agricultural use of antibiotics and antibiotic resistance

Chapter 8

AGRICULTURAL USE OF ANTIBIOTICS AND ANTIBIOTIC RESISTANCE Satoshi Koike1, Roderick Mackie2,3 and Rustam Aminov4* 1

Research Faculty of Agriculture, Hokkaido University, Sapporo, Japan 2 Department of Animal Sciences, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA 3 Carl R. Woese Institute for Genomic Biology, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA 4 School of Medicine and Dentistry, University of Aberdeen, Aberdeen, UK

ABSTRACT Antibiotics are widely used in food animal production systems for treatment of infectious disease as well as for metaphylactic and growth promoting purposes. This selective pressure results in the selection of bacteria that are able to acquire resistance to antibiotics, either by mutation(s) or via horizontal gene transfer. There are several estimates comparing the quantity of antibiotics used for human and animal consumption, and these quantities are predicted to grow rapidly in coming years. Estimates show that a substantial proportion of antibiotics used in agriculture are allocated to the subtherapeutic use. This type of use creates almost ideal conditions for selection and dissemination of antibiotic resistance. In this chapter, we review our previous work on molecular ecology of antibiotic resistance in agricultural settings. In particular, how the agricultural use of antibiotics contribute to selection and dissemination of antibiotic resistance genes into a wider environment, and how different agricultural practices may affect this process.

* Corresponding Author: Rustam Aminov, School of Medicine and Dentistry, University of Aberdeen, Aberdeen, United Kingdom, Email: [email protected].

2

Satoshi Koike, Roderick Mackie and Rustam Aminov

Keywords: animal production, antibiotic use, antibiotic resistance genes, dissemination

1. INTRODUCTION The growth-promoting effect of low-dose antibiotics on animals was discovered accidentally. In the late 1930s - beginning 1940s many companies initiated antibiotic discovery programmes. One of such programmes, at Lederle Laboratory Division of American Cyanamid Company, resulted in the discovery of chlortetracycline (Aureomycin) by Benjamin Minge Duggar (Duggar, 1948). His colleagues, animal nutritionist Robert Stokstad and biochemist Thomas Jukes were testing the effect of the producer strain (Streptomyces aureofaciens) biomass on the growth of chicken. They found that the biomass remaining after the fermentation and extraction of Aureomycin had a growth-promoting effect on the chickens. Since the original aim of their programme was the search for sources of vitamin B-12 to use as a feed additive for the food animal industry, they initially thought that the effect was due to the presence of this vitamin in the spent biomass of the antibiotic producer strain. More detailed analysis, however, established that the growth-promoting effect was not due to the vitamin but because of the presence of the residual chlortetracycline in the cell debris of S. auerofaciens left after the antibiotic extraction. After verification of the growth promoting effect of low dose chlortetracycline, American Cyanamid Company swiftly initiated the development of this new antibiotic feed additive, and this successful example was quickly followed by many other companies and countries around the world. Several present-day estimates suggest that a substantial proportion of antibiotics produced are used in food animals for non-therapeutic purposes (Landers et al. 2012; Krishnasamy and Silbergeld, 2015). There is a substantial body of evidence suggesting an epidemiological link between the antibiotic use in food animals and antibiotic resistance in humans (Landers et al. 2012). It took a considerable amount of time and effort by researchers to reveal this link, and the corresponding legislative measures to limit and ban the non-therapeutic use of antibiotics were implemented in many

Agricultural Use of Antibiotics and Antibiotic Resistance

3

countries, primarily within the EU. The first country to implement such rules was Sweden, which prohibited the use of growth-promoting antibiotics in 1986, while other EU countries banned specific antibiotics in feedstuffs in the years before January 1, 2006, when all these antibiotics were deleted from the Community Register of authorized feed additives (EPC, 2005; Castanon, 2007). Despite this, the global use of antimicrobials in food animals is predicted to grow rapidly, by at least 67% from 2010 to 2030 (Van Boeckel et al. 2015). In the BRICS countries (Brazil, Russia, India, China, and South Africa) the antimicrobial consumption by livestock is predicted to double. The total amount of antibiotics produced for this purpose is predicted to reach 105,596 ± 3,605 tons in 2030 from 63,151 ± 1,560 tons in 2010 (Van Boeckel et al. 2015). It is estimated about a third of this increase will be due to the switch from more extensive farming practices to large-scale intensive farming operations that routinely use antimicrobials at sub-therapeutic concentrations (Van Boeckel et al. 2015). Two of the most frequently used antibiotics in animal production systems belong to the tetracycline and macrolide classes. The most recent numbers available indicate that in 2013, for example, tetracyclines were the most frequently used antibiotics in food animals in the USA; the quantity sold to the industry that year reached a staggering 6,514,779 kg of active ingredient (FDA, 2015). This accounted for 71% of all antibiotics sold for use in food-producing animals. Tetracyclines that are administered via medicated feed accounted for the majority of domestic sales and distribution of medically important antimicrobials approved for the use in foodproducing animals. They were also the top antibiotics for administration via drinking water (FDA, 2015). In addition, the non-clinical use of tetracyclines is extended to aquaculture and horticulture (Chopra and Roberts, 2001). Thus the scale of the use of tetracyclines in agriculture is overwhelming. This large-scale use of tetracyclines in agriculture is probably responsible for the widespread resistance that made the first-generation tetracyclines essentially useless for the treatment of human infectious diseases. The two main mechanisms of tetracycline resistance include the active efflux of the drug from the cell and the synthesis of alternative elongation

4


factors that prevent the binding of tetracyclines to the ribosomes (Chopra and Roberts, 2001; Roberts, 2005). Alternative elongation factors are widespread among Gram-positive and Gram-negative pathogenic and commensal bacteria and because of this they were chosen for our study of molecular ecology of antibiotic resistance genes (Aminov et al. 2001). Among the efflux pumps, those specific for Gram-negative bacteria were chosen for our analyses because of the broader diversity and higher occurrence in the pathogenic microbiota (Aminov et al. 2002). Among the less common mechanisms of tetracycline resistance are the enzymatic degradation of tetracycline (Speer et al. 1991) and presumably a metabolic mechanism protecting against the entry of the antibiotic into the cell (Kazimierczak et al. 2009). The former mechanism, which is encoded by the tet(X) gene, however, may have a potential impact in the future for the treatment by third-generation tetracyclines (Aminov, 2013). This resistance mechanism seems to be rapidly penetrating potentially pathogenic microbiota due to the agricultural and clinical use of the first-generation tetracyclines. One of the main mechanisms of acquired resistance to macrolides is the activity of specific methylases, which are encoded by a variety of the erm genes and which dimethylate the adenine residue A2058 in the 23S rRNA component of the 50S large subunit of the bacterial ribosome (Weisblum 1995). This residue is located within the conserved region of domain V, which is a binding site for structurally unrelated but functionally similar antibiotics belonging to the macrolides, lincosamides, and depsipeptides. As a consequence of this methylation, the bacterial host becomes cross-resistant to all three classes of antibiotics, with the phenotype, which is called MLSB (macrolide-lincosamin-streptogramin B) resistant (Leclercq, 2002). In our previous research, we performed complex analyses of various aspects of antibiotic resistance in agriculture. First, we performed phylogenetic analyses of antibiotic resistance genes conferring resistance to several classes of antibiotics. Besides the reconstruction of evolutionary history of antibiotic resistance genes, which is very interesting on its own, these analyses allowed the phylogenetically aided design and development of molecular tools such as PCR primers for conventional and quantitative


5

PCR. Second, the molecular ecology approach developed, together with classical microbiology techniques, were applied to several pig production operations with the subtherapeutic use of antibiotics to investigate the potential for dissemination of antibiotic resistance genes into a wider environment such as groundwater and soil. Third, in our follow up works we also examined the effect of different agricultural practices such as organic and conventional pig production systems on the occurrence and diversity of certain antibiotic resistance genes using a metagenomic approach. A brief overview of the results will be given in this chapter.

2. METHODS 2.1. Phylogenetic Analysis and Primer Design As mentioned earlier there are two main mechanisms of tetracycline resistance, the active efflux of the drug from the cell and the synthesis of alternative elongation factors that prevent the binding of tetracyclines to the ribosomes (Chopra and Roberts, 2001; Roberts, 2005). The genes encoding these two families of tetracycline resistance were downloaded from the GenBank database (Benson et al. 1999). As for the acquired resistance to macrolides, one of the main mechanisms of resistance is the activity of specific methylases, which are encoded by a variety of the erm genes (Weisblum 1995). These were also downloaded from GenBank. The corresponding sequences were aligned using the multiple sequence alignment software ClustalX ver. 1.83 (Thompson et al. 1997). Phylogenetic analyses were performed using the neighbour joining method (Saitou and Nei, 1987), with the two-parameter model of Kimura (1980). The statistical significance of branching was evaluated by bootstrap analysis (Felsenstein, 1985) involving the construction of 1000 trees from re-sampled data. Other methods of phylogenetic reconstruction included the maximum likelihood and parsimony analyses implemented in DNAML and DNAPARS programs of the PHYLIP package (Felsenstein, 1989). In more recent analyses, Bayesian inference was used to infer phylogenetic trees and estimate the clade support. Posterior probabilities were calculated using a Bayesian

6


Markov chain Monte Carlo method implemented in the MrBayes v.3.1.1 program (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003). A 4 by 4 nucleotide substitution model with gamma-distributed site variation and a proportion of invariable sites was used. The priors were left at the defaults (i.e., uninformative priors) in MrBayes. Ten or twenty million generations were run, with sampling every 100 generations. Based on phylogenetic analyses, the corresponding clusters in the alignment file were manually inspected to design PCR primers that are specific while cover all diversity within the sequences belonging to the same tet or erm gene. The oligonucleotides designed were initially tested in silico, followed by rigorous laboratory validation with positive and negative controls. Other antibiotic resistance genes of interest used in our phylogenetic analyses included the cluster of vancomycin resistance genes, vanHAX, quinolone resistance genes, qnr (Aminov, 2007), as well as tet(X) (Aminov, 2009; Aminov, 2013).

2.2. Description of Sites Groundwater and Waste Lagoon Samples Groundwater and waste lagoon samples were collected from two intensive practice pig farms in Illinois, USA. The hydrogeology, position of sampling wells and farm operation details for these two sites has been reported previously (Chee-Sanford et al. 2001; Mackie et al. 2006). In brief, the site designated as Site A was a finishing operation, in which a two-stage waste handling system was implemented. Most of the swine waste solids settled in a concrete basin, from which the supernatant liquid freely flowed into an earthen lagoon. Site C was a farrowing and nursery operation, with a single-stage waste lagoon. Sixteen wells at Site A and six at Site C were previously installed to measure the potential waste contamination impact from the lagoons to groundwater using electromagnetic terrain conductivity surveys (Larson et al. 1997). Both operations used chlortetracycline and tylosin (personal communication with facilities operators) but the specifics of antibiotic use were not available.


7

Faecal Samples Faecal samples were also collected from two pig farms in the UK, one certified as organic and the other a conventional farm. The organic farm was located in Aberdeenshire, Scotland, and the conventional farm in the Bristol area, England. Pigs on the organic farm did not receive antibiotics for growth promotion, and none had been used for disease treatment. On the conventional farm, therapeutic antibiotics were given to animals when necessary, and the sows were prophylactically given trimediazine, which is a mix of trimethoprim and sulfadiazine, during the lactation period. The faecal samples on this farm, however, were collected from dry sows that were not receiving the antibiotic at the time of sampling. The sampling was performed on May 10, 2006, five months after the ban of growth-promoting antibiotics in the EU countries.

2.3. Sampling, DNA Isolation, and PCR Groundwater and Waste Lagoon Samples Before groundwater sampling, the wells were purged by removing 1.5 to 3 well volumes. Lagoon samples were collected from selected locations around the lagoon including up gradient and down gradient wells and consisted of eight pooled 2 L subsamples. Samples were collected into sterile bottles and kept on ice during transportation to the laboratory. The samples were stored at 4˚C prior to analysis. DNA isolation and purification for PCR analyses was performed as described (Koike et al. 2007). Faecal Samples Faecal samples from five adult animals were collected from both organic and conventional farms and transported to the laboratory in the Cary Blair transport medium for anaerobic organisms (Oxoid, UK). Total DNA from five pooled faecal samples was immediately isolated as described before (Kazimierczak et al. 2009).

8


PCR Analysis of Tet and Erm Genes Detailed conditions of conventional and quantitative PCR are described in a number of our previous publications (Aminov et al. 2001; Aminov et al. 2002; Aminov et al. 2004; Koike et al. 2007; Koike et al. 2010).

2.4. Metagenomic Library Construction and Screening BAC libraries were constructed with the use of a CopyControl™ BAC cloning kit (EPICENTRE®) as described earlier (Kazimierczak et al. 2009). Recombinant clones were selected on LB agar plates containing 12.5 µg/ml of chloramphenicol (Cm). The resulting CmR clones were picked up using a BioRobotics BioPick colony picker (Genomic Solutions, USA) and arrayed into a 384-well microtiter plate containing 70 µl of the cryoprotective solution (2xLB medium supplemented with 10% glycerol). The library from the organic farm consisted of 9,000 clones, and the library from the conventional farm consisted of 10,400 clones). The cells were grown overnight at 37˚C and stored at -70˚C. The library was further screened by printing on LB agar plates containing 5 µg/ml tetracycline (Tc), 3 µg/ml minocycline (Mn), 3 µg/ml doxycycline (Dx), and 0.5 µg/ml tigecycline (Tg).

2.5. Sequencing and Bioinformatics Since the number of tetracycline resistant clones in the library from the organic farm was low (10 clones), all of them were sequenced using Sanger chemistry and primer walking. Sequencing reactions were read on an automated eight-capillary Beckman sequencer (Beckman, UK). Chromatograms were assembled with the Lasergene 6 package. Nucleotide and translated amino acid sequences were analysed by using on-line BLAST (http://www.ncbi.nlm.nih.gov/blast), PFAM (http://www.sanger. ac.uk/Software/Pfam), and PSORT (http://psort.nibb.ac.jp) programs.


9

A total of 132 stable clones resistant to at least one of these four selective antibiotics (except tigecycline) were identified, and these clones were sequenced using a different strategy. Specifically, this was performed with the combination of Sanger, 454 and Illumina technologies. Sanger sequencing was applied to sequence the insert ends with vector-specific primers. For next-generation sequencing, the BAC DNAs were mixed in equimolar quantities and the resulting DNA mixes were sequenced on Roche GS FLX and Illumina HiSeq 2000 machines by the GenePool (http://genepool.bio.ed.ac.uk). All sequence data were filtered for host and vector contamination and assembled to reconstitute the original cloned DNA sequences. Annotation of the resulting contigs was performed using the RAST server (Aziz et al. 2008).

3. RESULTS 3.1. Evolutionary History and Ecology of Antibiotic Resistance Genes Tetracycline Resistance The very first phylogenetic reconstruction attempted, with the genes encoding ribosomal proteins protecting against the action of tetracycline, revealed the ancient evolutionary history of these genes (Aminov et al. 2001). With a high confidence level, it has been established that this group of genes is monophyletic and branched very early in evolution from other elongation factors. Moreover, the tetracycline resistance genes currently residing in commensal and pathogenic microbiota have also branched fairly early in the evolution from the corresponding genes of antibiotic-producing strains. The latter sister clade is exemplified by the tet and otr(A) genes from the producer strains, that is from the antibiotic-producing strains, Streptomyces lividans and S. rimosus. Later, when more sequence data of genes encoding ribosomal proteins protection proteins became available, we have repeated these phylogenetic analyses with a large set of genes (Aminov and Mackie, 2007). Once again, our initial observations of 2001 have been

10


confirmed. Thus, there is no indication of recent horizontal transfer of these genes from antibiotic producers to pathogens and commensals in the “antibiotic era.” There have been indications though that, according to hybridization data, some mycobacteria may actually carry the tetracycline resistance genes homologous to those originally described in streptomycetes (Pang et al. 1994; Roberts, 1996). However, the Mycobacterium sp. strain isolated from a root canal, appeared to be carrying the tet(M) gene unrelated to the genes present in streptomycetes (Rossi-Fedele et al. 2006). PCR screening with otr(A)-specific primers of a large collection of environmental and clinical mycobacterial isolates did not produce any positive hit (Kyselková et al. 2012). According to this and some earlier reports (Aínsa et al. 1998; De Rossi et al. 1998) tetracycline resistance in mycobacteria is predominantly encoded by the efflux pump genes, tet(V) and tap. Thus there is no indication of acquisition by pathogens of tetracyline resistance genes from antibiotic producers.

Macrolide Resistance Phylogenetic analyses of other antibiotic resistance genes also support the conclusion that there is no proven evolutionary history of horizontal transfer of antibiotic resistance genes from antibiotic producing microorganisms to pathogenic and commensal microbiota. In particular, our further phylogenetic analysis involved the erm genes encoding dimethylases that protect the ribosomes from the action of macrolides (Aminov and Mackie, 2007). This analysis demonstrated the existence of two clades, one with the erm genes found in high-G+C bacteria, including antibioticproducing Streptomyces and some pathogens, and another in low-G+C bacteria including taxonomically diverse commensal, pathogenic, and environmental bacteria. Our further analyses with the use of a large dataset and the maximum likelihood and Bayesian inference phylogenetic reconstruction algorithms effectively confirmed our initial proposition (Koike et al. 2010).


11

Vancomycin Resistance The similar dichotomy can be observed among the genes encoding resistance to vancomycin (Aminov and Mackie, 2007). Resistance towards this glycopeptide antibiotic is due to the alteration in the peptidoglycan synthesis pathway terminating in D-alanyl-D-lactate instead of D-alanyl-Dalanine (Aminov, 2016). Several genes encode this alternative pathway of peptidoglycan synthesis, and phylogenetic analysis, therefore, was performed with the concatenated set of three genes encoding D-lactate dehydrogenase, D-Ala-D-Lac ligase and D,D-dipeptidase (Aminov and Mackie, 2007). According to the phylogenetic placement of the vanHMX and vanHNX genes, the glycopeptide producers, Amycolatopsis orientalis and Actinoplanes teichomyceticus, which produce the glycopeptide antibiotics vancomycin and teicoplanin, correspondingly, formed a separate cluster, together with the Streptomyces spp. The phylogeny of the corresponding van genes in a number of the low-G+C bacteria from various ecological compartments including commensal, pathogenic, and environmental, formed a separate cluster, distinct from the glycopeptide producers (Aminov and Mackie, 2007). The cluster included bacteria causing human health issues such as the enterococci and staphylococci as well as environmental soil bacteria such as the species of the Bacillus and Paenibacillus genera. Thus, with the van genes as well, no support for transfer of antibiotic resistance genes from antibiotic producers to other, none-antibiotic-producing, bacteria was obtained. Moreover, the corresponding van gene cluster from the antibiotic-producer Amycolatopsis coloradensis cannot confer any detectable vancomycin resistance when expressed in eneterococci, while the expression of the van gene cluster from the environmental Paenibacillus species in enterococci results in high-level vancomycin resistance (Hasman et al. 2006). Thus this serves as another proof that the antibiotic resistance gene pools of antibiotic producers and non-producers are separated, with no detectable horizontal gene exchange. Quinolone Resistance Our further phylogenetic investigation included the genes conferring resistance to synthetic antibiotics to exclude the effect of antibiotic

12


producers and the corresponding resistance genes affiliated with them. In this model, it is possible to monitor for the resistance development without referring to the corresponding indigenous pool of the resistance genes residing in antibiotic producers. For this, we have chosen resistance to the class of synthetic antibiotics called quinolones (fluoroquinolones). The first representative of this class of antibiotics, nalidixic acid, was discovered accidentally during the efforts directed towards the synthesis of chloroquine used for malaria treatment (Wentland, 1993). The bactericidal activity of quinolones is due to the formation of a DNA gyrase-quinolone-DNA complex, which blocks the progression of the bacterial chromosome replication fork thus leading to cell death (Hiasa and Shea, 2000). Initially, the descriptions of mechanisms of resistance to quinolones were largely limited to target modifications, e.g., mutations in bacterial gyrase or topoisomerase IV protecting them from binding of quinolones and making the replication fork insensitive to the action of these drugs (Wohlkonig et al. 2010). Then other mechanisms of resistance emerged; in particular, acquired resistance mediated by the horizontally transferred traits was confirmed in 1998 (Ruiz et al. 2012). The mechanisms included target modification such as conferred by the qnr genes (Tran and Jacoby, 2002). This mechanism involves the binding of the qnr-encoded protein to DNA gyrase during the formation of gyrase-DNA complex, thus lowering gyrase binding to DNA and reducing the amount of holoenzyme-DNA targets for quinolone inhibition (Tran et al. 2005). The qnr genes encode proteins belonging to the pentapeptide repeat family proteins (Tran and Jacoby, 2002). As for the original role of these proteins, it has been suggested for one of the family members, MfpA, that it may help to maintain DNA in a condensed state, thus preventing undesirable topological changes during the replicative senescence periods (Hegde et al. 2005). Earlier, it has been suggested that the intrinsic resistance of Mycobacterium smegmatis to fluoroquinolones may be due to the presence of this protein (Montero et al. 2001). There are indeed a number of homologous genes in the environmental bacteria such as Photobacterium profundum, Vibrio vulnificus, and V. parahaemolyticus and, once cloned and expressed in E. coli, these genes may confer resistance to quinolones (Poirel


13

et al. 2005; Saga et al. 2005). Thus the function of the enzyme that is normally involved in DNA metabolism in environmental bacteria could be extended to other functions such as resistance to the synthetic antibiotic class such as the quinolones (Aminov, 2016). We performed phylogenetic analysis with the known qnr genes and the closest homologs encoding the pentapeptide repeat family proteins (Aminov and Mackie, 2007). The well-defined qnr gene families already proven to confer quinolone resistance included qnrA, qnrB, qnrS, and qnrVV. Interestingly, the qnrA genes are represented in several marine environmental isolates of Shewanella algae as well as in pathogenic bacteria such as Klebsiella pneumoniae, K. oxytoca, Salmonella enterica serotype Enteriditis, and Proteus mirabilis. The close similarity between the qnrA genes from different ecological compartments may suggest recent horizontal gene transfer event(s). At the time of our phylogenetic analysis of the qnr genes (Aminov and Mackie, 2007), the presence of the qnrS genes was detected in K. pneumoniae, S. enterica, Shigella flexneri, and in plasmid pGNB2 isolated from a wastewater treatment plant (Bönemann et al. 2006). Currently, the representation of environmental genomic and metagenomic sequences in databases is quite substantial. Thus we performed amino acid similarity searches using various quinolone resistance pentapeptide repeat proteins as queries. One of the representative results with the use of QnrS2 as a query sequence is shown in Table 1. There are 37 QnrS2 sequences retrieved from GenBank, which are 100% identical and which are encountered in taxonomically and ecologically diverse microbiota, ranging from the typical gut commensals to aquatic and pathogenic bacteria. In 25 of them, the qnrS2-encoding genes are located on plasmids thus offering a potential explanation for the observed rapid expansion of the gene into the ecologically and taxonomically diverse groups of bacteria (Table 1). Presently, however, it is difficult to discern the direction of horizontal gene transfer in the past since the qnrS2 genes have become quite ubiquitous in various ecological niches, especially in aquatic ecosystems (Wen et al. 2016). One of the hotspots could be wastewater treatment plants, the

14


effluents of which frequently contain elevated copy numbers of antibiotic resistance genes including qnrS (Rodriguez-Mozaz et al. 2015).

Microevolutionary Processes To clarify what could be the direction of antibiotic resistance gene flow in the environment, we constructed the tet(W) gene libraries from the DNA samples extracted from waste lagoons and groundwater in the vicinity of two swine farms (Koike et al. 2007). These libraries were sequenced and the phylogenetic analysis of the tet(W) genes revealed that there are several clades associated or not with the contamination coming from the waste lagoons. In brief, the groundwater impacted by the waste lagoon leaks contained tet(W) sequences very similar to that found in waste lagoons, and the sequences from both ecological compartments actually formed highly supported phylogenetic clusters. These results suggest the mobility of the antibiotic resistance gene pool between the ecological compartments such as the waste lagoons and groundwater. It remains to be established, however, whether this is a result of the migration of antibiotic-resistant bacteria from the lagoons or transmission of antibiotic resistant genes to the indigenous microbiota. In our earlier work, we established that the typical groundwater resident microbiota at the sites affected may carry the tet(M) genes found in the waste lagoons but the sequence resolution was too low to make this evidence conclusive (Chee-Sanford et al. 2001). Another important conclusion resulting from this work is that, despite being quite geographically distant from each other (265 km apart), the sequences from both sites were closely related and formed a novel tet(W) cluster. This implies that the two swine confinement facilities were subjected to the same selective pressure resulting in microevolutionary processes leading to the selection of specific tet(W) genotypes. At the same time, there were two distinct clusters of tet(W) sequences from groundwater, which was not affected by lagoon seepage (Koike et al. 2007). Thus there are also specific local selection pressures on the sites, which were independent of the waste lagoon impact and drive local speciation of the tet(W) genes. Thus, the highly sensitive Bayesian inference phylogenetic reconstruction allowed us to reveal the mobility of the resistome as well as to suggest that there are


15

common and differential selective pressures driving the microevolutionary processes in situ. The ecology of antibiotic resistance genes, however, was even more complex than the proposed model for tet(W) (Koike et al. 2010). Correlation between the distribution patterns of the erm and tet genes was found to be weak suggesting that additional factors may be involved in the dynamics and flux of erm genes in the environment. The phygenetic analyses of the tet and erm genes allowed to perform phylogenetically-aided design of PCR primers for these genes (Aminov et al. 2001; Animov et al. 2002; Koike et al. 2010). After rigorous validation, these tools were used to investigate molecular ecology of the tet and erm genes in the surrounding environment of pig farms. Table 1. Identical quinolone resistance pentapeptide repeat proteins QnrS2 in bacteria from different ecological niches Bacterium Aeromonas hydrophila (plasmid) Aeromonas sobria Aeromonas hydrophila Uncultured bacterium (plasmid) S. enterica subsp. enterica serovar Anatum (plasmid)

Protein designated quinolone resistance protein qnrS2 QnrS2 QnrS2 QnrS2

GenBank ID YP_002221301.1 YP_005015986.1 YP_006961374.1 ABE98197.1 ABF47470.1

Aeromonas caviae Aeromonas caviae (plasmid) Aeromonas media (plasmid) Aeromonas hydrophila (plasmid) Aeromonas hydrophila (plasmid) Aeromonas sobria (plasmid) Shigella sp. S120 (plasmid) Pseudoalteromonas sp. Q29 (plasmid)

quinolone resistance protein quinolone resistance protein S2 quinolone resistance protein S2 quinolone resistance protein QnrS2 quinolone resistance protein QnrS2 quinolone resistance determinant QnrS2

ABV68901.1 ACA29528.1 ACA29529.1 ACH70070.1 ACJ67292.1 AEG74317.1 AEL33521.1 AEP38971.1

Pseudoalteromonas sp. QC44 (plasmid)

quinolone resistance determinant AEP38972.1 QnrS2

Pseudomonas sp. Q24 (plasmid)

quinolone resistance determinant AEP38975.1 QnrS2 quinolone resistance protein S2 CCF03522.1 qnrS2 AEW70678.1

Aeromonas rivipollensis (plasmid) Aeromonas hydrophila (plasmid)

16


Aeromonas sobria (plasmid) Serratia marcescens Escherichia coli Aeromonas hydrophila (plasmid) Aeromonas sp. C3 (plasmid) Gammaproteobacteria bacterium Q1 Uncultured bacterium (plasmid) Aeromonas hydrophila (plasmid) Aeromonas hydrophila (plasmid) Aeromonas sobria (plasmid) Enterobacter hormaechei subsp. steigerwaltii Klebsiella pneumoniae (plasmid) Klebsiella pneumoniae (plasmid) Vibrio parahaemolyticus Vibrio parahaemolyticus Aeromonas caviae (plasmid) Escherichia coli Escherichia coli Escherichia coli Escherichia coli (plasmid)

qnrS2 quinolone resistance protein fluoroquinolone resistance protein QnrS2 Quinolone resistance protein S2 QnrS2 fluoroquinolone resistance protein

AEW70682.1 AEZ36151.1 KIG48700.1

Qnr QnrS2 QnrS2 QnrS2 fluoroquinolone resistance protein Qnr Qnr fluoroquinolone resistance protein fluoroquinolone resistance protein hypothetical protein A6763_18995 hypothetical protein BK371_25690 hypothetical protein BK402_24240 fluoroquinolone resistance protein fluoroquinolone resistance protein

ALG87834.1 ALG87966.1 ALG87998.1 ALG88021.1 KTI69733.1

AJS09375.1 AJT46556.1 KMT66915.1

ALU64991.1 ALU64992.1 OCP43721.1 OCP50900.1 OCW45167.1 OJQ82797.1 OJS43004.1 APK17956.1 APK41752.1

3.2. Ecology of Tet and Erm Genes in the Vicinity of Pig Farms In our initial studies of molecular ecology of antibiotic resistance genes we investigated the environment of two swine farms located in the state of Illinois, USA. The waste lagoons and groundwater were screened for the occurrence and diversity of the tet and erm genes (Chee-Sanford et al. 2001; Koike et al. 2010). Longitudal and quantitative analyses of the tet genes were


17

also performed (Koike et al. 2007). Samples from the waste lagoons understandably demonstrated the highest diversity and concentration of the antibiotic resistance genes tested. In groundwater samples from the vicinity of the pig farms the most frequently encountered tetracycline resistance genes were tet(M), tet(O), tet(Q), tet(W), tet(C), tet(H), and tet(Z). The macrolide resistance genes erm(B), erm(C), erm(G), erm(F) and erm(Q) were also frequently detected in the same groundwater samples mainly originating from the wells impacted by the waste lagoons (Koike et al. 2010). Most of the groundwater samples positive for the antibiotic resistance genes were collected from the wells located down gradient of the waste lagoons. These wells were situated in the shallow sand layer and located near waste lagoon. On the other hand, wells located up gradient or at a greater distance down gradient and wells installed deeper in the ground showed lower diversities and concentrations of antibiotic resistance genes. There was a positive correlation between the occurrence and diversity of antibiotic resistance genes and chemical contamination (chloride, ammonia and electrical conductivity) in groundwater samples. These findings strongly suggest that the seepage from waste lagoons containing residual antibiotics and antibiotic resistant bacteria may contribute to dissemination of antibiotic resistance genes via the groundwater flow. The extent of this “genetic” contamination is affected by a variety of soil characteristics and distance from the source of contamination. Various degrees of the co-occurrence of the tet and erm genes were observed. The strongest pattern of co-occurrence was among the genes encoding ribosomal protection proteins (tet(M), tet(O), tet(Q), and tet(W)) and also among the genes encoding tetracycline efflux pumps (tet(C), tet(H), and tet(Z)). There was a tendency for erm(C) to co-occur with all the tet genes, while erm(B) co-occurred mainly with the tet genes encoding tetracycline efflux pumps. Other erm genes (erm(A), erm(F), erm(G), and erm(Q)), however, did not show strong associations with any other resistance genes. The overall lack of a strong association between the occurrence of the erm and tet genes suggests that these genes are unlikely to be genetically linked and co-selected. Differential persistence of

18


tetracyclines and macrolides in the environment may be responsible for differential selection of these classes of antibiotic resistance in situ. Other factors involved may be the differential host species for antibiotic resistance genes as well as a limited range of hosts for the mobile genetic elements carrying different antibiotic resistance genes. Indeed, as can be seen in part 3 of Results, the range of the main mobile genetic elements involved in the carriage of tetracycline resistance genes is characterized by a rather narrow host range such as phages and prophages.

3.3. Effect of Agricultural Practices on Antibiotic Resistance Genes There is a direct association between the level of antibiotic usage and antibiotic resistance (Goossens et al. 2005), and the use of antibiotics in agriculture certainly contributes to the selection and dissemination of antibiotic resistance genes. What would be a consequence of limiting the exposure of animals to antibiotics? Would this result in a lesser quantity of antibiotic resistance genes released into the environment? To answer this question, we constructed and compared faecal metagenomic libraries from pigs reared in an antibiotic-free environment (organic pigs) and in conventionally reared pigs (the descriptions are given in Materials and Methods). The occurrence of genes conferring resistance to various tetracyclines was initially assessed phenotypically via the screening of metagenomic libraries expressed in an E. coli host against the panel of different tetracycline antibiotics.

Organic Pigs The BAC library from the organic pigs consisted of ca. 9,000 clones, with an average size of inserts (estimated by restriction digest) being ~15 kb (Kazimierczak et al. 2009). Thus we estimated that the library represented approximately 44 bacterial genomes (assuming an average bacterial genome size of 3 Mb). The clones were tested on LB plates supplemented with tetracycline, doxycycline and minocycline at concentrations that completely


19

inhibited the growth of the host strain with an empty vector (5 µg/ml, 3 µg/ml and 3 µg/ml, correspondingly). The screening resulted in ten resistant clones, the majority of which were resistant to all three tetracyclines tested. Clones 9 and 15 exhibited MnRDxR and DxR phenotypes, respectively. Thus the estimates show that tetracycline resistance in this library is encountered approximately in one genome out of four. At the same time, the organic pig metagenomic library, which is equivalent to about 44 gut bacterial genomes sampled, still harbours a sizable number of antibiotic resistance clones and we performed further analyses of these clones by sequencing and transposon mutagenesis to find out what genes and genetic mechanisms may contribute to their maintenance in antibiotic-free animals. Among the ten clones, we encountered tet(C) four times, tet(40) three times, tet(W) two times, and also observed single incidences of two novel resistant determinants, galE-1 and galE-2 (Kazimierczak et al. 2009). The latter two confer resistance to the secondgeneration tetracyclines, minocycline and doxycycline. Since the role of these genes, which encode UDP-glucose 4-epimerases, in conferring resistance to minocycline and doxycycline was not obvious, we performed transposon mutagenesis and subcloning to confirm the function. Interestingly, the antibiotic resistance phenotype was expressed only when cloned into a low-copy number BAC vector suggesting that the high copy numbers can be lethal for the cell. The majority of tetracyline resistant clones carried the components of putative mobile genetic elements, which presumably contribute to the maintenance and dissemination of antibiotic resistance in the antibiotic-free environment. Although no mobility potential was detected in the vicinity of galE-1 and galE-2, the genetic background in the latter gene is quite interesting because it carries multiple antibiotic resistance genes. The most frequently encountered tet(C) gene was found within a sequence that displayed more than 99% nucleotide sequence identity to the well-characterized mobilisable IncQ plasmid pSC101. This may explain the high frequency of the occurrence of the tet(C) gene in the metagenomic library. Moreover, we were able to reconstruct the insert as an autonomously replicating plasmid and perform experiments with the aim of estimating the

20


fitness cost of maintaining such a plasmid in the antibiotic-free environment. For this, we transformed E. coli ATTC® 23734 strain with the reconstituted construct from the metagenomic library and tested the freshly transformed strain against the empty isogenic strain in an antibiotic-free mixed culture for 20 generations. The ratio of the plasmid-free and plasmid-harbouring cells was 86% and 14%, respectively, suggesting that the metabolic cost of carrying this plasmid is substantial. The plasmid-carrying strain was then subjected to selection by tetracycline (5 µg/ml) for 100 generations and the competition experiment was repeated. This time the ratio of plasmid-free and plasmid-carrying isogenic strains was 66% and 44%, respectively, suggesting that even this fairly short selection is sufficient for cells to 'adapt' and carry this plasmid without excessive metabolic burden. The genetic context of the tet(40) gene in one case was associated within a putative transposon and in another - within a putative plasmid. In the former case it was present as a single resistance determinant and in the latter was linked to tet(W). The downstream region of tet(40) possessed a short 160 bp sequence which is highly conserved among all the currently known tet(40) genes and, therefore, may serve as a hot spot for excision and integration of the gene thus contributing to its mobility. Similarly to tet(40), the flanking regions of tet(W) showed sequence conservation among the genes analyzed in this study and by others in databases. The presence of such conserved core sequences may be indicative of their role in horizontal gene transfer. The surrounding genetic context of the gene was also associated with the putative mobile genetic elements.

Conventional Pigs Phenotypically, among the 10,400 BAC clones from the conventional pigs, 98 were resistant to tetracycline, doxycycline and minocycline; 14 were resistant to tetracycline and doxycycline; and 21 were resistant to doxycycline and minocycline. Resistance to a single antibiotic was detected in three (tetracycline), five (minocycline) and 29 (doxycycline) clones. No resistance to the third-generation tetracycline, tigecycline, was encountered in this library. A total of 132 clones in this library were resistant to the first and second generation tetracyclines thus suggesting that the frequency of


21

this phenotype encountered in the gut microbiota of conventional pigs is at least ten-fold higher compared to the organic pigs. With the similar metagenomic coverage, we encountered 10 resistant clones in the organic pig library vs. 132 clones in conventional pigs. There could be certain biases imposed by the heterologous expression in E. coli, but this potential barrier is similar for both libraries and the potential resistance is checked in genetically uniform background. The results suggest that the average load of tetracycline resistance genes per average-sized gut bacterial genome of conventional pigs is about 2.54, e.g., every gut bacterium carries between 2 and 3 tetracycline resistance genes. In organic pigs, the load of tetracycline resistance genes is much lower, about 0.22 genes per genome, thus the majority of bacteria, ca. 78%, are essentially free from tetracycline resistance genes. We conclude the consequences of antibiotic usage (animals which are organically raised vs. those that until recently had received growth promoting antibiotics) has tremendous consequences in terms of the tetracycline resistance gene pool residing in gut bacteria. Different antibiotic usage practices result in more than a ten-fold difference in tetracycline resistance gene concentration and, probably, not only the genes conferring resistance to this particular class of antibiotics. Since the metagenome from conventional pigs is bigger, it may contain taxonomically relevant information, in particular the phylogenetic markers that are used to determine taxonomic affiliation. Thus the metagenome was subjected to the corresponding analysis using the RAST server (Aziz et al. 2008). Interestingly, the tetracycline resistance metagenome, sampled from the conventional pigs, is represented by the genomic fragments originating from bacilli and lactobacilli (Table 2). The reason for this narrow gut microbiota representation in the tetracycline resistance metagenome is not clear. The pig gut microbiota is dominated (>90%) by the representatives of the Bacteroidetes and Firmicutes phyla (Isaacson and Kim, 2012). Only the representatives of the latter contribute to the tetracycline resistome, and it is not clear why the representatives of the former dominant phylum are not detectable in the library. We suspect that the heterologous expression barriers in the E. coli host may have precluded successful expression of antibiotic resistance genes derived from the representatives of this phylum.

22


The BAC system used is designed to clone large DNA fragments and thus the expression of the genes cloned is driven almost exclusively by the native regulatory sequences, which may be not functioning properly in the heterologous host. Table 2. Closest neighbours to the tet resistome of conventional pigs Score 544 520 506 460 427 401 399 389 389 365 357 351 343 342 339 329 328 324 319 299 298 292 283 282 278 269 267 264 263 262

Genome Name Lactobacillus amylovorus GRL 1112 Lactobacillus acidophilus 30SC Lactobacillus ultunensis DSM 16047 Bacillus thuringiensis serovar berliner ATCC 10792 Bacillus thuringiensis serovar thuringiensis str. T01001 Lactobacillus acidophilus ATCC 4796 Bacillus cereus AH1134 Lactobacillus acidophilus NCFM Lactobacillus crispatus 125-2-CHN Lactobacillus crispatus ST1 Bacillus cereus F65185 Lactobacillus crispatus MV-1A-US Lactobacillus crispatus MV-3A-US Lactobacillus amylovorus GRL1118 Lactobacillus acidophilus NCFM Bacillus thuringiensis serovar finitimus YBT-020 Lactobacillus helveticus H10 Lactobacillus crispatus JV-V01 Bacillus thuringiensis serovar kurstaki str. T03a001 Bacillus cereus Rock4-2 Bacillus thuringiensis Bt407 Bacillus cereus ATCC 14579 Bacillus thuringiensis serovar huazhongensis BGSC 4BD1 Lactobacillus helveticus DPC 4571 Bacillus cereus 172560W Lactobacillus gasseri ATCC 33323 Bacillus cereus B4264 Bacillus cereus ATCC 10876 Bacillus thuringiensis serovar sotto str. T04001 Lactobacillus helveticus DSM 20075


23

We further investigated the functional component of the metagenome. Interestingly, the largest component of the metagenome is represented by mobile genetic elements, in particular, by phages and prophages (Figure 1). Other genes participating in horizontal gene transfer include the component of DNA uptake and competence such as encoding DNA Topoisomerase III and also dnaX, hyp2, recR, and hyp3. The overrepresentation of mobile genetic elements in the metagenome suggests the genetic linkage between the mobilome and resistome. Indeed, careful inspection of ORFs encoded in the contigs allowed seeing these close physical associations.

Figure 1. Subsystem Category Distribution in the faecal metagenome of conventional pigs according to the RAST classification (Aziz et al. 2008). The largest Subsystem Feature Counts belong to Phages, Prophages, Transposable elements, and Plasmids; Carbohydrates; and DNA Metabolism.

Surprisingly, we found a very small number of hits with the known tetracycline resistance genes, only six hits, with another five hits revealing multidrug resistance efflux pumps, which may provide non-specific efflux of tetracyclines from the cell. At the same time, the conventional pig gut metagenome revealed three hits with UDP-glucose 4-epimerase. Previously we have shown that the genes encoding these enzymes confer resistance to the second-generation tetracyclines (Kazimierczak et al. 2009). The genes

24


do not display any similarity to the known tetracycline resistance genes but share 65–72% aa identity with UGEs and, correspondingly, possess the conserved GalE domains characteristic for these proteins (COG1087, Fry et al. 2000; Thoden et al. 1997). The metabolic function of UGE is the interconversion of UDP-glucose and UDP-galactose, and the latter is used as a monomer in the biosynthesis of extracellular lipopolysaccharides and capsular polysaccharides. Interestingly, screening a sludge metagenomic library for resistance to menadione resulted in the cloning of the similar gene (Mori et al. 2008). Structurally, menadione is very different from minocycline and doxycycline, and it is unlikely that the resistance mechanism to such structurally different compounds may involve a similar enzymatic modification mechanism. Besides, using LC-MS/MS, we were unable to demonstrate any modification of doxycycline upon incubation with the recombinant E. coli cells expressing galE-1 or galE-2. We hypothesise therefore that, as suggested for menadione resistance (Mori et al. 2008), resistance to the second generation tetracyclines conferred by the galE genes may be due the enhanced permeability barrier. Since only few genetically proven mechanisms of tetracycline resistance are detected in the otherwise phenotypically tetracycline resistant clone library, the diversity of mechanisms conferring tetracycline resistance may be underestimated. Further work is necessary to reveal these resistance mechanisms. Several important conclusions can be drawn from the pig fecal metagenome work reported here. First, easing the pressure of antibiotic selection indeed results in a substantial decrease of the antibiotic resistance gene pool in the gut. Moreover, this potentially would result in a lesser release of antibiotics and antibiotic resistance genes into the environment. Second, despite the fact that the tetracycline resistance mechanisms are among the best-studied ones, there may be still other unknown mechanisms involved in tetracyline resistance, which need to be investigated. Third, perhaps not surprisingly, the antibiotic resistome is strongly associated with the mobilome, i.e., mobile genetic elements. They are responsible for the rapid dissemination of antibiotic resistance genes among and into a variety of ecological compartments (Aminov, 2011; Aminov, 2012).


25

CONCLUSION In this chapter, we reviewed our previous work that encompassed the evolution and ecology of antibiotic resistance and how the agricultural practices may affect the ecology and microevolution of antibiotic resistance genes. Certainly, besides the use in animal industry, there is, of course, the use of antibiotics for treatment of human infectious diseases; this is the primary role of antibiotics. The human antibiotic consumption and the use of last-resort antibiotics are growing rapidly (Van Boeckel et al. 2014). There are certainly problems with the improper use of antibiotics in humans such as over-the-counter availability in many countries, inappropriate use (for example, for treatment of viral respiratory infections), poor compliance, and others. But the focus of this chapter is on the use of antimicrobials in food animals, the human consumption of antibiotics is discussed in other chapters of this book. However, what is evident from the biological point of view is that the broader the selection imposed by antibiotics (extended to the combined human and animal microbiota), the chances are higher are that mutational or acquired antibiotic resistances would arise among the commensal and, subsequently, pathogenic microbiota (Aminov and Mackie, 2007; Aminov, 2009; Aminov, 2013). Lessons from the history of antibiotic use teach us that the emergence of antimicrobial resistance is just a question of time and the antimicrobial use practices involved. Currently, no exception to the scenario of the eventual appearance of antibiotic resistance is known (Aminov, 2010). But, evidently, later is better, so that antibiotics may have a more useful and effective lifespan. Thus, these extremely worrying antibiotic consumption trends, especially in the areas that are not vital for the containment of human morbidity and mortality imposed by bacterial pathogens, dictate the urgent need for the global and coordinated action to prevent the overuse of antibiotics and the associated rise of hard-to-treat antibiotic-resistant pathogens (Aminov, 2016). As mentioned before, the global use of antimicrobials in food animals is predicted to grow rapidly and about a third of this increase will be due to the switch from traditional agricultural practices of raising food animals to

26


large-scale intensive farming operations that routinely use antimicrobials at sub-therapeutic concentrations (Van Boeckel et al. 2015). It is understandable that there are certain economic drivers behind the strategy of development adopted by the animal food production industries globally. The economic and other costs, however, are presently considered only within the animal industry borders while the consequences in the form of antimicrobial resistance have much broader implications. We pose the question, should these considerations also include the costs that are eventually paid by society for the development of new antibiotics, for more extended and expensive treatment of antibiotic resistant infections, and for the higher morbidity and mortality rates imposed by multidrug-resistant infections?

