Plasticity: downstream of glutamate

Research Update

TRENDS in Neurosciences Vol.24 No.10 October 2001

553

Plasticity: downstream of glutamate Peter C. Kind and Paul E. Neumann Glutamate neurotransmission is an essential component of many forms of neuronal plasticity, however, the intracellular mechanisms that mediate plasticity are only beginning to be elucidated. The emerging image of the NMDA receptor complex reminds us that the similarity between mechanisms of plasticity in various model systems is greater than their apparent differences. For example, the cAMP-dependent protein kinase A signalling pathway is crucial for plasticity in a variety of neuronal systems and across a wide variety of species.

Experience-dependent plasticity is an important feature of nervous systems, especially for development, learning and memory. Recent findings in mammalian cerebral cortical systems have given insight into intracellular mechanisms by which glutamate neurotransmission causes alterations in synaptic strength, gene transcription, and neuronal morphology. In studies of neuronal plasticity in mammals, hippocampal LTP and LTD, ocular dominance (OD) columns and somatosensory barrels are some of the most commonly used model systems. The neurotransmitter glutamate plays a key role in formation of long-lasting potentiation of postsynaptic responses following high-frequency stimulation in hippocampal slices1,2. Metabotropic (mglu) and NMDA ionotropic glutamate receptors appear to mediate the effects of glutamate through activation of intracellular enzymes, many of which are regulated by levels of postsynaptic Ca2+. In the somatosensory cortex of mice, NMDA receptor blockade by APV application blocks the functional changes associated with mystacial whisker ablation during the first postnatal week3. Furthermore, ablation of the NR1 subunit specifically in excitatory neurons in the cortex, prevents the formation of cortical barrels, but not the segregation of thalamocortical afferent fibres representing the mystacial vibrissae4. In the cat visual cortex, the majority of neurons respond, to a variable degree, to input from both eyes. Monocular deprivation by eyelid suture during the http://tins.trends.com

first few months of life in cats causes a dramatic alteration of these response properties such that the vast majority of cells respond solely to input from the nondeprived eye5. This shift in OD is dependent on activation of NMDA receptors because selective blockade with either APV or MK-801 prevents this plasticity6,7. Opening the closed eye and closing the initially non-deprived eye can reverse the shift in OD caused by monocular deprivation8. The bidirectional changes in OD are mirrored by changes in synaptic strength at individual synapses9 and by changes in the ratio of NR2B to NR2A subunits in NMDA receptors10,11. The increase in NR2A subunitcontaining NMDA receptors following visual experience is associated with a decrease in the duration of the NMDA receptor-mediated postsynaptic currents12. These findings are consistent with the hypothesis that alteration in the NMDA receptor subunit composition, and concomitant changes in receptor-mediated postsynaptic current kinetics, contributes to neuronal plasticity during the critical period or to changes in the LTD or LTP modification threshold12. However, the subunit change might affect plasticity via a mechanism that is independent of the change in current kinetics. The end of the sensitive period for LTP induction at thalamocortical synapses in mouse somatosensory cortex is correlated with an alteration in NMDA receptor subunit composition, but not with changes in NMDA receptor-mediated postsynaptic current kinetics13. Therefore, prolonged NMDA receptor-mediated Ca2+ currents might be necessary for experiencedependent modification of cortical neurons, but they are not sufficient. In spite of the wealth of knowledge concerning the intracellular cascades involved in hippocampal LTP and LTD, research examining the molecular basis of cortical development and plasticity has been largely restricted to the neurotransmitter receptors that initiate the cascade of events that lead to functional alterations in neuronal phenotype. The recent isolation and initial characterization of NMDA receptor complexes using immunoprecipitation

followed by mass spectroscopy and immunoblotting has identified numerous candidate molecules that could have crucial roles in glutamate receptor signalling and plasticity14. NMDA receptor complexes consist of more than 80 different proteins, including receptor subunits (e.g. from ionotropic glutamate receptors, mglu receptors and kainate receptors, but not AMPA receptors), adaptor proteins, signalling proteins, cytoskeletal proteins and cell adhesion proteins. The signalling molecules in these complexes include protein kinases and phosphatases, H-Ras, Rap2 and GTPase-activating proteins. These results are consistent with previous reports suggesting that NMDA receptors activate multiple intracellular signalling pathways as determined by immunoprecipitation and/or two-hydrid screens with ionotropic receptor subunits, or by pharmacological inhibitor or targeted-mutagenesis studies. The emerging image of the NMDA receptor complex reminds us that the similarity between mechanisms of plasticity in various model systems is greater than their apparent differences. As further evidence of this unified view of the molecular basis of neuroplasticity, cAMP/protein kinase A (PKA) signalling pathways, which have been shown to be involved in hippocampal LTP, barrel formation, and in learning and memory paradigms in Drosophila and Aplysia15–17, have recently been shown to be involved in OD (Ref. 18). Similar to NMDA receptor blockade with APV or MK-801, osmotic minipump application of the cAMPdependent PKA inhibitor Rp-8-Cl-cAMPS to the visual cortex of 3.5–4.5-week-old cats completely prevented the shift in OD resulting from a subsequent five-day period of monocular deprivation18. This loss of OD shift was accompanied by a disruption of orientation selectivity also similar to that observed with APV blockade of NMDA receptors6. These findings indicate a crucial role for PKA in development and plasticity of receptive field properties of visual cortical neurons and suggest that it might be a crucial signalling molecule downstream of NMDA receptor activation. The phosphorylation

0166-2236/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0166-2236(00)01921-4

554

Research Update

status of a PKA site in AMPA receptor GluR1 subunits is associated with bidirectional changes in AMPA receptor channel conductance and trafficking, which are both involved in hippocampal LTP and LTD (Refs 19–21). Beaver et al.18 suggest that NMDA receptor activation might cause increased cAMP-induced PKA activity, as has been reported in hippocampal LTP (Ref. 22); however, the precise mechanism is unknown. It appears most likely at this point that Ca2+ stimulation of calmodulindependent adenylyl cyclases (AC1 and AC8), which play a role in hippocampal LTP (Ref. 23), is responsible for NMDA receptor-mediated activation of PKA. Interestingly, PKA catalytic and regulatory (RIIβ) subunits are members of the NMDA receptor complex14. PKA-RIIβ and protein phosphatase 1 (PP1) are linked to the NR1 subunit by YOTIAO (A-kinase anchoring protein 9)24,25. This linkage facilitates phosphorylationdependent modulation of NMDA receptor activity in agreement with findings that activation of PKA enhances NMDA receptor currents25. It is also possible that PKA is mediating its plasticity-promoting effects through other neurotransmitter receptors. PKA is a chief target of the second messenger, cAMP, whose levels are influenced by ligands binding to G-protein-coupled receptors, including mglu, acetylcholine, dopamine, noradrenaline and serotonin receptors. In addition, cAMP levels following mglu receptor activation parallel the sensitive period to altered visual experience26 and acetylcholine, noradrenaline and serotonin have been shown to play roles in synaptic plasticity and cortical development27,28. Furthermore, mice with a deletion of the gene encoding mglu5 show defects in barrel development beyond those observed in PLC-β1deficient mice, suggesting a role for cAMP signalling29. The results indicate a role for cAMP in the development of barrels and suggest that PKA might integrate signals from a variety of sources to determine the nature of the experience-dependent changes. Mice lacking the Adcy1 gene show defects in thalamic afferent segregation and fail to form barrels17, suggesting a possible presynaptic role for PKA. PKA is present in axon terminals and has been shown to affect neurotransmitter http://tins.trends.com

TRENDS in Neurosciences Vol.24 No.10 October 2001

release30,31. If PKA is crucial for the release of glutamate from thalamocortical terminals, then inhibition of PKA could simply reduce cortical activity and hence either prevent activity-dependent changes or cause a shift in OD towards the closed eye. However, this suggestion appears unlikely for several reasons: first, Beaver et al.18 found no change in overall levels of spontaneous and visually-induced activity in normal and experimental animals. Second, mice with a genetic deletion of PKA-RIIβ display a near complete loss of barrels in somatosensory cortex, in spite of good segregation of thalamocortical afferents in the posteromedial barrel subfield (R.M. Abdel-Majid et al. unpublished observations). The different results in ADCY1 and PKA-RIIβ mutant mice indicate a role for cAMP in thalamic afferent segregation, independent of PKA signalling. With the development of advanced biochemical and molecular techniques, including specific antagonists to second messenger enzymes, the intracellular signalling pathways subsequent to receptor activation are beginning to be elucidated; however, the search for specificity within neuronal responses has been mired in an attempt to untangle second messenger systems. Although the extent to which the composition of NMDA receptor complexes varies during development (as a result of altered sensory experience and learning) within individual neurons and between brain regions is not known, the complex itself provides further evidence of the integration of intracellular responses to glutamate neurotransmission. Neuronal responses to NMDA receptor stimulation might be limited, varying largely in a coordinated, multi-faceted graded manner, with various thresholds. If cellular responses are integrated, why are there so many signalling pathways? Research in this exciting field will undoubtedly increase our understanding of the mechanisms by which activity through different neurotransmitter receptors during the sensitive period is integrated, resulting in alterations in neuronal function. References 1 Cain, D.P. (1997) LTP, NMDA, genes and learning. Curr. Opin. Neurobiol. 7, 235–242 2 Malenka, R.C. and Nicoll, R.A. (1999) Long-term potentiation – a decade of progress? Science 285, 1870–1874

