Memory reinstatement after erasure

Loeb's 1900 survey of comparative brain physiology mentions Aplysia only in passing. He reports that Steiner tried to produce circus movements in Aplysia and other mollusks by dividing the oesophageal commissure on one side, and that the attempt succeeded no better than in Pterotrachea.¹ Loeb also called Steiner's claim that the pedal ganglion alone controls locomotion anthropomorphic. He held that the observation shows only that fibres linking skin and foot muscle pass through the ganglion.¹ In this early literature Aplysia is one of several mollusks used to test the function of ganglia, not a model of memory.

Pla-Mauri and Solé describe the classic Aplysia case. A slight touch to the siphon triggers gill withdrawal. After a tail shock the response is sensitized, and repeated light touch without reinforcement habituates it. They note that sensitization involves enhanced neurotransmitter release, while habituation results from reduced synaptic activity.⁹ Chen and colleagues describe PKM Apl III as a homolog of mammalian PKMζ, formed from atypical PKC Apl III by calpain-dependent cleavage.² They report that earlier work from their group found its inhibition abolished both consolidated long-term sensitization and LTF, the synaptic basis of the memory.³

Chen and colleagues state that their morphological results challenge the notion that sensitization memory depends on the persistence of particular facilitated synapses.⁴ They then tested whether the memory could be reinstated after reconsolidation blockade and PKM inhibition, two treatments previously shown to eliminate LTF.⁴ Bédécarrats and colleagues later summarized this line of work. They describe an early, protein synthesis-dependent priming component that persists independently of behavioral and synaptic change. It permits memory to be reinstated after reconsolidation blockade, or induced by partial training after retrograde amnesia.⁵

Bédécarrats and colleagues reported in 2018 that sensitization memory in Aplysia could be transferred by injecting RNA from sensitized animals into naïve ones. They proposed this as support for a nonsynaptic, epigenetic model of storage.⁶ Gershman and colleagues argue that if single cells can learn, they must use a non-synaptic form of memory storage.⁷ They note that earlier claims of memory transfer by brain extracts were controversial and failed careful replication, and that recent work has revisited them.⁷ Lyon observes that Aplysia findings on the molecular basis of memory are accepted as relevant to human memory, but have not yet been connected to prokaryotes.⁸