Heide, 2009c). Yet, even larger structural diversity can be generated when the power of genetic recombination experiments is combined with the possibilities of organic synthesis. This is e.g. achieved in mutasynthesis experiments, in which the biosynthesis of a genuine structural moiety of the antibiotic is blocked by an appropriate gene inactivation. Subsequently, analogs of this structural moiety are chemically synthesized and fed to the mutant strain. These analogs can then be incorporated into the antibiotic, leading to new variants of the drug. This technique has been further refined (Anderle et al., 2007), leading e.g. to novclobiocins 217 and 225 (Fig. 2). As shown in Table 1, these compounds are more potent inhibitors of gyrase than the most potent natural aminocoumarins. They both show IC50 values of 1 nM against S. aureus gyrase, and this translates into excellent antibacterial activity against S. aureus in vitro, with MIC values < 0.06 g/ml (Anderle et al., 2008). Therefore, even though the affinity of natural aminocoumarins for their target has been optimized over millions of years in evolution, this affinity can still be improved by suitable structural modifications.
cluster for clorobiocin biosynthesis, deficient in a gene required for the biosynthesis of the genuine substituted benzoic acid moiety of clorobiocin, was introduced into this strain. This synthetic biology approach resulted in the efficient formation of the desired catecholsubstituted aminocoumarin antibiotic novclobiocin 401 (Fig. 2) (Alt et al., 2011a). The structural modification did not reduce the activity of this compound against gyrase, compared to the genuine antibiotic clorobiocin (Table 1). However, it clearly increased the activity against an E. coli test strain, as shown by the larger inhibition zone in a disk diffusion assay (Fig. 3). In contrast, an E. coli test strain deficient in tonB (required for energizing the catechol transporters) was much less inhibited by novclobiocin 401 (Fig. 3), indicating that this compound was actively imported across the Gram-negative cell envelope by the TonB-dependent catechol siderophore transporters. Novclobiocin 401 provides an example for the power of modern synthetic biology techniques for the generation of new antibiotics, and for the possibility to improve the activity of aminocoumarin antibiotics against Gram-negative organisms by suitable structural modification.
A new aminocoumarin antibiotic is actively imported across the cell envelope of Gram-negative bacteria A principal shortcoming of the aminocoumarin antibiotics is their poor activity against Gram-negative bacteria. One important reason for this is their poor penetration across the outer membrane of the Gram-negative cell envelope. However, the outer membrane contains transporters for the active import of certain compounds, e.g. of so-called catechol siderophores, compounds which complex iron through their ortho-diphenol (= “catechol”) structures and thereby facilitate uptake of this vital mineral ion into the bacterial cell. The recently generated aminocoumarin antibiotic novclobiocin 401 (Fig. 2) contains the catechol structural moiety 3,4-dihydroxybenzoic acid (3,4-DHBA). Since it mimics the structure of the catechol siderophores, it is actively imported across the bacterial cell envelope (Alt et al., 2011a). In contrast to the successful mutasynthesis experiments described above, novclobiocin 401 could not be obtained by feeding of 3,4-DHBA to an appropriate mutant strain, due to quick degradation of this benzoic acid derivative during external feeding. However, 3,4-DHBA was generated successfully in the cell by a synthetic biology strategy. An artificial gene operon was chemically synthesized, encoding a twostep pathway from chorismic acid (an intermediate of aromatic amino acid biosynthesis) to 3,4-DHBA. This pathway combines two enzymes of completely unrelated organisms and does not exist in nature. The two genes were synthesized with an optimized sequence (i.e. optimized codon usage) for efficient translation in a Streptomyces host strain, and with a strong constitutive promoter for efficient transcription. This construct was introduced into a specific, genetically engineered host strain. In addition, a modified gene
