Phenotypic and Genotypic Characterization of Streptococcus mutans Strains Isolated from Endodontic Infections

Abstract

Streptococcus mutans plays an important role in caries etiology and eventually in systemic infections. However, it is often found in infected root canals, but the pathophysiological characteristics of strains residing in this site are largely unknown. Here, we characterized strains of S. mutans isolated from root canals of primary (PI) and secondary/persistent (SI) endodontic infections in relation to serotype and genotype; presence of genes coding for collagen binding proteins (CBPs); collagen binding activity and biofilm formation capacity; ability to withstand environmental stresses; systemic virulence in Galleria mellonella; and invasion of human coronary artery endothelial cells and human dental pupal fibroblasts. Samples from 10 patients with PI and 10 patients with SI were collected, and a total of 14 S. mutans isolates, belonging to 3 genotypes, were obtained. Of these, 13 were serotype c, and 1 was serotype k. When compared with the reference strains, the clinical isolates were hypersensitive to hydrogen peroxide. Remarkably, all 14 strains harbored and expressed the CBP-encoding gene cbm, showing increased binding to collagen, enhanced systemic virulence in G. mellonella, and ability to invade human coronary artery endothelial cells and human dental pupal fibroblasts when compared with CBP-negative strains. Whole genome sequence analysis of PI and SI isolates revealed that these strains are phylogenetically related but genetically distinct from each other. Our findings highlight the importance of CBPs in facilitating colonization and persistence of S. mutans in collagenous substrates such as root canals and their potential role in the pathogenesis of endodontic infections.

Keywords: Collagen binding proteins; Streptococcus mutans; dental pulp cavity; endodontic infections.

Figure 1.

Whole genome comparison analysis (panel A) showing pan genome results from Roary pipeline. (A) Phylogenies were estimated by maximum-likelihood across six S. mutans isolates from a multi-locus sequence typing alignment from Roary. Presence/Absence gene matrix of all isolates shows a clear correspondence between PI (P1 and P6) and SI (S1 and S4). (B) Heatmap of genomic distance and clusters (Neighbor Joining approach) again shows how close PI (P1 and P6), and SI (S1 and S4) genomes are, in contrast with references strains (LAR01 and UA159).

Figure 2.

Growth inhibition radius of S. mutans clinical isolates from primary (P) secondary (S) endodontic infections exposed to a filter paper disc soaked with 0.5% H₂O₂. Experiments were performed in triplicates. The asterisk indicates significant differences between clinical strains and reference strains (p <0.05).

Figure 3.

Invasion of human dental pulp fibroblasts (HDPF) (A) and of human coronary artery endothelial cells (HCAEC) (B) by S. mutans isolates from primary (P) and secondary (S) endodontic infections. Reference CPB-positive strains OMZ175 and LAR01 were used as positive controls whereas the CBP-negative UA159 was used as negative control. Experiments were performed at least in triplicates. The asterisk indicates significant differences (p <0.05) between clinical and reference strains.

Figure 4.

Survival of G. mellonella systemically infected with S. mutans isolated from primary (P) and secondary (S) endodontic infections. Reference strains UA159 (CBP-negative), OMZ175 (Cnm-positive) and LAR01 (Cbm-positive) and heat killed OMZ175 (HK) were used as controls. Experiments were carried out in triplicates for at least 48 hours. The asterisk indicates that all clinical strains killed significantly more G. mellonella larvae than UA159 at different time-points (p <0.05).