# Characterization and ex vivo modelling of endodontic infections from the Arabian Gulf region

## TL;DR

This study explores the bacterial diversity and antibiotic resistance in endodontic infections from the Arabian Gulf region and successfully models these infections as ex vivo biofilms.

## Contribution

The study introduces ex vivo biofilm models directly derived from clinical endodontic infection samples in the Arabian Gulf region.

## Key findings

- Endodontic samples showed high bacterial diversity with 366 species detected.
- Ex vivo biofilms developed from clinical samples were polymicrobial and reproducible.
- A significant proportion of isolated bacteria showed resistance to clinically relevant antibiotics.

## Abstract

The microbiota of endodontic infections in patients from the Arabian Gulf region (AGR) is largely unexplored. While research in different global regions has investigated the microbial composition of such infections, studies using shotgun metagenomic sequencing (SMS) alongside culture‐dependent techniques (CDT) are limited. There are also few in vitro biofilm models that reflect the microbial profiles of endodontic infections. Therefore, by employing SMS and CDT, this research aimed to explore compositional and functional microbial profiles of endodontic infections from the AGR. The research also sought to develop ex vivo biofilms directly from endodontic infection samples.

SMS and CDT were used to analyze 32 root canal samples from necrotic pulp. Patients' samples were categorized into two cohorts: symptomatic (n = 19) and asymptomatic (n = 13). Samples underwent sequencing followed by bioinformatic analysis to investigate microbial composition, resistome, virulome, and functional differences. Two representative samples (8R, 15R) were selected to develop ex vivo biofilms on hydroxyapatite coupons. Similarity between inoculum and developed biofilms was assessed using SMS and CDT. The reproducibility of developed biofilms was assessed based on microbial composition and relative abundance at the species level using correlation coefficient analysis.

Endodontic samples had high bacterial diversity, with a total of 366 bacterial species detected across the two cohorts. Several antibiotic resistance (n = 59) and virulence (n = 82) genes were identified, with no significant differences between the cohorts. CDT identified 28 bacterial species, with 71.4% of the isolated bacteria having phenotypic resistance to clinically relevant antibiotics. SMS showed that the ex vivo biofilms were polymicrobial. Biofilm derived from sample 15R had 9 species and was dominated by Enterococcus faecalis, while sample 8R had 12 species and was dominated by Streptococcus mutans. Pearson correlation analysis demonstrated a significant positive correlation between biological biofilm replicates, confirming the reproducibility of biofilm formation.

There was high bacterial diversity in root canal samples from necrotic pulp. Samples were shown to contain antibiotic resistance and virulence genes, with no differences evident between symptomatic and asymptomatic infections. A high number of isolated bacteria were resistant to clinically used antibiotics. Ex vivo biofilm models from clinical samples were successfully developed and reproducibly reflected a polymicrobial composition.