Molecular Typing and Antimicrobial Susceptibility Profile of Staphylococcus aureus Isolates Recovered from Bovine Mastitis and Nasal Samples

Simple Summary Staphylococcus aureus is a major, prevalent mastitis pathogen, representing a real issue for bovine udder health, with unquestionable importance in human and veterinary medicine. The present study thus aimed to determine the antimicrobial resistance and the diversity of S. aureus recovered from transient and persistent intramammary infections and from extramammary niches (e.g., nares/muzzles) in dairy cows. We found that a large proportion of S. aureus strains exhibited multidrug resistance to antimicrobials, including resistance to antimicrobials that are critically important to human health. S. aureus isolates from transient and persistent IMIs did not differ, suggesting that the persistence of bovine intramammary infections (IMIs) was mainly determined by host factors, although S. aureus isolated from extramammary niches are not an important source of S. aureus intramammary infections. Furthermore, a discrepancy in antimicrobial resistance between S. aureus strains isolated from nares/muzzles and intramammary infections was observed. Abstract In the present study, we aimed to determine the antimicrobial resistance and molecular typing of Staphylococcus aureus recovered from transient and persistent intramammary infections and nares/muzzles in dairy cows. We investigated the antimicrobial resistance of 189 S. aureus strains using a broad antimicrobial susceptibility profile. Furthermore, 107 S. aureus isolates were strain-typed using staphylococcal protein-A (spa) typing. A large proportion of strains exhibited multidrug resistance to antimicrobials, including resistance to critically important antimicrobials, although no methicillin-resistant S. aureus strains were found. Our study did not strengthen the idea that extramammary niches (i.e., nares/muzzles) are an important source of S. aureus for bovine mastitis. A discrepancy in the antimicrobial resistance between S. aureus strains isolated from nares/muzzles and milk samples was observed. Furthermore, S. aureus isolates from transient and persistent intramammary infections (IMIs) did not differ by spa typing, suggesting that the persistence of bovine IMIs was determined by cow factors. Thus, the high level of multidrug-resistant S. aureus found in the two herds, considered together with the predominance of a well udder-adapted S. aureus strain, may contribute to our knowledge of the history of the high prevalence of mastitis caused by S. aureus, which is of great concern for animal and public health.


Introduction
Bovine mastitis is the most prevalent and costly disease that affects dairy farming. Moreover, it has great implications for milk production, the quality of milk and dairy products, antimicrobial usage, animal welfare, the environment, and the image of the dairy sector. Among other mastitis pathogens, Staphylococcus aureus is a major, prevalent mastitis pathogen, which represents a significant issue for bovine udder health, with unquestionable importance in human and veterinary medicine [1]. Although antibiotic treatment is widely used to fight bovine mastitis, S. aureus resistance to antimicrobials not only complicates antimicrobial treatment, but also represents a huge challenge for public health and food security, as cows are the major reservoir for the emergence of S. aureus human epidemic clones [2].
Furthermore, from the epidemiological point of view, it is crucial to increase our understanding on the circulating bovine-associated S. aureus strains among the cow population. A variety of molecular methods have been extensively used for typing S. aureus isolates, of which staphylococcal protein-A (spa) typing is one among the most common. spa typing is suitable for epidemiological studies and produces reproducible, unambiguous, and easily interpreted results [3]. Currently, the growing spa typing database developed by Harmsen et al. [4] is the largest database for typing S. aureus, surpassing that of multilocus sequence typing.
Thus, the objective of the present study was to determine the antimicrobial resistance and molecular typing of S. aureus strains recovered from transient and persistent intramammary infections and nares/muzzles.

Materials and Methods
This study was approved by the Animal Research Ethics Committee of the Federal University of Minas Gerais, Minas Gerais, Brazil, under the protocol number 201/2011. All bacterial strains were collected from two dairy herds with approximately 125 lactating dairy cows per herd between January 2013 and January 2014. Both herds had a high bulk tank milk somatic cell count (≥5.0 × 10 5 cells mL −1 ) and a history of a high level of mastitis caused by S. aureus (as determined by a veterinarian and herd statistics). At the beginning of the study, 24.80% and 15.75% of lactating dairy cows had an intramammary infection (IMI) by S. aureus in herds A and B, respectively. The dairy farms located in Minas Gerais state, Brazil, are geographically distant (approximately 450 km apart).

S. aureus Isolates from Milk Samples
We used 182 S. aureus isolates previously identified by biochemical tests [1,5]. All S. aureus isolates were further identified by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS; Bruker Daltonics, Bremen, Germany), as previously described by Nonnemann et al. [6]. Furthermore, S. aureus identification was confirmed by polymerase chain reaction (PCR) analysis targeting a portion of the thermonuclease gene conserved in S. aureus (nuc) [7]. All S. aureus isolates from milk samples were analyzed by an antimicrobial susceptibility test.
The S. aureus strains isolated from milk samples were further categorized into those derived from persistent and transient IMIs. A quarter was defined as having an IMI caused by S. aureus if ≥100 colony-forming unit colonies mL −1 were detected in the milk. An IMI was regarded as transient if S. aureus was isolated at only one sampling of the consecutive samplings, in the period between the first and the last samplings. A persistent IMI was assumed if a quarter had an IMI at ≥3 consecutive samplings (at least a one-month interval) caused by the same S. aureus strain. From those isolates, we selected 100 S. aureus isolates from milk samples according to their epidemiological behavior (transient vs. persistent IMIs) for spa typing.

S. aureus Isolates from Muzzle/Nare Samples
From the same herds and period, the nares/muzzles of dairy cows were sampled by swabbing the muzzle and the inner nares with a single moistened sterile cotton swab, as previously described [8]. Then, the swabs were spread onto the surface of Baird-Parker agar (Oxoid) plates with 5% egg yolk tellurite emulsion and incubated at 37 • C for 24-48 h under aerobic conditions. From each plate, if present, three gray-to-black colonies surrounded by clear zones, as a result of proteolysis and lipolysis, and three gray-to-black colonies without clear zones were selected and transferred separately to microtubes containing 900 µL of brain heart infusion (BHI), and incubated overnight at 37 • C, then glycerol (10% final concentration) was added, and the samples were stored at 80 • C until identification. Afterwards, for bacterial identification, the bacterial isolates (n = 159) were spread onto BHI agar plates for 24-48 h at 37 • C. The bacterial isolates were first subjected to Gram staining and catalase and coagulase tests and further identified by MALDI-TOF MS [6]. Furthermore, S. aureus identification was confirmed by PCR analysis targeting a portion of the S. aureus nuc gene [7]. Among the 159 bacterial isolates, all S. aureus isolates identified by both MALDI-TOF MS and PCR as S. aureus were subjected to spa typing and antimicrobial susceptibility tests.

Spa Typing
First, deoxyribonucleic acid (DNA) extracted from the bacterial culture in BHI broth by means of a method adapted from the boiling methodology described by Fan et al. [9], where phosphate-buffered saline (PBS PO 4 0.01 M, NaCl 0.15 M, pH 7.2) was replaced with tris-Ethylenediaminetetraacetic acid (EDTA) buffer (TE tris-HCl 10 mM, EDTA 1 mM, pH 8.0). The repeated region of S. aureus protein A was amplified with the primers previously described by Harmsen et al. [4], and the following PCR conditions were used: heating at 95°C for 5 min, 35 cycles of amplification at 95°C for 45 s, annealing at 60°C for 45 s, extension at 72°C for 90 s, and a final extension at 72°C for 10 min. The DNA sequences were obtained with an ABI-3500 automatic sequencer (Applied Bisystems ® , Foster, CA, USA). Spa types were determined with the protocol recommended by the Ridom spa Server (http://www.spaserver.ridom.de). The obtained spa-type sequences were analyzed using the spa plugin included in Bionumerics 7.6 (Applied Maths NV, Sint-Martens-Latem, Belgium).

