Research Article

Targeting DnaN for tuberculosis therapy using novel griselimycins

See allHide authors and affiliations

Science  05 Jun 2015:
Vol. 348, Issue 6239, pp. 1106-1112
DOI: 10.1126/science.aaa4690

New for old—TB drug development

Tuberculosis (TB) is a global health threat for which there is only lengthy drug treatment. Patients need to consume multiple tablets over several months and frequently fail to complete their treatment. Consequently, drug-resistant strains of the pathogen have emerged, which add to the threat. Kling et al. revisited a natural product called griselimycin, extracted from the same organism that produced the prototype anti-TB drug, streptomycin. Unmodified griselimycin has poor pharmacological properties. However, one synthetic derivative had improved oral uptake and penetrated cells of the immune system that harbor the TB mycobacterium. In combination with other drugs, the griselimycin derivative showed high potency in mice with TB.

Science, this issue p. 1106

Abstract

The discovery of Streptomyces-produced streptomycin founded the age of tuberculosis therapy. Despite the subsequent development of a curative regimen for this disease, tuberculosis remains a worldwide problem, and the emergence of multidrug-resistant Mycobacterium tuberculosis has prioritized the need for new drugs. Here we show that new optimized derivatives from Streptomyces-derived griselimycin are highly active against M. tuberculosis, both in vitro and in vivo, by inhibiting the DNA polymerase sliding clamp DnaN. We discovered that resistance to griselimycins, occurring at very low frequency, is associated with amplification of a chromosomal segment containing dnaN, as well as the ori site. Our results demonstrate that griselimycins have high translational potential for tuberculosis treatment, validate DnaN as an antimicrobial target, and capture the process of antibiotic pressure-induced gene amplification.

The discovery of streptomycin, a natural antibiotic produced by Streptomyces griseus, marked the beginning of two formative disciplines within the field of infectious diseases—namely, the study of bacterial-derived (rather than fungal- or plant-derived) medicinal compounds and the drug treatment of tuberculosis (TB) (1). This achievement initiated decades of research in the discovery and use of anti-TB drugs, ultimately leading to the development of the 6-month, multidrug regimen currently used for the cure of drug-susceptible TB (2). Unfortunately, failures in the implementation of this curative regimen, which are partly due to the challenges of its complex and lengthy nature, have led to the development and transmission of drug-resistant strains of Mycobacterium tuberculosis. Today, TB remains an enormous global health burden, causing an estimated 1.3 million deaths and 8.7 million new cases in 2012, and a growing percentage of TB (more than 30% of new cases in some countries) is multidrug-resistant (3). Thus, new drugs addressing novel M. tuberculosis targets are needed to provide different therapy options for patients with drug-resistant TB and also to both shorten and simplify treatment of drug-sensitive TB. Ideally, these new drugs should be combined in regimens tackling both drug-sensitive and drug-resistant TB, representing a paradigm shift toward more universally useful TB treatment regimens.

Bacterial-derived natural products remain a rich source for antibacterial lead compounds. In fact, ~80% of the currently used antibiotics are either directly derived from bacterial metabolic pathways or represent structural derivatives of metabolites with improved pharmaceutical properties (4). However, due to the reduced interest in development of antibacterial drugs in the last decades of the 20th century, quite a number of promising natural product leads were not advanced to clinical development. Recently, natural product and antibiotic research has been revitalized, not only because of the urgent need to identify novel antibiotics but also owing to advanced technologies becoming available. Thus, researchers are now enabled to overcome hurdles in natural product research, such as target identification by deciphering the self-resistance mechanisms in producer strains through whole-genome sequencing and compound optimization by genetic engineering. Successful recent applications of these capabilities include the derivation of semisynthetic spectinamides found to be highly active against both drug-resistant and -susceptible M. tuberculosis strains (5) and the identification of InhA as the mycobacterial target of the Dactylosporangium fulvum–produced pyridomycin (6).

In a search for neglected antibiotics with high anti-TB potential, Sanofi reinvestigated griselimycin (GM) (Fig. 1), a cyclic peptide that was isolated from two strains of Streptomyces identified in the 1960s (7). GM was found to have antibacterial activity specifically against organisms within the Corynebacterineae suborder, notably including Mycobacterium species, which prompted the company Rhône-Poulenc to pursue development of GM as an anti-TB drug. The first human studies were promising but revealed poor pharmacokinetics of GM, in particular short plasma elimination half-life after oral administration (8, 9). Following elucidation of the compound’s structure (10, 11), a derivatization program was initiated to find GM analogs with improved pharmacokinetic properties (12, 13); however, this program was terminated in the 1970s after rifampin (RIF) became available for TB treatment. Because of earlier reports of the effectiveness of GM against drug-resistant M. tuberculosis (14, 15), we recently reinitiated studies on this natural product lead with the ultimate goal of introducing a highly active, stable, and safe derivative of this compound class into the TB drug development pipeline.

