- Email: [email protected]
Microbiome of Deep Dentinal Caries from Reversible Pulpitis to Irreversible Pulpitis
Basic Research—Biology
Microbiome of Deep Dentinal Caries from Reversible Pulpitis to Irreversible Pulpitis Jinxin Zheng, BDS,*a Zhou Wu, MDS,†a Kaijun Niu, PhD,‡ Yanan Xie, BDS,* Xiaoli Hu, BDS,* Jieni Fu, BDS,* Dongtao Tian, BDS,§ Kaiyu Fu, BDS,¶ Bo Zhao, BDS,* Weiyang Kong, BDS,* Cuicui Sun, BDS,* and Ligeng Wu, MDS* Abstract Introduction: This study examined the identity of the microbiome of deep dentinal caries and its correlation with the inflammation status of caries-induced pulpitis. Methods: Seventy-five cases were diagnosed based on the American Association of Endodontics’s diagnostic criteria and divided into 4 groups: normal pulp with deep caries (NP; n = 13), reversible pulpitis with only cold-evoked pain (CRP; n = 17), reversible pulpitis with both cold/heat-evoked pain (CHRP; n = 24), and symptomatic irreversible pulpitis (SIP; n = 21). Samples were sequenced by 16S rDNA. Alpha and beta diversity were determined. Linear discriminant analysis effect size (LEfSe) analysis was used to detect intergroup differences, and receiver operating characteristic (ROC) curves were generated to assess the role of the caries microbiome in caries-induced pulpitis. Results: The 16S rDNA sequencing yielded 9100 operational taxonomic units. Lactobacillus had the highest relative abundance at the genus level among the 4 groups. There were significant differences in the distribution of the microbiome among the groups. In an alpha diversity analysis, species richness differed between the CRP group and the other groups. In a beta diversity analysis, the distribution of microorganisms in the SIP group was significantly different from those in the other 3 groups. LEfSe analysis indicated substantial differences in the microbiome among the groups, and the areas under the ROC curves (AUC) were all high (AUC: 0.734– 0.952). Conclusions: Characterization of the caries microbiome has the potential to become an auxiliary method for the diagnosis of pulpitis. This finding may prompt new research on diagnostic strategies for caries-induced pulpitis. (J Endod 2019;45:302–309)
Key Words microbiome, dentinal caries, irreversible pulpitis, Lactobacillus, reversible pulpitis
C
aries is a common Significance dental disease mediMicrobial determination indicated that species and ated by a variety of microquantity of the microorganism were consistent with organisms (1–4). Without clinical diagnosis and were associated with the appropriate treatment, severity of pulpitis. Improved knowledge of the mimicroorganisms in deep crobiome may help to develop diagnostic tools. caries invade the dental pulp through the dentinal tubules or directly cause pulp exposure (5–7), leading to pulpitis (8–10), which is clinically categorized as reversible and irreversible pulpitis (11). Some studies have indicated that the bacterial species associated with deep caries differ substantially from those of inflamed pulp. The main pathogens of caries are oral Streptococcus, Lactobacillus, Actinomyces, Veillonella, Neisseria, and others (12), whereas the dominant genera associated with endodontic diseases are Fusobacterium, Porphyromonas, Peptostreptococcus, Prevotella, Parvimonas, Dialister, and others (13, 14). Other studies have shown that most bacterial species in carious dentin are also found in infected root canals (15) and that there are positive associations between specific bacteria in carious dentin and irreversible pulpitis (10), suggesting that bacteria present in the most advanced layers of dentinal caries are candidate pathogens for inducing pulpitis and initiating pulp inflammation (15). Microorganisms in deep caries are associated with dental pulp symptoms (10, 16–18). However, less is known about the characteristics of the microbiome at the forefront of carious lesions in teeth with different pulpal diagnoses. The aim of this study was to characterize the microbiome of the dentinal caries microenvironment and the microbial diversity in deep dentinal caries of teeth with pulpitis by 16S rDNA sequencing as well as to evaluate the correlation between the microbiome and the inflammation status of caries-induced pulpitis. We also preliminarily investigated whether the differences in the microbiome in deep dentinal caries can contribute to the diagnosis of caries-induced pulpitis.
From the *Department of Endodontics, School of Stomatology, and ‡Nutritional Epidemiology Institute and School of Public Health, Tianjin Medical University, Tianjin, China; †Department of Stomatology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China; §Department of Stomatology, Wuqing People Hospital, Tianjin, China; and ¶Department of Orthodontics, Tianjin Stomatology Hospital, Tianjin, China. a Co–first authors. Address requests for reprints to Ligeng Wu, MDS, Department of Endodontics, School of Stomatology, Tianjin Medical University, #12 Qi Xiang Tai Road, He Ping District, Tianjin, 300070, China. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2018 American Association of Endodontists. https://doi.org/10.1016/j.joen.2018.11.017
302
Zheng et al.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology Materials and Methods Study Population This study was approved by the ethics committee of the Stomatology Hospital of Tianjin Medical University (project number TMUDHhMEC2015010). In total, 75 patients with deep caries were enrolled at Stomatology Hospital of Tianjin Medical University between February 2015 and February 2018, and informed consent was obtained from all subjects. Patients ranged in age from 12 to 60 years. Inclusion Criteria and Exclusion Criteria Inclusion criteria: 1. All cases had deep caries lesions. Periapical radiographs showed that the root apices of all teeth had completed development and that the depth of deep caries of anterior or posterior teeth exceeded two-thirds of the depth of dentin. 2. Caries affected the proximal or occlusal surfaces of the teeth. 3. Teeth without previous restorations. Exclusion criteria: 1. 2. 3. 4. 5.
Pregnancy. Antibiotics, analgesics, or anesthetics usage before treatment. Moderate or severe periodontitis or furcation involvement. Cracked tooth, apical periodontitis. Teeth with subgingival caries that could not be isolated with rubber dam. 6. Teeth with no response to pulp tests (described later). 7. Periapical radiographs revealed a periapical shadow or significant widening of the periodontal ligament space. 8. Infectious diseases, serious systemic disease, mental disorder, or failure to cooperate with treatment.
Cold Testing A large ＃2 cotton pellet with a refrigerant spray and a temperature of 26.2 C (1, 1, 1, 2-tetrafluoroethane) (Endo-Frost; Coltene Whaledent, Cuyahoga Falls, OH) was sprayed and applied to the middle third of the buccal surface of the crown of the tooth for 3 seconds and the patients provided their response. Heat Testing A light layer of lubricant was placed on the surface of the tooth and a heated gutta-percha rod was placed on the middle third of the buccal surface of the crown of the tooth for 3 seconds, and the patients provided their response.
Electric Pulp Testing After clinical isolation with cotton rolls and drying of the tooth with air, the probe tip of an electric pulp tester (SybronEndo, Orange, CA) was coated with toothpaste and applied to the middle third of the buccal surface of the crown of the tooth. All teeth were tested 3 times, and the readings from the pulp tester were recorded. A tooth with a response at a level lower than 60 was considered vital. The adjacent tooth or contralateral "normal" tooth was also tested to establish a baseline response. Pulp Diagnosis The diagnosis of pulpitis was based on clinical and radiographic findings and according to the criteria of the American Association of Endodontists (19). Normal pulp: no spontaneous pain, the responses to cold and heat tests were the same as those of control teeth, and no lingering pain. Reversible pulpitis: no spontaneous pain, sensitive responses to cold and heat tests compared with control teeth, and no more than 30 seconds of lingering pain after the removal of the stimulus. Based on the result of the heat test, reversible pulpitis was subdivided into 2 stages: reversible pulpitis with only cold-evoked pain and reversible pulpitis with both cold/heat-evoked pain. Symptomatic irreversible pulpitis: spontaneous pain, sensitive responses to cold or heat tests compared with control teeth, and more than 30 seconds of lingering pain after the removal of the stimulus. Clinical Grouping In total, 75 teeth (2 anterior and 73 posterior teeth) with deep caries were included. The cavity wall was located at either the proximal (n = 62) or occlusal (n = 13) surface. Based on patient history and clinical pulp tests, the cases were divided into 4 groups according to the severity of pulpal inflammation: (1) normal pulp with deep caries (NP group; n = 13), (2) reversible pulpitis with only cold-evoked pain (CRP group; n = 17), (3) reversible pulpitis with both cold/heat-evoked pain (CHRP group; n = 24), and (4) symptomatic irreversible pulpitis (SIP group; n = 21). The groups and diagnoses are shown in Table 1. Sample Collection One clinician examined the cases and collected the samples under aseptic conditions. The selective removal of carious tissue (20) was performed to collect samples. The mouth was rinsed with hydrogen peroxide before caries collection, and after local anesthesia, the carious teeth and the adjacent teeth were isolated with a rubber dam (Kant, St. Gallen, Switzerland). The surfaces of the tooth were disinfected with a 2% chlorhexidine gluconate solution (Longly Biomedical Wuhan Co
TABLE 1. Group Assignments and Diagnosis of 75 Cases According to the American Association of Endodontists Diagnostic Criteria Groups
NP
CRP
Patient history Pulp thermal tests
SP ( ) Cold* ( ) Heat† ( )
SP ( ) Cold (+) lp < 30 s Heat ( )
Cases Numbers Diagnosis
13 NP1–13 Normal pulp
CHRP
SP ( ) Cold (+) lp < 30 s Heat (+) lp < 30 s 17 24 CRP1–17 CHRP1–24 Reversible pulpitis
SIP SP (+) Cold (+) lp > 30 s Heat ( )
Cold ( ) Heat (+) lp > 30 s
Cold (+) lp > 30 s Heat (+) lp > 30 s
21 SIP1–21 Symptomatic irreversible pulpitis
CHRP, reversible pulpitis with both cold/heat-evoked pain; CRP, reversible pulpitis with only cold-evoked pain; lp, lingering pain; NP, normal pulp with deep caries; SIP, symptomatic irreversible pulpitis; SP, spontaneous pain. *Cold test; +, sensitive; , same as control teeth (control tooth: an adjacent tooth or contralateral normal tooth of the same name). † Heat test; +, sensitive; , same as control teeth (control tooth: an adjacent tooth or contralateral normal tooth of the same name).
