Molecular Characteristics of Staphylococcus aureus Causing Bovine Mastitis between 2014 and 2015

Staphylococcus aureus is highly pathogenic and can cause diseases in both humans and domestic animals. In animal species, including ruminants, S. aureus may cause severe or sub-clinical mastitis. This study aimed to investigate the molecular profile, antimicrobial resistance, and genotype/phenotype correlation of 212 S. aureus isolates recovered from cases of bovine mastitis from 2014 to 2015 in the Shanghai and Zhejiang areas of China. Nineteen sequence types (STs) were determined by multi-locus sequence typing, while the dominant ST was ST97, followed by ST520, ST188, ST398, ST7, and ST9. Within 14 methicillin-resistant S. aureus (MRSA) isolates and 198 methicillin-susceptible S. aureus (MSSA) isolates, ST97 was the predominant MSSA clone and ST9-MRSA-SCCmecXII-spa t899 was the most common MRSA clone. The MRSA strains showed much higher rates of resistance to multiple antibiotics than did MSSA strains. Compared with other MSSA strains, MSSA ST398 was more resistant to clindamycin, erythromycin, and ciprofloxacin. No isolates were resistant to vancomycin, teicoplanin, or linezolid. The molecular profiles of the virulence genes varied in different strains. ST520 strains carried seg-sei-sem-sen-seo genes, and ST9 and ST97 harbored sdrD-sdrE genes. Virulence phenotype analysis showed diversity in different clones. Biofilm formation ability was significantly enhanced in ST188 and ST7, and red blood cell lysis capacity was relatively strong in all S. aureus strains of animal origin except ST7. Our results indicate that MSSA was the predominant S. aureus strain causing bovine mastitis in eastern regions of China. However, the presence of multidrug resistant and toxigenic MRSA clone ST9 suggests that comprehensive surveillance of S. aureus infection should be implemented in the management of animal husbandry products.


INTRODUCTION
Staphylococcus aureus is a common facultative pathogenic bacterium that has long been recognized as a challenge in both human and veterinary medicine (Nemeghaire et al., 2014). In cattle, it is responsible for approximately one-third of cases of clinical and subclinical mastitis (Bradley et al., 2007;Botrel et al., 2010), a disease that causes major economic loss in the dairy industry worldwide.
Since MRSA was first reported in the United Kingdom in 1961, it has become a global cause of hospital-associated (HA) and community-associated (CA) infections (Uhlemann et al., 2014). In the past few decades, a new subgroup of S. aureus, socalled livestock-associated S. aureus (LA-SA) has been described. There have been several reports of MRSA colonization and/or infections in dairy cattle since the first report of MRSA in mastitis in 1972 (Devriese et al., 1972;Monecke et al., 2007;Fessler et al., 2010;Huber et al., 2010). LA-SA ST398 was originally reported to have emerged in the Netherlands among pigs and pig farmers in 2003 (Voss et al., 2005), and was later found in Austria, Germany, and Denmark (Van Cleef et al., 2011;Bal et al., 2016). It developed into an overwhelmingly dominant lineage in Europe and North America (Vanderhaeghen et al., 2010;Cuny et al., 2013). Unlike those in Europe and North America, the epidemic LA MRSA clones in Asian countries have their own characteristics. LA MRSA ST9 has been predominantly isolated from pigs in China, Hong Kong, Taiwan, Thailand, and Malaysia. LA MRSA ST221 has been reported in Japan, and LA MRSA ST398 has emerged among pigs in Singapore and South Korea (Cui et al., 2009;Guardabassi et al., 2009;Neela et al., 2009;Baba et al., 2010;Ho J. et al., 2012;Larsen et al., 2012). Although the sequence type of LA MRSA isolates was mainly ST9 in most Asian countries, these isolates harbored different SCCmec elements in different areas, such as SCCmec III in China, SCCmec IVb, or V in Hong Kong, SCCmec V in Malaysia, SCCmec IX in Thailand and non-types I-VII in Taiwan Fang et al., 2014;Chuang and Huang, 2015). ST9 strains were usually associated with the MDR phenotype, with high resistance rates (>80%) to erythromycin, clindamycin, tetracycline, ciprofloxacin and gentamicin (Wan et al., 2013). To date, most studies on LA S. aureus have been performed in pigs, mainly with MRSA strains, even though the vast majority of S. aureus strains are host-specific. The data on molecular typing and antibiotic-related studies of bovine derived S. aureus are very limited. A relatively higher incidence of S. aureus mastitis has been reported in China than in other countries, which may cause a particular public threat (Li et al., 2011;He et al., 2014). According to previous studies performed in six Chinese provinces from March 2010 to August 2013, the most popular S. aureus clones causing bovine mastitis were primarily ST97 (51.9%), ST398 (13.6%) and ST2154 (8.6%), with ST97-MRSA-SCCmec IV representing the most common clone (50%; Wang et al., 2015). However, the characteristics of S. aureus from dairy cows in eastern regions of China have not yet been thoroughly discussed. To provide elementary evidence for developing the appropriate treatment and control measures for bovine mastitis, it is essential to understand the molecular epidemiology and the antibiotic resistance of S. aureus infections locally. Therefore, the present study was designed to analyze the molecular characteristics of S. aureus strains in bovine mastitis isolated from 2014 to 2015 in the Shanghai and Zhejiang areas.

Sample Collection and Bacterial Isolation
In total, 212 S. aureus isolates were isolated from dairy cows with mastitis from 2014 to 2015 in farms in the Shanghai and Zhejiang areas of China. Milk samples were taken from cows with clinical mastitis that had symptoms including color change in the milk, inflammation of the udder, and decreased milk production. For milk collection, the udders of the clinical mastitis cows were cleaned with water and dried. Cotton balls with 70% ethanol were used to disinfect the surfaces of the udder. The first few streams of milk were dropped. The collected milk was kept in a cooler with ice and transported to the laboratory within 8 h. The milk samples were cultured on 5% blood plate and inoculated at 37 • C for 24 h. S. aureus identification was based on Gram staining, classical microbiological tests, including catalase and coagulase activity, and were further characterized using VITEK 2 Compact GP ID Card (bioMérieux, Marcy l'Etoile, France). S. aureus ATCC43300 was used as a quality control organism. All S. aureus isolates were stored at −80 • C.
DNA Extraction, MRSA Identification, spa Typing, and SCCmec Typing DNA was extracted as previously described (Hartmann et al., 1997). MRSA identification and mecA gene detection were performed using a previously published triplex PCR (Maes et al., 2002). The strains were further characterized by spa typing, as previously described (Harmsen et al., 2003). The resulting spa type was assigned by submitting the data to the S. aureus spa type database (http://www.spaserver.ridom.de). Staphylococcal cassette chromosome mec (SCCmec) types were determined by means of two multiplex PCRs (M-PCRs) designed for the detection of the mec-complex and the ccr-complex (Kondo et al., 2007;Yan et al., 2016). Specific primers, as previously described  was used in isolates with SCCmec types that could not be determined using the SCCmec typing strategy as previously described (Kondo et al., 2007).

MLST Typing and goeBURST Algorithm
Isolates were screened using a previously described method (Enright and Spratt, 1999) to detect the following seven housekeeping genes: carbamate kinase (arcC), shikimate dehydrogenase (aroE), glycerol kinase (glp), guanylate kinase (gmk), phosphate acetyltransferase (pta), triosephosphate isomerase (tpi), and acetyl coenzyme A acetyltransferase (yqiL). The sequences of the PCR products were compared with the existing sequences available from the MLST website (http://www.mlst.net) for S. aureus, and the allelic number was determined for each sequence. The goeBURST algorithm (http://goeBURST.phyloviz. net) was used to infer the evolutionary relatedness of the STs.

PFGE Typing
PFGE was used to compare the genetic diversity of the 10 isolates of ST9 (one isolate of MSSA ST9 and nine isolates of MRSA ST9). Briefly, Sma I-digested DNA embedded in agarose plugs was subjected to PFGE analysis at 14 • C in a CHEF-MAPPER system (Bio-Rad) at 6 V/cm, in 0.5 × Tris-borate-EDTA buffer, in two stages: first stage, initial pulse, 5 s; final pulse, 15 s for 10 h; second stage, initial pulse, 15 s, final pulse, 60 s for 10 h; angle 120 • .

Semiquantitative Biofilm Assay
A semiquantitative biofilm assay was performed using 96-well tissue culture plates based on the method described by Wang et al. (Wang and Gao, 2008) with the following modification. After the cells were fixed in Bouin's fixative for 1 h, the cells were washed gently two times in phosphate-buffered saline and then stained with 0.1% crystal violet solution. The stain was washed off gently under slowly running water and the plates were dried at room temperature. The absorbance of the stained biofilm was measured at 560 nm using Synergy2 (Bio-Tek).