Fig. 1 Chemical structure of GM and derivatives.

Substitutions at Pro8 increased metabolic stability.

Development of GM analogs

Our primary optimization goals for GM were to increase its potency, metabolic stability, and exposure. Metabolic stability profiling of natural, less abundant analogs of GM identified Pro8 as a main site of metabolic degradation, supported by the finding that the methyl derivative [methylgriselimycin (MGM)] (Fig. 1) was markedly more stable than GM itself after incubation with human liver microsomes (Table 1). Because only very small amounts of MGM are produced naturally, a total synthesis route was elaborated to provide access to MGM and related analogs (see supplementary text). Structure-activity relationships of new synthetic GM analogs resulting from this approach confirmed that incorporation of substituents at Pro8 led to metabolically highly stable compounds and also indicated that increasing lipophilicity considerably increased in vivo exposure in plasma and lungs of mice, as well as in vitro activity against M. tuberculosis (see supplementary text). From these efforts, cyclohexylgriselimycin (CGM) (Fig. 1) was identified. The minimum inhibitory concentration (MIC) values of CGM were 0.06 and 0.2 μg/ml for the drug-susceptible M. tuberculosis strain H37Rv in broth culture and within macrophage-like (RAW264.7) cells, respectively (Table 1). CGM exhibited time-dependent bactericidal activity in vitro (Fig. 2A). Although the unbound fraction of CGM in plasma was low (0.3 and 0.4% in human and mouse plasma, respectively), the MIC shift of CGM in the presence of human or mouse sera was only around five- or sevenfold, respectively (Table 1). The MIC values of CGM were similar for a range of M. tuberculosis strains from different lineages (representing geographical and evolutionary diversity), as well as for strains monoresistant to first- or second-line anti-TB drugs (table S1), demonstrating a lack of cross-resistance. GM, MGM, and CGM were not active against M. tuberculosis under hypoxic conditions in which the bacteria were not actively multiplying (Table 1). CGM exhibited optimized adsorption, distribution, metabolism, and excretion properties; that is, high oral bioavailability (89%), moderate total plasma clearance (1.1 liter/hour per kilogram), and a large volume of distribution (5.5 liter/kg). Exposure and half-life in mice (supportive of once-daily dosing) were higher than for GM and MGM. In addition, over the 30- to 100-mg/kg dose range, CGM exposure in plasma and lung increased roughly with dose proportionality and continued to increase over the 100- to 600-mg/kg dose range (table S2). Moreover, CGM exhibited limited potential for drug-drug interactions, as neither CYP induction nor inhibition was observed, and the contribution of CYP to degradation was balanced, indicating that induction of CYPs by other drugs should not affect CGM exposures (Table 1). CGM (at concentrations up to 5000 μg/ml) did not increase the number of revertant colonies in Ames II testing with TA98 and mixed strains of Salmonella typhimurium, indicating a lack of mutagenicity. Additionally, no chromosomal aberrations were observed in CGM-exposed mammalian cells, as tested with L5178Y cells (at CGM concentrations up to 1000 μg/ml, with or without metabolic activation) and with CHO-K1 cells (fig. S1), indicating a lack of genotoxicity.

Table 1 Optimization parameters for GM and derivatives.

Cmax, maximum concentration; AUC, area under the concentration curve; Vss, volume of distribution; t½, half-life; nd, not determined. Pharmacokinetic parameters (Cmax, AUC, and t½) were determined after a single oral administration of 30 mg/kg of the test compound in mice. For oral bioavailability, a single oral dose was compared to a single intravenous dose of 3 mg/kg. F = [AUCoral]/[AUCiv], the ratio of exposure of an equivalent dose after nonintravenous (in this case, oral) and intravenous administration as a measure of bioavailability.

View this table:
Fig. 2 In vitro and in vivo CGM activity against M. tuberculosis.

