HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Hung, W.-C. Articles by Teng, L.-J. Search for Related Content PubMed PubMed Citation Articles by Hung, W.-C. Articles by Teng, L.-J. Agricola Articles by Hung, W.-C. Articles by Teng, L.-J.

Species identification of mutans streptococci by groESL gene sequence

^1,2,5Department of Clinical Laboratory Sciences and Medical Biotechnology¹, Center for Optoelectronic Biomedicine² and Department of Microbiology⁵, National Taiwan University College of Medicine, Taipei, Taiwan ^3,4,6Division of Neurosurgery, Department of Surgery³, Department of Laboratory Medicine⁴ and Department of Dentistry⁶, National Taiwan University Hospital, Taipei, Taiwan

Correspondence Lee-Jene Teng ljteng{at}ha.mc.ntu.edu.tw

Received June 2, 2005
Accepted June 10, 2005

The near full-length sequences of the groESL genes were determined and analysed among eight reference strains (serotypes a to h) representing five species of mutans group streptococci. The groES sequences from these reference strains revealed that there are two lengths (285 and 288 bp) in the five species. The intergenic spacer between groES and groEL appears to be a unique marker for species, with a variable size (ranging from 111 to 310 bp) and sequence. Phylogenetic analysis of groES and groEL separated the eight serotypes into two major clusters. Strains of serotypes b, c, e and f were highly related and had groES gene sequences of the same length, 288 bp, while strains of serotypes a, d, g and h were also closely related and their groES gene sequence lengths were 285 bp. The groESL sequences in clinical isolates of three serotypes of S. mutans were analysed for intraspecies polymorphism. The results showed that the groESL sequences could provide information for differentiation among species, but were unable to distinguish serotypes of the same species. Based on the determined sequences, a PCR assay was developed that could differentiate members of the mutans streptococci by amplicon size and provide an alternative way for distinguishing mutans streptococci from other viridans streptococci.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
The mutans group of streptococci is one of the five groups in viridans streptococci (Whiley & Beighton, 1998). Mutans streptococci consists of seven species that can be classified into eight serotypes: Streptococcus mutans (serotypes c, e and f), Streptococcus sobrinus (serotypes d and g), Streptococcus criceti (serotype a), Streptococcus downei (serotype h), Streptococcus ferus (serotype c), Streptococcus macacae (serotype c) and Streptococcus ratti (serotype b) (Facklam, 2002; Whiley & Beighton, 1998). According to previous reports, the most common species isolated from human sources are S. mutans and S. sobrinus. These two species have been consistently linked with the formation of dental caries, especially serotype c S. mutans (Cvitkovitch, 2001). S. criceti and S. ratti are rarely isolated. Other species are found mostly in animals and have less significance in humans.

Species identification of mutans streptococci has been based on conventional methods including complicated biochemical tests, but is time-consuming and sometimes unsatisfactory (Whiley & Beighton, 1998). Recently, several molecular methods have been developed, such as the PCR assay based on the glucosyltransferase genes (Oho et al., 2000), 16S rRNA gene PCR-RFLP analysis (Sato et al., 2003), the 5' nuclease-based real-time PCR assay for quantitative detection (Yoshida et al., 2003), random amplified polymorphic DNA analysis (Truong et al., 2000) and the multiplex PCR to determine three serotypes of S. mutans (Shibata et al., 2003).

The groES and groEL genes are evolutionarily conserved and are considered as an alternative for identification of micro-organisms. In earlier studies, Goh et al. (1996, 1998, 2000) developed reverse-checkerboard hybridization to identify Staphylococcus and Enterococcus species based on the partial groEL gene sequences. In our laboratory, we successfully determined groESL gene sequences and provided another approach for the identification of Enterococcus species and viridans streptococci (Teng et al., 2001, 2002; Tsai et al., 2005). In this study, we determined the nearly full-length groESL gene sequences of eight reference strains of mutans streptococci to analyse the phylogenetic relationship and tried to differentiate members of the group by the amplicon size from a PCR targeting the groESL genes.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Bacterial strains.

