Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Evolution of a Pathogen: A Comparative Genomics Analysis Identifies a Genetic Pathway to Pathogenesis in Acinetobacter

  • Jason W. Sahl ,

    jsahl@tgen.org

    Affiliation Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America

  • John D. Gillece,

    Affiliation Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America

  • James M. Schupp,

    Affiliation Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America

  • Victor G. Waddell,

    Affiliation Arizona Department of Health Services, Bureau of State Laboratory Services, Phoenix, Arizona, United States of America

  • Elizabeth M. Driebe,

    Affiliation Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America

  • David M. Engelthaler,

    Affiliation Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America

  • Paul Keim

    Affiliations Department of Pathogen Genomics, Translational Genomics Research Institute, Flagstaff, Arizona, United States of America, Center for Microbial Genetics and Genomics, Northern Arizona University, Flagstaff, Arizona, United States of America

Abstract

Acinetobacter baumannii is an emergent and global nosocomial pathogen. In addition to A. baumannii, other Acinetobacter species, especially those in the Acinetobacter calcoaceticus-baumannii (Acb) complex, have also been associated with serious human infection. Although mechanisms of attachment, persistence on abiotic surfaces, and pathogenesis in A. baumannii have been identified, the genetic mechanisms that explain the emergence of A. baumannii as the most widespread and virulent Acinetobacter species are not fully understood. Recent whole genome sequencing has provided insight into the phylogenetic structure of the genus Acinetobacter. However, a global comparison of genomic features between Acinetobacter spp. has not been described in the literature. In this study, 136 Acinetobacter genomes, including 67 sequenced in this study, were compared to identify the acquisition and loss of genes in the expansion of the Acinetobacter genus. A whole genome phylogeny confirmed that A. baumannii is a monophyletic clade and that the larger Acb complex is also a well-supported monophyletic group. The whole genome phylogeny provided the framework for a global genomic comparison based on a blast score ratio (BSR) analysis. The BSR analysis demonstrated that specific genes have been both lost and acquired in the evolution of A. baumannii. In addition, several genes associated with A. baumannii pathogenesis were found to be more conserved in the Acb complex, and especially in A. baumannii, than in other Acinetobacter genomes; until recently, a global analysis of the distribution and conservation of virulence factors across the genus was not possible. The results demonstrate that the acquisition of specific virulence factors has likely contributed to the widespread persistence and virulence of A. baumannii. The identification of novel features associated with transcriptional regulation and acquired by clades in the Acb complex presents targets for better understanding the evolution of pathogenesis and virulence in the expansion of the genus.

Introduction

Acinetobacter baumannii is a nosocomial pathogen implicated with septicemia, pneumonia, and death [1], [2], [3]. A. baumannii is truly a global pathogen, as it has been isolated from hospitals throughout the world [4], [5], [6], [7], as well as in wounded soldiers serving in Iraq [8], [9], [10] and Afghanistan [11]. Acinetobacter is a genus in the phylum Proteobacteria, family Moraxellaceae, consisting of 26 named species (http://www.bacterio.cict.fr/a/acinetobacter.html). Acinetobacter spp. are found in most soil and water samples [12], but are also a major source of nosocomial infections [13]; the natural environments for the pathogens A. baumannii and A. nosocomialis are not currently known [14]. Although most nosocomial Acinetobacter infections are associated with A. baumannii [15], other species have also been associated with human disease. For example, A. ursingii has also been associated with nosocomial bloodstream infections [16], [17].

Our understanding of the pathogenesis of A. baumannii is largely based on the ad-hoc analyses of putative virulence factors. Much of the pathogenesis research in A. baumannii has focused on biofilm formation [18], [19], although no conclusive link between biofilm formation and infection has been established [20]. Recent studies have demonstrated that iron acquisition systems in A. baumannii are important virulence factors [21]. Three distinct siderophore systems associated with iron acquisition have been described in A. baumannii [22], [23]; the most well characterized of these systems is acinetobactin, which shares homology to a plasmid-encoded siderophore in Vibrio anguillarum [24], [25]. A recent study demonstrated that the proteins BasD and BauA, which are necessary for acinetobactin synthesis, are required for the pathogen to persist and ultimately kill host epithelial cells [26].

One of the primary concerns of A. baumannii as a nosocomial pathogen is its resistance to antimicrobials [7]. Large antibiotic resistance islands (RI), including the 86-kb RI in strain AYE [27], have been characterized in A. baumannii. Broad-spectrum beta-lactamase genes [28], which are part of the carbapenem-hydrolysing class D beta-lactamases (CHDLs) subgroup [29], have been identified in A. baumannii, including blaOXA-51 [30], blaOXA-23 [31], and the plasmid-encoded gene blaOXA-58 [32]. In addition, the expression of efflux pump systems, such as AdeFGH [33], AdeABC [34], AdeIJK [35], and AdeM [36] has been associated with multidrug resistance in A. baumannii.

Much of the research on the genus Acinetobacter has focused on genomes in the Acinetobacter calcoaceticus-baumannii (Acb) complex [37]. This complex includes A. baumannii, A. nosocomialis (originally genomic species (gen. sp.) 13TU [38]), A. pittii (originally gen. sp. 3 [38]), A. sp. DR1 [39], and A. calcoaceticus. A. calcoaceticus and A. sp. DR1 have not been implicated in serious human infection [14]. A recent study of Acinetobacter bloodstream infections in the United States demonstrated that the most common nosocomial infections were caused by A. baumannii (63% of cases), A. nosocomialis (21% of cases), and A. pittii (8% of cases) [40].

From a genomics perspective, most analyses of whole genome sequence data have focused on multi-drug resistant (MDR) isolates [41], [42], [43], [44], [45], although additional studies have considered evolutionary relationships of A. baumannii [46], [47]. In this study, a large-scale whole genome sequence analysis was performed on a set of previously characterized (n = 69), as well as newly sequenced (n = 67), Acinetobacter genomes. These data were used to identify the genomic diversity of the genus Acinetobacter and understand the flow of genetic information between species in the genus. Prior to whole genome sequence analysis, studying the conservation of diverse genes across the genus was not possible. This information is not only pertinent from a pathogenesis perspective, but may also aid in the identification of targets important for diagnostic, therapeutic and vaccine development.

