1. Introduction
Staphylococcus aureus is a common opportunistic pathogen that causes a variety of infections due to the presence of many colonization factors and virulence factors. It is one of the most frequent causes of skin and soft tissue infections (SSTIs) such as skin abscesses, furuncles, impetigo, and wound infections. Some of them, especially in patients with risk factors (diabetics, patients during immunosuppressive or cancer therapy, patients with indwelling catheters, HIV/AIDS), may progress to severe infections and require hospitalization.
S. aureus is also a leading cause of serious infections, such as bacteremia (
S. aureus bacteremia, SAB) or infective endocarditis, which can have serious consequences for the patient. High morbidity and mortality are associated especially with the widespread occurrence of methicillin-resistant
S. aureus (MRSA) strains. Resistance to all β-lactams (except for the latest generation of cephalosporins) and other antibiotics commonly used limits therapeutic options for treating staphylococcal infections [
1,
2,
3].
Among virulence factors produced by this microorganism, the pore-forming toxins (e.g., hemolysins, leukotoxins), enzymes (e.g., nucleases, staphylokinase, coagulases), epidermolytic toxins, and superantigens (SAgs) can be distinguished.
Panton–Valentine leukocidin (PVL)—encoded by two genes,
luk-S-PV, and
luk-F-PV—is an extracellular protein with dermonecrotic and leucocidal functions. It has cytotoxic activity against mammalian neutrophils, monocytes, and macrophages. The toxin can be produced both by community-acquired MRSA (CA-MRSA) as well as community-acquired methicillin-susceptible
S. aureus (CA-MSSA) strains. PVL-positive strains usually cause SSTIs, but they are also associated with severe infections, including septic arthritis, bacteremia, necrotizing pneumonia, and purpura fulminans [
4,
5].
Exfoliative toxins (ETs) are specific serine proteases that recognize and hydrolyze desmosome proteins in the skin and, therefore, play a role in host colonization and the invasion of injured mucosa and skin. ETs are causative agents for localized epidermal infections and generalized disease—staphylococcal scalded skin syndrome (SSSS). Serotypes ETA, ETB, and ETD are also associated with septic shock and may increase the severity of the infection [
6,
7].
Toxic shock syndrome toxin-1 (TSST-1) and staphylococcal enterotoxins (SEs), with their superantigenic properties, may contribute to the development of serious infections due to the activation of enormous numbers of T lymphocytes. TSST-1 is a bacterial exotoxin produced by approximately 5–25% of
S. aureus strains isolated from samples of different origins [
8,
9]. It is responsible for menstrual and non-menstrual cases of toxic shock syndrome, a severe and potentially fatal disease in humans or for neonatal toxic shock syndrome-like exanthematous disease. SEs are subdivided into classical SEs with emetic activity and SE-like proteins with no or relatively low emetic activity [
10]. SEs play an important role in staphylococcal food poisoning, but these virulence factors are associated also with the pathogenesis of several other human diseases, including pneumonia, sepsis-related infections, or toxic shock syndrome. Futhermore, SEB is capable of inhibiting keratinocyte proliferation and migration and may delay wound closure [
11,
12].
Recent studies have shown that staphylococcal SAgs affect the site of infection to cause tissue pathology and are crucial in the development and progression of sepsis and infective endocarditis [
13]. SAgs produced by
S. aureus colonizing the wounds may also play a major inhibitory role in the healing of chronic wounds because of the recurrent exposure to those virulence factors [
14].
This study aimed to compare drug susceptibility and virulence patterns in S. aureus strains isolated from patients with SAB or a chronic wound.
3. Discussion
S. aureus invades different body sites and causes a wide range of infections from mild skin and soft tissue infections to life-threatening diseases. Bloodstream infections remain especially associated with significant mortality, despite the improvements in the treatment of staphylococcal infections [
15]. SAB may be associated with infective endocarditis, osteomyelitis, and other metastatic infections. Chronic wounds, often caused by this pathogen, can also be a source of bloodstream infections. Furthermore, non-healing wounds due to bacterial infections and biofilm formation are related to increased health costs [
16].
A significant concern for public health are MRSA strains. Although, the decline in the resistance percentage was noted in recent years in some countries (e.g., the United Kingdom, Ireland, Germany, Portugal), MRSA remains an important pathogen involved in community- and health care-associated infections [
17]. The frequency of isolation of MRSA strains from chronic wounds is varied and depends not only on the country, but also on the population of wound patients studied. Risk factors for infection with a multi-drug resistant microorganism include previous hospitalization or stay in a chronic care center (the possibility of cross-contamination of patients’ wounds or through the hands of medical staff), or previous antibiotic therapy. Moreover, chronic wounds are a good environment for the development of many different bacteria, which promotes horizontal gene transfer and selection of resistant strains [
18,
19]. MRSA bloodstream infection, in turn, is associated with high mortality, especially in intensive care patients. In the presented study, 21.0% of
S. aureus isolated from chronic wounds were found to be MRSA, whereas Shettigar et al. [
20] identified in diabetic foot ulcer patients an almost three times higher percentage of methicillin-resistant strains. The incidence of methicillin resistance that we have noted among strains isolated from blood was similar (23.0%) to that of wound strains. A statistically significant difference was indicated between the two groups of strains in resistance to amikacin (12.0% vs. 31.0%), gentamicin (1.0% vs. 13.0%) and tobramycin (14.0% vs. 32.0%). Moreover, among the strains isolated from blood, there was a lower percentage of strains resistant to norfloxacin, but this difference was not statistically significant. However, a similar percentage of blood (11.0%) and wound (14.0%) strains had a constitutive mechanism of MLS
B resistance. Compared to it and compared to strains isolated from blood, the inducible MLS
B mechanism was slightly more frequently reported in strains derived from chronic wounds (19.0 vs. 14.0%).
