A short course of dietary amoxicillin was associated with different short-term shifts in the fecal microbiome of three breeds of young stud boars. Large White and Landrace animals had their lowest measured microbial diversity while the antibiotic was being given, whereas Duroc boars showed no significant difference across the study time points. The researchers set out to characterize short-term changes in the animals’ intestinal flora and its recovery after treatment.
The study included 103 boars: 40 Large White, 47 Landrace and 16 Duroc animals. The boars were 50 to 55 days old at the starting measurement, and fecal samples were collected before treatment, during exposure and after the drug was withdrawn.
A closer look at the treatment window
Amoxicillin was mixed into feed at 100 milligrams per kilogram from study day 3 through day 7. Samples were taken on day 0, day 7 and day 14, allowing the researchers to compare the animals with their own pretreatment measurements and to examine the period after exposure ended.
The researchers used fecal 16S rRNA sequencing, a method that identifies bacterial groups through a targeted part of their genetic material. The dataset contained 279 fecal samples, 17,045,354 raw sequences, 8,110 sequence variants, 19 bacterial phyla and 327 genera.
The analysis filtered and merged sequence reads, removed likely chimeras and used DADA2 to generate the sequence variants. Taxonomy was assigned with the Silva database, while community diversity and structure were assessed with QIIME2, Bray-Curtis analysis, principal coordinates analysis and PERMANOVA. The study also used PICRUSt2 to infer possible microbial functions from the sequence data, rather than measuring gene content directly.
For statistical comparisons, pairwise tests used the Wilcoxon rank-sum method with false discovery rate correction, while multiple-group comparisons used Kruskal-Wallis tests in R.
Diversity dipped during exposure
In both Large White and Landrace boars, the number of observed bacterial types and the Shannon diversity index were lower on day 7 than on days 0 and 14. The reported difference had a p-value below 0.001. The reported day 0-to-day 14 comparisons were not significant for observed species in Large White boars or for either index in Landrace boars.
The Duroc results were different. Neither diversity index showed a significant difference among the three time points in that breed. The authors interpret the findings as evidence that the apparent response varied by breed, although the Duroc group was the smallest of the three.
The overall mix of bacterial communities also differed by breed and time point, according to principal coordinates analysis and PERMANOVA, with a reported p-value below 0.001. In Large White and Landrace boars, day 7 samples formed a distinct pattern from both day 0 and day 14 samples. In Duroc boars, day 0 samples did not overlap with day 7 or day 14 samples, while some overlap remained between the latter two time points.
The result needs caution
These findings show a time-linked association, not proof that amoxicillin caused every change. The study did not include a parallel untreated cohort, and randomization was not reported, so age, environment and husbandry changes over the same period could not be separated from exposure-associated changes.
The apparent return of diversity in Large White and Landrace boars by day 14 should therefore be read as a short-term temporal pattern, not proof of complete recovery. Fecal samples reflect distal-gut luminal microbes and do not describe every intestinal compartment. The sequencing results also cannot establish the loss of absolute bacterial numbers, the complete elimination of any bacterial group or a change in antibiotic-resistance risk.
The functional results are similarly predictive. PICRUSt2 estimates likely pathways from bacterial sequences, but those predictions are not direct measurements of gene content or resistance-gene abundance. Larger studies with untreated comparisons, more Duroc animals and direct metagenomic, resistance-gene and absolute-abundance measurements would be needed to test whether the breed-specific patterns hold.
Paper data and sources
Original title: Dynamic changes in gut microbiota of three breeds of stud boars pre- and post-amoxicillin administration
Authors: Wang L, He H, Huang T, Jiang Z
Journal/Repository: Frontiers in veterinary science
Status: Peer-reviewed
First online: 2026-10-19
DOI: Not available
Original paper · Full text