(A) Activity of CGM against M. tuberculosis in 7H9 broth culture. The number following the compound abbreviation indicates the concentration in micrograms per milliliter. (B) CGM dose-ranging in the mouse model of acute TB: CFU counts after 4 weeks of treatment. Data represent the mean lung log10 CFU counts (five mice per group). Day 0 indicates lung CFU counts at treatment initiation (the day after infection). The number following the compound abbreviation indicates the dose in milligrams per kilogram per day. (C) CGM dose-ranging in the mouse model of chronic TB: CFU counts after 4 weeks of treatment. Data represent the mean lung log10 CFU counts (5 mice per group). Day 0 indicates lung CFU counts at treatment initiation (28 days after infection). The number following the compound abbreviation indicates the dose in milligrams per kilogram per day. (D) CGM administered in combination with anti-TB drugs during the intensive phase of treatment in a mouse model of TB. Data represent the mean lung log10 CFU counts (5 to 10 mice per group). Drug doses: RIF, 10 mg/kg per day; INH, 10 mg/kg per day; PZA, 150 mg/kg per day; CGM, 100 mg/kg per day. w, weeks. Standard treatment is 8 weeks of daily RIF + INH + PZA, followed by daily RIF + INH. Error bars in all panels represent SD. Dotted lines in (B) and (C) indicate the CFU counts in the lungs when treatment was initiated.

CGM activity in mouse models of TB

To assess the in vivo activity of CGM, we conducted dose-ranging experiments using mouse models of acute and chronic TB (see supplementary materials and methods). In the acute model, which is used to test the antimicrobial activity of compounds and regimens against bacteria actively multiplying in vivo, mice were aerosol-infected with M. tuberculosis, and oral administration of CGM (at daily doses ranging from 10 to 600 mg/kg) was initiated on the day after infection. After 4 weeks, all of the untreated control mice had died, whereas all of the mice treated with any dose of CGM survived. As expected, mice treated with isoniazid (INH) at 10 mg/kg experienced a decrease of ~3 log10 colony-forming units (CFUs) in the lungs, whereas mice receiving RIF at 10 mg/kg, which is expected to have poor initial activity in this model (16), did not die but did experience bacterial growth in the lungs (Fig. 2B). In mice receiving 10 and 25 mg/kg of CGM, neither the development of gross lung lesions (fig. S2) nor bacterial growth could be prevented (Fig. 2B). Total prevention of bacterial growth and gross lung lesions occurred in mice treated with 50 mg/kg, defining the minimal effective dose of CGM. The CFU count decreased by ~2 log10 in the lungs compared with the CFU count at initiation of treatment with a daily 100-mg/kg dose, defining the minimal bactericidal dose. In mice treated with 200-, 400-, and 600-mg/kg doses, a dose-dependent decline in lung CFU counts was observed (P < 0.0001), and the lungs of mice receiving the 600-mg/kg daily dose were culture-negative after 4 weeks of treatment. In the chronic model, which is used to test for antimicrobial activity against a stable bacterial population in vivo, mice were aerosol-infected, achieving a low implantation of 2.21 log10 CFU per lung, and treatment was initiated 4 weeks later when a stable, host-contained infection was established at nearly 7 log10 CFU per lung. After 4 weeks of treatment, all doses of CGM exhibited some degree of activity, with doses of 50 mg/kg and higher resulting in statistically significant differences in lung CFU counts from the untreated control (P ≤ 0.01) (Fig. 2C and fig. S2). Treatment with CGM at 100 mg/kg resulted in a decrease in lung CFU counts similar to that observed with RIF at 10 mg/kg. In both of these experiments, no overt adverse effects of CGM administration were observed in treated mice.

