Impact of systemic antimicrobial therapy on the faecal microbiome in symptomatic dairy cows

Abstract

Antimicrobial resistance is a global threat to human and animal health, with the misuse and overuse of antimicrobials suggested as the main drivers of resistance. Antimicrobial therapy can alter the bacterial community composition and the faecal resistome in cattle. Little is known about the impact of systemic antimicrobial therapy on the faecal microbiome in dairy cows in the presence of disease. Therefore, this study aimed to assess the impact of systemic antimicrobial therapy on the faecal microbiome in dairy cows in the pastoral farm environment, by analysing faecal samples from cattle impacted by several different clinically-defined conditions and corresponding antimicrobial treatments. Analysis at the individual animal level showed a decrease in bacterial diversity and richness during antimicrobial treatment but, in many cases, the microbiome diversity recovered post-treatment when the cow re-entered the milking herd. Perturbations in the microbiome composition and the ability of the microbiome to recover were specific at the individual animal level, highlighting that the animal is the main driver of variation. Other factors such as disease severity, the type and duration of antimicrobial treatment and changes in environmental factors may also impact the bovine faecal microbiome. AmpC-producing Escherichia coli were isolated from faeces collected during and post-treatment with ceftiofur from one cow while no third-generation cephalosporin resistant E. coli were isolated from the untreated cow samples. This isolation of genetically similar plasmid-mediated AmpC-producing E. coli has implications for the development and dissemination of antibiotic resistant bacteria and supports the reduction in the use of critically important antimicrobials.

Fig 1. Timeline for faecal sampling.

Experimental sampling days are indicated on the figure (S1, S2 and S3) and the number of treated and control (untreated) cows is shown at the top of the figure. The cows re-entered the milking herd at the timepoint associated with sample S3. Healthy control cow samples were collected at the same time as those of the treated cows.

Fig 2. Diversity of faecal samples from treated (n = 30) and control (n = 24) cows.

Diversity of faecal samples from treated (n = 30) and control (n = 24) cows for A: Shannon diversity and B: Chao1 diversity. Pairwise comparison of the α-diversity of the faecal samples were compared for health status and sample order; **, p<0.01; ***, p<0.001; ****, p<0.0001; ns, not significant, p>0.05.

Fig 3. Principal component analysis for the faecal microbiome in treated and control cows.

Principal Component Analysis for the faecal microbiome in treated animals (red) and control animals (blue) over time (shapes). The percentage of variation explained in the Principal Component Analysis is indicated on the axis labels.

Fig 4. Procaine penicillin G case study.

Shannon (A) and Chao1 (B) α-diversity, and relative abundance of the phyla (C) from faecal samples from cows treated with procaine penicillin G (n = 3) and respective control cows (n = 3). Principal Component Analysis (D) for the faecal microbiome of treated and control cows at the individual animal level (colours) over the sampling period (shapes). The percentage of variation explained in the Principal Component Analysis is indicated on the axis labels.

Fig 5. Penethamate hydriodide case study.

Shannon (A) and Chao1 (B) α-diversity, and relative abundance of the phyla (C) from faecal samples from cows treated with penethamate hydriodide (n = 3) and respective control cows (n = 3). Principal Component Analysis (D) for the faecal microbiome of treated and control cows at the individual animal level (colours) over the sampling period (shapes). The percentage of variation explained in the Principal Component Analysis is indicated on the axis labels.

Fig 6. Marbofloxacin/penethamate hydriodide case study.

Shannon (A) and Chao1 (B) α-diversity of faecal samples from a cow treated with marbofloxacin/penethamate hydriodide (n = 1) and the respective control cow (n = 1). (C) Relative abundance of the phyla from faecal samples from a cow treated with marbofloxacin/penethamate hydriodide (n = 1) and the respective control cow (n = 1). Principal Component Analysis (D) for the faecal microbiome of the treated and control cow at the individual animal level (colours) over the sampling period (shapes). The percentage of variation explained in the Principal Component Analysis is indicated on the axis labels.

Fig 7. Ceftiofur case study.

Shannon (A) and Chao1 (B) α-diversity, and relative abundance of the phyla (C) from faecal samples from cows treated with ceftiofur (n = 3) and respective control cows (n = 3). Principal Component Analysis (D) for the faecal microbiome of treated and control cows at the individual animal level (colours) over the sampling period (shapes). The percentage of variation explained in the Principal Component Analysis is indicated on the axis labels.