In Vitro Susceptibilities of Burkholderia mallei in Comparison to Those of Other Pathogenic Burkholderia spp.

Abstract

The in vitro antimicrobial susceptibilities of isolates of Burkholderia mallei to 16 antibiotics were assessed and compared with the susceptibilities of Burkholderia pseudomallei and Burkholderia cepacia. The antibiotic susceptibility profile of B. mallei resembled that of B. pseudomallei more closely than that of B. cepacia, which corresponds to their similarities in terms of biochemistry, antigenicity, and pathogenicity. Ceftazidime, imipenem, doxycycline, and ciprofloxacin were active against both B. mallei and B. pseudomallei. Gentamicin was active against B. mallei but not against B. pseudomallei. Antibiotics clinically proven to be effective in the treatment of melioidosis may therefore be effective for treating glanders.

Burkholderia mallei is the causative agent of glanders, a rare disease of equines which can be transmitted to humans with fatal consequences. The incidence of the disease has declined worldwide, and it has disappeared in the Western world as the use of equines for work and transport has declined; nevertheless, cases do still occur in Asia, Africa, South America, and Central America (19, 21). As a consequence of its declining incidence, very little reference is made to the disease in modern microbiology texts, and data regarding antibiotic susceptibility, particularly susceptibility to modern antibiotics, are limited, with most of the published work appearing in Russian literature (1, 5, 6, 16, 18). The in vitro antibiotic susceptibilities of a number of strains of B. mallei were measured and compared to those of two other medically important Burkholderia species. Burkholderia pseudomallei is an opportunistic environmental pathogen, inhabiting soil, stagnant water, and rice paddies, that causes melioidosis, a major cause of human morbidity and mortality in certain areas of the tropics (9, 15). B. pseudomallei and B. mallei are similar, in that they share morphological, biochemical, and antigenic characteristics, and their disease manifestations are similar. Burkholderia cepacia is also an opportunistic environmental pathogen, and though it is virtually nonpathogenic in healthy patients, it causes respiratory infection in cystic fibrosis patients and occasionally nosocomial infection in patients in intensive care units (14).

All antibiotic powders used in this study were obtained from Sigma Chemical Co. (Poole, Dorset, United Kingdom), with the exception of ceftazidime (Glaxo, Uxbridge, Middlesex, United Kingdom), ciprofloxacin (Bayer UK, Newbury, United Kingdom), and azithromycin (Pfizer, Sandwich, Kent, United Kingdom). All antibiotics were prepared and stored in accordance with National Committee for Clinical Laboratory Standards (NCCLS) guidelines (20). For imipenem, Primaxin (Merck Sharp & Dohme, Hoddesdon, Hertfordshire, United Kingdom), which contains both imipenem and cilastatin (1:1, wt/wt), was used. For amoxicillin-clavulanic acid, Augmentin tablets (Beecham Research, Welwyn Garden City, Hertfordshire, United Kingdom) were crushed, dissolved in water, and then filtered. Although NCCLS guidelines indicate that parenteral preparations should not be used for susceptibility testing, the Augmentin tablets and Primaxin powder were used because they were available in-house at the time of the study. As this work had no implications for patients per se, and with the susceptibilities of B. pseudomallei and B. cepacia to imipenem and amoxicillin-clavulanic acid previously reported (11, 24), useful comparisons could be made. It was assumed that cilastatin alone had no activity against any of the test organisms. Ten strains of B. mallei were obtained from the American Type Culture Collection (ATCC), and a further seven strains were a kind gift from C. Wray at Central Veterinary Laboratories, Weybridge, United Kingdom. Clinical and environmental strains of B. pseudomallei were obtained from the National Collection of Type Cultures and the London School of Hygiene and Tropical Medicine, and one strain was a kind gift from P. Maynard at Blackburn Royal Infirmary, Blackburn, United Kingdom. All B. cepacia strains came from the ATCC and the National Collection of Type Cultures. Strains were stored at −80°C either in glycerol or on Protect beads (TSC Ltd., Heywood, Lancashire, United Kingdom). Bacteria were recovered prior to the experiment by either placing the glycerol stock in Mueller-Hinton broth or placing five or six beads in Mueller-Hinton broth and then incubating the cultures statically at 37°C for 24 h (48 h for B. mallei). All procedures with B. mallei and B. pseudomallei were carried out within biosafety level 3 containment.

A microtiter plate dilution method in accordance with NCCLS guidelines was used (20). Briefly, 96-well microtiter plates containing antibiotics ranging in concentration from 0.063 to 64 mg/liter were prepared in advance and stored at −20°C except for plates containing imipenem and ampicillin, which were freshly made on the day of the experiment. An inoculum of approximately 5 × 10⁵ CFU/ml (determined by using McFarland standard 5) in 100 μl was made from overnight or 48-h cultures in Mueller-Hinton broth and was added to all wells. Plates were incubated at 37°C for 18 to 20 h. Because of the organism’s slow-growing nature, it was necessary to incubate B. mallei plates for 36 h before reading the MICs.

Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 29213 (National Collection of Industrial and Marine Bacteria, Aberdeen, Scotland) were used as quality control standards. All MICs were within the range specified in reference 2.