ACKNOWLEDGMENTS This was supported, in the metagenomics part, by the DEFRA grant OD2014. The authors also would like to thank Dr. Katarzyna A. Kazimierczak for metagenomic library construction and Dr. A. J. Travis for bioinformatics support.

REFERENCES Aínsa JA, Blokpoel MC, Otal I, Young DB, De Smet KA, Martín C. 1998. Molecular cloning and characterization of Tap, a putative multidrug efflux pump present in Mycobacterium fortuitum and Mycobacterium tuberculosis. J Bacteriol. 180:5836-43. Aminov, R. I., N. Garrigues-Jeanjean, and R. I. Mackie. 2001. Molecular ecology of tetracycline resistance: development and validation of primers for detection of tetracycline resistance genes encoding ribosomal protection proteins. Appl. Environ. Microbiol. 67:22-32. Aminov R. I., J. C. Chee-Sanford, N. Garrigues, B. Teferedegne, I. J. Krapac, B. A. White, and R. I. Mackie. 2002. Development, validation, and application of primers for detection of tetracycline resistance genes


27

encoding tetracycline efflux pumps in gram-negative bacteria. Appl. Environ. Microbiol. 68:1786-1793. Aminov, R. I., J. C. Chee-Sanford, N. Garrigues, A. Mehboob, and R.I. Mackie. 2004. Detection of tetracycline resistance genes by PCR methods. Methods in Molecular Biology, v. 268, p. 3-14. Humana Press. Aminov, R. I. and Mackie, R. I. 2007. Evolution and ecology of antibiotic resistance genes FEMS Microbiol Lett. 271:147-161. DOI: 10.1111/j.1574-6968.2007.00757.x. Aminov, R.I. 2009. The role of antibiotics and antibiotic resistance in nature. Environ. Microbiol. 11: 2970-2988. DOI: 10.1111/j.14622920.2009.01972.x. Aminov, R. I. 2010. A brief history of the antibiotic era: lessons learned and challenges for the future. Front. Microbiol. 1:134. DOI: 10.3389/fmicb.2010.00134. Aminov, R. I. 2011. Horizontal gene exchange in environmental microbiota. Front. Microbiol. 2:158. doi: 10.3389/fmicb.2011.00158. Aminov R. I. 2012. The extent and regulation of lateral gene transfer in natural microbial ecosystems. In: Horizontal Gene Transfer in Microorganisms (Ed: M.P. Francino), Chapter 6. Horizon Scientific Press, Norwich NR9 3DB, UK. Aminov RI. 2013. Evolution in action: dissemination of tet(X) into pathogenic microbiota. Front Microbiol. 4:192. DOI: 10.3389/fmicb. 2013.00192. Aminov RI. 2016. History of antimicrobial drug discovery - Major classes and health impact. Biochem Pharmacol. 2016 Oct 5. pii: S00062952(16)30331-8. doi: 10.1016/j.bcp.2016.10.001. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass EM, Kubal M, Meyer F, Olsen GJ, Olson R, Osterman AL, Overbeek RA, McNeil LK, Paarmann D, Paczian T, Parrello B, Pusch GD, Reich C, Stevens R, Vassieva O, Vonstein V, Wilke A, Zagnitko O. 2008. The RAST Server: Rapid Annotations using Subsystems Technology. BMC Genomics 9:75. Bönemann G, Stiens M, Puhler A, Schluter A. 2006. Mobilizable IncQrelated plasmid carrying a new quinolone resistance gene, qnrS2,

28


isolated from the bacterial community of a wastewater treatment plant. Antimicrob Agents Chemother. 50:3075–80. Benson D. A., M. S. Boguski, D. J. Lipman, J. Ostell, B. F. Ouellette, B. A. Rapp, and D. L. Wheeler. 1999. GenBank. Nucleic Acids Res. 27:12-17. Castanon JI. 2007. History of the use of antibiotic as growth promoters in European poultry feeds. Poult Sci 86:2466–2471. 10.3382/ps.200700249. Chee-Sanford, J.C., Aminov, R.I., Krapac, I.J., Garrigues-Jeanjean, N., and Mackie, R.I. 2001. Occurrence and diversity of tetracycline resistance genes in lagoons and groundwater underlying two swine production facilities. Appl. Environ. Microbiol. 67: 1494-1502. Chopra, I. and M. Roberts. 2001. Tetracycline antibiotics: mode of action, applications, molecular biology, and epidemiology of bacterial resistance. Microbiol. Mol. Biol. Rev. 65:232-260. De Rossi E, Blokpoel MC, Cantoni R, Branzoni M, Riccardi G, Young DB, De Smet KA, Ciferri O. 1998. Molecular cloning and functional analysis of a novel tetracycline resistance determinant, tet(V), from Mycobacterium smegmatis. Antimicrob Agents Chemother. 42(8): 19317. Duggar, B. M. 1948. Aureomycin: a product of the continuing search for new antibiotics. Ann N Y Acad Sci. 51: 177–181. DOI: 10.1111/j.17496632.1948.tb27262.x. EPC. 2005. Ban on Antibiotics as Growth Promoters in Animal Feed Enters into Effect, European Commission – IP/05/1687. Available at: http://europa.eu/rapid/press-release_IP-05-1687_en.htm. FDA. 2015. Summary Report on Antimicrobials Sold or Distributed for Use in Food-Producing Animals. Available at: https://docs.google. com/viewer?url=http%3A%2F%2Fwww.fda.gov%2Fdownloads%2FF orIndustry%2FUserFees%2FAnimalDrugUserFeeActADUFA%2FUC M440584.pdf. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution. 39:783-791. Felsenstein, J. 1989. PHYLIP - Phylogeny Inference Package (Version 3.2). Cladistics 5: 164-166.


29

Fry, B. N., S. Feng, Y. Y. Chen, D. G. Newell, P. J. Coloe, and V. Korolik. 2000. The galE gene of Campylobacter jejuni is involved in lipopolysaccharide synthesis and virulence. Infect. Immun. 68:25942601. Goossens H, Ferech M, Vander Stichele R, Elseviers M; ESAC Project Group. 2005. Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet. 365:579-87. DOI: 10.1016/S0140-6736(05)17907-0. Hasman H, Aarestrup FM, Dalsgaard A and Guardabassi L. 2006. Heterologous expression of glycopeptide resistance vanHAX gene clusters from soil bacteria in Enterococcus faecalis. J Antimicrob Chemother 57: 648–653. Hegde SS, Vetting MW, Roderick SL, Mitchenall LA, Maxwell A, Takiff HE and Blanchard JS. 2005. A fluoroquinolone resistance protein from Mycobacterium tuberculosis that mimics DNA. Science 308: 14801483. Hiasa H and Shea ME. 2000. DNA gyrase-mediated wrapping of the DNA strand is required for the replication fork arrest by the DNA gyrasequinolone-DNA ternary complex. J Biol Chem 275: 34780-34786. DOI: 10.1074/jbc.M001608200. Huelsenbeck, J.P., and Ronquist, F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755. Isaacson R, Kim HB. 2012. The intestinal microbiome of the pig. Anim Health Res Rev. 13(1):100-9. doi: 10.1017/S1466252312000084. Kazimierczak, K. A., K. P. Scott, D. Kelly and R. I. Aminov. 2009. Tetracycline resistome of the organic pig gut. Appl. Environ. Microbiol. 75:1717-22. Kimura M. 1980. A simple model for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16:111-120. Koike S, Krapac IG, Oliver HD, Yannarell AC, Chee-Sanford JC, Aminov RI, and Mackie RI. 2007. Monitoring and source tracking of tetracycline resistance genes in lagoons and groundwater adjacent to swine

30


production facilities over a three-year period. Appl. Environ. Microbiol. 73: 4813-4823. Koike, S., R.I. Aminov, A.C. Yannarell, H. Gans, I.G. Krapac, J.C. CheeSanford, and R.I. Mackie. 2010. Molecular ecology of macrolidelincosamide-streptogramin B methylases in waste lagoons and subsurface waters associated with swine production. Microb Ecol. 59:487-98. Krishnasamy V, Otte J, Silbergeld E. 2015. Antimicrobial use in Chinese swine and broiler poultry production. Antimicrob Resist Infect Control. 4:17. DOI: 10.1186/s13756-015-0050-y. Kyselková M, Chroňáková A, Volná L, Nĕmec J, Ulmann V, Scharfen J, Elhottová D. 2012. Tetracycline resistance and presence of tetracycline resistance determinants tet(V) and tap in rapidly growing mycobacteria from agricultural soils and clinical isolates. Microbes Environ. 27(4):413-22. Landers TF, Cohen B, Wittum TE, Larson EL. 2012. A Review of Antibiotic Use in Food Animals: Perspective, Policy, and Potential. Public Health Rep. 127(1): 4-22. Larson, T.H., Krapac, I.G., Dey, W.S., and Suchomski, C.J. 1997. Electromagnetic terrain conductivity surveys used to screen swine confinement facilities for groundwater contamination. In Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems. Reno, NV, pp. 271-279. Leclercq R. 2002. Mechanisms of resistance to macrolides and lincosamides: nature of the resistance elements and their clinical implications. Clin Infect Dis. 34(4): 482-92. DOI: 10.1086/324626. Mackie, R.I., Koike, S., Krapac, I., Chee-Sanford, J., Maxwell, S., and Aminov, R.I. 2006. Tetracycline residues and tetracycline resistance genes in groundwater impacted by swine production facilities. Animal Biotechnology 17: 157-176. Montero C, Mateu G, Rodriguez R, and Takiff H. 2001. Intrinsic resistance of Mycobacterium smegmatis to fluoroquinolones may be influenced by new pentapeptide protein MfpA. Antimicrob Agents Chemother 45: 3387-3392.


31

Mori, T., H. Suenaga, and K. Miyazaki. 2008. A metagenomic approach to the identification of UDP-glucose 4-epimerase as a menadione resistance protein. Biosci. Biotechnol. Biochem. 72:1611-1614. Pang Y, Brown BA, Steingrube VA, Wallace RJ, Jr, Roberts MC. 1994. Tetracycline resistance determinants in Mycobacterium and Streptomyces species. Antimicrob Agents Chemother. 38:1408–1412. Poirel L, Liard A, Rodriguez-Martinez JM and Nordmann P. 2005. Vibrionaceae as a possible source of Qnr-like quinolone resistance determinants. J Antimicrob Chemother 56: 1118–1121. Roberts M.C. 1996. Tetracycline resistance determinants: mechanisms of action, regulation of expression, genetic mobility, and distribution. FEMS Microbiol. Rev. 19:1-24. Roberts, M. C. 2005. Update on acquired tetracycline resistance genes. FEMS Microbiol. Lett. 245:195-203. Rodriguez-Mozaz S, Chamorro S, Marti E, Huerta B, Gros M, SànchezMelsió A, Borrego CM, Barceló D, Balcázar JL. 2015. Occurrence of antibiotics and antibiotic resistance genes in hospital and urban wastewaters and their impact on the receiving river. Water Res. 69:23442. doi: 10.1016/j.watres.2014.11.021. Ronquist, F., and Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 15721574. Rossi-Fedele G, Scott W, Spratt D, Gulabivala K, Roberts AP. 2006. Incidence and behaviour of Tn916-like elements within tetracyclineresistant bacteria isolated from root canals. Oral Microbiol Immunol. 21:218–222. Ruiz J, Pons MJ, Gomes C. 2012. Transferable mechanisms of quinolone resistance. Int J Antimicrob Agents. 40(3):196-203. doi: 10.1016/ j.ijantimicag.2012.02.011. Saga T, Kaku M, Onodera Y, Yamachika S, Sato K, Takase H. 2005. Vibrio parahaemolyticus chromosomal qnr homologue VPA0095: Demonstration by transformation with a mutated gene of its potential to reduce quinolone susceptibility in Escherichia coli. Antimicrob Agents Chemother 49: 2144-2145.

32


Saitou N., and M. Nei. 1987. The neighbour-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425. Speer BS, Bedzyk L, Salyers AA. 1991. Evidence that a novel tetracycline resistance gene found on two Bacteroides transposons encodes an NADP-requiring oxidoreductase. J Bacteriol. 173(1):176-83. Thoden, J. B., A. D. Hegeman, G. Wesenberg, M. C. Chapeau, P. A. Frey, and H. M. Holden. 1997. Structural analysis of UDP-sugar binding to UDP-galactose 4-epimerase from Escherichia coli. Biochemistry 36:6294-6304. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., and Higgins, D.G. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25: 4876-4882. Tran JH and Jacoby GA. 2002. Mechanism of plasmid-mediated quinolone resistance. Proc Natl Acad Sci USA. 99: 5638-5642. DOI: 10.1073/pnas.082092899. Tran JH, Jacoby GA and Hooper DC. 2005. Interaction of the plasmidencoded quinolone resistance protein Qnr with Escherichia coli DNA gyrase. Antimicrob Agents Chemother. 49: 118-125. DOI: 10.1128/AAC.49.1.118-125.2005. Van Boeckel TP, Gandra S, Ashok A, Caudron Q, Grenfell BT, Levin SA, Laxminarayan R. 2014. Global antibiotic consumption 2000 to 2010: an analysis of national pharmaceutical sales data. Lancet Infect Dis. 14(8): 742-50. DOI: 10.1016/S1473-3099(14)70780-7. Van Boeckel TP, Brower C, Gilbert M, Grenfell BT, Levin SA, Robinson TP, Teillant A, Laxminarayan R. 2015. Global trends in antimicrobial use in food animals. Proc Natl Acad Sci U S A. 112(18): 5649-5654. DOI:10.1073/pnas.1503141112. Weisblum B. 1995. Erythromycin resistance by ribosome modification. Antimicrob Agents Chemother. 39:577-85. DOI: 10.1128/AAC. 39.3.577. Wen Y, Pu X, Zheng W, Hu G. 2016. High Prevalence of Plasmid-Mediated Quinolone Resistance and IncQ Plasmids Carrying qnrS2 Gene in


33

Bacteria from Rivers near Hospitals and Aquaculture in China. PLoS One. 11(7):e0159418. doi: 10.1371/journal.pone.0159418. Wentland MP. 1993. In memoriam: George Y. Lesher, Ph.D., in Hooper DC, Wolfson JS (eds): Quinolone antimicrobial agents, ed 2., Washington DC, American Society for Microbiology: XIII - XIV. Wohlkonig A., Chan P. F., Fosberry A. P., Homes P., Huang J., Kranz M., Leydon V. R., Miles T. J., Pearson N. D., Perera R. L., Shillings A. J., Gwynn M. N., Bax B. D. 2010. Structural basis of quinolone inhibition of type IIA topoisomerases and target-mediated resistance. Nat. Struct. Mol. Biol. 17: 1152–1153.