The term “mobilome” refers to all MGEs and genes including plasmids, integrons, transposons, insertion sequences (ISs), integrative and conjugative elements (ICEs), and resistance islands and/or genomic islands (GIs) in bacteria which can translocate within and transfer between genomes (Frost et al., 2005; Cabezón et al., 2017). MGEs play a key role in both the development and spread of AMR genes among the bacterial populations as well as in the evolution of bacterial genomes which rapidly respond to a selective pressure. These responses include alteration of antimicrobial exposure in different genera, species, and strains, including those of Salmonella isolates (Davies and Davies, 2010).

Understanding the transfer dynamics of AMR plasmids requires detailed monitoring of different pathogenic bacteria in both clinical and non-clinical environments. Plasmid-mediated transfer plays a critical role in the spread of AMR in S. enterica and related species (Han et al., 2012; San Millan et al., 2015; San Millan, 2018). Several studies and review articles have described the importance of plasmid dynamics in Gram-negative bacteria, in particular S. enterica, and many have utilized DNA sequencing to help examine resistance transfer (Zhao et al., 2003; Han et al., 2012; Lerminiaux and Cameron, 2019; McMillan et al., 2019). Baker et al. (2018) highlights the importance of WGS as a key technology for understanding the spatial dynamics of AMR evolution and for genomic epidemiology (Baker et al., 2018). Tables 1, 2 present multiple sequencing platforms and computational-based approaches that can be used for the detection of plasmid-based AMR genes.

In studies of MDR plasmid transfer in different Salmonella strains, isolates have been sequenced and diverse AMR genes associated with chloramphenicol, tetracyclines, ampicillin, streptomycin, kanamycin, and β-lactam antimicrobials have been identified (Han et al., 2012). When the rate of horizontal transfer outweighs the costs they impose on their bacterial hosts, plasmids are favored in the new hosts (Svara and Rankin, 2011). Stalder et al. (2017) reported that common mechanisms of plasmid stabilization in bacterial pathogens could influence plasmid-host coevolutionary dynamics (Stalder et al., 2017). Various concentrations of antimicrobials are encountered by bacteria in different environments and may affect the dynamics of the microbial population. For example, plasmid transfer between bacteria may lead to acquired resistance to different antimicrobials through the transfer of AMR determinants in response to selective pressure (Martinez, 2008). Svara and Rankin (2011) designed a model to examine the effect of antimicrobial treatment, specifically the dosage of antimicrobials and the interval between treatments, on the evolution of plasmid-borne resistance (Svara and Rankin, 2011). Their results showed that different treatment regimens (different interval between antimicrobial treatments and the dosage of antimicrobials) can select for either plasmid-carried or chromosome-carried resistance. In the absence of competing non-resistant plasmids, high transmission favors plasmid-borne resistance. While in the presence of other plasmids, plasmid-borne resistance was favored over chromosomal resistance in the low frequency antimicrobial treatments, but a high intensity of antimicrobial treatments is required for resistance plasmids to outcompete non-resistance plasmids (Svara and Rankin, 2011). Therefore, changes in natural ecosystems, including the release of large amounts of antimicrobials, might alter the population dynamics of antimicrobial resistant microorganisms, which will affect the evolution and dissemination of AMR in nature (Martinez, 2008). Taken together, these findings emphasize the necessity of new therapeutics to reduce the spread of AMR.

According to their transmissibility by conjugation, there are two classes of transmissible plasmids: (1) conjugative plasmids, which contain a full set of conjugation genes; and (2) mobilizable plasmids, which contain only a minimal set of genes that allow them to be mobilized by conjugation when they coexist in the same donor cell with a conjugative plasmid (Garcillan-Barcia et al., 2011). Conjugative plasmids possess a generally low copy number, while mobilizable plasmids tend to be a high copy number (Watve et al., 2010; van Hoek et al., 2011). Both conjugative and mobilizable plasmids can encode AMR genes and transfer them to a new host (van Hoek et al., 2011). Over the last decade, more than 1,000 unique plasmids have been identified in the family of Enterobacteriaceae, some of the most common ones include IncA/C, HI2, HI1, I1-γ, X, L/M, N, FIA, FIB, FIC, W, Y, P, T, K, B/O, FI1, U, R, ColE, and Q (Garcia-Fernandez and Carattoli, 2010). Among these, some Inc. types can be clustered into groups based on the genetic and pilus structure: (1) IncF group (consisting of IncF, D, J, and S); (2) IncI group (includes IncB, I, and K); and (3) IncP group (containing IncM, P, U, and W); and Ti (tumor inducing) that may have implications on plasmid transfer dynamics (van Hoek et al., 2011). Specifically, many of the plasmids that have been identified in Salmonella strains include representatives of IncA/C, F, H, I, L/M, N, R, and X groups (Fricke et al., 2009; Foley et al., 2013).

