Abstract
The GenBank/EMBL/DDBJ accession numbers for the sequences reported in this paper are: FJ195387 (Acinetobacter radioresistens strain B472 putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene/blaOXA-134 gene/ATPase-encoding gene); FJ195388 (Acinetobacter radioresistens strain B374 putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene/blaOXA-23 gene/ATPase-encoding gene); FJ195389 (Acinetobacter genomospecies 10 strain S1055 ISAba3-like transposase-encoding gene/blaOXA-58 gene/insertion sequence ISAba3/araC1 gene/lysE gene); FJ200197 (Acinetobacter genomospecies 9 strain A361 ISAba3-like transposase-encoding gene/blaOXA-58 gene/insertion sequence ISAba3/araC1 gene).
A table of primer sequences is available as supplementary material with the online version of this paper.
Acquisition of OXA carbapenemases has emerged as one of the predominant mechanisms for carbapenem resistance in Acinetobacter spp. in many parts of the world. Four main groups of OXA carbapenemase-encoding genes are found in this genus, namely, blaOXA-23-like, blaOXA-40-like, blaOXA-51-like and blaOXA-58 genes. Acinetobacter baumannii is the source of blaOXA-51-like genes, while more recently the reservoir for blaOXA-23-like genes has been identified as being Acinetobacter radioresistens (Turton et al., 2006; Poirel et al., 2008). However, carbapenem resistance is only manifested in isolates possessing strong promoters upstream of the blaOXA genes, such as insertion sequence ISAba1 (Turton et al., 2006). We undertook a study to investigate the silent carriage of blaOXA-58, blaOXA-23-like and blaOXA-40-like genes in carbapenem-susceptible isolates of uncommon Acinetobacter spp. and to characterize the genetic environments of such genes.
Non-duplicate Acinetobacter clinical isolates were collected in the Central Pathology Laboratory, St James's Hospital, from May 2005 to October 2007. They were speciated using sequencing of the ribosomal polymerase B subunit-encoding gene rpoB and flanking regions (La Scola et al., 2006). Detection of blaOXA genes and characterization of the genetic surroundings of such genes were performed by PCR of whole-cell DNA extracts of isolates using a PCR core kit (Qiagen). The primers are shown in Supplementary Table S1 (available with the online journal). Sequencing of amplicons was performed in the Genomics Core Facility, Queens University Belfast, Belfast, UK, using an ABI Prism 3130xl analyser (Applied BioSystems). To find out if the blaOXA genes were plasmid mediated, plasmid extraction was performed using a Qiagen midi kit and the plasmid was used for PCR with the appropriate blaOXA primers. PCR for the rpoB gene was also performed to rule out chromosomal DNA carry-over. To locate the actual plasmids carrying these genes, agarose gel electrophoresis was first used to visualize individual plasmids, and subsequently, DNA from individual plasmid bands was eluted using a gel extraction kit (Qiagen) for use in PCR with blaOXA primers. PCR for the rpoB gene was again performed to rule out chromosomal DNA carry-over. Susceptibility testing was also performed on blaOXA-positive isolates using the Etest method (bioMérieux) for meropenem, imipenem, cefepime, cefotaxime, ceftazidime, piperacillin–tazobactam, amoxicillin–clavulanate and ampicillin. A standard inoculum of 0.5 McFarland and a heavier inoculum of 2 McFarland were used.
A total of 37 isolates of various Acinetobacter spp. were collected during the 30 month period (the number of isolates of each species is shown in parentheses): Acinetobacter johnsonnii (12), Acinetobacter ursingii (8), Acinetobacter genomospecies 9 (AG9) (5), Acinetobacter genomospecies 11 (4), Acinetobacter lwoffii (2), A. radioresistens (2), Acinetobacter genomospecies 10 (AG10) (1), Acinetobacter haemolyticus (1), Acinetobacter schindleri (1) and Acinetobacter tjernbergiae (1).
