Species-Level Gut Microbiota Analysis after Antibiotic-Induced Dysbiosis in Horses

SIMPLE SUMMARY: DNA sequencing has been used to characterize the intestinal microbiota of horses. However, all published studies are based on Illumina technology to date because it is widely accessible, but a major limitation is the short bacterial DNA fragment sequenced. Other long-read sequencing technologies (i.e., PacBio) can classify bacteria at the species level and therefore are more meaningful for the identification of specific changes in the gut after the use of antimicrobials, for example. The objectives of this study were to characterize the gut microbiota of horses at the species level before and after trimethoprim sulfadiazine administration and to compare results with short-read Illumina sequencing. PacBio sequencing failed to classify the equine intestinal microbiota at the species level but detected greater richness and less unclassified bacteria compared to Illumina sequencing. Bacteroidetes were the most abundant bacteria in the feces, followed by Firmicutes and Fibrobacteres in this study. Further studies should focus on improving databanks for equine studies. ABSTRACT: All current studies have used Illumina short-read sequencing to characterize the equine intestinal microbiota. Long-read sequencing can classify bacteria at the species level. The objectives of this study were to characterize the gut microbiota of horses at the species level before and after trimethoprim sulfadiazine (TMS) administration and to compare results with Illumina sequencing. Nine horses received TMS (30 mg/kg) orally for 5 days twice a day to induce dysbiosis. Illumina sequencing of the V4 region or full-length PacBio sequencing of the 16S rRNA gene was performed in fecal samples collected before and after antibiotic administration. The relative abundance and alpha diversity were compared between the two technologies. PacBio failed to classify the equine intestinal microbiota at the species level but confirmed Bacteroidetes as the most abundant bacteria in the feces of the studied horses, followed by Firmicutes and Fibrobacteres. An unknown species of the Bacteroidales order was highly abundant (13%) and deserves further investigation. In conclusion, PacBio was not suitable to classify the equine microbiota species but detected greater richness and less unclassified bacteria. Further efforts in improving current databanks to be used in equine studies are necessary.