Shiga toxin-producing Escherichia coli O157 in piglets and food from backyard systems.

Piglets suffer from diarrhea caused by the Shiga toxin-producing Escherichia coli (STEC) and can be carriers of the bacteria, with public health consequences in developing countries. The aim of the present study was to study the prevalence of STEC O157 in feces of 465 piglets and 54 food mixes from backyard systems, the antimicrobial susceptibility of STEC and the frequency of genes encoding extended-spectrum β-lactamases. The E. coli was isolated from 75.90 % of the evaluated feces. The STEC strains were identified in 33.11% of the sampled population and in 43.60% of the piglets carrying E. coli. Among STEC strains, the stx1 gene was the most frequent (22.30%). The rfbO157 gene was amplified in 47.40% of the STEC strains. High frequencies of STEC strains were not susceptible to ampicillin, carbenicillin and tetracycline. The blaTEM gene (52) was the most frequent among strains not susceptible to ampicillin. Class 1 integrons were the most frequent in those strains. Of the identified STEC strains, 48.70% were considered as multi-drug resistant and 1.90% were considered extensively drug resistant. In the supplied food, STEC O157 strains were identified in 25.00% of the STEC strains. We conclude that the piglets from backyard systems are carriers of STEC O157 strains not susceptible to common antibiotics, including penicillins and tetracyclines. In addition, supplied food is a source of this type of pathogenic bacteria. Through their direct contact with humans, the piglets and food represent a potential source of bacterial dissemination capable of producing gastrointestinal infections in humans.

Introduction

Escherichia coli is a zoonotic agent with the greatest impact on swine systems, causing acute enteritis with watery diarrhea in post-weaning piglets.[1] The high mortality and medication of piglets cause economic losses ranging from 25.00 to 50.00% of the profits of swine systems. [2] The pathogenic strains of E. coli in pigs include Shiga toxin-producing E. coli (STEC).[3] The serotype O157:H7 of STEC has become an important pathogen causing diarrhea, hemorrhagic colitis and hemolytic uremic syndrome in humans throughout the world.[4] Currently, the main concern of researchers is the increasing spread of E. coli strains carrying a group of β-lactamases known as extended-spectrum β-lactamases (ESBLs) having the ability to cause resistance to new beta-lactam antibiotics and other antibiotic families.[5],[6]

Some studies have examined the prevalence of O157 and non-O157 STEC strains in piglets in Mexico.[7],[8] However, most of these studies were carried out with piglets of intensive and semi-intensive swine farms, in which there is a strict control of sanitary measures and not much has been studied in swine backyard systems.

Swine backyard farming is a production system characteristic of certain regions of Mexico and other countries in the world representing a source of income and animal protein for families in rural communities. Despite its benefits, this type of systems is associated with a high risk of contamination by pathogens and an inappropriate use of antibiotics. [9]

Whether the piglets and food supplied in backyard systems can carry STEC O157 with antibiotic resistance mediated by ESBL has not been previously investigated and it was the main motivation for this study. The goal of this study was to investigate the prevalence of STEC O157 in piglets’ feces and food supplied in backyard systems in Chiapas, Mexico, the antimicrobial susceptibility of STEC strains and the frequency of genes encoding ESBL.

Materials and Methods

Study population. The study population consisted of 465 healthy hybrid (Yorkshire x Duroc) piglets (from 1 to 6 weeks of age) being randomly selected. The study period lasted from the winter of 2016 to the summer of 2018. Animal studies were approved by the Ethics Committee of the University of Sciences and Arts of Chiapas (approval #049/02-2018). All piglets came from backyard farms being located in Chiapas, Mexico. The average number of pigs in each pen was between one and ten. These piglets were maintained in pens made of wooden or masonry walls having roofs made from metal sheets or locally found materials, dirt or concrete floors, simple drinkers and feeders made from hollowed-out trunks, with no waste treatment system in place. Agricultural residues and food waste from homes, restaurants and markets constitute the food source. Fifty-four samples (150 g each) of a food mix (tortillas, fruits and vegetables) were collected, placed in sterilized plastic bottles and transported to the laboratory.

