- Email: [email protected]
Is NMDA Receptor-Coincidence Detection Required for Learning and Memory?
Neuron
Previews Arendt, 2009; Milnerwood and Raymond, 2010).
Gubernator, N.G., Zhang, H., Staal, R.G., Mosharov, E.V., Pereira, D.B., Yue, M., Balsanek, V., Vadola, P.A., Mukherjee, B., Edwards, R.H., et al. (2009). Science 324, 1441–1444.
REFERENCES Agid, O., Kapur, S., Arenovich, T., and Zipursky, R.B. (2003). Arch. Gen. Psychiatry 60, 1228–1235. Arendt, T. (2009). Acta Neuropathol. 118, 167–179. Bear, M.F., Huber, K.M., and Warren, S.T. (2004). Trends Neurosci. 27, 370–377. Cousin, M.A., and Nicholls, D.G. (1997). J. Neurochem. 69, 1927–1935. Diamond, J.S., and Jahr, C.E. (1997). J. Neurosci. 17, 4672–4687.
Lisman, J.E., Coyle, J.T., Green, R.W., Javitt, D.C., Benes, F.M., Heckers, S., and Grace, A.A. (2008). Trends Neurosci. 31, 234–242. Milnerwood, A.J., and Raymond, L.A. (2010). Trends Neurosci. 33, 513–523. Ogata, N., and Narahashi, T. (1989). Br. J. Pharmacol. 97, 905–913. Raghavachari, S., and Lisman, J.E. (2004). J. Neurophysiol. 92, 2456–2467.
Rayport, S., and Sulzer, D. (1995). J. Neurochem. 65, 691–703. Sah, D.W., and Bean, B.P. (1994). Mol. Pharmacol. 45, 84–92. Sankaranarayanan, S., and Ryan, T.A. (2000). Nat. Cell Biol. 2, 197–204. Strange, P.G. 119–133.
(2001).
Pharmacol.
Rev.
53,
Tischbirek, C.H., Wenzel, E.M., Zheng, F., Huth, T., Amato, D., Trapp, S., Denker, A., Welzel, O., Lueke, K., Svetlitchny, A., et al. (2012). Neuron 74, this issue, 830–844.
Is NMDA Receptor-Coincidence Detection Required for Learning and Memory? Christopher J. Tabone1,* and Mani Ramaswami1,* 1Trinity College Institute for Neuroscience, Smurfit Institute of Genetics and the Department of Zoology, Trinity College Dublin, Dublin-2, Ireland *Correspondence: [email protected] (C.J.T.), [email protected] (M.R.) DOI 10.1016/j.neuron.2012.05.008
The Mg2+ block of NMDA-type glutamate receptors (NMDARs) is crucial to their function as synaptic coincidence detectors. An analysis of Drosophila expressing a Mg2+-independent NMDAR by Miyashita et al. (2012) in this issue of Neuron concludes that the Mg2+ block is required primarily for long-term memory. Donald Hebb first proposed that synapses between two neurons would be strengthened if they showed coincident activity. This idea was hugely influential because such ‘‘Hebbian’’ plasticity could theoretically explain how memories formed, particularly associations between temporally linked events. Subsequently, Bliss and Lomo (1973) discovered long-term potentiation (LTP), a phenomenon in which synaptic strength is enhanced following bursts of synaptic activity. Thus, LTP gained particular notoriety as one of the underlying mechanisms of learning and memory and considerable effort was focused on unraveling mechanisms of coincidence detection and the subsequent synaptic plasticity. From these studies, NMDAtype glutamate receptors (NMDARs) emerged as a class of ionotropic receptors whose pharmacological or genetic
perturbations disrupted both LTP and learning and memory (Traynelis et al., 2010). NMDARs are now understood as pivotal molecules required for coincidence detection, synaptic plasticity, and learning and memory in the central nervous system (CNS). A voltage-dependent Mg2+ block of NMDARs allows them to function as Hebbian coincidence detectors (Mayer et al., 1984; Nowak et al., 1984). Binding by glutamate alone is insufficient for channel activation as Mg2+ remains bound to a site in the channel pore, effectively blocking ion transport. Eviction of this Mg2+ ion additionally requires membrane depolarization. Thus, the coincidence of presynaptic glutamate release and strong depolarizing potential in the postsynaptic neuron is required for the opening of NMDAR channels. Subsequent Ca2+ influx through the
