- Email: [email protected]
2B or Not 2B: A Tail of Two NMDA Receptor Subunits
Neuron
Previews rodent studies) to neuroscience. A door has been opened by Evrard et al. (2012), but much exciting work remains to be done.
Craig, A.D. (2011). Ann. N Y Acad. Sci. 1225, 72–82.
Nimchinsky, E.A., Gilissen, E., Allman, J.M., Perl, D.P., Erwin, J.M., and Hof, P.R. (1999). Proc. Natl. Acad. Sci. USA 96, 5268–5273.
Damasio, A.R. (2010). Self Comes to Mind: Constructing the Conscious Brain (New York: Pantheon).
Palaniyappan, L., and Liddle, P.F. (2012). J. Psychiatry Neurosci. 37, 17–27.
Dehaene, S., and Changeux, J.P. (2011). Neuron 70, 200–227.
Seeley, W.W., Merkle, F.T., Gaus, S.E., Craig, A.D., Allman, J.M., Hof, P.R., and Economo, C.V. (2012). Cereb. Cortex 22, 245–250.
Butti, C., and Hof, P.R. (2010). Brain Struct. Funct. 214, 477–493.
Evrard, H., Forro, T., and Logothetis, N. (2012). Neuron 74, this issue, 482–489.
Seth, A.K., Suzuki, K., and Critchley, H.D. (2011). Front. Psychol. 2, 395.
Cauda, F., Torta, D.M., Sacco, K., D’Agata, F., Geda, E., Duca, S., Geminiani, G., and Vercelli, A. (2012). Brain Struct. Funct., in press. Published online January 29, 2012. 10.1007/s00429-0120382-9.
Fletcher, P.C., and Frith, C.D. (2009). Nat. Rev. Neurosci. 10, 48–58.
REFERENCES Allman, J.M., Tetreault, N.A., Hakeem, A.Y., and Park, S. (2011). Am. J. Hum. Biol. 23, 5–21.
Friston, K. (2010). Nat. Rev. Neurosci. 11, 127–138.
Singer, T., Critchley, H.D., and Preuschoff, K. (2009). Trends Cogn. Sci. (Regul. Ed.) 13, 334–340. Watson, K.K., Jones, T.K., and Allman, J.M. (2006). Neuroscience 141, 1107–1112.
2B or Not 2B: A Tail of Two NMDA Receptor Subunits Carlos Cepeda1 and Michael S. Levine1,* 1Intellectual and Developmental Disabilities Research Center, Semel Institute for Neuroscience and Human Behavior, Brain Research Institute, The David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA 90095, USA *Correspondence: [email protected] DOI 10.1016/j.neuron.2012.04.011
N-methyl-D-aspartate (NMDA) receptor activation can be neuroprotective or neurotoxic depending on receptor location. In this issue of Neuron, Martel et al. (2012) demonstrate that the C-terminal of NMDA receptor subunits also contributes critically to excitotoxicity. NMDA receptor subunits containing the GluN2B C-terminal are more lethal than those containing the GluN2A tails, regardless of location. Is the location of N-methyl-D-aspartate receptors (NMDARs) at synaptic or extrasynaptic sites the only, or even the primary, determinant of neuroprotective or neurotoxic effects of glutamate? While we thought this question had been settled, at least partially (Levine et al., 2010; Milnerwood et al., 2010; Okamoto et al., 2009), new work from the laboratory that raised into prominence the differential role of synaptic and extrasynaptic NMDARs and gave us a better understanding of the intracellular cascades that lead to excitotoxicity (Hardingham et al., 2002) now demonstrates that we were missing part of the equation, a little but important C-tail. In effect, the C-terminal domain (CTD) of the NMDAR subunit appears to play a critical role in the function of the receptor. In an elegant study published in this issue of Neuron, Martel et al. (2012) report
that the activation of proapoptotic cascades is determined not just by the location of the NMDAR, but also by the identity of the CTD. As pharmacological approaches commonly used to differentiate the two subunits are limited at best, the authors used genetic manipulations to engender chimeric receptors in which only the CTD from GluN2A and GluN2B receptor subunits is C-terminal replaced (CTR). Why focus on the CTD? It has been shown that the CTD of NMDAR subunits is the primary area of sequence divergence, and it is the site that primarily binds scaffolding proteins, providing a strong rationale for examining its role in excitotoxicity. In the first series of experiments, expression of chimeric GluN2B2A(CTR) receptors in transfected hippocampal neurons produced similar currents as wild-type (WT) subunits (GluN2BWT) and
