Background

Staphylococcus aureus (S. aureus) is an important Gram-positive pathogen that causes various infectious diseases including pneumoniae and bacteremia. A previous study showed that patients with S. aureus infections had an excess one-year mortality of 20.2% compared with matched uninfected inpatients [1]. The genotype of S. aureus has been reported to influence the complications, severity, and mortality of infection. One study showed that the strains clonal complex 5 (CC5) and CC30 exhibited a significant trend toward increasing levels of hematogenous complications [2]. Another study found that patients with S. aureus sequence type 121 (ST121) infections often required longer hospitalization and prolonged antimicrobial therapy [3], whereas bloodstream infections by CC398, a methicillin-susceptible Staphylococcus aureus (MSSA), were associated with high mortality [4]. Therefore, analysis of the molecular characteristics and virulence gene profiles of S. aureus is important for prognosis of infection.

The molecular characteristics of S. aureus vary with region. In many Asian countries including China and Thailand, ST239 has been found to be the most prevalent type [5,6,7,8], whereas in the United States, ST8 (USA300) and ST121 are the most frequently observed [3, 9]. Even within China, the molecular characteristics of S. aureus isolates differ among cities; the predominant types in Wenzhou are ST188 and ST7 [10], the major type in Dalian and Shenyang is ST5 [11], and in Chengdu, ST59 is prevalent [12]. The molecular characteristics of S. aureus are also reported vary over time. Since 2000, ST239-t030-SCCmecIII has rapidly replaced ST239-t037-SCCmecIII as the major clone of S. aureus isolates in Chinese tertiary hospital care [2], whereas ST239-t030-MRSA, which in 2013 was the predominant genotype among all methicillin-resistant S. aureus (MRSA) strains in China, had been replaced by ST59-t437-MRSA by 2016 [13]. In addition, it was reported that the predominant clones, ST239-t030 and ST239-t037, were being replaced by the continually growing ST5-t2460 clone in 2017 in Shanghai [14]. Therefore, when monitoring the molecular characteristics of S. aureus isolates, it is preferable to focus on a specific region of interest at a particular time.

Hainan, the southernmost province of China, is surrounded by the South China Sea, and has a uniquely tropical monsoon and marine climate that differs significantly from that on the mainland. The island has been called a “natural large greenhouse,” and the hot and humid climate is conducive to bacterial growth. Studies of the molecular characteristics and antibiotic resistance profiles of S. aureus isolates from China have been carried out over the last 10 years in provinces such as Zhejiang, Guangdong, and Guangxi [15,16,17,18]. To date, however, no study has focused on the molecular characteristics and virulence gene profiles of S. aureus isolates in Hainan, and no hospital in Hainan has been included in any multicenter studies concerned with those characteristics of S. aureus in China [13, 19, 20]. Not even the CHINET surveillance system includes any hospital from Hainan. Although the total area of Hainan is relatively small, its population has now reached 10 million, and moreover, its tropical monsoon and marine climate is unique in China. These are important motivations to investigate the molecular characteristics, virulence genes, and antibiotic resistance profiles of S. aureus isolates from Hainan.

Methods

S. aureus isolates and primers

A total of 227 consecutive and non-duplicate S. aureus isolates were collected from three hospitals in 2013–2014 (n = 91) and 2018–2019 (n = 136). Of the three hospitals in Haikou city, Hainan General Hospital is a large teaching hospital with more than 100,000 admissions per year in Xiuying district; Haikou People’s Hospital is a medium-sized teaching hospital with about 50,000 admissions per year in Meinan district; and First Hospital Affiliated to Hainan Medical College is a medium-sized teaching hospital with 50,000 admissions per year in Longhua district. These isolates were collected from inpatients who had cough, fever and other clinical symptoms related to infection and whose peripheral white blood cell and/or neutrophil counts were elevated. These isolates were derived from diverse clinical specimens, including cutaneous abscess and wound secretion (n = 110, 48.5%), sputum and pharynx swabs (n = 48, 21.1%), blood (n = 42, 18.5%), and others (catheter tip, marrow, pleural fluid, cerebrospinal fluid, cystic cavity fluid, drainage liquid, ascites, joint fluid, biopsy, and urine) (n = 27, 11.9%). Only the first positive culture in the course of infection was included for further analysis. These isolates were identified by conventional microbiological methods including Gram staining, catalase, and coagulase tests, and confirmed with a VITEK 2 Compact system and a VITEK 2 AST-GP67 Test Kit (bioMerieux, Inc., Durham, NC, USA). All isolates were stored at − 80 °C for further experiments. All primers used in this study were synthesized by Tianyihuiyuan (China) (Table 1). This study was approved by the Ethics Committee of Hainan General Hospital. This was a retrospective study that did not involve collection of clinical and personal information from patients, so informed consent was not required.

