|Home | About | Journals | Submit | Contact Us | Français|
Macrococcus is a bacterial genus that is closely related to Staphylococcus, which typically is isolated from animal skin and products. The genome analysis of multidrug-resistant Macrococcus caseolyticus strain JCSC5402, isolated from chicken, previously led to the identification of plasmid pMCCL2, which carries a transposon containing an unusual form of the Macrococcus mec gene complex (mecAm-mecR1m-mecIm-blaZm). In M. caseolyticus strain JCSC7096, this mec transposon containing the mec gene complex (designated Tn6045 in this study) was found integrated downstream of orfX on the chromosome. Tn6045 of JCSC7096 was bracketed by the direct repeat sequences (DR) specifically recognized by cassette chromosome recombinase (CCR). A non-mecA-containing staphylococcal cassette chromosome (SCC) element, designated SCC7096, was integrated next to the mec transposon and separated from the latter by a DR. Nested PCR experiments showed that the mec transposon not only was excised singly but also coexcised with SCC7096 from the chromosome at the DRs. The coexcised elements formed the extrachromosomal closed circular DNA of the SCCmec-like element. SCCmec is known to be the mobile element conveying methicillin (meticillin) resistance in staphylococci. However, its origin has been unknown. Our observation revealed a potential mechanism of the generation of a new SCCmec-like element in M. caseolyticus, a commensal bacterium of food animals.
Macrococcus caseolyticus formerly was described as Staphylococcus caseolyticus (22), and it was reclassified as a member of a separate genus, Macrococcus, in 1998 (15). M. caseolyticus typically is isolated from animal skin and food, such as milk and meat (2, 15, 22). So far, there are no reports of M. caseolyticus isolated from humans. Baba et al. have determined the whole-genome sequence of a methicillin (meticillin)-resistant M. caseolyticus strain, JCSC5402, which was isolated from a skin swab of chicken from Japan. The strain harbored a plasmid (pMCCL2), on which we found a 5.1-kbp mec gene complex with a peculiar structure, mecI-mecR1-mecA-blaZ (designated mecIm-mecR1m-mecAm-blaZm), which is different from mecI-mecR1-mecA present in SCCmec, where blaZ encodes a beta-lactamase (2). This structure was considered the ancestral form of the mec gene complex, generated by the integration of the mecA gene into the bla operon blaI-blaR1-blaZ (2, 24), the latter being widely distributed among Gram-positive bacteria (9, 26).
In 1961, soon after the introduction of methicillin, methicillin-resistant Staphylococcus aureus (MRSA) was reported in England (13). MRSA is generated when methicillin-susceptible S. aureus (MSSA) exogenously acquires a staphylococcal cassette chromosome mec (SCCmec) (11). There are various types and subtypes in SCCmec, but they all are made up of two essential components: (i) the mec gene complex, containing mecA encoding the penicillin-binding protein 2′ (PBP2′) with reduced affinity to beta-lactam antibiotics, and (ii) the ccr gene complex, encoding site-specific recombinase(s). A large number of SCCmec elements have been sequenced and classified according to the combination of types of mec and ccr gene complexes (10). By the action of ccr gene-encoded recombinases, SCCmec is excised from the chromosome and integrated site specifically at the 3′ end of orfX (14), an open reading frame (ORF) of unknown function located near the replication origin (oriC) (3, 16, 27). As a result of the integration of SCCmec, directly repeated nucleotide sequences (DR) are generated at both ends of the integrated copy of SCCmec. Conversely, SCCmec is spontaneously excised, leaving only one DR from the chromosome (11, 12, 14). ccr gene-encoded recombinases recognize DRs for the integration and excision. SCCmec is speculated to have originated from the genome of other bacterial species and introduced into the species S. aureus by horizontal gene transfer (1, 11). Although some mecA gene homologues have been found in Staphylococcus sciuri and Staphylococcus vitulinus, they were not identified as parts of the mec gene complex or of the SCCmec element (23, 28). Thus, the origin of SCCmec remains unclear.
In this study, we describe a new SCCmec-like element found in M. caseolyticus strain JCSC7096, revealing a potential mechanism of the generation of SCCmec elements in macrococci.
In this study, we used M. caseolyticus strains JCSC5402, JCSC7096, and JCSC7528, which were isolated from chicken from Japan, China, and Thailand, respectively. These strains were identified by the determination of the 16S rRNA, hsp60, or sodA sequence, as described in previous reports (17, 20, 25). M. caseolyticus strains ATCC 13548T, ATCC 29750, ATCC 51834, and ATCC 51835 also were used for antimicrobial susceptibility testing.
MICs of oxacillin (Sigma Chemical Co., Ltd., St. Louis, MO) were determined by the agar dilution method using Mueller-Hinton agar (Difco, Detroit, MI), as recommended by the CLSI guidelines for staphylococci (5). The MIC breakpoints for resistance were those recommended for staphylococci in the CLSI guidelines (6).
The preparation of macrococcal DNA, PCR, long PCR, nested PCR, and nucleotide sequence determination and alignments were based essentially on previously published protocols for staphylococci (11, 12, 14). PCR amplifications of the mecAm gene, the downstream region of the orfX gene, and the mec gene complex region carried on the plasmid were performed by using the sets of primers listed in Table Table1.1. The PCR products were sequenced by the primer-walking method. To identify the spontaneous excision of DNA fragments from the JCSC7096 chromosome, nested PCRs were performed using the following two sets of primers (for the first and the second round of amplification) for each of the target DNA fragments: orfXm1-SCC2R1 and orfXm2-SCC2R2 for the identification of the excision of ΨSCCmec7096 and Tn6045, orfXm1-MC28R and orfXm2-MC27R for the identification of the coexcision of ΨSCCmec7096 and SCC7096, and orfXm1-MC34R and orfXm2-MC33R for the identification of the coexcision of ΨSCCmec7096, SCC7096, and ΨSCC7096-2. To identify the extrachromosomal closed circular DNA of the excised SCCmec-like element, the nested PCRs were performed using three pairs of primers: SCC2F1-SCC2R1 for the first round of PCR amplification, SCC2F2-SCC1R2 for the second round of the PCR amplification of the head-to-tail ligation site of the closed circular DNA of coexcised ΨSCCmec7096 and SCC7096, and SCC3F1-SCC1R2 for the second round of the PCR amplification of the head-to-tail ligation site of the closed circular DNA of coexcised ΨSCCmec7096, SCC7096, and ΨSCC7096-2. The sequences and positions of primers used in the nested PCR experiments are listed in Table Table11 and shown in Fig. Fig.2b,2b, respectively.
For detecting and sizing large plasmids and localizing the mecAm gene, S1-PFGE (pulse-field gel electrophoresis) and the Southern transfer of DNA fragments from the gel to a nylon membrane were performed by following previously published protocols (4). The preparation and hybridization of a probe for the mecAm gene were done using the digoxigenin/antidigoxigenin system (Roche Diagnostics K. K., Tokyo, Japan) according to the procedure recommended by the manufacturer. The mecAm probe was prepared by using the same primers as those used for the detection of the mecAm gene by PCR.
Open reading frames were identified with the genome analysis-oriented software In Silico Molecular Cloning, version 1.5 (In Silico Biology, Yokohama, Japan). Homology searches were performed using the BLAST program at the National Center for Biotechnology Information (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The phylogenetic tree was generated by CLUSTAL W using the neighbor-joining method (http://clustalw.ddbj.nig.ac.jp/top-j.html). The tree was visualized with Tree View software (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html). The branch length indicates the distance, which is expressed as the number of substitutions per 100 bases. The nucleotide sequences of ccr, pbp, and mecA genes were obtained from the GenBank database (http://www.ncbi.nlm.nih.gov/GenBank/index.html).
Nucleotide sequences of the JCSC7096 SCC element, JCSC7528 orfX gene region, and JCSC7528 mec transposon have been deposited in the DDBJ/EMBL/GenBank databases under accession no. AB498756, AB498757, and AB498758, respectively.
M. caseolyticus strains JCSC5402, JCSC7096, and JCSC7528 carried the mecAm gene detected by PCR with specific primers and were resistant to oxacillin (MICs of 128, 64, and 32 μg/ml, respectively). M. caseolyticus strains ATCC 13548T, ATCC 29750, ATCC 51834, and ATCC 51835, which were negative for mecAm by PCR analysis, were susceptible to oxacillin (MICs, 0.125, 0.125, 0.125, and 0.25 μg/ml, respectively). The oxacillin-resistant phenotype correlated with mecAm positivity by PCR.
The S1-PFGE analysis indicated that JCSC5402, JCSC7096, and JCSC7528 harbored plasmids of approximately 80 (pMCCL2), 100, and 75 kbp, respectively (Fig. (Fig.1a).1a). Southern hybridization localized the mecAm gene on the plasmids of JCSC5402 and JCSC7528. However, mecAm of JCSC7096 was not located on the plasmid but on the chromosome (Fig. (Fig.1b1b).
The sequencing analysis of the mecAm gene region on the 75-kbp plasmid of JCSC7528 revealed the same (100% nucleotide identity) mec gene complex and two adjacent transposase genes, located just upstream of the complex, that were found previously on the plasmid of JCSC5402 (2) (Fig. (Fig.2a).2a). On the JCSC7528 chromosome, orfX and other orf genes encoding conserved hypothetical proteins, which were homologues of MCCL_0033 and MCCL_0034 in JCSC5402, were adjacent. A transposon-like element encoding daunorubicin resistance was identified between orfX and MCCL_0033 on the JCSC5402 chromosome (Fig. (Fig.2a2a).
By the long PCR of JCSC7096 chromosome, we identified a long element integrated between the orfX gene and the MCCL_0033 homologue. Figures 2a and b show the genetic organization of the region downstream of orfX in JCSC7096. The chromosomally located mec gene complex was found to be located near the SCC integration site in orfX. The mec gene complex of JCSC7096 shared 100% nucleotide identity with those of JCSC5402 and JCSC7528. In addition, the mec gene complex of JCSC7096 was accompanied by two transposase genes, which had 96 and 100% nucleotide identity to the corresponding transposase genes of JCSC5402 and JCSC7528, respectively. These genes are flanked by a set of DRs, DRtnL and DRtnR, which presumably are generated by the transposon integration. We designated the 6.7-kbp mec transposon Tn6045 in this study. We identified four DRs (DR1 to DR4) between orfX and the MCCL_0033 homologue, which were similar to the DRs of MRSA SCCmec and MSSA SCC elements (12) (Fig. (Fig.2b).2b). Imperfect IRs (inverted complementary sequences) also were found linked to the DRs, which were similar to those of the SCC elements of staphylococci (12). Tn6045 was bracketed by DR1 and DR2. We designated this region ΨSCCmec7096 because of the lack of ccr genes (10). The further sequencing analysis of the downstream region of the mec gene complex identified a novel ccr gene complex that was composed of new molecular types of ccrA and ccrB genes, designated ccrAm1 and ccrBm1. These genes were part of an SCC element of 18.1 kbp in size that was designated SCC7096 (10) and was flanked by DR2 and DR3. Another ΨSCC element flanked by DR3 and DR4 was found between SCC7096 and the homologue of MCCL_0033, named ΨSCC7096-2 (10). ΨSCCmec7096, SCC7096, and ΨSCC7096-2 contained 6, 17, and 6 ORFs, respectively. Their BLAST first-hit entries are listed in Table Table22.
Compared to the corresponding genes of MRSA strain N315, the amino acid similarity of the gene homologues of JCSC7096 were 64, 53, 72, 66, 59, and 68% to mecI, mecR1, mecA, blaZ, ccrA, and ccrB, respectively. Strong amino acid sequence similarity also was seen in the ORFs around the ccrABm1 genes orf6 to orf9 and orf12 in SCC7096, which ranged from 47 to 85% compared to the corresponding genes of MRSA strain N315 (11, 16) (Fig. (Fig.2a).2a). orf1 to orf3 of ΨSCC7096-2 were homologous to the ORFs of the JCSC5402 chromosome present downstream of orfX (Fig. (Fig.2a,2a, Table Table2).2). orf3 in ΨSCC7096-2, encoding a truncated transposase, was present on both JCSC5402 and JCSC7096 chromosomes, but the ORFs beyond it were different from each other (Fig. (Fig.2a,2a, Table Table2).2). There was no apparent DRtn (the DRs generated by the integration of a transposon) inside ΨSCC7096-2. No drug resistance gene was found in SCC7096 and ΨSCC7096-2 (Table (Table22).
In the case of JCSC7528, no ccrABm1 genes were detected by PCR analysis (data not shown).
The overall G+C content of the 29 ORFs on ΨSCCmec7096, SCC7096, and ΨSCC7096-2 was 32%, which was slightly lower than the value (38%) for the whole genome of M. caseolyticus JCSC5402 (2), and it was close to that of the MRSA SCCmec (33% in the case of type II SCCmec of MRSA strain N315 [11, 16]). In the six ORFs of ΨSCCmec7096, however, the G+C content was 28%, which was much lower than the value for the whole genome of M. caseolyticus JCSC5402. The G+C content of the 17 ORFs on SCC7096 was 33% and of the 6 ORFs on ΨSCC7096-2 was 33%. It is, therefore, likely that both the mec transposon Tn6045 and SCC7096 originated from other bacterial species with lower G+C content than that of M. caseolyticus, and they were found in the M. caseolyticus strains as acquired DNA.
