Mobile genetic elements and antibiotic resistance in mine soil amended with organic wastes

Science of the Total Environment 621 (2018) 725–733

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Mobile genetic elements and antibiotic resistance in mine soil amended with organic wastes Carlos Garbisu a,1, Olatz Garaiyurrebaso b,1, Anders Lanzén a, Itxaso Álvarez-Rodríguez b, Lide Arana b, Fernando Blanco a, Kornelia Smalla c, Elisabeth Grohmann d, Itziar Alkorta b,⁎ a

NEIKER-Tecnalia, Department of Conservation of Natural Resources, Soil Microbial Ecology Group, Berreaga 1, 48160 Derio, Spain Instituto BIOFISIKA (CSIC, UPV/EHU), Department of Biochemistry and Molecular Biology, University of the Basque Country, P.O. Box 644, 48080 Bilbao, Spain c Julius Kühn-Institut, Federal Research Centre for Cultivated Plants, Institute for Epidemiology and Pathogen Diagnostics, Messeweg 11-12, 38104 Braunschweig, Germany d Beuth University of Applied Sciences, Life Sciences and Technology, Department of Microbiology, Seestraße 64, 13347 Berlin, Germany b

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Mine soil can be a reservoir of mobile genetic elements. • Almost all putative E. coli transconjugants displayed a multiresistant phenotype. • One putative E. coli transconjugant harboured imipenem resistance. • Putative E. coli transconjugants displayed an altered fitness. • Genome sequencing revealed no additional genes in the putative transconjugants.

a r t i c l e

i n f o

Article history: Received 21 September 2017 Received in revised form 19 November 2017 Accepted 19 November 2017 Available online xxxx Editor: Jay Gan Keywords: Antibiotic resistance Competitive fitness Conjugative plasmids Imipenem Integrons

a b s t r a c t Metal resistance has been associated with antibiotic resistance due to co- or cross-resistance mechanisms. Here, metal contaminated mine soil treated with organic wastes was screened for the presence of mobile genetic elements (MGEs). The occurrence of conjugative IncP-1 and mobilizable IncQ plasmids, as well as of class 1 integrons, was confirmed by PCR and Southern blot hybridization, suggesting that bacteria from these soils have gene-mobilizing capacity with implications for the dissemination of resistance factors. Moreover, exogenous isolation of MGEs from the soil bacterial community was attempted under antibiotic selection pressure by using Escherichia coli as recipient. Seventeen putative transconjugants were identified based on increased antibiotic resistance. Metabolic traits and metal resistance of putative transconjugants were investigated, and whole genome sequencing was carried out for two of them. Most putative transconjugants displayed a multi-resistant phenotype for a broad spectrum of antibiotics. They also displayed changes regarding the ability to metabolise different carbon sources, RNA: DNA ratio, growth rate and biofilm formation. Genome sequencing of putative transconjugants failed to detect genes acquired by horizontal gene transfer, but instead revealed a number of nonsense mutations, including in ubiH, whose inactivation was linked to the observed resistance to aminoglycosides. Our results confirm that mine soils contain MGEs encoding antibiotic resistance. Moreover, they point out the role of spontaneous mutations in achieving low-level antibiotic resistance in a short time, which was associated with a trade-off in the capability to metabolise specific carbon sources. © 2017 Published by Elsevier B.V.

⁎ Corresponding author at: BIOFISIKA Institute (UPV/EHU, CSIC), Department of Biochemistry and Molecular Biology, University of the Basque Country, P.O. Box 644, 48080 Bilbao, Spain. E-mail address: [email protected] (I. Alkorta). 1 Both authors contributed equally to this study.

https://doi.org/10.1016/j.scitotenv.2017.11.221 0048-9697/© 2017 Published by Elsevier B.V.

726

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

1. Introduction In the soil ecosystem, the proximity of bacterial cells within microniches or biofilms favours the exchange of genetic determinants via horizontal gene transfer (HGT) (Christensen et al., 1998). Due to the administration of antibiotics to livestock, animal manure is often considered as a reservoir of bacteria carrying transferable antibiotic resistance genes (ARGs). The use of manure as organic fertilizer may increase HGT and transfer of ARG in environmental habitats (Götz and Smalla, 1997). Furthermore, high metal concentrations in soil can negatively affect soil microbial communities (Epelde et al., 2010); they can select for genes encoding metal resistance, which in turn can be associated with antibiotic resistance due to co-resistance (different resistance determinants or ARGs on the same genetic element) or cross-resistance (same genetic determinant for resistance to both antibiotics and metals) mechanisms (Seiler and Berendonk, 2012). Although acquisition of MGEs can provide bacterial hosts with an array of beneficial traits such as antibiotic or metal resistance (Heuer and Smalla, 2012), it can also result in metabolic costs and changes in ecological fitness of recipient bacteria, owing to the replication, transcription and translation of acquired genes (Martínez et al., 2009). As consequence of these fitness costs, in the absence of selection pressure for their maintenance, MGEs can be lost from the bacterial population or the proportion of cells carrying MGEs can decrease (Jechalke et al., 2013). Alternatively, recipient bacteria can adapt through mutations on the chromosome or on the MGE itself, or by integration of the beneficial determinants into the chromosome (Dahlberg and Chao, 2003; Dionisio et al., 2005). Thus, fitness costs play a crucial role in the evolutionary dynamics of resistance, as they generate selection against such resistance (Zur Wiesch et al., 2011). The aim of this work was to search for MGEs and ARGs in metal-contaminated mine soil treated 5 years before with manure-based fertilizers. To our knowledge, only very few studies have investigated the long-term effects of different organic amendments on the presence of MGEs and ARGs in metal contaminated mine soil. Besides, we studied the ability of an E. coli lab strain to acquire MGEs and ARGs from the soil bacterial community under antibiotic selection pressure. We hypothesized that, even five years after treatment, organically amended mine soil would (i) show a higher abundance of MGEs, (ii) trigger HGT of ARGs and metal resistance more efficiently than untreated soil and (iii) that acquired antibiotic resistance by recipient strains is an indication for successful conjugative transfer.

Table 1 Soil physicochemical properties based on dry weight (mean ± SE).

Classification pH (1:2.5 w/v, water) Total organic matter (g kg−1) Nitrogen total (g kg−1) Phosphorus (Olsen) (g kg−1) Calcium (g kg−1) Magnesium (g kg−1) Potassium (g kg−1) Cation exchange capacity (cmol kg−1) Cadmium (Cd) (mg kg−1) Lead (Pb) (mg kg−1) Zinc (Zn) (mg kg−1)

SITE 1

SITE 2

Sandy-loam 6.49 ± 0.11 199.8 ± 28.3 7.13 ± 0.87 0.007 ± 0.001 1.34 ± 0.09 0.29 ± 0.03 0.16 ± 0.03 19.98 ± 2.83 6.00 ± 0.31 16,285 ± 188 15,529 ± 179

Loam 6.71 ± 0.03 151.2 ± 9.0 4.27 ± 0.54 0.002 ± 0.000 0.76 ± 0.08 0.18 ± 0.01 0.08 ± 0.00 15.12 ± 0.90 13.60 ± 0.16 28,587 ± 330 61,007 ± 352

2.2. MGEs in soil samples Total community DNA was extracted from 0.5 g of soil using the Fast DNA™ SPIN kit (MP Biomedicals, USA) and purified by Geneclean Spin Kit (MP Biomedicals, USA). Polymerase chain reaction (PCR) and Southern blot DNA hybridisation were used to study the presence of (1) class 1 integron integrase (intl1) and quaternary ammonium compound resistance (qacEΔ1) genes, (2) plasmid backbone regions related to replication (trfA of IncP-1 subgroups, IncN rep and V216 rep) and (3) origins of replication (IncQ oriV). Amplification of intl1 and qacEΔ1 genes was performed as described by Sandvang et al. (1997); of trfA gene from IncP-1 subgroups, as described by Bahl et al. (2009); of IncQ oriV and IncN rep, as described by Götz et al. (1996); and of pV216 rep, as described by Heuer et al. (2009). Oligonucleotide sequences, reference plasmids (positive controls), annealing temperatures and amplicon sizes are listed in Supplementary Table S1. Southern blot DNA hybridisation was performed for all target genes, regardless of the detection of PCR products. Blotting and subsequent chemiluminescence detection of digoxigenin (DIG)-labelled DNA hybrids was performed according to the DIG System User's Guide for Filter Hybridisation (Roche, Germany). DIG-labelled probes were generated from PCR-amplified fragments obtained with the reference plasmids (Supplementary Table S1). PCR products (75 μl each) were separated by agarose gel electrophoresis in 1× TBE buffer at 50 V for 4–5 h. DNA was then denatured and transferred to a Hybond-N+ membrane. Detection was carried out with CDP star® ready to use solution (Roche, Germany). The membrane was exposed to high performance chemiluminescence film (GE Healthcare Life Science, UK) and then developed.

2. Materials and methods 2.3. Capturing transferable antibiotic resistance genes from soil bacteria 2.1. Experimental design The study area (43°13′ N, 3°26′ W) is located in an abandoned mine (Spain) (Barrutia et al., 2011). Within the mining area, two sites showing different levels of metal contamination (Cd, Pb, Zn) were selected (site 1 being less and site 2 being more contaminated; Table 1). From both sites, samples were collected from three randomly spatially distributed 1 m2 plots, where the following amendments had been applied five years before, as part of an aided phytostabilisation field trial (for more information, see Galende et al., 2014 and Garaiyurrebaso et al., 2017): cow slurry (COW), sheep manure (SHEEP), poultry manure (POULTRY), and paper mill sludge mixed with poultry manure (2:1, v/ v) (PAPER). For both sites, controls, where no amendment was added, were included. Composite samples (10 sub-samples; 0–10 cm depth) were randomly collected from treated and control soils. Samples were kept at 4 °C during transport to the laboratory. Prior to DNA extraction and determination of physicochemical properties, the samples were sieved with a sterile sieve (2 mm) and stored at 4 °C for less than one week.

The exogenous plasmid isolation technique was used to capture transferable resistances into rifampicin resistant E. coli 1030. Initially, bacteria were detached from the soil matrix by mixing 10 mg of soil and 1.5 ml of 7.5 mM sodium pyrophosphate and Tween 80 (0.5%, w/ v). Samples were shaken for 45 min at room temperature. The mixture was then allowed to settle for 5 min and the supernatant with detached bacteria served as donor. Recipient cells (E. coli 1030, RIFR) were grown overnight at 37 °C in 2 ml TSB medium supplemented with rifampicin (100 μg ml−1). One mililiter of the recipient culture, as well as 1 ml of donor solution were centrifuged at 4850 ×g at room temperature. Both recipient and donor supernatants were then discarded, pellets washed twice in 1 ml sterile 1/10 TSB solution and resuspended in 1 ml sterile 1/10 TSB. Subsequently, 500 μl each of recipient and donor were mixed and centrifuged at 4850 ×g at room temperature. Pellets were resuspended in 50 μl sterile 1/10 TSB and applied to a nitrocellulose filter (0.22 μm pore size; Millipore, USA). Filters were incubated overnight at 28 °C on TSA plates supplemented with cycloheximide (300 μg ml−1). The bacterial lawn was resuspended in 1 ml sterile NaCl solution (0.85%, w/v) and stirred at 450 rpm for 20 min, to release bacterial

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

cells from the filter. Serial 10-fold dilutions of detached bacteria were spread on TSA plates supplemented with cycloheximide (300 μg ml−1), rifampicin (100 μg ml−1) and one additional antibiotic, according to the following concentrations (in μg ml−1): 25 for ampicillin (AMP), 30 for kanamycin (KAN), 10 for tetracycline (TET), 10 for chloramphenicol (CHL), 45 for erythromycin (ERY), 25 for gentamicin (GEN) and 10 for streptomycin (STR). After 48 h of incubation at 28 °C, single colonies were selected for further characterization. The following code was used to refer to antibiotic-resistant strains: site number (1 or 2), followed by the amendment applied to the soil (COW, SHEEP, POULTRY, PAPER or CONTROL), and finally the antibiotic used for selection (AMP, KAN, TET, CHL, ERY, GEN or STR). BOX-PCR (Martin et al., 1992) was used to compare genome fingerprints of antibiotic resistant strains and the recipient strain, in order to discard false positives. For this purpose, cell lysis was carried out using the Genomic DNA Extraction kit (Qiagen, Germany) followed by DNA preparation using the Fermentas DNA Extraction Kit (Thermo Scientific, USA). PCR reactions were carried out as described by Dealtry et al. (2014). PCR products were visualized on 1 × TBE agarose gel (1.5%, w/v). 2.4. Minimum inhibitory concentrations (MICs) For each putative transconjugant, minimum inhibitory concentrations (MICs) for the following antibiotics were determined by the plate dilution method (NCCLS, 2004): ampicillin, chloramphenicol, erythromycin, gentamicin, imipenem, kanamycin, streptomycin, sulfadiazine and tetracycline (concentrations ranged from 1024 μg ml−1 to 0.0625 μg ml−1). Mueller-Hinton agar plates (Sigma Aldrich, USA) were spot inoculated with ca. 1 × 107 CFU ml−1 of putative transconjugant. Plates were incubated for 24 h at 37 °C. E. coli ATCC 25299 was used as internal reference. Similarly, for each putative transconjugant, MIC values for the following heavy metals were determined: Cd, Cu, Zn and Pb. MIC value was defined as the lowest metal concentration that inhibited cell growth, as measured by optical density at 600 nm. Ten microliter of a culture with OD600 between 0.08 and 0.10 were inoculated into 150 μl LB medium containing Cd, Cu, Zn or Pb [CdCl2, CuCl2, ZnCl2 or Pb(NO3)2 at concentrations ranging from 0.25 to 25 mM]. Cultures were incubated overnight at 37 °C under shaking. Cell growth was visualized with a Synergy HT Multi-Mode Microplate Reader (Biotek, USA). 2.5. Growth rate and RNA: DNA ratio To assess potential fitness costs, the growth of putative transconjugants was compared with the growth of the E. coli recipient strain. All cultures were adjusted to an initial OD600 of 0.10. Cells were incubated at 37 °C, OD600 was measured every hour for the first 8 h and finally after 24 h. Cell concentrations were calculated from OD600 values using the following equation: an OD600 of 1 = 8 × 108 cells ml−1 (Myers et al., 2013). Maximum growth rate (μmax) was estimated from the slope of the inflection point of each growth curve (Perni et al., 2005). The RNA: DNA ratio was determined by adjusting the culture to an initial OD600 of 0.10 in 25 ml LB medium supplemented with rifampicin (100 μg ml−1). Cultures were incubated at 37 °C until an OD600 of 2.0, and divided into two parts for DNA and RNA extraction. DNA extraction was performed with the DNeasy Blood & Tissue Kit (Qiagen, Germany), and RNA extraction was carried out as described above. Nucleic acid concentrations were determined using an ND3000 fluorospectrophotometer (Nanodrop Technologies Inc.). 2.6. Phenotype fingerprinting with GEN III MicroPlates™ GEN III MicroPlates™ (Biolog, USA) were used to compare putative transconjugants and the E. coli recipient strain at a phenotypic level. For each putative transconjugant, the following criterion was used to establish whether a specific carbon substrate had been used (positive

727

result) or not (negative result): positive result = absorbance of the putative transconjugant ≥ 50% of the absorbance of the negative control included in the GEN III MicroPlate™; negative result = absorbance of the putative transconjugant b 50% of the absorbance of the negative control included in the GEN III MicroPlate™. Finally, the Jaccard similarity index was calculated. 2.7. Biofilm formation Putative transconjugants and the E. coli recipient strain were tested for biofilm formation by a modified in vitro adherence assay (Christensen et al., 1985). Briefly, 200 μl TSB medium in 96-well flat-bottom polystyrene plates (Sigma Aldrich, USA) was inoculated with 10 μl of overnight culture and grown without shaking at 37 °C for 24 h. Planktonic bacteria were removed by washing 3 times with sterile distilled water. To stain biofilms, 125 μl of crystal violet solution (0.1%, w/v) was added to each well and incubated for 10 min at room temperature. Then, three washing steps with sterile distilled water were performed. To solubilize the dye, 200 μl of glacial acetic acid (33%, v/v) was added to each well and incubated for 10 min at room temperature. For each strain, OD570 was measured in triplicate in a microplate reader. 2.8. Genome sequencing of putative transconjugants For genome sequencing, apart from the E. coli 1030 recipient strain, we selected (i) the putative transconjugant 1PAPER-STR, due to its resistance to imipenem and its decreased growth rate, compared to the recipient strain, and (ii) 1PAPER-CHL, due to the highest number of acquired resistances, but lacking imipenem resistance (Table 2). E. coli 1030 recipient strain, 1PAPER-CHL and 1PAPER-STR were grown until OD600 = 0.5 prior to DNA extraction using the E.Z.N.A. Bacterial DNA kit (Omega BioTek, USA). Libraries were prepared from extracted DNA for barcoded paired-end shotgun sequencing using Illumina MiSeq v2 (300 cycles/2 × 150 bp read length, TruSeq library preparation kit) and sequenced at the Advanced Research Facilities (SGIker) of the University of the Basque Country. Sequence data (in FASTQ format) were submitted to the European Nucleotide Archive (study accession PRJEB18259, run accessions ERR1744102-4). Sequence data (in FASTQ format) were demultiplexed and, for each targeted strain, subjected to de novo genome assembly using SPAdes v. 3.7.1 (paired-end mode with default parameters) (Bankevich et al., 2012). Only contigs longer than 500 bp and with a similar sequence above 50 × were considered for further analysis. Quality assessment and identification of probable misassemblies were carried out using QUAST (Gurevich et al., 2013) and the genome sequence of E. coli K-12 substr. MG1655 (Genbank Accession NC000913.3). In order to identify genome insertions or deletions, and plasmids not present in the recipient strain, all resulting contigs from the two putative transconjugants were aligned to those from the recipient strain using MUMmer (v3.23; Kurtz et al., 2004) and visualized using mummerplot. To identify single base differences between the assembled genomes, we utilised NUCmer from the same software package. Alignments and single base differences were verified by carrying out the same alignments to the E. coli K-12 substr. MG1655 sequence. The annotation of this genome sequence was also used to identify any protein-coding gene sequences (CDSs) of identified point mutations, i.e. verified single base differences between putative transconjugant and recipient strains. To identify the effect of mutations (synonymous vs. non-synonymous), we utilised the BLASTX web interface of NCBI (https://blast.ncbi.nlm. nih.gov) to align cropped sequences corresponding to individual CDSs to the Genbank non-redundant protein sequence collection (nr). To verify an identified point mutation in the ubiH gene, the 5′ region of the gene was amplified using primers designed with Primer-BLAST (Ye et al., 2012; Forward: TAATCATCGTCGGTGGCGG, Reverse: AATCCGCCAGAGATTGCCAG) from the E. coli 1030 recipient strain, 1PAPER-CHL and 1PAPER-STR, and sequenced using Sanger sequencing

728

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

Table 2 Minimum inhibitory concentrations (MIC, in mg/l). Transconjugants exhibiting over two-fold higher resistance than the E. coli recipient strain are shown in bold.

E. coli 1030 (R) 1PAPER-AMP 1COW-CHL 1SHEEP-CHL 1PAPER-CHL 2COW-CHL 2SHEEP-CHL 2PAPER-CHL 2CONTROL-CHL 1SHEEP-ERY 1PAPER-ERY 1CONTROL-ERY 2COW-ERY 2SHEEP-ERY 2PAPER-ERY 2POULTRY-ERY 1PAPER-STR 2POULTRY-STR

Ampicillin

Chloramphenicol

Erythromycin

Gentamicin

Imipenem

Kanamycin

Streptomycin

Sulfadiazine

Tetracycline

4–8 32–64 8–16 8–16 8–16 8–16 8–16 8–16 8–16 4–8 4–8 4–8 2–4 2–4 4–8 2–4 4–8 4–8

2–4 2–4 16–32 16–32 16–32 16–32 16–32 16–32 16–32 4–8 4–8 4–8 4–8 2–4 4–8 1–2 1–2 1–2

32–64 32–64 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 128–256 512–1024 32–64 32–64

0.125–0.250 0.125–0.250 0.25–0.50 0.25–0.50 0.25–0.50 0.25–0.50 0.25–0.50 0.25–0.50 0.25–0.50 0.125–0.250 0.125–0.250 1–2 0.125–0.250 0.125–0.250 0.25–0.50 b0.125 1–2 0.5–1.0

0.25–0.50 0.25–0.50 0.25–0.50 0.125–0.250 0.125–0.250 0.125–0.250 0.125–0.250 0.125–0.250 b0.125 0.25–0.50 0.25–0.50 0.125–0.250 0.25–0.50 0.25–0.50 0.25–0.50 0.125–0.250 1–2 0.25–0.50

0.125–0.250 0.125–0.250 0.5–1.0 0.5–1.0 0.5–1.0 0.5–1.0 0.5–1.0 0.5–1.0 0.5–1.0 0.5–1.0 0.125–0.250 2–4 0.125–0.250 0.125–0.250 0.5–1.0 0.125–0.250 4–8 2–4

1–2 1–2 2–4 2–4 2–4 2–4 2–4 2–4 2–4 1–2 1–2 4–8 1–2 1–2 0.5–1.0 0.5–1.0 8–16 N1024

4–8 4–8 2–4 1–2 8–16 2–4 2–4 1–2 1–2 4–8 2–4 2–4 2–4 4–8 1–2 1–2 8–16 4–8

1–2 1–2 4–8 4–8 4–8 4–8 4–8 4–8 4–8 1–2 1–2 0.5–1.0 1–2 1–2 1–2 0.5–1.0 1–2 0.5–1.0

technology. Resulting sequence pairs were overlapped using Phred/ Phrap (Ewing and Green, 1998) and manually corrected using Consed (Gordon et al., 1998).

PAPER-treated samples from Site 1 resulted in the largest number of putative transconjugants (Table 2). 3.3. Characterization of putative antibiotic-resistant strains

2.9. Statistical analysis ANOVA was applied where more than two samples were compared. Post-hoc pairwise comparisons were performed with Tukey-Kramer adjustment (p b 0.05) to refine p-values. Statistical analyses were carried out with Statview software (SAS Institute Inc.). 3. Results 3.1. Presence of MGEs in soil samples Except for SHEEP plots in Site 2, intl1 and qacEΔ1 genes were detected in all soils in at least one of the replicates, including untreated control soils (Table 3). Backbone IncP-1 and IncQ regions were detected in both sites. Plasmids belonging to the incompatibility group IncN were not detected in any sample and pHHV216-like plasmids were detected only in Site 1 (Table 3).

Although originating from different sites and treatments, a number of putative transconjugants (1COW-CHL, 1SHEEP-CHL, 2COW-CHL, 2SHEEP-CHL, 2PAPER-CHL and 2CONTROL-CHL) shared the same antibiotic MIC characteristics: In comparison to the recipient strain, their susceptibility to ampicillin, gentamicin and streptomycin was reduced two-fold, to erythromycin, kanamycin and tetracycline four-fold, and to chloramphenicol eight-fold (Table 2). 1PAPER-CHL showed increased resistance to sulfadiazine compared to the recipient strain, thus having the highest number of acquired resistances (Table 2). Further, 1PAPER-STR was the only putative transconjugant showing low-level resistance to imipenem with a MIC of 1–2 (four-fold compared to the recipient strain). Furthermore, 1PAPER-STR exhibited lower susceptibility to sulfadiazine, gentamicin, streptomycin and kanamycin, compared to the recipient strain (Table 2). No putative transconjugant exhibited higher tolerance to metals compared to the recipient, whose MIC values were determined as (in mM) 1–2 for Cd, 2.5–5 for Cu, 1–2.5 for Zn and 5–7.5 for Pb.

3.2. Isolation of antibiotic-resistant strains 3.4. Growth rate and RNA: DNA ratio Altogether, 36 resistant strains were isolated from the conjugation experiments (one from each sample and under selection of each of the six different antibiotics). Out of these, 17 were confirmed as E. coli 1030 by BOX-PCR fingerprints (Supplementary Fig. S1), hereafter referred to as “putative transconjugants”. Putative transconjugants were obtained from all soils including untreated controls under the selection of at least one antibiotic (Table 2), besides rifampicin. No resistant bacteria were obtained under tetracycline selection from any of the soils.

Putative transconjugants were divided into three groups depending on whether they showed significantly (p b 0.05) higher, similar or lower maximum growth rate (μmax) than the recipient strain (Table 4 and Supplementary Fig. S2). Compared to the recipient strain, 2COW-CHL, 1SHEEP-ERY and 1PAPER-AMP showed higher μmax values. All putative transconjugants isolated under streptomycin, kanamycin or gentamicin selection showed significantly lower μmax values, except for 2POULTRY-

Table 3 Detection of intl1, qacEΔ1, IncP-1 (all subgroups) trfA, IncP-1α trfA, IncP-1β trfA, IncQ oriV, IncN rep, pV216 rep sequences by PCR and Southern Blot. Each + represents a positive result for one of the three replicates. SITE 1

intl1 qacEΔ1 IncP-1 (all subgroups) trfA IncP-1α trfA IncP-1β trfA IncQ oriV IncN rep pV216 rep

SITE 2

Cow

Sheep

Paper

Poultry

Control

Cow

+++ +++ +

++ +++

+++ +++

+ ++ +

+++ +++ ++

+ +++

+

+++ +

+ ++ +

+ + +

+

+

Sheep

Paper

Poultry

Control

++ +++

++ +

+ +

+

+

+ ++

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733 Table 4 Maximum growth rate (μmax) of putative transconjugants compared to the E. coli recipient strain. Putative transconjugants with a lower μmax value than the recipient strain were divided in three groups, depending on whether they reached the stationary phase, stayed in the exponential phase or in the lag phase after 24 h of incubation. Higher μmax

Similar μmax

Lower μmax Stationary

1PAPER-AMP 1COW-CHL 1SHEEP-CHL 1PAPER-CHL 2COW-CHL 2SHEEP-CHL 2PAPER-CHL 2CONTROL-CHL 1SHEEP-ERY 1PAPER-ERY 1CONTROL-ERY 2COW-ERY 2SHEEP-ERY 2PAPER-ERY 2POULTRY-ERY 1PAPER- STR 2COW-STR 2POULTRY-STR

Exponential

729

with OD570 values in the range of 0.12–0.24 (Fig. 2; Christensen et al., 1985; Di Rosa et al., 2006). The remaining strains were classified as non-biofilm formers. Among these, 1SHEEP-CHL, 2CONTROL-CHL and 1PAPER-STR had particularly low OD570 values below 0.07 (Fig. 2).

Lag

3.7. Genome sequencing

x x x x x x x x x x x x x x x x x x

STR (Table 4). Although not statistically significant, all putative transconjugants exhibited higher RNA: DNA values compared to the recipient strain (Fig. 1). 3.5. Phenotypic fingerprinting with GEN III MicroPlates™ All putative transconjugants, except for 1PAPER-CHL, 2COW-CHL and 1SHEEP-ERY, were able to use at least one carbon source in excess of the recipient strain. However, most putative transconjugants showed a loss or a reduction in the capacity to metabolise specific carbon substrates (Table 5). The putative transconjugant with the largest phenotypic change compared to the recipient strain was 1PAPER-STR, with a loss of usage or reduced usage of 15 carbon substrates, as well as the acquisition of the capacity to transform a new carbon source (inosine). 1PAPERAMP was able to use the largest number of carbon substrates, without losing the capacity to use any of the carbon sources used by the recipient. 3.6. Biofilm formation Putative transconjugants 1PAPER-AMP, 1PAPER-ERY, 1CONTROL-ERY and 2PAPER-ERY were classified as weak biofilm formers

Sequencing resulted in over 15 million read pairs with an average length (excluding barcodes and adaptors) of 127 bp, relatively evenly distributed between the three sequenced genomes (recipient strain, 1PAPER-CHL and 1PAPER-STR) (Supplementary Table S2). Assembly resulted in between 96 and 101 contigs longer than 500 bp (excluding possible contaminants with coverage below 50 ×), with a total length of 4.48–4.49 million bp (compared to the 4.64 million bp of the full genome for E. coli K-12 substr. MG1655 strain). All contigs from both putative transconjugants could be fully aligned to the recipient strain and to E. coli K-12 substr. MG1655; in addition, all contigs from the recipient strain, except for one 5 kbp contig, could be fully aligned to E. coli K-12 substr. MG1655. The latter aligned to the genome of Escherichia phage phiX174 added as part of the library preparation. No additional genes resulting from genome-integrated or free plasmids acquired via HGT could be identified in any of the putative transconjugants. Excluding contigs likely to result from assembly errors (Supplementary Table S2), a number of random point mutations were identified relative to the recipient strain. These mutations were confirmed by comparing the sequences with the genome of E. coli K-12 substr. MG1655. They were located in seven protein-coding genes in 1PAPERCHL and six protein-coding genes in 1PAPER-STR, with no overlap between the two putative transconjugants (Supplementary Table S2). All mutations in 1PAPER-CHL were synonymous, i.e. did not result in any amino acid change, except for one mutation in a hypothetical protein with unknown function (possibly not transcribed). In 1PAPER-STR, five out of six mutated genes instead contained non-synonymous mutations, three of which were located in hypothetical proteins. The remaining two were in ubiH encoding 2-octaprenyl-6-methoxyphenol hydroxylase, and mltA encoding the membrane-bound lytic murein transglycosylase A. The ubiH gene had a single 44 T N G missense mutation resulting in L14R (Leu N Arg at residue 14). The mltA gene contained four missense mutations. However, they were all located in the part of the gene coding for the inactive precursor peptide that is removed by post-translational modification (N15 K, C51R, R66W and V96A), upstream of the active protein itself.

Fig. 1. RNA: DNA ratio of putative transconjugants. Mean ± SE. R: E. coli 1030 recipient strain.

730

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

Table 5 Carbon substrate utilization patterns between putative transconjugants and E. coli recipient strain.

E. coli 1030 (R) 1PAPER-AMP 2POULTRY-STR 2PAPER-ERY 2COW-CHL 1SHEEP-ERY 1SHEEP-CHL 2SHEEP-ERY 2POULTRY-ERY 1COW-CHL 1PAPER-ERY 2COW-ERY 2CONTROL-CHL 1CONTROL-ERY 2SHEEP-CHL 1PAPER-CHL 2PAPER-CHL 1PAPER-STR

Jaccard index

No. of substrates used

– 0.92 0.92 0.92 0.91 0.91 0.89 0.87 0.86 0.85 0.83 0.83 0.77 0.77 0.77 0.74 0.71 0.67

32 35 33 33 29 29 32 31 34 30 29 27 26 26 24 21 22 18

4. Discussion Class 1 integrons are major contributors to the evolution and dissemination of antibiotic resistance. They are widely spread via linkage with transposons, heavy metal, disinfectant and antibiotic resistance genes (Gillings, 2017). Class 1 integrons harbour two conserved segments (CS): (i) the 5′-CS contains an intI gene that encodes an integrase (intI1), a recombination site (attI), and a promoter responsible for the expression of the captured gene cassettes; and (ii) the 3′-CS contains a gene conferring resistance to quaternary ammonium compounds (qacEΔ1), a sulfonamide resistance gene (sul1) and an ORF (orf5) of unknown function (Canal et al., 2016). Here, intl1 and qacEΔ1 genes were detected in most soils, including untreated controls. Addition of organic amendments did not lead to a noticeable increase in the abundances of these genes. In contrast, in their study on the prevalence of ARGs in soils continually treated with manure for 30 years, Peng et al. (2017) found that the abundance of the intl1 gene increased in pig manure treated soils and showed a strong positive correlation with the abundance of ARGs. In our study, since amendments were applied five years previous to soil sampling, it is possible that initially increased abundances declined by the time of sampling to levels similar to untreated controls, as observed by Nõlvak et al. (2016). In the mentioned study, intl1

abundance declined to a level similar to that of untreated controls in less than one year after treatment with cattle slurry. Kyselková et al. (2015) reported that some ARGs could persist in cow manure-amended soils for at least several months. Further, shotgun metatranscriptomic profiling of soil communities has revealed significantly higher levels of integrase expression in untreated soil samples from the more contaminated Site 2 compared to Site 1 (Epelde et al., 2014). However, Epelde et al. (2014) only determined expression levels relative to total community RNA. The presence of putative conjugative or mobilisable plasmids indicates that bacteria in the study sites have gene-mobilizing capacity, with implications for potential dissemination of antibiotic and metal resistance genes. It is possible that the IncP-1 plasmids detected in Site 1 and the IncQ family plasmids detected in both sites played a role in the adaptation of soil microbial communities to metals, as observed by De Lipthay et al. (2008) and Heuer et al. (2002). The discrepancies observed between replicates regarding the detection of MGEs (Table 3) may partly be due to local soil heterogeneity. The fact that a number of strains acquired antibiotic resistance via the conjugation experiments suggests that this approach was successful, especially for the putative transconjugants displaying high-level resistance with MIC values orders of magnitude above that of the recipient strain. We could not isolate any conjugative plasmids from the transconjugants (data not shown), possibly due to the fact that conjugative plasmids are typically present in low copy number (b10 copies/cell), which reflects the selective advantage of minimizing the metabolic burden on the host (Norman et al., 2009; Paulsson, 2002). Nonetheless, the BOX-PCR data, together with the fact that all putative transconjugants showed resistance to rifampicin (a rifampicin resistant E. coli 1030 strain was used as recipient), as well as resistance to other antibiotics, suggest that we obtained real E. coli 1030 transconjugants. In any case, since we were not able to isolate the conjugative plasmids, we refer to them herein as putative transconjugants. In spite of the antibiotic resistance and the BOX-PCR-based confirmation of the E. coli strains as derived from E. coli 1030, genome sequencing of 1PAPER-CHL and 1PAPER-STR failed to identify any new genes acquired by HGT. However, one gene in the former and five in the latter exhibited non-synonymous base changes compared to the recipient, suggesting point mutations. Stresses such as exposure to antibiotics and metals can induce responses that increase the rate of mutagenesis, thus fuelling rapid adaptations such as antibiotic resistance (Galhardo et al., 2007). All non-synonymous mutations were found in genes coding for hypothetical proteins with unknown function, except two genes in 1PAPER-STR, namely ubiH (with one missense

Fig. 2. Biofilm formation of putative transconjugants. The line marks an OD570 of 0.120, in order to discriminate between non-biofilm-formers (OD570 b 0.120) and weak biofilm-formers (0.120 b OD570 b 0.240). Mean ± SE. R: E. coli 1030 recipient strain.

