Molecular and Anti-Microbial Resistance (AMR) Profiling of Methicillin-Resistant Staphylococcus aureus (MRSA) from Hospital and Long-Term Care Facilities (LTCF) Environment

To provide evidence of the cross-contamination of emerging pathogenic microbes in a local network between long-term care facilities (LTCFs) and hospitals, this study emphasizes the molecular typing, the prevalence of virulence genes, and the antibiotic resistance pattern of methicillin-resistant Staphylococcus aureus. MRSA isolates were characterized from 246 samples collected from LTCFs, medical tubes of LTCF residents, and hospital environments of two cities, Chiayi and Changhua. Species identification, molecular characterization, and drug resistance analysis were performed. Hospital environments had a higher MRSA detection rate than that of LTCF environments, where moist samples are a hotspot of MRSA habitats, including tube samples from LTCF residents. All MRSA isolates in this study carried the exfoliative toxin eta gene (100%). The majority of MRSA isolates were resistant to erythromycin (76.7%), gentamicin (60%), and ciprofloxacin (55%). The percentage of multidrug-resistant MRSA isolates was approximately 50%. The enterobacterial repetitive intergenic consensus polymerase chain reaction results showed that 18 MRSA isolates belonged to a specific cluster. This implied that genetically similar isolates were spread between hospitals and LTCFs in Changhua city. This study highlights the threat to the health of LTCFs’ residents posed by hospital contact with MRSA.


Introduction
In the past few decades, long-term care facilities (LTCFs) have faced an extreme threat from multidrug resistant (MDR) infectious pathogens [1,2]. The most abundant opportunistic pathogen is methicillin-resistant Staphylococcus aureus (MRSA), which not

The Detection of Toxin Genes by PCR
All isolated MRSA strains carried the exfoliative toxin eta gene (100%, 60/60) ( Table 3). A lower number of enterotoxin genes were present, including entA, entB, and entC (3.3%, 16.7%, and 13.3%, respectively), but no samples carried the entD, entE, or tsst-1 genes. In Changhua samples, the entA (16.7%, n = 6) and eta (100%, n = 6) genes were detected in isolated MRSA strains from the hospital environment, whereas other toxin genes were not found in all other isolates from this area. However, for MRSA strains from LTCF environment, the only genes detected were entC and eta (8.3% and 100%; n = 12, respectively).

Antimicrobial Susceptibility of the MRSA Strains
Eight types of antibiotics were used to analyze the antibiotic resistance capability of the 60 isolated MRSA strains. Overall, the resistance was ranked in order of highest to lowest as erythromycin, gentamicin, ciprofloxacin, clindamycin, tetracycline, sulfamethoxazoletrimethoprim, rifampicin, and chloramphenicol (Table 4). MRSA isolates from Changhua city had higher antimicrobial resistance compared to the Chiayi city. In addition, the Changhua hospital environment's isolates had a higher percentage of resistance against most of the antibiotics, except chloramphenicol and rifampicin, compared to Changhua LTCF environment's isolates. However, a higher percentage of isolates from Changhua LTCF resident tube samples were capable of resisting gentamicin. All the isolates from the Chiayi city LTCF environment were resistant to ciprofloxacin and erythromycin (100%, n = 11), and only one isolate was resistant to tetracycline and gentamicin. In the case of the Chiayi hospital environment, although the percentages of isolates able to resist the ciprofloxacin and erythromycin (50% and 62.5%, n = 8) were observed, this value was significantly higher in Chiayi LTCF environment samples.
For MDR profiling (Table 5), six MRSA isolates (83%) from the Changhua hospital environment had the highest degree of MDR. In the Changhua LTCF environment, the degree of MDR reached 50% (n = 12). In addition, for tube samples, 65.2% (n = 23) of isolates were found to be MDR. For Chiayi, only 37.5% (n = 8) of the total isolates from the hospital environment were MDR, whereas in the LTCF environment, only one isolate was MDR (9.1%).

