Study Design and Bacterial Isolates
In this cross-sectional study, a total of 145 specimens including pus, exudates from lesions, and tissue biopsies were obtained during January 2017 to August 2017 from patients with Diabetic foot infection in the Diabetes Research Center, Endocrinology and Metabolism Clinical Sciences institute, Tehran, Iran. Samples were acquired by sterile swabbing from the ulcer base and tissue biopsies were obtained by scraping the ulcer with a sterile curette. Then S. aureus isolates were identified by standard biochemical and microbiological techniques (Gram’s stain, catalase, coagulase and DNase activities and mannitol fermentation on mannitol salt agar).
All participants gave written informed consent. The characterization and severity of DFU were assessed based on the Wagner Ulcer Classification System, which classified all DFUs in five major categories including a superficial diabetic ulcer, ulcer extension, deep ulcer with abscess or osteomyelitis, gangrene to a portion of forefoot and extensive gangrene of foot (6).
A clinico-demographic questionnaire was arranged for data collection. The questions were pertained to demographic information for each patient including age, sex, prior antibiotic usage (≤ 3 months), prior hospitalization, implantation of percutaneous medical device, experience of any surgery or prior residence in a long-term healthcare facility within the 6 months prior to the sampling date, the course of DM, the course of the ulcer, previous ulcers and amputation history, ulcer grade, HbAlC, lifestyle factors, presence of retinopathy, nephropathy, neuropathy, and peripheral vascular disease. Symptoms such as extensive gangrenous cellulitis in toes or in the rest of the foot, necrotizing fasciitis, sepsis, exudate formation and some amputation cases in which the infection persisted, were the inclusion criteria in this study.

DNA Extraction
Chromosomal DNA of MRSA strains were extracted using the Sinapure-DNA kit. Prior to DNA extraction, the isolates were sub-cultured in tryptic soy broth at 37°C for 24 h. Culture material were pelleted by centrifugation and were enzymatically lysed with resuspension in 100 L prelysis buffer and 20 L lyzosyme, and incubated for at least 30 min. At the lysis step, 10 L ributinase was added and temperature was increased to 55°C. After that a 30 min incubation procedure was performed according to instructions.

Detection of MRSA
Methicillin resistance was determined on Mueller–Hinton agar by using the 30 μg cefoxitin disk. S. aureus colony suspension, equivalent to 0.5 McFarland turbidity was inoculated on Mueller–Hinton agar and incubated at 33 to 35°C for 16 to 18 h. The interpretation of the results was performed according to Clinical and Laboratory Standards Institute (CLSI) guidelines. Resistance to methicillin was confirmed by the detection of the mecA gene by PCR.

Sccmec Typing by Multiplex PCR
Purified genomic DNA was used as the template for SCCmec typing, based on ccr and mec gene complex typing, without determining the differences in the junkyard region. Multiplex PCR was carried out using primers suggested by Jarraud et al. (7). Our PCR products were sequenced and aligned based on SCCmec sequenced reference strains including: type III (AB037671), type IV (AB063173) and type V (AB121219) (4).
PCR amplification was performed in a volume of 25 μL with SinaClon PCR Master Mix 2X.  The reaction mixtures consisted of: DNA template 2 μL, oligonucleotide primers (1-2 μL), 2× PCR buffer) 12.5 μL and H₂O to get a final reaction volume of 25 μL. PCR amplification was performed using SENCOQUEST labcycler and was programmed for identifying mecA and ccr genes as follows: initial denaturation at 94°C for 2 min, 35 cycles of denaturation (94°C for 2 min), annealing (60°C for 1:30 min), extension (72°C for 2 min) and a final elongation at 72°C for 2 min. PCR amplification for identifying mec gene complex classes was programmed as follows: initial denaturation at 94°C for 1 min, 30 cycles of denaturation (94°C for 1 min), annealing (60°C for 1 min), extension (72°C for 2 min) and a final elongation at 72°C for 2 min.
SCCmec types that could not be categorized with a set of primers and under any of these types were classified as untypeable (4).
Our nucleotide sequences were submitted to GenBank, which in turn provided GenBank accession numbers for our nucleotide sequences:
BankIt2083483 SCCmecA MG874129 to BankIt 2083483 luk MG874136.

Detection of Toxin Genes
Sequences specific for lukS-PV–lukF-PV, lukE, lukD encoding pvl components S and F; lukE, lukD respectively, were detected by PCR SENCOQUEST lab cycler with primers suggested by Jarraud et al. (7). These primer sequences correspond to 433bp of the pvl gene and 269bp of lukED gene. We used Staphylococcus aureus (GenBank accession number Y13225), and lukSI and lukF-I of Staphylococcus intermedius (GenBank accession number X79188) as a positive control. PCR amplification was performed as previously described (8).

Antibiotic Susceptibility Tests
The antibiotic susceptibility tests (ASTs) were performed by using the Kirby–Bauer disk diffusion method on Muller–Hinton agar plates and the e-test technique was used to determine the minimal inhibitory concentration (MIC) of vancomycin according to Clinical and Laboratory Standards Institute (CLSI). Susceptibility tests for gentamicin, doxycycline, ciprofloxacin, erythromycin, clindamycin, trimethoprim-sulfamethoxazole, chloramphenicol, rifampicin, linezolid, teicoplanin, mupirocin and vancomycin were performed in accordance with the CLSI guideline 2016. S. aureus strain ATCC 25923 was used as a control for susceptibility testing.
The control strain was stored at -70   and cultured on nutrient agar at 4-8 . Before testing, the strains were sub-cultured to agar plates. Any alteration in the mean zone diameters with control strains may be explained as a mutation or contamination. The zone diameters of our samples were within the standards of the control strain.
We also performed the standard control of antimicrobial discs for disc diffusion methods. Working stocks were kept below 8 and were protected from light. Further proceedings were done related to discs containing b- lactamase inhibitors or imipenem which are vulnerable to moisture.
S. aureus strain ATCC 25923 is commonly used as a control strain for susceptibility testing to antibiotics and as a quality control strain for commercial products.

