Antibiotic and Metal Resistance in Escherichia coli Isolated from Pig Slaughterhouses in the United Kingdom

Antimicrobial resistance is currently an important concern, but there are few data on the co-presence of metal and antibiotic resistance in potentially pathogenic Escherichia coli entering the food chain from pork, which may threaten human health. We have examined the phenotypic and genotypic resistances to 18 antibiotics and 3 metals (mercury, silver, and copper) of E. coli from pig slaughterhouses in the United Kingdom. The results showed resistances to oxytetracycline, streptomycin, sulphonamide, ampicillin, chloramphenicol, trimethoprim–sulfamethoxazole, ceftiofur, amoxicillin–clavulanic acid, aztreonam, and nitrofurantoin. The top three resistances were oxytetracycline (64%), streptomycin (28%), and sulphonamide (16%). Two strains were resistant to six kinds of antibiotics. Three carried the blaTEM gene. Fifteen strains (18.75%) were resistant to 25 µg/mL mercury and five (6.25%) of these to 50 µg/mL; merA and merC genes were detected in 14 strains. Thirty-five strains (43.75%) showed resistance to silver, with 19 possessing silA, silB, and silE genes. Fifty-five strains (68.75%) were resistant to 8 mM copper or above. Seven contained the pcoE gene. Some strains were multi-resistant to antibiotics, silver, and copper. The results in this study, based on strains isolated between 2007 and 2010, will aid understanding about the effects of strategies to reduce resistance and mechanisms of antimicrobial resistance (AMR).


Introduction
The discovery and introduction of antibiotics prior to the Second World War marked a pivotal point in the ability of humanity to combat bacterial infections [1]. For over half a century, antibiotics have been widely used in the animal production industry. They have been used not only for disease prevention and treatment, but also for promotion of growth [2]. It is estimated that half of the total antibiotics produced are used for veterinary purposes [3]. The overuse and misuse of antibiotics stimulated the more rapid emergence and spread of antibiotic-resistant bacteria (ARB) and antibiotic resistant genes (ARGs), reducing their therapeutic potential against human and animal pathogens [4,5]. The occurrence of multidrug-resistant microorganisms has brought emergent threats to human health.
Metals and metalloids have a long history of human usage in medicine and agriculture. Some toxic metal(loid) compounds, such as mercury and arsenic/antimony compounds, were still used as first-line drugs or preferred-choice chemotherapeutics or antimicrobials in the late twentieth century [6]. Some metals, such as copper and zinc, are still used in animal husbandry as antimicrobials and as essential micro-nutrients and growth stimulants [7][8][9]. However, the actual concentration of metals
As phenotypic resistance was shown to some β-lactamases by seven strains, the presence of the antibiotic resistance genes, blaSHV, blaTEM, blaOXA-1, and blaCTX-M, was examined by specific polymerase chain reaction (PCR). The results specifically showed that three strains, No. 4, No. 7, and No. 61, carried the blaTEM gene (Table 2). According to the phenotypic results for antibiotic resistances, these three strains were the top three resistant strains, which were resistant to six, six, and five antibiotics,   6  2  2  AMP *, S10, OT, SXT, S300, C   5  1  1  S10, OT, SXT, S300, C   4  2  1  OT, SXT, S300, C  1  OT, F, S300, C   3  4  1  AMP, S10, EFT  1  AMP, S10, OT  1  S10, OT, S300  1  AMP, ATM, OT   2  14  9  S10, OT  4 OT, S300 1 S10, S300 As phenotypic resistance was shown to some β-lactamases by seven strains, the presence of the antibiotic resistance genes, bla SHV , bla TEM , bla OXA-1 , and bla CTX-M , was examined by specific polymerase chain reaction (PCR). The results specifically showed that three strains, No. 4, No. 7, and No. 61, carried the bla TEM gene (Table 2). According to the phenotypic results for antibiotic resistances, these three strains were the top three resistant strains, which were resistant to six, six, and five antibiotics, respectively. All three strains were resistant to AMP, but only strain 61 showed resistance to AMC, which can be associated with carriage of a plasmid borne bla TEM [42], and none showed typical extended spectrum β-lactamase (ESBL) resistance. Strains were also examined for the presence of mcr-1. No strains were detected carrying the mcr-1 gene, confirming the phenotypic result of colistin sulphate sensitivity.

