Previous Article | Next Article ![]()
Journal of Clinical Microbiology, September 2003, p. 4107-4112, Vol. 41, No. 9
0095-1137/03/$08.00+0 DOI: 10.1128/JCM.41.9.4107-4112.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Preventive Dentistry, Kyushu University Faculty of Dental Science, Fukuoka 812-8582,1 Department of Oral Health Sciences,2 Division of Social Dentistry, Dental Research Center, Nihon University School of Dentistry, Tokyo 101-8310, Japan3
Received 28 April 2003/ Returned for modification 2 June 2003/ Accepted 27 June 2003
|
|
|---|
|
|
|---|
The RGPs are composed of
1,2- and
1,3-linked rhamnan backbones with glucose side chains linked to alternate rhamnoses. Each serotype-specific polysaccharide has unique linkages of its glucose side chains (serotype c,
1,2-linkage; serotype e, ß1,2-linkage; and serotype f,
1,3-linkage) (5, 10). Recently, we identified and characterized the genes required for glucose side chain formation of the serotype c-specific RGP (9). However, the loci responsible for the determination of the other serotypes have not yet been elucidated.
In this study, we identified the loci involved in the glucose side chain formation of RGP in serotypes e and f of S. mutans and confirmed that these regions determine serotype specificities. Furthermore, we designed three pairs of primers from specific DNA sequences within each serotype determinant locus and succeeded in developing a multiplex PCR assay to easily identify serotypes of S. mutans strains. In addition, we evaluated the clinical usefulness of the PCR assay by using epidemiological samples from preschool children.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Bacterial strains used in this study
|
Confirmation of serotype specificity of RGP. Serotype specificity of RGP produced in S. mutans transformants was analyzed by immunodiffusion analysis (8). Autoclaved extracts prepared from whole cells of S. mutans strains were examined with serotype c-, e-, and f-specific rabbit antisera and with rhamnan-specific rabbit antiserum in 1% Noble Agar in saline. The serotype c-, e-, and f-specific rabbit antisera were raised by three subcutaneous injections of whole-cell suspensions of MT8148 (serotype c [Table 1]), MT703R (serotype e [Table 1]), or OMZ175 (serotype f [Table 1]) in incomplete Freund's adjuvant at 2-week intervals. These antisera were adsorbed with whole S. mutans cells of another serotype. The rhamnan-specific rabbit antiserum was prepared as described previously (19).
PCR experiments. PCR experiments designed to discriminate between S. mutans and Streptococcus sobrinus targeted the gene encoding the water-insoluble glucan-synthesizing enzyme (GTF-I) and were done with two sets of primers (GTFB-F plus GTFB-R and GTFI-F plus GTFI-R), as shown in Table 2. Three sets of primers (SC-F plus SC-R, SE-F plus SE-R, and SF-F plus SF-R) were used in the PCR assay to identify S. mutans serotypes (Table 2). Rapidly isolated chromosomal DNA from colonies on strips of Dentocult SM (Orion Diagnostica, Espoo, Finland) was used as a template.
|
View this table: [in a new window] |
TABLE 2. PCR primers used in this study
|
For cloning purposes, PCR was performed with 0.05 U of LA Taq DNA polymerase (Takara Bio Inc.) per µl in 10 mM Tris-HCl buffer (pH 8.3) containing pairs of primers (0.5 µM each primer) drawn from primers A, B, C, D, E, and F, 0.4 mM each deoxyribonucleoside triphosphate, 50 mM KCl, and 2.5 mM MgCl2. LA Taq DNA polymerase was used to enhance the amplification of the large PCR fragment.
Epidemiological procedure. Caries experience in 3- to 4-year-old children (n = 432, consisting of 237 boys and 195 girls) from five nursery schools in Tokyo, Japan, which is in a nonfluoridated area, was examined by two experienced and calibrated examiners (Cohen's kappa = 0.88) in November 2001. The parents of the children in the study granted permission for dental examination and microbiological sampling. Caries per tooth surface was diagnosed on the basis of visual classification as described by the World Health Organization (14). Oral mutans streptococci were recovered from plaque and saliva by using Dentocult SM Strip Mutans or Site Strip (Orion Diagnostica), as specified by the supplier. Colonies on strips were used for PCR detection.
Statistical analysis. The differences in the number of tooth surfaces with caries experience (the number of decayed and filled surfaces [dfs] on deciduous teeth per person) between groups were examined using the Kruskal-Wallis test followed by Steel's multiple-comparison test, using the statistical program Excel Toukei version 5.0 (Esumi Co., Ltd., Tokyo, Japan).
