Ethnicity-specific microbiome in early childhood caries: from the functional perspectives of oral biofilm

doi:10.21203/rs.3.rs-5153245/v1

Download PDF

Article

Ethnicity-specific microbiome in early childhood caries: from the functional perspectives of oral biofilm

https://doi.org/10.21203/rs.3.rs-5153245/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

National surveillance data has long shown a significant disparity in tooth decay among young children (early childhood caries, ECC). While factors including household poverty level, culture, health insurance, and infrastructure have been studied, the biomedical perspective is less explored. Using RNASeq technology, our findings show that, besides Streptococcus mutans, which is most commonly associated with caries, several additional dental plaque bacteria are significantly overexpressed in caries lesions. Notably, the bacterial species and functional profiles differ markedly between African American and Latin American Hispanic children. In African American children, gene expression profiles linked to Pseudopropionibacterium propionicum and Cardiobacterium hominis; in contrast, in Latin American Hispanic children, gene expression profiles were dominated by Propionibacterium acidifaciens, Selenomonas sp., Rothia dentocariosa, Atopobium parvulum, and Streptococcus sanguinis. This study underscores the diverse metabolic pathways in plaque bacteria contributing to ECC in minority populations, identifying significant bacterial species beyond common cariogenic bacteria.

Main Text:

Biological sciences/Microbiology/Biofilms

Health sciences/Health care/Dentistry/Dental conditions/Plaque

Early childhood caries (ECC) is a severe and aggressive form of dental caries in children younger than six years of age. In the United States, it is the most common chronic disease among children and is more prevalent in socioeconomically disadvantaged populations (1). ECC can have many negative impacts on a child’s quality of life, including increased risk of additional caries lesions in both primary and permanent dentitions (2, 3), reduced appetite due to pain with eating, delayed physical growth and development, lack of sleep, irritability, low self-esteem, as well as poor attendance and performance in school (4). Children with ECC visit emergency rooms more often and have higher hospitalization rates (5, 6, 7), which increases the cost of medical care. If left untreated, fatal orofacial infections or brain abscesses can occur (8, 9). Moreover, young children suffering from severe ECC often require surgical intervention for full mouth rehabilitation, which can cost up to $25,000 per case (11, 12). In 2009, the total dental expenses for US children aged 5–17 years were approximately $20 billion and accounted for 17.7% of all healthcare expenses, with approximately 40% of dental costs paid out of pocket, compared to 17% for medical care (13).

Three key factors have been historically implicated in the development of caries: the presence of acidogenic biofilm bacteria, including Streptococcus mutans, dietary sugar, and family history. The acidogenic bacteria adhere to tooth surfaces, metabolize dietary carbohydrates, and produce excessive acids in the plaque environment. Over time, excessive acids promote the demineralization and decay of the tooth structure. In addition, the acidic environment promotes the shift of the plaque biofilm community, favoring the expansion of the acidogenic bacteria. S. mutans has been recognized as a major cariogenic bacterial species in humans and has been shown to induce dental caries in gnotobiotic animal models (14, 15, 16). S. mutans is critical in forming a stable acidic biofilm that recruits and maintains a cariogenic microbial community (17). The virulence of S. mutans has been attributed to its ability to 1) synthesize extracellular polymers of adhesive glucan from dietary sucrose that help establish colonization and accumulation on the tooth surface, 2) metabolize various carbohydrates into compounds that acidify the environment, demineralize the dental surface, and create favorable conditions for aciduric bacterial species, and 3) survive and thrive in stressful acidic environments (18, 19). As proposed by Socransky et al., different microbial complexes in dental plaque present at varying stages of the oral health condition (20, 21); this intermicrobial communication (i.e., social networking) has been shown to largely contribute to the disease process by enhancing metabolic function and biofilm structures that favor disease development as well as altering the host response to single microbial species challenges (22).

Epidemiological evidence suggests that S. mutans, while strongly linked with and often necessary for the development of ECC (23, 24), is not always present in children with caries (25, 26), and some of the most cariogenic strains are found in healthy patients (27, 28, 29). Other bacterial genera associated with caries patients and therefore potentially linked to oral diseases include Lactobacillus, Bifidobacterium, Scardovia, Parascardovia, Actinomyces, Fusobacterium, Porphyromonas, Selenomonas, Bacteroides, Haemophilus, Veillonella, Leptotrichia, and Thiomonas (24, 30). Although little is understood about the ecological and functional roles of these bacteria, it is increasingly clear that the cariogenic activity of an oral biofilm is most likely a complex and polymicrobial etiology rather than caused by a single species (32, 33, 34, 35). Therefore, it is important to investigate the microbial community to understand how intricate relationships and synergies impact the oral microbiome and cause disease.

Caries prevalence and the composition of caries-associated biofilm vary by race and ethnicity (36, 37, 38, 39, 40, 41). Total dental caries is highest for Hispanic youth (57.1%), followed by Non-Hispanic Black youth (48.1%), and the prevalence of untreated dental caries is highest for Non-Hispanic Black youth (17.1%) followed by Hispanic youth (13.5%) (1). Although other factors play a role in these racial and ethnic disparities (e.g., socioeconomic status, diet, and oral hygiene), substantial differences in microbiota composition are observed between racial and ethnic groups. Ethnicity-specific clustering of microbial communities in saliva and subgingival biofilms can allow ethnicity to be accurately predicted from microbial signatures (37, 39). There have also been significant differences in alpha and beta diversity, and taxa that were ethnic group-specific among children with caries were identified from supragingival samples collected from children (42). Interestingly, although the bacterial compositions are different, the pathogenicity of the oral microbiome is similar. Evidence suggests that S. mutans is a fundamental species that creates a favorable environment that recruits other species, including Lactobacillus spp. Streptococcus sobrinus, and Candida albicans. Although the species may differ based on host race or ethnicity, they fill a specific niche within the biofilm that promotes caries progression. Therefore, understanding the functional similarities and differences in children of different ethnicities can help identify factors that drive oral health disparities and reveal essential pathways for caries development.

