Role of brain-derived neurotrophic factor and presynaptic proteins in passive avoidance learning in day-old domestic chicks.

The consolidation of a one-trial passive avoidance learning task in the day-old chick involves a number of transient and longer-term biochemical processes, including increased release of glutamate. This study demonstrated that brain-derived neurotrophic factor, a proposed modulator of synaptic transmission and neurotransmitter release, is involved in the cascade associated with memory consolidation in the chick and that its actions were linked to modulation of expression of SNAP-25, syntaxin and synaptophysin, required for exocytosis. Intracerebral injections of 5 microl of antibodies to brain-derived neurotrophic factor into the left and right intermediate medial hyperstriatum ventrale resulted in a dose-dependent reduction in avoidance of an "aversive" bead by 3 h after training. Neurotrophin antibodies (0.5 microg/chick) administered between 1 h before, and up to 30 min after, training induced amnesia by 3 h which was sustained for at least 24 h. Injections of recombinant brain-derived neurotrophic factor (50 microg/ml; 0.5 microg/chick) just before training maintained avoidance in birds trained with a weaker aversant (10% methylanthranilate), such that chicks showed enhanced recall at times (24 h) beyond that when shorter-term forms of memory have decayed. In lysed synaptosomal membranes prepared from chicks injected with antibodies to brain-derived neurotrophic factor there was a decrease in expression of SNAP-25 and syntaxin in the left, but not the right, intermediate medial hyperstriatum ventrale, a region known to be associated with memory formation, which correlated with the decrease in neurotrophin concentration. Thus, these data indicate that brain-derived neurotrophic factor is involved in the formation of a long-term memory for an aversive stimulus and may function as a modulator of presynaptic proteins associated with exocytosis, enabling increases in neurotransmitter release.

Interleukin-1 induces lipid peroxidation and membrane changes in rat hippocampus: An age-related study.

BACKGROUND: The proinflammatory cytokine, interleukin-1, is traditionally associated with the immune response but recent evidence indicates that it plays a role in neuronal function. Its expression is increased in neurodegenerative conditions and preliminary evidence suggests that it is also increased with increasing age. Receptors for interleukin-1 are differentially distributed in the brain with a high density in the hippocampus, where interleukin-1beta exerts inhibitory effects on release and calcium channel function. OBJECTIVE: The aim of this study was to investigate the possibility that interleukin-1 might lead to age-related changes in membrane composition. METHODS: Lipid peroxidation was assessed in the presence or absence of interleukin-1beta in hippocampal tissue prepared from 4- and 22-month-old rats. These data were analysed in parallel with age-related changes in arachidonic acid and interleukin-1beta concentrations in the hippocampus. RESULTS: We report that interleukin-1beta increased lipid peroxidation in hippocampal tissue prepared from 4- but not 22-month-old rats, and that this effect was inhibited by alpha-tocopherol. The attenuated response to interleukin-1beta in tissue prepared from aged rats correlated with increased expression of endogenous interleukin-1beta. Thus, using an ELISA, we have demonstrated an age-related increase in the concentration of interleukin-1beta, which is accompanied by an age-related decrease in membrane arachidonic acid. CONCLUSION: We propose that increased interleukin-1beta expression impacts on membrane composition and therefore contributes to age-related impairments in neuronal function.

Molecular cloning and characterization of rat P2Y4 nucleotide receptor.

An intronless open reading frame encoding a protein (361aa in length) was isolated from a rat genomic library probed with a DNA fragment from rat heart. This protein showed 83% sequence identity with the human P2Y4 (hP2Y4) receptor and represents a homologue of the human pyrimidinoceptor. However, the rP2Y4 receptor is not selective for uridine nucleotides and, instead, shows an agonist potency order of ITP = ATP = ADP(pure) = UTP = ATPgammaS = 2-MeSATP = Ap4A > UDP(pure). ADP, ATPgammaS, 2-MeSATP and UDP are partial agonists. Thus, in terms of agonist profile, rP2Y4 is more like the P2U receptor subtype. The rP2Y4 receptor was reversibly antagonized by Reactive blue 2 but not by suramin which, otherwise, inhibits the hP2Y2 receptor (a known P2U receptor). Thus, rP2Y4 and the P2Y2 subtype appear to be structurally distinct forms of the P2U receptor (where ATP and UTP are equi-active) but can be distinguished as suramin-insensitive and suramin-sensitive P2U receptors, respectively.

