In Vitro Activities of Isepamicin, Other Aminoglycosides, and Capreomycin against Clinical Isolates of Rapidly Growing Mycobacteria in Taiwan

Gwan-Han Shen, Bo-Da Wu, Kun-Ming Wu, Jiann-Hwa Chen

Division of Respiratory and Critical Care Medicine, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, Republic of China, Institute of Molecular Biology, National Chung Hsing University, Taichung, Taiwan, Republic of China, Rui-Fu-Shi Medical Laboratory, Taichung,

The in vitro activities of isepamicin against 117 Mycobacteria abscessus, 48 Mycobacterium fortuitum, and 20 Mycobacterium chelonae isolates were evaluated by a microdilution test. Isepamicin MIC 90 s were 16 µg/ml for the three species. Isepamicin was as active as amikacin and kanamycin and more active than tobramycin, capreomycin, gentamicin, and streptomycin.

Rapidly growing mycobacteria (RGM) can cause a wide spectrum of disseminated or localized diseases, especially pulmonary, skin, or soft tissue infections ( 6 ). Mycobacterium abscessus, Mycobacterium chelonae, and Mycobacterium fortuitum are the three major pathogenic RGM species. The management of RGM remains very difficult, especially for the problems associated with infection caused by M. abscessus ( 12 ).

Aminoglycoside agents have the potential to be extremely active against RGM ( 1, 5, 15 ). Amikacin has shown excellent activities against RGM in several studies and currently is the most widely used aminoglycoside in the treatment of RGM ( 1, 5, 15, 18 ). Amikacin and isepamicin, an aminoglycoside used in Asia, were developed by introducing the ( S )-4-amino-2-hydroxybutyryl and ( S )-3-amino-2-hydroxypropionyl side chains into the 1-amino groups of kanamycin and gentamicin, respectively ( 8 ). Isepamicin has shown excellent activities against a wide range of bacteria ( 4 ). The cyclic peptide capreomycin is sometimes considered an aminoglycoside because of its actions on bacterial ribosomes ( 7 ). This study compared the activities of isepamicin with those of five other aminoglycosides (amikacin, gentamicin, kanamycin, tobramycin, and streptomycin) and capreomycin against RGM.

RGM isolates were collected between November 2005 and July 2006 and identified by the conventional biochemical methods ( 10 ). Some of these (136 isolates) were confirmed by PCR restriction enzyme analysis of the 65-kDa hsp gene ( 13 ). Totals of 117 M. abscessus, 48 M. fortuitum, and 20 M. chelonae nonduplicate clinical isolates were collected. Of them, 71 (61%), 12 (25%), and 7 (35%), respectively, were recovered from patients with probable RGM infections (in which cases identical RGM species were recovered from three or more specimens from the same patient).

Broth microdilution MIC testing was performed according to CLSI guidelines ( 11, 16 - 18 ). The isolates were subcultured on Trypticase soy agar plates with 5% sheep blood (BBL Microbiology Systems) and incubated at 30°C for 72 h. Bacteria on the agar plates were collected and adjusted to a final inoculum (5 x 10 5 CFU/ml) in cation-supplemented Mueller-Hinton broth (Difco, Detroit, MI). Serial double dilutions of the tested antimicrobial agents were prepared with the same broth, and the concentrations in the wells ranged from 0.25 to 128 µg/ml. The inoculated trays were incubated at 30°C, and MICs were recorded after 3 to 5 days.

RGM isolates with amikacin MICs of 64 µg/ml are interpreted as resistant to amikacin and those with amikacin MICs of 16 µg/ml as susceptible to amikacin according to the CLSI cutoff criteria ( 11 ). No interpretive criteria have been approved for the susceptibilities of RGM to the other six agents except for that of M. chelonae to tobramycin. Quality control strain Staphylococcus aureus ATCC 29213 was included, and the results were in the acceptable range (MICs of 1 to 4 µg/ml).