open channel serves as a trigger for synaptic plasticity. Mouse models with mutations specific to the NMDA Mg2+ block site result in developmental defects and/or defects in complex behavior, suggesting that coincidence detection is required for normal NMDAR function in vivo (Single et al., 2000; Rudhard et al., 2003). However, for two reasons, neither these studies nor the observations of abnormal LTP and learning in these mutant mice (Chen et al., 2009) directly address the role of coincidence detection in vivo. First, all known Mg2+ block mutations in murine NMDARs also decrease Ca2+ conductance. Thus, it is unclear whether the resultant phenotypes are due to Mg2+ block-specific effects or reduced calcium permeability. Second, because Mg2+ block mutants show severe developmental defects and early lethality, it is
Neuron 74, June 7, 2012 ª2012 Elsevier Inc. 767
Neuron
Previews Wild type
dNR1(N631Q)
Mg2+
dCREB 2-b
homer staufen activin
dCREB 2-b
poised for transcription
homer staufen activin
transcription inhibited
no LTM
permissive for LTM
Figure 1. The NMDA Receptor Mg2+ Block Regulates dCREB2-b Activity and Long-Term Memory Formation The Mg2+ block reduces baseline calcium entry and prevents basal activation of dCREB2-b, thereby keeping neurons in a state permissive for activity-induced transcription required for LTM. Removal of the NMDAR Mg2+ block in dNR1(N631Q) flies causes increased baseline calcium signaling that causes a 4-fold increase in dCREB2-b expression and a subsequent strong repression of the LTM-associated genes staufen, activin, and homer. This prevents CREB-dependent plasticity required for LTM formation. (Illustration is by Jens Hillebrand.)
difficult to exclude the possibility that defects in learning observed in NMDAR Mg2+ block mutants arise due to altered
nervous system development. Miyashita et al. (2012)’s experiments in Drosophila circumvent these confounding issues
768 Neuron 74, June 7, 2012 ª2012 Elsevier Inc.
and directly assess the role of the Mg2+ block in memory formation. Drosophila NMDARs, composed of two subunits, dNR1 and dNR2, are necessary for normal memory formation (Xia et al., 2005; Wu et al., 2007). In the relevant behavioral assay, flies learn to associate an olfactory cue with aversive electric foot shocks. The ability of the fly to demonstrate learned avoidance of the shock-associated odor is quantified as a ‘‘performance index’’ or PI. Singletraining trials produce short-term and middle-term memory (STM/MTM). Multiple, spaced training sessions additionally produce a protein synthesisdependent long-term memory (LTM) that requires the function of CREB2, the cAMP-regulatory element-binding transcription factor (Yin and Tully, 1996). The first behavioral study of the NMDAR in Drosophila indicated its requirement for olfactory learning: hypomorphic dNR1 mutants showed defects in learning and LTM (Xia et al., 2005). Further work showed that while NMDA receptors are required in the mushroom body for early phases of memory, they are additionally required in the ellipsoid body for LTM (Wu et al., 2007). In this intellectual context, Miyashita et al. (2012) investigate the role of coincidence detection specifically through the Mg2+ block mechanism of NMDAR for learning and memory. Miyashita et al. (2012) first constructed a dNR1 transgene encoding the N631Q mutant NMDAR, which specifically disrupts the Mg2+ block site in the encoded protein with no detectable effect on Ca2+ permeability, as assessed using electrophysiology and Ca2+imaging. Expression of the dNR1(N631Q) transgene in hypomorphic dNR1 flies rescued the mutant’s learning defect, indicating that the Mg2+ block was not required for NMDAR’s function in learning. However, in contrast, N631Q expression disrupted LTM formation after spaced training. The first conclusion from these experiments that the Mg2+ block is not essential for learning does not conflict with current models of NMDAR function. Coincidence detection is necessary to limit plasticity only to synapses that display coincident pre- and postsynaptic activity. Thus, by removing the second requirement for postsynaptic excitation in Mg2+ block mutants, plasticity may
