On the participation of hippocampal PKC in acquisition, consolidation and reconsolidation of spatial memory.

Memory consolidation involves a sequence of temporally defined and highly regulated changes in the activation state of several signaling pathways that leads to the lasting storage of an initially labile trace. Despite appearances, consolidation does not make memories permanent. It is now known that upon retrieval well-consolidated memories can become again vulnerable to the action of amnesic agents and in order to persist must undergo a protein synthesis-dependent process named reconsolidation. Experiments with genetically modified animals suggest that some PKC isoforms are important for spatial memory and earlier studies indicate that several PKC substrates are activated following spatial learning. Nevertheless, none of the reports published so far analyzed pharmacologically the role played by PKC during spatial memory processing. Using the conventional PKC and PKCmu inhibitor 12-(2-cyanoethyl)-6,7,12,13-tetrahydro-13-methyl-5-oxo-5H-indolo[2,3-a]pyrrollo[3,4-c]carbazole (Gö6976) we found that the activity of these kinases is required in the CA1 region of the rat dorsal hippocampus for acquisition and consolidation of spatial memory in the Morris water maze learning task. Our results also show that when infused into dorsal CA1 after non-reinforced retrieval, Gö6976 produces a long-lasting amnesia that is independent of the strength of the memory trace, suggesting that post-retrieval activation of hippocampal PKC is essential for persistence of spatial memory.

On brain lesions, the milkman and Sigmunda.

Lesion studies have been of historical importance in establishing the brain systems involved in memory processes. Many of those studies, however, have been overinterpreted in terms of the actual role of each system and of connections between systems. The more recent molecular pharmacological approach has produced major advances in these two areas. The main biochemical steps of memory formation in the CAI region of the hippocampus have been established by localized microinfusions of drugs acting on specific enzymes of receptors, by subcellular measurements of the activity or function of those enzymes and receptors at definite times, and by transgenic deletions or changes of those proteins. The biochemical steps of long-term memory formation in CAI have been found to be quite similar to those of long-term potentiation in the same region, and of other forms of plasticity. Connections between the hippocampus and the entorhinal and parietal cortices in the formation and modulation of short- and long-term memory have also been elucidated using these techniques. Lesion studies, coupled with imaging studies, still have a role to play; with regard to human memory, this role is in many ways unique. But these methods by themselves are not informative as to the mechanisms of memory processing, storage or modulation.

On the participation of hippocampal p38 mitogen-activated protein kinase in extinction and reacquisition of inhibitory avoidance memory.

Inhibitory avoidance (IA) learning relies on the formation of an association between stepping down from a platform present in a certain context (conditioned stimulus; CS) with an aversive unconditioned stimulus (US; i.e. a footshock). A single CS-US pairing establishes a robust long-term memory expressed as an increase in step-down latency at testing. However, repeated retrieval of the avoidance response in the absence of the US induces extinction of IA memory. That is, recurring presentation of the CS alone results in a new learning indicating that the CS no longer predicts the US. Although the signaling pathways involved in the consolidation of IA and other fear-motivated memories have been profusely studied, little is known about the molecular requirements of fear memory extinction. Here we report that, as happens with its consolidation, extinction of IA long-term memory requires activity of the p38 subfamily of mitogen-activated protein kinases (MAPK) in the CA1 region of the dorsal hippocampus. Moreover, we found that inhibition of hippocampal p38MAPK blocked memory reacquisition after extinction without affecting either the increase in IA memory retention induced by a second training session or animal's locomotor/exploratory activity and anxiety state.

The connection between the hippocampal and the striatal memory systems of the brain: a review of recent findings.

Two major memory systems have been recognized over the years (Squire, in Memory and Brain, 1987): the declarative memory system, which is under the control of the hippocampus and related temporal lobe structures, and the procedural or habit memory system, which is under the control of the striatum and its connections (Mishkin et al., in Neurobiology of Learning by G Lynch et al., 1984; Knowlton et al., Science 273:1399, 1996). Most if not all learning tasks studied in animals, however, involve either the performance or the suppression of movement. Animals acquire connections between environmental or discrete sensory cues (conditioned stimuli, CSs) and emotionally or otherwise significant stimuli (unconditioned stimuli, USs). As a result, they learn to perform or to inhibit the performance of certain motor responses to the CS which, when learned well, become what can only be called habits (Mishkin et al., 1984): to regularly walk or swim to a place or away from a place, or to inhibit one or several forms of movement. These responses can be viewed as conditioned responses (CRs) and may sometimes be very complex. This is of course also seen in humans: people learn how to play on a keyboard in response to a mental or written script and perform the piano or write a text; with practice, the performance improves and eventually reaches a high criterion and becomes a habit, performed almost if not completely without awareness. Commuting to school in a big city in the shortest possible time and eschewing the dangers is a complex learning that children acquire to the point of near-perfection. It is agreed that the rules that connect the perception of the CS and the expression of the CR change from their first association to those that take place when the task is mastered. Does this change of rules involve a switch from one memory system to another? Are different brain systems used the first time one plays a sonata or goes to school as compared with the 100th time? Here we will comment on: 1) reversal learning in the Morris water maze (MWM), in which the declarative or spatial component of a task is changed but the procedural component (to swim) persists and needs to be re-linked with a different set of spatial cues; and 2) a series of observations on an inhibitory avoidance task that indicate that the brain systems involved change with further learning.