Isolation of The samples were taken directly from the rectum of piglets using sterile swabs and placed in Stuart's medium (Copan Diagnostics, Murrieta, USA); the swabs and bottles were transported in cold chain for microbiological analysis. Sub-samples (100 g) from each sample of food mix were placed in a sterile plastic bag and lactose broth was added to reach a 1:10 (10-1) final dilution.[10] These sub-samples were mixed for 1 min. The swabs and foods dilution were simultaneously inoculated onto the eosin-methylene blue agar (Sigma-Aldrich, Steinheim, Germany) and incubated at 37.00 ˚C for 24 hr. The lactose-fermenting colonies were subjected to the series of biochemical tests to confirm the identity of E. coli strains. The identity of E. coli strains was also genetically confirmed by polymerase chain reaction (PCR) through the amplification of uidA gene.[11] MacConkey agar with sorbitol (Sigma-Aldrich) was used to detect the serotype O157:H7 of STEC.[12]

Identification of STEC by PCR. In the PCR, the E. coli strain ATCC® 25922TM was used as a negative control and the STEC EDL933 strain (O157:H7) as a positive control. The strains were provided by Dr. Teresa Estrada García of CINVESTAV, Mexico, and deposited in the bacterial collection of University of Sciences and Arts of Chiapas, Chiapas, Mexico.[13] To obtain DNA, bacterial lysates from each of the previously selected colonies were prepared, suspended in 1.00 mL of deionized water and then boiled for 10 min. The bacterial lysates were centrifuged at 10,000 rpm for 5 min; the supernatant containing DNA was removed and stored at – 80.00 ˚C. The gene primers specific for Shiga toxin-producing E. coli (stx1 and stx2 genes) were amplified by PCR.[14] Shiga toxin-producing E. coli O157 was studied by amplifying the rfb gene (specific O-polysaccharide).[15] The amplification of ESBL genes blaTEM, blaSHV, blaCTXM, blaOXA and blaCMY was carried out using the primers and conditions reported previously.16 The primers for amplification of class 1 and 2 integrons genes were used as described by Mazel et al., and White et al., respectively.[17],[18] The primer sequences used in this study are provided in Table 1. The PCR reactions were run in a thermal cycler C1000 (Bio-Rad Laboratories, Hercules, USA) and the PCR products were analyzed through agarose gel electrophoresis (2.00%) at 80.00 V for 1 hr. The agarose gels were stained with Sybr Green® (Invitrogen, Carlsbad, USA) and visualized with the Molecular Imager® Gel DocTM XR System (Bio-Rad). Lambda molecular weight markers (10 and 1000 bp; Invitrogen) were also used.

Table 1

Primers used in this study

Primer pair	Sequence (5’-3’)	Encoded protein	Size (pb)	Reference
uidA	F: AAAACGGCAAGAAAAAGCAG	β-glucuronidase	147
	R: ACGCGTGGTTAACAGTCTTGCG
Stx1	F: CTGGATTTAATGTCGCATAGTG	Shiga toxin 1	150
	R: AGAACGCCCACTGAGATCATC
Stx2	F: GGCACTGTCTGAAACTGCTCC	Shiga toxin 2	255
	R: TCGCCAGTTATCTGACATTCTG
rfb O157	F: CGGACATCCATGTGATATGG	Specific O-polysaccharide	259
	TTGCCTATGTACAGCTAATCC
blaTEM	F: ATAAAATTCTTGAAGACGAAA	beta-lactamase TEM	1080
	R: GACAGTTACCAATGCTTAATC
blaSHV	F: TTATCTCCCTGTTAGCCACC	beta-lactamase SHV	795
	R: GATTTGCTGATTTCGCTCGG
blaCTXM	F: CGCTTTGCGATGTGCAG	beta-lactamase SHV	550
	R: ACCGCGATATCGTTGGT
blaOXA	F: TCAACTTTCAAGATCGCA	beta-lactamase OXA	591
	R: GTGTGTTTAGAATGGTGA
blaCMY	F: GACAGCCTCTTTCTCCACA	beta-lactamase CMY	1000
	R: TGG AACGAAGGCTACGTA
IntI	F: GGGTCAAGGATCTGGATTTCG	intI1	483
	R: ACATGCGTGTAAATCATCGTCG
Int2	F: CGGGATCCCGGACGGCATGCACGATTTGTA	class 2 integron	variable
	R: GATGCCATCGCAAGTACGAG