ECC is a complex disease with multiple genetic, behavioral, and environmental risk factors and is contingent on a complex ecology of oral microbes. Because ethnic disparities are persistent after controlling for socioeconomic, dietary, and other environmental factors and because there are clear differences in microbiota composition between different ethnic groups, there are likely ethnicity-specific microbiome processes that impact caries risk. This study aimed to investigate diagnostic changes in transcriptionally active organisms in cavities versus healthy plaque and identify commonalities and differences between African American and Hispanic children. We used a paired sample experimental design (plaque from caries-active and non-caries sites within the same patient) that provided greater statistical power for capturing true differential expression at the same cross-sectional time point. We have used our previously published, novel transcript assignment approach to link differentially expressed genes to their taxonomic source at the species level (43). From our metatranscriptomic analysis, we show that the well-established cariogenic organisms were active in ECC in both African American (AA) and Latin American Hispanic (LAH) children, but there were also striking differences in active organisms and functions between the two groups. This demonstrates that the pathways to caries development can be diverse and ethnicity-specific and highlights the need for more targeted intervention and prevention strategies.

Study Overview and Population Demographics

In this study, paired dental plaque samples were collected from caries and non-caries lesions from AA (n = 12) and LAH (n = 7) subjects with ECC (Supplemental Table S1–ECC cohort). This cohort (n = 19) was 58% female and 42% male, with a mean age of 4.4 years. The caries status of the subjects was determined by the total number of primary teeth (dmft) and tooth surfaces (dmfs) that were decayed, missing (due to decay), and filled. The mean dmft and dmfs among AA and LAH subjects were 8.2 and 18.9, respectively. To establish a non-disease baseline, plaque samples were collected from age-matched, caries-free AA (n = 9) and LAH (n = 3) children (Supplemental Table S1–Caries-free cohort). This cohort was 58% female and 42% male, with a mean age of 3.9 years. Both ECC and caries-free children had similar low socioeconomic backgrounds (≤ 322% FPL).

Sequencing Depth and Quality

The total RNA extracted from plaque samples ranged from 100 ng to 1000 ng, and high sequencing coverage was achieved with low adaptor contamination (Supplemental Table S2). Species accumulation curves for the ECC and caries-free cohorts (Supplemental Figure S1), as well as the AA and LAH (Supplemental Figure S2) indicate that the number of bacterial organisms reached a point of diminishing returns (at approximately 32 samples in the ECC cohort and at 22 samples in the caries-free cohort) which suggests that the sequencing depth and the number of samples we collected sufficiently captured most dominant taxa.

Bacterial Species Abundance

The bacterial species composition of each sample was determined by alignment-based taxonomic classification of sequence reads using MTSv software package (43). Supplemental Table S3 shows the species with the most significant fold change (differences in gene expression between the paired plaque samples within the ECC cohort). In 13 of 19 children, the median fold change for Streptococcus mutans reads was nearly 500-fold higher in caries plaque. Other classical cariogenic bacterial species, including Lactobacillus rhamnosus and Streptococcus parasanguinis, were also identified in higher abundance in caries plaque. Interestingly, there were non-classical cariogenic bacterial species, including Parascardovia denticolens, Olsenella sp. oral taxon 807, Olsenella uli, Selenomonas sputigena, and Prevotella denticola, consistently more abundant in caries plaque of at least 7 ECC subjects.

Principle Component Analysis (PCA) was completed to identify the similarity of bacterial species composition of each sample type (caries plaque, non-caries plaque, and caries-free plaque). The non-caries plaque and the caries-free plaque samples generally clustered together, although two samples from a caries-free subject clustered near the caries plaque samples, suggesting a potential pre-caries stage (Supplemental Figure S3).

Shifting Metabolic Functions and Contributing Bacterial Species in Early Childhood Caries

To identify the changes in gene expression and bacterial species contributing to caries development, we further assigned the reads by gene and species using the same alignment-based approach (43). Differential gene expression analysis was carried out between caries lesions and matched non-caries plaque samples in the ECC cohort, as well as healthy, caries-free plaque samples in the caries-free cohort. Due to the similarities between the caries-free cohort samples and the non-caries plaque samples in the ECC cohort (Supplemental Figure S3), we primarily focused on the genes that were significantly differentially expressed (SDE) between the matched caries and non-caries plaque samples to elucidate the contributing taxonomic and putative metabolic mechanisms of cavity formation within the same mouth and to capture better the differences between the African American and Hispanic groups.