All isolates were negative for blaOXA-40-like genes. The two A. radioresistens isolates (strains B472 and B374) were positive for blaOXA-23-like genes. In both isolates, a putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene was present upstream of blaOXA-23-like (98.4 % sequence concordance with GenBank accession no. EU571228), while downstream of the blaOXA-23-like gene, the ATPase-encoding gene was present in both isolates [99 % (B472) and 100 % (B374) sequence concordance with GenBank accession no. EU131372]. Sequencing of the entire blaOXA-23-like genes revealed that B374 carried blaOXA-23 while B472 carried a novel OXA gene blaOXA-134. blaOXA-134 differed from blaOXA-23 by two amino acid substitutions (Δ10, valine→phenylalanine; Δ33, valine→alanine). blaOXA-58 was found in two isolates, A361 (AG9) and S1055 (AG10). In both isolates, an ISAba3-like transposase was present upstream of blaOXA-58 (Fig. 1⇓). The amino acid sequence of its C-terminus differed from the corresponding 23 amino acids of the ISAba3 transposase. Insertion sequences ISAba2, ISAba1 and IS18 were absent in our isolates. Downstream of blaOXA-58, the insertion sequence ISAba3, as well as a transcriptional regulator-encoding gene (araC1), were found in both isolates. The gene encoding the threonine efflux protein (lysE) was found downstream of the araC1 gene in AG10 but not in AG9. Additional genes (putative esterase-encoding est and transcriptional regulator-encoding araC2 genes) reported in the A. baumannii MAD strain were not found in either of our isolates (Poirel & Nordmann, 2006).
Genetic environment of blaOXA-58 in carbapenem-susceptible AG10 (a) and AG9 (b) isolates.
PCR of plasmid extracts of A361 and S1055 was positive for blaOXA-58, indicating that these genes were plasmid-mediated. PCR of plasmid extracts for blaOXA-23-like was negative for B374 and B472, suggesting the genes were chromosomally located. Four plasmids were visualized for S1055 (ranging from about 5 to 30 MDa), while six plasmids were visualized for A361 (ranging from about 1.5 to 25 MDa). The blaOXA-58 gene was found on the approximately 30 MDa plasmid and the approximately 22 MDa plasmid in S1055 and A361, respectively. Antimicrobial susceptibility results are shown in Table 1⇓. All four isolates were susceptible to meropenem and imipenem, and no carbapenem-resistant subpopulations were observed even with the heavier inocula.
β-Lactam antimicrobial susceptibility of blaOXA-23-like-positive and blaOXA-58-positive carbapenem-susceptible Acinetobacter isolates from St James's Hospital, Dublin
Our A. radioresistens findings support the hypothesis that this species may be the source of blaOXA-23-like genes (Poirel et al., 2008). Mobilization of such genes via plasmids has resulted in worldwide dissemination of blaOXA-23 in Acinetobacter spp. However, blaOXA-58 is mostly reported in A. baumannii isolates and also in an Acinetobacter genomospecies 3 isolate (Marti et al., 2008; Poirel et al., 2005; Poirel & Nordmann, 2006). Our isolates represent what are believed to be the first reported cases of blaOXA-58 carriage in AG9 and AG10 spp. Additional insertion sequences, ISAba2, ISAba1 and IS18, are frequently present upstream of blaOXA-58 in carbapenem-resistant Acinetobacter isolates, but were absent in our isolates (Poirel & Nordmann, 2006). Instead, the blaOXA-58 in our isolates is bracketed by ISAba3-like and ISAba3 transposases in a similar arrangement as that of a composite transposon. It is likely that a similar arrangement of genes preceded the insertion of ISAba2 in other A. baumannii isolates since direct repeats were present within the disrupted ISAba3-like transposase immediately upstream and downstream of ISAba2 (Poirel & Nordmann, 2006). Since ISAba2, ISAba1 and IS18 have promoter sequences to upregulate blaOXA-58, our study suggests that they are more effective than those found within the ISAba3-like transposase, given the difference in carbapenem susceptibility patterns (Poirel & Nordmann, 2006). Alternatively, other factors, such as plasmid copy number and the number of blaOXA-58 genes per plasmid, may also play a role in determining the carbapenem susceptibility, although increasing the inoculum concentrations of our isolates did not increase the carbapenem MICs appreciably (Bertini et al., 2007).
The presence of blaOXA-58 and blaOXA-23-like genes in carbapenem-susceptible Acinetobacter isolates highlights the threat of undetected reservoirs of carbapenemase-encoding genes, since laboratory detection of such genes and subsequent infection control measures in hospitals generally target phenotypically multidrug-resistant organisms. In A. radioresistens, blaOXA-23-like genes can be mobilized onto plasmids and transferred to other species, as exemplified by the prevalence of blaOXA-23 in carbapenem-resistant Acinetobacter isolates worldwide. Isolates carrying plasmid-mediated blaOXA-58 also have the potential of conferring carbapenem resistance if such genes are subsequently transferred to species possessing more potent insertion sequences. As such, standard infection control precautions targeting all organisms (such as hand hygiene and scrupulous environmental cleaning) should be stringently adhered to in order to reduce the risk of transfer of such resistance genes. Judicious use of carbapenems must also be advocated, as widespread use of such agents may also create an environment favouring the inter-species transfer as well as the overt expression of these genes.
