Previous Article | Next Article ![]()
Journal of Bacteriology, May 2003, p. 2711-2722, Vol. 185, No. 9
0021-9193/03/$08.00+0 DOI: 10.1128/JB.185.9.2711-2722.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Bacteriology, Faculty of Medicine, Juntendo University, Research InstituteInternational Medical Center of Japan,1 The Organization for Pharmaceutical Safety and Research,2 Division of Molecular Biology, Institute of Medical Science, University of Tokyo, Tokyo, Japan3
Received 3 December 2002/ Accepted 4 February 2003
|
|
|---|
|
|
|---|
The type of an SCCmec element is defined by the allotype of the ccr genes and the structural class of the mecA gene complex it possesses. Three allotypes of ccr genes (type 1, type 2, and type 3) and four classes of mecA gene complex (class A, class B, class C, and class D) have been identified (21). The class A mecA gene complex maintains the prototypic structure, mecI- mecR1-mecA-IS431mec, where mecI and mecR1 encode a repressor and a signal transducer of beta-lactam-induced mecA gene transcription, respectively (11, 16, 27, 31). In the other classes, the mecI and mecR1 genes are either partially or completely deleted in presumable association with an integrated copy of IS1272 (class B) or IS431 (class C). The class D complex found in Staphylococcus caprae JA186 has a deleted mecR1, but adjacent to the deletion point neither IS1272 nor IS431 was identified. The region abutting the deletion point of
mecR1 was identical to a stretch of sequences in the type II SCCmec element in the region distant from mecA gene complex, and we speculate that a homologous recombination might have caused the deletion (21). Type I SCCmec carries a type 1 ccr (ccr-1) and a class B mecA gene complex, type II SCCmec carries a type 2 ccr (ccr-2) and a class A mecA gene complex, type III SCCmec carries a type 3 ccr (ccr-3) and a class A mecA gene complex, and type IV carries a type 2 ccr (ccr-2) and a class B mecA gene complex. Among the three types of ccr genes, only the ccrB1 gene of type I SCCmec is inactive because of a frame-shift mutation (thus, the gene is generally designated ccrB1*). Because both ccrA and ccrB genes are required for the excision and integration events of SCCmec (20), the ccrB1 gene must have been intact prior to its integration into S. aureus chromosome (18). However, the search for the intact ccrB1 in MRSA strains as well as in strains of methicillin-resistant coagulase-negative staphylococci (C-NS) has been unsuccessful (20).
Since the class B mecA gene complex found in MRSA strains contains a truncated copy of IS1272 that is prevalent in Staphylococcus haemolyticus but not in S. aureus (2, 4), Archer et al. (2) and others (44, 48) proposed that the direction of interspecies transfer of the mecA gene complex is from C-NS to S. aureus. Various types of SCCmec are indeed widely distributed among staphylococcal species other than S. aureus. SCCmec elements are typically present in methicillin-resistant C-NS, and we have observed several methicillin-susceptible strains that harbor staphylococcal cassette chromosome (SCC)-like elements on their chromosomes. Structural comparison of extant SCCmec elements revealed that SCCmec is made up of two mutually independent genetic components (the ccr and mecA gene complexes) (18). We hypothesized, therefore, the existence of a family of mobile genetic elements (SCCs) that serve for interspecies transfer of any useful genetic information among staphylococcal species and that SCCmec is one of the SCCs that is specialized for the transfer of methicillin resistance (17). In this study, in order to prove this hypothesis, we looked for a staphylococcal strain carrying ccr genes but lacking the mecA gene. By screening methicillin-susceptible staphylococcus type strains, we identified a type I SCC element that carries an intact set of type 1 ccr genes (ccrA1 and ccrB1) but no mecA gene in the genome of Staphylococcus hominis type strain GIFU12263.
|
|
|---|
DNA preparation, PCR amplification, and nucleotide sequencing.
The DNA extraction and PCR, long-range PCR, and nested PCR amplification steps were performed as previously described (17, 20). To detect ccr genes among staphylococcus type strains, we carried out PCR amplifications with a set of primers (
c and ßc) that we designed by using the sequences common to the three ccr genes. The ccr genes and their surrounding chromosomal regions were amplified by long-range PCR (20), using seven primers (see Fig. 2b), followed by nucleotide sequence determination. The seven primers were designed based on the nucleotide sequences of type I SCCmec in strain NCTT10442 (DDBJ/EMBL/GenBank accession no. AB033763 [see reference 18]). By an inverse PCR (IPCR) procedure, DNA fragments corresponding to the regions upstream and downstream of the ccr genes were amplified, and the products were then sequenced by primer walking. The experimental procedure for IPCR was as follows. The chromosomal DNA of GIFU12263 was digested with EcoRI for cloning the right side and with HindIII for cloning the left side of the SCC12263 element (Takara Shuzo Co., Ltd., Shiga, Japan). Self-ligation of the EcoRI- and the HindIII-digested DNA was performed (DNA concentrations of 1 to 2 µg/ml) for 10 min with a ligation kit (Boehringer Mannheim Biochemica, Mannheim, Germany). The ligation mixtures were then used as the template DNAs for PCR amplification. The primer sets were hR-2 and hR-3 for the EcoRI self-ligation and hL-2 and hL-3 for the HindIII self-ligation (see Fig. 2b). Sequence analysis was performed with the big dye terminator cycle sequencing kit (Applied Biosystem Inc., Foster City, Calif.) on a DNA analyzer (model 3100; Applied Biosystem Inc.). The nested PCR procedure was performed as described previously (17). We used two sets of primers: hL1 and hR1 for the first round of PCR amplification and hL-4 and hR4 for the second round of PCR amplification. All primers used in this study are listed in Table 1.
![]() View larger version (25K): [in a new window] |
FIG. 2. Genomic structure of SCC12263. (a) ORFs in and around SCC12263. The ORFs corresponding to sequences more than 90 amino acids in size and twenty-six possible reading frames are indicated by squares. Those above and those below the line are the ORFs whose transcription is directed to the right and those whose transcription is directed to the left, respectively. The white squares indicate the ORFs that were conserved among three SCCmec elements. The black squares indicate the ORFs that showed no identity to the four SCCmec elements. The gray square indicates the ORF with identity to type II SCCmec. (b) The essential structure of SCC12263. Locations of the essential genes are illustrated. The only restriction sites indicated are those of of HindIII and EcoRI. orfXhom is shown in orange, the ccr1 genes are shown in red, and the arrow indicates the direction of transcription. Nucleotide sequences of the left and right boundaries of attL and attR in SCC12263 were aligned with those of attL in NCTC10442 and attR in N315, respectively. The orange arrows indicate inverted complementary repeats, IR-L and IR-R, at both extremities of SCC12263, and the blue arrows indicate those of MRSA. The direct repeats in SCC12263 are indicated by magenta arrows, and those of MRSA are indicated by green arrows. The sequences that are colored are identical to each other. (c) Boundary of SCC12263. Nucleotide sequences around the attL and attR, respectively, are aligned with the sequence of the presumptive integration site, attB, on the GIFU12263 SCC12263(-) chromosome. The orange arrows indicate inverted complementary repeats, IR-L and IR-R, at both extremities of SCC12263. The red arrows indicate direct repeats. The orange and red sequences are identical to each other.
|
|
View this table: [in a new window] |
TABLE 1. Synthetic oligonucleotide primers
|
Finally, major matching regions were extracted by combining neighboring shorter matching regions as follows. First, for a pair of nucleotide sequences of SCCs, fragments of homologous regions were enumerated by using bl2seq (45). Second, neighboring fragments were connected to each other to form homologous regions. For this, for each segment A, a segment B was defined to be in its right neighbor if (i) the segments were matches in the same direction; (ii) the left end point of B was to the right of the right end point of A, but within 1,000 bp; (iii) the vertical distance between the two parallel lines including A and B, respectively, was at most 100 bp; and (iv) among such segments, B had the closest left end point to the right end point of A. The distance between end points was measured as the sum of the difference between the horizontal and vertical axes. Finally, maximal collections of neighboring segments with slopes between the two very end points determined to be between -1.1 and -0.9 or between 0.9 and 1.1 were chosen as homologous regions.
Several types of BLAST analyses were carried out with software available at the NCBI site (http://www3.ncbi.nlm.nih.gov/BLAST/). TransQ analysis was carried out with software available at brown.ims.u-tokyo.ac.jp:8118/html/transq.html.
Construction of recombinant plasmid and excision assay. Two recombinant plasmids, pSR1hom and pSR1aur, harboring ccrA and ccrB genes of GIFU12263 and NCTC10442, respectively, were constructed by cloning the BamHI-digested DNA fragments containing ccr genes into the unique BamHI site of plasmid vector pYT3 (7), as described previously (20). The two primers Hom1 and Hom2 were used for the amplification and cloning of the ccrAB gene complexes from the chromosomal DNAs of the above-mentioned strains. The fidelity of the nucleotide sequences of the ccr genes harbored by the two plasmids pSR1hom and pSR1aur was confirmed before use. Electroporation was performed as described previously (20).
Nucleotide sequence accession number. The sequence of SCC12263 of strain GIFU12263 has been assigned DDBJ/EMBL/GenBank accession number AB063171.
|
|
|---|
ccrB1 gene (Fig. 1a). The N-terminal domain of about 20 residues, containing presumptive catalytic active sites of CcrA and CcrB (6, 18, 19, 21), were well conserved between the two polypeptides.
![]() ![]() View larger version (64K): [in a new window] |
FIG. 1. (a) The deduced amino acid sequences of CcrB1 of GIFU 12263 S. hominis were aligned with those of NCTC 10442 MRSA. A black arrowhead indicates the locus of the first amino acid residue changed by an addition of adenine in the nucleotide sequence of ccrB1* as described in the text. (b) Phylogenetic relationships among ccrA genes, ccrB genes, and four site-specific recombinases. The four site-specific recombinases that show high similarity to ccr genes and that were selected to investigate phylogenetic relationships are the integrase (int) of bacteriophage phi-FC1 found in E. faecalis (1,216 bp; DDBJ/EMBL/GenBank accession no. AF124258), the integrase of bacteriophage TP901-1 found in L. lactis (1,458 bp; DDBJ/EMBL/GenBank accession no. X85213), the site-specific recombinase found in C. acetobutylicum ATCC 824 (1,635 bp; DDBJ/EMBL/GenBank accession no. AE007636), and the site-specific recombinase found in S. pneumoniae R6 (1,680 bp; DDBJ/EMBL/GenBank accession no. AE008466). The nucleotide sequences of six ccrA genes, six ccrB genes, and four site-specific recombinases were aligned by using the PileUp program with a Genetics Computer Group default scoring matrix. Phylogenetic relationships were examined with the Paupsearch program by the neighbor-joining method by creating 1,000 bootstrap replicates and by selecting 70% of consensus. The tree was visualized with Tree View software, which was obtained from the Tree View website (http://taxonomy.zoology.gla.ac.uk/rod/treeview.htm). The branch length indicates the distance, which is expressed as the number of substitutions per 100 bases. (c) The alignment of deduced amino acid sequences between OrfXaur representative from NCTC 10442 MRSA strain and OrfXhom in GIFU12263.
|
S. hominis ccr genes are carried by an element similar to SCCmec. Figure 2 shows the S. hominis chromosomal region around the ccr genes bounded by the characteristic direct repeats (DRscc) of 14 bases (with 12 identical bases) found 11.3 kb upstream and 7.3 kb downstream of ccr genes, (Fig. 2b and c). Incomplete inverted repeats of 28 bases, IR-L (left) and IR-R (right), also found in typical SCCmec elements were localized at both extremities. The region designated SCC12263 was 21,589 bp in length and contained a total of 26 ORFs whose characterization is given in Table 2. Furthermore, the homologue of S. aureus orfXaur, designated orfXhom, was identified to the right of IR-R. orfX is an ORF of unknown function located near the origin of replication of the S. aureus chromosome, which contains the integration site (attBscc) of SCCmec (17, 26). The orfXhom was 82% identical to orfXaur in nucleotide sequences, and their deduced amino acid sequences were 91% identical to each other (Fig. 1c).