Antimicrobial susceptibility analysis. The disk diffusion method was performed following the recommendations of Clinical and Laboratory Standards Institute.[19] The following antimicrobial susceptibility discs (BD BBL™ Sensi-Disc™, Becton, Dickinson and Company, San Jose, USA) were used for different anti-microbial categories: β-lactamic: ampicillin (10.00 µg), carbenicillin (100 µg) and oxacillin (1.00 µg), aminoglycosides: amikacin (30.00 µg), netilmicin (30.00 µg) and gentamicin (10.00 µg), cephalosporins: cefalotin (30.00 µg) and cefotaxime (30.00 µg), quinolones: ciprofloxacin (5.00 µg) and norfloxacin (10.00 µg), phenicols: chloramphenicol (30.00 µg), folate inhibitors: trimethoprim-sulfamethoxazole (25.00 µg), furans: nitrofurantoin (300 µg) and tetracyclines: tetracycline (30.00 µg). The β-lactam-resistant strains were subsequently analyzed using the disc diffusion method with amoxicillin-clavulanic acid discs (20.00/10.00 µg). The E. coli strains (intermediate and resistant phenotypes) not susceptible to at least three antibiotics belonging to different antimicrobial categories were classified as multi-drug resistant strains (MDRs); while, the strains not susceptible to at least one antibiotic belonging to each of the tested antimicrobial categories were classified as extensively drug resistant (XDR).[20]

Results

A total of 353 (75.90%) strains of E. coli were isolated from fecal samples collected from 465 piglets. The STEC strains were identified in 33.11% (154/465) of the sampled piglets and detected in 43.60% (154/353) of the piglets carrying E. coli. Among STEC strains, 22.30% (79/154) strains were carriers of the stx1 gene, 6.80% (24/154) of the stx2 gene and 14.40% (51/154) of both the stx1 and the stx2 genes. The rfbO157 genetic marker was amplified by PCR in 47.40% (73/154) of the STEC strains. In the supplied food, STEC strains were isolated from 14.80% (8/54) of the sampled food mix and identified in 16.00% (8/50) of the sampled food mix carrying E. coli; while, the rfbO157 genetic marker was amplified in 25.00% (2/8) of the STEC strains.

Antimicrobial susceptibility profile. Susceptibility of STEC strains to antibiotics used to treat gastrointestinal infections caused by E. coli was evaluated. More than three thirds of the identified STEC strains were not susceptible to ampicillin and carbenicillin. Half of the strains were not susceptible to tetracycline. The susceptibility of O157 strain was also evaluated; more than three thirds of the strains were not susceptible to ampicillin, carbenicillin and tetracycline (Table 2). The frequency of genes encoding β-lactamase in all 142 STEC strains not susceptible to ampicillin was also assessed. The blaTEM gene (52) was the most frequent among STEC strains, followed by blaCTX (25) and blaSHV (8). Seventeen STEC strains not susceptible to ampicillin turned out to be carriers of both the blaTEM genes and blaCTX genes; while, five strains were carriers of the blaTEM, blaCTX and blaSHV genes (Table 3).

Table 2

Antimicrobial non-susceptibility profile of the Shiga toxin-producing Escherichia coli (STEC) strains

Antimicrobial	Percentage of non-susceptibility (n)
	STEC (n = 154)	stx 1 (n = 79)	stx 2 (n = 24)	stx 1 /stx 2 (n = 51)	O157 (n = 73)
Ampicillin	81.10 (125)	89.80 (71)	83.30 (20)	66.60 (34)	91.70 (67)
Amoxicillin-clavulanic acid	25.30 (39)	26.50 (21)	20.80 (5)	25.50 (13)	43.80 (32)
Carbenicillin	66.80 (103)	67.10 (53)	70.80 (17)	64.70 (33)	82.10 (60)
Oxacillin	5.80 (9)	3.80 (3)	8.30 (2)	7.80 (4)	10.90 (8)
Amikacin	20.70 (32)	17.70 (14)	25.00 (6)	23.50 (12)	36.90 (27)
Gentamicin	15.60 (24)	12.60 (10)	20.80 (5)	17.60 (9)	28.70 (21)
Netilmicin	11.70 (18)	11.40 (9)	16.70 (4)	9.80 (5)	19.10 (14)
Cefalotin	26.60 (41)	25.30 (20)	33.30 (8)	25.50 (13)	42.40 (31)
Cefotaxime	11.00 (17)	11.40 (9)	12.50 (3)	9.80 (5)	23.20 (17)
Ciprofloxacin	11.70 (18)	11.40 (9)	16.70 (4)	9.80 (5)	21.90 (16)
Norfloxacin	14.90 (24)	16.40 (13)	25.00 (6)	9.80 (5)	30.40 (22)
Chloramphenicol	27.90 (43)	32.90 (26)	20.80 (5)	23.50 (12)	34.20 (25)
Trimethoprim-sulfamethoxazole	21.40 (33)	25.30 (20)	29.00 (7)	11.70 (6)	35.60 (26)
Nitrofurantoin	8.40 (12)	10.30 (8)	8.30 (2)	3.90 (2)	12.30 (9)
Tetracycline	48.00 (74)	43.00 (34)	62.50 (15)	49.00 (25)	83.50 (61)

