under Neurophysiology of Learning and Memory

Figure 1. A simplified illustration of the constituents of a synapse. (http://upload.wikimedia.org/wikipedia/commons/3/30/Chemical_synapse_schema_cropped.jpg)

Discovering the mechanism underlying learning in the brain is a fundamental aspect of understanding how the brain works. Learning involves neuroplasticity, the ability of the nervous system to change structurally and functionally. This plasticity is fundamental for behaviours such as addiction (i.e. alcohol addiction, sex addiction, food addiction) (Hyman, 2006) and social interactions (Yang, 2012) while impairment to plasticity from factors like neurodegenerative disorders (i.e. Huntington's Disease) (Zeef, 2012) disrupts cognitive abilities like learning and memory (Lacor, 2007; Ondrejcak, 2010; Zeef, 2012). An important component of neuroplasticity involves the strengthening or weakening of the synapses.

Synapses, which can be considered the basic functional unit of the nervous system, can be altered through activity-dependent processes such as long-term potentiation (LTP), which causes long-lasting strengthening of synaptic response, and long-term depression (LTD), which causes long-lasting weakening of synaptic response (Bi, 1998). A significant amount of experimental results came from post-synaptic research. Specifically, post-synaptic receptor activation (Bi, 1998), calcium driven mechanisms (Ghosh, 1995), dendritic spine properties, and cytoskeletal changes (Luscher, 2000; Hussain, 2001; Irie, 2002; Star, 2002; Carlisle, 2005; Rex, 2007; Bennett, 2011) have been shown to be related post-synaptic plasticity and learning.

Table of Contents

1 Activation

Two kinds of excitatory receptors involved in plasticity that have undergone extensive research are N-methy-D-aspartate receptors (NMDAR) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) (Mayer, 2004). AMPAR requires glutamate binding to activate, allowing Na+flow into the neuron and causing an excitatory postsynaptic potential. NMDAR requires both ligand and voltage activation, where depolarization of the cell removes an extracellular magnesium ion block while glutamate binds to the receptor, inducing conformational change, allowing both Na+ and Ca2+ to enter (Ghosh, 1995).

NMDA from Willis Bilderback on Vimeo.

Video 1: NMDAR Activation (Animation by Pete Bilderback).
NMDAR is both ligand and voltage gated. The voltage change
in the video is shown by the change in membrane colour.

While both receptors lead to depolarization, NMDAR also brings about intracellular calcium ion concentration changes. Cytosolic calcium ion concentrations are very important in calcium dependent second messengers and pathways, which are involved in plasticity.

2 Calcium Dependent Pathways

Increased cytosolic Ca2+, from extracellular sources or internal calcium stores, can lead to several second messenger related cellular responses. Pathways leading to increased calcium include NMDAR activation, which affects concentrations both directly and indirectly through depolarization and opening of voltage-gated Ca2+ channels and cytosolic-calcium-mediated calcium stores in the endoplasmic reticulum (Ghosh, 1995). Calcium stores may also be opened by the binding of inositol triphosphate (IP3), which in turn can be generated from GTP-linked receptors (Kaplin, 1994).
Ca2+ binds to the protein calmodulin inside the cell. The complex, calcium-calmodulin (CaM), then can bind to a variety of other cellular components such as CaM-dependent protein kinases (CaMK) and adenylate cyclases, altering their activity (Lisman, 2002).

2.a CaM Kinases
CaMK is an important enzyme for phosphorylating and activating various signaling pathways. CaMKII and CaMKIV in particular have been shown to be localized in the nucleus and activate wide-acting proteins such as MAPK and cAMP response element-binding protein (CREB) (Bozon, 2003). These proteins in turn increase various plasticity related gene expressions, such as BDNF (Sun, 1994; Tao, 1998; Bramham, 2005; Rex, 2007), zif268 (Bozon, 2003), and PKMzeta (Hernandez, 2003).
Moreover, CaMK has been shown to be able to autophosphorylate, remaining active even when intracellular Ca2+levels have returned to normal (Ghosh, 1995). This powerful mechanism makes it a very important protein in postsynaptic plasticity.

2.b Adenylyl Cyclase
Another target of CaM is CaM-dependent adenylate cyclases. Adenylate cyclases can influence cAMP levels, which in turn has been shown to be related to memory in animal models, such as rodents, Drosophila and Aplysia (Ghosh, 1995).