The sequence analysis of nested PCR products detected four excision patterns of the integrated DNA fragments from the chromosome. First, the transposase genes and mec gene complex were coexcised as a mec transposon from the chromosome, leaving DR1 and DR2. The second pattern was the excision of the mec transposon using DR1 and DR2. Therefore, it seems more plausible to assume either that ΨSCCmec7096 carrying Tn6045 was cut out from elsewhere and integrated into orfX by ccr-mediated recombination or that Tn6045 integrated itself as a transposon into a ΨSCC element that was preexistent downstream of orfX. The third pattern was the excision of the mec transposon together with SCC7096 using DR1 and DR3. The fourth pattern was the same as that describe above but using DR1 and DR4 (Fig. (Fig.2b).2b). Thus, the mec gene complex either was excised singly as mec transposon Tn6045 or coexcised with the ccr gene complex. Furthermore, we identified the extrachromosomal closed circular DNA of the coexcised elements by nested PCR (14). The sequencing analysis of the PCR products showed that the head-to-tail ligation occurred with the coexcited neighboring elements, ΨSCCmec7096-SCC7096 and ΨSCCmec7096-SCC7096-ΨSCC7096-2. The data indicated the formation of a single extrachromosomal DNA molecule that carries both mec and ccr gene complexes; i.e., an SCCmec mobile element formed by the ccr-mediated excision (14).
The phylogenetic tree of ccr genes showed that ccrAm1 and ccrBm1 of JCSC7096 were closely related to ccrA1 to ccrA4 and ccrB1 to ccrB4, which are found in the MRSA SCCmec and MSSA SCC elements, respectively (Fig. (Fig.3a3a).
Figure Figure3b3b illustrates the phylogenetic tree of PBPs and mecAm-encoded PBP2′ (mecAm). Intriguingly, the addition of mecAm to the analysis revealed a close evolutionary link between staphylococcal mecA-encoded PBPs (MecA), including those found in S. sciuri and S. vitulinus (23, 28), and enterococcal PBPs such as PBP5 of Enterococcus faecium and Enterococcus hirae, PBP3r of E. hirae, and PBP4 of E. faecalis (7, 8, 18, 19). The enterococcal PBPs are known to encode low-affinity PBPs for beta-lactam antibiotics that are comparable to mecA-encoded PBP2′ (7, 8, 18, 19). Thus, the Enterococcus PBPs, MecAm and MecA, seem to constitute a MecA superfamily sharing the same ancestral PBP, presumably with a low affinity to beta-lactam antibiotics (Fig. (Fig.3b).3b). We consider that the unusual mec gene complex (mecIm-mecRm1-mecAm-blaZm) found in M. caseolyticus was generated by the insertion of the mecA gene from the chromosome of a bacterial species, which had evolved from the common ancestor, into a bla operon by homologous recombination (2, 24). Such an event might be possible, since it was reported that pbp3r of E. hirae strain S185R was found on a plasmid and was speculated to have been derived from the E. faecium chromosome (21).
In conclusion, we found in the Macrococcus genome a number of new genetic structures containing a mecA gene homolog (mecAm) that have never been observed in other bacterial species, namely, a mec gene complex of a unique operon structure with blaZ, the mec transposon Tn6045, and the plasmid carrying Tn6045. In M. caseolyticus strain JCSC7096, the mec transposon seems to have transposed to the chromosome in immediate proximity to an SCC element carrying the ccr gene complex, leading to the generation of an SCCmec-like element. Furthermore, the ccrABm1 genes were active in excising the two essential components of SCCmec together using the two DRs (DR1 and DR3) located outside the two elements in JCSC7096. If the DR present between the two elements (DR2) were deleted and the transposase of Tn6045 were inactivated by genetic alterations, a new SCCmec element in which mec and ccr gene complexes are stably linked would be formed. Further studies are under way to isolate new SCCmec elements of M. caseolyticus. We have taken a glimpse at the process of how new types of SCCmec are produced using an mecA homolog and ccr genes in the three strains of M. caseolyticus.
We thank Tengku Zetty Maztura Tengku Jamaluddin for her help with the editorial assistance of the manuscript.
This work was supported by a grant-in-aid for 21st century COE research and a grant-in-aid for scientific research (18590438) from The Ministry of Education, Science, Sports, Culture and Technology of Japan.
Published ahead of print on 19 January 2010.