Spatial memory impairment and changes in hippocampal morphology are triggered by high-fat diets in adolescent mice. Is there a role of leptin?

Recent evidence has established that consumption of high-fat diets (HFD) is associated with deficits in hippocampus-dependent memory. Adolescence is an important period for shaping learning and memory acquisition that could be particularly sensitive to the detrimental effects of HFD. In the current study we have administered this kind of diets to both adolescent (5-week old) and young adult (8-week old) male C57BL mice during 8 weeks and we have evaluated its effect on (i) spatial memory performance in the novel location recognition (NLR) paradigm, and (ii) spine density and neural cell adhesion molecule (NCAM) expression in hippocampal CA1 pyramidal neurons. In order to characterize the eventual involvement of central leptin receptors we have also investigated the functionality of leptin receptors within the hippocampus. Here we report that animals that started to consume HFD during the adolescence were less efficient than their control counterparts in performing spatial memory tasks. In contrast to that, mice that were submitted to HFD during the young adult period displayed intact performance in the NLR test. In mice receiving HFD from the adolescence, the behavioral impairment was accompanied by an increase of dendritic spine density in CA1 pyramidal neurons that correlated with the up-regulation of neural cell adhesion molecule (NCAM) in this area. Deficits in spatial memory occurred concomitantly with a desensitization of the proteinkinase B (Akt) pathway coupled to hippocampal leptin receptors. In contrast, the STAT3 pathway remained unaffected by HFD. All effects of HFD were long-lasting because they remained intact even after 5 weeks of food restriction. Our results provide further evidence of the susceptibility of the hippocampus to HFD in adolescent individuals and suggest that leptin signaling integrity in this brain area is pivotal for memory performance.

Periadolescent amphetamine treatment causes transient cognitive disruptions and long-term changes in hippocampal LTP depending on the endogenous expression of pleiotrophin.

Amphetamine treatment during adolescence causes long-term cognitive deficits in rats. Pleiotrophin (PTN) is a cytokine with important roles in the modulation of synaptic plasticity, whose levels of expression are significantly regulated by amphetamine administration. To test the possibility that the long-term consequences of periadolescent amphetamine treatment cross species and, furthermore, to test the hypothesis that PTN could be one of the factors involved in the adult cognitive deficits observed after periadolescent amphetamine administrations, we comparatively studied the long-term consequences of periadolescent amphetamine treatment (3 mg/kg intraperitoneal, daily during 10 days) in normal wild-type (PTN+/+) and in PTN genetically deficient (PTN-/-) mice. Within the first week after cessation of treatment, significant deficits in the passive avoidance and Y-maze tests were only observed in amphetamine-pretreated PTN-/- mice. However, 13 and 26 days after the last administration, we did not find significant differences in Y-maze between amphetamine- and saline-pretreated PTN-/- mice. In addition, we did not find any genotype- or treatment-related anxiogenic- or depressive-like behaviour in adult mice. Furthermore, we observed a significantly enhanced long-term potentiation (LTP) in CA1 hippocampal slices from saline-pretreated PTN-/- mice compared with saline-pretreated PTN+/+ mice. Interestingly, amphetamine pre-treatment during adolescence significantly enhanced LTP in adult PTN+/+ mice but did not cause any effect in PTN-/- mice, suggesting LTP mechanisms saturation in naïve PTN-/- mice. The data demonstrate that periadolescent amphetamine treatment causes transient cognitive deficits and long-term alterations of hippocampal LTP depending on the endogenous expression of PTN.

High-fat diets induce changes in hippocampal glutamate metabolism and neurotransmission.

Obesity and high-fat (HF) diets have a deleterious impact on hippocampal function and lead to impaired synaptic plasticity and learning deficits. Because all of these processes need an adequate glutamatergic transmission, we have hypothesized that nutritional imbalance triggered by these diets might eventually concern glutamate (Glu) neural pathways within the hippocampus. Glu is withdrawn from excitatory synapses by specific uptake mechanisms involving neuronal (EAAT-3) and glial (GLT-1, GLAST) transporters, which regulate the time that synaptically released Glu remains in the extracellular space and, consequently, the duration and location of postsynaptic receptor activation. The goal of the present study was to evaluate in mouse hippocampus the effect of a short-term high-fat dietary treatment on 1) Glu uptake kinetics, 2) the density of Glu carriers and Glu-degrading enzymes, 3) the density of Glu receptor subunits, and 4) synaptic transmission and plasticity. Here, we show that HF diet triggers a 50% decrease of the Michaelis-Menten constant together with a 300% increase of the maximal velocity of the uptake process. Glial Glu carriers GLT-1 and GLAST were upregulated in HF mice (32 and 27%, respectively), whereas Glu-degrading enzymes glutamine synthase and GABA-decarboxilase appeared to be downregulated in these animals. In addition, HF diet hippocampus displayed diminished basal synaptic transmission and hindered NMDA-induced long-term depression (NMDA-LTD). This was coincident with a reduced density of the NR2B subunit of NMDA receptors. All of these results are compatible with the development of leptin resistance within the hippocampus. Our data show that HF diets upregulate mechanisms involved in Glu clearance and simultaneously impair Glu metabolism. Neurochemical changes occur concomitantly with impaired basal synaptic transmission and reduced NMDA-LTD. Taken together, our results suggest that HF diets trigger neurochemical changes, leading to a desensitization of NMDA receptors within the hippocampus, which might account for cognitive deficits.