2.c AMPA Receptor
NMDAR activation have been shown to induce LTP and LTD through the exocytosis and endocytosis of AMPAR. With the activation of CaMKII and calcium-dependent Ras-ERK pathway, AMPAR carrying vesicles fuse with the membrane during LTP (Patterson, 2010; Lisman, 2012). However, the localization of exocytosed AMPAR to the spine head requires the likely CaMKII-dependent phosphorylation of stargazin, a member of the TARP family (Tomita, 2005). Stargazin interacts with scaffold protein PSD-95 to help trap and localize AMPAR at postsynaptic sites (Bats, 2007).
Membrane-bound AMPAR become endocytosed during LTD (Malinow, 2002; Bredt, 2003), where LTP is characterized by an increase in synaptic strength from a high number of NMDAR activation due to high frequency stimulation and LTD is characterized by a weakening of synaptic strength due to low frequency stimulation.

It makes sense that an increase of AMPAR density at the synapse is associated with LTP as more receptors can contribute to the excitatory postsynaptic potential on subsequent stimulations. Likewise, a decrease in AMPAR density would mean a weaker postsynaptic potential.

In addition, GluR1, a subunit of AMPAR, can be phosphorylated by PKC or CaMKII at specific sites (i.e. S831), which have been observed to increase conductance of AMPAR through conformational changes (Kristensen, 2011). However, increase in conductance due to GluR1 phosphorylation requires stargazin (Kristensen, 2011; Lisman 2012).


Figure 2. CaMKII Dependent Modifications of Postsynaptic AMPAR properties in LTP
(Lisman, 2012)

Video 2: Insertion of AMPAR with Fluorescent Microscopy (Yudowski, 2007).

3 Cytoskeletal and Spine Changes

Plasticity has been associated with morphological changes. Cytoskeletal changes are known to occur, such as growth of new and existing dendrites and splitting of existing synapses (Malenka, 2004) and they provide better, more effective synaptic connections. The high concentrations of actin in the spines are subject to polymerization and depolymerization through various pathways, leading to a highly dynamic cytoskeletal structure (Carlisle, 2005) that affects the synaptic connections of the postsynaptic neuron.

3.a Filopodial Protrusion
Filopodial protrusions are important during development as they extend and retract based on exposure to neurotransmitters to later form dendritic spines as protruding filopodia come into contact with growth cones (Carlisle, 2005). Studies of blockage of NMDAR seem to suggest filopodial growth depends on cytosolic Ca2+ (Luscher, 2000). LTP-inducing stimuli have also shown rapid filopodia extensions in mature neurons (Carlisle, 2005).

3.b Actin-Related Processes
Fluorescence resonance energy transfer(FRET) based techniques have shown high frequency titanic stimulation as typically used for LTP leads to polymerization of actin filaments in spines whereas low frequency stimulations as typically used for LTD leads to actin depolymerization (Carlisle, 2005). This supports the idea of creating new or altering existing spines.

3.b.a Rac - PAK - LIM Kinase Pathway
Recent hippocampal studies have suggested that actin dynamics in spines involves the Rac-PAK-LIMK-Cofilin pathway. There are multiple messengers that result in this same pathway; two well-studied pathways will be presented.
One pathway starts by activation of EphB receptors located on spines by physical contact with axonal membrane-bound ephrins. EphB then activatves Rho GEF kalirin, which in turn activates Rac. Rac is a part of the Rho family GTPase, known for their regulation of actin cytoskeleton in spines. Specifically, Rac activates p21-activated kinases (PAK), which activates LIM kinase 1 (LIMK1). LIMK1 deactivates, through phosphorylation, cofilin, which reduces depolymerization of actin filaments. As actin filaments are constantly undergoing both polymerization and depolymerization, reducing depolymerization enhances the polymerization effect, leading to rapid growth of the dendritic spine (Carlisle, 2005).

Another pathway that uses the Rac-PAK-LIMK-Cofilin pathway involves NMDAR. As NMDAR is activated, cytosolic Ca2+ is increased, CaM is generated, leading to activation of β CaMK Kinase (CaMKKβ) and CaMKI. CaMKKβ and CaMKI in turn phosphorylates Ser516 in βPIX, which interacts with Rac, turning on the Rac-PAK-LIMK pathway as described above (Saneyoshi, 2008).

