Synergistic activity of the novel des-fluoro(6) quinolone, garenoxacin (BMS-284756), in combination with other antimicrobial agents against Pseudomonas aeruginosa and related species.

Non-fermentative Gram-negative bacteria (Pseudomonas aeruginosa, Burkholderia cepacia, Stenotrophomonas maltophilia and Acinetobacter spp.) are intrinsically less susceptible to many antimicrobial agents. Two-drug combinations have been used to treat infections caused by less susceptible pathogens. In this study, the antibacterial activity of garenoxacin (GARX) with non-quinolones was examined. The non-quinolones evaluated were cefepime (CEPI), imipenem (IMIP), aztreonam (AZTR), piperacillin-tazobactam (PIPC/TZ), amikacin (AMK), ceftazidime (CTAZ), trimethoprim-sulphamethoxazole (TMP/SMX) and ticarcillin-clavulanate (TICC/CA). Synergism was determined by time-kill analysis using GARX (at 2 x its MIC, not to exceed 4 mg/l) and the second drug (at 1 x MIC, not to exceed its susceptible MIC breakpoint), and is defined as > or = 2 log(10) enhanced killing at 24 h with the combination. Partial synergy is defined as > or = 1.5 log(10) but < 2 log(10) enhanced killing with the drug combination. Synergy/partial synergy was observed most often with GARX plus: CEPI, AZTR, PIPC/TZ, IMIP (five strains each) or AMK (four strains) vs. eight P. aeruginosa; CTAZ, AZTR (five strains each) vs. six B. cepacia; TICC/CA (six strains), CEPI, CTAZ or AMK (five strains each) vs. eight S. maltophilia; and CEPI, AMK (three strains each) or CTAZ, TICC/CA (two strains each) vs. four Acinetobacter spp. In conclusion, synergistic killing was observed frequently with GARX plus a non-quinolone bactericidal agents against non-fermentative Gram-negative bacteria, including strains intermediately susceptible/resistant to one or both agents.

Activity of gatifloxacin against strains resistant to ofloxacin and ciprofloxacin and its ability to select for less susceptible bacterial variants.

Gatifloxacin is an 8-methoxy fluoroquinolone. On quinolones, this side chain imparts increased activity against Gram-positive bacteria and enhanced killing. Gatifloxacin was tested against ofloxacin non-susceptible (ofloxacin MIC>2 mg/l) strains of Streptococcus pneumoniae (gatifloxacin MIC(90), 1 mg/l) and methicillin-resistant Staphylococcus aureus (MRSA, gatifloxacin MIC(90), 4 mg/l), and to ciprofloxacin non-susceptible (ciprofloxacin MIC>1 mg/l) strains of Escherichia coli (gatifloxacin MIC(90),>16 mg/l) and ciprofloxacin non-susceptible (ciprofloxacin MIC>0.06 mg/l) Neisseria gonorrhoeae (gatifloxacin MIC(50), 0.12 mg/l and MIC(90), 0.5 mg/l). Though gatifloxacin showed some reduced susceptibility to these populations, the MIC(50) and MIC(90) values suggest that gatifloxacin may be useful against pneumococci and some gonococcal strains not susceptible to other fluoroquinolones. Gatifloxacin did not select for less susceptible variants of MRSA and pneumococci, in contrast to the 10- to 100-fold higher selection frequencies with ciprofloxacin and ofloxacin. The single-step E. coli mutants selected by gatifloxacin and the comparator quinolones had quinolone MICs within the susceptible range. These data suggest that gatifloxacin use may hinder the development of quinolone-resistance, particularly in Gram-positive bacteria.

Activity of gatifloxacin and ciprofloxacin in combination with other antimicrobial agents.

