Rapid stress-induced corticosterone rise in the hippocampus reverses serial memory retrieval pattern.

We previously showed that an acute stress (electric footshocks) induced both a rapid plasma corticosterone rise and a reversal of serial memory retrieval pattern in a contextual serial discrimination (CSD) task. This study is aimed at determining (i) if the rapid stress effects on CSD performance are mediated by the hippocampus; (ii) if hippocampal corticosterone membrane receptor activation is involved in the rapid stress effects on CSD performance. In experiment 1, microdialysis in the dorsal hippocampus (dHPC) was used to measure the stress-induced corticosterone rise; in parallel, the effect of acute stress on CSD performance was evaluated. In addition, the functional involvement of corticosterone in the behavioral effects of stress was assessed by administering metyrapone, a corticosterone synthesis inhibitor, before stress. In experiment 2, the involvement of hippocampal corticosterone membrane receptors in the stress-induced reversal of CSD performance was studied by injecting corticosterone-bovine serum albumin (BSA) (a membrane-impermeable complex) in the dHPC in non stressed mice. Results showed that (i) the acute stress induced a rapid (15 min) and transitory (90 min) corticosterone rise into the hippocampus dHPC, and a reversal of serial memory retrieval pattern; (ii) both the endocrinal and memory stress-induced effects were blocked by metyrapone; (iii) corticosterone-BSA injection into the dHPC in non stressed mice mimicked the effects of stress on serial retrieval pattern. Overall, our study is first to show that (i) a rapid stress-induced corticosterone rise into the dHPC transitorily reverses serial memory retrieval pattern and (ii) hippocampal corticosterone membrane receptors activation is involved in the rapid effects of acute stress on serial memory retrieval.

Prefrontal cortex or basolateral amygdala lesions blocked the stress-induced inversion of serial memory retrieval pattern in mice.

Previous data from our team have shown that pre-test stress in mice reversed the pattern of memory retrieval in a contextual serial spatial task (CSD; Celerier, A., Pierard, C., Rachbauer, D., Sarrieau, A., & Beracochea, D. (2004). Contextual and serial discriminations: A new learning paradigm to assess simultaneously the effects of acute stress on retrieval of flexible or stable information in mice. Learning and Memory, 11, 196-204). The present study is aimed at determining brain areas which might be critically involved in mediating the stress effect on memory retrieval in the CSD task. For that purpose, we studied hereby the effects of ibotenic acid lesions of either the prefrontal cortex (PFC) or the basolateral amygdala (BLA) in Stressed or Non-Stressed Balb/c mice on memory retrieval in the CSD task. In that task, mice learned two successive spatial discriminations (D1 and D2) within two different internal contexts in a four-hole board. The stressor (electric footshocks) was delivered 5 min before test, occurring 24 h after acquisition. During test, mice were relocated either on the floor of the first or of the second discrimination. Results showed that (i) spatial memory was substantial and remained unaffected both by lesions and stress; (ii) Non-Stressed controls as well as Non-Stressed or Stressed PFC and BLA-lesioned mice remembered accurately D1 but not D2; and (iii) in contrast, Stressed controls accurately remembered D2 but not D1. In parallel to behavioral experiments, we also showed that PFC and BLA lesions did not affect the stress-induced increase of plasma corticosterone levels. All together, PFC and BLA integrity are not necessary for retrieval processes per se; in contrast, the PFC and BLA are critically involved in the mediation of the deleterious stress effects on serial order memory retrieval.

Mechanism of the hypothermic effect of MPP+ administered centrally in mice.

The neurotoxin methyl phenyl pyridinium (MPP+) was administered intracerebroventricularly (i.c.v.) to mice. From the 1.25 microg dose per mouse, MPP+ elicited a dose-dependent hypothermic effect from doses as low as 1.25 microg per mouse. The minimal lethal dose was determined to be between 17.5 and 20 microg per mouse. The hypothermia induced by 2.5 microg MPP+ was unaffected by pretreatment with propranolol (8 mg/kg, i.p.), scopolamine (5 mg/kg, s.c.) and haloperidol (250 microg/kg, i.p.). It was decreased by yohimbine (4 mg/kg, s.c.), idazoxan (5 mg/kg, s.c.) and desipramine (20 mg/kg, i.p.). In mice injected i.c.v. with 6 hydroxydopamine (50 microg, 8 days before testing with MPP+ 2.5 microg), a significant reduction in the hypothermic effect of MPP+ was observed. A similar 6 OHDA injection has been shown previously to reduce by about 40% the DA striatal content of DA and by about 70% the hypothalamic content of NE. On the contrary, in mice injected with MPP+ (17.5 microg, 8 days before testing with 50 microg 6 OHDA) there was no modification in the hypothermic effect of 6 OHDA (50 microg). This injection of MPP+ reduced by about 40% the striatal content of DA but did not affect the hypothalamic content of NE. It is concluded that MPP+ decreases body temperature, at least in part, by acting as an indirect NE agonist, which stimulates alpha2 adrenoreceptors. In contrast, this agent in the present experimental conditions, does not destroy NE neurons in opposition to its action on DA neurons.