Staphylococcus aureus Clinical Isolates: Antibiotic Susceptibility, Molecular Characteristics, and Ability to Form Biofilm

Periodic monitoring of Staphylococcus aureus characteristics in a locality is imperative as their drug-resistant variants cause treatment problem. In this study, antibiograms, prevalence of toxin genes (sea-see, seg-ser, seu, tsst-1, eta, etb, and etd), PFGE types, accessory gene regulator (agr) groups, and ability to form biofilm of 92 S. aureus Thailand clinical isolates were investigated. They were classified into 10 drug groups: groups 1–7 (56 isolates) were methicillin resistant (MRSA) and 8–10 (36 isolates) were methicillin sensitive (MSSA). One isolate did not have any toxin gene, 4 isolates carried one toxin gene (seq), and 87 isolates had two or more toxin genes. No isolate had see, etb, or tsst-1; six isolates had eta or etd. Combined seg-sei-sem-sen-seo of the highly prevalent egc locus was 26.1%. The seb, sec, sel, seu, and eta associated significantly with MSSA; sek was more in MRSA. The sek-seq association was 52.17% while combined sed-sej was not found. Twenty-three PFGE types were revealed, no association of toxin genes with PFGE types. All four agr groups were present; agr group 1 was predominant (58.70%) but agr group 2 strains carried more toxin genes and were more frequent toxin producers. Biofilm formation was found in 72.83% of the isolates but there was no association with antibiograms. This study provides insight information on molecular and phenotypic markers of Thailand S. aureus clinical isolates which should be useful for future active surveillance that aimed to control a spread of existing antimicrobial resistant bacteria and early recognition of a newly emerged variant.


Introduction
Staphylococcus aureus, a gram positive coccal bacterium, is either commensal that colonizes healthy nasal mucosa [1] or pathogen of humans. As a pathogen, the bacteria cause a variety of community and hospital acquired diseases including skin abscess [2], food poisoning [3], pneumonitis [4], sepsis [5], and toxic shock syndrome [6]. This bacterium produces several virulent factors including adhesins (colonization factors), toxic proteins/enzymes (e.g., DNase for bacterial spread, coagulase, and catalase for host immunity evasion) and exotoxins including exfoliative toxins (ExTs), staphylococcal enterotoxins (SEs), and toxic shock syndrome toxin-1 (TSST-1). Patients infected with the ExT producing S. aureus may develop scalded-skin syndrome [7]. The SEs and TSST-1, besides causing food poisoning, are also superantigens (SAg) that can stimulate a relatively large fraction of peripheral blood T cells to release massive amounts of proinflammatory

Bacterial Strains.
Ninety-two strains of S. aureus isolated from clinical specimens were obtained from three hospitals. They were 43 strains (S1-S43) isolated in 2007 from patients of Prince of Songkla University Teaching Hospital and kept at the Department of Microbiology, Faculty of Science, Prince of Songkla University, Songkhla province, southern Thailand; 36 strains (P1-P36) from the patients of Prasat Neurological Institute, Bangkok, in 2010, and 13 strains (T1-T13) isolated in 2010 from patients of the Hospital for Tropical Diseases, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand. The bacteria were reconfirmed by Gram staining, biochemical testing (catalase, coagulase, and DNase), and mannitol fermentation. Their ability to produce protein A was detected by agglutination assay.

Antimicrobial Susceptibility
Testing. Disc diffusion method was used for antimicrobial susceptibility testing of the S. aureus isolates which was done according to CLSI guidelines [24]. Antibiotic discs were cefoxitin, ciprofloxacin, clindamycin, erythromycin, gentamycin, oxacillin, penicillin G, rifampin, tetracycline, sulfamethoxazole plus trimethoprim, and teicoplanin (Oxoid, UK). Cefoxitin disc (30 g) and oxacillin disc (1 g) were used for detecting methicillinresistant isolates. S. aureus ATCC 25923 was used as control. Reduction of vancomycin susceptibility of the isolates was also determined by observing the minimum inhibitory concentration (MIC) by agar dilution according to the CLSI guidelines [24].

