High diversity of genetic lineages and virulence genes in nasal Staphylococcus aureus isolates from donkeys destined to food consumption in Tunisia with predominance of the ruminant associated CC133 lineage

Background The objective of this study was to determine the genetic lineages and the incidence of antibiotic resistance and virulence determinants of nasal Staphylococcus aureus isolates of healthy donkeys destined to food consumption in Tunisia. Results Nasal swabs of 100 donkeys obtained in a large slaughterhouse in 2010 were inoculated in specific media for S. aureus and methicillin-resistant S. aureus (MRSA) recovery. S. aureus was obtained in 50% of the samples, being all of isolates methicillin-susceptible (MSSA). Genetic lineages, toxin gene profile, and antibiotic resistance mechanisms were determined in recovered isolates. Twenty-five different spa-types were detected among the 50 MSSA with 9 novel spa-types. S. aureus isolates were ascribed to agr type I (37 isolates), III (7), II (4), and IV (2). Sixteen different sequence-types (STs) were revealed by MLST, with seven new ones. STs belonging to clonal clomplex CC133 were majority. The gene tst was detected in 6 isolates and the gene etb in one isolate. Different combinations of enterotoxin, leukocidin and haemolysin genes were identified among S. aureus isolates. The egc-cluster-like and an incomplete egc-cluster-like were detected. Isolates resistant to penicillin, erythromycin, fusidic acid, streptomycin, ciprofloxacin, clindamycin, tetracycline, or chloramphenicol were found and the genes blaZ, erm(A), erm(C), tet(M), fusC were identified. Conclusions The nares of donkeys frequently harbor MSSA. They could be reservoirs of the ruminant-associated CC133 lineage and of toxin genes encoding TSST-1 and other virulence traits with potential implications in public health. CC133 seems to have a broader host distribution than expected.


Background
Staphylococcus aureus is a global problem for both humans and animals that affect hospitalized patients and also healthy individuals in the community. This microorganism is usually associated with skin and soft tissue infections. However, it is also able to cause serious diseases, such as pneumonia, meningitis, or septicaemia, among others. Additionally, the pathogenicity of S. aureus infections is facilitated by the expression of several virulence factors, which include cell wall-associated adhesins and several toxin groups [1,2]. The coordinated expression of these virulence factors is dependent on a global quorum-sensing regulator system, named agr (accessory gene regulator) [3]. Methicillin-resistant S. aureus (MRSA) is a well-recognized human pathogen that has also been identified as a veterinary and zoonotic pathogen. MRSA can cause animal infections and is also able to colonize the skin, nasal and oral mucosa of healthy animals. Several studies have reported the emergence of MRSA among farm animals [4][5][6][7]. Moreover, the emergence of MRSA in the equine population has been demonstrated [8][9][10] and, very recently, MRSA isolates have been detected in microbiota of healthy donkeys [11].
More than forty-four million donkeys do exist on our planet, of which 40 million (96%) inhabit in developing countries [12] where donkeys play an important role serving as draught animals (packing, carting, threshing, farm cultivation, riding) [12], as well as for feeding; this is the case of milk [13] or meat products [14]. In Tunisia, there are over a hundred twenty three thousand donkeys (http://www.onagri.tn), what represents over 65.5% of the total equine population and it is considered one of the countries with the largest number of donkeys [15,16]. Donkey food represents approximately 5% of total meat consumed in Tunisia and it is especially eaten in low income families (http://www.femmezoom.com/).
Few data do exist about the genetic lineages of S. aureus and the prevalence of MRSA that colonizes nares of healthy equids. So far, different clonal lineages such as CC8, CC22 or CC398 have been identified in horses [10,17]. Moreover, the isolates detected presented varied susceptibility to antimicrobials, different toxin profiles and diverse agr and SCCmec types depending on the clone found [10,17]. Remarkably, no tst and lukF/lukS-PV positive isolates have been detected in these animals; additionally, as far as we know, no previous studies of this type have been performed in Tunisia and there are very few reports in other African countries [12].
The objectives of this work were to analyse the prevalence of S. aureus in nasal samples of healthy donkeys destined for food consumption in Tunisia and to determine the genetic lineages and the presence of antimicrobial resistance and virulence genes in the recovered isolates.