Statistical Analysis
The results were analyzed by SPSS 23.0 (SPSS Inc., Chicago, IL., USA) using Fisher’s exact test, Pearson’s X 2 test and independent sample t test, where appropriate (demographic information, virulence factors and AST results). A P-value of <0.05 was considered to be statistically significant.

Characteristics of Patients Infected with MRSA and MSSA Strains
In 145 samples from infected DFUs, 83 (53.89%) were predominately aerobic gram-positive organisms, of which 71 (46.10%) were identified as S. aureus. Different stages of the study procedure were exhibited in the flow-chart (Figure 1). The demographic and clinical characteristics of patients with DFUs infected with S. aureus isolates are shown in Table 1. Statistical analysis showed that antibiotic usage (P=0.043) and prior hospitalization (P=0.003) had significantly affected MRSA isolation from diabetic patients. More than two-thirds of the patients with MRSA infections (25 from 30) had undergone recent antibiotic therapy (≤3 months), while patients with MSSA infections were currently undergoing antibiotic therapy, or did not undergo any antibiotic therapy at all.
 

Fig. 1. Flow of sample collection from patients with diabetic foot ulcer (DFU) through the study and methicillin-resistant detection of S. aureus strains. SCCmec types determination and toxins gene obtained on MRSA.

Table 1. Characteristics and demographic data of patients with DFUs infected with MRSA and MSSA strains

Virulence Profiles in Patients Infected with MRSA
Using PCR method for the detection of mecA gene in 71 S. aureus isolated from DFUs, 30 (42.25%) isolates were confirmed as MRSA and 41(57.74%) isolates were methicillin susceptible. Based on the SCCmec typing by multiplex-PCR assay, 14 (46.7%) strains belonged to SCCmec type IV, 9 (30.0%) strains to SCCmec type III, and 6 (20.0%) to SCCmec type V. One isolate (3.3%) was classified as untypeable and did not belong to any of the SCCmec types (Figures 2a and 2b).
The pvl and lukED genes encoding the bicomponent leukotoxin (LukS-LukF and LukE-LukD, respectively), were detected by PCR. Results indicated that 17 (56.7%) strains of MRSA isolates harbored pvl gene, 15 (88.2%) of which included CA-MRSA (P=0.023). Genes encoding for LukED were found in all 30 (100%) MRSA isolates. The pvl gene was often detected in strains isolated from ulcer grade 2 and 3 (P=0.026), whereas the lukED gene was found in all strains from ulcer grade 2-5. Statistical analyses revealed that PVL produced by MRSA does not associate with factors such as previous ulcer, ulcer duration, prior hospitalization, record of amputation, and antibiotic usage (Table 2)

Fig. 2. PCR products of SCCmec typing (agarose 1%): (a) Lane 1: MSM (1kb molecular size marker). Lane 2: Non-Template Control (NTC). Lane 3: Negative Control of mecA (MSSA). Lane 4: ccrA1 of ccr gene complex- Type I SCCmec (control strain). Lane 5: ccrA2 of ccr gene complex- Type II SCCmec (control strain). Lane 6: ccrA3 of ccr gene complex- Type III SCCmec (positive specimen of diabetic isolates). Lane 7: ccrA2 of ccr gene complex- Type IV SCCmec (positive specimen of diabetic isolates). Lane 8: ccrC of ccr gene complex- type V SCCmec (positive specimen of diabetic isolates). (b) Lane 1: MSM (1kb molecular size marker). Lane 2: Non-Template Control (NTC). Lane 3: Negeative Control of mecA (MSSA). Lane 4: class B mec gene complex- Type I SCCmec (control strain). Lane 5: class A mec gene complex - Type II SCCmec. Lane 6: class A mec gene complex - Type III SCCmec. Lane 7: class B mec gene complex- Type IV SCCmec. Lane 8: class C mec gene complex- Type V SCCmec.

Table 2. Molecular characteristics of MRSA isolates recovered from DFU infections

Antimicrobial Susceptibility Patterns in MRSA Strains
All MRSA strains were sensitive to linezolid, mupirocin and vancomycin. The highest resistance rate (100%) among the HA-MRSA isolates was seen for gentamicin, ciprofloxacin, erythromycin, clindamycin, and rifampin. However, the CA-MRSA isolates were more susceptible to gentamicin (100%), rifampin (90%), ciprofloxacin (60%) and doxycycline (55.0%). The sensitivity to doxycycline (P=0.013), ciprofloxacin (P=0.002) and rifampin (P<0.001) antibiotics was statistically significant between HA-MRSA and CA-MRSA. Both HA- and CA-MRSA isolates showed the same susceptibility pattern to the other study antibiotics. No vancomycin-resistant strain was found among MRSA isolates (Table 3).

Table 3. Antibiotic susceptibility pattern of HA-MRSA and CA-MRSA strains