Persistence of Multidrug-Resistant Acinetobacter baumannii Isolates Harboring bla_OXA-23 and bap for 5 Years

Abstract

The emergence and dissemination of carbapenemase-producing Acinetobacter baumannii isolates have been reported worldwide, and A. baumannii isolates harboring bla_OXA-23 are often resistant to various antimicrobial agents. Antimicrobial resistance can be particularly strong for biofilm-forming A. baumannii isolates. We investigated the genetic basis for carbapenem resistance and biofilm-forming ability of multidrug-resistant (MDR) clinical isolates. Ninety-two MDR A. baumannii isolates were collected from one university hospital located in the Chungcheong area of Korea over a 5-year period. Multiplex PCR and DNA sequencing were performed to characterize carbapenemase and bap genes. Clonal characteristics were analyzed using REP-PCR. In addition, imaging and quantification of biofilms were performed using a crystal violet assay. All 92 MDR A. baumannii isolates involved in our study contained the bla_OXA-23 and bap genes. The average absorbance of biomass in Bap-producing strains was much greater than that in non-Bap-producing strains. In our study, only three REP-PCR types were found, and the isolates showing type A or type B were found more than 60 times among unique patients during the 5 years of surveillance. These results suggest that the isolates have persisted and colonized for 5 years, and biofilm formation ability has been responsible for their persistence and colonization.

Introduction

Acinetobacter baumannii is implicated in a variety of opportunistic nosocomial infections, including bacteremia, epidemic pneumonia, secondary meningitis, and urinary tract infections [23]. Treatments of infections caused by epidemic strains of A. baumannii are often extremely difficult because of the widespread resistance of strains to diverse antimicrobial agents. Resistance to various antimicrobial agents by these bacteria has usually resulted from intrinsic factors or acquisition of genes encoding antimicrobial resistance determinants. The antimicrobial resistance mechanisms include production of β-lactamases, production of aminoglycoside-modifying enzymes, decreased expression of outer membrane proteins, mutations in topoisomerase or gyrase, and overexpression of efflux pumps [20].

Carbapenems have become the drugs of choice in the treatment for severe Acinetobacter infections. However, during the last few years, the emergence and dissemination of carbapenem-resistant A. baumannii isolates have been reported worldwide. One set of carbapenem resistance strategies depends on the carbapenemase metallo-β-lactamases (MBLs); another set involves the OXA-type carbapenem-hydrolyzing β-lactamases (CHDLs) [18]. Subtypes of both MBLs (IMP-, GIM-, SIM-, SPM-, and VIM-type MBLs) and CHDLs (OXA-23-, OXA-40-, OXA-58-, and OXA-143-type CHDLs) have been reported in Acinetobacter worldwide. Among A. baumannii strains, increased carbapenem resistance rates have markedly limited therapeutic options. Particularly vexing are A. baumannii isolates producing OXA-23, which in addition to carbapenems, are often resistant to non-carbapenems, including aminoglycosides, cephalosporins, and fluoroquinolones [7]. Unfortunately, A. baumannii isolates forming biofilms usually exhibit enhanced antimicrobial resistance [22].

Biofilms are densely packed communities of bacteria and an extracellular polymeric substance. The extracellular material is generally composed of extracellular DNA, polysaccharides, and proteins [10]. Biofilm formation enhances the persistence of bacterial pathogens within adverse environments, including diverse antimicrobial agents, disinfectants, and host immune defenses. In addition, biofilms serve as reservoirs for bacterial dissemination to favorable conditions, and may facilitate exchanges of antimicrobial resistance genes among the bacterial residents [11]. In hospital environments, bacterial biofilms have developed on diverse surfaces such as those of artificial heart valves, catheters, intubation tubes, and cleaning instruments [5]. Diverse gene products seem to play a role in A. baumannii biofilm formation in vitro. Acinetobacter baumannii biofilm-associated protein (Bap), a high-molecular-weight protein, is one of the gene products that presents on the bacterial surface. Acinetobacter baumannii Bap plays an important role in biofilm maturation and stabilization on different substrates [14].

Although biofilm formation is important in persistence, colonization, and chronic infection by A. baumannii, few studies of A. baumannii biofilms are reported from Korea. The present study aims to determine the genetic basis for carbapenem resistance and to characterize the clonal relationships between multidrug-resistant (MDR) A. baumannii isolates obtained in Chungcheong, Korea, over a 5-year period. We also investigated bap gene prevalence and in vitro biofilm formation of MDR clinical isolates and assessed the correlation between colonization and biofilm formation of dominant MDR isolates.