Sampling and microbiological isolation
Nasal swabs from 100 healthy donkeys were obtained during March-May 2010 in a large slaughterhouse that receives animals destined for human consumption from farms of all Tunisia. The director of the abattoir gave his permission for taking the samples and it was supervised by the veterinarians of the abattoir. The ARRIVE (Animal Research: reporting of in vivo experiments) guidelines were followed in this study.
Nasal swabs were incubated in Tryptone Soy Broth (TSB) for 48h and then subcultured on Baird-Parker agar (BP) and ORSAB medium (Oxacillin Resistance Screening Agar Base, Oxoid) for 24-48 hours for S. aureus and MRSA recovery, respectively. Suspected S. aureus colonies were initially identified by conventional methods [Gram-staining, catalase test, oxidase test, DNase production, and ability to coagulate rabbit plasma (BioRad)]. S. aureus identification was confirmed by amplification of the species-specific nuc gene and, although all isolates were susceptible to methicillin, the presence of the mecA gene was also tested [5].

Detection of antimicrobial resistance genes
The ribosomal methylases encoded by erm(A), erm(B) and erm(C) genes, which confer resistance to erythromycin and clindamycin, and the efflux pump encoded by msr(A) gene, conferring resistance to erythromycin, were studied by PCR in erythromycin-resistant isolates with primers and conditions as previously described [5]. In addition, tet(K), tet(M) and tet(L) genes, which confer resistance to tetracycline, blaZ gene to penicillin, fusB and fusC genes to fusidic acid, ant(6)-Ia, str and ant(3)(9)' genes to streptomycin, and cat pC221 , cat pC223 , and cat pC194 genes to chloramphenicol were studied by PCR in antimicrobial-resistant S. aureus isolates [4].
Mutations in elongation factor G were studied by sequence analysis of fusA gene in all fusidic acid-resistant isolates [19].
Molecular typing of S. aureus isolates Spa-typing was performed in all S. aureus isolates as described elsewhere [20]. The polymorphic X region of spa gene was amplified by PCR, and sequences were analyzed using Ridom Staph-Type software version 1.5.21 (Ridom GmbH), which automatically detects spa repeats and assigns a spa-type according to http://spaserver. ridom.de/. Identification of agr allele group (I-IV) was determined by PCR as earlier described [21].
Multilocus-sequence-typing (MLST) was performed in selected S. aureus isolates (one isolate of each detected spa-type): the allelic profile of each isolate was obtained by sequencing internal fragments of 7 unlinked housekeeping genes (arcC, aroE, glpF, gmk, pta, tpi, and yqiL), allowing the determination of the sequence-type (ST), by the MLST database (http://saureus.mlst.net/). In these selected isolates, the clonal complexes (CC) were assigned according to the obtained ST; in the remaining isolates, the CC were assumed depending on the spa-types.
All isolates that presented spa-types associated with the clonal complex CC133 were tested for their capacity to coagulate bovine plasma (Sigma-Aldrich) following standard methodology [22].

Detection of staphylococcal toxin genes
All isolates were tested by PCR for the presence of 18 genes coding for staphylococcal enterotoxins (sea, seb, sec, sed, see, seg, seh, sei, sej, sek, sel, sem, sen, seo, sep, seq, ser and seu), tst gene encoding the TSST-1 (Toxic Shock Syndrome Toxin) [23], lukF/lukS-PV genes encoding leukocidin PVL (Panton Valentine leukocidin), lukED genes encoding the bicomponent leukotoxin LukE-LukD, lukM gene coding for leukocidin M as well as eta and etb genes encoding exfoliative ETA and ETB toxins, respectively, and hla, hlb, hld, hlg and hlg v genes encoding haemolysin toxins. The presence of all these genes were tested using primers and conditions as previously described [24].

Field survey for S. aureus isolates from donkeys
No MRSA isolates were recovered in the 100 nasal samples of donkeys tested in this study when inoculated onto ORSAB plates. Nevertheless, S. aureus isolates were obtained from Baird-Parker agar plates in 50 of the 100 tested samples (50%), and one isolate per sample was further studied. The collection of 50 S. aureus isolates were cefoxitin and oxacillin susceptible, lacked the mecA gene, and then were confirmed methicillinsusceptible S. aureus (MSSA).

