Underrated Staphylococcus species and their role in antimicrobial resistance spreading

Abstract The increasing threat of antimicrobial resistance has shed light on the interconnection between humans, animals, the environment, and their roles in the exchange and spreading of resistance genes. In this review, we present evidences that show that Staphylococcus species, usually referred to as harmless or opportunistic pathogens, represent a threat to human and animal health for acting as reservoirs of antimicrobial resistance genes. The capacity of genetic exchange between isolates of different sources and species of the Staphylococcus genus is discussed with emphasis on mobile genetic elements, the contribution of biofilm formation, and evidences obtained either experimentally or through genome analyses. We also discuss the involvement of CRISPR-Cas systems in the limitation of horizontal gene transfer and its suitability as a molecular clock to describe the history of genetic exchange between staphylococci.


Introduction
Every year, hundreds of thousands of deaths around the world are attributed to the ever increasing problem of antimicrobial resistance (Laxminarayan et al., 2016). It is estimated that, if the issue is not properly addressed, by the year of 2050, more than 10 million annual deaths will be caused by antimicrobial-resistant microorganisms, surpassing deaths by cancer (O'Neill, 2014). Although the accuracy of this frightening estimate is questioned by some authors, since the future scenario of disease treatment may be considerably different from the current one, the clinical, economical, and public health burden associated with antimicrobial resistance is undeniable (Kraker et al., 2016;Robinson et al., 2016).
In the context of resistance dissemination, bacteria of the Staphylococcus genus, residents of the normal microbiota of human beings and animals, play a central role. Staphylococcus aureus is the main pathogen of the group, responsible for a variety of clinical infections in humans and a leading cause of bacteremia, endocarditis, and many infections related to invasive medical devices (Tong et al., 2015). Meanwhile, coagulase negative staphylococci (CoNS), especially S. epidermidis and S. haemolyticus, have emerged as recurrent causative agents of nosocomial infections, mainly those related to indwelling devices (Becker et al., 2014). They are a serious threat in the twilight of the multidrug resistance era, for actively participating in the horizontal transmission of resistance (Becker et al., 2014;Czekaj et al., 2015).
Genomic analyses indicate that many of the genetic determinants of resistance may have been exchanged between several staphylococcal species from different environments and hosts (Rolo et al., 2017;Rossi et al., 2017a;Kohler et al., 2018;). This includes species that are understudied and/or underestimated, either for having been recently discovered, rarely involved with infectious diseases, or for lacking canonical staphylococcal virulence factors. However, recent evidence shows that these underrated species, despite not being usual pathogens, may have an important role in the exchange of antimicrobial resistance genes, by acting as gene-reservoirs for more pathogenic species, such as S. aureus (Otto, 2013;Rossi et al., 2017aRossi et al., , 2019.

Antimicrobial resistance increase and the interconnectedness of its spreading
In the past decade, consumption of antibiotic drugs increased by 35%, with 76% of this growth concentrated in Brazil, Russia, India, China, and South Africa (Van Boeckel et al., 2014), with a projection of a rise in consumption in the next 15 years of 67% (Van Boeckel et al., 2015). The disturbing situation of antimicrobial resistance led the World Health Organization to elaborate the "Global action plan on antimicrobial resistance", with goals that include: (i) improving the understanding of the problem through communication, education and training, (ii) increasing surveillance and research, (iii) advancing in preventive actions to reduce the incidence of infections, and (iv) optimizing the use of antimicrobial drugs in human and veterinary health (WHO, 2017). Tied to this worldwide concern to restrain the dispersion of resistance, the emerging engagement of scientists and other professionals with the "One Health" agenda increases the acknowledgement of the need for global approaches as the only possible way to keep the predicted disaster of 2050 of more than 10 million annual deaths caused by resistant microorganisms from actually happening.
