Molecular typing of methicillin and vancomycin-resistant
Staphylococcus aureus isolated from clinical specimens by doublelocus
sequence typing (DLST) method

Methicillin-resistant Staphylococcus aureus (MRSA) strains are the essential cause of infections in communities and hospitals. The present study was conducted to determine the molecular typing of MRSA, isolated from hospitalized patients, using the double-locus sequence typing (DLST). In total, 280 S. aureus isolated from clinical specimens by phenotypic (catalase, coagulase, DNase, oxacillin, vancomycin screening agar and antibiotic disk diffusion), and molecular methods (PCR for determining the mecA, vanA and nuc genes). The DLST and sequencing was performed for MRSA containing mecA. Out of 280 specimens, confirmed as Staphylococcus aureus (S. aureus), 123 (43.9%) strains were MRSA. The highest resistance toward the erythromycin (15 μg), followed by ciprofloxacin (5 μg), clindamycin (2 μg), tetracycline (30 μg), gentamicin (10 μg) and rifampicin (5 μg), was 98.3%, 97.5%, 94.3%, 90.2%, 83.7% and 41.4%, respectively. Also, the least resistance (0%) was observed in each of teicoplanin (30 μg), linzolide (30 μg), and vancomycin (30 μg). All (100%) of MRSA strains had the mecA, and none of them have had the vanA. The results of DLST showed that the most common sequence types were BPH 2003 and 0217. The DLST type 18-32 was a significant cluster of MRSA. By sequencing MRSA and comparing the dominant types via the DLST, it is possible to establish the etiology of the disease in a much shorter time, and prevent the complications of the disease. Therefore, the combination of partial sequences of clfB and spa can serve as useful genetic markers for MRSA typing. It concluded that the MRSA in our region was relatively high, but no vancomycin resistance was found. The majority of the MRSA DLST type was 18–32.


Introduction
Staphylococcus aureus (S. aureus) causes a wide range of diseases, such as infective endocarditis, osteomyelitis, food poisoning, septicemia, skin infections, boils, carbuncles, soft tissue infections, and scalded skin syndrome in humans (Shittu and Lin, 2006;Gordon and Lowy, 2008). Today, resistance to antibiotics is rising due to excessive consumption, causing concerns all around the world. Also, due to the resistance of this bacterium against antimicrobial agents, many antibiotics have been proposed for treatment. Staphylococci have recently received much scholarly attention owing to the emergence of penicillin-resistant strains, methicillin and vancomycin-resistant Staphylococcus aureus strains (MRSA and VRSA), and the combination of these strains with the epidemic clashes of severe hospital infections (Walsh and Howe, 2002;Carvalho et al., 2009). Penicillin-resistant strains were first isolated in hospitals in 1942, and afterwards in communities. In addition, S. aureus strains have also become methicillin and vancomycinresistant by the acquisition of genes such as the mecA and vanA. The MecA is located on a mobile genetic segment (mobile genome island) that is known as staphylococcal cassette chromosome mec (SCCmec). The drug resistance is caused by the acquisition and replacement of SCCmec elements within the chromosomes of susceptible strains. The 2.1 kb mecA is located between the locus of the protein A coding genes (spa), and the protein responsible for purine biosynthesis. The mecR and mecI regulatory genes and mecA are located in the mec operon. This operon is carried on the SCCmec. The mecA, sensitive to the presence of betalactamase, is regulated by the MecI repressor protein, and the MecRI membrane transducer signal protein (Elements IWGotCoSCC, 2009;Turlej et al., 2011). The prevalence of MRSA strains varies in cities, countries and even throughout the continents, due to the proliferation of the local clones of MRSA strains (Ko et al., 2005;Goering et al., 2008). It is important to study the mecA, vanA, clfB and spa (encoding PBP2a, ligase enzyme, ClfB and protein A, respectively) in MRSA (Kuhn et al., 2007;Turlej et al., 2011).
