FIGURE 3.

Delay eyeblink conditioning (EBC) in adult cbln1^-/- mice improves after a cerebellar subarachnoid Cbln1 injection. (A) A diagram showing a sequence of experiments. HA-Cbln1 (1 μg/g body weight) was injected into the subarachnoid spaces over the bilateral HVI lobes of 7- to 10-week-old cbln1^-/- mice. The mice were trained using the delay EBC paradigm beginning 2 or 30 days after the injection. (B) HA-Cbln1 disappears from the cerebellum by 10 d after injection. Immunoblotting with an antibody against HA was used to detect injected HA-Cbln1 in the P2 fraction of the whole cerebellum before (-), 2 days, and 10 days after the injection. Arrows indicate bands corresponding to HA-Cbln1. The blot represents results from n = 3 mice. (C) Distribution of injected Cbln1. Cerebellar slices were stained with anti-HA antibody (red) to detect HA-Cbln1 24 h after the injection. Neurons visualized with fluorescent Nissl stains (Neurotrace; green). The injected Cbln1 was localized to the cerebellar cortex. Asterisks in the enlarged images indicate the deep cerebellar nucleus (DCN). Scale bars, 20 μm. (D) Cbln1 transiently rescued the impaired delay EBC in cbln1^-/- mice. The CR% increased significantly in cbln1^-/- mice trained 2 days after a Cbln1 injection [+Cbln1 (2d)] compared to those trained 30 days after [+Cbln1 (30 days)] the injection or those not treated with Cbln1 (no inj; p < 0.01, a two-way repeated measures ANOVA, n = 8 for each group). T1-T7, training sessions; Sp1 and Sp2, spontaneous eyeblink responses before training. The dotted line corresponds to the average percentage of CR-like activities on Sp2 to indicate the baseline for learned responses.