Due to the promising bactericidal activity of CGM when administered as monotherapy, we evaluated the activity of this compound when administered together with first-line anti-TB drugs in the mouse model of TB chemotherapy. Standard TB treatment comprises a 2-month intensive phase of daily RIF, INH, pyrazinamide (PZA), and ethambutol, followed by a 4-month continuation phase of daily RIF and INH. We designed an experiment to assess the bactericidal activity of CGM (at the minimal bactericidal dose of 100 mg/kg per day) used alone and in combination during the first 3 months of treatment. Mice were aerosol-infected with M. tuberculosis (implantation of 3.26 log10 CFU per lung), and treatment was initiated 2 weeks after infection, when the lung CFU count had reached 7.43 log10. In these experimental conditions, CGM alone was as active as INH (the most bactericidal first-line drug), reducing CFU counts in the lungs of mice by 3.78 log10 over 3 months, indicating that on a molar basis, CGM (1196 g/mol) had equivalent potency to INH (137 g/mol). However, at any time point, the combination of INH and CGM was as active as either INH or CGM alone. The combination of CGM with PZA was as active as the standard combination of INH, RIF, and PZA, reducing lung CFU counts in the mice by 5.5 log10 over 3 months (table S3). Additionally, the combination of CGM with RIF was much more active than the standard combination (additional 2 log10 CFU decline), nearly resulting in mouse lung culture conversion by week 12 of treatment. Even more impressive was the activity of the three-drug combination of RIF, PZA, and CGM that resulted in mouse lung culture conversion after only 8 weeks of treatment, representing killing of >7 log10 CFU in the lungs. In contrast and as expected, in the mice receiving the standard drug combination, lung CFU counts declined by ~4.4 and ~5.7 log10 by weeks 8 and 12, respectively (Fig. 2D and table S3). To assess the activity of CGM (at 100 mg/kg per day) during the continuation phase, mice received the standard regimen for the first 8 weeks of treatment, and then CGM was administered alone or in combination during weeks 8 to 20. The bacterial population remaining during the continuation phase had survived the initial 8 weeks of treatment and may have been enriched for organisms that are phenotypically tolerant to the antibiotics (i.e., the so-called “persisters”). The replacement of either of the drugs in the standard continuation regimen (RIF and INH) with CGM did not affect the time to mouse lung culture conversion (20 weeks), and no statistically significant differences were observed in lung CFU counts between these groups at any time point (table S4). The bactericidal activity of the combination of INH with CGM was not significantly different from that of CGM alone at any time point. However, this combination was significantly more bactericidal than INH alone after 16 (P ≤ 0.05) and 20 (P ≤ 0.0001) weeks of treatment, with the effect being additive. Thus, taken together, our data indicate that CGM is highly active against actively replicating M. tuberculosis, both in vitro and in vivo, and also exhibits bactericidal activity against nonreplicating bacteria in vivo.

Self-resistance to GM in Streptomyces

The demonstrated activity of CGM against both drug-susceptible and -resistant M. tuberculosis (14, 15) and the notable enhancement of the bactericidal activity of the standard TB drug regimen with the addition of CGM in mice suggested that the mechanism of action of CGM may be different from other anti-TB drugs. GM and MGM are naturally produced by some species of Streptomyces, and we identified the biosynthetic gene cluster responsible for GM and MGM production in S. sp. DSM-40835 (17), a strain naturally resistant to GM, and examined it for the presence of a possible resistance-conferring component. We found a homolog of the dnaN gene, which encodes the sliding clamp (also called the β clamp) of DNA polymerase, within the nonribosomal peptide megasynthetase cassette for GM synthesis (fig. S3A and table S5). This dnaN homolog, annotated as griR, encoded a protein with 51% identity to DnaN from the same strain (fig. S3, B and C). To examine the effect of GriR on susceptibility to GM, we introduced the gene into Streptomyces coelicolor A3(2), a strain susceptible to this natural product (and naturally lacking the entire GM biosynthesis cassette) (18). The presence of the griR-expressing plasmid allowed the strain to survive in the presence of GM (fig. S3D and table S6), suggesting that overexpression of griR mediated GM resistance. Comparison of the DnaN sequence from different microbial species revealed a clear cluster break between GM-sensitive and -resistant bacteria based on homology of the sliding clamp (fig. S3C and table S6). Taken together, these data suggest that GM and its derivatives target DnaN and may thus interfere with DNA replication.

GM resistance in mycobacteria

The intriguing relationship between dnaN sequence, copy number of the gene, and resistance to GM observed in Streptomyces compelled us to examine the mechanism(s) of resistance to this compound in mycobacteria. Using the fast-growing, nonpathogenic, and GM-sensitive (MIC: 4.5 μg/ml) M. smegmatis, we selected GM-resistant bacteria in vitro. The GM-resistant M. smegmatis (which occurred at an extremely low frequency; at a GM concentration of 10 μg/ml the frequency was 5 × 10−10) exhibited an altered, elongated cellular morphology (Fig. 3A), as has been previously observed in GM-exposed M. tuberculosis (9). The GM-resistant bacteria were cross-resistant to MGM and CGM, but not to RIF or any other tested antibiotic (table S7). Genome sequence comparisons between the wild-type (WT) parent strain and mutants generated via stepwise exposure to increasing concentrations of GM (up to 40 μg/ml) revealed amplification of a chromosomal segment containing dnaN, with a single point mutation in the Pribnow box of the promoter region (115 G > A) (19) (tables S8 and S9), suggesting that mycobacterial resistance to GM, like that of Streptomyces, is mediated by amplification of DnaN. A number of inconsistent single-nucleotide polymorphisms (SNPs) were also observed. All mutants analyzed contained head-to-tail repeats of a chromosomal segment, but these segments or amplicons varied in size (ranging from 12 to 28 kb in length) and copy number (ranging from 3 to 49 copies), and occurrence of the amplicons was observed at different steps in the selection process (Fig. 3B, fig S4, and table S8). In addition to dnaN and its promoter, the amplicons contained dnaA, recF, gyrB, and up to more than 20 additional genes (table S9); they also all included the ori site. Amplification of a chromosomal segment was confirmed by Southern blot (Fig. 3, C and D). Based on the high copy number of the amplicons that include the ori site, formation of an extrachromosomal element could not be ruled out. However, plasmids were not detected by either standard plasmid isolation methods or agarose gel analysis. Yet in both M. smegmatis and M. tuberculosis, plasmids containing the same ori as the chromosome are usually maintained at a relatively low copy number (20, 21), and naturally occurring plasmids in mycobacteria are known to be difficult to isolate (22).

