Chronic impairment of ERK signaling in glutamatergic neurons of the forebrain does not affect spatial memory retention and LTP in the same manner as acute blockade of the ERK pathway

Abstract

The ERK/MAPK signaling pathway has been extensively studied in the context of learning and memory. Defects in this pathway underlie genetic diseases associated with intellectual disability, including impaired learning and memory. Numerous studies have investigated the impact of acute ERK/MAPK inhibition on long-term potentiation and spatial memory. However, genetic knockouts of the ERKs have not been utilized to determine whether developmental perturbations of ERK/MAPK signaling affect LTP and memory formation in postnatal life. In this study, two different ERK2 conditional knockout mice were generated that restrict loss of ERK2 to excitatory neurons in the forebrain, but at different time-points (embryonically and post-natally). We found that embryonic loss of ERK2 had minimal effect on spatial memory retention and novel object recognition, while loss of ERK2 post-natally had more pronounced effects in these behaviors. Loss of ERK2 in both models showed intact LTP compared to control animals, while loss of both ERK1 and ERK2 impaired late phase LTP. These findings indicate that ERK2 is not necessary for LTP and spatial memory retention and provide new insights into the functional deficits associated with the chronic impairment of ERK signaling.

Keywords: ERK signaling; LTP; MAPK; NCFC syndromes; spatial memory.

FIGURE 1. ERK2 is absent in the entire hippocampus of CKO^Emx mice and primarily CA1 of CKO^CamK mice

(A) Immunostaining of phospho-ERK1/2(green) in the hippocampus of WT, CKO^Emx and CKO^CamK mice at P90 following theta-burst stimulation of CA1. Bottom row magnification of boxed areas showing CA1 region of the hippocampus in the WT, CKO^Emx and CKO^CamK mice. sp=stratum pyramidale, sr=stratum radiatum. (B) Western blot of hippocampal tissue from WT, CKO^Emx and CKO^CamK mice at P90. (C–F) Densitometric analysis of relative protein levels of (C) total ERK1, (D) ERK1 activity (phospho-ERK1 normalized to ERK1), (E,F) total ERK2 protein at P10 (E) and P90 (F) and (G) ERK2 activity (phospho-ERK2 normalized to ERK2 levels) in WT (n=16), CKO^Emx (n=8), and CKO^CamK (n=6). *p<0.05, **p<0.01, ***p<0.001 using t-test.

FIGURE 2. ERK2 CKO^Emx and CKO^CamK mice exhibit normal spine densities of apical dendrites in CA1

(A–D) Representative image of apical dendrite with analysis of (A) spine density, (B) spine surface area, (C) spine length and (D) spine volume in apical dendrites of CA1 pyramidal neurons and dentate gyrus granule cells in P90 WT (n=5) and CKO^Emx (n=5) mice. (F–I) Analysis of (B) spine density, (C) spine surface area, (D) spine length and (E) spine volume in apical dendrites of CA1 pyramidal neurons in P90 WT (n=4) and CKO^CamK (n=4) mice. (J–N) Relative protein levels of (J) PSD95, (K) GRIA2, (L) GRIA1, (M) GRIN2A and (N) GRIN2B in the P90 hippocampus of WT (n=9), CKO^Emx (n=6), CKO^CamK (n=5) using densitometry analysis of western blots.

FIGURE 3. Behavioral assessment of spatial memory, object recognition memory and anxiety in ERK2 CKO mice

(A–F) Morris Water Maze: (A) WT, CKO^Emx and CKO^CamK mice were assessed for the number of trials where they remained immobile for the entire trial indicative of passive coping. **p<0.01 (B) Evaluation of training during the Morris water maze by measuring escape latencies against 8 trial blocks conducted over 4 days. p-values noted are from t-tests of CKO animals compared to WT. (C–D) Time spent in each quadrant of the arena during a probe trial on day 4 (Probe) and on day 5 (24hr ret) to assess for memory retention. *p<0.05, **p<0.01, ***p<0.001 from post-hoc Bonferroni test. Probe trial on day 4 (probe) and day 5 memory retention (24hr ret) were measured by looking at (E) platform location crossings and (F) target quadrant duration. (G–H) Novel Object Recognition: (G) Novel object preference was assessed in P90 WT, CKO^Emx and CKO^CamK mice. t-test, *p<0.05 (compared to a hypothetical value of 0 or no preference), #p<0.05 (compared to WT). (H) Object exploration during the habituation phase was analyzed in mice as well showing elevated object exploration in CKO mice. **p<0.01 using t-test. (I–K) Elevated Plus Maze: (I) Total arm entrances made by WT and CKOCamK mice during the duration of the test. (J) Time spent in the open arms of the maze and (K) time spent in the closed arms of the maze. *p<0.05 using t-test.

FIGURE 4. LTP is intact in mice lacking ERK2 in the hippocampus

(A) Input/Output curves of CA3 to CA1 connection in WT (n=8), CKO^Emx (n=5), and CKO^CamK (n=9). (B) Paired pulse facilitation in WT (n=7), CKO^Emx (n=5) and CKO^CamK (n=5) mice at different inter-pulse intervals. (C) CA1 long-term potentiation (LTP) following a 4 train TBS stimulation after 20 minutes of baseline recordings from WT (n=12: 6males, 6females), CKO^Emx (n=5: 4 males, 1female) and CKO^CamK (n=6: 6 males) hippocampal slices. *p<0.05 (Genotype effect from 2-way ANOVA).

FIGURE 5. LTP is impaired in mice lacking ERK1 and ERK2 in the hippocampus

(A) Cresyl Violet staining of hippocampus in P90 WT and DKO^CamK mice. (B) Immunostaining of phospho-ERK1/2(green) in the hippocampus of WT and DKO^CamK mice at P90 following theta-burst stimulation of CA1. Bottom row magnification of boxed areas showing CA1 region of the hippocampus in the WT and DKO^CamK mice. sp=stratum pyramidale, sr=stratum radiatum. (C) Western blot of hippocampal tissue from WT and DKO^CamK mice at P90. (D,E) Densitometric analysis of relative protein levels of (C) total ERK1, (D) total ERK2 protein in WT and DKO^CamK mice. *p<0.05, ***p<0.001 using t-test. (F) Input/Output curves of CA3 to CA1 connection in WT (n=12) and DKO^CamK (n=5) mice. *p<0.05, ***p<0.001 from Bonferroni post-hoc test (significant genotype effect in two-way ANOVA). (G) Paired pulse facilitation in WT (n=7) and DKO^CamK (n=7) mice at different inter-pulse intervals. (H) CA1 long-term potentiation (LTP) following a 4 train TBS stimulation after 20 minutes of baseline recordings from WT (n=12: 6males, 6females) and DKO^CamK (n=5: 2males, 4females) hippocampal slices. DKO^CamK. *p<0.05 using one way ANOVA of minutes 84–90 (bar).