Detection of Enterotoxigenic Potential and Determination of Clonal Profile in Staphylococcus aureus and Coagulase-Negative Staphylococci Isolated from Bovine Subclinical Mastitis in Different Brazilian States

Epidemiological studies have identified Staphylococcus aureus as the most common agent involved in food poisoning. However, current research highlights the importance of toxigenic coagulase-negative staphylococci (CoNS) isolated from food. The aim of this study was to characterize Staphylococcus spp. isolated from cows with bovine subclinical mastitis regarding the presence of genes responsible for the production of staphylococcal enterotoxins and of the tst-1 gene encoding toxic shock syndrome toxin 1, and to determine the clonal profile of the isolates carrying any of the genes studied. A total of 181 strains isolated in different Brazilian states, including the South, Southeast, and Northeast regions, were analyzed. The sea gene was the most frequent, which was detected in 18.2% of the isolates, followed by seb in 7.7%, sec in 14.9%, sed in 0.5%, see in 8.2%, seg in 1.6%, seh in 25.4%, sei in 6.6%, and ser in 1.6%. The sej, ses, set, and tst-1 genes were not detected in any of the isolates. The typing of the isolates by pulsed-field gel electrophoresis revealed important S. aureus and S. epidermidis clusters in different areas and the presence of enterotoxin genes in lineages isolated from animals that belong to herds located geographically close to each other.


Introduction
Several microorganisms are involved in mammary gland infections in production animals. National and international epidemiological studies have demonstrated the presence of the genus Staphylococcus in approximately 50% of bovine mastitis cases, highlighting the role of this group as the main causative agent of this infection [1]. Although Staphylococcus aureus is the most common agent involved in subclinical mastitis, coagulase-negative staphylococci (CoNS) have gained importance as causative agents of intramammary infections. In a study conducted in Brazil, Rall et al. [2] found a prevalence of these microorganisms of 27.4%, while 28.6% of the isolates were S. aureus in a study conducted in Turkey [3].
Studies [4,5] have emphasized the importance of toxigenic CoNS isolated from food. Therefore, although S. aureus is the most common agent involved in food poisoning, there is current concern in the scientific community regarding CoNS, which have been recognized as opportunistic pathogens in human and animal infections, allied to risks of toxigenic lineages in cases of food poisoning in humans [5].
The staphylococcal enterotoxins (SE) type A (sea), B (seb), C (sec), D (sed), and E (see) are the classical staphylococcal toxins. These toxins have emetic activity and are usually associated with outbreaks of food poisoning [6]. In addition, recently described enterotoxins (ses and set), enterotoxin-like (SE-like) toxins that do exert emetic activity, and toxic shock syndrome toxin 1 (tst-1) are known [6,7]. The staphylococcal enterotoxin R (ser) was first described as an enterotoxin-like toxin by Omoe et al. [6], but experimental studies later proved its emetic activity [8].
The objective of this study was to characterize the enterotoxigenic potential of Staphylococcus spp. isolated from cattle with subclinical mastitis in six Brazilian states: Paraná (PR), Santa Catarina (SC), Rio Grande do Sul (RS), São Paulo (SP), Minas Gerais (MG), and Pernambuco (PE). Additionally, we determined the clonal profile and geographic distribution of the S. aureus and S. epidermidis strains that carried any of these genes.

Detection of Enterotoxin and tst-1 Genes
Seventy-six (41.9%) of the 181 strains were positive for at least one of the genes studied. The sea gene was detected in 33 (18.2%) of the isolates, seb in 14 (7.7%), sec in 27 (14.9%), sed in 1 (0.5%), see in 15 (8.2%), seg in 3 (1.6%), seh in 46 (25.4%), sei in 12 (6.6%), and ser in 3 (1.6%). The sej, ses, set, and tst-1 genes were not detected in any of the strains studied (Table 1). Figure 1 shows the comparison of toxin gene detection in the Staphylococcus aureus and CoNS isolates.  Figure 2 shows the geographic distribution of enterotoxin gene-positive Staphylococcus spp. isolates. The sea and seb genes were found to coexist with the other genes studied and were detected in Staphylococcus spp. isolates from all Brazilian states, except for the state of Minas Gerais where only isolates carrying the sec gene were detected.

Distribution of Staphylococcal Enterotoxin Genes According to the Region Studied
The sei and ser genes were only detected in Staphylococcus spp. from the southern states (PR, SC, and RS).    Figure 2 shows the geographic distribution of enterotoxin gene-positive Staphylococcus spp. isolates. The sea and seb genes were found to coexist with the other genes studied and were detected in Staphylococcus spp. isolates from all Brazilian states, except for the state of Minas Gerais where only isolates carrying the sec gene were detected.  Figure 2 shows the geographic distribution of enterotoxin gene-positive Staphylococcus spp. isolates. The sea and seb genes were found to coexist with the other genes studied and were detected in Staphylococcus spp. isolates from all Brazilian states, except for the state of Minas Gerais where only isolates carrying the sec gene were detected.

Distribution of Staphylococcal Enterotoxin Genes According to the Region Studied
The sei and ser genes were only detected in Staphylococcus spp. from the southern states (PR, SC, and RS).

Determination of Clonal Profile
Molecular typing by Pulsed-field gel electrophoresis (PFGE) was performed only on the S. aureus and S. epidermidis isolates that in which any enterotoxin genes had been detected. Therefore, among the 181 strains included in the study, 27 S. aureus isolates and 15 S. epidermidis isolates were analyzed, corresponding to the species with the highest prevalence of strains carrying enterotoxin genes. The sei and ser genes were only detected in Staphylococcus spp. from the southern states (PR, SC, and RS).

Determination of Clonal Profile
Molecular typing by Pulsed-field gel electrophoresis (PFGE) was performed only on the S. aureus and S. epidermidis isolates that in which any enterotoxin genes had been detected. Therefore, among the 181 strains included in the study, 27 S. aureus isolates and 15 S. epidermidis isolates were analyzed, corresponding to the species with the highest prevalence of strains carrying enterotoxin genes.
Analysis permitted the identification of three S. aureus clusters that simultaneously included ě3 isolates with similarity ě80% (Figure 3, Table 2). Similarity was 100% in all groups. Cluster A, consisting of seven strains, contained a larger number of isolates with enterotoxigenic potential. This cluster comprised six isolates from Paraná and one isolate from São Paulo. The other clusters (B and C) also contained isolates from the South region, but from the state of Santa Catarina ( Table 2). Cluster B comprised isolates originating from different states (SC and RS).   (7); seb (7); sec (1); seh (5) When the profile of S. epidermidis was analyzed, two clusters (A and B) were found. Each cluster contained three isolates from the South region (PR and SC, respectively) ( Figure 4).

Discussion
The sea gene, which exhibited the highest prevalence in this study (18.2%) and was detected in S. aureus and CoNS strains, is carried by a prophage [9] and can be easily disseminated among Staphylococcus spp. strains. Its product, enterotoxin A, is frequently associated with food poisoning since it is toxic at low concentrations [10,11]. Enterotoxin A is produced at the beginning of the exponential phase and its expression is not regulated by the accessory gene regulator (agr), different from enterotoxins B, C, and D, which depend on the agr system for maximum expression [11,12]. The sec gene is located on pathogenicity islands and can be divided into three subtypes (sec1, sec2, and sec3) based on antigenic differences and on the animal host with which it is associated. Some studies suggest that the heterogeneity of enterotoxin C is related to selection for modified sec sequences that facilitate the survival of S. aureus in their respective hosts [11,13]. In the present study, sec was the second most common classical enterotoxin after sea.
The sed gene was detected in only 0.5% of the strains studied, in an S. epidermidis isolate. This gene was not detected by Calsolari et al. [3] in toxigenic CoNS isolates, with only one S. aureus strain being positive. The sed gene is located on plasmid pIB485 [14] and enterotoxin D is the second most common toxin associated with food poisoning [11]. A small amount of this enterotoxin is able to cause illness, mainly in children and the elderly [15,16]. Nonetheless, the near absence of sed in the strains studied here suggests that it is scarcely related with Staphylococcus spp. isolates from mastitis cases in these regions of Brazil and consequently with possible events of food poisoning.
The data of a study on isolates from dairy products responsible for food poisoning in the state of Minas Gerais [16] are consistent with studies demonstrating that sea and seb are the most prevalent among the toxins identified [5]. The same study [16] showed the production of staphylococcal enterotoxins by CoNS using immunological assays.
The see gene was also detected in a small percentage (8.2%) and was not found in the study of Srinivasan et al. [17] who investigated S. aureus strains isolated from milk of cows with mastitis in the region of Tennessee, USA. None of the 78 isolates was positive for that gene. The see gene is carried by a prophage and studies have shown that the see, sed, and sea genes are closely related, with 81% sequence homology between see and sea [11,18].
Although the sei gene was only detected in a small percentage (6.6%), it was present in CoNS strains, which is an important fact rarely described in the literature. In contrast, in the study of Srinivasan et al. [17], 47 (60.3%) of the 78 S. aureus isolates studied carried the gene.
The ser gene was detected in three S. aureus isolates, all from the South regions. This gene was found to coexist with some other enterotoxins since two strains from SC carried the combination ser + sea or ser + sec, while in the third isolate from RS, ser was associated with the enterotoxin I gene (sei). These data highlight the importance of knowledge on movable genetic elements that can transfer these genes among Staphylococcus spp. The same has been described by Lawrynowicz-Paciorek et al. [19] who also evaluated the frequency of coexisting genes.
Regarding the clonal profile, several epidemiological studies on S. aureus in cattle have demonstrated the involvement of a large number of molecular profiles in the etiology of mastitis in the world. However, certain profiles tend to predominate in different geographic regions [19,20]. In the present study, a large number of S. aureus strains grouped with ě80% similarity were observed, even when the region studied included two different states, such as clusters A and B. In both cases, the source of one isolate differed from that of the others. This similarity among isolates obtained from animals that belong to herds from different states can be explained by the trade of animals between farms.
According to Buzolla et al. [21], strains with identical genotypes may possess characteristics that convey advantages for their survival in the environment, to colonize the udder, and/or to cause diseases. This fact was proven in the present study of enterotoxin genes, which demonstrated the presence of clusters with the same profile in different locations. Mechanical milking machines are important sources of transmission of staphylococci in dairy herds since they can be contaminated with microorganisms derived from the animal's skin and milk, or even from the hands of the operators [22].
Taken together, the results of clonal profile analysis showed the existence of important genetic similarity among the isolates, particularly among S. aureus isolated from intramammary infections of herds from different Brazilian states. Furthermore, PFGE was found to be an excellent technique to detect these clones, permitting the establishment of emergency control measures to prevent subsequent dissemination of virulent strains.

Origin and Identification of the Strains
The 181 Staphylococcus spp. strains isolated from cows with bovine subclinical mastitis were provided by Embrapa Dairy Cattle. The strains were isolated in six Brazilian states: Paraná (PR), Santa Catarina (SC), and Rio Grande do Sul (RS); São Paulo (SP) and Minas Gerais (MG); and Pernambuco (PE), corresponding to the South, Southeast, and Northeast regions of Brazil, respectively.
The phenotypic and genotypic identification of these strains was performed for confirmation of the genus and correct identification of the species. The genus Staphylococcus was identified as described by Koneman et al. [23] using coagulase and sugar (trehalose, maltose, and mannitol) fermentation tests. Strains belonging to the CoNS group were submitted to biochemical tests as proposed by Cunha et al. [24] for phenotypic identification of the species. Total DNA was extracted using the illustra ® kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Genotypic identification of CoNS was performed using primers targeting conserved sequences adjacent to the 16S and 23S genes by the internal transcribed spacer-polymerase chain reaction (ITS-PCR) described by Barry et al. [25] and Couto et al. [26], using primers G1 and L1. PCR using the Staur-4 and Staur-6 primers developed by Straub et al.

Detection of Enterotoxin and tst-1 Genes
PCR for the detection of the enterotoxin and tst-1 genes was performed according to the parameters described by Johnson et al. [28] and Cunha et al. [5].

Analysis by Pulsed-Field Gel Electrophoresis
The Staphylococcus spp. isolates that were positive for the enterotoxins by PCR were submitted to clonal profile analysis by pulsed-field gel electrophoresis (PFGE) according to the modified protocol of McDougal et al. [33].
The BioNumerics software (Version 7.0, Applied Maths, Sint-Martens-Latem, Belgium, 2015) was used for similarity analysis. Dice correlation coefficients were calculated and a dendrogram was generated using the UPGMA method (unweighted pair group method using arithmetic averages).
Band position tolerance and optimization were set at 1.25% and 1%, respectively. A similarity coefficient of 80% was chosen for the definition of clusters.

Spatial Distribution of Staphylococcal Enterotoxin Genes in Different Regions of Brazil
The addresses of all isolates were geocoded and added to the BioNumerics program (Version 7.0, Applied Maths, Sint-Martens-Latem, Belgium, 2015). Only strains carrying some staphylococcal enterotoxin genes or toxic shock syndrome were adopted as inclusion criteria on the map.

Conclusions
The enterotoxin A gene exhibited the highest prevalence in the present study, suggesting its easy dissemination among Staphylococcus spp. strains, since it was widespread in the georeferencing study. The recently described enterotoxin R gene is already found in isolates from some of the regions in Brazil and coexists with previously described enterotoxins. This fact poses a risk for the dissemination of this new type of enterotoxin among cattle herds. A limitation of the present study was the fact that we did not perform gene expression analysis since the simple presence of enterotoxin genes in Staphylococcus spp. isolates does not imply the occurrence of food poisoning. However, the present results contribute to the understanding of the enterotoxigenic profile of isolates from cows with bovine subclinical mastitis in different Brazilian herds, especially when considering that they are one of the most common microorganisms involved in intramammary infections in cattle and that milk is an excellent growth medium for Staphylococcus spp. The study also identified important S. aureus and S. epidermidis clusters in different regions and the presence of enterotoxin genes in lineages isolated from animals that belong to herds located geographically close to each other.