Table 3

Genes of extended spectrum β-lactamase producing Shiga toxin-producing Escherichia coli (STEC) strains isolated from piglets

STEC groups (n)	Non-susceptible profile β-lactamic: Ampicillin ‎	β-lactamase gene	Number of isolates
STEC stx1 (79)	71	TEM CTX SHV TEM+CTX TEM+CTX+SHV	3213582
STEC stx2 (24)	20	TEM CTX SHV TEM+CTX TEM+CTX+SHV	54132
STEC stx1/stx2 (51)	51	TEM CTX SHV TEM+CTX TEM+CTX+SHV	158261

Of the identified STEC strains, 48.70% (n = 75) were not susceptible to at least one antibiotic in three different anti-microbial categories; these strains were considered as MDR. Also, 1.90% (n = 3) of STEC strains, predominantly STEC stx2, were not susceptible to at least one antibiotic in all tested categories; these strains were considered XDR. Class 1 integrons were detected in 74 STEC strains from 142 isolates not susceptible to ampicillin. Class 2 integrons were not detected (Table 4). The susceptibility of STEC strains isolated in food mix was evaluated. All the identified STEC strains were not susceptible to ampicillin and carbenicillin. Eight STEC strains not susceptible to ampicillin turned out to be carriers of the blaTEM gene; while, two STEC strains not susceptible to ampicillin turned out to be carriers of both the blaTEM genes and blaCTX genes. Of the identified STEC strains (n = 8), two strains were considered as MDR, class 1 integrons were detected in four STEC strains and class 2 integrons were not detected (data not showed).

