Molecular characterization of methicillin-resistant Staphylococcus aureus among insulin-dependent diabetic individuals in Brazil

Background People with diabetes mellitus, especially insulin-dependent diabetic patients, are a risk group for staphylococcal infections. Asymptomatic infection with Staphylococcus aureus is common and favors dissemination of the microorganism, rendering these individuals a source of infection. This study aimed to characterize the resistance profile, clonal profile and sequence type, as well as to analyze the prevalence and risk factors for nasal and oropharyngeal carriage of methicillin-susceptible (MSSA) and methicillin-resistant S. aureus (MRSA) isolated from insulin-dependent diabetic individuals in the city of Botucatu, SP, Brazil. Methods Staphylococcus aureus was collected from the nasopharynx and oropharynx of 312 community-dwelling insulin-dependent diabetic individuals over a period of 3 years (October 2015 to December 2018). The isolates were characterized by susceptibility profiling, detection of the mecA gene, SCCmec typing, and molecular typing by PFGE and MLST. The risk factors associated with S. aureus and MRSA carriage were determined by logistic regression analysis. Results The overall prevalence of colonization with S. aureus and MRSA was 30.4% and 4.8%, respectively. Fifteen of the 112 S. aureus isolates carried the mecA gene; SCCmec type IV was identified in 10 isolates, SCCmec type I in three, and SCCmec type II in two. Among the 15 resistant isolates (MRSA), four were susceptible to oxacillin/cefoxitin by the disc diffusion method and one MSSA isolate was resistant to sulfamethoxazole/trimethoprim. The analysis of risk factors revealed a protective effect of age and lung disease, while lower-extremity ulcers were a risk factor for S. aureus. For MRSA, only male gender was significantly associated as a risk factor in multivariate analysis. Clonal profile analysis demonstrated the formation of clusters among MRSA isolates from different patients, with the identification of ST5-IV, ST5-I, and ST8-IV. Isolates carrying ST398 were identified among MSSA and MRSA (ST398-IV). Conclusion Our findings reinforce the importance of epidemiological studies of S. aureus carriage, especially in populations at high risk of infections such as diabetics. The data suggest widespread dissemination of MRSA in the population of insulin-dependent diabetic patients studied, as well as the emergence of important lineages among these individuals.

