Molecular Signatures and Mechanisms of Long-Lasting Memory Consolidation and Storage ⇑ Cynthia Katche A, Martín Cammarota B, Jorge H
Total Page:16
File Type:pdf, Size:1020Kb
Neurobiology of Learning and Memory 106 (2013) 40–47 Contents lists available at SciVerse ScienceDirect Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme Invited Review Molecular signatures and mechanisms of long-lasting memory consolidation and storage ⇑ Cynthia Katche a, Martín Cammarota b, Jorge H. Medina a,c, a Instituto de Biología Celular y Neurociencias, Facultad de Medicina, UBA, Paraguay 2155, 3° piso, Buenos Aires 1121ABG, Argentina b Instituto do Cérebro, Universidade Federal do Rio Grande do Norte, Av. Nascimento de Castro 2155, Natal RN59056-450, Brazil c Departamento de Fisiologia, Facultad de Medicina, UBA, Paraguay 2155, 7° piso, Buenos Aires 1121ABG, Argentina article info abstract Article history: A body of evidence emerged in the last decade regarding late posttraining memory processing. Most of Received 29 January 2013 this new information comes from aversively motivated learning tasks that mainly depend on hippocam- Revised 25 June 2013 pus, amygdala and insular cortex, and points to the involvement of long-lasting changes in gene expres- Accepted 26 June 2013 sion and protein synthesis in late stages of memory consolidation and storage. Here, we describe recent Available online 3 July 2013 advances in this field and discuss how recurrent rounds of macromolecular synthesis and its regulation might impact long-term memory storage. Keywords: Ó 2013 Elsevier Inc. All rights reserved. Memory persistence Hippocampus Long-lasting modifications Gene expression 1. Introduction defined as the transition of memory from protein synthesis and gene expression-dependence to independence in specific brain re- Long-term memory (LTM) is conventionally defined as that last- gions involved in acquisition of a particular learning experience ing more than several hours and there is a vast amount of informa- (Medina, Bekinschtein, Cammarota, & Izquierdo, 2008). tion regarding the participation of many biochemical pathways The second phase is slower, and entails the participation of neo- and brain circuits in LTM formation. In 1900, Muller and Pilzecker cortical regions and their interactions with medial temporal lobe (1900) proposed that formation of permanent memory takes time structures reorganizing the recently learned material (Squire, and that during this time, memory remains vulnerable to disrup- 1992) by gradually binding together the multiple cortical regions tion. The process of developing stable memory is referred to as that store memory for a whole event. This phase is called sys- ‘‘consolidation’’ (McGaugh, 1966, 2000). A dominant idea that tems-level consolidation and lasts several days to weeks or months emerged two decades ago is that memory consolidation comprises in most learning tasks studied so far (Frankland & Bontempi, 2005). two stages or phases: the initial stage, called cellular or synaptic Systems consolidation is usually referred to as the process by consolidation, is fast, lasting from several hours to a couple of days. which memory becomes independent of the hippocampus. This process is thought to take place at synapses of the neuronal The standard model of memory consolidation posits that the circuits encoding the experience-dependent internal representa- hippocampus is primarily involved in consolidating and recalling tion such as the hippocampus and related non-cortical structures recent episodic-like memories while some cortical regions, (Dudai, 2002). Cellular consolidation involves the activation of including prelimbic, orbitofrontal, and anterior cingulate areas, transcription factors to modulate gene expression, as well as syn- are mostly implicated in remote memory processing (Frankland, thesis, posttranslational modification and reorganization of pro- Bontempi, Talton, Kaczmarek, & Silva, 2004; Lesburgueres et al., teins at synapses and soma, which finally ends in synaptic 2011; Maviel, Durkin, Menzaghi, & Bontempi, 2004; Shan, Chan, remodeling that makes the memory trace stable and precise (Albe- & Storm, 2008). However, strong alternative hypotheses emerged rini, 2009; Lamprecht & LeDoux, 2004; Morris, 2006; Ruediger in the last few years stating that some cortical regions outside et al., 2011). In other words, cellular consolidation has been the temporal lobe have indeed a role in the very first moments of memory formation and participate in recalling both recent and remote memories (Einarsson & Nader, 2012; Katche et al., 2013; ⇑ Corresponding author. Address: Instituto de Biología Celular y Neurociencias, Leon, Bruno, Allard, Nader, & Cuello, 2010; Tse et al., 2011). Facultad de Medicina, Paraguay 2155, 3rd floor, Buenos Aires 1121ABG, Argentina. Fax: +54 