The influence of non-quinolone antimicrobial agents on the antibacterial activities of gatifloxacin and ciprofloxacin was determined using chequerboard, fractional inhibitory concentration, (FIC) and time-kill analysis methods. In the chequerboard method, the quinolones were tested in combination with ten antimicrobial agents (macrolides, aminoglycosides, beta-lactams, vancomycin, rifampicin and chloramphenicol) against five bacterial strains (one strain each of Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, Enterococcus faecalis and Streptococcus pneumoniae). In no incidence was antagonism (FIC > or = 4) or synergy (FIC < or = 0.5) observed; all dual drug combinations involving gatifloxacin or ciprofloxacin showed additivity/indifference (FIC > 0.5, < 4). By time-kill analysis, the strains were tested at a quinolone concentration equal to 8 x MIC in combination with a second antibiotic at 0.5xits MIC. These combinations killed non-enterococcal strains at rates similar to those with quinolones alone. However, rifampicin and chloramphenicol were often antagonistic (100-fold lesser killing) to the lethal action of gatifloxacin and ciprofloxacin against E. faecalis. These findings indicate that, with the exception of E. faecalis, the antibacterial activities of quinolones are generally additive/indifferent to those of other antimicrobial agents.

Antibacterial spectrum of a novel des-fluoro(6) quinolone, BMS-284756.

The in vitro spectrum of a novel des-fluoro(6) quinolone, BMS-284756, was compared with those of five fluoroquinolones (trovafloxacin, moxifloxacin, levofloxacin, ofloxacin, and ciprofloxacin). BMS-284756 was among the most active and often was the most active quinolone against staphylococci (including methicillin-resistant strains), streptococci, pneumococci (including ciprofloxacin-nonsusceptible and penicillin-resistant strains), and Enterococcus faecalis. BMS-284756 inhibited approximately 60 to approximately 70% of the Enterococcus faecium (including vancomycin-resistant) strains and 90 to 100% of the Enterobacteriaceae strains and gastroenteric bacillary pathogens at the anticipated MIC susceptible breakpoint (

Comparative killing rates of fluoroquinolones and cell wall-active agents.

Killing rates of fluoroquinolones, beta-lactams, and vancomycin were compared against Enterobacteriaceae, Staphylococcus aureus, pneumococci, streptococci, and Enterococcus faecalis. The times required for fluoroquinolones to decrease viability by 3 log(10) were 1.5 h for Enterobacteriaceae, 4 to 6 h for staphylococci, and >/=6 h for streptococci and enterococci. Thus, the rate of killing by fluoroquinolones is organism group dependent; overall, they killed more rapidly than beta-lactams and vancomycin.

In vitro antifungal activity of BMS-207147 and itraconazole against yeast strains that are non-susceptible to fluconazole.

The activities of itraconazole and the new triazole BMS-207147 were determined against Candida strains that were susceptible-dose dependent (fluconazole MICs 16 to 32 micrograms/mL) or resistant (MICs > or = 64 micrograms/mL) to fluconazole. These strains included clinical isolates of Candida krusei, Candida glabrata, and Candida albicans. In addition, 16 isogenic, genetically characterized isolates of C. albicans, with progressively decreased susceptibility to fluconazole, were tested. BMS-207147 MICs to C. krusei, a species considered intrinsically resistant to fluconazole, were at 0.13 to 0.5 microgram/mL. The population distribution of the fluconazole-nonsusceptible C. glabrata was bimodal with BMS-207147/itraconazole MICs at 0.5 to 2 micrograms/mL and > or = 16 micrograms/mL. The BMS-207147 MICs to the majority of fluconazole-nonsusceptible C. albicans strains tested were < or = 1 microgram/mL. The activity of BMS-207147 was minimally affected by overexpression of the gene encoding the efflux pump MDR1, but MIC increases were observed with changes in ERG11 and with overexpression of the CDR transporter gene. Nonetheless, BMS-207147 can be active against C. albicans mutants containing cumulative resistance mechanisms to azoles. In other words, fluconazole-resistant candidal strains may be susceptible to BMS-207147.

