RESISTANCE TO ANTIBIOTICS BY ENTERIC BACTERIA ASSOCIATED WITH THE SWINE INDUSTRY: IN SILICO EXPLORATION OF THE DISTRIBUTION OF RESISTANCE GENES

Multidrug-resistant (MDR) bacteria are a signi ﬁ cant contributor to the global antibiotic resistance crisis, which is predicted to kill more people than cancer by 2050. Livestock production is a contributing factor as it has been one of the fastest-growing industries in the previous century but has the most harmful effects on the environment and human health. The pig is the most widely raised and consumed food-producing animal globally, with an upward trend. The derived residues and the meat products constitute an important reservoir of antibiotic resistance genes (ARGs) that can be transmitted to humans through consumption, direct contact, the environment, or poor handling, leading to relevant zoonotic diseases, especially enteric ones. It is essential to know the diversity, abundance, and distribution of ARGs to have better control and monitoring of their dispersion. In the present study, the ARGs and Mobile Genetic Elements (MGEs) of ﬁ ve enteric and pathogenic species commonly present in the microbiota of both pigs and humans were examined by bioinformatic analysis. This analysis showed that 157 ARGs were distributed across 1869 genomes of ﬁ ve bacterial species, ranked from highest to lowest diversity of ARGs: Klebsiella pneumoniae , Escherichia coli , Enterococcus faecium , Salmonella enterica , and Enterococcus faecalis . This study contributes to better management of antibiotics, which directly impact the health of both humans and animals.


INTRODUCTION
Antibiotic resistance is a global problem that is increasing dangerously.It is caused by selection of bacterial strains through the continuous exposure to these drugs [1].The intensive use of antibiotics in livestock (Figure 1) and fi sh farming creates conditions for developing and transferring resistance genes between bacteria in animals, humans, and the environment.It is a common practice to administer antibiotics to healthy animals for prophylactic and growth promotion purposes [2].Livestock production is one of the principal causes of antibiotic resistance worldwide.Among the different types of livestock production, swine (Sus scrofa domesticus) becomes relevant since its great global magnitude (110.975 million tons forecast for 2023) [3] and for being the livestock industry that consumes the highest quantity of antibiotics [2][3][4].Also, a reasonably common practice throughout the world is the reuse of livestock biosolids as organic fertilizer in agriculture.However, it is estimated that between 75% and 90% of antibiotics are excreted without being metabolized by animals [5] with the inherent risk of spreading multi-resistant bacteria, resistance genes, and antibiotic residues that can spread across the environment [6,7].
Some studies demonstrate and d escribe the relationships between the overuse of antibiotics, the presence of antimicrobial resistance genes (ARG), and the range to which they can spread between different hosts and environments when carried by Mobile Genetic Elements (MGE) via horizontal transfer [8][9][10].In 2020, Yang et al. [11] highlighted the potential role of pig manure as an essential reservoir of ARGs and MGEs.They analyzed manure from pig farms in China by qPCR and revealed that, in both feces and treated manure, class 1 and 2 integrons as well as conjugative plasmids, were present.Furthermore, in the United Stat es, Muurinen et al. [12] investigated the infl uence of antibiotics and alternative growth promoters on the pig fecal resistome, the mobility of ARGs, and their relationship with MGEs.Given the fact that horizontal gene transfer is induced by subtherapeutic concentrations of antibiotics and the activation of bacterial stress response mechanisms.In addition, antibiotics and alternative growth promoters were also found to strengthen the relationships between ARGs and MGEs directly and proportionally.
It is essential to know that, despite some anatomical and physiological differences, the brain development, immunology, microbiota, and gastrointestinal tract of pigs and humans are similar although the pig is considered a gnotobiotic model for studying gastrointestinal diseases or the response of the normal intestinal microbiota to various factors [13,14].For instance, Wang et al. [15] performed a comparative metagenomic analysis to characterize the gut microbiota and resistome of pigs and humans.They found 27 shared ARGs, among which the most abundant (between 59 and 60%) were resistant to tetracycline, macrolides, and streptogramin B. Furthermore, it was revealed that MGEs promoted their dispersal.Most of the ARGs came from the Gamma proteobacteria class, which harbors many of the most widely described multidrugresistant bacteria.
To better understand the prevalence, diversity, and distribution of resistance genes, this study performed a descriptive and comparative bioinformatic analysis of fi ve enteric and pathogenic bacteria commonly found in porcine isolates of Enterococcus faecalis, Enterococcus faecium, Escherichia coli, Klebsiella pneumoniae, and Salmonella enterica, based in the information available in public databases of genomes, assemblies, or genomic readings.