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

mutation) and mltA (four missense mutations). The ubiH gene has been implicated in antibiotic resistance (Baisa et al., 2013) and a conformation change or disruption of its encoded protein could therefore be considered as a possible explanation for the antibiotic resistance observed here, at least in part. The product of the ubiH gene is 2-octaprenyl-6-methoxyphenol hydroxylase, an oxygenase involved in ubiquinone biosynthesis (Nakahigashi et al., 1992). Cells lacking this gene fail to produce ubiquinone and can only grow by using fermentable carbon substrates (Baisa et al., 2013; Nakahigashi et al., 1992), which agrees well with the decreased metabolic capacity (Table 5) and lower growth rate (Supplementary Fig. S2) of this strain. Destructive mutations of this gene have been shown to be involved in resistance to cycloserine (Baisa et al., 2013), probably due to decreased antibiotic uptake, which has also been observed for mutants in the ubiF gene, which is involved in the same pathway (Collis and Grigg, 1989; Muir et al., 1981). The latter has also been shown to induce low-level resistance to aminoglycosides, which was also the case for our ubiH-mutant 1PAPER-STR (showing 8-fold increased resistance to gentamicin and streptomycin, and 32-fold increased resistance to kanamycin). Thus, it is likely that the mutation in the ubiH gene contributed to the observed resistance of 1PAPER-STR to aminoglycosides. Transconjugants obtained from PAPER-treated soil appeared to be particularly resistant to antibiotics. As reported in Garaiyurrebaso et al. (2017), from a phytostabilization point of view, PAPER was the most efficient treatment in terms of reduction of metal (Cd, Pb and Zn) bioavailability in the soil. Besides, PAPER treatment displayed the most consistent recovery of soil quality. Interestingly, microbial activity, biomass and especially diversity in PAPER-treated soils were clearly enhanced due to the reduction in metal bioavailability and to the high input of easily biodegradable organic matter present in this amendment. These changes induced by the application of the PAPER amendment might have affected the presence and abundance of ARGs and MGEs, and hence the antibiotic resistance observed in transconjugants from PAPER-treated soil. The mltA gene encodes the membrane-bound lytic murein transglycosylase A, involved in the recycling of muropeptides (Lommatzsch et al., 1997). This protein interacts with the synthesis of penicillin-binding protein 1B (PBP1B) during cell wall synthesis (Vollmer et al., 1999) which is targeted by imipenem (Moyá et al., 2010). Although the observed mutations were located in the inactive precursor peptide of the protein, it is possible that the mutations interfered with the post-translational modification, thus inactivating the protein and inducing resistance to imipenem. Similarly, for 1PAPER-CHL, it is possible that the observed mutation of a hypothetical protein was related to the detected antibiotic resistance. The mutations in the two resistant strains sequenced here could also be co-incidental to ARGs taken up through HGT. In any case, as abovementioned, we failed to recover plasmids by DNA extraction or sequencing. However, according to the manufacturer, the extraction kit used (E.Z.N.A. Bacterial DNA kit, Omega Bio-Tek, USA) is able to recover plasmid DNA. It seems unlikely that all 17 putative transconjugants have achieved resistance through random mutations, this is especially unlikely for those exhibiting higher level resistance such as 2POULTRY-ERY and 2POULTRY-STR. Most putative transconjugants displayed a multi-resistance phenotype (except 2SHEEP-ERY and 2POULTRY-ERY). Those resistant to erythromycin, gentamicin, kanamycin and streptomycin were detected independently of the amendment, whereas tetracycline and ampicillin resistance was only detected in putative transconjugants isolated under chloramphenicol selection. This could indicate that tetracycline, ampicillin and chloramphenicol resistance might be associated to the same MGE in these putative transconjugants, as suggested by Krauland (2010). Putative transconjugants with sulfonamide resistance were only found in Site 1, although sulfonamide resistance genes sul1 and sul2 are widely spread on plasmids from different plasmid groups (sul2: IncQ, LowGC plasmids; sul1: IncP-1, IncN, IncW often linked to class 1 integrons) (Heuer et al., 2009, 2012).

731

Some antimicrobials, such as imipenem, are considered drugs of last resort for treating antibiotic-resistant pathogens. Therefore, the lowlevel resistance to this antibiotic observed in 1PAPER-STR is interesting to note. Although high-level imipenem resistance is still rare in nature, a few studies have reported resistant bacteria in soil, including manuretreated farm soil (Rossolini et al., 2001). In fact, plasmids tend to levy a fitness cost on their hosts, and then plasmid-free competitors can outcompete plasmid-containing cells, driving plasmid extinction from the population (Hall et al., 2017). Alternatively, although resistance plasmids can indeed impose a fitness cost upon their hosts, work with plasmids pBR322 and pACYC184 has shown that bacteria are able to develop compensatory chromosomal mutations during a period of co-evolution, after which the plasmid-carrying strain becomes fitter than the plasmid-free parent strain (Enne et al., 2004; Lenski et al., 1994; Helling et al., 1981; Johnsen et al., 2002; McDermott et al., 1993). Evidence of evolutionary benefits and increased fitness by HGT is overwhelming, but it has also been noticed (Park and Zhang, 2012) that acquired DNA often functions inefficiently within new genomic backgrounds, thus being energetically or physiologically costly (gain versus loss). HGT is known to contribute to bacterial evolution and adaptation to new niches, but, at the same time, it is a dangerous process, as newly acquired genetic material can burden the overall cell physiology; for instance, horizontally acquired DNA can disrupt genomic features, sequester cellular housekeeping machineries, and limit cellular resources (e.g., the accompanying cost for transcription and translation of new genes and system-level effects due to rewiring of regulatory pathways) (Blokesch, 2017). Furthermore, MGEs are known to encode functions that promote their own transfer and which can interfere with host cell physiology (Blokesch, 2017). Escherichia coli is not well adapted to live in soil and therefore any exogenous plasmid isolation is likely based on interspecies gene transfer events that may result in new genetic combinations, capable of disrupting regulatory and physiological networks within the host or impeding the cellular machinery from performing essential housekeeping processes (Baltrus, 2013). However, although 1PAPER-STR showed the greatest physiological change of all putative transconjugants with respect to the recipient, it is likely that this was not caused by HGT but rather by a mutation in the ubiH gene, which is essential for non-fermentative metabolism. It is also interesting to note that most putative transconjugants had a higher RNA: DNA ratio compared to the recipient strain. This ratio is a frequently used indicator of cell growth and nutritional status (Clemmesen et al., 2003; Dortch et al., 1983; Elser et al., 2006), as it provides an index of protein synthesis capacity during active growth. 1PAPER-AMP presented one of the highest μmax values, along with one of the lowest increments in the RNA: DNA ratio, and its metabolic capacity appeared to be very similar to that of the recipient strain. If its ampicillin resistance was achieved via HGT, we speculate that it acquired a small, well-adapted plasmid, possibly integrated into the chromosome. In this respect, it has been reported that positive selection for advantageous plasmid-borne accessory genes alone is unlikely to underlie prolonged plasmid maintenance, because accessory genes are typically able to recombine with and become captured by the chromosome, leaving the costly autonomous plasmid redundant (Hall et al., 2017). In conclusion, our results indicate that metal-contaminated mine soils harbour a significant reservoir of MGEs. In terms of the isolated transconjugants, it is also likely that these MGEs hosted multiple antibiotic resistance genes. However, we observed no significant difference between treated soils and untreated controls, and no cross- or co-resistance to heavy metals and antibiotics in the putative transconjugants. Genome sequencing of two putative transconjugants failed to recover horizontally acquired genes, instead in one transconjugant a missense mutation likely inducing the observed resistance to aminoglycosides was detected. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2017.11.221.