Considered as a "new professional imperative", One Health is a collaborative and multidisciplinary effort to achieve optimal health for people, animals and our environment from local to global scales (King et al., 2008). It recognizes that the welfare of these three domains is interconnected and this link, ignored for so long, is crucial for the spreading of antimicrobial resistance. The use of antimicrobial drugs in agriculture, for example, is the largest worldwide, thus being a major driver of resistance for several reasons, like exposure of bacteria to sub-therapeutic doses of the antibiotics and the exposure of human and animals to those drugs and microorganisms, either via consumption of products or environmental release (Silbergeld et al., 2008). Studies also point out that the use of antimicrobial drugs, particularly to maintain health, productivity, and promoting growth of food animals, contribute to the dispersion of resistant bacteria in livestock and human beings (Marshall and Levy, 2011;Van Boeckel et al., 2015). In aquaculture, it is estimated that around 80% of antimicrobials used reach the environment with their activity intact, thereby expanding the surroundings where selection of resistant microorganisms will take place (Cabello et al., 2013). Even insects commonly associated with food animals, like houseflies and cockroaches, are presumably vehicles of microorganisms from the farms to urban centers (and vice versa), as evidenced by multidrug-resistant clonal lineages carried by them, that were also isolated from different environments (Zurek and Ghosh, 2014).
Pet animals have been shown to act as reservoirs of resistant bacteria, which in turn act as reservoirs of mobile genetic elements that carry antimicrobial resistance genes (Guardabassi et al., 2004;Rossi et al., 2017a). In fact, the relationship between the animals and their owners signifi-cantly shapes the microbiota of both counterparts (Song et al., 2013). For that reason, the indiscriminate use of antimicrobials in veterinary practice represents a direct threat to human beings. More aggravating is the fact that some drugs that are either not recommended to be used in humans, or those that are considered as last resources, are heavily used to treat animals. Some examples include the polymixins and chloramphenicol and its derivatives, the latter presenting several adverse effects that limit its employment (Cabello et al., 2013;Poirel et al., 2017).
Consistently, multidrug resistant strains, like methicillin-resistant Staphylococcus aureus (MRSA) are ubiquitous, being isolated from humans, pets, food, other animals and the environment (Vanderhaeghen et al., 2010;Kock et al., 2013;Rossi et al., 2017b). Due to local variations in control practices and specific characteristics of circulating clones, the overall geographic distribution of MRSA, for example, can range from 1 to 5% of isolates in northeastern Europe to more than 50% in certain Latin American countries (Brazil, Uruguay, Venezuela, Bolivia, Peru, and Chile) and in Japan .

Variety and clinical significance of Staphylococcus spp.
Although S. aureus is the major bacterium of its genus, more than 50 Staphylococcus species are registered in the List of Prokaryotic Names with Standing in Nomenclature database, available at http://www.bacterio.net (Parte, 2018). However, DNA sequencing of complete genomes or housekeeping genes, phylogenetic analyses, DNA-DNA hybridization, protein profiles, and genotyping techniques have constantly led to reclassifications or proposals of new species and subspecies (Sasaki et al., 2007;Fitzgerald, 2009;Taponen et al., 2012). As these techniques advance and new sources of Staphylococcus are explored, especially in different animals, new species are also discovered, such as S. nepalensis, isolated from goats (Spergser et al., 2003), S. stepanovicii, isolated from wild small animals (Hauschild et al., 2010), S. pseudintermedius, isolated from several domestic animals, such as dogs, cats, horses and parrots (Devriese et al., 2005), and S. agnetis, isolated from bovines with subclinical and clinical mastitis (Taponen et al., 2012).