In the 1970s, IncA/C plasmids were initially described as large-molecular-weight (140–200 kb) and low copy number plasmids present in fish pathogens such as Aeromonas hydrophila and Vibrio spp. (Carraro et al., 2015; Ruggiero et al., 2018). Subsequently, IncA/C plasmids were found to encode resistance genes to antimicrobial agents in food animals and agricultural settings (Johnson and Lang, 2012). To date, research on conjugative IncA/C plasmids has determined that these plasmids play an essential role in the spread of MDR among the species of Enterobacteriaceae, including S. enterica and Vibrio cholerae, which is an important concern in the public health community (Carraro et al., 2015; Han et al., 2018). The core genetic backbones of IncA/C plasmids are generally highly conserved, with greater than 98% nucleotide identity across the plasmid backbone. Most examples of IncA/C plasmids among non-typhoidal S. enterica isolates in the United States carry transfer-associated genes which accelerate conjugal transfer (Welch et al., 2007; Fricke et al., 2009; Wiesner et al., 2011; Han et al., 2018; Ruggiero et al., 2018). Wiesner et al. (2011) described a MDR Salmonella Typhimurium ST213 strain in Mexico, which was associated with the carriage of IncA/C plasmids that carried a plasmid-borne bla_CMY-2 gene encoding resistance to extended-spectrum cephalosporins (Wiesner et al., 2011). As predicted by Han et al. (2018), many IncA/C plasmids encode their own conjugal machinery and are able to be transferred within bacterial communities. However, others without the full cohort of transfer genes required other conjugative plasmids in the bacterial cell for transfer of the plasmids (Han et al., 2018). Rankin et al. (2002) indicated that IncA/C plasmids were responsible for MDR phenotypes in Salmonella Newport isolates and their resistance genes could be transferred both to other Salmonella as well as other susceptible organisms under antimicrobial selection pressure in the same environments (Rankin et al., 2002). In S. enterica serovar Newport isolates, IncA/C plasmids emerged as the source of MDR phenotypes after 1980 (Johnson and Lang, 2012). Moreover, these plasmids are considered as broad host range plasmids due to their ability to spread by conjugative transfer among the bacterial communities, meanwhile under in vitro conditions some IncA/C plasmids exhibit variation in their ability to transfer between Escherichia coli and S. enterica (Johnson and Lang, 2012). In addition to their presence in food animal environments, cases have also been documented of plasmids spreading in hospital environments (Johnson and Lang, 2012). In some of these studies it was demonstrated that IncA/C plasmids have mobilizable genes conferring resistance to a number of different antimicrobial agents including β-lactams, aminoglycosides, chloramphenicol, sulfonamide, trimethoprim, quinolones, and tetracyclines (Carraro et al., 2015). These studies have been somewhat hampered by the presence of a large number of AMR elements carried on the plasmids. These AMR elements have been a challenge for studying their basic molecular biology and regulatory mechanisms because most of the molecular biology tools that are available rely on the antimicrobial selection for mutational studies (Johnson and Lang, 2012; Carraro et al., 2015).

Integrons are mobile genetic elements containing a site-recombination system that is able to integrate, express, and exchange specific DNA known as gene cassettes (Silva et al., 2012; Ravi et al., 2014). Gene cassettes are defined as the smallest mobile elements with an AMR gene; they are located between two recombination sites (attI and attC 59-base elements; Rodriguez et al., 2006; Gillings, 2014).