Fig. 3 Emergence of GM resistance in M. smegmatis by target amplification.

(A) Phase-contrast images of WT and GM resistant M. smegmatis cells. Scale bars, 10 μm. (B) Overview of the genetic changes associated with increased exposure to GM for M. smegmatis mutants 1.GM2.5 to 1.GM40 (subscripts indicate GM concentration in micrograms per milliliter when the mutant was obtained). Red stars represent the 115 G > A mutation in the Pribnow box of the dnaN promoter region. (C) Southern blot and (D) fragmentation pattern of EcoRI-digested genomic DNA of WT M. smegmatis and GM-resistant mutants. EcoRI cuts downstream of the dnaN gene inside the amplicon and upstream of the dnaN gene outside the amplicon, resulting in a 15.9-kb fragment for samples without amplicons and 15.9-kb and 10.2-kb fragments for samples with amplicons. EcoRI restriction sites are indicated in red; the binding site of the probe (in the dnaN gene) is indicated in blue.

Although the exact sequence of the amplicons varied among GM-resistant mutants, the 115 G > A transition mutation in the dnaN promoter occurred in all mutants at the 2.5-μg/ml GM exposure step, leading to an MIC upshift from the baseline 4 μg/ml to 8 to 16 μg/ml (table S8). This point mutation was not consistently maintained in all steps of increasing GM concentration, being either replaced or complemented in mutants exposed to higher GM concentrations by amplification of a dnaN-containing chromosomal segment, which led to resistance up to or greater than 64 μg/ml. Resistance to GM was accompanied by considerable fitness costs, whereby M. smegmatis growth was negatively correlated with increasing levels of resistance (fig. S5), in some cases growing so poorly that we could not determine MIC values (table S8). Our data indicate that when GM is removed from the growth media, these genetic and phenotypic modifications begin to be reversed, including a decrease in amplicon copy number, a downshift in MIC, and less growth retardation (fig. S5 and table S8).

The unprecedented mechanism of gaining resistance to GM in our in vitro experiments with M. smegmatis prompted us to determine the relevance of these findings during in vivo infection with M. tuberculosis. To select resistant bacteria in vivo, nude mice were aerosol-infected with the fully drug-susceptible M. tuberculosis strain H37Rv and then treated by monotherapy with either INH (5 mg/kg) or CGM (100 mg/kg). Without the pressure from a functional immune system, monotherapy (when administered once daily, 5 days per week) with even the most bactericidal drug (INH) cannot prevent bacterial multiplication in these mice when drug levels in the blood fall below the therapeutic concentration, which, based on the half-life of INH (~1.5 hours) (23) and CGM (~4 hours) (Table 1) in the mouse, should occur daily and over the weekends. As anticipated, after 4 weeks of treatment, the lung bacterial loads began to increase in mice receiving either INH or CGM (fig. S6A). CFUs isolated from the CGM-treated mice were more than 10 times less sensitive to CGM (MIC: 1 to 2 μg/ml) (see proportion of CFUs growing on CGM-containing agar plates in table S10). Using the agar proportion method, the CGM-resistant M. tuberculosis were found to be fully susceptible to INH and RIF, as well as moxifloxacin and streptomycin. Genomic analysis of CGM-resistant colonies revealed an amplification of a 10.3-kb chromosomal segment containing the dnaN gene, together with four other genes and the ori site (fig. S6B and table S11), which was present in the CGM-resistant M. tuberculosis but absent in the WT parent strain. No SNPs were observed in the CGM-resistant bacteria compared with the WT parent strain. Thus, the mechanism of resistance observed in vitro with M. smegmatis was also observed in vivo with M. tuberculosis.