Methods and Materials

Strain selection/clinical data

Sixty-seven Acinetobacter genomes were sequenced from a variety of human and environmental sources (Table S1). Isolates were selected to capture a broad range of genetic, geographic, and temporal diversity and were not chosen based on specific clinical outcomes. Species designations were applied by the position of the isolate in the phylogeny and were not based on clinical typing methods.

DNA isolation, sequencing, assembly

Acinetobacter isolates were grown on nutrient agar for 24 hours. Genomic DNA was extracted following the manufacturer's protocol for Gram-negative bacteria in the Qiagen DNeasy Blood and Tissue kit (Cat # 69504). DNA samples were prepared for multiplexed, paired end sequencing following the manufacturer's protocol. For each isolate, 1–5 ug of dsDNA in 200 ul was sheared and then purified using the QIAquick PCR Purification kit (Cat #28106, Qiagen,Valencia, CA). Enzymatic processing of the DNA followed the guidelines as described in the Illumina protocol, but enzymes for processing were obtained from New England Biolabs (Cat #E6000L, New England Biolabs, Ipswich, MA) and the oligonucleotides and adaptors were obtained from Illumina (Cat #PE-400-1001). After ligation of the adaptors, the DNA was run on a 2% agarose gel for 2 hours, after which a gel slice containing 500–600 bp fragments of each DNA sample was isolated and purified using the QIAquick Gel Extraction kit (Cat #28706, Qiagen, Valencia, CA). Individual libraries were quantified with qPCR on the ABI 7900HT (Part #4329001, Life Technologies Corporation, Carlsbad, CA) using the Kapa Library Quantification Kit (part # KK4832 or KK4835, Kapa Biosystems, Woburn, MA). Based on the individual library concentrations, equimolar pools of libraries were prepared at a concentration of at least 1 nM. The pooled libraries were sequenced on the Illumina GA-IIx using the “Genomic DNA sequencing primer V2” protocol for 36 cycles. A 100 bp paired-end run was used for all isolates.

Paired-end sequence reads were assembled with Velvet [48], in conjunction with the VelvetOptimiser (http://bioinformatics.net.au/software.velvetoptimiser.shtml). Contigs shorter than 200 nucleotides were filtered from the assembly. Errors in each assembly were corrected with iCORN [49]. Assembly statistics are detailed in Table S1.

Whole genome phylogeny of the genus Acinetobacter

136 genomes were processed with the kSNP analysis tool [50], which generated a multiple sequence alignment based on single nucleotide polymorphisms (SNPs). This alignment included SNPs from 69 reference genomes (Table S2) and 67 genomes sequenced in this study (Table S1). Singleton and homoplastic SNPs were removed from the alignment with noisy [51] using the “nogap” setting. A tree was inferred on this reduced alignment with FastTree2 [52], using the following parameters: -spr 4 -mlacc 2 –slownni. The tree was rooted by A. radioresistens in FigTree (http://beast.bio.ed.ac.uk/FigTree).

Blast score ratio (BSR) analysis

Contigs from draft assemblies were concatenated with a linker (NNNNNCACACACTTAATTAATTAAGTGTGTGNNNNN) inserted between each contig; this linker contains a start/stop codon in all 6 frames. Coding regions (CDSs) were predicted using Glimmer3 [53]. CDSs from each genome were concatenated and then de-replicated by clustering with USEARCH [54], using an ID of 0.8 and an IDDEF value of 3. Low-density clusters (n<4) were removed to reduce the size of the dataset. Each representative cluster was then translated with transeq [55]; the translated sequence was then aligned against each genome in the dataset with TBLASTN [56]. The query bit score for each genome alignment was divided by the maximum bit score in all genomes to obtain the blast score ratio (BSR) [57]; the BSR value can range from 1.0 (100% ID across 100% of the peptide) to 0 (no significant alignment).

In silico gene screen

To identify the conservation of specific genes associated with virulence (Table S3) across the genus Acinetobacter, the peptide sequence for each virulence factor was aligned against all sequenced genomes with TBLASTN. The BSR values across groups were visualized with the multi-experiment viewer [58]. Raw BSR values for each marker screened in this study are listed in Table S4.

Multi-locus sequence typing (MLST) phylogeny

For a comparison to the whole-genome phylogeny, a tree was inferred from concatenated MLST sequences. Sequences from MLST markers were downloaded (gltA, gyrB, gdhB, recA, cpn60, gpi, rpoD) from the A. baumannii pubMLST database (pubmlst.org/abaumannii). Sequences were extracted from BLAST alignments, concatenated, aligned with MUSCLE [59], and a phylogeny was inferred with FastTree2, as has been done previously [60].

16S rRNA gene sequence analysis

For an additional comparison to the whole genome phylogeny, a tree was inferred from an alignment of 16S rRNA gene sequences. Sequences were extracted from all genomes with SSU-ALIGN (http://selab.janelia.org/software.html), which is based on a covariance model implemented in Infernal [61]. These sequences were combined with reference sequences in the Greengenes database [62]. Gene sequences were exported from ARB [63] and all sequences were aligned with SSU-ALIGN. Homoplastic and singleton SNPs were removed from the alignment with noisy. A tree was inferred with FastTree2, using the same parameters as with the whole genome phylogeny. Monophyletic clades were collapsed in ARB. Sequences from a basal branch in the Gammaproteobacteria were used to root the phylogeny.

Results

Whole genome sequencing of 67 new Acinetobacter genomes

In order to expand the Acinetobacter phylogeny, 67 genomes were sequenced and analyzed. While most sequenced isolates (n = 47) were identified as A. baumannii, genomes (n = 20) from other species were included. These genomes represent a significant addition of genomic data from the genus and will serve as a valuable resource for future genomic studies.

Whole genome phylogeny of the genus Acinetobacter

A whole genome phylogeny was inferred on ∼200,000 single nucleotide polymorphisms (SNPs) identified by kSNP (Figure 1). This phylogeny represents the most comprehensive global phylogeny of Acinetobacter based on whole genome sequencing and is consistent with a recent study of Acinetobacter evolution [64]. The results confirm that A. baumannii genomes comprise a monophyletic clade that is part of the larger Acinetobacter calcoaceticus-baumannii (Acb) complex. In addition to pathogens, the Acb complex contains environmental isolates not associated with serious human disease [65]; the three isolates sequenced in this study (A. calcoaceticus TG19593, TG19585, TG19588) from the environmental clade, including A. calcoaceticus and A. sp. DR1, are soil isolates not associated with serious human infection.

thumbnail
Figure 1. A whole genome phylogeny of 136 sequenced genomes in the genus Acinetobacter.

The phylogeny was inferred with FastTree2 [52] on a single nucleotide polymorphism (SNP) matrix alignment calculated with kSNP [50] and filtered with noisy [51]. The phylogeny was rooted with A. radioresistens. Genomes sequenced in the current study are shown in red. Genomes in the Acinetobacter calcoaceticus-baumannii (Acb) complex are colored by clade.

https://doi.org/10.1371/journal.pone.0054287.g001

The phylogeny was rooted with A. radioresistens, which has also shown to be the root for a tree inferred from rpoB sequences [38] and from an amplified fragment length polymorphism (AFLP) analysis [66]. The 16S rRNA gene sequence phylogeny inferred in this study supports A. radioresistens as the most basal clade of previously sequenced Acinetobacter spp. (Figure S1). A phylogeny inferred from a concatenation of multi-locus sequence typing (MLST) markers revealed a similar topology to the whole genome phylogeny in the Acb complex (Figure S2).

Genome size differences in Acinetobacter species

To determine differences in genome size at specific nodes in the phylogeny, genome assembly sizes (Table S1) were compared between the Acinetobacter root (A. radioresistens), the Acb complex, and the remaining Acinetobacter clade (Figure 1). The radioresistens clade had the smallest average genome size (3.21, 95% CI+/−0.13 mb), while the Acb complex had the largest (3.94, 95% CI+/−0.05 mb); the remaining Acinetobacter clade was intermediate in size (3.5, 95% CI+/−0.13 mb).

Gene gain/loss in the Acb complex

Based on the average genome size, the Acb complex has acquired a complement of genes compared to all other sequenced Acinetobacter species. To identify genes that were gained and lost in the Acb complex, a blast score ratio (BSR) analysis with peptide sequences was performed. For a gene to be considered as acquired by a lineage, we required a BSR value ≥0.8 in >85% of targeted genomes and a BSR value <0.4 in >95% of non-targeted genomes; a BSR value = 0.8 is approximate to 80% peptide identity over 100% of the peptide length [67]. Genomic targets that met these criteria were tabulated (Table 1). The peptide sequence for each gene was aligned against all genomes (n = 136) with TBLASTN [56]. The results demonstrate that each gene is specific to each targeted group based on the BSR analysis (Figure 2).

thumbnail
Figure 2. A heatmap of blast score ratio (BSR) [57] values for branch specific regions in the Acb complex.

BSR values were visualized with the multi-experiment viewer [58]. Samples were clustered using an average linkage clustering algorithm. Numbers for each feature correlate with features described in Table 1. Raw data values are shown in Table S5.

https://doi.org/10.1371/journal.pone.0054287.g002

thumbnail
Table 1. Annotation details of lost and acquired genes in the evolution of A. baumannii.

https://doi.org/10.1371/journal.pone.0054287.t001

Genes unique to the Acb complex (Table 1, Figure 2) include a multidrug resistance protein, a multidrug resistance transporter, and an AraC-family transcriptional regulator. Genes either not acquired or lost by the complex include a tellurite-resistance gene and a urea ABC transporter (Table 1) that shares homology (>80% identity over 100% of the peptide length) to membrane transporters in other Pseudomonads. Additional genes shared homology to hypothetical proteins with no known function.

Genes were also identified that are conserved in the nosocomialis-baumannii clade, but are absent from other genomes in the Acb complex. Genes unique to this clade include a class A beta-lactamase (TEM-1), a GntR-family transcriptional regulator (ABAYE1931), and conserved hypothetical proteins (Table 1); the acquired transcriptional regulator shares homology (56% ID over 100% of the peptide length) with a transcriptional regulator (bgla_2g16570) in the plant pathogen, Burkholderia gladioli [68]. Genes lost by this sub-clade include a bile acid sodium symporter associated with resistance to arsenic compounds [69] (Table 1).

Genes unique to the A. baumannii clade include hypothetical proteins and a GntR family (FCD domain) transcriptional regulator (ABAYE1316); this regulator shares homology (39% ID over 96% of the peptide length) with a regulator (PST_2058) in the opportunistic pathogen, Pseudomonas stutzeri [70]. Genes lost in A. baumannii include a glyoxylase/bleomycin resistance gene and a specific ECF-type RNA polymerase sigma factor. The functions of acquired hypothetical proteins in the Acb complex are not currently known; however, several of these peptides share homology with genes in known pathogens (Table S5).

Siderophore distribution and conservation in the genus Acinetobacter

One identified pathogenic mechanism of A. baumannii infection is through the expression of siderophores [24]. However, the conservation of siderophore genes across the Acinetobacter genus is not currently known. The acinetobactin iron acquisition cluster, which consists of 25 kb of sequence in seven operons [71], in A. baumannii AYE (Table S3) [27], was aligned against all Acinetobacter genomes analyzed in this study with TBLASTN. Genes screened included BasD [25], and BasE, which activates the 2,3-dihydroxybenzoic acid molecule and transfers it to BasF [72]. The acinetobactin cluster was highly conserved in A. baumannii and the clade that includes A. pittii (Figure 3); homologous peptides were also observed in several A. haemolyticus isolates, as has been previously observed [72]. The A. baumannii clade also contains siderophore cluster 1 (ABAYE2001-2008) [21] that appears to be unique to A. baumannii (Figure 3). Components of the acinetoferrin system are conserved in the A. haemolyticus clade [73]; four components of the acinetoferrin cluster are homologous to peptides in A. baumannii (Figure 3). In addition to siderophore genes, the A. baumannii-nosocomialis clade also contains a unique TonB-dependent siderophore receptor protein (ABAYE1644) (Figure 3); these proteins are involved in sensing extra-cellular signals and responding through transcriptional regulation [74].

thumbnail
Figure 3. A heatmap of blast score ratio (BSR) [57] values for iron acquisition genes in Acinetobacter.

BSR values were visualized with the multi-experiment viewer [58]. Accession details for each gene in specific iron acquisition systems are shown in Table S3, with raw data shown in Table S5.

https://doi.org/10.1371/journal.pone.0054287.g003

Carbapenem-hydrolyzing oxacillinases (OXA)

Antibiotic resistance is common in nosocomial A. baumannii infections [75]. One mechanism of resistance to beta-lactams, including carbapenems, in A. baumannii is through the expression of OXA-type enzymes [14]. To determine their distribution, OXA genes (Table S3) were informatically screened against all sequenced Acinetobacter genomes. Only blaOXA-51-like [30] genes were found to be conserved across the majority of Acinetobacter spp. (Figure 4); genomes from the nosocomialis and baylyi clades lack a blaOXA-51 homolog (Table S4). The blaOXA-51-like genes are specific to each species and have been proposed as a method of positively identifying A. baumannii isolates [76]. An in silico screen demonstrated that blaOXA-51-like primers [76] not only align with all A. baumannii genomes (n = 89) analyzed in this study, but do not significantly align with any other Acinetobacter spp. (n = 47), including non-baumannii genomes in the Acb complex.

thumbnail
Figure 4. A heatmap of blast score ratio (BSR) [57] values for efflux pump and beta-lactamase genes identified in Acinetobacter.

BSR values were visualized with the multi-experiment viewer [58]. Accession details for each gene are shown in Table S3, with raw data shown in Table S5.

https://doi.org/10.1371/journal.pone.0054287.g004

In A. baumannii isolates that only contain blaOXA-51-like genes, the presence of the insertion element ISAba1 is required for carbapenem resistance [77]. For example, the multi-drug susceptible genomes AB058, A118, AB900 and AB307-0294 are all blaOXA-51-like positive and ISAba1 negative (Table S4). While ISAba1 appears to be conserved in A. baumannii, a BSR analysis demonstrated that the insertion element is also present in other Acinetobacter spp. (Figure 4).

Discussion

The genus Acinetobacter contains a mixture of non-pathogenic environmental isolates and wide-spread nosocomial pathogens [78]. Although other Acinetobacter spp. have been implicated in human infection, A. baumannii appears to be the most widespread and virulent. With the advent of whole-genome sequencing and large-scale comparative analyses, the genomic differences between A. baumannii and related isolates can now be identified. The current study represents the most comprehensive comparative analysis of the conservation and distribution of virulence factors across the Acinetobacter genus performed to date.

The Acinetobacter whole genome phylogeny confirms that the A. calcoaceticus-baumannii (Acb) complex is monophyletic (Figure 1). Isolates from this complex are frequently associated with nosocomial infections [40], [79]. However, isolates from the calcoaceticus clade (Figure 1) are typically environmental isolates not associated with serious human infection [14]. In addition, A. pittii strain PHEA-2, was isolated from industry wastewater [80]. From a clinical perspective, the identification of Acb complex isolates, which include both environmental isolates and human pathogens, may be misleading from a treatment perspective [14]. Genomic regions identified in this study and guided by a whole genome phylogeny may provide better discrimination between related stains of varying clinical significance. The identification of clinical isolates based on acquired genes on specific phylogenetic branches may help to develop diagnostics that can accurately classify infections from nosocomial Acinetobacter pathogens.

Whole genome sequence analysis of 136 Acinetobacter genomes has provided a comprehensive view of Acinetobacter evolution. Based on the analysis of several conserved genes, A. radioresistens is the most basal lineage of sequenced Acinetobacter genomes. In addition, the A. radioresistens clade was shown to have the smallest average genome size where the Acb complex has the largest. Although genome reduction is generally associated with a pathogenic lifestyle [81], genome expansion in pathogens has also been observed [82]. While the functional roles of all acquired genes in A. baumannii have not yet been determined, they may be associated with the persistence and virulence of A. baumannii in hospital environments.

A. baumannii thrives in hospital settings, largely due to its persistence on abiotic surfaces [83]. One mechanism for A. baumannii persistence in hospital settings is the presence of a putative tip adhesion gene, csuE. The csuE gene (ABK1_1276) is involved in pilus and biofilm formation [19] and is largely conserved (BSR>0.8) in the A. baumannii (>92% presence), A. nosocomialis (100% presence), and A. pittii (67% presence) clades; csuE is absent (BSR<0.4) from environmental isolates in the calcoaceticus clade as well as Acinetobacter genomes not in the Acb complex (Table S5). The presence of this gene has been associated with the persistence of A. baumannii on abiotic surfaces, such as plastic and glass [19]. The lack of a homologous gene in non-Acb isolates could explain why pathogens in the Acb complex persist in the hospital environment. Furthermore, A. baumannii can survive desiccation much better than most other Acinetobacter spp. [84]; however, A. radioresistens has been shown to be perhaps the most desiccation-tolerant Acinetobacter spp. [85]. This suggests a mechanism for attachment, as well as persistence, on abiotic hospital surfaces.

The role of siderophores in bacterial pathogenesis is well characterized [86]. When A. baumannii invades the host, one mechanism of persistence and toxicity is the acinetobactin iron-acquisition system [24]. Although not unique to A. baumannii, the acinetobactin cluster is well conserved in the species and is likely a contributing factor in A. baumannii survival and pathogenesis. Three unique iron-acquisition systems have been identified in A. baumannii including siderophore cluster 1, which appears to be unique to A. baumannii (Figure 3). In a study of gene expression in iron-limited media, all three iron acquisition systems identified in A. baumannii were up regulated [21]. The presence of multiple iron acquisition systems could provide a competitive advantage for the pathogen over other host microbiota.

