The increasing prevalence of methicillin-resistant Staphylococcus aureus (MRSA)1 and multi/extensively drug-resistant Mycobacterium tubcerulosis is a major global health problem.2, 3 The development of a new class of antibacterials based on a novel chemotype or mechanism of action, and with efficacy against both susceptible and resistant strains, would be an extremely useful addition to current therapeutic options.

Previously, we identified four novel bis-imidazolinylindole compounds MBX 1066, MBX 1090, MBX 1113 and MBX 1128, that represent a new antibacterial chemotype with potent activity against a broad spectrum of both Gram-positive and Gram-negative bacterial species.4 While these compounds are rapidly bactericidal, their precise bacterial target is unknown, but studies suggest that they bind to DNA and inhibit DNA synthesis. In vitro, these compounds exhibit potent activity against antibiotic-resistant strains such as ciprofloxacin-resistant B. anthracis Ames, MRSA and vancomycin-resistant enterococci. Members of this class of compounds also exhibit activity in vivo in mouse models of lethal infection with B. anthracis, Y. pestis and methicillin-susceptible S. aureus.4

A particularly important feature of this new chemotype is the very low potential for development of mutations to resistance. Specifically, no mutations to resistance were observed for MBX 1066 and MBX 1113 and MBX 1162, even after 20 days of growth of S. aureus4, 5 and Escherichia coli (unpublished data) in sublethal concentrations. However, for MBX 1090, mutant clones were able to grow in concentrations up to 16 × the compound MIC.4 Interestingly, the mutants were not cross-resistant to the other members of this class of compounds and the mutations responsible for MBX 1090 resistance were demonstrated to cause elevated production of an efflux pump for which other compounds in this class are not a substrate.5 In this study, we extended the analysis of the activities of the bis-imidazolinylindole class of antibacterials against two clinically important antibiotic-resistant pathogens.

As our previous studies demonstrated that the bis-imidazolinylindole compounds MBX 1066, MBX 1090, MBX 1113 and MBX 1128 (see Table 1 for the structures) are active against M. smegmatis,4 we investigated their effect on one susceptible (H37Rv) and seven multidrug-resistant isolates of M. tuberculosis. Table 1 summarizes the in vitro activities of the four compounds. MICs ranged from <0.02 to 1.3 μg ml−1. The compounds MBX 1066 and MBX 1113 exhibited more potent antimycobacterial activity than did MBX 1090 and MBX 1128. Next, the antimycobacterial activity of MBX 1066 was compared with its structural analog, MBX 1162 (see Table 2 for the structure), which has previously shown to exhibit potent activity against Gram-positive and Gram-negative bacteria.6 As shown in Table 2, MBX 1162 was highly active against all the multidrug-resistant isolates of M. tuberculosis and compared favorably with MBX 1066. MBX 1162 was assayed for cytotoxicity against HeLa cells as described previously4 and exhibited slightly increased toxicity (CC50=4 μg ml−1) compared with its parent MBX 1066 (CC50=33 μg ml−1).4

Table 1 Comparison of the MICs of bis-imidazolinylindole compounds against antibiotic-sensitive and multidrug-resistant isolates of M. tuberculosis
Table 2 MICs of the two most-potent bis-imidazolinylindole compounds against Beijing antibiotic-sensitive and multidrug-resistant clinical isolates of M. tuberculosis

Previously, we reported that the four bis-imidazolinylindole compounds examined herein are quite active in vitro against MRSA, with MIC values ranging from 0.2 to 5 μg ml−1.4 To investigate the in vivo efficacy of these compounds, ICR/Swiss Webster mice (n=10) were challenged via the i.p. route with S. aureus strain NRS384, a community-associated methicillin-resistant (cMRSA) clinical isolate. One hour post infection, mice were treated via the i.v. route with a single dose of compound MBX 1066 (1 or 10 mg kg−1), MBX 1090 (1 or 10 mg kg−1) or MBX 1162 (1 or 10 mg kg−1). A 70–100% survival rate was observed in the compound-treated mice (Table 3). The protective effects were similar to those observed with the reference antibiotic daptomycin. Furthermore, sampling of the spleen for bacterial load revealed a 2–4 log reduction in CFU (per tissue) in each of the groups compared with the untreated control (Table 3). Thus, the compounds are active in vivo against cMRSA infection.

Table 3 In vivo efficacy of bis-imidazolinylindole compounds against community-associated methicillin-resistant S. aureus

In summary, we have shown that several members of the bis-imidazolinylindole compounds exhibit very potent in vitro activity against both susceptible and multidrug-resistant isolates of M. tuberculosis as well as in vivo activity against cMRSA. A newer analog, MBX 1162, demonstrated equivalent potency and slightly increased mammalian cell toxicity compared with MBX 1066. Optimization of the scaffold proceeds with the goal of lessening toxicity while maintaining potency. These compounds represent a promising new scaffold for the treatment of drug-resistant bacterial pathogens.

Experimental procedure

Bacterial strains

The M. tuberculosis clinical isolates tested in this study are maintained at an in-house repository at the Southern Research Institute (Table 1) and the National Institute of Allergy and Infectious diseases7 (Table 2). The drug resistance profiles for each clinical isolate are listed in their respective tables. S. aureus NRS384, a cMRSA clinical isolate representative of the USA300 clone, was used in the in vivo efficacy studies. The isolate is both staphylococcal chromosomal cassette mec type IV positive and Panton–Valentine leucocidin positive. The isolate was obtained from the Network on Antimicrobial Resistance in S. aureus (NARSA, Herndon, VA, USA).

In vitro growth inhibition studies

MIC was determined using the broth microdilution method.8 Briefly, compounds were dissolved in dimethyl sulfoxide and then diluted with Middlebrook 7H9 broth (BD, Sparks, MD, USA). Twofold serially diluted compounds (0.1 ml per well) were added to 96-well microtiter plates. The plates were inoculated with appropriate bacterial cultures that were standardized turbidimetrically to a concentration of 5 × 105 per well. The plates were incubated at 37 °C for 7 days. MIC was determined visually as the lowest amount of drug that completely inhibited growth and was scored by comparing the growth in each well with growth in the negative-control wells (bacteria only) and positive-control wells (isoniazid).

In vivo efficacy in the mouse model of methicillin-resistant S. aureus infection

Female ICR/Swiss mice (Charles River Laboratories, Wilmington, MA, USA) were challenged i.p. with 7.4 × 107 CFU of cMRSA. After 1 h, mice were treated via i.v. route with 1 or 10 mg kg−1 of compounds. On day 2, mice that survived challenge were euthanized and spleen harvested. Homogenized spleen was serially diluted in sterile phosphatase-buffered saline and appropriate concentrations plated on Trypticase Soy Agar plates. The number of CFU per tissue was determined by quantitative colony counts.