Simocyclinone D8, a bifunctional aminocoumarin antibiotic simultaneously binding to two different sites of gyrase Simocyclinone D8 (Fig. 1) is an aminocoumarin antibiotic discovered by Fiedler and coworkers (Schimana et al., 2000). It is produced, together with several structural analogs, by a Streptomyces strain. Like the classical aminocoumarins, it contains a 3-amino-4,7-dihydroxy-coumarin moiety, but it lacks the substituted deoxysugar moiety bound to the 7-hydroxy group. Therefore, it cannot bind to the typical aminocoumarin binding site on the gyrase B subunit, and it does not inhibit gyrase B-catalyzed ATP hydrolysis. Unexpectedly, simocyclinone D8 was nevertheless found to be a very efficient inhibitor of gyrase (Flatman et al., 2005). X-ray crystallographic studies (Edwards et al., 2009) as well as mass spectrometric studies (Edwards et al., 2011) showed that simocyclinone D8 simultaneously binds to two different site of the gyrase A subunit. One binding pocket interacts with the aminocoumarin moiety, and the other binding pocket with the angucyclinone moiety of the antibiotic. Both pockets lie in the predicted DNA binding saddle of gyrase, and therefore occupancy of either site by the inhibitor would prevent DNA binding and thereby the very first step in the gyrase reaction. The mode of action of simocyclinone D8 is therefore different from that of the fluoroquinolones and from that of the classical aminocoumarins, and represents a completely new mode of action. The two binding sites for simocyclinone D8 are adjacent to but not overlapping with the binding site for fluoroquinolones. Therefore, the discovery of simocyclinone D8 raises the prospect of developing inhibitors acting on new sites of the gyrase
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
35
Fig. 3. Determination of the antibacterial activity of clorobiocin, novobiocin and the catechol-compound novclobiocin 401 against E. coli mutants with and without active TonB. Living cells were stained with 2,3,5-triphenyltetrazolium chloride to improve visualization. C = solvent control. Adapted from Alt et al. (2011a).
molecule, including bifunctional inhibitors which bind simultaneously to two different sites of their target. Initial studies reported that simocyclinone D8 was primarily active against Gram-positive bacteria, but a recent publication reported that it is also active against clinical isolates of certain Gram-negative strains (Richter et al., 2010). Furthermore, evidence was reported for the interaction of simocyclinone D8 with an additional binding site on the gyrase B subunit (Sissi et al., 2010), though the significance of this binding site is yet unclear. Combinations of fluoroquinolones and aminocoumarins reduce resistance development Fluoroquinolones like ciprofloxacin have become one of the most important drug classes in the fight against bacterial infections. However, their therapeutic usefulness in endangered by the alarming spread of fluoroquinolone-resistant pathogen strains. Resistance against fluoroquinolones arises mostly from point mutations close to the active site tyrosine in the gyrase A subunit, and may be rapidly spread in between bacteria by horizontal gene transfer, a process which frequently involves phage transduction and DNA recombination events. One important reason that especially fluoroquinolones suffer from rapid emergence of resistance is that fluoroquinolones lead to a strong activation of RecA, for example in S. aureus (Mesak et al., 2008). RecA is a key enzyme involved in DNA repair, DNA recombination and induction of the SOS response, a process which also involves expression of an error-prone DNA polymerase and thereby leads to an increased frequency of point mutations (Schröder et al., 2012). The activation of RecA is therefore central to the development of resistance against fluoroquinolones. Vickers et al. (2007) reported that resistance development against fluoroquinolones was strongly reduced when S. aureus was