Nucleotide sequence accession numbers. The 3,784-bp (S. mutans Xc), 14,730-bp (S. mutans MT6219), and 16,442-bp (S. mutans LM7) nucleotide sequences determined in this paper have been deposited to the DDBJ data bank (URL, http://www.ddbj.nig.ac.jp) under accession numbers AB108684, AB108685, and AB108686, respectively.
|
|
|---|
![]() View larger version (25K): [in a new window] |
FIG. 1. Comparison of the genetic organization of the rgp loci from S. mutans Xc, LM7, and MT6219. The lower part of the diagram indicates regions responsible for glucose side chain formation during RGP synthesis. The identity shown at the bottom indicates the amino acid identity of the corresponding genes between two serotypes or among three serotypes. The rgpA through rgpF genes and ORF12 were common to the three serotypes and showed greater than 98% identity.
|
Conversion of serotype. To confirm that the regions identified above are determine the serotype, this region of Xc was replaced with the same region from different serotype strains. First, the region (ORF7 through ORF11) responsible for glucose side chain formation of Xc was replaced with the spectinomycin resistance gene (Spcr); the resultant mutant was designated Xc81 and produced only a rhamnan backbone with no glucose side chains. When Spcr of Xc81 was replaced with ORF1e through ORF4e of LM7 plus the erythromycin resistance gene (Emr) or with ORF1f through ORF3f of MT6219 plus Emr, the resultant transformants produced serotype e- or f-specific RGPs, respectively. Previous studies showed that rgpE, which lies upstream from these regions, was also essential in glucose side chain formation of serotype c RGP (11, 19). Therefore, the role of rgpE in serotype e- and f-specific RGP syntheses was examined. rgpE of Xc81 was replaced with the kanamycin resistance gene (Kmr), and the resultant transformant (Xc82) was generated. Even though ORF1e through ORF4e of LM7 or ORF1f through ORF3f of MT6219 were introduced into Xc82, these transformants produced only polysaccharides reacting with rhamnan-specific antiserum but not any serotype-specific RGP.
Next, the region responsible for glucose side chain formation in LM7 (ORF1e through ORF4e) or in MT6129 (ORF1f through ORF3f) was replaced with Spcr to yield LM7-81 and MT6219-81, respectively; the resultant transformants produced only the rhamnan backbone with no glucose side chains. When the region responsible for glucose side chain synthesis in Xc (rgpH through ORF11) was introduced into LM7-81 and MT6219-81, the resultant mutants produced serotype c-specific RGP. These results indicated that the regions between rgpF and ORF12, except for ORF7, were responsible for serotype-specific RGP synthesis in S. mutans, while rgpE has a common function in glucose side chain formation in all serotypes.
PCR evaluation. A comparison of the deduced amino acid sequences of the genes newly identified in this study revealed no homology among rgpH, ORF3e, and ORF2f, whereas the other genes showed homology to one another (Fig. 1). Based on this result, serotype-specific primers (SC-F plus SC-R, SE-F plus SE-R, and SF-F plus SF-R) were designed from the DNA sequences of rgpH (serotype c), ORF3e (serotype e), and ORF2f (serotype f), respectively, as listed in Table 2. Each amplified product had a unique size between 316 and 727 bp (Table 2). Of the 16 strains of mutans streptococci, S. mutans strains produced single bands of 727 bp (serotype c), 517 bp (serotype e), or 316 bp (serotype f) (Fig. 2A, lanes 5, 6, and 9 to 12). Other mutans streptococci did not produce an amplified product (lanes 1 to 4, 7, 8, and 13 to 16). Of the 21 other gram-positive strains, no amplified product was detected (data not shown). These results demonstrate that each set of primers can discriminate between specific serotypes of S. mutans strains.
![]() View larger version (42K): [in a new window] |
FIG. 2. Agarose gel electrophoresis of PCR products amplified with multiplex primers. (A) The purified chromosomal DNA samples from strains of mutans streptococci (A through H) were used as the templates. Lanes: 1, S. cricetus E49; 2, S. cricetus HS1; 3, S. ratti BHT; 4, S. ratti FA1; 5, S. mutans MT8148; 6, S. mutans Xc; 7, S. sobrinus MT8145; 8, S. sobrinus OMZ176; 9, S. mutans LM7; 10, S. mutans MT703R; 11, S. mutans MT6219; 12, S. mutans OMZ175; 13, S. sobrinus 6715; 14, S. sobrinus OU8; 15, S. downei Mfe28; 16, S. downei S28; M, molecular size markers. The capital letters above the lanes indicate serotypes, which correspond to those referred to in the text with lowercase letters. (B) The chromosomal DNA extracted from colonies on strips of Dentocult SM was used as the template. Lanes: 1 and 8, a reference marker of mixed PCR products from all three serotypes; 2, subject 1; 3, subject 2; 4, subject 3; 5, subject 4; 6, subject 5; 7, subject 6; M, molecular size markers.
|
![]() View larger version (10K): [in a new window] |
FIG. 3. Effect of S. mutans infection on caries experience in preschool children with regard to serotype. A, subjects scoring 0 with Dentocult SM kit (n = 218); B, subjects with a monoinfection by a single serotype (n = 158); C, subjects with a mixed infection by multiple serotypes (n = 11). Vertical bars represent standard error. Differences in caries experience among the groups were analyzed by the Kruskal-Wallis test (P < 0.001). **, significant difference according to the Steel multiple-comparison test (P < 0.01).
|
|
|
|---|
We previously demonstrated that rgpE, which is located in the middle of the rgp locus responsible for rhamnan backbone synthesis, is involved in glucose side chain formation of serotype c RGP (11, 19). Recently, we identified serotype c-specific genes responsible for
1,2-linked glucose side-chain formation downstream from the rgp locus in S. mutans Xc (9). In this study, it was confirmed that the rgpE homologs participated in a glucose transfer during RGP synthesis in serotypes e and f strains as well. However, the deduced amino acid sequence of the rgpE homologs revealed greater than 99.5% identity among the three serotypes, and a common function was confirmed in all serotypes by the conversion experiment. On the other hand, we recognized considerable discrepancies in the sequences of the region downstream from rgpF among the three serotype strains, although the rgpA through rgpF locus was well conserved among them. These results strongly indicated that the loci responsible for the side chain formation of serotype e and f RGPs were located downstream from rgpF. The conversion analysis confirmed that the regions between rgpF and ORF12 determined the serotype of S. mutans.
On the basis of these results, each serotype strain of S. mutans seems to have acquired its own specific genes for the synthesis of its serotype-specific antigen, and none of the three serotypes could be defined as an ancestral strain. Based on its predominance, the serotype c RGP structure may have advantages for S. mutans colonization of the oral cavity. It is interesting that no S. mutans strain with defective RGP glucose side chains has been isolated from the oral cavity. These findings suggest that the glucose side chains on S. mutans RGP might be important for its colonization of the oral cavity. We need to investigate further the function of RGP in S. mutans colonization.
Furthermore, we developed a simple, rapid, and reliable PCR method to identify serotypes of S. mutans. Although these serotypes have been hitherto identified by immunological methods, such as immunodiffusion analysis or fluorescent-antibody technique, these techniques can sometimes be ambiguous and time-consuming. There is cross-reactivity among the prepared antigens, and preparation of specific antigens and antibodies for each serotype is not easy. In the present study, we isolated S. mutans from 198 children, and serotypes of all but one isolate could be genetically defined by the multiplex PCR method developed in this study. In fact, of the seven subjects who were not identified by the PCR discriminating between S. mutans and S. sobrinus, serotyping PCR successfully identified two, one with serotype e and the other with serotype f. These findings suggested that the regions used for designing the primers for serotyping PCR were well conserved and that PCR discrimination of the serotype was clinically adequate. Sequence analysis of 16S rRNA revealed that S. sobrinus was present in one of the seven subjects and Streptococcus cricetus was present in two. The remaining two samples could not be identified because heterogeneous PCR products were obtained (data not shown).
Our epidemiological survey revealed that serotype c predominated (84.8%), serotype e was the next most common (13.3%), and serotype f occurred rarely (1.9%) in Japanese preschool children. This result is consistent with the data of Grönroos et al. (4). Further analysis of the worldwide serotype distribution of S. mutans using serotyping PCR will be helpful to understanding the phylogeny of S. mutans. In addition, it is interesting that children with a mixed infection by multiple serotypes seem to have a greater experience of caries than those with a monoinfection by a single serotype. It was reported that children who drank from a nursing bottle were often colonized with more than one clonal type (1). Although the PCR method used in this study could not distinguish ribotypes within S. mutans, a mixed infection by multiple serotypes might be equivalent to that by multiple ribotypes.
Fujiwara et al. (3) recently reported that they had isolated four S. mutans strains from the peripheral blood of patients with bacteremia; two strains were determined to be serotypes e and f by immunodiffusion, and the other two isolates were untypeable isolates with RGP with very low glucose contents. Since serotype c S. mutans strains are most frequently found in oral cavities of Japanese children, we wondered why serotype c S. mutans is not isolated from the blood of patients with bacteremia. In addition, a high incidence of untypeable S. mutans is uncommon. To define the contribution of S. mutans to the pathogenesis of infectious endocarditis, it is necessary to identify the serotypes of many S. mutans strains isolated from patients with infectious endocarditis. The PCR method developed in this study will be a powerful technique for clarifying the clinical importance of serotyping S. mutans.
|
|
|---|
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»