Fig. 4: Improvement of cognition by nasal vaccination.

a Elevated plus maze test. Time spent on open arms was significantly increased in FTLD-tau mice treated with control-v, and this change was suppressed by tau-v treatment. One control-v-treated control mouse and one tau-v-treated control mouse dropped from the open arms, thus failing to complete the task [F(3, 59) = 15.461, p < 0.0001, control/control-v vs. FTLD-tau/control-v: p < 0.0001, FTLD-tau/control-v vs. FTLD-tau/tau-v: p = 0.0379]. Tested by one-way ANOVA, followed by Fisher’s LSD test (n = 15–17 mice in each group). b–d Barnes circular maze test. In the training trials b, an increase of latency to reach the target hole in the second half of the trials was observed in control-v-treated FTLD-tau mice, but was reversed by tau-v treatment (F(3, 61) = 4.506, p = 0.0064, control/control-v vs. FTLD-tau/control-v: p = 0.0064, FTLD-tau/control-v vs. FTLD-tau/tau-v: p = 0.0385). Tested with repeated measures ANOVA. In the probe trials c 24 h after the last (24th) training trial, tau-v also significantly attenuated the elongation of latency to reach the target hole in FTLD-tau mice (F(3, 61) = 5.877, p = 0.0014, control/control-v vs. FTLD-tau/control-v: p = 0.0003, FTLD-tau/control-v vs. FTLD-tau/tau-v: p = 0.0408). Tested with one-way ANOVA followed by Fisher’s LSD test. The decrease in time spent around the target hole by FTLD-tau mice d was significantly suppressed by tau-v treatment (*p = 0.0867). Time spent around the target versus the next holes (mean duration) was tested by paired t-test (n = 16–17 mice in each group). e–g Fear conditioning test. Freezing times (expressed as percent of total time) in each minute of conditioning e, context testing f, and cued testing g with altered context were analyzed. No significant abnormalities in FTLD-tau mice compared to control mice were observed in conditioning and context testing e, f. However, freezing time with an auditory cue was significantly increased in control-v-treated FTLD-tau mice, but was reversed to control level by tau-v treatment g (F(3, 61) = 3.887, p = 0.00131, control/control-v vs. FTLD-tau/control-v: p = 0.0257, FTLD-tau/control-v vs. FTLD-tau/tau-v: p = 0.0042). Tested with repeated measures ANOVA followed by Fisher’s LSD test (n = 16–17 mice in each group). All blots were derived from the same experiment and were processed in parallel.