Characterization and crystal structure analysis of the GM-DnaN interaction

To confirm our genomic-based findings that DnaN is the target of GM and its derivatives, we used surface plasmon resonance (SPR) to characterize the binding of GM, MGM, and CGM with DnaN from M. smegmatis, M. tuberculosis, and Escherichia coli, as well as with the human sliding clamp [proliferating cell nuclear antigen (PCNA)]. Binding was also characterized with GriR, the DnaN homolog from the GM biosynthesis cluster in Streptomyces. SPR analysis demonstrated binding of GM to the mycobacterial sliding clamps with high affinity (equilibrium dissociation constant KD: 8.3 × 10−11 M and 1.0 × 10−10 M for M. smegmatis and M. tuberculosis DnaN proteins, respectively) and a fast recognition rate and slow dissociation from the protein (M. smegmatis, association rate constant ka: 2.2 × 107 M−1 s−1, dissociation rate constant kd: 1.9 × 10−3 s−1; M. tuberculosis, ka: 8.6 × 106 M−1 s−1, kd: 8.4 × 10−4 s−1) (Table 2 and fig. S7). No binding was detected between GM and the human sliding clamp, but binding was observed at a significantly lower level with the E. coli clamp (KD: 6.5 × 10−7 M), despite the lack of in vitro activity of GM against this organism (table S6). This lack of activity could be associated with the fast dissociation of GM from the E. coli DnaN compared with the mycobacterial sliding clamps (fig. S7), or it may be caused by efflux and/or impaired penetration in E. coli. Binding of GM was also observed to the GM resistance-conferring GriR, but at a significantly lower level compared with the mycobacterial DnaN proteins. It is possible that this weaker binding may be sufficient for self-resistance in Streptomyces. Binding of GM to GriR showed a kinetic profile similar to that of the E. coli DnaN, with fast association and dissociation rates but with somewhat lower affinity. MGM and CGM exhibited kinetic profiles similar to GM when interacting with the M. tuberculosis sliding clamp, whereas the interaction of M. smegmatis DnaN with CGM was characterized by a slower dissociation rate (kd: 4.1 × 10−4 s−1) and a slightly slower recognition rate (ka: 4.5 × 106 M−1 s−1). Thus, it may be the increasing lipophilicity (thus influencing the ability to traverse the cell wall) of the GM derivatives, rather than changes in binding kinetics, that explains their more potent antimycobacterial activity compared with GM.

Table 2 SPR-based kinetic parameters of sliding clamp interactions with GM, MGM, and CGM.

For M. smegmatis and M. tuberculosis DnaN, the KD values were determined from the ratio between the kinetic rate constants (ka/kd), and the dissociative half-life t1/2 was calculated by ln2/kd. For interactions with fast on and off rates (E. coli and S. sp. DSM-40835 DnaN proteins), KD values were determined by steady-state affinity analyses from the dependence of steady-state binding levels on analyte concentrations. Data represent mean and SD from three independent experiments. SPR sensorgrams are presented in fig. S7. na, not applicable.

View this table:

To gain a better understanding of the molecular interactions responsible for GM-DnaN complex formation, cocrystal structures of GM and CGM bound to mycobacterial DnaN proteins were determined. The structures of M. smegmatis DnaN with GM and CGM were refined to resolutions of 2.1 and 2.3 Å, respectively, whereas the structures of the M. tuberculosis DnaN with bound GM and CGM were refined to resolutions of 2.2 and 1.9 Å, respectively (table S12). Both inhibitors were well defined in the electron density maps (fig. S8). GM and CGM bound to a hydrophobic cleft between domains II and III of the sliding clamp (Fig. 4A), which has been identified previously as the peptide interaction site responsible for the binding of DNA polymerases and other DNA-modifying enzymes (24, 25). Binding of GM is mainly accomplished via hydrophobic interactions. Only two amino acid residues are involved in direct hydrogen bonding: Arg181 (or Arg183) for M. smegmatis (or M. tuberculosis) forms two hydrogen bonds to the ligand, one via its backbone carbonyl oxygen and one via its guanidine moiety, whereas Arg394 (or Arg399) for M. smegmatis (or M. tuberculosis) is involved in hydrogen bonding through its backbone amide nitrogen. Additionally, a bridging water molecule is coordinated between the ligand and Pro392 (or Pro397) for M. smegmatis (or M. tuberculosis) (Fig. 4B and fig. S9). Subsite one of the peptide-binding pocket is occupied by the cyclic part of the ligand, whereas subsite two harbors the linear part of the GM molecule (Fig. 4C). This linear section and the adjacent half of the macrocycle superimpose very well for GM and CGM, whereas differences are visible on the opposite side of the macrocycle (figs. S10 and S11A). The ligands seem to be less tightly bound in this area, thus exhibiting a higher degree of flexibility (fig. S11B). The additional cyclohexyl moiety of CGM protrudes into the solvent and is involved in the formation of crystal contacts (fig. S12); the crystal packing might induce the observed differences in CGM binding as compared with GM and thus may not be physiologically relevant.