 

Materials and Methods

Bacterial Strains

Ninety-two consecutive and non-duplicate MDR A. baumannii isolates were collected from one university hospital laboratory located in the Chungcheong area of Korea, from 2012 to 2016. Acinetobacter baumannii was identified using the VITEK 2 Automated Instrument ID System (bioMérieux, France); strain identities were confirmed by sequencing the partial rpoB housekeeping gene, as described previously [9].

Antimicrobial Susceptibility Testing

All A. baumannii isolates were subjected to a susceptibility test against 12 antimicrobial agents on Müller-Hinton agar (Difco Laboratories, MI, USA) with the Kirby-Bauer disk diffusion method, according to the CLSI M100-S21 guidelines [3]. The following antimicrobial disks (BBL, MI, USA) were used: amikacin (30 μg), ampicillin/sulbactam (10/10 μg), cefepime (30 μg), cefotaxime (30 μg), ceftazidime (30 μg), ciprofloxacin (5 μg), gentamicin (10 μg), imipenem (10 μg), meropenem (10 μg), and piperacillin/tazobactam (100/10 μg). Escherichia coli strain ATCC 25922 was used as a reference strain. Multidrug resistance was defined as resistance to three or more representatives of the following five drug classes: cephalosporins, carbapenems, β-lactam and β-lactam inhibitor combinations, fluoroquinolones, and aminoglycosides [15]. Resistance to one drug class was defined as resistance to all member antimicrobials of that class.

Characterization of Antimicrobial Resistance Determinants

The 92 MDR A. baumannii isolates were subjected to four multiplex PCR assays for the detection of carbapenemase genes. The four multiplex PCR assays are referred to as multi-1, multi-2, multi-3, and multi-4; each reaction included a specific primer set (Table 1) [6,19,24]. Total DNA was obtained from each target strain by using a DNA purification kit (SolGent, Korea) in accordance with standard protocols. Each multiplex PCR was performed in a total volume of 25 μl, using 50 ng of total DNA, 2.5 μl of 10× Taq buffer, 0.5 μl of 10 mM dNTP mix, 20 pmol of each primer, and 0.7 U of Taq DNA polymerase (SolGent). Each multiplex PCR was performed in a GeneAmp PCR System 9600 thermal cycler (Perkin-Elmer Cetus Corp., CT, USA) by predenaturation of the reaction mixture at 95℃ for 5 min, followed by 35 cycles of 95℃ for 30 sec, 52℃ for 40 sec, and 72℃ for 30 sec, with a final extension at 72℃ for 5 min. The amplified products were separated via electrophoresis on 1.5% (w/v) agarose gels containing ethidium bromide, and visualized using a BioDoc-14 imaging system (UVP, UK). All positive isolates identified using multiplex PCR were subjected to PCR assays using specific primers to identify types of carbapenemase genes [13].

Table 1.F, sense primer; R, antisense primer; D, A/G/T; Y, C/T

To investigate the bap gene, the 92 A. baumannii isolates were also subjected to PCR using bap-specific primers, 5’-TGCTGACAGTGACGTAGAACCACA-3’ and 5’-TGCAACTAGTGGAATAGCAGCCCA-3’ [2]. The amplicons were purified with a PCR purification kit (SolGent); these purified amplicons were sequenced using a BigDye Terminator Cycle Sequencing Kit (PE Applied Biosystems, CA, USA) and an ABI PRISM 3730XL DNA analyzer (PE Applied Biosystems). The various DNA sequences were confirmed using the BLAST paired alignment functionality.

Biofilm Imaging and Quantification

Imaging and quantification of the biofilm were performed on sterile polystyrene-bottomed 96-well plates using the crystal violet assay with some modifications [17]. Each well was filled with 100 μl of inoculum in brain heart infusion (BHI) broth, which had been prepared by suspending each 37℃ overnight BHI agarcultured isolate at 1 × 107 CFU/ml. Subsequently, the isolates were inoculated in triplicates and incubated at 37oC. After 18 h, the culture medium of each well was aspirated, and each well was washed three times with 200 ml of sterile phosphate-buffered saline. The biofilm formed in each well was stained with 100 μl of 1% crystal violet solution. After 10 min of staining, each well was washed three times with deionized water (10 min per wash). After the last wash, the p late was a ir d ried for 3 0 min. For imaging, microscopic observation of biofilms was performed using a phase contrast microscope (Meiji Techno, CA, USA). Subsequently, 100 μl of 95% ethanol was added to each well, to dissolve its biofilm. The optical density (OD) of each well at 560 nm was recorded using a Multiskan GO microplate spectrophotometer (Thermo Fisher Scientific, Germany). Each assay in this study was performed in triplicates, and the average values were calculated. In addition, the same methods were used to analyze biofilms produced by six reference strains. These reference strains consisted of noncarbapenemase A. baumannii isolates. Three of the reference isolates contained bap, and three did not.

Repetitive Extragenic Palindromic-PCR (REP-PCR) for Clonality Assessment

REP-PCR was conducted with a 50 μl reaction mixture containing 100 ng of chromosomal DNA, 5 μl of 10× Taq buffer, 1.0 μl of 10mM dNTP mix, 1.5 U of Taq DNA polymerase (SolGent), and 50 pmol of each of the primers REP1 (5’-IIIGCGCCGICATCAGGC-3’) and REP2 (5’-ACGTCTTATCAGGCCTAC-3’) [1]. The cycling conditions were as follows: initial denaturation occurred at 95℃ for 5 min; amplification occurred as 30 cycles of a 92℃ step for 50 sec, a 48℃ step for 55 sec, and a 70℃ step for 5 min; and a final extension step occurred at 70℃ for 10min. The amplified products were separated via electrophoresis on 1.5% agarose gels containing ethidium bromide, and visualized using a BioDoc-14 imaging system (UVP). Isolates belonging to the same clonal clusters showed identical REP-PCR profiles with the same band density and size.

 

Results

Characterization of MDR Clinical Isolates

Ninety-two MDR A. baumannii isolates were obtained from 92 patients over 5 years; isolates 1–8 were obtained in 2012; nos. 9–25, 2013; nos. 26–49, 2014; nos. 50–70, 2015; and nos. 71–92, 2016 (Table 2). The isolates were separated from diverse clinical specimens, including sputum (64.1%), urine (10.9%), wound (9.8%), bronchial aspiration (6.5%), blood (1.1%), and other (7.6%). The “other” indicates eye/ear/skin swabs, body fluids, and tissue specimens. The majority of isolates (70.6%) were separated from the respiratory system as sputa and bronchial aspirate. All isolates involved in this study possessed the A. baumannii bap gene, as well as the blaOXA-23 and blaOXA-51-like genes. No isolates possessed other carbapenemase genes other than blaOXA-23.

Table 2.AMK, amikacin; SAM, ampicillin/sulbactam; CEP, cefepime; CTX, cefotaxime; CAZ, ceftazidime; CIP, ciprofloxacin; CM, gentamicin; IPM, imipenem; MEM, meropenem; TZP, piperacillin/tazobactam; Bron-Asp, bronchial aspirate; S, susceptible; I, intermediate resistant; R, resistant; OD, optical density. aOD560 : Biofilm formation was quantified by measuring optical absorbance (560 nm) using crystal violet.

Biofilm Imaging and Quantification

Microscopic visualization was performed to verify biofilm formation for three randomly selected MDR A. baumannii isolates containing bap and blaOXA-23 genes, and three bap or three non-bap reference isolates. Biofilms of MDR A. baumannii isolates containing bap and blaOXA-23 genes, and bap reference isolates, covered greater surface areas, were thicker, were more abundant in matrix-like structures, and contained many more cells than did non-bap reference isolates. Non-bap reference isolates formed meager biofilms (Fig. 1).

Fig. 1.Imaging and quantification of biofilm formation by A. baumannii isolates using a phase contrast microscope and a 96-well microplate spectrophotometer, after crystal violet staining.

Biofilm formation was quantified by measuring optical absorbance (560 nm) for all 92 MDR A. baumannii, and the six reference strains, using crystal violet (Table 2). The mean absorbance of MDR isolates was 1.5 with a range from 0.7 to 2.3; the mean absorbance of A. baumannii isolates containing only bap was 1.4 with a range from 0.8 to 2.8. In contrast, the mean absorbance of A. baumannii isolates containing no bap was 0.7 with a range from 0.6 to 0.7.

REP-PCR Clusters

Epidemiological typing of the 92 MDR A. baumannii isolates was performed by repetitive REP-PCR. Only three REP-PCR types, designated types A, B, and C, were observed (Fig. 2). Two predominant REP-PCR types were type A (56.5%) and type B (32.6%); type A and B MDR isolates were collected over all 5 years (2012-2016). However, type C MDR isolates (10.9%) were only collected over the latter 3 years (2014-2016).

Fig. 2.Repetitive element sequence-based (REP)-PCR banding pattern among three clonal clusters of 92 multidrug-resistant A. baumannii strains, with 1-kb DNA size markers in the M lanes. The three clonal clusters were designated types A, B, and C. Type A contained bands a and b that showed different band densities and band c. Type B contained bands a’ and b’ that showed the same band density. Type C contained bands e and d’. Band d’ showed lower density than band d in types A and B.

 

Discussion

Acinetobacter baumannii is an opportunistic pathogen associated with multidrug resistance and nosocomial infections, resulting from an ability to colonize and persist in the hospital environment. During infection, colonization and persistence provide an isolate with the opportunity to develop antimicrobial resistance [11]. Colonization enables resistance to desiccation, resistance to physical disinfection, resistance to chemical disinfection, and multiple-month survival on inanimate surfaces. Biofilms play a key role in microorganism colonization during infection [14]. We investigated carbapenemase genes and their correlations with biofilm formation and colonization in MDR A. baumannii isolates collected over 5 years.

The blaOXA-23 gene is the sole carbapenemase gene detectable in all 92 MDR A. baumannii isolates. The disseminations of OXA-23-producing A. baumannii isolates associated with MDR have been previously reported in Korea and throughout the world [4,21]. In particular, OXA-23-type CHDL has been reported as the most prevalent of carbapenem-resistant A. baumannii isolates in Korea [12]. Interestingly, the isolates evaluated in the present study harbor bap. A recent report in the USA has studied A. baumannii isolates carrying both blaOXA-23 and bap genes, whereas there has been no report about that in Korea so far [15]. This is the first report about OXA-23- and Bap-co-producing A. baumannii isolates in Korea. Previous studies reported higher bap frequencies in endemic isolates than in their sporadic counterparts. We recently found that biofilm formation ability at the solid-liquid interfaces of A. baumannii isolates is at least three times higher than the ability of other Acinetobacter isolates [16]. In Australia, 92% of A. baumannii isolates from a single outbreak possess bap; in the USA, 84% of hospital isolates possess this gene [8,15]. Previously reported percentages are slightly lower than our results, because we selected from a subset of MDR A. baumannii isolates for detection of blaOXA-23 and bap genes, rather than from the entire set of clinical isolates. As we determined using the crystal violet staining technique, A. baumannii isolates carrying the bap gene and those without bap vary significantly in their biofilm formation capabilities. The average biomass absorbance of Bap-producing strains is much larger than that of non-Bap-producing strains, regardless of blaOXA-23 presence.

Clonal characteristic analysis of all 92 MDR A. baumannii isolates by REP-PCR reveals only three REP-PCR types (A, B, and C). In our study, only three REP-PCR patterns were exhibited. We suggest that diverse clonal clusters were not detected in our study because MDR A. baumannii isolates were collected from only one university hospital. REP-PCR types A and B occurred more than 30 times among unique patients over the 5 years of surveillance. This occurrence pattern suggests that the two clones have persisted and colonized for 5 years at one university hospital in Chungcheong, Korea. In addition, 10 isolates showing REP-PCR type C have been collected over the course of 3 years since 2014. Eventually, the 92 MDR isolates harboring bap and blaOXA-23 genes involved in this study belonged to three clones, which are all considered the endemic clone. The literature indicates that endemic isolates carry bap and blaOXA-23 genes with a higher frequency, and form thicker biofilms, than their sporadic counterparts [15].

In summary, our surveillance isolates were all endemic, MDR, and included bap and blaOXA-23 genes. The isolates have persisted and colonized for 5 years. Currently, there seems to be little correlation between the biofilm-forming ability of clinical isolates and epidemicity or endemicity of the isolates. However, our results suggest that biofilm formation ability by MDR A. baumannii isolates has been responsible for the persistence and colonization of the MDR strains. Additional investigations of A. baumannii are needed to characterize the correlation between the ability of the pathogen to form biofilms and its propensity to cause outbreaks and colonization.