426 Neuron 74, May 10, 2012 ª2012 Elsevier Inc.
did not affect the proportion of synaptic and extrasynaptic receptors, thus preventing potential confounds arising from receptor location. Interestingly, NMDAinduced cell death was reduced in chimeric GluN2B2A(CTR) compared to GluN2BWT-containing receptors, suggesting that excitotoxicity is better promoted by CTD2B than CTD2A. Similarly, neurons expressing GluN2A2B(CTR) were more susceptible to cell death than neurons expressing GluN2AWT (Figure 1). Using a different approach, a knockin mouse was generated in which the protein-coding region of the C-terminal exon of the GluN2B subunit was exchanged for that of the GluN2A subunit, named GluN2B2A(CTR)/2A(CTR). Cultured cortical neurons from these mice displayed similar levels of viability, synaptic connectivity, proportion of extrasynaptic NMDARs, sensitivity to ifenprodil,
Neuron
Previews rundown of NMDA currents, that GluN2A and GluN2B and single channel conducsubunits affect a-amino-3tance, compared to those hydroxy-5-methyl-4-isoxazomice. from GluN2B+/+ lepropionic acid (AMPA) Notwithstanding these simireceptor trafficking in oppolarities, NMDA currents site ways, with GluN2A prowere about 30% lower in moting and GluN2B inhibiting GluN2B2A(CTR)/2A(CTR) than in surface expression of GluA1 GluN2B+/+ cells. By adjusting subunits. In the realm of excitotoxicexogenous NMDA concenity, it was previously demontrations to produce similar strated that activation of currents in both types of GluN2B-containing NMDA cells, the authors confirmed receptors, at either synaptic that normalized NMDA or extrasynaptic sites, leads current produced more to excitotoxicity, whereas death in GluN2B+/+ than in Figure 1. The C-Terminal Domain of GluN2B Subunits Is More Lethal GluN2B2A(CTR)/2A(CTR) cells. In activation of either synaptic Excess glutamate can be excitotoxic via NMDARs. However, the extent of cell consequence, a switch in the or extrasynaptic GluN2Adeath depends in part on the CTD of NMDAR subunits. Swapping the CTD mouse genome from GluN2B contaning NMDARs profrom GluN2A for GluN2B produces more cell death. Conversely, swapping the CTD from GluN2B for GluN2A reduces cell death. Similarly, cells from CTD for GluN2A reduces motes neuronal survival and genetically modified knockin mice, in which the CTD from GluN2B is replaced NMDA-dependent Ca2+ influx is neuroprotective (Liu et al., by the CTD of GluN2A, are less vulnerable than cells from GluN2B+/+ mice, due 2007). In this study, adminand excitotoxicity. However, 2B in part to increased phosphorylation of CREB. In contrast, CTD displays istration of glycine alone these differences only stronger coupling to the PSD-95/nNOS pathway, which suppresses CREB activation. This indicates that the GluN2B subunit, regardless of location at or in the presence of a occurred at moderate (15– synaptic or extrasynaptic sites, is more lethal than the GluN2A subunit. GluN2B antagonist attenu50 mM) NMDA concentrations. ated ischemic brain damage. When NMDA concentration was increased (e.g., 100 mM), the CTD known that nitric oxide (NO) is a key Understanding the mechanisms of subtype-specific vulnerability disap- regulator of CREB phosphorylation. NO NMDAR-mediated excitotoxicity is parapeared. These results were confirmed is produced when NMDAR-dependent mount to the development of better neuin vivo. Thus, excitotoxic lesions induced Ca2+ influx activates nNOS via PSD-95 roprotective tools in acute and chronic by stereotaxic injection of a small dose of association with GluN2 subunits. In addi- conditions. While great strides had been NMDA into the hippocampus (CA1-CA3 tional experiments, Martel et al. (2012) made, many of them by Hardingham’s region) induced smaller lesion volumes found that GluN2B+/+ neurons coupled group, the intimate mechanisms of exciin GluN2B2A(CTR)/2A(CTR) compared