The FASEB Journal is now also available on Wiley Online Library.
FJ Express SummariesFull Access

Prenatal choline supplementation advances hippocampal development and enhances MAPK and CREB activation

  • ,
  • ,
  • , and
    Published Online:8 Jan 2004https://doi.org/10.1096/fj.03-0877fje

    Abstract

    SPECIFIC AIMS

    Choline is an essential nutrient for animals and humans. Previous studies showed that supplementing maternal diet with choline during the second half of gestation in rats permanently enhances memory performance of the adult offspring. This study investigated the hypothesis that the activity of the hippocampal mitogen-activated protein kinase (MAPK)/cAMP-response element binding protein (CREB) pathway mediates the enhanced cognitive performance observed in prenatally choline-supplemented rats. We examined the ability of juvenile rats to use relational cues in a water maze task and the level of phosphorylation of both MAPK and CREB in hippocampal slices in response to a depolarizing concentration of potassium or stimulation by glutamate or N-methyl-D-aspartate (NMDA) in prenatally choline-supplemented, choline-deficient, and control rats.

    PRINCIPAL FINDINGS

    1. Prenatal choline-supplementation enhances the ability of juvenile rats to use relational cues in a water maze task

    The ability to navigate using relational cues is considered a developmental milestone signaling the onset of hippocampal function. We assessed both spatial (relational) and cued navigational abilities of 18-22-day-old choline-supplemented and control rats. During spatial training at postnatal days (P) 18–19, prenatally choline-supplemented rats were able to use relational cues to remember the platform location during the first 5 trials (i.e., they significantly shortened their escape latencies with successive trials), while control rats showed no spatial learning under this condition. When these rats were re-tested 3 days later on the same task, both choline and control rats shortened their escape latencies over the first 5 (acquisition) trials (data not shown). In contrast, at 18–19 days of age, both control and choline-supplemented rats could learn the location of a platform that was directly cued. Thus, prenatal choline-supplementation causes an advancement, by ∼3 days, in spatial (relational) navigational abilities.

    2. Prenatal choline-supplementation enhances K+-stimulated MAPK phosphorylation

    MAPK phosphorylation was stimulated by depolarization of the slices. This effect depended on age and prenatal availability of choline. In slices from control animals, depolarization increased MAPK phosphorylation twofold at P18, and this effect increased, reaching fourfold at P25 (Fig. 1⤻ A). Prenatal choline supplementation tended to cause a small increase in basal levels of MAPK phosphorylation that reached statistical significance only on P20 as compared with control animals (Fig. 1A⤻ ). Moreover, prenatal choline supplementation significantly up-regulated depolarization-evoked MAPK phosphorylation as compared with that observed in prenatally choline deficient rats at all ages examined. Prenatal choline deficiency did not significantly affect basal levels of MAPK phosphorylation, but it severely impaired depolarization-evoked MAPK phosphorylation (Fig. 1B,C⤻ ). When plotted as function of the amount of choline in the diet, MAPK phosphorylation shows a choline dose-dependence in response to K+-stimulation at all time points (P18 and P22 shown in Fig. 1D, E⤻ , respectively). Similarly, the phosphorylation of MAPK in response to glutamate and NMDA stimulation was dependent on the amount of choline consumed by the mother during pregnancy in a dose-dependent fashion (Fig. 1D, E)⤻ .

    Figure 1.

    Figure 1. (A–C) K+-stimulated MAPK phosphorylation is enhanced by prenatal choline-supplementation. During days 11–17 of pregnancy, rats were fed a deficient, control, or supplemented diet (0, 8, or 36 mmol choline/kg of diet, respectively). Hippocampal slices from offspring of postnatal age (P) 18, 20, 22, and 25 days were incubated with basal medium or a depolarizing concentration of KCl (40 mM) for 10 min. Western blotting was performed with an antibody specific to the Thr202/Tyr204 phosphorylated form of MAPK and quantified. Data are means ± se (n=6). The entire dataset is presented on three panels (A–C) to facilitate viewing of the relevant comparisons. A 3-way ANOVA with choline treatment, [K+] and age as factors revealed significant effects of all factors (P<0.0001 for each). There were significant interactions between age and K+ (P<0.0005), and choline and K+ (P<0.05). K+-induced phosphorylation of MAPK was lower in prenatally choline-deficient rats than in choline-supplemented animals (P<0.05). (D, E) Dose Response. Slices from P18 and P22 rats were incubated with either 10 μM glutamate, 100 μM NMDA, or 40 mM KCl for 10 min. There was a significant effect of choline on P18 (P<0.005) and P22 (P<0.001). Glutamate-induced phosphorylation of MAPK was lower in prenatally choline-deficient rats than in choline-supplemented animals on P18 (P<0.05). NMDA increased MAPK phosphorylation on P22 only (P<0.001). There was an interaction between choline availability and NMDA treatment (P<0.05) on P22. NMDA-stimulated MAPK phosphorylation was higher in prenatally choline-supplement rats than in choline-deficient rats on P22 (P<0.005). K+-induced MAPK phosphorylation was higher in prenatally choline-supplemented rats than in choline-deficient animals on both days (P<0.05 for each) and there was a interaction between K+ and choline on P22 (P<0.05).

    Download figure Download PowerPoint

    3. Prenatal choline-supplementation enhances K+-stimulated CREB phosphorylation

    Prenatal choline supplementation caused a small but statistically significant increase in basal levels of CREB phosphorylation as compared with control animals at younger ages (P18, 20, 22) (Fig. 2⤻ A). In control rats, K+ simulation produced a 50% increase in CREB phosphorylation above resting levels in P18 rats that rose to a 75% increase in P25 rats (Fig. 2A, B⤻ ). K+-stimulated CREB phosphorylation in prenatally choline-supplemented rats was significantly and consistently higher than control (1.8-fold) and prenatally choline-deficient animals (2.1-fold) (Fig. 2A, C)⤻ . Prenatally choline-deficient rats showed little to no CREB activation in response to glutamate, NMDA, or a depolarizing concentration of KCl (Fig. 2)⤻ . Similar to MAPK phosphorylation, the level of phosphorylation of CREB responded in a dose-dependent manner to the amount of choline available in utero (Fig. 2D, E⤻ ). The data show that the prenatal supply of choline influences development and magnitude of phosphorylation patterns of MAPK and CREB, two proteins critical to learning and memory.

    Figure 2.

    Figure 2. (A–C)Prenatal choline-supplementation increases K+-stimulated CREB phosphorylation. The experiments were performed as described in Fig. 1⤻ and phospho-CREB was analyzed with an antibody specific to the Serine-133 phosphorylated form of CREB. There was an overall effect of choline (P<0.0001) and K+ (P<0.0001), and an interaction between choline and K+ (P<0.005). There was no effect of age (P=0.65). K+-induced phosphorylation of CREB was higher in prenatally choline-supplemented rats than in control and choline-deficient animals (P<0.01 and P<0.005, respectively). There was no difference between the K+-stimulated CREB phosphorylation of control and prenatally choline-deficient rats (P=0.84). (D, E) Dose Response. Overall, choline availability increased CREB phosphorylation on P18 and P22 (P<0.00001 and P<0.0005, respectively). On P18 there was a significant effect of glutamate treatment (P<0.001) and a significant interaction between glutamate treatment and choline availability (P<0.05). Glutamate-induced CREB phosphorylation was higher in prenatally choline-supplemented rats than in choline-deficient and control animals on P18 only (P<0.01 and P<0.0005, respectively). On P18, there was a significant effect of NMDA treatment (P<0.05) and there was an interaction between choline availability and NMDA treatment (P<0.05). Prenatally choline-supplemented rats had a higher level of NMDA-stimulated CREB phosphorylation than control and choline-deficient rats on P18 (P<0.05 and P<0.005, respectively). At both ages, there was a significant effect of K+ treatment (P<0.005 for each) on CREB phosphorylation. K+-induced CREB phosphorylation was higher in prenatally choline-supplemented rats than in control (P<0.005 on P22) and choline-deficient animals (P<0.05 and P<0.005, P18 and P22 respectively).

    Download figure Download PowerPoint

    CONCLUSIONS AND SIGNIFICANCE

    Our behavioral data demonstrate that supplementation with choline during prenatal development advances the onset of navigational ability using spatial (relational) cues – a skill that requires an intact, functioning hippocampus. At 18–19 days of age, choline-supplemented rats showed evidence of a precocious capacity for spatial navigation, whereas control rats required an additional 3 days of maturation to acquire this skill. Although we have not yet determined if prenatal choline deficiency during a limited time frame of development actually slows maturation of hippocampal function, it is known that early life malnutrition slows attainment of hippocampally-dependent spatial navigation. Because both choline-supplemented and control rats were able to learn the location of the hidden platform marked directly by a cue at 18–19 days of age, the performance differences observed between the two groups of animals were unlikely due to differences in the maturation of visual or swimming ability. Similarly, early life malnutrition does not interfere with cued navigation. Once spatial navigational ability is acquired, prenatally choline supplemented and control rats learn at a similar rate. As adults and into old age, prenatally choline-supplemented rats outperform control rats on tasks that assess memory capacity and precision, indicating that in addition to the advancement in hippocampal development, prenatal choline treatment causes permanent organizational changes in the brain.

    Prenatally choline-supplemented rats displayed enhanced phosphorylation of both MAPK and CREB following glutamate, NMDA, or depolarization-evoked stimulation. In contrast, MAPK and CREB phosphorylation was diminished in hippocampal slices from prenatally choline-deficient rats. Treatment with a depolarizing concentration of KCl revealed the most dramatic differences and the highest overall level of stimulation-evoked phosphorylation of MAPK and CREB, possibly due to the fact that depolarization with K+ can induce neurotransmitter release from the pre-synaptic neuron, leading to activation of post-synaptic receptors. This treatment can also initiate calcium-activated signaling events within the post-synaptic cell via the opening of voltage-gated calcium channels, whereas glutamate and NMDA target only post-synaptic receptors (Fig. 3⤻ ). Because we found no effect of prenatal choline status on the levels of total (i.e., the sum of non-phosphorylated and phosphorylated) MAPK and CREB protein, differences in stimulated levels of MAPK and CREB phosphorylation among the three treatment groups of animals are most likely due to altered neuronal excitability or changes in signaling pathways that lead to protein phosphorylation, such as amount of neurotransmitter release from pre-synaptic neurons or the number and/or activity of post-synaptic receptors. The latter possibility is supported by other data that NMDA receptor-mediated population excitatory postsynaptic potentials were elevated in slices from prenatally choline-supplemented rats relative to control animals suggesting that prenatal choline supplementation enhances NMDA receptor-mediated neurotransmission.

    Figure 3.

    Figure 3. Schematic diagram. K+ depolarization causes glutamate release from a presynaptic neuron and causes Ca2+ channels on a postsynaptic neuron to open and allow Ca2+ into the cell. Glutamate can activate NMDA receptors to increase the level of intracellular Ca2+, which directly or indirectly (dashed arrows) activates Ca2+-regulated protein kinases (CAMK), protein kinase A (PKA), and protein kinase C (PKC). These kinases directly or indirectly activate MAPK and CREB via phosphorylation. Availability of choline prenatally affects levels of phosphorylation of both MAPK and CREB.

    Download figure Download PowerPoint

    Our observations that prenatal choline supplementation increased basal phosphorylation of MAPK and CREB in hippocampal slices and enhanced stimulation-evoked phosphorylation of these proteins may help explain the facilitated long-term potentiation (LTP) induction seen in prenatally choline supplemented animals. Moreover, the impairment of MAPK and CREB activation in prenatally choline-deficient rats correlates well with the reported high stimulus threshold for LTP induction, and frequent failure to induce any LTP in hippocampal slices from prenatally choline-deficient rats.

    Our findings are also consistent with the notion that CREB activation serves as a molecular switch in the biochemical pathways leading to long-term storage of memories. Opposing genetic manipulations of CREB produce opposing effects on long-term memory (LTM) and the amount of phosphorylated CREB and its transcriptional activity correlate with LTM. In rodents subjected to hippocampally-dependent associative learning tasks, increases in the levels of phosphorylated CREB and in CRE-mediated gene expression are observed in the hippocampus.

    These data show that supplementation with the essential nutrient choline during the prenatal period causes a reorganization of hippocampal function in association with up-regulation of the MAPK/CREB signaling cascade, a pathway central for memory processing. Thus, the basic biochemical mechanisms of memory are subject to long-lasting modulation by the availability of a single nutrient during critical periods in development.

    1To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-0877fje;

    back