In vitro activity of a new oral triazole, BMS-207147 (ER-30346)

The antifungal activity of BMS-207147 (also known as ER-30346) was compared to those of itraconazole and fluconazole against 250 strains of fungi representing 44 fungal species. MICs were determined by using the National Committee for Clinical Laboratory Standards (NCCLS)-recommended broth macrodilution method for yeasts, which was modified for filamentous fungi. BMS-207147 was about two- to fourfold more potent than itraconazole and about 40-fold more active than fluconazole against yeasts. With the NCCLS-recommended resistant MIC breakpoints of > or = 1 microg/ml for itraconazole and of > or = 64 microg/ml for fluconazole against Candida spp., itraconazole and fluconazole were inactive against strains of Candida krusei and Candida tropicalis. In contrast, all but 9 (all C. tropicalis) of the 116 Candida strains tested had BMS-207147 MICs of < 1 microg/ml. The three triazoles were active against about half of the Candida glabrata strains and against all of the Cryptococcus neoformans strains tested. The three triazoles were fungistatic to most yeast species, except for BMS-207147 and itraconazole, which were fungicidal to cryptococci. BMS-207147 and itraconazole were inhibitory to most aspergilli, and against half of the isolates, the activity was cidal. BMS-207147 and itraconazole were active, though not cidal, against most hyaline Hyphomycetes (with the exception of Fusarium spp. and Pseudallescheria boydii), dermatophytes, and the dematiaceous fungi and inactive against Sporothrix schenckii and zygomycetes. Fluconazole, on the other hand, was inactive against most filamentous fungi with the exception of dermatophytes other than Microsporum gypseum. Thus, the spectrum and potency of BMS-207147 indicate that it should be a candidate for clinical development.

Recent developments in pradimicin-benanomicin and triazole antibiotics.

Fungal infections are on the rise as the number of patients with compromised immune systems continues to increase. The need for safer and more effective antifungals has resulted in the search for novel drug classes and for modifications to existing classes, with the aim of enhancing their antifungal spectra and potency. In this review, two classes of antifungals are discussed: the pradimicin-benanomicin antibiotics and the newer triazole derivatives. These have activity against Candida spp., Cryptococcus neoformans and Aspergillus spp., as well as variable activity against other less commonly encountered fungi including Pneumocystis carinii. Pradimicins-benanomicins are generally fungicidal, whereas the newer azoles appear to be selectively fungicidal to Cryptococcus neoformans and Aspergillus spp. Pradimicin-benanomicin acts by binding to mannan and alters membrane integrity. One water-soluble pradimicin candidate, BMS-181184, has been selected for clinical development. The triazoles act by inhibiting cytochrome P450 sterol 14a-demethylase. Four triazoles either currently in clinical development (voriconazole and D0870) or being considered as clinical candidates (ER-30346 and Sch 56592) will be discussed. The antifungal spectra, pharmacokinetic and toxicologic data in animals, and efficacy results in experimental infection models will be reviewed for BMS-181184 and the four newer triazoles. Results from the early clinical trials for voriconazole and D0870 will also be discussed.

Differences in the resistant variants of Enterobacter cloacae selected by extended-spectrum cephalosporins.

The rates of development of resistance to ceftriaxone, ceftazidime, cefepime, and cefpirome in 10 strains of Enterobacter cloacae were determined by daily transfer for 7 days to fresh medium containing twofold serial dilutions of the antibiotics. Development of resistance to ceftriaxone was the most rapid; this was followed by ceftazidime, cefpirome, and cefepime. Resistant variants selected by ceftriaxone and ceftazidime were cross-resistant and produced very high levels of beta-lactamase. On the other hand, resistant variants selected by cefepime and cefpirome often had moderately high levels of beta-lactamase and diminished levels of the 39- to 40-kDa porin protein.

In vitro antifungal and fungicidal spectra of a new pradimicin derivative, BMS-181184.

