Memory reconsolidation and its maintenance depend on L-voltage-dependent calcium channels and CaMKII functions regulating protein turnover in the hippocampus

Abstract

Immediate postretrieval bilateral blockade of long-acting voltage-dependent calcium channels (L-VDCCs), but not of glutamatergic NMDA receptors, in the dorsal CA1 region of the hippocampus hinders retention of long-term spatial memory in the Morris water maze. Immediate postretrieval bilateral inhibition of calcium/calmodulin-dependent protein kinase (CaMK) II in dorsal CA1 does not affect retention of this task 24 h later but does hinder it 5 d later. These two distinct amnesic effects are abolished if protein degradation by proteasomes is inhibited concomitantly. These results indicate that spatial memory reconsolidation depends on the functionality of L-VDCC in dorsal CA1, that maintenance of subsequent reconsolidated memory trace depends on CaMKII, and these results also suggest that the role played by both L-VDCC and CaMKII is to promote the retrieval-dependent, synaptically localized enhancement of protein synthesis necessary to counteract a retrieval-dependent, synaptic-localized enhancement of protein degradation, which has been described as underlying the characteristic labilization of the memory trace triggered by retrieval. Thus, conceivably, L-VDCC and CaMKII would enhance activity-dependent localized protein renewal, which may account for the improvement of the long-term efficiency of the synapses responsible for the maintenance of reactivated long-term spatial memory.

Fig. 1.

Intrahippocampal infusion of NIFE immediately after nonreinforced retrieval hinders spatial memory retention as measured 24 h or 5 d after reactivation. Animals with infusion cannulae implanted in the CA1 region of the dorsal hippocampus were trained during 5 d in the spatial version of the MWM. Twenty-four hours after the last training session, the animals were randomly assigned to one of four experimental groups and submitted to a 60-s probe test in the absence of the escape platform (P1) (black bar). Immediately after P1, the animals received intrahippocampal infusions of vehicle (VEH) (white bar) or NIFE (10 nmol per side; gray bar). Memory retention was assessed in a second 60-s probe test (P2) carried out 24 h (A and C) or 5 d after P1 (B and D). Data are expressed as means (± SEM) of the latency to swim over the previous location of the escape platform (A and B) or as the percentage of swimming time spent in the target quadrant (TQ) (C and D). *P < 0.05 and **P < 0.01 vs. VEH in Student t test (n = 9–12 per group).

Fig. 2.

Intrahippocampal infusion of AIP immediately after nonreinforced retrieval hinders spatial memory retention only if measured 5 d after reactivation. Animals with infusion cannulae implanted in the CA1 region of the dorsal hippocampus were trained during 5 d in the spatial version of the MWM. Twenty-four hours after the last training session, the animals were randomly assigned to one out of four experimental groups and submitted to a 60-s probe test in the absence of the escape platform (P1) (black bar). Immediately after P1, the animals received intrahippocampal infusions of vehicle (VEH) (white bar) or AIP (1.0 nmol per side; gray bar). Memory retention was assessed in a second 60-s probe test (P2) carried out 24 h (A and C) or 5 d after P1 (B and D). Data are expressed as means (± SEM) of the latency to swim over the previous location of the escape platform (A and B) or as the percentage of swimming time spent in the target quadrant (TQ) (C and D). *P < 0.05 vs. VEH in Student t test (n = 9 per group).

Fig. 3.

Intrahippocampal infusion of AIP immediately after nonreinforced remote retrieval hinders spatial memory retention only if measured 5 d after reactivation. Animals with infusion cannulae implanted in the CA1 region of the dorsal hippocampus were trained during 5 d in the spatial version of the MWM. Five days after the last training session, the animals were randomly assigned to one of four experimental groups and submitted to a 60-s probe test in the absence of the escape platform (P1) (black bar). Immediately after P1, the animals received intrahippocampal infusions of vehicle (VEH) (white bar) or AIP (1.0 nmol per side; gray bar). Memory retention was assessed in a second 60-s probe test (P2) carried out 24 h (A and C) or 5 d after P1 (B and D). Data are expressed as means (± SEM) of the latency to swim over the previous location of the escape platform (A and B) or as the percentage of swimming time spent in the target quadrant (TQ) (C and D). *P < 0.05 and **P < 0.01 vs. vehicle in Student t test (n = 10–12 per group).

Fig. 4.

The amnesic effect induced by the intrahippocampal infusion of ANI, NIFE, or AIP is blocked by proteasome antagonist. Animals with infusion cannulae implanted in the CA1 region of the dorsal hippocampus were trained during 5 d in the spatial version of the MWM. Twenty-four hours after the last training session, the animals were randomly assigned to one out of eleven experimental groups and submitted to a 60-s probe test in the absence of the escape platform (P1) (black bar). Immediately after P1, the animals received intrahippocampal infusions of vehicle (VEH) (white bar), ANI (1.0 μmol per side; dark gray bar), clasto-lactacystin βLAC (200 pmol per side; light gray bar), βLAC plus ANI (200 pmol and 1.0 μmol per side, respectively; horizontally striped bar), βLAC plus NIFE (200 pmol and 10 nmol per side, respectively; vertically striped bar), or βLAC plus AIP (200 pmol and 1.0 nmol per side, respectively; diagonally striped bar). Memory retention was assessed in a second 60-s probe test (P2) carried out 24 h (A and C) or 5 d after P1 (B and D). Data are expressed as means (± SEM) of the latency to swim over the previous location of the escape platform (A and B) or as the percentage of swimming time spent in the target quadrant (TQ) (C and D). *P < 0.05 and ***P < 0.001 vs. vehicle in Dunnett’s test after one-way ANOVA (n = 8–9 per group).