Phosphorylated cAMP response element-binding protein as a molecular marker of memory processing in rat hippocampus: effect of novelty.

From mollusks to mammals the activation of cAMP response element-binding protein (CREB) appears to be an important step in the formation of long-term memory (LTM). Here we show that a 5 min exposure to a novel environment (open field) 1 hr after acquisition of a one-trial inhibitory avoidance training hinders both the formation of LTM for the avoidance task and the increase in the phosphorylation state of hippocampal Ser 133 CREB [phosphorylated CREB (pCREB)] associated with the avoidance training. To determine whether this LTM deficit is attributable to the reduced pCREB level, rats were bilaterally cannulated to deliver Sp-adenosine 3', 5'-cyclic monophosphothioate (Sp-cAMPS), an activator of PKA. Infusion of Sp-Adenosine 3',5'-cyclic monophosphothioate Sp-cAMPS to CA1 region increased hippocampal pCREB levels and restored normal LTM of avoidance learning in rats exposed to novelty. Moreover, a 5 min exposure to the open field 10 min before the avoidance training interferes with the amnesic effect of a second 5 min exposure to the open field 1 hr after avoidance training and restores the hippocampal levels of pCREB. In contrast, the avoidance training-associated activation of extracellular signal-regulated kinases (p42 and p44 mitogen-activated protein kinases) in the hippocampus is not altered by novelty. Together, these findings suggest that novelty regulates LTM formation by modulating the phosphorylation state of CREB in the hippocampus.

GABAA receptor modulation of memory: the role of endogenous benzodiazepines.

GABAA receptors are known to downregulate memory consolidation processes: picrotoxin and bicuculline enhance memory, and benzodiazepines and muscimol depress it. The discovery of naturally occurring benzodiazepines in the brain prompted a recent investigation of whether these compounds could act as physiological regulators of the GABAA receptors involved in memory modulation. Different forms of learning cause a rapid reduction of benzodiazepine-like immunoreactivity in septum, amygdala and hippocampus; microinjection of the benzodiazepine antagonist flumazenil into these regions, at the time that consolidation is taking place, enhances memory. Ivan Izquierdo and Jorge Medina suggest that these and other findings indicate that benzodiazepines released in the septum, amygdala and hippocampus do indeed physiologically downregulate memory storage processes; moreover, benzodiazepine release could be modulated by the anxiety and/or stress associated with each type of learning.

Memory consolidation induces N-methyl-D-aspartic acid-receptor- and Ca2+/calmodulin-dependent protein kinase II-dependent modifications in alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptor properties.

The N-methyl-D-aspartic acid (NMDA) receptor-dependent activation of Ca2+/calmodulin-dependent protein kinase II (CaMKII) is necessary for induction of the long-term potentiation of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptor-mediated responses in the CA1 region of the hippocampus, a putative model for learning and memory. We analyzed the interplay among NMDA receptor, CaMKII and AMPA receptor during consolidation of the memory for an inhibitory avoidance learning task in the rat. Bilateral intra-CA1 infusion of the NMDA receptor antagonist D-(-)-2-amino-5-phosphonopentanoic acid (AP5) or of the CaMKII inhibitor 2-[N-(2-hydroxyethyl)]-N-(4-methoxybenzenesulfonyl)] amino-N-(4-chlorocinnamyl)-N-methylbenzylamine) (KN-93) immediately after step-down inhibitory avoidance training hindered memory consolidation. Learning of the avoidance response induced the NMDA receptor-dependent translocation of alphaCaMKII to a postsynaptic density-enriched fraction isolated from dorsal CA1 and the autophosphorylation of this kinase at Thr-286. Step-down inhibitory avoidance training increased the quantity of GluR1 and GluR2/3 AMPA receptor subunits and the phosphorylation of GluR1 at Ser-831 but not at Ser-845 in CA1 postsynaptic densities. The intra-CA1 infusion of KN-93 and AP5 blocked the increases in GluR1 and GluR2/3 levels and the phosphorylation of GluR1 brought on by step-down inhibitory avoidance training. Our data suggest that step-down inhibitory avoidance learning promotes the learning-specific and NMDA receptor-dependent activation of CaMKII in the CA1 region of the dorsal hippocampus and that this activation is necessary for phosphorylation and translocation of AMPA receptor to the postsynaptic densities, similarly to what happens during long-term potentiation.

Learning twice is different from learning once and from learning more.

The rat hippocampus plays a crucial role in the consolidation of a variety of memories, including that for a one trial inhibitory avoidance learning task in which stepping down from a platform is associated with a footshock. Here we show that this is the case regardless of the intensity of the footshock used and hence, of the strength of the learned response. However, additional learning produced by a second training session in this task does not involve the hippocampus but, instead, the striatum. Memory consolidation of the second trial requires glutamate alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate, N-methyl-D-aspartate and metabotropic receptors, activation of signaling pathways, gene expression and protein synthesis in the striatum, as are required in the hippocampus during memory consolidation of the first trial.

Endogenous benzodiazepine modulation of memory processes.

The immediate posttraining administration of the GABA antagonist, bicuculline, or of the Cl-channel blockers, picrotoxin or Ro 5-4864, enhances memory. These drugs are effective when injected into the amygdaloid nucleus. Intraamygdala muscimol has an opposite effect. All this suggests that memory is modulated at the posttraining period by GABA-A receptors. The pre-, but not posttraining systemic administration of benzodiazepines hinders, and that of inverse agonists, or of the benzodiazepine antagonist, flumazenil enhances retention of diverse tasks. Flumazenil, at doses lower than those that cause an enhancement, antagonizes the effect of benzodiazepine agonists and inverse agonists. This suggests that memory is modulated during acquisition by endogenous benzodiazepine receptor ligands: possibly the diazepam that was recently discovered in brain. Pretraining intraamygdala muscimol administration depresses memory, at doses several times higher than those that are effective posttraining. Pretraining Ro 5-4864 has no effect. This suggests that the release of endogenous benzodiazepines during training may modulate a GABA-A receptor complex, possibly in the amygdala, making it more sensitive to muscimol or Ro 5-4864 in the immediate posttraining period.

Benzodiazepines in the brain. Their origin and possible biological roles.

Great progress has been made in the last 5 yr in demonstrating the presence of benzodiazepines (BDZs) in mammalian tissues, in beginning studies on the origin of these natural compounds, and in elucidating their possible biological roles. Many unanswered questions remain regarding the sources and biosynthetic pathways responsible for the presence of BDZs in brain and their different physiological and/or biochemical actions. This essay will focus on recent findings supporting that: (1) BDZs are of natural origin; (2) mammalian brain contains BDZs in concentrations ranging between 5 x 10(-10)-10(-8) M; (3) dietary source of BDZs might be a plausible explanation for their occurrence in animal tissues, including man; (4) the formation of BDZ-like molecules in brain is a possibility, experimentally supported; (5) BDZ-like molecules including diazepam and N-desmethyldiazepam are elevated in hepatic encephalopathy; and (6) natural BDZs in the brain are involved in the modulation of memory processes. Future studies using the full range of biochemical, physiological, behavioral, and molecular biological techniques available to the neuroscientist will hopefully continue to yield exciting and new information concerning the biological roles that BDZs might play in the normal and pathological functioning of the brain.

Cholinergic mechanisms within the caudate nucleus mediate changes in blood pressure.

Microinjections of 10 micrograms of carbachol into the caudate nucleus induced changes in the blood pressure of cats anesthetized locally, paralyzed and artificially respired. These responses were dependent on the site of injection. Carbachol, microinjected at rostral levels of the caudate nucleus, elicited pressor responses while a decrease in blood pressure was observed following injections at caudal levels. Both of these effects were blocked by prior microinjection of atropine. Microinjections of carbachol outside the caudate did not affect the resting blood pressure. However, injections of carbachol into the lateral ventricle always produced pressor responses independent of the site of injection along the antero-posterior extension of the ventricle. On the other hand, microinjections of dopamine (20 micrograms) into the caudate nucleus failed to modify blood pressure. From this study, it is concluded that within the caudate nucleus there are two different muscarinic mechanisms, which when activated, mediate changes in blood pressure, possibly through the sympathetic nervous system.

