Molecular Typing of Methicillin Resistant Staphylococcus aureus Clinical Isolates on the Basis of Protein A and Coagulase Gene Polymorphisms

Increased frequency of methicillin-resistant Staphylococcus aureus (MRSA) in hospitalized patients requires rapid and reliable characterization of isolates for control of MRSA spread in hospitals. This study evaluated polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) as a molecular typing technique for MRSA strains on the basis of protein A (spa) and coagulase (coa) gene polymorphisms to verify their ability in assessing the relatedness of isolates. Seventy-five MRSA isolates, from different ICUs of Alexandria University Main Hospital, were characterized using antibiotyping and PCR-RFLP analysis of coa and spa genes. Thirty-two antibiotypes were identified. coa gene PCR generated 3 types and 10 subtypes of band patterns. HaeIII restriction digestion of amplified coa gene products produced 5 major banding patterns and 12 subtypes. spa gene PCR products generated 4 major and 11 minor types, and their HaeII restriction digestion showed 5 major and 12 minor banding patterns. The combined coa and spa RFLP patterns generated 22 combined R types. Typing using coa PCR and PCR-RFLP had the same discriminatory index (DI) value (0.64), which was comparable to that of both spa PCR and PCR-RFLP techniques (0.68). The combined grouping increased the DI value to 0.836. The current study revealed that testing for multiple gene polymorphisms is more useful for local epidemiologic purposes.


Introduction
The health risks associated with MRSA infections warrant the implementation of monitoring programs to control its dissemination, given the potential of MRSA to produce invasive infections, particularly in vulnerable patients, and its multiple resistances to antibiotics, which limits the therapeutic options available [1].
Antibiotyping and molecular typing are key functions for epidemiological investigation of hospital-onset S. aureus infection [2].
The antibiogram has been the main typing tool in many hospital outbreaks since the technique is widely available and standardized, and it can be used with all microbial species. Its main disadvantage consists of the variability in resistance expression, which is also susceptible to instability due to horizontal transmission and loss of extrachromosomal genetic elements [1].
Molecular methods for MRSA typing as pulsed field gel electrophoresis (PFGE) have good discriminatory power and more reproducibility but are expensive, not widely available, and time consuming [3]. A method for molecular typing of MRSA that depends on polymerase chain reactionrestriction fragment length polymorphism (PCR-RFLP) has proven absolute typeability, reproducibility, and good discriminatory power [4].
Variations in the sequence of genes coding for two species-specific proteins, coagulase (coa) and staphylococcal 2 International Journal of Microbiology protein A (spa), have been the basis for the most widely used forms of PCR typing for S. aureus, showing a good correlation with PFGE typing [1].
The coagulase gene amplification discriminatory power relies on the heterogeneity of the region containing the 81 bp tandem repeats at the 3 coding region of the coagulase gene which differs both in the number of tandem repeats and the location of AluI and HaeIII restriction sites among different isolates [5]. Protein A has been coded by spa gene, where the repeated part is located at 3 end and identified as X region; the repetitive part of region X consists of up to 12 units each with a length of 24 nucleotides. This 24-nucleotide region is highly polymorphic with respect to the number and sequence of repeats. Diversity of X region causes protein A variation [6]. In addition to its use as a marker, the number of repeats in the region X of spa has been related to the dissemination potential of MRSA, with higher numbers of repeats associated with higher epidemic capability [1].
The aim of this study was to evaluate PCR-RFLP as a molecular typing technique for MRSA strains on the basis of protein A and coagulase gene polymorphisms and to verify their ability in assessing the relatedness of MRSA isolates.

Material and Methods
The material of the study consisted of a total of 100 S. aureus strains isolated from various clinical specimens obtained from patients attending different ICUs in Alexandria Main University Hospital (AMUH) and delivered to the routine laboratory of Microbiology Department at AMUH. All strains were identified as S. aureus according to the standard microbiological techniques [7].

MRSA Typing
(A) PCR of coa Gene [5]. The identified MRSA strains were subjected to PCR for coa gene detection, using the following primers: forward (5 -CGA GAC CAA GAT TCA ACA AG-3 ) and reverse (5 -AAA GAA AAC CAC TCA CAT CA-3 ), which were designed to amplify the 3 end hyper variable region containing 81 bp tandem repeats of coa gene. The amplification reaction consisted of an initial denaturation step at 94 ∘ C for 5 min, followed by 30 cycles of denaturation at 95 ∘ C for 30 sec, annealing at 55 ∘ C for 45 sec, extension at 72 ∘ C for 2 min, followed by final extension at 72 ∘ C for 7 min. [5]. Depending on the number of 81 bp repeats, a strain analysis of PCR RFLP products was performed with HaeIII restriction enzyme (New England BioLabs, Frankfurt, Germany), where 10 L of PCR product of coa gene was incubated with 6 U of the enzyme at 37 ∘ C for 1 h 45 min in a water bath.

