Antimicrobial Resistance and Clonal Lineages of Staphylococcus aureus from Cattle, Their Handlers, and Their Surroundings: A Cross-Sectional Study from the One Health Perspective

Staphylococcus aureus have been progressively identified in farm animals and in humans with direct contact with these animals showing that S. aureus may be a major zoonotic pathogen. Therefore, we aimed to isolate S. aureus from cows, their handlers, and their immediate surroundings, and to investigate the antimicrobial resistance and genetic lineages of the isolates. Mouth and nose swabs of 244 healthy cows (195 Maronesa, 11 Holstein-Friesians, and 28 crossbreeds), 82 farm workers, 53 water and 63 soil samples were collected. Identification of species was carried out by MALDI-TOF MS Biotyper. The presence of antimicrobial resistance genes and virulence factors was assessed based on gene search by PCR. All isolates were typed by multilocus sequence typing and spa-typing. From 442 samples, 33 (13.9%), 24 (29.3%), 1 (2%), and 1 (2%) S. aureus were recovered from cows, farm workers, water, and soil samples, respectively. Most of the isolates showed resistance only to penicillin. S. aureus isolates were ascribed to 17 sequence types (STs) and 26 spa-types. Some clonal lineages were common to both cows and farm workers such as ST30-t9413, ST72-t148, and ST45-t350. Through a One Health approach, this study revealed that there is a great diversity of clonal lineages of S. aureus in cows and their handlers. Furthermore, some S. aureus lineages are common to cows and handlers, which may suggest a possible transmission.


Introduction
Zoonotic disease events have highlighted the increasing effect of pathogens on human and animal health [1,2]. Therefore, in the past, the One Medicine concept was implemented, which aimed to address animal-human interactions and human and animal health [3].

Sample Collection
A total of 442 samples were collected from 64 farms in the North of Portugal, which comprises 244 cows (195 Maronesa breed, 11 Holstein-Friesian and 28 crossbreed), 82 farm workers, 53 water samples, and 63 soil samples from February to April 2019. Samples from cows and farm workers were collected with a nasal and mouth swab (one sample per individual). The farms are managed by families who are dedicated to agriculture and generally share the same household. Water samples were collected from the cows' drinkers using sterile 500 mL plastic bottles with sodium thiosulfate and preserved at 4-8 • C. All samples were filtered on the same day they were collected. Soil samples were collected from the farm grounds with a sterile plastic bag. The age of the cows ranged from 4 to 22 years with an average of 10 years and among the 244 cows, 229 were females, and 15 males (Supplementary Table S1).

S. aureus Isolation
The swabs and 2 g of soil sample were inserted into tubes containing 5 mL of Brain Heart Infusion (BHI) broth (LiofilChem, Via Scozia, Italy) with 6.5% of NaCl and incubated at 36 • C for 24 h. Then, the inoculum was seeded onto Baird-Parker agar (Oxoid, Basingstoke, UK) plates for S. aureus isolation. Water samples were filtered through a cellulose nitrate 0.45 µm pore membrane filter (Whatman, Maidstone, UK). The filters were then inserted into tubes BHI broth tubes 6.5% of NaCl and incubated at 37 • C for 24 h. After the incubation period, the inoculum was seeded onto Baird-Parker agar plates. Colonies, with S. aureus characteristics but showing morphological differences, were collected from each plate. S. aureus species identification was performed by biochemical tests (catalase, DNase and coagulase) and by MALDI-TOF MS Biotyper (Bruker Daltonics, Billerica, MA, USA).

Molecular Typing
All isolates were typed by MLST, spa-, and agr-typing. The spa region was amplified by PCR, the fragments sequenced, and the obtained sequences were analyzed using Ridom ® Staph-type software (version 1.5, Ridom GmbH, Würzburg, Germany) [35]. MLST genotyping was performed as previously described [36]. Allele and STs were determined using the Staphylococcus MLST database at https://pubmlst.org/ (accessed on 7 October 2021). Isolates were also characterized by agr-typing (I-IV) by PCR using specific primers and conditions [37].

Results
In this study, S. aureus strains were isolated from cows, farm workers, and the cows' surrounding environment (soil and water). S. aureus were found in 24 (37.5%) of the 64 farms included in this study. A total of 58 (13.1%) S. aureus were isolated from the 442 samples. From the 244 cows sampled, 32 (13.1%) were colonized by S. aureus. However, one cow co-carried two different lineages of S. aureus; thus, 33 S. aureus were isolated from cows. Among the three tested breeds, Maronesa, Holstein-Friesian and crossbreed, S. aureus were detected in 25. 2 and 6, respectively. Regarding the farm workers, 24 (29.3%) S. aureus were recovered from the 82 samples. Water and soil samples were collected from 53 cows' drinkers and 63 soil grounds and only one isolate of each origin was recovered. Table 1 shows the percentage of S. aureus isolates resistant and susceptible to each antibiotic. Farms with positive samples are listed in Table 2 as well as the resistance and virulence profiles, and clonal lineages of the isolates. In 8 (Farm 3, 10, 15, 17, 39, 42, 60 and 63) of the 24 positive farms, S. aureus was isolated from cows only and, in seven farms (Farm 6,20,25,46,48,56,and 58), it was isolated from farm workers only. Interestingly, in farms 14 and 55, S. aureus was isolated only from soil and water samples, respectively. In the remaining farms, S. aureus was isolated from both cows and farm workers. All isolates were characterized regarding their antimicrobial resistance, virulence, and clonal lineages. Nineteen isolates from cows (n = 15), farm workers (n = 2), soil (n = 1), and water (n = 1) were susceptible to all antibiotics tested ( Table 2). Multidrug resistance was found in one isolate from a cow (VS3222) and one from a human (VS3263). Resistance to penicillin was detected in 36 isolates and all carried the blaZ gene. Four isolates were resistant to aminoglycosides and harbored the aac(6 )-aph(2") (n = 4), aph(3 )-IIIa (n = 3) and str genes. Two isolates from cows and one from a worker showed resistance to tetracycline conferred by the tetK gene. Resistance to erythromycin was found in four S. aureus isolated from farm workers with two being co-resistant to clindamycin. Resistance to macrolides and lincosamides was encoded by the ermC (n = 2), ermT, and ermB. Only one isolate showed phenotypic resistance to fusidic acid, but none of the tested genes were present. Finally, two S. aureus isolated from a cow and its handler carried the cat pC221 gene, which is responsible for chloramphenicol resistance. Five isolates from cows and one from one farm worker were positive for the scn gene of the IEC system and were further investigated regarding the presence of the other IEC genes. The isolates were ascribed to IEC type B (n = 4), G (n = 2) and E. All isolates harbored the virulence genes hla and hld. As expected, the hlb gene was detected in all IEC-negative isolates (n = 56), and six isolates also carried the tst gene. All isolates were typed by MLST, spaand agr-type. The 58 isolates were affiliated to 18 STs and 26 different spa-types, with 6 and 12 distinct STs and spa-types for the bovine isolates and nine distinct STs and spa-types for human isolates ( Figure 1). The most common S. aureus lineage in cows was ST6-t18899 (n = 9/244) and in farm workers was ST30-t012 (n = 7/82). In general, S. aureus isolates from cows were ascribed to ST6 (n = 9), ST133 (n = 5), ST30 (n = 4), ST45 (n = 4), ST72 (n = 3), ST672 (n = 2), ST7464 (n = 2), ST352 (n = 2), ST1 and ST2328, and spa-types t16615 (n = 9), t9413 (n = 4), t18899 (n =3), t148 (n = 3), t959 (n = 2), t871 (n = 2), t3750, t2207, t7355, t7669, t4735, t350, t706, t015, t267, t359 and t563. Isolated from farm workers belonged to ST30 (n = 7), ST45 (n = 4), ST5 (n = 3), ST72, ST121, ST97, ST34, ST188, ST8 and ST398, and spa-types t012 (n = 5), t9413 (n = 2), t018 (n = 2), t045 (n = 2), t189 (n = 2), t148, t7669, t350, t015, t018, t002, t162, t414, t008 and t571. The two S. aureus isolates from water and soil were ascribed to ST30-t018 and ST6-t16615, respectively. As for agr-typing, the isolates were grouped into agr type I (n = 35), II (n = 3), III (n = 19), and IV (n = 1).  Evidence of a possible transmission of S. aureus between farm workers and cows is shown in Farms 16, 62, and 64. For instance, in Farm 16, the same clonal lineage ST30-t9413 in isolates from four cows and one worker, and all isolates have the same phenotype and genotype. In addition, isolates from two farm workers share the same linages which may also suggest a possible human-to-human transmission since workers from the same farm are related and share the same household, and the same is observed in Farm 13. Transmission between cows sharing the same environment may also occur. In Farms 10, 13, and 60, S. aureus from cows share the same clonal lineages among them, which possibly indicates a cow-to-cow transmission.

