Methicillin, the first semisynthetic penicillin to be developed, was introduced in 1959 to overcome the problem of penicillin-resistant Staphylococcus aureus
due to β-lactamase (penicillinase) production (17
). However, methicillin-resistant S. aureus
(MRSA) strains were identified soon after its introduction (2
). MRSA strains have acquired and integrated into their genome a 21- to 67-kb mobile genetic element, termed the staphylococcal cassette chromosome mec
), which harbors the methicillin resistance (mecA
) gene and other antibiotic resistance determinants (10
). Since then, strains of MRSA have spread and become established as major nosocomial pathogens worldwide (1
). Recently, these organisms have evolved and emerged as a major cause of community-acquired infections (16
). These newly emerging community-acquired MRSA strains possess novel, small, mobile SCCmec
type IV or V genetic elements which contain the mecA
gene with or without additional antibiotic resistance genes and are more easily transferred to other strains of S. aureus
than larger SCCmec
(types I, II, and III) elements (22
). The emerging spread of these community-acquired MRSA strains poses a significant threat to public health (16
is a mobile genetic element characterized by the presence of terminal inverted and direct repeats, two essential genetic components (the mec
gene complex and the ccr
gene complex), and the junkyard (J) regions (10
). The mec
gene complex is composed of IS431mec
, and intact or truncated sets of regulatory genes, mecR1
. The ccr
gene complex encodes the recombinases (ccr
) that mediate the integration of SCCmec
into and its excision from the recipient chromosome and are, therefore, responsible for its mobility. The rest of the SCCmec
element is comprised of J regions (J1, J2, and J3) that are located between and around the mec
complexes and contain various genes or pseudogenes whose does not appear to be essential or useful for the bacterial cell, although notable exceptions include plasmid- or transposon-mediated resistance genes for non-β-lactam antibiotics or heavy metals (13
). To date, there are three classes (A, B, and C) of mec
complex and four allotypes (types 1, 2, 3, and 5) of ccr
complex. Different combinations of these complex classes and allotypes generate various SCCmec
types (Table 1
elements are currently classified into types I, II, III, IV, and V based on the nature of the mec
gene complexes, and are further classified into subtypes according to differences in their J region DNA (10
A thorough understanding of the molecular epidemiology and evolution of MRSA is required to help detect, track, control and prevent human disease due to this organism. Full characterization of MRSA requires definition of not only the putative bacterial genetic background but also of the complex and heterologous SCCmec
typing is an important molecular tool and its importance in community clonal outbreaks is increasingly being recognized (16
). The new MRSA nomenclature scheme recently set by the International Union of Microbiology Societies incorporates SCCmec
typing information in conjunction with that provided by multilocus sequence typing (7
). Previously described SCCmec
traditional PCR typing schemes have required the use of many (20 to 30) primer sets and multiple individual PCR experiments (23
). The only previously described multiplex PCR assay for SCCmec
) is more difficult to interpret and is limited in its ability to detect SCCmec
subtypes IVa, b, c, and d plus the newly described type V, these groups being implicated in currently emerging community MRSA outbreaks (12
Here we report a recently developed novel multiplex PCR assay capable of detecting and classifying all currently described SCCmec types plus the major subtypes, with simultaneous discrimination of MRSA from methicillin-susceptible S. aureus (MSSA).
Identification and selection of unique and specific loci and primer design for SCCmec types and subtypes.
To design the SCCmec
type- and subtype-unique and specific primers, an extensive BLAST sequence similarity search was conducted and was followed by comprehensive analyses and alignments of the S. aureus
and MRSA genomes and SCCmec
sequences currently available in the GenBank database. These loci consisted of open reading frames (ORFs) or sequence fragments, including ORF E008 of strain NCTC10442 (AB033763), kdpE of strain N315 (D86934), ORF CZ049 of strain 85/2082 (AB37671), ORF CQ002 of strain CA05 (AB063172), ORF CM001 of strain 8/6-3P (AB063173), ORF CR002 of strain MR108 (AB096217), ORF CG001 of strain JCSC4469 (AB097677), and ORF V011 of strain JCSC3624 (AB12121), and were found to be unique and specific for SCCmec
types and subtypes I, II, III, IVa, IVb, IVc, IVd, and V, respectively (refer to Materials and Methods for details of gene targets, strains, and accession numbers). The corresponding SCCmec
type- and subtype-unique and specific primers were designed (Table 2
) and their uniqueness and specificity were further confirmed with a GenBank database BLAST search. Utilization of these primers in our novel M-PCR assay allowed us to specifically detect the currently described SCCmec
types and subtypes of MRSA strains and clinical isolates.
