In Vitro Activity of Novel Rifamycins against Rifamycin-Resistant Staphylococcus aureus
http://www.100md.com
《抗菌试剂及化学方法》
ABSTRACT
We describe novel rifamycin derivatives (new chemical entities [NCEs]) that retain significant activity against a comprehensive collection of Staphylococcus aureus strains that are resistant to rifamycins. This collection of resistant strains contains 21 of the 26 known single-amino-acid alterations in RpoB, the target of rifamycins. Some NCEs also demonstrated a lower frequency of resistance development than rifampin and rifalazil in S. aureus as measured in a resistance emergence test. When assayed for activity against the strongest rifamycin-resistant mutants, several NCEs had MICs of 2 μg/ml, in contrast to MICs of rifampin and rifalazil, which were 512 μg/ml for the same strains. The properties of these NCEs therefore demonstrate a significant improvement over those of earlier rifamycins, which have been limited primarily to combination therapy due to resistance development, and suggest a potential use of these NCEs for monotherapy in several clinical indications.
INTRODUCTION
ifamycins, such as rifampin (RIF), rifabutin (RFB), and rifapentine, inhibit bacterial RNA polymerases isolated from a wide variety of microorganisms (4, 8, 9, 10, 21, 37, 38). This class of drugs has been used in the clinic as one component of multiple drug therapy, predominantly to treat tuberculosis (27) and to treat serious gram-positive infections (15, 32, 48). More recently, rifalazil (RFZ) [3'-hydroxy-5'-(4-isobutyl-1-piperazinyl) benzoxazinorifamycin], also known in the literature as KRM-1648 or ABI-1648, was used in phase 2 clinical trials in a multidrug regimen for the treatment of tuberculosis (6, 26).
Clinically, rifamycins are used mainly in combination therapies because of their propensity to select for resistant mutants when used as single agents (1, 42, 44). Both in vivo and in vitro, this resistance is caused by mutations in the rpoB gene, which encodes the target of rifamycins, the subunit of RNA polymerase (7, 14, 18, 27, 45, 47). The amino acid residues in the RpoB subunit of Thermus aquaticus that interact with RIF have been revealed by X-ray crystallography studies (3). Single-nucleotide changes in rpoB that result in alterations of the amino acids involved in rifamycin binding (the rifampin resistance-determining region [RRDR]) are sufficient to confer high-level resistance to these drugs (27). Resistance-conferring mutants have been well characterized for Mycobacterium tuberculosis and Staphylococcus aureus, among other bacteria (Fig. 1). The high concentration of drug required to kill such mutants is not achievable at the site of infection, and the result is therapeutic failure. Considering the rifamycin resistance mutation frequency of approximately 10–6 to 10–9 per cell (16, 24, 29, 41), clinical resistance development is exacerbated by high-bioburden infections.
ifalazil is unique among rifamycins in that is has good activity against RIF-resistant mutants of Chlamydia spp. Chlamydia trachomatis and Chlamydia pneumoniae mutants that are highly resistant to RIF are susceptible to RFZ (17, 33). Consistent with this finding, RFZ monotherapy was successful in eradicating C. trachomatis in subjects treated for nongonococcal urethritis in a recent phase 2 human clinical trial (B. Batteiger, W. McCormack, W. Stamm, and the Rifalazil Study Group, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. L-992b, 2004). However, the evaluation of pathogenic clinical isolates of other species, including both M. tuberculosis (19, 26, 44) and S. aureus (40), indicates that RFZ shares with RIF the loss of activity against strains that have resistance-conferring alterations in the RRDR.
One possible avenue for the broader use of rifamycins as monotherapeutic agents against pathogens other than Chlamydia is the creation of rifamycins with good activity against known rifamycin-resistant mutants of other pathogenic bacteria. This strategy assumes that there are no additional mutable sites in the RpoB protein that would confer high-level resistance to new, more potent rifamycins. This assumption is plausible because the interaction between the ansa chain of rifamycins and the RpoB protein occurs through amino acid residues that are highly conserved in most bacterial species (3, 20).
In searching for novel rifamycins that overcome resistance to RIF, we chose S. aureus as the test organism, because it is an important and prevalent pathogen for which there is a growing need for improved antibacterial treatments (30) and because both laboratory-derived and clinically derived rifamycin-resistant mutants of S. aureus containing mutations in rpoB have been previously described (referenced in the legend for Fig. 1).
We have synthesized over 700 novel benzoxazinorifamycins (new chemical entities [NCEs]) (34, 35, 36) and screened them for antibacterial activity, propensity to elicit resistant mutants, and activity against rifamycin-resistant mutants of S. aureus. NCEs that exhibited diminished resistance development and greatly enhanced activity against rifamycin-resistant strains of S. aureus compared to RFZ and RIF were identified. These novel compounds have the potential to be used in monotherapy more broadly than have other rifamycins.
MATERIALS AND METHODS
Bacterial strains. Wild-type S. aureus strain 8325--4 and mutated derivatives were obtained from Leeds University (I. Chopra) and have been described previously (23). RIF-resistant and wild-type S. aureus W59532 strains were obtained from Ordway Research Institute, Albany, NY, and characterized as described below. Rifamycin-resistant derivatives of S. aureus ATCC 29213 and S. aureus Smith were isolated and characterized at ActivBiotics, Inc., as outlined below. Strains were stored in 20% glycerol-cation-adjusted Mueller-Hinton II (CAM) broth (Becton Dickinson, Sparks, MD).
Materials. Rifampin B was purchased from Sigma Chemical Co. (St. Louis, MO). Rifalazil [3'-hydroxy-5'-(4-isobutyl-1-piperazinyl) benzoxazinorifamycin] and NCEs were provided by ActivBiotics, Inc. (Lexington, MA). Rifabutin was obtained from U.S. Pharmacopeia (Rockville, MD). RFZ and NCE structures are shown in Fig. 2. CAM agar was from Becton Dickinson (Sparks, MD).
Isolation of rifamycin-resistant mutants of S. aureus and determination of resistance frequency. Rifamycin-resistant mutants of S. aureus strains ATCC 29213 and Smith were isolated by inoculating 5 x 109 bacteria from independent single cultures onto CAM agar plates containing the rifamycin of interest. Rifamycin concentrations of 1 and 0.03 μg/ml were used. Plates were incubated at 35°C for 24 h. One resistant mutant per independent culture was purified and subjected to further characterization as outlined below. The total rifamycin-resistant and -sensitive colonies per plate were also counted and resistance frequencies calculated.
Antimicrobial susceptibility testing. MICs of compounds were determined by the agar dilution method or the broth microdilution method according to the Clinical and Laboratory Standards Institute (formerly the National Committee for Clinical Laboratory Standards) (22). For broth microdilution, serial twofold dilutions were carried out in CAM broth in microtiter plates. Subsequent dilutions were mixed four times with a multichannel pipettor, and pipette tips were changed every 3 dilutions to ensure minimal carryover of these hydrophobic compounds. The trays were covered with Mylar plate sealing sheets (ThermoLabsystems, Franklin, MA) and incubated at 37°C for 16 to 20 h. The MIC was the lowest concentration of compound that resulted in no detectable growth by visual inspection.
esistance emergence test. In order to assess the relative emergence of resistance, a stock seed culture (stored in 20% glycerol-CAM broth at –80°C) containing 108 cells of S. aureus strain ATCC 29213 was used to inoculate 50 ml of CAM broth and grown to an optical density at 600 nm of 1.0. A portion of this culture equivalent to 5 x 109 cells was spread on 150-mm CAM agar plates containing either 1 μg/ml or 0.3 μg/ml of the compound to be tested. Plates were incubated at 35°C for 24 h and resistant colonies counted. NCEs were scored as follows: for 1-μg/ml plates, more than 10 colonies per plate scored "0," 1 to 10 colonies per plate scored "1," and no colonies scored "2." Fewer than 10 colonies on the 0.03-μg/ml plate scored "3."
Characterization of mutations in the rpoB gene. Total DNA from resistant S. aureus strains was isolated using a QIAamp DNA mini kit (QIAGEN, Valencia, CA) and used as a template for amplification of specific regions of the rpoB gene by PCR. Oligodeoxyribonucleotide primers (Table 1) were used to amplify the DNA encoding amino acid residues 418 to 594 of the RpoB protein. These residues encompass the RRDR of the RpoB protein. For some strains, the DNA segments comprising the entire rpoB coding region were similarly amplified. The resulting amplification products were isolated after gel electrophoresis, purified using a QIAGEN gel extraction kit according to the manufacturer's instructions (QIAGEN, Inc., Valencia, CA), and sequenced at the Tufts Core Sequencing facility (Tufts University, Boston, MA).