-
The GenBank/EMBL/DDBJ accession numbers for the sequences reported in this paper are: FJ195387 (Acinetobacter radioresistens strain B472 putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene/blaOXA-134 gene/ATPase-encoding gene); FJ195388 (Acinetobacter radioresistens strain B374 putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene/blaOXA-23 gene/ATPase-encoding gene); FJ195389 (Acinetobacter genomospecies 10 strain S1055 ISAba3-like transposase-encoding gene/blaOXA-58 gene/insertion sequence ISAba3/araC1 gene/lysE gene); FJ200197 (Acinetobacter genomospecies 9 strain A361 ISAba3-like transposase-encoding gene/blaOXA-58 gene/insertion sequence ISAba3/araC1 gene).
-
A table of primer sequences is available as supplementary material with the online version of this paper.
Acquisition of OXA carbapenemases has emerged as one of the predominant mechanisms for carbapenem resistance in Acinetobacter spp. in many parts of the world. Four main groups of OXA carbapenemase-encoding genes are found in this genus, namely, blaOXA-23-like, blaOXA-40-like, blaOXA-51-like and blaOXA-58 genes. Acinetobacter baumannii is the source of blaOXA-51-like genes, while more recently the reservoir for blaOXA-23-like genes has been identified as being Acinetobacter radioresistens (Turton et al., 2006; Poirel et al., 2008). However, carbapenem resistance is only manifested in isolates possessing strong promoters upstream of the blaOXA genes, such as insertion sequence ISAba1 (Turton et al., 2006). We undertook a study to investigate the silent carriage of blaOXA-58, blaOXA-23-like and blaOXA-40-like genes in carbapenem-susceptible isolates of uncommon Acinetobacter spp. and to characterize the genetic environments of such genes.
Non-duplicate Acinetobacter clinical isolates were collected in the Central Pathology Laboratory, St James's Hospital, from May 2005 to October 2007. They were speciated using sequencing of the ribosomal polymerase B subunit-encoding gene rpoB and flanking regions (La Scola et al., 2006). Detection of blaOXA genes and characterization of the genetic surroundings of such genes were performed by PCR of whole-cell DNA extracts of isolates using a PCR core kit (Qiagen). The primers are shown in Supplementary Table S1 (available with the online journal). Sequencing of amplicons was performed in the Genomics Core Facility, Queens University Belfast, Belfast, UK, using an ABI Prism 3130xl analyser (Applied BioSystems). To find out if the blaOXA genes were plasmid mediated, plasmid extraction was performed using a Qiagen midi kit and the plasmid was used for PCR with the appropriate blaOXA primers. PCR for the rpoB gene was also performed to rule out chromosomal DNA carry-over. To locate the actual plasmids carrying these genes, agarose gel electrophoresis was first used to visualize individual plasmids, and subsequently, DNA from individual plasmid bands was eluted using a gel extraction kit (Qiagen) for use in PCR with blaOXA primers. PCR for the rpoB gene was again performed to rule out chromosomal DNA carry-over. Susceptibility testing was also performed on blaOXA-positive isolates using the Etest method (bioMérieux) for meropenem, imipenem, cefepime, cefotaxime, ceftazidime, piperacillin–tazobactam, amoxicillin–clavulanate and ampicillin. A standard inoculum of 0.5 McFarland and a heavier inoculum of 2 McFarland were used.
A total of 37 isolates of various Acinetobacter spp. were collected during the 30 month period (the number of isolates of each species is shown in parentheses): Acinetobacter johnsonnii (12), Acinetobacter ursingii (8), Acinetobacter genomospecies 9 (AG9) (5), Acinetobacter genomospecies 11 (4), Acinetobacter lwoffii (2), A. radioresistens (2), Acinetobacter genomospecies 10 (AG10) (1), Acinetobacter haemolyticus (1), Acinetobacter schindleri (1) and Acinetobacter tjernbergiae (1).
