Genetic relatedness and molecular characterization of multidrug resistant Acinetobacter baumannii isolated in central Ohio, USA

Abstract

Background: Over the last decade, nosocomial infections due to Acinetobacter baumannii have been described with an increasing trend towards multidrug resistance, mostly in intensive care units. The aim of the present study was to determine the clonal relatedness of clinical isolates and to elucidate the genetic basis of imipenem resistance.

Methods: A. baumannii isolates (n = 83) originated from two hospital settings in central Ohio were used in this study. Pulsed-field gel electrophoresis genotyping and antimicrobial susceptibility testing for clinically relevant antimicrobials were performed. Resistance determinants were characterized by using different phenotypic (accumulation assay for efflux) and genotypic (PCR, DNA sequencing, plasmid analysis and electroporation) approaches.

Results: The isolates were predominantly multidrug resistant (>79.5%) and comprised of thirteen unique pulsotypes, with genotype VII circulating in both hospitals. The presence of bla(OXA-23) in 13% (11/83) and IS(Aba1) linked bla(OXA-66) in 79.5% (66/83) of clinical isolates was associated with high level imipenem resistance. In this set of OXA producing isolates, multidrug resistance was bestowed by bla(ADC-25), class 1 integron-borne aminoglycoside modifying enzymes, presence of sense mutations in gyrA/parC and involvement of active efflux (with evidence for the presence of adeB efflux gene).

Conclusion: This study underscores the major role of carbapenem-hydrolyzing class D beta-lactamases, and in particular the acquired OXA-23, in the dissemination of imipenem-resistant A. baumannii. The co-occurrence of additional resistance determinant could also be a significant threat.

Figure 1

PFGE profiles of selected strains. A. baumannii isolates representing various resistance profiles from different hospitals and different units were genotyped. Figure is the representation of PFGE fingerprints of thirty one selected isolates. Thirteen genotypes (Group I to XIII) were identified in this study. The percentage of similarities was determined by the Dice's coefficient and UPGMA clustering. Major clusters were formed at the 75% similarity level. * Source abbreviations are MC (Ohio State University Medical Center) and CH (All central Ohio isolates derived from the State Public Health laboratory). The various ICUs in MC include Ross Heart hospital, James Cancer Hospital, Rhodes and Doans hospital and the Emergency Department.

Figure 2

Schematic representation of different types of gene cassettes identified in the Class 1 integrons in A. baumannii strains. Dotted lines represent the gene cassettes, oval circles the 59-base elements. The attI recombination site is shown by the hatched box. Locations of the 5'CS and 3'CS of class 1 integrons and those of the primer pairs, qacEΔ1-F, Sul1B and in-F (5'CS), in-B (3'CS), are shown in the bottom panel. Different types of variable region were found in our collection of isolates: Type I: accC1-orfX-orfX'-aadA1; n = 3, Type II: aacA4-catB8-aadA1, n = 26, Type III: aadB-aadA2; n = 3 and Type IV: aadB; n = 1. accC1 (3-N-aminoglycoside acetyltransferase) and aadB (2'-O-adenylyltransferase) confers gentamicin resistance, aadA1 and aadA2 (adenyltransferase) confers resistance to spectinomycin and streptomycin, aacA4 (6'-N-acetyltransferase) confers resistance to amikacin, netilmicin and tobramycin, catB8 (chloramphenicol acetyltransferase) confers resistance to chloramphenicol. The diagram was not drawn to scale.

Figure 3

Accumulation studies with ciprofloxacin. The fluorescence of the supernatant was measured with spectroflourimeter (LS 55 Fluorescence Spectrometer, 120 V, Perkin-Elmer model) at an excitation 275 nm and emission 440 nm for ciprofloxacin. The results for six representative MDR strains, namely, AC0003, AC0007, AC0008, AC0050, AC0062, AC0089 and one sensitive strain AC0030 are shown here. The graphs reflect the difference in fluorescence displayed by the bacterial cell in the presence and absence of efflux pump inhibitor CCCP. The arrow indicates the time of addition of CCCP. All experiments were carried out at least three times.

Figure 4

in vitro analysis of growth kinetics. The ability of different MDR clinical isolates namely AC0050, AC0062, AC0089 and sensitive strain AC0030, to grow in the presence of different concentrations of ciprofloxacin (A), nalidixic acid (B), either alone or in the presence of 25 μg/ml of CCCP was monitored in LB broth using a spectrophotometer. Range of results obtained for duplicate experiments are shown by error bars.