MATERIALS AND METHODS
395 genomes of bacteria isolated from pigs were downloaded from the NCBI database.The main criterium for selecting the genomes was that they had gene/ protein annotations.32 genomes correspond to Enterococcus faecalis, 201 to Enterococcus faecium, 41 to Escherichia coli, 31 to Klebsiella pneumoniae, and 90 to Salmonella enterica.
As a fi rst step, a randomly selected subset of 11 genomes per species were scanned with ResFinder 4.1 [16] and MGEfi nder 1.0.3 [17] in order to obtain a representative group of sequences of ARGs and MGEs.The fi nal quantity of genomes to be scanned with Resfi nder and MGEfi nder was determined, adding one by one until no newer ARGs or MGEs were found.A total of 18 genomes per species were suffi cient to get the representative group of ARGs and MGEs.We used the following parameters in ResFinder: Threshold for %ID, 85%; Minimum length, 60%; Type of reads, Assembled genomes or contigs.
All sequences, a total of 365, were merged in a single fi le, and a multiple sequence alignment was performed to establish the similarity between the antibiotic resistance genes of the different bacterial strains and discard the redundant ones to diminish the number of sequences used in the BLAST search.
The remaining 106 unique sequences were used for a broad BLAST search in all the 395 downloaded genomes.We used a standalone BLAST version for Linux installed locally on a desktop computer, a Perl script to organize data, and a spreadsheet to analyze the results.The cut-off values were for E-value maximum accepted value of 1e-18; coverture minimum accepted value of 40%; and identity minimum accepted value of 65%.

RESULTS
From the local BLAST search, it was found a signifi cant amount of beta-lactamases (~50%) followed by enzymes related to target modifi cation (~35%) and effl ux pumps (~15%).The total richness of gene families among the fi ve species of bacteria was 36, with Klebsiella pneumoniae being the one with the highest relative abundance per gene family (28 different families), followed by Escherichia coli (18), Enterococcus faecalis (9), Enterococcus faecium (7) and fi nally Salmonella enterica with seven.The information regarding each of the bacterial strains that were examined is shown in the Supplementary Material.Figure 2 details the gene richness of families of ARGs among the fi ve species of bacteria and shows the presence or absence of a gene.
A general comparison analysis of the total ARGs in all analyzed strains was also conducted to determine the ten most abundant ones (Table 1).It is evident that tetracycline resistance is the antibiotic group more represented (by Tet genes) with three multi-resistant genes (oqxA, oqxB, and msrC).
Regarding the MGEs, Table 2 shows their presence and distribution by class within the fi ve bacterial species.These data are comparable with those of Wang et al. [15], whose results on MGEs were mainly the presence of conjugative elements (integrative and mobilizable) and insertion sequences.The fact that in the present investigation all the bacteria had the insertion sequences as predominant could be explained because they are short chains of genetic material (less than 2.5 Kb of DNA) that can be more easily inserted into the DNA than into other larger elements [18].When the diversity (the connection between richness and abundance) of ARGs harbored in the examined bacteria are compared, a pattern coincides with the number of various resistance genes each bacterial species contains.K. pneumoniae leads the list of gene richness as it has evolved resistance to 12 different groups of antibiotics and many associated genes (Figure 3).Although E. faecalis has a higher ARG richness, E. faecium has more abundance, which results in greater diversity.Therefore, the specifi c order of these two species differs slightly when abundance or richness is considered independently of their diversity.