Analysis of the effect of membrane arachidonic acid concentration on modulation of glutamate release by interleukin-1: an age-related study.

Aging is associated with a change in membrane composition that includes a decrease in membrane polyunsaturated fatty acids, including arachidonic acid, and an increase in membrane cholesterol. Alterations in membrane structure are likely to impact on transmitter release, which relies on the fusion of synaptic plasma and synaptic vesicle membranes, and it may therefore be the underlying cause of the age-related decrease in glutamate release in hippocampal preparations. Recent evidence indicates that interleukin-1, by binding with its receptor, inhibits glutamate release in hippocampal synaptosomes prepared from young but not aged rats. The age-related attenuated effect may be due to impaired ligand-receptor interactions arising from the change in membrane composition, which should theoretically be reversed by increasing membrane polyunsaturated fatty acid concentration. To test this hypothesis, we have investigated the effect of a diet supplemented with arachidonic acid and its precursor, gamma-linolenic acid, on membrane arachidonic acid concentration, glutamate release and on the release response to interleukin-1 in hippocampal tissue prepared from aged and young rats. We report that dietary supplementation reversed the age-related changes in membrane arachidonic acid and expression of IL-1beta. We also present data that indicate that the age-related decrease in glutamate release from hippocampal synaptosomes was reversed in aged animals that had been fed on the experimental diet. The data support the view that changes in membrane composition contribute to certain age-related deficits, in particular the decrease in glutamate release observed in hippocampal synaptosomes.

The ability of aged rats to sustain long-term potentiation is restored when the age-related decrease in membrane arachidonic acid concentration is reversed.

The ability of aged rats to sustain long-term potentiation in the dentate gyrus of the hippocampus is impaired and this impairment correlates with decreased release of glutamate and a decrease in membrane arachidonic acid concentration. Twenty-two-month-old rats receiving a diet supplemented with arachidonic acid and its precursor, gamma-linolenic acid, sustained long-term potentiation in a manner indistinguishable from four-month-old controls. Dietary supplementation also restored arachidonic acid concentrations in membranes prepared from hippocampus of these aged animals to levels observed in hippocampus of four-month-old rats. Glutamate release stimulated by depolarization was similar in dentate gyrus prepared from young rats and aged rats which received the experimental diet, but was markedly reduced in aged animals which received the control diet. In addition, the synergism between arachidonic acid and the metabotropic glutamate receptor agonist, trans-1-amino-cyclopentyl-1,3-dicarboxylate, on glutamate release, which was observed in hippocampal synaptosomes prepared from four-month-old rats, was also observed in hippocampal preparations obtained from aged rats which had been fed with the experimental diet, but was absent in hippocampal preparations obtained from aged animals which were fed with control diet. Thus, reversing the age-related decrease in membrane arachidonic acid concentration restored ability of aged animals to sustain long-term potentiation and reversed age-related changes in glutamate release.

Time-dependent increase in release of arachidonic acid following passive avoidance training in the day-old chick.

We have investigated the role of arachidonic acid, a putative retrograde messenger, in a one-trial aversive learning task in the day-old chick. The left and right intermediate medial hyperstriatum ventrale (IMHV) in the chick forebrain have previously been implicated in the formation of memory for this task. Using an ex vivo technique we have determined the concentrations of various fatty acids liberated from prisms prepared from these brain regions at different time points up to 24 h following passive avoidance training. At 30, 60, and 75 min posttraining the concentration of arachidonic acid, but not of other fatty acids, in prisms prepared from the left IMHV, but not the right IMHV, was enhanced compared with that in chicks trained on a nonaversive water-coated bead. To test whether arachidonic acid liberation from the left IMHV was receptor-stimulated we showed that (a) liberation of endogenous arachidonic acid from homogenate prepared from the left and right IMHV of untrained chicks was stimulated by depolarization with KCl (50 mM) and that (b) glutamate agonists of the NMDA and metabotropic subtypes of glutamate receptor stimulated release of preloaded [14C] arachidonic acid from prisms prepared from the left IMHV but not the right IMHV. These results indicate that arachidonic acid is liberated from the left IMHV following passive avoidance training in the day-old chick and may play a role as a retrograde messenger in this memory task.

Passive avoidance learning in the day-old chick is modulated by GABAergic agents.