Human Erythrocyte Lysis Assay
Human blood was obtained from healthy people during physical examinations in Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China with ethical consent. Supernatants were collected from bacterial cultures grown for 8 h. Hemolytic activities were determined by incubating samples with human erythrocytes (2% v/v in Dulbecco's phosphate buffered saline, PBS) for 1 h at 37 • C. Hemolysis was determined by measuring the optical density at 540 nm using an enzymelinked immunosorbent assay reader.
We used the χ 2 and Fisher's exact tests, as appropriate for the analysis of categorical data. Unpaired two-tailed Student's t-tests were performed to analyze the statistical significance of biofilm formation and human erythrocytes lysis in S. aureus strains. P < 0.05 was considered statistically significant.

Antimicrobial Susceptibility Profiles
The results of antibiotic susceptibility testing showed that all of the isolates were susceptible to vancomycin, teicoplanin, and linezolid. Although the majority of the strains were resistant to penicillin (73.1%), they were still susceptible to most of the antibiotics tested. The resistance rates to other antibiotics tested were 31.6% to gentamicin, 23.6% to erythromycin, 16.5% to clindamycin, 11.8% to ciprofloxacin, 6.6% to cefoxitin, or oxacillin, 4.7% to cefuroxime, 4.2% to sulfamethoxazole/trimethoprim, and 3.8% to cefazolin ( Table 2).
However, all the MRSA strains were found to be resistant to at least four antibiotics. The resistance profiles of S. aureus isolates differed by their STs (Table 2). ST9 strains displayed the MDR phenotype, with high resistance (≥80%) to nine of the antibiotics tested FIGURE 1 | Application of the eBURST algorithm to MLST data for the collection of 212 S. aureus isolates. All 19 STs are represented by a filled circle. Blue and red circles represent STs that are group and sub-group founders, respectively. CC includes the groups of connected STs, considering that STs have at least six alleles in common with at least one other ST inside a CC.

Detection of Virulence Genes
In the present study, 30 virulence genes were screened in six predominant STs. The distribution of virulence genes differed among the strains according to ST. Some genes were present in all of the strains, but some genes were not found in any strain. For example, all of the strains carried four hemolysin genes (hla, hlb, hld, hlg), one serine protease gene (sspA), and three adhesion genes (icaA, clfA, fnbA), but none of the strains harbored tst, arcA, eta, etb, or pvl. The carriage of staphylococcal enterotoxin genes was strongly influenced by MLST profiles. Thirteen staphylococcal enterotoxin genes (sea, seb, sec, sed, see, seg, seh, sei, sem, sen, seo, seq, sek) were detected among these strains. Some enterotoxin genes (sea, seb, sed, seh, sej, sek, seq) could not be found in any strain. No enterotoxin gene was found in ST97, ST188, or ST398 isolates. The see-sep genes were only present in the ST7 strain, but the seg-sei-sem-sen-seo genes were present in ST7, ST9, and ST520 strains. Compared to the other STs, ST520 had the highest carrying frequency. The seg-sei-semsen-seo genes were present in all 10 ST520 isolates (Table 3).
Leucotoxins are regarded as virulence factors of S. aureus. Only ST520 (70%) carried both lukE and lukM. LukM could not be found in other STs, but lukE was detected in various other STs, including ST97, ST188, and ST7. The serine-aspartate repeat proteins (sdr) are members of a family of surface proteins that contribute to the pathogenicity of S. aureus. ST97 and ST9, which were the predominant STs of MSSA and MRSA, respectively, contained both sdrD and sdrE. Furthermore, sdrD only existed in ST7, but sdrE was present in various other STs, such as ST520, ST188, and ST398.

Biofilm Formation and Human Erythrocyte Lysis in S. aureus Strains
The ability of S. aureus to form biofilms is considered an important virulence factor because biofilms can tolerate antimicrobial agents, making the bacterium extremely difficult to eradicate. The biofilm assay was performed in the ST97 (randomly selected, n = 39), ST520 (randomly selected, n = 18), ST188 (randomly selected, n = 10), ST398 (randomly selected, n = 10), ST7 (randomly selected, n = 10), and ST9 (n = 10) strains. Compared to ST97 isolates, biofilm formation ability was enhanced significantly in ST188 and ST7 strains (Figure 2A). Moreover, ST7 strains had stronger biofilm formation than ST188 isolates (P < 0.05). There were no significant differences among the other STs, including ST97, ST9, ST520, and ST398. Hemolysis (lysis of erythrocytes) is also a significant virulence determinant of S. aureus, representing a crucial means for the bacteria to acquire iron. Lysis of red blood cell capacity was relatively strong in all S. aureus strains of animal origin, except ST7 ( Figure 2B). The hemolytic capacities of ST7 and ST188 strains were weaker than in ST97 isolates. Among these strains, ST7 had the strongest biofilm and the hemolytic capacity.