Genetic Diversity Analysis by ERIC-PCR
The combination of ERIC-PCR analysis with strain information for the 60 MRSA isolates is shown in Figure 1. The MRSA isolates in this study could be divided into three clusters by ERIC-PCR analysis: cluster 1 (24 isolates), cluster 2 (35 isolates), and cluster 3 (one isolate). We further divided cluster 1 into two sub-clusters (Cluster 1-1 and Cluster 1-2). Cluster 1-1 contained more MRSA strains than Cluster 1-2, and maximum number of this cluster's isolates were isolated from Chiayi city samples predominantly consist of SCCmec type IV + PVL genes, classified as community associated, non-MDR MRSA clones, and only carried the eta toxin gene. The MRSA isolates that belonged to Cluster 1-2 (6 isolates) were related to HA-MRSA and contained SCCmec element III and the eta toxin gene. These isolates were MDR strains from Changhua city's LTCF resident samples. Cluster 2 included three sub-clusters: 2-1-1, 2-1-2, and 2-2. The MRSA isolates of sub-cluster 2-1-1 presented a similar genetic profile as cluster 1-2, with the exception of the entB gene, which was present only in cluster 2-1-1.

Discussion
MRSA is one of the most predominant multidrug-resistant pathogens worldwide, and Asia is among the regions with the highest incidence in the world [27][28][29]. The estimated percentage of MRSA in hospital samples varies from 28% to >70% in Asia [30]. Studies regarding hospital environmental MRSA showed that the average prevalence was 2.2% and 11.8% in Ireland and Canada, respectively [31,32]. Several studies have focused on the isolation of MRSA from hospital curtains with MRSA detection percentages ranging from 15.5% to 31.6% [26,29,33]. In our case, MRSA prevalence in hospital environment samples was 18.8% in Changhua city and 25% in Chiayi city. The differences in MRSA prevalence in hospital and LTCF environments may be due to the sample type and local hygiene conditions. A UK study showed that MRSA detection rates were 40% and 17% in Furthermore, all isolates from clusters 2-1-1 and 1-2 were isolated from the dwelling medical tubes of Changhua LTFC residents. Eighteen MRSA isolates belonging to Cluster 2-1-2 showed high MDR properties combined with the eta gene and all were isolated from the hospital and LTCF environments of Changhua. The MRSA isolates of Cluster 2-2 were isolated from a range of samples, and the HA-MRSA isolates that contained SCCmec III elements and carried more than one toxin gene (entB, entC, eta, etc.) were predominant in this cluster. Cluster 3, considered as the outgroup, contained only one MRSA isolate isolated from a Changhua LTCF resident's tube sample.
The chi-square test was used to evaluate the discriminatory power of ERIC-PCR associated with the factors of all isolates, for example, sampling location, sampling source, SCCmec typing, drug resistance, and toxin gene profile. This showed that only the sampling source (P > 0.05) had no significant association with ERIC-PCR analysis while all the remaining parameters had a significant association with cluster classification of ERIC-PCR.

