Elsevier

Neurobiology of Disease

Volume 93, September 2016, Pages 47-56
Neurobiology of Disease

GluN3A promotes NMDA spiking by enhancing synaptic transmission in Huntington's disease models

https://doi.org/10.1016/j.nbd.2016.04.001Get rights and content

Highlights

  • •

    GluN3A is required for the early multivariate synaptic phenotype in YAC128 striatum.

  • •

    GluN3A deletion prevents enhancement of both AMPAR and NMDAR components of EPSCs.

  • •

    Strong synaptic input to SPNs elicits electrogenic events termed NMDA spikes.

  • •

    Outsized EPSCs are caused by unclamped NMDA spikes, not extrasynaptic NMDARs.

  • •

    No evidence for more extrasynaptic NMDARs in YAC128 SPNs with new assay

Abstract

Age-inappropriate expression of juvenile NMDA receptors (NMDARs) containing GluN3A subunits has been linked to synapse loss and death of spiny projection neurons of the striatum (SPNs) in Huntington's disease (HD). Here we show that suppressing GluN3A expression prevents a multivariate synaptic transmission phenotype that precedes morphological signs at early prodromal stages. We start by confirming that afferent fiber stimulation elicits larger synaptic responses mediated by both AMPA receptors and NMDARs in SPNs in the YAC128 mouse model of HD. We then show that the enhancement mediated by both is fully prevented by suppressing GluN3A expression. Strong fiber-stimulation unexpectedly elicited robust NMDAR-mediated electrogenic events (termed “upstates” or “NMDA spikes”), and the effective threshold for induction was more than 2-fold lower in YAC128 SPNs because of the enhanced synaptic transmission. The threshold could be restored to control levels by suppressing GluN3A expression or by applying the weak NMDAR blocker memantine. However, the threshold was not affected by preventing glutamate spillover from synaptic clefts. Instead, long-lasting NMDAR responses interpreted previously as activation of extrasynaptic receptors by spilled-over glutamate were caused by NMDA spikes occurring in voltage clamp mode as escape potentials. Together, the results implicate GluN3A reactivation in a broad spectrum of early-stage synaptic transmission deficits in YAC128 mice; question the current concept that NMDAR mislocalization is the pathological trigger in HD; and introduce NMDA spikes as a new candidate mechanism for coupling NMDARs to neurodegeneration.

Introduction

Huntington's disease (HD) is an inherited neurodegenerative disorder characterized at late stages by massive loss of spiny projection neurons (SPNs) in the striatum (Vonsattel and DiFiglia, 1998); striatal SPNs are also known as medium spiny neurons and MSNs. A major challenge has been to map the chain of events that leads from the genetic mutation to disease and to identify the underlying molecular mechanisms so that disease progression could be slowed from early stages.

The earliest pathophysiological alteration identified to date downstream of mutant huntingtin (mHTT) expression is robust enhancement in synaptic transmission between glutamatergic afferents and SPNs. The phenotype is multivariate. Afferent synapses contain both NMDA- and AMPA-type glutamate receptors (i.e., NMDARs and AMPARs), and the enhancement in transmission is via both types (Cepeda, C, et al., 2003, Li, L, et al., 2004, Graham, RK, et al., 2009, Joshi, PR, et al., 2009). In addition, receptors expressed outside of synaptic clefts (i.e., extrasynaptic receptors) are thought to be involved (Milnerwood et al., 2010).

Of the two types of glutamate receptors, NMDAR hyperfunction has long been implicated in neurodegenerative diseases, and a specific role for subtypes containing the GluN3A subunit was recently identified in HD (Wesseling and Perez-Otano, 2015). GluN3A is ordinarily down-regulated during postnatal development, but less so in HD patients and mouse models because mHTT interferes with a mechanism for the selective endocytic removal of NMDARs that contain GluN3A (Perez-Otano, I, et al., 2006, Marco, S, et al., 2013). A causal role was demonstrated by knocking out GluN3A from the YAC128 mouse model, which prevented HD signs including synapse loss, cell death, and motor and cognitive decline (Marco et al., 2013); the YAC128 genome contains a yeast artificial chromosome encoding a pathogenic variant of mHTT under the control of the human promoter (Slow et al., 2003). Knocking out GluN3A additionally suppressed the early electrophysiological alterations attributed to greater numbers of extrasynaptic NMDARs (Marco et al., 2013).

The combination of observations is consistent with the current concept that glutamate spillover from synaptic clefts onto extrasynaptic NMDARs drives neurodegeneration. However, NMDARs expressed within synaptic clefts are involved in multiple cellular- and network-level phenomena that have also been linked to neurodegeneration (Raymond et al., 2011). It was therefore important to determine if the other aspects of the multivariate phenotype, specific for receptors within synaptic clefts, were suppressed by deleting GluN3A from YAC128 mice.

Finally, we introduce the concept of NMDA spikes (Major et al., 2013), which unexpectedly turned out to be key for understanding the totality of the multivariate phenotype. Although best known for gating long-term synaptic plasticity, NMDARs can additionally serve as voltage-gated cation channels in electrogenic events that are analogous to action potentials carried by voltage gated ion channels. For NMDA spikes, the Mg2 + block functions as the voltage gate, but NMDA spikes additionally require glutamate, which is typically supplied from afferent synaptic terminals. NMDA spikes had not been implicated in disease, nor identified previously in SPNs, but dendritic voltage upstates that might be driven by NMDA spikes have been identified in SPNs (Plotkin et al., 2011).

In the present study, we begin by showing that GluN3A is required for the enhanced neurotransmission via AMPARs and NMDARs located within synaptic clefts. We then show that strong afferent fiber stimulation thought to elicit glutamate spillover from synaptic clefts can induce robust NMDA spikes in SPNs. The effective threshold for induction was more than two-fold lower in the YAC128 model because of the enhanced synaptic transmission, and could therefore be restored to wild-type (WT) levels by deleting GluN3A. Memantine, which is thought to block extrasynaptic NMDARs, could reverse the lower threshold. Surprisingly, however, glutamate scavenger experiments indicated that glutamate spillover onto extrasynaptic receptors was not a key player. Instead, the electrophysiological evidence for greater numbers of extrasynaptic NMDARs was found instead to result from NMDA spikes that continued to occur in voltage clamp mode. Together, our results demonstrate that the entire multivariate synaptic phenotype seen at early stages in HD depends on GluN3A and does not include altered extrasynaptic NMDARs, as was previously thought. The results raise the possibility that the therapeutic efficacy of memantine may be by preventing NMDA spikes rather than by blocking extrasynaptic receptors.

Section snippets

Mouse breeding

Genetic background was homogenized using two different strategies, with mutually confirmatory results. The strategy used in (Marco et al., 2013) was used for the blind studies in Figs. 1A–C and 3. Briefly, YAC128 mice (line 55 homozygotes in a FVB/N background) (Graham et al., 2006) containing full-length mHTT with 128 CAG repeats were crossed with GluN3A −/− (also known as Grin3a −/−) homozygotes in a C57Bl6 background. The cross yielded a first generation of heterozygotes that were subsequently

Increased synaptic transmission mediated by both AMPARs and NMDARs in YAC128 mice requires GluN3A

To determine if GluN3A is required for the enhanced AMPAR component of synaptic responses to afferent fiber stimulation seen in YAC128 mice, we recorded excitatory post-synaptic currents (EPSCs) in patch-clamped SPNs in ex vivo coronal slices from YAC128 mice and control strains with and without GluN3A (Fig. 1A). We used 28–35 day old mice because the enhanced synaptic responses in YAC128 mice are already present at this stage, but alterations in neuropil morphology and connectivity that might

Discussion

A recent report suggested that reactivation of the expression of juvenile NMDARs containing GluN3A subunits plays a causal or at least permissive role in signs of HD progression such as synapse loss, neuron death, and motor deficits (Marco et al., 2013). Here we show that GluN3A is similarly permissive for earlier-stage multivariate enhancement of synaptic input to SPNs that was identified previously in the YAC128 and other mouse models of HD (Cepeda, C, et al., 2003, Li, L, et al., 2004,

Competing interests

All authors declare no competing interests.

Funding

This work was funded by the Unión Temporal de Empresas (UTE) project at the Centro de Investigación Médica Aplicada, Spanish Ministry of Science grants (BFU2009-12160 to J.F.W., SAF2010-20636 and CSD2008-00005 to I.P.-O., and SAF2013-48983R to I.P.-O. and J.F.W.), a Beca Josefina Garre, and a Fundació La Marató TV3 grant.

