Abstract

Objectives

Infections caused by multidrug-resistant (MDR) Acinetobacter baumannii are a challenging problem worldwide. Here, the molecular epidemiology and the genetic basis of antibiotic resistance in 111 MDR A. baumannii strains isolated from June 2005 to March 2009 from infected patients in 10 intensive care units (ICUs) in central Italy were investigated.

Methods

Epidemiological typing was performed by random amplification of polymorphic DNA, PCR-based sequence grouping and macrorestriction analysis. MICs of antibiotics were determined by the broth microdilution method. Genes for OXA carbapenemases, metallo-β-lactamases and the CarO porin were searched for by PCR.

Results

Molecular genotyping identified one predominant A. baumannii lineage, related to the international clonal lineage II, accounting for 95.6% of isolates. Isolates referable to this lineage were recovered from all ICUs surveyed and were resistant to nearly all classes of antimicrobials, with the exception of tigecycline and colistin. A high percentage (60.5%) of A. baumannii isolates showed elevated resistance to imipenem (MICs ≥ 128 mg/L), concomitant with resistance to meropenem. Carbapenem resistance was associated with the presence of either blaOXA-58-like (22.8%) or blaOXA-23-like (71.1%) carbapenemase genes. Molecular typing showed that the epidemic lineage encoding OXA-23 emerged in 2007 and displaced a genetically related clone encoding OXA-58 that had been responsible for previous ICU outbreaks in the same region.

Conclusions

Emergence of the OXA-23 epidemic lineage could result from selective advantage conferred by the blaOXA-23-like determinant, which provides increased resistance to carbapenems.

Introduction

Acinetobacter baumannii has emerged as an important nosocomial pathogen worldwide,1 especially in intensive care units (ICUs). Resistance to environmental stresses, persistence in the hospital setting and a propensity to develop stable antibiotic resistance make A. baumannii a common yet difficult-to-treat pathogen.2

Carbapenems have been widely used to treat serious infections associated with multidrug-resistant (MDR) A. baumannii, but a trend of increasing resistance to these drugs has been observed worldwide in the past decade.3 Several mechanisms are responsible for β-lactam resistance in A. baumannii, including: (i) the production of β-lactamases; (ii) alterations of porin-like proteins, resulting in decreased permeability to antibiotics; and (iii) the activity of efflux pumps, decreasing the intracellular concentration of antibiotics.4 Of the above mechanisms, carbapenem-hydrolysing β-lactamases (carbapenemases) belonging to the molecular class D OXA enzymes have emerged globally as the main mechanism responsible for carbapenem resistance in A. baumannii.5 Four groups of acquired class D β-lactamases have been described in A. baumannii, encoded by blaOXA-23-like, blaOXA-24-like, blaOXA-51-like and blaOXA-58-like genes, respectively.6 The blaOXA-51-like genes are chromosomally located in all A. baumannii strains.7 The blaOXA-58-like genes are found mostly on plasmids, whereas the blaOXA-23-like and blaOXA-24-like genes can be either chromosomally or plasmid located.6,8,9 Carbapenemase gene dosage can further contribute to determining the extent of carbapenem resistance in A. baumannii.8

Previous studies have reported the frequent isolation of OXA-58-producing A. baumannii in Italy, suggesting that blaOXA-58-carrying A. baumannii strains have become endemic.8,10–12 Conversely, reports concerning the blaOXA-24-like gene are limited to the description of two sporadic isolates recovered in 2000 and 2004 from two different Italian regions.9

In a previous survey we provided evidence of the emergence in 2004–05 of the blaOXA-58 gene in clinical A. baumannii isolates genetically related to the international lineage II and responsible for ICU outbreaks in the main hospitals of the Rome urban area.12 Later, the isolation of A. baumannii producing the OXA-23 carbapenemase from patients hospitalized in Rome13 and Genoa14 was reported.

The present study analysed the epidemiological evolution and changes in antibiotic susceptibility patterns for a large collection of A. baumannii isolates recovered during 2005–09 from 10 main ICUs in central Italy. Aims of the investigation were: (i) to evaluate the genetic relationships among the isolates and to strains representative of major international clonal lineages; (ii) to trace the evolution of antibiotic resistance; and (iii) to search for changes in OXA carbapenemase gene carriage among isolates.

Materials and methods

Bacterial isolates

As part of the surveillance activity of the Gruppo Romano Acinetobacter baumannii (GRAB),12 117 Acinetobacter isolates provisionally identified as A. baumannii were collected between 15 June 2005 and 24 March 2009 from 10 ICUs in central Italy and submitted to the coordinating laboratory at the National Institute for Infectious Diseases ‘Lazzaro Spallanzani’ (Rome) for further characterization (Table S1, available as Supplementary data at JAC Online). The geographical locations of the 10 hospitals from Central Italy (designated C, D, H, I, L, M, N, O, P and Q) are provided in Figure S1 (available as Supplementary data at JAC Online). Isolates were obtained either from epidemic clusters or sporadic cases. In total, 111 non-repetitive Acinetobacter isolates (i.e. one isolate per patient) were collected, all of which were from inpatients with clinically and laboratory-confirmed Acinetobacter infection. Six additional isolates were recovered from medical devices or the ICU environment of hospital N (Table S1). Acinetobacter isolates were not selected a priori because of MDR phenotype, though all but three of them were resistant to more than two of five antimicrobial classes (antipseudomonal cephalosporins, antipseudomonal carbapenems, β-lactam/β-lactamase inhibitor combinations, antipseudomonal fluoroquinolones and aminoglycosides) and were therefore considered to be MDR.2 In addition, seven index strains of A. baumannii (33, 50, 51, 57, 80, 82 and 105) from previously described outbreaks,12 two reference strains for international clonal lineages I (RUH875) and II (RUH134)15 and the prototype sequenced strain ACICU16 were included in the study for comparison purposes (Table S2, available as Supplementary data at JAC Online).