Detection of Genes Coding for Staphylococcal Enterotoxins,
TSST-1, and ExTs. Genomic DNA was extracted from each S. aureus isolate by DNA extraction kit (Geneaid, Taiwan) following the protocol for Gram-positive bacteria. Quality of each DNA preparation was assessed by determining the ratio of OD 260 nm /OD 280 nm . Twenty-two virulence genes were amplified including sea-see, seg-ser and seu, tsst-1 and eta, etb and etd, using specific oligonucleotide primer sequences listed in Table 1 [25,26]. The PCR reaction mixture (25 L) is composed of 1 mM of each primer, 1x Taq buffer PCR, 0.2 mM dNTP, 2 mM MgCl 2 , 1 unit of Taq DNA polymerase (Fermentas, Germany), and 100 ng of DNA template. The PCR reaction mixture was subjected to the thermal cycles: an initial denaturation of DNA at 95 ∘ C for 10 min prior to 35 cycles of denaturation at 95 ∘ C for 30 sec, 55 ∘ C for 30 sec, and 72 ∘ C for 30 sec, followed by a final extension of 10 min at 72 ∘ C using the Lifecycler (BioRad, USA). The amplified products were analyzed by 1.5% agarose gel electrophoresis and ethidium bromide staining. The DNA bands were observed under an UV transilluminator (Syngene, England). Control bacteria for the PCR included strains ATCC 19095 (sea, sec, seh, seg, sei, sel, sem, sen, seo, seu, and tst), ATCC 14458 (seb and sek), ATCC 23235 (sed, sej), and ATCC 27664 (see, seq, and sea). For eta, etb, and etd, the PCR amplicons were verified by DNA sequencing and the nucleotide sequences were aligned with the staphylococcal eta, etb, and etd sequences of the database (accession numbers: L25372.1, M17348.1, and AB057421.1, resp.).

Biofilm Formation.
Ability of the S. aureus isolates to form biofilm was determined according to the protocol described previously [28] with modification. Individual bacterial isolates were cultured in TSB (Oxoid) supplemented with 0.25% glucose at 35 ∘ C until the turbidity reached McFarland no. 0.5. Approximately 100 cfu of each culture were applied in triplicate into wells of 96-well flat-bottomed microplate containing 200 L of the TSB and 0.25% glucose. Wells added with cultured S. epidermidis (ATCC12228) served as negative controls. The plate was incubated for 24 h. The content of each well was then discarded and the wells were washed five times with sterile 0.9% NaCl solution. Each well surface was stained by adding 100 L of 0.3% (w/v) crystal violet (Merck) in water and kept for 5 min. After five washing with sterile distilled water and air dried. The biofilm fixed on each well surface was extracted with 100 L of 70% ethanol and measured the absorbance at OD 570 nm . The isolates with OD 570 nm values above the mean OD 570 nm values plus three standard deviations of the negative control (mean neg + 3 SD Neg ) were considered positive for biofilm formation.

Statistical
Analyses. SPSS Statistics 16.0 was used for statistical analysis. Chi-squared ( 2 ) test and -test were used to analyze the data sorted by MRSA and MSSA groups and frequencies of virulence genes and biofilm formation, respectively. A probability value ( ) < 0.05 was considered different significantly.  Table 2.

Discussion
Diseases caused by S. aureus are health hazard to human worldwide. Since the first recognition of methicillin-resistant S. aureus in 1961 [29], there has been an upsurge of infections caused by the S. aureus variants that resist not only methicillin, but also other -lactams and vancomycin, which are therapeutic drugs of choice [30][31][32], leading to treatment failure and increased case fatality rate. The methicillin and vancomycin resistance of the S. aureus are encoded by staphylococcal cassette chromosome mec (SCCmec) and vanA, respectively [30,31]. Association of the presence of S. aureus toxin genes with methicillin sensitivity and resistance among S. aureus has been reported previously [28,[33][34][35]. The association was found also in the present study; the prevalence of the seb, sec, sel, seu, and eta was associated significantly ( < 0.05) with the MSSA while sek was found more in MRSA.
The toxin genes carried by the 92 Thailand isolates varied from none to as many as 11 genes ( Table 2). Five of the S. aureus enterotoxin genes, that is, seg, sei, sem, sen, and seo, belonged to the highly prevalent egc locus [36,37]; thus, their coexistence was frequently reported. Coexistence of segsei in the same strain, either alone or in more combination with other toxin gene(s) (sea, sec, sed, seh, sej, and/or tst) was found in 55% of the 429 S. aureus isolates from Germany [38]. In Japan, the seg-sei alone or with seb, sec, or sed were 24, 2.7, 6.8, and 2.0%, respectively [39]. The combined seg-seisem-sen-seo with seu was 15.1% among the Chinese isolates [26]. In the present study, the combined seg-sei-sem-sen-seo with other toxin genes including sea, seb, sed, sej, sek, sel, sep, seq, ser, and/or eta was found in 24/92 isolates (26.1%). There were 3 isolates that carried seg-sei-sen-seo with sea, sec, sek, sel, and/or seq and 1 isolate with seg-sei-sem-sen and seb. The previously reported fixed association of sed-sej [38] was not found among the 92 Thailand isolates. The combined sek-seq with other toxin gene(s), that is, sea and/or seb, was 45.5% among the Chinese isolates [26]. In the present study, the sekseq association was found in 48 of the 92 isolates (52.17%), either the two genes alone (16.3%) or with the other toxin genes (35.86%).
The ability of the isolates to produce SEA, SEB, SEC, and SED and ETA, ETB, and TSST-1 was examined by using SET-RPLA, TST-RPLA, and EXT-RPLA test kits, respectively. Not all isolates harboring the genes expressed the respective toxins. The results were similar to the finding reported previously among S. aureus isolates from milk and milk products from Morocco [40]. The unconformed results between genotypes (by PCR) to phenotypes (by RPLA) could be due to the fact that toxin production of the bacteria can be affected by the growth conditions including temperature, pH, and water activity. The so-produced toxin levels might be lower than the detection limits of the immunoassay [40,41]. Alternatively, the toxin gene may not be expressed due to mutation either in the coding region or in a regulatory region, for example, agr [42,43]. No annotated data are available in the literature on association of the ability of toxin production and antibiograms of the S. aureus. Nevertheless, in this study, the frequency of toxin production is higher among the MRSA (48.86%) than the MSSA (30.55%) ( < 0.05).
There was no association between PFGE patterns with the MRSA and MSSA of the 92 Thai strains which conformed to the results reported elsewhere [44,45]. However, PFGE patterns 21 and 22 of MRSA strains predominated among isolates from Prince of Songkla Hospital and Prasat Neurological Institute, that is, 32.5 and 27.8%, respectively. Among the 7 isolates of PFGE pattern 21 of Songkla that could produce enterotoxins, 6 strains (85.7%) produced SEA. All 7 isolates of PFGE type 22 of Prasat Neurological Institute isolates produced SEB.
The polymorphism in the agr locus was first described by Ji et al. in 1997 [46]. To date, S. aureus isolates were classified into four different agr groups [25,46]. In this study, all agr groups were found; large proportion (58.6%) of the isolates was agr group 1 which was similar to the data reported previously [16]. Moreover, majority (38/54 isolates, 70%) of the agr group 1 were MRSA which conformed also to the previous report [47]. However, it is noteworthy that isolates of the agr group 2 in this study carried more number of enterotoxins genes, and most of the toxin producing strains belonged to this agr group. The data were different from elsewhere which showed that most toxin producing S. aureus strains were either agr groups 3 [46] or 4 [48].
Biofilm formation contributes to bacterial pathogenesis and resistance to antibiotics and harsh environment. S. aureus isolates did form biofilms [28,49]. More strains of MSSA produced biofilm compared to MRSA strains [28]. In this study, 72.83% of the S. aureus isolates formed biofilm but there was no association with their antibiotic patterns.
In conclusion, the results of this study provide insight information on molecular and phenotypic markers of S. aureus clinical isolates in Thailand which should be useful for future active surveillance that aimed to control a spread of existing antimicrobial resistant bacteria as well as early recognition of a newly emerged variant.