RFLP of coa Gene PCR Products
(B) PCR of spa Gene [4]. The identified MRSA strains were subjected to PCR for spa gene detection using the following primers: forward (5 -ATC TGG TGG CGT AAC ACC TG-3 ), and reverse (5 -CGC TGC ACC TAA CGC TAA TG-3 ) which were designed to amplify the polymorphic X region that contains a variable number of 24 bp tandem repeats of the spa gene coding for protein A. Amplification reaction consisted of an initial denaturation step at 94 ∘ C for 4 minutes,followed by 35 cycles of denaturation at 94 ∘ C for 1 minute, annealing at 56 ∘ C for 1 minute, extension at 72 ∘ C for 3 minutes, followed by final extension at 72 ∘ C for 5 minutes. [4,5]. Five L of each spa gene amplicon and 10 units of HaeII restriction enzyme (New England BioLabs, Frankfurt, Germany) were incubated at 37 ∘ C for 3 hours. The PCR products and restriction digest fragments were detected by electrophoresis in 2% agarose gel.

RFLP of spa Gene PCR Products
The interpretation criteria for identifying different strains were a single band difference. Unique PCR-RFLP patterns were assigned a genotype.

Statistical Analysis: Determination of Numerical
Index of Discrimination. The probability that two unrelated isolates sampled from the test population will be placed into different typing groups or clusters was assessed according to the Hunter-Gaston formula [11]. This probability, also called discriminatory index, is calculated using the following equation: number of types obtained, and = the number of isolates belonging to this type.

Results
A total of 100 S. aureus strains isolated from different clinical specimens and different ICUs were identified according to the standard microbiological techniques. MRSA represented 75% of total S. aureus isolates, while MSSA represented 25% of the strains. Differentiation was based on sensitivity testing using oxacillin and cefoxitin discs, confirmed by the detection of the amplified 310 bp mecA gene using PCR in 75 MRSA strains ( Figure 1).
(3) spa Gene Typing. spa gene PCR products of different sizes could be detected in 71 out of the 75 MRSA isolates. The size of the PCR products ranged from 144 to 1392 bp (144, 192, 240, 360, 408, 600, 1104, 1200, 1296, and 1392), reflecting the number of 24 bp repeat units contained in the spa gene. These PCR products generated 4 major and 11 minor different spa gene types. Sixty-three MRSA strains (84%) showed single PCR band (type S1), 7 (9.3%) had two bands (type S2), and only one (1.3%) strain showed 3 PCR bands (type S3) ( Table 2 and Figure 4). The absence of PCR product is considered a separate type and this was shown in 4 strains (S4). DI value of spa gene PCR typing was 0.68.

(4) spa-RFLP Typing and HaeII Restriction Enzyme Digestion.
Restriction digestion performed on the amplified spa PCR products with HaeII revealed bands observed to be multiples of 24. Five distinct banding patterns (designated as A, B, C, D, and E) and 12 subtypes designated as [A1, 2/B1, 2, 3, 4, 5/C1, 2, 3/D/E] were produced (Table 2 and Figure 5). Most strains (65/75 = 86.6%) belonged to pattern B especially subtypes B4 (49.3%) and B2 (28%). PCR products of 2 strains were not digested by HaeII and 4 strains had no spa gene band; therefore, this method had 92% typeability. DI value of spa gene RFLP typing method was 0.68. (Table 3). The combination of both RFLP patterns was able to identify  International Journal of Microbiology