Antibiotic resistance may be the most troubling aspect in the evolution of the Acb complex. A unique gene acquired in the evolution of A. baumannii, compared to other Acinetobacter spp., is a narrow-spectrum TEM-1 beta-lactamase, which is highly conserved in A. baumannii and A. nosocomialis; however, the clinical significance of this gene acquisition is unclear [14]. What is of great concern is the resistance of A. baumannii to broad-spectrum antibiotics, such as fluoroquinolones, especially in outbreak events [87]. Beta-lactamase genes are conserved in A. baumannii [88], which allow the bacterium to persist in the host and resist treatment therapies. Efflux pumps are also important mechanisms for antibiotic resistance in A. baumannii [42]; a comparative analysis demonstrated that A. baumannii contains an assortment of efflux pump genes that are not well conserved outside of the Acb complex (Figure 4). Although additional mechanisms of resistance, including 16S rRNA methylation [89], have been described for A. baumannii, they are more difficult to identify through comparative genomic analyses.

In addition to these known mechanisms of persistence, colonization, and infection, additional currently un-identified mechanisms likely play a role in the pathogenesis of A. baumannii and related isolates. In this study, transcriptional regulators have been identified that are unique to specific nodes in the Acb complex and have unknown regulatory function (Table 1); several of these regulators share homology with regulators in other pathogens, which suggests a role of these genes in A. baumannii pathogenesis. Characterization of these regulators with a global and an un-biased approach, such as RNA-sequencing [90], may be required in order to better understand the regulatory networks of A. baumannii; these experiments are currently on-going.

Population level whole genome sequencing has provided a comprehensive tool for the analysis of Acinetobacter evolution. Furthermore, the established whole genome phylogeny has provided insight into important evolutionary relationships that cannot be fully determined with single gene or MLST analysis. The comparative method described in this study demonstrates how whole genome sequence analysis can be used to study the flow of genomic information between species in a genus, thereby allowing for an understanding of the evolution of an environmental microbe into a nosocomial pathogen. Guided by a whole-genome phylogeny, this flow of information can then be used to focus studies on the functional characterization of novel genomic features and determine their role in pathogenesis. This method represents a new paradigm in the identification and characterization of emerging human pathogens, such as A. baumannii.

Supporting Information

Figure S1.

A phylogenetic tree inferred from an alignment of 16S rRNA gene sequences. The tree was inferred with FastTree2 with 1000 bootstrap replicates; bootstrap support values are shown at nodes. Clades were collapsed in ARB [63].

https://doi.org/10.1371/journal.pone.0054287.s001

(TIF)

Figure S2.

A comparison of phylogenetic trees inferred on either a whole genome alignment, or an alignment of concatenated multi-locus sequencing typing (MLST) (pubmlst.org/abaumannii) sequences informatically extracted from each genome analyzed in this study. Trees were inferred with FastTree2 [52]. Clades were colored based on assignments made from the whole genome phylogeny.

https://doi.org/10.1371/journal.pone.0054287.s002

(TIF)

Table S1.

Details of isolation, assembly, and accession of isolates sequenced in this study.

https://doi.org/10.1371/journal.pone.0054287.s003

(PDF)

Table S2.

Accession details of reference genomes analyzed in this study.

https://doi.org/10.1371/journal.pone.0054287.s004

(PDF)

Table S3.

Accession and annotation details for genes screened in this study.

https://doi.org/10.1371/journal.pone.0054287.s005

(PDF)

Table S4.

Raw blast score ratio (BSR) for each gene screened in this study.

https://doi.org/10.1371/journal.pone.0054287.s006

(PDF)

Table S5.

Homology of acquired genes in the Acinetobacter calcoaceticus-baumannii (Acb) complex to genes in other bacteria.

https://doi.org/10.1371/journal.pone.0054287.s007

(PDF)

Acknowledgments

We would like to thank Dr. Judith Noble-Wang, Division of Healthcare Quality Promotion, Centers for Disease Control and Prevention and Dr. Mike Saubolle and Bette Wojack of Laboratory Sciences of Arizona for providing isolates. In addition, we thank Jolene Bowers, Rebecca Rimsza, Andrew Knadler, Nathan Hicks, and Remy Hilsabeck for technical assistance.

Author Contributions

Conceived and designed the experiments: JWS EMD DME PK VGW. Performed the experiments: JDG EMD JMS. Analyzed the data: JWS JDG. Contributed reagents/materials/analysis tools: VGW JMS EMD. Wrote the paper: JWS.