A new pradimicin derivative, BMS-181184, was compared with amphotericin B and fluconazole against 249 strains from 35 fungal species to determine its antifungal spectrum. Antifungal testing was performed by the broth macrodilution reference method recommended by the National Committee for Clinical Laboratory Standards (document M27-P, 1992). BMS-181184 MICs for 97% of the 167 strains of Candida spp., Cryptococcus neoformans, Torulopsis glabrata, and Rhodotorula spp. tested were < or = 8 micrograms/ml, with a majority of MICs being 2 to 8 micrograms/ml. Similarly, for Aspergillus fumigatus and 89% of the 26 dermatophytes tested BMS-181184 MICs were < or = 8 micrograms/ml. BMS-181184 was fungicidal for the yeasts, dermatophytes, and most strains of A. fumigatus, although the reduction in cell counts was less for A. fumigatus than for the yeasts. BMS-181184 was active against Sporothrix schenckii, dematiaceous fungi, and some members of the non-Aspergillus hyaline hyphomycetes. BMS-181184, however, was not fungicidal against members of the family Dematiaceae. BMS-181184 lacked activity or had poorer activity (MICs, > or = 16 micrograms/ml) against Aspergillus niger, Aspergillus flavus, Malassezia furfur, Fusarium spp., Pseudallescheria boydii, Alternaria spp., Curvularia spp., Exserohilum mcginnisii, and the zygomycetes than against yeasts. The activity of BMS-181184 was minimally (twofold or less) affected by changes in testing conditions (pH, inoculum size, temperature, the presence of serum), testing methods (agar versus broth macrodilution), or test media (RPMI 1640, yeast morphology agar, high resolution test medium). Overall, our results indicate that BMS-181184 has a broad antifungal spectrum and that it is fungicidal to yeasts and, to a lesser extent, to filamentous fungi.

In vitro activity of BMS-181139, a new carbapenem with potent antipseudomonal activity.

The in vitro activities of the carbapenem BMS-181139 were determined in comparison with those of imipenem, meropenem, ciprofloxacin, ceftriaxone, and vancomycin. BMS-181139 was the most active against species of Pseudomonas and related genera Alteromonas and Burkholderia, with MICs for 147 of 149 isolates of < 4 micrograms/ml. Of 22 imipenem-resistant (MIC > 8 micrograms/ml) P. aeruginosa strains, only 1 required an MIC of BMS-181139 of > 4 micrograms/ml, compared with 14 requiring the same meropenem MIC. BMS-181139 was the most active carbapenem against the majority of other gram-negative species except members of the tribe Proteeae, against which meropenem was more active. Although imipenem was more active against gram-positive species, BMS-18139 MICs at which 90% of strain tested were inhibited were < 1 microgram/ml for these species. BMS-181139 was generally active against isolates resistant to ciprofloxacin or broad-spectrum cephalosporins, including those containing plasmid-encoded beta-lactamases or high levels of chromosome-encoded beta-lactamases, as well as anaerobes except Clostridium difficile. Inoculum effects were noted for all three carbapenems against Klebsiella pneumoniae, Enterobacter cloacae, and Serratia marcescens but not Escherichia coli, Pseudomonas aeruginosa, or Staphylococcus aureus. BMS-181139's inoculum effect tended to be more marked. BMS-181139 exhibited bactericidal activity at the MIC for some strains and up to four to eight times the MIC for others. The postantibiotic effect of BMS-181139 was equal to or less than that of imipenem and, like meropenem, exhibited intraspecies variability. BMS-181139 was 30-fold more stable than imipenem and 7-fold more stable than meropenem to hydrolysis by hog kidney dehydropeptidase.

Activity of carbapenem BMS-181139 against Pseudomonas aeruginosa is not dependent on porin protein D2.

The broad antipseudomonal spectrum of the carbapenem BMS-181139 includes clinical strains and laboratory mutants of Pseudomonas aeruginosa that are resistant to imipenem. Unlike other known carbapenems (meropenem, panipenem, biapenem, and BO-2727), which have reduced activity against imipenem-resistant strains of P. aeruginosa, BMS-181139 was equally active against imipenem-susceptible (D2-sufficient) and imipenem-resistant (D2-deficient) strains. Conversely, imipenem and meropenem activities were the same against the susceptible parental strains and their BMS-181139-resistant mutants. Whereas basic amino acids antagonized the antipseudomonal activities of imipenem and meropenem, they had no effect on the activity of BMS-181139. These results suggest that the uptake of BMS-181139 into pseudomonal cells occurs by a non-D2 pathway. Compared with imipenem and meropenem, BMS-181139 may have a slightly higher affinity for penicillin-binding protein 2 (PBP-2) of P. aeruginosa. The rates of resistance development to imipenem, meropenem, and BMS-181139 in P. aeruginosa strains were similar; resistance occurred at frequencies of approximately 10(-7) to 10(-8). Resistance to BMS-181139 in P. aeruginosa is presumed to be caused by its diminished permeability since no change in their penicillin-binding protein affinities or beta-lactamase levels could be detected. In summary, BMS-181139 is a new carbapenem which differs from other known carbapenems in its lack of cross-resistance with imipenem. This difference could be explained by the permeation of BMS-181139 through a non-D2 channel, compared to the preferential uptake of other carbapenems by the D2 porin.