Metatranscriptomic analysis showed that core microbes contribute to the SDE genes shared among all ECC subjects, supporting the well-established drivers of ECC. We identified 7,763 SDE genes among all subjects with caries (Fig. 1), representing 2,186 KEGG orthologues (KOs). Of these genes, 6,967 were upregulated, and 676 were downregulated. Streptococcus mutans and Lactobacillus rhamnosus were associated with many genes that were SDE in both the AA and LAH groups; however, we also identified SDE genes from Veillanella parvula and Propionibacterium acidifaciens, which are more rarely associated with ECC. Genes that were upregulated in both groups (Fig. 2 – shared) were associated with S. mutans (15%), Veillonella parvula (15%), Propionibacterium acidifaciens (9%), lactobacilli (Lactobacillus rhamnosus–9%, L. gasseri–6%, L. oris–4%), Olsenella sp. (8%), and Bifidobacterium longum (5%).

The top functions of the shared up and downregulated genes for each species are shown in Fig. 3. Upregulated SDE genes were associated with more diverse bacterial species, whereas the downregulated SDE genes were predominantly associated with three bacterial species: Pseudopropionibacterium propionicum (51%), Abiotrophia defective (30%), and Streptococcus sanguinis (9%). Many upregulated porphyrin metabolism genes and lipopolysaccharide biosynthesis genes were associated with V. parvula (Fig. 3A). Most of the shared downregulated genes were associated with Pseudopropionibacterium propionicum and were related to oxidative phosphorylation and propanoate metabolism, while downregulated genes in S. sanguinis were associated with alkali-generating pathways (Fig. 3B).

Differences in active organisms and functions between African American and Latin American Hispanic children with ECC

Apart from the shared SDE genes, the AA and LAH groups presented a significant difference in the organisms and gene functions that were SDE in caries plaque. We identified 4,970 genes (from 2,106 KOs) that were uniquely SDE in AA ECC children and 6,519 genes (from 2,236 KOs) uniquely SDE in LAH ECC children (Fig. 1). Genes unique in AA children were primarily associated with two bacterial species, Pseudopropionibacterium propionicum and Cardiobacterium hominis, and most genes (3,224/4,970–65%) were downregulated (Figs. 2, 4, and 5), suggesting that these species were not as prevalent or active in the disease (caries) status. In addition, other studies have linked both species to healthy caries-free status (45, 48). The upregulated genes in the AA children included several acid-utilizing and/or producing species that are commonly associated with caries such as Campylobacter curvus, Streptococcus sp., lactobacilli, B. longum, and Selenomonas sputigena (26, 49, 50), as well as Streptococcus parasanguis which has been shown to help buffer the pH of the biofilm through alkali production (47).

Genes that were only SDE in the LAH ECC children were associated with a more diverse group of microorganisms (Propionibacterium acidifaciens, Selenomonas sp., Rothia dentocariosa, Atopobium parvulum, and Streptococcus sanguinis) and most of the genes (5,683/6,519–87%) were upregulated (Figs. 2, 4, and 5). Figure 4 illustrates a distinctive pattern of upregulated genes in the LAH children, with the predominant species differing from those observed in the AA children. The predominant contributors were S. sanguinis and A. defectiva, exhibiting notably fewer downregulated genes, although some downregulated genes were associated with Escherichia coli.

Overall, our data showed that there were major differences in gene expression and the active organisms between the AA and LAH children in caries plaque. There were 795 KOs that were SDE expressed in only one of the groups: 413 (52%) in the AA group and 382 (48%) in the LAH group (Supplemental Figure S4). This suggests that unique functions or pathways are specific to each group that are active in caries development. Additionally, even when the same KOs are SDE in both groups, there is evidence that different organisms fill these functional roles. Figure 6 shows similar changes in expression among orthologous KOs but in genes associated with a different set of species in each group. This evidence underscores an ecological shift in the contributing bacterial species despite maintaining a consistent functional profile.

In the past few decades, the ecological plaque hypothesis (31) has shifted our understanding of the role that microorganisms play in the formation of dental caries and periodontal diseases. This hypothesis suggests that oral disease development results from an imbalance in the total oral microflora, with the resulting ecological shift leading to the enrichment of “oral pathogens” or disease-related microorganisms. This ecological shift causes changes in the environment, favoring disease outcomes. For dental caries, the ecological shift of oral microorganisms appears to promote increased acid metabolism, which lowers environmental pH and fosters the demineralization of the tooth enamel surfaces. Our study directly isolated and sequenced RNAs from plaque biofilm in both caries and non-caries teeth in high-risk minority children. This enabled us to observe the direct ecological shift in oral bacteria and the change of the metabolic activities from non-disease status to disease status and identify the similarities and differences between the two groups of children.

Our findings reconfirm that classical cariogenic bacteria, such as S. mutans and Lactobacillus gasseri are not only highly expressed in caries plaque but also shared among all subjects in the AA and LAH groups. In addition to the classical cariogenic bacteria, we also identified other bacterial species, such as Veillonella parvula, Olsenella sp., and Bifidobacterium longum, which other studies have identified as potentially cariogenic (23, 32, 33, 34). Many upregulated porphyrin metabolism genes and lipopolysaccharide biosynthesis genes were associated with Veillonella parvula, suggesting a potential influence on bacterial membrane structure and host-relevant biochemical pathways. In addition, most of the shared downregulated genes were contributed by three species: Pseudopropionibacterium propionicum (51%), Abiotrophia defective (30%), and Streptococcus sanguinis (9%). These organisms were previously reported to be more abundant in caries-free children (44, 45, 46). Most of the shared downregulated genes were associated with Pseudopropionibacterium propionicum and were related to oxidative phosphorylation and propanoate metabolism. This suggests that energy production through oxidative phosphorylation and utilization of propanoate may be less efficient in caries (Fig. 3B). The expression of genes associated with alkali-generating pathways in S. sanguinis were downregulated in caries samples, which suggests a reduced ability to buffer the pH of the biofilm (47).

