Molecular typing of Staphylococcus aureus collected from a Major Hospital in Amman , Jordan

Introduction: Over the past decade methicillin-resistant Staphylococcus aureus (MRSA) has been recognized as a major cause of healthcare associated infections. Recently, however, epidemiology of this pathogen has changed drastically with the emergence of new clones in the community. Efficient epidemiological typing methods are essential to monitor and limit the occurrence and spread of epidemic clones. Methodology: A total of sixty S. aureus isolates were collected from the Jordan University hospital in Amman-Jordan. All isolates were characterized using Staphylococcus protein A (spa) typing and pulsed-field gel electrophoresis (PFGE). Samples were tested for their susceptibility patterns against seven antimicrobial agents and for their potential to form biofilms. Results: spa typing showed that spa type t044 was the most common representing 28% of the isolates studied and 38% of the MRSA population. PFGE revealed fourty-six pulsotypes among the sixty tested isolates clustering similar spa types together. The predominant resistance was detected against levofloxacin, chloramphenicol and clindamycin. One MSSA isolate typed as spa t955 showed biofilm formation potential through protein deposition.. Conclusion: The study results are based on one hospital, but the findings of this and other studies conducted in the region indicate that there is an urgent need for standardized surveillances combined with the application of well-validated typing methods to assess the occurrence of MRSA and to control its spread.


Introduction
Staphylococcus aureus is the most commonly isolated human bacterial pathogen and is an important cause of skin and soft tissue infections (SSTIs) [1].Colonization is an important step in the pathogenesis of S. aureus infection and is contributory for acquiring nosocomial infections [2].Diseases ranging from superficial skin infections to severe endocarditis, pneumonia and bacteremia may result from an infection caused by S. aureus [3].Several virulence factors contribute to the pathogenicity of S. aureus including Panton-Valentine leukocidin (PVL), protein A, toxic shock syndrome toxin 1 (TSST-1), enterotoxins and exfoliative toxins [1].The induction of biofilm formation, by the adhesion of bacteria to a surface, followed by accumulation in multilayered cell clusters through intercellular adherence, is an additional vital infectious component [4].
Treatment of S. aureus infections is complicated due to the ability of this bacterial species to become resistant to antibiotics [5].Methicillin resistant S. aureus (MRSA) emergence and dissemination are of global significance in both health care and community settings [6].MRSA has been reported to cause serious invasive and life-threatening infections in young children [7].In addition, children may act as vectors helping S. aureus and specifically MRSA to spread in both community and hospital environments [6].
Efficient epidemiological typing methods are essential to monitor and limit the occurrence and spread of epidemic clones between hospitals; typing systems may facilitate the discrimination between unrelated isolates and the recognition of isolates that belong to the same clonal lineages.Different molecular approaches such as spa typing, multilocus sequence typing (MLST), SCCmec typing, and pulsedfield gel electrophoresis (PFGE) can be used to determine the genetic relatedness between isolates of clinical relevance [3].
In previous studies conducted in Jordan nasal carriage of MRSA was detected in 550 hospital staff members of four hospitals in north Jordan [8] and in 40% of the Jordanian healthy young adult population.MRSA represented 19% and 57% of the nasal and clinical isolates, respectively [9].Furthermore, a retrospective study conducted on positive blood cultures taken from 378 children younger than 15 years of age, showed that S. aureus was the most common isolate in bacteremic children [10].Finally, fourty-one MRSA and sixty-two MSSA (methicillinsensitive) isolates were collected from children in Jordan.Genotyping based on spa and MLST revealed fourty-eight different spa types and identified distinct allelic profiles or STs, with the majority belonging to ST80.PFGE of fifteen different spa types revealed eight different pulsotypes, while SCCmec showed the predominance (53%) of subtype IV [11].Taking into consideration the changing epidemiology of S. aureus, this study was conducted to determine the clinical, epidemiological, and molecular features of currently circulating MSSA and MRSA strains.Understanding the epidemiology of MSSA and MRSA infection, particularly in developing countries where the knowledge of S. aureus transmission dynamics is probably limited, is fundamental for devising effective prevention and control strategies.

Isolates
A total of sixty S. aureus isolates were obtained from clinical specimens in the period between 2009 and 2010.These isolates were kindly provided by Dr. Asem Shehabi from the Jordan University hospital in Amman-Jordan (JUH).The JUH provides a specialized and high quality health care for about half a million patients a year from all over the country, with 600-beds inpatient capacity.The isolates were recovered from different sites of infection including mainly blood, wounds, catheters, burns, and nasal and breast swabs.Patients were distributed along a wide age range (one day up to 43 years).Isolates included both MRSA and MSSA.The samples were streaked on Mannitol Salt Agar (MSA) to be further purified and stored in Cryobanks at -20°C and -80°C.The isolates were designated with letters JS D, followed by the assigned lab number.

