ادغام مجدد و خیانت حافظه در حلزون

Lymnaea stagnalis were operantly conditioned to not perform aerial respiratory behaviour in a specific context (i.e. context-1). The memory for this learned response was reactivated 3 days later in context-1. During the 1 h reconsolidation period following memory reactivation, randomly picked snails were either maintained in context-1 or exposed to a new context (i.e. context-2). One hour later in the post-reconsolidation period, snails in context-1 were placed for 1 h in context-2 and vice-versa. In neither the hypoxic reconsolidation nor the post reconsolidation periods did snails receive a reinforcing stimulus when they opened their pneumostome. All snails were blindly tested for memory 24 h later period in context-2. Only those snails that had been exposed to context-2 during the reconsolidation period exhibited ‘memory’ for context-2. That is, memory infidelity was observed. Snails exposed to context-2 in only the post-reconsolidation period did not show memory for context-2. The immediate cooling of snails after their exposure to the new context in the reconsolidation period blocked the formation the implanted memory. Snails trained in context-1 and exposed to context-2 in the consolidation period only, also did not have memory for context-2. However, the memory for context-1 could still be recalled following successful implantation of the ‘new’ memory. All data presented here are consistent with the notion that during the reconsolidation process memory can be updated.

Hebb (1949) and later Lewis (1979) noted that memory (i.e. the ability to encode and retrieve information) was a dynamic brain process. As Lewis (1979) pointed out memory existed in one of two states: (1) A labile (i.e. active), and (2) A stable (i.e. inactive) one. In the active state memory could be modified or even lost, while in the stable-inactive state memory was held relatively inviolate. It is widely held that the active, more easily modifiable memory is made into a stable permanent memory via a process requiring altered gene activity and new protein synthesis. This is referred to as the consolidation process ( Dudai, 2004, Dudai, 2006, McGaugh, 2000 and Squire, 1987). Recently there has been renewed interest in what happens after a formed stable memory has been retrieved (i.e. made active, Dudai, 2006 and Nader, 2003). In many instances an activated memory, re-enters a labile state and must go through a ‘reconsolidation’ process to again stabilize and make it again permanent. The initial observation that following recall a memory was again subject to disruption by electro-convulsive shock (ECS) was made by the Lewis lab in 1968 ( Misanin, Miller, & Lewis, 1968). Similar results regarding the disruptability of a memory made active by a number of different amnesiac agents, including protein synthesis blockers, has been demonstrated in rodents ( Nader et al., 2000, Przybyslawski and Sara, 1997 and Taubenfeld et al., 2001), chicks ( Anokhin, Tiunova, & Rose, 2002), crab ( Pedreira, Perez-Cuesta, & Maldonado, 2002), Hermissenda ( Child, Epstein, Kuzirian, & Alkon, 2003), and our model system, Lymnaea stagnalis ( Sangha, Scheibenstock, & Lukowiak, 2003a). In addition to preserving memory, the so-called reconsolidation process has been hypothesized to allow this now active, labile memory to be modified or updated ( Dudai, 2006, Nader, 2003 and Tronel et al., 2005). We explore here one of the potential consequences of the ability to update or alter the original memory. We hypothesize that as a result of memory-updating the ability to implant a new memory (e.g. memory infidelity) without further operant conditioning in Lymnaea is possible. In the pond snail, L. stagnalis, a 3-neuron central pattern generator (CPG), whose sufficiency and necessity have been demonstrated ( Syed et al., 1990 and Syed et al., 1992), drives aerial respiratory behaviour. This behaviour, which predominates over cutaneous respiration in a hypoxic environment ( Lukowiak, Ringseis, Spencer, Wildering, & Syed, 1996), can be operantly conditioned by applying a tactile stimulus to the respiratory orifice area (the pneumostome) as the snail attempts to open it to breathe. However, because respiration still occurs cutaneously snails trained not to perform this aerial respiration are not harmed. Following associative learning long-lasting memory forms and the duration of this associative memory is dependent on the specific training procedure used ( Lukowiak et al., 1996, Lukowiak et al., 1998 and Lukowiak et al., 2000). Both the consolidation and reconsolidation processes of long-term memory (LTM) in our model system have been shown to be dependent on altered gene activity (i.e. transcription) and de novo protein synthesis (i.e. translation) using both specific blockers (e.g. actinomycin D and anisomycin, respectively) as well as by cooling to 4 °C for 1 h ( Sangha, Scheibenstock, McComb, & Lukowiak, 2003c, 2002). Moreover, transcriptional and translational processes within one of the three CPG neurons, RPeD1, have been shown to be a necessary for associative LTM formation, ( Scheibenstock, Krygier, Haque, Syed, & Lukowiak, 2002) its reconsolidation ( Sangha et al., 2003a), memory extinction ( Sangha, Scheibenstock, Morrow, & Lukowiak, 2003b), and forgetting ( Sangha et al., 2005). In addition, Lymnaea also exhibits context-specific learning and memory ( Haney & Lukowiak, 2001). That is, snails trained in the one context perform as naïve snails do when tested in a different context. However, snails trained on consecutive days in two different contexts have the ability to remember in both contexts ( Haney & Lukowiak, 2001). Since we can easily change the context snails experience during the reconsolidation period following activation of the ‘original’ memory (i.e. for context-1); we may be able to determine if as a result of experiencing a new context (i.e. context-2) during the reconsolidation period snails’ perform as though they were trained in the new context. We hypothesize that as a consequence of memory reactivation in Lymnaea it is possible to cause memory infidelity. Infidelity in the sense that the snail has memory for something it did not undergo training for. That is, do they have memory for the context-2 even though they did not receive operant conditioning in context-2. We also will have the ability of determining whether the ‘old’ memory (i.e. LTM for context-1) has been obliterated by the new memory.