Mitogen-activated protein kinase signaling in the hippocampus and its modulation by corticotropin-releasing factor receptor 2: a possible link between stress and fear memory

Abstract

A coordinated activation of multiple interlinked signaling pathways involving cAMP-dependent protein kinase (PKA) and mitogen-activated extracellular signal-regulated kinases (Mek-1/2) regulates gene expression and neuronal changes underlying memory consolidation. In the present study we investigated whether these molecular cascades might mediate the effects of stress on memory formation. We also investigated the role of hippocampal corticotropin-releasing factor receptor 2 (CRF2) in stress-enhanced learning and molecular signaling mediated by PKA, Mek-1/2, and their downstream targets extracellularly regulated kinases 1 and 2 (Erk-1/2) and p90-ribosomal-s-kinase-1 (p90Rsk-1). Acute 1 hr immobilization was used as a stressful stimulus, and one-trial context-dependent fear conditioning was used as a model for associative learning. Training of BALB/c mice 3 hr after the end of immobilization resulted in an enhancement of conditioned fear, as indicated by significantly increased freezing behavior of stressed when compared with nonstressed mice. Interestingly, Erk-1/2 phosphorylation after conditioning of nonstressed and stressed mice depended on PKA and Mek-1/2, respectively. Intrahippocampal injection of the selective Mek-1/2 inhibitor U0126 or CRF2 antagonist antisauvagine-30 (aSvg-30) prevented stress-enhanced fear conditioning and Mek-1/2-dependent activation of Erk-1/2 and p90Rsk-1. aSvg-30 did not affect the phosphorylation of the PKA regulatory subunit II of stressed mice. The molecular and behavioral effects of CRF2 coincided with stress-induced upregulation of CRF2 mRNA. These results suggest that modulation of Mek-1/2-dependent signaling by hippocampal CRF2 can be selectively involved in the delayed effects of stress on memory consolidation.

Figure 1.

Activation of hippocampal Erk-1/2 after fear conditioning and stress-enhanced fear conditioning. A, Time course of Erk-1/2 phosphorylation after fear conditioning (FC) and stress-enhanced fear conditioning (Stress/FC). In both groups, a significant upregulation of pErk-1/2 was observed 1 hr after training. B, Representative immunoblots. Statistically significant differences: ^*p < 0.001 versus naive mice.

Figure 2.

Activation of the hippocampal MAPK pathway after fear conditioning and stress-enhanced fear conditioning. A, Experimental design used in subsequent studies. Mice of the FC group either were subjected 24 hr after training to a memory test or killed 1 hr after training to determine the activation of the MAP kinase pathway. Mice of the stress/FC group were treated identically except that training was performed 3 hr after the end of 1 hr immobilization. B, Mice of the stress/FC group froze significantly more than mice of the FC group during the contextual memory test. C, The level of pMek-1/2 was significantly increased in mice of the stress/FC, but not in mice of the FC group when compared with naive mice. pErk-1/2 and pp90 Rsk-1 were significantly upregulated in both FC and stress/FC groups; however, their levels in the stress/FC groups were significantly higher than both naive and FC groups. D, Representative immunoblots. Statistically significant differences: ^*p < 0.01 versus naive; ^ap < 0.01 versus FC group.

Figure 3.

PKA-dependent upregulation of pErk-1/2 in response to fear conditioning. A, Experimental design. Injection of the PKA inhibitor Rp-cAMPS (18 μg per mouse) into the dorsal hippocampus was performed 15 min before the context-dependent fear conditioning of mice of the FC and stress/FC groups. B, Rp-cAMPS significantly reduced freezing behavior when compared with vehicle-injected animals. C, Mice injected intrahippocampally with Rp-cAMPS exhibited significantly lower pErk-1/2 levels than vehicle-injected animals. D, Representative immunoblots. Statistically significant differences: ^*p < 0.01 versus corresponding FC- or stress/FC-vehicle group.

Figure 4.

Mek-1/2 activity mediates stress-enhanced Erk-1/2 phosphorylation and fear conditioning. Vehicle or U0126 (1 μg per mouse) were injected intrahippocampally 15 min before the training as described in Figure 3A. A, Intrahippocampal injection of U0126 prevented freezing behavior of the stress/FC group without affecting freezing of the FC group. B, Mice of the stress/FC group injected intrahippocampally with U0126 exhibited significantly lower pErk-1/2 levels than vehicle-injected animals. U0126 did not affect pErk-1/2 levels in mice of the FC group. C, Representative immunoblots. Statistically significant differences: ^*p < 0.01 versus stress/FC-vehicle group.

Figure 5.

Hippocampal CRF₂ is induced by acute immobilization. A, BALB/c mice (three per group) were immobilized for 1 hr, and brain section were prepared for in situ hybridization at the time points 0.5, 3, and 24 hr after the end of immobilization. Representative hippocampal micrographs of the in situ hybridizations using an antisense probe for CRF₂ are shown. B, Quantified signals for hippocampal CRF₂ mRNA. A significant upregulation of CRF₂ mRNA was observed within the hippocampal subregions CA1 and CA3 and the dentate gyrus (DG) at the time point 3 hr after the end of immobilization. Statistically significant differences: ^*p < 0.01 versus naive mice.

Figure 6.

aSvg-30 prevents stress-enhanced fear conditioning. A, Intrahippocampal microinfusion of aSvg-30 15 min before training (as described in Fig. 2A) significantly reduced freezing of mice of the stress/FC group when compared with their vehicle controls; however, aSvg-30 did not affect freezing behavior of mice of the FC group. B, Intrahippocampal microinfusion of aSvg-30 15 min before immobilization (4 hr before training) did not affect freezing of mice of the stress/FC group. Statistically significant differences: ^*p < 0.01 versus FC-vehicle group; ^ap < 0.01 versus stress/FC-vehicle group.

Figure 7.

aSvg-30 prevents the upregulation of the hippocampal MAP Kinase pathway after stress-enhanced fear conditioning. Intrahippocampal microinfusion of aSvg-30 15 min before training significantly reduced pMek-1/2 (A), pErk-1/2 (B), and pp90Rsk-1 (C) levels in the stress/FC group without affecting the phosphorylation of the PKA regulatory subunit II (D) when compared with the stress/FC-vehicle group. The levels of pMek-1/2 (A), pErk-1/2 (B), pp90Rsk-1 (C), and pPKA (D) were not affected by aSvg-30 injections in mice of the FC group. E, Representative immunoblots. Actin immunostaining was used to control for the total protein amount among samples. Statistically significant differences: ^*p < 0.01 versus FC-vehicle group; ^ap < 0.01 versus stress/FC-vehicle group.