- Email: [email protected]
Protein phosphatase 1 and memory: practice makes PP1 imperfect?
Update
TRENDS in Neurosciences Vol.26 No.3 March 2003
. 100-fold (, 40 dB) and, thus, establish that prestin is required for normal auditory function. Concluding remarks There is now compelling evidence that prestin is both necessary and sufficient to produce somatic electromotility. Yet, just as a piston requires an engine block to make a functional motor, prestin may require other players to make outer hair cells motile. One possible contributor might be the membrane-bound sugar transporter, GLUT-5 [15]. The finding that sugar transport alters the motile properties and non-linear capacitance of outer hair cells suggests that GLUT-5 might play a role, perhaps as a coassembly partner or as an abundant membrane protein [12], providing a stiff substrate for prestin to push against. Deletion of the gene that encodes GLUT-5 in outer hair cells could offer insight. Liberman et al. have clearly demonstrated that prestin is necessary for cochlear amplification. What is not yet clear is whether outer hair cell electromotility is sufficient to produce cochlear amplification on its own or if other elements are required. The cochlear amplifier acts through positive feedback (Figs 1b,c). If any link in the positive feedback loop is broken, or even if the gain is reduced just a little, amplification will not occur. Whether prestin is providing the cycle-by-cycle work of the cochlear amplifier remains to be seen. For example, it remains possible that prestin provides the necessary stiffness or length to the lateral membrane of the outer hair cell but that the actual cycle-by-cycle work is done by active processes in the hair bundle. One line of evidence that supports the involvement of hair bundle mechanics is the presence of cochlear amplification in birds and reptiles, all of which lack outer hair cells and somatic motility [16]. How might the contributions from these distinct mechanisms be teased apart in mammals? Generation, by targeted gene replacement, of a mouse that expresses normal levels of a mutant prestin protein that is non-motile but that provides the proper length and stiffness to the outer hair cell lateral membranes might provide the answer. Amplification that remained would have to be attributed to other sources. Lastly, prestin is likely to emerge as a candidate on the growing list of genes that cause non-syndromic deafness in humans. These results suggest that stop, frame-shift or
117
null mutations might confer a recessive phenotype that would include elevated auditory thresholds. Liberman et al. found that, although the heterozygotes had half the pres mRNA and half the outer hair cell motility of wildtype mice, they had minimal loss in auditory sensitivity (,6 dB). This raises the possibility that gene transfer of wild-type pres might restore function in the prestin null mice, thereby providing a valuable paradigm to test gene therapeutic approaches to treat inner ear dysfunction. References 1 Brownell, W.E. (1999) How the ear works – nature’s solutions for listening. Volta Review 99, 9 – 28 2 Liberman, M.C. et al. (2002) Prestin is required for electromotility of the outer hair cell and for the cochlear amplifier. Nature 419, 300 – 304 3 Davis, H. (1983) An active process in cochlear mechanics. Hear Res. 9, 79 – 90 4 Hudspeth, A.J. et al. (2000) Putting ion channels to work: mechanoelectrical transduction, adaptation, and amplification by hair cells. Proc. Natl. Acad. Sci. U. S. A. 24, 11765 – 11772 5 Fettiplace, R. et al. (2001) Clues to the cochlear amplifier from the turtle ear. Trends Neurosci. 24, 169– 175 6 Ashmore, J.F. (2002) Biophysics of the cochlea – biomechanics and ion channelopathies. Br. Med. Bull. 63, 59– 72 7 Frolenkov, G.I. et al. (1998) The membrane-based mechanism of cell motility in cochlear outer hair cells. Mol. Biol. Cell 9, 1961– 1968 8 Brownell, W.E. et al. (1985) Evoked mechanical responses of isolated cochlear outer hair cells. Science 227, 194– 196 9 Ashmore, J.F. (1990) Forward and reverse transduction in the mammalian cochlea. Neurosci. Res. (Suppl.) 12, S39 – S50 10 Kalinec, F. et al. (1992) A membrane-based force generation mechanism in auditory sensory cells. Proc. Natl. Acad. Sci. U. S. A. 89, 8671 – 8675 11 Zheng, J. et al. (2000) Prestin is the motor protein of cochlear outer hair cells. Nature 405, 149 – 155 12 Belyantseva, I.A. et al. (2000) Expression and localization of prestin and the sugar transporter GLUT-5 during development of electromotility in cochlear outer hair cells. J. Neurosci. 20, RC116 13 Oliver, D. et al. (2001) Intracellular anions as the voltage sensor of prestin, the outer hair cell motor protein. Science 292, 2340 – 2343 14 Dallos, P. and Fakler, B. (2002) Prestin, a new type of motor protein. Nat. Rev. Mol. Cell. Biol. 3, 104– 111 15 Ge´le´oc, G.S. et al. (1999) A sugar transporter as a candidate for the outer hair cell motor. Nat. Neurosci. 2, 713– 719 16 Manley, G.A. (2001) Evidence for an active process and a cochlear amplifier in nonmammals. J. Neurophysiol. 86, 541 – 549
0166-2236/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0166-2236(03)00030-4
Protein phosphatase 1 and memory: practice makes PP1 imperfect? Scott Waddell Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA 01605, USA.
Long-lasting memories are most efficiently formed by multiple training sessions separated by appropriately timed intervals. A recent study revealed that expression Corresponding author: Scott Waddell ([email protected]). http://tins.trends.com
of a transgene encoding an inhibitor of protein phosphatase 1 (PP1) in the forebrain enhanced memory formed during sub-optimal training. Thus, PP1 apparently constrains memory formation in the mouse. Furthermore, the report proposes that PP1 promotes forgetting.
118
Update
TRENDS in Neurosciences Vol.26 No.3 March 2003
A conserved feature of learning and memory is that appropriate spacing of training sessions improves memory. Subjects given time in between training sessions generally remember better than those whose training sessions were crammed together [1]. Furthermore, spaced training appears to be essential for the formation of long-lasting memories in several model systems [2– 4]. Studies suggest that spaced training regimes cause persistent activation of cAMP-dependent protein kinase (PKA) which, in turn, activates the cAMP-responsive transcription factor CREB and so drives the expression of genes necessary for the long-term synaptic changes that engender long-lasting memory [5]. However, other than PKA and CREB, what else does spaced training trigger? The phosphorylation status of a protein is a balance between its kinases and phosphatases. The serine/threonine kinases PKA, Ca2þ –calmodulin-dependent protein kinase II (CaMKII) and mitogen-activated protein kinase (MAPK) are crucial for learning and memory and for several forms of synaptic plasticity [5,6]. Phosphatases such as protein phosphatase 1 (PP1) and the Ca2þ –calmodulin-dependent protein phosphatase calcineurin (CN) might inhibit memory and synaptic plasticity. In a recent study, Genoux et al. [7] demonstrated that tilting the balance away from PP1 allows what is usually an insufficient training protocol to induce a strong memory. A previous landmark study revealed that transgenic inhibition of CN (also called PP2B) in the forebrain enhanced learning and memory, in addition to enhancing hippocampal long-term potentiation (LTP, a cellular model of memory) [8]. These enhancements were PKA-dependent [7]. These findings are relevant to the study by Genoux et al. [7] because both CN and PKA indirectly regulate PP1 by controlling the phosphorylation and dephosphorylation of a PP1-inhibitory protein called inhibitor 1 (I-1) (Fig. 1) [9]. Only phosphorylated I-1 inhibits PP1, so inhibiting CN activity is expected to inhibit PP1 activity (among other things). However, forebrain-specific knockout of the gene encoding CN has no effect on water-maze learning, although it does impair working memory and long-term synaptic depression (LTD) [10]. One might expect that memory requires both concomitant phosphorylation and
CN I-1
I-1 P
PKA
PP1
TRENDS in Neurosciences
Fig. 1. The known relationship between calcineurin (CN), cAMP-dependent protein kinase (PKA), inhibitor 1 (I-1) and protein phosphatase 1 (PP1). PKA phosphorylates I-1 and CN dephosphorylates I-1 at the same site. I-1 inhibits PP1 only when phosphorylated. Genoux et al. used a constitutively active I-1 variant, I-1*, that carries a threonine-to-aspartic-acid substitution at the PKA phosphorylation site to mimic phosphorylation by PKA. http://tins.trends.com
dephosphorylation, as well as LTP and LTD in different paths. So what does transgenic inhibition of PP1 do? Transgenic interference with PP1 Genoux et al. [7] engineered mice in which expression of a transgene encoding constitutively active inhibitor 1 (I-1*) could be switched on in the forebrain of adult animals by feeding them doxycycline. Expression of I-1* could subsequently be switched off if the mice were deprived of doxycycline. This technology permitted analysis of I-1 function in the adult mouse without any potentially confounding developmental effects. I-1* transgenic mice display improved performance in an object-recognition task The memory of the I-1* mice was tested in an objectrecognition task in which a mouse is allowed to explore repeatedly, and become familiar with, three objects in a box at defined times [11]. The mouse is later returned to the box, but one of the objects has been replaced with a novel object. The amount of time the mouse spends exploring the new object is indicative of its memory of the familiar objects. Genoux et