Injection of drugs directly into the intermediate medial hyperstriatum ventrale (IMHV) of day-old chicks, prior to training on a chrome bead dipped in either the strong aversant methyl anthranilate (MeA), or the weak aversant quinine, allows investigation of the effects of potential amnestic and memory-enhancing agents on retention of a passive avoidance task. Chicks were injected into the left and right IMHV, with either saline or muscimol (GABA agonist), 30 minutes before training on an MeA-coated bead. On test, either 10 min, 30 min or 24 h after training, birds were presented with a dry chrome bead. Normally, trained birds will avoid the test bead; however, significantly more muscimol-injected birds pecked the dry bead than did saline-injected chicks, indicating amnesia in the muscimol-injected birds. In chicks injected bilaterally into the IMHV with bicuculline, a GABAA antagonist, 30 minutes prior to training on a quinine-coated bead, avoidance scores were significantly improved on testing at 24 h compared with saline-injected control chicks, indicating enhanced retention in bicuculline-treated birds. These results suggest a role for the GABAergic system in the acquisition and retention of passive avoidance learning in the day-old chick.

omega-Conotoxin GVIA disrupts memory formation in the day-old chick.

Calcium channel antagonists have previously been shown to alleviate age-associated reductions in memory acquisition and enhance retention for a number of different tasks. We have investigated the effects of pretraining injections of nifedipine, nimodipine, and amlodipine (L-type calcium channel antagonists) and omega-conotoxin GVIA (an N-type calcium channel antagonist) on passive avoidance and visual discrimination learning in day-old Ross 1 chunky chicks. None of the L-type calcium channel antagonists at any time or dose tested caused amnesia for the passive avoidance task or enhanced weak learning, and nifedipine did not affect acquisition or retention of the visual discrimination task. However, pretraining bilateral intracerebral injections of 1.25 pmole/hemisphere omega-conotoxin GVIA produced amnesia for the passive avoidance task in chicks tested 30 min or 3 h after training. The same dose of omega-conotoxin GVIA significantly reduced the rate of acquisition of the visual discrimination task, but did not affect retention of the task when tested 30 min or 3 h after training. These data indicate that calcium uptake via presynaptic N-type omega-conotoxin-sensitive but not L-type dihydropyridine-sensitive calcium channels is necessary for memory formation in young chicks.

Increase in arachidonic acid concentration in a postsynaptic membrane fraction following the induction of long-term potentiation in the dentate gyrus.

We have determined the concentration of free fatty acids in membranes of slices prepared from the dentate gyrus following the induction of long-term potentiation in the anaesthetized rat. Compared to unpotentiated tissue, there was a significant increase in the concentration of free arachidonic acid 2.5 min, 45 min and 3 h after induction of long-term potentiation. There was no corresponding increase in oleic, stearic or palmitic acids. To account for the increase in free arachidonate, the activities of phospholipase A2, phospholipase A1 and phospholipase C were determined at the same three time intervals in control and potentiated tissue. Two-and-a-half minutes after the induction of long-term potentiation, activity of phospholipase A2 was enhanced, while at 45 min, and at 3 h phospholipase C activity was increased. These results suggest that the liberation of free arachidonate is due initially to phospholipase A2 activity, but that at later stages of long-term potentiation, control switches to phospholipase C. Subcellular fractionation experiments revealed an increase in free arachidonate in the postsynaptic density fraction 45 min after induction of long-term potentiation, without significant changes in synaptosomal- or glial-enriched fractions. These results are consistent with the hypothesis that arachidonic acid, released from a postsynaptic site, acts as a trophic retrograde synaptic signal in long-term potentiation in the dentate gyrus.

Long-term Activation of Phosphoinositide Turnover Associated with Increased Release of Amino Acids in the Dentate Gyrus and Hippocampus Following Classical Conditioning in the Rat.