Structure-activity relationships of carbapenems that determine their dependence on porin protein D2 for activity against Pseudomonas aeruginosa.

A number of carbapenem derivatives were examined to determine the structure-activity relationships required for dependence on porin protein D2 for activity against Pseudomonas aeruginosa. As suggested by J. Trias and H. Nikaido (Antimicrob. Agents Chemother. 34:52-57, 1990), carbapenem derivatives, such as imipenem and meropenem, containing a sole basic group at position 2 of the molecule utilize the D2 channel for permeation through the outer membrane of pseudomonads; they are more active against D2-sufficient strains of P. aeruginosa. Our results indicated that carbapenems with a basic group at position 1 or 6 of the molecule did not depend on the D2 channel for activity; i.e. they were equally active against D2-sufficient and D2-deficient pseudomonal strains. However, addition of a basic group at position 1 or 6 of a carbapenem derivative already containing a basic group at position 2 resulted in its lack of dependency on the D2 pathway. Comparison between meropenem and its 1-guanidinoethyl derivative, BMY 45047, indicated that they differed in their dependence on D2; while meropenem required the D2 channel for uptake, BMY 45047 activity was independent of D2. Meropenem and BMY 45047 had similar affinities for the penicillin-binding proteins of P. aeruginosa. However, BMY 45047 and meropenem differed in the morphological changes that they induced in pseudomonal cells. While meropenem induced filamentation, BMY 45047 induced filaments only in BMS-181139-resistant mutants and not in imipenem-resistant mutants or in carbapenem-susceptible P. aeruginosa strains. These results suggested that in Mueller-Hinton medium the uptake of BMY 45047 through the non-D2 pathway is more rapid than that of meropenem through the D2 porin. In summary, the presence of a basic group at position 2 of a carbapenem is important for its preferential uptake by the D2 channel. However the addition of a basic group at position 1 or 6 of a carbapenem already containing a basic group at position 2 dissociates its necessity for porin protein D2 for activity.

Ciprofloxacin-induced, low-level resistance to structurally unrelated antibiotics in Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus.

The effects of ciprofloxacin on the rates of development of low-level resistance to other antibiotics were determined in vitro. Three methicillin-resistant Staphylococcus aureus and two Pseudomonas aeruginosa clinical strains were grown overnight in Mueller-Hinton broth with or without subinhibitory concentrations (1/2, 1/4, and 1/8 MICs) of ciprofloxacin or an aminoglycoside and then quantitatively plated onto medium containing 4 or 8 times the MICs of various antibiotics. The spontaneous mutational frequencies were determined and compared with those of cells not exposed to ciprofloxacin. Exposure of methicillin-resistant S. aureus strains to ciprofloxacin resulted in a > 100-fold increase in the isolation of variants with decreased susceptibilities to ciprofloxacin, tetracycline, imipenem, fusidic acid, and gentamicin, but not vancomycin. Likewise, a > 100-fold increase in the isolation of variants with decreased susceptibilities to ciprofloxacin and imipenem (35-fold) in P. aeruginosa A21213 was observed, and a > 100-fold increase in the isolation of variants with decreased susceptibilities to ciprofloxacin, amikacin, and cefepime in P. aeruginosa A22379 was observed. On the other hand, exposure of these strains to an aminoglycoside did not influence the development of resistance to nonaminoglycoside drugs. These results indicate that exposure to subinhibitory levels of ciprofloxacin can promote the development of low-level resistance to antibiotics with different modes of action.

Activity of quinolones in the Ames Salmonella TA102 mutagenicity test and other bacterial genotoxicity assays.