Table 1 Primers used in this study, and the results of SCCmec types I–V

Antimicrobial susceptibility testing

Antimicrobial susceptibility tests were carried out using a VITEK 2 Compact system and a VITEK 2 AST-GP67 Test Kit (bioMerieux, Inc., Durham, NC, USA). Twelve antibiotics were tested, including cefoxitin (FOX), clindamycin (CLI), erythromycin (ERY), gentamicin (GEN), levofloxacin (LEV), linezolid (LZD), oxacillin (OXA), penicillin (PEN), rifampicin (RIF), trimethoprim/sulfamethoxazole (SXT), tetracycline (TET), and vancomycin (VAN). S. aureus ATCC 25923 and ATCC25913 were used as the quality control strains, and the results were interpreted in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI M100-S29) [21]. In addition, S. aureus isolates were further identified using PCR for amplification of mecA as described previously [22], in which MRSA N315 was used as the positive control strain. The mecA-positive and cefoxitin-resistant isolates (cefoxitin minimum inhibitory concentration ≥ 8 μg/mL) were identified as MRSA. Isolates resistant to three or more different antimicrobial classes were defined as multidrug-resistant (MDR).

Staphylococcal protein a (spa) typing

Chromosomal DNAs were extracted from S. aureus isolates as described previously [23]. The extracted chromosomal DNAs were stored at − 20 °C for spa, Staphylococcus chromosomal cassette mec (SCCmec), multilocus sequence typing, and detection of virulence genes. For spa typing, the variable repeat region of spa was amplified using oligonucleotide primers [23, 24] (see Table 1) followed by sequencing. The PCR mixture and conditions were similar to those described previously [23]. The resulting amplicons were purified and subjected to Sanger dideoxy DNA sequencing (Tianyihuiyuan, China) followed by analysis using the Ridom web server (http://spaserver.ridom.de). S. aureus isolates that could not be classified as any known spa type were defined as nontypable (NT).

Multilocus sequence typing (MLST)

MLST was carried out according to the protocol described previously [23, 25]. Seven housekeeping genes of S. aureusarcC, aroE, glpF, gmk, pta, tpi, and yqil—were adopted for MLST. Seven respective PCR assays were conducted to amplify these seven housekeeping genes. These amplicons were sequenced using Sanger dideoxy DNA sequencing (Tianyihuiyuan, China). The resulting sequences were compared with the known alleles in the MLST database (http://saureus.mlst.net), which was used to determine ST. S. aureus isolates that could not be assigned to any known ST were submitted to the MLST database and assigned to new STs. The clustering of related STs, which were defined as clonal complexes (CCs), was determined using eBURST .

Staphylococcus chromosomal cassette mec (SCCmec) typing

The MRSA isolates were subjected to SCCmec typing as previously described [26]. MRSA isolates with suspected SCCmecIV were recharacterized by additional multiplex PCR as subtypes IVa, IVb, IVc, and IVd as described by Zhang et al. [22]. MRSA isolates that could not be assigned to any above type were defined as NT. All primers are listed in Table 1.

Detection of virulence genes and antibiotic resistance genes

Eleven virulence genes, including the Panton-Valentine leukocidin (pvl), the staphylococcal enterotoxin genes (sea, seb, sec), the exfoliative toxin genes (eta, etb), the hemolysin genes (hla, hlb), and the adhesion factor genes (fnbA, fnbB, clfA) were detected using PCR assays. The PCR mixture and conditions were similar to those described previously [23]. The common ERY resistance genes (ermA, ermB, ermC), and the TET resistance genes (tetL, tetK, tetM, tetO) were examined using PCR assays as previously described [27, 28].

Statistical analysis

Statistical analyses were performed using SPSS Statistics 24.0 for Windows. Data were analyzed using the chi-square or Fisher’s exact tests. All statistical tests were two-tailed, and p < 0.05 or p < 0.01 (Fisher’s exact tests among three groups) was considered to indicate statistical significance.

Results

Antimicrobial susceptibility and antibiotic resistance genes

A total of 227 S.aureus isolates were tested for antimicrobial susceptibility. The antimicrobial resistance profiles of the S. aureus, MRSA, MSSA and MDR isolates are shown in Fig. 1. No S. aureus isolate was resistant to VAN or LZD, but most were resistant to GEN (14.1%), LEV (10.6%), and RIF (19.8%). Less than 50% of isolates were resistant to the remaining antibiotics, except for PEN, to which 92.5% had resistance. All of 76 FOX-resistant isolates, including an OXA susceptible-MRSA (OS-MRSA), were found to be mecA-positive, and were thus classified as MRSA isolates. Statistical analysis showed that the MRSA isolates had significantly higher resistance rates to PEN than the MSSA isolates (100.0% vs. 88.7%, p = 0.002), ERY (75.0% vs. 35.8%, p < 0.001), CLI (64.5% vs. 29.8%, p < 0.001), GEN (18.4% vs. 8.6%, p = 0.031), RIF (17.1% vs. 4.6%, p = 0.002), and LEV (15.8% vs. 6.6%, p = 0.028).

Fig. 1
figure 1

Antimicrobial resistance profiles of MRSA, MSSA, MDRs and S. aureus isolates

A total of 111 ERY-resistant S. aureus isolates were found and used to examine the presence of erm gene. The most prevalent erm gene was ermC (90.1%, 100/111), followed by ermB (38.7%, 43/111) and ermA (21.6%, 24/111). In the ERY-resistant MSSA isolates, the frequencies of the three ERY-resistant genes showed the same trend as in the ERY-resistant MRSAs. ermC (87.0%, 47/54) was the most prevalent gene, followed by ermB (38.9%, 21/54) and ermA (11.1%, 6/54). The ERY-resistant MRSA isolates had higher frequencies of ermA than the ERY-resistant MSSA isolates (χ2 = 6.855, p < 0.05). All of 85 TET-resistant isolates carried the TET-resistant genes tet. The prevalences of tetK, tetM, tetL and tetO were 91.8% (78/85), 67.1% (57/85), 23.5% (20/85) and 0.0% (0/85), respectively. In the TET-resistant MSSA isolates, the most prevalent TET-resistant gene was still tetK (93.2%, 55/59), followed by tetM (59.3%, 35/59), tetL (23.7%, 14/59), and tetO (0.0%, 0/59). TET-resistant MRSA isolates had higher frequencies tetM than the TET-resistant MSSA isolates(χ2 = 5.227, p < 0.05).

One hundred thirteen (49.8%) S. aureus isolates were found to be MDR, defined as having resistance to more than three classes of antibiotics (Table 2). The chi-square test showed that the prevalence of MDR was significantly higher in the MRSA isolates than in the MSSA isolates (Table 2)(χ2 = 26.115, p < 0.05). In addition, when comparing the S. aureus isolates collected in 2013–2014 with those from 2018 to 2019, the resistance rates to all antibiotics except SXT were broadly similar. Compared with those collected in 2018–2019, the S. aureus isolates from 2013 to 2014 had a higher resistance rate to SXT (64.8% vs. 5.9%, p < 0.05) and a greater prevalence of MDR (61.5% vs. 41.9%, p < 0.05).

Table 2 The frequency of MDRs, main STs, and virulence genes among MRSA and MSSA

MLST, spa, and SCCmec typing

Forty STs belonging to 19 CCs and 2 singletons were identified by eBURST. As shown in Table 3 and Fig. 2, ST398 (14.1%, 32/227) was the most prevalent, followed by ST188 (13.2%, 30/227) and ST45 (10.1%, 23/227). It was found that 78.1% (25/32) of the ST398 isolates, 80.0% (24/30) of the ST188 isolates, and 91.3% (21/23) of the ST45 isolates were derived from Hainan General Hospital. In addition, three isolates could not be assigned to any known ST, so these novel alleles were submitted to the MLST database, and three new STs, ST5489, ST5492 and ST5493, were assigned. By spa typing, 79 spa types were found. The most prevalent was t189 (12.3%, 28/227), followed by t437 (7.9%, 18/227), t116 (7.5%, 17/227), and t011 (6.6%, 15/227). When the STs and spa typing were combined, the predominant combinations were ST188-t189 (12.3%, 28/227), ST45-t116 (7.5%, 17/227), ST59-t437 (7.0%, 16/227), ST398-t011 (6.6%, 15/227), ST398-t034 (4.8%, 11/227), and ST7-t091 (4.8%, 11/227). A strong association was observed between certain STs and spa types: ST188 was primarily associated with t189 (93.3%, 28/30); ST45 was associated mainly with t116 (73.9%, 17/23); and ST59 was associated mainly with t437 (72.7%, 16/22).

Table 3 Molecular characteristics of S. aureus isolates collected in this study
Fig. 2
figure 2

Distribution of STs in the clonal complexes. The diagram generated by eBURST based on the MLST data of this study, representing the relationships of 227 S. aureus isolates identified by MLST typing. Each number implies an MLST ST, STs that are linked by a line belong to the same cluster and the dot area indicates the prevalence of the ST in the MLST data of this study

The major types of S. aureus collected in 2013–2014 were ST188 (14.3%), ST45 (14.3%), ST59 (8.8%), and ST88 (8.8%), whereas in 2018–2019, ST398 (19.9%), ST188 (12.5%), ST59 (10.3%), ST45 (7.4%), and ST7 (7.4%) were the top five types. Among the STs that exhibited OXA sensitivity, the two predominant types in 2013–2014 were ST188-MSSA (14.3%) and ST45-MRSA (12.1%), whereas in 2018–2019 they were ST398-MSSA (18.4%) and ST59-MRSA (8.1%). The prevalence of ST398-MSSA markedly increased from 2013 to 2014 (5.5%) to 2018–2019 (18.4%), and this increase was significant (p < 0.05).

Among the 76 MRSA isolates, 6 SCCmec types or subtypes, namely types I, II, III, IVa, IVc, and V, were found. The most common SCCmec type was IVa, which was found in 43 isolates (56.6%, 43/76), whereas types I, II, III, IVc, and V were found in 1, 3, 6, 5, and 9 isolates, respectively. Nine isolates, including OS-MRSA, were classified as NT for SCCmec typing. When the STs and SCCmec typing were combined, the predominant combination was ST45-SCCmec IVa (8.8%, 20/227), and no significant difference was found in the positive rate of ST45-SCCmec IVa between the S. aureus isolates collected in 2013–2014 and 2018–2019 (12.1% vs. 6.6%, p > 0.05) (Table 3).

Virulence gene profiles

The frequencies of the virulence genes identified in the 227 S. aureus isolates are listed in Table 4. ClfA was present in all S. aureus isolates, hla, hlb, and eta were detected in 98.7, 70.9, and 57.3% of these isolates, respectively, whereas the remaining ones were found in less than 50%. One hundred and twenty (52.9%) S. aureus isolates harbored six or more virulence genes. Of those 120 isolates, 11 contained 9 virulence genes, 31 had 8 such genes, 38 carried 7, and 40 carried 6. The frequencies of fnbA, sea, and sec were significantly higher in the MRSA isolates than in the MSSA isolates, but no significant difference was found in the likelihood of harboring six or more virulence genes between the MRSA and MSSA isolates (56.6% vs. 51.0%, p > 0.05). Compared with those collected in 2013–2014, the S. aureus isolates from 2018 to 2019 had higher frequency of pvl, fnbB, hlb, seb, eta, and etb and higher rates of harboring six or more virulence genes.

Table 4 The frequency of virulence genes among main types of S. aureus isolates and the comparison of two time periods

Characteristics of the major clones ST398, ST188, and ST45

The most abundant sequence type found in this study was ST398 (14.1%, 32/227) followed by ST188 (13.2%, 30/227) and ST45 (10.1%, 23/227). Most ST398 (93.8%, 30/32) and ST188 (96.7%, 29/30) isolates were MSSA, whereas most ST45 (87.0%, 20/23) isolates were MRSA and all ST45-MRSA isolates belonged to the SCCmec IVa type (Tables 2 and 3). The ST398 (χ2 = 17.685, p < 0.01) and ST188 isolates (p < 0.01) had higher resistance rates to TET than the ST45 isolates. In addition, no significant difference was seen in the resistance rate to any antibiotics between ST398, ST188 and ST45 isolates.

Of the 11 tested virulence genes, pvl and fnbB were more frequent in ST398 isolates than in ST45 (χ2 = 22.010 and χ2 = 30.457, respectively, p < 0.01) and ST188 isolates (χ2 = 12.790 and χ2 = 38.027, respectively, p < 0.01). The prevalence of sec in ST45 isolates was higher than that in ST398 (χ2 = 43.487, p < 0.01) and ST188 isolates (χ2 = 32.500, p < 0.01), whereas the prevalence of eta in ST45 isolates was higher than in ST188 isolates (χ2 = 14.339, p < 0.01). On the contrary, the positive rate of hlb in ST45 isolates was lower than in ST398 (χ2 = 7.118, p < 0.01) and ST188 isolates (χ2 = 7.248, p < 0.01). No significant difference was found in the positive rate of any other virulence genes between any two of the three STs (Table 4).

Discussion

A total of 227 S. aureus isolates were collected in 2013–2014 and 2018–2019 from three hospitals in Hainan province for investigation of their antimicrobial resistance, virulence gene profiles, and molecular characteristics. The results showed that all isolates were susceptible to VAN and LZD, in agreement with most previous studies in mainland China [29,30,31]. In addition, when comparing the S. aureus isolates collected in 2013–2014 and 2018–2019, no significant difference was found in the resistance rates to the remaining antibiotics except that to SXT. Therefore, both sets of isolates were combined for analysis, and the average resistance rates to PEN, ERY, CLI, TET, FOX, OXA, GEN, LEV, and RIF were found to be 92.5, 48.9, 41.4, 37.4, 33.5, 33.0, 11.9, 9.7, and 8.8%, respectively. For comparison, in mainland China in the first half of 2018, the corresponding average rates were reported to be 92.7, 64.5, 38.4%, unreported, 34.4, 34.4, 18.7, 22.4, and 5.2% (www.chinets.com). While in Turkey in 2017, the average rates to OXA, RIF, VAN and LZD were 23.0, 14.0, 0.0, and 0.0%, respectively. In the United States in 2017, the average rates to OXA, ERY, RIF, VAN, and LZD were 45.0, 41.0, 1.0, 0.0, and 0.0%, respectively, whereas in Russia in 2017, they were 16.0, 2.0, 0.0, and 0.0%, and in Australia they were 19.0, 1.0, 1.0, and 0.0%, respectively (resistancemap.cddep.org/AntibioticResistance.php). The S. aureus isolates from Hainan had resistance rates against some antibiotics similar to those from mainland China and other countries, but differences were found in the resistance rates to ERY and LEV. In addition, the resistance rate to SXT in the S. aureus isolates collected in 2018–2019 (5.9%) was significantly lower than for those collected in 2013–2014 (64.8%), whereas the resistance rate to SXT has remained stable in recent years in mainland China. 10.1% in 2014 and 14.3% in the first half of 2018 (http://www.chinets.com). The steep decline in resistance to SXT may be due to the reduced use of this antibiotic in recent years in Hainan.

ERY and TET resistance depend on the presence of resistance genes erm and tet, respectively. The predominant resistance gene in the ERY-resistant isolates was ermC, which differed from previous studies in which most of the ERY-resistant strains harboured ermA [27, 32]. In this study, ermB was present in 38.7% of ERY-resistant isolates, whereas in most previous studies, ermB was rare or not detected at all [27, 32]. Therefore it is concluded that S. aureus isolates in Hainan have characteristic resistance genes for erythromycin resistance. Most of the TET-resistanct isolates harbored tetM and tetK, which indicates that those variants were responsible for resistance to TET, consistent with previous studies [28, 32, 33]. Our study shows that the frequency of tetM was higher in TET-resistant MRSAs than in TET-resistant MSSAs, consistent with the previous finding that the resistance mechanism mediated by tetM is predominant among TET-resistant MRSAs [34].

MLST typing, spa typing, and SCCmec typing were performed to analyze the molecular characteristics of the S. aureus isolates. ST398, ST188, and ST45 were the predominant STs among the S. aureus isolates in this study, among which ST398 and ST45 were the predominant clones in the MSSA and MRSA isolates, respectively. In addition, the most common SCCmec type was IVa, and ST45-SCCmec IVa was the most prevalent combination of ST and SCCmec typing in the MRSA isolates. ST188 and ST239 were previously reported as the predominant STs in MSSA and MRSA isolates, respectively [11, 19, 20, 35, 36]. Two of these studies were multicenter studies that showed that ST239-SCCmec III was the predominant MRSA genotype, but no ST45 clones were observed [11, 20]. A recent study in Shanghai showed that ST239-t030 and ST239-t037 were being driven out by the continual growth of the ST5-t2460 clone [14]. Therefore, it can be concluded that the molecular characteristics of S. aureus isolates in Hainan differ significantly from those in mainland China. It is reasonable to speculate that the divergent molecular characteristics of S. aureus isolates in Hainan are associated with the special geographical location of Hainan.

ST398 MSSA was found to be the most prevalent in Hainan province, and the patients with the ST398 MSSA isolates had no history of contact with livestock, confirming that the ST398 MSSA isolates we collected are of human origin. In addition, in the short span of 5 years, the prevalence of ST398 MSSA increased from 5.5 to 18.4% in Hainan. Similar to the epidemic situation in Hainan province, ST398 MSSA have been increasingly reported as a cause of invasive infections in patients without livestock contact [4]. In cohorts of patients in France, the number of ST398 MSSA cases was shown to increase from zero in 1999 to 4.6% in 2010, including 13.8% of cases with S. aureus bloodstream infections [4, 37]. Another retrospective study in France found that only 1.9% of bone and joint infection (BJI) MSSA strains were screened to be ST398 in 2008, whereas in 2010–2012, 14.0% of BJI MSSA strains belonged to ST398 [38]. Therefore, ST398 MSSA has emerged as an invasive pathogen causes bloodstream infections, BJIs, and potentially other conditions. Evidence suggests that ST398 MRSA and ST398 MSSA belong to distinct lineages [39]. It is well known that the ST398 MRSA lineage, associated with livestock, has become a worldwide threat within the past decade [4]. However, ST398 MSSA is a frequent source of S. aureus infections between individuals in households. This contrasts with the limited transmissibility of livestock-associated ST398 MRSA strains between humans [40]. ST398 MSSA has enhanced adhesion to human skin keratinocytes and keratin and it is more closely linked with human infections than ST398 MRSA. In addition, it was reported that the 30-day all-cause mortality was higher for patients with ST398 MSSA bloodstream infection than for a control group with non-ST398 MSSA infection [4]. Because ST398 MSSA has become the most prevalent ST in S. aureus isolates from Hainan and may be linked to higher mortality, it is necessary to monitor the changes in the molecular characteristics of S. aureus to prevent the wider dissemination of that strain.

The virulence factors of S. aureus play an important role during pathogenesis [41, 42]. Similar to the majority of studies in mainland China [8, 43], almost all strains in our study were positive for clfA and hla, confirming that these were the most common virulence factors in S. aureus, and no regional difference was seen in their distribution. Notably, the frequencies of eta and pvl were 57.3 and 47.6%, much higher than those in mainland China [10, 16, 19]. ST45, a common type of S. aureus isolate in Hainan, was found to have an eta prevalence of 95.7% in our results. Meanwhile, ST398, a clone with a low prevalence of pvl in previous studies [37, 40], had a frequency of 81.3% in this study. Together, these findings indicate that S. aureus isolates in Hainan have somewhat higher positive rates of eta and pvl. Previous studies reported some virulence genes are linked to specific molecular types [43, 44]. For example, ST8 (USA300) was linked to the acquisition of the enterotoxin Q and K genes. ST36 (USA200) was associated with the acquisition of the enterotoxin A gene and the toxic shock syndrome toxin 1 gene [44]. Therefore, it is rational to speculate that the higher rates of eta and pvl could be associated with the different distribution of STs. This implies that the molecular characteristics of S. aureus isolates affect their virulence gene profiles, leading us to conclude that S. aureus isolates collected in Hainan have distinct virulence gene profiles compared with those collected in mainland China. In addition, compared with the 2013–2014 isolates, the S. aureus isolates collected in 2018–2019 carried more virulence genes, but their rate of MDR was lower. The opposite trend between antibiotic resistance and virulence may be related to balance the energetic requirements for expressing resistance and producing toxins, which suggests that the antibiotic resistance and virulence of pathogens are in competition during evolution [45, 46]. This study has some limitations. First, the small sample size limited the broad representativeness of the study. Second, we had no information about the relationship between clinical data (e.g. mortality, severity) and molecular characteristics of the isolates, which will be the focus in further research. These fields.

Conclusions

S. aureus isolates in Hainan have unique molecular characteristics and virulence gene profiles. ST398-MSSA was the most common type of MSSA isolate and ST45-SCCmec IVa was the predominant type of MRSA isolate, neither of which had been reported in China before. Differences were also found between the antibiotic resistance and virulence gene profiles of the ST398 and ST45 isolates. ST398-MSSA showed a clear growth trend from 2013 to 2014 to 2018–2019, which deserves attention from public health services.