- Top of page
Rapid detection and typing of methicillin-resistant Staphylococcus aureus are important components of infection control programmes. A protocol is described that enables sequencing of the spa gene fragment directly from a multiplex PCR targeting the clinically relevant mecA, pvl and spa genes, resulting in high-throughput characterisation of S. aureus. Implementation of the method in the Danish national reference laboratory has markedly reduced the use of reagents and the requirement for hands-on time, and has also provided fast typing results. In addition, the method reduces the risk of sample mishandling.
The emergence and spread of strains of methicillin-resistant Staphylococcus aureus (MRSA) is a major healthcare concern in many parts of the world. MRSA surveillance should include, as a minimum, detection of the mecA gene and typing of the isolates, and detection of the pvl genes encoding Panton–Valentine leukocidin is often of interest [1,2]. In Denmark, all MRSA isolates are referred to the Statens Serum Institut (SSI), Copenhagen, for typing. In 2007, spa typing replaced pulsed-field gel electrophoresis at SSI as the primary typing method for MRSA because of its high discriminatory power and other advantages with respect to speed, interpretation and communicability. The present study describes a method that, in addition to conventional spa typing, detects the presence of the mecA and pvl genes in a single multiplex PCR. As spa typing involves DNA sequencing, the method also introduces the concept of sequencing one PCR fragment directly from among a mixture of PCR amplicons. Direct sequencing from the multiplex PCR minimises the number of PCRs required and the risk of sample mishandling.
DNA templates were prepared according to Kumari et al. . Each PCR contained 0.45 μM mecA primers (mecA P4, 5′-TCCAGATTACA ACTTCACCAGG; mecA P7, 5′-CCACTTCA TATCTTGTAACG) , 0.18 μM spa primers (spa-1113f, 5′-TAAAGACGATCCTTCGGTGAGC; spa-1514r, 5′-CAGCAGTAGTGCCGTTTGCTT) , 1 μM pvl primers (pvl-FP, 5′-GCTGGACAAA ACTTCTTGGAATAT; pvl-RP, 5′-GATAGGACACCAATAAATTCTGGATTG) , 1 × Multiplex PCR Master Mix (Qiagen, Valencia, CA, USA), 0.5 × Q-Solution (Qiagen) and 1 μL of DNA template preparation. Amplification was performed in a DNA Engine DYAD (Bio-Rad, Hercules, CA, USA), with 15 min at 94°C, followed by 30 cycles of 30 s at 94°C, 1 min at 59°C and 1 min at 72°C, with a final 10 min at 72°C. PCR products were visualised on E-Gels 2% w/v (Invitrogen, Carlsbad, CA, USA). The PCR products were vacuum-purified using NucleoFast 96 PCR plates (Macherey-Nagel, Easton, PA, USA) according to the manufacturer’s instructions. Sequencing of the amplicons with the PCR primers was performed as described previously using an ABI 3130 sequencer . Ridom StaphType (Ridom, Münster, Germany) and BioNumerics v.4.6 (Applied Maths, Sint-Martens-Latem, Belgium) software, together with the multilocus sequence typing database (http://www.MLST.net), were used for analysis and annotation of the sequences generated from the isolates.
The protocol was validated against 70 known MRSA isolates, 25 isolates of coagulase-negative staphylococci, and the three reference strains ATCC 6538 (methicillin-susceptible S. aureus), ATCC 33591 (MRSA) and ATCC 51625 (methicillin-resistant Staphylococcus epidermidis). Thirty-five of the 70 MRSA isolates had previously been confirmed as pvl-positive according to the PCR method described by Lina et al. . The multiplex assay was initially optimised as a duplex PCR containing the mecA and spa primers, and was later expanded to include the pvl primers. Both PCRs were validated with the collection of isolates described above.
The MRSA isolates had been assigned previously to clonal complexes (CCs) by pulsed-field gel electrophoresis, spa typing and, in some instances, multilocus sequence typing. The isolates belonged to 11 CCs (CC5, CC8, CC9, CC22, CC30, CC45, CC59, CC72, CC78, CC80 and CC152) and harboured SCCmec types I–VI. The known pvl-positive isolates belonged to nine different CCs, including the USA300-0114 (ST8-MRSA-IV), European CA-MRSA (ST80-MRSA-IV) and Southwest Pacific (ST30-MRSA-IV) clones (Fig. 1). Of the coagulase-negative staphylococcus isolates tested (11 S. epidermidis, seven Staphylococcus hominis, two Staphylococcus lugdunensis, two Staphylococcus capitis, two Staphylococcus haemolyticus and one Staphylococcus warneri), 11 were mecA-positive. The mecA status and species identifications were confirmed using the EVIGENE MRSA detection kit (SSI, Copenhagen, Denmark), and the RapidStaph32 system (bioMérieux, Marcy-l’Étoile, France).
Figure 1. Examples of isolates typed by the spa, mecA and pvl multiplex PCR. Lanes: 1 and 12, 100-bp ladder; 2, H2O control; 3, Staphylococcus epidermidis DK-14; 4, S. epidermidis ATCC 51625; 5, methicillin-susceptible Staphylococcus aureus (MSSA) ATCC 6538; 6, MSSA DK-E2211; 7, methicillin-resistant Stapylococcus aureus (MRSA) ATCC 33591; 8, MRSA USA300-0114; 9, MRSA CC80 (282-01); 10, MRSA CC30 (3250-01); 11, MRSA DK-55599 (one spa repeat).
Download figure to PowerPoint
The results of the initial validation showed complete concordance with the results obtained previously according to mecA, spa and pvl PCRs performed individually. The mecA and pvl amplicons from the multiplex PCR were sequenced to confirm the identity of the amplicons and to test the feasibility of direct sequencing of a particular amplicon from a multiplex PCR mixture.
Following the initial validation of the PCR method with only the mecA and spa primers, the duplex protocol was introduced routinely at SSI during January 2007. Testing of 759 routine samples identified 686 MRSA and 73 methicillin-susceptible S. aureus isolates. Sequencing revealed 102 different spa types among the MRSA isolates, of which ten were identified as new types according to the Ridom spa server (http://www.ridom.de). Of the 102 spa types, 79 were assigned to 16 different CCs, while 23 spa types (42 isolates) could not be assigned to CC groups. Isolates containing only one spa repeat (spa type t693; r07) were still detected correctly (Fig. 1, lane 11). Nine isolates failed to yield the spa gene fragment, but were confirmed as S. aureus by coagulase tests and by detection of the nuc and femA genes by PCR [8–10].
The multiplex method described in this report provides an efficient protocol for detecting three of the most important determinants used in surveillance of MRSA. The method confirms the presence of the mecA gene and detects the pvl genes in the same PCR used for conventional spa typing, which makes the procedure less expensive and laborious than performing separate reactions. Like conventional spa typing, the method is a fast, sequence-based typing method for S. aureus. Typing results are currently available in 2 working days, but could be made available in 1 day if facilities allow. Since January 2007, typing results have been reported twice-weekly from the laboratory, thereby providing timely results for local infection control purposes. Although the pvl genes are not restricted solely to community-acquired MRSA, pvl is present in all major community-acquired MRSA lineages and enables, e.g., distinction of USA300 isolates from other ST8 isolates [1,2,11].
To our knowledge, spa-sequencing directly from the products of a multiplex PCR has not been described previously. Other MRSA sequence typing protocols have involved other genes encoding surface proteins, e.g., clfB, in addition to spa, in order to improve the discriminatory power . The concept presented here of direct sequencing of a single cohort of PCR products from among a mixture of different amplicons can be transferred easily to such protocols, thus enabling, e.g., spa and clfB to be sequenced from the same multiplex PCR.