Background Diabetes mellitus is a progressive chronic disease characterized by high blood glucose levels, which is one of the most prevalent diseases in modern societies. Its treatment is often inadequate or absent [1]. It is estimated that more than 370 million people worldwide have diabetes and more than 5 million deaths were attributed to the disease and its complications in 2017 [2,3]. People with diabetes are known to be more susceptible to infections because of their elevated blood glucose levels and suppression of the immune response. In addition, neuropathy and reduced blood flow to the extremities are common in these individuals. Consequently, wounds tend to heal more slowly, increasing the risk of amputations and death [4,5].
Staphylococcus aureus is one of the leading causes of infections in diabetic individuals. This microorganism plays a particularly important role in this scenario by causing infections that range from superficial to severe and potentially fatal systemic infections, in addition to its ability to acquire resistance to multiple drugs. In the past, the pathogen was mainly found in hospital settings; however, we are now witnessing an increase particularly of resistant isolates (MRSA) acquired in the community that are genetically different from traditional nosocomial strains [6][7][8]. Infections caused by methicillin-resistant S. aureus (MRSA) are associated with a higher mortality rate compared to infections caused by methicillinsusceptible S. aureus (MSSA) [2]. Furthermore, studies suggest that staphylococcal infections are preceded by colonization with these microorganisms [2,[9][10][11][12][13] and at least one third of colonized healthy adults are at risk of developing subsequent invasive infections [6,7].
Bacterial resistance has become a major global threat. In the study of Onanuga and Temedie [14], the prevalence of multidrug-resistant MRSA isolated from the anterior nares of diabetic patients was 52.5%. This prevalence was 40% among diabetic outpatients in the study of Kutlu et al. [15]. High antimicrobial resistance of MRSA isolated from diabetic patients has also been reported by Alizargar et al. [16], with rates of erythromycin, ciprofloxacin and clarithromycin resistance of 81.9%, 71.3% and 65.5%, respectively, and a multidrug resistance rate of 59%. Two vancomycin-resistant Staphylococcus aureus strains were isolated in that study [16].
Diabetic individuals are considered a risk group for skin infections such as those caused by S. aureus and are also more prone to developing severe systemic infections [13]. Studies have shown that the prevalence of nasal colonization with S. aureus .6%) and MRSA (1-7.3%) varies according to geographic location and that the use of insulin is a risk factor for nasal MRSA carriage in diabetic individuals [16]. However, little is known about the role of throat colonization in this population.
Staphylococcus aureus is characterized by a high adaptation potential demonstrated by its success in surviving in different environments and the constant emergence of new lineages that cause different clinical manifestations and exhibit rapid epidemiological dissemination, causing a significant impact on health [17]. Globally disseminated MRSA clones included healthcare-associated [HA]-MRSA. The prevalence of this microorganism has also been increasing among community-associated [CA]-MRSA and livestock-associated [LA]-MRSA infections [17]. However, studies have demonstrated an exchange of these lineages with CA-MRSA, causing outbreaks of healthcare-associated infections in the community, and HA-MRSA being isolated in community settings [18,19]. LA-MRSA have also been reported in individuals without any previous contact with animals [17].
Staphylococcal cassette chromosome (SCCmec) typing is a useful epidemiological tool since different types are more prevalent in hospital or community settings. Unlike HA-MRSA that harbor large SCCmec (SCCmec types I, II and III) and are frequently multidrug-resistant, CA-MRSA are associated with SCCmec types IV and V, which are smaller and only carry the mecA gene of methicillin resistance [20][21][22][23].
In view of their complex epidemiology, S. aureus and MRSA lineages have been identified by techniques such as pulsed-field gel electrophoresis (PFGE), an adequate tool for the study of local outbreaks. Studies investigating the clonality of S. aureus are frequently complemented by multilocus sequence typing (MLST), which permits comparisons between S. aureus sequences described in different parts of the world. MLST and SCCmec typing have been used for the identification of global MRSA clones, which resulted in the detection of five large MRSA clonal complexes (CC) including CC5, CC8, CC30, CC45, and CC398 [24,25]. population of insulin-dependent diabetic patients studied, as well as the emergence of important lineages among these individuals. Considering their impact on the health of diabetic individuals, a better understanding of the epidemiology of and risk factors for colonization with S. aureus and MRSA is necessary. The aim of this study was to characterize the resistance profile, clonal profile and sequence type, as well as to analyze the prevalence and risk factors for nasal and oropharyngeal carriage of MSSA and MRSA isolated from insulin-dependent diabetic individuals in the city of Botucatu, SP, Brazil.

Study design
This was a cross-sectional study conducted in the city of Botucatu, São Paulo, Brazil, whose estimated population is 146,497 inhabitants [26]. We calculated the sample size from a population of 1631 individuals using an expected proportion of 50% (the typical value used in situations in which the proportion is unknown), considering a power of 80% (i.e., a beta error of 20%) and an alpha error of 5%, with a design effect of 1.0 (since no subsampling was performed). The formula used for sample size calculation is given in Additional file 1: Appendix S1. Using this calculation for proportions, we obtained a suggested n of 312 subjects. The subjects were selected randomly from the database of the Municipal Health Department over a period of 3 years (October 2015 to December 2018). If possible, the subjects were recruited during home visits (n = 204); however, in view of the difficulty in locating these individuals at the addresses obtained, many of them were invited by telephone to the events promoted at Basic Health Units (n = 70). Some subjects were recruited at the doctor's offices (n = 23) and at the headquarter of the Botucatu Association for Diabetes Support (ABAD in the Portuguese acronym) (n = 15). A questionnaire including the following data was applied to subjects who agreed to participate in the study: demographic data (gender and age); type of diabetes (1 or 2); time since diagnosis (years); time of insulin use (years); clinical data (comorbidities); presence of ulcers or amputations; tattoos; hospitalizations or medical procedures in the last year; and use of antimicrobials in the last year. All data were obtained by interview with the patient and/or legal representative following ethical standards. The questionnaires were extensively reviewed for inconsistencies.

Inclusion and exclusion criteria
All insulin-dependent diabetic patients living in the city of Botucatu, who agreed to participate by signing the free informed consent form (Additional file 1: Table S1), were included in the study. Patients who did not consent to participate in the study, patients who did not use insulin at the time of data collection, patients who had died, and patients who could not be found based on their personal data were excluded (Additional file 1: Table S2).

Collection of microbiological specimens
Nasal and oral mucosa samples were collected from 312 insulin-dependent diabetic individuals living in the city of Botucatu, São Paulo, Brazil, using sterile swabs in transport medium. Samples from the anterior nares and oropharynx were obtained using one swab for each site. For the collection of nasal samples, the swab was immersed in 0.9% sterile saline and inserted into both nares, rotating it and gently pressing its end against the mucosa. The technique for oropharyngeal sampling consisted of immersion of the swab and passing it gently over the surface of the throat, avoiding contact of the examiner with the tongue.
The swabs were transported in Stuart medium to the Laboratory of Bacteriology, Department of Microbiology and Immunology, Institute of Biosciences, UNESP, and seeded onto plates containing Baird-Parker agar, a selective medium for Staphylococcus. After incubation for 48 h at 37 °C, the isolated microorganisms were identified.

Identification of Staphylococcus aureus
The microorganisms were submitted to Gram staining for observation of their morphology and specific staining. After confirmation of these features, catalase and coagulase tube tests and biochemical tests (maltose, trehalose, and mannitol) were carried out to differentiate S. aureus from other Staphylococcus species [27,28].
After DNA extraction with the Illustra Kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK), the S. aureus isolates were confirmed genotypically by detection of the 16S rRNA gene [29] and the DNA SA442 fragment specific for S. aureus [30]. Thus, 112 S. aureus isolated obtained from 312 patients included in the study were identified.

Antimicrobial susceptibility testing
All 112 isolates obtained were subjected to antimicrobial susceptibility testing by the disc diffusion method using impregnated discs according to the criteria of the Clinical Laboratory Standards Institute (CLSI) [31]. The inoculum was adjusted to a 0.5 McFarland standard and seeded onto Mueller-Hinton agar and the plates were incubated for 24 h at 35 ºC. The following drugs were used: oxacillin (1 µg), cefoxitin (30 µg), linezolid (30 µg), quinupristin/dalfopristin (15 μg), and sulfamethoxazole/trimethoprim (25 µg). Antimicrobial activity was evaluated by determining the diameter of the inhibition zone, which was interpreted according to the CLSI [31]. International reference strains were used as controls for MSSA (S. aureus ATCC 25923) and MRSA (S. aureus ATCC 33591).

Determination of the minimum inhibitory concentration
The minimum inhibitory concentration (MIC) of vancomycin was determined by the E-test in all 112 S. aureus isolates. This quantitative test uses inert and transparent plastic strips (60 mm long × 5.5 mm wide) with a predefined gradient of concentrations of the antimicrobial to be tested. The MIC results were classified as susceptible, intermediate, or resistant according to the definitions of the CLSI [31]. An international reference strain (S. aureus ATCC 29213) was included as control.

Detection of the mecA gene of methicillin resistance
All S. aureus isolates were analyzed regarding the presence of the mecA gene by PCR following the parameters described by Murakami et al. [32]. International reference strains were included as positive (S. aureus ATCC 33591) and negative (S. aureus ATCC 25923) controls in all reactions. All isolates in which the mecA gene was detected were classified as MRSA, regardless of the result of the oxacillin and cefoxitin disc diffusion tests.

Determination of staphylococcal cassette chromosome mec (SCCmec) type
SCCmec typing was performed by multiplex PCR as described by Oliveira and de Lencastre [33] and updated by Milheiriço et al. [34] on all mecA gene-positive S. aureus isolates. The following strains were used as controls: COL for SCCmec type I; N315 for SCCmec type IA; PER34 for SCCmec type II; AN546 for SCCmec type III; HU25 for SCCmec type IIIA, and MW2 for SCCmec type IV.

Visualization of amplified products
The efficiency of the amplifications was confirmed by electrophoresis on 2% agarose gel prepared in 0.5 M Tris-borate-EDTA (TBE) buffer. A 100-bp marker was used as molecular weight standard. The gel was stained with SYBR ® Safe and photographed under UV transillumination.

Pulsed field gel electrophoresis (PFGE)
The modified protocol of McDougal et al. [35] was used to type all 112 S. aureus isolates by PFGE. The isolates were grown in BHI broth for 24 h. Next, 400 µL of the sample was added to a microtube and centrifuged at 12,000 rpm for 50 s. The supernatant was discarded and 300 µL TE solution (10 mM Tris, 1 mM EDTA, pH 8.0) was added. The samples were kept in a water bath for 10 min at 37 °C. After vortexing, 5 µL lysostaphin (1 mg/ mL in 20 mM sodium acetate, pH 4.5) and 300 μL lowmelting agarose were added. The samples were transferred to plug molds. The plugs were allowed to solidify and then placed in a 24-well plate with 2 mL EC solution (6 mM Tris-HCl, 1 M NaCl, 100 mM EDTA, 0.5% Brij-58, 0.2% sodium deoxycholate, 0.5% sodium lauroyl sarcosine) and incubated at 37 °C for at least 4 h. The EC solution was removed and the plugs were washed four times with 2 mL TE at room temperature at intervals of 30 min.
The SmaI enzyme (Fast Digest SmaI, Life Science, Canada) was used for restriction of genomic DNA. Electrophoresis was carried out in a CHEF-DR III System (BioRad Laboratories, USA) on 1% agarose gel prepared with 0.5 M TBE (Pulsed Field Certified Agarose, BioRad Laboratories, USA) under the following conditions: pulse switch time of 5 to 40 s for 21 h using a linear ramp; 6 V/ cm; angle of 120°; 14 o C; 0.5 M TBE as running buffer. The Lambda PFG Ladder (New England BioLabs) was used as molecular marker. The gel was stained with Gel-Red (10,000× in water; Biotium, USA) for 1 h and photographed under UV transillumination.
Similarity was analyzed with the BioNumerics 7.6 software (Applied Maths, Belgium). The dendrogram was constructed by the UPGMA method (Unweighted Pair Group Method with Arithmetic Mean), with band position tolerance and optimization adjusted to 1.2% and 1%, respectively.
Forty-five isolates, including one MRSA, could not be typed with the SmaI enzyme and were therefore digested with the ApaI restriction enzyme.
A similarity coefficient ≥ 80% was chosen for the determination of clusters. A cluster was defined as a group of three or more isolates showing ≥ 80% similarity.

Multilocus sequence typing (MLST)
Lineages representative of the clusters obtained by PFGE were selected for MLST. Five MRSA isolates obtained PFGE-SmaI and four MSSA isolates obtained by PFGE-ApaI were chosen.
MLST was performed according to the protocol of Enright et al. [36] by amplification and sequencing of seven housekeeping genes (arcC, aroE, glpF, gmk, pta, tpi, and yqiL). The PCR products were purified using the HiYield ™ Gel/PCR Fragments Extraction Kit and the sequencing reactions were performed in an ABI3500 8-capillary sequencer (50 cm) using POP7 as polymer (Applied Biosystems). The BioNumerics 7.6 program (Applied Maths, Belgium) was used for visualization of the sequences (electropherogram). The sequences were analyzed and compared to an online database (http:// www.mlst.net) (2004). Multivariate analysis was performed with the SPSS 20.0 software (©SPSS, Inc.) using a logistic regression model. The outcomes of interest were the overall presence of S. aureus or the presence of MRSA irrespective of sampling site. The variables were selected using a backward stepwise strategy. The criterion for entry and permanence of the variables in the models was p < 0.1. Final statistical significance was set at p < 0.05.

Prevalence of Staphylococcus aureus and MRSA carriage
Ninety-five of the 312 subjects included in the study were colonized with S. aureus, corresponding to an overall prevalence of 30.4% (95%CI 25.6-35.8%). Fifteen of the 95 subjects were colonized with MRSA either in the nose and/or oropharynx, corresponding to a prevalence of 4.8% (95%CI 2.9-7.8%). Regarding the sampling site, S. aureus was isolated exclusively from the nose in 44 (14.1%) individuals, exclusively from the oropharynx in 34 (10.9%), and from both sites in 17 (5.4%). Analysis of MRSA demonstrated that eight of the 15 colonized individuals carried the microorganism exclusively in the nose and six exclusively in the oropharynx. Interestingly, one subject carried MSSA in the oral mucosa and MRSA in the nasal mucosa (Fig. 1).

