Extended-Spectrum β-Lactamase-Producing Escherichia coli Isolated from Food-Producing Animals in Tamaulipas, Mexico

Extended-spectrum β-lactamase (ESBL)-producing E. coli has become an important global problem for the public health sector. This study aims to investigate the E. coli antimicrobial resistance profile among living food-producing animals in Tamaulipas, Mexico. A total of 200 fecal samples were collected from bovines, pigs, chickens and sheep. A total of 5.0% of the strains were phenotypically confirmed as ESBL producers. A high percentage of phenotypic antimicrobial resistance was observed against gentamicin (93.3%), tetracycline (86.6%) and streptomycin (83.3%). The gentamicin-resistant strains showed MDR, distributed among 27 resistance patterns to different antimicrobials. The antimicrobial resistance gene tet(A) was detected in 73.3% of isolates, aadA1 in 60.0% and sul2 in 43.3% of strains. The blaCTX-M gene was found in 23.3% of strains. The virulence gene hlyA was detected in 43.3% of isolates; stx1 and stx2 were not detected in any strain. The phylotyping indicated that the isolates belonged to groups A (33.3%), B1 (16.6%), B2 (40.0%) and D (10.0%). These results show that food-producing animals might be a reservoir of ESBL-producing bacteria and may play a role in their spread.


Introduction
Antimicrobial resistance (AMR) is considered to be one of the main threats to public health worldwide [1]. In particular, the extended-spectrum β-lactamase (ESBL)-producing strains stand out due to their plasmid-encoded enzymes that are capable of providing resistance to most beta-lactam antibiotics, including penicillin, cephalosporins and monobactams. In addition, they are often resistant to other classes of antibiotics such as aminoglycosides, fluoroquinolones and trimethoprim/sulfamethoxazole, making them more difficult to treat [2][3][4]. Infections with ESBL-producing Enterobacteriaceae are associated with worse clinical outcomes, higher mortality rates, extended hospital stays and more significant expenses compared with similar infections with bacteria that do not produce ESBL [3]. Most of this problem can be attributed to the spread of ESBL-producing Escherichia coli and Klebsiella pneumonia. They are not only restricted to healthcare-associated isolates confined within clinical settings since these strains can be commonly isolated from water, soil, domestic animals and food-producing animals [5][6][7]. As a better approach to this health problem, derived from antibiotic-resistant bacteria and their transmission models to humans, the "One Health" concept was implemented. The One Health initiative aims to achieve optimal health outcomes by recognizing the interconnectedness between people, animals, plants and their shared environment [5]. Several potential sources of ESBL-producing E. coli (ESBL-EC) for human infection or colonization were identified, including food-producing animals [6]. Although several studies concerning ESBL strains were carried out in hospital settings, environmental reservoirs recently received attention, making it challenging to estimate their potential risk to public health and control their spread. Specifically, the isolation of ESBL-producing E. coli from livestock has been increasingly reported worldwide due to the continuous use of antimicrobials in livestock, promoting the emergence of multidrug-resistant pathogens [8]. For example, in Mexico, some studies reported the presence of ESBL strains in meat retailed to the public [9,10]; however, more information is needed regarding the prevalence of these strains during meat handling or in living livestock prior to slaughter. In previous studies on livestock from Mexico, a high percentage of antimicrobial resistance and MDR were reported; however, these studies did not include information concerning the presence of ESBL strains [11][12][13][14]. For this reason, the present study aims to investigate the prevalence of ESBL-producing E. coli (ESBL-EC) isolated from living food-producing animals and their antimicrobial resistance profiles in Tamaulipas, Mexico.

Identification of ESBL-EC
From the 200 samples tested, 600 E. coli strains were isolated in our study from 50 bovines, 50 chickens, 50 pigs and 50 sheep. The samples were collected in the central region of Tamaulipas, Mexico. The overall prevalence of ESBL-EC samples was 11.5% (95%, Cl: 6.0-16.0%), and the ESBL-positive strains as a percentage of the total number of E. coli strains was 5.0% (95%, Cl: 2.0-7.3%); the highest percentages of the ESBL strains were observed in samples from pigs and chickens, at 7.3% and 7.3%, respectively (Table 1).