DISCUSSION
S. aureus is the most common causative organism of bovine mastitis, which is considered a common, complicated and economically unbearable disease in dairy animals worldwide (Wyder et al., 2011). Dairy animals with mastitis frequently shed S. aureus into the milk supply, which can contribute to food poisoning in humans. This study intended to investigate  STs with < 10 isolates were not included when calculating the percentages of virulence genes.
Frontiers in Cellular and Infection Microbiology | www.frontiersin.org the molecular profile and antimicrobial resistance of S. aureus isolates recovered from bovine mastitis from 2014 to 2015 in eastern regions of China.
In this study, we characterized 212 S. aureus isolates from bovine mastitis milk samples collected from farms in Shanghai and Zhejiang areas and identified 6.6% as MRSA strains, which was lower than western countries and Hong Kong (16-21.3%; Guardabassi et al., 2009), but higher than Korea (21/657, 3.2%; Lim et al., 2012), Malaysia (1.4%; Neela et al., 2009), and Japan (0.9%; Baba et al., 2010). These different frequencies may be due to the different animal populations studied or the implemented methodologies, among other factors.
Our study identified 14 bovine MRSA strains, including nine ST9-SCCmecXII-t899 strains, four belonging to SCCmec IV type (two ST965-t062, one ST59-t437, one ST537-t2755), and one SCCmecV strain (ST630-t4549). The MRSA isolates from eastern regions of China belonged to different spa and SCCmec types, suggesting their genetic diversity. In those previous studies, CC9 was identified as the predominant MRSA associated with pig farming and harbored various SCCmec types in different areas. For example, SCCmec was type III in mainland China, types IVb or V in Hong Kong, non-types I-VII in Taiwan, type V in Malaysia and type IX in Thailand (Cui et al., 2009;Ho J. et al., 2012;Larsen et al., 2012). In addition, SCCmec XII was a new SCCmec element that was first identified in two ST9-MRSA-t899 strains from pigs in China (Ho P. L. et al., 2012;Wang et al., 2015;Wu et al., 2015). It included a rare class C2 mec gene complex and a ccrC gene complex. The newly identified SCCmec was 48.575 kb in length and contained 45 predicted open reading frames. Therefore, the emerging ST9-MRSA-SCCmec XII strains can be isolated in both pigs and bovines in China, suggesting the potential possibility of its crossspecies transmission. Notably, the CC9 MRSA isolates in the current study were shown to be multidrug-resistant, with high resistance rates to β-lactams and eight other antibiotics widely used in chemotherapy. The presence of multidrug resistant MRSA clone CC9 suggests an alarming situation and could be a serious challenge to therapy. The PFGE result in our study showed that the restriction profiles of 10 CC9 strains were identical with only one fragment difference between MRSA CC9 and MSSA CC9 (Supplementary Image 1). These observations support the notion that CC9 MSSA lineages may provide a stable genetic environment for the integration of SCCmec in favor of their infection and transmission in different areas.
Among the 212 S. aureus isolates, the prevalence of MSSA in this study was 93.4%. In addition, our study revealed more heterogeneous MSSA lineages, 16 distinct STs were identified in MSSA, but only five STs were found in MRSA. The most frequently represented lineage in MSSA was CC97 (39.2%, 83/212), which accounted for more than one third of all S. aureus isolates, followed by CC520 (16%, 34/212), CC188 (10%, 21/212), CC398 (7.1%, 15/212), and CC7 (5.2%, 11/212). CC97 was found to be the most predominant clone, which is consistent with a recent study on bovine mastitis infection that also identified CC97 as the dominant strain type among six provinces of China from March 2010 to August 2013 (Wang et al., 2015). However, the difference is that all CC97 isolates identified in our study were MSSA strains. In particular, CC97 strains have become the dominant lineage in Chile (Smith et al., 2005), Brazil (Aires- de-Sousa et al., 2007;Rabello et al., 2007), Japan (Hata et al., 2010), the Netherlands (Ikawaty et al., 2009), and the United States (Smith et al., 2005). We found that CC97 isolates were more resistant to gentamicin and penicillin than other MSSA isolates. This implies that appropriate drug selection based on different MSSA types may reduce the reservoir of drug-resistant bacteria.
The second most frequent MLST type shown here was CC520 (16%), which was only reported in a few studies and could indicate the emergence of a new epidemic clone. Compared to other lineages, CC520 carried more virulence elements but presented a lower drug resistance rate, which may facilitate its success as a pathogen spread in bovines. Moreover, in our study, 21 MSSA isolates belonged to CC188, which is a double locus variant (DLV) of CC1 that includes MW2/USA400, the highly virulent and first-known Panton-Valentine leukocidin (PVL)-positive MRSA strain (Centers for Disease Control and Prevention, 1999). In a previous study of human infection, CC188 was the predominant clone in community-onset MSSA infections (Chen et al., 2012). However, the contamination of bovine mastitis samples with S. aureus by humans cannot be excluded. We propose that these results represent a probable transient acquisition from human handling, which will require more attention in further studies.