11 59509626. Although much is known about LTM consolidation, what puts E-mail address: [email protected] (J.H. Medina). the ‘‘long’’ in LTM is its persistence over time. Most of the acquired 1074-7427/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nlm.2013.06.018 C. Katche et al. / Neurobiology of Learning and Memory 106 (2013) 40–47 41 information is bound to disappear or may leave an undetectable knockout mice in which NMDAr are temporarily switch off during trace. In addition, and although it is not the subject matter of this the storage state showed impairments in memory persistence. review, it is essential to point here that recent findings strongly Using a similar technology, Wang et al. (2003) demonstrated suggest that behavioral and neurohumoral conditions at the mo- that reactivation of alpha-CaMKII activity in the forebrain during ment of memory retrieval could play a key role in modifying al- the first, but not the second or third week after training mice in ready consolidated memories or, conversely, helping to strength CFC is crucial for remote memory consolidation. In addition, CaM- the reactivated memory trace or keep it stable for longer periods KII heterozygeous knockout mice showed neocortical alterations in of time through a protein synthesis-dependent reconsolidation plasticity and normal retention fear memory when tested 1 day process. Moreover, quite recently it has been shown that, as hap- posttraining but impaired fear memory 10–50 days after training pens during memory consolidation (see below), administration of (Frankland, O’Brien, Ohno, Kirkwood, & Silva, 2001). Together, protein synthesis inhibitors late after reactivation hampers the these findings suggest that persistent activity of NMDAr and al- persistence of fear memory (Nakayama et al., 2013). Therefore, pha-CaMKII for the first several days after training is important the matter of memory persistence is central in understanding the for establishing remote memories and that NMDAr also have a role neurobiology of learning and memory. in maintaining long-lasting LTM. The main goal of this review article is to describe what has been ERKs are protein kinases involved in memory consolidation published about long-lasting molecular changes in memory pro- (Atkins, Selcher, Petraitis, Trzaskos, & Sweatt, 1998). Some reports cessing, and specifically this article deals with those molecular showed two waves of ERKs activation after training. The first one is mechanisms involved in long-lasting LTM. Therefore, we will focus rapid and transient; the second is delayed and persistent, lasting on cellular consolidation mechanisms in hippocampus and related for at least 24 h in the CA3 region (Trifilieff, Calandreau, Herry, brain regions (including some cortical areas) involved in late post- Mons, & Micheau, 2007; Trifilieff et al., 2006). These modifications training consolidation periods that lead to persistent memories. coincide with changes in CREB activation (see later). It has been Several of long-lasting molecular changes shown below are good also shown that ERK2 expression increases 24 h in the dorsal hip- examples of molecular changes that seem to last well beyond pocampus after IA training and that ERK1/2 participates late after times suggested for the formation of memory. Unraveling the func- training to sustain memory storage of two different hippocam- tional role of these molecular signatures is, in most cases, in pro- pus-dependent learning tasks (Bekinschtein et al., 2008; Eckel-Ma- gress and permits us to envision new ways of thinking memory han et al., 2008). Interestingly, a circadian oscillation of the consolidation. phosphorylation state of ERK1/2 in the hippocampus has been implicated in memory persistence (Eckel-Mahan, 2012). Recently, it has been shown that inducible and targeted deletion of ERK5 re- 2. Long-lasting changes in glutamate receptors, protein kinases, sulted in attenuation of adult neurogenesis in the dentate gyrus protein synthesis, and gene expression as molecular signatures and specific impairment in remote IA memory (Pan, Chan, Kuo, of long-lasting LTM Storm, & Xia, 2012). Another example of signaling molecules involved in memory 2.1. Glutamate receptors and protein kinases persistence is represented by neuronal adenylyl ciclases. Knock- ing-down adenylyl ciclase 1 (AC1) and 8 (AC8) in mice down-reg- The first study that demonstrated a long-lasting process in ulates the expression of several genes in the hippocampus. the hippocampus required for LTM consolidation is that of Rie- Interestingly, most of these genes are up-regulated in wild-type del et al. (1999). They demonstrated that ongoing activity in the mice 48 h after behavioral training (Wieczorek, Maas, Muglia, Vogt, hippocampus is required for spatial memory consolidation dur- & Muglia, 2010). These data