Benzylpenicillin resistance was found in most strains in the present study and the studies by other authors [
20,
21,
22,
23,
24], regardless of the clinical samples from which they were recovered. Among the remaining antibiotics, the highest percentage of all strains tested showed resistance to clindamycin (30.0%), erythromycin (29.5%), and norfloxacin (27.0%). Similar findings (19.8%, 28.7%, and 27.7%) were reported by Pomorska-Wesołowska et al. [
25] in Southern Poland on strains isolated from SAB and pneumonia. Other studies conducted by Wang et al. [
22], Yu et al. [
21], and Liang et al. [
26] indicated that the frequencies of erythromycin and clindamycin resistance among isolates collected from blood were significantly higher and amounted to 65.0%, 61.8% and 46.7% for the former and 58.3%, 48.3% and 40.0% for the latter of these antibiotics. The widespread use of some macrolides against infections caused by various bacteria promotes among staphylococci colonizing the human body the selection of strains resistant to these antibiotics. In contrast, clindamycin is an alternative to beta-lactam antibiotics for patients allergic to penicillins and can also be used to treat infections caused by MRSA strains. Therefore, in recent years, an increase in the number of strains resistant to macrolides and lincosamides, in particular, the iMLS
B phenotype, has been observed [
27,
28].
Varshney et al. [
29] reported that among strains derived from wounds, 5.6% are susceptible to all antibiotics and 40.7% are susceptible to all antibiotics, except for penicillin. We found similar percentages (6.0% and 35.0%, respectively). However, in the above-mentioned studies, in the case of strains isolated from blood, these percentages were lower compared to the results of our research (6.1% and 23.2% vs. 14.0% and 46.0%).
S. aureus pathogenicity correlates with bacteria’s capability to produce and secrete a variety of virulence factors that contribute to colonization, invasion, and damage of the host tissue and bacterial spread [
30]. Many virulence factors and biofilm formation are under the control of the accessory gene regulator-quorum sensing system (
agr-QS). Toxins or proteases secreted by
S. aureus have an important role in the ability to cause disease, but may become redundant once a chronic infection is established. Because many virulence factors are targets for the immune system, their downregulation or loss of agr-qs prevents further recruitment of phagocytes so is crucial for bacterial survival and long-term persistence in host cells/tissue. Strains isolated from chronic infections may also rewire their metabolism to promote biofilm production or may survive in a metabolically inactive state forming small-colony variants [
31,
32,
33].
In this study, among all the enterotoxin-encoding genes tested,
sei and
seg were the most frequent. A combination of those two genes was also found to be the most common. This is in accordance with the previous study [
5,
15,
21,
34,
35,
36] and related to the close location of both genes on the enterotoxin gene cluster (
egc) [
37]. A significantly higher percentage of
sei,
seg and
sea was found in chronic wound strains than in blood strains (61.0%, 59.0% and 19.0% vs. 46.0%, 45.0% and 6.0%, respectively). A similar relationship was observed in another study [
38], among strains isolated from wounds (56.3%, 56.3% and 43.8%) and blood (41.2%, 41.2% and 35.3%), whereas Pérez-Montarelo et al. [
15] recorded a slightly lower percentage of strains causing SSTI harboring those genes compared to strains causing catheter-related bacteremia (67.7%, 68.2% and 30.8% vs. 72.8%, 75.6% and 31.4%, respectively). It is believed that
egc superantigenes facilitate the colonization of mucosal surfaces and promote bacterial survival but negatively correlate with the severity of infections [
39,
40]. We also observed a higher rate of
sea and
sed genes among wound strains, which is in accordance with other data [
41]. It is considered that
sea and
sei may be a biomarker to distinguish colonization and infection of a wound [
38]. SEA and SED, with their increased ability to induce local inflammatory responses, in turn, may perform a function in perpetuating chronic non-healing wounds. Moreover, Merriman [
11] demonstrated that superantigens can delay wound closure by altering cell proliferation and migration.
In our study, the distribution rate of genes other than
seg and
sei enterotoxin genes among blood strains was much lower (1.0–12.0%). These findings are inconsistent with those reported previously, where
sea [
15,
22,
39],
seb [
5,
35,
42], or
sec [
35] were also widely present. However, Jarraud et al. [
43] suggested that
seg-sei gene combination without any of the other SEs, TSST-1 and ETs may also be capable of causing diseases such as toxic shock syndrome or staphylococcal scalded skin syndrome.
Park et al. [
36] found
eta and
etb genes in 2.6% and 100% SAB strains, respectively, in a two-year study period, but many studies [
3,
24,
35,
38,
44] showed a higher distribution of
eta compared to
etb. This is in accordance with our results, where the
eta gene was detected in 13.0–14.0% of the
S. aureus strains, whereas only 0–1.0% strains harbored the
etb gene. Li et al. [
35] reported that both genes were more common in SAB isolates compared to SSTI isolates recovered from children. The frequency of occurrence of genes encoding exotoxins, however, varies among blood strains of
S. aureus and may also be equal to or near zero [
34,
42,
45,
46].
The
luk-F/S-PV gene is detected more often in SSTI isolates than in SAB isolates [
15,
26,
32,
45,
46,
47,
48], which confirms our results. The incidence of
luk-F/S-PV gene in this study, however, was not high (3.0%), unlike the results of other researchers, in which it reached about 80% [
35,
47]. These differences may result from testing only strains derived from chronic wounds compared to tests involving different types of SSTI. In the studies conducted on staphylococci isolated from patients with diabetic foot infections, Viquez-Molina et al. [
49] detected the
luk-F/S-PV gene in 6.9% of strains and, in another study [
50], this gene was not found in any of the isolates.
TSST-1 produced by
S. aureus has been associated with several acute diseases including toxic shock syndrome, but also with chronic diseases. Our data showed that the
tst gene was carried by a relatively low (9.0%) percentage of blood strains, compared to the results of previous studies in which this percentage usually exceeded 10% [
3,
15,
34,
36,
38,
45,
50,
51]. Compared with SAB strains, the carriage rate for the
tst gene in chronic wound strains was slightly higher (13.0%), although the difference was not significant. This is in agreement with the results obtained with strains isolated from DFU and other SSTI [
15,
20,
26,
38].
Compared to strains derived from chronic wounds, a high percentage (35.0%) of strains isolated from patients with SAB did not have any of the genes tested. A lower percentage (25.6% and 29.4%) of
S. aureus strains obtained from blood without the virulence genes investigated was found by Becker et al. [
34] and Demir et al. [
38], respectively. In the present study a considerable variety of virulence profiles was observed among all strains, and only 12 of the 39 profiles coincided in both groups. The possession of single genes was observed in 18.0% and 15.0% of wound and blood strains, respectively. A similar result was reported in the study by Demir et al. [
38].
We found that SEs genes were more frequent among MRSA than MSSA strains. Other studies also showed a significant association between methicillin resistance and the presence of the majority of SEs genes [
5,
21,
36,
42,
52]. In our study, among the tested genes encoding enterotoxins tested, only
sea was found in more MSSA strains than MRSA. Park et al. [
36] found no
sea gene in MRSA bacteremia isolates, whereas over 16% of MSSA strains carried this gene.
Our results support the results of other studies [
23,
36,
50,
53], in which a higher rate of
tst-carrying isolates was detected among methicillin-susceptible isolates compared to that for MRSA isolates, which indicates a possible, inversely proportional relationship between drug resistance of the strains and their virulence in the occurrence of virulence genes.
To our knowledge, scientists most often analyze the difference in the prevalence of virulence genes in MSSA and MRSA strains, while the relationship between the presence of genes encoding toxins and drug susceptibility is less frequently assessed. No significant difference between the antibiotic resistance of non-enterotoxigenic strains and strains producing SEA, SEB, SEC, and SED from mastitic milk was reported by Suleiman et al. [
54]. Corredor Arias et al. [
55] drew similar conclusions from research results in which 22 superantigen genes were detected in strains isolated from different clinical samples. Choopani et al. [
56] reported a relationship between SEG and SEI production and antibiotic resistance, as well as a correlation between
seb gene presence and resistance to antibiotics in MRSA strains isolated from a clinical sample in a hospital in Tehran. In our study, an interesting fact was the detection of antibiotic resistance associated with the
sed gene.
The association between the distribution of genes encoding virulence factors and the development of a range of staphylococcal infections is still unknown. Moreover, infections caused by multi-resistant S. aureus strains remain a challenge for clinicians due to the limited treatment options in patients with such infections. The present research investigated drug susceptibility and virulence patterns in S. aureus strains isolated from chronic wounds and blood. The limitation of the study was the inability to assess the expression of investigated virulence genes and perform molecular typing of the isolates, which would allow for a better interpretation of our findings. Often, antibiotic susceptibility pattern, virulence factors, and their combinations are associated with particular genetic backgrounds (clones) that carry specific pathogenicity islands or plasmids. Although those limitations exist, our results allow us to suggest that strains obtained from chronic wounds seem to be more often resistant to antibiotics and more virulent compared to strains isolated from blood. We observed a statistically significant difference in the incidence of sea and sei genes, and the percentage of strains resistant to amikacin, gentamicin and tobramycin. Further studies on the pathogenesis of S. aureus, including molecular typing, are required to elucidate the importance of various virulence factors in strains from patients with SAB and chronic non-healing wounds.