In the LAH group, more genes were SDE by a more diverse group of bacterial species than by the AA group. Several of these species have established associations with increased caries risk, including Propionibacterium acidifaciens, Selemonas sp., Lactobacillus oris, Rothia dentocariosa, Actinomyces sp., Atopobium parvulum, and Campylobacter concisus (45, 51, 52). Species largely associated with down-regulated genes include S. sanguinis and A. defective; both species have been linked to healthy, caries-free microbiomes. There were also downregulated genes from Escherichia coli, which may be antagonistic toward S. mutans (53).

Caries development is a complex and dynamic process as some oral microorganisms metabolize carbohydrates/sugars to produce lactic acids, other bacteria uptake lactate to produce minor, weak acids, such as propanoate and butanoate, under anaerobic conditions. The continuous production of weak acids by these non-classical cariogenic bacteria may be a key reaction contributing to the severity of the ECC presented in minority children. A recent study by Cho et al. (50) reported that Selenomonas sputigena could act as a pathobiont mediating spatial structure and potentially contribute to biofilm virulence in ECC children. These findings are consistent with our data, with S. sputigena being upregulated in caries in the AA ECC group and other Selenomonas species upregulated in the LAH ECC group. Other studies (61, 62, 63) have reported the interaction between fungi (Candida albicans) and S. mutans in promoting the development of ECC. However, in our study, Candida albicans was not consistently detected within the cavity or matched plaque samples - only 2 of 19 cavity samples (AA and LAH child) and 1 of 19 matched plaque samples had more than 1 million reads unambiguously assigned to C. albicans. In summary, our results confirm that there seems to be a shared core group of organisms that are necessary (but perhaps not sufficient) to bolster the development of caries. Additionally, other accessory organisms fill niches that contribute to caries progression, and these organisms differ between the AA and LAH groups.

A well-known challenge with human biofilm research is the variability among individuals and different sample types. Oral microbiome composition can vary based on several demographic and health factors and can differ significantly within an individual's saliva compared to plaque biofilm. However, within the same individual and sample type, the composition of the oral microbiome remains stable (64, 65). Therefore, we used paired caries and non-caries plaque samples within the same children to minimize variability and focus our statistical power on capturing significant functional and taxonomic changes between the disease and non-disease states.

Our novel gene alignment approach allowed us to count the number of reads assigned to each gene and unambiguously assign reads to genes from specific species. From this, we could identify which functions were differentially expressed in caries and which species expressed these genes. We showed that many of the same functions were overexpressed in caries in both the AA and LAH groups, but the organisms mainly responsible differed between the two groups. Similarly, major differences in the organisms associated with significantly downregulated genes suggest that the protective species also vary between the groups. With this approach, we can understand the functional changes that lead to caries development and identify which species will likely be involved in the changes.

Our study consisted of 38 samples from 19 ECC children and 24 samples from 12 caries-free children, which was sufficient (Supplementary Figure S1) to identify the major bacterial and functional components within the cavity microbiome, with the number of samples in each group exceeding the recommended number for our differential analysis (66). From each sample, we sequenced more than 50 million reads with a high percent remaining after human subtraction, and the species accumulation curves indicated that a saturation point was reached, so additional samples would not significantly alter the number of detected species. The geographic location of our study participants was restricted to the metropolitan Baltimore area in lower-income households. While it was helpful that the location and economic status of the participants were controlled in our study, our results may not reflect broader patterns, and further investigation is needed. Due to limited participants, we could not include non-Hispanic white (Caucasian) children; however, given the higher ECC prevalence and rate of untreated decay in African American and Hispanic children, it is appropriate to target these high-risk populations.

Despite the limitations, our study demonstrated that ethnicity-specific microbiomes contribute to early childhood caries in minority populations through different metabolic pathways and active oral bacteria. In addition to the classical cariogenic bacteria (S. mutans), our study identified other accessory bacterial species essential for ECC development. These findings could benefit the future development of innovative diagnostic assessments and personalized therapeutic and preventive treatments.

Study Population and Sample collection

Sample collection was approved by the Institutional Review Board of the University of Maryland at Baltimore (protocol #HP-00053225). For the study, children younger than 6 years of age at the pediatric dental clinic at the University of Maryland School of Dentistry (UMSOD) were enrolled. After a thorough explanation of the risks, benefits, and alternatives to the parent/legal guardian and confirmation of the eligibility, consent was obtained from the parent/legal guardian. The children were medically healthy, had not taken antibiotics or used corticosteroids in the preceding three months, and had no acute dental infection such as swelling or abscess at the time of the sample collection. Dental plaque from teeth with and without caries was collected using sterile toothpicks. The plaque samples were immediately transferred to RNAlater (0.5 ml per sample) and mixed before frozen at -80°C until further analyses were performed. In addition to plaque sample collection, descriptive data was also collected, including the patient’s age, sex, and dental condition, including dentition, the total number of decayed, missing, and filled primary teeth (dmft), and surfaces (dmfs).