Resistance to Mercury
Various methods for testing for metal resistance exist in the literature. The levels at which toxicity is shown is influenced by many factors, particularly medium composition. Therefore, in this study, known resistant and sensitive strains were included for comparison and the levels at which strains grew or did not grow were compared with these to determine if the test strains were sensitive or resistant.
Phenotypic mercury resistance was tested by growth of strains on media with different concentrations of HgCl 2 . The level of HgCl 2 in Luria-Bertani (LB) agar that completely inhibits the growth of the E. coli sensitive control strain is 25 µg/mL. For the tested strains, 15 (18.75%) were resistant to 25 µg/mL mercury and 5 (6.25%) were resistant to 50 µg/mL mercury ( Figure 3a). Especially for strain 38, resistance was shown as same as the positive control (50 µg/mL) (E. coli J53 pMG101).
Antibiotics 2020, 9, x FOR PEER REVIEW 5 of 20 respectively. All three strains were resistant to AMP, but only strain 61 showed resistance to AMC, which can be associated with carriage of a plasmid borne blaTEM [42], and none showed typical extended spectrum β-lactamase (ESBL) resistance. Strains were also examined for the presence of mcr-1. No strains were detected carrying the mcr-1 gene, confirming the phenotypic result of colistin sulphate sensitivity.

Resistance to Mercury
Various methods for testing for metal resistance exist in the literature. The levels at which toxicity is shown is influenced by many factors, particularly medium composition. Therefore, in this study, known resistant and sensitive strains were included for comparison and the levels at which strains grew or did not grow were compared with these to determine if the test strains were sensitive or resistant.
Phenotypic mercury resistance was tested by growth of strains on media with different concentrations of HgCl2. The level of HgCl2 in Luria-Bertani (LB) agar that completely inhibits the growth of the E. coli sensitive control strain is 25 µg/mL. For the tested strains, 15 (18.75%) were resistant to 25 µg/mL mercury and 5 (6.25%) were resistant to 50 µg/mL mercury (Figure 3a). Especially for strain 38, resistance was shown as same as the positive control (50 µg/mL) (E. coli J53 pMG101). The specific primers of merA, merC, and intl1 genes were used to look for genetic elements associated with mercury resistance and carriage of Tn21-like transposable elements carrying an integron. The merA and merC genes were detected in 14 of the 80 (17.5%) strains and intl1 was detected in 4 strains (5%). Two strains, 10 and 14, contained both merA, merC, and the intl1 genes ( Table 2), indictive of carrying a transposon. Combined with phenotypic resistance results, these 14 strains containing merA and merC genes all showed resistance to 25 μg/mL mercury. Strain 38, containing merA, merC, and astA, was even resistant to 50 μg/mL.

Resistance to Silver
LB agar without salt with different concentrations of AgNO3 was used for detecting phenotypic silver resistance. The results showed all strains could grow at 300 µM AgNO3. At 400 µM AgNO3, E. coli sensitive control strain was completely inhibited. Of the tested strains, 35 strains (43.75%) showed silver resistance to 400 µM AgNO3 (Figure 3b). All strains were sensitive to 500 µM AgNO3. Of the 35 phenotypically resistant strains, 19 strains possessed silA, silB, and silE genes ( Table 2).