Neuron
Previews occur (1) more easily and (2) in additional postsynaptic neurons that do not show synchronous depolarization, e.g., in mushroom-body neurons that are not activated by the odorant. Consistent with a prediction from (1), the authors observe slightly enhanced learning in flies expressing the mutant dNR1(N631Q) transgene. The prediction from (2), wherein plasticity occurs throughout a larger group of neurons, will require additional comprehensive testing, e.g., for increased odor generalization after olfactory aversive conditioning. How then does removal of the Mg2+ block interfere with the formation of LTM? Previous studies in Drosophila have indicated that LTM is modulated by the relative activity of the repressor isoform of CREB, dCREB2-b (Yin and Tully, 1996). Could enhanced calcium influx through NMDAR mutants freed from the Mg2+ block result in increased activity of dCREB2-b? To address this question, the authors examined transcripts of the repressor isoform dCREB2b extracted from heads of dNR1(N631Q) flies. They report a nearly 4-fold increase in the basal expression level of dCREB2b as compared to wild-type animals. This increase in dCREB2-b expression is also observed in individual wild-type brains cultured in Mg2+-free medium, indicating that increases in calcium influx in the absence of a Mg2+ block can directly lead to increased levels of dCREB2-b. If the absence of the Mg2+ block directly affects the expression of the dCREB2b repressor, then the expression of CREB2-target genes should be affected. The authors tested this hypothesis by examining expression levels of the genes activin, staufen, and homer, previously shown to be transcriptionally induced after spaced training (Dubnau et al., 2003). Remarkably, Miyashita et al. (2012) observed an absence of this upregulation in dNR1(N631Q) flies. The block in gene induction was cell autonomous: flies expressing dNR1(N631Q) in mushroom body neurons showed no increase in homer expression in mushroom
bodies but still displayed homer upregulation in other structures such as the protocerebral bridge. Miyashita et al. (2012) conclude that decreased transcription of LTM-induced genes is a result of increased dCREB2b repressor activity in dNR1(N631Q)expressing neurons (Figure 1). Wild-type flies forced to express dCREB2-b at similarly elevated levels also display a block in activity-dependent transcription of activin, staufen, and homer. Thus, dCREB2-b levels are enhanced by removal of the Mg2+ block and this enhancement is sufficient to mimic the observed memory phenotypes of flies expressing the dNR1(N631Q) mutant NMDAR. It is curious that these conclusive experiments on the role of coincidence detection by NMDARs have been conducted in Drosophila, which, like other insects, uses acetylcholine (Ach) as its major excitatory neurotransmitter. Indeed, one may ask, how does the NMDAR function in nonglutamatergic synapses? Although Drosophila NMDARs differ from mammalian NMDARs in their cytoplasmic domains, they are functionally similar to their mammalian homologs in terms of conductance and gating. We suggest that unlike mammalian central synapses in which AMPA-type glutamate receptors mediate postsynaptic depolarization, nicotinic acetylcholine receptors mediate depolarization in Drosophila synapses. Glutamate required for NMDAR activation could conceivably be released by a distinct, temporally coupled glutamatergic neuron. Alternatively, it may be coreleased by the presynaptic cholinergic neuron. Consistent with this idea, glutamate corelease is widely documented in the mammalian CNS and has been recently proposed as a contributing mechanism for plasticity in the Drosophila antennal lobe (El Mestikawy et al., 2011; Das et al., 2011). Thus, the NMDARs’ ability to function as a coincidence detector may have led to its widespread use for Hebbian synaptic plasticity in both gluta-