Eight reference strains of mutans streptococci were included in this study. Serotypes a (S. criceti E49), b (S. rattus EA1), e (S. mutans MT730R), f (S. mutans OME175) and h (S. sobrinus MFe28) were kindly provided by S. Hamada (Osaka University, Osaka, Japan). Serotype c (S. mutans GS-5) was kindly provided by H. K. Kuramitsu (University of Texas, San Antonio, Texas, USA) and serotypes d (S. sobrinus B13) and g (S. sobrinus 6715) were kindly provided by S. M. Michalek (University of Alabama, Birmingham, Alabama, USA). In addition, 13 clinical isolates of S. mutans (nine serotype c, three serotype e and one serotype f), which were mostly obtained from blood cultures, were collected between 1999 and 2004 from the Bacteriology Laboratory, National Taiwan University Hospital, a 2000 bed teaching hospital in northern Taiwan. The clinical isolates were identified with the API 32 STREP system (bioMérieux) and confirmed by 16S rRNA gene sequencing. The serotyping of S. mutans was carried out as described previously by Shibata et al. (2003).

groESL gene sequencing of mutans streptococci.

DNA was prepared from the bacteria by using a DNA isolation kit (Puregene; Gentra Systems) according to the instructions of the manufacturer. Two pairs of degenerate primers, Strep-ES-UP (5'-GACTATTTCTGAC CAAGTGAT-3', corresponding to positions groES –89––70) with gor600R (5'-TCNCCRAANCCNGGYGCNTTNACNGC-3', corresponding to positions groEL 842–817, where N = A + C + T + G, R = A + G and Y = C + T), and gor600F (5'-GGNGAYGGNACNAC NACNGCNACNGT-3', corresponding to positions groEL 253–278) with gor1907-1927 (5'-YTACATCATNCCNCCCATCAT-3', corresponding to positions groEL 1623–1603), were used to amplify two overlapping fragments of the groESL genes and to obtain nearly complete groESL sequences. The PCR was carried out in a DNA thermal cycler (MJ Research) with 30 cycles of denaturation (94 °C, 30 s), annealing (52 or 54 °C, 1 min) and extension (72 °C, 1 min 30 s), followed by a final extension step (72 °C, 7 min). The PCR products were purified and subsequently sequenced on a sequencing system (model ABI PRISM 3100; Applied Biosystems) with a Taq BigDye-Deoxy Terminator cycle sequencing kit (Applied Biosystems) according to the instructions of the manufacturer.

Amplification of the partial groESL genes and the gene spacer.

Based on the determined sequences, one pair of primers, ES.5-29F (5'-TAAAACCHTTAGGHGAHCGWRTBGT-3', corresponding to positions groES 5–29, where H = A + T + C, W = A + T and B = T + C + G) and EL.35-18R (5'-CKKGCATCTGCTGAAAAT-3', corresponding to positions groEL 35–18, where K = T + G), were designed to amplify a fragment containing a region of partial groES, partial groEL and groES–groEL spacer. The PCR was carried out with 30 cycles of denaturation (94 °C, 30 s), annealing (50 °C, 30 s) and extension (72 °C, 1 min), followed by a final extension step (72 °C, 7 min). The amplification products were subsequently subjected to 1.5 % agarose gel electrophoresis (FMC BioProducts), stained with ethidium bromide and photographed under UV light.

Phylogenetic relationships.

DNA sequences were aligned by using Gene-Works software (IntelliGenetics). The phylogenetic relationships among species were analysed by the neighbour-joining method listed in the MEGA (molecular evolutionary genetic analysis) analytical package (Kumar et al., 2001). For the neighbour-joining analysis, the distance between the sequences was calculated by using Kimura's two-parameter model. Levels of similarity were determined among species. Bootstrap values were obtained for 500 randomly generated trees.

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Comparisons of the groES, spacer and groEL sequences among eight reference strains

Table 1 lists the lengths of the groES gene sequences and the pairwise nucleotide identities of the eight reference strains of mutans streptococci. There are two different lengths of groES gene sequences: S. ratti (serotype b) and S. mutans (serotypes c, e and f) showed the same size, 288 bp, and S. criceti (serotype a), S. sobrinus (serotypes d and g) and S. downei (serotype h) also showed the same size, 285 bp. The groES sequence similarities among the eight reference strains tested ranged from 61.5 to 100 % for nucleotide sequence identity and from 53.7 to 100 % for deduced amino acid sequence identity. Bacterial strains of the same species, such as serotypes c, e and f of S. mutans or serotypes d and g of S. sobrinus, displayed the highest degree of similarities. In addition, bacterial strains that showed the same size groES gene sequences displayed >76 % nucleotide similarities, while bacterial strains with different sized groES sequences only displayed < 65 % nucleotide similarities.

Table 2 presents the pairwise nucleotide identities of the groEL genes of the eight reference strains of mutans streptococci. The groEL sequence similarities among the eight reference strains tested ranged from 77.7 to 99.9 % for nucleotide sequence and from 90.5 to 100 % for deduced amino acid sequence.

The size of the spacer between groES and groEL ranged from 111 to 310 bp. S. mutans had the smallest spacer size (111 bp), following by S. ratti (125 bp), S. criceti (200 bp), S. sobrinus (215 bp) and S. downei (310 bp). The strains of serotypes c, e and f, which belong to S. mutans, had identical sequences in their spacers. Identical sequences were also seen in the strains of serotypes d and g, which both belong to S. sobrinus. The lengths and sequences of the groES–EL spacer were found to be species-specific in the mutans group streptococci.

Phylogenetic relationships

Phylogenetic trees derived from the nucleotide sequences of groES and groEL are presented in Fig. 1. The phylogenetic analysis, based on either groES gene sequences or partial groEL gene sequences, revealed that members of the mutans group streptococci cluster into two groups. One group contained four serotypes, b (S. ratti), c (S. mutans), e (S. mutans) and f (S. mutans), with the strains of the same species clustering together. These four strains displayed the same lengths in the groES gene (288 bp) and had higher similarities in the groES gene sequences among themselves. The other group contained the four strains of serotypes a (S. criceti), d, g (S. sobrinus) and h (S. downei). These four serotypes also had the same lengths in the groES gene (285 bp) and had higher similarities in the groES gene sequences. The two groups did not cluster together in the phylogenetic tree constructed from the alignment of the nucleotide sequences of the groES gene (Fig. 1a), but did cluster together in the phylogenetic tree based on the groEL gene (Fig. 1b). In comparison the phylogenetic tree depicted by Facklam (2002), based on the analysis of 16S rRNA gene sequences, showed that S. downei, S. sobrinus, S. criceti and S. mutans clustered together away from S. ratti. Minor differences were observed in the phylogenetic trees derived from nucleotide sequences of groES, groEL or 16S rRNA genes.

View larger version (23K): [in this window] [in a new window]   Fig. 1. Unrooted neighbour-joining trees based on nucleotide sequences of (a) full-length groES and (b) nearly full-length groEL (nucleotide position 1–1566) showing the phylogenetic relationships of Streptococcus species, Staphylococcus species, Enterococcus faecalis and Lactococcus lactis. The phylogenetic trees were generated by the neighbour-joining method in the MEGA2 package. The numbers at the nodes are the confidence levels expressed as percentages of occurrence in 500 bootstrapped resamplings. The scale bar indicates the evolutionary distance between sequences determined by measuring the lengths of the horizontal lines connecting two organisms. GenBank accession numbers for groES or groEL genes are shown after the species names.

Intraspecies variation of groESL sequences in clinical isolates

In order to evaluate the applicability of the groESL sequence to species identification, the intraspecies variation of groES, spacer and groEL sequences from 13 clinical isolates of S. mutans was determined. Of the 13 clinical isolates, nine were identified as serotype c, three were identified as serotype e and one was identified as serotype f. Table 3 lists the intraspecies polymorphisms of groES and groEL in these S. mutans clinical isolates. The identities of the groES sequences among clinical isolates and reference strains of a serotype ranged from 99.0 to 100 %. The identities of the groEL sequences among clinical isolates and reference strains of a serotype ranged from 99.5 to 99.8 %. All the isolates within the same serotype shared >99 % sequence similarity with their corresponding type strain. However, the sequence similarities for bacterial strains between serotypes within a species were also high and ranged from 98.6 to 100 % in groES and from 99.0 to 100 % in groEL sequences. Therefore, the groES or groEL sequence variation is not sufficient for differentiating serotypes within S. mutans. No sequence variation was observed for the groES–groEL intergenic spacer in the 13 clinical isolates.

Differentiation of species among mutans streptococci by PCR

Because of the apparently different sizes of the groES–EL spacer length in members of mutans streptococci, we designed a pair of primers ES.5-29F and EL.35-18R to amplify a fragment containing this region. As indicated in Fig. 2, at least three distinctly different sizes of PCR products could be observed visually, due to the variable length of the spacer. The expected sizes of PCR products amplified from the reference strains were 516 bp from serotype a (S. criceti), 444 bp from serotype b (S. ratti), 430 bp from serotypes c, e and f (S. mutans), 531 bp from serotypes d and g (S. sobrinus), and 626 bp from serotype h (S. downei). Therefore, amplicons from S. ratti (serotype b) and S. mutans (serotypes c, e and f) showed similar sizes (444 and 430 bp), and so did S. criceti and S. sobrinus (516 and 531 bp). Only S. downei (serotype h) generated a 626 bp amplicon. The 13 clinical isolates of Streptococcus mutans, consisting of three different serotypes, all generated the same amplicon sizes, which were identical to those of the reference strains (data not shown). Thus, this PCR, which amplified part of groES, the full-length of the spacer and part of the groEL gene, could help us to differentiate species of the mutans streptococci.

View larger version (47K): [in this window] [in a new window]   Fig. 2. Gel electrophoresis of PCR products from the amplification of partial groESL containing spacer in the mutans group streptococci. Lane M, DNA size markers (100-bp ladder, Invitrogen); lane a, serotype a (S. criceti); lane b, serotype b (S. ratti); lane c, serotype c (S. mutans); lane d, serotype d (S. sobrinus); lane e, serotype e (S. mutans); lane f, serotype f (S. mutans); lane g, serotype g (S. sobrinus); lane h, serotype h (S. downei).

The primers ES.5-29F and EL.35-18R used in this study could also amplify the groESL genes from other viridans streptococci due to sequence conservation. However, the sizes of the PCR products amplified from mutans streptococci are different to those from other viridans streptococci. The amplicons from Streptococcus anginosus, Streptococcus constellatus, Streptococcus gordonii, Streptococcus mitis, Streptococcus oralis and Streptococcus sanguinis were all smaller than 400 bp. Data from our previous report (Teng et al., 2002) or from GenBank indicate that the lengths of groES–EL spacers in Streptococcus species are usually smaller than 67 nucleotides except in mutans streptococci, Streptococcus agalactiae (95 bp) and Streptococcus infantarius subsp. infantarius (342 bp). The groES–EL spacer sizes of viridans streptococci are listed in Table 4. These results indicate that the variable sizes of the intergenic spacer between the groES and groEL genes could provide an alternative way for distinguishing mutans streptococci from other viridans streptococci, although more data from clinical isolates should be collected to confirm this suggestion.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This work was supported by grant NSC 93-2314-B-002-069 from the National Science Council of Taiwan.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES