Long-term potentiation and memory

The discovery of long-term potentiation (LTP) transformed research on the neurobiology of learning and memory. This did not happen overnight, but the discovery of an experimentally demonstrable phenomenon reflecting activity-driven neuronal and synaptic plasticity changed discussions about what might underlie learning from speculation into something much more concrete. Equally, however, the relationship between the discovery of LTP and research on the neurobiology of learning and memory has been reciprocal; for it is also true that studies of the psychological, anatomical and neurochemical basis of memory provided a developing and critical intellectual context for the physiological discovery. The emerging concept of multiple memory systems, from 1970 onwards, paved the way for the development of new behavioural and cognitive tasks, including the watermaze described in this paper. The use of this task in turn provided key evidence that pharmacological interference with an LTP induction mechanism would also interfere with learning, a finding that was by no means a foregone conclusion. This reciprocal relationship between studies of LTP and the neurobiology of memory helped the physiological phenomenon to be recognized as a major discovery.     

Posttetanic potentiation and presynaptically induced long-term potentiation at the mossy fiber synapse in rat hippocampus

A form of long-term potentiation (LTP) is induced at the mossy fiber (MF) synapse in the hippocampus by highfrequency presynaptic stimulation (HFS). It is generally accepted that induction of this form of LTP (MF LTP) does not depend on postsynaptic Ca²⁺ current gated by N-methyl-D -aspartate receptors, but it has remained controversial whether induction depends on postsynaptic depolarization and voltage-gated entry of Ca²⁺. There are also contradictory data on the time course of both LTP and post-tetanic potentiation (PTP), a shorter duration form of potentiation observed at MF synapses immediately following HFS. It has been proposed that some of these differences in results may have arisen because of difficulties in isolating monosynaptic responses to MF input. In the present study, whole cell recording was used to observe excitatory postsynaptic currents (EPSCs) elicited in CA3 pyramidal cells by input from MFs. Postsynaptic cells were dialyzed with 1,2-bis(o-aminophenoxy)-ethane-N,N,N′,N′-tetraacetic acid (BAPTA) and F^? to inhibit postsynaptic mechanisms that required Ca²⁺, cells were under voltage clamp during HFS, and conditions were selected to minimize the likelihood of polysynaptic contamination. Under these conditions, HFS nevertheless induced robust LTP (mean magnitude, 62%). The possibility that EPSCs were contaminated by polysynaptic components was investigated by exposing the slices to a suppressing medium (one that partially blocked neurotransmission). EPSC waveforms did not change shape during suppression, indicating that contamination was absent. The LTP observed always was accompanied by prominent PTP that lasted through the first 5 to 15 min following HFS (mean decay time constant, 3.2 min). Induction of this LTP was not cooperative; there was no relationship between the size of responses and the magnitude of the LTP induced. LTP magnitude also was unrelated to the extent to which postsynaptic cells depolarized during HFS. These results show that high rates of presynaptic MF activity elicit robust LTP whether or not there is accompanying postsynaptic depolarization or increase in the concentration of postsynaptic Ca²⁺. High-frequency MF activity also results in a PTP that is unusually large and long. © 1995 John Wiley & Sons, Inc.     

Proteases involved in long-term potentiation

Much attention has been paid to proteases involved in long-term potentiation (LTP). Calpains, Ca-dependent cysteine proteases, have first been demonstrated to be the mediator of LTP by the proteolytic cleavage of fodrin, which allows glutamate receptors located deep in the postsynaptic membrane to move to the surface. It is now generally considered that calpain activation is necessary for LTP formation in the cleavage of substrates such as protein kinase Czeta, NMDA receptors, and the glutamate receptor-interacting protein. Recent studies have shown that serine proteases such as tissue-type plasminogen activator (tPA), thrombin, and neuropsin are involved in LTP. tPA contributes to LTP by both receptor-mediated activation of cAMP-dependent protein kinase and the cleavage of NMDA receptors. Thrombin induces a proteolytic activation of PAR-1, resulting in activation of protein kinase C, which reduces the voltage-dependent Mg2+ blockade of NMDA receptor-channels. On the other hand, neuropsin may act as a regulatory molecule in LTP via its proteolytic degradation of extracellular matrix protein such as fibronectin. In addition to such neuronal proteases, proteases secreted from microglia such as tPA may also contribute to LTP. The enzymatic activity of each protease is strictly regulated by endogenous inhibitors and other factors in the brain. Once activated, proteases can irreversibly cleave peptide bonds. After cleavage, some substrates are inactivated and others are activated to gain new functions. Therefore, the issue to identify substrates for each protease is very important to understand the molecular basis of LTP.     

