News Physiol Sci 18: 147-150, 2003;
doi:10.1152/nips.01422.2002
1548-9213/03 $5.00
News in Physiological Sciences, Vol. 18, No. 4, 147-150,
August 2003
© 2003 Int. Union Physiol. Sci./Am. Physiol. Soc.
Cell Adhesion Molecules: Key Players in Memory Consolidation?
Hans Welzl1 and
Oliver Stork2
1 Division of Neuroanatomy and Behavior, Institute of Anatomy, University of Zurich, CH-8057 Zurich, Switzerland; and
2 Institute of Physiology, University of Magdeburg, School of Medicine, 39016 Magdeburg, Germany
 |
Abstract
|
|---|
Experimental evidence implies that L1 and neural cell adhesion molecule (NCAM) are involved in long-term memory formation. Changes in their expression and glycosylation appear to modify the synaptic strength underlying memory consolidation. Interference with L1 and NCAM function in a variety of learning tasks in different species severely attenuates memory consolidation, indicating their involvement in an evolutionary conserved mechanism of neural plasticity.
|
Introduction
|
|---|
The formation of memory, on a cellular level, is based on the facilitation or attenuation of transmission at specific synapses. Over time, memory is thought to be transformed from an initial short-lived and labile phase (short-term memory) into a longer-lasting stable form (long-term memory), a process termed consolidation. Initial phases of memory formation are dependent on the modification of preexisting synaptic molecules (receptors, channels, enzymes), which instantaneously alters the efficiency of synaptic transmission (9). Long-term memories, on the other hand, depend on the de novo synthesis of proteins and structural modifications at the synaptic level (2). The development of molecular techniques and the introduction of simple vertebrate and invertebrate model systems for the study of learning processes have led to the discovery of a vast number of molecules that participate in memory formation. These molecules, however, are not specifically involved in memory formation but comprise important players in many forms of neural plasticity, such as growth cone guidance and synapse formation during development.
Cell adhesion molecules are well known for their involvement in these developmental processes. Recent evidence strongly suggests that they also participate in synaptic changes underlying memory formation in adult individuals. The involvement of two members of the immunoglobulin superfamily of cell adhesion molecules, L1 and neural cell adhesion molecule (NCAM), in memory formation is particularly well supported by experimental data. Some evidence also exists suggesting a possible contribution of other adhesion molecules, such as integrins and cadherins, to lasting modifications of synaptic connectivity.
L1 is a glycoprotein with an apparent molecular mass of 200 kDa that is widely expressed on neuronal cells in the developing and adult nervous system. NCAM appears in various isoforms with molecular weights of 120, 140, and 180 kDa, which are generated from a single gene through alternative splicing (Fig. 1
). However, the major isoforms differ in their intracellular parts, and functionally important variability also exists in the extracellular region of the molecule (e.g., through expression of the small variable alternatively spliced exon in the fourth immunoglobulin-like domain). The core protein is further modified posttranslationally through glycosylation with polysialic acid (PSA), a carbohydrate consisting of long homopolymers of sialic acid. Both L1 and NCAM mediate homo- and heterophilic cell-cell and cell-matrix interactions. They can also show a cis-interaction with each other, i.e., interaction on the same cell surface via a specific epitope. This already indicates that these molecules are not the passive elements of attachment that the name "adhesion molecule" may suggest. In fact, their expression and their functional state are precisely regulated, and their interaction with extracellular, membrane, and cytoplasmic elements evokes functional changes that alter membrane properties and intracellular signaling cascades, including the inositol phospholipid signaling pathway, intracellular Ca2+ and pH, as well as nonreceptor tyrosine kinases of the Src family.