Discussion
Transmission of S. aureus between cows and people working with dairy cattle has been reported in 2007 [38]. Since then, many studies have been published with dairy cattle and the possible transmission between cows and farm workers [29,[38][39][40]. However, the great majority of studies focus only on S. aureus as a cause of bovine mastitis or its presence in bovine milk. In fact, S. aureus causing mastitis and the transmission to and from farm workers through direct contact have been extensively studied [29,40]. Indeed, studies investigating the presence of S. aureus in healthy beef cattle and the animal-humanenvironment transmission in the One Health context are scarce. In our study, we collected a total of 442 samples from cows, farm workers, and the farm environment (soil and water). In our previous study, we reported the absence of MRSA in Maronesa cattle, and so this is the first study reporting the presence of S. aureus in Maronesa cattle, which is an important traditional Portuguese breed [32]. From the 244 cows sampled, 13.1% were colonized by S. aureus, which is higher than most studies conducted with healthy cattle. Other studies conducted with healthy cattle reported an S. aureus frequency of between 5% and 8% [41][42][43]. Moreover, a study carried out in Tunisia reported an even lower frequency of S. aureus of only 1.3% in healthy cattle [44]. Likewise, Garipcin et al. investigated the presence of S. aureus in healthy cattle and humans in close contact with these animals and found a prevalence of 3.2% and 29.3% in cattle and humans, respectively [45]. The results of this study, in relation to samples of human handlers, is the same as that obtained in our study (29.3%). In fact, it has been reported that S. aureus is part of the normal mucosa of around 30% of the human population [46]. In contrast, another study carried out with samples from cattle and their caretakers found S. aureus in 42.9% and 74.2% in cattle and caretakers, respectively, which is a much higher frequency than most studies including ours [47]. Finally, another study similar to ours, in which the presence of S. aureus was investigated in cattle, caretakers, and the farm environment, found S. aureus in 4% and 16.6% of animal and human nose samples, but no S. aureus was found in the environmental samples. S. aureus and MRSA have been reported as environment associated with livestock including pigs, cattle and even in the production chain of dairy products [39,48,49]. In our study, the frequency of S. aureus in soil and water was also very low (1.6 and 1.9%, respectively). However, we excepted a higher frequency of S. aureus in soil samples since studies have shown that environmental sampling of barns and farms may be used for S. aureus and MRSA surveillance in livestock [50,51]. Furthermore, however, there is little information about the survival time of S. aureus on soil, and the manure spread on the farm soil could be a source of S. aureus on soil surfaces. We also expected to find a higher prevalence of S. aureus in the water of the cows' drinkers since S. aureus is present in the mouth and nose of cows and can spread in the water. This low frequency may be due to the S. aureus survival rate in fresh water, which was reported to be an average of 2.71 days and 4.84 days at 20 • C and 13 • C, respectively [52].
Zoonotic transmission of S. aureus strains between livestock and humans have been reported, particularly, with humans living and working in close contact with a farm [29,39]. S. aureus transmission between cattle and farm workers may occur through direct contact or in indirect exposure through the farm environment [39]. In our study, farm environment contamination did not seem to promote S. aureus colonization in both cattle and farm workers since only two environmental S. aureus were isolated from different farms (farms 14 and 55), and no S. aureus was detected in the cows or in the workers of those farms. Potential transmission between cows and workers was detected in farms 16, 49, 53, and 62. In farm 16, all cows were colonized by S. aureus ST30-t9413 carrying the blaZ, hla, hlb, and hld genes, and one of the farm workers was also colonized by the same S. aureus clone harboring the same genes. In addition, two other workers also carried S. aureus ST30 but with a different spa-type (t018). S. aureus ST30 was the predominant clone found in this study and was detected in cows, humans, and soil samples. ST30 is primarily associated with humans but is also spread among livestock, including cows and pigs [53,54]. Furthermore, CC30 comprises the most common MSSA lineage in Europe and gave rise to important epidemic clones such as EMRSA-16 [55,56]. In this study, ST30 isolates were associated with three spa-types: t018, t9413, and t012. S. aureus from cows were exclusively typed as t9413, while S. aureus ST30 from humans were typed as t018, t9413, and t012. S. aureus ST30-t012 isolate may be related to the Southwest Pacific clone and was the most prevalent clone among community and hospital settings in Portugal between 1992 and 2011 [56]. ST30-t9413 has only been reported in Portugal in strains isolated from wild owls, superficial waters and one farm worker with close contact with cattle, and all studies were conducted in the same region as this study [19,32,57]. spa-type t9413 may be cattle-associated and the ST30-t9413 isolated from farm workers in this study may have an animal origin. Furthermore, CC30 isolates were the only ones carrying the virulence gene tst, but none of the ST30-t9413 harbored this gene. The carriage of tst, in addition to the hemolysins genes, is in accordance with other studies that have shown that S. aureus ST30 often carries pathogenicity islands including tst gene [58]. Other S. aureus isolated in this study belonged to CC30, such as S. aureus ST7464-t871 detected in two cows from farm 39 and S. aureus ST34-t414 isolated from a farm worker (farm 47). Another possible piece of evidence of S. aureus human-toanimal and animal-to-animal transmission was detected in farms 64 and 60, respectively. All S. aureus isolates were typed as ST72 (CC8) and spa-type t148. S. aureus ST72 was first described in South Korea and is a particularly rare clone elsewhere in the world [59]. However, it is mostly associated with MRSA strains frequently found in the community and hospitals [60]. However, MSSA ST72-t148 has also been reported as a common cause of blood infection in Korea [61]. S. aureus ST45 was detected in four cows and four farm workers in this study and associated with five spa-types: t015, t7669, t350, t563, and t706. S. aureus ST45-t7669 was detected in one cow and one farm worker from farm 62 and, since both isolates encode the same resistance and virulence genes, we can suggest a possible bacterial transmission. S. aureus ST45 is a human-associated clone and is a major global MRSA lineage [62]. Nevertheless, MSSA ST45 has been detected in cow mastitis and farm workers with direct contact [63,64]. Effelsberg et al. analyzed a large collection of ST45 isolates from six continents and reported that ST45 phylogeny is defined by two distinct sublineages which correlated with geographical origins of the isolates [62]. However, in our study, since 3 of the 4 ST45 isolates from cows carried the IEC system genes, we can suggest that it may indicate a human spillover rather than an animal-associated ST45 sublineage as previously stated [62]. S. aureus ST6-t16615 was the most prevalent lineage in cows and was not detected in human samples. This lineage has been reported among wild rats and owls in Portugal [57,65] and as the main lineage in livestock in Algeria [66]. Although considered a human clone with relatively high prevalence in Asian countries, this lineage seems to be widely disseminated among animals [67]. Furthermore, in our study, none of the isolates carried the IEC systems, which suggests a possible animal adaptation. Some of the remaining S. aureus lineages were only detected in cows: ST133, ST672, ST352, ST1, and ST2328. S. aureus ST133 and ST2328 belonging to CC133 and ST1 (CC1) are known to be livestock-associated and lately have emerged as important zoonotic lineages [68]. CC133 lineage is regarded as mostly ungulate-animal specific, but it has also been detected in wild animals and surface waters [19,57,69]. In fact, S. aureus CC133 has been reported as the most prevalent in bovine mastitis milk [70,71]. spa-type t18899, found in three ST133 isolates in our study, was only reported in milk samples [72]. ST672 lineage is an emerging strain from the Indian subcontinent often related with CA-MRSA and rarely found elsewhere [73]. In our study, both ST672 isolates carried the IEC genes and were ascribed to group G, which may confirm a human origin [68]. S. aureus ST352 belongs to CC97, which is an animal-specific lineage, but it has also been detected in one farm worker in this study. CC97 is a pandemic bovine S. aureus lineage that emerged as a zoonotic agent and has been reported as a human epidemic CA-MRSA after host adaptation [30,47]. Other S. aureus lineages were exclusively detected in farm workers such as ST5, ST121, ST188, ST8, and ST398. ST5, ST8, and ST188 classical human linages [47]. However, ST398-t571 is the most common livestock-associated S. aureus lineage in Europe [74]. As in animals, the spa-type t571 is the most common spa-type in MSSA ST398 in humans [75]. However, this isolate has characteristics typical of being of animal origin: it has resistance to tetracycline conferred by the tetK gene, which is known to be a livestock-associated marker, and lacks the IEC system genes, which is currently considered to be the marker for human host adaptation [76]. References [77][78][79][80][81][82][83][84][85][86][87][88][89][90][91] are cited in the Supplementary Materials.

