| Sign In to gain access to subscriptions and/or personal tools. |
Phenotypic Heterogeneity of Streptococcus mutans in Dentin
1 Department of Operative Dentistry, Periodontology and Preventive Dentistry, Saarland University, D-66421 Homburg, Germany; Correspondence: * corresponding author, Stefan.Rupf{at}uniklinikum-saarland.de
Information concerning phenotypic heterogeneity of Streptococcus mutans in carious dentin is sparse. Matrix-assisted laser-desorption/ionization-time-of-flight mass-spectrometry (MALDI-TOF-MS) facilitates the phenotypic differentiation of bacteria to the subspecies level. To verify a supposed influence of restorative treatment on the phenotypic heterogeneity of S. mutans, we isolated and compared a total of 222 S. mutans strains from dentin samples of 21 human deciduous molars during caries excavation (T1) and 8 wks (T2) after removal of the temporary restoration. Phenotypic heterogeneity was determined by MALDI-TOF-MS and hierarchical clustering. Thirty-six distinct S. mutans phenotypes could be identified. Although indistinguishable phenotypes were found in the same teeth at T1 and T2, as well as in different teeth of individual participants, the phenotypic heterogeneity increased significantly, from 1.4 phenotypes per S. mutans-positive dentin sample at T1 to 2.2 phenotypes at T2. We attribute this to an adaptation of S. mutans to the modified environment under the restoration following caries excavation.
Key Words: Streptococcus mutans • phenotypic heterogeneity • MALDI TOF-MS
Caries is an infectious process that results in a copious microbial colonization of the dentin and, finally, its destruction. During this process, complex ecosystems are formed, offering a protective habitat in which bacteria can grow. A high selection pressure is imposed through competition for limited habitats and nutrients within these ecosystems. Caries therapy causes a significant change in the living conditions of the micro-organisms involved in the caries progression process. When decayed dentin is removed, the complex microbial ecosystem is mechanically destroyed, disinfecting agents are applied, and when the cavities are filled, nutrients are withdrawn. However, the dentin is not sterile after caries excavation (Lager et al., 2003). The species Streptococcus mutans is undisputedly associated with caries progression (Loesche, 1986) and is able to survive in the hard, yet porous, inner dentin (Banerjee et al., 2000). Little information is available on the genotypic and phenotypic stability of mutans streptococci in the dentin during the carious process and after caries excavation and filling therapy. Despite the high heterogeneity of S. mutans, no particular genotype or phenotype could be associated with the progression of caries until now (Liu et al., 2004; Nascimento et al., 2004). Positive correlations have, however, been found between the presence of multiple serotypes of mutans streptococci and the caries experience (Shibata et al., 2003; Shemesh et al., 2007). It has been shown for S. mutans that different phenotypes, namely, different mutacin production profiles of the same genotype, can occur (Kamiya et al., 2005). The aim of the present study was to characterize the phenotypic heterogeneity of S. mutans to the subspecies level with strains harvested from hard dentin. For phenotyping, matrix-assisted laser-desorption/ ionization time-of-flight mass spectrometry (MALDI-TOF MS) was applied. This method has been shown to facilitate differentiation of bacteria to the species as well as to the subspecies level (Rupf et al., 2005; Fagerquist et al., 2006; Friedrichs et al., 2007).
Strains of S. mutans Two-hundred-twenty-two S. mutans strains, isolated from hard dentin of 21 deciduous molars in 16 eight-year-old children, were available for the study. Dentin samples were collected from the cavity floor by means of a sterile rose bur after chemo-mechanical caries excavation (baseline sample: T1, n = 112 S. mutans strains) with Carisolv gel (MediTeam, Sävedalen, Sweden). Teeth were temporarily filled by conventional glass-ionomer cement (Ketac-Bond, 3M-ESPE, Seefeld, Germany). After 8 wks, a re-entry was performed, and the second set of hard dentin samples was taken (control sample: T2, n = 110 S. mutans strains). During clinical procedures, teeth were isolated by rubber dam. Dentin samples were suspended in 0.9 % NaCl and plated onto mitis salivarius agar with Bacitracin (MSB, Gold et al., 1973) in ten-fold dilution series. The MSB plates were incubated anaerobically at 35 ± 2°C for 7 days (GasPak, Becton Dickinson, Cockeysville, MD, USA). The total numbers (colony-forming units, cfu) were counted at T1 and at T2 with a mean of 0.1 x 102 cfu per sample. These S. mutans strains were selected by typical colony morphology and confirmed by physiological and biochemical characteristics, membrane fatty acids, and species-specific polymerase chain-reactions (Kneist and Heinrich-Weltzien, 2001; Rupf et al., 2001). Ethics approval for this study was granted by the ethics committees of the Universities of Jena and Leipzig. Informed consent was obtained from the parents and children recruited into the study.
Reference Strains
MALDI TOF-MS Sample Preparation For MALDI-TOF MS, from 106 to 107 bacterial cells were washed twice in 200 µL of nanopure water and centrifuged at 15,000 rpm for 1 min. Bacterial cell pellets were re-suspended in 10 µL of matrix (alpha-cyano-4-hydroxycinnamic acid, 20 mg/mL in 0.1% trifluoroacetic acid [TFA]/acetonitrile 1:2), and a 2-µL quantity of the mix was applied to a stainless steel target (scout 384, Bruker Daltonics, Bremen, Germany). Samples were allowed to crystallize, washed twice with 0.1% TFA, and re-crystallized in 0.1% TFA/acetonitrile 1:2.
MALDI TOF-MS Parameters
Identification of Peaks in MALDI-TOF MS Spectra and Similarity Analysis The similarity of the two spectra was then characterized by the respective Jaccard coefficient:
In the equation, ni,j is the number of peaks which are present in both samples, while ni and nj are the number of peaks which occur in only one of the two spectra. By this, a symmetric matrix of pair-wise similarities was formed (peak-mass-based similarity matrix).
In addition, a similarity matrix (
The similarity of samples i and j was obtained by a summation over all peaks contributing to either sample i or sample j. wik represents the intensity of peak i in the spectrum k. Distance matrices (*i,j) were calculated from normalized similarity matrices according to:
and were analyzed by agglomerative hierarchical clustering algorithms with the use of an average linkage function (Unweighted Pair Group Method with Arithmetic mean, UPGMA). Results were visualized as rooted dendrograms.
Identification of Phenotypes
Statistical Analysis
A maximum of 10 strains was isolated from each S. mutans-positive dentin sample. On average, 8.1 strains could be harvested at T1 and 8.5 strains at T2. The ratio of S. mutans-positive dentin samples to the number of investigated dentin samples was 0.67 at T1 and 0.62 at T2. In 8 of the investigated dentin samples, S. mutans strains were available at both T1 and T2, 6 samples were positive at T1 only, and 5 only at T2. No S. mutans was found at either point in two samples (Table  ).
The reference strains S. mutans Ingbritt and JB 1600 were identified as the most related reference strains. They defined the  -level for the phenotypic discrimination of the isolated bacterial strains (Figs. 1A, 1B, 2, 3).
Based on the established -level, up to 3 different S. mutans phenotypes could be isolated from the dentin samples. Overall, 36 different phenotypes were observed. While 19 phenotypes were counted on the tooth level at T1, 29 phenotypes were identified at T2. The difference in the ratios of 1.4 phenotypes per S. mutans-positive dentin sample at T1 and 2.2 phenotypes at T2 was found to be statistically significant (p = 0.0025). Regarding the 8 paired dentin samples which were S. mutans-positive both at baseline as well as at control, an increased heterogeneity was found for 7 samples, and a decrease was noted for 1 sample. The ratio of 1.25 phenotypes/sample at T1 differed significantly from that of 2.3 phenotypes/sample at T2 (p = 0.01).
Indistinguishable phenotypes were found in the same teeth at baseline as well as at control in 5 of the available 8 paired dentin samples which were S. mutans-positive at T1 as well as at T2. All dentin samples obtained from different children contained different phenotypes, with only one exception (Table
The basis of current caries management is the arresting of caries progression, and the removal of carious dentin is often necessary in clinical dentistry. There is a broad consensus among dentists that carious dentin should be removed until the dentin is hard when probed. However, the latter is often not germ-free. Remnant bacteria can be found in the inner hard dentin layer of the treated cavities, independent of the caries excavation procedure applied, and a low level of infection of the dentin is accepted as clinically harmless to the pulp (Banerjee et al., 2000; Toi et al., 2003; Wicht et al., 2005). The remaining bacteria seem to be further reduced by sustainable filling techniques (Weerheijm et al., 1999; Kuramoto et al., 2005; Paddick et al., 2005; Takahashi et al., 2006). However, recurrent caries caused by marginal disintegration of the fillings and remaining bacteria continue to be a concern in dentistry (Mjör, 2005). A previous investigation demonstrated a reduction of bacterial settlement in the dentin under composite fillings (Paddick et al., 2005). Glass-ionomer cements, which are often used as base under composite fillings, have been proven to have direct antibacterial effects as well (Takahashi et al., 2006). In our study, the cavity floors displayed a low level of infection, and 64% of the dentin samples were S. mutans-positive. No differences as to the positive isolation of S. mutans from the dentin samples were noted between the sample dates. The average number of strains per positive dentin sample was also identical. The heterogeneity analysis of the S. mutans isolates by MALDI-TOF MS on the basis of the heterogeneity of the available reference strains resulted in 1 to 3 different phenotypes per cavity. The heterogeneity of the isolates increased slightly between baseline (T1) and control (T2). This effect became even more pronounced when the results of the dentin samples containing S. mutans on both occasions were compared. In only 1 instance was a decrease in the number of phenotypes noted, whereas an increase was apparent in 7 cases. We suppose that the average increase in the phenotypic heterogeneity was caused by selection pressure on the bacteria resulting from the interruption of nutrition. Indistinguishable phenotypes were present in the different teeth of individual children at baseline and control. These phenotypes may represent identical strains, indicating their perfect adaptation and resilience in the dentin over prolonged periods of time. The increasing number of phenotypes at T2 could be due to the fact that phenotypes underrepresented at T1 found niches for settlement created by the eradication of opponents. In this context, intact cell MALDI-TOF MS seems to be a suitable method to recognize changes in the heterogeneity of bacterial strains under changing environmental conditions. Our investigation confirms the substantial heterogeneity of S. mutans, and proves that S. mutans can survive under adverse conditions for a limited period of time. Further investigations have to clarify whether the phenotypes differ within different ecological niches of the oral cavity, e.g., saliva, plaque, and carious dentin, and whether selection pressures lead to an increase of phenotypic heterogeneity.
This study was supported by a junior scientist’s grant from the Medical Faculty of the University of Leipzig as well as by institutional funds from the Department of Preventive Dentistry of the University of Jena. We thank Ms. Annette-Kaetzke for technical assistance, as well as Ms. Liz Weinstein and Ms. Anja Niemeyer-Rupf for critical reading of the manuscript. Received for publication February 11, 2008. Revision received July 10, 2008. Accepted for publication September 23, 2008.
Journal of Dental Research, Vol. 87, No. 12,
1172-1176 (2008) This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
i,j) which considers both peak masses and differences in the peak intensities was calculated according to: 