Table 4

Non susceptible profiles in STEC strains isolates from piglets

Class (No.)		Non susceptible phenotype	No.	ESBL gene (No.)			Integron class	Genetic marker
STEC stx1 (n=79)
(0)		0	4
(1)		AMP	9	TEM (4), CTX (5)
(1)		CAR	1
(1)		AMP CAR	13	TEM (7), CTX (6)
(2)		AMP STX	1				Class 1
(2)		AMP CEF	1
(2)		AMP CHL	1
(2)		AMP NIT	1
(2)		AMP CAR CEF	2	TEM			Class 1
(2)		AMP AMK CAR	2	TEM			Class 1
MDR (3)		AMP STX TET	1	CTX			Class 1	O157
(2)		AMP CAR TET	1	TEM			Class 1
MDR (3)		CHL STX TET	1				Class 1	O157
(2)		AMP CAR CHL	3	TEM (3)
MDR (3)		AMP CHL TET	1	TEM			Class 1
MDR (3)		AMP CAR CHL TET	1	TEM			Class 1	O157
MDR (3)		CAR CEF CTX NOR	1					O157
MDR (3)		AMP AMC CHL TET	1	TEM			Class 1	O157
MDR (3)		AMP CAR STX TET	1	TEM			Class 1	O157
MDR (3)		AMP CAR CHL TET	2	SHV (1)			Class 1
MDR (4)		CAR CEF CHL TET	1				Class 1	O157
(2)		AMP AMC CAR CEF	1	TEM				O157
MDR (3)		AMP AMK CEF CTX	1	TEM				O157
MDR (4)		AMP CHL STX TET	1	CTX			Class 1	O157
MDR (3)		AMP CAR CTX TET	1	TEM			Class 1	O157
MDR (3)		AMP AMC CAR STX TET	1	TEM			Class 1	O157
MDR (3)		AMP AMC AMK CAR TET	1	CTX			Class 1	O157
MDR (3)		AMP CAR CEF CTX NET	1	SHV				O157
MDR (4)		AMP AMK CEF STX TET	1	TEM			Class 1	O157
MDR (5)		AMP CAR NOR CHL STX TET	1	TEM			Class 1	O157
MDR (3)		AMP AMC CAR CIP NOR TET	1	CTX			Class 1	O157
MDR (5)		AMP AMC AMK CEF STX TET	1	SHV			Class 1	O157
MDR (4)		AMP AMC CAR CEF CHL STX	1	TEM			Class 1
MDR (4)		AMP AMK CAR GEN NET CTX TET	1	TEM			Class 1	O157
MDR (5)		AMP AMC CAR AMK GEN CIP NOR	1	CTX				O157
MDR (5)		AMP AMC CAR AMK GEN CEF NET	1	TEM			Class 1	O157
MDR (6)		AMP AMC CAR CHL STX NIT TET	3	TEM+CTX+ SHV			Class 1
MDR (5)		AMP AMC CAR OXA CEF CHL STX TET	1	TEM+CTX			Class 1
MDR (6)		AMP NET CTX CIP NOR CHL TET	1	TEM			Class 1	O157
MDR (6)		AMP AMC CAR AMK GEN CEF CHL TET	1	CTX			Class 1	O157
MDR (6)		AMP AMC CAR CEF CHL STX NIT TET	1	TEM			Class 1	O157
MDR (7)		AMP CAR GEN CEF CIP NOR STX NIT TET	1	TEM+CTX			Class 1	O157
MDR (5)		AMP CAR GEN NET CEF CTX CIP NOR TET	1	TEM+CTX			Class 1	O157
MDR (6)		AMP AMC CAR AMK CIP NOR CHL NIT TET	1	TEM+CTX			Class 1	O157
MDR (5)		AMP AMC CAR AMK GEN NET CEF NOR TET	1	TEM+CTX+SHV			Class 1	O157
MDR (7)		AMP AMC CAR AMK NET NOR CHL STX NIT TET	1	TEM+CTX			Class 1	O157
MDR (7)		AMP AMC CAR OXA GEN NET CIP NOR CHL STX TET	1	TEM+CTX			Class 1	O157
MDR (7)		AMP AMC CAR OXA AMK GEN CEF CTX CIP NOR CHL STX TET	1	TEM+CTX			Class 1	O157
MDR (7)		AMP AMC CAR GEN NET CEF CTX CIP NOR CHL STX TET	1	TEM+CTX+SHV			Class 1	O157
STEC stx2 (24)
(0)		0	(2)
(1)		AMP	(4)
(1)		AMP CAR	(1)
MDR (3)		CAR CHL TET	(1)				Class 1	O157
(2)		AMP CAR TET	(2)	TEM			Class 1	O157
MDR (3)		CAR NOR CHL	(1)					O157
MDR (3)		AMP CAR NOR STX	(1)	TEM			Class 1	O157
MDR (3)	AMP CAR NET TET				(1)	CTX			Class 1	O157
MDR (3)	AMP CAR CEF TET				(2)	TEM (1), CTX (1)			Class 1
MDR (3)	AMP CAR SXT TET				(1)	SHV			Class 1
MDR (4)	AMP CEF SXT TET				(1)	CTX			Class 1	O157
MDR (3)	AMP CAR CEF TET				(1)	TEM			Class 1	O157
MDR (4)	AMP AMC CAR AMK GEN TET				(1)	TEM+CTX			Class 1	O157
MDR (4)	AMP AMK CAR GEN STX TET				(1)	CTX			Class 1	O157
MDR (6)	AMP AMC CAR OXA AMK GEN NET CEF CTX CIP NOR CHL TET				(1)	TEM+CTX			Class 1	O157
MDR (6)	AMP AMC CAR AMK GEN NET CEF CTX CIP NOR SXT TET				(1)	TEM+CTX+SHV			Class 1	O157
XDR (8)	AMP AMC CAR AMK NET CEF CIP NOR CHL SXT NIT TET				(1)	TEM+CTX			Class 1	O157
XDR (8)	AMP AMC CAR OXA AMK GEN CEF CTX CIP NOR CHL SXT NIT TET				(1)	TEM+CTX+SHV			Class 1	O157
STEC stx1/stx2 (n=51)
(0)	0				(10)
(1)	AMP				(3)
(1)	CHL				(2)
(1)	CAR				(2)
(1)	AMP CAR				(3)	TEM (3)
(2)	CEF TET				(1)
(1)	AMP AMK CAR				(1)	CTX				O157
(2)	AMP AMK GEN				(1)	CTX
(2)	AMP CAR GEN				(1)	CTX
(2)	AMP CAR TET				(1)	TEM			Class 1	O157
(2)	AMP CAR CHL				(1)	CTX				O157
(2)	AMP CAR GEN NET				(1)	CTX				O157
MDR (3)	AMP CAR CIP TET				(1)	SHV			Class 1	O157
MDR (3)	AMP CAR CEF TET				(1)	SHV			Class 1	O157
MDR (4)	CAR CHL STX TET				(1)				Class 1	O157
MDR (3)	AMP CAR CHL TET				(2)	TEM (2)			Class 1	O157
MDR (4)	AMP CAR AMK CEF TET				(1)	TEM			Class 1	O157
MDR (4)	AMP CAR CIP STX TET				(1)	TEM			Class 1	O157
MDR (4)	AMP CAR AMK GEN TET				(1)	CTX			Class 1	O157
MDR (3)	AMP AMC CAR OXA CIP TET				(1)	CTX			Class 1	O157
MDR (4)	AMP CAR CHL STX TET				(1)	TEM			Class 1	O157
MDR (3)	AMP CAR GEN NET CEF				(1)	TEM			Class 1	O157
MDR (4)	AMP CAR AMK CTX TET				(1)	CTX			Class 1	O157
MDR (3)	AMP AMC CAR AMK GEN TET				(1)	TEM			Class 1
MDR (4)	AMP AMC AMK CEF CTX TET				(1)	TEM			Class 1	O157
MDR (4)	AMP AMC CAR CHL NIT TET				(1)	TEM+CTX			Class 1	O157
MDR (5)	AMP AMC CAR CEF NOR STX TET				(1)	TEM			Class 1	O157
MDR (4)	AMP AMC CAR OXA CIP NOR CHL TET				(1)	TEM+CTX			Class 1	O157
MDR (5)	AMP AMC CAR AMK CEF CHL TET				(1)	TEM+CTX			Class 1	O157
MDR (5)	AMP AMC CAR AMK CEF NOR TET				(1)	TEM			Class 1	O157
MDR (4)	AMC AMK CAR CEF CTX CHL TET				(1)	TEM			Class 1	O157
MDR (5)	AMP AMC CAR GEN NET CEF STX TET				(1)	TEM+CTX			Class 1	O157
MDR (5)	AMP AMC CAR AMK CEF CTX NOR TET				(1)	TEM+CTX			Class 1	O157
MDR (5)	AMP AMC CAR OXA AMK GEN NET CEF NOR TET				(1)	TEM+CTX			Class 1	O157
XDR (8)	AMP AMC CAR OXA GEN NET CEF CTX CIP CHL STX NIT TET				(1)	TEM+CTX+SHV			Class 1	O157

MDR; Multi Drug-Resistant, XDR; Extensively Drug-Resistant; AMP: Ampicillin, AMC: Amoxicillin, AMK: Amikacin, CAR: Carbenicillin, OXA: Oxacillin, GEN: Gentamicin, NET: Netilmicin, CEF: Cefalotin, NOR: Norfloxacin, CTX: Cefotaxime CIP: Ciprofloxacin, CHL: Chloramphenicol, STX: Trimethoprim-sulfamethoxazole, NIT: Nitrofurantoin TET: Tetracycline.

Discussion

Swine backyard farming systems are common in Mexico and developing countries worldwide. However, important problems have been described in these production systems, such as the lack of adequate technologies and technical assistance, which leads to a high prevalence of diseases.[21] This is the first study conducted in Mexico that reports the presence of STEC (43.60%) carrying the stx1 and/or stx2 genes in piglets from backyard systems. Of these STEC strains, 47.40% amplified the rfbO157 genetic marker. In contrast, another study has shown low presence of the stx1 and stx2 genes (0.10% and 1.00%, respectively) in STEC strains isolated from piglets of farms located in the central region of Mexico.7 In this context, a low prevalence (2.10%) of E. coli O157 was reported in pigs from farms located in the central region of Mexico.[8] Unlike intensive and semi- intensive swine systems, in which there is a strict control of the personnel and application of sanitary measures, backyard systems are characterized by poor animal health management and, in many cases, no biosecurity measures, explaining the contrast in these results.