RNA Extraction and Next Generation Sequencing

Dental plaque samples were lysed using bead-beating (FastPrep instrument with 0.1 mm silica spheres), in the presence of acid phenol. Following chloroform extraction, total RNA was precipitated using isopropanol and resuspended in DEPC-treated water. After DNase treatment to remove residual DNA, RNA was purified using the RNeasy kit (Qiagen). The following steps occurred at Azenta (formally GENEWIZ) NGS facility in New Jersey, USA. Briefly, an rRNA depletion sequencing library was prepared by using probes from QIAGEN FastSelect rRNA 5S/16S/23S kits (Qiagen, Hilden, Germany). The NEBNext Ultra II RNA Library Preparation Kit for Illumina was used for RNA sequencing library preparation following the manufacturer’s recommendations (NEB, Ipswich, MA, USA). The sequencing libraries were multiplexed and sequenced on an Illumina HiSeq-4000 instrument outputting 2x150bp Paired End reads.

Quality controls and computational subtraction of human reads

Sequence quality was assessed using the fastp software package (54, 55). Average quality scores and adapter contamination levels were within acceptable thresholds – quality scores of 30 or higher, and adapter contamination of 5% or lower. The only notable exception was the caries lesion sample from patient 75, which exhibited ~ 10% adapter contamination. Contaminating human reads were identified by aligning reads to the human genome (version GRCh38) using the Bowtie2 software package (56) with default parameters and computationally subtracting matched reads using samtools package (57) with default parameters.

Taxonomic composition analysis

To identify the species composition of the samples, we performed taxonomic classification of the sequence reads using the previously published MTSv software package (43). MTSv is an alignment-based metagenomic classifier that is highly accurate and produces fewer false positives than other rapid taxonomic classifiers. The reference sequences were downloaded from the NCBI’s Genbank Database (accessed 10/28/2019). All reference sequences without a valid (species level or below) taxonomic classification or those labeled as “environmental” samples were discarded; all sub-species taxonomic classifications (e.g., strain level) were relabeled as their parent species classification. Reads were aligned against the reference database with MTSv using a seed size of 14bp with a 2bp interval between seeds and a minimum requirement of 4 seed hits for initiating a full alignment. A set of most prevalent taxa was compiled by taking the union of the top 150 taxa in each sample, where the top taxa were defined as those with the highest number of uniquely assigned reads at the species level (at least 1,000 in each sample). The total number of species in the combined set was 290 (Supplemental Table S4).

Gene reference database

All functionally annotated genes associated with the most prevalent taxa were downloaded from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (https://www.genome.jp/kegg/, accessed August 2020) (65). When no gene data was available for a particular taxon in the KEGG database (50 of 290 instances), genes from the closest annotated species from the same genus were used. A total of 892,674 gene sequences corresponding to 7,940 KEGG Orthologues were obtained. The resulting sequences were used to build a custom database for species-level gene classification.

Differential expression analysis for comparative metatranscriptomics

The metatranscriptomic reads were assigned using MTSv with a custom gene reference database that allowed us to classify reads by gene and species (43). To improve sensitivity, we only considered alignments in regions unique to each gene for each species. This allowed us to reduce false assignments in cases where sequence similarity between genes from different species is high. DESeq2 software for R (59) was used to estimate fold changes using counts (transcripts per million) of uniquely assigned reads. The “approximate posterior estimation for generalized linear models” (apeglm) (60) model and software for estimating the more stable log2 fold change shrinkage from Zhu et al. were used (60). The full linear model used for the matched caries lesion and non-caries plaque samples was count ~ Study:Study.n + Study*Site, which captured the sample pairings and the effects of population. In this model “Study” referenced the population sampled and “Site” the disease/non-disease state. The full linear model used for the analysis of the caries-free cohort vs the ECC cohort was count ~ sample type where “count” was the observed number of unique alignments for a gene and “sample type” referenced the cohort membership of each sample.

An adjusted p-value using the Benjamini and Hochberg method (58) of 0.05 or lower and a shrunken log2 fold change greater than 0.58 (log₂ 1.5) was considered to be significant differential expression (SDE). For downstream analysis, we extracted genes that were differentially expressed in the caries lesion vs. paired non-caries plaque in the African American and Hispanic groups (Supplemental Table S5). Genes that were SDE in both groups were labeled “Shared” while genes that were SDE in only one group were labeled as either “African American only” or “Hispanic only”. Genes that were SDE in one group but in the other group had a similar shrunken log₂ fold change (within 0.5) but fell below the cutoff for significance were labeled as “Shared” to avoid counting differences that were simply an artifact cutoff placement. Functional categories were assigned to genes based on the KEGG functional hierarchy for the KO associated with each gene. For KOs with multiple functional classifications, we chose the one that appeared most commonly in the dataset and the more specific one in the case of a tie.

Data and Codes availability

The RNA sequencing reads are available through NCBI’s Sequence Read Archive BioProject: PRJNA1024640. Analysis scripts and Supplemental Table S4 and S5 are available through FigShare 10.6084/m9.figshare.25439143.

Competing interests

The authors declare no competing interests.

Author Contribution

KLCH, RKE, VYF conceptualized, designed, and supervised the study. KLCH performed and supervised the clinical and laboratory work. IS, TF, VYF analyzed and visualized the data. All authors drafted the manuscript together. All authors read, edited and approved the final content and format.

Acknowledgement

We thank Dr. Mark E. Shirtliff for his conceptualization and effort at the initial stage of this project. We thank Dr. Jan Harro for providing the knowledge of biofilm and related molecular techniques. This research was fully or partially supported by Colgate C.A.R.E program (KLCH), University of Maryland School of Dentistry INSPIRE grant (KLCH, JH), University of Maryland School of Dentistry PhD program in Dental Biomedical Science (RKE, KLCH) and National Institute of Health grant 1U54MD012388-01 (VYF).

Data Availability

Fleming, E., & Afful, J. Prevalence of Total and Untreated Dental Caries Among Youth: United States, 2015–2016. NCHS data brief, (307), 1–8 (2018)
al-Shalan, T. A., Erickson, P. R., & Hardie, N. A. Primary incisor decay before age 4 as a risk factor for future dental caries. Pediatric Dentistry, 19(1), 37–41 (1997).
Peretz, B., Ram, D., Azo, E., & Efrat, Y. Preschool caries as an indicator of future caries: a longitudinal study. Pediatric Dentistry, 25(2), 114–118 (2003).
Casamassimo, P. S., Thikkurissy, S., Edelstein, B. L., & Maiorini, E. Beyond the dmft: the human and economic cost of early childhood caries. Journal of the American Dental Association (1939), 140(6), 650–657 (2009). https://doi.org/10.14219/jada.archive.2009.0250
Griffin, S. O. et al. Dental services, costs, and factors associated with hospitalization for Medicaid-eligible children, Louisiana 1996-97. Journal of Public Health Dentistry, 60(1), 21–27 (2000). https://doi.org/10.1111/j.1752-7325.2000.tb03287.x
Kanellis, M. J., Damiano, P. C., & Momany, E. T. Medicaid costs associated with the hospitalization of young children for restorative dental treatment under general anesthesia. Journal of Public Health Dentistry, 60(1), 28–32 (2000).
Oliva, M. G., Kenny, D. J., & Ratnapalan, S. Nontraumatic dental complaints in a pediatric emergency department. Pediatric Emergency Care, 24(11), 757–760 (2008). https://doi.org/10.1097/PEC.0b013e31818c2641
Kichenbrand, C. et al. Brain abscesses and intracranial empyema due to dental pathogens: Case series. International Journal of Surgery case reports, 69, 35–38 (2020). https://doi.org/10.1016/j.ijscr.2020.03.025
Roy, H. et al. Oral microbes and the formation of cerebral abscesses: A single-centre retrospective study. Journal of Dentistry, 128, 104366 (2023). https://doi.org/10.1016/j.jdent.2022.104366
Dye, B. A. et al. Trends in oral health status: United States, 1988–1994 and 1999–2004. Vital and health statistics. Series 11, Data from the National Health Survey, (248), 1–92 (2007).
Dental Health Foundation. Mommy, It Hurts to Chew. The California Smile Survey: An Oral Health Assessment of California’s Kindergarten and 3rd Grade Children. February 2006. Available at: “https://centerfororalhealth.org/wp-content/uploads/2018/02/Mommy-It-Hurts-To-Chew.compressed.pdf ”. Accessed November 8, 2018. (Archived by WebCite at: “http://www.webcitation.org/73mOFxXRz”
Saint Louis C. Oral infections causing more hospitalizations. The New York Times. August 30, 2013. Available at: https://well.blogs.nytimes.com/2013/08/30/oral-infections-causing-more-hospitalizations/. Accessed November 8, 2018. (Archived by WebCite at: “http://www.webcitation.org/mainframe.php”)
Kazeminia, M. et al. Dental caries in primary and permanent teeth in children's worldwide, 1995 to 2019: a systematic review and meta-analysis. Head & Face Medicine, 16(1), 22 (2020). https://doi.org/10.1186/s13005-020-00237-z
Fitzgerald R. J. Inhibition of experimental dental caries by antibiotics. Antimicrobial Agents and Chemotherapy, 1(4), 296–302 (1972). https://doi.org/10.1128/AAC.1.4.296
McGhee, M. et al. Effective immunity to dental caries: protection of gnotobiotic rats by local immunization with Streptococcus mutans. Journal of Immunology (Baltimore, Md.: 1950), 114(1 Pt 2), 300–305 (1975).
de Stoppelaar, J. D., König, K. G., Plasschaert, A. J., & van der Hoeven, J. S. Decreased cariogenicity of a mutant of Streptococcus mutans. Archives of Oral Biology, 16(8), 971–975 (1971). https://doi.org/10.1016/0003-9969(71)90186-5
Bowen W. H. Dental caries - not just holes in teeth! A perspective. Molecular Oral Microbiology, 31(3), 228–233 (2016). https://doi.org/10.1111/omi.12132
Spatafora, G. et al. The Evolving Microbiome of Dental Caries. Microorganisms, 12(1), 121 (2024). https://doi.org/10.3390/microorganisms12010121
Manchanda, S. et al. Is Mutans Streptococci count a risk predictor of Early Childhood Caries? A systematic review and meta-analysis. BMC Oral Health, 23(1), 648 (2023). https://doi.org/10.1186/s12903-023-03346-8
Socransky, S. S. et al. Microbial complexes in subgingival plaque. Journal of Clinical Periodontology, 25(2), 134–144 (1998). https://doi.org/10.1111/j.1600-051x.1998.tb02419.x
Haffajee, A. D., Socransky, S. S., Patel, M. R., & Song, X. Microbial complexes in supragingival plaque. Oral Microbiology and Immunology, 23(3), 196–205 (2008). https://doi.org/10.1111/j.1399-302X.2007.00411.x
Lamont, R. J., Hajishengallis, G., & Koo, H. Social networking at the microbiome-host interface. Infection and Immunity, 91(9), e0012423 (2023). https://doi.org/10.1128/iai.00124-23
Chase, I. et al. Clinical outcomes for Early Childhood Caries (ECC): the influence of salivary mutans streptococci levels. European Journal of Paediatric Dentistry, 5(3), 143–146 (2004).
Lapirattanakul, J. et al. Oral Lactobacilli Related to Caries Status of Children with Primary Dentition. Caries Research, 54(2), 194–204 (2020). https://doi.org/10.1159/000506468
Aas, J. A. et al. Bacteria of dental caries in primary and permanent teeth in children and young adults. Journal of Clinical Microbiology, 46(4), 1407–1417 (2008) https://doi.org/10.1128/JCM.01410-07
Gross, E. et al. Bacterial 16S sequence analysis of severe caries in young permanent teeth. Journal of Clinical Microbiology, 48(11), 4121–4128 (2010). https://doi.org/10.1128/JCM.01232-10
Nakano, K. et al. Serotype distribution of Streptococcus mutans a pathogen of dental caries in cardiovascular specimens from Japanese patients. Journal of Medical Microbiology, 56(Pt 4), 551–556 (2007). https://doi.org/10.1099/jmm.0.47051-0
Carletto-Körber, F. P., González-Ittig, R. E., Jimenez, M. G., & Cornejo, L. S. Serotype diversity of Streptococcus mutans and caries activity in children in Argentina. European Journal of Paediatric Dentistry, 16(3), 177–180 (2015).
Bedoya-Correa, C. M., Rincón-Rodríguez, R. J., & Parada-Sanchez, M. T. Acidogenic and aciduric properties of Streptococcus mutans serotype c according to its genomic variability. European Journal of Oral Sciences, 129(6), e12824 (2021). https://doi.org/10.1111/eos.12824
Fakhruddin, K. S., Ngo, H. C., & Samaranayake, L. P. (2019). Cariogenic microbiome and microbiota of the early primary dentition: A contemporary overview. Oral Diseases, 25(4), 982–995. https://doi.org/10.1111/odi.12932
Marsh P. D. Microbial ecology of dental plaque and its significance in health and disease. Advances in Dental Research, 8(2), 263–271 (1994). https://doi.org/10.1177/08959374940080022001
Grier, A. et al. Oral Microbiota Composition Predicts Early Childhood Caries Onset. Journal of Dental Research, 100(6), 599–607 (2021). https://doi.org/10.1177/0022034520979926
Wang, Y. et al. Oral Microbiome Alterations Associated with Early Childhood Caries Highlight the Importance of Carbohydrate Metabolic Activities. mSystems, 4(6), e00450-19 (2019). https://doi.org/10.1128/mSystems.00450-19
Tanner, A. C. R. et al. The Caries Microbiome: Implications for Reversing Dysbiosis. Advances in Dental Research, 29(1), 78–85 (2018). https://doi.org/10.1177/0022034517736496
Simón-Soro, A., & Mira, A. Solving the etiology of dental caries. Trends in Microbiology, 23(2), 76–82 (2015). https://doi.org/10.1016/j.tim.2014.10.010
Naorungroj, S. et al. Racial differences in periodontal disease and 10-year self-reported tooth loss among late middle-aged and older adults: the dental ARIC study. Journal of Public Health Dentistry, 77(4), 372–382 (2017). https://doi.org/10.1111/jphd.12226
Cruz, G. D., Chen, Y., Salazar, C. R., & Le Geros, R. Z. The association of immigration and acculturation attributes with oral health among immigrants in New York City. American Journal of Public Health, 99 Suppl 2(Suppl 2), S474–S480 (2009). https://doi.org/10.2105/AJPH.2008.149799
Yang, Y. et al. Racial Differences in the Oral Microbiome: Data from Low-Income Populations of African Ancestry and European Ancestry. mSystems, 4(6), e00639-19 (2019). https://doi.org/10.1128/mSystems.00639-19
Mason, M. R. et al. Deep sequencing identifies ethnicity-specific bacterial signatures in the oral microbiome. PloS One, 8(10), e77287 (2013). https://doi.org/10.1371/journal.pone.0077287
Li, J. et al. Comparative analysis of the human saliva microbiome from different climate zones: Alaska, Germany, and Africa. BMC Microbiology, 14, 316 (2014). https://doi.org/10.1186/s12866-014-0316-1
van Palenstein Helderman, W. H., Matee, M. I., van der Hoeven, J. S., & Mikx, F. H. Cariogenicity depends more on diet than the prevailing mutans streptococcal species. Journal of Dental Research, 75(1), 535–545 (1996). https://doi.org/10.1177/00220345960750010501
Premaraj, T. S. et al. Ethnic variation of oral microbiota in children. Scientific Reports, 10(1), 14788 (2020). https://doi.org/10.1038/s41598-020-71422-y
Furstenau, T. N. et al. MTSv: rapid alignment-based taxonomic classification and high-confidence metagenomic analysis. PeerJ, 10, e14292 (2022). https://doi.org/10.7717/peerj.14292
Zhu, B., Macleod, L. C., Kitten, T., & Xu, P. Streptococcus sanguinis biofilm formation & interaction with oral pathogens. Future Microbiology, 13(8), 915–932 (2018). https://doi.org/10.2217/fmb-2018-0043
Ma, C. et al. Comparison of oral microbial profiles between children with severe early childhood caries and caries-free children using the human oral microbe identification microarray. PloS One, 10(3), e0122075 (2015). https://doi.org/10.1371/journal.pone.0122075
Kanasi, E. et al. Clonal analysis of the microbiota of severe early childhood caries. Caries Research, 44(5), 485–497 (2010). https://doi.org/10.1159/000320158
Burne, R. A., & Marquis, R. E. Alkali production by oral bacteria and protection against dental caries. FEMS Microbiology Letters, 193(1), 1–6 (2000). https://doi.org/10.1111/j.1574-6968.2000.tb09393.x
Wu, T. T. et al. Machine Learning Approach Identified Multi-Platform Factors for Caries Prediction in Child-Mother Dyads. Frontiers in Cellular and Infection Microbiology, 11, 727630 (2021). https://doi.org/10.3389/fcimb.2021.727630
Caufield, P. W. et al. Oral Lactobacilli and Dental Caries: A Model for Niche Adaptation in Humans. Journal of Dental Research, 94(9 Suppl), 110S–8S (2015). https://doi.org/10.1177/0022034515576052
Cho, H. et al. Selenomonas sputigena acts as a pathobiont mediating spatial structure and biofilm virulence in early childhood caries. Nature Communications, 14(1), 2919 (2023). https://doi.org/10.1038/s41467-023-38346-3
Thomas, R. Z. et al. Shifts in the microbial population in relation to in situ caries progression. Caries Research, 46(5), 427–431(2012). https://doi.org/10.1159/000339482
Torlakovic, L. et al. Microbial community succession on developing lesions on human enamel. Journal of Oral Microbiology, 4, 10.3402/jom.v4i0.16125 (2012). https://doi.org/10.3402/jom.v4i0.16125
Thurnheer, T., & Belibasakis, G. N. Integration of non-oral bacteria into in vitro oral biofilms. Virulence, 6(3), 258–264 (2015). https://doi.org/10.4161/21505594.2014.967608
Chen, Shifu. Ultrafast one-pass FASTQ data preprocessing, quality control, and deduplication using fastp. iMeta 2, e107 (2023). https://doi.org/10.1002/imt2
Chen, S., Zhou, Y., Chen, Y., & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics (Oxford, England), 34(17), i884–i890 (2018). https://doi.org/10.1093/bioinformatics/bty560
Langmead, B., & Salzberg, S. L. (2012). Fast gapped-read alignment with Bowtie 2. Nature Methods, 9(4), 357–359. https://doi.org/10.1038/nmeth.1923
Danecek, P. et al. Twelve years of SAMtools and BCFtools. GigaScience, 10(2), giab008 (2021). https://doi.org/10.1093/gigascience/giab008
Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological). 57(1):289–300 (1995). doi:10.1111/j.2517-6161.1995.tb02031.x
Love, M. I., Huber, W., & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550 (2014). https://doi.org/10.1186/s13059-014-0550-8
Zhu, A., Ibrahim, J. G., & Love, M. I. Heavy-tailed prior distributions for sequence count data: removing the noise and preserving large differences. Bioinformatics (Oxford, England), 35(12), 2084–2092 (2019). https://doi.org/10.1093/bioinformatics/bty895
Xiao, J. et al. Association between Oral Candida and Bacteriome in Children with Severe ECC. Journal of Dental Research, 97(13), 1468–1476 (2018). https://doi.org/10.1177/0022034518790941
Garcia, B. A. et al. Association of Candida albicans and Cbp + Streptococcus mutans with early childhood caries recurrence. Scientific Reports, 11(1), 10802 (2021). https://doi.org/10.1038/s41598-021-90198-3
Liu, Y. et al. Addition of cariogenic pathogens to complex oral microflora drives significant changes in biofilm compositions and functionalities. Microbiome, 11(1), 123 (2023). https://doi.org/10.1186/s40168-023-01561-7
Lazarevic, V. et al. Study of inter- and intra-individual variations in the salivary microbiota. BMC Genomics, 11, 523 (2010). https://doi.org/10.1186/1471-2164-11-523
Sato, Y. et al. Inter-Individual Differences in the Oral Bacteriome Are Greater than Intra-Day Fluctuations in Individuals. PloS One, 10(6), e0131607 (2015). https://doi.org/10.1371/journal.pone.0131607
Li, D. et al. An evaluation of RNA-seq differential analysis methods. PloS One, 17(9), e0264246 (2022). https://doi.org/10.1371/journal.pone.0264246

No competing interests reported.

ECCmanuscriptSupplemental.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Ethnicity-specific microbiome in early childhood caries: from the functional perspectives of oral biofilm

Status:

Version 1

Abstract

Figures

Introduction

Results

Study Overview and Population Demographics

Sequencing Depth and Quality

Bacterial Species Abundance

Shifting Metabolic Functions and Contributing Bacterial Species in Early Childhood Caries

Discussion

Methods

Study Population and Sample collection

RNA Extraction and Next Generation Sequencing

Quality controls and computational subtraction of human reads

Taxonomic composition analysis

Gene reference database

Differential expression analysis for comparative metatranscriptomics

Data and Codes availability

Declarations

Competing interests

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1