Resources for Genetic and Genomic Analysis of Emerging Pathogen Acinetobacter baumannii

Abstract

Acinetobacter baumannii is a Gram-negative bacterial pathogen notorious for causing serious nosocomial infections that resist antibiotic therapy. Research to identify factors responsible for the pathogen's success has been limited by the resources available for genome-scale experimental studies. This report describes the development of several such resources for A. baumannii strain AB5075, a recently characterized wound isolate that is multidrug resistant and displays robust virulence in animal models. We report the completion and annotation of the genome sequence, the construction of a comprehensive ordered transposon mutant library, the extension of high-coverage transposon mutant pool sequencing (Tn-seq) to the strain, and the identification of the genes essential for growth on nutrient-rich agar. These resources should facilitate large-scale genetic analysis of virulence, resistance, and other clinically relevant traits that make A. baumannii a formidable public health threat.

Importance: Acinetobacter baumannii is one of six bacterial pathogens primarily responsible for antibiotic-resistant infections that have become the scourge of health care facilities worldwide. Eliminating such infections requires a deeper understanding of the factors that enable the pathogen to persist in hospital environments, establish infections, and resist antibiotics. We present a set of resources that should accelerate genome-scale genetic characterization of these traits for a reference isolate of A. baumannii that is highly virulent and representative of current outbreak strains.

FIG 1

Resistance islands in AB5075-UW. Resistance island 1 (TnAbaR5075) is 13.5 kbp long and, like most other A. baumannii resistance islands, is a TnAbaR1-like island inserted in the comM gene. The TnAbaR modules are labeled according to a previous convention (59). Compared to most known comM islands, RI-1 is small and contains one rather than two copies of the Tn6018 module. RI-1 carries predicted arsenic and antimony resistance genes (arsCRCBH), a cadmium resistance locus (cadAR), and a sulfate transporter gene (sup). Resistance island 2 is a 13.5-kbp island in plasmid p1 consisting of a complex class 1 integron truncated at both ends by direct repeats of a 439-bp miniature inverted-repeat transposable element (MITE)-like sequence (green). The indicated 1,720- and 3,728-bp segments (including the MITE-like sequences) are 100% identical to sequences defining the ends of different class 1 integrons in other Acinetobacter strains (see Text S1 in the supplemental material). The 5-bp target site duplication of the plasmid at the RI-2 insertion site is shown. RI-2 includes genes for resistance to β-lactams (bla_GES-14), aminoglycosides (aacA4, aadA2, aadB, and strAB), chloramphenicol (cmlA5), and trimethoprim (dfrA7). Circles, intact attC sites (“59-bp elements”); red, interrupted gene fragments; dark red, putative pseudogenes. Additional gene abbreviations follow a previous convention (59).

FIG 2

Web browser map of three-allele library mutants. A partial screen shot of the transposon locations in a representative region of the AB5075 genome from the Transposon Mutant Library Browser (http://tools.uwgenomics.org/tn_mutants) is shown. Transposons are represented as triangles, with positions above or below the line corresponding to their orientations in the genome. Filled triangles represent insertions whose locations were confirmed by multiple sequencings, and open triangles represent locations determined by single sequencings. Placing the mouse cursor over individual triangles in the browser window reveals information about the insertions and facilitates ordering the corresponding mutants. CDS, coding sequence.

FIG 3

Profile of Tn-seq reads from the high-saturation mutant pool. The genes in a representative segment of the genome are shown. Vertical bars in the lower part of the diagram represent log₂-transformed normalized read counts per insertion site (counts averaged from the six technical replicates), with bars above or below the line reflecting insertion in forward or reverse orientation relative to the genome, respectively. Scale maximum corresponds to a log₂ value of 12. The genes thiD, acpP, fabG, and rpmF, which encode a thiamine metabolism protein, acyl-carrier protein, a lipid metabolism protein, and a ribosomal protein, respectively, lack insertions and are candidate essential genes. The fabD gene (lipid metabolism) is also classified as a candidate essential gene based on its low density of insertions (see Materials and Methods; also see Fig. S3 in the supplemental material). The significance of the few insertions detected in fabD is unknown, but one possibility is that they are not fully inactivating, e.g., due to insertion in transiently duplicated regions.