Bacteria were routinely identified to the species level by the participating centres using the Phoenix (Becton Dickinson, Sparks, MD, USA) and Vitek 2 (bioMérieux, Marcy l’Étoile, France) commercial systems. The source of Acinetobacter isolates is provided in Table S1.

Antimicrobial susceptibility testing

For ampicillin/sulbactam, piperacillin, piperacillin/tazobactam, cefepime, ceftazidime, aztreonam, meropenem, trimethoprim/sulfamethoxazole and colistin, susceptibility testing was performed using the Vitek 2 system. MICs of imipenem, amikacin, gentamicin, ciprofloxacin, levofloxacin, tetracycline and tigecycline were determined by the broth microdilution method, according to the CLSI protocol.17 All antimicrobials were obtained from Sigma-Aldrich (Milan, Italy), except tigecycline (Pfizer, New York, USA). Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 29213 were used as internal quality controls. MICs were interpreted according to the CLSI breakpoint criteria.17 No breakpoints were available for tigecycline.

Species assignment by amplified ribosomal DNA restriction analysis (ARDRA)

ARDRA was carried out with restriction enzymes AluI, CfoI, MboI, MspI and RsaI (Roche, Monza, Italy) and ribosomal DNA restriction patterns were interpreted as described previously.18,19

Epidemiological typing

Random amplified polymorphic DNA (RAPD) analysis was performed with the primer M13 as described previously.20 Macrorestriction analysis (MRA) of ApaI-digested genomes and PFGE was carried out as described previously,21 using a CHEF mapper (Bio-Rad, Segrate, Milan, Italy). An ApaI digest of genomic DNA from A. baumannii strain ACICU was included as the standard for gel run normalization. Electropherograms were analysed using BioNumerics software (Applied Maths, Sint-Martems-Latem, Belgium) with the Dice coefficient and the unweighted pair group method with averages (UPGMA) with a 1% tolerance limit and 1% optimization. Isolates that clustered with a similarity of >85% were considered to belong to the same RAPD or MRA type.22

To define the sequence group (SG), two multiplex PCRs designed to selectively amplify group 1 or group 2 alleles of the ompA, csuE and blaOXA-51-like genes were performed, and allelic profiles were interpreted according to Turton et al.23 and Towner et al.22

Molecular detection of OXA carbapenemase, metallo-β-lactamase and carO genes

The presence of the four groups of OXA carbapenemase genes (blaOXA-23-like, blaOXA-24-like, blaOXA-51-like and blaOXA-58-like) was detected by PCR using a multiplex assay.24,25 The occurrence of the ISAba1 element upstream of the blaOXA-23-like, blaOXA-51-like and blaOXA-58-like genes was determined by PCR as described previously.26 The identity of blaOXA-58-like and blaOXA-23-like amplicons was confirmed by direct DNA sequencing. PCR detection of the metallo-β-lactamase (blaIMP-1, blaVIM and blaSIM) was carried out as described previously.27–29

The carO gene was amplified using PCR Ready-to-Go Beads (GE Healthcare, Milan, Italy) with primers carO-FW (5′-TGACAACTACAGCTTTACTTGC-3′) and carO-RV (5′-CAACTGGC AACCATTTGT-3′). PCR conditions for carO were 94°C for 3 min, followed by 30 cycles of 94°C for 1 min, 53°C for 1 min and 72°C for 1 min, with final extension for 10 min at 72°C.

Copy number determination of the blaOXA-58-like gene

Total DNA was extracted from selected A. baumannii isolates using the Wizard Genomic DNA purification kit (Promega, Milan, Italy) according to the manufacturer's protocol. PCR amplification of the origin of replication of plasmid pACICU1 (repAci1 gene) was performed as described previously.8 The blaOXA-58-like gene copy number was determined using a long-range PCR assay (Qiagen, Milan, Italy) with forward primer 5′-CGATCAGAATGTTCAAGCGC-3′ and reverse primer 5′-GCGCTTGAACATTCTGATCG-3′ based on the pACICU1 plasmid sequence (accession number NC_010605.1). PCR mixtures comprised a volume of 50 µL containing 20 ng of purified DNA, 5 µL of 10× reaction buffer, 2.5 mM MgCl2, 500 µM of each dNTP, 0.4 µM of each primer and 2 U of long-range PCR enzyme. Cycling conditions for long-range PCR comprised 3 min at 93°C followed by 35 cycles of 15 s at 93°C, 30 s at 55°C and 10 min at 68°C. Based on the pACICU1 sequence, isolates with one, two or three copies of the blaOXA-58-like gene cassette gave no amplicon, a 4034 bp amplicon or a 8048 bp amplicon, respectively.

Results

Identification of Acinetobacter genospecies

One hundred and fourteen of the 117 submitted Acinetobacter isolates were definitively identified as A. baumannii by ARDRA, showing the typical profiles ‘11123’ (110 isolates; 94.0%), ‘11121’ (2 isolates; study codes 122 and 133; 1.71%) and ‘1112(1+3)’ (2 isolates; study codes 215 and 170; 1.71%) with CfoI, AluI, MboI, RsaI and MspI, respectively (Table S3, available as Supplementary data at JAC Online). All these isolates were also PCR-positive for a blaoxa-51-like gene, confirming their identity as A. baumannii. Three isolates, provisionally identified as A. baumannii, were assigned to the Acinetobacter calcoaceticus genospecies by ARDRA, and were PCR-negative for a blaoxa-51-like gene (Table S1).

A. baumannii genotyping and correlation with international clonal lineages

Dendrograms constructed following RAPD analysis and ApaI MRA showed one major cluster including 109 A. baumannii isolates (95.6%), defined by a similarity threshold >94% and >88% by RAPD analysis and MRA, respectively (Figure 1). This dominant lineage was genetically related to the blaOXA-58-positive clone previously isolated from ICU outbreaks in the Rome urban area (33, 80 and ACICU in Table S2)12 and to the international clonal lineage II (RUH134). Five isolates showed variant profiles (study codes 122, 133, 138, 170 and 215) and were unrelated to any international clonal lineages (Figure 1).

Figure 1.

Molecular typing of A. baumannii isolates. Clustering relationships inferred from RAPD analysis with the primer M13 and MRA. The dendrograms were generated using BioNumerics software (Applied Maths) using UPGMA and the Dice coefficient. Broken lines denote isolates belonging to the same RAPD or MRA type.22 Study codes (see Table S1) are shown on the right. Reference strains (see Table S2) are highlighted with grey shading.

Figure 1.

Molecular typing of A. baumannii isolates. Clustering relationships inferred from RAPD analysis with the primer M13 and MRA. The dendrograms were generated using BioNumerics software (Applied Maths) using UPGMA and the Dice coefficient. Broken lines denote isolates belonging to the same RAPD or MRA type.22 Study codes (see Table S1) are shown on the right. Reference strains (see Table S2) are highlighted with grey shading.

Multiplex PCRs for SG assignment yielded the 111 allelic profile (corresponding to SG 1) for 101 of 109 A. baumannii isolates (92.7%), in line with their genetic relatedness to the international clonal lineage II. The remaining eight A. baumannii isolates, which were genetically related to the international clonal lineage II, belonged to SG 4 (study codes 162, 163 and 164) or showed a new combination of amplicons in SG analysis (study codes 207, 217, 221, 233 and 234). Table S3 summarizes the relationship between SG, RAPD type and MRA type for each isolate.

Antimicrobial susceptibility

The MDR phenotype was observed for 111 of 114 (97.4%) A. baumannii isolates. These were characterized by resistance to piperacillin (97.4%), piperacillin/tazobactam (95.6%), cefepime (97.4%), ceftazidime (97.4%), aztreonam (100%), imipenem (93.8%), meropenem (86.0%), amikacin (97.4%), gentamicin (77.2%), ciprofloxacin (97.4%), levofloxacin (97.4%) and trimethoprim/sulfamethoxazole (95.6%) (Figure 2a). Remarkably, a significant percentage of A. baumannii isolates was susceptible to ampicillin/sulbactam (48.3%), and all isolates were susceptible to colistin (Figure 2a). The distribution of MICs of tigecycline and imipenem for the 114 A. baumannii isolates is shown in Figure 2(b and c). MIC50 values of tigecycline and imipenem were 0.5 mg/L and 128 mg/L, respectively. Of note, 60.5% of carbapenem-resistant A. baumannii isolates showed high-level resistance to imipenem with MICs ≥128 mg/L. Ninety-eight A. baumannii isolates were resistant to both imipenem and meropenem (86.0%), but nine (7.9%) were susceptible to meropenem and resistant to imipenem while seven (6.1%) were susceptible to both drugs.

Figure 2.

Analysis of antimicrobial susceptibility. (a) Antibiotic resistance for 114 A. baumannii isolates. Isolates showing an intermediate level of susceptibility were classified as resistant. (b and c) Distribution of MIC values of TGC and IPM, respectively, for 114 A. baumannii isolates. SAM, ampicillin/sulbactam; PIP, piperacillin; TZP, piperacillin/tazobactam; FEP, cefepime; CAZ, ceftazidime; ATM, aztreonam; IPM, imipenem; MEM, meropenem; AMK, amikacin; GEN, gentamicin; CIP, ciprofloxacin; LVX, levofloxacin; SXT, trimethoprim/sulfamethoxazole; CST, colistin; TGC, tigecycline.

Figure 2.

Analysis of antimicrobial susceptibility. (a) Antibiotic resistance for 114 A. baumannii isolates. Isolates showing an intermediate level of susceptibility were classified as resistant. (b and c) Distribution of MIC values of TGC and IPM, respectively, for 114 A. baumannii isolates. SAM, ampicillin/sulbactam; PIP, piperacillin; TZP, piperacillin/tazobactam; FEP, cefepime; CAZ, ceftazidime; ATM, aztreonam; IPM, imipenem; MEM, meropenem; AMK, amikacin; GEN, gentamicin; CIP, ciprofloxacin; LVX, levofloxacin; SXT, trimethoprim/sulfamethoxazole; CST, colistin; TGC, tigecycline.

Genetic determinants for carbapenem resistance

Carbapenem resistance was associated with the presence of two carbapenem-hydrolysing oxacillinase genes, namely blaOXA-23-like (81 of 114 isolates; 71.1%) and blaOXA-58-like (26 of 114 isolates; 22.8%) (Table 1). None of the isolates carried both determinants (Table S3). The ISAba1 element was always present upstream of the blaOXA-23-like gene, but it was never detected upstream of the blaOXA-51-like and blaOXA-58-like genes. No amplicons were obtained with primers targeting the blaOXA-24-like, blaIMP-1, blaVIM and blaSIM genes. The seven carbapenem-susceptible A. baumannii isolates were negative for all β-lactamase genes, except for a blaOXA-51-like gene (Table 1).

Table 1.

Relationship between presence of blaOXA-23-like or blaOXA-58-like carbapenemase genes, MEM and IPM resistance, and presence of A. baumannii pACICU1 origin of replication

No. (%) of isolates with
 
 resistance phenotype
 
 
carbapenemase gene MEM IPM Ori pACICU1 
blaOXA-23-like, 81 (71.1) 81 (71.1) 81 (71.1) 81 (71.1) 
blaOXA-58-like, 26 (22.8) 17 (14.9) 26 (22.8) 26 (22.8) 
None, 7 (6.1) 2 (1.8) 
No. (%) of isolates with
 
 resistance phenotype
 
 
carbapenemase gene MEM IPM Ori pACICU1 
blaOXA-23-like, 81 (71.1) 81 (71.1) 81 (71.1) 81 (71.1) 
blaOXA-58-like, 26 (22.8) 17 (14.9) 26 (22.8) 26 (22.8) 
None, 7 (6.1) 2 (1.8) 

MEM, meropenem; IPM, imipenem; Ori pACICU1, origin of pACICU1 replication.

Isolates showing an intermediate level of susceptibility to IPM or MEM are reported as resistant.

Changes in carbapenemase gene carriage and influence on carbapenem resistance

Molecular typing revealed that blaOXA-23-like- and blaOXA-58-like-positive isolates had identical or similar RAPD and MRA patterns, and all were genetically related to international clonal lineage II (Figure 1). The distribution of blaOXA-23-like- and blaOXA-58-like-positive isolates per year is shown in Figure 3(a). In April 2007 the first A. baumannii isolate testing positive for a blaOXA-23-like gene was isolated from hospital N. The OXA-23 determinant became prevalent in 2008 and replaced the OXA-58 determinant thereafter. Only two blaOXA-58-like-positive isolates were recovered in 2008 and none in 2009 (Figure 3a).

Figure 3.

Evolution of carbapenemase gene carriage and influence on carbapenem resistance. (a) Distribution of blaOXA-23-like- and blaOXA-58-like-positive isolates per year and month. (b) Distribution of IPM MIC values during the 5 year investigation period. *Ratio of blaOXA-58-positive isolates that were meropenem susceptible and IPM resistant; all blaOXA-23-positive isolates were meropenem resistant and IPM resistant. IPM, imipenem.

Figure 3.

Evolution of carbapenemase gene carriage and influence on carbapenem resistance. (a) Distribution of blaOXA-23-like- and blaOXA-58-like-positive isolates per year and month. (b) Distribution of IPM MIC values during the 5 year investigation period. *Ratio of blaOXA-58-positive isolates that were meropenem susceptible and IPM resistant; all blaOXA-23-positive isolates were meropenem resistant and IPM resistant. IPM, imipenem.

The levels of imipenem and meropenem resistance associated with the two predominant carbapenemases were remarkably different. All 81 blaOXA-23-like-positive isolates were resistant to both imipenem and meropenem, whereas only 17 of 26 blaOXA-58-like-positive isolates were resistant to both drugs (Table 1). The presence of an OXA-23 carbapenemase determinant was associated more frequently with high-level imipenem resistance (MIC ≥ 128 mg/L for 63 of 81 isolates, 77.8%) compared with an OXA-58 determinant (MIC ≥128 mg/L for 6 of 26 isolates, 23.1%) (Figure 3b).

The origin of pACICU1 replication (repAci1) was detected in all A. baumannii isolates carrying a blaOXA-58- or blaOXA-23-like gene, as well as in two carbapenem-susceptible isolates that tested negative for carbapenemase genes (study codes 133 and 136) (Table 1).

Resistance to meropenem in the blaOXA-58-like-positive isolates is associated with increased levels of resistance to other antibiotics

An intriguing observation was the variable level of resistance to meropenem among the 26 blaOXA-58-like-positive imipenem-resistant isolates. One copy of the blaOXA-58-like gene was identified in 22 of 26 isolates, irrespective of their meropenem resistance or susceptibility (Table S3). Two or three copies of the blaOXA-58-like gene were detected in four isolates (study codes 110 and 174, and study codes 123 and 137, respectively) (Figure S2, panel A; available as Supplementary data at JAC Online). However, a multicopy blaOXA-58-like gene was not associated with a high level of imipenem resistance (MIC ≥128 mg/L) or with meropenem resistance. For instance, isolate 137 was susceptible to meropenem and showed an imipenem MIC of 32 mg/L although carrying three copies of a blaOXA-58-like gene (Table S3).

All blaOXA-58-like-positive isolates, whether susceptible or resistant to meropenem, yielded a carO PCR product of the expected size of 650 bp (data not shown), irrespective of meropenem susceptibility level. Thus, meropenem resistance could not be associated with loss of integrity of the carO porin gene.

It has been previously suggested that overexpression of drug efflux systems contributes to carbapenem resistance in A. baumannii.6,30,31 As shown in Table 2 and Figure S2 (panel B), resistance to meropenem in blaOXA-58-like-positive isolates was associated with increased MICs of imipenem, tigecycline, amikacin, gentamicin, levofloxacin, ciprofloxacin, tetracycline and ethidium bromide, all of which are known substrates of multidrug efflux pumps.6,30–32 These results suggest that overexpression of efflux systems could be contributing to meropenem resistance in the blaOXA-58-like-positive isolates.

Table 2.

Analysis of MICs (mean ± SD, median and mode values) of representative antimicrobials for 26 blaOXA-58-like-positive A. baumannii isolates showing MEM-R and MEM-S phenotypes

 MIC of antimicrobial
 
MEM susceptibility phenotype (no. of isolates) IPM TGC AMK GENa LVX CIP TET EtBr 
R (17)         
 mean ± SD 55.1 ± 43.2 1.44 ± 0.72 90.2 ± 79.6 10.2 ± 8.9 11.5 ± 4.2 48.0 ± 33.9 24.9 ± 8.4 64.0 ± 0 
 median 64 64 32 32 64 
 mode 32 32 32 32 64 
S (9)         
 mean ± SD 27.5 ± 9.0 1.22 ± 0.66 16.2 ± 20.1 3.4 ± 1.2b 8.9 ± 2.7 15.1 ± 7.4 16.0 ± 0 39.1 ± 14.1 
    59.9 ± 169.5c     
 median 32 16 16 32 
 mode 32 16 16 32 
 MIC of antimicrobial
 
MEM susceptibility phenotype (no. of isolates) IPM TGC AMK GENa LVX CIP TET EtBr 
R (17)         
 mean ± SD 55.1 ± 43.2 1.44 ± 0.72 90.2 ± 79.6 10.2 ± 8.9 11.5 ± 4.2 48.0 ± 33.9 24.9 ± 8.4 64.0 ± 0 
 median 64 64 32 32 64 
 mode 32 32 32 32 64 
S (9)         
 mean ± SD 27.5 ± 9.0 1.22 ± 0.66 16.2 ± 20.1 3.4 ± 1.2b 8.9 ± 2.7 15.1 ± 7.4 16.0 ± 0 39.1 ± 14.1 
    59.9 ± 169.5c     
 median 32 16 16 32 
 mode 32 16 16 32 

MEM, meropenem; R, resistant; S, susceptible; IPM, imipenem; TGC, tigecycline; AMK, amikacin; GEN, gentamicin; LVX, levofloxacin; CIP, ciprofloxacin; TET, tetracycline; EtBr, ethidium bromide.

aStudy code 138 (Table S1) showed a very high level of resistance to GEN (>512 mg/L).

bMIC of GEN calculated excluding the isolate with study code 138.

cMIC of GEN calculated including the isolate with study code 138.

Discussion

A progressive decline in susceptibility rates to β-lactams has been documented in A. baumannii during the last decade.33 This emphasizes the need to monitor the diffusion of MDR A. baumannii strains by means of molecular epidemiology tools.

Despite the urgent infection control measures recommended by GRAB as a consequence of the 2004–05 survey,12 the number of carbapenem-resistant A. baumannii isolates from 10 ICUs in central Italy continued to increase between 2005 and 2009. Epidemiological typing highlighted the emergence of an epidemic MDR lineage accounting for 95.6% of all isolates. Of note, the emerging lineage was genetically related to the carbapenem-resistant A. baumannii strains responsible for multiple outbreaks occurring in the Rome urban area during 2004–05,12 and belonged to the A. baumannii international clonal lineage II. Accordingly, all isolates assigned to the epidemic A. baumannii lineage belonged to SG 1. Moreover, the epidemic lineage was isolated from all ICUs surveyed and was characterized by resistance to nearly all classes of antimicrobials, in addition to carbapenems.

The present study revealed that carbapenem resistance was driven by the dissemination of two OXA-type carbapenemase genes, namely blaOXA-23 (71.1% of isolates) and blaOXA-58 (22.8% of isolates). However, by comparison with data from our 2004–05 survey,12 the prevalence of blaOXA-58-carrying isolates appeared to be drastically reduced (22.8% in 2005–09 versus 80.7% in 2004–05), while the OXA-23-like determinant become prevalent in 2008 and completely replaced the OXA-58-like determinant thereafter.

The blaOXA-23 carbapenemase gene has increasingly been reported worldwide.5 In several studies13,34,35 the blaOXA-23-like- or blaOXA-58-like-positive A. baumannii isolates belonged to apparently unrelated MRA types and were distinct from the worldwide clonal lineages I, II and III. Only recently, the blaOXA-23-like gene has been detected among A. baumannii isolates genetically related to the international clonal lineages I or II.14,36 The present study revealed that isolates carrying blaOXA-23-like or blaOXA-58-like genes have a strict genetic correlation and both belonged to the international clonal lineage II, as did the blaOXA-58-positive clone responsible for ICU outbreaks in the Rome urban area during 2004–05.12 It can therefore be speculated that acquisition of the blaOXA-23-like determinant by the new epidemic lineage may provide A. baumannii with a selective advantage over the pre-existing blaOXA-58-positive clone by increasing resistance to both imipenem and meropenem. Given the high carbapenemase activity of OXA-23, the presence of the blaOXA-23-like gene could determine full resistance to both imipenem and meropenem.37 Interestingly, all A. baumannii isolates carrying the blaOXA-23-like gene possessed the pACICU1 replication origin but lacked the plasmid-borne blaOXA-58-like gene. This holds true also for two carbapenem-susceptible isolates (study codes 133 and 136), and thus precludes the establishment of any association between the presence of the repAci1 origin of replication and the presence of the carbapenemase genes.

Analysis of imipenem susceptibility during the 5 year investigation period (2005–09) demonstrated a gradual yet significant increase in resistance, coherent with the spread of isolates positive for blaOXA-23-like carbapenemase, while the pattern and extent of resistance to other drugs did not show remarkable differences. Of note, OXA-23 clonal isolates showed 100% resistance to meropenem and high imipenem MICs (≥128 mg/L). Conversely, meropenem activity was superior to that of imipenem in nearly 35% of OXA-58 clonal isolates. Since the OXA-58 enzyme hydrolyses imipenem more efficiently than meropenem,37 differences in enzymatic activity can partly explain the higher in vitro activity of meropenem observed in the blaOXA-58-like-positive group. However, additional mechanisms can determine the extent of carbapenem resistance in A. baumannii, including modification of the CarO porin for carbapenem influx,29 and plasmid and/or carbapenemase gene copy number.6,8 In the present study, resistance to meropenem was not linked to blaOXA-58-like gene copy number per plasmid or to loss of integrity of the CarO porin, at least in the blaOXA-58-like-positive isolates. In addition, broad-specificity multidrug efflux systems can mediate resistance to a wide range of antibiotics,32 including carbapenems.6,30,31 Interestingly, we observed that resistance to meropenem was associated with increased levels of resistance to several antibiotics and ethidium bromide, all known for being substrates of multidrug efflux systems. Therefore, it can be hypothesized that overexpression of efflux pumps in the blaOXA-58-like-positive isolates would confer resistance to meropenem, while concomitantly increasing the extent of resistance to imipenem and other antibiotics.

By comparison with our previous survey (2004–05),12 significantly decreased resistance to tigecycline was observed (MIC50 2 mg/L in 2004–05 versus 0.5 mg/L in 2005–09), concomitant with increased resistance to gentamicin (50.0% resistance in 2004–05 versus 77.2% in 2005–09). Colistin continued to exert excellent activity (100% susceptibility), whereas ampicillin/sulbactam remained effective for approximately half of MDR A. baumannii isolates from central Italy.

In conclusion, this study updates current knowledge on the prevalence of MDR A. baumannii in Italy and highlights the emergence and epidemic spread of a carbapenem-resistant epidemic lineage carrying the blaOXA-23-like determinant. The emerging OXA-23 lineage replaced a pre-existing OXA-58 clone,12 probably because of the selective advantage provided by increased resistance to carbapenems. This temporal succession of carbapenemase gene carriage appears to parallel the more recent clinical use of meropenem compared with imipenem.

Funding

This work was supported by a grant from Ricerca Corrente 2008 of the National Institute for Infectious Diseases ‘Lazzaro Spallanzani’ to P. V.

Transparency declarations

None to declare.

Supplementary data

Tables S1 to S3 and Figures S1 and S2 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).

Acknowledgements

We thank Gruppo Romano Acinetobacter baumannii (M. Ballardini, S. Bartolini, E. Bordi, M. Galiè, R. Minniti, M. Meledandri, L. Pacciani, G. Parisi, G. Prignano, C. Santini, M. Valmarin, M. Venditti and S. Ziantoni) for providing A. baumannii isolates. Pfizer (New York, USA) kindly provided laboratory-grade tigecycline.

References

1
Munoz-Price
LS
Weinstein
RA
Acinetobacter infection
N Engl J Med
 , 
2008
, vol. 
358
 (pg. 
1271
-
81
)
2
Peleg
AY
Seifert
H
Paterson
DL
Acinetobacter baumannii: emergence of a successful pathogen
Clin Microbiol Rev
 , 
2008
, vol. 
21
 (pg. 
538
-
82
)
3
Garnacho-Montero
J
Amaya-Villar
R
Multiresistant Acinetobacter baumannii infections: epidemiology and management
Curr Opin Infect Dis
 , 
2010
, vol. 
23
 (pg. 
332
-
9
)
4
Perez
F
Hujer
AM
Hujer
KM
, et al.  . 
Global challenge of multidrug-resistant Acinetobacter baumannii
Antimicrob Agents Chemother
 , 
2007
, vol. 
51
 (pg. 
3471
-
84
)
5
Higgins
PG
Dammhayn
C
Hackel
M
, et al.  . 
Global spread of carbapenem-resistant Acinetobacter baumannii
J Antimicrob Chemother
 , 
2010
, vol. 
65
 (pg. 
233
-
8
Erratum in: J Antimicrob Chemother 2010; 65: 1317
6
Poirel
L
Nordmann
P
Carbapenem resistance in Acinetobacter baumannii: mechanisms and epidemiology
Clin Microbiol Infect
 , 
2006
, vol. 
12
 (pg. 
826
-
36
)
7
Turton
JF
Woodford
N
Glover
J
, et al.  . 
Identification of Acinetobacter baumannii by detection of the blaOXA-51-like carbapenemase gene intrinsic to this species
J Clin Microbiol
 , 
2006
, vol. 
44
 (pg. 
2974
-
6
)
8
Bertini
A
Poirel
L
Bernabeu
S
, et al.  . 
Multicopy blaOXA-58 gene as a source of high-level resistance to carbapenems in Acinetobacter baumannii
Antimicrob Agents Chemother
 , 
2007
, vol. 
51
 (pg. 
2324
-
8
)
9
D'Andrea
MM
Giani
T
D'Arezzo
S
, et al.  . 
Characterization of pABVA01, a plasmid encoding the OXA-24 carbapenemase from Italian isolates of Acinetobacter baumannii
Antimicrob Agents Chemother
 , 
2009
, vol. 
53
 (pg. 
3528
-
33
)
10
Giordano
A
Varesi
P
Bertini
A
, et al.  . 
Outbreak of Acinetobacter baumannii producing the carbapenem-hydrolyzing oxacillinase OXA-58 in Rome, Italy
Microb Drug Resist
 , 
2007
, vol. 
13
 (pg. 
37
-
43
)
11
Zarrilli
R
Vitale
D
Di Popolo
A
, et al.  . 
A plasmid-borne blaOXA-58 gene confers imipenem resistance to Acinetobacter baumannii isolates from a Lebanese hospital
Antimicrob Agents Chemother
 , 
2008
, vol. 
52
 (pg. 
4115
-
20
)
12
D'Arezzo
S
Capone
A
Petrosillo
N
, et al.  . 
Epidemic multidrug-resistant Acinetobacter baumannii related to European clonal types I and II in Rome (Italy)
Clin Microbiol Infect
 , 
2009
, vol. 
15
 (pg. 
347
-
57
)
13
Mendes
RE
Spanu
T
Deshpande
L
, et al.  . 
Clonal dissemination of two clusters of Acinetobacter baumannii producing OXA-23 or OXA-58 in Rome, Italy
Clin Microbiol Infect
 , 
2009
, vol. 
15
 (pg. 
588
-
92
)
14
Di Popolo
A
Giannouli
M
Triassi
M
, et al.  . 
Molecular epidemiological investigation of multidrug-resistant Acinetobacter baumannii strains in four Mediterranean countries with a multilocus sequence typing scheme
Clin Microbiol Infect
 , 
2010
 
doi:10.1111/j.1469-0691.2010.03254.x
15
Nemec
A
Dijkshoorn
L
van der Reijden
TJ
Long-term predominance of two pan-European clones among multi-resistant Acinetobacter baumannii strains in the Czech Republic
J Med Microbiol
 , 
2004
, vol. 
53
 (pg. 
147
-
53
)
16
Iacono
M
Villa
L
Fortini
D
, et al.  . 
Whole-genome pyrosequencing of an epidemic multidrug-resistant Acinetobacter baumannii strain belonging to the European clone II group
Antimicrob Agents Chemother
 , 
2008
, vol. 
52
 (pg. 
2616
-
25
)
17
Clinical and Laboratory Standard Institute
Performance Standards for Antimicrobial Susceptibility Testing: Seventeenth Informational Supplement M100-S17
 , 
2007
Wayne, PA, USA
CLSI
18
Dijkshoorn
L
Van Harsselaar
B
Tjernberg
I
, et al.  . 
Evaluation of amplified ribosomal DNA restriction analysis for identification of Acinetobacter genomic species
Syst Appl Microbiol
 , 
1998
, vol. 
21
 (pg. 
33
-
9
)
19
Vaneechoutte
M
Dijkshoorn
L
Tjernberg
I
, et al.  . 
Identification of Acinetobacter genomic species by amplified ribosomal DNA restriction analysis
J Clin Microbiol
 , 
1995
, vol. 
33
 (pg. 
11
-
5
)
20
Grundmann
HJ
Towner
KJ
Dijkshoorn
L
, et al.  . 
Multicenter study using standardized protocols and reagents for evaluation of reproducibility of PCR-based fingerprinting of Acinetobacter spp
J Clin Microbiol
 , 
1997
, vol. 
35
 (pg. 
3071
-
7
)
21
Seifert
H
Dolzani
L
Bressan
R
, et al.  . 
Standardization and interlaboratory reproducibility assessment of pulsed-field gel electrophoresis-generated fingerprints of Acinetobacter baumannii
J Clin Microbiol
 , 
2005
, vol. 
43
 (pg. 
4328
-
35
)
22
Towner
KJ
Levi
K
Vlassiadi
M
, et al.  . 
Genetic diversity of carbapenem-resistant isolates of Acinetobacter baumannii in Europe
Clin Microbiol Infect
 , 
2008
, vol. 
14
 (pg. 
161
-
7
)
23
Turton
JF
Gabriel
SN
Valderrey
C
, et al.  . 
Use of sequence-based typing and multiplex PCR to identify clonal lineages of outbreak strains of Acinetobacter baumannii
Clin Microbiol Infect
 , 
2007
, vol. 
13
 (pg. 
807
-
15
)
24
Poirel
L
Marque
S
Heritier
C
, et al.  . 
OXA-58, a novel class D β-lactamase involved in resistance to carbapenems in Acinetobacter baumannii
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
202
-
8
)
25
Woodford
N
Ellington
MJ
Coelho
JM
, et al.  . 
Multiplex PCR for genes encoding prevalent OXA carbapenemases in Acinetobacter spp
Int J Antimicrob Agents
 , 
2006
, vol. 
27
 (pg. 
351
-
3
)
26
Turton
JF
Ward
ME
Woodford
N
, et al.  . 
The role of ISAba1 in expression of OXA carbapenemase genes in Acinetobacter baumannii
FEMS Microbiol Lett
 , 
2006
, vol. 
258
 (pg. 
72
-
7
)
27
Zong
Z
X
Valenzuela
JK
, et al.  . 
An outbreak of carbapenem-resistant Acinetobacter baumannii producing OXA-23 carbapenemase in western China
Int J Antimicrob Agents
 , 
2008
, vol. 
31
 (pg. 
50
-
4
)
28
Lee
K
Yum
JH
Yong
D
, et al.  . 
Novel acquired metallo-β-lactamase gene, blaSIM-1, in a class 1 integron from Acinetobacter baumannii clinical isolates from Korea
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
4485
-
91
)
29
Mussi
MA
Limansky
AS
Viale
AM
Acquisition of resistance to carbapenems in multidrug-resistant clinical strains of Acinetobacter baumannii: natural insertional inactivation of a gene encoding a member of a novel family of β-barrel outer membrane proteins
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
1432
-
40
)
30
Higgins
PG
Wisplinghoff
H
Stefanik
D
, et al.  . 
Selection of topoisomerase mutations and overexpression of adeB mRNA transcripts during an outbreak of Acinetobacter baumannii
J Antimicrob Chemother
 , 
2004
, vol. 
54
 (pg. 
821
-
3
)
31
Héritier
C
Poirel
L
Lambert
T
, et al.  . 
Contribution of acquired carbapenem-hydrolyzing oxacillinases to carbapenem resistance in Acinetobacter baumannii
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
3198
-
202
)
32
Magnet
S
Courvalin
P
Lambert
T
Resistance-nodulation-cell division-type efflux pump involved in aminoglycoside resistance in Acinetobacter baumannii strain BM4454
Antimicrob Agents Chemother
 , 
2001
, vol. 
45
 (pg. 
3375
-
80
)
33
Karageorgopoulos
DE
Falagas
ME
Current control and treatment of multidrug-resistant Acinetobacter baumannii infections
Lancet Infect Dis
 , 
2008
, vol. 
8
 (pg. 
751
-
62
)
34
Gur
D
Korten
V
Unal
S
, et al.  . 
Increasing carbapenem resistance due to the clonal dissemination of oxacillinase (OXA-23 and OXA-58)-producing Acinetobacter baumannii: report from the Turkish SENTRY Program sites
J Med Microbiol
 , 
2008
, vol. 
57
 (pg. 
1529
-
32
)
35
Chu
YW
Cheung
TK
Chu
MY
, et al.  . 
OXA-23-type imipenem resistance in Acinetobacter baumannii in Hong Kong
Int J Antimicrob Agents
 , 
2009
, vol. 
34
 (pg. 
285
-
6
)
36
Mugnier
PD
Poirel
L
Naas
T
, et al.  . 
Worldwide dissemination of the blaOXA-23 carbapenemase gene of Acinetobacter baumannii
Emerg Infect Dis
 , 
2010
, vol. 
16
 (pg. 
35
-
40
)
37
Ikonomidis
A
Pournaras
S
Maniatis
AN
, et al.  . 
Discordance of meropenem versus imipenem activity against Acinetobacter baumannii
Int J Antimicrob Agents
 , 
2006
, vol. 
28
 (pg. 
376
-
7
)

Supplementary data