|
View this table: [in a new window] |
TABLE 2. ORFs in and around S. hominis SCC12263 with deduced products showing similarities to extant proteins
|
Certain ORFs shown in Fig. 2 were specific to SCC12263. However, most of them did not show homology with any of the extant genes (see "Cluster of restriction-modification enzyme gene homologues in right end of SCC12263"). It was also significant that there was no antibiotic resistance gene or mobile element found in SCC12263. The G+C content of SCC12263 was 29.52%, which, though not in itself significant, was lower than the range of G+C content reported for species S. hominis (30 to 36% [38]). There were nine regions (eight ORFs within SCC12263 and one ORF outside SCC12263) having substantially deviant G+C content values (under 29%). The most notable examples were orf22 (StsI methylase [M.StsI] homologue), orf23, and orf25, whose G+C contents were 23.77, 23.93, and 24.21%, respectively.
Further analysis of GC3, the G+C content of the third codon letter, whose deviation from the average value of the entire chromosome may be associated with genes acquired by horizontal transfer (26), revealed GC3 values lower than 24% for orf2 to -11 and -22 to -25. Especially low were the values of orf22 to -25, which were 13 to 15%. It was also noted that orfXhom, shown as ORF26, had a significantly high G+C content of 37.5%, as in the case with orfXaur in MRSA strains (39.41% [see reference 17]).
Figure 3 compares nucleotide sequences of SCC12263 and SCCmec types I, II, III, IVa, and IVb. Two regions, B (from nucleotide positions 1,117 to 4,890) and D (from nucleotide positions 8,600 to 16,198), of SCC12263 were found to be highly homologous to the extant SCCmecs. Another three regions, A (from nucleotide positions 1 to 1,116), C (from nucleotide positions 4,891 to 8,599), and E (from nucleotide positions 16,199 to 21,589), did not show significant homology. The extents of homology of the B and D regions with the corresponding regions of the SCCmec elements are indicated in Table 3. Region D contained orf11 to -19 that corresponded to the ccr gene complex (18). On the other hand, region B encompassed orf2 to -7, whose functions are unknown.
![]() View larger version (22K): [in a new window] |
FIG.3. Homologous regions of nucleotide sequences of SCC12263 and SCCmec type I, II, III, IVa, and IVb and their identities. The nucleotide positions are relative to the SCCs. The SCC region of SCC12263 corresponds to the region from nucleotide 770 to 22358 in GIFU12263, the type I SCCmec corresponds to the region from nucleotide 4504 to 38867 in AB033763, the type II SCCmec corresponds to the region from nucleotide 4687 to 57653 in D86934, the type III SCCmec corresponds to the region from nucleotide 899 to 67794 in AB037671, the type IVa SCCmec corresponds to the region from nucleotide 975 to 25222 in AB063172, and type IVb SCCmec corresponds to the region from nucleotide 301 to 21219 in AB063173. The regions A through E (see text for description of location) and how they correspond to the structure presented in Fig. 2a are shown at the bottom.
|
|
View this table: [in a new window] |
TABLE 3. Corresponding regions of SCC12263 and SCCmeca
|
It was also remarkable that the regions similar to B and D regions were located in reverse order in type II and III SCCmec. Furthermore, the region B was very similar to the corresponding region of type IVa (97% identity), whereas the region D was not (79%) (Table 3). These observations suggest that recombination occurred between the two regions during molecular diversification of the SCC elements. Phylogenic relationship was inferred by computing multiple alignments of the core homologous regions, shared by five SCCs for region B and six SCCs for region D (data not shown). The two phylogenetic trees CcrA and CcrB computed as described above were different from each other. As mentioned above, the GC3 of the ORFs in the two regions of SCC12263 was also discrepant: low in region B but not in region D. Therefore even SCC12263 might have been formed by recombination between two evolutionary unrelated DNA fragments. A very weakly homologous region was observed between the rightmost part of type III SCCmec and the region from 14.6 to 16.0 kb of SCC12263 in reverse direction. Another one is between the leftmost part of type VIa SCCmec and the region from 20.7 to 21.2 kb of SCC12263. We know that the rightmost part of the type III SCCmec contains a pseudo-ccr gene complex containing a ccrA2-like gene (18). However, because of the paucity of homology between ccrA gene of the SCC12263 (which should be classified as a ccrA1 gene) and a degenerate ccrA2-like gene, no apparent homology line was visualized with the set for the computation. Regions A, C, and E of SCC12263 did not have corresponding regions in four SCCmec types. However, the nucleotide sequence from positions 771 to 1474 in region A showed significant homology (98% identical) to that in the chromosome of S. epidermidis strain RP62A. Several restriction-modification enzyme gene homologues were identified in region E.
Cluster of restriction-modification enzyme gene homologues in right end of SCC12263. Several nucleotide sequences homologous to restriction-modification genes were identified in region E of SCC12263. ORF22 (the deduced peptide encoded by orf22) showed high homology to type IIS modification enzymes, M.StsI (the DNA methyltransferase of the StsI restriction-modification system) and M.FokI, that recognize an asymmetric sequence 5'-GGATG-3' and generate 6-mA on both the strands (23, 24). A motif search with pfam indicated that ORF22 carries two alpha-type domains (motif I [AdoMet binding]-target recognition domain [TRD]-motif IV [DPPY]) in tandem. As for M.FokI (41), this putative protein appears to be formed by fusion of two methyl transferase proteins, each methylating one strand of the recognition sequence.
In the upstream of ORF22, there was a 174-bp nucleotide stretch homologous to type III SCCmec in strain 85/2082. TransQ search revealed weak similarity of this sequence stretch with that encoding a part of HgaI, a type IIS restriction enzyme that recognizes 5'-GACGC-3' (40). Its two cognate modification enzymes methylate its internal C5 on the two strands (42). This region may represent a remnant of a type IIS restriction enzyme gene. We cannot conclude whether this was part of a once-active gene forming a restriction-modification system with orf22 or not. The carboxy end of ORF25 showed homology with several restriction enzymes (by TransQ and PSI-BLAST analyses). These included McrA, a C5 methyl-specific endonuclease; ml17993 in Mesorhizobium loti found adjacent to a DNA methylase homologue; and HP0052, a gene in a cluster of restriction-modification homologues in Helicobacter pylori (strain 26695) (1). A BLASTx search with ORF23 as a query criterion revealed weak homology with a putative NTPase/helicase of a virus (gi|9631396|ret|NP_048295.1|) and a gene in Sulfolobus solfataricus (gi|13815496|gb|AE006826.1|AE006826). PSI-BLAST analysis detected similarity of the Sulfolobus gene with SAM-dependent O-methyltransferases (eg., gi:13399467) (49). It is not clear whether ORF23 represents a remnant of a novel type of DNA methyltransferase that is homologous to O-methyltransferases.
Spontaneous precise excision of SCC12263. To test whether SCC12263 behaves like SCCmec as a mobile element, PCR experiments were designed to monitor spontaneous precise excision of SCC12263 occurring in the attL and attR regions. The two combinations of primers, hL-1 and hR-1 and hL-4 and hR-4, were used for the first and second rounds of PCR, as described previously (17). The primers hR-1 and hR-4 were located in orfXhom, and hL-1 and hL-4 were located in the chromosomal region outside the left boundary of SCC12263 (Fig. 2b). It has been reported that the sensitivity of the first round of PCR was such that it could detect attBscc sequence generated in 1 in 10,000 cells of the culture by precise excision of SCC element (17). At this level of sensitivity, no amplified band was detected when GIFU12263 DNA was used as a template for PCR. However, the second round of PCR (nested PCR) did amplify a 0.8-kb DNA fragment. The DNA fragment was purified and was subjected to nucleotide sequence determination. By comparing it with the nucleotide sequences of chromosome-SCC junction sequences, we could confirm occurrence of precise excision of the SCC and infer the precise cutting site as shown in Fig. 2. The result indicated that spontaneous precise excision of SCC12263 presumably mediated by ccr genes occurred in S. hominis strain GIFU12263, though at a low frequency of less than 1 in 104 cells (17).
