Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: J Physiol. 2020 Mar 12;599(2):431–441. doi: 10.1113/JP278702

Functional interactions of ion channels with the actin cytoskeleton: does coupling to dynamic actin regulate NMDA receptors?

Juliana E Shaw 1, Anthony J Koleske 1,2
PMCID: PMC7416480  NIHMSID: NIHMS1558866  PMID: 32034761

Abstract

Synapses are enriched in the cytoskeletal protein actin, which determines the shape of the pre- and post-synaptic compartments, organizes the neurotransmitter release machinery, and provides a framework for trafficking of components. In the post-synaptic compartment, interactions with actin or its associated proteins are also critical for the localization and activity of synaptic neurotransmitter receptors and ion channels. Actin binding proteins, including spectrin and alpha-actinin, serve as molecular linkages between the actin cytoskeleton and a diverse collection of receptors, including the NMDA receptor (NMDAR) and voltage gated sodium channels. The actin cytoskeleton can regulate neurotransmitter receptors and ion channels by controlling their trafficking and localization at the synapse and by directly gating receptor channel opening. We highlight evidence that synaptic actin couples physically and functionally to NMDAR and supports its activity. The molecular mechanisms by which actin regulates NMDARs are only just emerging, and recent advancements in light and electron microscopy-based imaging techniques should aide in elucidating these mechanisms.

Graphical Abstract.

Dendritic spines are enriched with actin and a wealth of literature suggests a direct interaction between this cytoskeleton component and neurotransmitter receptors. We would like to highlight the physical and functional link between NMDA receptors and actin in this review.

Introduction

Fast neurotransmission at excitatory glutamatergic synapses is mediated primarily by two types of postsynaptic glutamate receptors: α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) and N-methyl-D-aspartate receptors (NMDARs). In response to glutamate release at the synapse, AMPA receptors (AMPARs) exhibit fast opening and closing kinetics and are permeable to Na+. At resting membrane potentials, NMDA receptors (NMDARs) are blocked by Mg2+, which is removed by AMPAR-mediated membrane depolarization. Binding of glutamate and a glycine or serine coagonist yields a slower, longer duration opening of NMDARs, allowing them to conduct Na+ and Ca2+ ions. Influx of Ca2+ through these channels is a critical trigger for activation of Ca2+-dependent signaling and cytoskeletal regulatory proteins which orchestrate changes in synaptic structure and neurotransmitter receptor function (Seeburg et al., 1995; Sibarov & Antonov, 2018). Given the central role of Ca2+ in mediating these plastic changes in synapses, it is not surprising that NMDARs are essential regulators of synaptic plasticity and that disruption of their function impairs learning and memory (Traynelis et al., 2010).

The cytoskeleton supports almost every cell biological process in eukaryotes, including neuronal migration, axon and dendrite morphogenesis, and synaptogenesis. Synapses are particularly enriched in factors that are responsive to Ca2+ influx, such as kinases and phosphatases, and the actin cytoskeleton is a central signaling hub for these molecules (Morishita et al., 2001; Lisman et al., 2002). Actin filaments are perfectly poised to participate in dynamic changes in synaptic structure and function in response to neuronal activity. Macroscopic changes in synapse size and shape, powered by dynamic actin rearrangements, have been the focus of many pioneering studies (Bourne & Harris, 2008; Hotulainen & Hoogenraad, 2010). Actin filament networks are enriched in the pre- and post- synaptic compartments, where they help to organize the neurotransmitter release and neurotransmitter detection machinery, respectively (Benke et al., 1993; Borgdorff & Choquet, 2002; Kevenaar & Hoogenraad, 2015). Dendritic spines are very small compartments (typically 0.05–0.5 μm3) enriched in structural and signaling proteins, making them a crucible for a variety of biochemical interactions. The high concentration of actin in the spine positions it to act as a platform and regulator of a host of diverse biochemical events at the synapse.

A growing number of studies have begun to demonstrate direct physical and functional interactions between neurotransmitter receptors, or other ion channels, and the actin cytoskeleton. Here, we review the mechanisms by which actin filaments engage ion channels and their synaptic tethers to regulate channel function, with a specific focus on how actin interacts functionally with the NMDAR ion channels to regulate their activity.

Actin filament networks are highly dynamic at the synapse.

Actin is the most abundant cytoskeletal component found in mature synapses (Fifkova & Delay, 1982; Matus et al., 1982), where it plays diverse roles in synaptogenesis, synapse maintenance and synaptic plasticity. The spine actin network polymerizes from a freely diffusing pool of actin monomers into an elaborate network of actin filaments. At steady state, actin preferentially polymerizes at the plus end of the filament, in an ATP-dependent manner, while actin monomers are lost from the minus end (Fig. 1A). This difference in polymerization behavior at either end leads to net turnover of the filaments (Pollard & Cooper, 2009). A large collection of actin regulatory proteins act on this process to nucleate new actin filaments, stabilize and/or cross-link filaments, prevent actin monomer addition to the growing filament, and even sever filaments (Fig. 1B). The net action of these regulators creates an elaborate network of linear and branched polymers that support overall spine structure (Fig. 1C)(Racz & Weinberg, 2004, 2006, 2008; Korobova & Svitkina, 2010). Importantly, filaments in this actin network extend into parts of the neurotransmitter receptor-containing post-synaptic density (PSD) (Rostaing et al., 2006), the first hint that actin may play a role in regulating synaptic activity.

Figure 1. Regulation of actin and its organization in dendritic spines.

Figure 1.

