Fig. 2

The gut microbiota is necessary and sufficient for the effects of metformin on cognitive function. a The escape latency in the Abx and metformin-treated (Abx + Met) group was not different from that of the Abx-treated and control groups during the acquisition phase. b Latency to first reaching the platform. c Number of times mice crossed the platform. d Time that mice stay in target quadrant where the platform was located. e The mean velocity was not different among the Abx + Met, Abx, and control groups. f The FMT + Met group had a shorter latency in reaching the escape platform than the control, Abx-treated, and FMT-treated groups during the acquisition phase. g The latency to first crossing the previous platform location was significantly shorter in the FMT + Met group than that in the control, Abx, and FMT groups. h The number of times the previous platform location was crossed and i the time in the target quadrant were significantly increased in the FMT + Met group during the probe test. j The mean velocity during the probe test was not different among the four groups. k The number of new objects touched times was significantly increased in mice treated with metformin and decreased in the Abx + Met group. However, this behaviour was restored in Abx-treated specific pathogen-free mice who underwent standardization of the gut microbiota with that from mice in the same litter. Ctrl, control; Met, metformin. The overall significance among three or four groups was determined by one-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant