Convenient test for screening metallo-beta-lactamase-producing gram-negative bacteria by using thiol compounds

Abstract

A simple disk diffusion test was constructed for detection of IMP-1-type metallo-beta-lactamase-producing gram-negative bacteria. Two Kirby-Bauer disks containing ceftazidime (CAZ) and a filter disk containing a metallo-beta-lactamase inhibitor were used in this test. Several IMP-1 inhibitors such as thiol compounds including 2-mercaptopropionic acid, heavy metal salts, and EDTA were evaluated for this test. Two CAZ disks were placed on a Mueller-Hinton agar plate on which a bacterial suspension was spread according to the method recommended by the National Committee for Clinical Laboratory Standards. The distance between the disks was kept to about 4 to 5 cm, and a filter disk containing a metallo-beta-lactamase inhibitor was placed near one of the CAZ disks within a center-to-center distance of 1.0 to 2.5 cm. For IMP-1-producing strains, the growth-inhibitory zone between the two disks expanded, while no evident change in the shape of the growth-inhibitory zone was observed for CAZ-resistant strains producing serine beta-lactamases such as AmpC or SHV-12. As a result, 2 to 3 microliter of undiluted 2-mercaptopropionic acid or mercaptoacetic acid able to block IMP-1 activity gave the most reproducible and clearest results, and CAZ-resistant strains producing AmpC or extended-spectrum beta-lactamases were distinguishable from IMP-1 producers by this test. A similar observation was made with IMP-1-producing clinical isolates such as Serratia marcescens, Klebsiella pneumoniae, Escherichia coli, Enterobacter cloacae, Enterobacter aerogenes, Citrobacter freundii, Proteus vulgaris, Pseudomonas aeruginosa, Pseudomonas putida, Acinetobacter spp., and Alcaligenes xylosoxidans. The specificity and sensitivity of this test were comparable to those of PCR analysis using bla(IMP)-specific primers. Therefore, this convenient test would be valuable for daily use in clinical laboratories.

FIG. 1

Inhibitory effects of 2-mercaptopropionic acid (2-MPA) on IMP-1 producers and non–IMP-1 producers. Three CAZ-resistant strains belonging to the gram-negative bacterial species P. aeruginosa, S. marcescens, and K. pneumoniae and producing IMP-1 metallo-β-lactamase or serine-β-lactamases (SHV-12 or AmpC) were tested. For each IMP-1 producer, a distinct growth-inhibitory zone appeared between the KB disk containing CAZ and the filter disk containing 2-MPA (left column). No change is evident around the two KB disks containing CAZ with or without 2-MPA for each serine β-lactamase producer (right column).

FIG. 2

Appearance of growth-inhibitory zone in IMP-1-producing strains by use of CAZ and 2-mercaptopropionic acid (2-MPA). Various levels of growth inhibition were observed in the IMP-1-producing gram-negative bacterial species tested. Marked growth inhibitions were observed in Acinetobacter sp., Alcaligenes xylosoxidans, Enterobacter aerogenes, E. coli, Proteus vulgaris, and Pseudomonas putida, whereas weak and ambiguous growth inhibitions were observed in C. freundii and E. cloacae.

FIG. 3

(A) Inhibitory effects of FeCl₂ on IMP-1 producers and non–IMP-1 producers. A slight expansion of growth-inhibitory zones between two disks was observed for all three IMP-1 producers (arrowheads). No change in the shape of the growth-inhibitory zone was evident for any serine-β-lactamase producer. (B) Inhibitory effects of EDTA on IMP-1 producers. Growth-inhibitory zones between two disks appeared for all three IMP-1 producers (arrowheads) when 5 μl of 500 mM EDTA solution was added to the filter.

FIG. 3

(A) Inhibitory effects of FeCl₂ on IMP-1 producers and non–IMP-1 producers. A slight expansion of growth-inhibitory zones between two disks was observed for all three IMP-1 producers (arrowheads). No change in the shape of the growth-inhibitory zone was evident for any serine-β-lactamase producer. (B) Inhibitory effects of EDTA on IMP-1 producers. Growth-inhibitory zones between two disks appeared for all three IMP-1 producers (arrowheads) when 5 μl of 500 mM EDTA solution was added to the filter.