matergic and nonglutamatergic systems (El Mestikawy et al., 2011). In conclusion, the study by Miyashita et al. (2012) highlights an unexpected mechanism by which the NMDAR Mg2+ block regulates memory and points to wider and richer roles for NMDAR functions in nervous systems. REFERENCES Bliss, T.V., and Lomo, T. (1973). J. Physiol. 232, 331–356. Chen, P.E., Errington, M.L., Kneussel, M., Chen, G., Annala, A.J., Rudhard, Y.H., Rast, G.F., Specht, C.G., Tigaret, C.M., Nassar, M.A., et al. (2009). Learn. Mem. 16, 635–644. Das, S., Sadanandappa, M.K., Dervan, A., Larkin, A., Lee, J.A., Sudhakaran, I.P., Priya, R., Heidari, R., Holohan, E.E., Pimentel, A., et al. (2011). Proc. Natl. Acad. Sci. USA 108, E646–E654. Dubnau, J., Chiang, A.-S., Grady, L., Barditch, J., Gossweiler, S., McNeil, J., Smith, P., Buldoc, F., Scott, R., Certa, U., et al. (2003). Curr. Biol. 13, 286–296. El Mestikawy, S., Walle´n-Mackenzie, A˚., Fortin, G.M., Descarries, L., and Trudeau, L.-E. (2011). Nat. Rev. Neurosci. 12, 204–216. Mayer, M.L., Westbrook, G.L., and Guthrie, P.B. (1984). Nature 309, 261–263. Miyashita, T., Oda, Y., Horiuchi, J., Yin, J.C., Morimoto, T., and Saitoe, M. (2012). Neuron 74, this issue, 887–898. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A., and Prochiantz, A. (1984). Nature 307, 462–465. Rudhard, Y., Kneussel, M., Nassar, M.A., Rast, G.F., Annala, A.J., Chen, P.E., Tigaret, C.M., Dean, I., Roes, J., Gibb, A.J., et al. (2003). J. Neurosci. 23, 2323–2332. Single, F.N., Rozov, A., Burnashev, N., Zimmermann, F., Hanley, D.F., Forrest, D., Curran, T., Jensen, V., Hvalby, Ø., Sprengel, R., and Seeburg, P.H. (2000). J. Neurosci. 20, 2558–2566. Traynelis, S.F., Wollmuth, L.P., McBain, C.J., Menniti, F.S., Vance, K.M., Ogden, K.K., Hansen, K.B., Yuan, H., Myers, S.J., and Dingledine, R. (2010). Pharmacol. Rev. 62, 405–496. Wu, C.-L., Xia, S., Fu, T.-F., Wang, H., Chen, Y.-H., Leong, D., Chiang, A.-S., and Tully, T. (2007). Nat. Neurosci. 10, 1578–1586. Xia, S., Miyashita, T., Fu, T.-F., Lin, W.-Y., Wu, C.-L., Pyzocha, L., Lin, I.-R., Saitoe, M., Tully, T., and Chiang, A.-S. (2005). Curr. Biol. 15, 603–615. Yin, J.C., and Tully, T. (1996). Curr. Opin. Neurobiol. 6, 264–268.
Neuron 74, June 7, 2012 ª2012 Elsevier Inc. 769
Previews Arendt, 2009; Milnerwood and Raymond, 2010).
Gubernator, N.G., Zhang, H., Staal, R.G., Mosharov, E.V., Pereira, D.B., Yue, M., Balsanek, V., Vadola, P.A., Mukherjee, B., Edwards, R.H., et al. (2009). Science 324, 1441–1444.
REFERENCES Agid, O., Kapur, S., Arenovich, T., and Zipursky, R.B. (2003). Arch. Gen. Psychiatry 60, 1228–1235. Arendt, T. (2009). Acta Neuropathol. 118, 167–179. Bear, M.F., Huber, K.M., and Warren, S.T. (2004). Trends Neurosci. 27, 370–377. Cousin, M.A., and Nicholls, D.G. (1997). J. Neurochem. 69, 1927–1935. Diamond, J.S., and Jahr, C.E. (1997). J. Neurosci. 17, 4672–4687.
Lisman, J.E., Coyle, J.T., Green, R.W., Javitt, D.C., Benes, F.M., Heckers, S., and Grace, A.A. (2008). Trends Neurosci. 31, 234–242. Milnerwood, A.J., and Raymond, L.A. (2010). Trends Neurosci. 33, 513–523. Ogata, N., and Narahashi, T. (1989). Br. J. Pharmacol. 97, 905–913. Raghavachari, S., and Lisman, J.E. (2004). J. Neurophysiol. 92, 2456–2467.
Rayport, S., and Sulzer, D. (1995). J. Neurochem. 65, 691–703. Sah, D.W., and Bean, B.P. (1994). Mol. Pharmacol. 45, 84–92. Sankaranarayanan, S., and Ryan, T.A. (2000). Nat. Cell Biol. 2, 197–204. Strange, P.G. 119–133.
(2001).
Pharmacol.
Rev.
53,
Tischbirek, C.H., Wenzel, E.M., Zheng, F., Huth, T., Amato, D., Trapp, S., Denker, A., Welzel, O., Lueke, K., Svetlitchny, A., et al. (2012). Neuron 74, this issue, 830–844.