Antimicrobial Susceptibility
Among the 30 ESBL-EC strains that were analyzed for this study, 100% (30/30) showed resistance to at least one antibiotic tested. In contrast, 93.3% (28/30) of the isolates exhibited a multidrug-resistance phenotype (resistant to ≥3 groups of antimicrobials). In addition, 27 phenotypic resistance patterns were observed, of which 1 was repeated in four strains, and the remaining 26 were unique and different (  (Table 3). From the analysis used to detect the presence of genes related to antimicrobial resistance, we found that the tetA gene was the most prevalent gene in the 30 ESBL-EC strains; it was revealed that 73.3% (22/30) of the isolates had the aadA1 gene, and this gene determined a resistance to aminoglycosides in 60.0% (18/30) of cases (Table 3). On the other hand, the results show that 50% (15/30) of the analyzed ESBL-EC strains harbored class 1 integrons (intl1). Class 2 and 3 integrons (intl2 and intl3) were not detected in any of the analyzed strains.

Detection of Virulence Factors
All of the tested strains were negative in the analysis performed to detect the stx1 and stx2 genes. The hylA gene was identified in 43.3% (13/30) of the ESBL-EC strains, in which the amplified band of 569 pb was detected; most of these were isolated from the pigs, only five were isolated from the sheep, and one was isolated from the chicken samples.

Detection of Virulence Factors
All of the tested strains were negative in the analysis performed to detect the stx1 and stx2 genes. The hylA gene was identified in 43.3% (13/30) of the ESBL-EC strains, in which the amplified band of 569 pb was detected; most of these were isolated from the pigs, only five were isolated from the sheep, and one was isolated from the chicken samples.

Discussion
E. coli is an interesting study model given its omnipresence in nature as a commensal and pathogen, being one of the primary vehicles that can transmit resistance and virulence genes between different species [15]. The monitoring of antibiotic-resistant E. coli strains from food animals and products is vital to determine its potential risk to humans [16]. This study found that 5.0% of strains recovered from livestock samples (cattle, pigs, chickens and sheep) were ESBL-EC. These results are similar to those reported by Benavides et al. (2021) in Chile of 3.0% (cattle, pigs, chickens, sheep and goats) [5]. However, if the prevalence of ESBL-EC is considered only in chicken samples, Sanou et al. (2022) [17] found a 7.8% incidence of ESBL-EC in chickens, similar to our results (7.3% in chicken samples). However, the percentage incidence of ESBL-EC was much higher in most other studies, reportedly between 3 and 68% [17][18][19][20][21]. An example of this includes the results published by Li [20] or Sghaier et al. (2019) in Tunisia, who reported 51.6% [21]. On the other hand, considering only the pig samples in our current study, we detected a 7.3% incidence of ESBL-EC. However, in other similar studies, the prevalence was higher; for example, the results published by Sanou et al. (2022) in Africa reported an ESBL-EC incidence of 63% [17], Giufre et al. in Italy reported 27% [20] and Miltgen et al. (2022) in Reunion Island reported 28.2% [22]. The high ESBL percentages reported in some studies are not surprising, given that β-lactam antibiotics are widely used in livestock. It is necessary to consider that farm animals are frequently exposed to the use of antibiotics, such as β-lactams, therapeutically (to treat clinically sick animals), for prophylaxis (given to healthy animals at risk of infection to prevent it from occurring), for metaphylaxis (to treat diseased animals in the same group as healthy animals) and, in some countries, for growth-promoting purposes (as a feed additive) [23]. The percentages of antibiotic resistance can vary from region to region due to the legislative measures on the use of antibiotics in each region or the type of management to which the animals are subjected. Considering the comparison of the current results with similar studies in other countries, the samples included from Tamaulipas exhibited a low percentage of ESBL-EC strains. However, it is essential to remember that, although it may seem to be a low percentage, these strains can spread their antibiotic resistance to other bacteria and can represent a risk to public health if the food produced by infected livestock is improperly handled.
In addition to the antibiotics in the β-lactam group, among the antimicrobials most commonly used in animal production are the tetracyclines, phenicols (chloramphenicol) and aminoglycosides (streptomycin) [24]. Most ESBL-EC strains isolated in this study were shown to be multidrug resistant (resistant to ≥3 groups of antibiotics). This may be due to the excessive use of antibiotics in livestock that can induce and accelerate the development of resistance in bacteria. The ESBL-EC strains isolated in this study showed a high rate of co-resistance to antibiotics, such as gentamicin in 93.3% of strains (28/30), tetracycline in 86.6% of strains (26/30) and streptomycin in 83.3% of strains (25/30). This high percentage of gentamicin-resistant ESBL-EC strains may be of interest, considering that it is an alternative recommended antibiotic for the treatment of some types of ESBL infections [25]. Although carbapenem treatment is considered the "gold standard" for severe and invasive ESBL Enterobacteriaceae infections [26], a combination regimen with aminoglycosides is also used [27,28]. However, the activity of aminoglycosides against ESBL-producing Enterobacteriaceae varies according to the geographical region [28]. As far as our review goes, only two previous studies focused on resistant bacteria in livestock were published in Tamaulipas , which showed 6.5% ESBL-EC, and reference [10]. Although these percentages of antibioticresistant strains in meat are lower than those obtained in livestock for the current study, they cannot be considered entirely comparable. It should be noted that although the meat samples were for sale in Tamaulipas, not all originated from local livestock production, since some businesses bring the meat from other regions. However, the discussion regarding the presence of antibiotic-resistant strains in Tamaulipas, whether or not they originated from local livestock, is relevant, since they play an essential role in the distribution of antibiotic resistance genes (ARG). At the same time, ARGs with a food or animal origin can be transferred to other bacteria via horizontal gene transfer (HGT), which plays a key role in acquiring, accumulating and disseminating ARGs to the most virulent bacteria [23,29] that can cause infection in humans.
The health risk associated with the spread of antibiotic resistance in the environment is estimated using the multiple antibiotic resistance index (MARI) [16,30]. In this study, all of the values obtained show that the MARI was greater than 0.2, suggesting that the ESBL-EC strains originated from an environment with a high contamination or overuse of antibiotics. The resistance genes detected in the ESBL-EC isolates via PCR revealed that the CTX-M gene was the most prevalent β-lactam gene (23.3%; 7/30), which corresponds to the phenotypic resistance to the third and fourth generation cephalosporins. Usually, bla CTX-M genes are located on transferable plasmids, which could spread among animal, environmental and human E. coli isolates [31]. Regarding the genes associated with resistance to non-β-lactam antibiotics, tetA was the most prevalent gene in the ESBL-EC strains. The high percentage of phenotypic resistance to tetracycline (86.6%) and genotypic detection (tetA 73.3% and tetB 13.3%) in these ESBL-EC strains is not surprising, since it is an antibiotic that is widely used in animal and human infections due to its wide availability and low cost. On the other hand, although quinolones and β-lactams are widely used worldwide for the treatment of many infectious diseases [32], none of the ESBL strains contained the qnrA gene, and only 20% had the qnrB gene. The qnr genes are usually integrated multimers, and plasmids often harbor other antibiotic resistance genes such as ESBLs, favoring their selection and dissemination [32,33]. This is considered to be a promoter of the spread of multidrug resistance. In the resistance capacity to antibiotics developed by some bacteria, integrons play an important role. They are the genetic platform enabling the bacteria to capture, store and reorder antibiotic resistance cassettes through site-specific recombination, facilitating their dissemination [34]. This study shows that 93.3% (28/30) of the ESBL-EC strains were MDR; of these, 50% (15/30) contained class 1 integrons, indicating that they were strains capable of dispersing antibiotic resistance. The class-1 integron-positive isolates were resistant to aminoglycosides (aadA1 14/15, strB 8/15), tetracyclines (tetA 15/15) and sulfonamides (sul1 8/15, sul2 10/15). Integrons, as mobile elements, can transmit the resistance genes from one organism to another; this represents an extremely important challenge in livestock since the performance of infection control programs is poor [35].
To determine the potential pathogenicity of the strains, we used the phylogenetic characterization system developed by Clermont et al. [35] to classify the strains into four phylogroups (A, B1, B2 and D). The strains belonging to the various phylogroups differ in their genome size, variable gene content, disease association, ecological niche and life history characteristics [36]. For this reason, the use of the phylogroup classification was employed in the study of ecological niches and lifestyles in bacterial pathogens, and improves our understanding of the population structure, providing invaluable epidemiological information [37]. Studies have shown that strains associated with virulent extraintestinal infection usually belong to phylogeny groups B2 or D, and that the commensal E. coli isolates are generally affiliated with groups A and B1 [38][39][40][41].
In this study, the ESBL-EC strains were classified as follows: 33.3% (10/30) were classified as phylogroup A, which is strongly associated with human sources [37]. A further 16.6% (5/30) were classified as phylogroup B1, which has a significant relationship with food sources [37]. Moreover, 40% (12/30) of the strains were in phylogroup B2, which is associated with herbivorous and omnivorous mammals, and is considered the leading cause of extraintestinal infections in humans and diarrheal diseases [37]. Finally, 10% (3/30) of the strains were classified as phylogroup D, which is also related to human sources [37]. Furthermore, it is interesting to note that in phylogroup B, 66.6% (8/12) of the strains had class 1 integrons, and 50% (6/12) had the hlyA gene.
To our knowledge, this is the first study on ESBL-EC strains in livestock from northeast Mexico (Tamaulipas state). On the other hand, the samples are obtained only from the central zone of the state, which could be considered a limitation, since the findings cannot be generalized to the entire state. However, being the first data generated in the state, they serve as an indicator of the current situation and a basis for continuing with studies of this type.
In general, the ESBL-EC strains were identified in the four livestock species analyzed; although it may seem to be a low percentage (5.0%), the presence of these strains always represents a focus of alert due to their role in the spread of the resistance to antibiotics. Five strains isolated from pigs simultaneously presented a class 1 integron and the hlyA gene, and all belonged to the B2 phylogroup and were MDR; therefore, they stand out as a potential risk to public health. These results highlight the importance of carrying out Antibiotics 2023, 12, 1010 7 of 11 this type of surveillance study in livestock to determine the current situation of bacterial resistance, the virulence factors present and the risk they represent to the population.

Identification of ESBL-EC
The study was conducted between January and September 2021 in the central region of the state of Tamaulipas, Mexico. Fecal samples were collected from adult animals, including cattle, chickens, pigs and sheep. Bovine and sheep samples were acquired via rectal retrieval using disposable gloves, while chicken and pig samples were recovered from cloacal swabs. All samples collected were aseptically manipulated, labelled and stored individually for transport to the laboratory. To isolate E. coli, each sample was inoculated into lactose broth (BD Difco™), homogenized and incubated at 37 • C for 24 h. Subsequently, one loop of the culture was streaked onto eosin-methylene blue (EMB) agar (BD Becton Dickinson and Co., Mexico) plates and incubated at 37 • C for 18-24 h. After incubation, three presumptive colonies with characteristics corresponding to E. coli morphology were randomly selected. Each colony was transferred to tryptic soy agar (TSA) (BD Becton Dickinson and Co) plates and incubated for 24 h at 37 • C to obtain a pure culture. Standard biochemical tests were applied to confirm the identity of the E. coli, including methyl red, Voges-Proskauer, lactose and sugar fermentation, indole and motility production and citrate metabolism. The identification was made via MALDI-TOF mass spectrophotometry and PCR. The DNA was obtained from a pure culture on tryptic soy agar (BD Becton Dickinson and Co., Cuautitlán Izcalli, Mexico) via lysis of a bacterial cell suspension at 95 • C for 15 min, followed by centrifugation at 13,000× g for 3 min. The mdh gene was employed to confirm the identification of E. coli as described by Vasquez et al. [42]. All strains that were confirmed to be E. coli were subjected to the phenotypic screening of ESBL production using the double-disk synergy test according to the European Committee on Antimicrobial Susceptibility Testing guidelines [43].

Phylogenetic Groups
ESBL-producing isolates were classified in phylogenetic groups by amplifying fragments of the chuA (279 bp) and yjaA (211 bp) genes, and the DNA fragment TspE4.C2 (152 bp) as described by Clermont et al. [16]. Briefly, groups were assigned on the basis of different combinations of the presence and/or absence of the three amplicons (A, B1, B2 and D).

Conclusions
The presence of ESBL-EC strains in livestock used for food production warrants a focus of attention on the problem of bacterial resistance. This is particularly important as our study found that a high percentage of the tested ESBL-EC strains in livestock used for food production were MDR. In Mexico and in the northeast of the country, there have been few studies that address this issue in animals. Consequently, there is no thorough understanding of the potential transmission of these strains. An interesting point to highlight from the results of this study is the presence of ESBL-EC strains in combination with the class 1 integron and hlyA gene, and its classification within the B2 phylogroup (considered pathogenic). This may represent a potential risk to public health and a niche method of distribution of ARGs to other bacterial strains. Results such as these indicate the importance of monitoring bacterial resistance and its reservoirs, which contribute to a One Health vision that helps us to gain a better understanding of the transmission of bacterial resistance.