The antibiotic susceptibility of B. mallei was similar to that of B. pseudomallei, with resistance to a number of antibiotics (Table 1). All strains of both organisms appeared to be sensitive to imipenem and doxycycline, while most strains were susceptible to ceftazidime, ciprofloxacin, and piperacillin. B. mallei was additionally sensitive to gentamicin, and some strains were sensitive to azithromycin. Azithromycin, imipenem, ceftazidime, and gentamicin showed good bactericidal activity at concentrations close to their respective MICs.

TABLE 1.

Comparative MICs of selected antibiotics for Burkholderia spp.

Antimicrobial agent	MIC (μg/ml) of antimicrobial agent fora:
	B. mallei			B. pseudomallei			B. cepacia
	Range	50%	90%	Range	50%	90%	Range	50%	90%
Ampicillin	1–>64	>64	>64	32–>64	64	>64	0.125–>64	>64	>64
Piperacillin	4–16	8	16	2–4	4	4	1–>64	8	>64
Amoxicillin-clavulanic acid	1–8	4	8	4–8	8	8	0.25–>64	>64	>64
Cefuroxime	8–>64	>64	>64	32–>64	64	>64	4–>64	64	>64
Ceftazidime	2–16	4	8	2–>64	2	>64b	1–64	2	64
Imipenem	0.125–0.25	0.25	0.25	0.5–1	0.5	1	1–16	8	16
Chloramphenicol	1–>64	64	>64	4–16	8	16	4–64	8	64
Doxycycline	0.125–4	0.5	2	0.5–1	0.5	1	0.125–32	16	32
Ofloxacin	0.5–8	2	8	1–8	4	8	0.25–8	2	8
Ciprofloxacin	0.25–8	1	8	0.5–4	2	4	0.125–4	0.5	4
Gentamicin	0.063–0.5	0.25	0.5	0.125–64	32	64	0.5–>64	>64	>64
Azithromycin	0.25–16	1	4	1–>64	64	>64	0.5–>64	64	>64
Rifampin	1–16	8	16	4–16	8	16	0.125–>32	8	32
Sulfamethoxazole	1–>64	64	>64	16–>64	16	>64	4–>64	>64	>64
Trimethoprim	0.125–64	8	32	4–>64	16	>64	0.125–64	1	>64
Co-trimoxazole	0.063–>64	16	>64	4–>64	16	>64	0.25–>64	4	>64

^a50% and 90%, MICs at which 50 and 90% of the isolates were inhibited, respectively. 

^bOne strain (a clinical isolate) was resistant. 

Although B. pseudomallei is also an environmental organism, it differs from B. cepacia in that it can cause disease in apparently healthy individuals (15). The organism does share a number of characteristics with B. mallei, including, it appears, antibiotic sensitivities. The in vitro MICs suggest that imipenem, doxycycline, ciprofloxacin, piperacillin, and ceftazidime may be useful in the treatment of both melioidosis and glanders (Table 1).

Clinical experience with melioidosis, however, has shown that despite good in vitro activity, an antibiotic may be ineffective, in particular, fluoroquinolone therapy trials have been disappointing (7, 10). Additionally, the choice of antibiotic depends on the presentation of the disease. Doxycycline, for example, is used alone for localized infection (12) and in combination with chloramphenicol for disseminated disease (8, 21). Co-trimoxazole is also recommended, but resistance to it is now common (21) and was evidenced by some of the strains used in this study. Ceftazidime is the favored antibiotic for therapy of systemic disease (23), however, resistant strains are beginning to appear (11). A clinical isolate from a patient treated with ceftazidime was found to be resistant to the antibiotic (Table 1), although it is not clear whether resistance arose during treatment. Imipenem was highly effective in vitro, which concurs with the findings of Smith et al. (22), who suggest that carbapenem antibiotics may be suitable candidates for therapy. Piperacillin may also be a suitable future candidate, particularly as concentrations in serum exceeding 100 μg/ml can be achieved (13).

The similarities between the two pathogens and their respective diseases suggest that the antibiotics that seem effective against melioidosis in vitro and are furthermore proven clinically would be effective against glanders. Additionally, the low in vitro MICs of gentamicin and azithromycin indicate that these two antibiotics may also be suitable for use in the treatment of glanders. These MICs must be interpreted carefully as B. pseudomallei is an intracellular pathogen (17) and B. mallei is likely to share its pathogenic mechanism. Although the MIC of gentamicin is low, the antibiotic does not penetrate intracellularly. In the case of azithromycin, the excellent intracellular penetration by the antibiotic (cellular/extracellular ratio, >7,000) (13) makes this a possible candidate antibiotic for treating glanders, and possibly melioidosis, even in cases caused by strains that require a high MIC of the antibiotic. The contrasting susceptibility of B. mallei and resistance of B. pseudomallei (with the exception of one strain) to gentamicin may in addition have some utility in a diagnostic test to distinguish between the two organisms.

The similarities between B. mallei and B. pseudomallei extend to their in vitro antibiotic susceptibilities. Antibiotics useful in the treatment of melioidosis may therefore be useful in treating glanders, although the disparity between in vitro susceptibility and clinical success in the treatment of melioidosis should also be considered for glanders.