A Simple Sample Processing Protocol and Multiplex PCR for Direct Detection of MRSA from Uncultured Clinical Samples — A Pilot Study

Phenotypic tests have limited discrimatory power to identify closely related members of genus Staphylococcus and particularly for identification of S. aureus. 157 isolates of S. aureus obtained from different clinical specimens were included in our study. To present a demonstration of our method’s sensitivity and ability to correctly detect S. aureus from uncultured clinical specimen, 30 known S. aureus positive but leftover uncultured clinical specimens from clinical microbiology laboratory were processed by our protocol and analyzed. All the 30 clinical specimens were confirmed as S. aureus among which 26 specimen were identified as MRSA and the remaining 4 as MSSA. These 30 clinical specimens used in the study showed 100% correlation with coagulase test and Cefoxitin disc diffusion method. Though commercial molecular diagnostic kits are available for detecting MRSA from swabs, this is probably the first time that multiplex PCR is being demonstrated directly on a variety of uncultured clinical specimens.


Introduction
S. aureus is a common pathogen causing minor skin infections to major life threatening diseases such as osteomyelitis, pneumonia and septicaemia [1] [2].
Accurate and rapid identification of S. aureus and MRSA directly from clinical samples is essential for the proper management of patients with bacteraemia, poisoning, toxic shock syndrome and the scalded skin syndrome, 2) S. epidermidis, a member of the common skin flora, which causes infections associated with devices, such as catheters and prosthetics and 3) S. saprophyticus, which causes urinary tract infections [11].Single phenotypic tests are inefficient for the identification of S. aureus.Indeed, mannitol salt agar (MSA) positive CoNS (Staphylococcus caprae, S. hemolyticus and S. saprophyticus) have been reported in Nigeria and Japan [12] [13].However, a combination of methods like isolation on MSA and screening by DNase agar improves the outcome [14].Thus, in certain settings, if used individually to identify Staphylococcus aureus, common phenotypic tests may be inconclusive; some isolates will be misidentified.The use of MSA prior to tube coagulase/DNase is highly recommended due to the clumping factor negative and tube coagulase positive Staphylococci that are increasingly being recovered from human infections [15].These isolates also produce a heat stable DNase and can be misidentified as S. aureus.However, these strains can be differentiated from S. aureus by their failure to produce acid from maltose, lactose and mannitol.Furthermore, rare strains of S. aureus can be coagulase negative, some Staphylococcus isolates from animals (S. intermedius, S. hyicus, S. delphini and S. schleiferi subsp.coagulans) are clumping factor negative but tube coagulase positive [16] [17] differentiation of which requires isolation on MSA also.
Methicillin resistance in Staphylococci is conferred by mecA gene that produces altered PBP2a.Detection of mecA gene remains the gold standard for identification of methicillin resistance; however it does not confirm the species S. aureus [18].There is no consensus on the genomic target that could be used to confirm the S. aureus.A number of auxiliary factors which influence methicillin resistance by regulating cell wall metabolism have been used by different laboratories to identify S. aureus.Notable among them are the femA or femB and femX (factor essential for methicillin resistance) genes [19] [20].However, failure to confirm the species of S. aureus as reported by others earlier [19] [21] [22] and our own report downplays the reliability of femA or femB as genomic target in species identification [23].The exact reason for the false negative results with fem genotyping is not yet known.
The speed with which MRSA is detected has a significant role to play in any successful strategy to impede the pathogen from dissemination.Since MRSA detection by culture requires 2 -3 days, quick detection techniques using PCR Advances in Infectious Diseases methods have made headway.The execution of these rapid tests minimizes the time of detection of MRSA from 48 -72 [24] to 2 -5 h [25] [26].Clinical evaluation data have shown that MRSA can thus be detected with very high sensitivity.
Currently there are commercial kits which can only detect MRSA (S. aureus) from nasal swabs and positive blood cultures.Presently available molecular tests which are PCR based for MRSA detection include the HyplexStaphyloResist PCR (BAG, Lich, Germany), the GenoType MRSA direct assay (Hain Lifescience, Germany), LightCycler Staphylococcus and MRSA detection kit (LC assay; Roche Diagnostics, Germany), the IDI-MRSA assay (GenOhm, San Diego, BD Diagnostics), and the recently introduced GeneXpert MRSA assay (Cepheid, Sunnyvale).Rossney et al. [26] [27].All these kits which could detect MRSA rapidly and were easy to use had major limitations like being very expensive, could be performed only in swabs of nasal, groin and blood samples and results need to be compared with culture.In addition, the Xpert MRSA assay requires more interpretation than currently suggested by the manufacturer therefore more expertise is required.Table 1 shows the kits used in MRSA identification and their limitations in detail.It may be noted that many of these kits are not validated on clinical samples other than swabs.In this study we used nuc gene as genetic marker for PCR amplification to identify clinical isolates of S. aureus in comparison to some of the conventional phenotyping methods.We also designed a simple sample processing protocol and multiplex PCR using nuc as species marker instead of fem sequence.We have demonstrated in this study the potential of the sample processing protocol and multiplex PCR on 30 uncultured left over but characterised clinical samples as a pilot study.

