Disruption of fear memory consolidation and reconsolidation by actin filament arrest in the basolateral amygdala

Abstract

The dynamic re-arrangement of actin filaments is an essential process in the plasticity of synaptic connections during memory formation. In this study, we determined in mice effects of actin filament arrest in the basolateral complex of the amygdala (BLA) at different time points after memory acquisition and re-activation, using the fungal cytotoxin phalloidin. Our data show a selective disruption of auditory cued but not contextual fear memory, when phalloidin was injected 6 h after conditioning. In contrast, no effect was observed when phalloidin was applied after 24 h, ruling out an interference with the retrieval or expression of conditioned fear. A comparable result was obtained after memory re-activation, hence suggesting similar actin-dependent mechanisms to be active during consolidation and reconsolidation of auditory fear memory. Biochemical analysis showed that phalloidin-mediated filament arrest leads to a transient increase of highly cross-linked actin filaments in the BLA, evident 2 h after injection. Together, these observations indicate that dynamic re-arrangements of actin filaments in the BLA during a late phase of fear memory consolidation and reconsolidation are critical for fear memory storage.

Introduction

Dynamic changes of actin filaments play a critical role in structural plasticity that is believed to underlie information storage in neural networks (Lamprecht and LeDoux, 2004, Matus, 2000). Presumably through trafficking of neurotransmitter receptors and cytoskeletal restructuring of pre- and post-synaptic sites, actin filaments help recently acquired fear memories to be transformed into a more lasting state, a process called memory consolidation. Indeed, changes of synaptic actin filaments and their effect on post-synaptic spines have been demonstrated in cultivated neuronal cells upon induction of synaptic plasticity (Honkura, Matsuzaki, Noguchi, Ellis-Davies, & Kasai, 2008). These match with observations made in vivo of post-synaptic actin filament assembly (Fukazawa et al., 2003) and long recognized morphological changes at the synapse (e.g., Fifkova & Van Harreveld, 1977). In fact, interfering with actin filament dynamics disturbs long-term potentiation of neuronal activity in the hippocampus (Kim and Lisman, 1999, Krucker et al., 2000) and local injections of the actin depolymerizing toxins cytochalasin D or lantrunculin A into the amygdala or hippocampus disturb the formation of auditory cued and contextual fear memory (Mantzur, Joels, & Lamprecht, 2009).

Classical fear conditioning is very well suited for the investigation of such memory consolidation processes. Subjects very quickly learn to associate a previously neutral stimulus (conditioned stimulus, CS, such as a tone) or context with a coinciding aversive stimulus (unconditioned stimulus, US, such as electric foot shock). After a single training session, animals form a robust fear memory and to subsequent exposures to the CS respond with species-specific defensive behaviors, such as freezing (Blanchard & Blanchard, 1969). It is now well established that the basolateral complex of the amygdala (BLA) provides a core structure and site of neural plasticity in such classical fear conditioning. The BLA, comprising the lateral and basal amygdala subnuclei, serves as a convergence site for uni- and multimodal sensory signals with ascending nociceptive pathways and affective information processed in limbic brain circuits (Maren, 2003, Maren and Quirk, 2004). Importantly, neural plasticity has been observed in this region following behavioral training as well as after electrical stimulation of thalamic, hippocampal or cortical afferences (Rogan et al., 1997, Sah et al., 2008). Molecular and cellular processes that are critical for the acquisition and consolidation of fear memories have been identified and ample evidence has accumulated for an involvement of actin cytoskeleton dynamics (Diana et al., 2007, Lamprecht et al., 2002, Lamprecht, Farb, et al., 2006, Mantzur et al., 2009, Stork et al., 2004). Fear conditioning has also been used to investigate the phenomenon of memory reconsolidation, i.e., the modification and renewed storage of prior memories upon their retrieval. It was shown that although consolidation and reconsolidation share principle characteristics, neurochemical and molecular details are only partially overlapping (Alberini, 2005, Tronson and Taylor, 2007).

In the current study we tested how an arrest of actin filaments in the mouse BLA through bilateral application of the death cap (amanita phalloides) toxin phalloidin affects the consolidation and reconsolidation of Pavlovian fear memory. We focused on the proposed role of actin filaments in the morphological changes in synaptic structures, which typically become evident ca. 0.5 h after stimulation and are thought to provide a long-lasting memory correlate (Bonhoeffer and Yuste, 2002, Korkotian and Segal, 2001). To begin to dissect these processes, phalloidin was applied to the BLA in different animals at time points either 0.5 h, 6 h and 24 h after conditioning or retrieval. Our results not only provide evidence for a similar role of amygdalar actin filament dynamics in consolidation and reconsolidation of auditory fear memory, but also suggest that a temporally distinct actin-dependent cellular process in the BLA is involved in responses to a familiarized background context upon fear memory re-activation.