For many years prior to the 21st century, vancomycin had been used to treat the MRSA infections. However, in July 2002, the US Center for Disease Control and Prevention (CDC) released its first report on MRSA strain resistance to vancomycin. Reduced sensitivity to vancomycin is reported to be caused by changes in the cell wall due to the presence of vanA (Assadullah et al., 2003). Epidemiological studies are essential to determine the source of infection, recognize the transmission of cross-pathogens in hospitals and the prevalence of the disease, identify the major pathogenic strains, and to evaluate the effectiveness of control measures. The conventional epidemiological typing methods such as antibiograms, biotyping, serotyping and phage typing have been sometimes useful in describing the epidemiology of infectious diseases, but they reveal little variability. As a result, DNA-based typing methods are deemed necessary for the study of the epidemiology of most microbial pathogens (Van Belkum et al., 2007). Given the presence of S. aureus in the environment and in the normal flora of hospitalized patients, molecular typing studies are of great importance in controlling infections in communities and in hospitals. The epidemiology of S. aureus has been analyzed by a set of different molecular typing methods, such as pulsed-field gel electrophoresis (PFGE), multi-locus sequence typing (MLST) and gene sequencing of Staphylococcus Protein A (Spa). Among these, PFGE is considered as the golden standard method for its high differentiation power (van Belkum et al., 1998;Turlej et al., 2011). Recently, it has been shown that an epidemiological study on S. aureus is possible by sequencing small regions in only two highly variable loci (for the clfB and spa loci, respectively) (Kuhn et al., 2007;Senn et al., 2013). Considering the very limited number of studies in the world, using the double-locus sequence typing (DLST) technique, as only few studies are available in foreign countries Okon et al., 2009) and since no epidemiological study on S. aureus has so far been conducted in Iran, using the DLST technique, and the high prevalence of MRSA in different parts of the world, it is necessary to determine the kinds of clones and MRSA types to be able to implement the strain control strategies, both at the hospital and the community levels. Therefore, the aim of this study was to determine the MRSA and VRSA relative prevalence, antibiotic susceptibility and the molecular typing of S. aureus strains with methicillin-resistant mecA, isolated from the hospitalized patients, admitted to university hospitals in the city of Ahvaz, Khuzestan province, Iran, using the DLST technique.

Materials and Methods
Clinical specimens. During the 6 months (from October to March 2016), all Staphylococci isolated at the microbial division of laboratories of university hospitals (Razi, Taleghani, Imam and Golestan) in Khuzestan province, Iran, were included in our study. A total of 280 isolates of S. aureus, out of 693 Staphylococci specimens from clinical sources, were analyzed. The specimens were obtained from tracheal secretions, blood, wound, urine, and the tracheal tube of patients, collected from different wards of the hospital, including: Women Surgery, Men Surgery, Intensive Care Unit (ICU), Women Internal, Men Internal, Out patient Department (OPD), Orthopedic, Infectious, Dermatology and Adult Nephrology Departments. Samples were transported to the microbiology laboratory of the School of Medicine at Ahvaz Jundishapur University of Medical Sciences (AJUMS).
Phenotypic methods for isolation of S. aureus After the collection of the isolates as Staphylococci bacteria, they were identified and confirmed as S. aureus, using standard microbiological tests, including Gram-staining, catalase, coagulase, DNase and mannitol salt agar fermentation (McClure et al., 2006).
Antibacterial susceptibility pattern of S. aureus (Wayne, 2014) An antibiotic sensitivity test was performed, using the Kirby Bauer disk diffusion method (all disks were provided by MAST, UK).
The tests for isolation of MRSA strains 1. Oxacillin disk diffusion test. The isolation of the MRSA strains, using oxacillin disk (1 μg), and according to the standards of the Clinical and Laboratory Standards Institute (Wayne, 2014). 2. Identifying methicillin-resistant strains, using cefoxitin.