Open in a new tab

A. ATP-actin monomers are added to the rapidly growing plus end. Actin bound ATP hydrolyzes to ADP•Pi- actin, which releases Pi to create ADP- actin. ADP-actin monomers dissociate from the minus end.

B. Actin is regulated by a host of proteins that crosslink filaments, sever them, nucleate new actin branches, sequester monomers, and cap actin ends.

C. Diagram illustrates actin organization in the dendritic spine and its interactions with various key regulators. Actin in the center and base of the spine (stable core) undergoes lower rates of polymerization and depolymerization. Actin in the periphery undergoes more significant polymerization and depolymerization, forming a dynamic shell. New actin polymerization is especially high adjacent to the postsynaptic density.

Fluorescent imaging techniques have revealed the dynamic and heterogeneous behavior of the actin network within dendritic spines. Fluorescence recovery after photobleaching (FRAP) experiments of GFP-actin in dendritic spines surprisingly show that 85% of spine actin is highly dynamic and turns over very rapidly (~40 seconds) (Star et al., 2002; Honkura et al., 2008). Bulk flow measurements of photoactivatable (PA) GFP-actin have enabled the study of the portion of actin that turns over more slowly. Using this technique, Honkura and colleagues showed that stable pools of actin reside at the base of the spine head, turning over 25-fold slower than the dynamic pool (time constant of 17 minutes) (Honkura et al., 2008). Cytochalasins and latrunculins are toxins that prevent actin polymerization by blocking addition of actin monomers, via binding to the plus end or sequestering actin monomers, respectively. Consistent with bulk measurements of actin flux, these compounds cause significant depolymerization of actin filaments in spines, but have comparatively little effect on more stable filaments that turn over slowly at the spine base (Allison et al., 1998; Chazeau & Giannone, 2016).

Optical tracking of single tagged-actin particles has provided a higher resolution perspective on actin dynamics in spines. These analyses have revealed that actin polymerization is highly heterogeneous, in direction and speed, within the spine and is enriched in “hotspots” (Frost et al., 2010; Chazeau et al., 2014; Chazeau & Giannone, 2016). These single-particle tracking experiments corroborate early electron microscopy showing that some actin filaments are juxtaposed to the PSD, suggesting that there could be a physical or functional coupling of actin to neurotransmitter machinery.

Actin regulates ion channels in diverse ways.

Given its dynamic behavior and localization, it is not surprising that actin impacts ion channels via a diversity of mechanisms. This regulation may be indirect - alterations in actin clearly impact the delivery and clustering of channel proteins at the synapse (we expand on this in a following section). However, the actin cytoskeleton can also directly interact with ion channels in diverse ways to regulate their conductance, open probability, or inactivation (Janmey, 1998), as summarized in Figure 2.

Figure 2. Models for direct regulation of ion channels by actin.

Figure 2.

Open in a new tab

A. Model 1. Actin filaments may interact with an ion channel and keep it clamped in a closed conformation. Depolymerization of actin would relieve this constraint and allow activation of the channel.

B. Model 2. Actin filaments may bind the channel and promote its opening.

C. Model 3. Binding of actin-associated tethering proteins may bind to the channel and promote its opening.

In the case of the nonvoltage gated sodium channel, the capping of cortical actin filament plus ends with cytochalasin causes the receptors to become activated (Shumilina et al., 2003). These findings suggest a mechanism in which interactions with actin hold the receptor in an inactivated conformation that is relieved by loss of coupling to actin (Fig. 2A). In the case of the 305 pS chloride channel and the potassium channel Kv3.3 (Schwiebert et al., 1994; Zhang et al., 2016), actin disassembly inhibits ion conductance, supporting a mechanism in which actin positively supports an open-pore conformation of the receptors (Fig. 2B). While direct actin binding may influence activation/inactivation of particular receptors, the actin filaments also act as a binding platform for a number of signaling and scaffolding molecules that may impact ion channel activity. Therefore, it is also plausible that changes in localization of a cytoskeleton-bound signaling/scaffolding protein may impact the gating of the ion channel (Fig. 2C)(Prat et al., 1995).

Regulation of ion channel trafficking and localization by actin.

Synaptic neurotransmitter receptors are concentrated at the PSD, which closely appose the presynaptic active zone, to ensure optimal responsiveness to neurotransmitter release. The PSD is densely packed with neurotransmitter receptors, scaffolding proteins, cell-adhesion molecules and signaling proteins. Postsynaptic efficacy is altered by the dynamic modification of the number of surface receptors diffusing in and out of the PSD (Choquet & Triller, 2003). Actin filaments are known to insert into this PSD (Bloom et al., 2003; Rostaing et al., 2006), where they may play a passive, more steric, role in limiting diffusion in and out of the synapse. However, recent studies also suggest active roles for actin filaments in regulating receptor mobility.

An array of techniques has been used to show that synaptic neurotransmission receptors and ion channels form clusters at the synapse or at discrete extrasynaptic locations in neurons. These observations argue against a model in which these synaptic membrane proteins freely diffuse in the membrane, but suggest instead that they are anchored to an underlying substructure. This tether has been suggested to be actin itself or scaffolding proteins that couple to actin filaments. In this section, we summarize data suggesting different synaptic cell surface receptors and ion channels couple to actin and associated scaffolds.

