Streptococcus mutans was first described by J. Kilian Clarke in 1924. Kilian Clarke, a microbiologist, had performed his study supported by a research grant from the Dental Diseases Committee of the Medical Research Council. The money for the grant had been raised by British dentists amongst themselves, to support dental research. His task was to study the microbiology of dental caries disease. In deep dentin caries lesions, he found a small, chained coccobacillus which was more oval than spherical in shape. He suggested that these microorganisms were mutant streptococci and called them Streptococcus mutans (Clarke, 1924) . Clarke tried to prove the association of these streptococci with dental caries disease, but since other researchers did not support his hypothesis, interest in S. mutans waned. In the 1960s, the recently developed method of gnotobiotic animal research stimulated studies on the microbiology of dental caries disease, and S. mutans was convincingly connected to dental caries disease (Hamada, 1986a).
According to the classification in Bergey's Manual of Determinative Bacteriology, 9th ed. (Holt et al., 1994), the genus Streptococcus includes the pyogenic, oral and anaerobic groups of streptococci, as well as a group of other streptococci. The cells are spherical or ovoid, 0.5-2.0 m m in diameter, occurring in pairs or chains when grown in liquid media, and stain Gram-positive. Streptococci require nutritionally rich media for growth. The metabolism is fermentative, producing mainly lactate but no gas. The streptococci are catalase-negative, and they commonly attack red blood cells, with either greenish discoloration ( a -hemolysis) or complete clearing ( b -hemolysis). Optimum temperature for growth is 37 ° C, and growth is usually restricted to 25-45 ° C. Streptococci constitute a major population in the oral cavity, with several different species colonizing the various ecological niches of the mouth. Some of the species exhibit Lancefield serological group antigens. Differentiation between the pyogenic, oral and anaerobic groups may be laborious, and combined information is needed for classification (Holt et al., 1994).
Fig. 1. Streptococci commonly found in the human mouth; phylogenetic relationships among groups (information from Russell, 2000).
The oral group has sometimes been named the viridans streptococci, referring to the partial clearing of the erythrocytes around the colony. However, the terms are not interchangeable, as some species classified as viridans streptococci are not detected in the oral cavity. The current classification of the oral streptococci places the bacteria into four species groups; the anginosus, mitis, mutans and salivarius groups (Fig. 1). The classification is based on chemotaxonomic and genotypic data, especially DNA-DNA base pairing and 16S rRNA gene sequence analysis (Whiley and Beighton, 1998). The mutans group includes S. mutans, S. sobrinus, Streptococcus cricetus, Streptococcus rattus, Streptococcus downeii and Streptococcus macacae. Although the phylogenetic position of Streptococcus ferus has yet to be determined by 16S rRNA sequencing, other data indicate that S. ferus also belongs to the mutans group (Whiley and Beighton, 1998). The mutans streptococci represent eight serotypes (Maiden et al., 1992; Whiley and Beighton, 1998).
Table 1. Characteristics of the mutans streptococci group a
Information from Maiden et al., 1992;
Whiley and Beighton, 1998.
Differential characteristics of the mutans
S. mutans cells are about 0.5-0.75 m m in diameter. S. mutans occurs in pairs or in short- or medium-length chains, without capsules. Under acid conditions in broth and on some solid media, these cocci may form short rods 1.5-3.0 m m in length. Rod-shaped morphology may be evident on primary isolation from oral specimens. S. sobrinus are about 0.5 m m in diameter. S. sobrinus occurs in pairs and in chains. The word sobrinus means male cousin on mother's side and refers to the "distant relationship" between this species and S. mutans. S. rattus, S. ferus and S. cricetus are about 0.5 m m in diameter, occurring in pairs or chains (Hardie, 1986).
The differentiation within the mutans streptococci is based on differences in biochemical reactions (Table 2) and physicochemical surface properties (van der Mei et al., 1991), and on use of molecular techniques (Whiley and Beighton, 1998).
Epidemiological studies show that human populations world-wide carry mutans streptococci (Loesche, 1986). Mutans streptococci have been demonstrated in nearly all subjects in populations with high, low and very low prevalence of caries (Carlsson, 1988). These microorganisms are harboured by 33-75% of 4-year-old children (Carlsson et al., 1975; Alaluusua and Renkonen, 1984; Köhler et al., 1988; Caufield et al., 1993), 80-90% of adolescents, and virtually all adults (Loesche, 1986; Carlson, 1988; Alaluusua et al., 1990). Of the serotypes, c / e / f (representing S. mutans) and d / g (representing S. sobrinus) have been detected in humans with high frequency. Some reports of finding serotypes a (S. cricetus) and b (S. rattus) have also been published (Loesche, 1986; van der Mei et al., 1991). Serotype c is predominant in plaque and saliva samples from humans (Loesche, 1986). In samples from Finnish children, 75-90% of S. mutans isolates are serotype c , 10-20% serotype e , and only a few percent represent serotype f (Alaluusua et al., 1989, 1994). S. sobrinus is not as prevalent as S. mutans and is usually detected together with S. mutans (Loesche, 1986). The prevalence of S. sobrinus has been reported to range between a very low frequency and 30% in different populations. In selected subject groups, a higher frequency of S. sobrinus has been reported (Lindquist, 1991). Only a few studies report finding S. sobrinus as the sole mutans species, and subjects who harbour both S. mutans and S. sobrinus tend to have higher salivary mutans streptococcal counts than subjects harbouring S. mutans alone (Köhler et al., 1988).
Mutans streptococci are facultative anaerobes and their optimal growth is at 37 ° C (Ma and Marquis, 1997).
On MS agar, S. mutans colonies are small, raised, irregularly margined and adherent, while S. sobrinus colonies are surrounded by a zooglea with a gelatinous consistency (Hamada & Slade, 1980b). On sucrose-containing agar, most strains of S. mutans produce colonies of about 1 mm in diameter, with beads, droplets or puddles containing soluble extracellular polysaccharide (Hardie, 1986). On blood agar incubated anaerobically for two days, S. mutans colonies are white or gray, circular or irregular, 0.5-1.0 mm in diameter, sometimes tending to adhere to the surface of the agar (Hardie, 1986).
For primary isolation of mutans streptococci, the most frequently used medium is mitis salivarius bacitracin (MSB) agar (Gold et al., 1973), which is composed of mitis salivarius agar with sucrose, bacitracin and potassium tellurite. MSB agar is selective for S. mutans, S. sobrinus and S. rattus. TYCSB agar contains TYC agar (trypticase, yeast extract and cystine) with sucrose and bacitracin (van Palenstein-Helderman et al., 1983). GSTB agar contains a basal GS agar (trypticase and yeast extract) with glucose, sucrose, bacitracin and potassium tellurite (Tanzer et al., 1984). TSY20B agar contains trypticase soy agar, yeast extract, sucrose and bacitracin (Schaeken et al., 1986). MSKB medium is composed of mitis salivarius agar, sorbitol, kanamycin sulfate, bacitracin and potassium tellurite (Kimmel and Tinanoff, 1991). Differences in culture result when different media are used (Little et al., 1977; Jordan, 1986; Schaeken et al., 1986; Dasanayake et al., 1995). Little et al. (1977) found that the use of blood-sucrose media produced the highest recoveries of mutans steptococci as compared with the selective media MSB and the Carlsson medium. Schaeken et al. (1986) also reported that MSB agar is inhibitory to mutans streptococci, especially S. sobrinus , finding higher recoveries on TYCSB agar. Dasanayake et al. (1995) found higher counts on MSB agar than on GSTB agar.
In addition, chair-side methods for detection and enumeration of mutans streptococci have been developed. In the Dentocult SM® dip slide (Orion Diagnostica, Espoo, Finland), a special slide was coated with mitis salivarius agar containing 20% sucrose (Alaluusua et al., 1984). After inoculation of the slide with saliva, two discs containing bacitracin (5 m g) were placed on the agar surface, and the growth density of mutans streptococci was scored after incubation at 37 ° C for 48 h in an atmosphere created by CO 2 -generating tablets that were placed into the cover tubes of the slides. This type of chair-side slide had a short shelf-life because of the mitis salivarius-sucrose agar. The Dentocult-SM® Strip Mutans test was introduced in 1989 (Jensen and Bratthall, 1989). When performing this test, a bacitracin disc is added to the broth at least 15 min before use. After the test subject has chewed a piece of paraffin for at least 1 min, a plastic strip is turned around in the mouth. The strip is withdrawn through closed lips such that a thin layer of saliva remains on the strip. The strip is closed in the cover tube with selective broth and incubated at 37 ° C for 48 h. After incubation, the strip with attached colonies is compared with the model chart for counts 0 to 3. These strips can be dried and stored for long periods. For validation of method, these chair-side techniques have been thoroughly compared with conventional selective agar plate culture. The comparison yields a good correlation between the methods with regard to detection of mutans streptococci, as well as counts on agar and score values (Alaluusua et al., 1984; Jensen and Bratthall, 1989).
A number of other chair-side culture methods for mutans streptococci exist, including methods using the adherence ability of isolates and using tooth picks or wooden spatulas for sampling (Matsukubo et al., 1981; Jordan et al., 1987; Bratthall et al., 1996).
The identification of mutans streptococci is based on distinctive colonial morphology on selective and nonselective agar, Gram staining, distinctive cell shape on light microscopy, specific growth characteristics, and sugar fermentation and enzymatic patterns. The identification scheme used in the present study is presented in Table 2. The scheme is based on information in Bergey's Manual of Determinative Bacteriology (9th ed., 1994) and on additional information on phenotypic properties. S. mutans isolates can also be identified by the commercial biochemical test system API 20 Strep (Bio Mérieux, Marcy-l'Étoile, France).
Monoclonal antibodies (MAbs) directed against specific species of mutans streptococci, and also against cell markers and enzymes have been developed. The MAbs have been used in a number of studies on mutans streptococci, for detection in epidemiological studies and in studies on microbial mechanisms (de Soet et al., 1987, 1990a, 1990b; Takei et al., 1992; Fukushima et al., 1993; Shi et al., 1998).
A DNA probe for detection of S. mutans was developed by using glucosyltransferase B gene ( gtfB ) and fructosyltransferase gene ( ftf ) fragments, enabling detection without culture (Smorawinska and Kuramitsu, 1992). The dextranase gene (dexA) has also been used to construct a DNA probe for detection of S. mutans (Ida et al., 1998). For detection of S. sobrinus, a DNA fragment in the dextranase gene (SSB-3) has been suggested to be useful (Ida et al., 1999).
Typing of isolates is applied in epidemiological studies to determine bacterial occurrence and modes of transmission. Typing of isolates is also performed for evaluation of whether certain strains are associated with specific clinical disease conditions and to characterize the heterogeneity of infection, i.e., whether subjects are colonized by one or multiple types of the microorganism. Evaluation criteria for typing methods include typeability (ability to give an outcome for every isolate included), reproducibility (ability to give the same result when repeating the analysis) and discriminatory power (ability to differentiate between unrelated strains) (Arbeit, 1999). Two main types of epidemiological typing systems for microorganisms are available, the phenotypic and the genotypic methods.
Traditional methods of characterizing bacterial isolates have relied on measurement of characteristics expressed by the microorganisms, such as bacteriocin production and sensitivity to bacteriocins, serotype, biochemical properties, antibiotic resistance and bacteriophage type (Maslow and Mulligan, 1996).
One of the first epidemiological typing systems for oral streptococci was bacteriocin typing (Kelstrup et al., 1970). Bacteriocins are proteinaceous substances produced by the bacteria that inhibit the growth of other, mostly closely related, bacteria. The typing is performed by measuring the inhibiting effect on bacterial growth of certain indicator strains, and by measuring the sensitivity of the bacteria to be typed to bacteriocins from other strains (Jack et al., 1995). Heterogeneity among strains of mutans streptococci within one individual was first shown by bacteriocin typing (Kelstrup et al., 1970).
Using Ouchterlony immunodiffusion, Bratthall demonstrated five serological groups of mutans streptococci (Bratthall, 1970). A total of eight serotypes were subsequently recognized (Perch et al., 1974; Beighton et al., 1981). The classification is based on cell-wall carbohydrate antigen (Table 1). Serotyping by immunodiffusion, immunofluorescence or immunoelectrophoresis has been widely applied for typing of mutans streptococci.
In 1974, Shklair and Keene divided mutans streptococci into five biotypes (a-e) on the basis of fermentation characteristics, arginine hydrolysis and bacteriocin sensitivity, and they reported that these biotypes corresponded with the serotypes reported in 1970. Their later scheme (Shklair and Keene, 1976) also includes serotypes f and g .
Among other phenotypic methods are cellular fatty acid analysis, whole-cell protein analysis and multilocus enzyme electrophoresis (MEE). MEE is based on the relative electrophoretic mobility of metabolite cellular enzymes. MEE has been successfully applied in studies with many organisms, but only one report using MEE in strain identification of mutans streptococci has been published (Gilmour et al., 1987). Phenotypic typing is, in most cases, relatively inexpensive to perform and typeability of mutans streptococcal isolates is high, however, reproducibility and discriminatory power are usually somewhat poorer (Arbeit, 1999).
For isolate fingerprinting, molecular typing methods have a higher discriminatory ability and reproducibility since these methods do not examine the gene expression but rather the DNA of the microorganisms to be studied (Arbeit, 1999; Olive and Bean, 1999). Among these typing methods are plasmid analysis, restriction endonuclease analysis (REA), restriction fragment length polymorphism (RFLP) (including ribotyping), pulsed field gel electrophoresis (PFGE) and arbitrarily primed polymerase chain reaction (AP-PCR).
Plasmids are extrachromosomal circles of DNA that encode many properties, including antimicrobial resistance, many virulence traits and hydrocarbon metabolism (Madigan et al., 1997a). Plasmid analysis was the first DNA-based technique applied in epidemiological studies on mutans streptococci (Caufield et al., 1982). Because plasmids are infrequently detected in mutans streptococci, in only 5% of strains (Hamada and Slade, 1980b), plasmid analysis is not applicable to typing of these bacteria.
Restriction endonuclease analysis (REA)
In restriction endonuclease analysis (REA), bacterial chromosomal DNA is cut with a restriction endonuclease and separated by gel electrophoresis. The restriction endonucleases are enzymes that cut the DNA chain at specific recognition sequences. The restriction enzymes are nowadays synthetically fabricated, but were originally isolated from bacteria, with their original function being defence against other bacteria. After separation by gel electrophoresis, gels are stained with ethidium bromide and detected under UV light, whereby the banding patterns obtained for different strains are compared. Often the process results in fingerprints with many bands, thus the interpretation of the REA profiles can be complicated. REA has been applied for evaluation of relatedness of mutans streptococcal isolates (Caufield and Walker, 1989; Kulkarni et al., 1989).
Restriction fragment length polymorphism (RFLP), including
After cleaving the chromosomal DNA of the microorganisms to be studied, the separation products can be labelled with either DNA or RNA probes in the Southern blot technique (Southern, 1975). The use of a probe derived from the Escherichia coli ribosomal operon was introduced by Grimont and Grimont in 1986. They had discovered that variations of the genes encoding ribosomal ribonucleic acid (rRNA), and variations in sites flanking those loci, could serve as a means of typing strains since ribosomal sequences are highly conserved. In ribotyping an isolate, after the gel electrophoresis of the cleaved DNA, the fragments are hybridized with the rRNA probe. When detecting the hybrids, every fragment containing a ribosomal gene will be highlighted. The banding patterns obtained in ribotyping include only a small number of bands, thus rendering comparison of fingerprints among isolates easier than comparing REA patterns. The term ribotyping was introduced in 1988 (Stull et al., 1988), and ribotyping of mutans streptococci has been applied since 1993 (Saarela et al., 1993) mainly in studies on transmission of mutans streptococci and stability of infection (Alaluusua et al., 1994).
Pulsed field gel electrophoresis (PFGE)
In pulsed field gel electrophoresis, a variation of agarose gel electrophoresis, the orientation of the electric field across the gel is changed periodically ("pulsed"), thus larger bacterial DNA fragments can be analysed than by REA (Arbeit, 1999). PFGE is considered the "gold standard" of molecular typing methods, with excellent discriminatory power and reproducibility (Arbeit, 1999; Olive and Bean, 1999). This method has not been applied for typing of mutans streptococci.
Arbitrarily primed polymerase chain reaction (AP-PCR)
Of the genotyping methods thus far applied to mutans streptococci, AP-PCR is perhaps the least laborious (Olive and Bean, 1999). AP-PCR can be performed with a very small sample volume. For the polymerase chain reaction, the template is annealed to one or more short primers (typically 9-10 bp) at low stringency. Amplification results in an array of DNA fragments, often termed random amplified polymorphic DNA (RAPD), that can be resolved by gel electrophoresis. AP-PCR requires no previous knowledge of the DNA to be analysed (Welsh and McClelland, 1990; Williams et al., 1990). A limitation of the method is that it is very sensitive to even minor variations in technical factors such as temperature, Mg 2+ concentration and polymerase source. Interlaboratory comparison of typing results is impeded by only a fair reproducibility (Arbeit, 1999), and only isolates processed simultaneously and fingerprints obtained concomitantly can be compared. This fair reproducibility is, however, complemented by a good discriminatory power (Arbeit, 1999). AP-PCR typing has been shown to be well applicable to typing of mutans streptococci (Saarela et al., 1996; Li and Caufield, 1998).
Normally, before birth the foetus is sterile. The inoculation of the human oral cavity starts with the first tactile contacts with the mother and other persons present at the parturition, and contact with air and equipment. The oral cavity of the toothless child contains only epithelial surfaces and the first colonizers are species not requiring a nonshedding surface. Early colonizers include some streptococci, Veillonella, Actinomyces, Fusobacterium and a few Gram-negative rods (Könönen et al., 1994, 1999). Streptococcus salivarius is among the first permanent colonizers, colonizing the dorsum of the tongue in the edentulous infant (Socransky and Manganiello, 1971). In the first months, most of the detected strains are transient colonizers. Streptococcus sanguis and the mutans streptococci are stably colonized only after the first tooth has erupted (Berkowitz et al., 1975b).
In general, the acquisition of microorganisms by the human body is by transmission directly from one host to another, or indirectly by means of another living agent (vector). Pathogens can also be transmitted by inanimate objects and disease vehicles such as food and water (Madigan et al., 1997b). Saliva is regarded as the most important vehicle of transmission of mutans streptococci via physical contact (Duchin and van Houte, 1978; Köhler and Bratthall, 1978) or use of shared objects, e.g. spoons and forks (Bratthall, 1997); mutans streptococci can be recovered from a metal plate 24-48 h after inoculation (Köhler and Bratthall, 1978). The mother is considered to be the most important source of infection for the child, the first indication of this obtained by using a phenotypic typing technique of S. mutans , bacteriocin typing (Berkowitz and Jordan 1975a; Masuda et al., 1985). Li and Caufield (1995) detected by molecular typing of mutans streptococcal isolates a homology of strains in 71% of mothers and their children, and Kozai et al. (1999) found that 51.4% of mutans streptococcal genotypes found in children were also identified in their mothers. Fathers and infants have a far lower strain match (Rogers, 1981; Davey and Rogers, 1984; Kulkarni et al., 1989; Li and Caufield, 1995, 1998). Only Kozai et al. (1999) found an abundant similarity of strains in fathers and children, 31.4% of strains detected in children were in agreement with those in their fathers. The time period when most children gain mutans streptococci in their oral flora is when the primary teeth are erupting, i.e., between 8 months and 3 years of age (Caufield et al., 1993). The probability of colonization with mutans streptococci is high when inoculation with mutans streptococci is frequent and microbial cell count is at least 10 5 per ml saliva (Berkowitz et al., 1981). Another suggested prerequisite for early colonization is that the baby's diet includes frequent intake of refined carbohydrates (Alaluusua, 1991a). Recently, it has been shown that mothers can diminish the probability of transmitting mutans streptococci to their children by using xylitol chewing gum (Söderling et al., 2000). The exact mechanism of action regarding xylitol is unknown. The sugar substitute xylitol is not fermented by mutans streptococci into cariogenic acid end-products, and in the oral environment, xylitol presumably selects for mutans streptococci with a weakened virulence (Trahan, 1995). It has previously been shown that in mothers harbouring high numbers of mutans streptococci in their saliva, preventive measures aimed at decreasing mutans streptococcal counts also result in decreased caries counts in their children, even after 15 years (Köhler et al., 1982, 1983, 1994; Bratthall, 1997).
The primary habitat of S. mutans and S. sobrinus is the human dentition. In addition, after mutans streptococci have colonized the dentition, they can be detected in saliva, on the tongue, on oral mucous membranes, on denture surfaces and on surfaces of orthodontic appliances. When a mutans streptococcal strain has achieved stable colonization of the oral cavity, the strain usually persists for a long time period (Alaluusua et al., 1994). Mutans streptococci can also be detected in faeces (Hamada et al., 1980a), but otherwise, very few reports on finding mutans streptococci outside the oral cavity are available. The organism has been detected in endocarditis (Parker and Ball, 1976) and in purulent eyes of a neonate (Reeder et al., 1985).
The first tooth erupts at the age of about 8 months (± 2 months) (Kreiborg 1991), and only after that can there be stable colonization by mutans streptococci. Mutans streptococci can occur in the dentition on the tooth surfaces in dental plaque and in carious enamel or dentin.
In enamel caries, the subsurface tissue is invaded by microorganisms. The identification of the bacteria invading dental enamel is as yet incomplete, but in experiments with gnotobiotic rats, in combination with scanning electron microscopy, S. mutans and S. sobrinus have been shown to be able to invade the enamel (Luoma et al., 1987; Seppä et al., 1989).
In carious dentin, the number of bacteria recovered per mg of dentin has been reported to be higher in superficial layers of the lesion as compared with deeper layers, with regard to both mutans streptococci and total bacterial count (Edwardsson, 1974; Hoshino, 1985). Many bacterial species are present, but no evidence suggests that a specific combination of organisms acting in symbiosis are involved in the decomposition of dentin (Edwardsson, 1974).
The dental plaque consists of cells (about 70% of the overall volume) and plaque matrix (Jenkins, 1978). The water content ranges between 80% and 85%. Of the dry weight, 40-50% is protein, 13-17% carbohydrate, 10-14% lipid and 10% ash. The concentrations of calcium and phosphate are quite variable, but typical figures are calcium 8 m g per mg and inorganic phosphate (as P) 16 m g per mg (Jenkins, 1978). Plaque maximum thickness on smooth surfaces is 300 m m, on approximal surfaces 5 mm and in fissures 2 mm (Sissons, 1997). The microbial composition of dental plaque differs both qualitatively and quantitatively from the bacterial communities of other oral surfaces, and the microflora also varies over time. Plaque microorganisms exhibit a high genetic heterogeneity, with 30 to 300 species represented (Sissons, 1997). This conglomerate of bacteria and interbacterial substances can be regarded as a prototype for a biofilm, a bacterial community. The biofilm forms a dynamic link between the microbial flora, the tooth surfaces, and the different constituents of saliva, with the microorganisms living in symbiosis and antibiosis with each other (Costerton and Lewandowski, 1997). Survival of oral bacteria is enhanced by dental plaque formation (Bowden and Hamilton, 1998).
The frequency of isolation of mutans streptococci in dental plaque has been reported to be site-dependent, such that the isolation frequency steadily increases from approximal surfaces of mandibular incisors, approximal surfaces of maxillary incisors, approximal surfaces of molars, to fissures of molars (Loesche, 1986). S. mutans and S. sobrinus show a somewhat similar colonization pattern in the dentition, but S. sobrinus is more frequently isolated from posterior than anterior teeth. In subjects harbouring both S. mutans and S. sobrinus , the species colonize buccal surfaces in comparable numbers, but on all other surfaces, S. mutans is predominant (Lindquist, 1991).
Saliva is a complex mixture of several components (Whelton, 1996). Whole saliva (oral fluid) is formed primarily from salivary gland secretions, but also contains gingival fluid, desquamated epithelial cells, bacteria, leucocytes, and possibly food residues, blood and viruses (Dawes, 1996; Whelton, 1996). Saliva is essential for maintenance of healthy oral tissues; it coats the oral mucosa and protects against irritation, forms an ion reservoir for tooth remineralization, functions as a buffer, aids in swallowing, exerts antimicrobial action, participates in pellicle formation and enzymic digestion of starch with amylase, and also participates in taste sensation by acting as a solvent (Whelton, 1996). Antimicrobial components in saliva include immunoglobulins, lysozymes, lactoferrins, salivary peroxidases, myeloperoxidases, histatins, amylases and anionic proteins (Bowen, 1996; Tenovuo, 1998). Moreover, organic components in saliva, especially mucous glycoproteins, function as nutrients for many oral bacteria (Bowen, 1996). Salivary proteins can be degraded by proteases produced by, for example, S. mutans and S. sanguis (Bowen, 1996).
The bacterial content of saliva is estimated to approach 10 9 bacteria per ml (Bowen, 1996). Saliva helps to control invasion of the mouth by microorganisms, and lack of saliva results in increased numbers of bacteria in the mouth. Saliva can act as a selective medium for bacterial growth, but continuously repeated swallowing results in clearing of bacteria (Bowen, 1996).
Salivary mutans streptococcal counts rarely exceed 10 7 CFU per ml. A highly significant correlation has been demonstrated between the salivary numbers of mutans streptococci and their prevalence in the dentition, both in terms of the number of tooth surfaces colonized and the level of infection of tooth surfaces (Duchin and van Houte, 1978; Lindquist, 1991).
Dental caries disease includes a breakdown of enamel, the hardest material in the human body, and a subsequent breakdown of the underlying dentin. The disease is the most prevalent of the chronic diseases affecting the human race. The aetiology of dental caries disease is multifactorial in that simultaneous participation of multiple factors is required for caries to occur (Tanzer, 1992). The original still-prevailing theory explaining the disease process implicates carbohydrates, oral microorganisms, and acids as the main factors in the caries process. This acidogenic theory (or Miller's chemoparasitic theory, 1902) states that "Dental decay is a chemico-parasitic process consisting of two stages, the decalcification of enamel, which results in its total destruction, and the decalcification of dentin, as a preliminary stage, followed by dissolution of the softened residue. The acid which affects this primary decalcification is derived from the fermentation of starches and sugar lodged in the retaining centers of the teeth." Miller isolated many microorganisms from the human oral cavity, some acidogenic and some proteolytic. Miller believed that caries is caused by a variety of microorganisms. Two other theories explaining the disease process are the proteolytic theory and the proteolysis-chelation theory (Shafer et al., 1983).
Today, mutans streptococci are considered to be the main aetiological microorganisms in caries disease, with lactobacilli and other microorganisms participating in the disease progression (Tanzer, 1992). Occasionally, some other microorganisms have been traced as initiator microorganisms. Severe dental caries has been induced in hyposalivated rats infected with Lactobacillus fermentum (Ooshima et al., 1994). Acidogenesis at a low pH has also been reported for a group of non-mutans streptococci (van Houte et al., 1991).
When the caries disease has a very fast progression it is called rampant caries. A specific form of rampant caries involves the dentition of very young children. The condition includes a demineralization and cavitation of the labial surfaces of the maxillary primary incisors, followed by involvement of the first primary molars (Ripa 1988). The disease onset is between 1 and 2 years of age. A variety of names have been given to the condition; nursing caries, nursing bottle caries, early childhood caries, baby bottle tooth decay and baby bottle caries (Curzon and Pollard, 1994; Tinanoff, 1997). The prevalence of nursing caries in different populations varies between 1% and 80% of preschool children, the proportion of affected children being very low in many western societies (Curzon and Pollard, 1994; Milnes, 1996). Some reports have been made on increasing incidence of this condition in western countries (Duperon, 1995). In Finland, the prevalence can be estimated at about 1%, exact figures being unavailable (Paunio, 1993; Alaluusua, 1999). The aetiology of this condition is also considered multifactorial. In most cases, the affected child has frequently received fermentable carbohydrates as a drink when going to sleep, combined with a lack of toothbrushing with fluoride toothpaste (Curzon and Pollard, 1994). Moreover, the combination of frequent breast-feeding and low additional fluoride use are considered to be contributing factors in the process of nursing caries (Hallonsten et al., 1995; Weerheijm et al., 1998). In children with nursing caries, mutans streptococci can be so dominant that the plaque above the caries lesions consists almost entirely of these organisms (van Houte et al., 1982).
The fact that dental caries is a world-wide disease requiring vast economic resources and causing a great deal of discomfort has called upon attempts aimed at developing an accurate screening method for detection of the 5-20% of subjects comprising the high-caries-risk group (Pienihäkkinen, 1987; Roeters, 1994; Hausen, 1997). Because mutans streptococci are considered to be the predominant pathogens of dental caries disease, individuals heavily colonized by mutans streptococci were thought to automatically be at high risk for caries. Indeed, in young children, early mutans streptococcal colonization on tooth surfaces has been recognized as an indicator of later high scores of decayed, missing and filled surfaces in deciduous teeth (dmfs index) (Alaluusua and Renkonen, 1983; Köhler et al., 1984, 1988; Jokela, 1997). However, it has become evident that, although prevalence of the infection is indicative of the disease status on a population level, when it comes to an older child or adult, on the individual level, the caries risk rate cannot be accurately predicted on the basis of how heavily the subject is colonized by mutans streptococci (Stecksen-Blicks, 1985; Alaluusua et al., 1987, 1990; Alaluusua, 1993; Alanen et al., 1994; Hausen, 1997).
The term virulence describes the capacity of a parasite (a microorganism) to cause disease to its host, the organism it lives on or in. The property is quantitative and expresses the degree of pathogenicity, the ability to inflict damage to the host. The relationship between host and parasite is dynamic and depends on their individual characteristics and their interrelationship as well as on external factors. Virulence consists of bacterial properties required in the interaction between host and parasite, factors that promote the entry, colonization and growth of the pathogen within the host, including those required for opposing host defences and for nutrient acquisition (Madigan, 1997b). The virulence factors of microorganisms can be studied by the classical approach, by isolating bacteria from healthy and diseased subjects and comparing the phenotypic properties of these microorganisms. During the last two decades, approaches utilized in virulence studies have also included molecular biology methods, techniques for the manipulation of DNA in vitro (Hensel and Holden, 1996). Studies on the expression of the modified bacterial chromosome can subsequently be extended to cell culture and animal experiments.
As regards the mutans streptococci, properties that affect their ability to cause dental caries disease are virulence factors promoting their colonization and survival in the biofilm, the dental plaque, that covers the tooth surfaces. Recognized virulence factors of mutans streptococci are adhesin-like cell surface proteins, acid tolerance, acid production, and production of glucosyltransferases, mutacin and intracellular polysaccharides (Kuramitsu, 1993). In addition to the recognized virulence factors, other properties of the microorganism may influence virulence. One of the suggested virulence factors is the proteolytic activity of mutans streptococci (Homer et al., 1990; Harrington and Russell, 1994; Jackson et al., 1997). S. mutans has been shown to produce two extracellular proteases, possibly metalloproteases, capable of degrading both gelatin and collagen-like substrates (Harrington and Russell, 1994). Currently under debate is whether metalloproteases detected in connection with mutans streptococci are produced by the microorganism or whether they are host-derived (Tjäderhane et al., 1998). Regarding the sIgA protease activity of mutans streptococci, the presiding view is that the organism itself does not produce this protease (Marcotte and Lavoie, 1998). Many oral streptococci do produce sIgA protease, which impairs the host defence by cleaving the secretory IgA present, and apparently mutans streptococci benefit from protease produced by the primary colonizers (Marcotte and Lavoie, 1998). Another trait enabling survival is the ability of mutans streptococci to rapidly adapt to the environment by microbial genetics phenomena; this property has been suggested to be an essential element in the dominance of S. mutans in cariogenic dental plaque (Burne et al., 1997; Burne, 1998). As a rule, the ability of cells to take up exogenous DNA, the regulation of natural genetic competence in bacteria, is dictated by nutritional conditions and cell-to-cell signalling (Solomon and Grossman, 1996).
The mutans streptococci synthesize extracellular polysaccharides from sucrose to increase their stickiness. Surface proteins of S. mutans also participate in adherence. In S. sobrinus, the adherence is probably primarily mediated by extracellular polysaccharides, with a minor influence by surface proteins (Gibbons et al., 1986). In addition to the microbial properties, host factors may affect adherence, and salivary components can function as receptors in oral pellicles for microbial adhesion to host surfaces (Scannapieco, 1994).
Glucosyltransferases (GTFs) and fructosyltransferases (FTFs) catalyse the synthesis of water-soluble and water-insoluble glucan and fructan polymers from sucrose (Loesche, 1986). The nucleotide sequences of gtf genes from different oral streptococci comply with the same basic pattern, and the GTFs are approximately 1500 amino acids long (Russell, 1994). Streptococcal GTFs have two common functional domains. The amino-terminal portion, the catalytic domain, is responsible for the cleavage of sucrose, and the carboxyl-terminal portion, the glucan binding domain, is responsible for glucan binding (Colby and Russell, 1997). S. mutans produces at least one FTF and three GTFs (Sato and Kuramitsu, 1986; Russell, 1994). GTF-I and GTF-SI enzymes, products of gtfB and gtfC genes, primarily catalyse the synthesis of water-insoluble glucans, whereas GTF-S, the product of gtfD , mainly catalyses the synthesis of water-soluble glucans. S. sobrinus has four gtf genes (Russell, 1994). Of the four GTFs of S. sobrinus , GTF-I produces water-insoluble glucans, while the other three produce water-soluble glucans (Hanada et al., 1993). The three gtf genes from S. mutans have been thoroughly assessed by comparative sequence analysis, revealing that interstrain differences of gtfB and gtfD are limited, but gtfC exhibits significant interstrain variability (Fujiwara et al., 1998).
The in vitro effects of the GTFs have been extensively studied; however, few studies with clinical strains have been performed to date.
S. mutans cells express a predominant surface protein called P1 (or I/II, B, IF, SR or PAc), which functions in the binding of mutans streptococci to human salivary pellicle-coated surfaces (Crowley et al., 1999). The apparent functional equivalent to this protein is the spaA protein of S. sobrinus (Kuramitsu, 1993).
The mutans streptococci ferment many different sugars, and they appear to metabolize sucrose to lactic acid more rapidly than other oral bacteria. This is thought to be related to the multitude of enzyme systems catalysing the reactions of transport and metabolism of sucrose expressed by these organisms (Kuramitsu, 1993). These metabolic reactions render the dental plaque acidic in the presence of a fermentable carbon source, and the acid tolerance of the mutans streptococci enables them to continue metabolisms even at low pH. It has been demonstrated that strains of mutans streptococci are more acid tolerant than all other bacteria examined, with the exception of lactobacilli (Loesche, 1986). An inducible property exists in mutans streptococci which permits adaptation to acidic environments (Hamilton and Buckley, 1991; Birkhed et al., 1993). This property of acid tolerance (or acidurance) appears to be connected with the membrane-associated H + (proton)-translocating ATPase of these organisms (Bender at al., 1986).
Many bacteria produce bacteriocins, i.e. antibacterial peptides, to interfere with the growth of other closely related microorganisms (Jack et al., 1995). Bacteriocins are ribosomally synthesized and usually require extensive posttranslational modification for activity. The genes involved in the synthesis and modification of bacteriocins are often carried by a plasmid or a transposon (Madigan et al., 1997a). Bacteriocins are frequently named according to the bacterial species producing them; bacteriocin produced by mutans streptococci is called mutacin. Mutacin production is usually not plasmid encoded (Caufield et al., 1990). When bacteriocin activity is plasmid encoded, the plasmid generally also confers bacteriocin immunity to the microorganism (Jack et al., 1995). Some bacteriocins also have a commercial value. Nisin, which is produced by Lactococcus lactis strains, has been used for more than 30 years as a preservative in the food industry (Jack et al., 1995).
Several mutacins have been purified and biochemically characterized (Fukushima et al., 1982; Ikeda et al., 1982; Hamada et al., 1986b; Loyola-Rodriguez et al.,1992; Novak et al. 1994; Chikindas et al., 1995). In replacement experiments, strains producing increased amounts of mutacin have been shown to colonize more easily (van der Hoeven and Rogers, 1979; Hillman et al., 1987).
Most strains of S. mutans and S. sobrinus produce intracellular iodine-staining polysaccharides (IPS) from sucrose, which, according to results from experiments with rats, may contribute to their virulence (Kuramitsu, 1993). Because of this intracellular polysaccharide storage, these cariogenic bacteria have the ability to continue fermentation in the absence of exogenous food supplies (Loesche, 1986).
The traditional preventive treatment approaches for dental caries disease include dietary counselling on reduced intake of refined sugars, oral hygiene instruction and topical application of fluoride. To aid these preventive measures, antimicrobial preparations, such as chlorhexidine (CHX), can be used to reduce the numbers of mutans streptococci colonizing an individual (Emilson and Fornell, 1976; Twetman and Petersson, 1999). The outcome of all these measures, including CHX use, is largely dependent on patient compliance.
Chlorhexidine, developed in the 1950's, is a bisbiguanide that typically exerts its action on Gram-positive bacteria, particularly on mutans streptococci (Hennessey, 1973; Emilson, 1994; Järvinen et al., 1995). Clinically, the reduction in mutans streptococcal counts not only reduces the caries increment, but may also reduce the probability of transmission of mutans streptcocci from mothers to their young children (Emilson, 1994; Köhler and Andréen, 1994). The antibacterial action of CHX is based on its adsorption on bacterial surfaces. CHX can be administered in solutions, dental gels, varnish or even chewing tablets (Luoma, 1992; Nuuja, 1992). The combination of CHX with sodium fluoride is a more potent inhibitor of acidogenic streptococci than either CHX or fluoride alone (Luoma, 1972; McDermid et al., 1985). The inhibiting effect of CHX on plaque formation has been reported to persist in human subjects even upon 2 years of continuous use (Gjermo and Eriksen, 1974). Surfaces which are heavily infected by mutans streptococci are more rapidly recolonized after antimicrobial treatment. In practice, posterior teeth are recolonized more readily than other tooth surfaces (Lindquist, 1991). Adverse effects of CHX are rare, but both IgE-mediated local responses and anaphylactic shock reactions may occur (Ebo et al., 1998).
Bacterial resistance to antiseptics and disinfectants may be an inherent or acquired property of an organism. Antimicrobial resistance can evolve through acquisition of genetic material, such as plasmids or transposons, or by mutation. Thus far, laboratory tests have failed to conclusively demonstrate the possibility of "training" organisms to become chlorhexidine-resistant (Russell and Day, 1993). Many bacterial species, e.g. Proteus and Providencia , have an intrinsic chlorhexidine resistance (Russell and Day, 1993).