Responses of the Resident Oral Microbiome of Children to Amoxicillin Therapy

Abstract

Background: The World Health Organisation considers antibiotic resistance an increasing global threat that creates a clinical and financial burden on healthcare systems worldwide. The British Prime Minister in July 2014 announced that antibiotic resistance is not a distant threat, with 25,000 people a year already dying from infections resistant to antibiotic drugs in Europe alone. Dentists account for 7 to 10% of all antibiotic prescriptions. Several paediatric dentistry guidelines recommend amoxicillin for the management of dental abscesses. Antibiotics undoubtedly impact our microbiome. The knowledge of the effect of antibiotic use on the oral microbiome is limited, especially in children. A better understanding of the effect of antibiotics and their disturbance to the oral microbiome is required. Aims: This longitudinal cohort study examines the hypothesis that the oral microbiome is altered in function and in composition through amoxicillin administration to children and the oral microbiome is a reservoir for antibiotic resistance genes (ARGs) from both oral and non-oral organisms. Method: A total of 14 children aged 4-11years old attending the Leeds Dental Hospital who were indicated for antibiotic prescription were recruited. Saliva samples and clinical information were collected from children before, two-week and three-month after amoxicillin therapy. Oral microbial profiles (taxonomic and functional) were obtained using a shotgun metagenomic and meta-transcriptomic approaches. Unique reads were annotated using blastp against the CARD database (e-value ≤ 1e-5) to characterise ARGs. Comparison analysis was used to evaluate the composition, function, and resistome of the oral microbiome at different timepoints. Results: There were great inter-individual differences in taxonomic, functional, and resistome profiles. The ubiquitous oral microbiome was identified at different taxonomy levels before and after antibiotic exposure and included bacterial, fungal, archaeal, and viral components. Streptococcus was found to be the most abundant (18.9%) and active (26.3%) genus in the oral microbiome and showed resilience, where it was depleted in the two-week timepoint and recovered after three months of amoxicillin therapy. The statistically significant changes in the composition of the oral microbiome occurred at the two-week timepoint at the phylum, genus, and species levels (q-value < 0.05). KEGG pathway annotations indicated high proportions of genes involved in carbohydrate and amino acid metabolisms (32,453 and 25,712 genes respectively) in all samples. The functional metagenomic and metatranscriptomic investigations of the KEGG and eggNOG analyses generally agreed. CAZy analyses revealed an enrichment of the six CAZy families in all the samples, with glycoside hydrolases being the most enriched family. The statistically significant changes in the function profile of the oral microbiome occurred at the two-week and three-month timepoints. The study identified 68 ARG types in the samples. A statistically significant increase in ARGs number after three months compared to the baseline was found. Additionally, thirty ARGs showed statistically significant abundance changes between timepoints, including the ARGs conferring resistance to amoxicillin (TEM-1, OXA-85, OprM, golS, and mtrE). The dominant antibiotic resistance mechanisms of the identified ARGs were antibiotic efflux (44.2%) followed by antibiotic inactivation (19.1%). Conclusion: Differences in the metagenome, metatranscriptome and resistome of children is showing following treatment with amoxicillin. The study indicates a substantial but incomplete recovery of the oral microbiome following three months of amoxicillin therapy.