to more strongly to NMDA-induced NO totoxicity had been missing. Although production and concluded that stronger the increased presence of GluN2B GluN2B+/+ mice. Which signaling cascades contribute CTD2B coupling to PSD-95, NO produc- subunits in extrasynaptic locations is still to differential susceptibility of CTDs to tion, and nNOS-dependent CREB inacti- a matter of debate, the role of these excitotoxic insults? One obvious target, vation leads to enhanced vulnerability to subunits in NMDAR-mediated toxicity is based on previous work by Hardingham’s excitotoxic insults. Finally, the basis for well supported by experimental evidence. Huntington’s disease (HD) is one good and other groups, is NMDA-dependent the stronger association of PSD-95 with activation of CREB. While basal levels of GluN2BWT compared to GluN2B2A(CTR) example in which the present findings CREB (serine 133) phosphorylation was explored. An internal region of the can open new venues for therapies. were unaltered in GluN2B2A(CTR)/2A(CTR) CTD2B (1086–1157), when deleted, re- Recent studies (Milnerwood et al., 2010; compared to GluN2B+/+ neurons, CREB sulted in a decrease in PSD-95 asso- Okamoto et al., 2009), partly promoted phosphorylation was prolonged in ciation, whereas overexpression of this by Hardingham’s previous work, have GluN2B2A(CTR)/2A(CTR) neurons when chal- region led to reduced NMDA-induced demonstrated enhanced extrasynaptic lenged with NMDA. This observation was cell death. This region thus could be impli- NMDAR-mediated activity in HD mouse supported by the fact that blockade of cated in NMDAR signaling leading to cell models and the effectiveness of memanCRE-mediated gene expression (with death. tine (an NMDAR antagonist used as ICER, an inhibitory CREB family member) The idea that GluN2A and GluN2B a more selective extrasynaptic receptor increased NMDA-induced cell death, indi- subunits play different roles in diverse blocker) for the treatment of some HD cating that differential CREB activation processes such as synaptic plasticity, symptoms. Lynn Raymond’s laboratory contributes to CTD subtype-dependent intracellular signaling, and excitotoxicity in Vancouver has demonstrated the regulation of excitotoxicity. has often been entertained. In the important role that the GluN2B subunit What makes the two CTDs different? case of synaptic plasticity, experimental plays in striatal cell death in HD. ExpresThe answer appears to rely, in part, on evidence is not conclusive, and it appears sion of mutant huntingtin (htt) has been enhanced coupling of CTD2B to the that both subunits are necessary (Mu¨ller hypothesized to alter striatal NMDAR PSD-95/nNOS signaling cassette. It is et al., 2009). However, there is evidence signaling (Raymond et al., 2011). In the Neuron 74, May 10, 2012 ª2012 Elsevier Inc. 427
Neuron
Previews early stages of the disease, studies in HD genetic mouse models have shown increased NMDAR-induced currents (Starling et al., 2005). Importantly, this increase appears to be mediated by NMDAR-containing GluN2B subunits, as enhanced currents and toxicity in cultured neurons and acute slices are abolished by ifenprodil or memantine (Kaufman et al., 2012). Thus, experimental evidence supports the idea that mutant htt enhances cell death by modulating GluN2B subunits. In agreement, dramatic exacerbation of striatal neuronal loss was reported when HD knockin mice were crossed with GluN2B-overexpressing mice (Heng et al., 2009). Does the presence and relative abundance of GluNR2B subunits make neurons more vulnerable? A recent study showed that mediumsized spiny neurons (MSNs) of the indirect striatal output pathway, i.e., the neurons that are believed to be more affected in the early stages of HD, express more functional GluN2B-containing NMDARs (Jocoy et al., 2011). In contrast, MSNs