Differential response to prolonged amoxicillin treatment: long-term resilience of the microbiome versus long-lasting perturbations in the gut resistome

ABSTRACT

The collateral impact of antibiotics on the microbiome has attained increasing attention. However, the ecological consequences of long-term antibiotic exposure on the gut microbiome, including antibiotic resistance, are still limited. Here, we investigated long-term exposure effects to amoxicillin on the human gut microbiome and resistome. Fecal samples were collected from 20 patients receiving 3-months of amoxicillin or placebo treatment as part of a Norwegian multicenter clinical trial on chronic low back pain (AIM study). Samples were collected at baseline, last day of treatment, and 9 months after antibiotic cessation. The abundance and diversity of microbial and resistome composition were characterized using whole shotgun and functional metagenomic sequencing data. While the microbiome profiles of placebo subjects were stable over time, discernible changes in diversity and overall microbiome composition were observed after amoxicillin treatment. In particular, health-associated short-chain fatty acid producing species significantly decreased in proportion. However, these changes were short-lived as the microbiome showed overall recovery 9 months post-treatment. On the other hand, exposure to long-term amoxicillin was associated with an increase in total antimicrobial resistance gene load and diversity of antimicrobial resistance genes, with persistent changes even at 9 months post-treatment. Additionally, beta-lactam resistance was the most affected antibiotic class, suggesting a targeted response to amoxicillin, although changes at the gene level varied across individuals. Overall, our results suggest that the impact of prolonged amoxicillin exposure was more explicit and long-lasting in the fecal resistome than in microbiome composition. Such information is relevant for designing rational administration guidelines for antibiotic therapies.

KEYWORDS:

• Antimicrobial resistance
• microbiome
• resistome
• antimicrobial resistance gene
• amoxicillin
• antibiotics
• chronic low back pain
• Modic changes
• functional metagenomics
• shotgun

Acknowledgments

The sequencing service was provided by the Norwegian Sequencing Centre (www.sequencing.uio.no), a national technology platform hosted by the University of Oslo and supported by the “Functional Genomics” and “Infrastructure” programs of the Research Council of Norway and the Southeastern Regional Health Authorities. The sequencing data analysis and storage were performed on SAGA & NIRD resources provided by Sigma2 - the National Infrastructure for High Performance Computing and Data Storage in Norway. Also, we would like to thank Marius Trøseid, Martin Kummen and Johannes ER Hov for invaluable advice with planning collection of fecal samples and Knut Morten Huneide, Marianne Thorsø, Anne Julsrud Haugen and Lars Grøvle for help with sample collection.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability

Clean shotgun metagenomic reads after removal of human DNA and assembled functional metagenomic contigs are available at NCBI SRA under BioProject ID: PRJNA894204. Additional participants metadata (deidentified) will be available upon request addressed to [email protected], in accordance with local registration and ethical approval.

Contributions

FCP conceived the study design, experiments and analyses. LCHB collected the fecal samples and clinical characteristics from participants. K Stu., RJ, and GS were involved in metagenomic DNA extraction from feces and prepared shotgun metagenomic sequencing libraries. HA created functional metagenomic libraries, performed functional selections and prepared functional metagenomic sequencing libraries. AB and GS also contributed to the functional metagenomics. K Sto. and JAZ planned and performed the original AIM trial. DB contributed with the concept and idea of collecting fecal samples. AD performed the computational analyses of shotgun metagenomic sequencing and functional metagenomic data. AD interpreted the results and drafted the manuscript and figures with critical revision performed by FCP. RJ and AD edited the final figures in the manuscript. All authors made a substantial contribution to the revision of the manuscript and approved its final version.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19490976.2022.2157200

Additional information

Funding

This work was supported by the Norwegian Research Council (NFR) (project number: 273833, 274867), Olav Thon Foundation (project number: 421258), by governmental organizations Helse Sør-Øst RHF (grant number: 2015090) and KLINBEFORSK (grant number: 2017201).