View larger version (15K):
[in this window]
[in a new window]
|
FIGURE 1. A: structure of neural cell adhesion molecule (NCAM) and L1. The extracellular part of all NCAM isoforms consists of 5 immunoglobulin-like and 2 fibronectin type III-like domains, whereas that of L1 contains 6 immunoglobulin-like and 5 fibronectin type III-like domains. The 120-kDa NCAM isoform is linked to the membrane through a phospholipid anchor and thus can be easily released into the extracellular milieu. The 140-kDa NCAM, 180-kDa NCAM, and L1, in contrast, are anchored to the cell membrane and have specific intracellular domains, through which they can interact with the cytoskeleton or components of the postsynaptic density. Blue dots demarcate the attachment sites for polysialic acid (PSA) polymers on NCAM. B: cell-cell adhesion due to the presence of NCAM molecules in membranes of opposing cells can be weakened by 1) internalization of NCAM and/or by 2) replacement of NCAM with PSA-glycosylated NCAM. In the snail Aplysia californica, internalization of its specific cell adhesion molecules begins at coated pits; the molecules are then either degraded or reinserted at distant sites into the membrane. Binding rates between PSA-NCAMs are three- to fourfold weaker than those between 2 NCAMs without PSA. The attached clouds of PSA polymers impede membrane-membrane contact and the interaction between NCAMs as well as other membrane-bound molecules.
|
|
The critical involvement of glycoproteins in memory formation has been suggested on the basis of experiments showing that avoidance learning in the chick (Gallus domesticus) can be prevented through the inhibition of protein glycosylation in the brain. In support of this, the isotopic labeling of novel glycoproteins revealed an increased generation of cell surface molecules after learning (for review, see Ref. 13). Neural cell adhesion molecules were considered prime candidates for these processes, in particular since isotope incorporation in learning-specific proteins labeled proteins of a molecular weight equivalent to that of the synapse-specific 180 kDa NCAM isoform. More direct evidence for a role of L1 and NCAM in memory formation came from studies that interfered with the function of these molecules in different learning paradigms.
|
Antibodies against L1 and NCAM can impair memory consolidation
|
|---|
Memory consolidation in chicks and in rats could be impaired by application of antibodies that interfere with the function of L1 and NCAM. In the chick, intracranial injection of antibodies against L1 or NCAM prevented memory consolidation when done immediately after visual discrimination training or 4–6 h later; injections in between these two time windows or after the second time window, however, had no such effect (for review, see Ref. 14). In rats, the intraventricular injection of anti-NCAM (6) or a synthetic peptide that prevents NCAM internalization (8) during or 6–8 h after training inhibited memory consolidation in a passive avoidance task. Consolidation was also distinctly attenuated when anti-L1 or anti-NCAM was chronically infused into the ventricle during spatial learning in a Morris water maze (1).
These in vivo findings are matched by effects of anti-L1 and anti-NCAM antibodies on neural plasticity in situ. Long-term potentiation, an electrical stimulation-induced long-lasting increase in synaptic efficacy, is considered to be a model for synaptic plasticity underlying memory formation. It can be studied in vivo as well as in brain slices, the latter being more easily accessible to experimental manipulations. Interfering with L1 or NCAM function in such brain slices through bath application of antibodies against these cell adhesion molecules could indeed prevent the development of hippocampal long-term, but not short-term, potentiation (10).
|
Learning is followed by altered expression and glycosylation of L1 and NCAM
|
|---|
The above-reported findings are supported by the observation that glycoprotein synthesis, including L1 and NCAM, increases after passive avoidance training and spatial learning in rats (7, 12). A similar increase in L1 and NCAM glycosylation after visual discrimination training in chicks appeared in two waves, one immediately after training and one ~6 h after training. The possibility of a third wave of synthesis at 15–18 h after training has also been suggested (18). Likely as a result, the distribution of PSA-NCAM is altered after passive avoidance training in the chick (15). These data complement the above-cited observation of two sensitive time windows for the amnesic effect of drugs interfering with L1 and NCAM function or with protein synthesis in general. They stress the fact that consolidation is not completed 1–2 h after learning but continues for possibly up to 18 h or more.
|
NCAM glycosylation and/or internalization is necessary for memory consolidation
|
|---|
Adhesion properties of L1 and NCAM can be efficiently regulated through glycosylation. In the brain, PSA seems to be selectively associated with NCAM. It is thought that the negatively charged clouds of PSA sugar groups at pre- and postsynaptic sites reject each other and thus lead to a measurable increase in size of the synaptic cleft. The resulting decrease in adhesive strength appears to be necessary for the remodeling of synapses so that a long-lasting memory trace in the form of a facilitated transmission ensues. Increased polysialation of NCAM has been observed 12 and 24 h after one-trial passive avoidance learning (7). When PSA was continually clipped off NCAM with intracranially injected endoneuraminidase, however, spatial learning in rats was attenuated. The same enzyme when added to brain slices in vitro also blocked hippocampal long-term potentiation (4).
The interaction of cell adhesion molecules can also be reduced by internalization of the complete molecule. In a model system of memory formation in the sea snail Aplysia californica, the NCAM homologues of this organism (Aplysia cell adhesion molecules) were removed from the cell surface through an increased endocytosis. This internalization involved an activation of the MAP kinase pathway (3). A similar mechanism may also be active in vertebrates, because an increased ubiquitination of NCAM was observed after avoidance training in rats. In the same task, preventing NCAM internalization with an NCAM-specific synthetic peptide also interfered with memory consolidation (8). Thus regulation of cell adhesion at synapses seems to be an evolutionary conserved mechanism of memory consolidation.
|
Targeted disruption of the NCAM gene results in learning deficits
|
|---|
Finally, the role of NCAM in memory formation was confirmed through the analysis of mice with targeted disruption of the NCAM gene. These NCAM-null mutants develop normally with only minor anatomic aberrations in the olfactory bulb and hippocampal mossy fiber system. However, their performance was inferior to wild-type littermates in a spatial learning task and in fear conditioning (Refs. 5 and 16; Fig. 2). These deficits were once more matched by the finding of a reduced long-term potentiation (11). Memory deficits in NCAM-null mutants were accompanied by alterations in emotional/motivational behavior and a hyperresponsiveness of the serotonergic system. However, these appear to relate to different NCAM functions, because transgenic reexpression of 180-kDa NCAM rescued emotional behavior and serotonin response, but not the learning deficits, in the null mutants (16, 17). The data support the hypothesis of a crucial involvement of NCAM, and especially its 180-kDa isoform, in various behavioral processes, including memory formation.
View larger version (16K):
[in this window]
[in a new window]
|
FIGURE 2. Behavioral deficits in NCAM mutant mice. A: null mutation of the NCAM gene results in reduced contextual and auditory-cued fear memory, as indicated by a reduced freezing response during reexposure in response to the shock context and acoustic fear cue. These deficits could not be overcome through reexpression of 180 kDa NCAM as a transgene in the brain of NCAM null mutant mice. B: anxiety-like behavior, in contrast, could be rescued through the transgenic 180 kDa NCAM, indicating an independent cellular mechanism. *P < 0.05 and **P < 0.01 vs. wild-type. ++P < 0.01 vs. NCAM-null mutant.
|
|
|
Concluding remarks and future outlook
|
|---|
Research during the past decade has provided strong evidence for the participation of the immunoglobulin-type cell adhesion molecules L1 and NCAM in memory consolidation. The involvement of cell adhesion molecules in synaptic plasticity during development and memory consolidation appears to be an evolutionary preserved mechanism that can be found in a wide range of invertebrate and vertebrate species. Modification of L1- and NCAM-mediated cell recognition is a necessary step in the formation of stable changes in synaptic communication and is achieved through coordinated changes in gene expression, posttranslational modification, and internalization of the protein lasting 18 h or more after a learning situation. There is evidence that some of these processes are controlled by glucocorticoids: central administration of corticosterone not only enhanced retention of weak learning but also stimulated a wave of glycoprotein synthesis several hours after training, which seems to include NCAM (14).
|
Acknowledgments
|
|---|
This work was supported by National Competence Center of Research "Plasticity and Repair" and the Deutsche Forschungsgemeinschaft.
|
References
|
|---|
- Arami S, Jucker M, Schachner M, and Welzl H. The effect of continuous intraventricular infusion of L1 and NCAM antibodies on spatial learning in rats. Behav Brain Res 81: 81–87, 1996.[ISI][Medline]
- Bailey CH, Bartsch D, and Kandel ER. Toward a molecular definition of long-term memory storage. Proc Natl Acad Sci USA 93: 13445–13452, 1996.[Abstract/Free Full Text]
- Bailey CH, Kaang B-K, Chen M, Martin KC, Lim C-S, Casadio A, and Kandel ER. Mutation in the phosphorylation sites of MAP kinase blocks learning-related internalization of apCAM in Aplysia sensory neurons. Neuron 18: 913–924, 1997.[ISI][Medline]
- Becker CG, Artola A, Gerardy-Schahn R, Becker T, Welzl H, and Schachner M. The polysialic acid modification of the neural cell adhesion molecule is involved in spatial learning and hippocampal long-term potentiation. J Neurosci Res 45: 143–152, 1996.[ISI][Medline]
- Cremer H, Lange R, Christoph A, Plomann M, Vopper G, Roes J, Brown R, Baldwin S, Kraemer P, Scheff S, Barthels D, Rajewsky K, and Wille W. Inactivation of the NCAM gene in mice results in size-reduction of the olfactory bulb and deficits in spatial learning. Nature 367: 455–459, 1994.[Medline]
- Doyle E, Nolan PM, Bell R, and Regan CM. Intraventricular infusions of anti-neural cell adhesion molecules in a discrete posttraining period impair consolidation of a passive avoidance response in the rat. J Neurochem 59: 1570–1573, 1992.[ISI][Medline]
- Doyle E, Nolan PM, Bell R, and Regan CM. Hippocampal NCAM180 transiently increases sialylation during the acquisition and consolidation of a passive avoidance response in the adult rat. J Neurosci Res 31: 513–523, 1992.[ISI][Medline]
- Foley AG, Hartz BP, Gallagher HC, Ronn LC, Berezin V, Bock E, and Regan CM. A synthetic peptide ligand of neural cell adhesion molecule (NCAM) IgI domain prevents NCAM internalization and disrupts passive avoidance learning. J Neurochem 74: 2607–2613, 2000.[ISI][Medline]
- Kandel ER. The molecular biology of memory storage: a dialogue between genes and synapses. Science 294: 1030–1038, 2001.[Abstract/Free Full Text]
- Lüthi A, Laurent, SP, Figurov A, Müller D, and Schachner M. Hippocampal long-term potentiation and neural cell adhesion molecules L1 and NCAM. Nature 372: 777–779, 1994.[Medline]
- Müller G, Wang C, Skibo G, Toni N, Cremer H, Calaora V, Rougon G, and Kiss JZ. PSA-NCAM is required for activity-induced synaptic plasticity. Neuron 17: 413–422, 1996.[ISI][Medline]
- O’Connell AW, Fox GB, Barry T, Murphy KJ, Fichera G, Foley AG, Kelly J, and Regan CM. Spatial learning activates neural cell adhesion molecule polysialylation in a corticohippocampal pathway within the medial temporal lobe. J Neurochem 68: 2538–2546, 1997.[ISI][Medline]
- Rose SPR. Glycoproteins and memory formation. Behav Brain Res 66: 73–78, 1995.[Medline]
- Rose SPR. Cell-adhesion molecules, glucocorticoids and long-term-memory formation. Trends Neurosci 18: 502–506, 1995.[ISI][Medline]
- Rusakov DA, Davies HA, Krivko IM, Stewart MG, and Schachner M. Training in chicks alters PSA-N-CAM distribution in forebrain cell membranes. Neuroreport 5: 2469–2473, 1994.[Medline]
- Stork O, Welzl H, Wolfer D, Schuster T, Mantei N, Stork S, Hoyer D, Lipp H, Obata K, and Schachner M. Recovery of emotional behaviour in neural cell adhesion molecule (NCAM) null mutant mice through transgenic expression of NCAM180. Eur J Neurosci 12: 3291–3306, 2000.[ISI][Medline]
- Stork O, Welzl H, Wotjak CT, Hoyer D, Delling M, Cremer H, and Schachner M. Anxiety and increased 5-HT1A receptor response in NCAM null mutant mice. J Neurobiol 40: 343–355, 1999.[ISI][Medline]
- Tiunova A, Anokhin KV, Schachner M, and Rose SPR. Three time windows for amnestic effect of antibodies to cell adhesion molecule L1 in chicks. Neuroreport 9: 1645–1648, 1998.[ISI][Medline]
Copyright © 2003 by the Int. Union Physiol. Sci./Am. Physiol. Soc.
Top
Introduction
Antibodies against L1 and...