Discussion
Very limited data on the nasal carriage of S. aureus in donkeys are available. In this sense, a study conducted in Ethiopia described the detection of S. aureus in the upper respiratory tract of 13% of tested donkeys [12]. Other study performed in Italy detected S. aureus in 6% of donkey milk samples [13]. The high recovery rate of S. aureus detected among nasal samples of healthy donkeys in the present report (50% of tested samples) is relevant, and represents the first study of this type in Tunisia. The high diversity of genetic lineages among the S. aureus recovered is noteworthy. However, it should be noted that all our strains were methicillin susceptible. Elevated clonal variety has been already detected in other studies among MSSA strains and it seems that MSSA of human show a higher genetic diversity than MRSA [25]. The 16 different STs identified by MLST among S. aureus isolates, with seven of them new, were distributed in eight clonal complexes (CC133, CC1, CC6, CC522, CC72, CC7, CC15 and CC22), and five singletons. Among these, CC133, CC1 and CC6 were predominant and grouped 68% of typed isolates.
Twenty two of our isolates (44%) corresponded to the major ruminant lineage CC133 [6,22], being the major detected lineage. Several previous studies have reported the existence of CC133 in clinical S. aureus isolates of cattle, goats and sheep [26][27][28]; nevertheless, this clone was not detected in an earlier study carried out in Tunisia on healthy sheep [29]. It is interesting to remark that the clonal complex CC133 was associated with different spa-types in donkey isolates in our study (t1166, t1403, t2420, t3583, t7718, t8836) or in sheep and goat isolates (t544, t2678, t3495, t4560, t5592, t7294, t7296, t7297, t7298, t7300) in previous studies [28]. It has been reported that S. aureus isolates of the clonal complex CC133 have been also responsible for most cases of mastitis in dairy farms [30], and it seems that this clone may have a broad geographic distribution. The agr-type I detected among the CC133 isolates in our study was coincident with former reports [31]. Some authors propose that strains of lineage CC133 could have evolved and adapted to small ruminants derived from humans due to an adaptive genome diversification resulted from allelic variation, gene loss, and horizontal acquisition of mobile genetic elements containing virulence genes with attenuated or enhanced activity in ruminants. The capacity to coagulate bovine plasma detected in our isolates belonging to this CC is a characteristic previously reported for isolates adapted to small ruminants and bovine [22]; and the acquisition of a novel staphylococcal pathogenicity island (SaPIov2) carrying a novel von Willebrand factor-binding protein (vWBP) with ruminant-specific coagulase activity has been described among isolates of this lineage [22]. On the other hand, a very recent study undergone in a Danish Zoo [32] has revealed that ST133 may have a broader host distribution since it was detected in S. aureus of a wide variety of animal species. However, to the best of our knowledge, this is the first description of this genetic lineage in donkeys [6].
Seven MSSA isolates were ascribed to the clonal complex CC1 (those of ST1 and ST1738) which is a common lineage among human isolates [26]. This data is in accordance with earlier reports [33] where S. aureus isolated from equines were more likely to cluster into human associated lineages. In the referred study performed in United Kingdom, equine-associated S. aureus isolates were assigned to the major human lineage CC1. Although this CC seems to be related to human isolates, a possible animal origin has also been suggested in other studies, given that its presence in other animal species is not exceptional [34].
The lineage CC6, less frequently detected among our isolates (5 isolates), has been previously detected in both human and animal S. aureus isolates [35,36]. Our CC6 MSSA isolates were classified into the agr group I, which is in harmony with previous studies [37].
Three MSSA isolates were ascribed to the clonal complex CC72; this clone has been detected in humans in different countries [38,39]. One isolate belonged to the clonal complex CC15, which has been before detected in humans at both community and hospital settings [40,41]. Another study conducted in five African countries [42] showed that 12% of S. aureus obtained from humans in the community corresponded to the CC15 clone. In our study, the characteristics of our CC15 isolate were different from other previous reports in relation to the spa-types, agr-types and virulence determinants [40,42].
One MSSA isolate was ascribed to lineage CC22, associated to the United Kingdom EMRSA-15, which is commonly isolated from pet animals, especially dogs [35,43]. This result is consistent with that of [17] where the MRSA isolates of equines belonged to the same clone EMRSA-15 (CC22) and harbored the egc clusterlike comprising the enterotoxin genes: seg, sei, sem, sen, seo, and seu [37].
Several virulence genes were detected in our MSSA isolates. The detection of the tst gene in 12% of isolates is interesting, although this percentage was lower than the one detected among S. aureus isolates from healthy sheep (78%) or healthy humans (20%) in Tunisia [29,44]. Nevertheless, the tst gene was not detected among S. aureus isolates of donkey milk in a previous report [13]. Most of our MSSA isolates harboured haemolysin genes, where the hla and hld genes were present in all our isolates. Similarly, a significant number of strains harboured the hlb gene, while the hlg and hlg v genes were only present in one isolate. High occurrence of the hld gene in nasal S. aureus isolates of healthy sheep in Tunisia has also been reported [29]. Remarkably, most of our isolates (90%) carried the lukED genes, which are commonly present in cattle causing bovine mastitis infections [5]. In addition, a high number of enterotoxin genes, with 3 distinct egc-cluster-like variants, were observed. With this regard, some strains presented some but not all of the genes comprising the egc-cluster or egc-cluster-like. The absence of one or more genes in the egc-cluster has been previously reported [45,46]. Even though S. aureus associated food poisoning outbreaks are normally due to human isolates [47], the presence of a wide variety of enterotoxin genes in S. aureus from donkey reflects the adaptation of enterotoxigenic strains to different mammalian species. The presence of the exfoliatin etb gene in one isolate, which also revealed novel genetic characteristics (t8842-ST2152), is remarkable.
Most of our MSSA showed susceptibility to the antimicrobials tested (60%) with several exceptions. The low frequency of penicillin resistance detected among our isolates contrasts with the high frequency of this type of resistance reported for human isolates, even in commensal strains [44,45]; however, it is in agreement with isolates of sheep origin [29]. With regard to fusidic acid resistance, the rate found in different countries is very variable, with percentages ranging from <1% to >50% [48]. No mutations in elongation factor G have been detected in the fusA sequences of the analyzed fusidic acid resistant S. aureus isolates and the fusC gene was only detected in one strain. The resistance mechanism of the remaining fusidic acid resistant MSSA strains of this study remains unknown. In addition, 2 streptomycin-and one chloramphenicol-resistant isolates did not present any of the tested resistance genes.