In general, staphylococci are natural inhabitants of skin and mucous membranes of human beings and animals, while the prevalence of species widely varies according to the host. S. felis, for example, is typically isolated from feline, either healthy or presenting signs of lower urinary tract disease, otitis externa, and ocular disease (Rossi et al., 2017b;Worthing et al., 2018); S. pseudintermedius is prevalent in domestic dogs, healthy or related to diseases like pyoderma and otitis externa (Ruscher et al., 2009;Rossi et al., 2018); S. caprae is involved with intramammary infections in dairy goats (Moroni et al., 2005), among others. Regardless of their source, infections caused by unusual human pathogens are sporadically reported (Morfin-Otero et al., 2012;Novakova et al., 2006a,b), with special emphasis on those caused by S. pseudintermedius, with most cases indicating the contact of domestic dogs with their owners as the probable source of infection (Van Hoovels et al., 2006;Riegel et al., 2011;Lozano et al., 2017). Given their great adaptability to unfavorable conditions (Uribe-Alvarez et al., 2016;Rossi et al., 2017c), staphylococci isolated from the surrounding environment are also responsible for human acquisition of relevant pathogens, including MRSA (Hardy et al., 2006;Sexton et al., 2006).
The production of coagulase and its plasma-clotting activity is a central diagnostic feature to distinguish staphylococcal strains in coagulase-positive staphylococci (CoPS) and coagulase-negative staphylococci (CoNS) (Becker et al., 2014). In addition to being key to diagnostics and group differentiation of Staphylococcus in CoPS and CoNS, coagulase is a virulence factor that leads to the cleavage of soluble fibrinogen to produce a fibrin coat in the surface of the bacteria, thus protecting it from phagocytosis and other host defenses (Powers and Wardenburg, 2014). Moreover, the polymorphisms of the coagulasecoding gene, coa, allows it to be explored in molecular typing techniques (Salehzadeh et al., 2016;Javid et al., 2018). However, because some populations of CoPS may not have the coa gene, while some CoNS present this gene, the applications of these methods are limited (Almeida et al., 2018). These coagulase-variable strains are more frequently found in some species than others, but their misdiagnosis may lead to unsuitable treatment of infections and control measures. This is especially significant when the detection of the pathogenic S. aureus relies on coagulase production and strains of these species do not produce coagulase, leading to isolate misidentification (Sundareshan et al., 2017). Strains of the CoNS S. chromogenes, S. xylosus, S. cohnii and S. agnetis have been reported to clot plasma, leading to misidentification of the pathogens causing mastitis in dairy animals (Taponen et al., 2012;Santos et al., 2016;Almeida et al., 2018). For that reason, researchers have relied on more accurate identification methods, including the sequencing of the 16S rRNA and the housekeeping genes rpoB, encoding the b subunit of RNA polymerase, as well as tuf, encoding the EF-tu elongation factor (Ghebremedhin et al., 2008;Li et al., 2012). Given its simplicity to perform and its cost effectiveness, matrix assisted laser desorption ionizationtime of flight mass spectrometry (MALDI-TOF) is emerging as a potential tool for microbial identification and diagnosis (Singhal et al., 2015). MALDI-TOF identification of staphylococci shows a good correlation with sequencing results, although the lack of standards for uncommon and recently identified species is still a bottleneck (Rossi et al., 2017b).
CoPS, with a special emphasis on S. aureus, responsible for several clinical infections, from those in skin and soft tissues to systemic disease processes like sepsis (Tong et al., 2015), are considered to be more pathogenic than CoNS. A plethora of virulence factors have been described and extensively reviewed for S. aureus, comprising molecules involved in tissue adhesion, immune evasion and host cell injury (Powers and Wardenburg, 2014). Proteins covalently anchored to the cell wall peptidoglycan may participate in not only biotic and abiotic surface adhesion, but also in biofilm formation and iron acquisition, among other functions (Foster et al., 2014). Moreover, a wide variety of toxins can be secreted, aiming to evade the defense mechanisms of the host (Otto, 2014). Other relevant pathogenic CoPS belong in the Staphylococcus indermedius group (SIG). This group includes zoonotic pathogens typically associated with dog bites, i.e., S. intermedius, S. pseudintermedius, the recently described S. cornubiensis (Murray et al., 2018), and S. delphini, first isolated from skin lesions of dolphins (Varaldo et al., 1988).