Our hypothesis is that the acquisition and dissemination of STEC O157 and non-O157 strains in piglets from back-yard systems could be related to the origin of the food provided to pigs and to direct contact between pigs, humans and pets. Unlike specialized farms, in backyard systems the pigs diet is based on fruit and vegetable waste, stale tortillas and bread, etc. This variety of ingredients is associated with a greater variability of the intestinal bacterial population, which is considered beneficial to the health of host.[22] However, there is a high risk of conta-mination with the pathogens present in the pigs’ food due to poor sanitary management. Pathotypes of diarrheagenic E. coli, including STEC, have been identified in ready-to-eat cooked vegetable salads (1.40%) distributed by restaurants in Mexico[23] and tomatoes (6.00%) purchased from public markets in Pachuca, Mexico.[24] These results are consistent with the findings of our work, suggesting that STEC could be acquired and disseminated through the vegetable waste provided as a feed to pigs. Although the sample size analyzed here was small, we did detect STEC O157 in supplied food, indicating that these foods represent a potential source of bacterial dissemination for piglets. Swine backyard farming systems are characterized by the involvement of women and other family members in animal management activities as well as people outside the family during the sale process. Direct contact between humans and animals is a major factor in the spread of STEC, especially in developing countries with a high prevalence of gastrointestinal infections in humans caused by diarrheagenic E. coli.[25],[26] Moreover, the presence of pets and other domestic animals (cattle, sheep and birds) inside the house as well as proliferation of harmful fauna are also factors involving in the spread of harmful germs, since this type of animals are important reservoirs of diarrheagenic E. coli, including the O157:H7 serotype, participating in gastrointestinal infections in humans.[27],[28]

In this work, STEC strains were resistant mainly to ampicillin (81.10%), followed by carbenicillin (66.80%) and tetracycline (48.00%). This trend was similar for STEC O157. In addition, about half of the identified STEC strains were resistant to at least one antibiotic in three of the tested antimicrobial categories (MDR). The results of the present work showed a high frequency of resistance mainly to β-lactams in STEC strains isolated form piglets of backyard systems, similar to what being reported by other authors in intensive and semi-intensive swine systems. For example, high resistance to tetracycline (79.57%) and ampicillin (48.79%) was reported in STEC strains isolated from pig feces of farms located in the city of Chongqing, China.[29] Recently, the phenotype of resistance to ampicillin (99.50%) and carbenicillin (99.00%) was identified in STEC strains isolated from pigs of farms located in central Thailand.[30] It has been reported that some beta-lactam antibiotics, such as penicillin and ampicillin, lose viability when use as a first-line of choice during chemotherapeutic treatment of an infectious process affecting pigs throughout the world due to the acquisition of resistance mechanisms.[5] The present study demonstrated the presence of genes encoding β-lactamase in STEC strains isolated from piglets of backyard systems, mainly the blaTEM (52), blaCTX (25) and blaSHV (8) genes. Moreover, class 1 integrons were also identified. These findings are consistent with those reported by Samanta et al.[31] The presence of E. coli with high resistance to ampicillin is common in piglets due to the presence of ESBLs blaCTX-M and blaTEM and class 1 integrons.[32],[33] The results confirm that backyard piglets can be a carrier of ESBL- producing E. coli; however, further studies regarding the presence of specific bla profile are suggested. Bacteria carrying class 1 integrons play a role in the spread of resistance genes and pose a serious health risk to humans if transmitted to them.[34]

The present study showed that piglets from backyard systems are carriers of STEC O157 and non-O157 strains not susceptible to penicillins and tetracyclines. It also showed that the most of these strains have genes that code ESBLs, mainly blaTEM, blaCTX and blaSHV. In addition, we showed that STEC O157 and non-O157 could be acquired and disseminated through the food mix provided to pigs. These results could be used for the development of more efficient preventive measures, diagnostic methods and antimicrobial alternatives in swine backyard farming systems, in order to reduce a risk for public health.

Conflicts of interest

The authors declare that the study was carried out in the absence of commercial or financial relationships that could be interpreted as a potential conflict of interest and all persons gave their informed consent prior to their inclusion in the study.