Frequency and Characterization of Antimicrobial Resistance and Virulence Genes of Coagulase-Negative Staphylococci from Wild Birds in Spain. Detection of tst-Carrying S. sciuri Isolates

The objective of this study was to determine the prevalence and diversity of coagulase-negative staphylococci (CoNS) species from wild birds in Spain, as well as to analyze the antimicrobial resistance phenotype/genotype and the virulence gene content. During 2015–2016, tracheal samples of 242 wild birds were collected in different regions of Spain for staphylococci recovery. The species identification was performed using MALDI-TOF. The antimicrobial resistance phenotype and genotype was investigated by the disk diffusion method and by PCR, respectively. The presence of the virulence genes lukF/S-PV, tst, eta, etb, etd and scn was investigated by PCR. Moreover, CoNS carrying the mecA gene were subjected to SCCmec typing. Of the tested animals, 60% were CoNS-carriers, and 173 CoNS isolates were recovered from the 146 positive animals, which belonged to 11 species, with predominance of S. sciuri (n = 118) and S. lentus (n = 25). A total of 34% of CoNS isolates showed a multidrug resistance phenotype, and 42 mecA-positive methicillin-resistant CoNS (MRCoNS) were detected. The isolates showed resistance to the following antimicrobials (percentage of resistant isolates/antimicrobial resistance genes detected): penicillin (49/ blaZ, mecA), cefoxitin (24/ mecA), erythromycin and/or clindamycin (92/ erm(B), erm(C), erm(43), msr(A), mph(C), lnu(A), lsa(B), vga(A) and sal(A)), gentamicin and/or tobramycin (5/ aac(6′)-Ie-aph(2″)-Ia, ant(4′)-Ia), streptomycin (12/str), tetracycline (17/ tet(K), tet(L), tet(M)), ciprofloxacin (4), chloramphenicol (1/ fexA), fusidic acid (86/ fusB, fusD) and trimethoprim–sulfamethoxazole (1/ dfrK). None of the isolates harbored the lukF/S-PV, eta, etb, etd and scn genes, but two S. sciuri isolates (1%) carried the tst gene. Wild birds are frequently colonized by CoNS species, especially S. sciuri. We identified scavenging on intensively produced livestock and feeding on landfills as risk factors for CoNS carriage. High proportions of MRCoNS and multidrug resistant CoNS were detected, which coupled with the presence of important virulence genes is of concern.


Introduction
Staphylococci can be divided in two major groups based on their capacity to produce the enzyme coagulase and, hence, their ability to clot the blood plasma: coagulase-positive (CoPS) and coagulase-negative staphylococci (CoNS). CoNS constitute a very heterogeneous group that comprise over 40 commensal species of the mucous membranes and skin of humans and other animals, especially mammals and birds [1][2][3][4]. However, in the last decade, CoNS have also been recognized as important causative agents of nosocomial infections, especially the species S. haemolyticus, S. epidermidis, S. saprophyticus and S. lugdunensis [3,4]. Moreover, CoNS can also cause disease in animals [1,5], including dermatitis and endocarditis in poultry [6,7].
Methicillin-resistant CoNS (MRCoNS), mostly mediated by the mecA gene, have gained interest in recent years because of their implications in human and veterinary medicine [8]. Besides methicillin resistance, CoNS have been postulated as an important reservoir of antimicrobial resistance genes that are often located on mobile genetic elements, and, therefore, could be transferred to more pathogenic bacteria, such as S. aureus, by horizontal gene transfer [3,4,9,10]. In fact, the CoNS belonging to the Staphylococcus sciuri group, which includes S. sciuri, S. fleurettii, S. lentus, S. vitulinus and S. stepanovicii, are especially relevant because of their role in the origin, evolution and spread of the mecA gene [4,11]. Regarding the pathogenicity of CoNS, former studies have reported major virulence factors of S. aureus (e.g., Panton-Valentine leukocidin, toxic shock syndrome toxin and exfoliative toxins) in CoNS recovered from humans, livestock and the livestock environment, although their detection is still highly unusual [1,2,12,13].
Several studies have investigated the molecular characteristics of CoNS isolated from food items, companion animals, livestock and clinical samples [1,5,8,[13][14][15][16][17], but CoNS from wildlife remains largely unattended. In this regard, birds have been postulated as sentinels, reservoirs and potential disseminators of antimicrobial resistance due to their interaction with the human interface, their diverse ecological niches and their capacity to migrate long distances [18]. Thus, the objective of the present study was to evaluate the prevalence and diversity of CoNS species in wild birds, to determine their antimicrobial susceptibility pattern and to analyze their virulence gene content.
Most of the species sampled are obligate scavengers (bearded vulture, Egyptian vulture and cinereous vulture) or facultative scavengers that primarily feed on carrion (red kite). Magpies, black headed gulls and western marsh harriers are occasional scavengers, while honey buzzards and northern bald ibis feed primarily on insects (wasps in particular in honey buzzards). All samples, except magpies and black-headed gulls, were obtained as part of conservation programs, either during banding or radio-tagging of chicks in the nest (cinereous vulture, Egyptian vulture, honey buzzard, northern bald ibis and bearded vulture), pre-reintroduction health checks of captive raised chicks (northern bald ibis and bearded vulture), or adults captured at vulture restaurants for banding and radio-tagging (red kites, bearded vultures and western marsh harrier). Magpies were collected from hunting estates after capture and euthanasia as part of authorized control programs, and black-headed gulls were juveniles dead from trauma in a thunderstorm collected near a landfill.
Chicks were sampled at the end of the breeding season, between May and July, specifically: None of the birds was specifically captured for the purpose of the present study and capture and handling of the birds was authorized in each case by permits from the regional government. Handling and sampling were carried out following all applicable international, national, and/or institutional guidelines for the care and ethical use of animals, specifically directive 2010/63/EU and All samples were maintained at 4 • C until arrival at the laboratory and frozen at −80 • C until further analysis. These samples were tested in parallel for the presence of CoPS [19] and also for CoNS; this last one in the present study. In the previous work, CoPS were recovered from 20 of the 242 samples analyzed (8.3%) and they were identified as S. aureus (n = 9) and S. delphini (n = 12) [19].