Voltage gated sodium channels

The earliest evidence for the role for actin in localized clustering came from the study of voltage-gated sodium channels (VGSC) in axons (Angelides et al., 1988). VGSCs are critical for the initiation and propagation of action potentials (Offord & Catterall, 1989). VGSCs are diffusely localized on the cell body, but exhibit punctate localization at the axon hillock, presynaptic terminals, and at focal points of the axon known as nodes of Ranvier. Interestingly, VGSC diffusion within the membrane on the cell body is an order of magnitude faster than their diffusion at the axon hillock, and channels are prevented from diffusing between the cell body and the axon proper, suggesting a coupling of VGSCs to the underlying cytoskeleton at the hillock (Angelides et al., 1988). In mice that have a genetic defect in Schwann cell development and myelinogenesis, neurons contain patches of receptors of similar size and distribution to that of normal membranes, indicating that the receptors are confined by internal cellular constraints and not by Schwann cells or axon:glial junctions (Ellisman, 1979).

Indeed, ankyrin co-purifies with VGSCs in native complexes from brain tissue, providing a link between VGSCs and the actin cytoskeleton (Srinivasan et al., 1988). Although ankyrin is widely associated with the membrane, its binding to VGSCs is high affinity, stoichiometric, and specific, as ankyrin does not bind GABA receptors or DHP receptors (Srinivasan et al., 1988). Indeed, the II-III linker of VSGC binds ankyrin and is essential for ankyrin-dependent sorting to these clusters (Garrido et al., 2003). These observations provide direct evidence for a linkage to actin via ankyrin, suggesting this critical link localizes VGSCs to the nodes.

AMPAR

There are two well-accepted pathways that concentrate AMPARs in the synapse – one path regulates AMPAR exo/endocytosis and the other regulates AMPAR movement through the membrane from extrasynaptic sites to the PSD. Exchange by diffusion into and out of the synapse is thought to account for the fastest changes in AMPAR number, allowing for rapid removal of damaged receptors and rapid changes in neurotransmission as a spine undergoes synaptic plasticity (Borgdorff & Choquet, 2002; Tardin et al., 2003). However, since multiple recycling pathways exist, only a minor population of these receptors are probably confined and diffusion-limited by actin, with confinement increasing with maturity of the neuron (Borgdorff & Choquet, 2002). This perspective is also supported by the observation that 75% of AMPARs are extractable with the non-ionic detergent Triton X-100, usually capable of only solubilizing plasma membrane and free receptors, leaving behind cytoskeleton-associated receptors. This contrasts notably with the behavior of NMDARs, which are virtually non-extractable with this method (Allison et al., 1998).

There is good evidence, however, that clustering of a portion of AMPARs at synapses depends on actin. Latrunculin A-mediated actin depolymerization causes 40% of synapses to lose synaptic clusters of AMPARs, suggesting that actin serves as a tether to anchor AMPARs at the synapse (Allison et al., 1998). Quantum dot tracking of AMPARs show that diffusion of the receptor is faster extrasynaptically than synaptically, owing to the fact that synaptic receptors are tightly enmeshed in the PSD. On the other hand, dynamic actin promotes fast diffusion of extrasynaptic receptors, since conditional knock out of the actin severing protein n-cofilin stabilizes actin filaments and subsequently slows diffusion of the extrasynaptic receptors, with no change in diffusion within the synapse. Furthermore, the use of actin interfering drugs leads to changes in synaptic and extrasynaptic receptor diffusion, demonstrating the complexity of controlling AMPAR availability (Rust et al., 2010). These data support a model in which dynamic cortical actin supports rapid AMPAR diffusion and stabilization of this network results in stabilization of receptor diffusion. This dynamic behavior helps to rapidly fine tune AMPAR localization at the synapse.

NMDAR

NMDARs belong to the same class of glutamate receptors as AMPARs, but differ significantly in their gating properties, ion selectivity, regulation by coagonists, and tethering to actin. Synaptic stimulation of NMDARs activates signaling cascades that are critical for synaptic plasticity, while activation of extrasynaptic NMDAR receptors can cause excitotoxic cell death (Paoletti & Ascher, 1994; Hardingham & Bading, 2003; Hansen et al., 2014). Live imaging of NMDARs labeled with a fluorescent conantoxin derivative provided the first evidence of physical NMDAR coupling to the cytoskeleton. NMDARs clustered to synapses and FRAP experiments revealed that most of the NMDARs do not freely exchange, indicating immobilization at the synapse and suggesting direct coupling to the cytoskeleton (Benke et al., 1993). The dependence of this direct coupling on actin was demonstrated by Allison et al. who showed that Latrunculin A-mediated actin depolymerization caused NMDAR clusters to disperse from synapses and coalesce into large non-synaptic cell body clusters (Allison et al., 1998). Residual NMDAR receptor clusters may be held together via the PSD-95 synaptic scaffolding protein, which colocalizes with these non-synaptic clusters following actin depolymerization. The actin binding protein, α-actinin-2, which binds the NMDAR GluN1 and GluN2B subunit cytoplasmic tails (Wyszynski et al., 1997; Krupp et al., 1999), was dispersed from NMDARs upon actin depolymerization (Allison et al., 1998), suggesting that this protein is not essential for receptor clustering, but more likely acts to modulate receptor function in the spine. The dispersion of these receptors from synaptic contacts following actin depolymerization demonstrates the key role for actin in NMDAR tethering in spines (Fig. 3). The remainder of this review focuses on the direct and indirect interactions of NMDARs with actin. We highlight how these interactions impact NMDAR tethering at synapses and also discuss the evidence that actin may directly modulate NMDAR receptor activity.

Figure 3. Coupling of NMDA receptors to the cytoskeleton.

Figure 3.