Resistance to Copper
In this study, E. coli ED8739 and E. coli ED8739 pRJ1004 were used as negative and positive controls, respectively. The results showed that E. coli ED8739 could grow at 6 mM CuSO4 or below, The specific primers of merA, merC, and intl1 genes were used to look for genetic elements associated with mercury resistance and carriage of Tn21-like transposable elements carrying an integron. The merA and merC genes were detected in 14 of the 80 (17.5%) strains and intl1 was detected in 4 strains (5%). Two strains, 10 and 14, contained both merA, merC, and the intl1 genes ( Table 2), indictive of carrying a transposon. Combined with phenotypic resistance results, these 14 strains containing merA and merC genes all showed resistance to 25 µg/mL mercury. Strain 38, containing merA, merC, and astA, was even resistant to 50 µg/mL.

Resistance to Copper
In this study, E. coli ED8739 and E. coli ED8739 pRJ1004 were used as negative and positive controls, respectively. The results showed that E. coli ED8739 could grow at 6 mM CuSO 4 or below, but could not grow at 7 mM copper. However, E. coli ED8739 pRJ1004 expressed extremely strong resistance to copper, being resistant to 20 mM CuSO 4 . Compared with E. coli ED8739, most strains showed resistance to only 7 mM copper ( Figure 3c). Fifty five strains (68.75%) were resistant to 8 mM copper, twenty nine (36.25%) strains were resistant to 9 mM copper, and two strains were resistant to 10 mM copper. Over 10 mM, no strains showed resistance.
At the same time, we examined all strains for carriage of pcoE. The pcoE gene was detected in seven strains (Table 2). Two strains (No. 50 and No. 77) with the pcoE genes showed phenotypic resistance to 8 mM and 9 mM copper, while one strain (No. 78) showed resistance to 10 mM ( Table 2). The other four strains (No. 14, No. 63,No. 74,and No. 75) containing the pcoE gene did not show resistance to 8 mM copper, only to 7 mM CuSO 4 , slightly above the basal copper tolerance of laboratory E. coli strains ( Table 2 and Table S1). Most other strains phenotypically copper resistant at 8 mM did not contain a pcoE gene (Table S1).

Discussion
Antimicrobial resistance (AMR) has been characterized as a global public health crisis that must be managed with the utmost urgency [43]. The evolution of AMR is a complex, multifactorial issue [44]. The lack of understanding for the potential dangers of inappropriate use, as well as the widespread access to antimicrobials, have led to the current AMR crisis [45]. Several strategies, such as antimicrobial stewardship programs, preventing the misuse of antibiotics, and promoting the discovery of new antimicrobial agents, have been developed in recent years to slow this progression [44]. Antimicrobial resistance surveillance at different periods will aid people to know about the effects of these strategies and pathologic mechanisms of AMR. To the best of our knowledge, this is the first study to examine Antibiotics 2020, 9, 0746 8 of 19 the prevalence of the co-occurrence of both antibiotic resistance and metal resistance in E. coli from slaughterhouses in the United Kingdom.
In this study, we focused on AMR of E. coli strains isolated from pig slaughterhouses in 2007, 2009, and 2010. We used the strain stocks stored on Microbank TM beads at −80 • C. This method is often used for long-term storage of microorganisms [36,46,47]. In this study, of the 81 strains, which were stored in Microbank TM in 2009 and 2010, 80 live strains were recovered, giving a recovery rate of 98.8%, which indicated that Microbank TM could be used to store E. coli for at least seven years. However, during storage, some characteristics appeared to have changed. For example, when originally isolated, all strains were detected with the plasmid borne astA gene, which is found among certain categories of diarrhoeagenic E. coli [48]. After storage, only 70 strains were detected carrying the astA gene, suggesting plasmid loss on recovery. Three strains were indole negative, although originally characterised as indole positive. These results showed that long-term storage may lead to the instability of some E. coli strain characteristics.
