Intense training prevents the amnestic effect of inactivation of dorsomedial striatum and induces high resistance to extinction

Abstract

A large body of evidence has shown that treatments that interfere with memory consolidation become ineffective when animals are subjected to an intense learning experience; this effect has been observed after systemic and local administration of amnestic drugs into several brain areas, including the striatum. However, the effects of amnestic treatments on the process of extinction after intense training have not been studied. Previous research demonstrated increased spinogenesis in the dorsomedial striatum, but not in the dorsolateral striatum after intense training, indicating that the dorsomedial striatum is involved in the protective effect of intense training. To investigate this issue, male Wistar rats, previously trained with low, moderate, or high levels of foot shock, were used to study the effect of tetrodotoxin inactivation of dorsomedial striatum on memory consolidation and subsequent extinction of inhibitory avoidance. Performance of the task was evaluated during seven extinction sessions. Tetrodotoxin produced a marked deficit of memory consolidation of inhibitory avoidance trained with low and moderate intensities of foot shock, but normal consolidation occurred when a relatively high foot shock was used. The protective effect of intense training was long-lasting, as evidenced by the high resistance to extinction exhibited throughout the extinction sessions. We discuss the possibility that increased dendritic spinogenesis in dorsomedial striatum may underly this protective effect, and how this mechanism may be related to the resilient memory typical of post-traumatic stress disorder (PTSD).

Fig 1. Histological results.

(A) Diagrams illustrating the location of infusion needle tips in dorsomedial striatum of the rats included in the statistical analyses. The number of needle tips is less than the actual number because of overlapping sites. (B) The photomicrograph shows a representative coronal slice with the tracks left by the cannulae (large arrow) and injector needle tips (small arrow). Numbers in the diagrams represent distance from bregma. Redrawn from Paxinos and Watson [40].

Fig 2. Effect of pre-training infusions of TTX into the dorsomedial striatum on memory consolidation of one-trial step-through inhibitory avoidance trained with 1.0 mA.

Timeline of experimental procedure is shown on top of figure. Data represent median step-through latencies, in seconds, with interquartile ranges. There were no significant differences between the TTX and VEH groups in training (A) and escape (B) scores. A significant retention deficit was found after administration of TTX (C). In this and in the following figures, the small circles represent individual data. *** p < 0.0001 vs. VEH; n = 9 rats per group.

Fig 3. Test for state-dependency.

Timeline of experimental procedure is shown on top of figure. Data represent median step-through latencies, in seconds, with interquartile ranges. TTX or VEH was infused into the dorsomedial striatum both before the training and retention sessions. Training was carried out using 1.0 mA. There were no significant differences between the TTX and VEH groups in training (A) and escape (B) scores, but a significant retention deficit was found after administration of TTX (C). ** p < 0.01 vs. VEH; number of rats per group: VEH = 11, TTX = 9.

Fig 4. Effect of pretraining TTX administration into the dorsomedial striatum on short-term and long-term retention latencies.

Timeline of experimental procedure is shown on top of figure. Data represent median step-through latencies, in seconds, with interquartile ranges. Training was carried out using 1.0 mA. There were no significant differences between the TTX and VEH groups in short-term memory (30 min) retention scores, but a significant long-term (24 h) retention deficit was produced by TTX. ** p < 0.01 vs. VEH; n = 9 per group. The number of groups in this experiment and in the remaining experiments was relatively high and for this reason in this figure and the following figures training and escape latencies were omitted to avoid overfilling. In all cases, training and escape latencies of the experimental groups did not differ significantly from VEH controls (Fig 4).

Fig 5. Effects of pre-training infusions of TTX into the dorsomedial striatum on retention during the first extinction session of one-trial inhibitory avoidance trained with no foot shock (0.0 mA), or with low (0.5 mA), moderate (1.0 mA), and high (3.0 mA) foot shock intensities.

Timeline of experimental procedure is shown on top of figure. Data represent median step-through latencies, in seconds, with interquartile ranges. There were no significant differences between the VEH and TTX 0.0 mA groups. When compared with their respective control (VEH) group, a highly significant retention deficit was produced by TTX after training with 0.5 and 1.0 mA, but an optimal retention score was evident in the 3.0 mA group. *** p < 0.0001 vs. VEH; number of rats per group: VEH 0.0 mA = 10, TTX 0.0 mA = 10, VEH 0.5 mA = 9, TTX 0.5 mA = 10, VEH 1.0 mA = 10, TTX 1.0 mA = 9, VEH 3.0 mA = 11, TTX 3.0 mA = 11. The number of animals in each group is the same for Figs 6 and 7.

Fig 6. Median retention latencies across the daily extinction sessions (E1–E7) shown by the groups trained in one-trial step-through inhibitory avoidance with 0.0, 0.5, 1.0, and 3.0 mA.

Timeline of experimental procedure is shown on top of figure. Note the high resistance to extinction displayed by the VEH and TTX groups that had been trained with 3.0 mA. See results of this section for details of statistical analyses.

Fig 7. Effect of pre-training infusions of TTX into the dorsomedial striatum on extinction of one-trial inhibitory avoidance trained with 0.0 mA, low (0.5 mA), moderate (1.0 mA), and high (3.0 mA) foot shock intensities.

The bars represent the median of latency scores (with interquartile ranges) of the seven extinction sessions for each group. When compared with their respective control (VEH) group, a significant retention deficit was produced by TTX after training with 0.5 and 1.0 mA, but no significant differences in retention scores appeared between the groups that had been trained with 0.0 or 3.0 mA. ** p < 0.01, *** p < 0.0001 vs. VEH.