exposed to a combination of fluoroquinolones and aminocoumarins. Under these conditions, the frequency of resistance development was so low that, under a combination therapy with both classes of gyrase inhibitors, resistance may no longer be selected for (Vickers et al., 2007). The biological basis of this phenomenon has recently been unraveled in a study by Schröder et al. (2012). The authors first confirmed the strong (i.e. 13-fold) activation of recA expression by the fluoroquinolone ciprofloxacin in S. aureus. Subsequently, the authors proved that the aminocoumarin novobiocin led to a 17-fold repression of recA expression. Notably, the inhibitory effect of novobiocin was clearly dominant over the activating effect of ciprofloxacin: simultaneous application of novobiocin and ciprofloxacin repressed rather
than induced recA expression. Thereby, novobiocin inhibited the ciprofloxacin-induced SOS response in S. aureus. Novobiocin also reduced the spontaneous mutation frequency, as evidenced by the reduced emergence of non-hemolytic variants of S. aureus. Furthermore, novobiocin reduced the rate of phage transduction of chromosomal markers, an indicator of recombination frequency. As expected from these observations, resistance development against ciprofloxacin was reduced when S. aureus was exposed to a combination of ciprofloxacin and novobiocin, as compared to ciprofloxacin alone. The aminocoumarin antibiotics clorobiocin and simocyclinone D8 showed similar effects on recA expression as novobiocin (Schröder et al., 2012). Also for coumermycin A1 , advantages of its combination with fluoroquinolones have been described (van der Auwera et al., 1987). Therefore, the combination of aminocoumarins with fluoroquinolones may offer a strategy to provide effective antibacterial therapy with a reduced risk of resistance development. Conclusions and perspectives Aminocoumarin antibiotics act on gyrase, one of the bestestablished targets in antibacterial drug therapy. They show extremely high affinity to this target, and strong antibacterial activity against multiresistant Gram-positive strains including MRSA. Their binding sites to gyrase are different from those of the fluoroquinolones, and therefore they are also active against fluoroquinolone-resistant strains. The recent finding that aminocoumarins efficiently repress the fluoroquinolone-induced activation of RecA and thereby reduce resistance development may open new prospects for combination therapies using both fluoroquinolones and aminocoumarins. Furthermore, the discovery that the new aminocoumarin simocyclinone D8 binds at different sites of gyrase than the classical aminocoumarins, and acts by a different mechanism, opens new routes for the development of gyrase inhibitors. Besides the fluoroquinolones, the aminocoumarins are the only class of gyrase inhibitors which have been introduced into clinical anti-infective therapy. However, their clinical success has been limited for several reasons. One of these is their eukaryotic toxicity, which at least in part results from their action on eukaryotic Hsp90. As both bacterial gyrase and eukaryotic Hsp90 are well-characterized targets and readily accessible to X-ray crystallographic studies, it appears quite possible to rationally develop inhibitors with selective activity against either gyrase or Hsp90, for use as either antibacterial of anticancer drugs, respectively.
36
L. Heide / International Journal of Medical Microbiology 304 (2014) 31–36
Another shortcoming of the aminocoumarins is their poor activity against Gram-negative bacteria, partly due to their poor penetration across the outer membrane. As described, the development of novclobiocin 401 which is actively imported across the Gram-negative cell envelope may open prospects to overcome this problem. A pharmaceutical problem of the aminocoumarins (especially of coumermycin A1 ) is their low solubility. However, with the recent possibility to generate a multitude of derivatives of the aminocoumarins by genetic engineering techniques, as well as with the recent advances in chemical synthesis of aminocoumarin analogs, the