DNA extraction
Isolates were grown overnight on MSA at 37°C.DNA extraction was carried out using Nucleospin Tissue (Macherey-Nagel, Düren, Germany) and BIORAD InstaGene Matrix (Bio-Rad, Hercules, USA) according to the manufacturer's instructions.The extracted DNA was stored at -20°C until needed.

Multiplex PCR for detection of 16S rRNA, PVL, and mecA genes
Amplification of the 16S rRNA, PVL and mecA genes was done as previously described by McClure et al. [12].PVL negative MRSA (N315) and PVL positive MSSA (ATCC 49775) were used to determine assay sensitivity.PCR products (7 µl) were resolved in a 1.5% Seakem gold agarose (Lonza, Rockland, USA) gel in 0.5% Tris-borate-EDTA buffer (Bio-Rad, Hercules, USA) at 6V/cm for one hour and were visualized with ethidium bromide.

spa typing Based Upon Repeat Pattern (BURP) analysis
The polymorphic X region of protein A gene (spa) was amplified as described previously [3,13].By applying the BURP algorithm implemented by the software, spa types with more than five repeats were clustered into different groups, with the calculated cost between the members of the group being  6 [14].The spa type was assigned using the Ridom StaphType software version 2.2.1 (Ridom GmbH, Wüzburg, Germany).

Pulsed-field gel electrophoresis (PFGE)
All the sixty isolates were typed using PFGE technique to reveal a banding pattern on the gel.The protocol was adapted from Georing and Winters 1992 [15].A bacteriophage lambda ladder PFG marker (New England BioLabs, Hitchin, UK) was included in each gel.Cluster analysis was performed with Gel-Compar II software version 6.5 (Applied Maths, Sint-Martens-Latem, Belgium) using dice coefficient and visualized as a dendrogram by the unweighted pair group method (UPGMA).

Biofilm assay
All isolates were grown in 96 well plates (Corning, New York, USA) according to the protocol adapted from Izano et al. [4].Biofilm stability was then tested against DNase I, proteinase K, and sodium metaperiodate (Sigma Aldrich, Saint Louis, USA) enzymes, which were then added to separate wells for the same isolate leaving the fourth as a control (BHI + 0.25% glucose).Optical density OD 590nm was read with ELISA reader (Multiskan FC Microplate reader, Thermo Fisher Scientific, Waltham, USA).Staphylococcus epidermidis RP62A (ATCC 35984) and TM300, kindly donated by Dr. Barbara Kahl from University of Münster, Germany, were used as positive and negative controls, respectively.

Results
Results showed that 68% (41 samples) of the isolates were MRSA, with 70% (43 samples) being PVL negative.Out of the eighteen PVL positive isolates, fifteen were MRSA (83%).All sixty isolates were typeable by spa typing, which showed the presence of twenty-five different spa types; the most common being t044 representing 28% of the isolates followed by t037, t1149 and t318, t159 (Figure 1).Within MRSA, 38% were from spa type t044 and 12% spa type t037.Using BURP algorithm, the twenty-five spa types were clustered so that each spa type belonged to one cluster and was related to at least one other spa type within the cluster.There were two major spa-CCs, eight singletons, two clusters with no founder, and one spa type (t386) was excluded.
All sixty isolates were genotyped by PFGE after SmaI enzyme digestion of chromosomal DNA.Fortysix pulsotypes were identified using 80% similarity as a cutoff value, and seven groups had more than one isolate.It is noteworthy that most isolates belonging to the same spa type such as t044, t386, and t037, were clustered together (Figure 2).
S. epidermidis RP62A (ATCC 35984) showed an OD of 0.812 in the control well at  590nm .This value did not vary significantly upon the addition of DNase I and proteinase K. On the other hand, addition of sodium metaperiodate changed the OD significantly to 0.213.Tested S. aureus isolates showed an OD value between 0.160 and 0.216, similar to that of the blank, in the control well.JS D51, an MSSA with spa type t955, showed an absorbance value of 0.893 similar to that of the positive control.The OD however, dropped to 0.167 upon proteinase K addition, indicating the potential ability of this isolate to form biofilm through protein deposition (Table 1).
Antibiotic susceptibility was tested against seven antimicrobial categories including: aminoglycosides, ansamycins, penicillin, fluroquinolones, lincosamides, macrolides, and phenicols.According to the definitions proposed by Magiorakos et al. [17], 25% of all tested isolates were multidrug-resistant (MDR) and 67% were extensively drug resistant (XDR).MRSA and MSSA isolates exhibited resistance patterns to levofloxacin, chloramphenicol, clindamycin, erythromycin and gentamicin.Figure 3 demonstrates the percentage resistance to the tested antimicrobial agents for MRSA and MSSA, and the percentage of strains being MDR among MRSA and MSSA.Dendrogram of all 60 S. aureus isolates following digestion with SmaI restriction enzyme, using GelCompar with percent similarity calculated by Dice coefficient (tolerance: 1%, optimization: 0.5%) and represented by unweighted-pair group method (UPGMA) settings.Lambda ladder PFG marker (New England BioLabs, UK) was included in each lab for alignment of bands.Similarity cutoff of 80% is represented by a vertical line and PFGE groups are numbered from 1 to 46.Specimen origin, presence of mecA and PVL genes as well as spa type for each sample is also included.The scale bar shows percentage similarity.