New M-PCR for typing and subtyping SCCmec types I to V, and simultaneous detection of methicillin resistance (mecA gene).
We developed a new and simple single M-PCR assay to determine (classify) SCCmec types and subtypes I, II, III, IVa, IVb, IVc, IVd, and V, and simultaneously discriminate MRSA from MSSA. The M-PCR assay targeted the unique and specific loci of SCCmec types and subtypes I, II, III, IVa, IVb, IVc, IVd, and V, with concomitant mecA gene detection, the last serving as a determinant of methicillin resistance but also serving as an internal positive control for the assay.
To ensure the individual primer pairs were adequate for the amplification of all nine loci (gene fragments), the single-target PCR protocol with each individual primer pair was conducted prior to the M-PCR optimization, using eight SCCmec
standard control strains: type I (NCTC10442), type II (N315), type III (85/2082), type IVa (CA05), type IVb (8/6-3P), type IVc (MR108), type IVd (JCSC4469) and type V (WIS [WBG8318]-JCSC3624) (Tables 1
). Each individual PCR amplification reaction yielded the fragment of the expected size, i.e., 613, 398, 280, 776, 493, 200, 881, 325, and 147 bp for the unique and specific loci of SCCmec
types and subtypes I, II, III, IVa, IVb, IVc, IVd, and V, and mecA
gene in their corresponding strains, respectively. The optimized M-PCR condition as described above was obtained through assaying different primer concentrations and other PCR components. Amplification in a single M-PCR produced distinct bands corresponding to their respective molecular sizes that were easily recognizable in agarose gels stained with ethidium bromide (Fig. 1
Sensitivity of M-PCR.
The sensitivity of our M-PCR assay was examined in 8 SCCmec standard control strains for type I (NCTC10442), type II (N315), type III (85/2082), type IVa (CA05), type IVb (8/6-3P), type IVc (MR108), type IVd (JCSC4469) and type V (WIS [WBG8318]-JCSC3624). This assay was capable of detecting, with reproducibility, a band in ethidium bromide-stained gels at dilutions corresponding to 6 × 104 CFU per PCR for all eight type- and subtype-specific genes. However, the sensitivity for the internal control mecA gene varied slightly depending on the strains examined, being 6 × 105 CFU per PCR for the strains NCTC10442 (type I), JCSC4469 (type IVd) and WIS (type V), and 6 × 104 CFU per PCR for all other type or subtype strains [N315 (type II), 85/2082 (type III), CA05 (type IVa), 8/6-3P (type IVb), and MR108 (type IVc)]. This sensitivity is quite compatible with the single target PCR assay (1 × 104 to 6 × 105) (data not shown), suggesting that our M-PCR assay is sufficiently robust.
Validation of M-PCR assay.
To validate the M-PCR assay, we simultaneously compared our assay with the traditional PCR SCCmec
typing scheme including mec
gene complex typing and a previously described M-PCR assay (24
). Validation of our assay was performed by testing a total of 54 well-characterized MRSA strains with known SCCmec
types including type I (n
= 3), type II (n
= 14), type III (n
= 9), type IVa (n
= 18), type IVb (n
= 4), type IVc (n
= 3), type IVd (n
= 2), type V (n
= 1). We found a 100% concordance in typing SCCmec
types I-IV between the PCR results of our M-PCR, traditional SCCmec
typing method, and Oliveira's assay (Table 3
) except for one type V strain. However, in the WIS strain (type V), both our assay and the traditional SCCmec
typing method correctly identified this strain as SCCmec
type V, but Oliveira's M-PCR falsely categorized the strain as SCCmec
type III (Table 3
). In addition, our assay had more discriminatory power and was able to further classify type IV strains into subtypes IVa, b, c, and d (Table 3
To address our assay's ability in differentiating MRSA from MSSA, we tested 150 randomly selected local clinical MSSA isolates, in addition to the above 54 MRSA control strains and the 453 clinical MRSA isolates (below) and found a mecA gene band (147 bp) in all MRSA isolates but not in any MSSA isolates, hence being 100% concordant with phenotypic susceptibility (VITEK) and conventional mecA gene PCR test results.
Applicability and accuracy of M-PCR.
To assess the applicability and accuracy of the M-PCR assay, we further applied our SCCmec typing assay to test a total of 453 local clinical MRSA isolates randomly selected from our clinical stock collection for the 16-year period from 1989 to 2004. Among them, 235 (51.88%), 122 (26.93%), 74 (16.34%), 5 (1.1%), and 4 (0.88%) isolates belonged to SCCmec types and subtypes II, III, IVa, IVb, and IVc, but no SCCmec types and subtypes I, IVd, or V were found among the isolates tested. However, there were 13 (2.87%) isolates that were not typeable using our assay, with five (1.10%) isolates having multiple bands and eight (1.77%) isolates with amplification of only the mecA gene.
These not-typeable isolates were further characterized using the traditional PCR SCCmec
typing method and Oliveira's M-PCR assay. In five multiple-band isolates, one isolate presenting two bands of 200 bp and 280 bp (corresponding to types IVc and III by our new assay) was also not-typeable by the traditional PCR but was found to be type III by Oliveira's M-PCR, while the other four isolates with bands of 398 bp and either 613 bp or 200 bp (corresponding to types II and either type I or IVc by our new assay) were typed as types II in both other assays (Table 4
), and may represent undescribed new variant subtypes of SCCmec
type II. However, among the other eight isolates with amplification of only the mecA
gene, only one isolate (mecA
-band 8) was determined to be type IV by both the traditional PCR SCCmec
typing method and Oliveira's M-PCR assay, while the remaining (7 isolates) had incongruent typing results among the two other typing methods (Table 4
), potentially representing new types or variant subtypes.
typing is one of the most important molecular tools available for understanding the epidemiology and clonal strain relatedness of MRSA, particularly with the emerging outbreaks of community-acquired MRSA occurring on a worldwide basis. However, due to the very complex and diverse structure of the SCCmec
typing is usually achieved by DNA sequence analysis (21 to 67 kb) (10
), Southern blot analysis using three or more restriction enzymes and several key probes specific for each SCCmec
), and by PCR. Previously described traditional PCR SCCmec
typing schemes target the individual regions of the classes of the mec
, and mecI
), the allotypes of the ccr
, and ccrC
), and individual subtypes of the J regions, and therefore require the use of many primer sets and PCR experiments (12
). These methods are laborious, time-consuming and expensive, resulting in limited utility for clinical and surveillance purposes.
Oliveira and de Lencastre developed a multiplex PCR strategy for mec
element type assignment and defined types of SCCmec
based on genes located within the J-regions of SCCmec
elements as follows: locus A, located downstream of the pls
gene and is specific for SCCmec
type-1; locus B, internal to the kdp operon, which is specific for SCCmec
type II; locus C, internal to the mecI
gene present in SCCmec
types II and III; locus D, internal to the dcs region present in type I, II, and IV; locus E, located in the region between integrated plasmid pI258
and transposon Tn554
, specific for SCCmec
type III; locus F, which is also specific for SCCmec
type-III located in the region between Tn554
; locus G, the left junction between IS431
and pUB110; and locus H, the left junction between IS431
and pT181 (24
). This is the only single-step multiplex PCR assay published to date, but it too has its limitations. Since it is much simpler and easier to perform than the traditional (nonmultiplex) PCR assays for SCCmec
typing, it has been increasingly used in favor of the traditional method. As a result, different SCCmec
types are named according to the standard SCCmec
type definition first established by Hiramatsu's group (T. Ito, X. Ma, Y. Kondo, P. Changtrakool, S. Traklsomboon, C. Tiensasitorn, M. Jamklang, T. Chavalit, J. Song, and K. Hiramastsu, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 115, 2004).
In addition to hampered interpretation due to the presence of multiple bands for each SCCmec
type (because of non-type-specific targets) and difficulties in assay optimization, Oliveira's assay (24
) has limitations in detecting the newly described SCCmec
type V, misclassifying them as type III (Table 3
), while failing to discriminate type IV into subtypes a, b, c, and d (24
). Since the newer SCCmec
types IV and V have recently been associated with community-acquired infection (12
), detecting type V and discriminating type IV into subtypes IVa, b, c, and d may play an important role in the understanding of the epidemiology and ultimate prevention and control of currently emerging community MRSA clonal outbreaks. Therefore, a more robust and simpler SCCmec
typing assay is urgently required.
