1. Background
Staphylococcus aureus and especially methicillin-resistant S. aureus (MRSA) is a major human and animal pathogen that causes both nosocomial and community-acquired (CA) infections. S. aureus is responsible for a variety of diseases, from mild skin infections such as abscesses and postsurgical wound infections to severe life-threatening respiratory infections such as sinusitis (1, 2). During the years, these bacteria have been known as one of the most common causes of hospital-acquired (HA) infections, which can acquire resistance to a variety of antibiotics such as oxacillin and vancomycin that enable them to be present and persist in hospitals and communities in different geographical regions worldwide (3, 4).
Pathogenic strains of S. aureus often promote infections by producing exotoxins such as enterotoxins, staphylokinase (SAK), toxic shock syndrome toxin-1 (TSST-1) β-lysin, capsular polysaccharides, lipase, exfoliative toxin, and Panton-Valentine leukocidin (PVL), as well as expressing cell-surface proteins such as microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) (2, 5). Most of the S. aureus virulence factors are encoded by accessory genetic elements, including plasmids, prophages, and pathogenicity islands or by genes located next to the staphylococcal cassette chromosome (SCC) implicated in methicillin resistance (1, 6).
Prophages are widespread among MRSA isolates and are responsible to encode and disseminate potent staphylococcal virulence factors that convert non-virulent isolates to virulent ones, indicating their key role in virulence and pathogenesis. Although staphylococcal prophages have been studied extensively (6-8), nowadays they are used to type MRSA isolates (7, 9, 10).
2. Objectives
We have previously reported the prevalence, SCCmec types, antibiotic resistance patterns and clonal dissemination of 491 MRSA strains in a referral hospital in Tehran, Iran (3), but in the present study using the same bacterial isolates, we investigated the relationship between virulence factors and the presence of different prophage types in MRSA strains.
3. Methods
3.1. Sampling and Identification of MRSA
We have previously described the sampling procedure. Briefly, during three years from July 2011 to September 2014, 2103 suspected strains were isolated from clinical samples of inpatients and outpatients in a referral hospital and its clinic in Tehran, Iran, which were identified as S. aureus using species-specific primers for nucA gene. All the 491 MRSA strains were confirmed according to their resistance to oxacillin (1 µg) and presence of mecA gene (3). In this study, all the S. aureus strains were tested for resistance to cefoxitin (30 µg) as recommended by the Clinical and Laboratory Standard Institute (CLSI) (11).
3.2. DNA Extraction
DNAs of all the MRSA isolates were extracted using QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany). The DNA concentration of each isolate was measured by the Nanodrop ND-1000.
3.3. Prophage Typing
A multiplex-PCR assay consisting of specific primers for SGA, SGB, SGF, SGD and SGL prophage serogroups and SGFa and SGFb as two prophage subtypes was employed for the prophage typing of MRSA strains as described previously (10).
3.4. Detection of Virulence Genes
PCR reactions and specific primers for genes encoding ETA and ETB (exfoliative toxin A and B) (12), TSST (13), PVL (14), SAK (15) and beta-haemolysin (HLB) (15) were employed to test the presence of different virulence factor genes among MRSA strains. The PCR cycles and conditions were the same as published previously.
Moreover, a wide range of staphylococcal enterotoxin genes (sea-seq) were detected using specific primers in separate PCR reactions according to the protocols reported previously (16).
4. Results
4.1. Prevalence of MRSA
Among all the 2103 S. aureus isolates tested, 491 (23.3%) strains showed resistance to cefoxitin and were identified as MRSA. In a previous publication, we also showed that all these 491 strains were resistant to oxacillin and harbored mecA gene, which confirmed them as MRSA strains and were analyzed further.
4.2. Prophage Types and Prophage Patterns
Results of prophage typing of MRSA strains revealed that except for SGD prophage types, all types of prophages were detected in this study and all the strains contained at least one prophage serogroup (SGF) and two subgroups (SGFa and SGFb) (Table 1). Amongst all the four different prophage patterns detected, prophage pattern 3 consisting of SGB, SGF, SGFa and SGFb prophage serogroups was the predominant one (81%; n = 398), and SGA and SGL prophage types were only detected in pvl-positive CA-MRSA strains.
