Prenatal ethanol exposure persistently impairs NMDA receptor-dependent activation of extracellular signal-regulated kinase in the mouse dentate gyrus

Abstract

The dentate gyrus (DG) is the central input region to the hippocampus and is known to play an important role in learning and memory. Previous studies have shown that prenatal alcohol is associated with hippocampal-dependent learning deficits and a decreased ability to elicit long-term potentiation (LTP) in the DG in adult animals. Given that activation of the extracellular signal-regulated kinase 1/2 (ERK1/2) signaling cascade by NMDA receptors is required for various forms of learning and memory, as well as LTP, in hippocampal regions, including the DG, we hypothesized that fetal alcohol-exposed adult animals would have deficits in hippocampal NMDA receptor-dependent ERK1/2 activation. We used immunoblotting and immunohistochemistry techniques to detect NMDA-stimulated ERK1/2 activation in acute hippocampal slices prepared from adult fetal alcohol-exposed mice. We present the first evidence linking prenatal alcohol exposure to deficits in NMDA receptor-dependent ERK1/2 activation specifically in the DG of adult offspring. This deficit may account for the LTP deficits previously observed in the DG, as well as the life-long cognitive deficits, associated with prenatal alcohol exposure.

Fig. 1

NMDA receptor-dependent activation of ERK1/2 mediated by PKA and PKC. Synaptic NMDA receptors can induce ERK1/2 activation via increases in intracellular calcium and subsequent activation of calcium/calmodulin dependent adenylyl cyclase (AC) which produces cAMP and activates PKA, or intracellular calcium can activate calcium-dependent protein kinase C (PKC). PKA activates ERK1/2 via Rap1-B-Raf-MEK and PKC activates ERK1/2 via Ras-Raf1-MEK (see text). PKA can inhibit Raf-1 and potentially hinder PKC-mediated ERK1/2 activation. Extrasynaptic NMDA receptors do not activate ERK1/2; rather, they inhibit ERK1/2 via activation of calcium-dependent phosphatases, possibly striatal-enriched protein tyrosine phosphatase (STEP).

Fig. 2

NMDA-dependent activation of ERK2 in saccharin control (Sacc) and fetal alcohol-exposed (FAE) adult offspring hippocampal slices, n=8. (A) Representative phospho-ERK1/2 (pERK1/2; top) and total ERK1/2 (bottom) Western blots for both Sacc and FAE hippocampal slices treated with 0, 25, 50 and 100 μM NMDA. (B) Levels of pERK2 in Sacc and FAE samples at various concentrations of NMDA. Two-way ANOVA of the pERK2 data showed a significant effect of prenatal diet (p<0.02). (C) Total ERK2 levels in Sacc and FAE hippocampal slices with increasing NMDA concentration. There were no significant differences between groups or with NMDA treatment on total ERK2 levels. (D) ERK2 activation state expressed as a ratio of pERK2 to total ERK2 in Sacc and FAE hippocampal slices treated with 0, 25, 50 and 100 μM NMDA. Two-way ANOVA revealed significant main effects of prenatal diet (p=0.010) and NMDA (p<0.01), but no significant interaction. All levels of pERK2 and total ERK2 are expressed as optical densities normalized to a hippocampal standard preparation loaded in all gels in order to compare across gels. Sacc and FAE samples were run on the same gel in order to compare between the groups and blots were done in duplicate. Total ERK2 immunoreactivity was evaluated on all blots and used to generate pERK2/total ERK2 ratio for each sample.

Fig. 3

ERK2 activation in saccharin control (Sacc) and fetal alcohol-exposed (FAE) adult hippocampal slices treated with 100 μM NMDA in the presence and absence of MEK inhibitor U0126 or NMDA receptor inhibitors MK801 & APV. (A) Representative phospho-ERK1/2 (pERK1/2; top) and total ERK1/2 (bottom) Western blots for both Sacc and FAE hippocampal slices treated with 100 μM NMDA in presence and absence of MEK inhibitor 20 μM U0126. (B) Representative pERK1/2 (top) and total ERK1/2 (bottom) Western blots for both Sacc and FAE hippocampal slices treated with 100 μM NMDA in presence and absence of NMDA inhibitors 100 μM APV & 50 μM MK801. (C) ERK2 activation data expressed as a ratio of pERK2 to total ERK2 for Sacc and FAE slices treated, or untreated, with 100 μM NMDA (n=8) and pretreated, or not, with 50 μM MK801 / 100 μM APV (n=4), or with 20 μM U0126 (n=4). Two-way ANOVA revealed a significant effect of treatment (p<0.01). Post hoc analysis showed a significant increase in ERK2 activation in +NMDA compared to -NMDA Sacc slices (**p<0.01) and a significant difference between +NMDA versus +NMDA + MK801 / APV data in Sacc slices (*p<0.05). Comparison of the no NMDA control versus +NMDA and the +NMDA versus NMDA + MK801/APV in FAE slices revealed the difference was not significant (ns). U0126 pretreated slices were significantly different from untreated control slices in Sacc and FAE groups (# p<0.001). Data from +U0126 versus +NMDA + U0126 slices were not significantly different.

Fig. 4

ERK2 phosphorylation in response to treatment with phorbol 12,13 diacetate (PDA), n=6, or forskolin (FSK), n=8, in saccharin control (Sacc) and fetal alcohol exposed (FAE) adult hippocampal slices. (A) Representative phospho-ERK1/2 (pERK1/2; top) and total ERK1/2 (bottom) Western blots for both Sacc and FAE hippocampal slices treated without or with 3 μM PDA. (B) ERK2 data expressed as a ratio of pERK2 to total ERK2 in Sacc and FAE hippocampal slices treated without or with 3 μM PDA. Two-way ANOVA of the data revealed a significant effect of PDA treatment (p<0.01). Post hoc analysis showed a significant ERK2 activation upon PDA treatment in both groups, **p<0.01. (C) Representative phospho-ERK1/2 (pERK1/2) top and total ERK1/2 (bottom) Western blots for both Sacc and FAE hippocampal slices treated without or with 50 μM FSK. (D) ERK2 data expressed as a ratio of pERK2 to total ERK2 in Sacc and FAE hippocampal slices treated with 50 μM FSK. Two-way ANOVA of the data revealed a significant effect of FSK treatment ( p<0.01). Post hoc analysis showed a significant ERK2 activation upon FSK treatment in both groups, **p<0.01.

Fig. 5

NMDA-induced activation of ERK1/2 in re-sectioned hippocampal slices stimulated in vitro (scale bar = 300 μm). (A) Representative 4X images of Sacc and FAE slices showing anti-phospho-ERK1/2 immunoreacitivity (red) in 16 um hippocampal slices without NMDA stimulation (-NMDA) or after exposure to 100μM NMDA for 3 min (+NMDA). Regions indicated are CA1, dentate gyrus (DG) and CA3. NMDA-mediated ERK1/2 activation was shown to occur mainly in CA1 region for both groups. (B) No significant difference in area of ERK1/2 activation found between Sacc (n=5) and FAE (n=6) slides. DAPI (blue) was used to stain nuclei.

Fig. 6

NMDA induced activation of ERK1/2 in re-sectioned hippocampal slices stimulated in vitro. Representative 10X images of CA1, dentate gyrus (DG) and CA3 regions without NMDA (-NMDA) or after exposure to 100μM NMDA for 3 min (+NMDA) for Sacc (A) and FAE (B) groups (scale bar = 100 μm). The majority of ERK1/2 activation was found to occur in CA1 pyramidal dendrites within the stratum radiatum (s.r.) for both Sacc and FAE images. Stratum pyramidale (s.p.) and stratum oriens (s.o.) also shown. ERK1/2 activation in the DG region is depicted by white arrows on Sacc 10X image (A). DAPI (blue) was used to stain nuclei.

Fig. 7

ERK1/2 activation in CA1 (A) and dentate gyrus (DG) (B) for Sacc (n=5) and FAE (n=6) re-sectioned slices (-NMDA and +NMDA). (A) Two-way ANOVA of CA1 data showed a significant effect of NMDA treatment (p<0.0001). Post hoc analysis revealed a significant increase in both Sacc (***p<0.0001) and FAE (**p<0.01) groups upon NMDA stimulation in the CA1 region. (B) Two-way ANOVA of DG data showed a significant effect of NMDA treatment (p<0.05) and a significant effect of prenatal diet (p<0.05), however there was not a significant interaction. Post hoc analysis revealed a significant increase in ERK1/2 activation in the DG of Sacc slices (*p<0.05), but not in FAE slices.