Bacterial Isolation and Identification
The tracheal swab samples were inoculated into brain heart infusion (BHI; Condalab, Madrid, Spain) broth supplemented with 6.5% NaCl and incubated for 24 h at 37 • C. An aliquot of 30 µL was seeded on mannitol salt agar (MSA; Condalab, Madrid, Spain) and oxacillin resistance screening agar base (ORSAB; Oxoid, Hampshire, UK) with 2 mg/L oxacillin and incubated for 24 h at 37 • C, for CoNS and MRCoNS recovery. A maximum of five mannitol non-fermenting colonies per sample were selected and identified by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF; Bruker Daltonics, Bremen, Germany) using the standard extraction protocol recommended by Bruker [20]. Isolates belonging to different CoNS species of each sample were further studied.
The presence of the following resistance genes was tested by single PCRs: blaZ, mecA, mecB  (Table S1). The physical linkage of tet(L)-dfrK was investigated by PCR (Table S1). Positive controls from the collection of the Universidad de La Rioja were included in all PCR assays. In addition, mutations leading to amino acid changes in the GyrA and GrlA proteins were investigated in ciprofloxacin-resistant isolates by PCR and amplicon sequencing (Table S1). The respective sequences of S. lentus strain NCTC12102 (GenBank accession number UHDR01000002), S. sciuri strain NCTC12103 (GenBank accession number LS483305) and S. xylosus strain NCTC11043 (GenBank accession number UHEI01000002) were used as reference for the amino acid changes detection.

Staphylococcal Cassette Chromosome mec (SCCmec) Characterization
All CoNS isolates carrying the mecA gene were subjected to SCCmec typing by multiplex PCRs targeting the ccr gene complex and the mec gene complex, as previously described [24] (Table S1).

Virulence Gene Content
The presence of the genes encoding the virulence determinants Panton-Valentine leukocidin (lukF/S-PV), toxic shock syndrome toxin 1 (tst), and the exfoliative toxins A (eta), B (etb) and D (etd) was studied by single PCRs and confirmed by amplicon sequencing (Table S1). Moreover, the presence of the scn gene, the marker of the human immune evasion cluster (IEC), was investigated in all CoNS isolates (Table S1). Positive controls from the collection of the Universidad de La Rioja were included in all PCR assays.

Statistical Analysis
Pearson's chi-square test was used to explore colonization of the tested birds by S. sciuri and S. lentus. Specifically, we compared the carriage of S. sciuri and S. lentus between obligate or facultative scavengers and species less likely to feed on carrion (especially livestock carrion), and in the case of S. lentus carriage, in red kites captured during and on spring (return) migration (February 2016, n = 25) and kites arriving at the vulture restaurant on Southbound migration (November 2015, n = 13). Analyses were carried out using SPSS statistical software version 24.0 (IBM ® , SPSS Inc., Chicago, IL, USA) and significance was set at p ≤ 0.05.

Virulence Gene Content
None of the isolates carried the lukS/F-PV, eta, etb, etd or scn genes. Interestingly, two S. sciuri isolates recovered from cinereous vultures carried the tst gene (Table 3).

SCCmec Typing
Among the 47 mecA-positive CoNS isolates, ten S. sciuri and two S. lentus isolates were typed as SCCmec-III and one S. epidermidis as SCCmec-IV. No consensus for the SCCmec type was determined for the remaining mecA-carrying CoNS, either because they were not ascribed to a previously known SCCmec type, or because they were non-typeable with the primers used (Table S3).

Discussion
The current work represents the largest study characterizing CoNS recovered from healthy free-ranging birds, and provides novel information about the frequency, diversity of species, antimicrobial resistance phenotype/genotype and the virulence profile among CoNS from wild animals.
The CoNS tracheal carriage rate detected in wild birds (60%) was higher than that detected in a previous study conducted among wild boars in Spain (37.7%) [25], but similar to that among birds of prey in Portugal (75%) [26]. These results suggest that birds (or at least those consuming vertebrate or invertebrate prey) are more frequently colonized by CoNS than mammals; however, further studies in other animal species need to be assessed to corroborate these data. In this study, a high diversity of CoNS species was detected, S. sciuri being the predominant one accounting for nearly 70% of the isolates recovered. This was to be expected since this species is known to has broad host range and is adapted to very different habitats [4,8,12,27]. Moreover, although it has been formerly found causing infections in animals [15], S. sciuri is the most common CoNS species colonizing healthy wild animals, including birds [25,26]. S. lentus was the second species most frequently recovered in this study and was especially prevalent among red kites. This staphylococcal species is commonly detected among farm animals and people with professional exposure to livestock [16,28,29]. More frequent colonization of obligate and frequent scavengers by S. sciuri and S. lentus suggests that despite being respiratory tract colonizers, carrion feeding may increase the exposure. In particular, carriage of S. lentus by red kites was higher in February than in November when kites arriving from breeding grounds in northern and central Europe were captured. Both the difference in prevalence and the high rate of tetracycline resistance detected in this staphylococcal species provide circumstantial evidences that S. lentus may be acquired from livestock carrion, particularly from slaughterhouse remains (pork) and deceased chickens from commercial layer and broiler farms that are employed as food at the vulture restaurant where the birds were captured. The species S. xylosus and S. kloosii were also recovered from birds of prey in Portugal [26]. Other CoNS species that are frequently detected as causative agents of diverse diseases in humans, such as S. epidermidis, S. saprophyticus, S. succinus and S. capitis [4], were also found among healthy wild birds at very low rates.
The antimicrobial resistance rates detected for some antimicrobials among the isolates investigated are of great concern, especially because wild birds are not supposed to be under the selective pressure of antimicrobial agents; this contrasts with the high susceptibility rates detected among CoNS from wild boars in a previous Spanish study [25]. In our work, 24% of the CoNS isolates showed methicillin resistance, which is far higher than the rate observed in mammals (2.5%) and in birds of prey (0%) [25,26]. Methicillin resistance was mediated in all cases by the mecA gene, which was not surprising, as mecC-carrying MRCoNS isolates are still scarce [30], and, as far as we know, the mecB gene has not been previously reported in CoNS species. However, in the previous study that characterized the CoPS isolates of these samples, mecC-positive methicillin-resistant S. aureus (MRSA) were detected [19]. Moreover, five S. sciuri or S. fleurettii isolates harbored the mecA gene but were phenotypically susceptible to cefoxitin, which has been formerly reported among CoNS of diverse origins [12,31]. mecA gene homologues that are not always associated with methicillin resistance have been found in S. sciuri and S. fleurettii [11], but the primers used in this study do not amplify these mecA homologues. Hence, the methicillin-susceptible phenotype could be attributed to the heterogenous expression of the mecA gene. We could not identify the mechanism implicated in penicillin resistance in 26 isolates, which has been formerly reported among CoNS [26,27]. This fact suggests that other unknown resistance mechanisms are present in the isolates investigated or that the breakpoints for this antimicrobial are not accurate for CoNS. The extremely high resistance rate to clindamycin detected (90%) is worrisome as this antimicrobial is widely used in clinical and veterinary medicine. Among the S sciuri isolates, clindamycin resistance was mainly mediated by the presence of the sal(A) gene, which confers intrinsic resistance to lincosamides and streptogramin A antimicrobials in this species [32]. Although the sal(A) gene has been previously described in non-S. sciuri CoNS species from pets in China [33], in this work, it was only detected in S. sciuri. Macrolides and lincosamides resistance mediated by the erm(43) gene was detected in seven S. lentus isolates. This gene has been formerly detected among isolates belonging to the species S. lentus and S. sciuri of diverse origins, including healthy wild animals [25,34]. It is necessary to note the high tetracycline resistance rate (60%) observed among the S. lentus isolates with respect to the overall tetracycline rate of CoNS (17%), which was mostly mediated by the tet(K) gene. Similar resistance rates in S. lentus were detected in isolates recovered from livestock [29,35], which could be attributable to the extended use of this antimicrobial in veterinary medicine. The location of dfr(K) next to the tet(L) gene in one S. lentus isolate suggests their location in a pKKS2187-like plasmid previously reported among MRSA-CC398 and MRCoNS of pig origin [29,36]. This supports the hypothesis that red kites could acquire CoNS from livestock carrion at the vulture restaurant via direct contact or the food chain. In cinereous vulture chicks, the potential route of colonization is less obvious. Cinereous vultures generally feed on medium-sized native carrion and, in the case of this colony situated in a National Park close to Madrid, no feeding station (vulture restaurant) is in the range of foraging of the vultures during chick raising. However, observation of birds ringed at the colony foraging on solid urban waste landfills [37] suggests that one of the potential sources for colonization by CoNS could be exposure of the adults in the landfill environment and subsequent colonization of the nestlings, or exposure of the latter from landfill foraged food, as has been shown for CoPs in white stork nestlings [38]. However, to date, no detailed information on the degree of exposure of the cinereous vulture chicks to landfill foraged food is available. Fusidic acid resistance rates above 50% were also reported in CoNS from livestock and the livestock environment and birds of prey [8,9,12,26,29], which suggests the presence of mutations in the fusA gene or intrinsic resistance genes like the fusD in S. saprophyticus and fusE in S. cohnii subsp. urealyticus [39]. Fortunately, all CoNS isolates from wild birds exhibited susceptibility to linezolid, which is considered as a last resort antimicrobial in human medicine used to treat serious infections caused by multidrug resistant Gram-positive bacteria, including MRCoNS.
In this study, two S. sciuri isolates harbored the tst gene encoding the pyrogenic toxin superantigen TSST-1 that is considered one of the most important virulence factors produced by S. aureus. It is located on staphylococcal pathogenicity islands (SaPIs) and its mobilization is assisted by different phages [40]. This gene has been formerly described among clinical CoNS isolates but also among those recovered from bovine milk [2,13,41]; however, to best of our knowledge, this is the first description of tst-carrying CoNS isolates from wild animals. Apart from the enterotoxigenic potential, few studies that explore the prevalence of major virulence factors of S. aureus in CoNS exist.
As previously reported by other authors, the SCCmec type IV and, especially, type III are the most common types detected among CoNS isolates of animal origin [8,28,42]. However, the majority of mecA-carrying isolates could not be SCCmec typed with the primers used, which highlights the high diversity of SCCmec types among CoNS and suggests the presence of novel SCCmec elements different from those found in MRSA isolates.

Conclusions
Free-living predatory birds are frequently colonized by CoNS species, especially S. sciuri. In this study we have demonstrated that wild birds are a reservoir of CoNS carrying not only important antimicrobial resistance genes, but also major virulence factors traditionally associated with S. aureus. Scavenging on livestock from intensive production (pork and poultry) and foraging on landfills are a potential source of CoNS isolates recovered from wild birds. The detection of two S. sciuri isolates carrying the tst gene in wildlife is of great concern.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-2607/8/9/1317/s1. Table S1: Primer pairs used for the characterization of CoNS recovered from wild birds; Table S2: Antimicrobial resistance phenotype and genotype, and virulence gene content of the 173 CoNS recovered from free-ranging birds; Table S3: SCCmec typing of the 47 CoNS isolates from wild birds that carried the mecA gene. Funding: This work was partially supported by project SAF2016-76571-R from the Agencia Estatal de Investigación (AEI) of Spain and the Fondo Europeo de Desarrollo Regional (FEDER) of EU, project RTA2015-00088-C03-02 from the National Institute for Agricultural Research and Technology INIA) and Sierra de Guadarrama National Park in the monitoring of the black vulture colony Alto Lozoya. Laura Ruiz-Ripa has a pre-doctoral fellowship from the Universidad de La Rioja. Rosa Fernández-Fernández has a predoctoral fellowship FPU from the Ministerio de Ciencia, Innovación y Universidades of Spain (FPU18/05438). Maria Cruz Camacho was supported by a predoctoral fellowship from the regional government of Castilla -La Mancha (JCCM).