generation of compounds with better solubility as well as improved target selectivity should be possible. Thereby, new prospects arise for the development of new and improved aminocoumarins which may be used alone, or in combination with fluoroquinolones, as gyrase inhibitors in antibacterial therapy. Acknowledgement The author gratefully acknowledges financial support from the Deutsche Forschungsgemeinschaft (SFB 766). References Alt, S., Burkard, N., Kulik, A., Grond, S., Heide, L., 2011a. An artificial pathway to 3,4-dihydroxybenzoic acid allows generation of new aminocoumarin antibiotic recognized by catechol transporters of E. coli. Chem. Biol. 18, 304–313. Alt, S., Mitchenall, L.A., Maxwell, A., Heide, L., 2011b. Inhibition of DNA gyrase and DNA topoisomerase IV of Staphylococcus aureus and Escherichia coli by aminocoumarin antibiotics. J. Antimicrob. Chemother. 66, 2061–2069. Anderle, C., Hennig, S., Kammerer, B., Li, S.M., Wessjohann, L., Gust, B., Heide, L., 2007. Improved mutasynthetic approaches for the production of modified aminocoumarin antibiotics. Chem. Biol. 14, 955–967. Anderle, C., Stieger, M., Burrell, M., Reinelt, S., Maxwell, A., Page, M., Heide, L., 2008. Biological activities of novel gyrase inhibitors of the aminocoumarin class. Antimicrob. Agents Chemother. 52, 1982–1990. Bonitz, T., Alva, V., Saleh, O., Lupas, A.N., Heide, L., 2011. Evolutionary relationships of microbial aromatic prenyltransferases. PLoS ONE 6, e27336. Burlison, J.A., Neckers, L., Smith, A.B., Maxwell, A., Blagg, B.S., 2006. Novobiocin: redesigning a DNA gyrase inhibitor for selective inhibition of Hsp90. J. Am. Chem. Soc. 128, 15529–15536. Cheenpracha, S., Vidor, N.B., Yoshida, W.Y., Davies, J., Chang, L.C., 2010. Coumabiocins A-F, aminocoumarins from an organic extract of Streptomyces sp. L-4-4. J. Nat. Prod. 73, 880–884. Chopra, S., Matsuyama, K., Tran, T., Malerich, J.P., Wan, B., Franzblau, S.G., Lun, S., Guo, H., Maiga, M.C., Bishai, W.R., Madrid, P.B., 2012. Evaluation of gyrase B as a drug target in Mycobacterium tuberculosis. J. Antimicrob. Chemother. 67, 415–421. Collin, F., Karkare, S., Maxwell, A., 2011. Exploiting bacterial DNA gyrase as a drug target: current state and perspectives. Appl. Microbiol. Biotechnol. 92, 479–497. Donnelly, A., Blagg, B.S., 2008. Novobiocin and additional inhibitors of the Hsp90 C-terminal nucleotide-binding pocket. Curr. Med. Chem. 15, 2702–2717. Dutta, R., Inouye, M., 2000. GHKL, an emergent ATPase/kinase superfamily. Trends Biochem. Sci. 25, 24–28. Edwards, M.J., Flatman, R.H., Mitchenall, L.A., Stevenson, C.E., Le, T.B., Clarke, T.A., McKay, A.R., Fiedler, H.P., Buttner, M.J., Lawson, D.M., Maxwell, A., 2009. A crystal structure of the bifunctional antibiotic simocyclinone D8, bound to DNA gyrase. Science 326, 1415–1418. Edwards, M.J., Williams, M.A., Maxwell, A., McKay, A.R., 2011. Mass spectrometry reveals that the antibiotic simocyclinone D8 binds to DNA gyrase in a “bentover” conformation: evidence of positive cooperativity in binding. Biochemistry 50, 3432–3440. Eustáquio, A.S., Gust, B., Luft, T., Li, S.M., Chater, K.F., Heide, L., 2003. Clorobiocin biosynthesis in Streptomyces, Identification of the halogenase and generation of structural analogs. Chem. Biol. 10, 279–288. Flatman, R.H., Eustáquio, A., Li, S.M., Heide, L., Maxwell, A., 2006. Structure-activity relationships of aminocoumarin-type gyrase and topoisomerase IV inhibitors obtained by combinatorial biosynthesis. Antimicrob. Agents Chemother. 50, 1136–1142. Flatman, R.H., Howells, A.J., Heide, L., Fiedler, H.P., Maxwell, A., 2005. Simocyclinone D8, an inhibitor of DNA gyrase with a novel mode of action. Antimicrob. Agents Chemother. 49, 1093–1100. Fujimoto-Nakamura, M., Ito, H., Oyamada, Y., Nishino, T., Yamagishi, J., 2005. Accumulation of mutations in both gyrB and parE genes is associated with high-level resistance to novobiocin in Staphylococcus aureus. Antimicrob. Agents Chemother. 49, 3810–3815. Gellert, M., Mizuuchi, K., O’Dea, M.H., Nash, H.A., 1976. DNA gyrase: an enzyme that introduces superhelical turns into DNA. Proc. Natl. Acad. Sci. U.S.A. 73, 3872–3876.
