Limited Day-to-Day Variation in the Canine Gut Microbiota: Implications for Microbiome Studies

David Atuahene^1,2Luke Wolfe²Oona Elisabet Vanhatalo³Benjamin T Veenstra³Joseph H Skarlupka²Katie L Anderson³Garret Suen²Giorgia Meineri¹Jessica Pritchard^3*

• ¹Department of Veterinary Sciences, School of Agriculture and Veterinary Medicine, University of Turin, 10095 Grugliasco, Turin, Italy
• ²Department of Bacteriology, University of Wisconsin-Madison, Wisconsin-Madison WI 53706, United States
• ³Department of Medical Sciences, University of Wisconsin-Madison School of Veterinary Medicine, WI 53706, Madison, United States

The gut microbiome is vital for health and affects gastrointestinal, systemic, and neurological functions. In dogs, fecal samples provide an effective mechanism to assess the gut microbiota as they are non-invasive, easily obtained, and representative of the gut microbiome. However, traditional methods usually require sampling across three consecutive days per time point to minimize the presumed variation in the gut microbiome. Here, we sought to investigate whether the gut microbiome obtained from a single-day fecal sampling is reflective of the microbiome obtained from three-day collections. To accomplish this, we collected fecal samples from 12 dogs over three days and compared each single-day microbiome against the combined microbiotas of the three-day samples. We found no significant daily variation in the gut microbiota, as determined by two one-sided tests of equivalence (TOST) analysis, indicating that there are little to no day-to-day changes in the microbiota. Further microbial comparisons using PERMANOVA (p = 0.98) and non-metric multidimensional scaling also showed no significant differences in the microbial composition across the sampled days. Taken together, our findings suggest that a single sample can represent the gut microbiome as accurately as samples obtained across three consecutive days. As such, a single-day sampling approach can be used in dog microbiome studies, which would reduce both labor and costs while preserving overall data quality.