Introduction
Environmental bacteria represent a huge resource of antibiotic-resistant species awaiting characterization [8]. The majority of antibiotic-resistant bacteria in the environment are derived from clinical and veterinary places, enter rivers from sewage and waste water, and can then be delivered to, and accumulate in, soil [13]. To understand the nature of antibiotic resistance, previous studies have sought to characterize antibiotic resistance genes from soil using agar-based culture-dependent protocols under laboratory conditions [1,35]. For example, artificial agar media containing target antibiotics have been used to isolate and characterize specific bacterial species that are resistant to particular antibiotics, including streptomycin-resistant Pseudomonas fluorescens, penicillin-resistant P. fluorescens and Burkholderia cepacia, and chloramphenicol-resistant Flavobacterium sp. and Streptomyces venezuelae [1,35]. However, a number of recent studies indicate that only 0.1 to 1% of bacteria from environmental samples, such as soil, are culturable [3]. These data suggest that many antibiotic resistance genes remain undiscovered and have not yet been characterized, providing a potential source of information relevant to human and animal health.
To overcome the limitations of culture-dependent methods, a next-generation sequencing-based metagenomic approach has been employed and evaluated for its ability to identify unknown antibiotic resistance genes [40]. Beyond a sequencing-based metagenome approach, function-based (functional) metagenome analysis allowed metagenome clones to be selected and characterized by biological validation [2,10,17,46,47]. Clearly, unknown antibiotic resistance genes cannot be detected through database searches; therefore, functional metagenomic techniques have been developed to address this problem. Functional metagenome libraries are generally constructed in a surrogate host (usually Escherichia coli) to facilitate expression of fragmented environmental DNA [6,30]. Screening of functional metagenome libraries has led to the identification of a number of novel antibiotic resistance genes from soil, including beta-lactamases, which mediate resistance to carbenicillin and amoxicillin [11], acetyltransferases for resistance to kanamycin [38], drug resistance transporters for florfenicol and chloramphenicol [23,44], and tetracycline efflux proteins mediating resistance to tetracycline [40]. Despite the fact that gentamicin is a powerful broad-spectrum antibiotic, gentamicin resistance genes are poorly represented in previous metagenome mining reports [14,27]. Aminoglycoside (including gentamicin) resistance genes play critical roles in attenuating antibiotic uptake or decreasing cell permeability, altering ribosomal-binding sites and production of aminoglycoside-modifying enzymes, resulting in poor ribosome binding [9,42]. Enzymatic modification, such as ATP-dependent O-phosphorylation by phosphotransferases (APH), ATP-dependent O-adenylation by nucleotidyltransferases, and acetyl CoA-dependent N-acetylation by acetyltransferases, is the most common type of aminoglycoside resistance [43].
Here, we applied a functional metagenomic approach to isolate unknown gentamicin resistance genes from environmental samples. We employed a metagenome library from soil bacteria communities into E. coli and isolated four clones conferring gentamicin resistance. All four clones encoded a putative RNA ligase that was similar to T4 rnlA, a known bacterial toxin [29,32]. We then characterized the molecular mechanism underlying the antibiotic resistance mediated by this gene. To the best of our knowledge, this is the first study to demonstrate that toxin-like genes from a soil metagenome can confer gentamicin resistance.
Materials and Methods
Bacterial Strains and Culture Conditions
E. coli strain EPI300 was used for metagenome library construction and strain DH5α was used for shot-gun library construction. E. coli cells were routinely cultured in Luria-Bertani (LB) medium (Affymetrix, Cleveland OH, USA) at 37℃, and plating media contained 2% agar. When appropriate, media were supplemented with 20 μg/ml chloramphenicol and 100 μg/ml ampicillin for strain and plasmid maintenance, respectively. Antibiotics were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Metagenome Library Construction
Metagenome libraries were obtained from the Metagenome Bank of the Microbial Genomics Frontier Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB) [24,25]. Briefly, the libraries were constructed from various soil samples that included 38 spots in South Korea, such as mud flat, compost, paddy soils, upland soils, and rhizosphere soils. The fosmid libraries consisted of extracted DNA fragments from each soil source (35–40 kb), which were cloned into the vector pCC1fos. Fosmid clones were deposited into single wells of 96-well microtiter plates (a single pool per well) with an average of 1,000 clones per single pool. The number of total clones were mathematically 5.76 × 105.