Determination of in vitro antimicrobial susceptibility
All 112 isolates identified as S. aureus were submitted to in vitro antimicrobial susceptibility testing as previously described. The disc diffusion method revealed 11 strains that were resistant to oxacillin and cefoxitin, including seven resistant to both drugs and four that were resistant only to cefoxitin (inhibition zone ≤ 21). It should be noted that four of the isolates that were susceptible to both drugs exhibited resistance by the genotypic method.
Among all isolates, only one was resistant to sulfamethoxazole/trimethoprim and there was no case of resistance to quinupristin/dalfopristin or linezolid. In addition, the MIC50 and MIC90 for vancomycin were 0.50 and 1.0 μg/mL, respectively, and all isolates were susceptible (Supplementary Material Table S3).

Detection of the mecA gene and characterization of SCCmec
Fifteen of the 112 S. aureus isolates carried the mecA gene of methicillin resistance. There was a predominance of the SCCmec type IV among isolates (n = 10), but three isolates harboring SCCmec type I and two harboring SCCmec type II were also identified. Most individuals harboring MRSA isolates reported no contact with the hospital environment in the last year, except for one subject with SCCmec type II and one with SCCmec type IV.   The study of risk factors for MRSA carriage revealed only male gender as a risk factor in uni-and multivariate analysis (OR = 3.64, 95% CI 1.12-11.78, p = 0.03).

Determination of the clonal profile of S. aureus and MRSA by pulsed-field gel electrophoresis (PFGE)
A total of 112 S. aureus isolates were analyzed by PFGE. Forty-five of these isolates could not be typed repeatedly with SmaI, including one MRSA isolate. However, molecular typing of all 45 isolates was possible using the ApaI enzyme.
For clonal profile analysis, one dendrogram was constructed for susceptible S. aureus isolates (MSSA) and one for resistant isolates (MRSA) using SmaI and ApaI, which permitted to identify clusters with similarity ≥ 80% in both groups. Figure 2 shows the dendrogram of the PFGE-SmaI and PFGE-ApaI profiles of MRSA isolates, as well as their in vitro susceptibility profile to oxacillin and cefoxitin, presence of the mecA gene, SCCmec type and MLST analysis. Analysis of PFGE-SmaI isolates revealed the presence of two clusters (A and B). Cluster A contained five isolates, four of them showing 100% similarity (554O, 555 N, 659 N, and 665 N). All of them were isolated from

Table 1 Uni-and multivariate (logistic regression) analysis of predictors of Staphylococcus aureus carriage in diabetic individuals
All values are reported as number (%) unless otherwise specified. Significant associations are indicated in italic OR odds ratio, CI confidence interval, CNS central nervous system   Fig. 2b. None of the isolates grouped with international clones. Two clusters (A and C) containing four strains and two clusters (B and D) containing strains were obtained for the MSSA isolates (Supplementary Material Figure  S1). Cluster A contained two isolates from the nasal and oropharyngeal mucosa of the same subject, demonstrating colonization of different sites with the same isolate. Similarly, in cluster B, two isolates from the nasal and oropharyngeal mucosa of the same subject were grouped with one nasal isolate of another subject. On the other hand, the four isolates of cluster C were obtained from unrelated individuals and in cluster D all isolates were from the oral mucosa of different subjects. These findings suggest widespread dissemination of S. aureus in the community.
Nine (52.9%) of the 17 subjects colonized at both sampling sites (nose/oropharynx) carried the same isolate in the nasal and oral mucosa, while eight (47.0%) carried different S. aureus strains in the nose and throat. Interestingly, two of the 17 subjects concomitantly colonized with S. aureus in the nose and throat had their strains (nasal and oropharyngeal) typed after digestion with different restriction enzymes. Isolates 691O and 747 N were typed with SmaI, while 691 N and 747O could only be typed with ApaI, confirming that they are different S. aureus strains. There was also one patient colonized with MRSA (735 N) in the nasal mucosa and with MSSA (735O) in the oropharyngeal mucosa, both typed with ApaI.
Analysis of typable isolates with ApaI revealed three major clusters (A, C, and D) and three minor clusters of 3 isolates (B, E, and F). In cluster A which contained 14 isolates with 85.6% similarity, two isolates were from the nasal and oropharyngeal mucosa of the same patient and the remaining from different subjects. The same was observed for clusters C and D. In cluster D, the isolates were grouped with a strain previously identified in another study as ST398 (strain 76 N). These findings are shown in Supplementary Material Figure S2.

Table 2 Uni-and multivariate (logistic regression) analysis of predictors of MRSA carriage in diabetic individuals
All values are reported as number (%) unless otherwise specified. Significant associations are indicated in italic OR odds ratio, CI confidence interval, CNS central nervous system

Molecular typing of S. aureus and MRSA by multilocus sequence typing (MLST)
Based on the clusters obtained by PFGE, nine S. aureus isolates (four MSSA and five MRSA) were selected for molecular typing by the MLST technique. Typing of the MRSA isolates revealed a predominance of sequence type ST5 in three of the four isolates analyzed and one isolate with ST8. Regarding SCCmec type, isolates of the following lineages were obtained: ST5-IV (n = 2), ST5-I (n = 1), and ST8-IV (n = 1). In addition, the strain typed with ApaI by PFGE was characterized as ST398-IV (Fig. 2).
Among the clusters of the MSSA isolates obtained by PFGE that could not be typed with SmaI and that were digested with ApaI, four lineages were selected for MLST. There was a predominance of ST398 (n = 3). However, one isolate exhibited divergence in the allele of the arcC gene and was sent to the curator of the MLST database (https ://pubml st.org/) for identification of the ST. This isolate was identified as ST 6133 (Fig. 3).

Discussion
The prevalence of colonization with S. aureus and MRSA among insulin-dependent diabetic individuals was 30.4% and 4.8%, respectively. Similar data have been reported by Hart et al. [37] who analyzed 258 patients and found a prevalence of 39.1% for S. aureus and 3.1% for MRSA among diabetic individuals. In a population-based survey   conducted in the same city, Pires et al. [38] found a similar prevalence among community-dwelling individuals, with an overall prevalence of S. aureus of 32.7%. However, the prevalence of MRSA was higher in our study (4.8% vs 0.9%). In a recent study, Lin et al. [2] found a lower prevalence of S. aureus and MRSA than that obtained in the present study (16.4% S. aureus and 2.8% MRSA). The authors suggested that the presence of different microorganisms in the microbiota of these individuals causes competition for the same site, which could explain the variation found in prevalence studies. None of the studies included oral mucosa as a potential site of colonization.
It should be noted that 10.9% (n = 34) of the subjects were colonized exclusively in the oral mucosa, six of them with MRSA. This finding reinforces the suggestion of Partida et al. [39] that colonization of the oral mucosa can compromise control measures of pathogen dissemination since the throat is not part of routine screening.
The MRSA isolates were identified by phenotypic methods (disc diffusion) and by PCR for detection of the mecA gene. We found four isolates carrying the mecA gene that did not exhibit phenotypic resistance to cefoxitin or oxacillin. Although mecA gene resistance is present in all cells of a population with intrinsic resistance, it may only be expressed by a small proportion of these cells, a fact that results in the so-called heteroresistance [40]. Isolates carrying the mecA gene but that are susceptible to oxacillin/cefoxitin have been reported worldwide and are called oxacillin-susceptible MRSA (OS-MRSA) [41][42][43][44][45]. According to Andrade-Figueiredo & Leal-Balbino [46], this phenomenon may be due to partial excision of SCCmec in multidrug-resistant MRSA isolates or chromosomal integration of the cassette chromosome, resulting in MSSA isolates that contain SCCmec segments.
High antimicrobial resistance of MRSA isolated from diabetic patients has been reported in different studies [14][15][16]. In our study, there were no multidrug-resistant isolates and none of the isolates was resistant to vancomycin, although one MRSA isolate had a MIC of 1.5 μg/ mL, indicating a potential therapeutic risk [47][48][49]. Two other MSSA isolates had a vancomycin MIC of 1.5 μg/ mL.
In the present study, the analysis of risk factors revealed an association only with lower-extremity ulcers, which is consistent with literature findings showing that the same isolate colonizing the nares was present in foot ulcers and wounds [50,51].
Age was associated with a lower risk of S. aureus colonization, with a 2% decrease in the risk of colonization for each additional year of age. Similar data have been reported by Pereira-Franchi et al. [52]. It is believed that, with increasing age, individuals are exposed to factors (not analyzed in our study) that can prevent colonization with S. aureus, in addition to increased ecological competition with other microorganisms.
Lung disease was also a protective factor against the acquisition of S. aureus, a fact that might be related to colonization of the respiratory tract with other microorganisms that are competing with S. aureus. The nasopharyngeal microbiota changes over time; the level of bacterial colonization is higher during upper respiratory infection [53] and other species such as Streptococcus pneumoniae and Haemophilus influenzae may thus interfere with the capacity of S. aureus to persist in the nasal mucosa. Mueller et al. [54] also found a protective effect of age but, in contrast to our findings, lung disease was a risk factor for S. aureus colonization.