From phenotypic results, 78% of analysed strains were resistant to between one and six antibiotics. Multi-drug resistance prevalence was 28.8%, which was similar to reports from other places at the same time. For instance, Sheikh et al. from Canada (with pork, beef, poultry, and turkey samples) and Llorente et al. from Buenos Aires (with beef samples) reported a multi-drug resistance prevalence of 28.1% and 27.8%, respectively [49,50]. From the Tadesse et al. review from 1950 to 2002 (of human and food samples from beef, pork, poultry), a prevalence of multi-drug resistance of 59.1% in beef and 53.7% in pork was found [51]. The results in the current study showed that the multi-drug resistance prevalence in pigs in the United Kingdom between 2007 and 2010 was very low. In contrast, recent investigations in Tamaulipas, Mexico showed that the prevalence of antimicrobial resistance amounted to 92% in retail samples of beef and pork [30].
Tetracyclines, as a group of antibiotics, are a cause for special attention because large quantities are applied in the therapy in human and animal infections as well as in some countries in animal feed as growth promoters. However, only a small amount of tetracyclines can be metabolized or absorbed by humans and animals. Residues of these antibiotics are accumulated in the environment and may lead to the occurrence of resistant species [52]. Oxytetracycline (OT), belonging to the tetracycline group of antibiotics, is a broad-spectrum antibiotic, active against a wide variety of bacteria [53]. In this study, of all resistances, OT resistance was predominant (n = 51; 64%), with some strains showing only tetracycline resistance, but others showing multi-drug resistance ( Figure 1 and Table 1). Similar results were reported in ready-to-eat meat (66.7%) [54], pork (43.8%) [55], beef (74%) [56], and chicken (97%) [57]. The AMR prevalence in Enterobacteriaceae isolates from slaughtered pigs has been shown to be 31.8% in Salmonella in 2004-2005 in Ethiopia [58] and 56% in 2013-2014 in Italy [59]; 22% of E. coli in 2007 in Denmark [60]; and 78.2% in Yersinia enterocolitica in 2013-2014 in Italy [61]. In this current study, 64% of all E. coli strains showing resistance was higher than that in Denmark in 2007. In another report, for Enterococcus faecium and E. faecalis isolated from pigs, the resistance level of 82% in the United Kingdom was higher than the average level (67.4%) in the European Union (EU) in 2013-2014 [62]; the reason for this was unclear. A recent report on resistance in wild pigs showed that, of 115 E. coli strains, none were resistant to OT. The abundant prevalence of tetracycline resistance represents a useful marker to monitor antimicrobial-resistant bacteria [63,64]. These results of OT resistance may represent a genetic archaeology of the use of veterinary antimicrobials [38]. Streptomycin (S10) (n = 26; 33%) and sulfonamide (S300) (n = 13; 16%) were the other two antibiotics to which a higher percentage of all strains were resistant in this study. This result was similar to previous reports [55,65,66].
For the genotypic detection of β-lactam resistance, PCR amplification of the bla TEM , bla SHV , and bla CTX-M genes is often used because it is straightforward and cost-effective [68] and, in one report, the bla TEM gene was detected as the most prevalent resistance gene in isolates from poultry, beef, and pork in Czech Republic [55]. In this study, the presence of bla TEM , bla SHV , and bla CTX-M genes was determined. The result showed that the TEM gene existed in three strains ( Table 2). Between 2007 and 2010, a large number of ESBL-containing resistant strains were identified in the EU [70,71]. Different results in our study might be because of differences in the antibiotics used in animal husbandry, which may influence resistance carriage. While some characteristics of strains changed during storage, particularly the loss of the astA gene, which is usually plasmid carried, it could be suggested that resistance plasmids had also been lost. However, the loss of astA was seen in only 12.5% of strains, thus plasmid loss is unlikely to account for the low level of ESBL resistance seen here. The bla TEM gene encodes the production of β-lactamase enzymes that hydrolyze the β-lactam ring and inactivate those β-lactams [72]. Beta-lactamases are the principal mechanism of bacterial resistance to beta-lactam antibiotics. AmpC beta-lactamases and/or extended-spectrum beta-lactamases (ESBLs) are of particular concern for the prevalence of multidrug-resistant Gram-negative bacteria [73]. In E. coli, ESBLs were plasmid-encoded [74], while AmpC beta-lactamase production was chromosomally mediated [73]. Combined with the result in this study, we deduced that the β-lactam resistance seen in the other strains may be conferred by upregulation of the chromosomal ampC gene found in all E. coli [73], which concurs with the results seen in other studies [38].
According to the phenotypic results, two of the three strains (No. 4 and No. 7) containing the TEM gene showed multi-drug resistance to six antibiotics (AMP, S10, OT, S300, SXT, and C). The other one (No. 61) was resistant to five antibiotics (AMP, AMC, OT, S300, and SXT) ( Table 2). However, none of these strains showed typical phenotypic ESBL resistance patterns associated with carriage of the TEM gene. The presence of the bla TEM gene might be associated with the co-occurrence of genes for resistance to AMP, OT, S300, and SXT. At the same time, strains No. 4 and No. 7 contained the intl1 gene (Table 2). According to the report of Fernandez-Alarcon, et al. [75], IncA/C plasmids carrying bla TEM and intl1 and genes conferring resistance to gentamicin, kanamycin, streptomycin, trimethoprim/sulfamethoxazole, sulfoxazole, chloramphenicol, and tetracycline have been described. In this study, whether strains No. 4 and No.7 contain IncA/C plasmids needs to be further researched.
Apart from antibiotics, a similar issue of resistance also exists for metals. Metals also exhibit antimicrobial features. Even when no antimicrobial compounds are used, certain metals can maintain or even increase bacterial resistance against certain agents [76]. In medicine and agriculture, the most commonly used metals have probably been mercury (Hg), copper (Cu), silver (Ag), arsenic (As), and antimony (Sb) [6]; zinc and copper are frequently used as feed supplements for pigs. In this study, we examined E. coli resistance to Hg, Cu, and Ag. For Hg, 15 strains (18.75%) were phenotypically resistant to 25 µg/mL HgCl 2 . Of these, 14 strains (93.3%) contained both merA and merC genes ( Table 2). The mercuric ion reductase (MerA), an enzyme that catalyzes the conversion of the thiol-avid Hg (II) to volatile, uncharged Hg (0) that lacks significant affinity for any liganding functional groups, is the heart of the mercury resistance (HgR) mechanism [77]. MerC is an operon-encoded transmembrane protein, particularly found in Tn21-family transposons, which pumps ionic mercury into the cytosol [78]. MerC may be needed under conditions of very high Hg (II) exposure [79] and is characteristic of Tn21-family transposons. Intl is a gene encoding an integrase, which is a necessary component of an integron, and many class 1 integrons (intl1) carrying antibiotic resistance cassettes are found on Tn21-like transposons [80]. In this study, there were two strains possessing merA, merC, and intl1 genes ( Table 2). The results indicated that a Tn21-like transposon is likely to be present in these two strains.
In this study, to test bacterial silver resistance, silver concentrations of 0~600 µM were set up according to previous research [81]. The results showed that the sensitive strain E. coli J53 can grow up to 300 µM of AgNO 3 , while the resistant strain E. coli J53 with plasmid pMG101 (J53 pMG101) continues to grow up to 600 µM. Gupta et al. reported that E. coli J53 could grow up to 100 µM of AgNO 3 [81]. Although these differences in resistance levels shown between two studies are possibly due to experimental conditions and operators, resistance at 400 µM of AgNO 3 was significantly correlated to the presence of genotypic resistances genes (r = 0.6), and thus could be taken as the silver concentration that denotes phenotypic resistance. The genes for Ag+ resistance are a cluster of seven genes (and two open reading frames of unknown function), which is organized into three divergently transcribed units [82]. SilE is a periplasmic metal-binding protein that has been purified and measured for Ag + -binding properties and is highly expressed when silver ions are present. SilS and SilR are a presumed two-component membrane sensor and transcriptional responder, and SilCBA and SilP are a presumed Ag + efflux chemiosmotic cation-proton antiporter and a P-type ATPase, respectively. In this study, three Ag + resistance genes silA, silB, and silE were examined. The results showed that 23.75% of strains possessed these three genes and expressed phenotypic silver resistance, which may suggest that silver compounds have been widely used as antimicrobials [83]. Of these 23.75% of strains, there were eight strains that also possessed mercury-resistance genes ( Table 2). That means these strains possess multi-metal resistances. In addition, there were another 20% of strains that showed phenotypic resistance without silA, silB, and silE genes.
Copper is an essential metal to aerobic forms of life, being involved in donating or accepting electrons in redox-active enzymes, or in the electron transport chain [6,84]. It is commonly used in the farming of pigs, but the actual concentration of copper in the feed is 3-9 times higher than that of the growth requirements [10]. In Europe, copper sulphate in pig feed is permitted at 125-250 mg/kg feed (equivalent to 0.78-1.56 mM) and has been shown to stimulate the growth rate of piglets and, at these levels, is known to modify the gastrointestinal microbiota composition [85]. In this study, three strains were resistant to more than 8 mM copper and one of them (No. 78) was resistant to 10 mM copper (Table 2). However, the phenotypic results of most strains for copper resistance did not match the genotypic ones as seven strains (8.75%) possessed the pcoE gene, which is an indicator for carriage of the pco copper resistance system, but did not show copper resistance at the level that the positive control (E. coli ED8739 pRJ1004) demonstrated (Figure 3c). This result may be due to copper resistance being conferred after a succession of events from a combination of both chromosomal and plasmid-encoded genes [86], or that an incomplete pco system is present. The pco and sil operons have regularly been found together in an identical arrangement in plasmids, but also on the chromosome [6]. In this study, we further found that all strains possessing the pcoE gene also contained the sil genes.
Noticeably, one strain (No. 14) containing mer, intl1, sil, pcoE, and astA genes was detected ( Table 2). This strain was phenotypically resistant to mercury (25 µg/mL), silver (400 µM), and antibiotics (oxytetracycline and sulfonamide). That means there were co-resistant strains in the food chain. The result also indicated there are high-risk, multi-drug (metal and antibiotics) resistant strains in the food chain that carry genes for potential pathogenicity. Fortunately, no phenotypic copper resistance was detected in this strain.
It is of particular concern that exposure to heavy metals could favor co-selection of antimicrobial resistances [87]. The determination of correlations occurring between antibiotic and heavy metal resistances is considered an important tool to gain new insight into the not yet fully characterized mechanism(s) leading to co-selection [88]. In this study, correlation of metal resistance and antibiotic resistance presence was examined (Table 2). Oxytetracyline was common in all metal resistant strains; 13 of 19 strains with silver resistance showed OT resistance and 14 of 14 strains with mercury resistance showed OT resistance. However, OT resistance was also common in non-metal resistant strains and, for the 80 strains, the Pearson correlation coefficient of OT with silver was r = 0.054, showing no specific correlation; for OT with mercury, it was r = 0.34, showing weak correlation. However, for all silver resistant strains, OT resistance with silver was r = 0.79 and that of OT with mercury was r = 1. The results showed that a correlation existed for OT and silver and a strong correlation existed for OT and mercury. These results also indicate co-selection between antibiotic and metal might exist in E. coli isolates in this study.
The strains used in this study were collected between 2007 and 2010. There have since been very significant reductions in antibiotic usage in pig husbandry in the United Kingdom, including in use of human critical antibiotics such as fluoroquinolones, 3/4 generation cephalosporins, and colistin [89], and there are plans to remove zinc as a supplement in pig feed in the EU. These data, therefore, provide an historic baseline against which the results of these changes in antimicrobial usage can be measured. The retention of resistances to antimicrobials no longer in active use may demonstrate a genetic archaeology of the use of veterinary antimicrobials and provide a better understanding of the drivers for their presence.

Isolate Recovery and Re-confirmation
Brain Heart Infusion agar (Sigma, UK) was used to recover the isolates stored in Microbank (Pro-Lab, UK) beads at −80 • C. The inoculated plates were incubated at 37 • C for 24 h. LB agar (Sigma, UK) was used to cultivate the isolates. Growth on TBX agar (Sigma, UK) was used to re-confirm the isolates as E. coli. An oxidase test was performed using a strip (MB0266A, Oxoid, UK) as directed by the manufacturer. An indole test was performed using the RapID Spot Indole Reagent (R8309002, SLS, UK). Several drops of Indole Spot Reagent were placed on a piece of filter paper. With an inoculating loop a portion of an 18-24 hour growth isolated colony was picked up and rubbed onto the reagent saturated area. A positive reaction was denoted by the appearance of a blue to blue-green colour change on the bacterial smear within 10 seconds. Negative reactions remained colourless or light pink.

Antibiotic Sensitivity Tests
The disc diffusion antibiotic sensitivity tests were carried out according to Clinical and Laboratory Standards Institute (CLSI) guidelines [40,41]. Four or five bacterial colonies were taken from LB plates, which had been incubated overnight at 37 • C. The colonies were suspended in 5 mL of Mueller-Hinton broth (Oxoid, UK) to a 0.5 McFarland standard concentration. Then, 100 µL of the culture was pipetted onto the surface of a 25 mL Mueller-Hinton agar plate (Oxoid, UK), and the inoculum was distributed by spreading using a sterile swab. The plates were left to dry at room temperature (for no more than 15 min), and the antibiotic discs were applied using a disc dispenser (Pro-lab, UK). The plates were then incubated at 37 • C for 18-24 h, and the results were recorded by measuring the inhibition zone diameter across the disc and interpreted according to standard measurement tables [40,41]. The seventeen antibiotics used in the tests are listed in Table 3; discs were supplied by Oxoid (UK), except for Cefquinome (Bioconnection, UK), and the quality control strains used were E. coli ATCC 25922 (ESBL negative) and E. coli NCTC 13353 (ESBL positive CTX-M-15). For phenotypically colistin-resistant isolates, the agar dilution method was used according to the CLSI procedure [39]. Isolates showing growth above the concentration of 8 mg/L colistin were regarded as resistant [90].

PCR Detection of Genetic Elements Carrying Metal and Antibiotic Resistance Genes
Bacterial DNA was extracted by dispersing two colonies of E. coli from an overnight culture on LB agar into 100 µL of sterile 1 × TE buffer (10 mM Tris-Cl, 1 mM EDTA buffer, pH 7.6). The suspension was heated to 100 • C for 30 min to rupture bacterial cells (Eppendorf Thermomixer Comfort, Germany) and then centrifuged (Sigma, UK) at 13,000 × g for 15 min. The crude DNA in the supernatants was transferred into sterile microcentrifuge tubes and stored at -20 • C until use (total DNA). All primers in this study were synthesized by Sigma-Aldrich Life Science, UK. The primer sequences and expected PCR product sizes are given in Table 4. Unless indicated otherwise, E. coli ATCC 25922 was used as a negative control. A 100 bp DNA size marker (NEB, UK) was used in each gel, except where indicated otherwise.
For the detection of β-lactamase genes, PCR was used to examine for bla SHV , bla TEM , bla OXA-1 , and bla CTX-M genes [96]. Presence of the mcr-1 resistance gene was used to determine resistance to colistin sulfate. The reaction mixture was as described above. E. coli NCTC 13353 was used as a positive control for CTX, E. coli NCTC 13352 as a positive control for TEM, E. coli strain A2.5 [38] as a positive control for OXA, and Klebsiella pneumoniae NCTC 13368 as a positive control for SHV. E. coli NCTC 13846 was used as a positive control for mcr-1. Except for mcr-1, PCR amplification was one cycle of denaturation for 5 min at 94 • C followed by 30 cycles of 30 s at 94 • C, 30 s at 55 • C, and 1 min at 72 • C, with a final extension of 7 min at 72 • C [38]. The program for mcr-1 was 10 min at 94 • C followed by 25 cycles of 30 s at 94 • C, 1 min at 58 • C, and 1 min at 72 • C, with a final extension of 10 min at 72 • C.
For silver resistance, silA, silB, and silE genes were screened [97]. E. coli J53 and E. coli J53 pMG101 were negative and positive controls, respectively. PCR reactions were carried out under the following conditions: one cycle of denaturation for 3 min at 95 • C followed by 30 cycles of 30 s at 95 • C, 30 s at 54 • C, and 1 min at 72 • C, with a final extension of 5 min at 72 • C. The markers with 100 bp (for silB and silE) and 1000 bp (silA) DNA size were used in each gel (Table 4).
For mercury resistance of merA and merC genes [99], as well as intl1 [100] detection, 25 µL reaction mixture contained 2.5 µL 10 × Buffer, 2 µL (2.5 mM) of each of the dNTPs, 0.5 µL (10 µM) of each primer, 0.125 µL (5 U/µL) DreamTaq (ThermoFisher Scientific, UK), 1 µL DNA template, and 18.375 µL of molecular biological grade water. The strains, E. coli J53 and E. coli J53 pMG101, were used as negative and positive controls, respectively. PCR reactions were carried out under the following conditions: one cycle of denaturation for 5 min at 95 • C followed by 30 cycles of 30 s at 95 • C, 30 s at 67 • C, and 1 min at 72 • C, with a final extension of 5 min at 72 • C.
For copper resistance of pcoE gene detection, total DNA (1 µL) was used in a 25 µL reaction mixture that contained 2.5 µL 10 × Buffer, 2 µL (2.5 mM) of each of the dNTPs, 2 µL (10) µM of each primer, 0.125 µL (5 U/µL) DreamTaq (ThermoFisher Scientific, UK), and 15.375 µL of molecular biological grade water. The strains, E. coli ED8739 and E. coli ED8739 pRJ1004, were as negative and positive controls, respectively. PCR reactions were carried out under the following conditions: one cycle of denaturation for 5 min at 95 • C followed by 30 cycles of 30 s at 95 • C, 30 s at 48 • C, and 1 min at 72 • C, with a final extension of 5 min at 72 • C.

Conclusions
In this study, we have examined the phenotypic and genotypic resistances to 18 antibiotics and 3 metals (mercury, silver, and copper) of E. coli isolated between 2007 and 2010 from pig slaughterhouses in the United Kingdom. The results showed resistances to oxytetracycline, streptomycin, sulphonamide, ampicillin, chloramphenicol, trimethoprim-sulfamethoxazole, ceftiofur, amoxicillin-clavulanic acid, aztreonam, and nitrofurantoin. The top three resistances were oxytetracycline (64%), streptomycin (28%), and sulphonamide (16%). Two strains were resistant to six kinds of antibiotics. Three carried the bla TEM gene. Fifteen strains (18.75%) were resistant to 25 µg/mL mercury and five (6.25%) of these to 50 µg/mL; merA and merC genes were detected in 14 strains. Thirty-five strains (43.75%) showed resistance to silver, with 19 possessing silA, silB, and silE genes. Fifty-five strains (68.75%) were resistant to 8 mM copper or above. Seven contained the pcoE gene. Some strains were multi-resistant to antibiotics, silver, and copper. These results indicate the potential for high-risk, multi-drug (metal and antibiotics) resistant strains to enter the food chain and co-selection between antibiotic and metal might exist in E. coli isolates, and will aid understanding about the effects of strategies to reduce resistance and mechanisms of antimicrobial resistance (AMR).