The AbaR antibiotic resistance islands found in Acinetobacter baumannii global clone 1 – Structure, origin and evolution
Introduction
Multiply antibiotic resistant Acinetobacter spp. have become an increasingly important cause of nosocomial infections (Clark et al., 2016; Perez et al., 2007; Towner, 2009; Wong et al., 2017) and are included in the ESKAPE group of nosocomial pathogens where antibiotic resistance is a serious problem (Boucher et al., 2009; Pendleton et al., 2013). In Europe, the increase in antimicrobial resistance was attributed to the spread of three clones of A. baumannnii, designated European clones I, II and III (Dijkshoorn et al., 1996; Hamouda et al., 2010; Nemec et al., 2008; Turton et al., 2007; van Dessel et al., 2004; Zarrilli et al., 2013). However, in the late 2000s, as typing was simplified and information about multiply antibiotic resistant A. baumannii isolates from other continents gradually became available, it became clear that the reach of isolates from two of these groups, namely I and II, was much greater as they were also found on other continents such as North America (Adams et al., 2010, 2008), Australia (Post and Hall, 2009; Post et al., 2010), South America (Grosso et al., 2011; Ramírez et al., 2013) and Asia (Fu et al., 2010; Kim et al., 2013; Lee et al., 2011; Peymani et al., 2012; Solomennyi et al., 2015; Sung et al., 2012) as well as in various other countries (Diancourt et al., 2010; Higgins et al., 2010; Mugnier et al., 2010). Consequently, in this review the terms Global Clone 1 (GC1) and Global Clone 2 (GC2) will be used to replace European Clones I and II or International Clones I and II, as this better reflects the global distribution of isolates belonging to these clonal complexes. GC1 and GC2 correspond to the clonal complexes CC1 and CC2 in the Institut Pasteur multilocus sequence typing (MLST) scheme (Diancourt et al., 2010).
Multiple resistance in Acinetobacter species and even resistance to all antibiotics available at the time was first noted in the 1970s (Perez et al., 2007; Towner, 2009). However, detailed examination of the context of acquired antibiotic resistance in A. baumannii began late relative to studies of other Gram negative pathogens. The first large antibiotic resistance gene cluster found in an A. baumannii isolate to be completely sequenced was from a multiply resistant strain, AYE, from France (Fournier et al., 2006). AYE contains an 86 kb resistance island that was designated AbaR1 (A. baumannii resistance) (Fournier et al., 2006). AbaR1 is a complex transposon that is located in the chromosome rather than in a plasmid. It is includes both antibiotic and heavy metal resistance genes and the 18 antibiotic resistance genes, some of which are present in multiple copies, confer resistance to most of the known classes of antibiotics. The AYE isolate was later shown to be sequence group SG2 (Turton et al., 2007), corresponding to European clone I or GC1.
A number of complex transposon types that carry resistance genes have been found in the same location as AbaR1 and though they include related components and are likely to have a similar movement mechanism, each type has a separate evolutionary history. In this review, we have drawn together what is known about the structure of one type, the AbaR-type genomic resistance islands that have since been found in the majority of antibiotic resistant GC1 isolates, and occasionally in GC2 isolates. How they are distinguished from other resistance island types is addressed. Then their arrangements are compared. Their origin, evolution and movement are also considered.
Section snippets
Overview of resistance islands found in comM in GC1 and GC2 isolates
AbaR1 was originally defined as a region that had inserted into the chromosome, creating a duplication of the 5 bp target site (Fournier et al., 2006) consistent with incorporation by transposition. Imperfect inverted repeats of 26 bp were later found at the transposon boundaries (Post and Hall, 2009). The chromosomal insertion site is in a gene known as comM that encodes an ATPase and is located about 280 kb from the replication origin.