Is NMDA Receptor-Coincidence Detection Required for Learning and Memory? Christopher J. Tabone1,* and Mani Ramaswami1,* 1Trinity College Institute for Neuroscience, Smurfit Institute of Genetics and the Department of Zoology, Trinity College Dublin, Dublin-2, Ireland *Correspondence: [email protected] (C.J.T.), [email protected] (M.R.) DOI 10.1016/j.neuron.2012.05.008
The Mg2+ block of NMDA-type glutamate receptors (NMDARs) is crucial to their function as synaptic coincidence detectors. An analysis of Drosophila expressing a Mg2+-independent NMDAR by Miyashita et al. (2012) in this issue of Neuron concludes that the Mg2+ block is required primarily for long-term memory. Donald Hebb first proposed that synapses between two neurons would be strengthened if they showed coincident activity. This idea was hugely influential because such ‘‘Hebbian’’ plasticity could theoretically explain how memories formed, particularly associations between temporally linked events. Subsequently, Bliss and Lomo (1973) discovered long-term potentiation (LTP), a phenomenon in which synaptic strength is enhanced following bursts of synaptic activity. Thus, LTP gained particular notoriety as one of the underlying mechanisms of learning and memory and considerable effort was focused on unraveling mechanisms of coincidence detection and the subsequent synaptic plasticity. From these studies, NMDAtype glutamate receptors (NMDARs) emerged as a class of ionotropic receptors whose pharmacological or genetic
perturbations disrupted both LTP and learning and memory (Traynelis et al., 2010). NMDARs are now understood as pivotal molecules required for coincidence detection, synaptic plasticity, and learning and memory in the central nervous system (CNS). A voltage-dependent Mg2+ block of NMDARs allows them to function as Hebbian coincidence detectors (Mayer et al., 1984; Nowak et al., 1984). Binding by glutamate alone is insufficient for channel activation as Mg2+ remains bound to a site in the channel pore, effectively blocking ion transport. Eviction of this Mg2+ ion additionally requires membrane depolarization. Thus, the coincidence of presynaptic glutamate release and strong depolarizing potential in the postsynaptic neuron is required for the opening of NMDAR channels. Subsequent Ca2+ influx through the
open channel serves as a trigger for synaptic plasticity. Mouse models with mutations specific to the NMDA Mg2+ block site result in developmental defects and/or defects in complex behavior, suggesting that coincidence detection is required for normal NMDAR function in vivo (Single et al., 2000; Rudhard et al., 2003). However, for two reasons, neither these studies nor the observations of abnormal LTP and learning in these mutant mice (Chen et al., 2009) directly address the role of coincidence detection in vivo. First, all known Mg2+ block mutations in murine NMDARs also decrease Ca2+ conductance. Thus, it is unclear whether the resultant phenotypes are due to Mg2+ block-specific effects or reduced calcium permeability. Second, because Mg2+ block mutants show severe developmental defects and early lethality, it is
Neuron 74, June 7, 2012 ª2012 Elsevier Inc. 767
Neuron
Previews Wild type
dNR1(N631Q)
Mg2+
dCREB 2-b
homer staufen activin
dCREB 2-b
poised for transcription
homer staufen activin
transcription inhibited
no LTM
permissive for LTM
Figure 1. The NMDA Receptor Mg2+ Block Regulates dCREB2-b Activity and Long-Term Memory Formation The Mg2+ block reduces baseline calcium entry and prevents basal activation of dCREB2-b, thereby keeping neurons in a state permissive for activity-induced transcription required for LTM. Removal of the NMDAR Mg2+ block in dNR1(N631Q) flies causes increased baseline calcium signaling that causes a 4-fold increase in dCREB2-b expression and a subsequent strong repression of the LTM-associated genes staufen, activin, and homer. This prevents CREB-dependent plasticity required for LTM formation. (Illustration is by Jens Hillebrand.)
difficult to exclude the possibility that defects in learning observed in NMDAR Mg2+ block mutants arise due to altered
nervous system development. Miyashita et al. (2012)’s experiments in Drosophila circumvent these confounding issues
768 Neuron 74, June 7, 2012 ª2012 Elsevier Inc.
and directly assess the role of the Mg2+ block in memory formation. Drosophila NMDARs, composed of two subunits, dNR1 and dNR2, are necessary for normal memory formation (Xia et al., 2005; Wu et al., 2007). In the relevant behavioral assay, flies learn to associate an olfactory cue with aversive electric foot shocks. The ability of the fly to demonstrate learned avoidance of the shock-associated odor is quantified as a ‘‘performance index’’ or PI. Singletraining trials produce short-term and middle-term memory (STM/MTM). Multiple, spaced training sessions additionally produce a protein synthesisdependent long-term memory (LTM) that requires the function of CREB2, the cAMP-regulatory element-binding transcription factor (Yin and Tully, 1996). The first behavioral study of the NMDAR in Drosophila indicated its requirement for olfactory learning: hypomorphic dNR1 mutants showed defects in learning and LTM (Xia et al., 2005). Further work showed that while NMDA receptors are required in the mushroom body for early phases of memory, they are additionally required in the ellipsoid body for LTM (Wu et al., 2007). In this intellectual context, Miyashita et al. (2012) investigate the role of coincidence detection specifically through the Mg2+ block mechanism of NMDAR for learning and memory. Miyashita et al. (2012) first constructed a dNR1 transgene encoding the N631Q mutant NMDAR, which specifically disrupts the Mg2+ block site in the encoded protein with no detectable effect on Ca2+ permeability, as assessed using electrophysiology and Ca2+imaging. Expression of the dNR1(N631Q) transgene in hypomorphic dNR1 flies rescued the mutant’s learning defect, indicating that the Mg2+ block was not required for NMDAR’s function in learning. However, in contrast, N631Q expression disrupted LTM formation after spaced training. The first conclusion from these experiments that the Mg2+ block is not essential for learning does not conflict with current models of NMDAR function. Coincidence detection is necessary to limit plasticity only to synapses that display coincident pre- and postsynaptic activity. Thus, by removing the second requirement for postsynaptic excitation in Mg2+ block mutants, plasticity may
Neuron
Previews occur (1) more easily and (2) in additional postsynaptic neurons that do not show synchronous depolarization, e.g., in mushroom-body neurons that are not activated by the odorant. Consistent with a prediction from (1), the authors observe slightly enhanced learning in flies expressing the mutant dNR1(N631Q) transgene. The prediction from (2), wherein plasticity occurs throughout a larger group of neurons, will require additional comprehensive testing, e.g., for increased odor generalization after olfactory aversive conditioning. How then does removal of the Mg2+ block interfere with the formation of LTM? Previous studies in Drosophila have indicated that LTM is modulated by the relative activity of the repressor isoform of CREB, dCREB2-b (Yin and Tully, 1996). Could enhanced calcium influx through NMDAR mutants freed from the Mg2+ block result in increased activity of dCREB2-b? To address this question, the authors examined transcripts of the repressor isoform dCREB2b extracted from heads of dNR1(N631Q) flies. They report a nearly 4-fold increase in the basal expression level of dCREB2b as compared to wild-type animals. This increase in dCREB2-b expression is also observed in individual wild-type brains cultured in Mg2+-free medium, indicating that increases in calcium influx in the absence of a Mg2+ block can directly lead to increased levels of dCREB2-b. If the absence of the Mg2+ block directly affects the expression of the dCREB2b repressor, then the expression of CREB2-target genes should be affected. The authors tested this hypothesis by examining expression levels of the genes activin, staufen, and homer, previously shown to be transcriptionally induced after spaced training (Dubnau et al., 2003). Remarkably, Miyashita et al. (2012) observed an absence of this upregulation in dNR1(N631Q) flies. The block in gene induction was cell autonomous: flies expressing dNR1(N631Q) in mushroom body neurons showed no increase in homer expression in mushroom
bodies but still displayed homer upregulation in other structures such as the protocerebral bridge. Miyashita et al. (2012) conclude that decreased transcription of LTM-induced genes is a result of increased dCREB2b repressor activity in dNR1(N631Q)expressing neurons (Figure 1). Wild-type flies forced to express dCREB2-b at similarly elevated levels also display a block in activity-dependent transcription of activin, staufen, and homer. Thus, dCREB2-b levels are enhanced by removal of the Mg2+ block and this enhancement is sufficient to mimic the observed memory phenotypes of flies expressing the dNR1(N631Q) mutant NMDAR. It is curious that these conclusive experiments on the role of coincidence detection by NMDARs have been conducted in Drosophila, which, like other insects, uses acetylcholine (Ach) as its major excitatory neurotransmitter. Indeed, one may ask, how does the NMDAR function in nonglutamatergic synapses? Although Drosophila NMDARs differ from mammalian NMDARs in their cytoplasmic domains, they are functionally similar to their mammalian homologs in terms of conductance and gating. We suggest that unlike mammalian central synapses in which AMPA-type glutamate receptors mediate postsynaptic depolarization, nicotinic acetylcholine receptors mediate depolarization in Drosophila synapses. Glutamate required for NMDAR activation could conceivably be released by a distinct, temporally coupled glutamatergic neuron. Alternatively, it may be coreleased by the presynaptic cholinergic neuron. Consistent with this idea, glutamate corelease is widely documented in the mammalian CNS and has been recently proposed as a contributing mechanism for plasticity in the Drosophila antennal lobe (El Mestikawy et al., 2011; Das et al., 2011). Thus, the NMDARs’ ability to function as a coincidence detector may have led to its widespread use for Hebbian synaptic plasticity in both gluta-
matergic and nonglutamatergic systems (El Mestikawy et al., 2011). In conclusion, the study by Miyashita et al. (2012) highlights an unexpected mechanism by which the NMDAR Mg2+ block regulates memory and points to wider and richer roles for NMDAR functions in nervous systems. REFERENCES Bliss, T.V., and Lomo, T. (1973). J. Physiol. 232, 331–356. Chen, P.E., Errington, M.L., Kneussel, M., Chen, G., Annala, A.J., Rudhard, Y.H., Rast, G.F., Specht, C.G., Tigaret, C.M., Nassar, M.A., et al. (2009). Learn. Mem. 16, 635–644. Das, S., Sadanandappa, M.K., Dervan, A., Larkin, A., Lee, J.A., Sudhakaran, I.P., Priya, R., Heidari, R., Holohan, E.E., Pimentel, A., et al. (2011). Proc. Natl. Acad. Sci. USA 108, E646–E654. Dubnau, J., Chiang, A.-S., Grady, L., Barditch, J., Gossweiler, S., McNeil, J., Smith, P., Buldoc, F., Scott, R., Certa, U., et al. (2003). Curr. Biol. 13, 286–296. El Mestikawy, S., Walle´n-Mackenzie, A˚., Fortin, G.M., Descarries, L., and Trudeau, L.-E. (2011). Nat. Rev. Neurosci. 12, 204–216. Mayer, M.L., Westbrook, G.L., and Guthrie, P.B. (1984). Nature 309, 261–263. Miyashita, T., Oda, Y., Horiuchi, J., Yin, J.C., Morimoto, T., and Saitoe, M. (2012). Neuron 74, this issue, 887–898. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A., and Prochiantz, A. (1984). Nature 307, 462–465. Rudhard, Y., Kneussel, M., Nassar, M.A., Rast, G.F., Annala, A.J., Chen, P.E., Tigaret, C.M., Dean, I., Roes, J., Gibb, A.J., et al. (2003). J. Neurosci. 23, 2323–2332. Single, F.N., Rozov, A., Burnashev, N., Zimmermann, F., Hanley, D.F., Forrest, D., Curran, T., Jensen, V., Hvalby, Ø., Sprengel, R., and Seeburg, P.H. (2000). J. Neurosci. 20, 2558–2566. Traynelis, S.F., Wollmuth, L.P., McBain, C.J., Menniti, F.S., Vance, K.M., Ogden, K.K., Hansen, K.B., Yuan, H., Myers, S.J., and Dingledine, R. (2010). Pharmacol. Rev. 62, 405–496. Wu, C.-L., Xia, S., Fu, T.-F., Wang, H., Chen, Y.-H., Leong, D., Chiang, A.-S., and Tully, T. (2007). Nat. Neurosci. 10, 1578–1586. Xia, S., Miyashita, T., Fu, T.-F., Lin, W.-Y., Wu, C.-L., Pyzocha, L., Lin, I.-R., Saitoe, M., Tully, T., and Chiang, A.-S. (2005). Curr. Biol. 15, 603–615. Yin, J.C., and Tully, T. (1996). Curr. Opin. Neurobiol. 6, 264–268.
Neuron 74, June 7, 2012 ª2012 Elsevier Inc. 769