Molecular Characteristics and Changing Trend of Methicillin-Resistant Staphylococcus Aureus From Invasive Infections between 2012 and 2018

Background Methicillin-resistant Staphylococcus aureus (MRSA) is a major global problem. The analysis of the molecular characteristics and changing trend of MRSA is essential for the control and treatment of diseases caused by the pathogen. Methods A total of 162 MRSA isolates from invasive infections between 2012 and 2018 were collected, molecular typing and antimicrobial susceptibility tests to explore its molecular epidemiologic change in a hospital were performed. Results All of the 162 MRSA isolates (86.4% HA-MRSA and 13.6% CA-MRSA) were divided into 16 different ST and 30 Spa types. The major STs were ST5 (96/162, 59.3%) and the predominant spa type was t311(83/162, 51.2%). Five SCCmec types were found and the most common SCCmec type was type II (101/162, 61.7%). The prevalence of ST5 MRSA gradually declined from 2014 to 2018 but the prevalence of ST59 MRSA signicantly increased. At the same time, livestock-associated methicillin-resistant S.aureus ST239 and ST9 were detected. 28 isolates were Panton-valentine leucocidin (pvl) gene positive (28/162, 17.3%). The most prevalent pvl-positive clone was ST59-IVa-t437. Comparing with HA-MRSA, CA-MRSA had a lower probability of ST5 (9.1% vs, 67.1%, P=0.000) but a higher probability of ST59 (63.6% vs. 11.4%, P=0.000), not only that, it was more likely to carrying pvl-positive gene (36.4% vs. 14.3%, P=0.028). S. 25923 for quality control. DNA extraction was performed from pure S. aureus cultures after 24 h of incubation at 37°C on Columbia agar + 5% sheep blood (bioMérieux) using QIAamp DNA Mini Kit (QIAGEN, CA, USA) according to the manufacturer’s specications. Extracted DNA was stored at − 20°C for further analyses. All isolated DNA was used as template for all PCR reactions.


Background
Staphylococcus aureus is a notorious pathogen which is able to cause widespread infections such as pneumonia, sepsis, endocarditis, toxic shock syndrome, or necrotizing fasciitis 1 . Besides, Methicillin-resistant S. aureus (MRSA) infections, which threaten the public health safety of the world, are associated with higher mortality and higher health care costs than infections caused by methicillin-susceptible S. aureus (MSSA) 2 . According to the surveillances, the prevalence of MRSA in hospital is as high as 70%-80% in many Asian countries 3,4 . In China, although a decrease of MRSA prevalence was observed in recent years, the high prevalence around 30% of MRSA were still causes major problems 5 . Not only that, due to the continuous evolution of strains, the constant change of medications and the increasing frequency of personnel migration among regions, the MRSA infection becomes more complicated and diverse, which may bring new challenges to the infection management.
By the microbiological and clinical characteristics, community-acquired (CA)-MRSA infection differed from healthcare-associated (HA)-MRSA infection. Over the past decade, CA-MRSA infection have been increasing while HA-MRSA infection declined 6 . In terms of the molecular epidemiology, most CA-MRSA isolates carry SCCmec IV or V compared with HA-MRSA 7,8 . And CA-MRSA are considered more virulent because they possess speci c virulence factors including Panton-Valentine leucocidin (pvl) 8 . In addition, the molecular characteristics of MRSA also have signi cant differences in different regions and change over time. For example, in the United States, ST8 (USA300) has been the most prevalent type 9 while in many Asian countries, ST5 and ST239 were the most common clones which have been observed 10,11 . Since 2000, the Asian-dominant HA-MRSA ST239 clones have persisted and adaptively evolved in hospital environments for decades 12 . However, a previous study conducted in a general teaching hospital in Shanghai, China has demonstrated that the predominant HA-MRSA clones, ST239-t030 and ST239-t037, were being replaced by the continually growing ST5-t2460 clone in 2017 13 . Therefore, it is valuable to keep abreast of the prevalent strains, the changing trend and drug resistance of MRSA infection.
In this study, we reported an in-depth epidemiological and genomic investigation of MRSA from invasive infections between 2012 to 2018 in a teaching hospital in east China. The aim of this work was to explore the molecular characteristics and changing trend of MRSA isolates so that provide a basis for prevention and treatment of MRSA infections.

Bacterial isolates
This study was conducted at the First A liated Hospital, College of Medicine, Zhejiang University, a 2500-bed teaching hospital in Eastern China. A total of 162 isolates collected between January 2012 to December 2018 were isolated from invasive infections, including the blood (n = 111), pleural effusion (n = 11), abdominal uid (n = 19), cerebrospinal uid (n = 5) and other sterile sources (n = 16). All isolates were con rmed to be S. aureus by MALDI-TOF MS (Bruker Daltonics, Bremen, Germany) and were de ned as methicillin resistance by cefoxitin disk diffusion method according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) standards 14 . The investigation was a retrospective study and all patients were anonymized; informed consent was waived.
According to the Centers for Disease Control and Prevention(CDC) criteria of CA-MRSA, cases of CA-MRSA infection de ned as cultured <48h after hospital admission or in the outpatient setting, having no history of MRSA infection or colonization, admission to healthcare facility, dialysis, surgery or insertion of indwelling devices in the past one year. Patients were de ned as HA-MRSA infection when cultured >48h after hospital admission and contacted with previous healthcare which means MRSA linked to a hospitalization but presenting in the community or at hospital readmission 15,16 .

Antimicrobial susceptibility testing
The antimicrobial susceptibility of all isolates was evaluated using the microdilution broth method. Results were interpreted in accordance with the CLSI guidelines 14  DNA extraction DNA extraction was performed from pure S. aureus cultures after 24 h of incubation at 37°C on Columbia agar + 5% sheep blood (bioMérieux) using QIAamp DNA Mini Kit (QIAGEN, CA, USA) according to the manufacturer's speci cations. Extracted DNA was stored at − 20°C for further analyses. All isolated DNA was used as template for all PCR reactions.
Con rmation of MRSA and detection of pvl All S. aureus isolates con rmed to be methicillin-resistant by cefoxitin disk diffusion method were collected to test the presence of mecA or mecCgene as previous described by the means of PCR 17,18 . All isolates were screened for the Panton-Valentine leucocidin (PVL) using the PCR method as previous described 19 .

Molecular typing methods
Three typing methods were used for all MRSA strains. MLST was carried out by PCR ampli cation and sequencing of 7 housekeeping genes (arcC, aroE, glpF, gmk, pta, tpi, yqiL) according to the method as previous described 20,21 . The sequences of PCR products were compared with existing sequences available on the MLST website (http://saureus.mlst.net) for S.aureus, and the allelic number was determined for each sequence. Clustering of related STs that were de ned as cloned complexes (CCs) was performed by using the program eBURST algorithm 22 .
We used a multiplex PCR method to determine the SCCmec types 17 . SCCmec types were assigned in accordance with the combination of the cassette chromosome recombinase (ccr) type and mec class. All MRSA isolates which could not be con rm with any type from I to V were de ned as non-typable (NT). Spa typing was carried out by ampli cation and sequencing of polymorphic X region of the protein A gene using the method of PCR 23 . Spa types were assigned by spa database website (http://spa.ridom.de) .

Statistical Analysis
Statistical analyses were performed using SPSS Version 23.0 (IBM Corporation, Armonk, NY, USA). The chi-square test was used to compare the proportion of ST59 and ST5 strains, the probability of bloodstream source infection and distribution of virulence genes between CA-MRSA and HA-MRSA. P-values ≤0.05 were considered signi cant.

Antimicrobial susceptibility
A total of 162 MRSA isolates (140 HA-MRSA and 22 CA-MRSA) were tested for antimicrobial susceptibility. The results showed that all of the isolates were susceptible to vancomycin, tigecycline, and linezolid, high resistance to erythromycin, uoroquinolones, tetracycline, and clindamycin, lower resistance to gentamicin, rifampin, and sulfamethoxazole/trimethoprim (Table 1).

Discussion
MRSA represented a major problem for public health systems, as they resulted in high incidences both in hospital and community settings 24 . Previous studies have shown that MRSA infections were mainly caused by a few predominant clones causing outbreak in hospital 25 . Not only that, the predominant MRSA clones have their regional characters and can change over time 13,26 . Therefore, the studies analyzed the molecular typing and changing trend of MRSA can clarify the local epidemic clones and provide a basis for rational treatment and effective control of drug resistance.
In this study, one of the most important ndings was the molecular type changed among MRSA isolates. The most common MRSA clone was ST5-II-t311 in our study, but the prevalence of ST5 is getting down; at the same time, ST59 is increasing gradually, and ST239 is getting rare.