732

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733

Acknowledgements This work has been supported by the Spanish Ministry of Economy and Competitiveness (AGL2015-64481-C2-1-R and AGL2016-76592R), the Basque Government (IT1018-16) and the Interreg SUDOE Programme (PhytoSUDOE-SOE1/P5/EO189). Olatz Garaiyurrebaso was a PhD student supported by the Basque Government and by the Fundación Biofísica Bizkaia. The authors acknowledge the technical support provided by Eneritz Bilbao and Iker Martín. References Bahl, M.I., Burmølle, M., Meisner, A., Hansen, L.H., Sørensen, S.J., 2009. All IncP-1 plasmid subgroups, including the novel ε subgroup, are prevalent in the influent of a Danish wastewater treatment plant. Plasmid 62, 134–139. Baisa, G., Stabo, N.J., Welch, R.A., 2013. Characterization of Escherichia coli D-cycloserine transport and resistant mutants. J. Bacteriol. 195, 1389–1399. Baltrus, D.A., 2013. Exploring the costs of horizontal gene transfer. Trends Ecol. Evol. 28, 489–495. Bankevich, A., Nurk, S., Antipov, D., Gurevich, A.A., Dvorkin, M., Kulikov, A.S., Lesin, V.M., Nikolenko, S.I., Pham, S., Prjibelski, A.D., Pyshkin, A.V., Sirotkin, A.V., Vyahhi, N., Tesler, G., Alekseyev, M.A., Pevzner, P.A., 2012. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J. Comput. Biol. 19, 455–477. Barrutia, O., Artetxe, U., Hernández, A., Olano, J.M., García-Plazaola, J.I., Garbisu, C., Becerril, J.M., 2011. Native plant community growing on an abandoned lead-zinc mining area in northern Spain: implications for phytoremediation and germplasm preservation. Int. J. Phytorem. 13, 256–270. Blokesch, M., 2017. In and out - contribution of natural transformation to the shuffling of large genomic regions. Curr. Opin. Microbiol. 38, 22–29. Canal, N., Meneghetti, K.L., Ponzi de Almeida, C., da Rosa Bastos, M., Muner Otton, L., Corcão, G., 2016. Characterization of the variable region in the class 1 integron of antimicrobial-resistant Escherichia coli isolated from surface water. Braz. J. Microbiol. 47, 337–344. Christensen, G.D., Simpson, W.A., Younger, J.J., Baddour, L.M., Barrett, F.F., Melton, D.M., Beachey, E.H., 1985. Adherence of coagulase-negative staphylococci to plastic tissue culture plates: a quantitative model for the adherence of staphylococci to medical devices. J. Clin. Microbiol. 22, 996–1006. Christensen, B.B., Sternberg, C., Andersen, J.B., Eberl, L., Moller, S., Givskov, M., Molin, S., 1998. Establishment of new genetic traits in a microbial biofilm community. Appl. Environ. Microbiol. 64, 2247–2255. Clemmesen, C., Bühler, V., Carvalho, G., Case, R., Evans, G., Hauser, L., Hutchinson, W.F., Kjesbu, O.S., Mempel, H., Moksness, E., Otteraa, H., Paulsen, H., Thorsen, A., Svaasand, T., 2003. Variability in condition and growth of Atlantic cod larvae and juveniles reared in mesocosms: environmental and maternal effects. J. Fish Biol. 62, 706–723. Collis, C.M., Grigg, G.W., 1989. An Escherichia coli mutant resistant to phleomycin, bleomycin, and heat inactivation is defective in ubiquinone synthesis. J. Bacteriol. 171, 4792–4798. Dahlberg, C., Chao, L., 2003. Amelioration of the cost of conjugative plasmid carriage in Escherichia coli K12. Genetics 165, 1641–1649. De Lipthay, J.R., Rasmussen, L.D., Oregaard, G., Simonsen, K., Bahl, M.I., Kroer, N., Sorensen, S.J., 2008. Acclimation of subsurface microbial communities to mercury. FEMS Microbiol. Ecol. 65, 145–155. Dealtry, S., Ding, G.C., Weichelt, V., Dunon, V., Schlüter, A., Martini, M.C., Del Papa, M.F., Lagares, A., Amos, G.C., Wellington, E.M., Gaze, W.H., Sipkema, D., Sjöling, S., Springael, D., Heuer, H., van Elsas, J.D., Thomas, C., Smalla, K., 2014. Cultivation-independent screening revealed hot spots of IncP-1, IncP-7 and IncP-9 plasmid occurrence in different environmental habitats. PLoS One 2, e89922. Di Rosa, R., Creti, R., Venditti, W., Amelio, R.D., Arciola, C.R., Montanaro, L., Baldassari, L., 2006. Relationship between biofilm formation, the enterococcal surface protein and gelatinase in clinical isolates of Enterococcus faecalis and Enterococcus faecium. FEMS Microbiol. Lett. 256, 145–150. Dionisio, F., Conceição, I.C., Marques, A.C., Fernandes, L., Gordo, I., 2005. The evolution of a conjugative plasmid and its ability to increase bacterial fitness. Biol. Lett. 1, 250–252. Dortch, Q.F., Roberts, T.L., Clayton Jr., J.R., Ahmed, S.I., 1983. RNA/DNA ratios and DNA concentrations as indicators of growth rate and biomass in planktonic marine organisms. Mar. Ecol. Prog. Ser. 13, 67–71. Elser, J.J., Watts, T., Bitler, B., Markow, T.A., 2006. Ontogenetic coupling of growth rate with RNA and P contents in five species of Drosophila. Ecology 2, 846–856. Enne, V.I., Bennett, P.M., Livermore, D.M., Hall, L.M.C., 2004. Enhancement of host fitness by the sul2-coding plasmid p9123 in the absence of selective pressure. J. Antimicrob. Chemother. 53, 958–963. Epelde, L., Becerril, J.M., Kowalchuk, G.A., Deng, Y., Zhou, J.Z., Garbisu, C., 2010. Impact of metal pollution and Thlaspi caerulescens growth on soil microbial communities. Appl. Environ. Microbiol. 76, 7843–7853. Epelde, L., Lanzén, A., Blanco, F., Urich, T., Garbisu, C., 2014. Adaptation of soil microbial community structure and function to chronic metal contamination at an abandoned Pb-Zn mine. FEMS Microbiol. Ecol. 91, 1–11. Ewing, B., Green, P., 1998. Basecalling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8, 186–194. Galende, M.A., Becerril, J.M., Barrutia, O., Artetxe, U., Garbisu, C., Hernandez, A., 2014. Field assessment of the effectiveness of organic amendments for aided phytostabilization of a Pb-Zn contaminated mine soil. J. Geochem. Explor. 145, 181–189.

Galhardo, R.S., Hastings, P.J., Rosenberg, S.M., 2007. Mutation as a stress response and the regulation of evolvability. Crit. Rev. Biochem. Mol. Biol. 42, 399–435. Garaiyurrebaso, O., Garbisu, C., Blanco, F., Lanzén, A., Martín, I., Epelde, L., Becerril, J.M., Jechalke, S., Smalla, K., Grohmann, E., Alkorta, I., 2017. Long-term effects of aided phytostabilization on microbial communities of metal-contaminated mine soil. FEMS Microbiol. Ecol. Mar;93 (3). https://doi.org/10.1093/femsec/fiw252 (Epub 2016 Dec 22). Gillings, M.R., 2017. Class 1 integrons as invasive species. Curr. Opin. Microbiol. 38, 10–15. Gordon, D., Abajian, C., Green, P., 1998. Consed: a graphical tool for sequence finishing. Genome Res. 8, 195–202. Götz, A., Smalla, K., 1997. Manure enhances plasmid mobilization and survival of Pseudomonas putida introduced into field soil. Appl. Environ. Microbiol. 63, 1980–1986. Götz, A., Pukall, R., Smit, E., Tietze, E., Prager, R., Tschäpe, H., van Elsas, J.D., Smalla, K., 1996. Detection and characterization of broad-host-range plasmids in environmental bacteria by PCR. Appl. Environ. Microbiol. 62, 2621–2628. Gurevich, A., Saveliev, V., Vyahhi, N., Tesler, G., 2013. QUAST: quality assessment tool for genome assemblies. Bioinformatics 29, 1072–1075. Hall, J.P.J., Brockhurst, M.A., Dytham, C., Harrison, E., 2017. The evolution of plasmid stability: are infectious transmission and compensatory evolution competing evolutionary trajectories? Plasmid 91, 90–95. Helling, R.B., Kinney, T., Adams, J., 1981. The maintenance of plasmid-containing organisms in populations of Escherichia coli. J. Gen. Microbiol. 123, 129–141. Heuer, H., Smalla, K., 2012. Plasmids foster diversification and adaptation of bacterial populations in soil. FEMS Microbiol. Rev. 36, 1083–1104. Heuer, H., Krögerrecklenfort, E., Wellington, E.M.H., Egan, S., van Elsas, J.D., van Overbeek, L., Collard, J.M., Guillaume, G., Karagouni, A.D., Nikolakopoulou, T.L., Smalla, K., 2002. Gentamicin resistance genes in environmental bacteria: prevalence and transfer. FEMS Microbiol. Ecol. 42, 289–302. Heuer, H., Kopmann, C., Bihn, C., Top, E., Smalla, K., 2009. Spreading antibiotic resistance through spread manure: characteristics of a novel plasmid type with low %G+C content. Environ. Microbiol. 11, 937–949. Heuer, H., Binh, C., Jechalke, S., Kopmann, C., Zimmerling, U., Krögerrecklenfort, E., Ledger, T., González, B., Top, E., Smalla, K., 2012. IncP-1ε plasmids are important vectors of antibiotic resistance genes in agricultural systems: diversification driven by class 1 integron gene cassettes. Front. Microbiol. 3, 2. Jechalke, S., Kopmann, C., Richter, M., Moenickes, S., Heuer, H., Smalla, K., 2013. Plasmidmediated fitness advantage of Acinetobacter baylyi in sulfadiazine-polluted soil. FEMS Microbiol. Lett. 348, 127–132. Johnsen, P.J., Simonsen, G.S., Olsvik, O., Midtvedt, T., Sundsfjord, A., 2002. Stability, persistence, and evolution of plasmid-encoded VanA glycopeptide resistance in enterococci in the absence of antibiotic selection in vitro and in gnotobiotic mice. Microb. Drug Resist. 8, 161–170. Krauland, M., 2010. Novel integron gene cassette arrays identified in a global collection of multi-drug resistant non-typhoidal Salmonella enterica. Curr. Microbiol. 60, 217–223. Kurtz, S., Phillippy, A., Delcher, A.L., Smoot, M., Shumway, M., Antonescu, C., Salzberg, S.L., 2004. Versatile and open software for comparing large genomes. Genome Biol. 5, R12. Kyselková, M., Kotrbová, L., Bhumibhamon, G., Chroňáková, A., Jirout, J., Vrchotová, N., Schmitt, H., Elhottová, D., 2015. Tetracycline resistance genes persist in soil amended with cattle feces independently from chlortetracycline selection pressure. Soil Biol. Biochem. 81, 259–265. Lenski, R.E., Simpson, S.C., Nguyen, T.T., 1994. Genetic analysis of a plasmid-encoded, host genotype-specific enhancement of bacterial fitness. J. Bacteriol. 176, 3140–3147. Lommatzsch, J., Templin, M., Kraft, A., Vollmer, W., Höltje, J., 1997. Outer membrane localization of murein hydrolases: MltA, a third lipoprotein lytic transglycosylase in Escherichia coli. J. Bacteriol. 179, 5465–5470. Martin, B., Humbert, O., Camara, M., Guenzi, E., Walker, J., Mitchell, T., Andrew, P., Prudhomme, M., Alloing, G., Hakenbeck, R., Morrison, D.A., Boulnois, G.J., Claverys, J.P., 1992. A highly conserved repeated DNA element located in the chromosome of Streptococcus pneumonia. Nucleic Acids Res. 20, 3479–3483. Martínez, J.L., Sánchez, M.B., Martínez-Solano, L., Hernandez, A., Garmendia, L., Fajardo, A., Alvarez-Ortega, C., 2009. Functional role of bacterial multidrug efflux pumps in microbial natural ecosystems. FEMS Microbiol. Rev. 33, 430–449. McDermott, P.J., Gowland, P., Gowland, P.C., 1993. Adaptation of Escherichia coli growth rates to the presence of pBR322. Lett. Appl. Microbiol. 17, 139–143. Moyá, B., Zamorano, L., Juan, C., Ge, Y., Oliver, A., 2010. Affinity of the new cephalosporin cxa-101 to penicillin-binding proteins of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 54, 3933–3937. Muir, M.E., Hanwell, D.R., Wallace, B.J., 1981. Characterization of a respiratory mutant of Escherichia coli with reduced uptake of aminoglycoside antibiotics. Biochim. Biophys. Acta (BBA) Bioenergetics 638, 234–241. Myers, J.A., Curtis, B.S., Curtis, W.R., 2013. Improving accuracy of cell and chromophore concentration measurements using optical density. BMC Biophys. 6, 4. Nakahigashi, K., Miyamoto, K., Nishimura, K., Inokuchi, H., 1992. Isolation and characterization of a light-sensitive mutant of Escherichia coli K-12 with a mutation in a gene that is required for the biosynthesis of ubiquinone. J. Bacteriol. 174, 7352–7359. NCCLS, 2004. Performance Standards for Antimicrobial Susceptibility Testing; Fourteenth Informational Supplement. 24. The National Committee for Clinical Laboratory Standards, Wayne, Pennsylvania, pp. 96–130. Nõlvak, H., Truu, M., Kanger, K., Tampere, M., Espenberg, M., Loit, E., Raave, H., Truu, J., 2016. Inorganic and organic fertilizers impact the abundance and proportion of antibiotic resistance and integron-integrase genes in agricultural grassland soil. Sci. Total Environ. 562, 678–689. Norman, N., Hansen, L.H., Sørensen, S.J., 2009. Conjugative plasmids: vessels of the communal gene pool. Philos. Trans. R. Soc. B 364, 2275–2289. Park, C., Zhang, J., 2012. High expression hampers horizontal gene transfer. Genome Biol. Evol. 4, 523–532.

C. Garbisu et al. / Science of the Total Environment 621 (2018) 725–733 Paulsson, J., 2002. Multileveled selection on plasmid replication. Genetics 161, 1373–1384. Peng, S., Feng, Y., Wang, Y., Guo, X., Chu, H., Lin, X., 2017. Prevalence of antibiotic resistance genes in soils after continually applied with different manure for 30 years. J. Hazard. Mater. 340, 16–25. Perni, S., Andrew, P.W., Shama, G., 2005. Estimating the maximum growth rate from microbial growth curves: definition is everything. Food Microbiol. 22, 491–495. Rossolini, G., Condemi, M.A., Pantanella, F., Docquier, J.D., Amicosante, G., Thaller, M.C., 2001. Metallo-β-lactamase producers in environmental microbiota: new molecular class B enzyme in Janthinobacterium lividum. Antimicrob. Agents Chemother. 45, 837–844. Sandvang, D., Aarestrup, F.M., Jensen, L.B., 1997. Characterisation of integrons and antibiotic resistance genes in Danish multiresistant Salmonella enterica Typhimurium DT104. FEMS Microbiol. Lett. 157, 177–181.

733

Seiler, C., Berendonk, T., 2012. Heavy metal driven co-selection of antibiotic resistance in soil and water bodies impacted by agriculture and aquaculture. Front. Microbiol. 3, 399. https://doi.org/10.3389/fmicb.2012.00399. Vollmer, W., von Rechenberg, M., Höltje, J.-V., 1999. Demonstration of molecular interactions between the murein polymerase PBP1B, the lytic transglycosylase MltA, and the scaffolding protein MipA of Escherichia coli. J. Biol. Chem. 274, 6726–6734. Ye, J., Coulouris, G., Zaretskaya, I., Cutcutache, I., Rozen, S., Madden, T., 2012. PrimerBLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinf. 13, 134. Zur Wiesch, P.A., Kouyos, R., Engelstädter, J., Regoes, R.R., Bonhoeffer, S., 2011. Population biological principles of drug-resistance evolution in infectious diseases. Lancet Infect. Dis. 11, 236–247.