Long-term effects of brief acute stress on cellular signaling and hippocampal LTP

Abstract

In a previous study, we reported that a brief exposure to swim stress transforms an electrically induced, protein synthesis-independent early long-term potentiation (early LTP) into a protein synthesis-dependent late LTP ["reinforcement of LTP" in the hippocampal dentate gyrus (DG)] (Korz and Frey, 2003). This transformation depends on activation of mineralocorticoid receptors (MRs) by corticosterone, and on intact basolateral amygdala (BLA) function. Here, we demonstrate that a brief swim experience results in lasting changes in levels of hippocampal cellular signaling molecules that are known to be involved in the induction of late LTP. Within the DG, MRs were rapidly upregulated, whereas glucocorticoid receptor (GR) levels were elevated with a 3 h delay. Levels of phosphorylated mitogen-activated protein kinase 2 (pMAPK2) and p38 MAPK, as well as phosphorylated calcium/calmodulin-dependent protein kinase II (pCaMKII) were enhanced shortly after swim stress and remained elevated until 24 h, whereas levels of phosphorylated cAMP response element-binding protein (pCREB) remained unchanged. MR and GR were upregulated with a longer delay within the CA1 region, whereas levels of pMAPK2 and p38MAPK were rapidly increased, but the former returned to basal levels after 3 h. Levels of pCREB and pCaMKII were maintained in an enhanced state after swim stress. DG-LTP reinforcement requires a serotonergic but not dopaminergic heterosynaptic receptor activation that probably mediates the BLA-dependent modulation of LTP under stress. Thus, molecular alterations induced by specific stress resemble late LTP-related molecular changes. These changes, in interaction with stress-specific heterosynaptic processes, may support the transformation of early LTP into late LTP. The results contribute to the understanding of the rapid consolidation of cellular and possibly systemic memories triggered by stress.

Figure 1.

A, LTP after a 2 min (n = 9) and a 10 min (n = 7) swim, 15 min after application of a weak tetanus. Swimming animals showed significantly prolonged LTP as compared with nonswimming controls (n = 10) (A). B, Time course of serum corticosterone levels after a 2 min swim. Levels are significantly increased 15 min and 8 h and significantly reduced 24 h after swimming (n = 6 for controls; n = 8 for each experimental group). Given are the means and SEM. Asterisks indicate significant differences for the respective time points as compared with controls. The data set in A (15 min group) was partially adopted from Korz and Frey (2003) and complemented additional experiments.

Figure 2.

LTP is significantly reinforced after a 2 min swim, 1 h before (A) (n = 5) and 15 min (B) (n = 9) or 1 h after (C) (n = 6), but not 4 h after (D) (n = 9) a weak tetanus (sample controls, n = 10). Given are the means and SEM.

Figure 3.

Regulation of protein levels and levels of phosphorylated proteins within the dentate gyrus of the hippocampus at different time points (n = 4, each) after a 2 min swim for mineralocorticoid receptors (A), glucocorticoids receptors (B), pMAPKs (C), p38MAPK (D), αpCaMKII (E), and pCREB (F). Given are the means and SEM. Asterisks indicate significant differences for the respective time points as compared with controls.

Figure 4.

A–F, Regulation of protein levels and levels of phosphorylated proteins within the CA1-region of the hippocampus at different time points (n = 4, each) after a 2 min swim for mineralocorticoid receptors (A), glucocorticoids receptors (B), pMAPKs (C), p38MAPK (D), αpCaMKII (E), and pCREB (F). Given are the means and SEM. Asterisks indicate significant differences for the respective time points as compared with controls.

Figure 5.

LTP reinforcement by a 2 min swim is inhibited by blockade (dihydroergocristine) of serotonin receptors (A) (treated group, n = 8; controls, n = 9), but not by dopaminergic (Schering 23390) blockade (B) (treated group, n = 9; controls, n = 8) as compared with time-matched controls. Given are the means and SEM. The insets show representative analog traces of population spike amplitudes for indicated time points for individual animals of the different groups of experiments A–C (from left to right).