3.b.b Intersectin-cdc42-N-Wasp-ARP2/3 pathway
EphB receptors have also been shown to interact and accumulate intersectin-1 in the spine. The accumulated intersectin recruits N-Wasp, which enhances intersectin’s GEF activity. This allows intersectin to activate Cdc42, which causes actin polymerization through N-WASP and ARP2/3 (Hussain, 2001; Irie, 2003).

3.b.c Actin regulation via NMDAR-dependent proteins (Gelsolin and Profilin)
NMDAR activation causes an influx of Ca2+, which activates gelsolin. Gelsolin is an important calcium-dependent protein involved in binding and capping actin at dendritic sites (Hussain, 2001). It is thus an important regulator of cytoskeletal structure involved in plasticity.

NMDAR is also involved in the regulation of Rho A. NMDAR activation leads to activation of ROCK, the specific kinase of Rho A, which leads to the release of profilin at the targeted spine (Star, 2002). Profilin catalyzes ADP-ATP exchange on ADP-actin monomers, which makes the monomers available for actin polymerization (Bennett, 2011). It is thus also an important regulator of cytoskeletal structure involved in plasticity.

Figure 3. EphB and NMDA receptor dependent actin-modification pathways in dendritic spine.

Disruption by ADDLs


Lacor's paper (2007) showed that solube Abeta-derived oligomers (ADDLs), suspected of being a key component in the neurodegenerative disorder known as Alzheimer’s disease, damage synapses. By using the separation process of fractionation, they showed that ADDLs bind to postsynaptic complexes with NMDAR. However, they did not seem to attach to GABAergic neurons under the same technique, showing specificity towards glutaminergic pyramidal cells.

By treating mature hippocampal neurons with ADDLs, longer exposure led to a decreased density of immunolabeled NR1 and NR2B, subunits of NMDAR, and EphB receptors, labeled with N-terminal antibodies.

Long exposure of ADDLs to hippocampal neurons seemed to result in decreased spine density, abnormal spine morphology such as thin filopodia-like and abnormally long (>2μm) spines and the decrease of cytoskeletal proteins like drebrin. These results strongly suggest ADDLs role in hindering post-synaptic plasticity and supports the idea that ADDLs play a role in Alzheimer's disease, causing decreased cognitive ability.


Bats, C. (2007). The Interaction between Stargazin and PSD-95 Regulates AMPA Receptor Surface Trafficking. Neuron (Cambridge, Mass.), 53(5), 719-734.

Bennett, M. (2011). A Model of NMDA Receptor Control of F-actin Treadmilling in Synaptic Spines and Their Growth. Bulletin of mathematical biology, 73(9), 2109-2131.

Bi, G. Synaptic modifications in cultured hippocampal neurons: Dependence on spike timing, synaptic strength, and postsynaptic cell type. The Journal of neuroscience, 18(24), 10464-10472 (1998).

Bozon, B. (2003). MAPK, CREB and zif268 are all required for the consolidation of recognition memory. Philosophical transactions. Biological sciences, 358(1432), 805-814.

Bramham, C. (2005). BDNF function in adult synaptic plasticity: The synaptic consolidation hypothesis. Progress in neurobiology, 76(2), 99-125.

Bramham, C. (2007). Dendritic mRNA: transport, translation and function. Nature reviews. Neuroscience, 8(10), 776-789.

Bredt, D. (2003). AMPA Receptor Trafficking at Excitatory Synapses. Neuron (Cambridge, Mass.), 40(2), 361-379.

Carlisle, H. (2005). Spine architecture and synaptic plasticity. Trends in neurosciences (Regular ed.), 28(4), 182-187.

Ghosh, A. (1995). Calcium signaling in neurons: molecular mechanisms and cellular consequences. Science (New York, N.Y.), 268(5208), 239-247.

Harvey,C.D. (2008). The spread of Ras activity triggered by activation of a single dendritic spine. Science 321:136140.

Hernandez, A. (2003). Protein Kinase M Synthesis from a Brain mRNA Encoding an Independent Protein Kinase C Catalytic Domain: Implications for the Molecular Mechanism of Memory. The Journal of biological chemistry, 278(41), 40305-40316.

Hussain, N. (2001). Endocytic protein intersectin-I regulates actin assembly via Cdc42 and N-WASP. Nature cell biology, 3(10), 927-932.