References

  1. 1. Cisneros JM, Reyes MJ, Pachon J, Becerril B, Caballero FJ, et al. (1996) Bacteremia due to Acinetobacter baumannii: epidemiology, clinical findings, and prognostic features. Clin Infect Dis 22: 1026–1032.
  2. 2. Cisneros-Herreros JM, Garnacho-Montero J, Pachon-Ibanez ME (2005) Nosocomial pneumonia due to Acinetobacter baumannii. Enferm Infecc Microbiol Clin 23 Suppl 3: 46–51.
  3. 3. Zurawski DV, Thompson MG, McQueary CN, Matalka MN, Sahl JW, et al. (2012) Genome sequences of four divergent multidrug-resistant Acinetobacter baumannii strains isolated from patients with sepsis or osteomyelitis. J Bacteriol 194: 1619–1620.
  4. 4. Perez F, Hujer AM, Hujer KM, Decker BK, Rather PN, et al. (2007) Global challenge of multidrug-resistant Acinetobacter baumannii. Antimicrobial agents and chemotherapy 51: 3471–3484.
  5. 5. Lolans K, Rice TW, Munoz-Price LS, Quinn JP (2006) Multicity outbreak of carbapenem-resistant Acinetobacter baumannii isolates producing the carbapenemase OXA-40. Antimicrobial agents and chemotherapy 50: 2941–2945.
  6. 6. Tognim MC, Andrade SS, Silbert S, Gales AC, Jones RN, et al. (2004) Resistance trends of Acinetobacter spp. in Latin America and characterization of international dissemination of multi-drug resistant strains: five-year report of the SENTRY Antimicrobial Surveillance Program. International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases 8: 284–291.
  7. 7. Van Looveren M, Goossens H (2004) Antimicrobial resistance of Acinetobacter spp. in Europe. Clin Microbiol Infect 10: 684–704.
  8. 8. Yun HC, Murray CK, Roop SA, Hospenthal DR, Gourdine E, et al. (2006) Bacteria recovered from patients admitted to a deployed U.S. military hospital in Baghdad, Iraq. Military medicine 171: 821–825.
  9. 9. Schafer JJ, Mangino JE (2008) Multidrug-resistant Acinetobacter baumannii osteomyelitis from Iraq. Emerging infectious diseases 14: 512–514.
  10. 10. Scott P, Deye G, Srinivasan A, Murray C, Moran K, et al. (2007) An outbreak of multidrug-resistant Acinetobacter baumannii-calcoaceticus complex infection in the US military health care system associated with military operations in Iraq. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 44: 1577–1584.
  11. 11. Tien HC, Battad A, Bryce EA, Fuller J, Mulvey M, et al. (2007) Multi-drug resistant Acinetobacter infections in critically injured Canadian forces soldiers. BMC infectious diseases 7: 95.
  12. 12. Baumann P (1968) Isolation of Acinetobacter from soil and water. Journal of bacteriology 96: 39–42.
  13. 13. Dijkshoorn L, Nemec A, Seifert H (2007) An increasing threat in hospitals: multidrug-resistant Acinetobacter baumannii. Nat Rev Microbiol 5: 939–951.
  14. 14. Peleg AY, Seifert H, Paterson DL (2008) Acinetobacter baumannii: emergence of a successful pathogen. Clin Microbiol Rev 21: 538–582.
  15. 15. Seifert H, Baginski R, Schulze A, Pulverer G (1993) The distribution of Acinetobacter species in clinical culture materials. Zentralblatt fur Bakteriologie : international journal of medical microbiology 279: 544–552.
  16. 16. Horii T, Tamai K, Mitsui M, Notake S, Yanagisawa H (2011) Blood stream infections caused by Acinetobacter ursingii in an obstetrics ward. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases 11: 52–56.
  17. 17. Nemec A, De Baere T, Tjernberg I, Vaneechoutte M, van der Reijden TJ, et al. (2001) Acinetobacter ursingii sp. nov. and Acinetobacter schindleri sp. nov., isolated from human clinical specimens. International journal of systematic and evolutionary microbiology 51: 1891–1899.
  18. 18. Russo TA, Luke NR, Beanan JM, Olson R, Sauberan SL, et al. (2010) The K1 capsular polysaccharide of Acinetobacter baumannii strain 307-0294 is a major virulence factor. Infection and immunity 78: 3993–4000.
  19. 19. Tomaras AP, Dorsey CW, Edelmann RE, Actis LA (2003) Attachment to and biofilm formation on abiotic surfaces by Acinetobacter baumannii: involvement of a novel chaperone-usher pili assembly system. Microbiology 149: 3473–3484.
  20. 20. Wand ME, Bock LJ, Turton JF, Nugent PG, Sutton JM (2012) Acinetobacter baumannii virulence is enhanced in Galleria mellonella following biofilm adaptation. Journal of medical microbiology 61: 470–477.
  21. 21. Eijelkamp BA, Hassan KA, Paulsen IT, Brown MH (2011) Investigation of the human pathogen Acinetobacter baumannii under iron limiting conditions. BMC genomics 12: 126.
  22. 22. Dorsey CW, Tolmasky ME, Crosa JH, Actis LA (2003) Genetic organization of an Acinetobacter baumannii chromosomal region harbouring genes related to siderophore biosynthesis and transport. Microbiology 149: 1227–1238.
  23. 23. Mihara K, Tanabe T, Yamakawa Y, Funahashi T, Nakao H, et al. (2004) Identification and transcriptional organization of a gene cluster involved in biosynthesis and transport of acinetobactin, a siderophore produced by Acinetobacter baumannii ATCC 19606T. Microbiology 150: 2587–2597.
  24. 24. Yamamoto S, Okujo N, Sakakibara Y (1994) Isolation and structure elucidation of acinetobactin, a novel siderophore from Acinetobacter baumannii. Archives of microbiology 162: 249–254.
  25. 25. Dorsey CW, Tomaras AP, Connerly PL, Tolmasky ME, Crosa JH, et al. (2004) The siderophore-mediated iron acquisition systems of Acinetobacter baumannii ATCC 19606 and Vibrio anguillarum 775 are structurally and functionally related. Microbiology 150: 3657–3667.
  26. 26. Gaddy JA, Arivett BA, McConnell MJ, Lopez-Rojas R, Pachon J, et al. (2012) Role of acinetobactin-mediated iron acquisition functions in the interaction of Acinetobacter baumannii strain ATCC 19606T with human lung epithelial cells, Galleria mellonella caterpillars, and mice. Infection and immunity 80: 1015–1024.
  27. 27. Vallenet D, Nordmann P, Barbe V, Poirel L, Mangenot S, et al. (2008) Comparative analysis of Acinetobacters: three genomes for three lifestyles. PLoS One 3: e1805.
  28. 28. Livermore DM (1995) beta-Lactamases in laboratory and clinical resistance. Clinical microbiology reviews 8: 557–584.