Antimicrobial Susceptibility Tests
A broad antimicrobial susceptibility profile was analyzed by the automated Vitek 2 ® compact system (BioMériux, Inc., Durham, NC, USA) by determining minimum inhibitory concentration (MIC) using veterinary susceptibility AST-GP69 card (BioMériux, Inc., Durham, NC, USA) panels for Gram-positive bacteria. The following antimicrobials were tested by the Vitek 2 ® kit: ampicillin/sulbactam, benzylpenicillin, cefoxitin screen, chloramphenicol, clindamycin, inducible resistance to clindamycin, enrofloxacin, erythromycin, fusidic acid, gentamicin, imipenem, kanamycin, marbofloxacin, mupirocin, nitrofurantoin, oxacillin, rifamycin, tetracycline, trimethoprim/sulfamethoxazole, and vancomycin. S. aureus isolates were regarded as multidrug-resistant if they were not susceptible to three or more classes of distinct antimicrobials. Furthermore, S. aureus resistance to penicillin was excluded from the definition of multidrug-resistance because of the widespread resistance of S. aureus to this antimicrobial agent, as proposed by Magiorakos et al. [10]. All isolates were also tested by the Kirby-Bauer disk diffusion technique using disks for oxacillin (1 µg) and cefoxitin (10 UI/30 µg) for the prediction of methicillin-resistant S. aureus (MRSA). All antimicrobial susceptibility criteria were interpreted according to the Clinical and Laboratory Standards Institute [11,12] and the European Committee on Antimicrobial Susceptibility Testing [13].
In addition, all S. aureus isolates phenotypically regarded as oxacillin-and/or cefoxitin-resistant by MIC using the Vitek 2 ® compact system or the Kirby-Bauer disk diffusion technique were further investigated by Etest ® (bioMérieux, Basingstoke, UK) and the presence of the methicillin resistance gene (mecA and mecC). PCR analysis for the detection of the mecA and mecC genes was performed as previously described by Mehrotra et al. [14] and Paterson et al. [15], respectively.
The antimicrobial susceptibility results by the Vitek 2 ® Compact system are summarized in Table  S1. Our results showed that 46.56% (n = 88) of S. aureus isolates were not susceptible to at least three distinct classes of antimicrobials. Beyond that, even if we excluded resistance to penicillin, 30 (15.87%) isolates were regarded as multidrug-resistant S. aureus. We found eight and four oxacillin-resistant and cefoxitin-resistant S. aureus isolates from milk samples using the Vitek 2 ® Compact system, respectively. Therefore, although most of them (seven and four oxacillin-resistant and cefoxitin-resistant S. aureus isolates, respectively) were confirmed by the Kirby-Bauer disk diffusion technique, the resistance to methicillin was not confirmed by the E-test ® or the presence of the mecA and mecC genes, suggesting that none of them could be regarded as MRSA. Although a limited number of S. aureus strains were isolated from nasal/muzzle samples and characterized in this study, a discrepancy in antimicrobial resistance between S. aureus isolated from nares/muzzles and milk samples was observed. For instance, while a great proportion of the S. aureus isolates from milk samples were resistant to benzylpenicillin (94.50%), gentamycin (80%), and tetracycline (43.96%), none of the S. aureus isolates from nares/muzzles were resistant to these antimicrobials. In contrast, the S. aureus isolates from nares/muzzles showed resistance to clindamycin and erythromycin (Table S1).