Hyman, S. (2006). Neural Mechanisms of Addiction: The Role of Reward-Related Learning and Memory. Annual review of neuroscience,, 565-598.

Irie, F. (2002). EphB receptors regulate dendritic spine development via intersectin, Cdc42 and N-WASP. Nature neuroscience, 5(11), 1117-1118.

Kaplin, A. (1994). Purified Reconstituted Inositol 1,4,5-Trisphosphate Receptors - Thiol Reagents Act Directly on Receptor Protein. The Journal of Biological Chemistry, 269(46), 28972-28978.

Kristensen, A. (2011).Mechanism of Ca2 /calmodulin-dependent kinase II regulation of AMPA receptor gating. Nature neuroscience, 14(6), 727-U327.

Lacor, P. (2007). A Oligomer-Induced Aberrations in Synapse Composition, Shape, and Density Provide a Molecular Basis for Loss of Connectivity in Alzheimer's Disease. The Journal of neuroscience, 27(4), 796-807.

Lisman, J. (2002). The Molecular Basis of Camkii Function in Synaptic and Behavioural Memory. Nature reviews. Neuroscience, 3(3), 175-190.

Lisman, J. (2012). Mechanisms of CaMKII action in long-term potentiation. Nature reviews. Neuroscience,13(3), 169-182.

Lüscher, C. (1999). Role of AMPA Receptor Cycling in Synaptic Transmission and Plasticity. Neuron (Cambridge, Mass.), 24(3), 649-658.

Lüscher, C. (2000). Synaptic plasticity and dynamic modulation of the portsynaptic membrane. Nature neuroscience, 3(6), 545-550

Malenka, R. (2004). LTP and LTD. Neuron (Cambridge, Mass.), 44(1), 5-21.

Malinow, R. (2002). AMPA R T S P. Annual review of neuroscience, 25(1), 103-126.

Mayer M. (2004). Structure and Function of Glutamate Receptor Ion Channels. Annual review of physiology,66:161-181.

Ondrejcak, T. (2010). Alzheimer’s Disease Amyloid β-Protein and Synaptic Function. Neuromolecular medicine, 12(1), 13-26.

Patterson, M. (2010). AMPA receptors are exocytosed in stimulated spines and adjacent dendrites in a Ras-ERK-dependent manner during long-term potentiation. PNAS : Proceedings of the National Academy of Sciences, 107(36), 15951-15956.

Rex, C. (2007). Brain-Derived Neurotrophic Factor Promotes Long-Term Potentiation-Related Cytoskeletal Changes in Adult Hippocampus. The Journal of neuroscience, 27(11), 3017-3029.

Saneyoshi, T. (2008). Activity-Dependent Synaptogenesis: Regulation by a CaM-Kinase Kinase/CaM-Kinase I/βPIX Signaling Complex. Neuron (Cambridge, Mass.), 57(1), 94-107.

Shi, S. (1999). Rapid Spine Delivery and Redistribution of AMPA Receptors After Synaptic NMDA Receptor Activation. Science (New York, N.Y.), 284(5421), 1811-1816.

Star, E. (2002). Rapid turnover of actin in dendritic spines and its regulation by activity. Nature neuroscience, 5(3), 239-246.

Sun, P. (1994). Differential activation of CREB by Ca2 /calmodulin-dependent protein kinases type II and type IV involves phosphorylation of a site that negatively regulates activity. Genes & development, 8(21), 2527-2539.

Tao, X. (1998). Ca2 Influx Regulates BDNF Transcription by a CREB Family Transcription Factor-Dependent Mechanism. Neuron (Cambridge, Mass.), 20(4), 709-726.

Tomita, S. (2005). Bidirectional synaptic plasticity regulated by phosphorylation of stargazin-like TARPs. Neuron, 45, 269–277.

Yang, Y. (2012). EPAC Null Mutation Impairs Learning and Social Interactions via Aberrant Regulation of miR-124 and Zif268 Translation. Neuron (Cambridge, Mass.), 73(4), 774-788.

Yudowski, G. (2007). Real-Time Imaging of Discrete Exocytic Events Mediating Surface Delivery of AMPA Receptors. The Journal of neuroscience, 27(41), 11112-11121.

Zeef, D. (2012). Memory deficits in the transgenic rat model of Huntington's disease. Behavioural brain research, 227(1), 194-198.