Heide, L., 2009a. The aminocoumarins: biosynthesis and biology. Nat. Prod. Rep. 26, 1241–1250. Heide, L., 2009b. Aminocoumarins: mutasynthesis, chemoenzymatic synthesis, and metabolic engineering. Methods Enzymol. 459, 437–455. Heide, L., 2009c. Genetic engineering of antibiotic biosynthesis for the generation of new aminocoumarins. Biotechnol. Adv. 27, 1006–1014. Hong, D.S., Banerji, U., Tavana, B., George, G.C., Aaron, J., Kurzrock, R., 2012. Targeting the molecular chaperone heat shock protein 90 (Hsp90): Lessons learned and future directions. Cancer Treat. Rev.. Kim, C.G., Lamichhane, J., Song, K.I., Nguyen, V.D., Kim, D.H., Jeong, T.S., Kang, S.H., Kim, K.W., Maharjan, J., Hong, Y.S., Kang, J.S., Yoo, J.C., Lee, J.J., Oh, T.J., Liou, K., Sohng, J.K., 2008. Biosynthesis of rubradirin as an ansamycin antibiotic from Streptomyces achromogenes var. rubradiris NRRL3061. Arch. Microbiol. 189, 463–473. Laponogov, I., Pan, X.S., Veselkov, D.A., McAuley, K.E., Fisher, L.M., Sanderson, M.R., 2010. Structural basis of gate-DNA breakage and resealing by type II topoisomerases. PLoS ONE 5, e11338. Lesher, G.Y., Froelich, E.J., Gruett, M.D., Bailey, J.H., Brundage, R.P., 1962. 1,8Naphthyridine derivatives. A new class of chemotherapeutic agents. J. Med. Pharm. Chem. 91, 1063–1065. Lewis, R.J., Singh, O.M.P., Smith, C.V., Skarzynski, T., Maxwell, A., Wonacott, A.J., Wigley, D.B., 1996. The nature of inhibition of DNA gyrase by the coumarins and the cyclothialidines revealed by X-ray crystallography. EMBO J. 15, 1412–1420. Maxwell, A., Lawson, D.M., 2003. The ATP-binding site of type II topoisomerases as a target for antibacterial drugs. Curr. Top. Med. Chem. 3, 283–303. Mesak, L.R., Miao, V., Davies, J., 2008. Effects of subinhibitory concentrations of antibiotics on SOS and DNA repair gene expression in Staphylococcus aureus. Antimicrob. Agents Chemother. 52, 3394–3397. Morgan-Linnell, S.K., Hiasa, H., Zechiedrich, L., Nitiss, J.L., 2007. Assessing sensitivity to antibacterial topoisomerase II inhibitors. Curr. Protoc. Pharmacol. 39, 3.13.1113.13.26. Pan, X.S., Fisher, L.M., 1999. Streptococcus pneumoniae DNA gyrase and topoisomerase IV: overexpression, purification, and differential inhibition by fluoroquinolones. Antimicrob. Agents Chemother. 43, 1129–1136. Pojer, F., Li, S.M., Heide, L., 2002. Molecular cloning and sequence analysis of the clorobiocin biosynthetic gene cluster: new insights into the biosynthesis of aminocoumarin antibiotics. Microbiology 148, 3901–3911. Pommier, Y., Leo, E., Zhang, H., Marchand, C., 2010. DNA topoisomerases and their poisoning by anticancer and antibacterial drugs. Chem. Biol. 17, 421–433. Pozzi, R., Simone, M., Mazzetti, C., Maffioli, S., Monciardini, P., Cavaletti, L., Bamonte, R., Sosio, M., Donadio, S., 2011. The genus Actinoallomurus and some of its metabolites. J. Antibiot. (Tokyo) 64, 133–139. Richter, S.N., Frasson, I., Palumbo, M., Sissi, C., Palu, G., 2010. Simocyclinone D8 turns on against Gram-negative bacteria in a clinical setting. Bioorg. Med. Chem. Lett. 20, 1202–1204. Samuels, D.S., Garon, C.F., 1993. Coumermycin A1 inhibits growth and induces relaxation of supercoiled plasmids in Borrelia burgdorferi, the Lyme disease agent. Antimicrob. Agents Chemother. 