Discussion
MRSA is the major cause of nosocomial mortality and morbidity; it is commonly found in the community and hospital environment especially in the ICUs [12].
The present study was conducted on patients attending different ICU wards in AMUH. The backgrounds of the patients (ICU, age, and sex) were randomized to reduce the possibility of contact transmission and various specimens were used for MRSA isolation (respiratory system fluid or secretions, pus, blood, urine, and nasal swabs). Among 100 S. aureus isolated strains, 75 were proven to be MRSA by disc diffusion detection of mecA gene by PCR. These isolates were selected for further genotypic characterization.
None of our isolates was resistant to vancomycin or teicoplanin, but all were resistant to penicillin, with variable resistance patterns to the used antimicrobial panel. Using a different panel with few similar antibiotics, Baddour et al. [14] study showed different susceptibility patterns. The difference noted in the susceptibility of the MRSA isolates from different hospitals probably reflects the different patterns of antibiotic usage and thus development of resistance in these hospitals. Antibiotic resistance patterns are influenced by the local environment, selective antibiotic pressure, acquisition, and loss of plasmids carrying resistance genes and various other genetic mechanisms [15]. Sixteen of our isolates could be classified as extensively drug-resistant (XDR). The remaining 59 are multidrug-resistant (MDR) and none was pandrugresistant (PDR). Other studies documented the association of recovery of MDR-MRSA strains from inpatient clinical samples rather than from outpatients [16,17].
MRSA typing is an essential component of an effective surveillance system to describe epidemiological trends and infection control strategies. Antibiogram typing has been successfully used for screening of epidemic strains [15]. In the present study, 32 antibiotypes were identified.
Janwithayanuchit et al. [18] from Thailand identified 9 different antibiotypes using a panel of 10 antimicrobial agents. Antibiotypes 1 and 2 represented 44.2% (57/129) and 35.6% (46/129) of their isolates, respectively. Those were nearly similar to our antibiotypes 1 and 2 except in sensitivity to trimethoprim-sulfamethoxazole. They concluded that antimicrobial susceptibility testing is the simplest epidemiologic typing method, the results are easy to interpret, and minimal laboratory skills and equipment are required. Nevertheless, its main disadvantage consists of the variability in resistance expression, which is susceptible to instability due to loss of extra chromosomal genetic elements. However, antibiogram typing cannot be used as the only typing method for MRSA because of its poor discriminatory power [1]. 8 International Journal of Microbiology  Genotypic techniques to type MRSA must be particularly discriminatory, as MRSA strains probably originate from a single clone or at least a few strain types [13].
The coa gene is the principal criterion for the identification of S. aureus isolates [1]. Its 3 end contains an 81 bp tandem short sequence repeat (SSR) series, the number of which differs between strains [19]. The coa gene was identified in all of MRSA isolates in this study, indicating 100% typeability. The coa PCR showed 100% sensitivity and specificity in Tiwari et al. (2008) [20] study, confirming the fact that this gene is present in all S. aureus isolates.
In the present study, 9 different amplicons of the coa gene with band sizes ranging from 81 to 1215 bp were found, generating 3 different types and 10 subtypes of coa band patterns. The majority (54/75) of MRSA strains showed double bands, 16 (21.3%) showed single band, and the remaining 5 (6.7%) had three bands. The 405 bp band was the most common and was found in 57/75 of the isolates (76%). The presence of more than one band has been explained by the existence of more than one allelic form of coagulase gene, allowing one strain to produce one or more of these variants [21]. This gene polymorphism might be due to deletion or insertion mutations by which a portion of the 3 end region of the coa gene is deleted or several nucleotides are inserted and as a consequence change the coa gene size [22].
Himabindu et al. [5], using the same primer, showed that the sizes of coa PCR products were classified into 3 band classes. The majority of isolates belonged to the band class of 812 bp, which was close to our study results, where 69.3% of our isolates belonged to the same band class (810 bp). The difference in coagulase types was found to be subject to geographical variation [23].
According to Lawrence et al. [24], the high discriminative power of coa gene typing in their study was associated with the recovery of a 402 bp PCR product, which was always connected with detection of MRSA in culture. This was in agreement with our finding, where the majority (57 = 76%) of our isolates produced a 405 bp PCR band using the same primers.
In the present study, HaeIII restriction enzyme was chosen for coa gene PCR-RFLP, aiming at a more detailed characterization of the MRSA isolates. It generated 5 RFLP banding patterns (A, B, C, D, and E) and 12 subtypes [A1, 2, 3/B/C1, 2, 3/D1, 2, 3/E1, 2]. Four strains were not digested by the enzyme, indicating that not all the isolates genomic DNA contained HaeIII restriction sites; accordingly, this method had 94.6% typeability.
Lawrence et al. study [24] recovered a 402 bp PCR product which was cut into three fragments of 176 bp, 146 bp, and 81 bp after HaeIII digestion. This was different from our findings where the digestion of the PCR product of band size 405 bp by HaeIII enzyme was unsuccessful, along with the production of other bands (1215 bp, 891, 486, 243, 81) that also remained uncut which might explain the fewer number of genotypes obtained by coa PCR-RFLP in our study.