LA MRSA was originally reported to have emerged in 2003 among healthy pigs and pig farmers in France and the Netherlands (Witte et al., 2007). In European countries and in North America, CC398 was the overwhelmingly dominant LA MRSA, whereas CC9 predominated in most Asian countries (Cortimiglia et al., 2016;Sharma et al., 2016). In our study, all CC398 strains were identified as MSSA strains. Additionally, the observed spa type distribution of MSSA CC398 mirrored the most frequently detected spa types for MRSA CC398: t034. In contrast to the large number of reports on MRSA CC398 (Fessler et al., 2010;Schijffelen et al., 2010;Smith and Pearson, 2010;Crombé et al., 2012;Price et al., 2012), limited attention has been given to the ecology of MSSA CC398 in livestock farms (Hasman et al., 2010). MSSA CC398 is a clone of clinical importance that affects humans and livestock in different geographic regions. Price et al. analyzed 89 isolates and concluded that LA-MRSA likely originates from humans as MSSA. They theorized that the recent emergence of LA-MRSA can be seen as a reintroduction of MSSA that acquired methicillin-resistance to the original host (Price et al., 2012). Therefore, CC398-MSSA strains in this study may be a risk factor for the transmission of MRSA to people in close contact with the infected animals. Additionally, the present study showed that CC398 was more resistant to clindamycin, erythromycin, and ciprofloxacin. Our results may contribute greatly to the understanding of the emergence and transmission of MRSA CC398 among livestock because they may indicate the existence of a reservoir from which MRSA CC398 strains can emerge.
S. aureus is the most common causative organism of bovine mastitis. However, the treatment and control of this infection often fails due to the complex nature of the S. aureus strains. The development of drug resistance is a significant feature of these organisms. The results of this research revealed that all MRSA isolates were resistant to oxacillin, penicillin and cefoxitin. All MSSA isolates were susceptible to cefazolin, cefuroxime, and oxacillin. In general, MRSA strains showed much higher resistance rates to all the tested antibiotics than MSSA strains ( Table 3). All the MRSA strains were found to be resistant to at least four antibiotics. It was apparent that CC9 strains had significantly higher multiple antibiotic resistance profiles compared with other lineages. The widespread resistance of cattle-derived S. aureus will be a serious challenge to bovine mastitis therapy. The ability of bacterial pathogens to produce biofilms is regarded as a major cause of resistance to antibiotics and as responsible for persistent infections and innate immune defense for both animal and human infectious diseases (Cucarella et al., 2004;Bjarnsholt, 2013). The biofilm formed in multilayer growth of bacteria, avoiding bacterial clearance by the host immune system. In our study, we detected biofilm formation in different lineages. The results showed that the biofilm formation ability of CC7 and CC188 strains was significantly higher than other lineages. However, the resistance rates to different antibiotics in CC7 and CC188 were not higher than other lineages. This may indicate that the mechanism of drug resistance in bovine mastitisassociated S. aureus may depend on other factors than biofilm formation. While biofilm formation increases drug resistance and infection time, it can reduce bacterial virulence. Hemolysis (lysis of erythrocytes) is a significant virulence determinant of S. aureus, representing a crucial way for the bacteria to acquire iron. As shown in our study, lysis of red blood cell capacity was relatively strong in all S. aureus strains of animal origin, except CC7 and CC188. Therefore, different types of S. aureus have different potential risks to human beings.
In conclusion, the prevailing S. aureus strains in bovine mastitis were mainly MSSA strains in eastern regions of China. However, the presence of the multidrug-resistant and toxigenic MRSA clone ST9-t899 SCCmec XII suggests an alarming situation and might be a serious challenge to therapy, creating a public health concern. There were differences in the resistance genes, virulence genes and biological activities of the different molecular types of strains. As S. aureus from livestock may emerge as a threat to public health, comprehensive surveillance should be taken against S. aureus infection in the management of animal husbandry products.

ETHICS STATEMENT
Blood of healthy individuals for the lysis of erythrocytes by bacterial culture was collected using a standard method in accordance with a protocol approved by the ethics committee of Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai, China. All individuals provided written informed consent prior to donating blood.

AUTHOR CONTRIBUTIONS
ML, JS, and XW planned and supervised the experiments and wrote the paper. TL, HL, XW, QG, and YD performed the experiments and/or analyzed the data.