We designed eight new sets of SCCmec type- and subtype-unique and specific primers and 1 new set of methicillin resistance (mecA gene-based) primers based on comprehensive analyses and alignments of the MSSA and MRSA genomes and SCCmec sequences, and have successfully developed a novel multiplex PCR SCCmec typing assay (in a single multiplex PCR with a single band for each type or subtype) capable of classifying MRSA isolates into SCCmec types and subtypes I, II, III, IVa, IVb, IVc, IVd, and V, according to the current updated SCCmec typing system, while simultaneously being able to discriminate MRSA from MSSA.
Simultaneous comparison of our new assay with the traditional PCR SCCmec typing method (including mec and ccr gene complex typing) and Oliveira's assay demonstrated 100% sensitivity and specificity when testing a large number of control strains. Further application of our assay in randomly selected local clinical isolates confirmed its feasibility and practicality. This novel assay offers a rapid, simple, and feasible method for SCCmec typing of MRSA, and may serve as a useful tool for clinicians and epidemiologists in their efforts to prevent and control infections caused by these organisms. It has been used in our regional MRSA surveillance program and helped identify and confirm the emergence of an outbreak of community-acquired MRSA infection in Calgary, Canada in 2004, with resultant expedient implementation of prevention and control measures (J. Conly, M. Gilbert, K. Zhang, D. Gregson, S. Elsayed, M. Mulvey, K. Laupland, L. Louie, H. Rabin, B. Baylis, and P. Boiteau, Abstr. AMMI Canada-CACMID 2005 Annual Conference, Abstr. and Presentation G5, 2005; M. Gilbert, J. Suishansian, J. MacDonald, D. Gregson, S. Elsayed, K. Zhang, K. Laupland, M. Louie, T. Louie, D. Nielsen, G. Keays, A. Honish, D. Gravel, M. Mulvey, J. Gillespie, J. Conly, Abstr. AMMI Canada-CACMID 2005 Annu. Conf., abstr. and presentation G2, 2005).
It has only recently been shown that some methicillin-susceptible staphylococci, including MSSA and methicillin-susceptible coagulase-negative staphylococci, could harbor SCC elements that contain the essential features of SCCmec
but lack the mecA
). These SCC elements serve as a vehicle of transfer for various genetic markers including genes mediating antibiotic resistance or virulence. The potential role of SCC for mediating gene movement in staphylococci is awaiting further investigation. Hence, our multiplex assay (incorporating a concomitant mecA
gene into specific SCCmec
typing system) may play a critical role in this regard.
Both the traditional PCR SCCmec typing scheme and Oliveira's multiplex PCR technique are PCR methods targeting unique loci. Not-typeable MRSA isolates are encounted when using the traditional PCR SCCmec typing scheme and Oliveira's multiplex PCR technique but the nontypeability rate is variable. Ito et al. used their traditional PCR typing method to type 617 MRSA isolates from Asian countries and found 5 (0.81%) strains were not-typeable (T. Ito, X. Ma, Y. Kondo, P. Changtrakool, S. Traklsomboon, C. Tiensasitorn, M. Jamklang, T. Chavalit, J. Song, and K. Hiramastsu, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 115, 2004).
Perez-Roth et al. (27
) found 11 not-typeable clones out of 375 isolates (2.93%) (due to unmatching patterns) when typing MRSA clinical isolates during a 5-year period (1998 to 2002) in a Spanish hospital, and Chung et al. (3
) found 4 out of 113 isolates (3.54%) were not-typeable when typing MRSA strains recovered at a Florida hospital, when both groups of investigators used Oliveira's assay. We used our newly developed assay to type 453 local clinical randomly selected isolates and found 13 (2.87%) not-typeable isolates. Except for one isolate (mecA
-band 8), the remaining 12 isolates (Table 4
) are potentially new types or subtypes. The explanation for these observations, as quoted by others, may be related to the presence of new structural types and subtypes or structural rearrangements and recombination of the mec
). Further investigations, including sequencing the mec
element, are needed in order to characterize these currently not-typeable isolates.
Our newly described assay was designed to target the SCCmec type- and subtype-unique and specific gene loci, based on the currently available sequence data of the MRSA and MSSA genomes and variable SCCmec type and subtype sequences in the GenBank database. This new assay is more robust and has greater simplicity in its ability to classify all known types in a single PCR compared to currently available typing methodologies. However, we acknowledge the potential future limitations of our assay because of the possibility that as more SCCmec elements from newly emerging clinical isolates are sequenced, new SCCmec types and subtypes will be discovered, necessitating a reconfiguration of our assay to allow for the detection of these newer types.