Table 1 shows the MIC ranges, the MIC 50 s and MIC 90 s, and the percentages of isolates with MICs of 16, 32, and 64 µg/ml for the seven antimicrobial agents against the 185 RGM isolates. It is clear that amikacin, isepamicin, and kanamycin had excellent activities against RGM (MIC 50 s, 1 to 16 µg/ml; MIC 90 s, 87% of the isolates of each of the three RGM species had MICs of 16 µg/ml. When MIC 50 s were compared, isepamicin was found to be onefold more active than amikacin against M. abscessus and M. chelonae and as active as kanamycin against the 185 RGM isolates but sevenfold less active than amikacin against M. fortuitum. When MIC 90 s were compared, isepamicin was found to be onefold more active than amikacin against M. abscessus, onefold more active than kanamycin against M. fortuitum, and as active as amikacin and kanamycin against M. chelonae but onefold less active than kanamycin against M. abscessus and threefold less active than amikacin against M. fortuitum. Gentamicin exhibited limited activities (MIC 50 s, 16 to 32 µg/ml; MIC 90 s, 32 to 64 µg/ml) and streptomycin poor activities (MIC 50 s, 64 to 128 µg/ml; MIC 90 128 µg/ml) against each of the three RGM species. Tobramycin showed excellent activity against M. abscessus and limited to good activities against M. fortuitum and M. chelonae. Capreomycin showed good activity against M. fortuitum but poor activities against M. abscessus and M. chelonae, which is consistent with the results of Lévy-Frébault et al. ( 9 ).

TABLE 1. In vitro inhibitory activities of amikacin, isepamicin, kanamycin, tobramycin, gentamicin, streptomycin, and capreomycin against 117 isolates of M. abscessus, 48 isolates of M. fortuitum, and 20 isolates of M. chelonae

While none of the M. chelonae isolates tested had amikacin or isepamicin MICs of 32 µg/ml, 15 (13%) M. abscessus and 3 (6%) M. fortuitum isolates had amikacin and/or isepamicin MICs of 32 µg/ml (Table 2 ). Two M. abscessus isolates (CH10 and R31) were essentially resistant to all of the seven agents tested (MICs, 128 µg/ml). For the remaining 13 M. abscessus isolates, isepamicin was either as active as (2 isolates) or one- to threefold more active than (11 isolates) amikacin. Similar phenomena were observed with kanamycin and tobramycin. For the three M. fortuitum isolates with isepamicin MICs of 32 µg/ml, amikacin was 7- or 15-fold more active than isepamicin (Table 2 ). Isepamicin may be a good therapeutic option for RGM isolates that are nonsusceptible to amikacin, and vice versa.

TABLE 2. In vitro inhibitory activities of amikacin, isepamicin, kanamycin, tobramycin, gentamicin, streptomycin, and capreomycin against M. abscessus and M. fortuitum isolates that had amikacin MICs of 64 or 32 µg/ml and/or isepamicin MICs of 64 or 32 µg/ml

Because of the high prevalence of antimicrobial resistance in RGM in Taiwan ( 18 ), the use of a single agent for treatment is not recommended. Our study indicates that isepamicin, amikacin, and kanamycin exhibited excellent activities against RGM, and tobramycin exhibited excellent activity against M. abscessus. These antimicrobial agents can be used in the combination regimens for RGM. Isepamicin is particularly important since animal and clinical trials have shown that isepamicin is one of the less toxic aminoglycosides ( 3, 14 ). The activities of isepamicin against five M. chelonae and M. fortuitum strains were previously reported ( 2 ).

Ho et al. ( 7 ) found poor activities for amikacin, kanamycin, tobramycin, gentamicin, streptomycin, and capreomycin against M. chelonae and for kanamycin, tobramycin, streptomycin, and capreomycin against M. fortuitum. Only amikacin and gentamicin had good activities against M. fortuitum. Our results are largely different from theirs. The discrepancies may be due to differences in the methods of in vitro testing or the RGM strains used in their studies.


ACKNOWLEDGMENTS

We thank Po-Ren Hsueh for providing M. chelonae isolates.

This study was supported by grants from the Center for Disease Control (DOH 95-DC-1106) and the National Science Foundation (NSC 91-2316-B-005-003-CC3) of Taiwan.