of the direct pathway appear to express relatively greater levels of GluN2A subunits and are less affected. While these studies are indicative of contrasting roles of NMDAR subunits, it was not until the present work by Martel et al. (2012) that the precise locus and mechanisms have been unraveled. Based on their findings, the GluN2B/PSD-95/ nNOS axis represents an attractive target for therapeutic intervention. Indeed, as the authors indicate, results from a series of studies demonstrating antiexcitotoxic effects of TAT-NR2B9c, PSD-95 knockdown, or disruption of the PSD-95-nNOS
interface can now be explained. In addition, the translational potential is great and is supported by recent evidence that administration of TAT-NR2Bc, even hours after stroke, can prevent neuronal damage and neurological deficits (Cook et al., 2012). While the role of NO in disease processes such as HD remains to be established, neuroprotective or neurotoxic effects can occur depending on a number of factors (Deckel, 2001). Although the new findings of Martel et al. (2012) are revealing, more studies will be necessary to understand how identity and location of GluN2 type subunits at synaptic and extrasynaptic sites contribute to excitotoxicity. In particular, visualization of NMDAR surface mobility in and out of the synapse in native conditions will be extremely useful. For example, using single-particle and singlemolecule tracking approaches, NMDAR mobility has been shown to depend on the identity of GluN2-type subunits, as NMDARs containing GluN2B subunits are less stable than those containing GluN2A subunits (Groc et al., 2006). In conclusion, the present contribution will certainly become another classic in the field of NMDAR-mediated neurotoxicity, with far-reaching scientific and clinical implications. As the GluN2 subunit saga moves on, the ‘‘tail’’ of 2B or not 2B remains an important component of the question. REFERENCES Cook, D.J., Teves, L., and Tymianski, M. (2012). Nature 483, 213–217. Deckel, A.W. (2001). J. Neurosci. Res. 64, 99–107.
428 Neuron 74, May 10, 2012 ª2012 Elsevier Inc.
Groc, L., Heine, M., Cousins, S.L., Stephenson, F.A., Lounis, B., Cognet, L., and Choquet, D. (2006). Proc. Natl. Acad. Sci. USA 103, 18769– 18774. Hardingham, G.E., Fukunaga, Y., and Bading, H. (2002). Nat. Neurosci. 5, 405–414. Heng, M.Y., Detloff, P.J., Wang, P.L., Tsien, J.Z., and Albin, R.L. (2009). J. Neurosci. 29, 3200–3205. Jocoy, E.L., Andre´, V.M., Cummings, D.M., Rao, S.P., Wu, N., Ramsey, A.J., Caron, M.G., Cepeda, C., and Levine, M.S. (2011). Front. Syst. Neurosci. 5, 28. Kaufman, A.M., Milnerwood, A.J., Sepers, M.D., Coquinco, A., She, K., Wang, L., Lee, H., Craig, A.M., Cynader, M., and Raymond, L.A. (2012). J. Neurosci. 32, 3992–4003. Levine, M.S., Cepeda, C., and Andre´, V.M. (2010). Neuron 65, 145–147. Liu, Y., Wong, T.P., Aarts, M., Rooyakkers, A., Liu, L., Lai, T.W., Wu, D.C., Lu, J., Tymianski, M., Craig, A.M., and Wang, Y.T. (2007). J. Neurosci. 27, 2846–2857. Martel, M.-A., Ryan, T.J., Bell, K.F.S., Fowler, J.H., McMahon, A., Al-Mubarak, B., Komiyama, N.H., Horsburgh, K., Kind, P.C., Grant, S.G.N., et al. (2012). Neuron 74, this issue, 543–556. Milnerwood, A.J., Gladding, C.M., Pouladi, M.A., Kaufman, A.M., Hines, R.M., Boyd, J.D., Ko, R.W., Vasuta, O.C., Graham, R.K., Hayden, M.R., et al. (2010). Neuron 65, 178–190. Mu¨ller, T., Albrecht, D., and Gebhardt, C. (2009). Learn. Mem. 16, 395–405. Okamoto, S., Pouladi, M.A., Talantova, M., Yao, D., Xia, P., Ehrnhoefer, D.E., Zaidi, R., Clemente, A., Kaul, M., Graham, R.K., et al. (2009). Nat. Med. 15, 1407–1413. Raymond, L.A., Andre´, V.M., Cepeda, C., Gladding, C.M., Milnerwood, A.J., and Levine, M.S. (2011). Neuroscience 198, 252–273. Starling, A.J., Andre´, V.M., Cepeda, C., de Lima, M., Chandler, S.H., and Levine, M.S. (2005). J. Neurosci. Res. 82, 377–386.
