Molecular Characteristics of Methicillin-Resistant Staphylococci Clinical Isolates from a Tertiary Hospital in Northern Thailand

Methicillin-resistant staphylococci are now recognized as a major cause of infectious diseases, particularly in hospitals. Molecular epidemiology is important for prevention and control of infection, but little information is available regarding staphylococcal infections in Northern Thailand. In the present study, we examined antimicrobial susceptibility patterns, detection of antimicrobial resistance genes, and SCCmec types of methicillin-resistant S. aureus (MRSA) and methicillin-resistant coagulase-negative staphylococci (MR-CoNS) isolated from patients in a hospital in Northern Thailand. The species of MRSA and MR-CoNS were identified using combination methods, including PCR, MALDI-TOF-MS, and tuf gene sequencing. The susceptibility pattern of all isolates was determined by the disk diffusion method. Antimicrobial resistance genes, SCCmec types, and ST239 were characterized using single and multiplex PCR. ST239 was predominant in MRSA isolates (10/23). All MR-CoNS (N=31) were identified as S. haemolyticus (N=18), S. epidermidis (N=3), S. cohnii (N=3), S. capitis (N=6), and S. hominis (N=1). More than 70% of MRSA and MR-CoNS were resistant to cefoxitin, penicillin, oxacillin, erythromycin, clindamycin, gentamicin, and ciprofloxacin. In MRSA isolates, the prevalence of ermA (78.3%) and ermB (73.9%) genes was high compared to that of the ermC gene (4.3%). In contrast, ermC (87.1%) and qacA/B genes (70.9%) were predominant in MR-CoNS isolates. SCCmec type III was the dominant type of MRSA (13/23), whereas SCCmec type II was more present in S. haemolyticus (10/18). Ten MRSA isolates with SCCmec type III were ST239, which is the common type of MRSA in Asia. This finding provides useful information for a preventive health strategy directed against methicillin-resistant staphylococcal infections.


Introduction
Staphylococcus is recognized as an important cause of nosocomial infection. e most prominent pathogen of the genus is the coagulase-positive Staphylococcus aureus, which causes osteomyelitis, endocarditis, septic arthritis, pneumonia, and skin infections [1]. However, coagulase-negative staphylococci (CoNS) such as S. epidermidis, S. haemolyticus, S. lugdunensis, S. cohnii, S. capitis, and S. hominis are also associated with various infections with possible fatal outcomes in newborns or immunocompromised patients [2]. It is well established that staphylococcal infections in hospitals show an increasing prevalence of methicillinresistant S. aureus (MRSA) and methicillin-resistant coagulase-negative staphylococci (MR-CoNS) isolates [3,4]. Methicillin resistance in staphylococci results from the recombinase-mediated insertion of the staphylococcal chromosomal cassette mec (SCCmec), the mobile genetic element that carries mecA and various antibiotic resistance genes.
e mecA gene encodes penicillin-binding protein PBP2a that has a low affinity for β-lactam antibiotics [5]. To date, eleven SCCmec types (I-XI) have been identified. SCCmec types I, II, and III have been associated more frequently with hospital-acquired MRSA (HA-MRSA), while SCCmec types IV and V are the most dominant in MRSA infections acquired in the community (CA-MRSA) [6]. Previous studies reported the prevalence rate of these major clones varies markedly in different geographic regions; the predominant HA-MRSA clone in Asian countries is MRSA-ST239-III [7]. S. epidermidis has been found to harbor SCCmec types I, II, III, IV, and V. Likewise, SCCmec types II, III, and V have been discovered in S. haemolyticus [8]. It is generally accepted that the tolerance of chlorhexidine in S. aureus is associated with the family of the qac (qacA/B) gene, which encodes proton-motive forcedependent export pumps [9]. Recently, a study suggested that qacA/B carriage might contribute to an increasing global dominance of CC22 and ST239 clones [10]. Erythromycin resistance in staphylococci is predominantly caused by erythromycin resistance RNA methylase, whose action also affects resistance to other macrolides, lincosamides, and streptogramin B (MLS B ). is resistance is mediated by the erm-type genes, caused almost exclusively by ermA or ermC [11]. Little information is available on the molecular epidemiology of MRSA and MR-CoNS in Northern ailand. is study was designed to characterize the antimicrobial resistance genes and SCCmec types of MRSA and MR-CoNS isolated from a hospital in Chiangrai Province located in Northern ailand. ese data will provide insights into the epidemiology of the MRSA and MR-CoNS in this region.

Bacterial Isolates.
A total of 54 clinical isolates of staphylococci were collected from November 2015 to October 2016 from patients who were admitted to Chiangrai Prachanukroh Hospital, Chiangrai. e hospital is a (756bed) teaching hospital that handles ∼3,500 admissions per day, located in the north of ailand. e isolates were collected from blood (39 isolates, 72.2%), pus (10 isolates, 18.5%), sputum (4 isolates, 7.4%), and other body fluids (1 isolate, 1.9%). e bacteria were initially identified by colony morphology, mannitol fermentation, Gram characteristics, catalase test, coagulase test, and DNase activity. e phenotypic methicillin resistance was assessed using the cefoxitin disk diffusion method in accordance with the Clinical and Laboratory Standard Institute guidelines (CLSI M100-S24) at our clinical laboratory, which has been accredited by the College of American Pathologists [12]. S. aureus NCTC10442, S. aureus JCSC10442, and S. aureus WIS were used as reference strains for SCCmec typing. S. aureus COL was used as a positive control for mecA gene detection.

Species Identification of Methicillin-Resistant Staphylococci.
All isolates were confirmed as staphylococci by a PCR method based on the 16S rRNA gene [13]. e mecA gene was detected in all isolates to confirm the methicillin resistance [14]. MRSA was identified using PCR for detecting the nuc gene as previously described by Sasaki et al. [15]. e species level of MR-CoNS was identified by MALDI-TOF-MS [16] and tuf gene sequencing [17]. e direct colony of MALDI-TOF-MS analysis was analyzed as previously described [15]. e score identification criteria were used as follows: a score of 2.000 to 3.000 indicated species-level identification, a score of 1.700 to 1.999 indicated genus-level identification, and a score <1.700 indicated an unreliable identification [18].

Determination of SCCmec Types.
Multiplex PCR was carried out as described by Zhang et al. [19]. Amplification was performed in a total volume of 25 µl containing 3 µl of 10x buffer with 15 mM of Mg 2+ , 2.5 µl of 2.5 mM dNTP, 0.2 µl of 5 U Taq polymerase, various concentrations of each primer, and 3 µl of the DNA template. e condition for thermal cycler was set as follows: denaturation at 94°C for 4 min followed by 30 cycles at 94°C for 20 sec, 55°C for 30 sec, and 72°C for 30 min and a final extension at 72°C for 5 min. All PCR products were visualized using gel electrophoresis with 1% agarose gel stained with ethidium bromide.

ST239 Identification.
e ST239 was determined by the PCR method using two oligonucleotide primer sets as previously described by Feil et al. [20]. Amplification reaction was performed with the following condition: 1 cycle of predenaturation at 95°C for 15 min followed by 30 cycles at 95°C for 30 sec, 55°C for 30 sec, and 72°C for 30 sec and a final extension at 72°C for 7 min.
e primer sets are shown in Supplementary Material 1. All PCR products were visualized using gel electrophoresis with 1% agarose gel stained with ethidium bromide. e absence of bias was ensured by the sequencing of each gene in the representative isolates.

Species Distribution of Staphylococci.
e species of all isolates were identified by combined methods, including biochemical tests, PCR, MALDI-TOF-MS, and DNA sequencing. All 23 MRSA isolates were confirmed by detection of the nuc gene, and all species of MR-CoNS isolates were confirmed by tuf gene sequencing. e species distribution of MR-CoNS is given in Figure 1. e species included methicillin-resistant S. haemolyticus (n � 18), methicillinresistant S. epidermidis (n � 3), methicillin-resistant S. cohnii (n � 3), methicillin-resistant S. capitis (n � 6), and methicillin-resistant S. hominis (n � 1).

Distribution of SCCmec Types and ST239 Type Detection.
All 54 staphylococci were mecA-positive isolates. SCCmec types of all isolates were assigned by multiplex PCR according to the procedures and primer sets listed. As shown in Table 1, all MRSA isolates could be classified into six types of SCCmec elements: types I (n � 6), II (n � 1), III (n � 13), IVa (n � 1), IVb (n � 1), and V (n � 1). e distribution of SCCmec types in all MR-CoNS used in this study was ranked as types I (n � 3), II (n � 10), III (n � 5), IVa (n � 3), IVc (n � 2), and V (n � 2). SCCmec type II was the predominant clone (55.6%) in S. haemolyticus. e distribution of SCCmec types in each species is given in Table 1. Interestingly, using the multiplex PCR method, we could detect ST239 in 10 isolates of MRSA, and all of them were of SCCmec type III.

Discussion
Methicillin-resistant staphylococci have dispersed worldwide and continue to be among the most common hospital pathogens. e prevalence and characterization of MRSA and MR-CoNS in hospitals have been reported from different parts of the world [24,25]. However, the increase of antibiotic resistance in nosocomial isolates of MRSA and MR-CoNS aggravates this problem and poses a great challenge for the management of hospital-acquired infections. In the present study, we found that the 54 staphylococcal isolates belonged to 6 different species. e species distribution identification by MALDI-TOF-MS was consistent with the species identified by tuf gene sequencing, with the exception of one isolate (SP33) (Figure 1). Using MALDI-TOF-MS, this isolate was identified as S. epidermidis, but tuf gene sequencing identified it as S. haemolyticus. We assumed that the species assigned by tuf gene sequencing was more accurate because the score of MALDI-TOF-MS was only at the level of genus identification. Moreover, MALDI-TOF-MS could not identify 3 isolates of MR-CoNS. ese 3 isolates were identified as S. cohnii by tuf gene sequencing. is result was consistent with a previous study reporting that MALDI-TOF-MS could not identify S. cohnii to the species level [26]. Additionally, a phylogenetic tree based on tuf gene sequencing was compared with the MALDI-TOF dendrogram for all 31 isolates of MR-CoNS ( Figure 1). Interestingly, if the disagreement for one isolate (SP 33) was not considered, the structure of each species was broadly in alignment. Only S. hominis was located in different structures of both phylogenetic trees. To the best of our knowledge, this is the first comparison between phylogenetic tree based on tuf gene sequencing and MALDI-TOF dendrogram of MR-CoNS. We found that MRSA and MR-CoNS isolates were resistant to multiple antibacterial agents ( Figure 2). Among MR staphylococci isolates, 82.6% were resistant to 7-10 antibiotics (96.8% of MR-CoNS and 60.9% of MRSA). is result is similar to the findings in China and France with a high rate of antibiotic resistance within MRSA clinical isolates [27,28]. In this study, all MRSA and MR-CoNS isolates were sensitive to vancomycin and linezolid. us, these drugs remain suitable options for the treatment of serious infections caused by MRSA and MR-CoNS. e mecA gene, encoding a PBP variant which confers resistance to methicillin, was detected in 100% of staphylococci isolated in this study. mecA is carried by the mobile genetic element SCCmec. e distribution of different SCCmec types in methicillin-resistant staphylococci varied depending on the host species, bacterial clones, and possibly geographical locations [29]. SCCmec typing has become essential for the epidemiological characterization of MRSA and MR-CoNS clones. In this study, 54 methicillin-resistant staphylococci were investigated for their SCCmec types; SCCmec type III was found to be predominant, with a proportion of 56.5% (13/23) of MRSA isolates. Our results are in agreement with Chongtrakool et al., who reported SCCmec type III as the predominant type in many Asian countries such as Saudi Arabia, India, Sri Lanka, Singapore, Indonesia, ailand, Vietnam, Philippines, and China, whereas SCCmec type I of MRSA isolates which shows high prevalence in Iran (56.9%) was found to be only 26.1% in our study [29,30].
We found that ST239 was detected in 43.5% (10/23) MRSA isolates, and all positive clones carried SCCmec type III (ST239-SCCmec III). Previous studies have demonstrated that ST239-SCCmec III is the endemic HA-MRSA in many Asian countries, although a recent study showed that this clone is being steadily displaced by emerging CA-MRSA clones [24]. ST239-SCCmec III was also reported to be the    (77.1%), ST239-SCCmec II was accounted for as the most dominant nosocomial MRSA clone in 18 hospitals in China [31]. It has been reported that ST239-SCCmec III was detected in at least 90% of HA-MRSA isolates in Sappasithiprasong Hospital, Northeast ailand [20]. ese dominant types were resistant to many antibiotics such as erythromycin, gentamicin, sulfamethoxazole/trimethoprim, and ciprofloxacin (  [38]. We found that 26.1% of MRSA isolates carried the qacA/B gene. Its prevalence in the present study is higher than that in a previous report by Lu et al., who found 25 (7.8%) of the 321 MRSA isolates harboring qacA/B [39]. On the contrary, 70.9% of all MR-CoNS isolates in the present study carried the qacA/B gene. is prevalence was higher than the rate of the qacA/B gene carried by CoNS isolated from surgical sites (37.9%) [40], nurses (56.7%), and the general population in Hong Kong (13.5%) [23]. e increased proportion of the qacA/B gene in MR-CoNS indicates that hospital-acquired infections could exert selective pressure for carriage of these strains.
In summary, most of the MRSA isolates in the present study were typed as ST239-SCCmec type III, while different MR-CoNS species carry various SCCmec types. is finding provides useful information for a preventive health strategy to combat methicillin-resistant staphylococcal infections.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Acknowledgments
We would like to thank microbiology staff from Chiangrai Prachanukroh Hospital for specimen collection. We also acknowledge Dr. Keiichi Hiramatsu and Dr. Teruyo Ito for  providing SCCmec-type strains. We thank Prof. Dr. Gavin Reynolds for editing the manuscript. is work was supported by a grant from the National Research Council of ailand (R2560B064) to SS.