There is a high degree of heterogeneity in the coa gene of S. aureus making PCR-RFLP of coa gene not suitable as a single typing method [18]. The polymorphisms of the spa and coa genes were not independent, and combination use of them was found by many researchers to be a valid instrument for epidemiological studies [25].
The spa gene is not consistently expressed in all S. aureus isolates [26]. In the present study, spa gene was expressed in 71 isolates (94.6%). X region of spa gene is polymorphic, containing a varying number of 24 bp repeats, whose number and sequence differ among strains [26].
The sizes of the spa PCR products in the current study were classified into 11 classes, creating 4 major and 11 minor different spa gene types according to band number: 63 strains showed single PCR band, 7 had two bands, and only one strain showed 3 PCR bands. The absence of PCR product is considered a separate type and this was shown in 4 strains. This was also shown in Shakeri et al. [6] and Adesida et al. [27] studies, where spa gene was absent in 3.8% and 5% of their S. aureus isolates, respectively. Similar results were obtained by Schmitz et al. [13] and Montesinos et al. [1], who distinguished, by spa gene PCR, 5 types amongst the 183 and 4 types amongst 124 of their MRSA isolates, respectively.
In addition to its use as a marker, the number of repeats in the region X of spa has been related to the dissemination potential of MRSA, with higher numbers of repeats associated with higher epidemic capability [26]. S. aureus strains with shorter length of protein A cannot adhere to the surface of nasal epithelium and are discharged by breath, sneezing, and coughing [6]. This observation might explain our findings that MRSA strains with longer spa bands (>1200 bp = 1200-1392 bp) were isolated from respiratory secretions (85.6%) more than other samples, reflecting the role of spa protein in adherence to the respiratory epithelium, which protects the isolate from being discharged by breath, sneezing, and coughing.
There is also controversy over the proposed correlation between the number of repeats and the MRSA dissemination potential, with most of the epidemic strains reported as having more than seven repeats. The spread of MRSA strains between hospitals has been recognized in several studies, perhaps as a consequence of the prolonged carrier status and the increased mobility of the population [1]. In the present study, all but one spa gene positive MRSA isolates had more than 7 repeats (>168 bp), and all isolates were disseminated in different ICU wards in the hospital.
HaeII was the restriction enzyme used to digest spa PCR products in the present study, showing 5 distinct spa banding RFLP patterns (A, B, C, and D & E) and 12 minor patterns. The middle sized band (240 bp) remained the same in 67/75 isolates (89.3%).
Closer results were reported by Mehndiratta et al. [28], who reported 5 types of spa gene PCR-RFLP patterns of 1150-1420 bp fragments. However, their smallest fragment that remained the same in all patterns was of size 243 bp.
In Wichelhaus et al. study [4], the discriminatory power of PCR-RFLP of coa and spa genes using HaeII enzyme was found to be sufficiently high and corresponded reasonably well with PFGE. Wilailuckana et al. [29] also reported that typing results of MRSA isolates using combinations of different molecular methods provided a considerably usable tool for discrimination.
In the present study, the combination of both coa and spa genes PCR-RFLP patterns was able to identify 22 combined R types (R1-R22), where types R2 and R1 were the most frequent, including 27 and 13 MRSA isolates, respectively.
A similarly designed study by Mitani et al. [30] reported 8 spa and 6 coa types using HaeII restriction enzyme for coa and spa genes PCR-RFLP. Combination of these 2 techniques revealed 10 R types, where types R1-R4 were relatively frequent.
In the present study, typing using both coa PCR and coa PCR-RFLP had the same DI value (0.64), which is comparable to DI value of typing using both spa PCR and spa PCR-RFLP (0.68). Moreover, the combined grouping of coa PCR-RFLP and spa PCR-RFLP increased the discriminatory index of the typing procedure to 0.836 which is higher when compared to the highest DI of the individual typing technique. So it is desirable that RFLP of both genes should be used for better and reliable discrimination.
The DI value of the antibiotyping in the current study was the highest among all techniques used (0.94). The antibiotic abuse and the lack of antibiotic stewardship in our hospitals could explain the lack of correlation between antibiotypes and genotypes of our MRSA isolates.
This study demonstrates that, although PFGE remains the gold standard for MRSA typing, the typeability of PCR-RFLP products in the genotyping of MRSA strains was an attractive tool for routine epidemiological surveillance and infection control. We therefore recommend this method as a reliable and rapid system in epidemiological investigations similar to this study. Our study revealed that no single gene analysis is efficient in distinguishing strains within a heterogeneous species and testing for multiple gene polymorphisms is more useful for local epidemiologic or outbreak investigation purposes. The polymorphisms of the spa and coa genes are not independently discriminative, and the combined use of both was found to be a valid instrument for epidemiological studies. The variability in genes that are built from repetitive moieties might go along with enhanced speed of evolution. This typing system, therefore, is intended mainly for routine epidemiological surveillance, where isolates obtained within a narrow time frame are subject for typing analysis.