The increasing recognition of the importance of S. pseudintermedius as a zoonotic pathogen has boosted investigations on virulence factors involving infections caused by this bacterium. Among them, pore-forming toxins seem to play a pivotal role in the characteristic skin infections (Abouelkhair et al., 2018;Maali et al., 2018). Because many of these virulence factors are encoded in mobile genetic elements (MGE) with extensive variation in gene content, different strains strongly vary in their virulence arsenal (Otto, 2014;Moon et al., 2015).
The CoNS constitute the vast majority of staphylococci, comprising more than 80% of the species described to date (Becker et al., 2014;Parte, 2018). Historically considered as harmless inhabitants of the human and animal microbiota, in the past two decades CoNS have emerged as the major nosocomial pathogens, mostly associated with invasive medical devices (Becker et al., 2014), being particularly threatening to immunocrompromised individuals (Morfin-Otero et al., 2012). Among these opportunistic pathogens, S. epidermidis is the most frequent cause of nosocomial infections (Gomes et al., 2014), followed by S. haemolyticus (Czekaj et al., 2015). The infections caused by these species are particularly important because they are difficult to treat, since device colonization is usually related to biofilm formation, which can lead to complications, including sepsis, endocarditis, and a wide variety of local infections derived from the bloodstream spreading of bacteria (Chang et al., 2018;Oliveira et al., 2018).

Commensal and opportunistic staphylococci acting as gene reservoirs
Staphylococcus species that are usually considered as harmless inhabitants of the microbiota of animals and humans beings, like most CoNS, lack nearly all of the virulence factors described for S. aureus and do not encode many known specific factors, apart from a limited amount of toxins and exoenzymes (Zhang et al., 2003). Their emergent threat comes from the fact that these bacteria may carry a huge amount of antimicrobial resistance genes located in MGEs (Gomes et al., 2014;Czekaj et al., 2015;Hosseinkhani et al., 2018).
Multidrug-resistant CoNS have been increasingly linked to infections outbreaks in healthcare units (Chang et al., 2018;Li et al., 2018), including strains that are not only isolated from patients, but also from healthcare workers, and the environment (Widerstrom et al., 2016). Microbiome studies reveal clonal staphylococcal strains wide ability to colonize diverse hosts and surrounding environments (Song et al., 2013;Lax et al., 2014). Their widespreadness can lead to infections caused by multidrug resistant strains in both humans and animals by species that are atypical pathogens in one of the hosts. Dutta et al. (2018), for example, have recently isolated Staphylococcus pettenkoferi strains causing peritonitis from a domestic cat. This species was discovered in 2002 in various human patients showing multiple clinical manifestations, but had never been isolated from animals before. Likewise, as aforementioned, the typical canine staphylococcal species S. pseudintermedius can occasionally cause disease in human beings as well (Riegel et al., 2011;Lozano et al., 2017).
This long overlooked exchange of microorganisms and antimicrobial resistance between different hosts and environments is now clear, and strains isolated from humans and domestic animals carry several resistance genes in common (Schwarz et al., 2018). Even though some antimicrobials have their use restricted to treat infections in animals, many multidrug resistance genes in staphylococci isolated from them confer resistance to antimicrobial agents that are highly important in human medicine (Wendlandt et al., 2015). Then, even if one staphylococcal species is not a common pathogen, it can be a potential threat, because it may be capable of transferring antimicrobial resistance genes to more pathogenic species, such as S. aureus, thereby enhancing its capacity to resist drug therapy (Haaber et al., 2017). For that reason, some CoNS have been suggested to act as antimicrobial genes reservoirs within the Staphylococcus genus (Cafini et al., 2016;Rossi et al., 2017a).