Conclusions
In conclusion, the nares of healthy donkeys could be a reservoir of S. aureus isolates of the small ruminant associated CC133 lineage and also of isolates carrying the toxic shock syndrome related gene tst and enterotoxin genes responsible for food poisoning outbreaks.
Remarkably, CC133 seems to have a broader host distribution having been detected in different animal species. More studies should be performed in the future to gain knowledge in the genetic lineages of S. aureus circulating among healthy animals, as well as in the capacity of these strains to produce virulence factors, due to the risk of animal-to-human bacterial transfer and the acquisition and dissemination of the SCCmec element, which is responsible for genetic background common to endemic methicillin resistant S. aureus.

Competing interests
Non-financial competing interests do exist.
Authors' contributions HG and RBS took the animal samples, isolated the S. aureus strains and identified them and carried out important part of the genetic analysis. They participated in the writing of the manuscript. KBS participated in the design of the study, writing of the paper and in the collection of samples and isolation of microorganisms. EGS and CL contributed to the genetic analysis of the virulence factors in the collection of isolates and in the revision of the paper. AJ and NK participated in the identification of S. aureus isolates and in the study of the resistance genes of the microorganisms. MZ contributed to the design of the study, the general discussion of the manuscript and the writing of the paper. CT and AB conceived the study, participated in its design and carried out the general coordination of it and helped to draft the manuscript. All authors read and approved the final manuscript.
Authors' information EGS, CL, MZ and CT belong to the Research Group on -Antimicrobial resistance, virulence and molecular epidemiology in bacteria of animals, food, humans and the environment of the University of Rioja that is coordinated by CT (PhD in Pharmacy and professor of the University of Rioja) and with important contribution of MZ (PhD in Veterinary and Associate Prof. of the Univ. of Rioja). HG, RBS, KBS, AJ, NK and AB belong to the research group on Surveillance of antimicrobial resistance and virulence in animals, food and the environment and implication in human health of the Univ. of Tunis (in Tunisia), coordinated by AB (PhD in Biology and professor of that University).