Not long after AbaR1 was described, regions related to but
Nomenclature for A. baumannii resistance islands
Currently, some confusion is arising from the naming of the various resistance island types. To ensure that the two main distinct types of resistance island found in comM are not confused, we propose that AbaR should be retained exclusively for the configuration common in GC1. The term AbGRI1 has been proposed for the complex transposon type found in comM in most GC2 isolates (Nigro et al., 2013) and, different AbGRI1 can be numbered AbGRI1-1, etc. The AbGRI1 variants including ones that
AbaR0 is the progenitor of AbaR variant forms
Though the 86,244 bp AbaR1 found in AYE (GenBank accession no. CT025832 for AbaR1 and CU459141 for the AYE chromosome), and described in detail in section 8.1, was the first AbaR to be sequenced (Fournier et al., 2006), it has features that set it apart from the majority of AbaR. The AbaR1 MARR is longer and includes many more resistance genes than MARR that have been examined subsequently. Instead, the AbaR in the majority of GC1 isolates more closely resemble the configuration of the 63 kb
General structure of AbaR
The general structure for AbaR genomic islands shown in Fig. 2A was proposed (Post et al., 2010) after comparing the structures of AbaR1 (Fournier et al., 2006), AbaR3 (Adams et al., 2008), AbaR5 (Post and Hall, 2009), and AbaR6 and AbaR7 (Post et al., 2010), and the components making up this arrangement were named. The same general structure was later seen in further GC1 isolates (Hamidian et al., 2014a, b; Holt et al., 2016; Krizova et al., 2011; Krizova and Nemec, 2010; Nigro et al., 2011;
The MARR of AbaR0 and AbaR3
AbaR3, originally found in the chromosome of the multiply antibiotic resistant A. baumannii strain AB0057 (Adams et al., 2008) (GenBank accession no. CP001182.2) was later found in other GC1 isolates from Australia (Hamidian et al., 2014a; Holt et al., 2016). AbaR0 shown in Fig. 3 was first found in an Australian GC1 isolate from 1998 (Hamidian et al., 2014b) but is likely to be present in the 1977 isolate HK302 (Devaud et al., 1982; Krizova and Nemec, 2010). These two AbaR have the most
Origin of the MARR
The characteristic features of the MARR described above have allowed the origin of the bulk of the MARR to be identified as they are found in the multiple antibiotic resistance region of the IncM1 plasmid R1215 (Blackwell et al., 2016). The source of R1215 is not well documented. However, it was recovered from a Serratia marcescens strain some time prior to 1980 and was submitted to the UK Collection of National Type Cultures by Dr. Naomi Datta. This timing correlates well with the estimated
Variants of AbaR0 and AbaR3
The structures of all of the AbaR variants that have been reported to date are listed in Table 1 together with their origin in AbaR0 or AbaR3 where it can be deduced from available information. Some of these variant structures were generated using PCR mapping strategies combined with sequencing of any novel boundary indicative of a deletion (Krizova et al., 2011; Krizova and Nemec, 2010) but available information for AbaR20 does not allow an arrangement to be deduced. In several cases, contigs
Relatives of the Tn6019 backbone transposon
The A. baumannii strain ATCC 17978 (Smith et al., 2007) (GenBank accession no. CP000512), which is only resistant to sulphonamides (Nigro and Hall, 2011), carries a cryptic transposon in the same position as AbaR type resistance islands in the comM gene. This transposon (Fig. 8B), designated Tn6021 (Post and Hall, 2009), was the first found to be related to Tn6019, the AbaR backbone transposon. Tn6021 is closely related to Tn6022, which makes up the backbone of AbaR4 (Hamidian and Hall, 2011)
Conclusions
Most multiply or extensively antibiotic resistant GC1 isolates examined to date, including the earliest known isolate, carry a single complex transposon of the AbaR type located at a specific location in the chromosome. Given the complexity of this transposon it is likely that it has been acquired independently only once and the source has been identified as a close relative of the resistance region of an IncM plasmid R1215. This region has then evolved in situ via exchanges of gene cassettes
Acknowledgements
R.M.H., M.H. and this work were supported in part by NHMRC Project grants GNT1026189 and GNT1079616. MH is currently supported by a UTS Chancellors Fellowship (PRO17-4005).
References (85)
- et al.
Transposon targeting determined by resolvase
FEMS Microbiol. Letts.
(2000) - et al.
Carbapenem-non-susceptible Acinetobacter baumannii of sequence type 92 or its single-locus variants with a G428T substitution in zone 2 of the rpoB gene
J. Antimicrob. Chemother.
(2011) - et al.
Arsenate reduction: thiol cascade chemistry with convergent evolution
J. Mol. Biol.
(2006) - et al.
Characterisation and clonal dissemination of OXA-23-producing Acinetobacter baumannii in Tabriz, northwest Iran
Int. J. Antimicrob. Agents
(2012) - et al.
Extensively drug-resistant Acinetobacter baumannii belonging to the international clonal lineage I in a Russian burn intensive care unit
Int. J. Antimicrob. Agents
(2015) Acinetobacter: an old friend, but a new enemy
J. Hosp. Infect.
(2009)- et al.
Use of sequence-based typing and multiplex PCR to identify clonal lineages of outbreak strains of Acinetobacter baumannii
Clin. Microbiol. Infect.
(2007) - et al.
Identification of a new geographically widespread multiresistant Acinetobacter baumannii clone from European hospitals
Res. Microbiol.
(2004) - et al.
Global evolution of multidrug-resistant Acinetobacter baumannii clonal lineages
Int. J. Antimicrob. Agents
(2013) - et al.
Genomewide analysis of divergence of antibiotic resistance determinants in closely related isolates of Acinetobacter baumannii
Antimicrob. Agents Chemother.
(2010)