Detection of sasX Gene and Distribution of SCCmec Types in Invasive and Non-invasive Coagulase-negative Staphylococci

Background: Coagulase-negative staphylococci, which belong to the normal microbiota of the skin and mucous membranes, are opportunistic pathogens. sasX, a newly described protein, is thought to play an important role in nasal colonization and methicillin-resistant Staphylococcus aureus virulence, and it may be acquired from coagulase-negative staphylococci by horizontal gene transfer. It has been considered that understanding the function of sasX gene may help clarify the relevance of the different adhesion mechanisms in the pathogenesis of infections associated with biofilm. Aims: To investigate the sasX gene presence, staphylococcal cassette chromosome mec types, and antimicrobial resistance patterns of invasive and noninvasive coagulase-negative staphylococci isolates. Study Design: Cross-sectional study. Methods: The study included a total of 180 coagulase-negative staphylococci strains. Non-invasive isolates (n=91) were obtained from the hands of healthy volunteers who do not work at the hospital (n=30), the nasal vestibule of healthy volunteer hospital workers (n=26), and central venous catheter (n=35). Invasive isolates (n=89) were isolated from peripheral blood cultures of inpatients who do not have catheters. All isolates were identified by conventional microbiological methods, automated systems, and, if needed, with matrix-assisted laser desorption/ionization-time of flight. Staphylococcal cassette chromosome mec typing, sasX and mec gene detection, antibiotic susceptibility, and sasX gene sequence analysis were performed. Results: Peripheral blood, central venous catheter colonization, and nasal vestibule isolates were positive for the sasX gene, whereas hand isolates were negative. sasX gene was present in 17 isolates, and no statistical significance was found between invasive and noninvasive isolates (p=0.173). Sequence analysis of the sasX genes showed high homology to related proteins of Staphylococcus phage SPbeta-like and Staphylococcus epidermidis RP62A. staphylococcal cassette chromosome mec type V was the most prevalent regardless of species. staphylococcal cassette chromosome mec type II was more frequent in invasive isolates and found to be statistically important for invasive and noninvasive S. epidermidis isolates (p=0.029). Staphylococcus haemolyticus isolates had the overall highest resistance rates. Resistance to ciprofloxacin, trimethoprim-sulfamethoxazole, and erythromycin was found to be higher in isolates from catheter and blood culture. Staphylococcus hominis isolates had the highest rate for inducible clindamycin resistance. None of the isolates were resistant to vancomycin, teicoplanin, and linezolid. Conclusion: The sasX gene is detected in 9.44% of the isolates. There is no statistical difference between the sasX-positive and -negative isolates in terms of antibacterial resistance and the presence of sasX and SCCmec types. Further studies about the role of sasX at virulence in coagulase-negative staphylococci, especially from clinical samples such as tracheal aspirate and abscess isolates, and distribution of staphylococcal cassette chromosome mec types are needed.

Coagulase-negative staphylococci (CoNS), which belong to the normal microbiota of the skin and mucous membranes, are opportunistic pathogens, particularly in patients who are immunocompromised or have indwelling devices (1). Especially, Staphylococcus epidermidis and Staphylococcus haemolyticus are considered to be the major nosocomial pathogens, as well as others like Staphylococcus saprophyticus (2,3). These bacteria are also the cause of bacteremia in nearly 30% of bloodstream-related infections and responsible for sepsis episodes in immunocompromised patients (4).
Resistance to methicillin and almost all beta-lactam antibiotics in staphylococci is mostly caused by the acquisition of the mecA gene, encoding an additional altered low-affinity penicillin-binding protein, which is located on a mobile genetic island, termed the staphylococcal cassette chromosome mec (SCCmec). SCCmec contains mec gene complex and ccr gene complex of varying size and genetic structure (5). There is strong evidence suggesting the diversity of SCCmec in CoNS. New combinations of a mec complex and ccr complex, differences from the amplification of the ccr gene, and multiple presence of ccr gene complex make the typing difficult in CoNS (6)(7)(8)(9). sasX, a homolog of sesI, is a newly described protein that is thought to play an important role in nasal colonization and methicillinresistant Staphylococcus aureus (MRSA) virulence, and sasX gene is located on a ϕSPβ-like prophage (10). sesI is one of the S. epidermidis surface (ses) proteins that can be expressed during biofilm expression (11) and is encoded by ϕSPβ prophage (12). The sequence similarity of sesI and sasX genes indicates that the latter may be acquired from CoNS by horizontal gene transfer, and it has been considered that understanding the function of sasX gene may help clarify the relevance of the different adhesion mechanisms in the pathogenesis of infections associated with biofilm (13). This study aimed to investigate the sasX gene carriage, which was thought to play a role at attachment, pathogenicity, and distribution of the SCCmec types of invasive and noninvasive CoNS.