ESULTS
ationale for analysis of rifampin resistance. Although RIF and RFZ possess excellent antibacterial activity against a broad range of bacterial species, they readily elicit highly resistant mutants (19, 26, 40, 44). We synthesized a library of novel benzoxazinorifamycins (NCEs) related to RFZ with the goal of identifying those with improved activity against rifamycin-resistant bacteria compared with RIF or RFZ. In order to evaluate activity against rifamycin-resistant mutants, two methods were employed. In one, a comprehensive collection of rifamycin-resistant mutants of S. aureus ATCC 29213 was isolated and characterized with respect to rpoB gene sequence. As a part of the process of assembling this otherwise isogenic mutant collection, NCEs, RIF, and RFZ were utilized as selective agents, providing an opportunity to determine whether mutations conferring resistance vary as a function of the selective agent. Once the comprehensive panel of mutants was assembled, the antibacterial activities of NCEs against these mutants were determined. During the process of assembly and characterization of the panel of rifamycin-resistant mutants, we instituted a second, technically simpler method, the resistance emergence test. In this approach, the frequency of emergence of rifamycin-resistant mutants from S. aureus ATCC 29213 cultures was determined directly by inoculating media containing NCEs with 5 x 109 CFU of the strain and then enumerating resistant colonies that emerged. A diagrammatic description of the screening flow and numbers of NCEs that were subjected to each method is shown in Fig. 3.
Comprehensive collection of rifamycin-resistant S. aureus strains. S. aureus ATCC 29213 or Smith isolates were inoculated onto CAM agar plates containing RIF, RFZ, or one of six NCEs at various concentrations as described in Materials and Methods. Resistant strains were isolated, and the nucleotide sequence of their rpoB genes was determined. The results from one representative experiment described in Table 2 show that the mutants obtained from ABI-0043, ABI-0094, and ABI-0299 selections contained mutations identical to mutations derived from RIF or RFZ selections (Table 2) and mutations described previously in the literature (see the legend for Fig. 1 for references). This same result was obtained when other NCEs were the selective agents (ABI-306, ABI-418, and ABI-420 [data not shown]). In all, over 200 such independently isolated mutants were characterized by sequence analysis of their RRDRs. As the isolation and analysis of resistant mutants progressed, no additional mutations were identified, suggesting genetic saturation under the conditions used. To augment our own rifamycin-resistant mutant collection with mutants conferring weaker resistance, we added to the panel a series of well-characterized strains derived from S. aureus 8325-4 (23). Finally, we also characterized a set of RIF-resistant mutants of S. aureus W59532 in anticipation of future testing of NCEs in animal models of infection. The latter strain background has been used successfully in such models (A. Louie, W. Liu, M. R. Deziel, M. Drusano, L. Turner, T. Gumbo, and G. L. Drusano, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. A-1297, 2004).
ifamycin susceptibility of the rifamycin-resistant S. aureus collection. The panel of resistant S. aureus mutants was further analyzed for susceptibility to rifamycins, including RIF, RFZ, and several NCEs (Table 3). RIF, RFB, and RFZ had identical MICs for wild-type S. aureus 29213. Rifamycin-resistant mutants fell into two classes, those that conferred high-level resistance to RIF (MIC of 64 μg/ml) and those that conferred moderate to low-level resistance to RIF (MIC of between 0.125 and 16 μg/ml) (Table 3). For low-level RIF-resistant mutants, the MICs of RFZ were between 4- and 64-fold lower than the MICs of RIF. Despite improved activity against this class of mutants, RFZ did not have better antibacterial activity against strains containing mutations that conferred high-level RIF resistance, except for the strain containing the A477D change (Table 3). These results are consistent with previously published results (40). In marked contrast, several RFZ-based NCEs had antibacterial activity that was up to 256-fold better than that of RIF against strongly RIF-resistant mutants (Table 3). Indeed, for the best NCEs, the maximum MIC observed for any member of the mutant collection was 2 to 4 μg/ml (Table 3) (see below). In general, particular mutations in different strain backgrounds all had similar susceptibilities to NCEs and rifamycins (within a twofold concentration). The few instances where rpoB mutations conferred four- to eightfold different susceptibilities to different strain backgrounds are denoted in Table 3 as separate MIC data lines. For instance, the A477D alteration imparted fourfold differences in the MICs of ABI-0418 and ABI-0420 depending on the S. aureus strain background.
Analysis of the antibacterial activity of NCEs against low-level RIF-resistant mutants uncovered a structure-activity relationship. Pairwise comparison of NCEs that differed only at position 25 of their ansa chains (data not shown) revealed that NCEs containing an O-acetyl group at this position had moderately lower (four- to eightfold) MICs against the low-level-resistant mutants than RFZ. Corresponding NCEs that contained a hydroxyl group at position 25 had roughly the same MIC against these mutants as RFZ.
To facilitate screening of the large number of NCEs in this study, a panel that consisted of representative highly RIF-resistant mutants, including alleles that had been previously identified from clinical S. aureus isolates, was assembled (H481Y, Q468K, S486L, and H481D) (1, 41). Good activity against these mutants correlated well with activity against the other highly RIF-resistant mutants in the complete panel (Table 3). Also included in the screening panel were two strains with D471Y and S464P changes in RpoB that conferred weaker RIF resistance. Of the 431 NCEs assayed for activity against the screening panel, 260 retained good activity, with MICs less than or equal to 8 μg/ml for each strain in the panel. Of these, 199 had maximum MICs of 4 μg/ml or less against screening panel strains (Fig. 3).
esistance emergence analysis. A second approach to identifying NCEs with good antibacterial potency against rifamycin-resistant mutants consisted of subjecting 758 NCEs to a resistance emergence test. S. aureus cells (109) were applied to agar medium containing either 1 or 0.3 μg/ml NCE, as described in Materials and Methods. A score of "0" in this assay indicates similarity to RFZ in terms of capacity to elicit resistant mutants, whereas scores of "1," "2," or "3" indicate progressive improvement in the ability to suppress the emergence of resistant mutants (see Materials and Methods) (Fig. 3).
Six hundred fifty-four NCEs (86.2%) scored "0" in this test, that is, they were as likely to elicit rifamycin-resistant mutants as RFZ. Forty-six NCEs (6.1%) scored "1," 55 NCEs (7.3%) scored "2," and 3 NCEs (0.4%) scored "3" (Fig. 3). The 13.4% of total NCEs that scored above "0" therefore represent a class of rifamycins that are less prone to resistance emergence. Of this group, all 3 NCEs that scored "3" and 45 of 55 NCEs that scored "2" had MICs against members of the mutant screening panel of 4 μg/ml. Of the 46 NCEs that scored "1," 39 had MICs against members of the mutant screening panel of 4 μg/ml. Thus, the resistance emergence test was a rapid and valid method for identifying NCEs that were potent against rifamycin-resistant strains. The structures of ABI-0418, ABI-0597, and ABI-0720, which scored "3" in the resistance emergence test, are shown in Fig. 2.
esistance emergence and its relation to structure at position 25. Analysis of resistance emergence data for all NCEs tested revealed that there was a distinct difference in performance of NCEs that depended on the chemical group at position 25 of the ansa chain. Those NCEs that were O-acetylated at this position scored better for resistance emergence than their hydroxylated counterparts. This phenomenon is most clearly illustrated when considering the group of 199 NCEs with high activity (MIC 4 μg/ml) against the mutant screening panel. Thirty-two of 119 (27%) 25-O-acetyl NCEs in this group had a resistance emergence score of 2, and an additional 23 compounds (19%) scored 1. For NCEs in this high-activity group that were hydroxylated at position 25, only 7 of 80 (9%) scored 2 and 13 (16%) scored 1.
Contradictions between resistance emergence score and MIC testing against RIF-resistant mutants. Several NCEs that suppressed the emergence of resistant mutants (resistance emergence score of "2") had MICs against resistant mutants in the screening panel that were relatively high. For example, ABI-0299, when present in the agar medium used to assess resistance emergence at 1 μg/ml, did not give rise to any resistant colonies. One might assume from this result that the MIC of ABI-0299 for any rifamycin-resistant mutants in our collection would be equal to or lower than 1 μg/ml. However, the MIC of ABI-0299 for the most resistant strains in the mutant screening panel was in the 2- to 4-μg/ml range (Table 3). One possible explanation for this discrepancy is that there can be a difference between MICs for NCEs tested by the agar dilution method and MICs for NCEs tested by the broth microdilution method. To determine whether this was the case, five representative NCEs were retested by these two methods (Table 4). For four of the five NCEs tested, including ABI-0299, MICs for the most resistant mutants in the screening panel were between two- and fourfold lower when tested by the agar dilution method (Table 4). This result suggests that NCEs can be more potent in an agar milieu than in broth and explains why some NCEs prevent growth of resistant mutants at 1 μg/ml in agar medium.
DISCUSSION
In this study we have identified novel rifamycins that have both the ability to suppress the emergence of rifamycin-resistant mutants and increased activity against mutants resistant to other rifamycins. This advance is important because the NCEs described here have the potential to be used as monotherapeutic agents; historically, rifamycins have been used mainly in the clinic as a part of combination antibiotic therapy due to the problem of resistance development (1, 12, 25, 42, 43).
We utilized two independent methods for evaluating NCEs for susceptibility to resistance development (Fig. 3). The resistance emergence assay is a rapid two-plate test that estimates the resistance frequency at fixed high concentrations of NCE (roughly 30- and 500-fold higher than their MICs). We used this method to screen a large number of NCEs and identify those that are improved, compared with RIF, with respect to resistance emergence. When the same NCEs were tested for resistance development at lower concentrations relative to their MICs, i.e., at less than 10x the MICs of NCEs, resistance frequencies were similar to those observed with RIF and RFZ: approximately 10–8 per cell per generation (data not shown).
