Search

There is growing evidence that lipids play a fundamental role in neuronal plasticity and learning and memory. Effects of nutrition on brain lipid composition and neuronal functioning are known, but the feeding interventions are often severe and may not reflect nutritional effects below clinical relevance. Therefore, we tested two commercially available rat feeding diets with only moderate differences in the food compositions, a standard diet (gross energy metabolizable 12.8 MJ/kg) and a energy reduced diet (gross energy metabolizable 8.9 MJ/kg) on possible effects upon dentate gyrus lipid composition, spatial learning and memory in a water maze and corticosterone release (blood serum concentrations) in adult male rats. Rats were fed with the standard diet up to an age of 8 weeks. One group was further fed with the standard and another with the energy reduced diet until an age of 5 months. We did not found differences in serum corticosterone levels. We found group differences in a variety of lipids in the hippocampal dentate gyrus.. Most of the lipid levels were lower in energy reduced diets, namely glycerophosphoethanolamines, sphingomyelins and hexosyceramides, whereas some ceramides (Cer18:0 and Cer24:1) and glycerophosphocholines (PC34:3 and PC36:2) were upregulated compared to the standard diet group. The performance in a common reference memory water maze task was not different between groups, however during reversal learning (platform in a different position) after the initial training, the standard diet fed rats learned better and spatial memory was improved compared to the energy reduced diet group. Thus, moderate differences in feeding diets have effects specifically upon spatial cognitive flexibility. Possible relations between differences in lipid composition and cognitive flexibility are discussed.

Cognitive processes require striatal activity. The underlying molecular mechanisms are widely unknown. For this reason the striatal transcriptome of young (YM), aged cognitively impaired (OMB), and unimpaired (OMG) male rats was analyzed. The global comparison of transcripts reveal a higher number of differences between OMG and YM as compared to OMB and YM. Hierarchical clustering detects differences in up- and down-regulated gene clusters in OMG and OMB when compared to YM. In OMG we found more single genes to be specifically regulated in this group than in OMB when compared to young. These genes were considered as cognition specific, whereas genes shared in OMG and OMB were considered as age specific. OMB specific up-regulated genes are related to negative control of cell differentiation and transcription (Hopx), to phagocytosis (Cd202) and cell adhesion (Pcdhb21), whereas down-regulated genes are related to associative learning, behavioral fear response and synaptic transmission (Gabra5). OMG specific up-regulated genes are in the context of maintenance of transcription and estrogen receptor signaling (Padi2, Anxa3), signal transduction [Rassf4, Dock8)], sterol regulation (Srebf1), and complement activity (C4a, C4b). Down-regulated genes are related to lipid oxidation reduction processes (Far2) and positive regulation of axon extension (Islr2). These relations were supported by pathway analysis, which reveals cholesterol metabolism processes in both aged group and cholesterol biosynthesis specifically in OMG; adipogenesis and focal adhesion in OMB. In OMG glucuronidation, estrogen metabolism, inflammatory responses and TGF beta signaling where detected as specific for this group. Signal transduction of the sphingosine-1-phospate-receptor (S1P) receptor was the main pathway difference in the comparison of OMB and OMG with downregulated genes in the first group. This difference could also be observed in the OMB vs. YM comparison but not in the OMG vs. YM analysis. Thus, an up-regulation of cognition related genes could be observed in OMG compared to OMB rats. The S1P pathway discriminated between OMB and OMG as well as between OMB and OMG. Since this pathway has been described as essential for cognitive processes in the striatum of mice, it may, among steroid hormone signaling, significantly contribute to the maintenance of cognitive processes in OMG.

Nutrition can have significant effects on behavior and cognitive processes. Most of the studies related to this use extremely modified diets, such as high fat contents or the exclusion of distinct components needed for normal development and bodily homeostasis. Here we report significant effects of diets with moderate differences in compositions on food rewarded spatial learning in young (3-4 months), adult (6-7 months), and aged (17-18 months) rats. Young rats fed with a lower energy diet showed better performance only during aquisition of the spatial task when compared to rats fed with a standard diet. Adult rats (6-7 months) fed with a standard diet performed less well in the spatial learning task, than rats fed with lower energy diet. Aged rats fed with a lower energy diet (from 13 to 18 months of age) performed better during all training phases, as in a previous test when they were adult and fed with a standard diet. This difference could only be partly explained by lower motivation to search for food in the first test. Correspondingly, the variability of individual performance was significantly higher and increased over trials in adult rats fed with the standard diet as compared to adult rats fed with lower energy diet. Thus, moderate changes in feeding diets have large effects on motivation and cognition in elderly and less in young rats in a food rewarded spatial learning task. Therefore, nutrition effects upon food rewarded spatial learning and memory should be considered especially in aging studies.

The dopaminergic system is known to be involved in working memory processed by several brain regions like prefrontal cortex (PFC), hippocampus, striatum. In an earlier study we could show that Levodopa but not Modafinil enhanced working memory in a T-maze only during the early phase of training (day 3), whereas the later phase remained unaffected. Rats treated with a higher dose performed better than low dose treated rats. Here we could more specifically segregate the contributions of dopamine type 1- and 2- like receptors (D1R; D2R) to the training state dependent modulation of spatial working memory by intracerebroventricular (ICV) application of a D1R-like (SKF81297) and D2R-like agonist (Sumanirole) and antagonist (SCH23390, Remoxipride) at a low and high dose through 3 days of training. The D1R-like-agonist at both doses enhanced working memory at day 1 but only in the low dose treated rats enhancement persists over training compared to control rats. Rats treated with a high dose of a D1R-like-antagonist show persistent enhancement of working memory over training, whereas in low dose treated rats no statistical difference at any time point could be determined compared to controls. The D2R-like-agonist at both doses does not show an effect at any time point when compared to control animals, whereas the D2R-like antagonist at a low dose enhanced working memory at day 2. For the most effective D1R-like agonist, we repeated the experiments in a water maze working memory task, to test for task dependent differences in working memory modulations. Treated rats at both doses did not differ as compared to controls, but the temporal behavioral performance of all groups was different compared to T-maze trained rats. The results are in line with the view that spatial working memory is optimized within a limited range of dopaminergic transmission, however suggest that these ranges vary during spatial training.