Generally, integrons contain three important major elements: (1) the integrase, which is a gene encoding integration into the host genetic backbone, (2) the recombination site, which serves as the primary site to capture the resistance gene cassettes, and (3) the promoter. Each of these elements is required for transcription and drives expression of gene cassettes (Rodriguez et al., 2006; Silva et al., 2012; Gillings, 2014). Integrons are divided into two different types: (1) mobile integrons (MIS), which carry a limited number of AMR gene cassettes (these include the class I integron) and are involved in the dissemination of AMR; and (2) chromosomal integrons (CIS), which have a different number of gene cassettes and are not involved in the dissemination of AMR (Domingues et al., 2012). Moreover, integrons are found in a wide diversity of clinical bacterial strains and environmental isolates in both chromosomes and within MGE (Gillings, 2013). The first discovered chromosomally-located superintegrons were identified in V. cholerae isolates (Ravi et al., 2014). These superintegrons are located on the chromosome and typically carry more than 20 gene cassettes with unknown function (Ravi et al., 2014). Integrons are immobile on their own and yet have been observed to transfer across numerous bacterial genomes (Gillings, 2014). Integrons are divided into five classes based on the type of gene capture mechanism, gene cassettes, and the similarity of sequence which encodes integrases to facilitate insertion of the target DNA (Gillings, 2013, 2014; Ravi et al., 2014). Each of the five different classes of integrons (class I–V) encodes a distinct integrase gene. The class I integrons play an important role in the spread of AMR genes and are the most-studied types from clinical samples (Gillings, 2013). They are referred to as broad host range as they have been found both in commensal and pathogenic bacteria (Ravi et al., 2014) and are widely found across different S. enterica isolates (Rodriguez et al., 2006).

Recent comparative studies described how integrons contribute to MDR phenotypes (Gillings, 2014; Ravi et al., 2014; Deng et al., 2015). The integron location within MGEs and the chromosome has a unique advantage for generating genomic and phenotypic diversity (Gillings, 2014). Three different classes of mobile integrons (class I, II, and III) can contribute to the MDR phenotypes (Mazel, 2006). For example, class I and class II integrons are associated with the bacteria from animals and have an impact on the human gastrointestinal microbiota by transmission of AMR through the food supply. The horizontal transfer of class I integrons to commensal organisms via conjugation facilitates the development and spread of resistant bacteria (de Paula et al., 2018; Miller and Harbottle, 2018). Class I integrons can also play a key role in the transmission of AMR in the aquatic environment and in fish pathogens, as well as in terrestrial pathogens such as S. enterica, E. coli, and other bacteria species (Miller and Harbottle, 2018). Evershed et al. (2009) discovered IncA/C plasmids, which play an important role in the dissemination of AMR genes, have an unusual cassette in their class I integrons, with most carrying the same set of resistance genes including: aacU, aphA, hph, sul2, and tetA(D) (Evershed et al., 2009). Similar integrons have also been identified in S. enterica serovars Senftenberg and Ohio. IncA/C plasmids have a broad host range, having been recovered from numerous species, their conjugal transfer capacities are variable (Evershed et al., 2009; Wiesner et al., 2013).

Moreover, Villa and Carattoli (2005) reported that some plasmids in MDR S. Typhimurium strains carry sprC, rck, and pefA virulence genes, in addition to two class I integrons carrying AMR genes within the Tn22 and Tn1696 transposons (Villa and Carattoli, 2005). Thus, it was shown that the association between MDR and virulence determinants may contribute to virulence plasmid evolution (Villa and Carattoli, 2005). Likewise, Rodriguez et al. (2006) reported that the class II integrons are present on self-transferable plasmids of Salmonella (Rodriguez et al., 2006). It is well-known, that the diversity of integrons plays a critical role in the spread of AMR genes and can drive evolution in S. enterica. Several integrons (both class I and class II) have been detected in clinical or environment isolates of Salmonella (Yeh et al., 2018) and other Enterobacteriaceae from different geographical locations (Adelowo et al., 2018; Gomi et al., 2018; Leungtongkam et al., 2018; Phoon et al., 2018).