The acquisition of antimicrobial resistance genes is mainly credited to the occurrence of conjugation and bacteriophage transduction and the presence of dozens of insertion sequences in staphylococcal genomes, whose rearrangements contribute to genome plasticity and strains' phenotypic diversification (Takeuchi et al., 2005;Haaber et al., 2017). For a while, bacteriophage transduction was perceived as the primary route of horizontal gene transfer between staphylococci, while conjugation was believed to play only a minor role in the evolution of this genus, given the scarcity of mobilization and conjugation loci in staphylococcal plasmids (Ramsay et al., 2016;Haaber et al., 2017). In fact, only 5% of sequenced staphylococcal plasmids harbor genes required for autonomous transfer by conjugation, which is contrasting with the abundant evidences for horizontal transfer of plasmids between different lineages and species of Staphylococcus (Ramsay et al., 2016).
However, new mechanisms of transference of those types of plasmids have been discovered and probably explain the extensive plasmid transfer within the genus. These mechanisms include conjugation mediated by integrative and conjugative elements (ICEs), also referred to as conjugative transposons (Lee et al., 2012), in trans recognition of multiple variants of the canonic origin of transfer (oriT) by some conjugative plasmids (O'Brien et al., 2015b), and the activity of novel conjugative plasmids described in community isolates of S. aureus that are capable of mobilizing unrelated non-conjugative plasmids (O'Brien et al., 2015a).
The Staphylococcal Cassette Chromosome mec (SCCmec), a genomic island that encodes resistance to methicillin and nearly all other beta-lactam antibiotics, is also a protagonist in the emergence of resistant strains. Analyses of numerous SCCmec sequences indicate that this mobile genetic element evolved by recombination and assembly events involving an ancestral SCCmec III cassette between strains of the S. sciuri group and the species S. vitulinus and S. fleurettii, which were then transferred to S. aureus and other species (Rolo et al., 2017), with CoNS acting as their central reservoirs (Saber et al., 2017).
Our group has demonstrated the transference of high molecular weight plasmids carrying the mupA gene for mupirocin resistance from S. epidermidis, S. aureus, and S. haemolyticus strains, from either human or canine origin, to another S. aureus strain (Bastos et al., 1999;Rossi et al., 2016Rossi et al., , 2018. Mupirocin resistance spreading is alarming, since this drug, which is used as an intranasal ointment by healthcare workers, can significantly reduce the occurrence of nosocomial infections caused by MRSA (Patel et al., 2009). Similarly, Cafini et al. (2016) demonstrated the transfer of linezolid resistance mediated by the cfr gene through plasmids, between S. epidermidis, S. aureus, and Enterococcus spp. isolates from Japan. Enterococcus spp. is also a protagonist of nosocomial infections and has been pointed out to participate in plasmid exchange with MRSA, through which resistance to the last-resource antibiotic vancomycin appears to be acquired (Kohler et al., 2018). A study performed by Meric et al. (2015) with hundreds of genomes of S. aureus and S. epidermidis showed that these species share only about half of their pan genome, but there was a considerable sharing of mobile genetic elements between the two species, in particular genes associated with pathogenic islands and the SCCmec.
Biofilm formation, a characteristic feature of many CoNS (Barros et al., 2015;Buttner et al., 2015), seems to provide the ideal environment for the occurrence of horizontal gene transfer (Madsen et al., 2012).

Biofilms as the perfect place for horizontal gene transfer among staphylococci
Bacterial biofilms are dense surface-associated cellular communities embedded in a protective self-produced matrix of exopolysaccharides, whose development confers new properties to its inhabitants (Black andCosterton, 2010, Flemming et al., 2016). Biofilms feature a social cooperation between bacteria that enhances nutrient uptake and distribution and, given its protective matrix, hinders the exposure to antimicrobials and host defenses, thus increasing survival (Flemming et al., 2016). Hence, biofilm formation plays a fundamental role in virulence.
Although the biofilm mechanisms involved in resistance to host defenses are not entirely understood, they include spatially limiting the access of leukocytes and their products to the target cells, suppression of leukocyte effector functions and cell-cell communication to increase resistance (Leid, 2009). Reducing membrane permeability also contributes to limit the entrance of antimicrobials (Liu et al., 2000). However, the presence of antibiotics that affect the Staphylococcus cell wall have been shown to modulate natural transformation in a process that seems to be dependent of the alternative sigma fator H, SigH (Thi et al., 2016). In fact, the expression of SigH-controlled genes makes S. aureus cells competent for transformation by plasmids or chromosomal DNA (Morikawa et al., 2012), which in turn increases the probability of plasmid exchange between the cells within the biofilm. Liu et al. (2017) demonstrated that the presence of many antibiotics induces the expression of the ccrC1 gene, involved with the excision of the SCCmec from the bacterial chromosome, thus triggering its transfer.
Overall, the high cellular density of biofilms, increased concentration of exogenous DNA, enhanced cell competence, stabilization of cell-cell contact by the matrix and even the biofilm architecture, that facilitates the dispersion of MGEs, may contribute to horizontal gene transfer (Madsen et al., 2012;Savage et al., 2013). The fact that more transconjugants are produced after filter-matings, when compared with planktonic cells mating, indicates the importance of biofilms for the transfer of MGEs (Madsen et al., 2012). It has been shown that S. aureus biofilms can drastically increase the rates of conjugation and transformation of plasmids containing resistance to multiple drugs (Savage et al., 2013).
As observed for many bacteria, during biofilm maturation S. aureus can suffer lysis and release its genomic DNA, producing the so-called external DNA (eDNA), which is a major component of the matrix of biofilms established by this bacterium (Sugimoto et al., 2018). The eDNA adsorbs to the surface of a single cell in long loop structures, which act as an adhesive substance that facilitates cell attachment, in addition to influence the hydrophobicity of the bacterial cell surface (Okshevsky and Meyer, 2015). Furthermore, in S. epidermidis, eDNA production reduces the depth of vancomycin penetration, as the matrix-embed and negatively charged DNA interacts with positive charges of the antimicrobials (Doroshenko et al., 2014). Moreover, the eDNA released by the cells to constitute the biofilm matrix is also available for horizontal gene transfer by transformation to competent cells in the community (Hannan et al., 2010;Vorkapic et al., 2016).
The dispersion of MGEs and eDNA is likely facilitated by the empty spaces within biofilms, which also function as channels that allow the flowing of fluids and the consequent dispersion of nutrients and oxygen to all cells. Studies indicate that this biofilm architecture is influenced by the production of biosurfactant compounds by the adherent bacteria (Davey et al., 2003). Consistent with this, we have recently shown that S. haemolyticus strains are capable of producing biosurfactants that affect biofilm formation on abiotic surfaces (Rossi et al., 2016). Likewise, surfactant peptides produced by S. epidermidis control biofilm maturation and detachment from colonized catheters (Wang et al., 2011).
While, as abovementioned, CoNS lack most of the S. aureus virulence factors, biofilm formation is a major feature of S. epidermidis and S. haemolyticus, as well as of other CoNS, such as S. saprophyticus, S. hominis and S. cohnii. This phenotype is positively correlated with widespread antimicrobial resistance (Allori et al., 2006;Czekaj et al., 2015).

The CRISPR paradox in gene-reservoirs' staphylococci
Clustered regularly interspaced short palindromic repeats (CRISPRs), allegedly present in the genomes of close to 90% of archaea and 50% of bacteria, are a prokaryotic evolutionary defense response to the preponderance of bacteriophages in the biosphere, acting as an interference system against foreign nucleic acids (Horvath and Barrangou, 2010). The CRISPR locus is composed of direct repeats of palindromic sequences interspersed with small sequences called spacers, which are fragments of exogenous sequences ( Figure 1A). A functional CRISPR system has adjacent proteins (CRISPR-associated proteins, Cas), responsible for the process of recognition of foreign nucleic acids and the interference against invasive genetic elements (Horvath and Barrangou, 2010).
CRISPR interference begins with the entrance of an exogenous DNA or RNA in the bacterial cell, which is processed into small fragments by repair proteins, like those of the RecBCD complex (Amitai and Sorek, 2016), and later recognized by a complex of universal Cas1-2 proteins, which incorporates a fragment of the invader nucleic acid in the CRISPR locus as a new spacer ( Figure 1B). The CRISPR locus is expressed as a long transcript that is processed, either by Cas proteins or intrinsic RNases, into small RNAs referred to as CRISPR-RNAs (crRNAs). Each crRNA is made of a fragment of the original direct repeat and the spacer, whose complementarity to foreign mobile genetic elements is the basis of the interference process, fulfilled with the activity of another Cas protein (like Cas9), or Cas complexes (van der Oost et al., 2014).
Marraffini and Sontheimer (2008) described a S. epidermidis CRISPR containing a spacer whose sequence was homologous to a region of the nes gene, coding a nickase found in virtually every staphylococcal conjugative plas-Resistance exchange in staphylococci mid sequenced. Consistently, this CRISPR prevented conjugation and transformation from happening, while the removal of the spacer allowed plasmid uptake. Later, Hatoum-Aslan et al. (2014) constructed mutants for all the nine cas/csm genes of the type III-A system in Staphylococcus epidermidis RP62a and showed that many mutations affected interference by impacting the interference complex formation or the crRNA biogenesis. At least 4% of the spacers within staphylococcal CRISPRs are identical to publicly available plasmid sequences, thus demonstrating their antiplasmid activity (Rossi et al., 2017a). This percentage may be underrated due to under-sampling of mobile genetic elements and, as a consequence, their relative scarcity in public databases (Mojica et al., 2005).
Since CRISPR-Cas systems were believed to exist in roughly 50% of bacteria, and given its antiplasmid activity, it is supposed to highly impact horizontal gene transfer among staphylococci. However, a study performed by Cao et al. (2016) with hundreds of S. aureus clinical isolates from China revealed that less than 1% of them harbored a complete arrangement of cas genes. Later, our group analyzed the genomes of dozens of CoNS of 15 species and also found that less than 15% of them carried CRISPRs and cas genes (Rossi et al., 2017a). Regardless of how abundant CRISPRs really are among Staphylococcus isolates, they are clearly rarer than previously thought, which is consistent with the role of staphylococcal as reservoirs of antimicrobial resistance genes.
Because spacers are incorporated in the CRISPR locus following an organized and thus chronological order (van der Oost et al., 2014), their sequences can be explored as molecular clocks to reveal the history of genetic invasion of a given isolate, and even to study the epidemiologic connection between different strains carrying a CRISPR of a common origin. With that prerogative, we explored the spacer sequences of CRISPRs located within SCCmec elements from different Staphylococcus species. As indicated by the high sequence identity between the SCCmecs, the identical sequence and organization of spacers evidenced that the mobile genetic element had been transferred between strains of canine S. pseudintermedius and S. schleiferi, and then to strains of S. capitis and S. aureus isolated from humans (Rossi et al., 2017a;Rossi et al., 2019).

Concluding remarks
Most Staphylococcus species lack many of the S. aureus virulence factors and, because they are less frequently isolated from infectious processes, their impact in pathogenesis is usually overlooked. However, sequence and experimental evidences of horizontal transfer of antimicrobial resistance genes between them, favored by their capacity of forming biofilms and the scarcity of restrictive CRISPR-systems, show that these bacteria participate actively in the process of drug resistance spreading. Moreover, the fact that we exchange microbiota with our surroundings, makes it imperative that epidemiologic stud- 6 Rossi et al. Steps of a CRISPR system activity, from foreign nucleic acid recognition and spacer incorporation (a step called adaptation) to interference.
ies and strategies to control the advance of antimicrobial resistance consider the integrated nature of the relationship between human beings, animals, and the environment.

Acknowledgments
The authors thank the financial support from Fun-

Conflict of Interest
The authors declare no conflict of interests.

Author Contributions
CCR and MGM conceived and designed the study, CCR, MFP and MGM analyzed the data and wrote the manuscript. All authors read and approved the final version.