Loop-Mediated Isothermal Amplification Assay for the Rapid Detection of Staphylococcus aureus

Staphylococcus aureus, including methicillin-resistant S. aureus (MRSA), is an important human pathogen that produces a variety of toxins and causes a wide range of infections, including soft-tissue infections, bacteremia, and staphylococcal food poisoning. A loop-mediated isothermal amplification (LAMP) assay targeting the arcC gene of S. aureus was developed and evaluated with 119 S. aureus and 25 non-S. aureus strains. The usefulness of the assay was compared with the PCR method that targets spa and arcC genes. The optimal temperature for the LAMP assay was 58.5°C with a detection limit of 2.5 ng/μL and 102 CFU/mL when compared to 12.5 ng/μL and 103 CFU/mL for PCR (spa and arcC). Both LAMP and PCR assays were 100% specific, 100% sensitive, 100% positive predictive value (PPV), and 100% negative predictive value (NPV). When tested on 30 spiked blood specimens (21 MRSA, eight non-S. aureus and one negative control), the performance of LAMP and PCR was comparable: 100% specific, 100% sensitive, 100% PPV, and 100% NPV. In conclusion, the LAMP assay was equally specific with a shorter detection time when compared to PCR in the identification of S. aureus. The LAMP assay is a promising alternative method for the rapid identification of S. aureus and could be used in resource-limited laboratories and fields.


Introduction
Staphylococcus aureus, including methicillin-resistant S. aureus (MRSA), is an important bacterial pathogen associated with community and healthcare S. aureus infections in Malaysia and worldwide. ey are known to produce a variety of virulence factors, including the staphylococcal enterotoxins (SEs) and the toxic shock syndrome toxin (TSST) which are responsible for staphylococcal food poisoning [1,2].
Polymerase chain reaction (PCR) and real-time PCR have been used for rapid identi�cation of S. aureus, particularly for MRSA [3,4]. Both methods require the use of special equipment, that is, PCR thermocycler or real-time PCR, respectively.
Loop-mediated isothermal ampli�cation (LAMP) which is based on autocycling strand displacement DNA synthesis using the Bst DNA polymerase enzyme was developed by Notomi et al. [5]. is Bst DNA polymerase is derived from Bacillus stearothermophilus, which possesses a 5 ′ → 3 ′ exonuclease activity that needs high concentration of magnesium for optimum activity [6]. is enzyme can be inactivated by incubation at 80 ∘ C for 15 min. In the LAMP method, four (B3, F3, FIP, and BIP) to six (B3, F3, FIP, BIP, Loop-F, and Loop-R) primers are used for ampli�cation, and the reactions are carried out at isothermal condition [7,8]. Loop primers that bind to loop structures are used to shorten the reaction time of the LAMP assay [9]. e LAMP assay has been developed for rapid identi�cation of a wide variety of bacteria, including S. aureus, Vibrio parahaemolyticus, Campylobacter jejuni, Campylobacter coli, Leptospira species, Salmonella Typhi, and Escherichia coli [10][11][12][13][14][15].
e objective of this study was to develop and determine the usefulness of the LAMP assay in comparison with PCR for rapid identi�cation of S. aureus.
All MRSA and MSSA strains were previously isolated from inpatients admitted to a local tertiary hospital. e MSSA strains were previously con�rmed by BBL Staphyloslide-Latex Test (BD, USA) [17] while MRSA strains were con�rmed by standard biochemical tests [16] with minor modi�cations. In brief, all MRSA strains were streaked on blood and mannitol salt agars and incubated overnight at 35 ∘ C. Strains that showed -hemolysis in blood agar and appeared as yellow colonies in mannitol salt agar were subjected to coagulase test and cefoxitin disk diffusion test. e strain was con�rmed as MRSA strain if it gave a positive result in the coagulase test and has zone diameter of less 22 mm in cefoxitin disk diffusion test [16].

Preparation of DNA Template.
Crude DNA from Gramnegative bacteria such as E. coli, Salmonella Typhimurium, and Shigella sonnei was obtained by direct boiled cell lysate. Brie�y, a loopful of bacterial culture was mixed with 100 L of sterile-deionized water, and the suspension was boiled in 100 ∘ C for 5 min. Aer boiling, the cell lysate was snapped cool for 5 min in ice and centrifuged at 10,000 ×g for 2 min. e supernatant was transferred to a clean microfuge tube and used as DNA template for LAMP and PCR analyses. Crude DNA from Gram-positive S. aureus and S. epidermidis was obtained by using lysostaphin-lysis method while crude DNA for L. monocytogenes was obtained by using a lysozyme-lysis method. Procedures for lysostaphin and lysozyme-lysis methods were very similar to the boiling method except for the addition of lysostaphin (2 g/mL) or lysozyme (20 g/mL) in the bacterial culture and incubation at 37 ∘ C for 10 min. e DNA was quanti�ed by using Eppendorf BioPhotometer (Eppendorf Ltd, Germany).
Meanwhile, crude DNA from spiked blood samples was obtained by centrifugation and washing the residue twice with deionized water followed by direct cell lysis. Brie�y, 100 L of spiked blood sample was mixed with 900 L of sterile deionized water and the suspension was centrifuged at 10,000 ×g for 3 min. e supernatant was discarded and the pellet was resuspended in 1 mL of sterile deionized water before they were centrifuged at 10,000 ×g for 3 min. e supernatant was removed and the pellet was mixed with 300 L sterile deionized water. Lysostaphin (2 g/mL) was added to the cell suspension and was incubated at 37 ∘ C for 5 min, followed by heating at 100 ∘ C for 10 min. Aer boiling, the cell lysate was snapped cool for 5 min in ice. e cell lysate was then centrifuged and the supernatant was transferred to a clean microfuge tube and used as a DNA template.

Optimization of the LAMP Assay Using Different Temperatures.
Optimization of the LAMP assay was performed on a reference strain MRSA0807-7 by using Loopamp DNA ampli�cation kit (Eiken Chemical Co. Ltd., Tokyo, Japan). e kit is based on the use of four kinds of primers (two inner and two outer) that recognize six distinct regions of a target gene in the presence of Bst DNA polymerase with strand-displacement activity along with substrates, mixture of samples with incubation at a constant temperature (http://www.loopamp.eiken.co.jp/e/lamp/index.html).
Brie�y, 25 L of the reaction mixture containing 4 L of each FIB and BIP primer (equivalent to 40 pmol of each FIB and BIP primer), 0.5 L of each F3 and B3 primer (equivalent to 5 pmol of each F3 and B3 primer), 12.5 L of 2X reaction mixture (provided in the kit), 1 L of Bst DNA polymerase (provided in the kit), 1.5 L of sterile deionize water, and 1 L of DNA template (5 ng) was used.
e reaction mixtures were incubated at different temperatures ranging from 56 ∘ C to 63 ∘ C for 80 min, followed by enzyme inactivation at 80 ∘ C for 2 min in the Loopamp real-time turbidimeter (LA-320, Teramecs, Co., Ltd., Kyoto, Japan). is Loopamp real-time turbidimeter is speci�cally designed for real-time monitoring of the LAMP reaction. e reaction is considered positive when the turbidity reached 0.1 within 60 min at 650 nm. e time needed for the turbidity of each tested sample to exceed OD 650 nm at 0.1 is referred to as the threshold time (Tt) [19].

Evaluation of the LAMP Assay on Bacterial Cultures.
e optimized LAMP assay was performed on all 79 MRSA, 20 MSSA, 5 S. epidermidis, 5 Salmonella Typhimurium, 5 Shigella sonnei, 5 E. coli, and 5 L. monocytogenes by using Loopamp DNA ampli�cation kit (Eiken Chemical Co. Ltd., Tokyo, Japan). Following the optimization temperature (refer to result), the ampli�cation of LAMP assay was performed at 58.5 ∘ C for 80 min and followed by 80 ∘ C for 2 min.
In addition, a positive result could also be determined by direct visualization of turbidity of the mixtures or by the formation of a white precipitate at the bottom of the microfuge tube aer centrifugation at 10,000 ×g at 3 min.

PCR Detection of spa and arcC Genes.
In parallel, PCR targeting spa and arcC genes was performed on 79 MRSA, 20 MSSA, 5 S. epidermidis, 5 E. coli, 5 Salmonella Typhimurium, 5 Shigella sonnei, and 5 L. monocytogenes as previously described by Harmsen et al. [20] and Enright et al. [18], respectively. Brie�y, two Monoplex PCR were carried out in a 25 L volume containing 1.4 mM MgCl 2 , 35 mM each deoxynucleoside triphosphate, 0.3 M of each 1095F and 1517R primers (for spa gene) or 0.3 M of each arcC-1 and arcC-2 primers (for arcC gene), 1X Green buffer, 0.5 U Taq DNA polymerase (Promega, Madison WI, USA), and 5 L of DNA template. Following PCR, 5 L aliquots of each sample were subjected to electrophoresis on 1.5% agarose gels. Representative amplicons of spa and arcC genes were puri�ed by using �iagen DNA puri�cation kit (�iagen GmBH, Germany) and sequenced to validate their identities.

Determination of the Detection Limit (Sensitivity) in
LAMP and PCR Assays Using Pure Culture. To determine the detection limit (sensitivity) of LAMP and PCR assays, DNA template of MRSA0807-7 was serially diluted 10-fold with sterile water to 10 −1 to 10 −8 concentrations. One L of genomic DNA from each dilution was used as a DNA template for the LAMP and another 5 L of genomic DNA was used as a DNA template for the PCR assay.
To determine the detection limit in terms of CFU, a 6 hr culture of MRSA0807-7 was serially diluted 10-fold with 0.85% normal saline. An aliquot of 100 L of each dilution was plated on a Tryptone Soy Agar (TSA) plate to enumerate the bacterial count and another 100 L was used for DNA preparation for LAMP and PCR tests.

Preparation of Spiked Blood Samples and LAMP Assay.
Blood samples used in this study were obtained from a healthy donor. e blood samples were stored at 4 ∘ C in K2-EDTA anticoagulated blood tube. Different microorganisms, including 21 MRSA, 4 Salmonella Typhimurium, 2 Klebsiella pneumoniae, 2 E. coli, and 1 sterile distilled water without bacterial DNA were used for sensitivity and speci�city evaluation.
An aliquot of 100 L of a 6 hr culture growth (OD 600 nm at 1) was serially diluted 10-fold dilutions in 0.85% of normal saline. An aliquot of 100 L of each dilution was plated on a Tryptone Soy Agar (TSA) plate to enumerate the bacterial count. In parallel, an aliquot of 100 L of each dilution was spiked into a 900 L of blood samples and incubated at 37 ∘ C overnight. An aliquot of 100 L of spiked blood samples was then plated on TSA plate for the determination of CFU count and another 100 L was used for DNA preparation as described earlier.

Optimization of the Conditions for the LAMP Assay.
Crude DNA from the reference MRSA0807-7 strain was used as the target template in order to determine the optimal condition for the LAMP assay. Ampli�cation with primers F3, B3, FIP, and BIP by LAMP assay yielded a positive result at 58 min under isothermal condition of 60 ∘ C when measured by using real-time turbidimeter (LA-320, Teramecs, Co., Ltd., Kyoto, Japan). No difference was observed when the LAMP assays were performed under isothermal condition between 60 ∘ C and 62 ∘ C. No ampli�cation was observed at 63 ∘ C.
e product ampli�ed at 58.5 ∘ C yielded positive result at 52 min when compared to other temperatures. Although a positive reaction was detected using different reaction temperatures between 60 ∘ C and 62 ∘ C, we found out that 58.5 ∘ C was the optimum temperature for the LAMP assay since positive reaction was detected at 52 min when compared to 58 min using reaction temperatures of 60 ∘ C to 62 ∘ C. Hence, a reaction temperature of 58.5 ∘ C was selected as the optimum temperature for LAMP assay.

Detection Limit (Sensitivity) and Speci�city of LAMP
and PCR Assays. e sensitivity of LAMP and PCR assays was done by using both template DNA and minimal CFU of bacteria. e detection limit for the LAMP assay was 2.5 ng/ L and 10 2 CFU/mL while the detection limit for the PCR assay was at 12.5 ng/ L (since 5 L of DNA template was used in the PCR assay when compared to 1 L of DNA template used in the LAMP assay) and 10 3 CFU/mL.
As for the speci�city test, LAMP assays were performed on all 124 isolates including MRSA, MSSA, S. epidermidis, L. monocytogenes, Salmonella Typhimurium, and Shigella sonnei. All 99 S. aureus strains gave positive results in LAMP assay with Tt values of 52 min 48 sec to 55 min. No Tt value was observed for all non-S. aureus strains indicating negative results.
Aer centrifugation, all the reaction tubes that contained MRSA and MSSA showed a white precipitate at the bottom of the tube. No white precipitate was observed in tubes containing E. coli, S. epidermidis, Shigella sonnei, L. monocytogenes, and Salmonella Typhimurium.

3.�. Sensitivity and Speci�city of the LAMP Assay and PCR on
Spiked Blood Samples. irty known bacterial cultures were blind-coded and spiked in blood samples. ese samples were tested with the optimized LAMP and PCR assays.
Different concentrations of bacterial cell cultures ranging from of 1.5 × 10 2 CFU/mL to 1.5 × 10 8 CFU/mL were spiked into each blood sample. e minimal detectable concentration for LAMP was 1.76 × 10 2 CFU/mL and for PCR was 1.76 × 10 3 CFU/mL. Twenty-one S. aureus strains yielded positive results with Tt values between 53 and 55 min while non-S. aureus strains yielded negative results.
e data was compared with PCR results and the identity of the strains. Both LAMP and PCR were 100% speci�c, 100% sensitive, 100% positive predictive value (PPV), and 100% negative predictive value (NPV).