Fig. 4 Crystal structure of the M. smegmatis sliding clamp (DnaN) in complex with GM.

(A) Crystal structure of the M. smegmatis DnaN dimer in complex with GM at a resolution of 2.1 Å. GM binds to a hydrophobic pocket between domains II and III, which is known as the protein-protein interaction site responsible for the recruitment of DNA polymerases by the sliding clamp. The cartoon representation of the left half of the homodimeric ring is colored from the N terminus (blue) to the C terminus (red). The second half of the ring is shown as surface representation colored according to the electrostatic surface potential, ranging from –5 kT/e (red) to +5 kT/e (blue) (k, Boltzmann’s constant; T, temperature; e, charge of an electron). (B) Interactions between the ligand and the protein. The residues involved in hydrophobic contacts with GM are represented as gray sticks. The residues also involved in hydrogen bonding with the ligand are represented as green sticks. Hydrogen bonds are indicated by dashed green lines. F, Phe; R, Arg; M, Met; P, Pro; T, Thr; L, Leu; V, Val. (C) Surface representation of the binding pocket.[pe] The surface is colored according to the electrostatic surface potential. Both subsites of the binding pocket are indicated by circles.

Binding of GM to the peptide interaction site of DnaN should lead to inhibition of the DnaN interaction with the polymerase III α subunit (DnaE1). To test this, we used SPR to assess the effect of GM on the binding of DnaN with a DnaE1 peptide that contains the internal DnaN binding motif and is known to be essential for processive replication in E. coli (26). Binding of this peptide decreased by half when the sliding clamp was saturated with GM (fig. S13). Inhibition of DnaE1 binding to DnaN should result in inefficient replication with decreased processivity and may lead to DNA strand breaks. Induction of the SOS response was observed in GM-exposed M. smegmatis (fig. S14).

Several functional and structural studies have indicated that the sliding clamp may be a feasible antibacterial target that can be addressed by small molecules and peptidic inhibitors (24, 2729). Although inhibition of DNA replication does not necessarily lead to immediate bacterial cell death, blocking the replication process may induce other lethal damage within the cell, such as induction of the SOS response (fig. S14) and/or toxin-antitoxin–like stress, as in the case of the quinolones, which block DNA replication by targeting topoisomerases but kill bacteria rapidly (30). In fact, DnaN was recently shown to be targeted by a toxin-antitoxin system in Caulobacter crescentus (31). The cell-elongation phenotype of GM-exposed mycobacteria indicates that blocking DNA replication induces a wider cellular response; this cell-elongation effect has also been reported in M. fortuitum upon exposure to the fluoroquinolone ofloxacin (32). Furthermore, in both Gram-positive and Gram-negative bacteria, the sliding clamp has been demonstrated to interact with DNA repair proteins, including many induced during the SOS response (33). Thus, the interaction of GMs with DnaN could also affect DNA repair, contributing to the bactericidal activity of the compounds. Whereas previously reported DnaN binders inhibited bacterial growth with only moderate potency, GM and its derivatives represent the first DnaN inhibitors with picomolar affinity to the target and high efficacy (MIC values ≤1.0 μg/ml) against mycobacteria. Known small-molecule DnaN binders targeted the deep hydrophobic pocket of subsite one, whereas subsite two remained empty. Peptidic inhibitors and the natural interacting peptides target both subsites; the linear segment and part of the macrocycle of GM superimposed very well with their backbone, thereby mimicking the conformation of a linear peptide bound to DnaN (fig. S15). The specific inhibition of mycobacteria by GM, which should leave commensal microbiota intact, constitutes an additional advantage with respect to the selection of resistance.