Conclusions
In this study, both cows and farm workers are carriers of S. aureus strains. However, S. aureus was isolated from only one soil and one water sample, which may suggest a low survival of S. aureus in the environment. Several cow isolates that belonged to classical human genetic lineages were indistinguishable from S. aureus isolated from farm workers in close contact with the cows, which suggests a possible transmission from humans as previously evoked. Animal-to-human transmission may have also occurred, although in a smaller number of cases, which indicates an acquisition through occupational contact. Moreover, our results also provide the evidence of S. aureus transmission among cows and among humans sharing the same household, although the direction of transfer could not be proven.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10050941/s1, Table S1: Farms, date, and location of sample collection and distribution of the S. aureus isolates among cows, farmers, and environment samples; Table S2: Primer pairs used for molecular typing and detection of antimicrobial resistance genes in S. aureus strains.
Funding: This work was funded by the R&D Project CAREBIO2: Comparative assessment of antimicrobial resistance in environmental biofilms through proteomics-towards innovative theranostic biomarkers, with references NORTE-01-0145-FEDER-030101 and PTDC/SAU-INF/30101/2017, financed by the European Regional Development Fund (ERDF) through the Northern Regional Operational Program (NORTE 2020) and the Foundation for Science and Technology (FCT). This work was supported by the Associate Laboratory for Green Chemistry-LAQV, which is financed by national funds from FCT/MCTES (UIDB/50006/2020 and UIDP/50006/2020) and by the projects UIDB/CVT/00772/2020 and LA/P/0059/2020 funded by the Portuguese Foundation for Science and Technology (FCT). Vanessa Silva is grateful to FCT (Fundacão para a Ciência e a Tecnologia) for financial support through the PhD grant SFRH/BD/137947/2018.

Institutional Review Board Statement:
The study was conducted according to the Helsinki Declaration (ICH-GCP principles), in compliance with Schedule Y/ICMR Guidelines, the Oviedo Convention, and was approved by the Ethics Committee of University of Trás-os-Montes e Alto Douro (EC-UTAD, 8 November 2019).