Lesion of forebrain nuclei reveals possible presynaptic cholinergic muscarinic receptors in rat cerebral cortex.

In rats, three days after unilateral lesion of magnocellular basal forebrain nuclei, binding of L-[3H]-quinuclidinyl benzilate, and acetylcholinesterase activity decreased significantly in the ipsilateral and, to a lesser extent, in the contralateral cerebral cortex. This result suggests the existence of presynaptic muscarinic receptors in the cortical projections of these nuclei. After 14 days, the binding increased on both sides while the level of acetylcholinesterase activity remained low. These findings suggest that deafferentation causes ultimately an increase in postsynaptic receptors.

The role of the caudate-putamen nucleus in salivary secretion induced by L-DOPA.

Experiments were performed in rats of the Wistar strain anesthetized with alpha-chloralose (100 mg/kg). Electrolytic lesion of either components of the striopallidal complex (corpus striatum, globus pallidus or entopeduncular nucleus) reduced the sensory response to L-DOPA in the contralateral submaxillary glands. Damage to other neural structures, directly or indirectly related to the striopallidal system, left the salivary response unaffected. These structures were: substantia nigra, cerebral cortex, ventromedial and center median-parafascicular thalamic nuclei, nucleus accumbens and posterior hypothalamic areas, including the medial forebrain bundle and lateral habenular nucleus. However, lesions placed in H1-H2 fields of Forel and reticular formation, lateral to the periaqueductal gray, reduced the salivary response in the contralateral glands. This effect was similar to that observed in animals with lesions of the striopallidal complex. From this study, it is concluded that the striatum is the target area for the central effect of L-DOPA on salivary secretion, by activation of pathways descending through the fields of Forel and mesencephalic reticular formation to the contralateral lower brain stem.

Benzodiazepine receptors in the rat hippocampal formation: action of catecholaminergic, serotoninergic and commissural denervation.

The problem of benzodiazepine receptor localization in the rat hippocampal formation has been approached using several methods of selective deafferentation, followed by [3H]flunitrazepam binding studies. The intraventricular injection of 6-hydroxydopamine reduced, after 14 days, the norepinephrine content of the hippocampal formation by 68.4%, and decreased the number of binding sites by 32%, without change in affinity. The intraventricular injection of 5,6 dihydroxytryptamine reduced the serotonin content by 61.5% but did not alter the [3H]flunitrazepam binding. The intraventricular bilateral injection of 0.5 micrograms kainic acid selectively destroyed the pyramidal neurons in area CA3 of both hippocampi and produced an increase of 28% in [3H]flunitrazepam binding, without change in affinity. These results are discussed in relation to our previous observations about benzodiazepine receptor changes after fimbria-fornix transection. The reduction in [3H]flunitrazepam binding after administration of 6-hydroxydopamine suggests the possible localization of the benzodiazepine receptors on adrenergic presynaptic terminals. The increase in binding sites after destruction of CA3 pyramidal cells, which are the site of origin of commissural fibers, is tentatively interpreted as resulting from the sprouting of mossy fibers that replace the associational-commissural projections.

Cholinergic muscarinic receptors in rat cerebral cortex, basal ganglia and cerebellum undergo rapid and reversible changes after acute stress.

Rats, submitted to forced swimming for a period of 15 min (stress), were killed immediately, 60 min or 24 h thereafter. There was an initial decrease in [3H]quinuclidinyl benzilate specific binding in membranes of cerebral cortex (-27%) that returned to normal levels after 60 min. In basal ganglia (striatum and globus pallidus) there was a significant decrease (-14%) in Bmax after 60 min that recovered at 24 h. In cerebellum, an increase in [3H]quinuclidinyl benzilate binding occurred at 60 min (+41%) which was reversed at 24 h. In all cases there were no changes in affinity. These results are discussed in relation to the possible mechanisms that could be involved in the rapid, reversible and selective changes of cholinergic muscarinic receptors in response to acute stress.

Benzodiazepine receptors in rat cerebral cortex and hippocampus undergo rapid and reversible changes after acute stress.

Rats were submitted to forced swimming and were killed 15 min after initiation of the stress and at 1 h, 1 day and 4 days thereafter. Immediately after the stress there was a decrease of 30% in the density of [3H]flunitrazepam binding sites in the cerebral cortex and of 27% in the hippocampal formation, with no changes in all the other brain areas studied. These changes in the number of benzodiazepine receptors were also corroborated by the binding of [3H]ethyl-beta-carboline carboxylate. For both ligands there were no changes in affinity. These effects were selective for the benzodiazepine receptors and no changes in alpha 1, alpha 2 and beta adrenoceptors and in dopaminergic receptors were observed. One hour after the stress, the number of benzodiazepine receptors had recovered in the cerebral cortex (8% above the control) and had increased greatly in the hippocampal formation (+53%). One day after the stress, the [3H]flunitrazepam binding in the cerebral cortex reached the normal level but it was still slightly elevated (+16%) in the hippocampus. These results are discussed in relation to some contradictory findings in the literature and to the fact that the hippocampal formation is related to neural mechanisms underlying behavior and neuroendocrine regulation.