Eight quinolones were examined for their bacterial mutagenicity in the Ames Salmonella TA102 assay and for their effects in other bacterial genotoxicity assays. In the quantitative Ames plate incorporation assay, all eight quinolones induced His+ deletion reversion in Salmonella tester strain TA102, with maximum reversion observed at about two to eight times the MIC. The quinolones also induced the SOS response. At quinolone concentrations close to the MIC, SOS cell filamentation gene sulA was induced in sulA::lacZ fusion strain Escherichia coli PQ37. RecA-mediated cleavage of lambda repressor in lambda::lacZ fusion strain E. coli BR513 was measurable at about 10 times the MIC, though no induction occurred at 20 micrograms of nalidixic or oxolinic acid per ml. Genotoxicity of quinolones also was observed in the Bacillus subtilis DNA repair assay, in which the mutant strain M45 (recA) was more susceptible to quinolones than its parent strain, H17 (rec+). The results from these analyses indicate that quinolones induce SOS functions and are mutagenic in bacteria; these properties correspond to their antimicrobial activities.

Oral bioavailability and anti-simian varicella virus efficacy of 1-beta-D-arabinofuranosyl-E-5-(2-bromovinyl)uracil (BV-araU) in monkeys.

BV-araU (1-beta-D-arabinofuranosyl-E-5-[2-bromovinyl]uracil) has potent antiviral activity against varicella zoster virus in cell culture and is undergoing clinical evaluation. In the present study, pharmacokinetic parameters and the efficacy of BV-araU against infection with simian varicella virus (SVV) were evaluated in African green monkeys. Pharmacokinetic parameters were determined by analysis of the BV-araU content of sera obtained after oral and intravenous administration to normal monkeys. Peak serum concentrations showed dose proportionality, with the 0.1 mg/kg dose resulting in a peak serum concentration of 0.05 micrograms/ml, the approximately ED50 value for the SVV inoculum in cell culture. BV-araU administered orally twice daily at 0.1 mg/kg for 10 days starting 48 h after intratracheal SVV infection prevented vesicular rash development and suppressed viremia. Effective therapy could be initiated 96 h after infection. Taken together, these results indicate that BV-araU is effective oral therapy at doses that achieve peak serum levels equivalent to the ED50 for SVV in cell culture.

Tissue distribution of amphotericin B lipid complex in laboratory animals.

Amphotericin B lipid complex (ABLC), under development for the treatment of serious fungal disease, is not a true liposome but a complex of amphotericin B, dimyristoyl phosphatidylcholine and dimyristoyl phosphatidylglycerol with a particle size range of 1.6-6.0 microns. Tissue distribution of ABLC was determined in mice and rats after i.v. or i.p. administration. ABLC resembles typical liposomal preparations with amphotericin B concentrating in the reticuloendothelial system. After a single i.v. treatment with ABLC, amphotericin B was present in high concentrations in liver, lung and spleen of mice and rats while plasma levels were consistently low. Mouse liver contained 48% of the administered dose 1 h after treatment and always contained the largest amount of amphotericin B after ABLC treatment. In mice treated once daily for 7 consecutive days with 10 mg kg-1 ABLC, liver amphotericin B concentration reached 377 micrograms g-1. Tissue concentrations of amphotericin B were substantially lower when ABLC was given i.p. instead of i.v. with reticuloendothelial tissues containing 2- to 7-fold more after i.v. treatment. Animals treated with 10 mg kg-1 ABLC for 14 consecutive days showed no overt signs of toxicity and had only transient changes in liver and kidney function after treatment.

Emergence of homogeneously methicillin-resistant Staphylococcus aureus.

Forty-seven clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA), collected between 1986 and 1990 from 29 institutions, were analyzed for susceptibility to various antibiotics. Twenty-six strains were homogeneously methicillin resistant (i.e., greater than or equal to 10% of the cells in these strains were able to grow on Mueller-Hinton agar containing 50 micrograms of methicillin per ml). The MICs of gentamicin, clindamycin, trimethoprimsulfamethoxazole, methicillin, and imipenem for homogeneous MRSA strains were higher than those for heterogeneously resistant strains. Both types of strains were, for the most part, susceptible to vancomycin and trimethoprim-sulfamethoxazole. Ciprofloxacin-resistant MRSA strains were not isolated prior to 1988 but made up 40% of the post-1987 strains. The level of methicillin resistance correlated well with the imipenem MIC, suggesting that susceptibility to imipenem may serve as a marker to identify and monitor the prevalence of homogeneous MRSA strains.