The second, more labor-intensive screening method was susceptibility testing using a rifamycin-resistant mutant panel. In order to ensure a complete panel, we performed independent in vitro selections using RIF, RFZ, and six different NCEs as selective agents. It was significant that no new resistant alleles of RpoB were found despite an exhaustive search using the six NCEs as selective agents. Thus, resistance to NCEs is primarily confined to the 19 alleles derived from our selections that were found repeatedly in our analysis and that have been described in previous reports in which rifampin was the selective agent. The fact that virtually all of the published single-nucleotide mutations that confer high rifamycin resistance in clinical isolates were represented (1, 41) (Fig. 1) indicates that our panel was comprehensive and relevant. We also supplemented the collection with nine strains from the Chopra group (23), including two weakly resistant mutants. In the context of our panel, these mutants, containing L466S or D471E changes in RpoB, were in fact the least resistant to rifamycins, with MICs of RIF equal to 0.125 μg/ml (Table 3). Because we did not characterize mutants from selections at low levels of RIF (0.03 μg/ml), it is not surprising that we did not isolate these two alleles.
Our screens identified NCEs that fell into two classes: (i) NCEs (n = 199) with activity (MIC of 4 μg/ml) against rifamycin-resistant S. aureus and (ii) NCEs (n = 104) that suppressed the emergence of rifamycin-resistant S. aureus strains, most of which were also members of the first class. Many members of the first class, however, failed to suppress resistance emergence. The most likely explanation for this failure, given that no new RpoB alleles were identified, is that these NCEs are less active in the agar-based medium used in the resistance emergence assay than NCEs that scored better than RFZ.
A structure-activity relationship emerged from the analysis of NCEs based on the 25 position of the ansa chain. Although there was no clear advantage in activity against wild-type S. aureus for NCEs having an O-acetyl group or hydroxyl moiety at position 25, the NCEs which contained a 25-hydroxyl had 4- to 32-fold less antibacterial activity than their 25-O-acetyl counterparts when assayed against weakly resistant mutants in the panel (MIC of 16 μg/ml for RIF) (Table 3). This phenomenon was observed for an additional 30 pairs of molecules differing only at position 25 (data not shown). One possible explanation is that NCEs lacking an O-acetyl group at the 25 position have one less binding site in the RRDR and thus are weaker binders that are more susceptible to mutations that lead to normally weak effects for O-acetylated compounds. In fact, X-ray crystallographic data from RIF bound to Thermus aquaticus RNA polymerase define a hydrogen bond between the oxygen portion of the 25-O-acetyl group and a conserved Phe residue within the RRDR which would be lost for 25-hydroxyl NCEs (3). We are currently in the process of testing this hypothesis by determining the 50% inhibitory concentrations of 25-O-acetyl/hydroxyl pairs of NCEs against isolated mutant RNA polymerase. It is curious that NCEs had similar MICs against the strongest resistant mutants (Table 3), despite the fact that 25-hydroxyl NCEs had higher MICs against the weaker mutant class.
This superior performance of the 25-O-acetyl NCEs compared with the 25-hydroxyl NCEs was also apparent when the 34 pairs of NCEs differing only at position 25 were analyzed for resistance emergence. Twenty-eight 25-O-acetyl NCEs had a higher score than their cognate hydroxyl NCEs; only three 25-hydroxyl NCEs had a higher score, and three NCEs had the same score (data not shown). These differences could be negated if hydrolysis of the 25-O-acteylated NCEs occurred readily in vivo to yield 25-hydroxylated NCEs. However, the stability in humans of RFZ (data not shown), a 25-O-acetylated benzoxazinorifamycin, suggests that the in vitro differences described above may be indicative of their in vivo performance.
Operationally, each of the two methods used to identify NCEs with good activity against RIF-resistant S. aureus mutants had advantages. The resistance emergence test was simple and rapid and reliably identified a subset of NCEs with lower MICs against the mutant panel. However, we included susceptibility testing against a panel of resistant strains because (i) it was important to assemble a comprehensive mutant panel using NCEs, in addition to RIF, as selective agents, (ii) the resistance emergence test could select for NCEs specifically having enhanced activity in agar medium compared with that in liquid medium (Table 4), and (iii) the NCEs (especially 25-hydroxyl NCEs) that have resistance emergence scores of 0 but good MICs (4 μg/ml) against rifamycin-resistant mutants may merit further testing, for example, in animal models of infection. In any case, our methods have proved valuable in evaluating NCEs for the very important property of having increased activity against resistant strains, especially because monotherapy in the clinical setting is a goal of our program.
In evaluating the attractive NCEs defined in this study for potential clinical use, additional information provided by measuring efficacy against mutant strains in an in vivo setting would be valuable. To consider using these NCEs successfully as monotherapies, it will be essential to show efficacy against the resistant-mutant subpopulations in addition to simply killing or preventing growth of the wild-type strain.
ACKNOWLEDGMENTS
We thank Sam Bettis, Kara Brown, Debra Buxton, Courtney Calabria-Tanzi, and Julie Ormsby for technical assistance and Ian Chopra and George Drusano for bacterial strains. We also thank Linc Sonenshein and Andrew Wright for helpful discussions and for critically reading the manuscript.
EFERENCES
Aubry-Damon, H., C.-J. Soussy, and P. Courvalin. 1998. Characterization of mutations in the rpoB gene that confer rifampin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 42:2590-2594.
Billington, O. J., T. D. McHugh, and S. H. Gillespie. 1999. Physiological cost of rifampin resistance induced in vitro in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 43:1866-1869.
Campbell, E. A., N. Korzheva, A. Mustaev, K. Murakami, S. Nair, A. Goldfarb, and S. A. Darst. 2001. Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104:901-912.
Chazan, L. L., and S. T. Bayley. 1973. Some properties of a DNA-dependent RNA polymerase from Halobacterium cutirubrum. Can. J. Biochem. 51:1297-1304.[Medline]
Coll, P., L. M. Aragon, F. Alcaide, M. Espasa, M. Garrigo, J. Gonzalez, J. M. Manterola, P. Orus, and M. Salvado. 2005. Molecular analysis of isoniazid and rifampin resistance in Mycobacterium tuberculosis isolates recovered from Barcelona. Microb. Drug Resist. 11:107-114.[CrossRef][Medline]
Dietze, R., L. Teixeira, L. M. Rocha, M. Palaci, J. L. Johnson, C. Wells, L. Rose, K. Eisenach, and J. J. Ellner. 2001. Safety and bactericidal activity of rifalazil in patients with pulmonary tuberculosis. Antimicrob. Agents Chemother. 45:1972-1976.
Dreses-Werringloer, U., I. Padubrin, L. Kohler, and A. P. Hudson. 2003. Detection of nucleotide variability in rpoB in both rifampin-sensitive and rifampin-resistant strains of Chlamydia trachomatis. Antimicrob. Agents Chemother. 47:2316-2318.
Fujii, K., A. Tsuji, S. Miyazaki, K. Yamaguchi, and S. Goto. 1994. In vitro and in vivo antibacterial activities of KRM-1648 and KRM-1657, new rifamycin derivatives. Antimicrob. Agents Chemother. 38:1118-1122.
Harshey, R. M., and T. Ramakrishnan. 1976. Purification and properties of DNA-dependent RNA polymerase from Mycobacterium tuberculosis H37RV. Biochim. Biophys. Acta 432:49-59.[Medline]
Hartmann, G., K. O. Honikel, F. Knusel, and J. Nuesch. 1967. The specific inhibition of the DNA-directed RNA synthesis by rifamycin. Biochim. Biophys. Acta 145:843-844.[Medline]
Heep, M., B. Brandstatter, U. Rieger, N. Lehn, E. Richter, S. Rusch-Gerdes, and S. Niemann. 2001. Frequency of rpoB mutations inside and outside the cluster I region in rifampin-resistant clinical Mycobacterium tuberculosis isolates. J. Clin. Microbiol. 39:107-110.
Heep, M., S. Odenbreit, D. Beck, J. Decker, E. Prohaska, U. Rieger, and N. Lehn. 2000. Mutations at four distinct regions of the rpoB gene can reduce the susceptibility of Helicobacter pylori to rifamycins. Antimicrob. Agents Chemother. 44:1713-1715.
Herrera, L., S. Jimenez, A. Valverde, M. A. Garcia-Aranda, and J. A. Saez-Nieto. 2003. Molecular analysis of rifampicin-resistant Mycobacterium tuberculosis isolated in Spain (1996-2001). Description of new mutations in the rpoB gene and review of the literature. Int. J. Antimicrob. Agents 21:403-408.
Kadurugamuwa, J. L., L. V. Sin, J. Yu, K. P. Francis, T. F. Purchio, and P. R. Contag. 2004. Noninvasive optical imaging method to evaluate postantibiotic effects on biofilm infection in vivo. Antimicrob. Agents Chemother. 48:2283-2287.