Multivariate analysis identified male gender as a risk factor for colonization with MRSA. Similar results were verified in a study of the prevalence of S. aureus and MRSA in bedridden individuals and residents of Long-Term Care Institutions for the Elderly (ILPIs) in the same city [55]. In this study developed by Silva [55], the male gender was associated with increased risk for the carrying of S. aureus and was also the only variable that showed to be a risk factor for carrying MRSA. Nillius et al. [56] reported that male ILPI residents had an almost double risk for carrying MRSA when compared to women, probably because they had more risk factors than they did.
With respect to clonality of the MRSA isolates, lineages belonging to the most widespread clonal complexes were identified, including CC5-ST5-IV (639O) and CC8-ST8-IV (72O) isolated from individuals with colonized oropharyngeal mucosa. This fact reinforces the importance of throat colonization, which could be a route of transmission within the population examined. Other studies involving individuals from the same city and region also found CC5-ST5-IV and CC8-ST8-IV, suggesting that these strains are prevalent in the region [57][58][59][60][61][62].
Studies suggest a high clonal diversity among S. aureus isolates, particularly among MSSA [63]. Among the isolates that could not be typed with SmaI, four were typed by MLST and were characterized as ST398. This fact was also observed by de Souza [61]. The ST398 clonal lineage has been associated with infection and colonization of humans and domestic animals, such as dogs, horses and pigs, in many countries around the world [8]. This lineage is called livestock-associated S. aureus and was described for the first time among both MSSA and MRSA on pig farms in France [64,65]. Since then, ST398 has spread rapidly to other animals and has been increasingly related to infections not only in rural workers but also in people and animals without risk factors [31,66]. Although susceptible to oxacillin, this S. aureus lineage is associated with severe infections, as reported by Bonesso et al.  [67] in patients with ventilator-associated pneumonia in whom the infection was fatal in most cases. Our finding demonstrated a predominance of SCCmec type IV among isolates, in agreement with the findings of other prevalence studies on non-diabetic individuals conducted in the State of São Paulo [57,61]. However, SCCmec types I and II were also detected, which are commonly found circulating in health services. This fact has also been reported by Pereira-Franchi et al. [62] and Silveira et al. [68] who found a higher prevalence of isolates harboring SCCmec type II, which was attributed to a history of hospitalization.
It is worth mentioning that patients in hospital-community settings, such as bedridden or institutionalized older adults with chronic infections, have a higher prevalence of SCCmec types I and II [62,68]; in addition, hospitalized patients frequently carry isolates that harbor SCCmec type III. These SCCmec are larger and carry plasmids and transposons with other resistance genes, often multidrug-resistant genes. On the other hand, community-dwelling patients are associated with SCCmec types IV and V, which are smaller and carry only the mecA gene of methicillin resistance [38].
One limitation of the present study is the small number of MRSA isolates (n = 15), which may result in a low statistical power of the analyses. In addition, the patients studied are not typically community-dwelling since they often need to seek health services (primary care) because of their diabetes.
The present study provides important data about the epidemiology of S. aureus and MRSA in a population of insulin-dependent diabetic individuals. The isolates analyzed had a low rate of resistance to the tested drugs, with only one isolate being resistant to sulfamethoxazole-trimethoprim; however, the prevalence of MRSA was higher than that found in a population-based study conducted in the same city on healthy individuals [38]. Within this context, screening for oral colonization is extremely important since some individuals were colonized only at this body site. In the population studied here, clones were detected among the MSSA and MRSA isolates and an important clonal lineage (ST398) was identified. These data suggest widespread dissemination of MRSA in the population of insulindependent diabetic patients studied, as well as the emergence of important S. aureus lineages in these individuals.
Additional file 1. They were included as tables of the characteristics of the individuals included and excluded from the study. In addition to the dendrograms of the PFGE-SmaI and PFGE-ApaI profiles of MSSA isolated from insulin-dependent diabetic individuals.