Screening for Gentamicin-Resistant Clones
The fosmid libraries were pooled and spread on LB agar containing 10, 20, and 40 μg/ml gentamicin for isolating different capacities of antibiotic resistance. These concentrations of gentamicin were chosen because they inhibit the growth of the host E. coli strain EPI300. The library was plated in triplicates and plates were incubated aerobically at 37℃ for 24–48 h. Fosmid clones pA1, pA3, pC1, and pG1 that were resistant on three replicate plates were selected, their resistance phenotype was verified by re-streaking onto agar containing 20 μg/ml chloramphenicol (pCC1fos vector marker gene), and stocks were prepared for further analysis. To screen for genes conferring gentamicin resistance, high-molecular-weight DNA from fosmid clone pA1 isolated from paddy soils from Asan-si, Gongju-si, and Suncheon-si was sheared [39] and a shot-gun library (2–3 kb) was constructed in the pUC118 vector by a commercial company (Genotech, Daejeon, Korea). The shot-gun library fragments were transformed into DH5α competent cells (Enzynomics, Daejeon, Korea). The shot-gun library was plated onto LB agar with 10, 20, and 40 μg/ml gentamicin and incubated aerobically at 37℃ for 24–48 h. Shot-gun clones pA1b, pA1g, pA1h, and pA1j that were resistant in three replicate plates were selected, their resistance phenotype was verified by re-streaking onto agar containing 100 μg/ml ampicillin (pUC118 vector marker gene), and stocks were prepared for further analysis. Plasmid DNA was isolated from individual resistant clones and used for repeat transformation of E. coli DH5α cells to confirm that the plasmid was responsible for the resistance phenotype.
Spot Assays
Cultures containing fosmid clones, vector control (pCC1fos), and positive control (pDONR207: containing a known gentamicin resistance gene, acetyltransferases, acc) were grown to an OD600 of 0.3 in LB containing 20 μg/ml chloramphenicol at 37℃ and allowed to grow for some time in the presence of 100 μM IPTG to induce expression of the cloned gene product. Ten microliters of 10-fold serial dilutions (1:10 to 1:104) of each bacterial culture was spotted onto LB agar plates containing 0, 1.25, 2.5, 5, 10, 20, and 40 μg/ml gentamicin. The plates were incubated aerobically at 37℃ for 24 h.
Minimum Inhibitory Concentration Assays
Minimum inhibitory concentrations (MICs) of antibiotics for shot-gun clones were determined by serial dilution assay of the relevant antibiotic in LB broth. E. coli DH5α cultures containing the pUC118 vector were grown to an OD600 of 0.3 in LB containing 100 μg/ml ampicillin with shaking at 37°C. After inoculation of cultures into LB broth (0.5%), the suspensions were serially diluted from the suspensions containing 50 μg/ml gentamicin [49]. Positive (broth +inoculum only) and negative (broth + 50 μg/ml gentamicin) controls were included in each assay. Tubes were incubated aerobically at 37°C for 24 h on a shaker at 200 rpm. MICs were measured and compared with that of the vector control (pUC118) to determine the relative increase in resistance. For comparison, spot assays were conducted for each population of clones. Ten microliters of 10-fold serial dilutions of resistant cells grown in the presence of 3 or 12 μg/ml gentamicin was applied to LB agar plates containing the same concentrations of gentamicin. The MIC50 and MIC90 values were calculated by probit analysis [12,37]. Determination of log10 gentamicin concentration values were taken with the probit points 5.0000 (MIC50) and 6.2816 (MIC90).
Sequencing of Fosmid and Shot-Gun Clones
Plasmids were extracted from bacterial cultures using the QIAprep Spin Mini Prep kit (Qiagen, Sussex, UK) and sequenced by a commercial company (Genotech, Korea). Protein-coding genes were identified using Prodigal v2.6.1 [18]. Predicted coding sequences (CDSs) were BLAST-searched against UniProt [51], Pfam [4], and Clusters of Orthologous Groups (COG) [45] databases to gain insight into the molecular functions and family classifications of predicted genes. Signal peptides and transmembrane helices were predicted using SignalP v4.1 [34] and TMHMM v2.0 [20]. rRNA, tRNA, and other miscellaneous features were predicted using RNAmmer v1.2 [22], tRNAscan-SE v1.21 [26], and Rfam v12.0 [16]. The graphic circular map of the genome was constructed and visualized using Circos v0.67 [21]. Automatic detection of clustered regularly interspaced short palindromic repeats (CRISPR) was performed using MinCED v0.2.0 [5]. Shot-gun clones were sequenced using M13 forward and reverse primers and data analyzed using NCBI BLAST and NCBI ORF finder to identify resistance genes. Both nucleotide and amino acid sequence alignments were calculated using BLAST (http:// blast.ncbi.nlm.nih.gov), and putative protein functions were annotated based on similarities to sequences in BLAST2go v3.2 [7] and Prodigal [18].
Metagenome Fragment Analyses
The organism identification of the pA1 fosmid clone was predicted using PhylopythiaS, a bioinformatics search engine [28]. (http://binning.bioinf.mpi-inf.mpg.de/).
Results
Soil Metagenome Screening for Gentamicin-Resistant Fosmid Clones
Screening of a 576-fosmid-clone metagenome library constructed from soil DNA led to the selection of four clones stably expressing a resistance phenotype on LB plates containing 10 and 20 μg/ml gentamicin. The MICs of gentamicin were determined for the four fosmid clones, pA1, pA3, pC1, and pG1 (Fig. 1); all showed stable growth on LB agar plates containing 20 μg/ml gentamicin, whereas the pCC1fos vector control was susceptible at 20 μg/ml gentamicin. Clones pA1, pA3, pC1, and pG1, therefore, demonstrated an 8-fold higher resistance to gentamicin than the pCC1fos fosmid vector control. E. coli cells with plasmid pDONR207 (containing a known gentamicin resistance gene, acc) were used as a positive control as they conferred 16-fold greater gentamicin resistance than the pCC1fos negative control (Fig. 1C). We chose clone pA1 for further characterization, due to its consistent phenotype occurrence.
Fig. 1.Assessment of the gentamicin resistance of fosmid clones. Spot assay of four selected fosmid clones. Ten-fold serial dilutions of a log-phase culture were spotted onto LB agar media containing 0, 1.25, 2.5, 5, 10, 20, or 40 μg/ml gentamicin. Representative (A) 2.5 and (B) 20 μg/ml gentamicin plates. (C) Data plotted as a graph. pCC1fos and pDONR207 were used as negative and positive controls for gentamicin resistance, respectively.
Screening of Shot-Gun Clones to Identify Gentamicin Resistance Genes
To identify specific genes associated with gentamicin resistance, the pA1 fosmid clone (38 kb) was fragmented and a shot-gun library of the resulting fragments constructed. The average size of the cloned DNA fragments was 2–3 kb. The MICs of gentamicin were determined for four shot-gun clones originating from fosmid pA1; pA1b, pA1g, pA1h, and pA1j (Fig. 2A). All four clones grew in the presence of 12–25 μg/ml gentamicin, representing a concentration 4- to 8-fold higher than that in which the vector control grew. This indicated that the resistance to gentamicin conferred by the shot-gun clones was similar to that conferred by the fosmid clone, whereas cultures containing the pUC118 vector control were susceptible to 3 μg/ml gentamicin (Fig. 2A). Consistent with the results, all four shot-gun clones showed resistance to both 3 and 12 μg/ml gentamicin when grown on LB agar, whereas the control vector grew on plates containing 3 μg/ml but not 12 μg/ml gentamicin (Figs. 2B and 2C). Likewise, pA1 showed resistance to both 3 and 12 μg/ml gentamicin, whereas the fosmid vector control pCC1fos showed growth inhibition at 12 μg/ml gentamicin (Figs. 2B and 2C). MICs for 50% (MIC50) and 90% (MIC90) of shot-gun and fosmid clones were calculated by log-probit analysis (Table 1). The MIC50 values were 0.2, 1.1, 0.9, and 1.0 μg/ml for pA1b, pA1g, pA1h, and pA1j, respectively; the MIC90 values were 1.3, 1.6, 1.4, and 1.4 μg/ml for pA1b, pA1g, pA1h, and pA1j, respectively. The MIC50 value of pA1b (0.2) out of the four selected clones was the lowest, whereas the MIC90 value of pA1b (1.3) was not much different than that of the other clones (1.4 and 1.6) (Table 1). The MIC50 and MIC90 of the fosmid clone pA1 were close to pA1b.