Conclusion

Total synthesis of new GMs led to the optimized CGM, which demonstrated a clear potential for TB treatment by targeting the DNA polymerase sliding clamp. Currently, CGM and a few other GM derivatives are being further profiled to select the best drug candidate to move forward in development, with CGM being one of the lead compounds pursued in late preclinical studies involving broad toxicity testing. In this work, we have demonstrated the validity of DnaN as a drug target and have identified direct small-molecule–DnaN interactions that probably interrupt essential polymerase and DNA repair activities in mycobacteria, leading to killing both in vitro and in vivo. Through the selection of GM- or CGM-resistant bacteria, we discovered that resistance to these compounds is possible, albeit at a very low frequency, and is mediated through multiplication of dnaN, whereby the resistance to GMs increases with increasing dnaN amplification. The nature of the amplicons, consisting of clonal repeats that differ slightly between bacterial isolates but all include the ori site, and the observed decrease in target amplification in the absence of GM suggest that the amplification is connected to severe fitness costs, as observed by the slow growth of the mutants. As exposure to GM triggered chromosomal duplications in a concentration-dependent manner in vitro, GM could possibly serve as a tool to study the mechanism of chromosomal duplication events in mycobacteria that have contributed to the natural variation between different lineages. Because there is no preexisting resistance to GMs due to the different mode of action compared with that of existing TB drugs, and because resistance occurs at an extremely low frequency and is associated with a severe fitness cost, this new series has the potential to contribute to drug regimens with utility for patients with both drug-sensitive and drug-resistant TB. In addition, our work has confirmed the sliding clamp as an antimicrobial target and revealed an unusual mechanism of conferring drug resistance that is applicable to the wider antibiotic discovery field.

Supplementary Materials

www.sciencemag.org/content/348/6239/1106/suppl/DC1

Materials and Methods

Supplementary Text

Figs. S1 to S15

Tables S1 to S12

References (3476)

References and Notes

  1. Acknowledgments: We thank the staff at beamlines P11 at PETRA III (DESY, Hamburg, Germany), X06DA at the Swiss Light Source (PSI, Villigen, Switzerland), and BL14.1 at BESSY II (Berlin, Germany) for assistance during data collection. R.M.’s research laboratory was supported by the German Ministry for Education and Research (BMBF) with grant fZ031598. Sanofi funded its contribution. D.V.A., N.C.A., and J.H.G. are supported by the KwaZulu-Natal Research Institute for Tuberculosis and HIV and the Howard Hughes Medical Institute. D.V.A., S.T., E.N., and J.H.G. were supported by two grants from Sanofi-Aventis (Amd. 14, grant 111937, and Amd. 7, grant 103671). We thank C. Wylegalla for technical assistance. We acknowledge the GMAK group at HZI for assistance with genome sequencing of bacterial mutants; S. Franzblau from the University of Illinois at Chicago for performing the in vitro MIC assays both under hypoxic conditions and with different strains of M. tuberculosis (testing agreement) and for the gift of luciferase-expressing M. tuberculosis; and P. Brodin from the Pasteur Institute, Lille, France, for the gift of GFP-expressing M. tuberculosis H37Rv. We thank A. Upton from the TB Alliance for analysis of the manuscript. The sequence of the Streptomyces sp. DSM-40835 GM biosynthetic gene cluster has been deposited in GenBank (accession number KP211414). Atomic coordinates and structure factors for the reported crystal structures have been deposited in the Protein Data Bank (identification numbers listed in table S12). A.B., E.F., S.S., M.B.B., F.B.-P., M.B., G.C., M.G., P.H., S.K., C.L., H.M., C.K., K.B., M.Ger., H.H., M.Ko., M.Ku., L.F., and S.L. are employed by Sanofi-Aventis R&D. Sanofi has filed patent applications on GM derivatives. Author contributions: A.K. performed resistance studies and DnaN expression and physicochemical analysis of the interaction of the protein with GM derivatives. P.L. performed expression of M. smegmatis DnaN and all structural biology experiments. P.L. and D.W.H. were involved in structural analyses and data interpretation. P.H. and F.B.P. identified the GM series from the Rhône-Poulenc archives. G.C., M.Ges., H.H., A.B., and E.F. designed and synthesized all GM derivatives. S.S., M.B.B., E.F., and A.B. managed the drug optimization program. S.L., M.B., and L.F. supervised the overall GM program at Sanofi. C.L. and M.B.B. designed and performed the in vitro experiments with M. tuberculosis H37Rv. S.K. and M.Ko. designed and performed the pharmacokinetic evaluations. M.Ko. and K.B. elucidated the metabolic degradation of GM and its derivatives. M.Ger. and M.Ku. performed analytics and structure confirmation of GM and its derivatives. M.Ku. and H.M. performed conformational analysis and designed models for GM derivatives with improved pharmacokinetic properties. D.V.A., N.C.A., E.N., S.T., and J.H.G. designed and performed all animal infection experiments, including associated assays, and analyzed all resulting data. N.Z. analyzed all genome data. J.H. performed activity assays and identified and analyzed resistant M. smegmatis strains. S.C.W. and C.K. identified and characterized the GM biosynthetic gene cluster. A.K., P.L., A.B., N.C.A., S.L., M.B., and R.M. conceived the studies and wrote the paper. A.K. and P.L. contributed equally to the study. All authors discussed the results and commented on the manuscript.
View Abstract

Navigate This Article