Discussion
Apart from the few available studies that have focused on the ecology of S. aureus [8,16] and those that have investigated non-aureus staphylococci isolated from body sites [17], information on the frequency of bacterial isolates from nares/muzzles in dairy cows is limited. Although S. aureus colonization of nares represents a primary reservoir of this pathogen in humans, nares/muzzles did not appear to be a major reservoir for S. aureus in dairy cows, corroborating the findings of Capurro et al. [16]. Rather, nares/muzzles appeared to be mainly colonized by non-aureus staphylococci in dairy cows.
Furthermore, our study did not strengthen the idea that extramammary niches (i.e., nares/muzzles) are an important source of S. aureus strains that cause persistent IMIs in dairy cows. In agreement with our results, Leuenberger et al. [18] reported that the S. aureus genotype B is highly associated with the mammary gland, whereas other S. aureus genotypes (e.g., C and R) that cause sporadic and noncontagious IMI were the most abundant in extramammary niches, such as the hocks, teat skin, nostrils, and perineum. The only S. aureus strain (ST89) originating from the nose was also detected in milk samples, and the latter was associated with a transient IMI in another cow, indicating that if it does cause an IMI, it is likely associated with a sporadic and noncontagious IMI. Moreover, one S. aureus strain isolated from herd A was the same strain that caused persistent IMI, suggesting that this strain might adapt to extramammary niches and colonize them, as previously suggested for teat skin [19]. Thus, even though S. aureus can colonize extramammary niches of dairy cows (i.e., nares/muzzles) [8,16], our study showed that the udder is the most important reservoir of these bacteria, which also substantiates their contagious behavior.
We also showed that one S. aureus strain (spa type t605) was widespread in the two herds investigated, although the herds are geographically distant. This S. aureus strain was responsible for almost all IMIs, probably because it is well adapted to the udder, leading to the persistence of IMIs. Thus, our results are in agreement with other reports that indicated that a few udder-adapted S. aureus types with broad geographic distribution are responsible for most of the IMIs in dairy herds [20].
Another important finding was that the herd that only had one S. aureus strain was a closed herd, containing no cows that had been purchased from other herds in recent decades, in contrast to the other herd, an open herd, in which new S. aureus strains could be introduced by animals from foreign herds. We identified an additional S. aureus strain in the open herd. Although the owner of the open herd acquired several dairy cows from distinct dairy herds (personal communication), the most udder-adapted S. aureus strain might spread more efficiently, preventing the spread of other less well-udder adapted (i.e., opportunistic) S. aureus strains. In the present study, the S. aureus ST605 clone was the most important clone related to bovine IMIs, which is in agreement with another Brazilian study [21]. To the best of our knowledge, this study is the first report of the isolation of the S. aureus ST098 clone from bovine samples. In this study, the S. aureus ST098 clone was the most common S. aureus strain isolated from nasal samples. We also demonstrated that the persistence of bovine IMIs was probably mainly determined by host factors, most likely related to innate and adaptative immune responses, as the same S. aureus strain (ST605 clone) caused persistent and transient IMIs.
The antimicrobial susceptibility results from the Vitek 2 ® Compact system are summarized in Table  S1. Our results showed that 46.56% (n = 88) of S. aureus isolates were not susceptible to at least three distinct classes of antimicrobials. Beyond that, even if we excluded resistance to penicillin, as proposed by Magiorakos et al. [10], 30 (15.87%) isolates were regarded as multidrug-resistant S. aureus. A high proportion of S. aureus isolates originated from IMIs harboring resistance to tetracycline and penicillin has been previously observed in Brazil [22,23], which corroborates our findings. Nevertheless, our study showed a high level of resistance to gentamycin in S. aureus originating from IMIs, although gentamycin resistance frequencies reported are generally low [22,23]. The high level of multidrug resistant S. aureus found here, together with the predominance of an udder-adapted strain, may contribute to our knowledge of the history of the high level of mastitis caused by this pathogen, as the existence of penicillin-resistant S. aureus is related to low cure rates for S. aureus IMI treatment with either β-lactam or non-β-lactam antimicrobials [24]. Although no MRSA was detected here, our study showed that the S. aureus isolates from IMIs had considerable resistance to antimicrobials, including resistance to critically important antimicrobials, such as macrolides (e.g., erythromycin) and glycopeptides (e.g., vancomycin). Moreover, some intermediate resistance to critically important antimicrobials such as the quinolone group (e.g., marbofloxacin and enrofloxacin) was also found. Altogether, our data emphasize that S. aureus IMIs are concerning to animal and public health.
We hypothesize that the divergence in antimicrobial resistance could be attributed to, at least in part, the distinct history of exposure to antimicrobials between S. aureus isolated from noses/muzzles compared to those isolated from milk samples. For instance, S. aureus isolates from noses/muzzles may have been exposed to antimicrobials (e.g., macrolides) used systemically since the early stages of life (e.g., for respiratory diseases and diarrhea treatment), resulting in exposure for a longer time (since colonization). These antimicrobials are totally different from the antimicrobials which are mainly used locally for mastitis treatment (e.g., β-lactams) after the first parturition.

Conclusions
Overall, we can conclude that S. aureus strains isolated from noses/muzzles displayed distinct strain-typing and antimicrobial resistance profiles from those isolated from IMIs. Nonetheless, S. aureus isolated from transient and persistent S. aureus IMIs did not reveal divergent strain typing patterns, suggesting that the persistence of S. aureus IMIs is mainly determined by host factors. Table S1 is available online at http://www.mdpi.com/2076-2615/10/11/2143/s1, Table  S1. Antimicrobial susceptibility profile of the Staphylococcus aureus isolates from milk and nasal/muzzle swab samples determined using the automated Vitek 2 ® compact system by minimum inhibitory concentration.