Antimicrobial Resistance, SCCmec, Virulence and Genotypes of MRSA in Southern China for 7 Years: Filling the Gap of Molecular Epidemiology

As the prevalence of Staphylococcus aureus infections is of worldwide concern, phenotype and genotype in prevalent MRSA strains require longitudinal investigation. In this study, the antibiotic resistance, virulence gene acquisition, and molecular type were determined on a large scale of nosocomial S. aureus strains in Southern China during 2009–2015. Bacterial identification and antimicrobial susceptibility to 10 antibiotics were tested by Vitek-2. Virulence genes encoding staphylococcal enterotoxins (SEA, SEB, SEC, SED, and SEE), exfoliative toxins (ETA and ETB), Panton–Valentine leukocidin (PVL), and toxic shock syndrome toxin (TSST) were detected by PCR, with SCCmec typing also conducted by multiplex PCR strategy. Genotypes were discriminated by MLST and spaA typing. MLST was performed by amplification of the internal region of seven housekeeping genes. PCR amplification targeting the spa gene was performed for spa typing. No resistance to vancomycin, linezolid, or quinupristin and increase in the resistance to trimethoprim/sulfamethoxazole (55.5%) were identified. A total of nine SCCmec types and subtypes, thirteen STs clustered into thirteen spa types were identified, with ST239-SCCmec III-t037 presenting the predominant methicillin-resistant S. aureus (MRSA) clone. Typically, SCCmec type IX and ST546 were emergent types in China. Isolates positive for both pvl and tsst genes and for both eta and etb genes were also identified. Important findings in this study include: firstly, we have provided comprehensive knowledge on the molecular epidemiology of MRSA in Southern China which fills the gap since 2006 or 2010 from previous studies. Secondly, we have presented the correlation between virulence factors (four major groups) and genotypes (SCCmec, ST and spa types). Thirdly, we have shown evidence for earliest emergence of type I SCCmec from 2012, type VI from 2009 and type XI from 2012 in MRSA from Southern China.


Introduction
As a major human pathogen and important "superbug", methicillin-resistant Staphylococcus aureus (MRSA) is responsible for a wide range of infections and diseases due to possession of numerous virulence factors and toxins [1][2][3][4]. Currently, indiscriminate use of existing antibiotics which lead to the spread of antibiotic resistance, poses a dilemma for the treatment of bacterial infections [5][6][7][8][9]. Despite a variety of novel therapy and drug discovery, antibiotic resistance in microbes remains a major public health concern for the treatment of infectious diseases [10][11][12][13][14]. MRSA is one of such infectious bacteria that has become resistant to β-lactam antibiotics [15][16][17][18]. As China remains one of the worst areas for antibiotic abuse [19], it is necessary to raise general concerns regarding the surveillance and investigation of antibiotic resistance mechanism involved in clinical MRSA in China. To better understand the molecular evolution and epidemiological characterization of S. aureus, the analysis of SCCmec and genetic backgrounds mainly involved in randomly amplified polymorphic DNA (RAPD) types, sequence types (STs), clone complexes (CCs), and spa types is necessary [20][21][22].
In China, the first evidence of molecular types in MRSA dates back to 2003. However, this study had been primarily conducted using a large number of samples (118) from a hospital in Taiwan with a much smaller number of samples (14) from a hospital in Nanjing in mainland China [23]. In 2007, we reported on ST239-MRSA and strains carrying class 1 integrons isolated in 2005. Ref. [5], for which we had further provided the SCCmec (type III), spa, and coa types in a subsequent study [24]. In 2008, we reported on the presence of the SCCmec of four different types of methicillin-resistant coagulase-negative staphylococci (MRCNS) [25]. In 2011, we conducted a molecular epidemiology study on MRSA isolates from 2001-2006, including SCCmec, RAPD, MLST, spa, and coa typing, as well as carriage of class 1 integrons [26]. In a subsequent study, we had provided evidence on the molecular epidemiology of MRSA during 2001-2010 [27]. As found from all above studies conducted in Guangzhou representative of Southern China, ST239-MRSA-III was highly prevalent, with only a few exceptions carrying type II SCCmec and none of other ST than ST239. Only one distinct coa type HIJKL and 2 spaA types (WGKAOMQ-t037 and WGKAQQ-t030 were identified during the study period of 2001-2010. However, since the types of SCCmec has increased from six types (I to VI) in 2006 to fourteen (I to XIV) to date, the concern about the evolution of SCCmec and molecular epidemiology in MRSA since 2006 or 2010 in Southern China remains unclear but important [28,29]. Consequently, in this study, numerous S. aureus strains isolated in Southern China during 2009-2015, were subjected to antimicrobial resistance, virulence genes identification, SCCmec, MLST, and spa typing.

Carriage of Virulence Genes
According to the results, SEs genes were commonly detected, with the identification rate of sea, seb, sec, sed, and see found to be 67.8% (352/519), 33.0% (171/519), 42.6% (221/519), 1.9% (10/519), 35.8% (186/519), respectively. For exfoliative toxins genes, 12.4% (61/493) and 9.6% (49/511) of the strains were positive for eta and etb. In addition, tsst and pvl were detected in 28.6% (146/511) and 14.9% (76/511) of strains.    I  0  0  0  1  3  1  7  12  IA  0  0  0  1  3  1  2  7  II  1 a  2  in each year were listed, with total amount of isolates in the rightmost column. # The amount of S. aureus isolates resistance to each antibiotic (oxacillin, trimethoprim/sulfamethoxazole, erythromycin, ciprofloxacin, rifampin, moxifloxacin, penicillin, tetracycline, levofloxacin, and clindamycin) in each year were listed. The data were adapted to chi-square test between years and the superscripted "abcde" in the table is the significant difference letter marking method. (Arrange all the averages from large to small, label the largest average with the letter "a", subtract that average from the following to obtain the range, and label any range less than 0.05 with the letter "a", until the range is greater than or equal to 0.05, or a significant difference from an average, the average is superscript with the letter "b", and the average labeled "b" is used as the criterion, compared with the above average which is larger than it, any difference which is not significant is marked with the letter "b", and the maximum average marked with "b" is used as the standard, compared with the unmarked average below, those that are not significantly different continue to be labeled with the letter "b" until an average that is significantly different is labeled with the letter "c". The comparisons are repeated until the smallest average has a marked letter).

Discussion
A remarkably high multi-drug resistance rate (88.7%) was obtained in the tested strains, with the majority as MRSA. Noteworthily, increase in trimethoprim/sulfamethoxazole resistance (from 20.  (Table 2). An increase in trimethoprim/sulfamethoxazole resistance within S. aureus isolates had also been reported in our previous study and other reports [30][31][32][33]. A decrease in ciprofloxacin in this study was also in accordance with the previous surveillance, whereas the resistance rate of ciprofloxacin was found to be 69.2% during 2006-2010 and 44.0% in 2011-2015. Concerning the correlation between antibiotic resistance and SCCmec types, isolate type II SCCmec showed significantly lower resistance to erythromycin (8.3%) than other types (average 81.5%). Type V SCCmec (23.5%) was the least resistant to trimethoprim/sulfamethoxazole, followed by type II (42.9%) and type III (44.4%). Low resistance to rifampin was also identified in type II SCCmec (18.3%) As a major human pathogen, MRSA has a number of virulence genes in its genome, which contributes to its pathogenicity. However, very few studies had touched upon the relevance between its molecular epidemiology and virulome (virulence genes profile). In S. aureus, extracellular protein toxins are major toxins that could significantly enhance its pathogenicity, including Staphylococcal enterotoxins (SEs), toxic shock syndrome toxin 1 (TSST-1), exfoliative toxins (ETs) and Panton-Valentine leukocidin (PVL) [34,35]. In addition, there are a few subtypes of such toxins, especially SEs. In this study, four types of such toxins with a total of nine targets were investigated, followed by a comprehensive analysis of the relatedness. For the occurrence of virulence genes, high diversity was found in different years and subtypes. For SEs, sea, seb, and sec had been commonly detected  Table 2). The other four virulence genes were also commonly detected during the studied period. Importantly, first emerging in 2010, the carriage of eta changed from 7.0% during 2009-2012 to 29.4% in 2015 [36]. In this study, we further performed a comprehensive analysis of the relatedness between virulence genes and other microbial traits, including antibiotic susceptibility, SCCmec, MLST, and spa types. However, no significant correlation was identified between virulence gene carriage and antibiotic susceptibility. Table 4 had comprehensively shown the relatedness between the carriage of each specific virulence gene, with SCCmec, ST types, and spa types separately. For SEs, a higher relatedness rate with type I SCCmec was found, compared with a much lower rate with hospital-associated MRSA (type II and III SCCmec), especially for seb and sed. In addition, it was found that sea showed a higher prevalence in ST239 than other SEs. For tsst and pvl, the higher occurrence was found in CA-MRSA.
Previously in Southern China, type III SCCmec was the dominant with only a few strains carrying type II SCCmec or untypeable [26,27]. However, in this study, diversity in SCCmec was obtained, with type I/IA, II, III/IIIA, IV, V, VI and IX SCCmec totaling 7 types and 2 subtypes within 490 MRSA isolates (Table 2). Despite the high occurrence of type III SCCmec, declination in hospital associated types (70.4% with 345/490 for I, II and III), increase in community-associated types (17.6% with 86/490 for IV and V), as well as diversity in detected SCCmec, had been shown in the present study. Remarkably, the five types (I, IV, V, VI and IX) and one subtype (IA) newly found in this study, had represented the first evidence of type I (mainly in Japan and Korea) and IX (the second identification after Thailand) in China. In addition, prevalence of type III had dropped from higher than 90% to 48.0%, with an increase in type II to 18.6%. Further analysis was conducted in regards with the correlation between virulence factor and SCCmec. Frequently identified in community-associated MRSA (CA-MRSA) strains [37], pvl was detected in only 14.9% of tested strains, of which type II (35.5%) and III (34.2% including 7.9% IIIA) SCCmec were predominant followed by type IV (19.7%), VI (3.9%) and I (2.6%). Remarkably, 24 and 12 S. aureus isolates were pvl + tsst + and eta + etb + , respectively. As genotyping was concerned, the strategy we used was to combine MLST and spa typing. In comparison with previous reports, higher diversity was found in this study, with 7 types and 2 subtypes of SCCmec, 13 STs, and 13 spa types identified from 490 MRSA strains within a study period of 7 years and 2 medical centers. Within STs and spa types, despite ST239 remaining as the dominant type, common SCCmec types such as ST5, ST45, and ST59 had also been frequently detected. Significantly, ST546 (2.1%, 10/467) has been first identified in Asia during 2010-2012, and the carriage rate of ST239 dropped from 76.9% in 2011 to 58.5% in 2015. For clonal complex (CC) as shown in Table 4, ST239-SCCmec II/III/IV, ST5/45-SCCmec III, ST59-SCCmec II, ST546-SCCmec II/III/IV were the prevalent types. In Southern China, ST239-SCCmec III was the only predominant MRSA clone during 2001-2006 (93.8%) and had changed to 37.7% during 2009-2015. Diversity of both SCCmec and ST were also found, with ST239-SCCmec IV, ST5-SCCmec III, ST45-SCCmec III, ST59-SCCmec II, ST1-SCCmec III, ST546-SCCmec II, III and ST546-SCCmec IV identified for the first time. In combination with SCCmec ST and spa types, ST239-SCCmec II/III-t037/t030/t1080 was the predominant MRSA clone accounting for 30% of all tested MRSA strains (Table 3).

Bacterial Identification and Antimicrobial Susceptibility Testing (AST)
For the clinical sample, blood agar plate (Huankai Biotech, Guangzhou, China) was used to specifically identify S. aureus strains. After acquisition of yellow/golden yellow colonies, bacterial identification on all S. aureus strains was performed to the species level according to standard procedures [38], including colony morphology, Gram staining, catalase test, Vitek-2 automated system, and PCR amplification on 16S rRNA (Staphylococcus specific) and femA (S. aureus specific) genes. Methicillin resistance was determined by PCR on mecA using primers M1 and M2. S. aureus strain ATCC29212 carrying 16S rRNA, mecA, and femA genes served as positive control. AST was conducted by Vitek-2 [39] with 10 studied antibiotics including ciprofloxacin, clindamycin, erythromycin, levofloxacin, linezolid, moxifloxacin, oxacillin, quinupristin, rifampin, trimethoprim/sulfamethoxazole, and vancomycin. All results were interpreted according to criteria of Clinical and Laboratory Standards Institute (CLSI 2021). Strain resistance to >2 drugs was considered to be multi-drug resistance [40].

Detection of Virulence Genes
A total of 9 virulence genes were selected to test in this study (Table 2), including 5 SEs genes (sea, seb, sec, sed and see), 2 exfoliative toxins genes (eta and etb), 1 panton-valentine leukocidin gene (pvl) and 1 toxic shock syndrome toxin gene (tsst). Genomic DNA from S. aureus strains for PCR amplification was prepared from overnight cultures according to the instruction of DNA extraction kit (Dongsheng Biotech Co., Ltd., Guangzhou, China). Briefly, harvested cells were orderly treated with lysozyme and proteinase K to obtain lysates. Suspension was purified after the removal of proteins and salts etc. The highly purified DNA was strictly stored at −20 • C. A total of 9 virulence genes were detected by PCR in all S. aureus strains, with tested genes encoding staphylococcal enterotoxins (SEA, SEB, SEC, SED and SEE), exfoliative toxins (ETA and ETB), Panton-Valentine leukocidin (PVL) and toxic shock syndrome toxin (TSST) [36,41]. All PCR assays were performed in triplicate using the primers listed in Table 5 as described previously [42][43][44]. PCR amplification was performed in a volume of 50 µL with 2× PCR Master Mix (Dongsheng Biotech Co., Ltd., Guangzhou, China). The DNA Thermal Cycler EDC-810 (Eastwin Biotech Co., Ltd., Beijing, China) was programmed as follows: the first denaturation at 94 • C for 5 min, denaturation at 94 • C for 30 s, annealing at respective temperature listed in Table 5 for 30 s, and an extension at 72 • C for 90 s for 30 cycles and at last the final extension at 72 • C for 7 min. Three positive strains for each virulence gene were adapted to Sanger sequencing to confirm the accuracy of the amplicons. Strains with correct amplicons were subsequently used as positive controls.

