Third-Generation Cephalosporin- and Tetracycline-Resistant Escherichia coli and Antimicrobial Resistance Genes from Metagenomes of Mink Feces and Feed

Abstract

American mink (Neovison vison) is a significant source of global fur production. Except for a few studies from Denmark and Canada reporting antimicrobial resistance in bacteria isolated from clinical cases, studies from the general mink population are scarce and absent in the United States. Mink feces (n = 42) and feed (n = 8) samples obtained from a mink farm were cultured for the enumeration and detection of tetracycline-resistant (TET^r)- and third-generation cephalosporin-resistant (TGC^r)-Escherichia coli. Isolates were characterized phenotypically for their resistance to other antibiotics and genotypically for resistance genes. TET^r E. coli were detected from 98% of feces samples (mean concentration = 6 log₁₀) and from 100% of feed samples (mean concentration = 3.2 logs). Among TET^r E. coli isolates, 44% (n = 41) of fecal- and 50% (n = 8) of feed isolates were multidrug resistant (MDR; resistance to ≥3 antimicrobial classes), and 96% (n = 49) of TET^r isolates were positive for tet(A) and/or tet(B). TGC^r E. coli were detected from 95% of feces and 75% of feed samples with 78% (n = 40) of fecal isolates, and all six of the feed isolates were MDR. Nearly two-thirds (65%) of the TGC^r E. coli isolates (n = 46) were positive for bla _CMY-2; the remaining 35% were positive for bla _CTX-M, with the bla _CTX-M-14 being the predominant (75%, n = 16) variant detected. Metagenomic DNA was extracted directly from feces and feed samples, and it was tested for 84 antimicrobial resistance genes by using quantitative polymerase chain reaction (PCR) array; selected genes were also quantified by droplet digital PCR. The genes detected from the fecal samples belonged mainly to five antimicrobial classes: macrolide–lincosamide–streptogramin B (MLS_B; 100% prevalence), TETs (88.1%), β-lactams (71.4%), aminoglycosides (66.7%), and fluoroquinolones (47.6%). β-Lactam, MLS_B, and TET resistance genes were also detected from feed samples. Our study serves as a baseline for further studies and to streamline antimicrobial use in mink production in accordance with current regulations as in food animals.

Introduction

American mink (Neovison vison) is farmed for its skin (pelt) production, and it plays a leading role in the global fur industry accounting about 80% of international fur trade (Hansen, 2014). The United States, on average, produced 3.2 million mink pelts per year over the past 45 years, which was valued at $126 million per year, based on annual data released by the National Agricultural Statistics Service from 1975 to 2018 (Supplementary Fig. S1). Mink are seasonal breeders that occur yearly between late February and early March, and the females give birth up to 5–8 kits between late April and mid-May. Mink are weaned between 6 and 10 weeks of age and harvested for fur in November and December. Mink are kept individually in wire meshed cages that are placed in rows inside uninsulated barns (Hovland et al., 2017).

Mink are carnivores and their feed comprises animal products obtained from dairy, beef, poultry, and fisheries operations (FCUSA, 2019). Animal source feed can be a source of infection to the mink if not properly cooked (Jensen et al., 2016, 2017). Antimicrobials are used for the treatment of various infections that affect mink health (Compo et al., 2017), exerting selection pressure and resulting in the propagation of antimicrobial-resistant (AMR) bacteria in the mink farms and beyond (Pedersen et al., 2009; Nikolaisen et al., 2017).

The AMR bacteria can be introduced into mink farms through feed, propagates under intensive animal production, and spreads into the environment through animal manure (Travis and Aulerich, 1978). Other than a few studies, mainly in Denmark (Pedersen et al., 2009; Nikolaisen et al., 2017) that reported AMR mostly in pathogenic bacteria isolated from clinical cases, AMR studies in the general mink population are scarce, and particularly absent in the United States. The objectives of this study were to investigate the occurrence of tetracycline (TET^r)- and third-generation cephalosporin (TGC^r)-resistant E. coli by culture, and antimicrobial resistance genes (ARGs) in mink fecal and feed metagenomes.

Materials and Methods

Institutional animal care and use committee review was not required since fecal samples were collected from under the cages of the mink with no manipulation of the animals. In the summer of 2017, the opportunity arose to get access to a mink farm after oral consent was obtained. Accordingly, 42 fecal samples from seven houses (6 samples per house) and 8 bulk feed samples were collected from a mink farm that had multiple open-air houses with 600–4500 mink per house. Feed samples were not cooked at the time of collection, and the farm imported feed from an outside source.

Isolation and characterization of TET^r and TGC^r E. coli

Ten grams of samples were transferred to filter bags (Nasco Whirl-Pack, Fort Atkinson, WI), suspended in 90 mL of buffered peptone water (BPW; Becton, Dickinson and Company [BD], Franklin Lakes, NJ), and homogenized in a laboratory blender (Stomacher 400 Circulator; Seward Laboratory Systems, Inc., Islandia, NY) at 200 rpm for 15 s. After an 8-mL aliquot was removed for enumeration, the remaining suspension was incubated at 25°C for 2 h followed by 42°C for 6 h and held at 4°C for secondary enrichment (Agga et al., 2016b).

Generic- and TET^r-E. coli were enumerated by plating 50 μL of appropriate dilutions by using an Eddy Jet 2 spiral plater (Neutec Group Inc., Farmingdale, NY) on modified membrane thermo-tolerant E. coli (m-TEC) agar (BD), and CHROMagar E. coli (DRG International, Inc., Springfield, NJ) plates supplemented with 16 mg/L TET (CEC + TET), respectively. The m-TEC plates were incubated at 35°C for 2 h and then at 44°C for 22 h; CEC + TET plates were incubated at 44°C for 24 h according to the manufacturer's instructions. After incubation, red or magenta colonies on m-TEC and blue colonies on CEC + TET were counted by using a SphereFlash^® automated colony counter (Neutec Group, Inc.) as presumptive generic- and TET^r-E. coli, respectively. Secondary enrichments were done to detect generic- and TET^r-E. coli from three and two enumeration negative samples, respectively, and TGC^r E. coli. BPW pre-enrichments (0.5 mL) were transferred to 2.5 mL of MacConkey broth (MCB; BD), MCB supplemented with 16 mg/L TET (MCB + TET), and MCB supplemented with 2 mg/L of cefotaxime (MCB + CTX). After 18–24 h of incubation at 42°C, MCB, MCB + TET, and MCB + CTX broths were streaked onto m-TEC, CEC + TET, and m-TEC plates supplemented with 2 mg/L CTX (m-TEC + CTX), respectively. Antibiotics were obtained from Millipore Sigma (St. Louis, MO), and the Clinical Laboratories Standards Institute (CLSI) resistance breakpoint concentrations (CLSI, 2020) were used (Table 1).

Table 1.

Concentration Ranges and Clinical Laboratories Standards Institute Resistance Breakpoints of Fourteen Antimicrobials Used to Determine Susceptibilities of Tetracycline-Resistant Escherichia coli Isolated from Mink Feces and Feed Samples

CLSI class	Antimicrobial agent	Concentration tested (μg/mL)	Resistance breakpoint (μg/mL)
Aminoglycosides	Gentamicin	0.25–16	≥16
Aminoglycosides	Streptomycin	2–64	≥32
β-Lactam/β-lactamase inhibitor combinations	Amoxicillin/clavulanic acid	1/0.5–32/16	≥32/16
Carbapenems	Meropenem	0.06–4	≥4
Cephems	Cefoxitin	0.5–32	≥32
Cephems	Ceftriaxone	0.25–64	≥4
Folate pathway inhibitors	Sulfisoxazole	16–256	≥512
Folate pathway inhibitors	Trimethoprim/sulfamethoxazole	0.12/2.38–4/76	≥4/76
Macrolides	Azithromycin	0.25–32	≥32
Penicillins	Ampicillin	1–32	≥32
Phenicols	Chloramphenicol	2–32	≥32
Quinolones	Ciprofloxacin	0.015–4	≥1
Quinolones	Nalidixic acid	0.5–32	≥32
Tetracyclines	Tetracycline	4–32	≥16

Breakpoints were adopted from the CLSI M100-ED30 document (CLSI, 2020). Streptomycin has no CLSI breakpoint, and the NARMS consensus breakpoint was applied (CDC, 2018).

CLSI, Clinical Laboratories Standards Institute; NARMS, National Antimicrobial Resistance Monitoring System.

From each positive sample, two colonies were inoculated into 96 deep wells (VWR, Radnor, PA) containing tryptic soy broth (BD), and they were incubated overnight at 37°C. DNA was isolated by using the BAX lysis method following the manufacturer's instructions (DuPont Qualicon, Inc., Wilmington, DE). Isolates were confirmed by multiplex polymerase chain reaction (PCR) as described (Horakova et al., 2008). E. coli ATCC 25922 was used as a positive control. TET^r- and TGC^r-E. coli isolates were tested against 14 antimicrobials (Table 1) by broth microdilution using Sensititre system (Trek Diagnostics, Thermo Fisher Scientific, Inc., Cleveland, OH). TGC^r E. coli were also tested for additional 10 broad-spectrum β-lactam drugs by using the ESB1 panel (Table 2). E. coli ATCC 25922 was included for quality control. Isolates were tested for resistance genes by PCR (Table 3). bla _CTX-M-positive isolates were tested for bla _CTX-M-14 and bla _CTX-M-15 by quantitative PCR (qPCR) using Streck ARM-D multiplex kit (La Vista, NE) and by PCR for bla _CTX-M-2.

Table 2.

Concentration Ranges and Clinical Laboratories Standards Institute Resistance Breakpoints of Sixteen Antimicrobials on ESB1 Sensititre Panel Used to Determine Susceptibilities of Third-Generation Cephalosporin-Resistant Escherichia coli

Antimicrobial class	Antimicrobial agent	Concentration tested (μg/mL)	Resistance breakpoint (μg/mL)
Penicillin	Ampicillin	8–16	≥32
Cephems	Cefazolin	8–16	≥8
	Cefepime	1–16	≥16
	Cefotaxime	0.25–64	≥4
	Cefoxitin	4–64	≥32
	Cefpodoxime	0.25–32	≥8
	Ceftazidime	0.25–128	≥16
	Ceftriaxone	1–128	≥4
	Cephalothin^a	8–16	≥32
Fluroquinolone	Ciprofloxacin	1–2	≥1
Aminoglycosides	Gentamicin	4–16	≥16
Carbapenem	Imipenem	0.5–16	≥4
Carbapenem	Meropenem	1–8	≥4
Beta lactam/beta lactamase inhibitor	Piperacillin/tazobactam constant 4	4/4–64/4	≥128/4
	Cefotaxime/clavulanic acid	0.12/4–64/4	≥8/4
	Ceftazidime/clavulanic acid	0.12/4–128/4	≥64/4

Breakpoints were adopted from the CLSI M100-ED30 document (CLSI, 2020).

CDC breakpoint for E. coli was used https://cdc.gov/narms/antibiotics-tested.htm

Table 3.

Primer Sequences Used for The Polymerase Chain Reaction Detection of Antimicrobial Resistance Genes of Escherichia coli Isolated from Mink Feces and Feed

Target gene	Primer	Primer sequence (5′-3′)^a	Tm ^b (°C)	PCR product (bp)^c	References
bla _CMY-2	F	GACAGCCTCTTTCTCCACA	55.0	1000	Kozak et al. (2009)
bla _CMY-2	R	TGGACACGAAGGCTACGTA	55.0	1000	Kozak et al. (2009)
bla _CTX-M	F	CCGCTGCCGGTYTTATC	55.0	500	Cottell et al. (2013)
bla _CTX-M	R	ATGTGCAGYACCAGTAA	55.0	500	Cottell et al. (2013)
bla _CTX-M-2	F	GCGCAGACCCTGAAAAATCT	55.0	59	Cormier et al. (2019)
bla _CTX-M-2	R	TGTGCCCGCTGAGTTTCC	55.0	59	Cormier et al. (2019)
tet(A)	F	GCTACATCCTGCTTGCCTTC	55.0	210	Ng et al. (2001)
tet(A)	R	CATAGATCGCCGTGAAGAGG	55.0	210	Ng et al. (2001)
tet(B)	F	TTGGTTAGGGGCAAGTTTTG	55.0	659	Ng et al. (2001)
tet(B)	R	GTAATGGGCCAATAACACCG	55.0	659	Ng et al. (2001)
tet(C)	F	CTTGAGAGCCTTCAACCCAG	55	418	Ng et al. (2001)
tet(C)	R	ATGGTCGTCATCTACCTGCC	55	418	Ng et al. (2001)
tet(E)	F	AAACCACATCCTCCATACGC	55	278	Ng et al. (2001)
	R	AAATAGGCCACAACCGTCAG

Primer concentration of 400 nm.

Tm (°C) is the annealing temperature of the PCR reaction.

bp is base-pairs.

PCR, polymerase chain reaction.

Detection and quantification of antibiotic resistance genes

Metagenomic DNA was extracted from 500 mg of feces and feed samples by using the FastDNA Spin kit for soil (MP Biomedical, Solon, OH), and it was purified by using Genomic DNA Clean and Concentrator-25 kit (Zymo Research, Irvine, CA). Individual fecal DNA and four pooled DNA from feed samples were analyzed by using microbial DNA qPCR array (QIAGEN, Valencia, CA). The array detects up to 84 genes (Supplementary Table S1) that belong to aminoglycosides (5 genes), β-lactams (56), macrolide–lincosamide–streptogramin B (MLS_B; 6), fluoroquinolones (11), TETs (2), glycopeptides (2), and multidrug resistance (2). Selected ARGs were quantified by Droplet Digital PCR (QX200; Bio Rad) with Eva green chemistry and QuantaSoft data analysis software (Bio Rad) using published primers and protocols (Table 4).

Table 4.

Primers Used in the Droplet Digital Polymerase Chain Reaction for Quantification of Resistance Genes from Mink Feces and Feed Samples

Gene	Forward primer^a	Reverse primer^a	Annealing Tm (°C)^b	References
aacC4	GAGTTGATGGCAAAGGTTCC	AAGGCTCTTCTCCTTGAGCC	58	Szczepanowski et al. (2009)
aadA1	GCGAGCTTTGATCAACGACC	ATGTCATTGCGCTGCCATTC	60	Vikram et al. (2017)
bla _NDM	CAGCGCAGCTTGTCG	TCGCGAAGCTGAGCA	52	Peirano et al. (2011)
bla _ACC-1	CGCTGATGCAGAAGAATA	CGCTAACCCATAGTTATAAATG	55	Geyer et al. (2012)
bla _ACC-3	GCATGACGCACACCTATCTT	CGTGGATCGGCTCATTCTTT	57	This study
bla _OXA-58	GCAATTGCCTTTTAAACCTGA	CTGCCTTTTCAACAAAACCC	58	Szczepanowski et al. (2009)
bla _CMY-2	TTCTCCGGGACAACTTGACG	GCATCTCCCAGCCTAATCCC	60	Vikram et al. (2017)
bla _CTX-M	GATACCACTTCACCTCGGGC	CCTTAGGTTGAGGCTGGGTG	60	Vikram et al. (2017)
bla _KPC-2	AATGAGCTGCACAGTGGGAA	ATCGCCGTCTAGTTCTGCTG	60	Vikram et al. (2017)
bla _SHV	CCAGATCGGCGACAACGT	ACCCGCCTTGACCG	60	Cormier et al. (2019)
bla _TEM	GTGCGGTATTATCCCGTGTTG	CGCCGGGCAAGAG	60	Cormier et al. (2019)
qnrB-5	GATCGTGAAAGCCAGAAAGG	ACGATGCCTGGTAGTTGTCC	53	Robicsek et al. (2006)
qnrD	CGAGATCAATTTACGGGGAATA	AACAAGCTGAAGCGCCTG	50	Cavaco et al. (2009)
ermA	AGTCAGGCTAAATATAGCTATC	CAAGAACAATCAATACAGAGTCTAC	56	Koike et al. (2010)
ermB	TCACCGAACACTAGGGTTGC	CTGTGGTATGGCGGGTAAGT	60	Vikram et al. (2017)
ermC	AATCGTGGAATACGGGTTTGC	CGTCAATTCCTGCATGTTTTAAGG	62	Koike et al. (2010)
mefA	GGAGCTACCTGTCTGGATGG	CAACCGCCGGACTAACAATA	58	Szczepanowski et al. (2009)
msrA	CAAATGGCACAAGCATCATC	TGTGGTTTTTCAACTTCTTCCA	58	Szczepanowski et al. (2009)
tetA	CGGCAATCATTCCGAGCATG	ATTCTGCATTCACTCGCCCA	60	Vikram et al. (2017)
tetB	TGGTGGTGGGATCGCTTTAC	AATGGGCCAATAACACCGGT	60	Vikram et al. (2017)
tetM	GTGCCGCCAAATCCTTTCTG	GCATCCGAAAATCTGCTGGG	60	Vikram et al. (2017)
vanC1	TGCTTGTGATGCGATTTCTC	ATCGCTCCTTGATGGTGAC	60	Rashidi et al. (2019)
vanC2/3	GGGAAGATGGCAGTATCCAA	GCAGCAGCCATTTGTTCATA	60	Rashidi et al. (2019)
mecA	GGCATTGTAGCTAGCCATTCC	TGGCAGACAAATTGGGTGGT	60	Vikram et al. (2017)
sul1	CGCACCGGAAACATCGCTGCAC	TGAAGTTCCGCCGCAAGGCTCG	56	Pei et al. (2006)
sul2	TCCGGTGGAGGCCGGTATCTGG	CGGGAATGCCATCTGCCTTGAG	61	Pei et al. (2006)

Primer concentration of 400 nM.

Tm (°C) is the annealing temperature of the PCR reaction.

Bacterial enumeration and gene quantification data were back calculated to colony-forming units (CFUs) or gene copies per gram of feces or feed, respectively, and analyzed by the negative binomial regression model. Expected counts, expressed as mean log₁₀ values, were obtained. Binary data were expressed as the proportion of total samples tested. Data were analyzed by using Stata 16.1 (Stata Corporation, College Station, TX).