Previews rodent studies) to neuroscience. A door has been opened by Evrard et al. (2012), but much exciting work remains to be done.
Craig, A.D. (2011). Ann. N Y Acad. Sci. 1225, 72–82.
Nimchinsky, E.A., Gilissen, E., Allman, J.M., Perl, D.P., Erwin, J.M., and Hof, P.R. (1999). Proc. Natl. Acad. Sci. USA 96, 5268–5273.
Damasio, A.R. (2010). Self Comes to Mind: Constructing the Conscious Brain (New York: Pantheon).
Palaniyappan, L., and Liddle, P.F. (2012). J. Psychiatry Neurosci. 37, 17–27.
Dehaene, S., and Changeux, J.P. (2011). Neuron 70, 200–227.
Seeley, W.W., Merkle, F.T., Gaus, S.E., Craig, A.D., Allman, J.M., Hof, P.R., and Economo, C.V. (2012). Cereb. Cortex 22, 245–250.
Butti, C., and Hof, P.R. (2010). Brain Struct. Funct. 214, 477–493.
Evrard, H., Forro, T., and Logothetis, N. (2012). Neuron 74, this issue, 482–489.
Seth, A.K., Suzuki, K., and Critchley, H.D. (2011). Front. Psychol. 2, 395.
Cauda, F., Torta, D.M., Sacco, K., D’Agata, F., Geda, E., Duca, S., Geminiani, G., and Vercelli, A. (2012). Brain Struct. Funct., in press. Published online January 29, 2012. 10.1007/s00429-0120382-9.
Fletcher, P.C., and Frith, C.D. (2009). Nat. Rev. Neurosci. 10, 48–58.
REFERENCES Allman, J.M., Tetreault, N.A., Hakeem, A.Y., and Park, S. (2011). Am. J. Hum. Biol. 23, 5–21.
Friston, K. (2010). Nat. Rev. Neurosci. 11, 127–138.
Singer, T., Critchley, H.D., and Preuschoff, K. (2009). Trends Cogn. Sci. (Regul. Ed.) 13, 334–340. Watson, K.K., Jones, T.K., and Allman, J.M. (2006). Neuroscience 141, 1107–1112.
2B or Not 2B: A Tail of Two NMDA Receptor Subunits Carlos Cepeda1 and Michael S. Levine1,* 1Intellectual and Developmental Disabilities Research Center, Semel Institute for Neuroscience and Human Behavior, Brain Research Institute, The David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA 90095, USA *Correspondence: [email protected] DOI 10.1016/j.neuron.2012.04.011
N-methyl-D-aspartate (NMDA) receptor activation can be neuroprotective or neurotoxic depending on receptor location. In this issue of Neuron, Martel et al. (2012) demonstrate that the C-terminal of NMDA receptor subunits also contributes critically to excitotoxicity. NMDA receptor subunits containing the GluN2B C-terminal are more lethal than those containing the GluN2A tails, regardless of location. Is the location of N-methyl-D-aspartate receptors (NMDARs) at synaptic or extrasynaptic sites the only, or even the primary, determinant of neuroprotective or neurotoxic effects of glutamate? While we thought this question had been settled, at least partially (Levine et al., 2010; Milnerwood et al., 2010; Okamoto et al., 2009), new work from the laboratory that raised into prominence the differential role of synaptic and extrasynaptic NMDARs and gave us a better understanding of the intracellular cascades that lead to excitotoxicity (Hardingham et al., 2002) now demonstrates that we were missing part of the equation, a little but important C-tail. In effect, the C-terminal domain (CTD) of the NMDAR subunit appears to play a critical role in the function of the receptor. In an elegant study published in this issue of Neuron, Martel et al. (2012) report
that the activation of proapoptotic cascades is determined not just by the location of the NMDAR, but also by the identity of the CTD. As pharmacological approaches commonly used to differentiate the two subunits are limited at best, the authors used genetic manipulations to engender chimeric receptors in which only the CTD from GluN2A and GluN2B receptor subunits is C-terminal replaced (CTR). Why focus on the CTD? It has been shown that the CTD of NMDAR subunits is the primary area of sequence divergence, and it is the site that primarily binds scaffolding proteins, providing a strong rationale for examining its role in excitotoxicity. In the first series of experiments, expression of chimeric GluN2B2A(CTR) receptors in transfected hippocampal neurons produced similar currents as wild-type (WT) subunits (GluN2BWT) and
426 Neuron 74, May 10, 2012 ª2012 Elsevier Inc.
did not affect the proportion of synaptic and extrasynaptic receptors, thus preventing potential confounds arising from receptor location. Interestingly, NMDAinduced cell death was reduced in chimeric GluN2B2A(CTR) compared to GluN2BWT-containing receptors, suggesting that excitotoxicity is better promoted by CTD2B than CTD2A. Similarly, neurons expressing GluN2A2B(CTR) were more susceptible to cell death than neurons expressing GluN2AWT (Figure 1). Using a different approach, a knockin mouse was generated in which the protein-coding region of the C-terminal exon of the GluN2B subunit was exchanged for that of the GluN2A subunit, named GluN2B2A(CTR)/2A(CTR). Cultured cortical neurons from these mice displayed similar levels of viability, synaptic connectivity, proportion of extrasynaptic NMDARs, sensitivity to ifenprodil,
Neuron
Previews rundown of NMDA currents, that GluN2A and GluN2B and single channel conducsubunits affect a-amino-3tance, compared to those hydroxy-5-methyl-4-isoxazomice. from GluN2B+/+ lepropionic acid (AMPA) Notwithstanding these simireceptor trafficking in oppolarities, NMDA currents site ways, with GluN2A prowere about 30% lower in moting and GluN2B inhibiting GluN2B2A(CTR)/2A(CTR) than in surface expression of GluA1 GluN2B+/+ cells. By adjusting subunits. In the realm of excitotoxicexogenous NMDA concenity, it was previously demontrations to produce similar strated that activation of currents in both types of GluN2B-containing NMDA cells, the authors confirmed receptors, at either synaptic that normalized NMDA or extrasynaptic sites, leads current produced more to excitotoxicity, whereas death in GluN2B+/+ than in Figure 1. The C-Terminal Domain of GluN2B Subunits Is More Lethal GluN2B2A(CTR)/2A(CTR) cells. In activation of either synaptic Excess glutamate can be excitotoxic via NMDARs. However, the extent of cell consequence, a switch in the or extrasynaptic GluN2Adeath depends in part on the CTD of NMDAR subunits. Swapping the CTD mouse genome from GluN2B contaning NMDARs profrom GluN2A for GluN2B produces more cell death. Conversely, swapping the CTD from GluN2B for GluN2A reduces cell death. Similarly, cells from CTD for GluN2A reduces motes neuronal survival and genetically modified knockin mice, in which the CTD from GluN2B is replaced NMDA-dependent Ca2+ influx is neuroprotective (Liu et al., by the CTD of GluN2A, are less vulnerable than cells from GluN2B+/+ mice, due 2007). In this study, adminand excitotoxicity. However, 2B in part to increased phosphorylation of CREB. In contrast, CTD displays istration of glycine alone these differences only stronger coupling to the PSD-95/nNOS pathway, which suppresses CREB activation. This indicates that the GluN2B subunit, regardless of location at or in the presence of a occurred at moderate (15– synaptic or extrasynaptic sites, is more lethal than the GluN2A subunit. GluN2B antagonist attenu50 mM) NMDA concentrations. ated ischemic brain damage. When NMDA concentration was increased (e.g., 100 mM), the CTD known that nitric oxide (NO) is a key Understanding the mechanisms of subtype-specific vulnerability disap- regulator of CREB phosphorylation. NO NMDAR-mediated excitotoxicity is parapeared. These results were confirmed is produced when NMDAR-dependent mount to the development of better neuin vivo. Thus, excitotoxic lesions induced Ca2+ influx activates nNOS via PSD-95 roprotective tools in acute and chronic by stereotaxic injection of a small dose of association with GluN2 subunits. In addi- conditions. While great strides had been NMDA into the hippocampus (CA1-CA3 tional experiments, Martel et al. (2012) made, many of them by Hardingham’s region) induced smaller lesion volumes found that GluN2B+/+ neurons coupled group, the intimate mechanisms of exciin GluN2B2A(CTR)/2A(CTR) compared to more strongly to NMDA-induced NO totoxicity had been missing. Although production and concluded that stronger the increased presence of GluN2B GluN2B+/+ mice. Which signaling cascades contribute CTD2B coupling to PSD-95, NO produc- subunits in extrasynaptic locations is still to differential susceptibility of CTDs to tion, and nNOS-dependent CREB inacti- a matter of debate, the role of these excitotoxic insults? One obvious target, vation leads to enhanced vulnerability to subunits in NMDAR-mediated toxicity is based on previous work by Hardingham’s excitotoxic insults. Finally, the basis for well supported by experimental evidence. Huntington’s disease (HD) is one good and other groups, is NMDA-dependent the stronger association of PSD-95 with activation of CREB. While basal levels of GluN2BWT compared to GluN2B2A(CTR) example in which the present findings CREB (serine 133) phosphorylation was explored. An internal region of the can open new venues for therapies. were unaltered in GluN2B2A(CTR)/2A(CTR) CTD2B (1086–1157), when deleted, re- Recent studies (Milnerwood et al., 2010; compared to GluN2B+/+ neurons, CREB sulted in a decrease in PSD-95 asso- Okamoto et al., 2009), partly promoted phosphorylation was prolonged in ciation, whereas overexpression of this by Hardingham’s previous work, have GluN2B2A(CTR)/2A(CTR) neurons when chal- region led to reduced NMDA-induced demonstrated enhanced extrasynaptic lenged with NMDA. This observation was cell death. This region thus could be impli- NMDAR-mediated activity in HD mouse supported by the fact that blockade of cated in NMDAR signaling leading to cell models and the effectiveness of memanCRE-mediated gene expression (with death. tine (an NMDAR antagonist used as ICER, an inhibitory CREB family member) The idea that GluN2A and GluN2B a more selective extrasynaptic receptor increased NMDA-induced cell death, indi- subunits play different roles in diverse blocker) for the treatment of some HD cating that differential CREB activation processes such as synaptic plasticity, symptoms. Lynn Raymond’s laboratory contributes to CTD subtype-dependent intracellular signaling, and excitotoxicity in Vancouver has demonstrated the regulation of excitotoxicity. has often been entertained. In the important role that the GluN2B subunit What makes the two CTDs different? case of synaptic plasticity, experimental plays in striatal cell death in HD. ExpresThe answer appears to rely, in part, on evidence is not conclusive, and it appears sion of mutant huntingtin (htt) has been enhanced coupling of CTD2B to the that both subunits are necessary (Mu¨ller hypothesized to alter striatal NMDAR PSD-95/nNOS signaling cassette. It is et al., 2009). However, there is evidence signaling (Raymond et al., 2011). In the Neuron 74, May 10, 2012 ª2012 Elsevier Inc. 427
Neuron
Previews early stages of the disease, studies in HD genetic mouse models have shown increased NMDAR-induced currents (Starling et al., 2005). Importantly, this increase appears to be mediated by NMDAR-containing GluN2B subunits, as enhanced currents and toxicity in cultured neurons and acute slices are abolished by ifenprodil or memantine (Kaufman et al., 2012). Thus, experimental evidence supports the idea that mutant htt enhances cell death by modulating GluN2B subunits. In agreement, dramatic exacerbation of striatal neuronal loss was reported when HD knockin mice were crossed with GluN2B-overexpressing mice (Heng et al., 2009). Does the presence and relative abundance of GluNR2B subunits make neurons more vulnerable? A recent study showed that mediumsized spiny neurons (MSNs) of the indirect striatal output pathway, i.e., the neurons that are believed to be more affected in the early stages of HD, express more functional GluN2B-containing NMDARs (Jocoy et al., 2011). In contrast, MSNs of the direct pathway appear to express relatively greater levels of GluN2A subunits and are less affected. While these studies are indicative of contrasting roles of NMDAR subunits, it was not until the present work by Martel et al. (2012) that the precise locus and mechanisms have been unraveled. Based on their findings, the GluN2B/PSD-95/ nNOS axis represents an attractive target for therapeutic intervention. Indeed, as the authors indicate, results from a series of studies demonstrating antiexcitotoxic effects of TAT-NR2B9c, PSD-95 knockdown, or disruption of the PSD-95-nNOS
interface can now be explained. In addition, the translational potential is great and is supported by recent evidence that administration of TAT-NR2Bc, even hours after stroke, can prevent neuronal damage and neurological deficits (Cook et al., 2012). While the role of NO in disease processes such as HD remains to be established, neuroprotective or neurotoxic effects can occur depending on a number of factors (Deckel, 2001). Although the new findings of Martel et al. (2012) are revealing, more studies will be necessary to understand how identity and location of GluN2 type subunits at synaptic and extrasynaptic sites contribute to excitotoxicity. In particular, visualization of NMDAR surface mobility in and out of the synapse in native conditions will be extremely useful. For example, using single-particle and singlemolecule tracking approaches, NMDAR mobility has been shown to depend on the identity of GluN2-type subunits, as NMDARs containing GluN2B subunits are less stable than those containing GluN2A subunits (Groc et al., 2006). In conclusion, the present contribution will certainly become another classic in the field of NMDAR-mediated neurotoxicity, with far-reaching scientific and clinical implications. As the GluN2 subunit saga moves on, the ‘‘tail’’ of 2B or not 2B remains an important component of the question. REFERENCES Cook, D.J., Teves, L., and Tymianski, M. (2012). Nature 483, 213–217. Deckel, A.W. (2001). J. Neurosci. Res. 64, 99–107.
428 Neuron 74, May 10, 2012 ª2012 Elsevier Inc.
Groc, L., Heine, M., Cousins, S.L., Stephenson, F.A., Lounis, B., Cognet, L., and Choquet, D. (2006). Proc. Natl. Acad. Sci. USA 103, 18769– 18774. Hardingham, G.E., Fukunaga, Y., and Bading, H. (2002). Nat. Neurosci. 5, 405–414. Heng, M.Y., Detloff, P.J., Wang, P.L., Tsien, J.Z., and Albin, R.L. (2009). J. Neurosci. 29, 3200–3205. Jocoy, E.L., Andre´, V.M., Cummings, D.M., Rao, S.P., Wu, N., Ramsey, A.J., Caron, M.G., Cepeda, C., and Levine, M.S. (2011). Front. Syst. Neurosci. 5, 28. Kaufman, A.M., Milnerwood, A.J., Sepers, M.D., Coquinco, A., She, K., Wang, L., Lee, H., Craig, A.M., Cynader, M., and Raymond, L.A. (2012). J. Neurosci. 32, 3992–4003. Levine, M.S., Cepeda, C., and Andre´, V.M. (2010). Neuron 65, 145–147. Liu, Y., Wong, T.P., Aarts, M., Rooyakkers, A., Liu, L., Lai, T.W., Wu, D.C., Lu, J., Tymianski, M., Craig, A.M., and Wang, Y.T. (2007). J. Neurosci. 27, 2846–2857. Martel, M.-A., Ryan, T.J., Bell, K.F.S., Fowler, J.H., McMahon, A., Al-Mubarak, B., Komiyama, N.H., Horsburgh, K., Kind, P.C., Grant, S.G.N., et al. (2012). Neuron 74, this issue, 543–556. Milnerwood, A.J., Gladding, C.M., Pouladi, M.A., Kaufman, A.M., Hines, R.M., Boyd, J.D., Ko, R.W., Vasuta, O.C., Graham, R.K., Hayden, M.R., et al. (2010). Neuron 65, 178–190. Mu¨ller, T., Albrecht, D., and Gebhardt, C. (2009). Learn. Mem. 16, 395–405. Okamoto, S., Pouladi, M.A., Talantova, M., Yao, D., Xia, P., Ehrnhoefer, D.E., Zaidi, R., Clemente, A., Kaul, M., Graham, R.K., et al. (2009). Nat. Med. 15, 1407–1413. Raymond, L.A., Andre´, V.M., Cepeda, C., Gladding, C.M., Milnerwood, A.J., and Levine, M.S. (2011). Neuroscience 198, 252–273. Starling, A.J., Andre´, V.M., Cepeda, C., de Lima, M., Chandler, S.H., and Levine, M.S. (2005). J. Neurosci. Res. 82, 377–386.