ccr genes from SCC12263 promote excision of SCCmec. In order to examine if the ccr genes on SCC12263 encode active recombination or not, the ccrA1 and ccrB1 genes of S. hominis GIFU12263 and S. aureus NCTC 10442 were cloned into a shuttle vector, pYT3 (7). The recombinant plasmids obtained were pSR1hom and pSR1aur, respectively. Four plasmids; pYT3 (plasmid vector), pSR1hom, pSR1aur, and pSR2 (pYT3 harboring ccr2 genes was formerly designated pSR [20]) were introduced by electroporation into strain 85/1940, which has a type I SCCmec with a mutated ccrB1 gene that is indistinguishable from that of NCTC10442 (18), followed by selection of transformants on the tetracycline plates. (The reason that we did not use NCTC10442, a representative MRSA strain carrying type I SCCmec (17), was that the strain was tetracycline resistant). The strains 85/1940, 85/1940(pYT3), 85/1940(pSR2), 85/1940(pSR1hom), and 85/1940(pSR1aur) were then cultivated in drug-free broth for 9 days at 30°C with one passage per day, after which the proportion of the cells that had lost SCCmec was evaluated by replicating them onto agar plates with and without ceftizoxime. On the 9th day, the cultures of strains 85/1940, 85/1940(pYT3), and 85/1940(pSR1aur) yielded essentially equal numbers of colonies on the BHI agar plates with or without ceftizoxime (Table 4). On the other hand, the 9 days passage of the strains 85/1940(pSR2) and 85/940(pSR1hom) generated ceftizoxime-susceptible cells, which constituted 6 and 4% of the entire cell population, respectively (Table 4). This demonstrates that the ccrB1 of the S. hominis strain were active in excising the type I SCC mec. To determine the site of excision mediated by the ccr genes of the S. hominis SCC12263, we amplified the attBscc region on the chromosome in 85/1940(pSR1hom) and N315(pSR1hom) by PCR with cR2 and cL3 as the primers (17). As controls, we used the chromosomes of 85/1940, 85/1940(pSR2), 85/1940(pSR1aur), N315, N315(pSR2), and N315(pSR1aur) strains. All strains were cultivated in drug-free broth for 18 h, and then their chromosomal DNAs were extracted. Figure 4 shows that bands were amplified with 85/1940(pSR2), 85/1940(pSR1hom), N315(pSR2), and N315(pSR1hom), but not with 85/1940(pSR1aur) or N315(pSR1aur). The DNA fragments were purified and were subjected to nucleotide sequence determination, and we compared the sequences with those of 85/1940 and N315. The nucleotide sequences in 85/1940(pSR1hom) and N315(pSR1hom) were identical to those in 85/1940(pSR2) and N315(pSR2), respectively. The identity of the excision mediated by pSR2 and pSR1hom indicated that the ccr genes on SCC12263 encoded the recombinases of the same specificities of those of the extant ccr gene products of type I and type II SCC mec.
|
View this table: [in a new window] |
TABLE 4. Excision of SCCmec mediated by plasmids carrying ccr genes
|
![]() View larger version (82K): [in a new window] |
FIG. 4. PCR for detection of attBscc region on the chromosome. Amplification was carried out with a set of primers cL3 and cR2 specific for the regions to the left and right of the chromosome-SCCmec junction point. Lanes: 1, 1-kb ladder as a molecular weight marker; 2, N315; 3, N315(pSR2); 4, N315(pSR1hom); 5, N315(pSR1aur); 6, 85/1940; 7, 85/1940(pSR2); 8, 85/1940(pSR1hom); 9, 85/1940(pSR1aur).
|
|
|
|---|
This study was done as a part of our pursuit of a molecular predecessor of the SCCmec elements, which are distributed widely among clinical strains of various staphylococcal species (21). Sequencing analysis of the chromosome regions of the S. hominis strain surrounding the ccr genes identified a genomic island at the boundaries of which were found DR and IR sequences and orfXhom having high homology to orfXaur of S. aureus in which an attachment site for SCCmec is located. The structure of the boundaries indicated that the island was integrated at the attachment site attBscc found in the 3' end of orfXhom, whose sequence was indistinguishable from the attBscc into which SCCmec is integrated site specifically. Thus, we designated the island SCC as the representative of hypothesized SCC family in which the staphylococcal mobile elements share the same structural and functional features with extant SCCmec elements (possession of ccr genes, as well as DR and IR sequences, and site-specific integration into the attBscc) except for carriage of mecA gene complex.
The SCC12263 appears to be a type I SCC element (i.e., it is part of a subgroup of the SCC family as defined by the possession of type 1 ccr gene complex). An SCC belonging to this subfamily should have served as a recipient of class B mecA gene complex in the E region to generate type I SCCmec in the past. However, since strong homologies between SCC12263 and type I SCCmec were confined only to regions B and D, SCC12263 may not be the immediate precursor SCC for the generation of type I SCC mec. It may be that the SCC family is more diverse than we have expected, and continued search may eventually identify an SCC more closely related to the precursor SCC, with a homology extending, for example, to region A as well. The precursor SCCs for type II and type III SCCmec may also be identified in C-NS strains, since some mecA-negative strains that hybridize these types of ccr genes have been identified (Y. Katayama et al., unpublished data).
Besides the structural diversities in SCC elements, there seem to be significant diversities in the biological functions which they carry. We have started to observe that SCC is not confined to the transmission of methicillin resistance alone; that it is a well-developed vehicle for genetic exchange of any useful genes among staphylococcal species is also supported by other recent observations. We have previously described a genetic element, IE25923, integrated at the attBscc in the chromosome of methicillin-susceptible S. aureus type strain ATCC 25923 (18). Although it is considered as a remnant of SCC (it lacks ccr genes), neither the mecA gene complex nor any other antibiotic resistance gene was found among dozens of the ORFs present on it (18). Even extant SCCmec contains some genes whose functions are unrelated to antibiotic resistance: e.g., type I SCCmec contains SD repeat (a repeat domain comprising mainly aspartate and serine)-rich surface proteins that interfere the host cell's capacity to bind to fibronectin and fibrinogen (47), and type II SCCmec contains KDP operons that are involved in the ATP-dependent potassium transfer (43). A recent illustrative case in this line of observation is that found in the genome of an MRSA strain MW2 the sequencing of whose whole genome revealed the presence of an SCC-like structure integrated in tandem with a copy of SCC mec, and in association with this SCC-like structure a staphylococcal enterotoxin gene seh was identified (3). This genetic organization suggests the possibility that the seh gene, which is not present at the corresponding locus in any of the other five sequenced S. aureus chromosomes, might have been acquired by SCC-mediated lateral gene transfer (3). Recently, an SCC without the mecA gene complex has also been identified in S. aureus strain M, which, designated SCCcap1, carries a cap operon instead of antibiotic resistance genes (28). Therefore, it would be quite plausible to consider SCC a ubiquitous mobile genetic element whose function in staphylococcal species is interspecies transfer of genetic information useful to the host cells living in various stressful environments. The environmental selective pressure is not limited to antibiotics alone but could be any number of things, such as competition with other microorganisms, assault from the human immune system, poor or biased nutritional resources, and even pollutants such as cadmium and mercury, against which specific resistance genes in type III SCCmec have been identified (18). In this regard, detection of SCC and identification of the genes it carries would be useful not only for us to learn a bacterium's capability to survive antibiotic chemotherapy but also to track the individual life history of the staphylococcal strain (i.e., the series of environmental stresses that the given organism has undergone).
This work was supported by Grant-in-Aid for Scientific Research on Priority Areas (13226114) from The Ministry of Education, Science, Sports, Culture and Technology of Japan and the Core University System Exchange Programme under the Japan Society for the Promotion of Science, coordinated by the University of Tokyo Graduate School of Medicine and Mahidol University. The study was also partly supported by a Grant for International Health Cooperation Research (11C-4) from the Ministry of Health and Welfare and by the Research for the future Program of the Japan Society for the Promotion of Science. The work in IMS was supported by grants to I.K. from MEXT, NEDO, and the Uehara Memorial Foundation.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»