All isolates were negative for blaOXA-40-like genes. The two A. radioresistens isolates (strains B472 and B374) were positive for blaOXA-23-like genes. In both isolates, a putative O-sialoglycoprotein endopeptidase-encoding (gcp) gene was present upstream of blaOXA-23-like (98.4 % sequence concordance with GenBank accession no. EU571228), while downstream of the blaOXA-23-like gene, the ATPase-encoding gene was present in both isolates [99 % (B472) and 100 % (B374) sequence concordance with GenBank accession no. EU131372]. Sequencing of the entire blaOXA-23-like genes revealed that B374 carried blaOXA-23 while B472 carried a novel OXA gene blaOXA-134. blaOXA-134 differed from blaOXA-23 by two amino acid substitutions (Δ10, valine→phenylalanine; Δ33, valine→alanine). blaOXA-58 was found in two isolates, A361 (AG9) and S1055 (AG10). In both isolates, an ISAba3-like transposase was present upstream of blaOXA-58 (Fig. 1⇓). The amino acid sequence of its C-terminus differed from the corresponding 23 amino acids of the ISAba3 transposase. Insertion sequences ISAba2, ISAba1 and IS18 were absent in our isolates. Downstream of blaOXA-58, the insertion sequence ISAba3, as well as a transcriptional regulator-encoding gene (araC1), were found in both isolates. The gene encoding the threonine efflux protein (lysE) was found downstream of the araC1 gene in AG10 but not in AG9. Additional genes (putative esterase-encoding est and transcriptional regulator-encoding araC2 genes) reported in the A. baumannii MAD strain were not found in either of our isolates (Poirel & Nordmann, 2006).
Genetic environment of blaOXA-58 in carbapenem-susceptible AG10 (a) and AG9 (b) isolates.
PCR of plasmid extracts of A361 and S1055 was positive for blaOXA-58, indicating that these genes were plasmid-mediated. PCR of plasmid extracts for blaOXA-23-like was negative for B374 and B472, suggesting the genes were chromosomally located. Four plasmids were visualized for S1055 (ranging from about 5 to 30 MDa), while six plasmids were visualized for A361 (ranging from about 1.5 to 25 MDa). The blaOXA-58 gene was found on the approximately 30 MDa plasmid and the approximately 22 MDa plasmid in S1055 and A361, respectively. Antimicrobial susceptibility results are shown in Table 1⇓. All four isolates were susceptible to meropenem and imipenem, and no carbapenem-resistant subpopulations were observed even with the heavier inocula.
β-Lactam antimicrobial susceptibility of blaOXA-23-like-positive and blaOXA-58-positive carbapenem-susceptible Acinetobacter isolates from St James's Hospital, Dublin
Our A. radioresistens findings support the hypothesis that this species may be the source of blaOXA-23-like genes (Poirel et al., 2008). Mobilization of such genes via plasmids has resulted in worldwide dissemination of blaOXA-23 in Acinetobacter spp. However, blaOXA-58 is mostly reported in A. baumannii isolates and also in an Acinetobacter genomospecies 3 isolate (Marti et al., 2008; Poirel et al., 2005; Poirel & Nordmann, 2006). Our isolates represent what are believed to be the first reported cases of blaOXA-58 carriage in AG9 and AG10 spp. Additional insertion sequences, ISAba2, ISAba1 and IS18, are frequently present upstream of blaOXA-58 in carbapenem-resistant Acinetobacter isolates, but were absent in our isolates (Poirel & Nordmann, 2006). Instead, the blaOXA-58 in our isolates is bracketed by ISAba3-like and ISAba3 transposases in a similar arrangement as that of a composite transposon. It is likely that a similar arrangement of genes preceded the insertion of ISAba2 in other A. baumannii isolates since direct repeats were present within the disrupted ISAba3-like transposase immediately upstream and downstream of ISAba2 (Poirel & Nordmann, 2006). Since ISAba2, ISAba1 and IS18 have promoter sequences to upregulate blaOXA-58, our study suggests that they are more effective than those found within the ISAba3-like transposase, given the difference in carbapenem susceptibility patterns (Poirel & Nordmann, 2006). Alternatively, other factors, such as plasmid copy number and the number of blaOXA-58 genes per plasmid, may also play a role in determining the carbapenem susceptibility, although increasing the inoculum concentrations of our isolates did not increase the carbapenem MICs appreciably (Bertini et al., 2007).
The presence of blaOXA-58 and blaOXA-23-like genes in carbapenem-susceptible Acinetobacter isolates highlights the threat of undetected reservoirs of carbapenemase-encoding genes, since laboratory detection of such genes and subsequent infection control measures in hospitals generally target phenotypically multidrug-resistant organisms. In A. radioresistens, blaOXA-23-like genes can be mobilized onto plasmids and transferred to other species, as exemplified by the prevalence of blaOXA-23 in carbapenem-resistant Acinetobacter isolates worldwide. Isolates carrying plasmid-mediated blaOXA-58 also have the potential of conferring carbapenem resistance if such genes are subsequently transferred to species possessing more potent insertion sequences. As such, standard infection control precautions targeting all organisms (such as hand hygiene and scrupulous environmental cleaning) should be stringently adhered to in order to reduce the risk of transfer of such resistance genes. Judicious use of carbapenems must also be advocated, as widespread use of such agents may also create an environment favouring the inter-species transfer as well as the overt expression of these genes.