al. [7] investigated the involvement of I-1 in spaced training by varying the number and duration of training episodes. Wild-type mice showed greatest preference for the novel object under a long inter-trial interval training regime consisting of five bouts of 5 min at 15 min intervals. A block training session of 25 min of continuous training, or a short inter-trial interval training regime offive bouts of 5 min at 5 min intervals, produced only moderate discrimination between new and old objects in control mice. Surprisingly, the short-interval regime trained the I-1* transgenic mice as efficiently as did training control or transgenic mice with the long-interval regime. This result is unique because there was no apparent difference between wild-type and transgenic mice given block training or longinterval training. This enhancement persisted for 24 h and was reversed by inhibiting transgene expression. The authors show that, as expected, I-1* expression significantly reduced PP1 activity in the hippocampus and cortex but had no effect on activity of either calcineurin or protein phosphatase 2A. Thus, enhanced memory correlates with a transgene-dependent reduction in PP1 activity. But does training normally alter PP1 activity? Genoux et al. [7] tested this by measuring cortical PP1 enzymatic activity in wild-type mice following block, short-interval and longinterval training. Strikingly, only the long-interval training significantly reduced global cortical PP1 enzyme activity. These results are consistent with the hypothesis that PP1 constitutes a memory constraint that is removed by training protocols sufficient to induce long-term memory. Which memory-relevant targets does PP1 dephosphorylate? One obvious target is CREB, and Genoux et al. [7] tested CREB activation with a CREB-responsive lacZ reporter gene [12]. Counting the number of lacZ-positive cells per unit area of cortex revealed that CREB-mediated gene expression is efficiently induced by long-interval training in both I-1*transgenic and control mice. However, short-interval training increased the number of lacZ-positive cells only in the I-1*-transgenic mice [7]. Therefore, enhanced memory following short-interval training correlated with increased
Update
TRENDS in Neurosciences Vol.26 No.3 March 2003
CREB transcription activity. This implies that induction of I-1* permits a step involved in formation of long-term memory to proceed in short-interval-trained mice that would normally only occur in long-interval-trained mice. However, it seems unlikely that new gene expression contributes to enhancing memory only 5 min after training. In addition, because CREB activation appeared to be widespread across the cortex, it would be informative to know which of these brain areas are relevant for long-term memory in the object recognition task employed. I-1* transgenic mice apparently remember for longer Genoux et al. [7] also tested I-1* transgenic mice in the Morris water maze, in which mice learn the location of an escape platform in a tank of opaque water, aided by spatial cues [13]. I-1* transgenic mice learned faster than control mice, needing fewer training trials before they reached their quickest escape time, consistent with their enhanced memory for objects. This improved performance in I-1* transgenic mice correlated with a modest increase in the phosphorylation of CaMKII and the GluR1 subunit of the AMPA-type glutamate receptor – two factors more likely to be involved in the immediate effects of training. There was no difference between I-1*-transgenic and control mice when they were trained repetitively with a spaced protocol in the water maze. However, there was a startling effect on the persistence of memory. I-1* mice displayed a transgene-expression-dependent reduction in memory decay – that is, they remember for longer or forget less. This reduced decay was also evident if I-1* was induced only after training, suggesting that PP1 plays a role in maintenance of memory. However, because the memory decay is monitored by testing the same mice over and over again for the platform location (without it being there) it is also plausible that I-1* transgenic mice are impaired in extinction of the spatial memory. Put differently, perhaps they are unable to learn that the platform is no longer where it used to be. Older animals were also tested for water-maze learning and memory. Aged I-1* mice learned slightly faster than similar-age-control mice. Furthermore, aged control mice had no apparent memory for the platform location one week after training, whereas aged I-1* mice remembered for a month. The authors suggested that this is further evidence that PP1 promotes forgetting. If this role of PP1
119
were normally relevant to memory decline, aged control mice would be expected to exhibit higher PP1 activity. The authors mention that I-1 might affect PP1 independent processes, which is certainly a caveat of the study. Concluding that all observed effects are PP1dependent is a small leap of faith. Nevertheless, the results provide good evidence for a role of I-1 itself in memory, which is supported by previous data showing that PP1 regulates LTD [14]. It will be interesting to determine whether improved memory of I-1* transgenic mice also correlates with altered neuronal plasticity. Perhaps physiology will further enlighten us as to why practice makes perfect. References 1 Spear, N.E. (1978) The Processing of Memories: Forgetting and Retention, Erlbaum 2 Dash, P.K. (1990) Injection of the cAMP-responsive element into the nucleus of Aplysia sensory neurons blocks long-term facilitation. Nature 345, 718 – 721 3 Tully, T. et al. (1994) Genetic dissection of consolidated memory in Drosophila. Cell 79, 35– 47 4 Muller, U. (2000) Prolonged activation of cAMP-dependent protein kinase during conditioning induces long-term memory in honeybees. Neuron 27, 159 – 168 5 Mayford, M. and Kandel, E.R. (1999) Genetic approaches to memory storage. Trends Genet. 15, 463– 470 6 McGaugh, J.L. (2000) Memory – a century of consolidation. Science 287, 248 – 251 7 Genoux, D. et al. (2002) Protein phosphatase 1 is a molecular constraint on learning and memory. Nature 418, 970– 975 8 Malleret, G. et al. (2001) Inducible and reversible enhancement of learning, memory, and long-term potentiation by genetic inhibition of calcineurin. Cell 104, 675 – 686 9 Oliver, C.J. and Shenolikar, S. (1998) Physiologic importance of protein phosphatase inhibitors. Front Biosci. 3, D961 – D972 10 Zeng, H. et al. (2001) Forebrain-specific calcineurin knockout selectively impairs bidirectional synaptic plasticity and working/ episodic-like memory. Cell 107, 617– 629 11 Save, E. et al. (1992) Object exploration and reactions to spatial and nonspatial changes in hooded rats following damage to parietal cortex or hippocampal formation. Behav. Neurosci. 106, 447 – 456 12 Impey, S. et al. (1996) Induction of CRE-mediated gene expression by stimuli that generate long-lasting LTP in area CA1 of the hippocampus. Neuron 16, 973– 982 13 Morris, R.G. et al. (1982) Place navigation impaired in rats with hippocampal lesions. Nature 297, 681– 683 14 Morishita, W. et al. (2001) Regulation of synaptic strength by protein phosphatase 1. Neuron 32, 1133– 1148 0166-2236/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0166-2236(03)00029-8
Receptor tyrosine kinase transactivation: fine-tuning synaptic transmission Stephen S.G. Ferguson Cell Biology Research Group, Robarts Research Institute, Department of Physiology and Pharmacology, University of Western Ontario, 100 Perth Drive, PO Box 5015, London, Ontario, Canada, N6A 5K8
G-protein-coupled receptors generate signals that promote gene transcription through the ‘transactivation’ of Corresponding author: Stephen S.G. Ferguson ([email protected]). http://tins.trends.com
receptor tyrosine kinases (RTKs) and activation of the mitogen-activated protein kinase (MAPK) cascade – a process that involves RTK autophosphorylation and endocytosis. Pioneering work now suggests that
TRENDS in Neurosciences Vol.26 No.3 March 2003
. 100-fold (, 40 dB) and, thus, establish that prestin is required for normal auditory function. Concluding remarks There is now compelling evidence that prestin is both necessary and sufficient to produce somatic electromotility. Yet, just as a piston requires an engine block to make a functional motor, prestin may require other players to make outer hair cells motile. One possible contributor might be the membrane-bound sugar transporter, GLUT-5 [15]. The finding that sugar transport alters the motile properties and non-linear capacitance of outer hair cells suggests that GLUT-5 might play a role, perhaps as a coassembly partner or as an abundant membrane protein [12], providing a stiff substrate for prestin to push against. Deletion of the gene that encodes