To identify the MRSA strains based on the cefoxitin disk (30 μg), Mueller-Hinton agar was used according to the CLSI guideline; the resistant strains to cefoxitin (30 μg) were considered as MRSA (Wayne, 2014). 3. Oxacillin agar screening. Mueller-Hinton agar (MHA) plates, containing 4% NaCl and 6 µg/mL of oxacillin were prepared. The plates were inoculated with 10 µL of the 0.5 McFarland turbidity standard suspension of the isolate by streaking in one quadrant, and incubated at 35°C for 24 h. Then, the plates were observed carefully by transmitted light, any growth after 24 h was considered as oxacillin resistant (Wayne, 2011).

Vancomycin agar screening
Based on the CLSI guideline, Brain Heart Infusion agar (BHI), containing 4% NaCl and 6 µg/mL of vancomycin powder (Sigma) was prepared. A 10 µl inoculum of 0.5 McFarland suspensions of each specimen spotted into the medium (Wayne, 2011).

Molecular methods
DNA extraction by the kit (Sinaclon, Iran). The colonies, grown after 24 h were taken, and the DNA extraction was performed according to the manufacturer's protocol (Sinaclon Co., Tehran, Iran). Lysis buffer (400 µl) was added to the sample and vortexed at max speed for 20 seconds, then 300 μl precipitation solution was added and vortexed at max speed for 5 s. Subsequently, the solution was transferred to a spin column with collection tube, and centrifuged for 1 min. Then, wash buffer (400 μl) was added to the spin column, and centrifuged for 1 min. Afterwards, the spin column was washed with 400 μl of wash buffer II by centrifugation for 1 min (twice), and the spin column was placed in the collection tube, centrifuged for 1 min, and carefully the column was transferred to a new 2 ml tube, and 30 μl of 65°C pre-heated elution buffer was placed in the center of the column, and incubated for 3-5 min at 65°C. Then, the DNA was eluted after 1 min centrifugation.
The reconfirmation of S. aureus by molecular method. S. aureus isolates, identified and confirmed by culture and phenotypic methods, were reconfirmed by amplifying nuclease gene (nuc). The primer sequences for the nuc have been shown in Tab. 1.
PCR reaction for the nuc. The PCR was performed in the final volume of 20 μl, consisting of 2 μl of DNA template, 2 μl of PCR buffer (1x), 1.5 mM of MgCl 2 (50 mM), 0.2 μl of dNTPs (10 mM), 0.4 μM of each nuc1 and nuc2 primers (10 pmol), 1U of Taq DNA polymerase (5 u/μL) and 13.2 μl of double distilled water. PCR amplifications were done in a thermocycler (Eppendorf, Germany), under the subsequent cycle circumstances: an initial denaturation step at 94°C for 5 min; 30 cycles of denaturation at 94°C for 1 min, annealing at 50°C for 1 min, extension at 72°C for 2 min, and a final extension step at 72°C for 10 min.
Identification of the mecA and vanA in MRSA specimens by PCR method. The primer sequences for the mecA and vanA have been shown in Tab. 1. The composition of the reagents in master mix and the temperature has been indicated under the PCR reaction for nuc, annealing temperature for mecA and vanA were 52°C and 56°C, respectively. The PCR was performed on the thermocycler (Eppendorf, Germany).
Electrophoresis of the PCR Products. The PCR products were electrophoresed on 1.5% agarose gel, and then the gels were stained with 5 µg/mL ethidium bromide. The standard strains ATCC 33591, 25923 and 52199 were used as the positive control for the mecA, nuc and vanA, respectively, and distilled water as the negative control for all PCR reaction (Kalorey et al., 2007;Japoni et al., 2011;Havaei et al., 2012).
The DLST for isolates with mecA. Methicillin-resistant strains were examined, using the DLST. After DNA extraction, PCR was performed on two highly variable loci of clfB and spa at 800-1300 bp and 1000-1500 bp, respectively. The primers have been listed in the following Table. The standard electrophoresis and gel staining with ethidium bromide were utilized (Sinaclon, Iran).
In this study, bioinformatics software programs such as MEGA 06 (with UPGMA method, based on the Dice coefficient), Bioedite and Gene runner were used to analyze the DNA sequences and determine the homology of the sequences. Inconcise, the extracted DNAs of MRSA were used for the PCR amplification, using specific primers. The standard gel electrophoresis was applied, and the gels were stained with ethidium bromide (Sinaclon, Iran). Electrophoresis was carried out on 1.2% agarose gel, at 85 V for 1 h. The length of the PCR product was variable among the isolates. The PCR products were purified and sequenced by the Bioneer Corporation (Bioneer, Daejeon, South Korea). The nucleotide sequences were analyzed by BLAST, and then saved as a FASTA format for sequence alignment and phylogenetic purposes. The S. aureus DLST website (http://www.dlst.org/S.aureus/) was used to assign numbers to each distinct allele within a locus, after submitting good quality trace file sequences for each of clfB and spa loci per each tested isolates.

Results
According to phenotypic methods, our results were indicative of the fact that a total of 693 specimens was collected as staphylococci, 280 specimens were confirmed as S. aureus, and the rest were coagulase negative staphylococci. According to the oxacillin disk (1 μg) diffusion test, 121 strains of S. aureus were identified as MRSA. Also, based on the cefoxitin disk (30 μg) diffusion, 123 isolates of S. aureus were identified as MRSA. According to the oxacillin agar screening test, 122 strains were MRSA. The number and frequency of MRSA isolates, according to the type of   The highest number of isolated MRSA 51 (41.5%) was observed at Taleghani Hospital; also, ICU wards accounted for the majority of isolated MRSAs. Based on the antibiotic resistance pattern, the results revealed that erythromycin (98.3%) accounted for the highest resistance of MRSAs isolated. The minor resistances were belonged to teicoplanin (0%), linzolide (0%) and vancomycin (0%). Based on the vancomycin agar screening test, all of MRSA strains were sensitive to vancomycin. In this study, the 280 specimens identified as S. aureus, using the phenotypic method were reconfirmed as S. aureus by nuc gene as mentioned below (Fig. 1). Furthermore, 100% of all 123 (43.9%) MRSA strains, identified by the cefoxcitin disk diffusion method showed the presence of mecA by the molecular examination (Fig. 2). The sensitivity for oxacillin disk diffusion and oxacillin screening agar was 98% and 99%, respectively. However, the sensitivity and specificity of cefoxitin and PCR were 100%; it was also demonstrated that the specificity of all four methods was 100%. In addition, 30 MRSA strains (24.4%) were selected randomly for DLST. The sequences were submitted to the DLST website for allele assignment. The allele profiles were compared and clustered, using an online data analysis service. The results of the DLST showed a high similarity with the results obtained from the MLST method. The most common sequence types were BPH 2003 and 0217. The type 18-32 was a significant cluster of MRSA isolates, using DLST. The DLST showed a great typeability and discriminatory power.
The evaluation of the electrophoresis induced by the amplification of clfB and Spa, illustrated the presence of bands with a length of 1200 bp and 1100 bp, respectively (Figs. 3 and 4). The isolate, sample/ward, DLST types and antibiotic pattern of MRSA isolates have been presented in Tab. 3. Also, the phylogenetic relationships of MRSA isolates have been indicated in Fig. 5.

Discussion
S. aureus is one of the most important human pathogen, causing a wide range of infections. Furthermore, the prevalence of MRSA strains, particularly in hospital infections, is a global clinical challenge. Many epidemiological studies revealed that more than 50% of S. aureus infections are caused by MRSA strains (Stefani and Varaldo, 2003;Schito, 2006). The appearance of methicillinresistant isolates in MRSA, carrying SCCmec with multiple drug-resistant genes, is problematic in the treatment of S. aureus infections.
Methicillin resistance in MRSA strains is caused by mecA acquisition (Deurenberg and Stobberingh, 2008). The mecA is located on a mobile genomic island called the staphylococcal cassette chromosome mec (SCCmec). So far, 11 major SCCmec types (I-XI types) have been identified. The SCCmec elements are one of the unique genomic islands that contain the ccr (Cassette chromosome recombinase) gene complex and the mec gene complex. All MRSA strains contain SCCmec elements that are located in a specific locus of S. aureus chromosomes called attBscc (Bacterial chromosomal attachment site) (Turlej et al., 2011;Hiramatsu et al., 2013). The present study was shown that the mean prevalence of S. aureus and MRSA were 280 (40.94%) and 123 (43.9%), respectively. The prevalence rates of MRSA were 71% in Egypt and 26.9% in South Africa (El Kholy et al., 2003;Shittu and Lin, 2006). Therefore, the average outbreak is lower than that of hospitals in Egypt and higher, compared with South African hospitals. The highest prevalence was observed at Taleghani Hospital 51(41.5%) and the highest number of isolated MRSAs was seen in ICU wards (33.3%). It is worth mentioning that the frequency of MRSA in the Scandinavian countries is less than 2% and more than 40% in the Mediterranean countries (Stefani and Varaldo, 2003).
The results of the current study indicated that no resistance of MRSAs to teicoplanin (0%), linzolide (0%), and vancomycin (0%) was seen. The resistance to vancomycin was reported 12% in the north-west of Iran, 3.7% and 15% in other countries ( (Fig. 4). Electrophoresis was carry out using 1.2% agarose gel at 85 V for 1 h. Lane M100 bp DNA ladder, lane 1 Positive control, lane 2 Negative control, lane 3 to 12 Spa gene. Ali et al., 2014). In another study, out of 783 S. aureus strains, two strains were found to be vancomycin and teicoplanin resistant (Tiwari and Sen, 2006). In another report, 12 patients with resistance to linzolide (LRSA) were identified (García et al., 2010). Hospitalization and prolonged use of antibiotics can cause colonization and infection with the MRSA strains. The infection can be transmitted from patient to patient, the staff to patient, or from the environment to patient (Hiramatsu et al., 2001). Recognition, classification and evaluation of the phylogenetic correlation among MRSA strains are essential for epidemiological researches. The DLST design was effectively used for genotyping; and, the local epidemiology of the MRSA isolates in the Khuzestan province were explored. Hence, this study revealed an important role in disease control and identification as well. Recently, several patterns based on the variable number tandem repeat typing (e.g., MLVA, SCCmec and MLST) has been described in S. aureus, but all of these methods are costly and timeconsuming. PFGE is also considered as the golden standard method for epidemiological studies, due to its high differentiation power. It is one of the most important molecular methods for epidemiological studies and typing of S. aureus (van Belkum et al., 1998).
The main problem with this method and all other methods based on comparing patterns of DNA fragments, is the difficulty to compare the results obtained in the laboratory (Narukawa et al., 2009). In contrast, sequencebased methods, which are common have definitive results and provide a good repeatability. Also, several typographic designs have become available, which are based on the analysis of regions, having a variable repetitive sequence and the sequences of nucleic acids, including MLST and Spa sequencing. Such designs are used for epidemiological studies, as well as comparisons of S. aureus strains in a geographical region with strains of other parts of the world. Each isolate is characterized by a number of repetitive sequences, observed in many loci. Having multiple and repetitive sequences, complicates the typing. In addition, the MLST is less suitable for the study of the nuances of strains. Although Spa classification better reveals such small genomic differences, it is less likely to be implemented due to genomic recombination events. The MLST uses the nucleotide data of some seven housekeeping genes. Despite the fact that this method is useful for understanding the structure of the overall population of S. aureus, it is both costly and timeconsuming; that is why it is not very practical. Furthermore, it has a low differentiation power for epidemiological investigations, and is used for longitudinal studies as well as phylogeny studies and the evolutionary process of species (Robinson and Enright, 2004;Kuhn et al., 2007). Therefore, the DLST method was performed based on small regions of the 3' sequence. The DLST for S. aureus involves multiplication and sequencing, and then analyzing two highly variable gene loci, clfB and spa (Kuhn et al., 2007;Senn et al., 2013). In 2014, this technique was first introduced, using three variable loci for Pseudomonas aeruginosa. In 2015, this technique was designed, using two highly variable loci (Basset and Blanc, 2014). In recent reviews, the comparison of the DLST with MLST and PFGE has shown that the differentiation power indexes of DLST with regard to S. aureus are almost identical to those of MLST, and both techniques are capable of detecting highrisk epidemic clones.