Results

Concentration and prevalence of TET^r- and TGC^r-E. coli

Generic- and TET^r-E. coli were detected from 98% (n = 42) of feces samples, with the mean counts of 6.1 (95% confidence interval [CI]: 5.8–6.3) and 5.7 (95% CI: 5.5–5.9) log₁₀ CFU/g feces, respectively. All feed samples (n = 8) were positive for generic- and TET^r-E. coli, with mean counts of 3.7 (95% CI: 3.6–3.8) and 3.2 (95% CI: 3.1–3.2) log₁₀ CFU/g feed, respectively. TGC^r E. coli were detected from 95% of the fecal (n = 42) and 75% of the feed (n = 8) samples. Fecal TET^r E. coli isolates were predominantly co-resistant to aminoglycosides, sulfonamides, and penicillins, ranging from 32% to 59%. Resistance to other antimicrobial classes was below 3% (Table 5). Overall, 13 resistance profiles were detected; 44% of TET^r E. coli isolates were multidrug resistant (MDR), defined as being resistant to ≥3 CLSI antimicrobial classes, including TET. More than half (58.5%) of TET^r E. coli isolates were resistant to TET alone, or co-resistant to streptomycin, sulfisoxazole, or both drugs (Supplementary Table S2). Feed TET^r E. coli isolates were co-resistant to four classes (aminoglycosides, sulfonamides, macrolides, and penicillins) (Table 5). Four resistance profiles were detected among TET^r E. coli feed isolates, with half of the isolates being MDR; the remaining half were resistant to TET alone (Supplementary Table S2). Cross- and co-resistance among TGC^r E. coli isolates is shown in Table 5. Half of the feed isolates and 68% of the fecal isolates were cross-resistant to second-generation cephalosporin and potentiated β-lactams. The predominant co-resistance was to TET. Among fecal TGC^r E. coli isolates, nine resistance profiles were observed, with 77.5% of the isolates being MDR. Among the feed TGC^r E. coli, four resistance profiles were detected, with all isolates being MDR (Supplementary Table S3).

Table 5.

Percentage Resistant (Number) to Other Antibiotics of Tetracycline- and Third-Generation Cephalosporin-Resistant Escherichia coli Isolated from Mink Feces (n = 42) and Feed (n = 8) Samples

CLSI class	Antimicrobial agent	TET^r E. coli		TGC^r E. coli
CLSI class	Antimicrobial agent	Feces (n = 41)	Feed (n = 8)	Feces (n = 40)	Feed (n = 6)
Aminoglycosides	Gentamicin	31.7 (13)	25.0 (2)	0	33.3 (2)
Aminoglycosides	Streptomycin	58.5 (24)	50.0 (4)	17.5 (7)	50.0 (3)
β-Lactam/β-lactamase inhibitor combinations	Amoxicillin/clavulanic acid	2.4 (1)	0	67.5 (27)	50.0 (3)
	Piperacillin/tazobactam constant 4^a			2.5 (1)	0
	Cefotaxime/clavulanic acid^a			67.5 (27)	50.0 (3)
	Ceftazidime/clavulanic acid^a			7.5 (3)	16.7 (1)
Carbapenems	Meropenem	0	0	0	0
	Imipenem^a			2.5 (1)	0
Cephems	Cefoxitin	2.4 (1)	0	67.5 (27)	50.0 (3)
Cephems	Ceftriaxone	2.4 (1)	0	100 (40)	100 (6)
	Cefazolin^a			100 (40)	100 (6)
	Cefepime^a			0	0
	Cefotaxime^a			100 (40)	100 (6)
	Cefpodoxime^a			100 (40)	100 (6)
	Ceftazidime^a			67.5 (27)	50.0 (3)
	Cephalothin^a,b			100 (40)	100 (6)
Folate pathway inhibitors	Sulfisoxazole	39.0 (16)	50.0 (4)	10.0 (4)	33.3 (2)
Folate pathway inhibitors	Trimethoprim/sulfamethoxazole	7.3 (3)	25.0 (2)	2.5 (1)	0
Macrolides	Azithromycin	0	12.5 (1)	0	0
Penicillins	Ampicillin	39.0 (16)	25.0 (2)	100 (40)	100 (6)
Phenicols	Chloramphenicol	2.4 (1)	0	5.0 (2)	33.3 (2)
Quinolones	Ciprofloxacin	0	0	0	0
Quinolones	Nalidixic acid	2.4 (1)	0	5.0 (2)	0
Tetracyclines	Tetracycline	100 (41)	100 (8)	67.5 (27)	83.3 (5)

In addition to the 14 antibiotics on the gram-negative (CMV4AGNF) panel, TGC^r E. coli were also tested on an additional 10 antimicrobials on the ESB1 panel. The blank spaces corresponding to the 10 additional drugs indicate that the TET^r E. coli were not tested against these drugs.

The NARMS consensus breakpoint was applied.

TET^r, tetracycline resistant; TGC^r, third generation cephalosporin resistant.

Ninety-six percent of TET^r E. coli (n = 49) were positive for tet(A) (81.6%), tet(B) (24.5%), or both (10.2%). The remaining two isolates (4.0%) were also negative for tet(C) and tet(E); one of the isolates was resistant only to TET, whereas the other isolate was resistant to gentamycin, streptomycin, and TET. About two-thirds (65.2%) of the TGC^r E. coli isolates (n = 46) were positive for bla _CMY-2. The remaining 34.8% were positive for bla _CTX-M, consisting of 12 isolates positive for bla _CTX-M-14, 3 for bla _CTX-M-15, and 1 for bla _CTX-M-2. Only one (2.0%) TET^r E. coli isolate was positive for bla _CMY-2, and no bla _CTX-M gene was detected among the TET^r E. coli isolates. On the contrary, about 70% of the TGC^r E. coli isolates (n = 46) were positive for tet(A) (56.5%), tet(B) (17.4%), or both (4.4%).

Prevalence and quantity of ARGs from metagenomic DNA of fecal and feed samples

Thirty-three ARGs conferring resistance to aminoglycosides (4 genes), β-lactams (16), fluoroquinolones (5), MLS_B (5), TETs (2), and glycopeptides (1) were detected from the fecal samples (Table 6). The number of ARGs detected per sample ranged from 2 to 19, with a mean and median of nine (Fig 1). The ARGs detected from the fecal samples (n = 42) belonged mainly to five antimicrobial classes: MLS_B (100% prevalence), TETs (88.1%), β-lactams (71.4%), aminoglycosides (66.7%), and fluoroquinolones (47.6%). The most prevalent aminoglycoside resistance gene was aadA1 (62%), followed by aacC4 (36%). Class C β-lactamases were detected in >50% of the fecal samples predominated by the bla _ACC genes; bla _OXA and bla _SHV were detected in 29% and 17% of the fecal samples, respectively. The qnrB-5 (38%), followed by qnrD (19%), was the predominant fluoroquinolone resistance gene detected from the fecal samples. MLS_B resistance was mainly represented by the commonly encountered erm (100%), mef(A) (91%), and msr(A) (41%) genes. TET^r genes tet(A) and tet(B) were detected with a similar frequency. All four pooled feed DNA samples carried four genes, with each representing three antimicrobial classes: bla _ACC-3, two erm and mef(A), and tet(A) genes. Mean gene copies of ARGs ranged from 2.4 (bla _CMY-2) to 6.3 [tet(M)] log₁₀/g of feces. For feed samples, it ranged from 2.8 (aacC4) to 7.1 [tet(M)] log₁₀/g feed (Table 7).

FIG. 1.

The number of resistance genes detected per sample from mink feces by quantitative polymerase chain reaction array.

Table 6.

Prevalence of Antibiotic Resistance Genes Detected from Metagenomic DNA Extracted from Mink Feces and Feed Samples

Antibiotic class	Antibiotic resistance genes	Feces (n = 42)		Feed (n = 4)^a
Antibiotic class	Antibiotic resistance genes	Number positive	% Positive	Feed (n = 4)^a
Aminoglycosides		28	66.7
	aacC2	4	9.5
	aacC4	15	35.7
	aadA1	26	61.9
	aphA6	8	19
Beta-lactams		30	71.4	2
Class A beta-lactamases		7	16.7
	bla _SHV (156D)	1	2.4
	bla _SHV (156G)	7	16.7
	bla _SHV (238G240E)	6	14.3
	bla _SHV (238G240K)	1	2.4
	bla _SHV (238S240E)	1	2.4
	bla _SHV (238S240K)	1	2.4
Class B beta-lactamases	bla _NDM	1	2.4
Class C beta-lactamases		24	57.1	2
	bla _ACC-1	24	57.1
	bla _ACC-3	15	35.7	2
	bla _DHA	2	4.8
	bla _LAT	2	4.8
Class D beta-lactamases		12	28.6
	bla _OXA-10	1	2.4
	bla _OXA-2	1	2.4
	bla _OXA-55	1	2.4
	bla _OXA-58	9	21.4
Methicillin resistance	mecA	6	14.3
Fluoroquinolones		20	47.6
	aac (6)-Ib-cr	2	4.8
	qnrB-1 group	1	2.4
	qnrB-5 group	16	38.1
	qnrD	8	19.0
	qnrS	1	2.4
Macrolide-lincosamide-streptogramin B		42	100	4
	ermA	17	40.5
	ermB	42	100	4
	ermC	30	71.4	2
	mefA	38	90.5	4
	msrA	17	40.5
Tetracyclines		37	88.1	4
	tetA	34	81	4
	tetB	37	88.1
Glycopeptides	vanC	5	11.9

Number of pooled feed samples positive.

Table 7.

Mean Counts (log 10/g) of Antibiotic Resistance Gene Copies from Metagenomic DNA Extracted from Mink Feces and Feed Samples

Antibiotic resistance class	Antimicrobial resistance genes	Feces (n = 42)		Feed (n = 8)
Antibiotic resistance class	Antimicrobial resistance genes	Count	95% CI	Count	95% CI
Aminoglycosides	aacC4	3.2	3.0–3.5	2.8	1.8–3.8
Aminoglycosides	aadA1	4.3	4.0–4.5	3.0	2.0–4.0
Class A beta-lactamases	bla _CTX-M	2.7	2.5–2.9	3.5	2.7–4.3
	bla _KPC-2	2.7	2.5–2.8	3.6	3.5–3.7
	bla _SHV	2.6	2.3–2.8	3.1	2.1–4.1
	bla _TEM	4.7	4.5–5.0	3.5	2.6–4.3
Class B beta-lactamases	bla _NDM	2.5	1 sample	—	—
Class C beta-lactamases	bla _CMY-2	2.4	2.1–2.8	3.0	2.2–3.7
	bla_ACC-1	4.1	3.9–4.3	4.1	3.9–4.4
	bla _ACC-3	3.7	3.3–4.0	3.3	2.5–4.1
Class D beta-lactamases	bla _OXA-58	3.1	2.8–3.5	3.0	1.3–4.8
Fluoroquinolones	qnrB-5	3.0	2.8–3.2	4.4	4.3–4.6
Fluoroquinolones	qnrD	3.7	3.5–4.0	4.7	4.6–4.8
Macrolide-lincosamide-streptogramin B	ermA	3.7	3.4–4.1	3.6	1.7–5.4
	ermB	5.8	5.6–6.0	6.5	6.3–6.6
	ermC	5.6	5.2–5.9	3.4	2.9–3.8
	mefA	4.6	4.3–4.8	4.4	4.3–4.5
	msrA	3.8	3.5–4.0	2.7	1.5–4.0
Tetracyclines	tetA	4.7	4.2–4.9	3.7	3.6–3.8
	tetB	4.1	3.8–4.3	3.8	3.2–4.5
	tetM	6.3	6.0–6.5	7.1	6.9–7.2
Glycopeptides^a	vanC1	3.1	2.9–3.3	—
Glycopeptides^a	vanC2/3	3.6	2.9–4.2	—
Other beta-lactam genes	mecA	3.0	2.8–3.2	3.5	3.1–3.9
Sulfonamides	sul1	5.0	4.7–5.3	4.1	3.7–4.6
Sulfonamides	sul2	4.7	4.4–5.0	3.6	3.4–3.8

Five samples positive for vanC gene by qPCR array were quantified by primers that targeted C1 and C2/3 variants.

CI, confidence interval; qPCR, quantitative polymerase chain reaction.

Discussion

Although a great deal of information exists on food animals, humans, and the environment (Graham et al., 2019), it is also important to investigate AMR in other settings such as commercial mink production. A few studies reported that AMR of bacteria of mink originated from diagnostic submissions (Pedersen et al., 2009; Chriél et al., 2012; Hansen et al., 2017; Nikolaisen et al., 2017; Qiu et al., 2019). Here, we report AMR bacteria and genes from a commercial mink farm. As our first objective we examined the presence and level of TET (approved in 1945 to represent older antibiotics)- and TGC (approved in 1980 to represent newer antibiotics)-resistant E. coli by the culture method. TET^r E. coli was detected from almost all fecal and feed samples at 6 and 4 logs, respectively. TGC^r E. coli were also detected from 95% and 75% of fecal and feed samples, respectively. In Denmark, E. coli isolated at diagnostic labs from mink necropsy samples showed varying degrees of resistance to different antibiotics, including TET and β-lactams (Pedersen et al., 2009; Chriél et al., 2012; Nikolaisen et al., 2017). In China, the antibiotic resistance of E. coli obtained from the feces of healthy mink raised on a commercial farm ranged from 2% resistance to cefoxitin to ≥90% resistance to ampicillin and TET (Qiu et al., 2019).

Molecular analysis showed that TET^r and TGC^r in E. coli are mainly conferred by tet(A) and tet(B), and bla _CMY-2 or bla _CTX-M genes, respectively. The bla _CTX-M was detected in one-third of the TGC^r E. coli isolates, indicating that it is an important extended spectrum β-lactamase (ESBL) gene in E. coli. We previously reported that the use of ceftiofur (a TGC) in cow-calf operation was significantly associated with an increased prevalence of bla _CTX-M among the TGC^r E. coli (Agga et al., 2016a). A few TET^r E. coli isolates carried the bla genes, whereas the majority of TGC^r E. coli isolates carried the tet genes. This finding indicates a unidirectional selection in which TET^r is selected for perhaps through the use of other antibiotics, and that TET use may not necessarily select for resistance to other antibiotics. Although the mechanism behind this phenomenon requires further study, it can be related to the widespread occurrence of TET^r, which also is more likely related to its widespread use for a longer time than the relatively newer antibiotics (FDA, 2019), and that the fitness cost of bacteria carrying TET^r genes may be minimal (Nguyen et al., 1989). Phenotypically, although ampicillin resistance was observed in 39% of TET^r isolates, more than two-thirds of the TGC^r E. coli isolates were also TET^r (Table 5; Supplementary Tables S2 and S3). Although novobiocin is the only antibiotic approved by the United States Food and Drug Administration for use in mink (FDA, 2020), extra label use of other antibiotics such as ampicillin, penicillin G, florfenicol, erythromycin, neomycin, and trimethoprim sulfamethoxazole for disease treatment and prevention is common (FCUSA, 2019). A study in Denmark showed that penicillins, aminoglycosides, sulfonamides/trimethoprim, macrolides, lincosamides, and TETs were the most frequently used antibiotics during 2001–2006 (Pedersen et al., 2009). In another Denmark study between 2007 and 2016, TETs, aminopenicillins, macrolides, lincosamides, and sulfonamides/trimethoprim were the most frequently prescribed antibiotics for mink production (Nikolaisen et al., 2017).

In the second objective, we examined the presence of a broader range of ARGs in the metagenomes of mink feces and feed. The ARGs detected by qPCR array from the fecal metagenome of mink belonged mostly to five antibiotic classes: MLS_B, TET, β-lactam, aminoglycosides, and fluoroquinolones, most of which are used either as injectable for the treatment of individual mink or as in-feed for the treatment of the entire herd (FCUSA, 2019). We note that although the first part of this article reports on E. coli, total resistance genes detected through metagenomic DNA can be attributed to any bacteria in the samples, including non-cultured bacteria. The ARGs conferring resistance to MLS_B classes of antibiotics belong to mechanisms mediated by target modification through erythromycin ribosome methylation (erm) or efflux (mef: macrolide efflux and msr: macrolide-streptogramin resistance) genes (Roberts et al., 1999). The two most prevalent MLS_B genes erm(B) and mef(A) were reported by using a similar method from livestock and environmental sources (Agga et al., 2015; Vikram et al., 2017) and are commonly found in both gram-positive and gram-negative bacteria (Roberts, 2011). Although tet(A) and tet(B) are exclusively reported from gram-negative bacteria, tet(M) was equally reported from both gram-positive and gram-negative bacteria (Roberts and Schwarz, 2016).

The β-lactamase genes commonly detected from mink feces belonged to the three groups and four molecular classes (Bush and Jacoby, 2010). The group 1 β-lactamases of molecular class C (bla _CMY-2, bla _ACC) are plasmid-borne ampC type cephalosporinases that hydrolyze cephalosporins. The bla _CMY-2 gene is a major resistance gene detected among mink fecal TGC^r E. coli isolates. Ambler class C-1 (ACC-1) β-lactamase, first identified from Klebsiella pneumoniae from a patient in Germany (Bauernfeind et al., 1999), has variants that have been reported from enteric bacteria of human origin (Philippon et al., 2002). The bla _ACC-1 and bla _ACC-3 were detected from 57% and 36% of mink fecal samples, respectively, with a mean concentration of 4 logs. Subgroup 2be of molecular class A (some variants of bla _TEM and bla _SHV, and bla _CTX-M) that we identified represents ESBLs together with some variants of bla _OXA, a subgroup 2d of molecular class D (Bradford, 2001; Bush, 2018). The bla _SHV gene variants were detected at 17% prevalence with a mean concentration of 2.6 logs from mink feces (Tables 5 and 6), and they were also commonly detected in human wastewater samples (Agga et al., 2015). The presence of bla _TEM at 5 logs in the present study (Table 6) and its high prevalence (63.2%) in feedlot cattle in Canada (Cormier et al., 2020) may suggest its widespread occurrence in concentrated animal production, including mink farming.

The prevalence and concentration of the predominant aminoglycoside resistance gene aadA1 are similar to that of livestock and human waste (Agga et al., 2015) and feedlot cattle (Vikram et al., 2017; Miller et al., 2018). Quinolone resistance genes qnrB-5 and qnrD commonly observed in the mink feces (Tables 6 and 7) were rarely detected from livestock and municipal waste (Agga et al., 2015) or feedlot cattle (Vikram et al., 2017). Factors driving this widespread occurrence of plasmid-borne quinolone resistance genes on a mink farm compared with livestock or human population need further investigation. vanC, which was detected and subsequently quantified from five fecal samples, is chromosomally encoded and confers intrinsic low-level vancomycin resistance in certain Enterococcus species (Rashidi et al., 2019).

Feed quality is paramount in mink production since the diet of mink is based on animal products, which, if not properly cooked or re-contaminated, can predispose mink to various infectious diseases that can lead to more antibiotic use (Jensen et al., 2016, 2017). All feed samples analyzed were positive for TET^r E. coli, although concentrations were lower than those of the fecal samples. Further, TGC^r E. coli were isolated from six of the eight feed samples tested. TET^r E. coli and TGC^r E. coli were isolated from feed samples collected from feedlot cattle (Agga et al., 2016b), suggesting that feed is an important vehicle for the introduction of AMR bacteria into animal production facilities.

Conclusions

Similar fecal concentrations of AMR bacteria and ARGs from food production animals and environments were found on a mink farm. The high prevalence of TGC^r E. coli and ARGs, including ESBLs, conferring resistance to critically important and highest-priority drugs for human medicine such as TGC and fluroquinolones is alarming. The study also indicated the importance of mink feed as a vehicle of dissemination of AMR bacteria, including TGC^r E. coli and genes of public health importance to mink farms and their potential spread to the environment. The results of this study may serve as a basis for more studies to develop recommendations for the prudent use of antimicrobials in mink production.

Footnotes

Acknowledgments

The authors thank Rohan Parekh, Anna Wilkin, and Maggie Mesker for their technical support.

Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by the U.S. Department of Agriculture, Agricultural Research Service (Project No. 5040-12630-006-00D).

Supplementary Material

Supplementary Figure S1

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

References

Agga

, Arthur

, Durso

, Harhay

, Schmidt

. Antimicrobial-resistant bacterial populations and antimicrobial resistance genes obtained from environments impacted by livestock and municipal waste. PLoS One, 2015; 10:e0132586.

Agga

, Schmidt

, Arthur

. Antimicrobial-resistant fecal bacteria from ceftiofur-treated and nonantimicrobial-treated comingled beef cows at a cow-calf operation. Microb Drug Resist, 2016a;22:598–608.

Agga

, Schmidt

, Arthur

. Effects of in-feed chlortetracycline prophylaxis in beef cattle on animal health and antimicrobial-resistant Escherichia coli . Appl Environ Microbiol, 2016b;82:7197–7204.

Bauernfeind

, Schneider

, Jungwirth

, Sahly

, Ullmann

. A novel type of AmpC beta-lactamase, ACC-1, produced by a Klebsiella pneumoniae strain causing nosocomial pneumonia. Antimicrob Agents Chemother, 1999; 43:1924–1931.

Bradford

PA.

Extended-spectrum β-lactamases in the 21st century: Characterization, epidemiology, and detection of this important resistance threat. Clin Microbiol Rev, 2001; 14:933–951.

Bush

Past and present perspectives on beta-lactamases. Antimicrob Agents Chemother, 2018; 62:ee01076-18.

Bush

, Jacoby

. Updated functional classification of beta-lactamases. Antimicrob Agents Chemother, 2010; 54:969–976.

Cavaco

, Hasman

, Xia

, Aarestrup

. qnrD, a novel gene conferring transferable quinolone resistance in Salmonella enterica serovar Kentucky and Bovismorbificans strains of human origin. Antimicrob Agents Chemother, 2009; 53:603–608.

[CDC] The Centers for Disease Control and Prevention. National Antimicrobial Resistance Monitoring System for Enteric Bacteria (NARMS): Human isolates Surveillance Report for 2015 (Final Report). Atlanta, GA: U.S. Department of Health and Human Services, 2018.

10.

Chriél

, Harslund

, Jensen

, Jakobsen

, Hammer

, Lassen-Nielsen

. Antimicrobials in mink-consumption and resistance patterns. In: Larsen PF, et al., eds. Proceedings of the Xth International Scientific Congress in fur animal production. Wageningen, Netherlands: Wageningen Academic Publishers, 2012.

11.

CLSI. Performance Standards for Antimicrobial Susceptibility Testing. 30^th ed. CLSI supplement M100. Wayne, PA: Clinical and Laboratory Standards Institute, 2020.

12.

Compo

, Pearl

, Tapscott

, et al. On-farm biosecurity practices and causes of preweaning mortality in Canadian commercial mink kits. Acta Vet Scand, 2017; 59:57.

13.

Cormier

, Chalmers

, Cook

, et al. Presence and diversity of extended-spectrum cephalosporin resistance among Escherichia coli from urban wastewater and feedlot cattle, in Alberta, Canada. Microb Drug Resist, 2020; 26:300–309.

14.

Cottell

, Kanwar

, Castillo-Courtade

, et al. bla _CTX-M-32 on an IncN plasmid in Escherichia coli from beef cattle in the United States. Antimicrob Agents Chemother, 2013; 57:1096–1097.

15.

FCUSA. Book 1: Standard Guidelines for the Operation of Mink Farms in the United States. Fur Commission USA, 2019. Available at: https://furcommission.com/wp-content/uploads/2019/10/SGOM-Book-1_Criteria-Forms-ProtocolsG.pdf (accessed December 10, 2019 ).

16.

FDA. 2018 Summary Report on Antimicrobials Sold or Distributed for Use in Food-Producing Animals. Center for Veterinary Medicine, U.S. Food and Drug Adminstration (FDA), Department of Health and Human Services, 2019.

17.

FDA. Animal drugs at FDA. 2020. Available at: https://animaldrugsatfda.fda.gov/adafda/views/#/home/previewsearch/012-375 (accessed February 19, 2020 ).

18.

Geyer

, Reisbig

, Hanson

. Development of a TaqMan multiplex PCR assay for detection of plasmid-mediated ampC beta-lactamase genes. J Clin Microbiol, 2012; 50:3722–3725.

19.

Graham

, Bergeron

, Bourassa

, et al. Complexities in understanding antimicrobial resistance across domesticated animal, human, and environmental systems. Ann N Y Acad Sci, 2019; 1441:17–30.

20.

Hansen

HO.

The global fur industry: Trends, globalization and specialization. J Agric Sci Technol A, 2014; 4:543–551.

21.

Hansen

, Larsen

, Skov

, et al. Livestock-associated methicillin-resistant Staphylococcus aureus is widespread in farmed mink (Neovison vison). Vet Microbiol, 2017; 207:44–49.

22.

Horakova

, Mlejnkova

, Mlejnek

. Specific detection of Escherichia coli isolated from water samples using polymerase chain reaction targeting four genes: Cytochrome bd complex, lactose permease, beta-d-glucuronidase, and beta-d-galactosidase. J Appl Microbiol, 2008; 105:970–976.

23.

Hovland

A-L

, Ahola

, Malmkvist

Behaviour of foxes and mink kept for fur production. In: The Ethology of Domestic Animals: An Introductory Text. Jensen P, ed. Wallingford, United Kingdom: CABI, 2017, pp. 255–269.

24.

Jensen

, Sommer

, Struve

, Clausen

, Chriel

. Factors associated with usage of antimicrobials in commercial mink (Neovison vison) production in Denmark. Prev Vet Med, 2016; 126:170–182.

25.

Jensen

, Sommer

, Struve

, Clausen

, Chriel

. A cross-sectional field study on potential associations between feed quality measures and usage of antimicrobials in commercial mink (Neovison vison). Prev Vet Med, 2017; 143:54–60.

26.

Koike

, Aminov

, Yannarell

, et al. Molecular ecology of macrolide–lincosamide–streptogramin B methylases in waste lagoons and subsurface waters associated with swine production. Microb Ecol, 2010; 59:487–498.

27.

Kozak

, Boerlin

, Janecko

, Reid-Smith

, Jardine

. Antimicrobial resistance in Escherichia coli isolates from swine and wild small mammals in the proximity of swine farms and in natural environments in Ontario, Canada. Appl Environ Microbiol, 2009; 75:559–566.

28.

Miller

, Vikram

, Agga

, Arthur

, Schmidt

. Effects of in-feed chlortetracycline prophylaxis in beef cattle on antimicrobial resistance genes. Sci Rep, 2018; 15:689–697.

29.

, Martin

, Alfa

, Mulvey

. Multiplex PCR for the detection of tetracycline resistant genes. Mol Cell Probes, 2001; 15:209–215.

30.

Nguyen

, Phan

, Duong

, Bertrand

, Lenski

. Effects of carriage and expression of the Tn10 tetracycline-resistance operon on the fitness of Escherichia coli K12. Mol Biol Evol, 1989; 6:213–225.

31.

Nikolaisen

, Lassen

DCK

, Chriel

, Larsen

, Jensen

, Pedersen

. Antimicrobial resistance among pathogenic bacteria from mink (Neovison vison) in Denmark. Acta Vet Scand, 2017; 59:60.

32.

Pedersen

, Hammer

, Sorensen

, Heuer

. Usage of antimicrobials and occurrence of antimicrobial resistance among bacteria from mink. Vet Microbiol, 2009; 133:115–122.

33.

Pei

, Kim

S-C

, Carlson

, Pruden

. Effect of river landscape on the sediment concentrations of antibiotics and corresponding antibiotic resistance genes (ARG). Water Res, 2006; 40:2427–2435.

34.

Peirano

, Ahmed-Bentley

, Woodford

, Pitout

. New Delhi metallo-beta-lactamase from traveler returning to Canada. Emerg Infect Dis, 2011; 17:242–244.

35.

Philippon

, Arlet

, Jacoby

. Plasmid-determined AmpC-type beta-lactamases. Antimicrob Agents Chemother, 2002; 46:1–11.

36.

Qiu

, Jiang

, Ju

, et al. Molecular and phenotypic characteristics of Escherichia coli isolates from farmed minks in Zhucheng, China. Biomed Res Int, 2019; 2019:3917841.

37.

Rashidi

, Zhu

, Kaiser

, et al. Vancomycin-resistance gene cluster, vanC, in the gut microbiome of acute leukemia patients undergoing intensive chemotherapy. PLoS One, 2019; 14:e0223890.

38.

Roberts

MC.

Environmental macrolide-lincosamide-streptogramin and tetracycline resistant bacteria. Front Microbiol, 2011; 2:40.

39.

Roberts

, Schwarz

. Tetracycline and phenicol resistance genes and mechanisms: Importance for agriculture, the environment, and humans. J Environ Qual, 2016; 45:576–592.

40.

Roberts

, Sutcliffe

, Courvalin

, Jensen

, Rood

, Seppala

. Nomenclature for macrolide and macrolide-lincosamide-streptogramin B resistance determinants. Antimicrob Agents Chemother, 1999; 43:2823–2830.

41.

Robicsek

, Strahilevitz

, Sahm

, Jacoby

, Hooper

. qnr prevalence in ceftazidime-resistant Enterobacteriaceae isolates from the United States. Antimicrob Agents Chemother, 2006; 50:2872–2874.

42.

Szczepanowski

, Linke

, Krahn

, et al. Detection of 140 clinically relevant antibiotic-resistance genes in the plasmid metagenome of wastewater treatment plant bacteria showing reduced susceptibility to selected antibiotics. Microbiology, 2009; 155:2306–2319.

43.

Travis

, Aulerich

Mink require savvy to raise; market is a roller coaster. In: Living on a Few Acress. Hayes J, ed. Washington, D.C.: U.S. Department of Agriculture, 1978, pp. 330–333.

44.

Vikram

, Rovira

, Agga

, et al. Impact of “raised without antibiotics” beef cattle production practices on occurrences of antimicrobial resistance. Appl Environ Microbiol, 2017; 83:AEM.01682-17.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB

0.02 MB

Third-Generation Cephalosporin- and Tetracycline-Resistant Escherichia coli and Antimicrobial Resistance Genes from Metagenomes of Mink Feces and Feed

Abstract

Introduction

Materials and Methods

Isolation and characterization of TETr and TGCr E. coli

Detection and quantification of antibiotic resistance genes

Results

Concentration and prevalence of TETr- and TGCr-E. coli

Prevalence and quantity of ARGs from metagenomic DNA of fecal and feed samples

Discussion

Conclusions

Footnotes

Acknowledgments

Disclosure Statement

Funding Information

Supplementary Material

References

Supplementary Material

Isolation and characterization of TET^r and TGC^r E. coli

Concentration and prevalence of TET^r- and TGC^r-E. coli