# Resistome and microbiome profiling of bovine milk following antimicrobial dry cow therapy: insights from short- and long-read metagenomic sequencing

## TL;DR

This study examines how antimicrobial dry cow therapy affects the milk microbiome and antibiotic resistance in cows, finding no major changes in microbiota but increased resistance gene diversity.

## Contribution

The study reveals that antimicrobial dry cow therapy increases resistome diversity and highlights methodological challenges in resistome profiling using different sequencing platforms.

## Key findings

- Antimicrobial dry cow therapy increased the diversity and abundance of resistance genes in milk.
- Short-read and long-read sequencing showed limited agreement in resistance gene detection.
- Host DNA content limited sequencing depth and sensitivity in milk samples.

## Abstract

Selective antimicrobial dry cow therapy (DCT) is implemented as part of mastitis control programs, particularly in dairy cows with recent clinical episodes or elevated somatic cell counts. In this study, we investigated the effects of the use of antimicrobials at drying-off on the milk microbiota and resistome by comparing treated (T, n=18) and untreated (NT, n=13) cows. Milk samples from all animals were analyzed using short-read Illumina shotgun sequencing and a subset of 10 samples were also subjected to long-read Oxford Nanopore Technologies (ONT) sequencing. No significant differences in microbial composition or diversity were observed between treated and untreated groups with either technique, indicating that antimicrobial DCT may not induce long-term shifts in the milk microbiota. However, cows receiving antibiotic treatment showed a higher diversity and abundance of genetic determinants of resistance (GDRs) in their milk resistome. Findings from the two sequencing platforms revealed limited concordance in antimicrobial resistance gene content, highlighting that sequencing platform and bioinformatic pipeline choices substantially influence resistome profiling outcomes. Furthermore, the high proportion of host DNA limited sequencing depth and sensitivity, underscoring the need for improved host DNA depletion or targeted enrichment strategies. This study provides insights into the biological and methodological challenges of milk resistome characterization, particularly in low-biomass, host-DNA-rich samples and demonstrates the lack of standardized analytical approaches in resistome studies. Overall, our findings support the prudent use of antibiotics and highlight the need for further longitudinal studies to clarify the temporal dynamics of antimicrobial DCT effects on the milk resistome and microbiota.