Molecular typing of MRSA isolates by spa and PFGE

1 Introduction

Methicillin resistant S. aureus (MRSA) gains tremendous attention in medical practice and microbial research because of the ascending challenge of therapy and control. MRSA causes annual deaths higher than any other infectious agent in USA (Klevens et al., 2007). In 2011 the Center for Disease Control and Prevention (CDC) estimated that more than 80,000 invasive infections and more than 11,000 deaths were caused by MRSA (Dantes et al., 2013). Furthermore, patients infected with MRSA required longer period of hospitalization and higher health care cost compared to patients infected with other staphylococcus species (Cosgrove et al., 2005). Meta-analysis study comparing MRSA infection to methicillin sensitive S. aureus (MSSA) bacteremic infection revealed that MRSA caused markedly higher mortality, which was attributed to prescribing ineffective early therapy (Cosgrove et al., 2005).

Considering the spread of MRSA strains across the borders, stringent strategies are required to control their dissemination and trace their nosocomial reservoirs including active surveillance and general infection control measures such as screening, decolonization and contact isolation. However, to ensure the effectiveness of these strategies, knowledge of the common MRSA genotypes, especially of the ones circulating in the population, is required. Several molecular typing methods employing DNA banding patterns are available such as pulsed-field gel electrophoresis (PFGE) or utilizing DNA sequencing such as the staphylococcal protein A (spa) typing and multilocus sequence typing (MLST).

Staphylococcal cassette chromosome which harbor mec (SCCmec) typing method is a polymerase chain reaction (PCR)-based and detect the type of SCCmec cassette but not its structure and is therefore less discriminative (Boye et al., 2007; Zhang et al., 2005). PFGE is a very discriminative approach for S. aureus typing because it is based on fragmentation of the bacterial chromosome using Serratia marcescens endonuclease I (SmaI) enzyme and separation of the digested DNA fragments on agarose gel using alternating pulses. However, it is laborious, time consuming and lacks the reproducibility among different laboratories in addition to its cost ineffectiveness. Furthermore, efforts to harmonize standard protocols were only successful on the national levels in some countries but not internationally (Murchan et al., 2003; Tenover et al., 1994; van Belkum et al., 1998). Sequence-based typing methods are preferable to DNA-size typing methods. MLST typing was established to analyze the allelic sequence of short DNA fragments (450–500 bp) of seven S. aureus housekeeping genes (Enright et al., 2000). This method is an excellent tool for screening and very useful in studying the molecular evolution of S. aureus. However, it is expensive, laborious, and time consuming and therefore not suitable for diagnostic purposes as for each S. aureus isolate seven PCR reactions and 14 sequencing reactions should be performed apart from the data analysis.

Spa typing method targets region X of the spa gene, which contains a tandem repeats (about 24 bp in length) (Frenay et al., 1996). Different spa types vary by the type of repeats and their copy numbers. Spa typing became very popular since it is based on sequencing of a single locus, less expensive and less time consuming than other methods. Moreover, it has more discriminative power compared to MLST (Malachowa et al., 2005). The availability of the spa types on the central spa server (http://spaserver.ridom.de) and its unified nomenclature provided another important advantage. The use of these typing tools contributes to understanding clonal diversity and transmission of MRSA in the hospital and community settings.

The aim of our study is to genotype MRSA isolates using spa typing to determine the circulating MRSA strains, which will enable a better design of surveillance protocols and control strategies. The study also aims at comparing the spa typing approach to that based on PFGE.

2 Methods:

2.2

2.2 Spa typing:

Extraction of DNA was performed using Qiagen DNA extraction kit following the standard protocol with addition of lysozymes (Qiagen, Germany). DNA concentration and quality assessment was done by NanoDrop (life technologies, USA). The spa1095F (5′-AAAGACGATCCTTCGGTGAGC-3′) and spa1517R (5′-GCTTTTGCAATGTCATTTACTG-3′) primers were used to amplify the target fragment of the spa gene by PCR according to the protocol previously described (Harmsen et al., 2003). The amplified DNA product was separated on 1% agarose by gel electrophoresis and the amplicons were purified by gel extraction kit from Qiagen (Qiagen, Germany). The purified DNA product was sequenced in both forward and reverse directions using the same PCR primers in combination with the big dye termination mix kit and the ABI genetic analyzer 3500 (Applied Biosystems, USA). All DNA sequences were trimmed to the required length and the full fragment sequence was obtained by aligning the forward and reverse sequences for each isolate using the DNAGear software (F et al., 2012). Spa types were allocated to each isolate according to the repeats listed in the spa typing website (http://www.spaserver.ridom.de/) using the DNAGear software (F et al., 2012). Fig. 1 illustrates the flow of spa typing method (Fig. 1).

Close

Fig. 1

Illustration of the procedure and flow of spa typing method.

Export to PPT

3 Results

In total, thirty-five spa types were identified in our cohort. The most common spa type in our study was t044 constituting 30.18% of the isolates (Table 1). Spa types t127 and t304 were the second most common types (5.6% each), followed by t363 (4.6%), and t1200 and t002 (3.8% each) (Table 1). The rest of the isolates were clustered either in less frequent spa types or as singletons (Table 1). Seven isolates had no matching spa type in the database. Three of the seven isolates had new repeats and four isolates had known repeats but the arrangement of the repeats didn’t match with previously identified spa types (Table 1).