37, 46–50. Sanyal, G., Doig, P., 2012. Bacterial DNA replication enzymes as targets for antibacterial drug discovery. Expert Opin. Drug Discov. 7, 327–339. Schimana, J., Fiedler, H.P., Groth, I., Süssmuth, R., Beil, W., Walker, M., Zeeck, A., 2000. Simocyclinones, novel cytostatic angucyclinone antibiotics produced by Streptomyces antibioticus Tü 6040. I. Taxonomy, fermentation, isolation and biological activities. J. Antibiot. (Tokyo) 53, 779–787. Schmutz, E., Hennig, S., Li, S.M., Heide, L., 2004. Identification of a topoisomerase IV in actinobacteria: purification and characterization of ParYR and GyrBR from the coumermycin A1 producer Streptomyces rishiriensis DSM 40489. Microbiology 150, 641–647. Schmutz, E., Mühlenweg, A., Li, S.M., Heide, L., 2003. Resistance genes of aminocoumarin producers: Two type II topoisomerase genes confer resistance against coumermycin A1 and clorobiocin. Antimicrob. Agents Chemother. 47, 869–877. Schröder, W., Goerke, C., Wolz, C., 2012. Opposing effects of aminocoumarins and fluoroquinolones on the SOS response and adaptability in Staphylococcus aureus. J. Antimicrob. Chemother., doi: 10.1093/jac/dks456. Sissi, C., Vazquez, E., Chemello, A., Mitchenall, L.A., Maxwell, A., Palumbo, M., 2010. Mapping simocyclinone D8 interaction with DNA gyrase: evidence for a new binding site on GyrB. Antimicrob. Agents Chemother. 54, 213–220. Steffensky, M., Li, S.M., Heide, L., 2000. Cloning, overexpression, and purification of novobiocic acid synthetase from Streptomyces spheroides NCIMB 11891. J. Biol. Chem. 275, 21754–21760. van der Auwera, P., Vandermies, A., Grenier, P., Klastersky, J., 1987. Comparative in vitro activity of CI934, a new fluoroquinolone, alone and in combination with coumermycin, against gram-positive bacteria. Drugs Exp. Clin. Res. 13, 125–132. Vickers, A.A., O’Neill, A.J., Chopra, I., 2007. Emergence and maintenance of resistance to fluoroquinolones and coumarins in Staphylococcus aureus: predictions from in vitro studies. J. Antimicrob. Chemother. 60, 269–273. Wang, Z.X., Li, S.M., Heide, L., 2000. Identification of the coumermycin A1 biosynthetic gene cluster of Streptomyces rishiriensis DSM 40489. Antimicrob. Agents Chemother. 44, 3040–3048. Wohlkonig, A., Chan, P.F., Fosberry, A.P., Homes, P., Huang, J., Kranz, M., Leydon, V.R., Miles, T.J., Pearson, N.D., Perera, R.L., Shillings, A.J., Gwynn, M.N., Bax, B.D., 2010. Structural basis of quinolone inhibition of type IIA topoisomerases and target-mediated resistance. Nat. Struct. Mol. Biol. 17, 1152–1153.