Glimpse into the genome sequence of a multidrug-resistant Acinetobacter pittii ST950 clinical isolate carrying the bla OXA-72 and bla OXA-533 genes in China

The development of carbapenem-resistant Acinetobacter species is of serious concern in the hospital settings and naturally occurring oxacillinase genes (bla OXA) have been identified in several Acinetobacter species. In this study, we report the genome sequence of A. pittii TCM178 belongs to ST950, a multidrug-resistant isolate that harbored the bla OXA-72 and bla OXA-533 genes simultaneous. The genome size was estimated to be 3,789,564 bp with 3,501 predicted coding regions, and G+C content is 37.60%. Our findings have raised awareness of the possible constitution of a reservoir for peculiar carbapenemase genes in A. pittii that may spread among other Acinetobacter species in China.

Acinetobacter pittii, formerly named Acinetobacter genomic species (gen. sp.) 3, is frequently associated with hospital-acquired infections and outbreaks that poses a particular concern due to its ability to acquire multidrug resistance to a wide range of antibiotics (Wang et al. 2014). Carbapenems have potent activity against Acinetobacter spp. and are often used as the last resort for the treatment of infections due to the high level of antimicrobial resistance. The resistance rates to carbapenems among Acinetobacter spp., mainly caused by intrinsically possessed an extensive arsenal of chromosomal genes encoding carbapenem-hydrolysing class D-lactamases (CHDLs), have increased dramatically in the last decade. To date, six phylogenetic subgroups of this enzyme have been identified in Acinetobacter spp., namely OXA-23, OXA-24/40, OXA-51, OXA-58, OXA-143 and OXA-235 (Evans & Amyes 2014). The gene bla  , one of the bla OXA-40-like genes, was first identified in an A. baumannii strain isolated in Thailand, 2004. After that, this enzyme was reported in Acinetobacter spp. clinical isolates from different regions of the world. Genes encoding naturally occurring oxacillinases have been identified in several Acinetobacter species, such as bla OXA-23-like (A. radioresistens), bla OXA-51-like (A. baumannii), bla    and bla OXA-228-like (A. bereziniae), which could be utilised as a tool for rapid species identification (Montealegre et al. 2012, Kamolvit et al. 2015.
In the present study, we determined the whole genome sequence of a two OXA-type oxacillinases-producing A. pittii TCM178 isolate in China. The potential antibiotic resistance genes in TCM178 were predicted and a novel variant allele of the intrinsic bla OXA-533 gene has been found. These findings may improve our understanding of the antibiotic resistance mechanisms in A. pittii, and provide a clinical guidance for the therapy of A. pittii infections.
The strain A. pittii TCM178 was recovered from a blood sample of a male hospitalised patient with pneumonia at Hangzhou, Zhejiang province, China, in 2013. The strain was grown overnight at 37ºC in Mueller-Hinton broth (Oxoid, Hampshire, United Kingdom). It was identified according to rpoB gene sequencing analysis (La Scola et al. 2006). Antimicrobial susceptibility tests were conducted using the agar diffusion method and Etest (AB bioMérieux, Solna, Sweden), and interpreted according to Clinical and Laboratory Standards Institute (CLSI) 2017 guidelines (M100-S27).
Genomic DNA was extracted using a QIAamp DNA minikit (Qiagen, Valencia, CA) according to the protocol recommended by the manufacturer. Quality of the extracted DNA was examined using NanoDrop spectrophotometer (Thermo Scientific, Waltham, MA, USA) and a Qubit version 2.0 fluorometer (Life Technologies, Carlsbad, CA, USA). Subsequently, DNA library was prepared using Nextera™ DNA Sample Preparation kit (Nextera, USA), while the library quality was validated by Bioanalyzer 2100 high sensitivity DNA kit (Agilent Technologies, Palo Alto, CA) prior to sequencing. The DNA libraries with insert sizes of 300 bp were constructed from 5 μg DNA using an Illumina paired-end DNA sample prep kit (Illumina Inc., Cambridge, United Kingdom) and sequenced using the 2×150 bp paired-end sequencing strategy. The genome sequence of strain TCM178 was sequenced using an Illumina HiSeq 2500 platform (Illumina Inc., Madison, WI, USA). After sequencing, all sequence reads were preprocessed to re-move low-quality or artifactual bases. The FastQC 0.11.5 was used for assessing the quality of the raw data, and the AlienTrimmer 0.4.0 was used to trim and discard the reads with a Phred quality score below 20. Finally, the fqduplicate 1.1 was used to discard every duplicate pairedend read. The trimmed reads were de novo assembled using CLC Genomics Workbench 9.0 (Qiagen, Valencia, CA). The optimal k-mer size was automatically determined using KmerGenie 1.7039. The contigs were linked and placed into scaffolds or supercontigs. The analysis was carried out using the SSPACE Premium scaffolder 2.3. The gapped regions within the scaffolds were partially closed in an automated manner using GapFiller 1.10.