Open in a new tab

Enlargement of the postsynaptic density shows NMDARS with distinct receptor compositions. Receptors containing GluN2B can bind to PSD95 and related PDZ domain-containing scaffold proteins, which in turn interact with GKAP. Actin appears to be important for GKAP interactions with Shank1.

NMDA receptors form extensive interactions with the actin cytoskeleton

The finding that Latrunculin A decreased clustering of NMDARs and α-actinin-2 at synaptic sites provided direct evidence that NMDAR tethering at the synapse required coupling to an underlying actin network (Allison et al., 1998; Sattler et al., 2000). Indeed, early protein interaction screens identified the actin-binding protein α-actinin-2 as a direct binder of NMDAR GluN1 and GluN2B subunit cytoplasmic tails (Wyszynski et al., 1997; Krupp et al., 1999), while gel overlay assays indicated binding of spectrin, another actin-binding protein, to GluN1, GluN2A, and GluN2B (Wechsler & Teichberg, 1998) (Figure 4 illustrates the known interactions of actin binding proteins to the GluN1, GluN2A, and GluN2B cytoplasmic tails, as well as potential sites of regulation by phosphorylation). Subsequent proteomic analyses of NMDAR associated proteins by mass spectrometry has affirmed that NMDARs associate with actin and a rich network of cytoskeletal proteins, including myosin IIB, paxillin, ezrin, and cortactin in addition to α-actinin-2 and spectrin (Husi et al., 2000).

Figure 4. Protein interaction and phosphorylation sites on the NMDA receptor subunit cytoplasmic tails.

Figure 4.

Open in a new tab

The diagram highlights the proteins that are known to bind to the intracellular tails of the GluN1, GluN2A, and GluN2B subunits, as well as the known phosphorylation sites. The intracellular tails are zoomed in 2 times compared to the rest of the domains to enhance resolution of binding partners and phosphorylation sites. The GluN2A and GluN2B cytoplasmic tails share 32% amino acid identity and 48% similarity. The boxes aligned below each tail depict the region of the tail known to mediate interactions with each specific binding protein. Certain splice variants of the GluN1 subunit and each of the GluN2 subunits has C-terminal Serine - Aspartate - Valine (SDV) motif that can bind the indicated PDZ domain-containing synaptic scaffolding proteins. See text for more details.

Many parallel observations also support the concept that the actin cytoskeleton is critical for proper localization of NMDAR in the spine. In addition to the aforementioned dependence of NMDAR clustering on actin, disruption of actin with Latrunculin A reduces the coupling between the NMDAR-associated PSD scaffold proteins GKAP/DLGAP1 (guanylate kinase associated protein/Discs large-associated protein 1) and Shank1 (SH3 and multiple ankyrin repeat domains protein 1) and increases their turnover in spines, as revealed by FRAP (Kuriu et al., 2006). These data indicate that actin filaments help to keep these scaffolding proteins anchored to one another and properly localized within the spine. Further, actin polymerization hotspots lie just adjacent to the NMDAR-containing postsynaptic density (Frost et al., 2010). All together, these findings suggest that NMDARs couple via scaffolding proteins to a dynamic pool of actin filaments.

More recent experiments affirm that NMDAR coupling to the synaptic cytoskeletal scaffold is dynamic and may be differentially regulated, depending on receptor subunit composition. NMDARs are composed of two GluN1 subunits and two of any combination of GluN2 subunits (Traynelis et al., 2010; Paoletti et al., 2013; Hansen et al., 2014). In the cortex and hippocampus GluN1:GluN2B diheteromers predominate in immature synapses, but GluN2A expression increases during maturation such that GluN1:GluN2A diheteromers and GluN1:GluN2A:GluN2B triheteromers predominate in mature synapses.

Different NMDA receptors employ distinct synaptic localization mechanisms, depending on their GluN2 subunit composition. When overexpressed in cerebellar granule neurons, GluN2 subunits can assemble with endogenous GluN1 into functional receptors that can be detected at the synapse by an enhancement of NMDAR currents and changes in NMDAR-EPSC decay rates. Mutation of a key residue (S1480A) within a motif that mediates binding to actin-associated, PDZ domain-containing, scaffolding proteins prevents GluN2B localization to the synapse (Prybylowski et al., 2005). Similar manipulations to the PDZ binding motif in GluN2A do not impact its incorporation under these conditions, indicating that GluN1:GluN2A and GluN1:GluN2B receptors tether to the synapse via distinct anchoring mechanisms. This differential tethering of distinct NMDARs within the synapse has also been affirmed by live imaging studies. Single molecule quantum dot labeling of individual GluN2A or GluN2B molecules in immature synapses reveals that both receptors diffuse slowly within the confines of the postsynaptic density. Surprisingly, chemical treatments that induce enhancement of AMPAR-mediated transmission (aka long-term potentiation or LTP) cause GluN2B to diffuse at higher rates into regions that lie outside the PSD (Dupuis et al., 2014). Antibody crosslinking of GluN2B-containing receptors to restrict diffusion, block LTP, suggesting that GluN2B diffusion may play a role in signaling events or recruiting other essential components to the synapse required for LTP.

Actin may also be important for NMDAR trafficking to the synapse.