Author contributions

KM, MKR, and JFW performed experiments. KM, IPO and JFW designed experiments. SM, RMT, IPO and JFW designed the mouse breeding strategy. SM and RMT created and maintained the mouse colony and conducted genotyping. SM, RMT, and JFW provided mice in a blind fashion. KM and IPO helped edit the manuscript. JFW analyzed the data and wrote the manuscript.

Acknowledgments

We thank Aitor Zandueta for technical assistance and Drs. Juan Lerma, Guy Major, Michael Ehlers, Karl Magleby, Stephen Traynelis, Jon Johnson, Samuel Wang, Donald Lo, Tomás Aragón, and Montse Arrasate for discussions.

References (60)

  • S.M. Antonov et al.

    Permeant ion regulation of N-methyl-d-aspartate receptor channel block by Mg(2 +)

    Proc. Natl. Acad. Sci. U. S. A.

    (1999)
  • D. Bar-Yehuda et al.

    Space-clamp problems when voltage clamping neurons expressing voltage-gated conductances

    J. Neurophysiol.

    (2008)
  • E. Bezard et al.

    Pathophysiology of levodopa-induced dyskinesia: potential for new therapies

    Nat. Rev. Neurosci.

    (2001)
  • C. Cepeda et al.

    Transient and progressive electrophysiological alterations in the corticostriatal pathway in a mouse model of Huntington's disease

    J. Neurosci.

    (2003)
  • H.S. Chen et al.

    Open-channel block of N-methyl-d-aspartate (NMDA) responses by memantine: therapeutic advantage against NMDA receptor-mediated neurotoxicity

    J. Neurosci.

    (1992)
  • A.C. Costa

    The glutamatergic hypothesis for Down syndrome: the potential use of N-methyl-d-aspartate receptor antagonists to enhance cognition and decelerate neurodegeneration

    CNS Neurol. Disord. Drug Targets

    (2014)
  • J.T. Coyle et al.

    Oxidative stress, glutamate, and neurodegenerative disorders

    Science

    (1993)
  • D.M. Cummings et al.

    Alterations in striatal synaptic transmission are consistent across genetic mouse models of Huntington's disease

    ASN Neuro

    (2010)
  • M. Day et al.

    Differential excitability and modulation of striatal medium spiny neuron dendrites

    J. Neurosci.

    (2008)
  • R.K. Graham et al.

    Differential susceptibility to excitotoxic stress in YAC128 mouse models of Huntington disease between initiation and progression of disease

    J. Neurosci.

    (2009)
  • M.T. Harnett et al.

    Synaptic amplification by dendritic spines enhances input cooperativity

    Nature

    (2012)
  • N.A. Hessler et al.

    The probability of transmitter release at a mammalian central synapse

    Nature

    (1993)
  • P.R. Joshi et al.

    Age-dependent alterations of corticostriatal activity in the YAC128 mouse model of Huntington disease

    J. Neurosci.

    (2009)
  • S. Kang et al.

    CaV1.3-selective l-type calcium channel antagonists as potential new therapeutics for Parkinson's disease.

    Nat. Commun.

    (2012)
  • H.W. Kessels et al.

    Metabotropic NMDA receptor function is required for beta-amyloid-induced synaptic depression

    Proc. Natl. Acad. Sci. U. S. A.

    (2013)
  • G.J. Klapstein et al.

    Electrophysiological and morphological changes in striatal spiny neurons in R6/2 Huntington's disease transgenic mice

    J. Neurophysiol.

    (2001)
  • M.E. Larkum et al.

    Synaptic integration in tuft dendrites of layer 5 pyramidal neurons: a new unifying principle

    Science

    (2009)
  • R.S. Larsen et al.

    NR3A-containing NMDARs promote neurotransmitter release and spike timing-dependent plasticity

    Nat. Neurosci.

    (2011)
  • M.S. Levine et al.

    Enhanced sensitivity to N-methyl-d-aspartate receptor activation in transgenic and knockin mouse models of Huntington's disease

    J. Neurosci. Res.

    (1999)
  • L. Li et al.

    Enhanced striatal NR2B-containing N-methyl-d-aspartate receptor-mediated synaptic currents in a mouse model of Huntington disease

    J. Neurophysiol.

    (2004)

    • Cited by (21)

    • Design, synthesis and evaluation of novel 1-phenyl-pyrrolo[1,2-b]isoquinolin-3-one derivatives as antagonists for the glycine binding site of the NMDA receptor

      2023, European Journal of Medicinal Chemistry

    • Combined analysis of cytoarchitectonic, molecular and transcriptomic patterns reveal differences in brain organization across human functional brain systems.

      2022, NeuroImage
      Citation Excerpt :

      Such switches are responsible for synaptic plasticity and synaptic strength (Collingridge, Isaac, & Wang, 2004; Das et al., 1998; Gambrill & Barria, 2011; Lau & Zukin, 2007). A disturbance in the balance of receptors and their subunits is often associated with diseases, e.g., schizophrenia, epilepsy, neuropathic pain, Alzheimer's and Huntington's disease (Baulac et al., 2001; Hashimoto et al., 2008; Hines, Davies, Moss, & Maguire, 2012; Limon, Reyes-Ruiz, & Miledi, 2012; Mahfooz et al., 2016). The receptor subunit gene expression varied between different rodent brain regions, which seems to be relevant for the functional differentiation of the brain (Hörtnagl et al., 2013; Pirker, Schwarzer, Wieselthaler, Sieghart, & Sperk, 2000; Sieghart & Sperk, 2002; Wisden, Laurie, Monyer, & Seeburg, 1992).

    • Altered cortical processing of sensory input in Huntington disease mouse models

      2022, Neurobiology of Disease
      Citation Excerpt :

      In order to compare evoked EPSCs without large amplitude events, the lowest stimulation intensity required to evoke synaptic responses was used (not different between genotypes). However, more YAC128 neurons than WT neurons showed these large amplitude events, that have previously been described as “NMDAR spikes” (Poleg-Polsky, 2015; Palmer et al., 2014; Mahfooz et al., 2016) (*p = 0.04, Chi-square = 4.295, df = 1, n = 10 (6) YAC128 and n = 11 (7) WT) following minimal stimulation (Fig. 4A,B). Under basal conditions without glutamate spillover, YAC128 neurons showed a variable proportion of stimulation trials that evoked large events (Fig. 4C).

    • RNAi-Based GluN3A Silencing Prevents and Reverses Disease Phenotypes Induced by Mutant huntingtin

      2018, Molecular Therapy
      Citation Excerpt :

      Third, the approach targets one of the earliest disease mechanisms when intervention is more likely to be efficacious and before a point of no return has been reached.27 Electrophysiological and morphological evidence of early dysfunction and loss of MSN synapses is extensive in HD mouse models9,18 and humans.11,12 Likewise, longitudinal imaging and functional studies in humans report significant striatal atrophy years prior to diagnosable HD,28 which is strongly correlated with time to disease onset, performance, and clinical progression.14,15

    • Kinetic models for activation and modulation of NMDA receptor subtypes

      2018, Current Opinion in Physiology
      Citation Excerpt :

      Therefore, diheteromeric GluN1/GluN3 channels are glycinergic and insensitive to glutamate or other GluN2-specific ligands. No kinetic model is yet available for GluN1/GluN3 receptors, although they play important roles in synaptogenesis and neurodegeneration [30,31]. Results reported to date indicate that their macroscopic behaviors and mechanisms are distinct in many aspects from their glutamatergic GluN1/GluN2 brethren and are likely more similar to those of AMPA and Kainate receptors [32].

    • Synaptopathic mechanisms of neurodegeneration and dementia: Insights from Huntington's disease

      2017, Progress in Neurobiology
      Citation Excerpt :

      Although targeting NR2B could be beneficial, subunit-selective pharmacology is generally imperfect (reviewed in Köhr, 2006). In accordance with the differences between Syn-NMDAR and Ex-NMDAR function, low doses of memantine, a voltage-dependent, uncompetitive NMDAR antagonist (Parsons et al., 2007), can block Ex-NMDARs, ameliorate neuropathological and behavioural manifestations (Okamoto et al., 2009; Milnerwood et al., 2010) and normalise the over-activated NMDAR-mediated transmission in the striatum of YAC128 mice, possibly via GluN3A blockade (Mahfooz et al., 2016). This effect was lost using high doses of memantine, which blocks both Syn- and Ex-NMDARs.

    View all citing articles on Scopus
    View full text
    © 2016 Elsevier Inc. All rights reserved.