Table 1.Prophage Types and Prophage Patterns of Methicillin-Resistant Staphylococcus aureus Strains
| Pattern | SGA | SGB | SGF | SGFa | SGFb | SGL | No. (%) |
|---|---|---|---|---|---|---|---|
| 1 | + | + | + | + | + | + | 6 (1) |
| 2 | + | - | + | + | + | + | 22 (5) |
| 3 | - | + | + | + | + | - | 398 (81) |
| 4 | - | - | + | + | + | - | 65 (13) |
4.3. Prevalence of Virulence Genes
4.3.1. Enterotoxins
Amplification of staphylococcal enterotoxin genes revealed the presence of 11 enterotoxin genes among MRSA strains, in which all (100%) the strains were positive for sea, sek and seq genes (Table 2). Moreover, sel and seg genes were detected in 32.8% and 20.3% of the MRSA strains, respectively, and the prevalence of sem, sei and sen genes was limited to 12.6%, 8.3% and 4.1% of the strains, respectively. On the other hand, none of the MRSA tested was found positive for genes encoding enterotoxins B, D, E and H, and the frequency of sep, sec and seo genes varied from 0.4% to 2.6%.
Table 2.The Frequency and Enterotoxin Patterns of Methicillin-Resistant Staphylococcus aureus Strainsa
| Pattern | SEA | SEC | SEG | SEI | SEK | SEL | SEM | SEN | SEO | SEP | SEQ | No. (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | + | - | - | - | + | - | - | - | - | - | + | 265 (54) |
| 2 | + | - | - | - | + | - | - | - | - | + | + | 2 (0.4) |
| 3 | + | - | - | - | + | - | - | + | - | - | + | 3 (0.6) |
| 4 | + | - | - | - | + | - | + | - | - | - | + | 12 (2.4) |
| 5 | + | - | - | - | + | + | - | - | - | - | + | 98 (20) |
| 6 | + | - | - | - | + | - | + | + | - | - | + | 2 (0.4) |
| 7 | + | - | + | - | + | - | + | + | - | - | + | 2 (0.4) |
| 8 | + | - | + | - | + | - | + | - | - | - | + | 33 (6.7) |
| 9 | + | - | + | + | + | - | - | - | - | - | + | 4 (0.8) |
| 10 | + | + | + | + | + | + | + | + | + | - | + | 3 (0.6) |
| 11 | + | - | + | + | + | + | + | + | + | - | + | 3 (0.6) |
| 12 | + | - | + | + | + | - | + | + | + | - | + | 7 (1.4) |
| 13 | + | + | + | + | + | + | - | - | - | - | + | 3 (0.6) |
| 14 | + | + | - | - | + | + | - | - | - | - | + | 2 (0.4) |
| 15 | + | - | - | + | + | + | - | - | - | - | + | 7 (1.4) |
| 16 | + | - | + | + | + | + | - | - | - | - | + | 14 (2.8) |
| 17 | + | - | + | - | + | + | - | - | - | - | + | 31 (6.3) |
| N | 491 | 8 | 100 | 41 | 491 | 161 | 62 | 20 | 13 | 2 | 491 | |
| % | 100 | 1.6 | 20.3 | 8.3 | 100 | 32.8 | 12.6 | 4.1 | 2.6 | 0.4 | 100 |
a Enterotoxins B, D, E, H, R, S and T were not detected.
According to the results of enterotoxin genes detection, 17 different patterns of enterotoxin were detected among the strains (Table 2), in which enterotoxin pattern 1 consisting of sea, sek and seq genes was the predominant pattern and was detected in 54% (n = 265) of the strains, followed by pattern 5 (n = 98; 20%). In addition, enterotoxin pattern 10, which harbored all the detected genes, except for sep gene, had the lowest frequency and was found in 3 (0.6%) strains.