Effects of chronic cocaine administration on spatial learning and hippocampal spine density in two genetically different strains of rats.

Lewis and Fischer-344 rats have been proposed as an addiction model because of their differences in addiction behaviour. It has been suggested that drug addiction is related to learning and memory processes and depends on individual genetic background. We have evaluated learning performance using the eight-arm radial maze (RAM) in Lewis and Fischer-344 adult rats undergoing a chronic treatment with cocaine. In order to study whether morphological alterations were involved in the possible changes in learning after chronic cocaine treatment, we counted the spine density in hippocampal CA1 neurons from animals after the RAM protocol. Our results showed that Fischer-344 rats significantly took more time to carry out test acquisition and made a greater number of errors than Lewis animals. Nevertheless, cocaine treatment did not induce changes in learning and memory processes in both strains of rats. These facts indicate that there are genetic differences in spatial learning and memory that are not modified by the chronic treatment with cocaine. Moreover, hippocampal spine density is cocaine-modulated in both strains of rats. In conclusion, cocaine induces similar changes in hippocampal neurons morphology that are not related to genetic differences in spatial learning in the RAM protocol used here.

Lewis and Fischer 344 rats as a model for genetic differences in spatial learning and memory: Cocaine effects.

Lewis (LEW) and Fischer 344 (F344) rats are considered a model of genetic vulnerability to drug addiction. We previously showed important differences in spatial learning and memory between them, but in contrast with previous experiments demonstrating cocaine-induced enhanced learning in Morris water maze (MWM) highly demanding tasks, the eight-arm radial maze (RAM) performance was not modified either in LEW or F344 rats after chronic cocaine treatment. In the present work, chronically cocaine-treated LEW and F344 adult rats have been evaluated in learning and memory performance using the Y-maze, two RAM protocols that differ in difficulty, and a reversal protocol that tests cognitive flexibility. After one of the RAM protocols, we quantified dendritic spine density in hippocampal CA1 neurons and compared it to animals treated with cocaine but not submitted to RAM. LEW cocaine treated rats showed a better performance in the Y maze than their saline counterparts, an effect that was not evident in the F344 strain. F344 rats significantly took more time to learn the RAM task and made a greater number of errors than LEW animals in both protocols tested, whereas cocaine treatment induced deleterious effects in learning and memory in the highly difficult protocol. Moreover, hippocampal spine density was cocaine-modulated in LEW animals whereas no effects were found in F344 rats. We propose that differences in addictive-like behavior between LEW and F344 rats could be related to differences in hippocampal learning and memory processes that could be on the basis of individual vulnerability to cocaine addiction.

MDMA enhances hippocampal-dependent learning and memory under restrictive conditions, and modifies hippocampal spine density.

OBJECTIVES: Addictive drugs produce forms of structural plasticity in the nucleus accumbens and prefrontal cortex. The aim of this study was to investigate the impact of chronic MDMA exposure on pyramidal neurons in the CA1 region of hippocampus and drug-related spatial learning and memory changes. METHODS AND RESULTS: Adolescent rats were exposed to saline or MDMA in a regime that mimicked chronic administration. One week later, when acquisition or reference memory was evaluated in a standard Morris water maze (MWM), no differences were obtained between groups. However, MDMA-exposed animals performed better when the MWM was implemented under more difficult conditions. Animals of MDMA group were less anxious and were more prepared to take risks, as in the open field test they ventured more frequently into the central area. We have demonstrated that MDMA caused an increase in brain-derived neurotrophic factor (BDNF) expression. When spine density was evaluated, MDMA-treated rats presented a reduced density when compared with saline, but overall, training increased the total number of spines, concluding that in MDMA-group, training prevented a reduction in spine density or induced its recovery. CONCLUSIONS: This study provides support for the conclusion that binge administration of MDMA, known to be associated to neurotoxic damage of hippocampal serotonergic terminals, increases BDNF expression and stimulates synaptic plasticity when associated with training. In these conditions, adolescent rats perform better in a more difficult water maze task under restricted conditions of learning and memory. The effect on this task could be modulated by other behavioural changes provoked by MDMA.

Shift of circadian feeding pattern by high-fat diets is coincident with reward deficits in obese mice.

Recent studies provide evidence that high-fat diets (HF) trigger both i) a deficit of reward responses linked to a decrease of mesolimbic dopaminergic activity, and ii) a disorganization of circadian feeding behavior that switch from a structured meal-based schedule to a continuous snacking, even during periods normally devoted to rest. This feeding pattern has been shown to be a cause of HF-induced overweight and obesity. Our hypothesis deals with the eventual link between the rewarding properties of food and the circadian distribution of meals. We have investigated the effect of circadian feeding pattern on reward circuits by means of the conditioned-place preference (CPP) paradigm and we have characterized the rewarding properties of natural (food) and artificial (cocaine) reinforcers both in free-feeding ad libitum HF mice and in HF animals submitted to a re-organized feeding schedule based on the standard feeding behavior displayed by mice feeding normal chow ("forced synchronization"). We demonstrate that i) ad libitum HF diet attenuates cocaine and food reward in the CPP protocol, and ii) forced synchronization of feeding prevents this reward deficit. Our study provides further evidence that the rewarding impact of food with low palatability is diminished in mice exposed to a high-fat diet and strongly suggest that the decreased sensitivity to chow as a positive reinforcer triggers a disorganized feeding pattern which might account for metabolic disorders leading to obesity.