Longitudinal study of the short- and long-term effects of hospitalisation and oral trimethoprim-sulfadiazine administration on the equine faecal microbiome and resistome

Abstract

Background: Hospitalisation and antimicrobial treatment are common in horses and significantly impact the intestinal microbiota. Antimicrobial treatment might also increase levels of resistant bacteria in faeces, which could spread to other ecological compartments, such as the environment, other animals and humans. In this study, we aimed to characterise the short- and long-term effects of transportation, hospitalisation and trimethoprim-sulfadiazine (TMS) administration on the faecal microbiota and resistome of healthy equids.

Methods: In a longitudinal experimental study design, in which the ponies served as their own control, faecal samples were collected from six healthy Welsh ponies at the farm (D0-D13-1), immediately following transportation to the hospital (D13-2), during 7 days of hospitalisation without treatment (D14-D21), during 5 days of oral TMS treatment (D22-D26) and after discharge from the hospital up to 6 months later (D27-D211). After DNA extraction, 16S rRNA gene sequencing was performed on all samples. For resistome analysis, shotgun metagenomic sequencing was performed on selected samples.

Results: Hospitalisation without antimicrobial treatment did not significantly affect microbiota composition. Oral TMS treatment reduced alpha-diversity significantly. Kiritimatiellaeota, Fibrobacteres and Verrucomicrobia significantly decreased in relative abundance, whereas Firmicutes increased. The faecal microbiota composition gradually recovered after discontinuation of TMS treatment and discharge from the hospital and, after 2 weeks, was more similar to pre-treatment composition than to composition during TMS treatment. Six months later, however, microbiota composition still differed significantly from that at the start of the study and Spirochaetes and Verrucomicrobia were less abundant. TMS administration led to a significant (up to 32-fold) and rapid increase in the relative abundance of resistance genes sul2, tetQ, ant6-1a, and aph(3")-lb. lnuC significantly decreased directly after treatment. Resistance genes sul2 (15-fold) and tetQ (six-fold) remained significantly increased 6 months later.

Conclusions: Oral treatment with TMS has a rapid and long-lasting effect on faecal microbiota composition and resistome, making the equine hindgut a reservoir and potential source of resistant bacteria posing a risk to animal and human health through transmission. These findings support the judicious use of antimicrobials to minimise long-term faecal presence, excretion and the spread of antimicrobial resistance in the environment. Video Abstract.

Keywords: Antimicrobial resistance; Antimicrobial resistance genes; Aph(3”)-lb; Horse; Microbiota; Shotgun metagenomic sequencing; ant6-1a; lnuC; sul2; tetQ.

Fig. 1

Relative abundance of the most abundant bacterial communities at (I) phylum and (II) class level in faecal samples collected from Welsh ponies (A–F) at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (D27–D211)

Fig. 2

Mean relative abundance of specified phyla over time. Mean relative abundance of phyla in the faecal microbiota of Welsh ponies at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (D27–D211). Scaling on the y-axis varies according to relative abundance

Fig. 3

Alpha-diversity (Shannon) over time. Alpha-diversity of the faecal microbiota of Welsh ponies at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (D27–D211). The blue solid line is the mean, and the grey area represents one standard deviation

Fig. 4

Compositional differences in faecal microbiota over time. Bray-Curtis non-metric multidimensional scaling (NMDS) plot of beta-diversity of the faecal microbiota of Welsh ponies at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26) and after discharge from the hospital (short-term follow-up D27–D34) up until 6 months after hospitalisation and antimicrobial treatment (long-term follow-up D41–D211). Overall, PERMANOVA and pairwise PERMANOVA indicate significant differences between all groups (p<0.001), with six out of ten comparisons having significant differences in beta dispersion (p<0.05)

Fig. 5

Random forest analysis of microbiota composition over time. Random forest analysis of faecal microbiota composition of Welsh ponies (A–F) at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (D27–D211). “Start of study” samples (D0-D13-1) and “TMS treatment” samples (D22–D26) were used as classes for training. 1 = 100% similarity with samples at the start of the study

Fig. 6

Resistome abundance in the faecal microbiome. Stacked bar chart of the relative abundance of resistance genes clustered at the antimicrobial class level observed in the faecal microbiome of Welsh ponies (A–F) at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (D27–D211)

Fig. 7

Relative abundance of resistance genes over time. Relative abundance of resistance genes in the faeces of Welsh ponies at the farm (D0–D13-1), during hospitalisation without treatment (D14–D21), during hospitalisation and treatment with TMS (D22–D26), and after discharge from the hospital up until 6 months after hospitalisation and antimicrobial treatment (long-term D27–D211). I sul2, II tetQ, III ant(6)-la, IV aph(3”)-lb, V lnuC. ALR: additive log-ratio