Memory  >   Memory Consolidation: Filopodia Guide Axonal Branching

A large body of evidence suggests that memory consolidation involves synaptogenesis within potentiated dendritic branches (Chapter 26). An example is the synaptic connection between a hippocampal CA1 neuron and a neuron in the medial prefrontal cortex (mPFC). Chapter 27 further shows that CRMP2 could play a key role in memory consolidation by directing neurite growth, such as the axonal branching from a CA1 neuron to an mPFC dendritic branch that has been potentiated by learning. A question arises: how does CRMP2 know which dendritic branches have been potentiated? The answer may lie in dendritic filopodia which are the precursors of dendritic spines (Ozcan, 2017).


Figure 28-1. A filopodium results from nucleation of actin filaments, which may be induced by Ca2+. [Adapted from: Leondaritis and Eickholt, 2015]

A filopodium is a thin, finger-like and highly motile protrusion from the dendritic shaft, resulting from the nucleation of actin filaments (F-actin) (Figure 28-1). Actin nucleation is controlled by actin nucleators which, in turn, are regulated by specific signals. Ca2+ is one of the signals capable of initiating the generation of filopodia (Leondaritis and Eickholt, 2015). Since NMDA receptors (NMDARs) are permeable to Ca2+, they could play a role in the generation of filopodia. Indeed, blocking NMDARs has been shown to reduce the density of filopodia in dendritic shafts (Portera-Cailliau et al., 2003).

Learning-induced memory formation is known to arise from the opening of NMDARs, which are located in both synaptic and extrasynaptic membrane. Synchronous opening of NMDARs within the same memory unit may produce NMDA plateaus, leading to accumulation of Ca2+ ions in a dendritic branch that constitutes a memory unit. This should promote the generation of filopodia in the dendritic shaft of the memory unit. Some of them may mature into spines by repeated stimulation (Figure 28-2).


Figure 28-2. Typical stages in the growth of new dendritic spines. Depending on the persistence of neural activity, the growth process may result in a new synaptic connection or failure to form one. [Adapted from: Ozcan, 2017]

The Role of Ephs and Ephrins in Synaptogenesis

Filopodia were suggested to play an exploratory role in searching axonal partners nearly two decades ago (Jontes and Smith, 2000). This notion is now supported by mounting evidence. The molecules that guide the axon terminal toward filopodia turn out to be Ephs and ephrins (Hruska and Dalva, 2012; Dines and Lamprecht, 2016). Ephs (EphA and EphB) are transmembrane tyrosine kinase receptors and ephrins (ephrin-A and ephrin-B) are their ligands. Each group has several members. The ephrin-A group consists of five members, designated as ephrin-A1 to ephrin-A5. It has been shown that the interaction between ephrin-A5 and EphA5 plays a necessary and activity-independent role in the initiation of the early phase of synaptogenesis (Akaneya et al., 2010).

The growth cone, located at the tip of an axon terminal, may advance, pause, retraction, or turn. These types of movement are regulated by axon guidance molecules such as semaphorins and ephrins. Binding of CRMP2 to tubulin promotes microtubule polymerization and axon growth, but the signaling cascades triggered by semaphorins and ephrins may alter the phosphorylation state of CRMP2, resulting in growth cone collapse (Arimura et al., 2005). Recently, it has been shown that the microtubule motor proteins, kinesin and dynein, drive the turn of growth cones (Kahn and Baas, 2016). CRMP2 could also be involved, as it can interact with both kinesin (Kimura et al., 2005) and dynein (Arimura et al., 2009). Strikingly, ephrin-A5 has also been shown to trigger the signaling cascades, leading to growth cone collapse via phosphorylation of both CRMP2 and myosin light chain (MLC) (Arimura et al., 2005).

Filopodia are intrinsically highly motile on the dendritic shaft to increase the probability of contact with axonal partners. They may contain EphA5. Once EphA5 binds ephrin-A5 on the distal axon, the interaction could immobilize filopodia and stop axon growth, thereby establishing the prototype of synapses.


Author: Frank Lee
First published: April, 2018
Last updated: July, 2018