4.3.2. Virulence Factor Genes
The hlb gene was detected in 100% of MRSA strains, and 93% (n = 457) of the strains were positive for sak gene encoding SAK. Moreover, none of the MRSA tested was positive for etb gene, and eta (exfoliative toxins A) gene was present in 81% (n = 398) of the strains. tst and pvl genes had the lowest frequencies compared to other virulence factors and were detected in 15.9% (n = 78) and 5.7% (n = 28) of the MRSA strains tested. The presence of pvl gene was only limited to CA-MRSA strains.
5. Discussion
In this study, we described the prevalence of different prophage types among MRSA strains isolated from a referral hospital in Tehran, Iran, during three years. We have reported previously that prophage typing could be a useful method for typing of MRSA strains (7, 9, 17). Our findings revealed that all the prophage types, except for SGD, were present among the strains and SGF prophage type was the predominant one, which is consistent with previous reports during 2012 - 2016 in Iran (7, 9, 10, 16, 18). On the other hand, four prophage patterns were reported in this study, which is similar to previous reports from Iran, in which prophage pattern 3 was reported as the dominant pattern. Dominance of some prophage patterns in the current and previous studies further suggests the circulation of nearly the same MRSA clones in hospitals and communities. The pvl gene was only detected in SGA prophage-positive CA-MRSA strains, which is in line with previous reports (10, 17, 19).
In the present study, 11 different staphylococcal enterotoxins and five virulence genes were detected among all the MRSA strains. These results indicated that there was no special relationship between clinical samples and types of virulence factors. Unlike the previous study (16), we could detect different enterotoxins and virulence factor genes among the isolates. The frequency of sea, sek, seq and hlb genes was higher than that in other reports from Iran and other countries (18, 20-23). Moreover, the enterotoxin pattern 1 consisting of enterotoxin A, K and Q was the most frequent enterotoxin pattern. Different prevalence rates of enterotoxin genes have been reported worldwide (16, 18, 24-29). The discrepancy in the prevalence of different genes is most likely due to the origin of the isolates and the genetic structure of each isolate. The lower prevalence of enterotoxin genes (sea, seb, seg, sei and sej) among MRSA isolates was reported in the Czech Republic (30). The low prevalence of some enterotoxin genes such as seg, sei, sem, sen and seo (so-called egc) in the present study could be due to the theory that indicated these enterotoxin genes are more frequent among commensal strains compared to pathogenic ones (31).
Exfoliative toxin is a causative agent of staphylococcal scalded skin syndrome (SSSS), and documents showed that there were no significant differences in the prevalence of this toxin among MRSA and methicillin-sensitive S. aureus (MSSA) strains. Sila et al. revealed that 3% of MSSA strains carried the eta gene, whereas 10% of MRSA were eta-positive (30). In the present study, we could not find any relationship between the presence of eta gene and patients’ age, but it has been shown previously that exfoliative toxin is more frequent among children compared to adults (32).
ETA is an SGB prophage-encoded virulence factor and was present among all MRSA strains harboring this prophage type. SAK is a 16-kDa prophage-encoded protein, which acts as a fibrin-specific activator of human plasminogen produced by certain S. aureus strains, which indicates the proteolytic activity of MRSA strains (33). We found that 93% of MRSA strains isolated from different clinical samples such as urine, wound, sputum and cerebrospinal fluid (CSF) were positive for sak gene, which is higher than the previous report from Iran (18). Moreover, the tst gene was found in 16% of the MRSA strains. Although different frequencies of tst gene have been reported previously (20, 34-37), our finding in the present study is significantly higher than other reports.
In conclusion, our findings illustrated the presence of highly virulent MRSA strains in a referral hospital in Tehran. The increasing rate of virulence factors may be due to the study of higher number of MRSA strains or the presence of more virulent strains and higher risk patients in hospitals. The presence of different bacteriophages encoding virulence factors among MRSA strains enables them to produce a broad spectrum of diseases, which highlights the potential threat to patients. Bacteriophage typing and the identification of different prophage types could be useful for the prediction of such virulence factors.