Bacterial strains and identification
A total of 180 CoNS strains from İbni Sina Hospital Central Microbiology Laboratory Biobank, which were previously approved by the ethical committee for clinical investigations of Ankara University School of Medicine -21-680-12), were included in the study. Noninvasive isolates (n=91) were obtained between 2013 and 2014 from the hands of healthy volunteers who do not work at the hospital without chronic diseases (n=30), noses of healthy volunteers (nurses, doctors, and hospital staff) who worked at the university hospital (902 beads) without chronic diseases (n=26), and central venous catheter colonization (n=35). Catheter colonization is defined as the growth of an organism, using the semiquantitative roll-plate culture technique greater than 15 colony-forming units from the tip or the subcutaneous segment of the removed catheter in patients who have no growth at their blood cultures, which were taken simultaneously (14). On the other hand, invasive isolates (n=89) were isolated from peripheral blood cultures of inpatients who do not have catheters. In our study, we defined isolates from sterile body sites, including blood (which were causative agents of bloodstream infections according to guidelines) as invasive and isolates from hands, nasal vestibules, and central venous catheter colonization as noninvasive. Isolates were identified by conventional microbiological methods (Gram staining, catalase, and coagulase tests), BD Phoenix (Becton Dickinson, USA) automated systems, and, if needed, Bruker Microflex MS (Bruker Daltonics, Bremen, Germany).

Sequence analysis of sasX gene
Sequencing of sasX genes was performed using ABI Prism 310 genetic analyzer (Applied Biosystems, Foster City, CA, USA) with BigDye fluorescent terminator chemistry. Sequence data obtained were analyzed with the basic local alignment search tool of the National Center for Biotechnology Information (18).

Statistical analysis
The Kruskal-Wallis variance analysis compared the vancomycin MIC values, and chi-square or Fisher's exact tests for all other comparisons, as appropriate. All analyses were performed using SPSS Statistics for Windows, Version 15.0. A p-value <0.05 was considered statistically significant.

Antimicrobial susceptibility patterns
Methicillin resistance was found in 135 of the isolates using the cefoxitin disk diffusion test. When overall antibiotic resistance was considered, S. haemolyticus isolates had the highest resistance rates except for clindamycin and tetracycline ( Table 2). Resistance to ciprofloxacin, trimethoprim-sulfamethoxazole, and erythromycin was found to be higher in isolates from catheter colonization (71.4%, 71.4%, and 80%, respectively) and blood culture (70.8%, 68.5%, and 85.5%, respectively). Considerable proportions of isolates from the nose were found to be susceptible to ciprofloxacin (96.2%), trimethoprim-sulfamethoxazole (96.2%), and rifampicin (96.2 %). S. hominis isolates had the highest inducible clindamycin resistance rates (36.9%), followed by S. capitis (23.07%), S. epidermidis (11.7%), and S. haemolyticus (8.3%). None of the isolates were found to be resistant to vancomycin, teicoplanin, and linezolid. The resistance patterns of the isolates are given in Table  2. Vancomycin MIC50, MIC90, and MIC ranges were 1, 4, and 0.125-4 μg/mL, respectively, by broth microdilution method, and the minimum, median, and maximum values of vancomycin for the CoNS spp. are given in Table 3. MIC ranges of the samples are given in Table 4. No difference was found between vancomycin MIC values and the isolates (p=0.477) and SCCmec types (p=0.277). No difference was found between sasX-positive and negative isolates in terms of antibiotic resistance patterns.

Presence of sasX and mecA genes
sasX gene was detected in 9.44% (17/180) of the isolates. Of the strains isolated, 12.36% (n=89) from peripheral blood, 11.42% (n=35) from catheter colonization, and 7.69% (n=26) from nasal vestibules were positive for the sasX gene, whereas strains from healthy hands (n=30) were negative. sasX carriage rate distribution and p-values according to the sample types are presented in Table  5.

