Latent memory facilitates relearning through molecular signaling mechanisms that are distinct from original learning

Highlights

• Latent memory supports savings in multiple temporal domains of memory formation.

• Savings learning can be studied in a reduced behavioral preparation of Aplysia.

• Savings learning mechanisms are distinct from original learning requirements.

Abstract

A highly conserved feature of memory is that it can exist in a latent, non-expressed state which is revealed during subsequent learning by its ability to significantly facilitate (savings) or inhibit (latent inhibition) subsequent memory formation. Despite the ubiquitous nature of latent memory, the mechanistic nature of the latent memory trace and its ability to influence subsequent learning remains unclear. The model organism Aplysia californica provides the unique opportunity to make strong links between behavior and underlying cellular and molecular mechanisms. Using Aplysia, we have studied the mechanisms of savings due to latent memory for a prior, forgotten experience. We previously reported savings in the induction of three distinct temporal domains of memory: short-term (10 min), intermediate-term (2 h) and long-term (24 h). Here we report that savings memory formation utilizes molecular signaling pathways that are distinct from original learning: whereas the induction of both original intermediate- and long-term memory in naïve animals requires mitogen activated protein kinase (MAPK) activation and ongoing protein synthesis, 2 h savings memory is not disrupted by inhibitors of MAPK or protein synthesis, and 24 h savings memory is not dependent on MAPK activation. Collectively, these findings reveal that during forgetting, latent memory for the original experience can facilitate relearning through molecular signaling mechanisms that are distinct from original learning.

Introduction

Memory is typically assessed through the measurement of overt changes in behavior. However, studies of learning and memory have also clearly established that memory can also develop in a latent, non-expressed form following a learning experience and, although inaccessible through tests of recall, is revealed through the ability to influence (promote or inhibit) subsequent learning and memory formation in a wide variety of paradigms (e.g., priming (Antzoulatos et al., 2006, Parsons and Davis, 2012, Philips et al., 2006), savings (Ebbinghaus, 1885/1913), latent inhibition (Tolman & Honzik, 1930)). In many instances, latent memory can outlast the overt expression of memory for an experience to provide a platform for “savings” (a reduction in the number of study trials or time required for relearning; Ebbinghaus, 1885/1913, Nelson, 1971). While well established as a behavioral feature of memory, the mechanistic nature of the latent memory trace and the means by which savings occurs during relearning remains unclear.

Despite its ubiquitous nature, very few studies of latent memory and savings have been initiated in invertebrate model organisms (Antzoulatos et al., 2006, Matzel et al., 1992, Parvez et al., 2005, Philips et al., 2006, Susswein and Schwarz, 1983), wherein strong links can be made between behavior and the underlying cellular and molecular mechanisms. With its large, readily identifiable neurons and simple nervous system, the invertebrate mollusk Aplysia californica has proven to be an advantageous model for elucidating the cellular and molecular mechanisms of memory formation. Moreover, learning in Aplysia recapitulates critical features of the human savings phenomenon (Antzoulatos et al., 2006, Philips et al., 2006, Susswein and Schwarz, 1983). We previously showed that a latent memory outlasts the initial forgetting of long-term memory (LTM) for sensitization of the tail-elicited siphon withdrawal reflex (T-SWR) for at least two days, and supports the facilitated induction of three distinct temporal phases of memory during relearning: short-term (10 min), intermediate-term (2 h) and long-term savings memories (24 h) (Philips et al., 2006). Importantly, savings was not observed when retraining was delayed for four days after initial signs of forgetting, identifying a strong parallel to the human learning phenomenon, in which the benefit of prior experience is time-limited (Ebbinghaus, 1885/1913). The demonstration of savings in Aplysia has established a unique opportunity to study the cellular and molecular features of the latent memory trace and its facilitation of subsequent memory formation in simple neural circuits.

In the present study, we examined the molecular features of savings in Aplysia. First, we replicated our earlier observation of savings for the induction of sensitization memories within the T-SWR (Philips et al., 2006). We then showed that the facilitated induction of 2 h and 24 h memory requires the interaction between two spaced training trials and is supported by plasticity in the previously described narrow temporal window for two-trial LTM formation in naïve animals (Philips, Tzvetkova, & Carew, 2007). Savings could also be observed when retraining reduced behaving preparations of previously trained Aplysia. In molecular studies, we found that the induction of 2 h savings memory was unique from the induction requirements for a comparable phase of ITM formation in naïve animals, in that (i) it no longer required activation of the highly conserved mitogen activated protein kinase (MAPK) signaling pathway and (ii) it was not disrupted by the inhibition of ongoing protein synthesis. We additionally showed that 24 h savings memory was distinct from the LTM induced in naïve animals, in that its induction was independent of MAPK activation, but still required protein synthesis. Collectively, these findings indicate that latent memory for prior sensitization facilitates relearning across multiple temporal phases of memory by engaging distinct molecular rules for subsequent memory formation.