Fig. 2.Characterization of gentamicin-resistant shot gun clones. (A) Minimum inhibitory concentrations of gentamicin for each of the four shot-gun clones were determined in liquid culture using 10-fold serial dilution assays. Cultures were incubated aerobically at 37℃ for 24h, and pUC118 was used as a vector control. Spot assay of four shot-gun clones and fosmid clone pA1. Ten-fold serial dilutions of each culture grown in gentamicin were spotted onto LB agar media containing the same concentrations, (B) 3 and (C) 12 μg/ml of gentamicin. pUC118 and pCC1fos were used as vector controls for shot-gun and fosmid clones, respectively.Variation in the resistance profiles of the shot-gun clones. Spot assays of four shot-gun clones and the pUC118 (vector control). Ten-fold serial dilutions of a log-phase culture were spotted onto LB agar media containing (D) 50 μg/ml kanamycin, (E) 100 μg/ml streptomycin, and (F) 25 μg/ml spectinomycin.
Table 1.aMICs were calculated by log-probit analysis.
Resistance of the Shot-Gun Clones to Various Antibiotics
To characterize the resistance profile of the shot-gun clones, their susceptibility to a wide range of antibiotics classes, including cephalosporin (cephalothin), aminoglycosides (kanamycin and streptomycin), MLSK (spectinomycin), tetracycline, lipopeptides (colistin), quinolones (ciprofloxacin), fluoroquinolones (lomefloxacin), and trimethoprim, was examined. The MIC of each antibiotic for cultures containing the pUC118 control vector was first determined and shot-gun clones were subsequently tested for susceptibility. The shot-gun clones were resistant to streptomycin, kanamycin, and spectinomycin (Figs. 2D, 2E, and 2F); however, their susceptibility to all other antibiotics was similar to that of the vector control (data not shown). Streptomycin, kanamycin, and gentamicin are all aminoglycosides, which inhibit protein synthesis and, although not of the same class (Figs. 2D and 2E), spectinomycin also works by the same mechanism (Fig. 2F). These results strongly suggest that a gene sequence in the shot-gun clones plays an important role in blocking protein synthesis inhibition.
Sequencing of the Full-Length pA1 Fosmid Sequence
The general features of the pA1 fosmid clone sequence are shown in Table 2 and Fig. 3A. The complete sequence of the fosmid insert comprised 38,400 bp with a G+C content of 67.50% (Table 2). The coding regions cover 86.74% of the fosmid (33,309 bp) and encode 43 proteins (Table 2). Of these, 22 protein-coding genes (51.16%) were assigned to COGs (Fig. 3A). The most abundant COG categories were “nucleotide transport and metabolism” (six proteins), followed by “Amino acid transport and metabolism” (three proteins), “Lipid transport and metabolism” (three proteins), and “Replication, recombination, and repair” (two proteins). Signal peptides and transmembrane helices were found in 2 and 10 protein-coding genes, respectively (Table 2). The accession number of pA1 is KU240005.
Table 2.General features of the pA1 fosmid.
Fig. 3.Profiling of pA1 and the rnlA-like gene. (A) Graphical circular map of the pA1 fosmid. From outside to center: Genes on the forward strand (color coded according to COG categories); genes on the reverse strand (color coded according to COG categories); GC content (black); and GC skew (light green/orange). (B) Schematic diagram of shot-gun clone sequences compared with that of fosmid pA1. Alignment of pA1 and four shot-gun clones using NCBI BLAST. Among the 43 CDS identified in pA1, eight (the 33–40th in the region 29,300–34,700 bp) are present in the four shot-gun clones. The numbers indicate individual similar proteins as follows. 33: Ham1 family protein (3.50E-68, 68.15%); 34: n-methyl-D-aspartate receptor NMDAR2C subunit (3.29E-60, 58.35%); 38: glycine-rich protein family protein (2.96E-26, 57.90%); 40: uridine kinase (7.75E-38, 51.05%); 35, 37, and 39: hypothetical proteins.
Sequencing of Shot-Gun Clones
To determine genes involved in gentamicin resistance, the isolated resistant shot-gun clones were sequenced. The genetic organization, sizes, and similarities of the clone sequences are summarized in Table 3. The inserts of resistant clones were 2–3 kb in size. The nucleotide sequences and encoded proteins of the clones did not show similarity to any known gentamicin resistance genes (acc, ant, and aph) when the sequences were compared using NCBI BLAST. Surprisingly, all the clones showed greatest similarity to T4 RnlA, with 33% amino acid sequence identity (Table 3). rnlA encodes a type II toxin and is, therefore, likely to be expressed in response to stress. These results suggest that the resistance conferred by these clones is due to an rnlA-like gene, although the mechanism of action is unknown.