Transposons, commonly referred to as “jumping gene systems,” provide functionality for the transfer of a segment of bacterial DNA to and from chromosomal or plasmid DNA and often carry AMR genes (Bennett, 2008; Giedraitiene et al., 2011; Sultan et al., 2018). Transposons are made up of a transposase and inverted repeat sequences on their ends. The inverted repeat sequences can accelerate target recognition and recombination; while the transposase catalyzes the movement of DNA segments to another part of the genome by a cut-and-paste mechanism, or by a replicative transposition mechanism, and functional genes (such as AMR genes; Giedraitiene et al., 2011). Furthermore, transposable elements (TEs) consist of a set of mobile elements that include small cryptic elements, IS elements, transposons, and transposing bacteriophages. The bacteriophage is a bacterial virus, which is also known as a transposing phage, that infects and replicates its own sequences within bacteria through either a lysogenic or lytic cycle (Balcazar, 2014). In the lysogenic cycle, the phage is referred to as temperate or non-virulent, and occurs following the injection of bacteriophage DNA into the bacteria cell. During this phase, the bacteriophage integrates its genome into the host genome via the phage-encoded integrases to form prophage, where it replicates passively as part of the host genome. As the phage excises from the host chromosome, it may transport bacterial genes from one bacteria strain to another via a process called transduction (Bennett, 2008; Giedraitiene et al., 2011). In the lytic cycle, also referred to as virulent infection, immediately after injecting into the host cell, the bacteriophage genome synthesizes what are termed early proteins instead of integrating its DNA into the host genome. These early proteins function to break down the host DNA allowing the bacteriophage to control the host’s cellular machinery to synthesize the proteins required to build new phage particles. Eventually, the destruction of the infected bacteria cell results in the release of the new phage progeny. Transduction mediated by temperate bacteriophages can serve as a horizontal gene transfer mechanisms and has played an important role in the dissemination of AMR genes among bacterial populations (Bennett, 2008; Giedraitiene et al., 2011; Colavecchio et al., 2017). Multiple bacteriophage and prophages have been reported in Salmonella strains, such as Fels-1, Gifsy-2, P22, and FelixO1 (Switt et al., 2013; Colavecchio et al., 2017). Among them, Salmonella phage P22 has been widely used in molecular biology for its ability to introduce foreign genes, including AMR genes into recipient cells by transduction (Switt et al., 2013). Since its emergence in the 1990s, the MDR ACSSuT-type (resistance to ampicillin, chloramphenicol, streptomycin, sulfonamide, and tetracycline) S. Typhimurium phage type DT104, has become widely distributed worldwide. Studies showed that the ACSSuT-type region of DT104 was derived from two separate evolutionary events. One was the integration of a 43-kb Salmonella genomic island (SGI) 1 that carries multiple AMR genes and the other is the integration of a P22-like phage into the chromosome to form prophage PDT17 or ST104 (Chiu et al., 2006). More information about the contribution of bacteriophages to the spread of AMR genes among foodborne pathogens can be found in a review article wrote by Colavecchio et al. (2017).

The sizes of transposon sequences range from 3 to 40 kb depending on the variability of the genes present in the transposons (Pagano et al., 2016). Figure 3 shows a representative transposon (Tn21) that is located within an IncA/C plasmid and carries a class 1 integron as well as multiple antimicrobial and disinfectant resistance genes (Han et al., 2012). There are two major types of TEs: (1) composite transposons (class I) and (2) complex transposons (class II). Class I transposons contain a variety of resistance genes and have little DNA homology in their central regions, which are flanked by IS elements (Pagano et al., 2016; Sultan et al., 2018). In S. enterica serovars, such as Typhimurium and Choleraesuis, class I transposons often reside in conjugative plasmids, such as IncHI2, and are responsible for AMR gene dissemination (Sultan et al., 2018). In contrast, class II transposons, constituting of three dissimilar, but interrelated families (Tn3, Tn21, and Tn2501), have more diverse genetic structures than class I transposons (Pagano et al., 2016; Sultan et al., 2018). A wide diversity of class I transposons (for example, Tn1696, Tn10, Tn6088, Tn3, Tn9, and Tn 1721-like transposon) carrying a variety of different AMR genes have been identified in Salmonella serovar (Frech and Schwarz, 1999; Orman et al., 2002; Cain et al., 2010; Cain and Hall, 2012; Chen et al., 2020). The TEs that are transferred include: (i) the integrative and conjugative elements (ICEs), (ii) the integrative and mobilizable elements (IMEs), and (iii) decayed elements derived from ICEs or IMEs, such as cis-mobilizable elements (CIMEs; Lao et al., 2020). CIMEs are decayed elements originating from ICEs or IMEs that have lost their integration and transfer genes, but retained attL and attR sites (Guedon et al., 2017). The CIMEs are known to be major vehicles for acquisition of a broad spectrum of AMR genes among bacteria and many other genes that can be advantageous for their hosts (Guedon et al., 2017). The recombination between related or unrelated ICEs, IMEs, and CIMEs, likely integrated in tandem, plays a major role in the evolution and plasticity of ICEs and IMEs (Bellanger et al., 2014; Lao et al., 2020).

In general, IS elements are small mobile elements (at least 0.5 kb) that perform as a complex type of transposable element and typically carry one or two transposase (Tnp) genes (Pagano et al., 2016; Partridge et al., 2018). These elements are defined as the simplest autonomous mobile elements in bacterial genomes (He et al., 2015). They are classified into different groups based on active site motifs in Tnp, designated by the key amino acids which are present in the active site including DDE (Asp and Glu), DEDO and H4H (a His split by a large collection of hydrophobic amino acids); and/or on whether transposition is a conservative, cut-and-paste mechanism, where the IS is simply excised from the donor and inserted into the recipient, or is replicative (Seth-Smith et al., 2012). The transposition can be replicated by a copy-and-paste mechanism or copy-out-paste-in mechanism. In the copy-and-paste mechanism, the IS is replicated to join the donor and recipient in a co-integrate, which is subsequently resolved to provide the original donor plus the recipient with the IS; while in the copy-out-paste-in mechanism, the IS is replicated into a double-stranded circular intermediate that then integrates into the recipient (Partridge et al., 2018). IS elements are widespread and can occur in very high numbers in bacterial genomes and have had an important impact on genome structure and function (Siguier et al., 2014). IS-mediated gene inactivation can affect bacterial virulence, AMR and metabolism (Vandecraen et al., 2017). For example, IS1 or IS10 insertion can up-regulate the AcrAB-TolC efflux pump in Salmonella, resulting in increased AMR (Ramirez-Santos et al., 1992; Siguier et al., 2014). IS1, which is approximately 768 bp, can be detected in the majority of the family Enterobacteriaceae including E. coli, Salmonella, Shigella, and Klebsiella (Ramirez-Santos et al., 1992). IS1 can be found in the chromosome and plasmids, respectively, and some strains carried IS1 in both the chromosome and plasmids (Ramirez-Santos et al., 1992). Ramirez-Santos et al. (1992) reported that the frequency of IS1 was higher in Salmonella than in E. coli and Shigella.

The relatively high prevalence of IS1 in plasmids of MDR clinical isolates suggests its role in the dissemination of AMR genes (Ramirez-Santos et al., 1992). ISPa12, a member of IS4 family, can induce expression of ESBL PER-1 in a series of Gram-negative isolates, including Salmonella, resulting in resistance to penicillins, cefotaxime, ceftibuten, ceftazidime, and aztreonam (Poirel et al., 2005). IS200 is a Salmonella-specific IS that is found in almost all Salmonella serotypes examined, but is absent from most other enteric bacteria, thus it has been used as fingerprinting tool that is applied for strain discrimination (Beuzon et al., 2004). IS elements have been reviewed thoroughly in previous publications (Siguier et al., 2014; Vandecraen et al., 2017).

Resistance islands are genomic islands that contain multiple resistance genes. They have been described mainly in the Proteobacteria, including Salmonella, Shigella flexneri, V. cholerae, Staphylococcus aureus, and Acinetobacter baumannii (Pagano et al., 2016). Several AMR islands, also called SGIs, have been identified in Salmonella strains (Miriagou et al., 2006; Silva et al., 2012). The best known is SGI1, which was initially described in the epidemic MDR Salmonella Typhimurium phage-type DT104 (Boyd et al., 2001; Doublet et al., 2005). Since then, several variants of SGI1 have been described in a wide variety of Salmonella serovars, including Typhimurium, Agona, Paratyphi B, Albany, Meleagridis, and Newport, which indicates its horizontal transfer potential (Levings et al., 2005). SGI1 is 43 kb and contains 44 open reading frames (ORFs), many of which show no homology to known gene sequences. As mentioned above, SGI1 is an integrative mobilizable element that contains a complex class 1 integron named In104, located within the AMR gene cluster which is 13 kb and is located at the 3′ end of SGI1 (Doublet et al., 2005; Levings et al., 2005). The SGI1 antimicrobial resistance gene cluster contains a complex class 1 integron that harbors five resistance genes [aadA2, sul1, floR, tetA(G), blaPSE-1] encoding ampicillin, chloramphenicol, florfenicol, streptomycin/spectinomycin, sulphonamides, and tetracycline (ACSSuT) resistance, the so-called ACSSuT resistance type (Doublet et al., 2005; Levings et al., 2005; de Curraize et al., 2021).

Quorum-sensing is an intercellular cell–cell communication mechanism that allows bacteria to control the expression of genes involved in a variety of cellular processes and plays a critical role in the adaption and survival of bacteria in their environment (Kendall and Sperandio, 2007). QS is mediated by secreted chemical signals called autoinducers (AIs), which are small diffusible molecules that are synthesized and released from bacterial cells in accordance with cell number and accumulate in the external environment (Taga et al., 2003). When bacterial cell density reaches to a certain level that AIs concentration is over a minimal threshold, AIs bind to cognate receptors to alter and regulate bacterial gene expression accordingly in response to their population density (Taga et al., 2003; Kendall and Sperandio, 2007). A large number of physiological processes in bacteria, including biofilm formation, virulence factor production, bioluminescence, sporulation, motility, and antibiotic production, are regulated by QS systems (Rutherford and Bassler, 2012; Wang et al., 2019a).

While most QS signals are species-specific, AI-2 has been observed throughout the bacterial kingdom and is considered as a universal QS signal (Wang et al., 2019a). It is involved in interspecies cell-to-cell communication, as AI-2 produced by one species can influence gene expression in another. AI-2 is a metabolic byproduct of a luxS gene-encoded synthase, which is an enzyme that has been found in more than 55 bacterial species and is involved primarily in the conversion of ribosyl-homocysteine into homocysteine and 4,5-dihydroxy-2,3-pentanedione (DPD), the precursor of AI-2 (Vendeville et al., 2005). The LuxS/AI-2 QS system is biologically important as it is involved in numerous physiological processes such as altering bacterial growth characteristics, biofilm formation, conjugation, virulence regulation, AMR, and metabolism in Salmonella (Taga et al., 2001; Choi et al., 2007; Widmer et al., 2007; Chen et al., 2008; Karavolos et al., 2008; Jesudhasan et al., 2010; Ju et al., 2018; Shi et al., 2018). It contributes to AMR through different mechanisms: MGE, efflux pumps, the VraSR two-component system, inhibition of the folate synthesis pathway, and biofilm formation. Detailed information on the relationships between the LuxS/AI-2 system and drug resistance has been reviewed by Wang et al. (2019a). In brief, the LuxS/AI-2 system has been shown to impact the transfer of resistance plasmids and the LuxS/AI-2-based QS enhances the expression of genes involved in conjugation-related activities, leading to increased conjugation frequency (Lu, 2004).

In the presence of AI-2, the conjugation frequency of the plasmid RP4 carrying the tet(A) gene in bacteria exposed to tetracycline increased significantly compared to a non-exposed control (Lu, 2004). Increased AI-2 concentrations led to increased expression of plasmid-transfer associated gene trbC. Similar studies conducted by another group also showed that exposure to sub-MIC tetracyclines could facilitate the conjugative transfer of plasmid RP4 in E. coli and this process could be enhanced by AIs, but inhibited by quorum sensing inhibitors (QSIs; Zhang et al., 2018). Therefore, the LuxS/AI-2 could be a potential target for preventing the spread of bacterial resistance.

The SOS response is a global stress response to DNA damage in which the cell cycle is arrested and DNA repair and mutagenesis is induced (Beaber et al., 2004; Baharoglu et al., 2010; Zgur-Bertok, 2013; Qin et al., 2015). It is an inducible DNA repair pathway controlled by two key regulators, LexA, a repressor, and RecA, an inducer (Podlesek and Žgur Bertok, 2020; Figure 4). In bacteria, upon DNA damage, SOS is induced when an increase in intracellular single-stranded DNA (ssDNA) is generated when DNA polymerase III stalls at a lesion while helicase continues unwinding DNA (Podlesek and Žgur Bertok, 2020). Stimulated by ssDNA, RecA is activated by binding to ssDNA (Maslowska et al., 2019). The formation of a ssDNA/RecA nucleofilament stimulates auto-proteolysis of the LexA repressor, leading to de-repression of genes comprising the SOS regulon (Baharoglu et al., 2010; Podlesek and Žgur Bertok, 2020). In addition to several endogenous triggers (for example, UV irradiation, oxidants, and chemical mutagens) that cause the accumulation of ssDNA, thus inducing the SOS response, ssDNA is also produced by several mechanisms of exogenous DNA uptake involved in lateral gene transfer, including conjugation, transformation, and occasionally transduction (Baharoglu et al., 2010). The SOS response can also be induced by numerous antimicrobials, presumably because these antimicrobials cause the production of ssDNA. At least two classes of antimicrobials, fluoroquinolones (such as ciprofloxacin) and dihydrofolate reductase inhibitors (such as trimethoprim), can activate the SOS response (Beaber et al., 2004; Qin et al., 2015). Besides being initially recognized as a regulator of DNA damage repair, it has been noted that the SOS response plays a much broader role in bacteria. As an error-prone repair system, the SOS response contributes significantly to DNA changes observed in a wide range of species by promoting an elevated mutation rate which generates genetic diversity and adaptation, including the development of resistance to antimicrobials (Podlesek and Žgur Bertok, 2020). The SOS response also plays a key role in AMR by stimulating gene transfer, inducing mutation and genomic rearrangements, and the formation of biofilms which are highly recalcitrant to antimicrobials (Podlesek and Žgur Bertok, 2020).

The SOS response has been shown to promote horizontal dissemination of resistance genes in the presence of antimicrobial agents (Beaber et al., 2004; Hastings et al., 2004). Guerin et al. (2009) observed that the antimicrobial compounds found in manure can activate the SOS response in bacteria, which then increases the activity of integrases and transposases, and thereby transposition rates of gene cassettes and other translocative elements (Guerin et al., 2009). They concluded that LexA controlled expression of most integron integrases and consequently regulated cassette recombination. Baharoglu et al. (2010) found that conjugative transfer of plasmids R388, R6Kdrd, and RP4 in E. coli and V. cholerae induces the SOS response and up-regulates integrase expression, triggering cassette recombination in recipient cells (Baharoglu et al., 2010). Additional studies have shown that the SOS response can lead to an increase in cell membrane permeability and downregulation of the korA repressor of various transfer associated genes in the RP4 plasmid (Wang et al., 2019b).

The SOS induction leads to genetic diversification of these mobile elements and their transfer to surrounding bacteria, resulting in the dissemination of AMR genes (Baharoglu et al., 2010). It was also observed that ICEs could utilize the SOS response to mobilize themselves from the bacterial chromosome and infect other cells (Hastings et al., 2004). The transfer of SXT, an approximately 100 kb ICE derived from V. cholerae that encodes a variety of antimicrobial resistance genes, is markedly enhanced when the SOS response is induced by two DNA-damaging antimicrobials (mitomycin C and ciprofloxacin), resulting in the transfer of MDR genes (Hastings et al., 2004).

Other studies have outlined the connections between horizontal gene transfer, bacterial stress response (SOS), and recombination of gene cassettes in integrons, which provides new insights into the development of the AMR within a population (Guerin et al., 2009; Baharoglu et al., 2010). Since several commonly used antimicrobials can induce the SOS response, which in turn can enhance the conjugative transfer of the plasmids, the use of certain antimicrobial agents, either clinically or in agricultural settings, might potentiate the horizontal dissemination of AMR genes to a broad range of bacterial species and hasten genetic change and the evolution to resistance in pathogenic populations.