Culture Isolation and Characterization
157 isolates of S. aureus obtained from different clinical specimens were included in our study.S. aureus colonies were identified by standard microbiological tests which included isolation on mannitol salt agar, coagulase test and DNase test.All S. aureus isolates were screened with Cefoxitin (30 µg) and Oxacillin (1 µg) disc on Mueller Hinton Agar to identify MRSA and MSSA.Thermonuclease activity was measured by streaking isolates on Methyl Green DNase Agar (Himedia Pvt Ltd) plates, and measure the zone of clearance after 24 h at 37˚C.Advances in Infectious Diseases 3) Not commercially available yet. [29]

Genotyping of Clinical Isolates of S. aureus
DNA was isolated [28] from a few isolates of S. aureus for initial optimization of PCR, precipitated with isopropanol and finally dissolved in 10 mM Tris-EDTA buffer (pH8.0).For subsequent screening of all the isolates, cell free DNA lysate was prepared by TEX (Tris buffer pH 8.0-EDTA-Triton X-100) method [29] Primers and the thermal cycling conditions are detailed out in Table 2. Staphylococcus genus was confirmed with 16S rRNA [30] and MRSA was confirmed by the detection of mecA [31] while nuc was evaluated as genomic target for species identification [32] [33] against microbiology methods.

Processing Uncultured Specimens
A total of 30 clinical specimens confirmed to contain S. aureus including swabs, endotracheal secretions and pus samples collected from Microbiology laboratory.These were the left over samples and collected after 24 h of storage at 4˚C.
All of them were processed by the TEX method and the DNA lysates were stored at −20˚C until use.The flowchart of processing the samples is as shown in the scheme Figure 1.Advances in Infectious Diseases  Three sets of primers were used, one for nuc (species specific gene), one for 16S rRNA (genus specific gene) and another one for mecA (methicillin resistance gene).The reaction conditions for the multiplex PCR are described in Table 1.The PCR reaction mixture in a volume of 20 µL contained the following;

Results
All isolates were screened and confirmed by coagulase test.157 coagulase positive isolates were included as S. aureus in this study.We screened these isolates for methicillin resistance (MRSA) by Cefoxitin and Oxacillin Disc diffusion test, 105 isolates were MRSA and 52 were MSSA.However, in genotyping, 106 isolates were identified as MRSA (mecA-PCR positive) and the remaining 51 isolates were classified as MSSA (mecA-PCR negative).One isolate which was indeterminate (22 mm) by Cefoxitin disc method was found to harbour mecA.The Oxacillin disc diffusion test did not compare well with mecA PCR test; 67% sensitivity, 94% specificity, 96% positive predictive value (PPV) and 58% negative predictive value (NPV).Cefoxitin disc diffusion test compared well with mecA PCR with sensitivity of 99.06%, specificity 100%, Positive Predictive Value of 100%; and a Negative Predictive Value of 98%.The results are presented in Table 3(a) and Table 3(b).

Genotyping of S. aureus (n = 30) from Uncultured Clinical Specimen
We optimised a triplex PCR for detection of mecA, nuc and 16S rRNA with 31 uncultured left over specimen from microbiology laboratory.This multiplex PCR produced distinct amplicons of expected size for mecA (162 bp), nuc (270 bp) and 16s rRNA (886 bp) when analysed on agarose gel (Figure 3).Multiplex PCR result of other bacterial isolates is presented in Figure 4 and Figure 5. Results of our multiplex PCR were compared with microbiology laboratory results.
All the uncultured specimens were correctly identified as S. aureus, 26 of them tested positive for mecA correlating with phenotype (Cefoxitin disc diffusion) as MRSA.This is probably the first time that multiplex PCR is being demonstrated directly on a variety of uncultured clinical specimens.

Discussion
Thermostable nuclease gene nuc was reported to have 100% sensitivity and specificity for the identification of S. aureus isolate [32] [33].A few studies have employed femA and nuc along with mecA as molecular targets for identification of S. aureus and characterization of MRSA [34].Some commercial kits are available for directly detecting MRSA directly from nasal swabs and blood cultures [34] (Table 1).Several kits target orfX and sequence near to SCCmec region in the genome while one kit (BDMaxStaphSR kit, Becton Dickinson) also targets nuc in addition to mecA and orfX [35].We reported poor sensitivity when femA was used as species identification genetic marker in PCR [23].Therefore, we evaluated nuc as species specific marker along with mecA and 16S rRNA simultaneously to identify MRSA from methicillin resistance non-S.
aureus species.A non-S.aureus methicillin resistant species is indicated when our multiplex PCR result is negative for nuc but positive for mecA and 16S rRNA targets.uncultured clinical samples 2 were endotracheal secretions, which are usually known to harbor mixed microbial populations.Our protocol worked with these endotracheal secretions also and identified correctly the pathogen as S. aureus.
We first wanted to demonstrate that our protocol works well with uncultured clinical specimen with results comparable to conventional microbiology.Our intention was to show that adopting such a protocol would enable same day reporting (6 -8 h) of MRSA status of a given sample to the clinician thus facilitating a quick therapeutic decision making.Our protocol requires an extensive evaluation and validation to determine the diagnostic sensitivity, specificity and predictive values which are in progress.

Figure 1 .
Figure 1.Flow chart for processing of uncultured clinical samples.

1. 5 mM
MgCl 2 , 30 pmol of each primer of mecA and nuc, 10 pmol of primers of 16S rRNA, 200 µmol of dNTP along with 1 U of Taq (KAPA Taq DNA Polymerase, KAPA Biosystems Inc.) and 10 µL of cell free lysate as DNA template.S. aureus ATCC 6538P was used as positive control MRSA (PC).Other gram posi-
evaluated Xpert MRSA assay, which is run on the

Table 1 .
Comparison of commercial MRSA detection kits.

Table 2 .
Primers used for study.