Table 3.Predicted ORFs from the gentamicin-resistant shot-gun clones.
The putative gene responsible for the resistance phenotype was identified within the full-length pA1 fosmid insert by alignment of shot gun clone sequences using Prodigal. We found that among the 43 CDSs in pA1, CDS 36 was the only sequence common to all four shot-gun clones. Strikingly, CDS no. 36 encoded an rnlA-like gene (Fig. 3B). We attempted to clone the specific region but failed to construct the specific area conferring the rnlA-like gene from CDS 36. The possible explanation can be overexpression of the toxin gene resulting in loss of robustness in a single cell and due to high G+C contents.
Genus Identification of the pA1 Fosmid Clones
To determine the source organism of the pA1 fosmid clone containing the rnlA-like gene, we applied a bioinformatics search engine, PhylopythiaS. The full-length sequence was similar to that derived from a single genus, Slackia (data not shown).
Discussion
In this study, we used functional metagenomics to identify gentamicin resistance genes in soil. A metagenome library containing soil DNA in surrogate host E. coli cells was screened, and all four selected gentamicin resistance clones encoded a protein with a deduced amino acid sequence with 33% identity to the RNA ligase, T4 RnlA.
RnlA is a toxin of the type II toxin-anti-toxin (TA) system; both toxin and anti-toxin are proteins and the anti-toxin neutralizes the toxin by direct interaction [15]. For many years, toxins belonging to the TA system were thought to cause persistent cell formation, leading to antibiotic resistance. HipA was the first toxin found to be involved in persistence in E. coli [29]; later, it was shown to be neutralized by the HipB anti-toxin [41]. Although rnlA-mediated cell persistence has not been extensively studied, rnlA overexpression in E. coli results in persistent cell formation and assists survival under antibiotic selection pressure [33]. These data led us to hypothesize that the mechanism underlying the gentamicin resistance of shot-gun clones pA1b, pA1, pA1h, and pA1j is similar to that conferred by rnlA. However, further experiments demonstrated that the rnlA-like gene, which confers resistance to gentamicin in E. coli, did not restore the RnlA activity of an E. coli rnlA mutant (data not shown). Alternatively, an unknown protein in E. coli may act as an “adapter anti-toxin” that undergoes protein-protein interaction with the metagenome library-derived rnlA-like gene because no rnlB-like gene exists in the pA1 fosmid clone near the rnlA-like gene [50].
PhylopythiaS can predict the organism of origin from short metagenomic DNA fragments; no previously available method can do this [28]. Our results indicate that the fosmid clone pA1 is likely derived from an organism belonging to the genus Slackia. Genome data (GenBank Accession Numbers NZ_ACUX00000000, NZ_ADMD00000000, and NC_013165) for Slackia exigua, S. piriformis, and S. heliotrinireducens are currently available [36]. To validate the existence of the rnlA-like gene, we attempted to search for it in the Slackia genomes, but failed to find any, indicting that no homologous gene exists or different Slackia species contained the homologous gene. Slackia is a genus of family Coriobacteriaceae within the Actinobacteria phylum, and its members are gram-positive, non-motile, obligate anaerobes [48]. Slackia may accumulate in soil contaminated by sewage, since Slackia species are generally isolated from human feces, wound infections, and abscesses [19,31]. Slackia species are resistant only to methoxazole-trimethoprim at >32 μg/ml; [19] however, the clone containing the rnlA-like gene is susceptible to 20 μg/ml trimethoprim. Studies of gentamicin resistance have yet to be performed in Slackia. Therefore, our data indicating that Slackia may be resistant to gentamicin is important, since Slackia species are known to be pathogenic to humans [19].
Our results presented herein identify an rnlA-like gene-mediated gentamicin resistance mechanism, using functional metagenome screening, which may be a novel mechanism of antibiotic resistance mediated by a protein with a function previously unassociated with this trait. These data contribute to our knowledge about the presence and diversity of antibiotic resistance genes in environmental bacteria. Future work will involve characterization of the mechanism by which the rnlA-like gene confers resistance to gentamicin and examination of its role in resistance to other antibiotics.