SmaI digestion and PFGE identified the presence of 34 pulsogroups designated as A to T (Fig. 2). PFGE group J was the largest and comprised 24 of the studied isolates. Other pulsogroups (A, B, E, I, P, and S) comprised 4–7 isolates each (Fig. 2).

Close

Fig. 2

Pulsed field gel electrophoresis groups based on 80% Dice coefficient.

Export to PPT

There was no significant clonal association between spa types and PFGE groups. There was clustering of some spa types within some of the pulsogroups. However, it was statistically non-significant, most probably due to the small numbers of isolates of such spa types. Additionally, the seven isolates that potentially represent new spa types clustered within the identified pulsogroups.

No association was found between neither of the typing methods and the clinical data nor demographic data such as gender, age group, source of isolates, or antibiotic susceptibility pattern. However, some spa types (t002 and t363) were confined to isolates recovered from a set of infection sites (wound, abscess, and endotracheal aspirate), while others were more frequently associated with isolates recovered from colonization sites (t304 and t690). There was no relation between spa types and severity of infection.

4 Discussion

Effective strategies are required to control the dissemination of MRSA worldwide such as active surveillance and institutional preventive measures, which requires insight into the molecular epidemiology of the circulating genotypes. MRSA prevalence in Saudi Arabia ranges between 12 and 49% depending on the region (Austin et al., 2003; Baddour et al., 2006; Bukharie and Abdelhadi, 2001; El Amin and Faidah, 2012). The estimated carriage rate of MRSA among health care workers is too high reaching 67% (Iyer et al., 2014) leading to an increase in the risk of hospital-acquired infections. Two studies were recently published using the spa typing approach but had either a small number of isolates or studied a selected group of patients (Abou Shady et al., 2015; Alreshidi et al., 2013). Based on that, we have initiated this study, which revealed a clear genotypic diversity within the MRSA population.

More than thirty types were identified using spa or PFGE typing approaches among 106 isolates. This genotypic diversity of MRSA strains could be attributed to the diversity in the population living in the area; people from different nationalities work and live in the region and hence importing the diversity observed. Additionally, it could be attributed to low rate of nosocomial cross-transmission due to stringent MRSA screening program adopted for more than a decade, and the selective pressure effect of antimicrobials over various circulating strains (Argudin et al., 2009).

Spa typing method relies on the PCR amplification followed by sequencing of the highly variable X fragment of the staphylococcal protein A, which makes it simple and fast compared to the MLST approach. All of the isolates, except seven, had spa types matching a known type on the database. Our isolates with new spa type were obtained from skin and nasal samples.

Spa type t044 (30.18%), which was the most common in this study, was previously reported as being the second most common after type t037 in Riyadh (Alreshidi et al., 2013). This indicated the high prevalence of type t044 in Saudi Arabia with some regional variation. However, in another study from Buraydah city in Al-Qassim district near to Riyadh none of the two former common spa types were detected, which could be attributed to the low sample size (15 MRSA isolate) (Abou Shady et al., 2015).

Type t044 represents the so called ST80 MRSA, which is very common in the Middle East. It was detected in other neighboring countries at variable frequencies such as Jordan (Al-Bakri et al., 2013; Bazzoun et al., 2014); however, it was only reported in adults and healthcare workers but not in children (Aqel et al., 2015). This was in contrast to our results, as type t044 was detected in all age groups in our study. It was also reported from Lebanon (Harastani et al., 2014), United Arab Emirate (Sonnevend et al., 2012), Kuwait (Udo and Al-Sweih, 2017), Iran (Goudarzi et al., 2016), and in one but not all studies from Tunisia (Ben Nejma et al., 2013; Ben Said et al., 2016; Kechrid et al., 2011), and widely disseminated in Europe, Australia (Larsson et al., 2014), and USA (Fluit et al., 2015). No correlation between type t044 and the source of the isolate was reported. However, some spa types were confined to isolates obtained from a specific site of infection; for example t002 and t363 were isolated from wound, abscess, and endotracheal aspirate. However, the small number of isolates (5 each) precludes firm conclusion. There was no relation between spa types and severity of infection.

Three new spa types were reported, and four isolates had known repeats but the repeat sequence did not match any previously identified spa types on database. The X region of the spa protein is very polymorphic, with 761 repeats being so far registered on the Ridom spa server. The random arrangement of these repeats generated 7643 spa types. New spa types are generated by the rearrangement of the repeats in a recombination event or by generation of new repeats due to DNA polymerase error, deletion, or duplication events.

Despite its limitations, spa typing has shown, in this work, its utility to study local MRSA epidemiology and to help in clustering isolates. It also showed a better discriminative power. The lack of association between the spa type and PFGE pattern could be attributed to the large genetic diversity of the studied isolates.

5 Conclusion

The data presented showed high diversity of MRSA in our population providing a baseline for further molecular characterization in the Region. Spa-typing can be considered a practical approach to investigate and manage MRSA-linked outbreaks because of its simplicity, relative low cost, high throughput and discriminative power besides the standardized nomenclature and portability into an international database. Infrequently when misclassification or non-typeability is a concern, a sophisticated typing tool, such as PFGE in this study, is an informative supplementary addition.