DISCUSSION
The bacterial species with the highest abundance of resistance genes was Klebsiella pneumoniae, followed by Escherichia coli, Enterococcus faecium, Enterococcus faecalis, and Salmonella enterica (Figure 2).Consistently, except for Enterococcus spp., the gene richness of ARGs among the bacterial species analyzed corresponded equally to their number of genes per genome, resulting in a very similar order of richness by species from highest to lowest.Although analyzed separately, these two variables constitute the local genetic diversity of studied species.
The fi ndings on the richness and abundance of ARGs among the fi ve bacterial species in this study are comparable to the work of Yang et al. [11].They detected and determined that the genes tetO, tetW, tetQ, sul1, sul2, oqxB, qnrS, and ermC are prevalent in liquid pig manure.In addition, Jarat et al. [19] reported genes belonging to the tet, qnr, and erm classes.However, the abundance of an ARG is not necessarily proportional to the frequency of the group to which it belongs (Table 1).This may be explained because there are numerous gene families that, however, can transfer the same or similar resistance.Moreover, the potential for spreading ARGs into the environment is mainly through water, air, or land routes.Interestingly, by interconnected processes involving agricultural and fi sh farming activities, the environment, humans, and wildlife as host and propagation vectors [2] are also spreading potentials.For example, Xiong et al. [20] revealed by HPLC-MS/MS and qPCR the result of the selective pressure of antibiotics on the ARGs of bacterial communities in freshwater sediments polluted with pig manure.They found that selective pressure favoured bacterial genera such as Escherichia and Salmonella.In descending order, tetracyclines, fl uoroquinolones, and sulfonamides were found in water sediment samples.Due to the chemical stability and partitioning capabilities of antibiotics, the sequence was likely reversed in water.When comparing the gene diversity (Figure 3), it is appreciated that Klebsiella pneumoniae is the leading species, with a higher abundance of each gene class.It is known that K. pneumoniae is one of the bacteria with the highest frequency and diversity of resistance genes.Even in this species, many ARGs were discovered before being identifi ed in other pathogens, and current scientifi c evidence suggests that spreading ARGs by horizontal transfer to other bacterial groups and environments is essential; consequently, this bacterium represents a signifi cant and constantly growing threat to public health [21,22].Similarly, together with K. pneumoniae, the bacterium E. faecalis, E. faecium, and E. coli belongs to the group of pathogens known as ESKAPE, which contains some of the known pathogens with the greatest multi-resistance to antibiotics.So, at this point, it is appropriate to notice the similarities between the two Enterococcus species, mainly at the richness level, presumably due to shared genes within the same taxonomic genus.
Concerning E. coli, the second most ARG-diverse bacteria observed (Figure 3), shows fl uoroquinolone resistance through the qnr genes, like K. pneumoniae.It also highlights the role these genes could play given their plasmid-mediated nature, which can be conjugative or integrative and, therefore, spread via horizontal transfer to other bacterial species, even with different ARGs [23].Likewise, gene mobility between groups of Enterobacterales and Salmonella, among others, has been described [24].This can be explained as the combined result of the selective pressure action exerted by the most frequently used antibiotics (such as tetracyclines, penicillins, macrolides, sulfonamides, and aminoglycosides) and the individual contribution made by each bacterial species.According to Li et al. [25], who tracked the most common ARGs in the air in samples around the world and found the blaTEM genes (to be the most frequent, Klebsiella pneumoniae), was the one with the most remarkable diversity of resistance genes and, in whose resistome, the bla genes stood out.
In 2020, Capita et al. [26] analyzed preparations of beef, pork, and poultry samples and found that, of 126 total strains, 63% were multi-resistant, 23% resistant, and only 14% were sensitive.Among the bacteria found to be resistant were E. coli, K. pneumoniae, and S. enterica.Torres et al. [27] determined that E. faecium and E. faecalis resist beta-lactams, streptogramins, glycopeptides, lincosamides, phenicols, macrolides and oxazolidinones carried by farm animals destined for human consumption.Nevertheless, many countries still use antibiotics such as tetracyclines, sulfonamides, streptogramins, and aminoglycosides as growth promoters in animal husbandry; better regulation or absolute restriction, as in developed countries, would try to minimize ARGs.
Hu et al. [28] previously conducted a comparative bioinformatic study using the Resfi nder platform between the ARGs harbored in the microbiota of humans, cattle, pigs, and chickens and found that 41 genes were transferred between animals and humans.
Besides that, 11 genes corresponded to the tetracycline class, ten to aminoglycosides, nine to macrolide-lincosamide-streptogramin B, fi ve to chloramphenicol, three to beta-lactams, and three to sulfonamides.They also observed that ARGs spread more readily among closely related bacteria in their phylogeny.They suggested that antibiotic exposure is the cause of selection, the environment is the physical barrier, and phylogeny determines the ability for horizontal transfer to occur through crossfertilization.This is similar to the results obtained in the present study regarding the classes of antibiotics for which the most signifi cant number of resistance genes were found and the similarity in the composition of ARGs between different species with a high phylogenetic relationship, as in the case of E. faecalis and E. faecium.The presence of the tet(M) and erm(B) genes, as well as the bla, aac, dfr, and van gene families were found to be consistent in addition to the results presented by Wang et al. [29] and the outcomes of our study.Despite the limitations of our fi ndings, this fact suggests the relevance of the horizontal transfer of ARGs, the overuse of the various antibiotic classes, and the interspecifi c scope that it can have, which is essential for humans and the ecological stability of the environment.Nevertheless, it is recommended to analyze the plasmids detected in these strains that interfere with carrying resistance genes to understand better.
Regarding the individual analysis of ARGs by class of antibiotics (Figure 2), these seem to be proportional to their greater or lesser use in the livestock industry (Figure 1).Tetracyclines (Tet genes) are the most widely used class of antibiotics in the world, and one of those with the oldest industrial and clinical development (1950); penicillins (blaTEM genes) are the second most used class in the European Union [30].Consequently, it is inferred and verifi ed [31] that the greater the overuse of antibiotics, the more selective pressure is established to enhance antimicrobial resistance.
Currently, antibiotics in animal husbandry are used mainly to promote growth and not for therapeutic purposes, which generates continuous and biologically unnecessary overexposure to them [32].Therefore, one of the measures with the most signifi cant impact in preventing the spread of multi-resistance to antibiotics should be adequate control, legislation, and practical limitations regarding their sale and administration in humans and for small and large-scale veterinary use.Despite the existence of various and recent restrictions on its use as growth promotion in cattle raising (European Union, 2006; the United States, 2017; Brazil and China, 2020), its use in this sector in some countries is around 80% of the total antibiotics consumed [12,33,34].Through the European Medicines Agency (EMA) and the Food and Drugs Administration (FDA), Europe and the United States are the political regions with the most stringent administration and regulation of the use of antibiotics [3,35].
The metagenomic study of six different habitats by Zhang et al. [36] reported the presence of 2561 ARGs, which confer resistance to 24 classes of antibiotics from 4572 samples.This work verifi ed that 23.78% of the ARGs found are risky to health, particularly those that confer multi-resistance.Likewise, this comprehensive analysis exposed the distribution patterns of ARGs in the world.In contrast, First World regions such as Europe and the United States of America stood out for being the ones that concentrated them the most, a fact that concurs with being the countries with the highest livestock production.
In Mexico, there is evidence about the clinical importance and resistance to antibiotics associated with the bacteria discussed in the current article such that K. pneumoniae is reported as a producer of extended-spectrum beta-lactamases, which confers resistance to beta-lactams third-generation cephalosporins.Moreover, E. faecalis and E. faecium, have disclosed resistance to ampicillin (beta-lactam), imipenem (beta-lactam), and vancomycin (glycopeptide), as well as E. coli to quinolones [37].In addition, no law governs the presence of antibiotics in meat products since the one that existed was dissolved in 2014; however, it does not currently encompass a prohibition on the use of antibiotics as growth promoters [38].As a result, it is recommended to raise awareness among the relevant authorities, encouraging them to become involved in the situation and to take part in decision-making processes that infl uence the effi cient and comprehensive containment of multidrug-resistant bacteria.
Ipso facto, despite the existence of bioinformatic research about ARGs with similar proceedings, this study constitutes an innovative investigation when comparing the resistance genes present in different bacterial species and by linking them up to the use of antibiotics for which they generate resistance; thus, it probably contributes to evidencing the relevance of these bacteria regarding antimicrobial resistance in the world.

CONCLUSION
Antibiotics continue to be widely used as growth enhancers in the livestock industry.
This study provides a bioinformatic analysis examining the genetic diversity of strains of fi ve major enteric and pathogenic bacteria isolated from porcine origins.In addition, evidence was provided that suggests an apparent correlation between the more outstanding livestock administration of the classes of antibiotics and the higher resistance reported against them by bacteria that contain ARGs against them.Despite the need for further confi rmatory studies, it is strongly recommended that nations evaluate and promote their legislation on the veterinary use of antibiotics to limit the spread of antibiotic resistance.

Figure 1 .
Figure 1.Percentage distribution of antibiotic classes approved for sale in animal production in the US (A) and Europe (B) (adapted from FDA, 2020 & EMA, 2020).

Figure 3 .
Figure 3. Diversity of ARGs in fi ve bacterial species.The richness and crossed abundance of ARGs in the fi ve bacterial species analyzed.Several genes provide multi-resistance, so the number of genes shown is relative to the antimicrobial class for which it generates resistance.

Table 1 .
ARGs most frequently detected in all bacterial strains, from highest to lowest.

Table 2 .
Comparison between the types of MGEs present in the bacterial strains analyzed in this study.