In addition to tethering, one recent study suggests that actin cytoskeleton-based mechanisms are critical for activity-dependent trafficking of NMDARs to the synapse. In this model, chemical activation of protein kinase C (PKC) activity increased functional NMDAR density at synapses. Interestingly, this process was inhibited by knockdown or inhibition of myosin IIB as well as disruption of actin with Latrunculin A (Bu et al., 2015). Given that myosin IIB is not known to transport cargoes, it is unlikely that it is directly shuttling NMDARs in this case. Instead, it is possible that myosin IIB might locally increase tension on the actin cytoskeleton, an interaction that could modulate NMDAR activation. PKC-mediated phosphorylation of GluN2A at S1291 and S1312 is also known to increase NMDAR currents (Chen & Roche, 2007), but it is unclear whether this stimulation depends on myosin IIB-based mechanisms. While the evidence points to a role for the actin-based cytoskeleton in modulating NMDAR levels at the synapse, the exact mechanisms that govern this process are not yet well understood.

Activity of NMDA receptors is dependent on the integrity of the actin cytoskeleton.

Longitudinal recordings of NMDARs in hippocampal neurons under patch clamp has demonstrated that their activity is critically linked to actin cytoskeletal structure. Repeated stimulation of NMDARs over 10s of minutes in cultured hippocampal neurons leads to a long-term reduction in peak NMDAR-mediated currents as measured by patch clamp, a phenomenon known as “rundown” (Legendre et al., 1993; Rosenmund & Westbrook, 1993; Rosenmund et al., 1995). Ca2+ influx via NMDARs is essential for rundown and this phenomenon is not observed in kainate or AMPA glutamate receptors, indicating that (1) there is a tight spatial coupling between Ca2+ influx via NMDARs and the rundown process and (2) one or more Ca2+-dependent processes is a key mediator. Many lines of evidence identified actin depolymerization as a key mediator of rundown. First and foremost, treatment of neurons with the actin stabilizing drug phalloidin was sufficient to prevent rundown, while treatment of neurons with cytochalasin D, to inhibit actin polymerization, was sufficient to induce rundown (Rosenmund & Westbrook, 1993). A major target of Ca2+ during rundown appears to be gelsolin, an actin severing protein that is required for NMDAR rundown (Furukawa et al., 1997). Rundown also appears to require the activity of the actin regulatory RhoA GTPase, as rundown can be inhibited by the C3 toxin that inhibits RhoA (Norenberg et al., 1999).

A recent study from our lab strongly suggests that coupling to dynamic actin networks might be particularly important for activity of GluN2B subunit-containing NMDARs (GluN2B-NMDARs). Our group found that phospho-Y1252 in the GluN2B cytoplasmic tail serves as a binding site for the Nck2 adaptor protein, which partners with N-WASp to activate actin branch nucleation by the Arp2/3 complex (Levy et al., 2018). Interestingly, hyperactivating mutations in the SHP2 tyrosine phosphatase associated with Noonan Syndrome, dephosphorylate phospho-Y1252 in GluN2B. This causes Nck2 to become untethered from GluN2B, reducing the levels of Nck2 found in the post-synaptic compartment and ultimately causing a reduction of dynamic actin in the spine. While GluN2B levels at the neuronal cell surface are unaffected, loss of phospho-Y1252 and uncoupling from Nck2 are associated with a nearly complete loss of GluN2B-NMDAR activity at the synapse. These defects likely are a contributing factor to the cognitive impairments associated with Noonan Syndrome.

Another example of actin regulating NMDAR activity comes from studies of the Abl2 nonreceptor tyrosine kinase, which binds to and stabilizes actin filaments in vitro (Courtemanche et al., 2015). Loss of Abl2 function in mice is associated with exuberant GluN2B-NMDAR-mediated currents at a subset of hippocampal synapses. These synapses enlarge and have even larger currents as the mice mature and this is accompanied by a net loss of synapses (Xiao et al., 2016). One possibility is that the growing synapses outcompete their neighbors for some limiting building block, possibly actin or a factor that couples the receptor to the actin cytoskeleton. Alternatively, enlargement could be a result of homeostatic scaling as an attempt of the synapse to normalize NMDAR function. These are just a few examples of likely several mechanisms by which actin may impact NMDAR activity.

Why would actin regulate receptor function?

The findings that NMDARs are coupled physically and functionally to the actin cytoskeleton raises the key question of why NMDARs are coupled to actin. NMDARs are mechanosensitive – disruption of tension on the receptor eliminates its ability to function (Paoletti & Ascher, 1994). This tension may be essential for the receptor to undergo conformational changes required for gating of the conductance pore.

By bridging multiple receptors, networks of actin filaments may enable NMDARs to communicate with each other within a cluster. Here, it is important to note that tetanic NMDA receptor stimulation can increase actin polymerization at the synapse (Okamoto et al., 2004). It is possible that this local actin polymerization may act as a feed forward signal to prime NMDARs at the synapse to be more sensitive to activation by glutamate. This increased sensitivity may involve spatial constraint – the actin may promote clustering of NMDARs to optimize their activation by presynaptically released glutamate. It is also tantalizing to speculate that newly polymerized actin might promote NMDA receptor opening. This could occur in a manner similar to integrin receptors, in which protein interactions with the intracellular tail of integrins evoke wholesale changes in the conformation of the receptor that alter their affinities for extracellular binding ligands (Calderwood et al., 2013).

It is also possible that the coupling of actin to NMDARs limits their activity under pathological states. Under situations where elevated Ca2+ influx via NMDARs might provoke excitotoxicity, Ca2+-mediated actin depolymerization might serve to protect the neuron from this insult. In support of this hypothesis, oxygen glucose deprivation causes NMDAR-dependent excitotoxicity via elevated synaptic glutamate release (Sattler et al., 2000). Depolymerization of actin with Latrunculin A significantly reduces excitotoxicity resulting from this energy stress.

Ongoing and future challenges in studying how actin regulates NMDAR function

The biggest challenges in understanding how actin interacts functionally with NMDARs will be to define the key underlying molecular interactions and measure how they impact NMDAR localization and function. While multiple actin-interacting proteins bind NMDAR cytoplasmic tails or are found in complexes with the receptor (Husi et al., 2000), it is not clear which of these are essential for NMDAR localization or function. Given the very high actin polymerization rates identified just adjacent to the post-synaptic membrane, it is likely that NMDARs and other components of the postsynaptic density are coupled in an extraordinarily dynamic fashion, mediated by multiple dynamic interactions. Super-resolution and single particle light microscopy approaches offer the potential to identify how manipulations of key molecules impact NMDAR localization and function. Advances in detector technology and sample throughput in cryo-electron tomography may offer the potential to directly visualize these precise molecular networks. This could set the stage to understand how these NMDAR:actin connections regulate NMDAR function and how they are dynamically altered throughout the life of a synapse to impact synaptic activity and plasticity.

Acknowledgments

Work in our lab is supported by NIH Grants R01-NS105640 and R01-MH115939 (to A.J.K.). J.E.S. was supported by NIH training grant T32GM007223 and is currently supported by F31MH116571. We thank members of the Koleske lab for daily interactions and especially Josie Bircher, Amanda Jeng, and Yevheniia Ishchenko for helpful comments on the manuscript. We dedicate this work to our colleague Jim Howe.

Abbreviations:

AMPAR

α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor

FRAP

fluorescence recovery after photobleaching

LTP

Long-term potentiation

NMDAR

N-methyl-D-aspartate receptor

PA

photoactivation

PKC

Protein kinase C

PSD

post-synaptic density

VGSC

voltage-gated sodium channels

Footnotes

The authors declare no competing financial interests.

References

  1. Allison DW, Gelfand VI, Spector I & Craig AM. (1998). Role of actin in anchoring postsynaptic receptors in cultured hippocampal neurons: differential attachment of NMDA versus AMPA receptors. J Neurosci 18, 2423–2436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Angelides KJ, Elmer LW, Loftus D & Elson E. (1988). Distribution and lateral mobility of voltage-dependent sodium channels in neurons. J Cell Biol 106, 1911–1925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Benke D, Marti T, Heckendorn R, Rehm H, Kunzi R, Allgeier H, Angst C & Mohler H. (1993). Photoaffinity labeling of the NMDA receptor. Eur J Pharmacol 246, 179–180. [DOI] [PubMed] [Google Scholar]
  4. Bloom O, Evergren E, Tomilin N, Kjaerulff O, Low P, Brodin L, Pieribone VA, Greengard P & Shupliakov O. (2003). Colocalization of synapsin and actin during synaptic vesicle recycling. J Cell Biol 161, 737–747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Borgdorff AJ & Choquet D. (2002). Regulation of AMPA receptor lateral movements. Nature 417, 649–653. [DOI] [PubMed] [Google Scholar]
  6. Bourne JN & Harris KM. (2008). Balancing structure and function at hippocampal dendritic spines. Annual review of neuroscience 31, 47–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bu Y, Wang N, Wang S, Sheng T, Tian T, Chen L, Pan W, Zhu M, Luo J & Lu W. (2015). Myosin IIb-dependent Regulation of Actin Dynamics Is Required for N-Methyl-D-aspartate Receptor Trafficking during Synaptic Plasticity. J Biol Chem 290, 25395–25410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Calderwood DA, Campbell ID & Critchley DR. (2013). Talins and kindlins: partners in integrin-mediated adhesion. Nat Rev Mol Cell Biol 14, 503–517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chazeau A & Giannone G. (2016). Organization and dynamics of the actin cytoskeleton during dendritic spine morphological remodeling. Cell Mol Life Sci. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chazeau A, Mehidi A, Nair D, Gautier JJ, Leduc C, Chamma I, Kage F, Kechkar A, Thoumine O, Rottner K, Choquet D, Gautreau A, Sibarita JB & Giannone G. (2014). Nanoscale segregation of actin nucleation and elongation factors determines dendritic spine protrusion. EMBO J 33, 2745–2764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen BS & Roche KW. (2007). Regulation of NMDA receptors by phosphorylation. Neuropharmacology 53, 362–368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Choquet D & Triller A. (2003). The role of receptor diffusion in the organization of the postsynaptic membrane. Nat Rev Neurosci 4, 251–265. [DOI] [PubMed] [Google Scholar]
  13. Courtemanche N, Gifford SM, Simpson MA, Pollard TD & Koleske AJ. (2015). Abl2/Abl-related gene stabilizes actin filaments, stimulates actin branching by actin-related protein 2/3 complex, and promotes actin filament severing by cofilin. J Biol Chem 290, 4038–4046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dupuis JP, Ladepeche L, Seth H, Bard L, Varela J, Mikasova L, Bouchet D, Rogemond V, Honnorat J, Hanse E & Groc L. (2014). Surface dynamics of GluN2B-NMDA receptors controls plasticity of maturing glutamate synapses. EMBO J 33, 842–861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ellisman MH. (1979). Molecular specializations of the axon membrane at nodes of Ranvier are not dependent upon myelination. J Neurocytol 8, 719–735. [DOI] [PubMed] [Google Scholar]
  16. Fifkova E & Delay RJ. (1982). Cytoplasmic actin in neuronal processes as a possible mediator of synaptic plasticity. J Cell Biol 95, 345–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Frost NA, Shroff H, Kong H, Betzig E & Blanpied TA. (2010). Single-molecule discrimination of discrete perisynaptic and distributed sites of actin filament assembly within dendritic spines. Neuron 67, 86–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Furukawa K, Fu W, Li Y, Witke W, Kwiatkowski DJ & Mattson MP. (1997). The actin-severing protein gelsolin modulates calcium channel and NMDA receptor activities and vulnerability to excitotoxicity in hippocampal neurons. J Neurosci 17, 8178–8186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Garrido JJ, Giraud P, Carlier E, Fernandes F, Moussif A, Fache MP, Debanne D & Dargent B. (2003). A targeting motif involved in sodium channel clustering at the axonal initial segment. Science 300, 2091–2094. [DOI] [PubMed] [Google Scholar]
  20. Hansen KB, Ogden KK, Yuan H & Traynelis SF. (2014). Distinct functional and pharmacological properties of Triheteromeric GluN1/GluN2A/GluN2B NMDA receptors. Neuron 81, 1084–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hardingham GE & Bading H. (2003). The Yin and Yang of NMDA receptor signalling. Trends Neurosci 26, 81–89. [DOI] [PubMed] [Google Scholar]
  22. Honkura N, Matsuzaki M, Noguchi J, Ellis-Davies GC & Kasai H. (2008). The subspine organization of actin fibers regulates the structure and plasticity of dendritic spines. Neuron 57, 719–729. [DOI] [PubMed] [Google Scholar]
  23. Hotulainen P & Hoogenraad CC. (2010). Actin in dendritic spines: connecting dynamics to function. J Cell Biol 189, 619–629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Husi H, Ward MA, Choudhary JS, Blackstock WP & Grant SG. (2000). Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci 3, 661–669. [DOI] [PubMed] [Google Scholar]
  25. Janmey PA. (1998). The cytoskeleton and cell signaling: component localization and mechanical coupling. Physiol Rev 78, 763–781. [DOI] [PubMed] [Google Scholar]
  26. Kevenaar JT & Hoogenraad CC. (2015). The axonal cytoskeleton: from organization to function. Front Mol Neurosci 8, 44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Korobova F & Svitkina T. (2010). Molecular architecture of synaptic actin cytoskeleton in hippocampal neurons reveals a mechanism of dendritic spine morphogenesis. Mol Biol Cell 21, 165–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Krupp JJ, Vissel B, Thomas CG, Heinemann SF & Westbrook GL. (1999). Interactions of calmodulin and alpha-actinin with the NR1 subunit modulate Ca2+-dependent inactivation of NMDA receptors. J Neurosci 19, 1165–1178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kuriu T, Inoue A, Bito H, Sobue K & Okabe S. (2006). Differential control of postsynaptic density scaffolds via actin-dependent and -independent mechanisms. J Neurosci 26, 7693–7706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Legendre P, Rosenmund C & Westbrook GL. (1993). Inactivation of NMDA channels in cultured hippocampal neurons by intracellular calcium. J Neurosci 13, 674–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Levy AD, Xiao X, Shaw JE, Sudarsana Devi SP, Katrancha SM, Bennett AM, Greer CA, Howe JR, Machida K & Koleske AJ. (2018). Noonan Syndrome-Associated SHP2 Dephosphorylates GluN2B to Regulate NMDA Receptor Function. Cell reports 24, 1523–1535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lisman J, Schulman H & Cline H. (2002). The molecular basis of CaMKII function in synaptic and behavioural memory. Nat Rev Neurosci 3, 175–190. [DOI] [PubMed] [Google Scholar]