SCCmec Typing
A multiplex PCR strategy is the most available method for SCCmec typing, and it can be used to study the evolution of MRSA [45]. Different SCCmec types were determined by specific primers listed in Table 5. PCR amplification was performed in a volume of 50 µL with 2× PCR Master Mix (Dongsheng Biotech Co., Ltd., Guangzhou, China). The DNA Thermal Cycler EDC-810 (Eastwin Biotech Co., Ltd., Beijing, China) was programmed as follows: the first denaturation at 94 • C for 5 min, denaturation at 94 • C for 30 s, annealing at respective temperature listed in Table 5 for 30 s, and an extension at 72 • C for 90 s for 30 cycles and at last the final extension at 72 • C for 7 min. PCR products were analyzed by electrophoresis on 1.5% agarose gel. S. aureus strains 10442 (carrying ccr1 and IS1272-mecA), N315 (carrying ccr2 and mecI-mecR1), JP25 (carrying ccr3), and WIS (carrying IS431-mecI-mecA) were served as positive controls.

MLST
Strains adapted for MLST and spa typing [20,[49][50][51]. MLST was performed by amplification of the internal region of seven housekeeping genes, including arc, aroE, glpF, gmk, pta, tpi, and yqiL. The amplicons were 456, 456, 465, 417, 474, 402, and 516 bp, respectively. The PCR products were sequenced with ABI Prism 377 DNA Sequencer (PE Applied Biosystems, USA) and compared with the existing sequences available in the MLST website (http://www.pubmlst.org, accessed on 1 November 2022) for S. aureus, and the allelic number was determined for each sequence. The sequence type (ST) was determined according to the pattern of the combination of the seven alleles, and the clonal complex (CC) was defined by the BURST (based upon related sequence types) program v3.0 by accessing the MLST website.

Spa Typing
Protein A, which is encoded by spa gene, is a surface protein originally found in the cell wall of S. aureus and it has been used in biochemical research because of its ability to bind immunoglobulins. According to the number, characteristics, and arrangement of repeated sequences in X region, which is a highly repeated sequence in protein A, the spa typing can reliably and accurately present the polymorphism of S. aureus [52]. PCR amplification targeting spa gene with primers spa-Up and spa-Dn (Table 5) was performed for spa typing. The PCR was programmed as follows: the first denaturation at 94 • C for 5 min, denaturation at 94 • C for 30 s, annealing at 60 • C for 45 s, and an extension at 72 • C for 90 s for 30 cycles, and at last the final extension at 72 • C for 10 min. PCR products were analyzed by electrophoresis on 1.5% agarose gel. Purified amplicons were sequenced with ABI Prism 377 DNA Sequencer (PE Applied Biosystems, Foster City, CA, USA) and repetitive sequences were collected to align to Ridom spa-server (https://spaserver.ridom. de/) (accessed on 4 October 2017) in which 16150 spa types, 709 repeats, and 350761 strains are available.

Statistical Analysis
In this study, antimicrobial susceptibility results and organization were managed in WHONET (version 5.6) (Boston, MA, USA). The chi-square test or Fisher's exact test was applied in the correlation analyses, if appropriate. A p value < 0.05 was defined as statistically significant.

Conclusions
In conclusion, this study has provided comprehensive knowledge on the correlation among antimicrobial resistance, SCCmec, ST, and spa types from a large scale of clinical S. aureus during a lengthy period. An increase in the resistance to trimethoprim/sulfamethoxazole was identified. A total of nine SCCmec types and subtypes, thirteen STs clustered into thirteen spa types were identified, with ST239-SCCmec III-t037 presenting the predominant MRSA clone. Typically, SCCmec type IX and ST546 were emergent types in China. Isolates positive for both pvl and tsst genes and for both eta and etb genes were also identified. The results yielded from this study will aid in further surveillance of molecular epidemiology and evolutionary discrepancy on MRSA.