Discussion
S. aureus, including MRSA, is a persistent human pathogen responsible for a variety of infections ranging from sotissue infections to bacteremia [21]. us, an accurate and rapid detection of S. aureus is much needed to reduce the risk factor caused by this organism. e identi�cation of S. aureus by a conventional bacterial culture test oen requires 1 to 2 days with the plating on blood agar and a series of biochemical tests, including the coagulase test. Although PCR assay shortens the identi�cation time to 4 to 5 hours, this technique requires special equipment such as PCR thermocycler, electrophoresis set, and gel documentation system.
In this study, LAMP assay was used for the rapid iden-ti�cation of S. aureus from pure cultures and spiked blood specimens. Two pairs of primers (inner and outer) were used in the LAMP assay, and the whole reactions occur in a single tube containing of Bst DNA polymerase, DNA templates, and reaction buffers under a constant temperature. erefore, denaturation of DNA template could be omitted in the LAMP assay.
e amplicons from the LAMP assay could be quantitatively pro�led and measured by using the Loopamp turbidity meter which revealed the ampli�cation kinetics of the tested strains. We also found out that the LAMP assay performance could be carried out at wider reaction temperatures ranging from 58 ∘ C to 62 ∘ C using 60 min as the cut-off time (standard isothermal ampli�cation protocol).
In this study, Loopamp real-time turbidimeter was used in the LAMP assay whereas PCR thermocycler, gel electrophoresis, and gel documentation systems are needed for PCR analysis. Even though both approaches use special equipment, the LAMP assay holds an advantage as the positive result of the LAMP assay could be viewed through the Loopamp real-time turbidimeter screen. Furthermore, Loopamp real-time turbidimeter was set to measure the concentration of the tube at six-second intervals, and therefore, we are able to con�rm the ampli�cation threshold time on the screen even while the reaction is still ongoing. Unlike LAMP, the PCR assay requires conventional PCR ampli�cation using a PCR thermocycler followed by agarose gel electrophoresis and staining with either ethidium bromide or gel-red before we are able to view the result through the gel documentation system. erefore the endpoint detection of the LAMP assay is simpler and more direct.
Although both LAMP and PCR assays showed 100% speci�city, the LAMP assay produced higher sensitivity when compared to the PCR assay as the detection limit of the LAMP assay was 2.5 ng/ L which was approximately �ve times more sensitive than PCR assay. e whole process of the LAMP assay from the preparation of DNA template to endpoint detection only required 80 min even whereas the conventional PCR required 4 hours starting from DNA template to the visualization of the amplicons using agarose gel electrophoresis. Furthermore, the present LAMP method also allowed rapid identi�cation of S. aureus in positive blood samples within 80 min which is shorter than 2 hours as reported earlier by Misawa et al. [22]. e detection limit of LAMP assay using the spiked blood sample was comparable with pure cultures. is is not surprising as Kaneko et al. [23] have reported that LAMP assay is more tolerant towards inhibitory substances in the clinical samples. However, Wang et al. [24] reported that the usefulness of the LAMP assay could be limited by the presence of inhibitors in raw milk.
In addition, positive results could be observed aer 52 min even without the use of loop primers indicating that it was not mandatory to include loop primers in the assay. is reaction time was shorter than the 60 min reported by Wang et al. [15] since the whole reaction of the LAMP assay was carried out under isothermal condition; time loss due to thermal changes could be prevented.
Another advantage of this LAMP assay was that it only requires a heating block, which is easily available in most laboratories. Furthermore, the result or product from the LAMP assay could be determined by unaided eye without requiring special equipment. White precipitate could also be easily seen at the bottom of the tube aer centrifugation at 10,000 ×g at 3 min indicating the positive result. ese white precipitates contained white magnesium pyrophosphate, which is generated during the process of strand displacement of auto-cycling reaction [14].
In conclusion, the LAMP assay was equally speci�c when compared to PCR in the identi�cation of S. aureus. By using optimum conditions, it is not necessary to include loop primers in the lamp assay. LAMP assay is a rapid, �exible, and simple tool for the identi�cation of S. aureus isolates.
�on�ict of �nterests e authors declare that there is no con�ict of interests.