3 Schlagger, B.L. et al. (1993) Postsynaptic control of plasticity in developing somatosensory cortex. Nature 364, 623–626 4 Iwasato, T. et al. (2000) Cortex-restricted disruption of NMDAR1 impairs neuronal patterns in the barrel cortex. Nature 406, 726–731 5 Mitchell, D.E. and Timney, B. (1984) Postnatal development of function in the mammalian visual system. In Handbook of Physiology Section I: The Nervous System, Vol. 3, Part 1 Sensory Processes. I. Darian-Smith (Ed.). American Physiological Society, Bethesda, pp. 507–555 6 Bear, M.F. et al. (1990) Disruption of experiencedependent synaptic modifications in striate cortex by infusion of an NMDA receptor antagonist. J. Neurosci. 10, 909–925 7 Daw, N.W. et al. (1999) Injection of MK-801 affects ocular dominance shifts more than visual activity. J. Neurophysiol. 81, 204–215 8 Blakemore, C. and Van Sluyters, R.C. (1974) Reversal of the physiological effects of monocular deprivation in kittens: further evidence for a sensitive period. J. Physiol. 237, 195–216 9 Kirkwood, A. et al. (1996) Experience-dependent modification of synaptic plasticity in visual cortex. Nature 381, 526–528 10 Quinlan, E.M. et al. (1999) Bidirectional, experience-dependent regulation of N-methyl-Daspartate receptor subunit composition in the rat visual cortex during postnatal development. Proc. Natl. Acad. Sci. U. S. A. 96, 12876–12880 11 Quinlan, E.M. et al. (1999) Rapid, experiencedependent expression of synaptic NMDA receptors in visual cortex in vivo. Nat. Neurosci. 2, 352–357 12 Philpot, B.D. et al. (2001) Visual experience and deprivation bidirectionally modify the composition and function of NMDA receptors in visual cortex. Neuron 29,157–169 13 Barth, A.L. and Malenka, R.C. (2001) NMDAR EPSC kinetics do not regulate the critical period for LTP at thalamocortical synapses. Nat. Neurosci. 4, 235–236 14 Husi, H. et al. (2000) Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat. Neurosci. 3, 661–669 15 Willis, E.P. et al. (1997) PKA isoforms, neural pathways, and behavour: making the connection. Curr. Opin. Neurobiol. 7, 397–403 16 Xia, Z. and Storm, D.R. (1997) Calmodulinregulated adenylyl cyclases and neuromodulation. Curr. Opin. Neurobiol. 7, 391–396 17 Abdel-Majid, R.M. et al. (1998) Loss of adenylyl cyclase I activity disrupts patterning of mouse somatosensory cortex. Nat. Genet. 19, 289–291 18 Beaver, C.J. et al. (2001) Cyclic AMP-dependent protein kinase mediates ocular dominance shifts in cat visual cortex. Nat. Neurosci. 4, 159–163 19 Banke, T.G. et al. (2000) Control of GluR1 AMPA receptor function by cAMP-dependent protein kinase. J. Neurosci. 20, 89–102 20 Ehlers, M.D. (2000) Reinsertion or degradation of AMPA receptors determined by activitydependent endocytic sorting. Neuron 28, 511–525 21 Lee, H.K. et al. (2000) Regulation of distinct AMPA receptor phosphorylation sites during bidirectional synaptic plasticity. Nature 405, 955–959 22 Roberson, E.D. and Sweatt, J.D. (1996) Transient activation of cyclic AMP-dependent protein kinase

Research Update

23

24

25

26

during hippocampal long-term potentiation. J. Biol. Chem. 271, 30436–30441 Wong, S.T. et al. (1999) Calcium-stimulated adenylyl cyclase activity is critical for hippocampus-dependent long-term memory and late phase LTP. Neuron 23, 787–798 Feliciello, A. et al. (1999) Yotiao protein, a ligand for the NMDA receptor, binds and targets cAMPdependent protein kinase II(1). FEBS Lett. 464, 174–178 Westphal, R.S. et al. (1999) Regulation of NMDA receptors by an associated phosphatase-kinase signaling complex. Science 285, 93–96 Reid, S.N. et al. (1996) cAMP levels increased by activation of metabotropic glutamate receptors correlate with visual plasticity. J. Neurosci. 16, 7619–7626

TRENDS in Neurosciences Vol.24 No.10 October 2001

27 Bear, M.F. and Singer, W. (1986) Modulation of visual cortical plasticity by acetylcholine and noradrenaline. Nature 320, 172–176 28 Gu, Q. and Singer, W. (1995) Involvement of serotonin in developmental plasticity of kitten visual cortex. Eur. J. Neurosci. 7, 1146–1153 29 Hannan, A.J. et al. (2001) Phsopholipase C-(1, activated via mGluRs, mediates activitydependent differentiation in cerebral cortex. Nat. Neurosci. 4, 282–288 30 Chavez-Noriega, L.E. and Stevens, C.F. (1994) Increased transmitter release at excitatory synapses produced by direct activation of adenylate cyclase in rat hippocampal slices. J. Neurosci. 14, 310–317 31 Trudeau, L.E. (1996) Direct modulation of the secretory machinery underlies PKA-dependent

555

synaptic facilitation in hippocampal neurons. Neuron 17, 789–797

Peter C. Kind Dept of Biomedical Sciences, Edinburgh University, Hugh Robson Building, George Square, Edinburgh, UK EH8 9XD. e-mail: [email protected] Paul E. Neumann Dept of Anatomy and Neurobiology, Faculty of Medicine, Dalhousie University, Sir Charles Tupper Medical Building, Halifax, Nova Scotia B3H 4H7 Canada.

Dual use of the transcriptional repressor (CtBP2)/ribbon synapse (RIBEYE) gene: how prevalent are multifunctional genes? Joram Piatigorsky Vertebrates have ribbon synapses in the retina and in other sensory structures that are specialized for rapid, tonic release of synaptic vesicles1. The lamellar sheets of the ribbon situated at right angles to the plasma membrane are lined with synaptic vesicles that undergo exocytosis under the influence of Ca2++. Synaptic ribbons act as a conveyer belt to accelerate the release of this ready supply of synaptic vesicles at the presynaptic membranes. Although the protein composition of the terminals of ribbon synapses is generally similar to that of ordinary synapses in nervous tissue, much less is known about the composition of the ribbons themselves. RIM, a universal component of presynaptic active zones that interacts with rab3 on the synaptic vesicle, has been localized to the ribbons2. In addition, the kinesin motor protein, KIF3A, is associated with the ribbons and other organelles in presynaptic nerve terminals3. Recently, a ~120 kDa protein called RIBEYE has been identified in purified ribbons of bovine retina. The RIBEYE cDNA was cloned and its gene identified in the database.

RIBEYE turns out to contain a novel N-terminal domain (domain A) fused to the transcriptional repressor, CtBP2 (domain B); both CtBP2 and RIBEYE are encoded in the same gene (see Fig. 1). RIBEYE is expressed in the retina by utilization of a tissue-specific promoter http://tins.trends.com

within intron 1 of the CtBP2/RIBEYE gene, whereas CtBP2 is expressed ubiquitously by using a different 5′ promoter. Thus, exon 1 encoding the Nterminal sequence of CtBP2 is not expressed when the CtBP2/RIBEYE gene produces RIBEYE (Fig. 1b). Alternative RNA splicing eliminates the exon encoding domain A of RIBEYE when this gene expresses CtBP2 (Fig. 1b). Schmitz et al.1 also showed that CtBP2 is homologous to D-isomer-specific 2hydroxyacid dehydrogenase and binds NAD+, consistent with the still unproven possibility that it has a functionally significant enzymatic activity. Thus, the two proteins encoded in the CtBP2/RIBEYE gene appear to have derived from a dehydrogenase family member. Much interest has focussed on sequencing the genomes of different species. An unexpected finding is the relatively low estimate of protein-coding genes in humans (~35 000), indeed, this is only approximately twice that of worms or flies (see Refs 4,5). The current data indicate that alternative splicing leads to greater protein diversity in humans compared with invertebrates. Although not unique (see Ref. 6), the CtBP2/RIBEYE gene is a fascinating example of alternative promoters and alternative splicing generating two functionally different proteins in different

tissues: one (CtBP2) is used ubiquitously for transcriptional control and the other (RIBEYE) is expressed in a tissue-specific fashion and used for synaptic vesicle transport and exocytosis. Such dual use of a gene increases estimations of gene numbers as defined by functional products. Should CtBP2/RIBEYE be considered as two overlapping genes in spite of the integrated nature of the coding sequences? There is a growing list of genes fulfilling more than one role (see Ref. 7). These multifunctional genes do not always invoke alternative RNA splicing. The diverse lens crystallins that have been recruited from stress proteins and metabolic enzymes are, with rare exception, examples of identical proteins whose functions are affected strictly by tissue differences in gene expression in the absence of alternative splicing (see Refs 8,9). In contrast to the CtBP2/RIBEYE gene, the genes encoding crystallins generally use the same transcription initiation site and exons regardless of the function of the encoded protein, which might be an enzyme or stress protein in non-ocular tissues and a refractive protein in the lens. This dual use of the identical protein has been called gene sharing (see Ref. 10) and shows, similar to RIBEYE, the importance of gene regulation for innovating and directing protein function.

0166-2236/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0166-2236(00)01894-4