Discussion
Infections due to MRSA are becoming a major health issue in the Middle East, yet questions on the epidemiology and infection remain unanswered.This study focuses on spa typing and PFGE of S. aureus samples obtained from one hospital in Amman, Jordan.Although the results of this study cannot be generalized to the whole country, they revealed the distribution of fourty-one MRSA isolates over eighteen spa types, as opposed to fourteen spa types for the nineteen MSSA isolates.This reflects a higher diversity in the genetic profile of MSSA compared to MRSA, which was an observation previously reported also by Grundmann et al. [18].The most commonly occurring spa types were t044 and t037.The high prevalence of spa type t044 was in concordance with several other studies including: Vindel et al. [19] who showed that this type was widely disseminated in Europe, Tokajian et al. [20] reporting that 38% of S. aureus isolates belonged to spa type t044 in Lebanon, and Khalil et al. [11] revealing that 18% of MRSA recovered from stool and nose specimens from children in Jordan were of spa type t044.spa type t044 belongs to the one of the recently emerging PVLpositive MRSA strains (ST80-MRSA-IV), known to occur and predominate in certain regions in Europe and in the Middle East [21].
Using BURP algorithm, all spa types were clustered into two spa-CCs, eight singletons, and three groups with no founders.Both spa-CC 044 and 037 contained more than one spa type being both MSSA and MRSA.These overlapping clonal complexes, which included both MRSA and MSSA, could be indicative of the extension of MRSA from MSSA through the acquisition of the SCCmec [22].
The sixty isolates were analyzed by PFGE, and fourty-six pulsotypes were obtained when 80% similarity cutoff value was used.PFGE has clustered similar spa types together showing the high discriminatory power of PFGE as previously described by Cookson et al. and Bosch et al. [23,24].
The ability to form biofilm was determined through measuring the absorbance values of all of the sixty isolates and comparing them to that of S. epidermidis (positive control).Only one MSSA isolate belonging to spa t955 and which clustered as a singleton, showed an initial absorbance of 0.893, which was comparable to the positive control.This value decreased to 0.167 upon proteinase K addition and remained high in the presence of sodium metaperiodate and DNase.Since sodium metaperiodate, proteinase K, and DNase indicated the presence of polysaccharide intercellular adhesion (PIA), proteinaceous factors, and genomic eDNA, respectively, we could infer that this isolate was capable of forming biofilm through protein deposition [25].The ability of S. aureus to form more pronounced biofilms was previously correlated to the site of infection.The potential biofilm former obtained in this study was isolated from the skin, which was in harmony with the finding that S. aureus isolated from skin has a better ability to form completely recognized biofilms compared to those taken from other sites of the body [26].This might be explained by the tendency of S. aureus to use the biofilm formation as an advantage to occupy the unstable environment of the human skin, placing the host at risk if the skin is breached [27].Bacterial biofilms are structured communities and the formation of these sessile communities and their inherent resistance to antibiotics and immune attack are at the root of many persistent chronic bacterial infections [28].

Conclusion
Given the great diversity observed within the MSSA and MRSA clinical isolates, it seems likely that many different strains circulate into the hospital and in the region.To obtain a better view of what is circulating in the region, and what the dimensions of the problem are, it is recommended that surveillance programs combined with dedicated typing methods should be performed.

Figure 1 .
Figure 1.Percentage distribution of the recovered spa types in this study.

Figure 2 .
Figure 2.Dendrogram of all 60 S. aureus isolates following digestion with SmaI restriction enzyme, using GelCompar with percent similarity calculated by Dice coefficient (tolerance: 1%, optimization: 0.5%) and represented by unweighted-pair group method (UPGMA) settings.Lambda ladder PFG marker (New England BioLabs, UK) was included in each lab for alignment of bands.Similarity cutoff of 80% is represented by a vertical line and PFGE groups are numbered from 1 to 46.Specimen origin, presence of mecA and PVL genes as well as spa type for each sample is also included.The scale bar shows percentage similarity.

Figure 3 .
Figure 3. Percentage of S. aureus isolates that are resistant to the tested antimicrobial agents for MRSA and MSSA, along with the percentage of strains that were MDR and being MRSA or MSSA.Levofloxacin (LEV), Chloramphenicol (C), Clindamycin (DA), Erythromycin (ERY), and Gentamicin (GEN).

Table 1 :
Determination of the OD 590 nm using ELISA reader for the studied isolates before and after the addition of one of the following: DNase I, proteinase K, and sodium metaperiodate