Novel potential biomarkers for predicting childhood caries via metagenomic analysis

Abstract

Background: Dental caries is a prevalent global health issue, particularly among children, with significant oral and overall health implications. The oral microbiome is considered a critical factor in caries development, with various microbial species implicated in the disease process.

Objectives: This study aims to explore the changes and interactions of oral microbiota in childhood caries using metagenomic analysis, and identify potential biomarkers for early caries detection and treatment.

Methods: Saliva samples were collected from 241 children aged 6 to 9 years, categorized into caries-free (CF), low-caries (CL), and caries-severe (CS) groups. Metagenomic sequencing was performed to analyze the oral microbiome, followed by a series of statistical and functional analyses to characterize microbial diversity and function.

Results: The study revealed significant differences in the microbial community composition among the groups, with the CS group exhibiting higher alpha and beta diversity than that of the CF group. Numerous unclassified microorganisms, such as Campylobacter SGB19347 and Catonella SGB4501, are intimately linked to dental caries and display intricate interaction networks, suggesting the potential formation of a distinct ecological network. In functional assessment, we identified a possible link between pectin and caries, suggesting that microorganisms that produce pectinase enzymes might play a role in the advancement of severe dental caries. Additionally, we identified 16 species as the best marker for severe dental caries, achieving an impressive AUC of 0.91.

Conclusion: The role of microbiota in dental caries is multifaceted, involving a complex interplay of microbial species and functions. Our findings enhance the understanding of the microbial basis of dental caries and offer potential diagnostic and therapeutic targets. The predictive capacity of the identified biomarkers warrants further investigation for early caries detection and intervention.

Clinical significance: The identification of novel biomarkers through metagenomic analysis enables early detection and targeted intervention for childhood caries, potentially transforming children dental care and significantly improving long-term oral health outcomes.

Keywords: biomarkers; children; dental caries; metagenomic sequencing; oral microbiome.

Figure 1

Description of taxa between CF, CL and CS group at each taxonomic level. (a) the microbial composition analysis of the CF, CL and CS group at the genus. (b) the microbial composition analysis of the CF, CL and CS group at the species. (c) the collection of the CF, CL and CS group in the genus. (d) the upset plot of the CF, CL and CS group in the genus.

Figure 2

Association between microbiota composition and caries status. (a) Shannon and Simpson index between the CF group and the CL group (p = 0.13, p = 0.082). (b) Shannon and Simpson index between the CF group and the CS group (p = 0.043, p = 0.022). (c) Shannon and Simpson index between the CL group and the CS group (p = 0.6, p = 0.57). (d) PCoA analysis of the CF, CL and CS group based on Bray-Curtis distance between the bacterial communities present in all specimens (R²= 0.03%, p = 0.001).

Figure 3

Potential salivary microbial biomarkers associated with severe caries. (a) microbial biomarkers associated with severe caries in the saliva samples at FDR < 0.001. (b) thirty-one species were observed to be closely associated with severe caries and their FDR values ​​were taken and ranked. Red sectors represent for severe caries associations and blue sectors represent for caries-free associations. Dark red and dark blue represent the first expression pattern, and light red and light blue represent the second expression pattern (c) the AUC index of thirty-one species. (d) the Foldchange index of thirty-one species.

Figure 4

The network between potential salivary microbial biomarkers (value≥0.5). Shows communities (namely, right for caries-free and left for Severe caries) of bacterial species and their positive (blue Pearson coefficient) or negative (red Pearson coefficient) abundances correlation. Alternatively, edge thickness holds an inverse relationship to the Pearson p-value and is color-coded based on the Pearson coefficient, with blue representing positive correlations and red, negative ones. A dashed line delineates a “structural gap”, indicating a significant number of negative Pearson correlation edges between caries free and severe caries communities.

Figure 5

Functional metagenomic signatures between the CS and CF group. differentially expressed KEGG pathways between the CS and CF group (a, b). (a) pectate lyase (K01728). (b) uncharacterized protein (K07071). differentially expressed Metacyc pathways between the CS and CF group (c-i). (c) the starch degradation III pathway (PWY-6731). (d) the superpathway of menaquinol-9 biosynthesis (PWY-5845). (e) the superpathway of demethylmenaquinol-9 biosynthesis (PWY-5862). (f) L-citrulline (L-cit) biosynthesis (CITRULBIO-PWY). (g) urea cycle (PWY-4984). (h) norspermidine (NSPD) biosynthesis (PWY-6562). (i) uroporphyrinogen-III (PWY0-1415). (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 6

Potential diagnostic biomarker for severe caries. (a) consortia of bacteria predict severe caries status. Receiver Operating Characteristic (ROC) with tenfold cross-validation was employed to assess the best oral bacterial species, and the best species for predicting severe caries and its AUC value are shown. orange line represents AUC of the top 16 species-level predictors, dark red line represents AUC of the top severe caries enriched predictive species (Streptococcus mutans, S mutans) and blue line represent AUC of the top severe caries enriched predictive species (Campylobacter SGB19347, C SGB19347). (b) the top 16 species-level predictors (determined by mean decrease GINI) for a random forest model predicting severe caries. (c) variation in relative abundance of the top severe caries enriched predictive species (Streptococcus mutans, S mutans) with dmft index. (d) variation in relative abundance of the top caries-free enriched predictive species (Campylobacter SGB19347, C SGB19347) with dmft index.