GLUT-5 in outer hair cells could offer insight. Liberman et al. have clearly demonstrated that prestin is necessary for cochlear amplification. What is not yet clear is whether outer hair cell electromotility is sufficient to produce cochlear amplification on its own or if other elements are required. The cochlear amplifier acts through positive feedback (Figs 1b,c). If any link in the positive feedback loop is broken, or even if the gain is reduced just a little, amplification will not occur. Whether prestin is providing the cycle-by-cycle work of the cochlear amplifier remains to be seen. For example, it remains possible that prestin provides the necessary stiffness or length to the lateral membrane of the outer hair cell but that the actual cycle-by-cycle work is done by active processes in the hair bundle. One line of evidence that supports the involvement of hair bundle mechanics is the presence of cochlear amplification in birds and reptiles, all of which lack outer hair cells and somatic motility [16]. How might the contributions from these distinct mechanisms be teased apart in mammals? Generation, by targeted gene replacement, of a mouse that expresses normal levels of a mutant prestin protein that is non-motile but that provides the proper length and stiffness to the outer hair cell lateral membranes might provide the answer. Amplification that remained would have to be attributed to other sources. Lastly, prestin is likely to emerge as a candidate on the growing list of genes that cause non-syndromic deafness in humans. These results suggest that stop, frame-shift or
117
null mutations might confer a recessive phenotype that would include elevated auditory thresholds. Liberman et al. found that, although the heterozygotes had half the pres mRNA and half the outer hair cell motility of wildtype mice, they had minimal loss in auditory sensitivity (,6 dB). This raises the possibility that gene transfer of wild-type pres might restore function in the prestin null mice, thereby providing a valuable paradigm to test gene therapeutic approaches to treat inner ear dysfunction. References 1 Brownell, W.E. (1999) How the ear works – nature’s solutions for listening. Volta Review 99, 9 – 28 2 Liberman, M.C. et al. (2002) Prestin is required for electromotility of the outer hair cell and for the cochlear amplifier. Nature 419, 300 – 304 3 Davis, H. (1983) An active process in cochlear mechanics. Hear Res. 9, 79 – 90 4 Hudspeth, A.J. et al. (2000) Putting ion channels to work: mechanoelectrical transduction, adaptation, and amplification by hair cells. Proc. Natl. Acad. Sci. U. S. A. 24, 11765 – 11772 5 Fettiplace, R. et al. (2001) Clues to the cochlear amplifier from the turtle ear. Trends Neurosci. 24, 169– 175 6 Ashmore, J.F. (2002) Biophysics of the cochlea – biomechanics and ion channelopathies. Br. Med. Bull. 63, 59– 72 7 Frolenkov, G.I. et al. (1998) The membrane-based mechanism of cell motility in cochlear outer hair cells. Mol. Biol. Cell 9, 1961– 1968 8 Brownell, W.E. et al. (1985) Evoked mechanical responses of isolated cochlear outer hair cells. Science 227, 194– 196 9 Ashmore, J.F. (1990) Forward and reverse transduction in the mammalian cochlea. Neurosci. Res. (Suppl.) 12, S39 – S50 10 Kalinec, F. et al. (1992) A membrane-based force generation mechanism in auditory sensory cells. Proc. Natl. Acad. Sci. U. S. A. 89, 8671 – 8675 11 Zheng, J. et al. (2000) Prestin is the motor protein of cochlear outer hair cells. Nature 405, 149 – 155 12 Belyantseva, I.A. et al. (2000) Expression and localization of prestin and the sugar transporter GLUT-5 during development of electromotility in cochlear outer hair cells. J. Neurosci. 20, RC116 13 Oliver, D. et al. (2001) Intracellular anions as the voltage sensor of prestin, the outer hair cell motor protein. Science 292, 2340 – 2343 14 Dallos, P. and Fakler, B. (2002) Prestin, a new type of motor protein. Nat. Rev. Mol. Cell. Biol. 3, 104– 111 15 Ge´le´oc, G.S. et al. (1999) A sugar transporter as a candidate for the outer hair cell motor. Nat. Neurosci. 2, 713– 719 16 Manley, G.A. (2001) Evidence for an active process and a cochlear amplifier in nonmammals. J. Neurophysiol. 86, 541 – 549
0166-2236/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0166-2236(03)00030-4
Protein phosphatase 1 and memory: practice makes PP1 imperfect? Scott Waddell Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA 01605, USA.
