Weaning has been described as one of the most stressful events in the life of horses. Given the importance of the interaction between the gut-brain axis and gut microbiota under stress, we evaluated (i) the effect of two different weaning methods on the composition of gut microbiota across time and (ii) how the shifts of gut microbiota composition after weaning affect the host. A total of 34 foals were randomly subjected to a progressive (P) or an abrupt (A) weaning method. In the P method, mares were separated from foals at progressively increasing intervals every day, starting from five min during the fourth week prior to weaning and ending with 6 h during the last week before weaning. In the A method, mares and foals were never separated prior to weaning (0 d). Different host phenotypes and gut microbiota composition were studied across 6 age strata (days â30, 0, 3, 5, 7, and 30 after weaning) by 16S rRNA gene sequencing. Results revealed that the beneficial species belonging to Prevotella, Paraprevotella, and Ruminococcus were more abundant in the A group prior to weaning compared to the P group, suggesting that the gut microbiota in the A cohort was better adapted to weaning. Streptococcus, on the other hand, showed the opposite pattern after weaning. Fungal loads, which are thought to increase the capacity for fermenting the complex polysaccharides from diet, were higher in P relative to A. Beyond the effects of weaning methods, maternal separation at weaning markedly shifted the composition of the gut microbiota in all foals, which fell into three distinct community types at 3 days post-weaning. Most genera in community type 2 (i.e., Eubacterium, Coprococcus, Clostridium XI, and Blautia spp.) were negatively correlated with salivary cortisol levels, but positively correlated with telomere length and N-butyrate production. Average daily gain was also greater in the foals harboring a community type 2 microbiota. Therefore, community type 2 is likely to confer better stress response adaptation following weaning. This study identified potential microbial biomarkers that could predict the likelihood for physiological adaptations to weaning in horses, although causality remains to be addressed.