Kissling, M., and N. Bergamini. 1981. Rifampicin in free combination with other antimicrobial drugs in non-Tb infections. Clinical data on 650 patients (a review). Chemotherapy 27:368-402.
Kunin, C. M., D. Brandt, and H. Wood. 1969. Bacteriologic studies of rifampin, a new semisynthetic antibiotic. J. Infect. Dis. 119:132-137.
Kutlin, A., S. Kohlhoff, P. Roblin, M. R. Hammerschlag, and P. Riska. 2005. Emergence of resistance to rifampin and rifalazil in Chlamydophila pneumoniae and Chlamydia trachomatis. Antimicrob. Agents Chemother. 49:903-907.
Mandell, G. L., and D. R. Moorman. 1980. Treatment of experimental staphylococcal infections: effect of rifampin alone and in combination on development of rifampin resistance. Antimicrob. Agents Chemother. 17:658-662.
Moghazeh, S. L., X. Pan, T. Arain, C. K. Stover, J. M. Musser, and B. N. Kreiswirth. 1996. Comparative antimycobacterial activities of rifampin, rifapentine, and KRM-1648 against a collection of rifampin-resistant Mycobacterium tuberculosis isolates with known rpoB mutations. Antimicrob. Agents Chemother. 40:2655-2657.
Morse, R., K. O'Hanlon, and M. D. Collins. 2002. Phylogenetic, amino acid content and indel analyses of the beta subunit of DNA-dependent RNA polymerase of gram-positive and gram-negative bacteria. Int. J. Syst. Evol. Microbiol. 52:1477-1484.
Morse, R., K. O'Hanlon, M. Virji, and M. D. Collins. 1999. Isolation of rifampin-resistant mutants of Listeria monocytogenes and their characterization by rpoB gene sequencing, temperature sensitivity for growth, and interaction with an epithelial cell line. J. Clin. Microbiol. 37:2913-2919.
National Committee for Clinical Laboratory Standards. 2003.
Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th ed., vol. 23. Approved standard M7-A6. National Committee for Clinical Laboratory Standards, Wayne, Pa.
O' Neill, A., B. Oliva, C. Storey, A. Hoyle, C. Fishwick, and I. Chopra. 2000. RNA polymerase inhibitors with activity against rifampin-resistant mutants of Staphylococcus aureus. Antimicrob. Agents Chemother. 44:3163-3166.
O'Neill, A. J., J. H. Cove, and I. Chopra. 2001. Mutation frequencies for resistance to fusidic acid and rifampicin in Staphylococcus aureus. J. Antimicrob. Chemother. 47:647-650.
Padayachee, T., and K. P. Klugman. 1999. Molecular basis of rifampin resistance in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 43:2361-2365.
Park, Y. K., B. J. Kim, S. Ryu, Y. H. Kook, Y. K. Choe, G. H. Bai, and S. J. Kim. 2002. Cross-resistance between rifampicin and KRM-1648 is associated with specific rpoB alleles in Mycobacterium tuberculosis. Int. J. Tuberc. Lung Dis. 6:166-170.
othstein, D. M., A. D. Hartman, M. H. Cynamon, and B. I. Eisenstein. 2003. Development potential of rifalazil. Expert Opin. Investig. Drugs 12:255-271.[CrossRef][Medline]
Sajduda, A., A. Brzostek, M. Poplawska, E. Augustynowicz-Kopec, Z. Zwolska, S. Niemann, J. Dziadek, and D. Hillemann. 2004. Molecular characterization of rifampin- and isoniazid-resistant Mycobacterium tuberculosis strains isolated in Poland. J. Clin. Microbiol. 42:2425-2431.
Schmitz, F.-J., A. C. Fluit, D. Hafner, A. Beeck, M. Perdikouli, M. Boos, S. Scheuring, J. Verhoef, K. Khrer, and C. Von Eiff. 2000. Development of resistance to ciprofloxacin, rifampin, and mupirocin in methicillin-susceptible and -resistant Staphylococcus aureus isolates. Antimicrob. Agents Chemother. 44:3229-3231.
Segreti, J. 2005. Efficacy of current agents used in the treatment of gram-positive infections and the consequences of resistance. Clin. Microbiol. Infect. 11:29-35.
Siddiqi, N., M. Shamim, S. Hussain, R. K. Choudhary, N. Ahmed, Prachee, S. Banerjee, G. R. Savithri, M. Alam, N. Pathak, A. Amin, M. Hanief, V. M. Katoch, S. K. Sharma, and S. E. Hasnain. 2002. Molecular characterization of multidrug-resistant isolates of Mycobacterium tuberculosis from patients in North India. Antimicrob. Agents Chemother. 46:443-450.
Smith, S. M., R. H. Eng, and F. Tecson-Tumang. 1989. Ciprofloxacin therapy for methicillin-resistant Staphylococcus aureus infections or colonizations. Antimicrob. Agents Chemother. 33:181-184.
Suchland, R. J., A. Bourillon, E. Denamur, W. E. Stamm, and D. M. Rothstein. 2005. Rifampin-resistant RNA polymerase mutants of Chlamydia trachomatis remain susceptible to the ansamycin rifalazil. Antimicrob. Agents Chemother. 49:1120-1126.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, and J. Raker. Feb. 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0043298.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, X. Y. Yu, J. M. Siedlecki, and Y. Yang. June 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0137189.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, J. Raker, X. Y. Yu, J. M. Siedlecki, and Y. Yang. Sept. 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0197333.
Wehrli, W., J. Nüesch, F. Knüsel, and M. Staehelin. 1968. Action of rifamycins on RNA polymerase. Biochim. Biophys. Acta 157:215-217.
Wehrli, W. 1983. Rifampin: mechanisms of action and resistance. Rev. Infect. Dis. 5:S407-S411.
Wichelhaus, T. A., B. Bddinghaus, S. Besier, V. Schfer, V. Brade, and A. Ludwig. 2002. Biological cost of rifampin resistance from the perspective of Staphylococcus aureus. Antimicrob. Agents Chemother. 46:3381-3385.
Wichelhaus, T. A., V. Schfer, V. Brade, and B. Bddinghaus. 2001. Differential effect of rpoB mutations on antibacterial activities of rifampicin and KRM-1648 against Staphylococcus aureus. J. Antimicrob. Chemother. 47:153-156.
Wichelhaus, T. A., V. Schfer, V. Brade, and B. Bddinghaus. 1999. Molecular characterization of rpoB mutations conferring cross-resistance to rifamycins on methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemother. 43:2813-2816.
Williams, D. L., L. Spring, L. Collins, L. P. Miller, L. B. Heifets, P. R. Gangadharam, and T. P. Gillis. 1998. Contribution of rpoB mutations to development of rifamycin cross-resistance in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 42:1853-1857.
Williams, D. L., C. Waguespack, K. Eisenach, J. T. Crawford, F. Portaels, M. Salfinger, C. M. Nolan, C. Abe, V. Sticht-Groh, and T. P. Gillis. 1994. Characterization of rifampin resistance in pathogenic mycobacteria. Antimicrob. Agents Chemother. 38:2380-2386.
Yang, B., H. Koga, H. Ohno, K. Ogawa, M. Fukuda, Y. Hirakata, S. Maesaki, K. Tomono, T. Tashiro, and S. Kohno. 1998. Relationship between antimycobacterial activities of rifampicin, rifabutin and KRM-1648 and rpoB mutations of Mycobacterium tuberculosis. J. Antimicrob. Chemother. 42:621-628.
Yu, J., J. Wu, K. P. Francis, T. F. Purchio, and J. L. Kadurugamuwa. 2005. Monitoring in vivo fitness of rifampicin-resistant Staphylococcus aureus mutants in a mouse biofilm infection model. J. Antimicrob. Chemother. 55:528-534.
Yue, J., W. Shi, J. Xie, Y. Li, E. Zeng, and H. Wang. 2003. Mutations in the rpoB gene of multidrug-resistant Mycobacterium tuberculosis isolates from China. J. Clin. Microbiol. 41:2209-2212.
Zarrouk, V., B. Bozdogan, R. Leclercq, L. Garry, C. Feger, C. Carbon, and B. Fantin. 2001. Activities of the combination of quinupristin-dalfopristin with rifampin in vitro and in experimental endocarditis due to Staphylococcus aureus strains with various phenotypes of resistance to macrolide-lincosamide-streptogramin antibiotics. Antimicrob. Agents Chemother. 45:1244-1248.
Zimmerli, W., A. F. Widmer, M. Blatter, R. Frei, P. E. Ochsner, et al. 1998. Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. JAMA 279:1537-1541.
Present address: Novartis Institutes for Biomedical Research, Inc., Infectious Disease, 100 Technology Sq., Cambridge, MA 02139.
Present address: Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, 15 Vassar St., Cambridge, MA 02139.
Present address: Predix Pharmaceuticals, Inc., 4 Maguire Rd., Lexington, MA 02421.
ActivBiotics, Inc., Lexington, Massachusetts 02421
1 Tufts University School of Medicine, 136 Harrison Ave., Boston, Massachusetts 02111
2 Veterans Affairs Medical Center and State University of New York, Upstate Medical University, 800 Irving Ave., Syracuse, New York 132103(Christopher K. Murphy,Ste)
We describe novel rifamycin derivatives (new chemical entities [NCEs]) that retain significant activity against a comprehensive collection of Staphylococcus aureus strains that are resistant to rifamycins. This collection of resistant strains contains 21 of the 26 known single-amino-acid alterations in RpoB, the target of rifamycins. Some NCEs also demonstrated a lower frequency of resistance development than rifampin and rifalazil in S. aureus as measured in a resistance emergence test. When assayed for activity against the strongest rifamycin-resistant mutants, several NCEs had MICs of 2 μg/ml, in contrast to MICs of rifampin and rifalazil, which were 512 μg/ml for the same strains. The properties of these NCEs therefore demonstrate a significant improvement over those of earlier rifamycins, which have been limited primarily to combination therapy due to resistance development, and suggest a potential use of these NCEs for monotherapy in several clinical indications.
INTRODUCTION
ifamycins, such as rifampin (RIF), rifabutin (RFB), and rifapentine, inhibit bacterial RNA polymerases isolated from a wide variety of microorganisms (4, 8, 9, 10, 21, 37, 38). This class of drugs has been used in the clinic as one component of multiple drug therapy, predominantly to treat tuberculosis (27) and to treat serious gram-positive infections (15, 32, 48). More recently, rifalazil (RFZ) [3'-hydroxy-5'-(4-isobutyl-1-piperazinyl) benzoxazinorifamycin], also known in the literature as KRM-1648 or ABI-1648, was used in phase 2 clinical trials in a multidrug regimen for the treatment of tuberculosis (6, 26).
Clinically, rifamycins are used mainly in combination therapies because of their propensity to select for resistant mutants when used as single agents (1, 42, 44). Both in vivo and in vitro, this resistance is caused by mutations in the rpoB gene, which encodes the target of rifamycins, the subunit of RNA polymerase (7, 14, 18, 27, 45, 47). The amino acid residues in the RpoB subunit of Thermus aquaticus that interact with RIF have been revealed by X-ray crystallography studies (3). Single-nucleotide changes in rpoB that result in alterations of the amino acids involved in rifamycin binding (the rifampin resistance-determining region [RRDR]) are sufficient to confer high-level resistance to these drugs (27). Resistance-conferring mutants have been well characterized for Mycobacterium tuberculosis and Staphylococcus aureus, among other bacteria (Fig. 1). The high concentration of drug required to kill such mutants is not achievable at the site of infection, and the result is therapeutic failure. Considering the rifamycin resistance mutation frequency of approximately 10–6 to 10–9 per cell (16, 24, 29, 41), clinical resistance development is exacerbated by high-bioburden infections.
ifalazil is unique among rifamycins in that is has good activity against RIF-resistant mutants of Chlamydia spp. Chlamydia trachomatis and Chlamydia pneumoniae mutants that are highly resistant to RIF are susceptible to RFZ (17, 33). Consistent with this finding, RFZ monotherapy was successful in eradicating C. trachomatis in subjects treated for nongonococcal urethritis in a recent phase 2 human clinical trial (B. Batteiger, W. McCormack, W. Stamm, and the Rifalazil Study Group, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. L-992b, 2004). However, the evaluation of pathogenic clinical isolates of other species, including both M. tuberculosis (19, 26, 44) and S. aureus (40), indicates that RFZ shares with RIF the loss of activity against strains that have resistance-conferring alterations in the RRDR.
One possible avenue for the broader use of rifamycins as monotherapeutic agents against pathogens other than Chlamydia is the creation of rifamycins with good activity against known rifamycin-resistant mutants of other pathogenic bacteria. This strategy assumes that there are no additional mutable sites in the RpoB protein that would confer high-level resistance to new, more potent rifamycins. This assumption is plausible because the interaction between the ansa chain of rifamycins and the RpoB protein occurs through amino acid residues that are highly conserved in most bacterial species (3, 20).
In searching for novel rifamycins that overcome resistance to RIF, we chose S. aureus as the test organism, because it is an important and prevalent pathogen for which there is a growing need for improved antibacterial treatments (30) and because both laboratory-derived and clinically derived rifamycin-resistant mutants of S. aureus containing mutations in rpoB have been previously described (referenced in the legend for Fig. 1).
We have synthesized over 700 novel benzoxazinorifamycins (new chemical entities [NCEs]) (34, 35, 36) and screened them for antibacterial activity, propensity to elicit resistant mutants, and activity against rifamycin-resistant mutants of S. aureus. NCEs that exhibited diminished resistance development and greatly enhanced activity against rifamycin-resistant strains of S. aureus compared to RFZ and RIF were identified. These novel compounds have the potential to be used in monotherapy more broadly than have other rifamycins.
MATERIALS AND METHODS
Bacterial strains. Wild-type S. aureus strain 8325--4 and mutated derivatives were obtained from Leeds University (I. Chopra) and have been described previously (23). RIF-resistant and wild-type S. aureus W59532 strains were obtained from Ordway Research Institute, Albany, NY, and characterized as described below. Rifamycin-resistant derivatives of S. aureus ATCC 29213 and S. aureus Smith were isolated and characterized at ActivBiotics, Inc., as outlined below. Strains were stored in 20% glycerol-cation-adjusted Mueller-Hinton II (CAM) broth (Becton Dickinson, Sparks, MD).
Materials. Rifampin B was purchased from Sigma Chemical Co. (St. Louis, MO). Rifalazil [3'-hydroxy-5'-(4-isobutyl-1-piperazinyl) benzoxazinorifamycin] and NCEs were provided by ActivBiotics, Inc. (Lexington, MA). Rifabutin was obtained from U.S. Pharmacopeia (Rockville, MD). RFZ and NCE structures are shown in Fig. 2. CAM agar was from Becton Dickinson (Sparks, MD).
Isolation of rifamycin-resistant mutants of S. aureus and determination of resistance frequency. Rifamycin-resistant mutants of S. aureus strains ATCC 29213 and Smith were isolated by inoculating 5 x 109 bacteria from independent single cultures onto CAM agar plates containing the rifamycin of interest. Rifamycin concentrations of 1 and 0.03 μg/ml were used. Plates were incubated at 35°C for 24 h. One resistant mutant per independent culture was purified and subjected to further characterization as outlined below. The total rifamycin-resistant and -sensitive colonies per plate were also counted and resistance frequencies calculated.
Antimicrobial susceptibility testing. MICs of compounds were determined by the agar dilution method or the broth microdilution method according to the Clinical and Laboratory Standards Institute (formerly the National Committee for Clinical Laboratory Standards) (22). For broth microdilution, serial twofold dilutions were carried out in CAM broth in microtiter plates. Subsequent dilutions were mixed four times with a multichannel pipettor, and pipette tips were changed every 3 dilutions to ensure minimal carryover of these hydrophobic compounds. The trays were covered with Mylar plate sealing sheets (ThermoLabsystems, Franklin, MA) and incubated at 37°C for 16 to 20 h. The MIC was the lowest concentration of compound that resulted in no detectable growth by visual inspection.
esistance emergence test. In order to assess the relative emergence of resistance, a stock seed culture (stored in 20% glycerol-CAM broth at –80°C) containing 108 cells of S. aureus strain ATCC 29213 was used to inoculate 50 ml of CAM broth and grown to an optical density at 600 nm of 1.0. A portion of this culture equivalent to 5 x 109 cells was spread on 150-mm CAM agar plates containing either 1 μg/ml or 0.3 μg/ml of the compound to be tested. Plates were incubated at 35°C for 24 h and resistant colonies counted. NCEs were scored as follows: for 1-μg/ml plates, more than 10 colonies per plate scored "0," 1 to 10 colonies per plate scored "1," and no colonies scored "2." Fewer than 10 colonies on the 0.03-μg/ml plate scored "3."
Characterization of mutations in the rpoB gene. Total DNA from resistant S. aureus strains was isolated using a QIAamp DNA mini kit (QIAGEN, Valencia, CA) and used as a template for amplification of specific regions of the rpoB gene by PCR. Oligodeoxyribonucleotide primers (Table 1) were used to amplify the DNA encoding amino acid residues 418 to 594 of the RpoB protein. These residues encompass the RRDR of the RpoB protein. For some strains, the DNA segments comprising the entire rpoB coding region were similarly amplified. The resulting amplification products were isolated after gel electrophoresis, purified using a QIAGEN gel extraction kit according to the manufacturer's instructions (QIAGEN, Inc., Valencia, CA), and sequenced at the Tufts Core Sequencing facility (Tufts University, Boston, MA).