Page 5/11
ST5-II MRSA (NEW York/Japan) clone was one of the most predominant genotypes in many countries 27,28 . Also, previous study showed that ST5 was the predominant HA-MRSA clone in Zhejiang, China from 2012 to 2013 29 . Moreover, some investigators have expressed different opinions on the most main spa type of ST5. It has been reported that t002 was main spa type in ST5 clone isolates in previous study 30 , but ST5-II-t311 has become the predominant HA-MRSA clone in Hangzhou, Zhejiang Province between 2012 to 2013. In our study, we demonstrated that t311 to be dominant in ST5 clonal isolates while t002 was the second spa type, which was similar to previous study 31 . Notably, the prevalence of ST5-II-t311 decreased from 2014 (12/18, 66.7%) to 2018 (6/28, 21.4%) while ST5-II-t2460 slightly increased. Li et al. 32 even showed that ST5-t2460 was the most common clone in S. aureus causing bloodstream infection, which had never been reported in China before. Therefore, whether the ST5-II-t2460 clone has been conferred more competitive advantages remains bewildering.
ST59 was a major CA-MRSA and the second common MRSA in ours results. Our nding indicated the distinctions between CA-MRSA and HA-MRSA isolates have become blurred and they could be a mutual integration 29 . Previous studies indicated that ST59 MRSA might spread into hospital from community 29,33 . Similarly in the study, the ST59 clone showed a gradually increasing trend to replace ST5, higher SCCmecIV/V and pvl positive prevalence (14/28, 50%) in ST59 fatherly con rmed the stipulation 34 . More attention to the dissemination of ST59 in hospitals need.
Of note, previous studies have demonstrated ST239-III-t030 and ST239-III-t037 to be the most prevalent clones of MRSA in China 34 .
However, in our study, we only found 3 isolates with ST239. Similarly, some scholars have found that ST239 was gradually decreasing in China 31 . The reasons need to be elucidated. Some studies inferred that ST239-t030 MRSA displayed lower growth rate and lower competitive advantage compared to ST59-t437 MRSA. These may also be one of the reasons why ST239 gradually decreased and ST59 gradually increased, although ST239 is a high resistant strain.
A number of STs that were rare in previous reports also caught our attention. As we all know, MRSA sequence type ST398 referred to as LA-MRSA (livestock-associated methicillin-resistant S.aureus), from pigs, farmers and environment 35 has been reported in Europe 36 and North America 37 . Interesting, we found eight ST398 (4.9%) isolates identi ed in our study, which was the third highest among all STs. Not only that, infections caused by ST9 which also belonged to LA-MRSA were seldom reported from human, but mostly in pigs. However, one ST9 isolate were identi ed by molecular typing in our study, which may suggest that cross-species transmission of the emerging ST9 MRSA between swine and humans.
We acknowledge several limitations in our study. As it was a retrospective study, this study had not collected the clinical data including prognosis, severity of disease and others, which may cause that this study had no information about the relationship between clinical data and molecular characteristics. Secondly, the single center isolates analyzed may in uence the representativeness of the study. Finally, we could do more in-depth analysis such as WGS (Whole Genome Sequencing) for special strains including ST398 and ST9.

Conclusion
In conclusion, ST5-II-t311 was the predominant clone of MRSA isolate in our hospital, but showed a downward trend generally from 2012 to 2018. Community-associated ST59 MRSA strains were revealed to have spread into hospitals and there may be an upward trend in the future. Some LA-MRSA may spread into human being. We should regularly monitor the prevalence and drug resistance of methicillin-resistant Staphylococcus aureus and take effective measures to prevent and control the occurrence of nosocomial infection.

Declarations
Author contributions:THX designed the study. YYZ conducted the correlation analysis and prepared the drafts of the manuscript. YW, HX and TTX provided assitance with the study design and statistical analysis. KY, STZ YZZ were responsible for the results interpretation and manuscript review. JRJ and PS helped the identi cation of bacteria. All authors agree to be accountable for all aspects of the work.

Ethics: Ethics approval was submitted and approved through Research Ethics Committee of the First A liated Hospital, College of
Medicine, Zhejiang University. The consent to participate was waived by our institutional review board since this study was retrospective data collection.
Transparency declarations: No potential con ict of interest was reported by the author(s).  Tables   Table 1 Antibiotic resistance