Long-term potentiation and 4-aminopyridine

Long-term potentiation (LTP) of excitatory postsynaptic potentials (epsp's) was investigated with extracellular field potential recording in hippocampal slices from rats. In the presence of 100 microM 4-aminopyridine (4-AP) the probability of eliciting LTP was unchanged or increased; the extent of potentiation was not significantly different from normal. During LTP saturation, 4-AP further enhanced the epsp. These data are inconsistent with an involvement of A-current reduction in LTP.     

Long-term potentiation in the Eocene

The first ten years of long-term potentiation (LTP) research are reviewed. Surprisingly, given the intensity of current interest, the discovery paper did not trigger a wave of follow-on experiments. Despite this, the initial work laid out what ultimately became standard questions and paradigms. The application of the then still novel hippocampal slice technique oriented LTP towards basic neuroscience, perhaps somewhat at the cost of lesser attention to its functional significance. The use of slices led to the discovery of the events that trigger the formation of LTP and provided some first clues about its extraordinary persistence. Signs of the intense controversy over the nature of LTP expression (release vs receptors) emerged towards the end of the first decade of work. What appears to be lacking in the literature of that time is a widespread concern about LTP and memory. This may reflect a somewhat different attitude that neurobiologists then had towards memory research and a perceived need to integrate the new potentiation phenomenon into the web of established science before advancing extended arguments about its contributions to behaviour.     

Quantal analysis and long-term potentiation

Quantal analysis is useful for assessing the pre- and/or post-synaptic locus of the expression of long-term tetanic potentiation with the condition, however, that the studied synaptic potentials have been evoked by single cell stimulations, as is the case with paired recordings of identified neurons. The application of this methodology, primarily with indirect criteria, has produced conclusions which dance back and forth across the synaptic cleft.     

Calcium signals in long-term potentiation and long-term depression

We describe postsynaptic Ca2+ signals that subserve induction of two forms of neuronal plasticity, long-term potentiation (LTP) and long-term depression (LTD), in rat hippocampal neurons. The common induction protocol for LTP, a 1-s, 50-Hz tetanus, generates Ca2+ increases of about 50-Hz in dendritic spines of CA1 neurons. These very large increases, measured using a low affinity indicator (Mg fura 5), were found only in the spines and tertiary dendrites, and were dependent upon influx through N-methyl-D-aspartate (NMDA) gated channels. High affinity Ca2+ indicators (e.g., fura 2) are unable to demonstrate these events. In acute slices, neighboring dendritic branches often showed very different responses to a tetanus, and in some instances, neighboring spines on the same dendrite responded differently. LTD in mature CA1 neurons was induced by a low frequency stimulus protocol (2 Hz, 900 pulses), in the presence of GABA- and NMDA-receptor blockers. This LTD protocol produced dendritic Ca2+ increases of <1 microM. Duration of the Ca2+ increase was approximately 30 s and was due to voltage-gated Ca2+ influx. Finally, the ability of synaptically addressed Ca2+ stores to release Ca2+ was studied in CA3 neurons and was found to require immediate preloading and high intensity presynaptic stimulation, conditions unlike normal LTP-LTD protocols.     

Postsynaptic potentiation and desensitization in frog neuromuscular junction

The fractional increase in ACh responses that occurs at the beginning of each train of iontophoretically applied ACh pulses has been examined at the frog neuromuscular junction at room temperature, in the presence of active cholinesterase, during desensitization produced by a rapid sequence (every 20 s) of short (5 Hz, 5 s) iontophoretic trains of ACh. The fractional increase in ACh responses, which is used as an indicator of postsynaptic potentiation, becomes progressively greater with ACh application, often markedly (greater than 100%), although ACh responses are greatly reduced (as much as 90%) owing to desensitization. Clearly postsynaptic potentiation can exist concomitantly with desensitization. In addition, the dose-response curve is shifted to the right and its maximal response is diminished. The shift in the dose-response curve to the right, which can explain greater postsynaptic potentiation, is unlikely to be caused by accumulation of "monoligand-bound ACh receptor complexes," since experiments were done with active cholinesterase. The shift probably results from a greater number of desensitized receptors which, because of their large affinity for ACh molecules, serve as "high affinity traps." A small decrease of the maximal dose-response suggests only a small fractional decrease in the number of activable receptors, whereas a large shift to the right indicates a large fractional increase in the number of desensitized receptors. It appears that prior to ACh application only a small fraction of all receptors are desensitized. Alternatively, the shift to the right occurs because the cooperative action of ACh on receptors increases during desensitization.     