JOE — Volume 45, Number 3, March 2019
Characterization of the Caries Microbiome
303
Basic Research—Biology Ltd, Wuhan, China). Undermined enamel, debris, and superficial carious tissue were removed using sterile high-speed burs (Ningbo Xinyuan Dental Equipment Co, Ltd, Zhejiang, China) under water cooling. The cavity was irrigated with 5 mL 0.9% sterile saline (China Otsuka Pharmaceutical Co, Ltd, Tianjin, China). The lift-draw method was used to remove the superficial layer of the caries using a low-speed ball bur (Ningbo Xinyuan Dental Equipment Co, Ltd, Zhejiang, China). Using a sterile spoon excavator, samples per tooth of the deepest layer of the caries were transferred to a sterile Eppendorf tube (EP tube) containing 0.5 mL sterile saline. The EP tubes with samples were numbered according to group and registration order and stored immediately at 80 C in a refrigerator. The EP tube was removed and stored in dry ice before 16S rDNA sequencing.
Sample Quality Control
1. 2. 3. 4. 5.
To ensure quality in the sampling, the following control measures were used: The surface of the tooth was disinfected using chlorhexidine before sampling. Teeth that could not be isolated with rubber dam were excluded. Only one experienced clinician examined the cases and collected the samples to ensure consistency in all samples. Eppendorf tubes were sterile. The whole operation was performed under sterile conditions.-
16S rDNA Sequencing The total genomic DNA from samples was extracted using the cetyltrimethylammonium ammonium bromide/sodium dodecyl sulfonate method. DNA concentration and purity were monitored on 1% agarose gels. To meet sequencing requirements, samples with less than 150 ng DNA were eliminated. DNA was diluted to 1 ng/mL using sterile water. The 16S rDNA gene regions (16S V3–V4) were amplified using specific primers with barcodes. All polymerase chain reactions (PCRs) were performed using Phusion High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA) under the following conditions: 95 C for 5 minutes, followed by 34 cycles of 94 C for 1 minute, 57 C for 45 seconds and 72 C for 1 minute, and then a final elongation step at 72 C for 10 minutes and 16 C for 5 minutes. An equal volume of 1 loading buffer (containing SYB green) was added to the PCR products, followed by separation by electrophoresis on a 2% agarose gel for detection. Samples with a bright main strip at 400 to 450 base pairs (bp) were chosen for further experiments. PCR products were mixed in equal ratios. Then, mixed PCR products were purified using the Qiagen Gel Extraction Kit (Qiagen, Hilden, Germany). DNA libraries were generated using the TruSeq DNA PCR–Free Sample Preparation Kit (Illumina, San Diego, CA). The library was sequenced using the HiSeq2500 PE250 sequencer (Illumina) and 250-bp paired-end reads were generated. Then, the reads of each sample were spliced to obtain the original tag data. After filtering, the high-quality tag data were obtained and compared with sequences in the database (Gold database, http:// drive5.com/uchime/uchime_download.html). Operational taxonomic units (OTUs) were obtained using Uparse v7.0.1001 (http://drive5. com/uparse/) for species annotation and microbial diversity analysis. Statistical Analysis Alpha diversity was estimated to analyze the richness and diversity of microbial species in the samples based on observed species. Results were calculated using Qiime (version 1.9.1), and rarefaction curves and 304
Zheng et al.
species accumulation boxplot were obtained using R software (Version 2.15.3). Beta diversity was analyzed to evaluate differences among samples in species complexity. Weighted UniFrac distances were calculated using Qiime, and trees were generated using the unweighted pair-group method with arithmetic mean (UPGMA). Anosim analysis was used to compare differences among groups with differences within groups. Linear discriminant analysis (LDA) effect size (LEfSe) analysis was performed, setting the LDA score to 4 by default. The accuracy of microbiome characterization in predicting the status of caries-induced pulpitis was analyzed using receiver operating characteristic (ROC) curves.
Results Taxonomic Identification and Relative Abundance A total of 9100 OTUs were detected in 75 samples, belonging to 3 kingdoms (bacteria, archaea, and unknown), 56 phyla, 145 classes, 222 orders, 429 families, and 1132 genera. In total, 8934 OTUs were assigned to bacteria, including 51 phyla, and 162 OTUs were assigned to archaea, including 5 phyla. Taxonomic assignments revealed 10 phyla in the NP, CRP, CHRP, and SIP groups, 4 of which had relatively high abundances (Fig. 1A): Firmicutes (45.9%, 36.3%, 37.1%, and 63.6%, respectively), Actinobacteria (32.0%, 34.5%, 36.2%, and 16.0%), Proteobacteria (11.1%, 10.7%, 10.2%, and 9.3%), and Bacteroidetes (7.5%, 12.6%, 12.2%, and 8.6%). The relative abundance of Lactobacillus was higher in the SIP group than in the other groups; Olsenella and Actinomyces had similar abundances in the CRP and CHRP groups. The relative abundances of Streptococcus, Pseudoramibacter, and Acinetobacter were higher in the NP group than in the other 3 groups (Fig. 1B, C). Microbial Community Rarefaction curves are shown in Figure 2A and Figure 2B. The curve for the CRP group was relatively steep, and the tail of the CRP7 curve was the steepest. The remaining 74 curves were relatively flat. In general, the CRP group differed from the other groups in terms of species richness. Species accumulation boxplot of the 75 samples is shown in Figure 2C. As the sample size gradually increased, microbial species did not increase significantly, indicating that the sample size was sufficient to capture species diversity. The UPGMA tree for the 4 groups at the phylum level revealed that the distributions of microbiome in the reversible pulpitis groups (CRP and CHRP groups) were similar, whereas the distribution of the microbiome in the SIP group was significantly different from those in the other 3 groups. The relative abundance of Firmicutes was the highest in the SIP group. Within the remaining 3 groups, the relative abundances of Firmicutes and Actinobacteria were higher than those of the other phyla (Fig. 2D). Analysis of Intergroup Differences in Species According to ANOSIM (analysis of similarities), differences among groups were greater than differences within groups (R > 0; data not shown). The differential species among the groups are shown in Figure 3 and Supplemental Table S1. The genera Lactobacillus, Actinomyces, and Unidentified_Veillonellaceae differed among the groups. The relative abundance of Lactobacillus was the highest in the SIP group, and Actinomyces and Unidentified_Veillonellaceae were relatively highly abundant in the CHRP group. JOE — Volume 45, Number 3, March 2019
Basic Research—Biology
Figure 1. Differences in the microbial composition and distribution among the 4 groups. (A) Average of the top 10 most abundant phyla identified in each group. (B) Members of the top 10 genera. (C) Heatmap of the top 10 genera. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
Associations Between the Microbiome and Reversible/ Irreversible Pulpitis ROC curves of the differential species among the groups were used to evaluate associations between the microbiome in deep dentinal caries and specific pulpal diagnoses. The area under the ROC curve (AUC) can be used to compare the accuracy of the identity of the microbiome and clinical examination in the diagnosis of caries-induced pulpitis. The AUCs for NP versus CRP, CRP versus CHRP, CRP versus SIP, and CHRP versus SIP were 0.864 (95% confidence interval [CI], 0.71– 1), 0.851 (95% CI, 0.71–0.99), 0.952 (95% CI, 0.88–1), and 0.800 (95% CI, 0.63–0.97), respectively (Fig. 4A). Curves for reversible pulpitis and irreversible pulpitis compared with normal pulp are shown in Figure 4B. The AUCs for CRP, CHRP, and SIP from NP were 0.864 (95% CI, 0.71 –1), 0.734 (95% CI, 0.54–0.93), and 0.833 (95% CI, 0.67– 0.99), respectively. Random forest histograms for the intergroup differences in the microbiome of deep caries were generated (Fig. 4C–E). The analysis suggested that Lactobacillus in the SIP group was typically different from those in the other groups.
Discussion The 16S rDNA is a useful indicator for bacterial phylogenic analysis and taxonomic identification. Compared with traditional microbial JOE — Volume 45, Number 3, March 2019
identification methods, 16S rDNA sequencing identifies taxa based on differences in a variable region sequence (21). The approach is relatively reliable and is not restricted by the culture environment. In this study, 16S rDNA sequencing was used to detect more than 9000 OTUs, far more than the 700 known bacteria in oral caries (22). The reasons for the large number of OTUs found in this study compared with the previous studies (15, 17) are as follows: first, the caries samples in this study were recovered from asymptomatic teeth in addition to teeth with reversible pulpitis and irreversible pulpitis; in previous studies, the caries samples were recovered only from teeth with irreversible pulpitis; second, the sample size was larger than those of the previous studies; and third, the sample sequencing yielded many low-abundance species, and there was no culling and filtering of low-abundance species in the standard analysis. Genera identified in the caries of anterior teeth were also detected in posterior teeth caries. Similarly, genera in occlusal caries were also detected in proximal caries. However, considering that the amount of anterior tooth and occlusal caries samples was small in addition to the fact that there are presently no relevant reports that the different types or surfaces of teeth with caries can influence the composition of the microbiome, further studies are needed to explore the correlation between the microbiome of caries located in different types, or on different surfaces, of teeth with pulpitis.
Characterization of the Caries Microbiome
305
Basic Research—Biology
Figure 2. Diversity analysis. (A) Rarefaction curves for alpha diversity in the 4 groups. (B) Rarefaction curves for alpha diversity in the 75 samples. (C) Species accumulation boxplot of the 75 samples. (D) Beta diversity shown by UPGMA (unweighted pair-group method with arithmetic mean) of the top 10 most abundant phyla. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); OTU, operational taxonomic unit; SIP, symptomatic irreversible pulpitis (n = 21).
Rarefaction curves can directly reflect the rationality of sequencing data and can indirectly reflect species richness in a sample. In a comparison of the rarefaction curves for the 4 groups, the CRP group curve was relatively steep (Fig. 2A), indicating that there may be undetected
microorganisms in this group. The curves for the 75 samples showed substantial variation (Fig. 2B), suggesting that the microbial species richness varied greatly. The curve for CRP7 increased rapidly and did not exhibit a flat tail, suggesting that the microbial species in the sample
Figure 3. Linear discriminant analysis (LDA) effect size analysis of the 4 groups. LDA score >4 indicates a biomarker with statistical differences between groups. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
306
Zheng et al.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology
Figure 4. A classification system for identifying caries-induced pulpitis. (A, B) ROC for the random forest classifier based on the intergroup differences in species. (C, D, E) Intergroup differences in species based on the random forest model. Lengths of bar in the histogram represent the mean decrease in accuracy, which indicates the importance of the intergroup differences in species for classification. The color denotes the enrichment of intergroup species differences in NP (red), CRP (purple), CHRP (blue), and SIP (green) according to the odds ratio score. AUC, area under the receiver operating characteristic (ROC) curves; CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
were highly abundant and that some microorganisms were undetected. The curves for CHRP7, CHRP2, CHRP1, and CRP4 exhibited rapid increases and flat tails, suggesting that the abundance of microbial species was high and that detection was relatively complete. The rarefaction curves for the other 70 samples increased slowly and had flat tails, indicating low abundances and complete detection. Overall, among the 75 samples, the detection of the microbial species of 74 samples was relatively complete. In the species accumulation boxplot of the 75 samples (Fig. 2C), as the sample size increased, the trend of the boxplot gradually became flat, suggesting that the microbial species did not increase significantly; that is, even if the sample size continues to increase, only 1 or 2 additional types of microorganism may be detected; and all the samples of the 4 groups met the requirement of at least 3 biological duplications per group (23), basically achieving the 16S rDNA analysis requirement, with a certain degree of repeatability. This suggests that the sample size was sufficient and that reliability was high. According to the thermal examination, reversible pulpitis was subdivided into 2 stages in our study: reversible pulpitis with only coldevoked pain and reversible pulpitis with both cold/heat-evoked pain. Previous studies have shown that it is the A fibers that have a relatively low threshold of excitability to external stimuli rather than the C fibers JOE — Volume 45, Number 3, March 2019
that are activated by hydrodynamic stimuli (eg, heat, cold) (24, 25). However, the C fibers may generate a response if heat is applied to increase the temperature of the dentin-pulp border, and pulpitis characterized by pain is more likely to be associated with C fiber activity, which is indicative of pulpal tissue injury (26). Therefore, reversible pulpitis with both cold/heat-evoked pain may be more advanced than reversible pulpitis with only cold-evoked pain. The UPGMA trees for the 4 groups showed that the distributions of microbiome were similar at the 2 stages of reversible pulpitis (Fig. 2D), indicating consistency with respect to clinical diagnosis, grouping, and microbial detection. In the normal pulp with deep caries and reversible pulpitis, both Firmicutes and Actinobacteria were highly abundant. In symptomatic irreversible pulpitis, the abundance of Firmicutes was the highest (Fig. 1A). These results might indicate that Actinobacteria decreases and Firmicutes increases as caries progresses. Similar results were obtained in a longitudinal study of young children (12). Firmicutes and Actinobacteria accounted for most of the sequences detected in the 4 groups. The former was mostly represented by Lactobacillus, Pseudoramibacter, and Streptococcus, whereas the most common representative genera of the latter were Propionibacterium, Olsenella, and Actinomyces, suggesting their strong association
Characterization of the Caries Microbiome
307
Basic Research—Biology with deep dentinal caries as well as reversible and irreversible pulpitis. This result confirms the results of previous studies showing that microorganisms present in the front line of dentinal carious lesions are likely to be involved in the development of pulp inflammation (10, 27, 28). According to the clinical diagnosis, the status of the pulp was divided into 4 phases: normal pulp with deep caries, reversible pulpitis with only cold-evoked pain, reversible pulpitis with both cold/heatevoked pain, and symptomatic irreversible pulpitis. LEfSe analysis indicated that the microbial species differed among these statuses (Fig. 3). At the genus level, the potential bacteria associated with reversible pulpitis with both cold/heat-evoked pain were Actinomyces and Unidentified_Veillonellaceae. Studies have shown that Actinomyces are associated with carious lesions (29) and that white spot formation occurs in the earliest stage of caries, a sign of a high abundance of Actinomyces (30), suggesting an important role for Actinomyces in caries initiation. However, few studies have confirmed the correlation between Actinomyces and reversible pulpitis. Compared with reversible pulpitis with only cold-evoked pain, Actinomyces abundance was different in reversible pulpitis with both cold/heat-evoked pain. A previous study has shown that Actinomyces are associated with cold sensitivity (16). This means that Actinomyces is also likely to be related to the heat sensitivity of reversible pulpitis. Unlike the normal pulp with deep dentinal caries and the 2 stages of reversible pulpitis, the dominant genus in symptomatic irreversible pulpitis was Lactobacillus. Next-generation sequencing analysis has also revealed that half of advanced carious lesions with symptomatic irreversible pulpitis are dominated by Lactobacillus (15). Species of Lactobacillus do not produce substantial extracellular polysaccharides (31), and their affinity for the tooth surface is poor, making them difficult to find on smooth tooth surfaces. However, in retentive areas, such as cavities, species of Lactobacillus are frequently found (32). Therefore, Lactobacillus is not the initial factor in dental caries but is involved in the progression of caries (33). During the progression of dental caries, Lactobacillus and Streptococcus mutans form a synergistic relationship, which contributes to the colonization of Lactobacillus (34), which are acidogenic and aciduric bacteria. In the acidic environment of carious lesions, bacterial diversity is diminished, favoring the growth of aciduric bacteria, such as Lactobacillus (35). As the growth of Streptococcus is inhibited, Lactobacillus becomes the dominant bacteria. A study of the relationship between the microbiota of the most advanced layers of dentinal caries and irreversible pulpitis revealed that Lactobacillus occurs at significantly higher levels in cases with continuous pain (17). Therefore, Lactobacillus may be one of the pathogenic microbes that cause symptomatic irreversible pulpitis. As pulpitis gradually progresses, the microbial species and quantity change. As shown in the random forest histograms (Fig. 4C–E), there were differences in the microbial distribution between the SIP group and the other groups. ROC curves were used to evaluate associations between characteristics of the microbiome in deep dentinal caries and specific pulpal diagnoses; the AUC, which ranges from 0.5 for a useless test to 1.0 for a perfect test, provides a useful summary of the overall accuracy. The AUCs were all high (AUC: 0.734 – 0.952), suggesting that the identity of the microbiome in deep caries is a major indicator of caries-induced pulpitis (Fig. 4A,B). A limitation of the study is the potential for contamination during sampling, despite attention to isolation and disinfection of the tooth surface before sampling. In addition, the various pulpitis diagnoses were determined by using only the American Association of Endodontists clinical diagnostic criteria and not confirmed histologically, an invasive diagnosis and treatment procedure, which is contrary to the minimally invasive pulp capping treatment. 308
Zheng et al.
In conclusion, the microorganisms associated with reversible pulpitis with both cold/heat-evoked pain were Actinomyces and Unidentified_Veillonellaceae; Actinomyces may be related to the heat sensitivity of reversible pulpitis. In addition, Lactobacillus is the dominant taxon in symptomatic irreversible pulpitis. Overall, knowledge about the microbiome associated with reversible pulpitis- and irreversible pulpitisassociated microbial composition can provide a basis for pulpitis diagnosis whereby the identity of the microbiome could potentially become an auxiliary method for the diagnosis of caries-induced pulpitis. However, further studies are needed to reveal the value and clinical application of this diagnostic approach.
Acknowledgments This work was supported by the Development of Science and Technology Plan Projects of Wu Qing District, Tianjin (grant number WQKJ201619). The authors deny any conflict of interest related to this study.
Supplementary Material Supplementary material associated with this article can be found in the online version at www.jendodon.com (https://doi. org/10.1016/j.joen.2018.11.017).
References 1. Marsh PD. Microbial ecology of dental plaque and its significance in health and disease. Adv Dent Res 1994;8:263–71. 2. Kianoush N, Adler CJ, Nguyen KA, et al. Bacterial profile of dentine caries and the impact of pH on bacterial population diversity. PloS One 2014;9:e92940. 3. Munson MA, Banerjee A, Watson TF, et al. Molecular analysis of the microflora associated with dental caries. J Clin Microbiol 2004;42:3023–9. 4. Chhour KL, Nadkarni MA, Byun R, et al. Molecular analysis of microbial diversity in advanced caries. J Clin Microbiol 2005;43:843–9. 5. Hoshino E, Ando N, Sato M, et al. Bacterial invasion of non-exposed dental pulp. Int Endod J 1992;25:2–5. 6. Brittan JL, Sprague SV, Macdonald EL, et al. In vivo model for microbial invasion of tooth root dentinal tubules. J Appl Oral Sci 2016;24:126–35. 7. Reeves R, Stanley HR. The relationship of bacterial penetration and pulpal pathosis in carious teeth. Oral Surg Oral Med Oral Pathol 1966;22:59–65. 8. Seltzer S, Farber PA. Microbiologic factors in endodontology. Oral Surg Oral Med Oral Pathol 1994;78:634–45. 9. Bergenholtz G. Inflammatory response of the dental pulp to bacterial irritation. J Endod 1981;7:100–4. 10. Martin FE, Nadkarni MA, Jacques NA, et al. Quantitative microbiological study of human carious dentine by culture and real-time PCR: association of anaerobes with histopathological changes in chronic pulpitis. J Clin Microbiol 2002;40:1698–704. 11. Levin LG, Law AS, Holland GR, et al. Identify and define all diagnostic terms for pulpal health and disease states. J Endod 2009;35:1645–57. 12. Gross EL, Beall CJ, Kutsch SR, et al. Beyond Streptococcus mutans: dental caries onset linked to multiple species by 16S rRNA community analysis. PloS One 2012;7:e47722. 13. Sakamoto M, Rocas IN, Siqueira JF, et al. Molecular analysis of bacteria in asymptomatic and symptomatic endodontic infections. Oral Microbiol Immunol 2006;21: 112–22. 14. Keskin C, Demiryurek EO, Onuk EE. Pyrosequencing analysis of cryogenically ground samples from primary and secondary/persistent endodontic infections. J Endod 2017;43:1309–16. 15. R^oc¸as IN, Alves FRF, Rachid CTCC, et al. Microbiome of deep dentinal caries lesions in teeth with symptomatic irreversible pulpitis. PLoS One 2016;11:e0154653. 16. Hahn CL, Falkler WA Jr, Minah GE. Correlation between thermal sensitivity and microorganisms isolated from deep carious dentin. J Endod 1993;19:26–30. 17. Rocas IN, Lima KC, Assuncao IV, et al. Advanced caries microbiota in teeth with irreversible pulpitis. J Endod 2015;41:1450–5. 18. Massey WL, Romberg DM, Hunter N, et al. The association of carious dentin microflora with tissue changes in human pulpitis. Oral Microbiol Immunol 1993;8:30–5. 19. Glickman GN. AAE Consensus Conference on Diagnostic Terminology: background and perspectives. J Endod 2009;35:1619–20. 20. Schwendicke F, Frencken JE, Bjorndal L, et al. managing carious lesions: consensus recommendations on carious tissue removal. Adv Dent Res 2016;28:58–67.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology 21. Tian Y, Li YH. Comparative analysis of bacteria associated with different mosses by 16S rRNA and 16S rDNA sequencing. J Basic Microbiol 2017;57:57–67. 22. Aas JA, Paster BJ, Stokes LN, et al. Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 2005;43:5721–32. 23. Auer PL, Doerge RW. Statistical design and analysis of RNA sequencing data. Genetics 2010;185:405–16. 24. Narhi M, Jyvasjarvi E, Virtanen A, et al. Role of intradental A- and C-type nerve fibres in dental pain mechanisms. Proc Finn Dent Soc 1992;88(Suppl 1):507–16. 25. Matthews B, Vongsavan N. Interactions between neural and hydrodynamic mechanisms in dentine and pulp. Arch Oral Biol 1994;39(Suppl):87S–95S. 26. Narhi M, Yamamoto H, Ngassapa D, et al. The neurophysiological basis and the role of inflammatory reactions in dentine hypersensitivity. Arch Oral Biol 1994; 39(Suppl):23S–30S. 27. Martin FE. Carious pulpitis: microbiological and histopathological considerations. Aust Endod J 2003;29:134–7. 28. Hahn CL, Falkler WA Jr, Minah GE. Microbiological studies of carious dentine from human teeth with irreversible pulpitis. Arch Oral Biol 1991;36:147–53.
JOE — Volume 45, Number 3, March 2019
29. Aas JA, Griffen AL, Dardis SR, et al. Bacteria of dental caries in primary and permanent teeth in children and young adults. J Clin Microbiol 2008;46:1407–17. 30. Becker MR, Paster BJ, Leys EJ, et al. Molecular analysis of bacterial species associated with childhood caries. J Clin Microbiol 2002;40:1001–9. 31. Minah GE, Loesche WJ. Sucrose metabolism by prominent members of the flora isolated from cariogenic and non-cariogenic dental plaques. Infect Immun 1977;17: 55–61. 32. Van Houte J, Gibbons RJ, Pulkkinen AJ. Ecology of human oral lactobacilli. Infect Immun 1972;6:723–9. 33. Loesche WJ. Role of Streptococcus mutans in human dental decay. Microbiol Rev 1986;50:353–80. 34. Willcox MD, Patrikakis M, Harty DW, et al. Coaggregation of oral lactobacilli with streptococci from the oral cavity. Oral Microbiol Immunol 1993;8: 319–21. 35. Marsh PD. Dental plaque as a biofilm: The significance of pH in health and caries. Compend Contin Educ Dent 2009;30:76–8. 80, 3–7; quiz 8, 90.
Characterization of the Caries Microbiome
309
Basic Research—Biology SUPPLEMENTARY TABLE 1. ANOSIM (Analysis of Similarities) Group
R-value*
NP - SIP CRP - SIP CHRP - SIP CRP - NP CHRP - NP CHRP - CRP
0.1008 0.05864 0.1924 0.02131 0.02669 0.08332
*R-value is between ( 1, 1). R-value is greater than 0, indicating that the differences among groups is greater than the differences within groups. R-value is less than 0, indicating that the differences within groups is greater than the differences among groups.
309.e1
Zheng et al.
JOE — Volume 45, Number 3, March 2019
Microbiome of Deep Dentinal Caries from Reversible Pulpitis to Irreversible Pulpitis Jinxin Zheng, BDS,*a Zhou Wu, MDS,†a Kaijun Niu, PhD,‡ Yanan Xie, BDS,* Xiaoli Hu, BDS,* Jieni Fu, BDS,* Dongtao Tian, BDS,§ Kaiyu Fu, BDS,¶ Bo Zhao, BDS,* Weiyang Kong, BDS,* Cuicui Sun, BDS,* and Ligeng Wu, MDS* Abstract Introduction: This study examined the identity of the microbiome of deep dentinal caries and its correlation with the inflammation status of caries-induced pulpitis. Methods: Seventy-five cases were diagnosed based on the American Association of Endodontics’s diagnostic criteria and divided into 4 groups: normal pulp with deep caries (NP; n = 13), reversible pulpitis with only cold-evoked pain (CRP; n = 17), reversible pulpitis with both cold/heat-evoked pain (CHRP; n = 24), and symptomatic irreversible pulpitis (SIP; n = 21). Samples were sequenced by 16S rDNA. Alpha and beta diversity were determined. Linear discriminant analysis effect size (LEfSe) analysis was used to detect intergroup differences, and receiver operating characteristic (ROC) curves were generated to assess the role of the caries microbiome in caries-induced pulpitis. Results: The 16S rDNA sequencing yielded 9100 operational taxonomic units. Lactobacillus had the highest relative abundance at the genus level among the 4 groups. There were significant differences in the distribution of the microbiome among the groups. In an alpha diversity analysis, species richness differed between the CRP group and the other groups. In a beta diversity analysis, the distribution of microorganisms in the SIP group was significantly different from those in the other 3 groups. LEfSe analysis indicated substantial differences in the microbiome among the groups, and the areas under the ROC curves (AUC) were all high (AUC: 0.734– 0.952). Conclusions: Characterization of the caries microbiome has the potential to become an auxiliary method for the diagnosis of pulpitis. This finding may prompt new research on diagnostic strategies for caries-induced pulpitis. (J Endod 2019;45:302–309)
Key Words microbiome, dentinal caries, irreversible pulpitis, Lactobacillus, reversible pulpitis
C
aries is a common Significance dental disease mediMicrobial determination indicated that species and ated by a variety of microquantity of the microorganism were consistent with organisms (1–4). Without clinical diagnosis and were associated with the appropriate treatment, severity of pulpitis. Improved knowledge of the mimicroorganisms in deep crobiome may help to develop diagnostic tools. caries invade the dental pulp through the dentinal tubules or directly cause pulp exposure (5–7), leading to pulpitis (8–10), which is clinically categorized as reversible and irreversible pulpitis (11). Some studies have indicated that the bacterial species associated with deep caries differ substantially from those of inflamed pulp. The main pathogens of caries are oral Streptococcus, Lactobacillus, Actinomyces, Veillonella, Neisseria, and others (12), whereas the dominant genera associated with endodontic diseases are Fusobacterium, Porphyromonas, Peptostreptococcus, Prevotella, Parvimonas, Dialister, and others (13, 14). Other studies have shown that most bacterial species in carious dentin are also found in infected root canals (15) and that there are positive associations between specific bacteria in carious dentin and irreversible pulpitis (10), suggesting that bacteria present in the most advanced layers of dentinal caries are candidate pathogens for inducing pulpitis and initiating pulp inflammation (15). Microorganisms in deep caries are associated with dental pulp symptoms (10, 16–18). However, less is known about the characteristics of the microbiome at the forefront of carious lesions in teeth with different pulpal diagnoses. The aim of this study was to characterize the microbiome of the dentinal caries microenvironment and the microbial diversity in deep dentinal caries of teeth with pulpitis by 16S rDNA sequencing as well as to evaluate the correlation between the microbiome and the inflammation status of caries-induced pulpitis. We also preliminarily investigated whether the differences in the microbiome in deep dentinal caries can contribute to the diagnosis of caries-induced pulpitis.
From the *Department of Endodontics, School of Stomatology, and ‡Nutritional Epidemiology Institute and School of Public Health, Tianjin Medical University, Tianjin, China; †Department of Stomatology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China; §Department of Stomatology, Wuqing People Hospital, Tianjin, China; and ¶Department of Orthodontics, Tianjin Stomatology Hospital, Tianjin, China. a Co–first authors. Address requests for reprints to Ligeng Wu, MDS, Department of Endodontics, School of Stomatology, Tianjin Medical University, #12 Qi Xiang Tai Road, He Ping District, Tianjin, 300070, China. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2018 American Association of Endodontists. https://doi.org/10.1016/j.joen.2018.11.017
302
Zheng et al.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology Materials and Methods Study Population This study was approved by the ethics committee of the Stomatology Hospital of Tianjin Medical University (project number TMUDHhMEC2015010). In total, 75 patients with deep caries were enrolled at Stomatology Hospital of Tianjin Medical University between February 2015 and February 2018, and informed consent was obtained from all subjects. Patients ranged in age from 12 to 60 years. Inclusion Criteria and Exclusion Criteria Inclusion criteria: 1. All cases had deep caries lesions. Periapical radiographs showed that the root apices of all teeth had completed development and that the depth of deep caries of anterior or posterior teeth exceeded two-thirds of the depth of dentin. 2. Caries affected the proximal or occlusal surfaces of the teeth. 3. Teeth without previous restorations. Exclusion criteria: 1. 2. 3. 4. 5.
Pregnancy. Antibiotics, analgesics, or anesthetics usage before treatment. Moderate or severe periodontitis or furcation involvement. Cracked tooth, apical periodontitis. Teeth with subgingival caries that could not be isolated with rubber dam. 6. Teeth with no response to pulp tests (described later). 7. Periapical radiographs revealed a periapical shadow or significant widening of the periodontal ligament space. 8. Infectious diseases, serious systemic disease, mental disorder, or failure to cooperate with treatment.
Cold Testing A large ＃2 cotton pellet with a refrigerant spray and a temperature of 26.2 C (1, 1, 1, 2-tetrafluoroethane) (Endo-Frost; Coltene Whaledent, Cuyahoga Falls, OH) was sprayed and applied to the middle third of the buccal surface of the crown of the tooth for 3 seconds and the patients provided their response. Heat Testing A light layer of lubricant was placed on the surface of the tooth and a heated gutta-percha rod was placed on the middle third of the buccal surface of the crown of the tooth for 3 seconds, and the patients provided their response.
Electric Pulp Testing After clinical isolation with cotton rolls and drying of the tooth with air, the probe tip of an electric pulp tester (SybronEndo, Orange, CA) was coated with toothpaste and applied to the middle third of the buccal surface of the crown of the tooth. All teeth were tested 3 times, and the readings from the pulp tester were recorded. A tooth with a response at a level lower than 60 was considered vital. The adjacent tooth or contralateral "normal" tooth was also tested to establish a baseline response. Pulp Diagnosis The diagnosis of pulpitis was based on clinical and radiographic findings and according to the criteria of the American Association of Endodontists (19). Normal pulp: no spontaneous pain, the responses to cold and heat tests were the same as those of control teeth, and no lingering pain. Reversible pulpitis: no spontaneous pain, sensitive responses to cold and heat tests compared with control teeth, and no more than 30 seconds of lingering pain after the removal of the stimulus. Based on the result of the heat test, reversible pulpitis was subdivided into 2 stages: reversible pulpitis with only cold-evoked pain and reversible pulpitis with both cold/heat-evoked pain. Symptomatic irreversible pulpitis: spontaneous pain, sensitive responses to cold or heat tests compared with control teeth, and more than 30 seconds of lingering pain after the removal of the stimulus. Clinical Grouping In total, 75 teeth (2 anterior and 73 posterior teeth) with deep caries were included. The cavity wall was located at either the proximal (n = 62) or occlusal (n = 13) surface. Based on patient history and clinical pulp tests, the cases were divided into 4 groups according to the severity of pulpal inflammation: (1) normal pulp with deep caries (NP group; n = 13), (2) reversible pulpitis with only cold-evoked pain (CRP group; n = 17), (3) reversible pulpitis with both cold/heat-evoked pain (CHRP group; n = 24), and (4) symptomatic irreversible pulpitis (SIP group; n = 21). The groups and diagnoses are shown in Table 1. Sample Collection One clinician examined the cases and collected the samples under aseptic conditions. The selective removal of carious tissue (20) was performed to collect samples. The mouth was rinsed with hydrogen peroxide before caries collection, and after local anesthesia, the carious teeth and the adjacent teeth were isolated with a rubber dam (Kant, St. Gallen, Switzerland). The surfaces of the tooth were disinfected with a 2% chlorhexidine gluconate solution (Longly Biomedical Wuhan Co
TABLE 1. Group Assignments and Diagnosis of 75 Cases According to the American Association of Endodontists Diagnostic Criteria Groups
NP
CRP
Patient history Pulp thermal tests
SP ( ) Cold* ( ) Heat† ( )
SP ( ) Cold (+) lp < 30 s Heat ( )
Cases Numbers Diagnosis
13 NP1–13 Normal pulp
CHRP
SP ( ) Cold (+) lp < 30 s Heat (+) lp < 30 s 17 24 CRP1–17 CHRP1–24 Reversible pulpitis
SIP SP (+) Cold (+) lp > 30 s Heat ( )
Cold ( ) Heat (+) lp > 30 s
Cold (+) lp > 30 s Heat (+) lp > 30 s
21 SIP1–21 Symptomatic irreversible pulpitis
CHRP, reversible pulpitis with both cold/heat-evoked pain; CRP, reversible pulpitis with only cold-evoked pain; lp, lingering pain; NP, normal pulp with deep caries; SIP, symptomatic irreversible pulpitis; SP, spontaneous pain. *Cold test; +, sensitive; , same as control teeth (control tooth: an adjacent tooth or contralateral normal tooth of the same name). † Heat test; +, sensitive; , same as control teeth (control tooth: an adjacent tooth or contralateral normal tooth of the same name).
JOE — Volume 45, Number 3, March 2019
Characterization of the Caries Microbiome
303
Basic Research—Biology Ltd, Wuhan, China). Undermined enamel, debris, and superficial carious tissue were removed using sterile high-speed burs (Ningbo Xinyuan Dental Equipment Co, Ltd, Zhejiang, China) under water cooling. The cavity was irrigated with 5 mL 0.9% sterile saline (China Otsuka Pharmaceutical Co, Ltd, Tianjin, China). The lift-draw method was used to remove the superficial layer of the caries using a low-speed ball bur (Ningbo Xinyuan Dental Equipment Co, Ltd, Zhejiang, China). Using a sterile spoon excavator, samples per tooth of the deepest layer of the caries were transferred to a sterile Eppendorf tube (EP tube) containing 0.5 mL sterile saline. The EP tubes with samples were numbered according to group and registration order and stored immediately at 80 C in a refrigerator. The EP tube was removed and stored in dry ice before 16S rDNA sequencing.
Sample Quality Control
1. 2. 3. 4. 5.
To ensure quality in the sampling, the following control measures were used: The surface of the tooth was disinfected using chlorhexidine before sampling. Teeth that could not be isolated with rubber dam were excluded. Only one experienced clinician examined the cases and collected the samples to ensure consistency in all samples. Eppendorf tubes were sterile. The whole operation was performed under sterile conditions.-
16S rDNA Sequencing The total genomic DNA from samples was extracted using the cetyltrimethylammonium ammonium bromide/sodium dodecyl sulfonate method. DNA concentration and purity were monitored on 1% agarose gels. To meet sequencing requirements, samples with less than 150 ng DNA were eliminated. DNA was diluted to 1 ng/mL using sterile water. The 16S rDNA gene regions (16S V3–V4) were amplified using specific primers with barcodes. All polymerase chain reactions (PCRs) were performed using Phusion High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA) under the following conditions: 95 C for 5 minutes, followed by 34 cycles of 94 C for 1 minute, 57 C for 45 seconds and 72 C for 1 minute, and then a final elongation step at 72 C for 10 minutes and 16 C for 5 minutes. An equal volume of 1 loading buffer (containing SYB green) was added to the PCR products, followed by separation by electrophoresis on a 2% agarose gel for detection. Samples with a bright main strip at 400 to 450 base pairs (bp) were chosen for further experiments. PCR products were mixed in equal ratios. Then, mixed PCR products were purified using the Qiagen Gel Extraction Kit (Qiagen, Hilden, Germany). DNA libraries were generated using the TruSeq DNA PCR–Free Sample Preparation Kit (Illumina, San Diego, CA). The library was sequenced using the HiSeq2500 PE250 sequencer (Illumina) and 250-bp paired-end reads were generated. Then, the reads of each sample were spliced to obtain the original tag data. After filtering, the high-quality tag data were obtained and compared with sequences in the database (Gold database, http:// drive5.com/uchime/uchime_download.html). Operational taxonomic units (OTUs) were obtained using Uparse v7.0.1001 (http://drive5. com/uparse/) for species annotation and microbial diversity analysis. Statistical Analysis Alpha diversity was estimated to analyze the richness and diversity of microbial species in the samples based on observed species. Results were calculated using Qiime (version 1.9.1), and rarefaction curves and 304
Zheng et al.
species accumulation boxplot were obtained using R software (Version 2.15.3). Beta diversity was analyzed to evaluate differences among samples in species complexity. Weighted UniFrac distances were calculated using Qiime, and trees were generated using the unweighted pair-group method with arithmetic mean (UPGMA). Anosim analysis was used to compare differences among groups with differences within groups. Linear discriminant analysis (LDA) effect size (LEfSe) analysis was performed, setting the LDA score to 4 by default. The accuracy of microbiome characterization in predicting the status of caries-induced pulpitis was analyzed using receiver operating characteristic (ROC) curves.
Results Taxonomic Identification and Relative Abundance A total of 9100 OTUs were detected in 75 samples, belonging to 3 kingdoms (bacteria, archaea, and unknown), 56 phyla, 145 classes, 222 orders, 429 families, and 1132 genera. In total, 8934 OTUs were assigned to bacteria, including 51 phyla, and 162 OTUs were assigned to archaea, including 5 phyla. Taxonomic assignments revealed 10 phyla in the NP, CRP, CHRP, and SIP groups, 4 of which had relatively high abundances (Fig. 1A): Firmicutes (45.9%, 36.3%, 37.1%, and 63.6%, respectively), Actinobacteria (32.0%, 34.5%, 36.2%, and 16.0%), Proteobacteria (11.1%, 10.7%, 10.2%, and 9.3%), and Bacteroidetes (7.5%, 12.6%, 12.2%, and 8.6%). The relative abundance of Lactobacillus was higher in the SIP group than in the other groups; Olsenella and Actinomyces had similar abundances in the CRP and CHRP groups. The relative abundances of Streptococcus, Pseudoramibacter, and Acinetobacter were higher in the NP group than in the other 3 groups (Fig. 1B, C). Microbial Community Rarefaction curves are shown in Figure 2A and Figure 2B. The curve for the CRP group was relatively steep, and the tail of the CRP7 curve was the steepest. The remaining 74 curves were relatively flat. In general, the CRP group differed from the other groups in terms of species richness. Species accumulation boxplot of the 75 samples is shown in Figure 2C. As the sample size gradually increased, microbial species did not increase significantly, indicating that the sample size was sufficient to capture species diversity. The UPGMA tree for the 4 groups at the phylum level revealed that the distributions of microbiome in the reversible pulpitis groups (CRP and CHRP groups) were similar, whereas the distribution of the microbiome in the SIP group was significantly different from those in the other 3 groups. The relative abundance of Firmicutes was the highest in the SIP group. Within the remaining 3 groups, the relative abundances of Firmicutes and Actinobacteria were higher than those of the other phyla (Fig. 2D). Analysis of Intergroup Differences in Species According to ANOSIM (analysis of similarities), differences among groups were greater than differences within groups (R > 0; data not shown). The differential species among the groups are shown in Figure 3 and Supplemental Table S1. The genera Lactobacillus, Actinomyces, and Unidentified_Veillonellaceae differed among the groups. The relative abundance of Lactobacillus was the highest in the SIP group, and Actinomyces and Unidentified_Veillonellaceae were relatively highly abundant in the CHRP group. JOE — Volume 45, Number 3, March 2019
Basic Research—Biology
Figure 1. Differences in the microbial composition and distribution among the 4 groups. (A) Average of the top 10 most abundant phyla identified in each group. (B) Members of the top 10 genera. (C) Heatmap of the top 10 genera. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
Associations Between the Microbiome and Reversible/ Irreversible Pulpitis ROC curves of the differential species among the groups were used to evaluate associations between the microbiome in deep dentinal caries and specific pulpal diagnoses. The area under the ROC curve (AUC) can be used to compare the accuracy of the identity of the microbiome and clinical examination in the diagnosis of caries-induced pulpitis. The AUCs for NP versus CRP, CRP versus CHRP, CRP versus SIP, and CHRP versus SIP were 0.864 (95% confidence interval [CI], 0.71– 1), 0.851 (95% CI, 0.71–0.99), 0.952 (95% CI, 0.88–1), and 0.800 (95% CI, 0.63–0.97), respectively (Fig. 4A). Curves for reversible pulpitis and irreversible pulpitis compared with normal pulp are shown in Figure 4B. The AUCs for CRP, CHRP, and SIP from NP were 0.864 (95% CI, 0.71 –1), 0.734 (95% CI, 0.54–0.93), and 0.833 (95% CI, 0.67– 0.99), respectively. Random forest histograms for the intergroup differences in the microbiome of deep caries were generated (Fig. 4C–E). The analysis suggested that Lactobacillus in the SIP group was typically different from those in the other groups.
Discussion The 16S rDNA is a useful indicator for bacterial phylogenic analysis and taxonomic identification. Compared with traditional microbial JOE — Volume 45, Number 3, March 2019
identification methods, 16S rDNA sequencing identifies taxa based on differences in a variable region sequence (21). The approach is relatively reliable and is not restricted by the culture environment. In this study, 16S rDNA sequencing was used to detect more than 9000 OTUs, far more than the 700 known bacteria in oral caries (22). The reasons for the large number of OTUs found in this study compared with the previous studies (15, 17) are as follows: first, the caries samples in this study were recovered from asymptomatic teeth in addition to teeth with reversible pulpitis and irreversible pulpitis; in previous studies, the caries samples were recovered only from teeth with irreversible pulpitis; second, the sample size was larger than those of the previous studies; and third, the sample sequencing yielded many low-abundance species, and there was no culling and filtering of low-abundance species in the standard analysis. Genera identified in the caries of anterior teeth were also detected in posterior teeth caries. Similarly, genera in occlusal caries were also detected in proximal caries. However, considering that the amount of anterior tooth and occlusal caries samples was small in addition to the fact that there are presently no relevant reports that the different types or surfaces of teeth with caries can influence the composition of the microbiome, further studies are needed to explore the correlation between the microbiome of caries located in different types, or on different surfaces, of teeth with pulpitis.
Characterization of the Caries Microbiome
305
Basic Research—Biology
Figure 2. Diversity analysis. (A) Rarefaction curves for alpha diversity in the 4 groups. (B) Rarefaction curves for alpha diversity in the 75 samples. (C) Species accumulation boxplot of the 75 samples. (D) Beta diversity shown by UPGMA (unweighted pair-group method with arithmetic mean) of the top 10 most abundant phyla. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); OTU, operational taxonomic unit; SIP, symptomatic irreversible pulpitis (n = 21).
Rarefaction curves can directly reflect the rationality of sequencing data and can indirectly reflect species richness in a sample. In a comparison of the rarefaction curves for the 4 groups, the CRP group curve was relatively steep (Fig. 2A), indicating that there may be undetected
microorganisms in this group. The curves for the 75 samples showed substantial variation (Fig. 2B), suggesting that the microbial species richness varied greatly. The curve for CRP7 increased rapidly and did not exhibit a flat tail, suggesting that the microbial species in the sample
Figure 3. Linear discriminant analysis (LDA) effect size analysis of the 4 groups. LDA score >4 indicates a biomarker with statistical differences between groups. CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
306
Zheng et al.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology
Figure 4. A classification system for identifying caries-induced pulpitis. (A, B) ROC for the random forest classifier based on the intergroup differences in species. (C, D, E) Intergroup differences in species based on the random forest model. Lengths of bar in the histogram represent the mean decrease in accuracy, which indicates the importance of the intergroup differences in species for classification. The color denotes the enrichment of intergroup species differences in NP (red), CRP (purple), CHRP (blue), and SIP (green) according to the odds ratio score. AUC, area under the receiver operating characteristic (ROC) curves; CHRP, reversible pulpitis with both cold/heat-evoked pain (n = 24); CRP, reversible pulpitis with only cold-evoked pain (n = 17); NP, normal pulp with deep dentinal caries (n = 13); SIP, symptomatic irreversible pulpitis (n = 21).
were highly abundant and that some microorganisms were undetected. The curves for CHRP7, CHRP2, CHRP1, and CRP4 exhibited rapid increases and flat tails, suggesting that the abundance of microbial species was high and that detection was relatively complete. The rarefaction curves for the other 70 samples increased slowly and had flat tails, indicating low abundances and complete detection. Overall, among the 75 samples, the detection of the microbial species of 74 samples was relatively complete. In the species accumulation boxplot of the 75 samples (Fig. 2C), as the sample size increased, the trend of the boxplot gradually became flat, suggesting that the microbial species did not increase significantly; that is, even if the sample size continues to increase, only 1 or 2 additional types of microorganism may be detected; and all the samples of the 4 groups met the requirement of at least 3 biological duplications per group (23), basically achieving the 16S rDNA analysis requirement, with a certain degree of repeatability. This suggests that the sample size was sufficient and that reliability was high. According to the thermal examination, reversible pulpitis was subdivided into 2 stages in our study: reversible pulpitis with only coldevoked pain and reversible pulpitis with both cold/heat-evoked pain. Previous studies have shown that it is the A fibers that have a relatively low threshold of excitability to external stimuli rather than the C fibers JOE — Volume 45, Number 3, March 2019
that are activated by hydrodynamic stimuli (eg, heat, cold) (24, 25). However, the C fibers may generate a response if heat is applied to increase the temperature of the dentin-pulp border, and pulpitis characterized by pain is more likely to be associated with C fiber activity, which is indicative of pulpal tissue injury (26). Therefore, reversible pulpitis with both cold/heat-evoked pain may be more advanced than reversible pulpitis with only cold-evoked pain. The UPGMA trees for the 4 groups showed that the distributions of microbiome were similar at the 2 stages of reversible pulpitis (Fig. 2D), indicating consistency with respect to clinical diagnosis, grouping, and microbial detection. In the normal pulp with deep caries and reversible pulpitis, both Firmicutes and Actinobacteria were highly abundant. In symptomatic irreversible pulpitis, the abundance of Firmicutes was the highest (Fig. 1A). These results might indicate that Actinobacteria decreases and Firmicutes increases as caries progresses. Similar results were obtained in a longitudinal study of young children (12). Firmicutes and Actinobacteria accounted for most of the sequences detected in the 4 groups. The former was mostly represented by Lactobacillus, Pseudoramibacter, and Streptococcus, whereas the most common representative genera of the latter were Propionibacterium, Olsenella, and Actinomyces, suggesting their strong association
Characterization of the Caries Microbiome
307
Basic Research—Biology with deep dentinal caries as well as reversible and irreversible pulpitis. This result confirms the results of previous studies showing that microorganisms present in the front line of dentinal carious lesions are likely to be involved in the development of pulp inflammation (10, 27, 28). According to the clinical diagnosis, the status of the pulp was divided into 4 phases: normal pulp with deep caries, reversible pulpitis with only cold-evoked pain, reversible pulpitis with both cold/heatevoked pain, and symptomatic irreversible pulpitis. LEfSe analysis indicated that the microbial species differed among these statuses (Fig. 3). At the genus level, the potential bacteria associated with reversible pulpitis with both cold/heat-evoked pain were Actinomyces and Unidentified_Veillonellaceae. Studies have shown that Actinomyces are associated with carious lesions (29) and that white spot formation occurs in the earliest stage of caries, a sign of a high abundance of Actinomyces (30), suggesting an important role for Actinomyces in caries initiation. However, few studies have confirmed the correlation between Actinomyces and reversible pulpitis. Compared with reversible pulpitis with only cold-evoked pain, Actinomyces abundance was different in reversible pulpitis with both cold/heat-evoked pain. A previous study has shown that Actinomyces are associated with cold sensitivity (16). This means that Actinomyces is also likely to be related to the heat sensitivity of reversible pulpitis. Unlike the normal pulp with deep dentinal caries and the 2 stages of reversible pulpitis, the dominant genus in symptomatic irreversible pulpitis was Lactobacillus. Next-generation sequencing analysis has also revealed that half of advanced carious lesions with symptomatic irreversible pulpitis are dominated by Lactobacillus (15). Species of Lactobacillus do not produce substantial extracellular polysaccharides (31), and their affinity for the tooth surface is poor, making them difficult to find on smooth tooth surfaces. However, in retentive areas, such as cavities, species of Lactobacillus are frequently found (32). Therefore, Lactobacillus is not the initial factor in dental caries but is involved in the progression of caries (33). During the progression of dental caries, Lactobacillus and Streptococcus mutans form a synergistic relationship, which contributes to the colonization of Lactobacillus (34), which are acidogenic and aciduric bacteria. In the acidic environment of carious lesions, bacterial diversity is diminished, favoring the growth of aciduric bacteria, such as Lactobacillus (35). As the growth of Streptococcus is inhibited, Lactobacillus becomes the dominant bacteria. A study of the relationship between the microbiota of the most advanced layers of dentinal caries and irreversible pulpitis revealed that Lactobacillus occurs at significantly higher levels in cases with continuous pain (17). Therefore, Lactobacillus may be one of the pathogenic microbes that cause symptomatic irreversible pulpitis. As pulpitis gradually progresses, the microbial species and quantity change. As shown in the random forest histograms (Fig. 4C–E), there were differences in the microbial distribution between the SIP group and the other groups. ROC curves were used to evaluate associations between characteristics of the microbiome in deep dentinal caries and specific pulpal diagnoses; the AUC, which ranges from 0.5 for a useless test to 1.0 for a perfect test, provides a useful summary of the overall accuracy. The AUCs were all high (AUC: 0.734 – 0.952), suggesting that the identity of the microbiome in deep caries is a major indicator of caries-induced pulpitis (Fig. 4A,B). A limitation of the study is the potential for contamination during sampling, despite attention to isolation and disinfection of the tooth surface before sampling. In addition, the various pulpitis diagnoses were determined by using only the American Association of Endodontists clinical diagnostic criteria and not confirmed histologically, an invasive diagnosis and treatment procedure, which is contrary to the minimally invasive pulp capping treatment. 308
Zheng et al.
In conclusion, the microorganisms associated with reversible pulpitis with both cold/heat-evoked pain were Actinomyces and Unidentified_Veillonellaceae; Actinomyces may be related to the heat sensitivity of reversible pulpitis. In addition, Lactobacillus is the dominant taxon in symptomatic irreversible pulpitis. Overall, knowledge about the microbiome associated with reversible pulpitis- and irreversible pulpitisassociated microbial composition can provide a basis for pulpitis diagnosis whereby the identity of the microbiome could potentially become an auxiliary method for the diagnosis of caries-induced pulpitis. However, further studies are needed to reveal the value and clinical application of this diagnostic approach.
Acknowledgments This work was supported by the Development of Science and Technology Plan Projects of Wu Qing District, Tianjin (grant number WQKJ201619). The authors deny any conflict of interest related to this study.
Supplementary Material Supplementary material associated with this article can be found in the online version at www.jendodon.com (https://doi. org/10.1016/j.joen.2018.11.017).
References 1. Marsh PD. Microbial ecology of dental plaque and its significance in health and disease. Adv Dent Res 1994;8:263–71. 2. Kianoush N, Adler CJ, Nguyen KA, et al. Bacterial profile of dentine caries and the impact of pH on bacterial population diversity. PloS One 2014;9:e92940. 3. Munson MA, Banerjee A, Watson TF, et al. Molecular analysis of the microflora associated with dental caries. J Clin Microbiol 2004;42:3023–9. 4. Chhour KL, Nadkarni MA, Byun R, et al. Molecular analysis of microbial diversity in advanced caries. J Clin Microbiol 2005;43:843–9. 5. Hoshino E, Ando N, Sato M, et al. Bacterial invasion of non-exposed dental pulp. Int Endod J 1992;25:2–5. 6. Brittan JL, Sprague SV, Macdonald EL, et al. In vivo model for microbial invasion of tooth root dentinal tubules. J Appl Oral Sci 2016;24:126–35. 7. Reeves R, Stanley HR. The relationship of bacterial penetration and pulpal pathosis in carious teeth. Oral Surg Oral Med Oral Pathol 1966;22:59–65. 8. Seltzer S, Farber PA. Microbiologic factors in endodontology. Oral Surg Oral Med Oral Pathol 1994;78:634–45. 9. Bergenholtz G. Inflammatory response of the dental pulp to bacterial irritation. J Endod 1981;7:100–4. 10. Martin FE, Nadkarni MA, Jacques NA, et al. Quantitative microbiological study of human carious dentine by culture and real-time PCR: association of anaerobes with histopathological changes in chronic pulpitis. J Clin Microbiol 2002;40:1698–704. 11. Levin LG, Law AS, Holland GR, et al. Identify and define all diagnostic terms for pulpal health and disease states. J Endod 2009;35:1645–57. 12. Gross EL, Beall CJ, Kutsch SR, et al. Beyond Streptococcus mutans: dental caries onset linked to multiple species by 16S rRNA community analysis. PloS One 2012;7:e47722. 13. Sakamoto M, Rocas IN, Siqueira JF, et al. Molecular analysis of bacteria in asymptomatic and symptomatic endodontic infections. Oral Microbiol Immunol 2006;21: 112–22. 14. Keskin C, Demiryurek EO, Onuk EE. Pyrosequencing analysis of cryogenically ground samples from primary and secondary/persistent endodontic infections. J Endod 2017;43:1309–16. 15. R^oc¸as IN, Alves FRF, Rachid CTCC, et al. Microbiome of deep dentinal caries lesions in teeth with symptomatic irreversible pulpitis. PLoS One 2016;11:e0154653. 16. Hahn CL, Falkler WA Jr, Minah GE. Correlation between thermal sensitivity and microorganisms isolated from deep carious dentin. J Endod 1993;19:26–30. 17. Rocas IN, Lima KC, Assuncao IV, et al. Advanced caries microbiota in teeth with irreversible pulpitis. J Endod 2015;41:1450–5. 18. Massey WL, Romberg DM, Hunter N, et al. The association of carious dentin microflora with tissue changes in human pulpitis. Oral Microbiol Immunol 1993;8:30–5. 19. Glickman GN. AAE Consensus Conference on Diagnostic Terminology: background and perspectives. J Endod 2009;35:1619–20. 20. Schwendicke F, Frencken JE, Bjorndal L, et al. managing carious lesions: consensus recommendations on carious tissue removal. Adv Dent Res 2016;28:58–67.
JOE — Volume 45, Number 3, March 2019
Basic Research—Biology 21. Tian Y, Li YH. Comparative analysis of bacteria associated with different mosses by 16S rRNA and 16S rDNA sequencing. J Basic Microbiol 2017;57:57–67. 22. Aas JA, Paster BJ, Stokes LN, et al. Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 2005;43:5721–32. 23. Auer PL, Doerge RW. Statistical design and analysis of RNA sequencing data. Genetics 2010;185:405–16. 24. Narhi M, Jyvasjarvi E, Virtanen A, et al. Role of intradental A- and C-type nerve fibres in dental pain mechanisms. Proc Finn Dent Soc 1992;88(Suppl 1):507–16. 25. Matthews B, Vongsavan N. Interactions between neural and hydrodynamic mechanisms in dentine and pulp. Arch Oral Biol 1994;39(Suppl):87S–95S. 26. Narhi M, Yamamoto H, Ngassapa D, et al. The neurophysiological basis and the role of inflammatory reactions in dentine hypersensitivity. Arch Oral Biol 1994; 39(Suppl):23S–30S. 27. Martin FE. Carious pulpitis: microbiological and histopathological considerations. Aust Endod J 2003;29:134–7. 28. Hahn CL, Falkler WA Jr, Minah GE. Microbiological studies of carious dentine from human teeth with irreversible pulpitis. Arch Oral Biol 1991;36:147–53.
JOE — Volume 45, Number 3, March 2019
29. Aas JA, Griffen AL, Dardis SR, et al. Bacteria of dental caries in primary and permanent teeth in children and young adults. J Clin Microbiol 2008;46:1407–17. 30. Becker MR, Paster BJ, Leys EJ, et al. Molecular analysis of bacterial species associated with childhood caries. J Clin Microbiol 2002;40:1001–9. 31. Minah GE, Loesche WJ. Sucrose metabolism by prominent members of the flora isolated from cariogenic and non-cariogenic dental plaques. Infect Immun 1977;17: 55–61. 32. Van Houte J, Gibbons RJ, Pulkkinen AJ. Ecology of human oral lactobacilli. Infect Immun 1972;6:723–9. 33. Loesche WJ. Role of Streptococcus mutans in human dental decay. Microbiol Rev 1986;50:353–80. 34. Willcox MD, Patrikakis M, Harty DW, et al. Coaggregation of oral lactobacilli with streptococci from the oral cavity. Oral Microbiol Immunol 1993;8: 319–21. 35. Marsh PD. Dental plaque as a biofilm: The significance of pH in health and caries. Compend Contin Educ Dent 2009;30:76–8. 80, 3–7; quiz 8, 90.
Characterization of the Caries Microbiome
309
Basic Research—Biology SUPPLEMENTARY TABLE 1. ANOSIM (Analysis of Similarities) Group
R-value*
NP - SIP CRP - SIP CHRP - SIP CRP - NP CHRP - NP CHRP - CRP
0.1008 0.05864 0.1924 0.02131 0.02669 0.08332
*R-value is between ( 1, 1). R-value is greater than 0, indicating that the differences among groups is greater than the differences within groups. R-value is less than 0, indicating that the differences within groups is greater than the differences among groups.
309.e1
Zheng et al.
JOE — Volume 45, Number 3, March 2019