References
- Allen HK, Donato J, Wang HH, Cloud-Hansen KA, Davies J, Handelsman J. 2010. Call of the wild: antibiotic resistance genes in natural environments. Microbiol. Rev. 8: 251-259.
- Allen HK, Moe LA, Rodbumrer J, Gaarder A, Handelsman J. 2009. Functional metagenomics reveals diverse beta-lactamases in a remote Alaskan soil. ISME J. 3: 243-251. https://doi.org/10.1038/ismej.2008.86
- Amann RI, Ludwig W, Schleifer KH. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59: 143-169.
- Bateman A, Coin L, Durbin R, Finn RD, Hollich V, Griffiths-Jones S, et al. 2004. The Pfam protein families database. Nucleic Acids Res. 32: D138-D141. https://doi.org/10.1093/nar/gkh121
- Bland C, Ramsey TL, Sabree F, Lowe M, Brown K, Kyrpides NC, Hugenholtz P. 2007. CRISPR recognition tool (CRT): a tool for automatic detection of clustered regularly interspaced palindromic repeats. BMC Bioinformatics 8: 209. https://doi.org/10.1186/1471-2105-8-209
- Chistoserdovai L. 2010. Functional metagenomics: recent advances and future challenges. Biotechnol. Genet. Eng. Rev. 26: 335-352. https://doi.org/10.5661/bger-26-335
- Conesa A, Gotz S, Garcia-Gomez JM, Terol J, Talon M, Robles M. 2005. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21: 3674-3676. https://doi.org/10.1093/bioinformatics/bti610
- D’Costa VM, Griffiths E, Wright GD. 2007. Expanding the soil antibiotic resistome: exploring environmental diversity. Curr. Opin. Microbiol. 10: 481-489. https://doi.org/10.1016/j.mib.2007.08.009
- Davies J, Wright GD. 1997. Bacterial resistance to aminoglycoside antibiotics. Trends Microbiol. 5: 234-240. https://doi.org/10.1016/S0966-842X(97)01033-0
- Demanche S, Sanguin H, Poté J, Navarro E, Bernillon D, Mavingui P, et al. 2008. Antibiotic-resistant soil bacteria in transgenic plant fields. Proc. Natl. Acad. Sci. USA 105: 3957-3962. https://doi.org/10.1073/pnas.0800072105
- Donato JJ, Moe LA, Converse BJ, Smart KD, Berklein FC, McManus PS, Handelsman J. 2010. Metagenomic analysis of apple orchard soil reveals antibiotic resistance genes encoding predicted bifunctional proteins. Appl. Environ. Microbiol. 76: 4396-4401. https://doi.org/10.1128/AEM.01763-09
- Finney DJ. 1947. Probit Analysis; A Statistical Treatment of the Sigmoid Response Curve. Cambridge University Press.
- Forsberg KJ, Reyes A, Wang B, Selleck EM, Sommer MO, Dantas G. 2012. The shared antibiotic resistome of soil bacteria and human pathogens. Science 337: 1107-1111. https://doi.org/10.1126/science.1220761
- Gad GF, Mohamed HA, Ashour HM. 2011. Aminoglycoside resistance rates, phenotypes, and mechanisms of gram-negative bacteria from infected patients in upper Egypt. PLoS One 6: e17224. https://doi.org/10.1371/journal.pone.0017224
- Gerdes K, Christensen SK, Lobner-Olesen A. 2005. Prokaryotic toxin-antitoxin stress response loci. Microbiol. Rev. 3: 371-382.
- Griffiths-Jones S, Moxon S, Marshall M, Khanna A, Eddy SR, Bateman A. 2005. Rfam: annotating non-coding RNAs in complete genomes. Nucleic Acids Res. 33: D121-D124. https://doi.org/10.1093/nar/gki081
- Henriques I, Moura A, Alves A, Saavedra MJ, Correia A. 2006. Analysing diversity among beta-lactamase encoding genes in aquatic environments. FEMS Microbiol. Ecol. 56: 418-429. https://doi.org/10.1111/j.1574-6941.2006.00073.x
- Hyatt D, Chen GL, Locascio PF, Land ML, Larimer FW, Hauser LJ. 2010. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics 11: 119. https://doi.org/10.1186/1471-2105-11-119
- Kim KS, Rowlinson MC, Bennion R, Liu C, Talan D, Summanen P, Finegold SM. 2010. Characterization of Slackia exigua isolated from human wound infections, including abscesses of intestinal origin. J. Clin. Microbiol. 48: 1070-1075. https://doi.org/10.1128/JCM.01576-09
- Krogh A, Larsson B, von Heijne G, Sonnhammer EL. 2001. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 305: 567-580. https://doi.org/10.1006/jmbi.2000.4315
- Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. 2009. Circos: an information aesthetic for comparative genomics. Genome Res. 19: 1639-1645. https://doi.org/10.1101/gr.092759.109
- Lagesen K, Hallin P, Rodland EA, Staerfeldt HH, Rognes T, Ussery DW. 2007. RNAmmer: consistent and rapid annotation of ribosomal RNA genes. Nucleic Acids Res. 35: 3100-3108. https://doi.org/10.1093/nar/gkm160
- Lang KS, Anderson JM, Schwarz S, Williamson L, Handelsman J, Singer RS. 2010. Novel florfenicol and chloramphenicol resistance gene discovered in Alaskan soil by using functional metagenomics. Appl. Environ. Microbiol. 76: 5321-5326. https://doi.org/10.1128/AEM.00323-10
- Lee MH, Lee CH, Oh TK, Song JK, Yoon JH. 2006. Isolation and characterization of a novel lipase from a metagenomic library of tidal flat sediments: evidence for a new family of bacterial lipases. Appl. Environ. Microbiol. 72: 7406-7409. https://doi.org/10.1128/AEM.01157-06
- Lee MH, Oh KH, Kang CH, Kim JH, Oh TK, Ryu CM, Yoon JH. 2012. Novel metagenome-derived, cold-adapted alkaline phospholipase with superior lipase activity as an intermediate between phospholipase and lipase. Appl. Environ. Microbiol. 78: 4959-4966. https://doi.org/10.1128/AEM.00260-12
- Lowe TM, Eddy SR. 1997. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25: 955-964. https://doi.org/10.1093/nar/25.5.0955
- Martin J, Barras M, Yui N, Kirkpatrick C, Kubler P, Norris R. 2012. Gentamicin monitoring practices in teaching hospitals - time to undertake the necessary randomised controlled trial. J. Clin. Toxicol. 2: 146.1-146.5.
- McHardy AC, Martin HG, Tsirigos A, Hugenholtz P, Rigoutsos I. 2007. Accurate phylogenetic classification of variable-length DNA fragments. Nat. Methods 4: 63-72. https://doi.org/10.1038/nmeth976
- Moyed HS, Bertrand KP. 1983. hipA, a newly recognized gene of Escherichia coli K-12 that affects frequency of persistence after inhibition of murein synthesis. J. Bacteriol. 155: 768-775.
- Mullany P. 2014. Functional metagenomics for the investigation of antibiotic resistance. Virulence 5: 443-447. https://doi.org/10.4161/viru.28196
- Nagai F, Watanabe Y, Morotomi M. 2010. Slackia piriformis sp. nov. and Collinsella tanakaei sp. nov., new members of the family Coriobacteriaceae, isolated from human faeces. Int. J. Syst. Evol. Microbiol. 60: 2639-2646. https://doi.org/10.1099/ijs.0.017533-0
- Otsuka Y, Ueno H, Yonesaki T. 2003. Escherichia coli endoribonucleases involved in cleavage of bacteriophage T4 mRNAs. J. Bacteriol. 185: 983-990. https://doi.org/10.1128/JB.185.3.983-990.2003
- Otsuka Y, Yonesaki T. 2012. Dmd of bacteriophage T4 functions as an antitoxin against Escherichia coli LsoA and RnlA toxins. Mol. Microbiol. 83: 669-681. https://doi.org/10.1111/j.1365-2958.2012.07975.x
- Petersen TN, Brunak S, von Heijne G, Nielsen H. 2011. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat. Methods 8: 785-786. https://doi.org/10.1038/nmeth.1701
- Popowska M, Rzeczycka M, Miernik A, Krawczyk-Balska A, Walsh F, Duffy B. 2012. Influence of soil use on prevalence of tetracycline, streptomycin, and erythromycin resistance and associated resistance genes. Antimicrob. Agents Chemother. 56: 1434-1443. https://doi.org/10.1128/AAC.05766-11
- Pukall R, Lapidus A, Nolan M, Copeland A, Glavina Del Rio T, Lucas S, et al. 2009. Complete genome sequence of Slackia heliotrinireducens type strain (RHS 1). Stand. Genomic Sci. 1: 234-241. https://doi.org/10.4056/sigs.37633
- Rath S, Sahu MC, Dubey D, Debata NK, Padhy RN. 2011. Which value should be used as the lethal concentration 50 (LC(50)) with bacteria? Interdiscipl. Sci. 3: 138-143. https://doi.org/10.1007/s12539-011-0081-x
- Riesenfeld CS, Goodman RM, Handelsman J. 2004. Uncultured soil bacteria are a reservoir of new antibiotic resistance genes. Environ. Microbiol. 6: 981-989. https://doi.org/10.1111/j.1462-2920.2004.00664.x
- Sambrook J, Russell DW. 2006. Fragmentation of DNA by sonication. CSH Protoc. 2006. DOI: 10.1101/pdb.prot4538.
- Schmieder R, Edwards R. 2011. Insights into antibiotic resistance through metagenomic approaches. Future Microbiol. 7: 73-89. https://doi.org/10.2217/fmb.11.135
- Schumacher MA, Piro KM, Xu W, Hansen S, Lewis K, Brennan RG. 2009. Molecular mechanisms of HipA-mediated multidrug tolerance and its neutralization by HipB. Science 323: 396-401. https://doi.org/10.1126/science.1163806
- Shakil S, Khan R, Zarrilli R, Khan AU. 2008. Aminoglycosides versus bacteria - a description of the action, resistance mechanism, and nosocomial battleground. J. Biomed. Sci. 15: 5-14. https://doi.org/10.1007/s11373-007-9194-y
- Shaw KJ, Rather PN, Hare RS, Miller GH. 1993. Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglycoside-modifying enzymes. Microbiol. Rev. 57: 138-163.
- Tao W, Lee MH, Wu J, Kim NH, Kim JC, Chung E, et al. 2012. Inactivation of chloramphenicol and florfenicol by a novel chloramphenicol hydrolase. Appl. Environ. Microbiol. 78: 6295-6301. https://doi.org/10.1128/AEM.01154-12
- Tatusov RL, Fedorova ND, Jackson JD, Jacobs AR, Kiryutin B, Koonin EV, et al. 2003. The COG database: an updated version includes eukaryotes. BMC Bioinformatics 4: 41. https://doi.org/10.1186/1471-2105-4-41
- Tringe SG, von Mering C, Kobayashi A, Salamov AA, Chen K, Chang HW, et al. 2005. Comparative metagenomics of microbial communities. Science 308: 554-557. https://doi.org/10.1126/science.1107851
- Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, Eisen JA, et al. 2004. Environmental genome shotgun sequencing of the Sargasso Sea. Science 304: 66-74. https://doi.org/10.1126/science.1093857
- Wade WG, Downes J, Dymock D, Hiom SJ, Weightman AJ, Dewhirst FE, et al. 1999. The family Coriobacteriaceae: reclassification of Eubacterium exiguum (Poco et al. 1996) and Peptostreptococcus heliotrinireducens (Lanigan 1976) as Slackia exigua gen. nov., comb. nov. and Slackia heliotrinireducens gen. nov., comb. nov., and Eubacterium lentum (Prevot 1938) as Eggerthella lenta gen. nov., comb. nov. Int. J. Syst. Bacteriol. 49: 595-600. https://doi.org/10.1099/00207713-49-2-595
- Wiegand I, Hilpert K, Hancock RE. 2008. Agar and broth dilution methods to determine the minimal inhibitory concentration (MIC) of antimicrobial substances. Nat. Protoc. 3: 163-175. https://doi.org/10.1038/nprot.2007.521
- Williams JJ, Hergenrother PJ. 2012. Artificial activation of toxin-antitoxin systems as an antibacterial strategy. Trends Microbiol. 20: 291-298. https://doi.org/10.1016/j.tim.2012.02.005
- Wu CH, Apweiler R, Bairoch A, Natale DA, Barker WC, Boeckmann B, et al. 2006. The Universal Protein Resource (UniProt): an expanding universe of protein information. Nucleic Acids Res. 34: D187-D191. https://doi.org/10.1093/nar/gkj161