Discussion
MRSA is one of the most predominant multidrug-resistant pathogens worldwide, and Asia is among the regions with the highest incidence in the world [27][28][29]. The estimated percentage of MRSA in hospital samples varies from 28% to >70% in Asia [30]. Studies regarding hospital environmental MRSA showed that the average prevalence was 2.2% and 11.8% in Ireland and Canada, respectively [31,32]. Several studies have focused on the isolation of MRSA from hospital curtains with MRSA detection percentages ranging from 15.5% to 31.6% [26,29,33]. In our case, MRSA prevalence in hospital environment samples was 18.8% in Changhua city and 25% in Chiayi city. The differences in MRSA prevalence in hospital and LTCF environments may be due to the sample type and local hygiene conditions. A UK study showed that MRSA detection rates were 40% and 17% in LTCF environmental samples in 2011 and 2013, respectively [34]. This study also implied that the lower frequency in 2013 might be due to improved infrastructure in the new LTCF. This agreed with our results that showed a lower MRSA detection percentage in Chiayi LTCF, which had a better hygiene environment compared to that of Changhua LTCF. Liu et al. demonstrated a disparity in MRSA detection rates between six LTCFs and concluded that improved infrastructure and a good hygiene environment constitute a powerful approach for reducing MRSA prevalence [27]. Our results implied that MRSA strains were present with high occurrence in moist environments, which agreed with other investigations, as moist samples act as prominent hotspots of MRSA distribution [7,8].
Here, the highest detection rate of MRSA was in nasogastric tubes, which are frequently used in LTCF residents.
The predominant SCCmec types were SCCmec III and IV, which agrees with studies of LTCFs in Taiwan [27,29]. However, the predominant SCCmec types in the two hospital environments of this study were different. The dominant type was SCCmec III in Chunghua hospital but was SCCmec IV in Chiayi hospital. HA-MRSA (31 isolates) was the dominant classification in this study, where HA-MRSA typically belongs to SCCmec I, II, and III [30]. Clinical studies in Taiwan also indicated that most of the HA-MRSA strains from hospitals belonged to SCCmec III [30,35,36]. Between 17.1% and 21.1% of LA-MRSA strains were isolated from hospitals and CTCFs, respectively, and all LA-MRSA isolated here belonged to SCCmec IV+PVL. This type is the same as our previous LA-MRSA study in the river basin of Chiayi [37]. Huang and Che's study also concluded that more than 80% of Asian-specific LA-MRSA strains in Taiwan carried SCCmec IV and PVL genes, which strengthens our results [38].
The eta and etb enterotoxin-associated genes and tsst-1 are frequently prevalent virulence genes that vary between countries [30,33,39]. All the MRSA strains isolated in this study carry the eta gene; eta is reported to be the most privileged virulence factor-encoded gene, although this is contradicted in some reports [33,[40][41][42]. Additionally, in contrast to other studies, we did not detect any entD, entE, and tsst-1-positive strains [33,42]. This showed that the distribution of toxin profiles of MRSA was highly diverse and implied geographic differences in these distributions. The MRSA isolates carrying the entB or entC genes all belonged to SSCmec III, except for one CA-MRSA isolate that carried two toxin genes, whereas all other isolates carried the eta gene only. This result is similar to the marginal differences identified in the distribution of toxin genes with SCCmec types reported by Fooladi et al. [33].
Antibiotic susceptibility testing revealed that MRSA isolates were more resistant to ciprofloxacin, erythromycin, and gentamicin than to other antibiotics, and 50% of these were related to MDR. MRSA from the hospital environment had more MDR ability compared with that from the LTCF environment. Furthermore, the prevalence of MDR MRSA was higher in Changhua city than in Chiayi city, and these were mainly HA-MRSA strains, followed by CA-MRSA from hospital environmental samples and LTCF resident samples, which agreed with a previous report [43]. Twenty-six of a total of 30 MDR MRSA isolates were identified in Changhua city and, among them, seven isolates were resistant to six antibiotics; this is an urgent and concerning issue and implies a crosscontamination problem.
ERIC-PCR is one of the most effective DNA-based molecular typing methods used to categorize MRSA isolates associated with epidemiological modeling and helps to track their spreading route [44]. In our ERIC-PCR fingerprinting, most of the MRSA isolates carrying the SCCmec IV + PVL genes (CA-MRSA) were from Chiayi LTCF environment samples and belonged to Cluster 1-1. Sub-clusters 1-1 and 2-1-1 were all from the Changhua LTFC resident samples, indicating that there were two main types of MRSA distributed in these LTCF residents.
In contrast, isolates in sub-cluster 2-1-1 from Changhua hospital and LTCFs generally consisted of SCCmec III (HA-MRSA) with MDR isolates. This finding implied that a genetically similar type of MRSA had spread between LTCFs and hospitals. ERIC-PCR aided the discrimination of resistance patterns among SCCmec types reported in our previous study [37]. ERIC-PCR is a popular tool for determining the genetic relatedness of different multidrug-resistant pathogenic bacteria from environmental samples, and previous analyses are consistent with our results [45,46]. This conclusion is further supported by the chi-square statistical hypothesis, where the individual significance score was < 0.05 with ERIC-PCR results. A similar analysis, adopted by Akindolire et. al. to distinguish the genetic diversity of isolated MRSA strains from their samples, strongly supported our ERIC-PCR results [47]. Therefore, using the ERIC-PCR figure combined with more detailed information, including toxin profiles, sampling location, other typing methods, and drug-resistant profiles, is a favorable approach to provide a complete understanding of pathogen transmission, microbial contamination, and surveillance spot information.