ESULTS
ationale for analysis of rifampin resistance. Although RIF and RFZ possess excellent antibacterial activity against a broad range of bacterial species, they readily elicit highly resistant mutants (19, 26, 40, 44). We synthesized a library of novel benzoxazinorifamycins (NCEs) related to RFZ with the goal of identifying those with improved activity against rifamycin-resistant bacteria compared with RIF or RFZ. In order to evaluate activity against rifamycin-resistant mutants, two methods were employed. In one, a comprehensive collection of rifamycin-resistant mutants of S. aureus ATCC 29213 was isolated and characterized with respect to rpoB gene sequence. As a part of the process of assembling this otherwise isogenic mutant collection, NCEs, RIF, and RFZ were utilized as selective agents, providing an opportunity to determine whether mutations conferring resistance vary as a function of the selective agent. Once the comprehensive panel of mutants was assembled, the antibacterial activities of NCEs against these mutants were determined. During the process of assembly and characterization of the panel of rifamycin-resistant mutants, we instituted a second, technically simpler method, the resistance emergence test. In this approach, the frequency of emergence of rifamycin-resistant mutants from S. aureus ATCC 29213 cultures was determined directly by inoculating media containing NCEs with 5 x 109 CFU of the strain and then enumerating resistant colonies that emerged. A diagrammatic description of the screening flow and numbers of NCEs that were subjected to each method is shown in Fig. 3.
Comprehensive collection of rifamycin-resistant S. aureus strains. S. aureus ATCC 29213 or Smith isolates were inoculated onto CAM agar plates containing RIF, RFZ, or one of six NCEs at various concentrations as described in Materials and Methods. Resistant strains were isolated, and the nucleotide sequence of their rpoB genes was determined. The results from one representative experiment described in Table 2 show that the mutants obtained from ABI-0043, ABI-0094, and ABI-0299 selections contained mutations identical to mutations derived from RIF or RFZ selections (Table 2) and mutations described previously in the literature (see the legend for Fig. 1 for references). This same result was obtained when other NCEs were the selective agents (ABI-306, ABI-418, and ABI-420 [data not shown]). In all, over 200 such independently isolated mutants were characterized by sequence analysis of their RRDRs. As the isolation and analysis of resistant mutants progressed, no additional mutations were identified, suggesting genetic saturation under the conditions used. To augment our own rifamycin-resistant mutant collection with mutants conferring weaker resistance, we added to the panel a series of well-characterized strains derived from S. aureus 8325-4 (23). Finally, we also characterized a set of RIF-resistant mutants of S. aureus W59532 in anticipation of future testing of NCEs in animal models of infection. The latter strain background has been used successfully in such models (A. Louie, W. Liu, M. R. Deziel, M. Drusano, L. Turner, T. Gumbo, and G. L. Drusano, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. A-1297, 2004).
ifamycin susceptibility of the rifamycin-resistant S. aureus collection. The panel of resistant S. aureus mutants was further analyzed for susceptibility to rifamycins, including RIF, RFZ, and several NCEs (Table 3). RIF, RFB, and RFZ had identical MICs for wild-type S. aureus 29213. Rifamycin-resistant mutants fell into two classes, those that conferred high-level resistance to RIF (MIC of 64 μg/ml) and those that conferred moderate to low-level resistance to RIF (MIC of between 0.125 and 16 μg/ml) (Table 3). For low-level RIF-resistant mutants, the MICs of RFZ were between 4- and 64-fold lower than the MICs of RIF. Despite improved activity against this class of mutants, RFZ did not have better antibacterial activity against strains containing mutations that conferred high-level RIF resistance, except for the strain containing the A477D change (Table 3). These results are consistent with previously published results (40). In marked contrast, several RFZ-based NCEs had antibacterial activity that was up to 256-fold better than that of RIF against strongly RIF-resistant mutants (Table 3). Indeed, for the best NCEs, the maximum MIC observed for any member of the mutant collection was 2 to 4 μg/ml (Table 3) (see below). In general, particular mutations in different strain backgrounds all had similar susceptibilities to NCEs and rifamycins (within a twofold concentration). The few instances where rpoB mutations conferred four- to eightfold different susceptibilities to different strain backgrounds are denoted in Table 3 as separate MIC data lines. For instance, the A477D alteration imparted fourfold differences in the MICs of ABI-0418 and ABI-0420 depending on the S. aureus strain background.
Analysis of the antibacterial activity of NCEs against low-level RIF-resistant mutants uncovered a structure-activity relationship. Pairwise comparison of NCEs that differed only at position 25 of their ansa chains (data not shown) revealed that NCEs containing an O-acetyl group at this position had moderately lower (four- to eightfold) MICs against the low-level-resistant mutants than RFZ. Corresponding NCEs that contained a hydroxyl group at position 25 had roughly the same MIC against these mutants as RFZ.
To facilitate screening of the large number of NCEs in this study, a panel that consisted of representative highly RIF-resistant mutants, including alleles that had been previously identified from clinical S. aureus isolates, was assembled (H481Y, Q468K, S486L, and H481D) (1, 41). Good activity against these mutants correlated well with activity against the other highly RIF-resistant mutants in the complete panel (Table 3). Also included in the screening panel were two strains with D471Y and S464P changes in RpoB that conferred weaker RIF resistance. Of the 431 NCEs assayed for activity against the screening panel, 260 retained good activity, with MICs less than or equal to 8 μg/ml for each strain in the panel. Of these, 199 had maximum MICs of 4 μg/ml or less against screening panel strains (Fig. 3).
esistance emergence analysis. A second approach to identifying NCEs with good antibacterial potency against rifamycin-resistant mutants consisted of subjecting 758 NCEs to a resistance emergence test. S. aureus cells (109) were applied to agar medium containing either 1 or 0.3 μg/ml NCE, as described in Materials and Methods. A score of "0" in this assay indicates similarity to RFZ in terms of capacity to elicit resistant mutants, whereas scores of "1," "2," or "3" indicate progressive improvement in the ability to suppress the emergence of resistant mutants (see Materials and Methods) (Fig. 3).
Six hundred fifty-four NCEs (86.2%) scored "0" in this test, that is, they were as likely to elicit rifamycin-resistant mutants as RFZ. Forty-six NCEs (6.1%) scored "1," 55 NCEs (7.3%) scored "2," and 3 NCEs (0.4%) scored "3" (Fig. 3). The 13.4% of total NCEs that scored above "0" therefore represent a class of rifamycins that are less prone to resistance emergence. Of this group, all 3 NCEs that scored "3" and 45 of 55 NCEs that scored "2" had MICs against members of the mutant screening panel of 4 μg/ml. Of the 46 NCEs that scored "1," 39 had MICs against members of the mutant screening panel of 4 μg/ml. Thus, the resistance emergence test was a rapid and valid method for identifying NCEs that were potent against rifamycin-resistant strains. The structures of ABI-0418, ABI-0597, and ABI-0720, which scored "3" in the resistance emergence test, are shown in Fig. 2.
esistance emergence and its relation to structure at position 25. Analysis of resistance emergence data for all NCEs tested revealed that there was a distinct difference in performance of NCEs that depended on the chemical group at position 25 of the ansa chain. Those NCEs that were O-acetylated at this position scored better for resistance emergence than their hydroxylated counterparts. This phenomenon is most clearly illustrated when considering the group of 199 NCEs with high activity (MIC 4 μg/ml) against the mutant screening panel. Thirty-two of 119 (27%) 25-O-acetyl NCEs in this group had a resistance emergence score of 2, and an additional 23 compounds (19%) scored 1. For NCEs in this high-activity group that were hydroxylated at position 25, only 7 of 80 (9%) scored 2 and 13 (16%) scored 1.
Contradictions between resistance emergence score and MIC testing against RIF-resistant mutants. Several NCEs that suppressed the emergence of resistant mutants (resistance emergence score of "2") had MICs against resistant mutants in the screening panel that were relatively high. For example, ABI-0299, when present in the agar medium used to assess resistance emergence at 1 μg/ml, did not give rise to any resistant colonies. One might assume from this result that the MIC of ABI-0299 for any rifamycin-resistant mutants in our collection would be equal to or lower than 1 μg/ml. However, the MIC of ABI-0299 for the most resistant strains in the mutant screening panel was in the 2- to 4-μg/ml range (Table 3). One possible explanation for this discrepancy is that there can be a difference between MICs for NCEs tested by the agar dilution method and MICs for NCEs tested by the broth microdilution method. To determine whether this was the case, five representative NCEs were retested by these two methods (Table 4). For four of the five NCEs tested, including ABI-0299, MICs for the most resistant mutants in the screening panel were between two- and fourfold lower when tested by the agar dilution method (Table 4). This result suggests that NCEs can be more potent in an agar milieu than in broth and explains why some NCEs prevent growth of resistant mutants at 1 μg/ml in agar medium.
DISCUSSION
In this study we have identified novel rifamycins that have both the ability to suppress the emergence of rifamycin-resistant mutants and increased activity against mutants resistant to other rifamycins. This advance is important because the NCEs described here have the potential to be used as monotherapeutic agents; historically, rifamycins have been used mainly in the clinic as a part of combination antibiotic therapy due to the problem of resistance development (1, 12, 25, 42, 43).
We utilized two independent methods for evaluating NCEs for susceptibility to resistance development (Fig. 3). The resistance emergence assay is a rapid two-plate test that estimates the resistance frequency at fixed high concentrations of NCE (roughly 30- and 500-fold higher than their MICs). We used this method to screen a large number of NCEs and identify those that are improved, compared with RIF, with respect to resistance emergence. When the same NCEs were tested for resistance development at lower concentrations relative to their MICs, i.e., at less than 10x the MICs of NCEs, resistance frequencies were similar to those observed with RIF and RFZ: approximately 10–8 per cell per generation (data not shown).
The second, more labor-intensive screening method was susceptibility testing using a rifamycin-resistant mutant panel. In order to ensure a complete panel, we performed independent in vitro selections using RIF, RFZ, and six different NCEs as selective agents. It was significant that no new resistant alleles of RpoB were found despite an exhaustive search using the six NCEs as selective agents. Thus, resistance to NCEs is primarily confined to the 19 alleles derived from our selections that were found repeatedly in our analysis and that have been described in previous reports in which rifampin was the selective agent. The fact that virtually all of the published single-nucleotide mutations that confer high rifamycin resistance in clinical isolates were represented (1, 41) (Fig. 1) indicates that our panel was comprehensive and relevant. We also supplemented the collection with nine strains from the Chopra group (23), including two weakly resistant mutants. In the context of our panel, these mutants, containing L466S or D471E changes in RpoB, were in fact the least resistant to rifamycins, with MICs of RIF equal to 0.125 μg/ml (Table 3). Because we did not characterize mutants from selections at low levels of RIF (0.03 μg/ml), it is not surprising that we did not isolate these two alleles.
Our screens identified NCEs that fell into two classes: (i) NCEs (n = 199) with activity (MIC of 4 μg/ml) against rifamycin-resistant S. aureus and (ii) NCEs (n = 104) that suppressed the emergence of rifamycin-resistant S. aureus strains, most of which were also members of the first class. Many members of the first class, however, failed to suppress resistance emergence. The most likely explanation for this failure, given that no new RpoB alleles were identified, is that these NCEs are less active in the agar-based medium used in the resistance emergence assay than NCEs that scored better than RFZ.
A structure-activity relationship emerged from the analysis of NCEs based on the 25 position of the ansa chain. Although there was no clear advantage in activity against wild-type S. aureus for NCEs having an O-acetyl group or hydroxyl moiety at position 25, the NCEs which contained a 25-hydroxyl had 4- to 32-fold less antibacterial activity than their 25-O-acetyl counterparts when assayed against weakly resistant mutants in the panel (MIC of 16 μg/ml for RIF) (Table 3). This phenomenon was observed for an additional 30 pairs of molecules differing only at position 25 (data not shown). One possible explanation is that NCEs lacking an O-acetyl group at the 25 position have one less binding site in the RRDR and thus are weaker binders that are more susceptible to mutations that lead to normally weak effects for O-acetylated compounds. In fact, X-ray crystallographic data from RIF bound to Thermus aquaticus RNA polymerase define a hydrogen bond between the oxygen portion of the 25-O-acetyl group and a conserved Phe residue within the RRDR which would be lost for 25-hydroxyl NCEs (3). We are currently in the process of testing this hypothesis by determining the 50% inhibitory concentrations of 25-O-acetyl/hydroxyl pairs of NCEs against isolated mutant RNA polymerase. It is curious that NCEs had similar MICs against the strongest resistant mutants (Table 3), despite the fact that 25-hydroxyl NCEs had higher MICs against the weaker mutant class.
This superior performance of the 25-O-acetyl NCEs compared with the 25-hydroxyl NCEs was also apparent when the 34 pairs of NCEs differing only at position 25 were analyzed for resistance emergence. Twenty-eight 25-O-acetyl NCEs had a higher score than their cognate hydroxyl NCEs; only three 25-hydroxyl NCEs had a higher score, and three NCEs had the same score (data not shown). These differences could be negated if hydrolysis of the 25-O-acteylated NCEs occurred readily in vivo to yield 25-hydroxylated NCEs. However, the stability in humans of RFZ (data not shown), a 25-O-acetylated benzoxazinorifamycin, suggests that the in vitro differences described above may be indicative of their in vivo performance.
Operationally, each of the two methods used to identify NCEs with good activity against RIF-resistant S. aureus mutants had advantages. The resistance emergence test was simple and rapid and reliably identified a subset of NCEs with lower MICs against the mutant panel. However, we included susceptibility testing against a panel of resistant strains because (i) it was important to assemble a comprehensive mutant panel using NCEs, in addition to RIF, as selective agents, (ii) the resistance emergence test could select for NCEs specifically having enhanced activity in agar medium compared with that in liquid medium (Table 4), and (iii) the NCEs (especially 25-hydroxyl NCEs) that have resistance emergence scores of 0 but good MICs (4 μg/ml) against rifamycin-resistant mutants may merit further testing, for example, in animal models of infection. In any case, our methods have proved valuable in evaluating NCEs for the very important property of having increased activity against resistant strains, especially because monotherapy in the clinical setting is a goal of our program.
In evaluating the attractive NCEs defined in this study for potential clinical use, additional information provided by measuring efficacy against mutant strains in an in vivo setting would be valuable. To consider using these NCEs successfully as monotherapies, it will be essential to show efficacy against the resistant-mutant subpopulations in addition to simply killing or preventing growth of the wild-type strain.
ACKNOWLEDGMENTS
We thank Sam Bettis, Kara Brown, Debra Buxton, Courtney Calabria-Tanzi, and Julie Ormsby for technical assistance and Ian Chopra and George Drusano for bacterial strains. We also thank Linc Sonenshein and Andrew Wright for helpful discussions and for critically reading the manuscript.
EFERENCES
Aubry-Damon, H., C.-J. Soussy, and P. Courvalin. 1998. Characterization of mutations in the rpoB gene that confer rifampin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 42:2590-2594.
Billington, O. J., T. D. McHugh, and S. H. Gillespie. 1999. Physiological cost of rifampin resistance induced in vitro in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 43:1866-1869.
Campbell, E. A., N. Korzheva, A. Mustaev, K. Murakami, S. Nair, A. Goldfarb, and S. A. Darst. 2001. Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104:901-912.
Chazan, L. L., and S. T. Bayley. 1973. Some properties of a DNA-dependent RNA polymerase from Halobacterium cutirubrum. Can. J. Biochem. 51:1297-1304.[Medline]
Coll, P., L. M. Aragon, F. Alcaide, M. Espasa, M. Garrigo, J. Gonzalez, J. M. Manterola, P. Orus, and M. Salvado. 2005. Molecular analysis of isoniazid and rifampin resistance in Mycobacterium tuberculosis isolates recovered from Barcelona. Microb. Drug Resist. 11:107-114.[CrossRef][Medline]
Dietze, R., L. Teixeira, L. M. Rocha, M. Palaci, J. L. Johnson, C. Wells, L. Rose, K. Eisenach, and J. J. Ellner. 2001. Safety and bactericidal activity of rifalazil in patients with pulmonary tuberculosis. Antimicrob. Agents Chemother. 45:1972-1976.
Dreses-Werringloer, U., I. Padubrin, L. Kohler, and A. P. Hudson. 2003. Detection of nucleotide variability in rpoB in both rifampin-sensitive and rifampin-resistant strains of Chlamydia trachomatis. Antimicrob. Agents Chemother. 47:2316-2318.
Fujii, K., A. Tsuji, S. Miyazaki, K. Yamaguchi, and S. Goto. 1994. In vitro and in vivo antibacterial activities of KRM-1648 and KRM-1657, new rifamycin derivatives. Antimicrob. Agents Chemother. 38:1118-1122.
Harshey, R. M., and T. Ramakrishnan. 1976. Purification and properties of DNA-dependent RNA polymerase from Mycobacterium tuberculosis H37RV. Biochim. Biophys. Acta 432:49-59.[Medline]
Hartmann, G., K. O. Honikel, F. Knusel, and J. Nuesch. 1967. The specific inhibition of the DNA-directed RNA synthesis by rifamycin. Biochim. Biophys. Acta 145:843-844.[Medline]
Heep, M., B. Brandstatter, U. Rieger, N. Lehn, E. Richter, S. Rusch-Gerdes, and S. Niemann. 2001. Frequency of rpoB mutations inside and outside the cluster I region in rifampin-resistant clinical Mycobacterium tuberculosis isolates. J. Clin. Microbiol. 39:107-110.
Heep, M., S. Odenbreit, D. Beck, J. Decker, E. Prohaska, U. Rieger, and N. Lehn. 2000. Mutations at four distinct regions of the rpoB gene can reduce the susceptibility of Helicobacter pylori to rifamycins. Antimicrob. Agents Chemother. 44:1713-1715.
Herrera, L., S. Jimenez, A. Valverde, M. A. Garcia-Aranda, and J. A. Saez-Nieto. 2003. Molecular analysis of rifampicin-resistant Mycobacterium tuberculosis isolated in Spain (1996-2001). Description of new mutations in the rpoB gene and review of the literature. Int. J. Antimicrob. Agents 21:403-408.
Kadurugamuwa, J. L., L. V. Sin, J. Yu, K. P. Francis, T. F. Purchio, and P. R. Contag. 2004. Noninvasive optical imaging method to evaluate postantibiotic effects on biofilm infection in vivo. Antimicrob. Agents Chemother. 48:2283-2287.
Kissling, M., and N. Bergamini. 1981. Rifampicin in free combination with other antimicrobial drugs in non-Tb infections. Clinical data on 650 patients (a review). Chemotherapy 27:368-402.
Kunin, C. M., D. Brandt, and H. Wood. 1969. Bacteriologic studies of rifampin, a new semisynthetic antibiotic. J. Infect. Dis. 119:132-137.
Kutlin, A., S. Kohlhoff, P. Roblin, M. R. Hammerschlag, and P. Riska. 2005. Emergence of resistance to rifampin and rifalazil in Chlamydophila pneumoniae and Chlamydia trachomatis. Antimicrob. Agents Chemother. 49:903-907.
Mandell, G. L., and D. R. Moorman. 1980. Treatment of experimental staphylococcal infections: effect of rifampin alone and in combination on development of rifampin resistance. Antimicrob. Agents Chemother. 17:658-662.
Moghazeh, S. L., X. Pan, T. Arain, C. K. Stover, J. M. Musser, and B. N. Kreiswirth. 1996. Comparative antimycobacterial activities of rifampin, rifapentine, and KRM-1648 against a collection of rifampin-resistant Mycobacterium tuberculosis isolates with known rpoB mutations. Antimicrob. Agents Chemother. 40:2655-2657.
Morse, R., K. O'Hanlon, and M. D. Collins. 2002. Phylogenetic, amino acid content and indel analyses of the beta subunit of DNA-dependent RNA polymerase of gram-positive and gram-negative bacteria. Int. J. Syst. Evol. Microbiol. 52:1477-1484.
Morse, R., K. O'Hanlon, M. Virji, and M. D. Collins. 1999. Isolation of rifampin-resistant mutants of Listeria monocytogenes and their characterization by rpoB gene sequencing, temperature sensitivity for growth, and interaction with an epithelial cell line. J. Clin. Microbiol. 37:2913-2919.
National Committee for Clinical Laboratory Standards. 2003.
Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th ed., vol. 23. Approved standard M7-A6. National Committee for Clinical Laboratory Standards, Wayne, Pa.
O' Neill, A., B. Oliva, C. Storey, A. Hoyle, C. Fishwick, and I. Chopra. 2000. RNA polymerase inhibitors with activity against rifampin-resistant mutants of Staphylococcus aureus. Antimicrob. Agents Chemother. 44:3163-3166.
O'Neill, A. J., J. H. Cove, and I. Chopra. 2001. Mutation frequencies for resistance to fusidic acid and rifampicin in Staphylococcus aureus. J. Antimicrob. Chemother. 47:647-650.
Padayachee, T., and K. P. Klugman. 1999. Molecular basis of rifampin resistance in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 43:2361-2365.
Park, Y. K., B. J. Kim, S. Ryu, Y. H. Kook, Y. K. Choe, G. H. Bai, and S. J. Kim. 2002. Cross-resistance between rifampicin and KRM-1648 is associated with specific rpoB alleles in Mycobacterium tuberculosis. Int. J. Tuberc. Lung Dis. 6:166-170.
othstein, D. M., A. D. Hartman, M. H. Cynamon, and B. I. Eisenstein. 2003. Development potential of rifalazil. Expert Opin. Investig. Drugs 12:255-271.[CrossRef][Medline]
Sajduda, A., A. Brzostek, M. Poplawska, E. Augustynowicz-Kopec, Z. Zwolska, S. Niemann, J. Dziadek, and D. Hillemann. 2004. Molecular characterization of rifampin- and isoniazid-resistant Mycobacterium tuberculosis strains isolated in Poland. J. Clin. Microbiol. 42:2425-2431.
Schmitz, F.-J., A. C. Fluit, D. Hafner, A. Beeck, M. Perdikouli, M. Boos, S. Scheuring, J. Verhoef, K. Khrer, and C. Von Eiff. 2000. Development of resistance to ciprofloxacin, rifampin, and mupirocin in methicillin-susceptible and -resistant Staphylococcus aureus isolates. Antimicrob. Agents Chemother. 44:3229-3231.
Segreti, J. 2005. Efficacy of current agents used in the treatment of gram-positive infections and the consequences of resistance. Clin. Microbiol. Infect. 11:29-35.
Siddiqi, N., M. Shamim, S. Hussain, R. K. Choudhary, N. Ahmed, Prachee, S. Banerjee, G. R. Savithri, M. Alam, N. Pathak, A. Amin, M. Hanief, V. M. Katoch, S. K. Sharma, and S. E. Hasnain. 2002. Molecular characterization of multidrug-resistant isolates of Mycobacterium tuberculosis from patients in North India. Antimicrob. Agents Chemother. 46:443-450.
Smith, S. M., R. H. Eng, and F. Tecson-Tumang. 1989. Ciprofloxacin therapy for methicillin-resistant Staphylococcus aureus infections or colonizations. Antimicrob. Agents Chemother. 33:181-184.
Suchland, R. J., A. Bourillon, E. Denamur, W. E. Stamm, and D. M. Rothstein. 2005. Rifampin-resistant RNA polymerase mutants of Chlamydia trachomatis remain susceptible to the ansamycin rifalazil. Antimicrob. Agents Chemother. 49:1120-1126.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, and J. Raker. Feb. 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0043298.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, X. Y. Yu, J. M. Siedlecki, and Y. Yang. June 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0137189.
van Duzer, J. H., A. F. Michaelis, W. B. Geiss, D. G. Stafford, J. Raker, X. Y. Yu, J. M. Siedlecki, and Y. Yang. Sept. 2005. Rifamycin analogs and uses thereof. U.S. patent 2005/0197333.
Wehrli, W., J. Nüesch, F. Knüsel, and M. Staehelin. 1968. Action of rifamycins on RNA polymerase. Biochim. Biophys. Acta 157:215-217.
Wehrli, W. 1983. Rifampin: mechanisms of action and resistance. Rev. Infect. Dis. 5:S407-S411.
Wichelhaus, T. A., B. Bddinghaus, S. Besier, V. Schfer, V. Brade, and A. Ludwig. 2002. Biological cost of rifampin resistance from the perspective of Staphylococcus aureus. Antimicrob. Agents Chemother. 46:3381-3385.
Wichelhaus, T. A., V. Schfer, V. Brade, and B. Bddinghaus. 2001. Differential effect of rpoB mutations on antibacterial activities of rifampicin and KRM-1648 against Staphylococcus aureus. J. Antimicrob. Chemother. 47:153-156.
Wichelhaus, T. A., V. Schfer, V. Brade, and B. Bddinghaus. 1999. Molecular characterization of rpoB mutations conferring cross-resistance to rifamycins on methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemother. 43:2813-2816.
Williams, D. L., L. Spring, L. Collins, L. P. Miller, L. B. Heifets, P. R. Gangadharam, and T. P. Gillis. 1998. Contribution of rpoB mutations to development of rifamycin cross-resistance in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 42:1853-1857.
Williams, D. L., C. Waguespack, K. Eisenach, J. T. Crawford, F. Portaels, M. Salfinger, C. M. Nolan, C. Abe, V. Sticht-Groh, and T. P. Gillis. 1994. Characterization of rifampin resistance in pathogenic mycobacteria. Antimicrob. Agents Chemother. 38:2380-2386.
Yang, B., H. Koga, H. Ohno, K. Ogawa, M. Fukuda, Y. Hirakata, S. Maesaki, K. Tomono, T. Tashiro, and S. Kohno. 1998. Relationship between antimycobacterial activities of rifampicin, rifabutin and KRM-1648 and rpoB mutations of Mycobacterium tuberculosis. J. Antimicrob. Chemother. 42:621-628.
Yu, J., J. Wu, K. P. Francis, T. F. Purchio, and J. L. Kadurugamuwa. 2005. Monitoring in vivo fitness of rifampicin-resistant Staphylococcus aureus mutants in a mouse biofilm infection model. J. Antimicrob. Chemother. 55:528-534.
Yue, J., W. Shi, J. Xie, Y. Li, E. Zeng, and H. Wang. 2003. Mutations in the rpoB gene of multidrug-resistant Mycobacterium tuberculosis isolates from China. J. Clin. Microbiol. 41:2209-2212.
Zarrouk, V., B. Bozdogan, R. Leclercq, L. Garry, C. Feger, C. Carbon, and B. Fantin. 2001. Activities of the combination of quinupristin-dalfopristin with rifampin in vitro and in experimental endocarditis due to Staphylococcus aureus strains with various phenotypes of resistance to macrolide-lincosamide-streptogramin antibiotics. Antimicrob. Agents Chemother. 45:1244-1248.
Zimmerli, W., A. F. Widmer, M. Blatter, R. Frei, P. E. Ochsner, et al. 1998. Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. JAMA 279:1537-1541.
Present address: Novartis Institutes for Biomedical Research, Inc., Infectious Disease, 100 Technology Sq., Cambridge, MA 02139.
Present address: Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, 15 Vassar St., Cambridge, MA 02139.
Present address: Predix Pharmaceuticals, Inc., 4 Maguire Rd., Lexington, MA 02421.
ActivBiotics, Inc., Lexington, Massachusetts 02421
1 Tufts University School of Medicine, 136 Harrison Ave., Boston, Massachusetts 02111
2 Veterans Affairs Medical Center and State University of New York, Upstate Medical University, 800 Irving Ave., Syracuse, New York 132103(Christopher K. Murphy,Ste)