SCCmec types
With the protocol of Kondo et al. (17), 130 of the 135 mecApositive isolates were typed. For all isolates, regardless of species, type V (34%) was the most prevalent, followed by type I (23.7%), type II (17%), type VI (10.4%), type III (4.4%), type I-II (1.5%), type I-III (1.5%), and type I-V (3.7%). Five of the isolates (3.7%) could not be typed. SCCmec distribution of the isolates is given in Table 6. When all invasive and noninvasive isolates were taken into consideration, type II was more frequent than the other types in invasive isolates and found to be statistically significant (p=0.014). Also, the type II ratio was statistically significant (p=0.029) between invasive and noninvasive S. epidermidis isolates.    TOTAL, n (%); p-value the nasal colonization, the possible role of this gene in healthcarerelated settings for nasal colonization should be investigated. When our findings are considered in respect of invasive and noninvasive isolates, the carriage of sasX is found in 12.36% of isolates from peripheral blood cultures, and 11.43%, 7.69%, and 0% of isolates from central venous catheter (catheter colonization), nasal vestibule, and hands, respectively. Our peripheral blood and nasal isolate results were in concordance with the results of Li et al. (10). The high sasX ratio in catheter colonizers seems to support the role of sasX in adherence to surfaces, as in nasal isolates. The small number of CoNS rather than S. epidermidis, S. haemolyticus, and S. hominis isolates and the absence of tracheal aspirate and abscess samples were the limitations of our study.
Also, Li et al. (21) stated that invasive infections caused by sasX-positive isolates were higher than sasX-negative isolates in accordance with our isolates from blood cultures. There was no sasX carriage in isolates from hands, and this finding also supports the results of Li et al. (10,20), which underlined the potential role of sasX in invasive rather than noninvasive isolates.    that the phages of CoNS rather than S. epidermidis isolates were also highly conserved.
In a study about MRSA ST239 isolates, Holden et al. (19) stated that the sasX gene is located at the 3′ end region of a ϕSPβ-like prophage, which had coding sequences for aminoglycoside resistance. In our study, no difference was found for gentamycin and other antimicrobial resistance between sasX-positive and -negative isolates in accordance with Li et al. (21). In our study, type V SCCmec (34%) was the most prevalent type in all isolates regardless of species, except S. hominis ( Table 6). The SCCmec types differ among CoNS species and between studies. McManus et al. (24) reported that type IVc is the most common type in methicillin-resistant S. epidermidis isolated from cerebrospinal fluid and external ventricular drains in deviceassociated meningitis patients. Garza-González et al. (25) found that types III and IVa were the predominant types in S. epidermidis isolated from blood samples, whereas Saffari et al. (26) and Chen et al. (27) found type IV and type III/SCCHg, respectively. Pinheiro et al. (28) characterized S. epidermidis (n=79) blood culture isolates and found SCCmec types III (53.2%) and II (29.1%) as the most prevalent types. The diversity of the types found in these studies may depend on the geographical differences, and the clinical samples from which the isolates were obtained, the epidemiologic similarity of the isolates, and the methodological limitations were used to detect SCCmec types in CoNS. In our study, no statistical difference was found between the presence of the sasX gene and the SCCmec type. In our study, with respect to invasive isolates, SCCmec type II was more frequent than the other types when compared with noninvasive isolates, and it was found to be statistically important (p=0.014). Also, the type II ratio was found to be statistically important (p=0.029) between invasive and noninvasive S. epidermidis isolates. In some studies, it is seen that there is no significant difference in invasive and noninvasive isolates in terms of SCCmec type (9,29). It is thought that this difference may be due to the number of isolates and diversity of sample origin (cat, pig, human isolates, etc.) used in the studies. Some CoNS isolates could not be typed in our study with the multiplex PCR method of Kondo et al. (17). As Garza-González et al. (30) also stated in their study, new methodologies are needed to be designed because of the inconsistent results for SCCmec typing of CoNS strains. The usage of two or more different multiplex PCR protocols/conditions (new primer designs for conserved gene sequences) directed to different targets of the SCCmec complex may be helpful to solve this issue. In this study, we characterized our invasive and noninvasive CoNS isolates from different samples of human origin and for the presence of sasX gene and SCCmec types. For sasX, we found 9.44% positivity. Remarkably, the sasX gene was found to be negative from strains isolated from healthy volunteers' hands, which indicate the possible role of this gene in invasive isolates.
No statistical difference was found between sasX-positive and -negative isolates in respect to antibacterial resistance and between sasX presence and SCCmec types. SCCmec type V is the most prevalent type of S. haemolyticus (47.7%) and S. epidermidis (35%) isolates. As far as we know, this is the first data about the distribution SCCmec types and sasX carriage of different CoNS from our country. As a result, more large scale and comprehensive studies are needed to define the SCCmec type distribution and to find out the molecular epidemiological characteristics of sasX and its possible role at colonization and infection in particular species. Also, more clinical samples of CoNS are needed.