Long-lasting memories are most efficiently formed by multiple training sessions separated by appropriately timed intervals. A recent study revealed that expression Corresponding author: Scott Waddell ([email protected]). http://tins.trends.com
of a transgene encoding an inhibitor of protein phosphatase 1 (PP1) in the forebrain enhanced memory formed during sub-optimal training. Thus, PP1 apparently constrains memory formation in the mouse. Furthermore, the report proposes that PP1 promotes forgetting.
118
Update
TRENDS in Neurosciences Vol.26 No.3 March 2003
A conserved feature of learning and memory is that appropriate spacing of training sessions improves memory. Subjects given time in between training sessions generally remember better than those whose training sessions were crammed together [1]. Furthermore, spaced training appears to be essential for the formation of long-lasting memories in several model systems [2– 4]. Studies suggest that spaced training regimes cause persistent activation of cAMP-dependent protein kinase (PKA) which, in turn, activates the cAMP-responsive transcription factor CREB and so drives the expression of genes necessary for the long-term synaptic changes that engender long-lasting memory [5]. However, other than PKA and CREB, what else does spaced training trigger? The phosphorylation status of a protein is a balance between its kinases and phosphatases. The serine/threonine kinases PKA, Ca2þ –calmodulin-dependent protein kinase II (CaMKII) and mitogen-activated protein kinase (MAPK) are crucial for learning and memory and for several forms of synaptic plasticity [5,6]. Phosphatases such as protein phosphatase 1 (PP1) and the Ca2þ –calmodulin-dependent protein phosphatase calcineurin (CN) might inhibit memory and synaptic plasticity. In a recent study, Genoux et al. [7] demonstrated that tilting the balance away from PP1 allows what is usually an insufficient training protocol to induce a strong memory. A previous landmark study revealed that transgenic inhibition of CN (also called PP2B) in the forebrain enhanced learning and memory, in addition to enhancing hippocampal long-term potentiation (LTP, a cellular model of memory) [8]. These enhancements were PKA-dependent [7]. These findings are relevant to the study by Genoux et al. [7] because both CN and PKA indirectly regulate PP1 by controlling the phosphorylation and dephosphorylation of a PP1-inhibitory protein called inhibitor 1 (I-1) (Fig. 1) [9]. Only phosphorylated I-1 inhibits PP1, so inhibiting CN activity is expected to inhibit PP1 activity (among other things). However, forebrain-specific knockout of the gene encoding CN has no effect on water-maze learning, although it does impair working memory and long-term synaptic depression (LTD) [10]. One might expect that memory requires both concomitant phosphorylation and
CN I-1
I-1 P
PKA
PP1
TRENDS in Neurosciences
Fig. 1. The known relationship between calcineurin (CN), cAMP-dependent protein kinase (PKA), inhibitor 1 (I-1) and protein phosphatase 1 (PP1). PKA phosphorylates I-1 and CN dephosphorylates I-1 at the same site. I-1 inhibits PP1 only when phosphorylated. Genoux et al. used a constitutively active I-1 variant, I-1*, that carries a threonine-to-aspartic-acid substitution at the PKA phosphorylation site to mimic phosphorylation by PKA. http://tins.trends.com
dephosphorylation, as well as LTP and LTD in different paths. So what does transgenic inhibition of PP1 do? Transgenic interference with PP1 Genoux et al. [7] engineered mice in which expression of a transgene encoding constitutively active inhibitor 1 (I-1*) could be switched on in the forebrain of adult animals by feeding them doxycycline. Expression of I-1* could subsequently be switched off if the mice were deprived of doxycycline. This technology permitted analysis of I-1 function in the adult mouse without any potentially confounding developmental effects. I-1* transgenic mice display improved performance in an object-recognition task The memory of the I-1* mice was tested in an objectrecognition task in which a mouse is allowed to explore repeatedly, and become familiar with, three objects in a box at defined times [11]. The mouse is later returned to the box, but one of the objects has been replaced with a novel object. The amount of time the mouse spends exploring the new object is indicative of its memory of the familiar objects. Genoux et al. [7] investigated the involvement of I-1 in spaced training by varying the number and duration of training episodes. Wild-type mice showed greatest preference for the novel object under a long inter-trial interval training regime consisting of five bouts of 5 min at 15 min intervals. A block training session of 25 min of continuous training, or a short inter-trial interval training regime offive bouts of 5 min at 5 min intervals, produced only moderate discrimination between new and old objects in control mice. Surprisingly, the short-interval regime trained the I-1* transgenic mice as efficiently as did training control or transgenic mice with the long-interval regime. This result is unique because there was no apparent difference between wild-type and transgenic mice given block training or longinterval training. This enhancement persisted for 24 h and was reversed by inhibiting transgene expression. The authors show that, as expected, I-1* expression significantly reduced PP1 activity in the hippocampus and cortex but had no effect on activity of either calcineurin or protein phosphatase 2A. Thus, enhanced memory correlates with a transgene-dependent reduction in PP1 activity. But does training normally alter PP1 activity? Genoux et al. [7] tested this by measuring cortical PP1 enzymatic activity in wild-type mice following block, short-interval and longinterval training. Strikingly, only the long-interval training significantly reduced global cortical PP1 enzyme activity. These results are consistent with the hypothesis that PP1 constitutes a memory constraint that is removed by training protocols sufficient to induce long-term memory. Which memory-relevant targets does PP1 dephosphorylate? One obvious target is CREB, and Genoux et al. [7] tested CREB activation with a CREB-responsive lacZ reporter gene [12]. Counting the number of lacZ-positive cells per unit area of cortex revealed that CREB-mediated gene expression is efficiently induced by long-interval training in both I-1*transgenic and control mice. However, short-interval training increased the number of lacZ-positive cells only in the I-1*-transgenic mice [7]. Therefore, enhanced memory following short-interval training correlated with increased
Update
TRENDS in Neurosciences Vol.26 No.3 March 2003
CREB transcription activity. This implies that induction of I-1* permits a step involved in formation of long-term memory to proceed in short-interval-trained mice that would normally only occur in long-interval-trained mice. However, it seems unlikely that new gene expression contributes to enhancing memory only 5 min after training. In addition, because CREB activation appeared to be widespread across the cortex, it would be informative to know which of these brain areas are relevant for long-term memory in the object recognition task employed. I-1* transgenic mice apparently remember for longer Genoux et al. [7] also tested I-1* transgenic mice in the Morris water maze, in which mice learn the location of an escape platform in a tank of opaque water, aided by spatial cues [13]. I-1* transgenic mice learned faster than control mice, needing fewer training trials before they reached their quickest escape time, consistent with their enhanced memory for objects. This improved performance in I-1* transgenic mice correlated with a modest increase in the phosphorylation of CaMKII and the GluR1 subunit of the AMPA-type glutamate receptor – two factors more likely to be involved in the immediate effects of training. There was no difference between I-1*-transgenic and control mice when they were trained repetitively with a spaced protocol in the water maze. However, there was a startling effect on the persistence of memory. I-1* mice displayed a transgene-expression-dependent reduction in memory decay – that is, they remember for longer or forget less. This reduced decay was also evident if I-1* was induced only after training, suggesting that PP1 plays a role in maintenance of memory. However, because the memory decay is monitored by testing the same mice over and over again for the platform location (without it being there) it is also plausible that I-1* transgenic mice are impaired in extinction of the spatial memory. Put differently, perhaps they are unable to learn that the platform is no longer where it used to be. Older animals were also tested for water-maze learning and memory. Aged I-1* mice learned slightly faster than similar-age-control mice. Furthermore, aged control mice had no apparent memory for the platform location one week after training, whereas aged I-1* mice remembered for a month. The authors suggested that this is further evidence that PP1 promotes forgetting. If this role of PP1
119
were normally relevant to memory decline, aged control mice would be expected to exhibit higher PP1 activity. The authors mention that I-1 might affect PP1 independent processes, which is certainly a caveat of the study. Concluding that all observed effects are PP1dependent is a small leap of faith. Nevertheless, the results provide good evidence for a role of I-1 itself in memory, which is supported by previous data showing that PP1 regulates LTD [14]. It will be interesting to determine whether improved memory of I-1* transgenic mice also correlates with altered neuronal plasticity. Perhaps physiology will further enlighten us as to why practice makes perfect. References 1 Spear, N.E. (1978) The Processing of Memories: Forgetting and Retention, Erlbaum 2 Dash, P.K. (1990) Injection of the cAMP-responsive element into the nucleus of Aplysia sensory neurons blocks long-term facilitation. Nature 345, 718 – 721 3 Tully, T. et al. (1994) Genetic dissection of consolidated memory in Drosophila. Cell 79, 35– 47 4 Muller, U. (2000) Prolonged activation of cAMP-dependent protein kinase during conditioning induces long-term memory in honeybees. Neuron 27, 159 – 168 5 Mayford, M. and Kandel, E.R. (1999) Genetic approaches to memory storage. Trends Genet. 15, 463– 470 6 McGaugh, J.L. (2000) Memory – a century of consolidation. Science 287, 248 – 251 7 Genoux, D. et al. (2002) Protein phosphatase 1 is a molecular constraint on learning and memory. Nature 418, 970– 975 8 Malleret, G. et al. (2001) Inducible and reversible enhancement of learning, memory, and long-term potentiation by genetic inhibition of calcineurin. Cell 104, 675 – 686 9 Oliver, C.J. and Shenolikar, S. (1998) Physiologic importance of protein phosphatase inhibitors. Front Biosci. 3, D961 – D972 10 Zeng, H. et al. (2001) Forebrain-specific calcineurin knockout selectively impairs bidirectional synaptic plasticity and working/ episodic-like memory. Cell 107, 617– 629 11 Save, E. et al. (1992) Object exploration and reactions to spatial and nonspatial changes in hooded rats following damage to parietal cortex or hippocampal formation. Behav. Neurosci. 106, 447 – 456 12 Impey, S. et al. (1996) Induction of CRE-mediated gene expression by stimuli that generate long-lasting LTP in area CA1 of the hippocampus. Neuron 16, 973– 982 13 Morris, R.G. et al. (1982) Place navigation impaired in rats with hippocampal lesions. Nature 297, 681– 683 14 Morishita, W. et al. (2001) Regulation of synaptic strength by protein phosphatase 1. Neuron 32, 1133– 1148 0166-2236/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0166-2236(03)00029-8
Receptor tyrosine kinase transactivation: fine-tuning synaptic transmission Stephen S.G. Ferguson Cell Biology Research Group, Robarts Research Institute, Department of Physiology and Pharmacology, University of Western Ontario, 100 Perth Drive, PO Box 5015, London, Ontario, Canada, N6A 5K8
G-protein-coupled receptors generate signals that promote gene transcription through the ‘transactivation’ of Corresponding author: Stephen S.G. Ferguson ([email protected]). http://tins.trends.com
receptor tyrosine kinases (RTKs) and activation of the mitogen-activated protein kinase (MAPK) cascade – a process that involves RTK autophosphorylation and endocytosis. Pioneering work now suggests that