The results of the DLST showed great typeability and discriminatory power (ID = 0.91), the results of this study are consistent with those reported in other countries (Kuhn et al., 2007;Basset et al., 2009). The most common sequence types were BPH 2003 and 0217. The DLST type, 18-32 was a significant cluster (10/30; 33%) of MRSA isolates. In this study, the DLST results showed that these strains were originated from the hospital. Drug-resistant isolates confirmed that patients were in relatively high antibiotic resistance risk. The S. aureus strains, which have been secluded from diverse geographical and infection origins, have illustrated a variety in their DLST types. It can be due to their variety, adaptive capability to diverse geographical niches and illness causes.
In 2010, Basset et al. conducted the DLST technique based on the sequencing of two genes, clfB and spa on 1242 MRSA isolates, isolated from Western Switzerland. Then, they compared them with a collection of isolates typed by PFGE. The results showed 88% of congruence between DLST and PFGE clones. Therefore, the DLST can be included in a routine monitoring plan. In this study, the alleles were classified as DLST-types with clfB and spa alleles. All isolates derived from each DLST cluster belonged to two clonal complexes (cc). Thus, these two genes are good markers for epidemiological studies. The MRSA isolates were studied by Kuhn et al., (2017) and it was shown that the sequencing of small areas, around 500 bp of only two highly variable clfB and spa loci, allowed studying this pathogen epidemiologically. It has also been shown that unlike the PFGE method, this technique has a high stability, and yields the similar results over time (over 3 years). Also, (Vogel et al., 2012) reported a gradual transition of ST228 clone to MRSA with a high transmission capability in the hospital care units in Switzerland. They determined the Finally, it was determined that with the development of new sequencing technologies, qualitative and quantitative information is obtained from a single strain, but more data are needed for their interpretation and important investments are required in technology and routine research. Therefore, using this new method proved to be a standard unambiguous typing of S. aureus strains. Moreover, this technique has a high power in determining new types. The results of this study are consistent with those reported in other countries Huang et al., 2017). This method is easy to accomplish and helps to reduce costs and save time. Hence, this method can be used as a scientific and practical approach to be easily applied in monitoring programs, including the typology of hundreds of isolates. In addition, this method proves high typeability, differentiation and repeatability. Unambiguous identification of types makes this method useful and effective in epidemiological studies. Also, this method yielded results with a high degree of similarity with the results of the MLST method, while being more practical and requiring much lower cost and time. Therefore, the DLST method was proposed as a valuable typing tool in epidemiological surveillance (Kuhn et al., 2007;Basset et al., 2010). Infection-inducing clones are limited in number and belonged to a number of pandemic strains. Therefore, in order to prevent the spread of infection with multiple strains, appropriate measures regarding the consumption of antibiotics and the infection controls should be taken. This study can be a step forward in identifying and implementing a useful, practical, and scientific molecular typing method for epidemiological studies in Iran, for S. aureus. The results of this study illustrated that the prevalence of MRSAs in Ahvaz hospitals was relatively high, but these strains were totally sensitive to vancomycin, teicoplanin and linzolide antibiotics. The results of the DLST showed that the most common sequence types were BPH 2003 and 0217. The DLST type 18-32 was the significant cluster of MRSA. Therefore, the combination of partial sequences of clfB and spa can serve as useful genetic markers for the MRSA typing. By sequencing MRSA and comparing the dominant types via the DLST, it is possible to establish the etiology of the disease in a much shorter time and prevent the complications of the disease.
Funding Statement: The authors received no specific funding for this study.
Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the present study.