Long-term potentiation in cultured hippocampal neurons

Studies performed on low-density primary neuronal cultures have enabled dissection of molecular and cellular changes during N-methyl-D-aspartate (NMDA) receptor-dependent long-term potentiation (LTP). Various electrophysiological and chemical induction protocols were developed for the persistent enhancement of excitatory synaptic transmission in hippocampal neuronal cultures. The characterisation of these plasticity models confirmed that they share many key properties with the LTP of CA1 neurons, extensively studied in hippocampal slices using electrophysiological techniques. For example, LTP in dissociated hippocampal neuronal cultures is also dependent on Ca(2+) influx through post-synaptic NMDA receptors, subsequent activation and autophosphorylation of the Ca(2+)/calmodulin-dependent protein kinase II (CaMKII) and an increase in alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptor insertion at the post-synaptic membrane. The availability of models of LTP in cultured hippocampal neurons significantly facilitated the monitoring of changes in endogenous postsynaptic receptor proteins and the investigation of the associated signalling mechanisms that underlie LTP. A central feature of LTP of excitatory synapses is the recruitment of AMPA receptors at the postsynaptic site. Results from the use of cell culture-based models started to establish the mechanism by which synaptic input controls a neuron's ability to modify its synapses in LTP. This review focuses on key features of various LTP induction protocols in dissociated hippocampal neuronal cultures and the applications of these plasticity models for the investigation of activity-induced changes in native AMPA receptors.     

突触长时程增强形成机制的研究进展

高等动物脑内突触传递的可塑性是近30年来神经科学研究的热点,突触传递长时程增强（long-term potentiation,LTP)是神经元可塑性的反映,其形成主要与突触后机制有关。过去关于LTP机制的研究主要集中于N－甲基－D门冬氨酸（NMDA）受体的特征及该受体被激活后的细胞内级联反应,现认为脑内存在只具有NMDA受体而不具有α－氨基羟甲基恶唑丙酸（AMPA）受体的“静寂突触（silent synapse)”,这一概念的提出,使人们认识到AMPA受体在LTP表达的突触后机制中的重要作用。     

Heat potentiation of radiation damage versus radiation potentiation of heat damage

The enhanced lethality of mammalian cells after combined treatment with hyperthermia and radiation is usually attributed to heat potentiation of radiation damage. However, it has been suggested that the situation may be reversed and that radiation may act as a modifier for heat damage. To test this hypothesis, BP-8 murine sarcoma cells were subjected to sequential radiation and heat treatments and the kinetics and extent of cell death were evaluated with the [125I]-iododeoxyuridine prelabeling assay. Cell death after heating was rapid and essentially complete within 2 days after heat exposure, whereas radiation death was slow and became apparent only after a delay period of 3 days. Combined exposure of cells to radiation and heat caused a pronounced increase in the delayed component of cell death, that is, the radiation component of death. Irradiation of cells before heating did not change the early heat component of cell death even in cells that were exposed to massive radiation doses of up to 300 Gy prior to heating. These results indicate that the increased cell death observed in hyperthermia/radiation-treated cells results from heat potentiation of radiation damage, not radiation potentiation of heat damage.     

AMPA receptor trafficking and long-term potentiation

Activity-dependent changes in synaptic function are believed to underlie the formation of memories. A prominent example is long-term potentiation (LTP), whose mechanisms have been the subject of considerable scrutiny over the past few decades. I review studies from our laboratory that support a critical role for AMPA receptor trafficking in LTP and experience-dependent plasticity.     

Long-term potentiation and the ageing brain

Ageing is associated with learning and memory impairments. Data are reviewed that suggest that age-related impairments of hippocampal-dependent forms of memory, may be caused, in part, by altered synaptic plasticity mechanisms in the hippocampus, including long-term potentiation (LTP). To the extent that the mechanisms responsible for LTP can be understood, it may be possible to develop therapeutic approaches to alleviate memory decline in normal ageing.