Sampling Information and Collection Method
Samples were taken from the hospital and LTCF environment of two cities in Tai Table S1. Sterile cotton swabs were used to collect bacterial samples from the surface of each sampling point. Swabs were soaked in 5 mL of sterile phosphate buffer saline (PBS) in a centrifuge tube for storage and transported, at low temperature, to the laboratory for analysis.

Isolation Method of MRSA
To isolate MRSA from collected samples a two-step selective culture process was used. For the growth of MRSA colonies, we used CHROMagar™ MRSA (TPM ready-to-use media) and Baird-Parker agar (TPM ready-to-use media). The colonies that were grown on the respective cultured agar plate after the incubation period (30 • C for 24 h), were transferred into a sterile test tube containing brain heart infusion broth for pure culture [48]. Species identification, molecular characterization, and drug resistance analysis of isolated colony pure were performed using 300 µL broth from the enrichment culture.

PCR Identification of MRSA Strains
An aliquot of 300-600 µL of well-grown bacterial culture was centrifuged for 5 min at 10,000 rpm for DNA extraction. Genomic DNA was extracted using a commercial bacterial DNA extraction kit (MagPurix Bacterial DNA Extraction Kit, ZP02006, Taiwan, China). A reference bacterial genomic DNA (MRSA ATCC 29213) was used as a positive control (extracted by the same method and kit) [37]. After gDNA extraction, the PCR reaction mixture was prepared using 300 µg of gDNA with primers and master mix (Fast-RunTM Taq Master Mix with Dye). The total PCR reaction volume was 25 µL, and PCR reaction conditions are described in Supplemental Table S2. ERIC-PCR was used for PCR-based typing, and BioNumerics software was used to analyze kinship typing [37,49]. The nuc and mecA genes were used for the identification of MRSA strain types [50]. SCCmec and Panton-Valentine leukocidine (PVL) were used to confirm the classification of mec elements [51]. Some enterotoxins, toxic shock syndrome toxin-1 (TSST-1), and exfoliative toxins (ETs) genes were targeted for detection within the S. aureus. Finally, all PCR products were assessed by electrophoresis (110 V, 30 min, 1.5% agarose gel) to check the respective gene-specific amplicons.

Conclusions
This study showed that MRSA occurs at relatively higher rates in moist samples, especially in LTCF resident dwelling tubes. The eta gene was commonly found in all MRSA isolates. The hospital environmental MRSA isolates from Chiayi city carried the highest amount of enterotoxin genes, such as entA, entB, and entC, compared with those from Changhua city. Additionally, the MRSA isolates from the hospital environment and LTCF resident dwelling tubes exhibited high percentages of multidrug resistance to the following antibiotics: ciprofloxacin, clindamycin, erythromycin, sulfamethoxazole-trimethoprim, and tetracycline. The multidrug resistance problem of healthcare-associated MRSA is more severe in Changhua city than in Chiayi city. The SCCmec III containing HA-MRSA was identified as rank one and PVL + SCCmec IV element carrying CA-MRSA was rank two group of MRSA strain among 60 isolated S. aureus clones. ERIC-PCR was an effective tool for epidemiological characterization of MRSA isolates. The presence of genetically similar and multidrug-resistant MRSA in hospitals and their affiliated LTCFs is a threat to public health and needs to be closely monitored and controlled.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/antibiotics10060748/s1, Table S1: Details of samples location, source, types and number, Table S2: MRSA strain identification (ERIC-PCR), SCCmec typing and enterotoxin gene detection condition of PCR with primers information.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare that they have no conflict of interest.