The genome sequence was automatically annotated by the NCBI Prokaryotic Genomes Annotation Pipeline (PGAP) and Rapid Annotation using Subsystem Technology (RAST) server (Aziz et al. 2008, Seemann 2014. Classification of predicted proteins in clusters of orthologous groups (COG) functional categories was analysed with the WebMGA web server (Wu et al. 2011). PathogenFinder 1.1, ResFinder 2.1, CARD 2017, Viru-lenceFinder 1.5, PlasmidFinder 1.3, MLST 1.8 (Multi-Locus Sequence Typing), and BacWGSTdb were used to estimate the number of pathogenicity determinants, acquired antibiotic resistance genes, virulence genes, plasmids and the sequence type (ST) using the assembled ge- nome (Larsen et al. 2012, Zankari et al. 2012, Cosentino et al. 2013, Carattoli et al. 2014, Joensen et al. 2014, Ruan & Feng 2016, Jia et al. 2017. Minimum spanning tree (MSTree) analysis is a graph-based clustering method that can generate graphical results from MLST profile data. The PHYLOViZ 2.0 program can link the allele designations within the PubMLST database and draw an MSTree (Nascimento et al. 2017). The MSTree was calculated by Prim's algorithm and was incorporated with the eBURST algorithm to obtain a comprehensive and precise graphical result. The graphical results included different types of lines were used to illustrate the number of shared alleles, and the length of branches reflect the distance between different types. Further bioinformatics analysis, such as identification of genomic islands, insertion elements (IS), prophage sequences, clustered regularly interspaced short palindromic repeat (CRISPR) sequences and secondary metabolite gene clusters, were predicted by application of IslandViewer 3, ISfinder 1.0, PHASTER 2016, CRISPRFinder 1.0 and antiSMASH 4.0.0 tools with default parameters, respectively (Siguier et al. 2006, Grissa et al. 2007, Dhillon et al. 2015, Weber et al. 2015, Arndt et al. 2016. The draft genome sequence of A. pittii TCM178 consisted of 43 contigs which comprised 3,789,564 bases, and PGAP server predicted a total of 3,501 protein-cod-ing sequences. The overall G+C content of this strain amounted to 37.60%. In total, 63 tRNA genes and 3 rRNA operons were identified, respectively (Fig. 1). The distribution of genes into COGs functional categories is summarised in Table. We also identified the aminoglycoside resistance genes strA, strB and aph(3')-VIa, betalactam resistance genes bla ADC-25 , bla OXA-72 and bla OXA-213-like , macrolide resistance genes msr(E) and mph(E), and tetracycline resistance gene tet(39). The allele of bla-OXA-213-like gene is a novel variant allele (271/273 identities) of intrinsic bla OXA-533 gene of A. pittii strain UKK_0145 (bla OXA gene for OXA-213 family carbapenem-hydrolysing class D beta-lactamase OXA-533, complete CDS, NCBI Reference Sequence: NG_051483.1). The genome also contains at least 20 genomic islands and several IS elements: the majority belonging to the IS3, IS5, and IS110 families. Similarly, there is one prophage sequence and three CRISPR sequences can be predicted in the genome. The presence of four putative secondary metabolite gene clusters, including the arylpolyne, bacteriocin, hserlactone and siderophore biosynthetic gene clusters can also be predicted. In addition, genome sequencing identified the gene bla OXA-72 was flanked by XerC/XerD-binding sites, a structure-specific recombination system implicated with its mobilisation. Upstream of the gene bla OXA-533 , an N-acetyltransferase-  encoding gene was identified; whereas the gene encoding a putative suppressor of F exclusion of phage T7 ( fxsA) was found downstream. Genomic DNA digested with S1-nuclease and plasmid DNA from this strain were separated by PFGE, and subsequent Southern blot and hybridisation with bla OXA -specific probe showed that the bla OXA-72 gene was located on a plasmid of ~ 10 kb and the gene bla OXA-533 was located on the chromosome. The draft genome sequence of A. pittii TCM178 was also annotated using RAST server, with a total of 3,515 proteincoding sequences were determined, 2,141 of these were categorised into 451 subsystems, and 66 were RNAs. In the RAST-annotated genome, most of the genes were assigned into amino acids and derivatives (15.5%) followed by Carbohydrates (10.4%) and cofactors, vitamins, prosthetic groups and pigment (9.7%) subsystems. Genes encoding iron acquisition system, multidrug resistance efflux pumps and type IV secretion system were also identified in the genome. The MLST analysis showed that A. pittii TCM178 belongs to ST950, according to the MLST scheme of A. baumannii developed by Institute Pasteur. An MSTree algorithm was used to predict the clonal relationship and topological arrangement between A. pittii TCM178 and all 61 isolates in the PubMLST database. Only three clonal complexes were identified by the MSTree algorithm indicating limited genetic diversity among these isolates (Fig. 2).
In summary, these findings have raised awareness of the emergence of a multidrug-resistant ST950 A. pittii isolate producing OXA-72 and OXA-533 carbapenemases in China. The possible emergence of two OXA-type enzymes is worrying and must be monitored to avoid their major spread to more clinically relevant bacterial species. Further studies on the comparative genomic analysis of a large-scale sampling of A. pittii strains from a wide spatial and temporal range to get a more comprehensive picture of genomic epidemiological characteristics are warranted. To our knowledge, this is the first draft genome sequence of a multidrug-resistant OXA-72 and OXA-533 carbapenemase-producing A. pittii isolate in China.
The genome sequence of A. pittii TCM178 (Biosample ID: SAMN04452413) can be accessed at DDBJ/ENA/ GenBank under the accession number LSAM00000000. The version described in this paper is the first version, LSAM01000000. The genome sequences data are available in FASTA, annotated GenBank flat file, graphical and ASN.1 formats.The genome project data are also available at GenBank under the genome Bioproject ID PRJNA310419.

AUTHORS' CONTRIBUTION
ZR and JFW designed and coordinated all the experiments; YC and JFW performed cultivation, DNA extraction and purification; ZR performed the genome assembly, gene prediction, gene annotation and comparative genomic analysis; ZR and JFW drafted the manuscript. All authors participated in data interpretation and manuscript preparation, provided essential reagents, read and approved the final manuscript.