Participation of hippocampal metabotropic glutamate receptors, protein kinase A and mitogen-activated protein kinases in memory retrieval.

The ability to recall past events is a major determinant of survival strategies in all species and is of paramount importance in determining our uniqueness as individuals. In contrast to memory formation, the information about the molecular mechanisms of memory retrieval is surprisingly scarce and fragmentary. Here we show that pretest inhibition of the specific upstream activator of mitogen-activated protein kinase kinase, or of protein kinase A in the hippocampus, blocked retrieval of long-term memory for an inhibitory avoidance task, a hippocampal-dependent learning task. An activator of protein kinase A enhanced retrieval. Mitogen-activated protein kinase activation increased in the hippocampus during retrieval, while protein kinase A activity remained unchanged. Pretest intrahippocampal blockade of metabotropic glutamate receptors or alpha-amino-3-hydroxy-5-methyl-4-isoxazolone propionic acid/kainate receptors, but not N-methyl-D-aspartate receptors or calcium/calmodulin dependent-protein kinase II, impaired retrieval. Thus, recall of inhibitory avoidance activates mitogen-activated protein kinase, which is necessary, along with metabotropic glutamate receptors, alpha-amino-3-hydroxy-5-methyl-4-isoxazolone propionic acid/kainate receptors, and protein kinase A, for long-term memory expression. Our results indicate that memory formation and retrieval may share some molecular mechanisms in the hippocampus.

Chrysin (5,7-di-OH-flavone), a naturally-occurring ligand for benzodiazepine receptors, with anticonvulsant properties.

Chrysin (5,7-di-OH-flavone) was identified in Passiflora coerulea L., a plant used as a sedative in folkloric medicine. Chrysin was found to be a ligand for the benzodiazepine receptors, both central (Ki = 3 microM, competitive mechanism) and peripheral (Ki = 13 microM, mixed-type mechanism). Administered to mice by the intracerebroventricular route, chrysin was able to prevent the expression of tonic-clonic seizures induced by pentylenetertrazol. Ro 15-1788, a central benzodiazepine receptor antagonist, abolished this effect. In addition, all of the treated mice lose the normal righting reflex which suggests a myorelaxant action of the flavonoid. The presence in P. coerulea of benzodiazepine-like compounds was also confirmed.

In vivo and in vitro modulation of central type benzodiazepine receptors by phosphatidylserine.

The in vivo and in vitro modulation of central benzodiazepine binding sites (BDZ-R) by phosphatidylserine purified from bovine cerebral cortex (BC-PS) was studied. Five days i.p. administration of 15 mg/kg/day of BC-PS liposomes increased the maximal number of binding sites (Bmax) for [3H]flunitrazepam in cerebral cortical membranes. In contrast, the density of hippocampal benzodiazepine recognition binding sites decreased. In cerebellar membranes, BC-PS treatment did not alter the characteristics of [3H]flunitrazepam binding. Similar experiments using phosphatidylcholine extracted from bovine brain (BC-PC) resulted in no changes in the [3H]flunitrazepam binding in the 3 neural structures studied. Confirming previous results, rats submitted to an acute swimming stress showed a decrease in the density of cerebral cortex BDZ-R. Animals treated with BC-PS liposomes before stress showed cortical [3H]flunitrazepam binding significantly below treated, unstressed animals but not below controls. The effects of BC-PS liposomes appeared to be selective for the central type of BDZ-R since no changes were observed in [3H]RO 5-4864 binding, a radioligand specific for the peripheral type BDZ-R. Preincubation of cerebral cortical and cerebellar synaptosomal membranes with BC-PS liposomes (1-300 micrograms per assay) significantly increased in a concentration-dependent manner (up to 100 micrograms) the [3H]flunitrazepam binding. Scatchard analysis revealed changes in the apparent affinity without alterations in the Bmax. Very similar results were obtained using a purified PS from spinal cord. BC-PC, phosphatidylinositol, phosphatidic acid and the lyso derivatives of PS and PC (lysoPS and lysoPC) were found to be ineffective.(ABSTRACT TRUNCATED AT 250 WORDS)