  29. 29. Nordmann P, Poirel L (2002) Emerging carbapenemases in Gram-negative aerobes. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 8: 321–331.
  30. 30. Brown S, Young HK, Amyes SG (2005) Characterisation of OXA-51, a novel class D carbapenemase found in genetically unrelated clinical strains of Acinetobacter baumannii from Argentina. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 11: 15–23.
  31. 31. Donald HM, Scaife W, Amyes SG, Young HK (2000) Sequence analysis of ARI-1, a novel OXA beta-lactamase, responsible for imipenem resistance in Acinetobacter baumannii 6B92. Antimicrobial agents and chemotherapy 44: 196–199.
  32. 32. Poirel L, Marque S, Heritier C, Segonds C, Chabanon G, et al. (2005) OXA-58, a novel class D {beta}-lactamase involved in resistance to carbapenems in Acinetobacter baumannii. Antimicrobial agents and chemotherapy 49: 202–208.
  33. 33. Coyne S, Rosenfeld N, Lambert T, Courvalin P, Perichon B (2010) Overexpression of resistance-nodulation-cell division pump AdeFGH confers multidrug resistance in Acinetobacter baumannii. Antimicrob Agents Chemother 54: 4389–4393.
  34. 34. Magnet S, Courvalin P, Lambert T (2001) Resistance-nodulation-cell division-type efflux pump involved in aminoglycoside resistance in Acinetobacter baumannii strain BM4454. Antimicrob Agents Chemother 45: 3375–3380.
  35. 35. Damie-Piolle L, Magnet S, Bremont S, Lambert T, Courvalin P (2008) AdeIJK, a resistance-nodulation-cell division pump effluxing multiple antibiotics in Acinetobacter baumannii. Antimicrob Agents Chemother 52: 557–562.
  36. 36. Su XZ, Chen J, Mizushima T, Kuroda T, Tsuchiya T (2005) AbeM, an H+-coupled Acinetobacter baumannii multidrug efflux pump belonging to the MATE family of transporters. Antimicrobial agents and chemotherapy 49: 4362–4364.
  37. 37. Gerner-Smidt P, Tjernberg I, Ursing J (1991) Reliability of phenotypic tests for identification of Acinetobacter species. Journal of clinical microbiology 29: 277–282.
  38. 38. Nemec A, Krizova L, Maixnerova M, van der Reijden TJ, Deschaght P, et al. (2011) Genotypic and phenotypic characterization of the Acinetobacter calcoaceticus-Acinetobacter baumannii complex with the proposal of Acinetobacter pittii sp. nov. (formerly Acinetobacter genomic species 3) and Acinetobacter nosocomialis sp. nov. (formerly Acinetobacter genomic species 13TU). Research in microbiology 162: 393–404.
  39. 39. Kang YS, Jung J, Jeon CO, Park W (2011) Acinetobacter oleivorans sp. nov. is capable of adhering to and growing on diesel-oil. Journal of microbiology 49: 29–34.
  40. 40. Wisplinghoff H, Paulus T, Lugenheim M, Stefanik D, Higgins PG, et al. (2012) Nosocomial bloodstream infections due to Acinetobacter baumannii, Acinetobacter pittii and Acinetobacter nosocomialis in the United States. The Journal of infection 64: 282–290.
  41. 41. Adams MD, Chan ER, Molyneaux ND, Bonomo RA (2010) Genomewide Analysis of Divergence of Antibiotic Resistance Determinants in Closely Related Isolates of Acinetobacter baumannii. Antimicrob Agents Chemother 54: 3569–3577.
  42. 42. Fournier PE, Vallenet D, Barbe V, Audic S, Ogata H, et al. (2006) Comparative genomics of multidrug resistance in Acinetobacter baumannii. PLoS Genet 2: e7.
  43. 43. Hornsey M, Loman N, Wareham DW, Ellington MJ, Pallen MJ, et al. (2011) Whole-genome comparison of two Acinetobacter baumannii isolates from a single patient, where resistance developed during tigecycline therapy. The Journal of antimicrobial chemotherapy 66: 1499–1503.
  44. 44. Iacono M, Villa L, Fortini D, Bordoni R, Imperi F, et al. (2008) Whole-genome pyrosequencing of an epidemic multidrug-resistant Acinetobacter baumannii strain belonging to the European clone II group. Antimicrob Agents Chemother 52: 2616–2625.
  45. 45. Sahl JW, Johnson JK, Harris AD, Phillippy AM, Hsiao WW, et al. (2011) Genomic comparison of multi-drug resistant invasive and colonizing Acinetobacter baumannii isolated from diverse human body sites reveals genomic plasticity. BMC genomics 12: 291.
  46. 46. Imperi F, Antunes LC, Blom J, Villa L, Iacono M, et al. (2011) The genomics of Acinetobacter baumannii: insights into genome plasticity, antimicrobial resistance and pathogenicity. IUBMB life 63: 1068–1074.
  47. 47. Di Nocera PP, Rocco F, Giannouli M, Triassi M, Zarrilli R (2011) Genome organization of epidemic Acinetobacter baumannii strains. BMC microbiology 11: 224.
  48. 48. Zerbino DR, Birney E (2008) Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res 18: 821–829.
  49. 49. Otto TD, Sanders M, Berriman M, Newbold C (2010) Iterative Correction of Reference Nucleotides (iCORN) using second generation sequencing technology. Bioinformatics 26: 1704–1707.
  50. 50. Gardner SN, Slezak T (2010) Scalable SNP analyses of 100 bacterial or viral genomes. J Forensic Res 1 .
  51. 51. Dress AW, Flamm C, Fritzsch G, Grunewald S, Kruspe M, et al. (2008) Noisy: identification of problematic columns in multiple sequence alignments. Algorithms for molecular biology : AMB 3: 7.
  52. 52. Price MN, Dehal PS, Arkin AP (2010) FastTree 2–approximately maximum-likelihood trees for large alignments. PLoS One 5: e9490.
  53. 53. Delcher AL, Bratke KA, Powers EC, Salzberg SL (2007) Identifying bacterial genes and endosymbiont DNA with Glimmer. Bioinformatics 23: 673–679.
  54. 54. Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26: 2460–2461.
  55. 55. Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular Biology Open Software Suite. Trends in genetics : TIG 16: 276–277.
  56. 56. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402.
  57. 57. Rasko DA, Myers GS, Ravel J (2005) Visualization of comparative genomic analyses by BLAST score ratio. BMC Bioinformatics 6: 2.
  58. 58. Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, et al. (2006) TM4 microarray software suite. Methods Enzymol 411: 134–193.
  59. 59. Edgar RC (2004) MUSCLE: a multiple sequence alignment with reduced time and space complexity. BMC Bioinformatics 5: 113–113.
  60. 60. Sahl JW, Steinsland H, Redman JC, Angiuoli SV, Nataro JP, et al. (2011) A comparative genomic analysis of diverse clonal types of enterotoxigenic Escherichia coli reveals pathovar-specific conservation. Infect Immun 79: 950–960.
  61. 61. Nawrocki EP, Kolbe DL, Eddy SR (2009) Infernal 1.0: inference of RNA alignments. Bioinformatics 25: 1335–1337.
  62. 62. DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, et al. (2006) Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl Environ Microbiol 72: 5069–5072.
  63. 63. Ludwig W, Strunk O, Westram R, Richter L, Meier H, et al. (2004) ARB: a software environment for sequence data. Nucl Acids Res 32: 1363–1371.
  64. 64. Chan JZM, Halachev MR, Loman NJ, Constantinidou C, Pallen MJ (2012) Defining bacterial species in the genomic era: insights from the genus Acinetobacter. BMC Microbiol Accepted and in press.
  65. 65. Higgins PG, Lehmann M, Wisplinghoff H, Seifert H (2010) gyrB multiplex PCR to differentiate between Acinetobacter calcoaceticus and Acinetobacter genomic species 3. Journal of clinical microbiology 48: 4592–4594.
  66. 66. Dijkshoorn L, Nemec A (2008) The diversity of the genus Acinetobacter. In: Gerischer U, editor. Acinetobacter molecular microbiology. Norfolk, United Kingdom: Caister Academic Press. pp. 1–34.
  67. 67. Rasko DA, Rosovitz MJ, Myers GS, Mongodin EF, Fricke WF, et al. (2008) The pangenome structure of Escherichia coli: comparative genomic analysis of E. coli commensal and pathogenic isolates. J Bacteriol 190: 6881–6893.
  68. 68. Seo YS, Lim J, Choi BS, Kim H, Goo E, et al. (2011) Complete genome sequence of Burkholderia gladioli BSR3. Journal of bacteriology 193: 3149.
  69. 69. Bobrowicz P, Wysocki R, Owsianik G, Goffeau A, Ulaszewski S (1997) Isolation of three contiguous genes, ACR1, ACR2 and ACR3, involved in resistance to arsenic compounds in the yeast Saccharomyces cerevisiae. Yeast 13: 819–828.
  70. 70. Lalucat J, Bennasar A, Bosch R, Garcia-Valdes E, Palleroni NJ (2006) Biology of Pseudomonas stutzeri. Microbiology and molecular biology reviews : MMBR 70: 510–547.
  71. 71. Penwell WF, Arivett BA, Actis LA (2012) The Acinetobacter baumannii entA gene located outside the acinetobactin cluster is critical for siderophore production, iron acquisition and virulence. PloS one 7: e36493.
  72. 72. Drake EJ, Duckworth BP, Neres J, Aldrich CC, Gulick AM (2010) Biochemical and structural characterization of bisubstrate inhibitors of BasE, the self-standing nonribosomal peptide synthetase adenylate-forming enzyme of acinetobactin synthesis. Biochemistry 49: 9292–9305.
  73. 73. Okujo N, Sakakibara Y, Yoshida T, Yamamoto S (1994) Structure of acinetoferrin, a new citrate-based dihydroxamate siderophore from Acinetobacter haemolyticus. Biometals : an international journal on the role of metal ions in biology, biochemistry, and medicine 7: 170–176.
  74. 74. Koebnik R (2005) TonB-dependent trans-envelope signalling: the exception or the rule? Trends in microbiology 13: 343–347.
  75. 75. Maragakis LL, Perl TM (2008) Acinetobacter baumannii: epidemiology, antimicrobial resistance, and treatment options. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 46: 1254–1263.
  76. 76. Turton JF, Woodford N, Glover J, Yarde S, Kaufmann ME, et al. (2006) Identification of Acinetobacter baumannii by detection of the blaOXA-51-like carbapenemase gene intrinsic to this species. Journal of clinical microbiology 44: 2974–2976.
  77. 77. Turton JF, Ward ME, Woodford N, Kaufmann ME, Pike R, et al. (2006) The role of ISAba1 in expression of OXA carbapenemase genes in Acinetobacter baumannii. FEMS microbiology letters 258: 72–77.
  78. 78. Bergogne-Berezin E, Towner KJ (1996) Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clinical microbiology reviews 9: 148–165.
  79. 79. Koh TH, Tan TT, Khoo CT, Ng SY, Tan TY, et al. (2012) Acinetobacter calcoaceticus-Acinetobacter baumannii complex species in clinical specimens in Singapore. Epidemiology and infection 140: 535–538.
  80. 80. Zhan Y, Yan Y, Zhang W, Yu H, Chen M, et al. (2011) Genome sequence of Acinetobacter calcoaceticus PHEA-2, isolated from industry wastewater. Journal of bacteriology 193: 2672–2673.
  81. 81. Moran NA (2002) Microbial minimalism: genome reduction in bacterial pathogens. Cell 108: 583–586.
  82. 82. Raffaele S, Kamoun S (2012) Genome evolution in filamentous plant pathogens: why bigger can be better. Nature reviews Microbiology 10: 417–430.
  83. 83. Brossard KA, Campagnari AA (2012) The Acinetobacter baumannii biofilm-associated protein plays a role in adherence to human epithelial cells. Infection and immunity 80: 228–233.
  84. 84. Jawad A, Heritage J, Snelling AM, Gascoyne-Binzi DM, Hawkey PM (1996) Influence of relative humidity and suspending menstrua on survival of Acinetobacter spp. on dry surfaces. Journal of clinical microbiology 34: 2881–2887.
  85. 85. Jawad A, Snelling AM, Heritage J, Hawkey PM (1998) Exceptional desiccation tolerance of Acinetobacter radioresistens. The Journal of hospital infection 39: 235–240.
  86. 86. Neilands JB (1995) Siderophores: structure and function of microbial iron transport compounds. The Journal of biological chemistry 270: 26723–26726.
  87. 87. Dijkshoorn L, Aucken H, Gerner-Smidt P, Janssen P, Kaufmann ME, et al. (1996) Comparison of outbreak and nonoutbreak Acinetobacter baumannii strains by genotypic and phenotypic methods. Journal of clinical microbiology 34: 1519–1525.
  88. 88. Poirel L, Nordmann P (2006) Carbapenem resistance in Acinetobacter baumannii: mechanisms and epidemiology. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 12: 826–836.
  89. 89. Doi Y, Adams JM, Yamane K, Paterson DL (2007) Identification of 16S rRNA methylase-producing Acinetobacter baumannii clinical strains in North America. Antimicrobial agents and chemotherapy 51: 4209–4210.
  90. 90. Wang Z, Gerstein M, Snyder M (2009) RNA-Seq: a revolutionary tool for transcriptomics. Nature reviews Genetics 10: 57–63.