The release of amino acids and the hydrolysis of inositol phospholipids were examined in parallel in three hippocampal areas following classical conditioning. Paired or unpaired tone(CS) - shock(US) presentations were given to animals engaged in a previously acquired food-motivated lever-pressing task. Conditioned suppression of lever-pressing was the behavioural measure of conditioning. Twenty-four hours after the last conditioning session, the dentate gyrus and areas CA3 and CA1 of the hippocampus were removed bilaterally from conditioned and pseudoconditioned animals, and slices cut and stored in liquid nitrogen for subsequent analysis. Crude synaptosomal pellets were prepared to investigate: (i) potassium-stimulated release of preloaded [3H]glutamate and [14C]aspartate in the presence and absence of extracellular Ca2+; (ii) [3H]inositol labelling of phosphoinositides and inositol phosphates; and (iii) [14C]arachidonic acid labelling of 1,2-diacylglycerol (1,2-DG). Potassium-stimulated, Ca2+-dependent release of [3H]glutamate in synaptosomes prepared from the dentate gyrus and area CA3 was significantly greater in conditioned animals than in pseudoconditioned animals. In area CA1, K+-stimulated, Ca2+-dependent release of [14C]aspartate was significantly increased in conditioned animals. These results confirm in synaptosomes, and extend to a period of 24 h our previous report of an increased release of transmitter in the dentate gyrus and hippocampus associated with classical conditioning. In parallel with the increased release of amino acids, learning was associated with a significant increase in labelling of phosphoinositides and inositol phosphates by [3H]inositol and a significant increase in labelling of 1,2-DG by [14C]arachidonic acid in the three hippocampal areas examined. It is suggested that a long-lasting presynaptic activation of inositol lipid metabolism may contribute to the learning-dependent increase in the capacity of hippocampal terminals to release transmitter and hence to the maintenance of a neurochemical trace which may, at least in part, underlie lasting changes in synaptic function built up during associative learning.

Time-related Changes in Basal Phosphoinositide Turnover after Induction of Long-term Potentiation in the Dentate Gyrus are Blocked by Commissural Stimulation.

We have examined basal phosphoinositide turnover in synaptosomes obtained from the dentate gyrus of anaesthetized rats in which long-term potentiation was induced unilaterally in perforant path-granule cell synapses. Relative to the unpotentiated side, [3H]myo-inositol labelling of inositol phosphates was significantly enhanced 45 min and 3 h after induction of long-term potentiation, but reduced after 2.5 min. Similarly, [14C]arachidonic acid labelling of 1,2-diacylglycerol was increased 45 min and 3 h after induction of long-term potentiation, but reduced after 2.5 min. In a second series of experiments, induction of long-term potentiation was blocked by stimulation of the commissural projection to granule cells. In synaptosomes prepared from this tissue, there was no difference in phosphoinositide turnover between tetanized and control sides at any of the three post-tetanic intervals. We conclude that in the dentate gyrus, long-term potentiation is associated with an increase in phosphoinositide turnover which is established between 2.5 min and 45 min post-tetanus and which persists for at least 3 h.

Increases in glutamate release and phosphoinositide metabolism associated with long-term potentiation and classical conditioning.

Long-term potentiation (LTP) is a widely studied model of the kind of activity-dependent modulation of synaptic efficacy which is assumed to provide the physical basis for learning. Whether LTP, in the hippocampus or elsewhere in the brain, does in fact serve such a role is still a matter for debate. One approach to answering this question is to identify physiological or biochemical changes which are common to both learning and LTP; in the hippocampus, for example, one can ask whether the biochemical changes associated with LTP are also associated with learning. In this chapter we summarize the results which we have obtained in a study of glutamate release and phosphoinositide turnover in the dentate gyrus of rats trained in a classical conditioning task. The similarity between the changes occurring after classical conditioning and those associated with LTP is consistent with the hypothesis that LTP is one of the mechanisms by which a neural trace of the learned association is formed. We discuss this interpretation in the light of the observation that classical conditioning does not appear to affect synaptic responses in the hippocampus.

Increased hydrolysis of phosphatidylinositol-4,5-bisphosphate in long-term potentiation.

Inositol phospholipid hydrolysis was examined in slices and synaptosomes prepared from area CA3 of control hippocampi and hippocampi in which long-term potentiation (LTP) was induced in vivo. In both synaptosomes and slices, LTP was associated with an increase in [3H]inositol labelling of inositol phosphates but not phosphoinositides. Glutamate (10(-3) M) significantly increased labelling of inositol phosphates in slices obtained from control tissue but had no effect either in slices obtained from potentiated tissue or in synaptosomes obtained from control or potentiated tissue. The finding that glutamate had no significant effect in slices prepared from potentiated tissue suggests that glutamate-mediated stimulation of inositol phospholipid hydrolysis in a postsynaptic compartment may be saturated in LTP. The possibility that the increase in phospholipid hydrolysis associated with LTP which we report here may be linked with the previously reported increase in transmitter release is discussed.