  33. Matus A, Ackermann M, Pehling G, Byers HR & Fujiwara K. (1982). High actin concentrations in brain dendritic spines and postsynaptic densities. Proc Natl Acad Sci U S A 79, 7590–7594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Morishita W, Connor JH, Xia H, Quinlan EM, Shenolikar S & Malenka RC. (2001). Regulation of synaptic strength by protein phosphatase 1. Neuron 32, 1133–1148. [DOI] [PubMed] [Google Scholar]
  35. Norenberg W, Hofmann F, Illes P, Aktories K & Meyer DK. (1999). Rundown of somatodendritic N-methyl-D-aspartate (NMDA) receptor channels in rat hippocampal neurones: evidence for a role of the small GTPase RhoA. Br J Pharmacol 127, 1060–1063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Offord J & Catterall WA. (1989). Electrical activity, cAMP, and cytosolic calcium regulate mRNA encoding sodium channel alpha subunits in rat muscle cells. Neuron 2, 1447–1452. [DOI] [PubMed] [Google Scholar]
  37. Okamoto K, Nagai T, Miyawaki A & Hayashi Y. (2004). Rapid and persistent modulation of actin dynamics regulates postsynaptic reorganization underlying bidirectional plasticity. Nat Neurosci 7, 1104–1112. [DOI] [PubMed] [Google Scholar]
  38. Paoletti P & Ascher P. (1994). Mechanosensitivity of NMDA receptors in cultured mouse central neurons. Neuron 13, 645–655. [DOI] [PubMed] [Google Scholar]
  39. Paoletti P, Bellone C & Zhou Q. (2013). NMDA receptor subunit diversity: impact on receptor properties, synaptic plasticity and disease. Nat Rev Neurosci 14, 383–400. [DOI] [PubMed] [Google Scholar]
  40. Pollard TD & Cooper JA. (2009). Actin, a central player in cell shape and movement. Science 326, 1208–1212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Prat AG, Xiao YF, Ausiello DA & Cantiello HF. (1995). cAMP-independent regulation of CFTR by the actin cytoskeleton. Am J Physiol 268, C1552–1561. [DOI] [PubMed] [Google Scholar]
  42. Prybylowski K, Chang K, Sans N, Kan L, Vicini S & Wenthold RJ. (2005). The synaptic localization of NR2B-containing NMDA receptors is controlled by interactions with PDZ proteins and AP-2. Neuron 47, 845–857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Racz B & Weinberg RJ. (2004). The subcellular organization of cortactin in hippocampus. J Neurosci 24, 10310–10317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Racz B & Weinberg RJ. (2006). Spatial organization of cofilin in dendritic spines. Neuroscience 138, 447–456. [DOI] [PubMed] [Google Scholar]
  45. Racz B & Weinberg RJ. (2008). Organization of the Arp2/3 complex in hippocampal spines. J Neurosci 28, 5654–5659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rosenmund C, Feltz A & Westbrook GL. (1995). Calcium-dependent inactivation of synaptic NMDA receptors in hippocampal neurons. J Neurophysiol 73, 427–430. [DOI] [PubMed] [Google Scholar]
  47. Rosenmund C & Westbrook GL. (1993). Calcium-induced actin depolymerization reduces NMDA channel activity. Neuron 10, 805–814. [DOI] [PubMed] [Google Scholar]
  48. Rostaing P, Real E, Siksou L, Lechaire JP, Boudier T, Boeckers TM, Gertler F, Gundelfinger ED, Triller A & Marty S. (2006). Analysis of synaptic ultrastructure without fixative using high-pressure freezing and tomography. Eur J Neurosci 24, 3463–3474. [DOI] [PubMed] [Google Scholar]
  49. Rust MB, Gurniak CB, Renner M, Vara H, Morando L, Gorlich A, Sassoe-Pognetto M, Banchaabouchi MA, Giustetto M, Triller A, Choquet D & Witke W. (2010). Learning, AMPA receptor mobility and synaptic plasticity depend on n-cofilin-mediated actin dynamics. EMBO J 29, 1889–1902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sattler R, Xiong Z, Lu WY, MacDonald JF & Tymianski M. (2000). Distinct roles of synaptic and extrasynaptic NMDA receptors in excitotoxicity. J Neurosci 20, 22–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Schwiebert EM, Mills JW & Stanton BA. (1994). Actin-based cytoskeleton regulates a chloride channel and cell volume in a renal cortical collecting duct cell line. J Biol Chem 269, 7081–7089. [PubMed] [Google Scholar]
  52. Seeburg PH, Burnashev N, Kohr G, Kuner T, Sprengel R & Monyer H. (1995). The NMDA receptor channel: molecular design of a coincidence detector. Recent Prog Horm Res 50, 19–34. [DOI] [PubMed] [Google Scholar]
  53. Shumilina EV, Negulyaev YA, Morachevskaya EA, Hinssen H & Khaitlina SY. (2003). Regulation of sodium channel activity by capping of actin filaments. Mol Biol Cell 14, 1709–1716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Sibarov DA & Antonov SM. (2018). Calcium-Dependent Desensitization of NMDA Receptors. Biochemistry (Mosc) 83, 1173–1183. [DOI] [PubMed] [Google Scholar]
  55. Srinivasan Y, Elmer L, Davis J, Bennett V & Angelides K. (1988). Ankyrin and spectrin associate with voltage-dependent sodium channels in brain. Nature 333, 177–180. [DOI] [PubMed] [Google Scholar]
  56. Star EN, Kwiatkowski DJ & Murthy VN. (2002). Rapid turnover of actin in dendritic spines and its regulation by activity. Nat Neurosci 5, 239–246. [DOI] [PubMed] [Google Scholar]
  57. Tardin C, Cognet L, Bats C, Lounis B & Choquet D. (2003). Direct imaging of lateral movements of AMPA receptors inside synapses. EMBO J 22, 4656–4665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Traynelis SF, Wollmuth LP, McBain CJ, Menniti FS, Vance KM, Ogden KK, Hansen KB, Yuan H, Myers SJ & Dingledine R. (2010). Glutamate receptor ion channels: structure, regulation, and function. Pharmacol Rev 62, 405–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Wechsler A & Teichberg VI. (1998). Brain spectrin binding to the NMDA receptor is regulated by phosphorylation, calcium and calmodulin. EMBO J 17, 3931–3939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Wyszynski M, Lin J, Rao A, Nigh E, Beggs AH, Craig AM & Sheng M. (1997). Competitive binding of alpha-actinin and calmodulin to the NMDA receptor. Nature 385, 439–442. [DOI] [PubMed] [Google Scholar]
  61. Xiao X, Levy AD, Rosenberg BJ, Higley MJ & Koleske AJ. (2016). Disruption of Coordinated Presynaptic and Postsynaptic Maturation Underlies the Defects in Hippocampal Synapse Stability and Plasticity in Abl2/Arg-Deficient Mice. J Neurosci 36, 6778–6791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zhang Y, Zhang XF, Fleming MR, Amiri A, El-Hassar L, Surguchev AA, Hyland C, Jenkins DP, Desai R, Brown MR, Gazula VR, Waters MF, Large CH, Horvath TL, Navaratnam D, Vaccarino FM, Forscher P & Kaczmarek LK. (2016). Kv3.3 Channels Bind Hax-1 and Arp2/3 to Assemble a Stable Local Actin Network that Regulates Channel Gating. Cell 165, 434–448. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES