Exploring the complex interplay between oral Fusobacterium nucleatum infection, periodontitis, and robust microRNA induction, including multiple known oncogenic miRNAs

Abstract

Fusobacterium nucleatum (F. nucleatum) is an oral commensal bacterium that can become pathogenic and is associated with periodontitis, adverse pregnancy outcomes, and colorectal cancer (CRC). MicroRNAs are conserved, non-coding RNA molecules that regulate gene expression and are detected in microbial infections. This study aims to characterize the microRNA expression kinetics in the mandibles of C57BL/6J mice infected with F. nucleatum for 8 and 16 weeks and to identify miRNAs as potential biomarkers using NanoString nCounter miRNA panels. Mice were divided into four groups: 8 weeks of infection, 16 weeks of infection, and their respective sham infection. F. nucleatum-infected mice showed 100% bacterial colonization on the gingival surface, along with a significant increase in alveolar bone resorption (P < 0.0001) and intravascular dissemination to the heart, indicating its invasive potential. Out of 577 miRNAs analyzed, seven miRNAs were upregulated, and two miRNAs were downregulated in the 8 weeks of infection group. In the 16 weeks of infection group, seven miRNAs were upregulated while 13 miRNAs were downregulated. Notably, miR-205, miR-210, and miR-199a-3p were differentially expressed at 8 weeks as well as miR-28 at 16 weeks and have been previously reported in human periodontitis. The 13 miRNAs induced by F. nucleatum (e.g., miR-361-5p) are linked to 13 multiple malignancies. In addition, miR-126-5p has been identified as a potential biomarker for patients with PD and cardiovascular disease. These results indicate that F. nucleatum induces several PD-related miRNAs and links them to systemic comorbidities. Furthermore, this study revealed F. nucleatum induction of 14 oncogenic miRNAs, and specifically, 12 miRNAs were linked with CRC.IMPORTANCEOur study investigated oral commensal Fusobacterium nucleatum, a critical bacterium associated with gum disease, adverse pregnancy outcomes, and enriched several tumors, including colorectal cancer (CRC). Recently, microRNAs have emerged as critical players in the interactions between host and microbes, and many host functions have been reported to be regulated by miRNAs during infection. F. nucleatum oral infection in mice induced gum disease, disseminated to the heart, lungs, and several miRNAs. Elevated miR-361 expression was linked to multiple cancers. In addition, miR-126-5p expression has been reported as a potential biomarker in patients with periodontitis and coronary artery disease, indicating F. nucleatum's virulence potential. The 13 miRNAs induced by F. nucleatum are linked to 13 multiple malignancies, including CRC. These results indicate that F. nucleatum acts as a potent cancer-causing bacterium. This study opens new avenues for exploring F. nucleatum's role in gum disease and its link with cancer.

Keywords: CRC miRNAs; F. nucleatum; machine learning models; miRNAs; periodontal disease.

Fig 1

Intraoral infection with F. nucleatum significantly induced ABR. (A) Schematic diagram of the experimental design depicting the monobacterial infection with F. nucleatum (4 days per week on every alternate week), plaque sampling for PCR, and euthanasia. (B) Representative images showing horizontal ABR in the mandible (lingual view) of F. nucleatum-infected and sham-infected mice, with the ABR area outlined from the alveolar bone crest (ABC) to the cementoenamel junction (CEJ). (C) Morphometric analysis of the mandible and maxillary ABR at 8 and 16 weeks post-infection. A significant increase in ABR was seen in F. nucleatum-infected mice compared to sham-infected mice (****P < 0.0001; **P < 0.01; *P < 0.05; ordinary two-way ANOVA). Data points and error bars are represented as mean ± SEM (n = 10).

Fig 2

DE miRNAs in F. nucleatum-infected mandibles (8 and 16 weeks). (A) The volcano plot depicts the upregulated (green) and downregulated (red) miRNAs that showed a fold difference of ±1.1 with a P-value of <0.05. The log2 fold change is on the x-axis, and the negative log of the P-value is on the y-axis. The black dots stand for the miRNAs that do not pass the filter parameters. Seven significant upregulated miRNAs and two downregulated miRNAs were found in 8 weeks of F. nucleatum-infected mice compared to 8 weeks of sham-infected mice (n = 10). Seven significant upregulated miRs and 13 downregulated miRs were found in 16 weeks of F. nucleatum-infected mice compared to 16 weeks of sham-infected mice (n = 10). (B) Venn diagram analysis illustrates the distribution of DE miRNAs in 8 weeks and 16 weeks of infections with F. nucleatum. (C and D) Predicted functional pathway analysis of DE miRNAs from F. nucleatum-infected mandibles. Bubble plot of KEGG analysis on predicted target genes of DE miRNAs in F. nucleatum-infected mice at 16 weeks of infection compared to sham-infected mice. The KEGG pathways are displayed on the y-axis, and the x-axis represents the false discovery rate (FDR), which means the probability of false positives in all tests. The size and color of the dots represent the number of predicted genes and the corresponding P-value, respectively.

Fig 3

DE miRNAs and mirPath V.3—KEGG predicted functional pathways (KEGG pathway # mmu05200 and mmu05100). (A) A total of 37 genes (identified by KEGG) were involved in the pathways in cancer and (B) 10 genes were involved in the bacterial invasion of epithelial cells signaling pathway during 16 weeks of F. nucleatum infection. Red boxes indicate genes with significantly increased expression during pathways in cancer and bacterial invasion of epithelial cells. Green boxes indicate no change in gene expression. Many pathogenic bacteria can invade phagocytic and non-phagocytic cells and colonize them intracellularly, then become disseminated to other cells. Invasive bacteria induce their uptake by non-phagocytic host cells (e.g., epithelial cells) using two mechanisms referred to as the zipper model and trigger model. Listeria, Staphylococcus, Streptococcus, and Yersinia are examples of bacteria that can enter using the zipper model. These bacteria express proteins on their surfaces that interact with cellular receptors, starting signaling cascades that result in close apposition of the cellular membrane around the entering bacteria. Shigella and Salmonella are examples of bacteria entering cells using the trigger model. An arrow indicates a molecular interaction and a line without an arrowhead indicates a molecular interaction that results in inhibition.

Fig 4

F. nucleatum upregulated miRNAs are linked with 13 multiple malignancies, including CRC. F. nucleatum-induced miR-361-5p alone is associated with 13 types of tumor, and 12 out of 14 miRNAs are associated with CRC.

Fig 5

A summary of the most prominent features in the five machine learning models using SHapley Additive exPlanations (SHAP) values. In panels A–E, feature importance is ranked from the top (the most important) to the bottom (the least important). The x-axis shows the impact that a feature has on the model. The bar charts show the overall impact of a feature, whereas the swarm plot shows both the positive and negative impacts. In the swarm plots, each dot represents an instance of a miRNA variable, and the color bar shows the variable’s value (high to low). The three study groups are (1) mice assessed at 8 weeks, (2) mice assessed at 16 weeks, and (3) 8 and 16 weeks together.