Electron microscopic immunocytochemical methods were used to determine the localization, subcellular distribution and expression of activity-regulated cytoskeletal protein (Arc/Arg3.1) in dentate gyrus after unilateral induction of long-term potentiation (LTP) in the perforant pathway of anaesthetized rats. At 2 h post-induction, immunoreaction product was visible in the dentate gyrus in both the granule cell and molecular layers. Arc expression was higher in the potentiated than the unstimulated contralateral hemisphere. Single-section electron microscopy analysis in unstimulated tissue and in tissue prepared 2 and 4 h after LTP induction showed Arc immunoreactivity (Arc-IR) in dendrites, dendritic spines and glia. Arc-IR was associated with synaptic and non-synaptic plasma membrane apposed to axon terminals and with cytoplasmic organelles, including the cytoskeleton. Arc-IR was also present in neuronal perikarya and there was occasional labelling of nuclei and axons. At 2 h post-LTP induction, there were significant increases in Arc-IR within the granule cell and molecular layers of the dentate gyrus and particularly within the middle molecular layer relative to the inner and outer molecular layers. This increase in Arc expression 2 h after LTP induction was blocked by the N-methyl-D-aspartate receptor antagonist (RS)-3-2-carboxypiperazin-4-yl-propyl-1-phosphonic acid. In animals killed 4 h after LTP induction, Arc expression had declined and differences between the potentiated and unpotentiated hemispheres were no longer significant. Our data provide ultrastructural evidence for a transient LTP-associated increase in the expression of Arc protein in the middle molecular layer of the dentate gyrus, with preferential targeting to dendrites, dendritic spines and glia.

No abstract

The amyloid beta/A4 protein precursor (APP) has been shown to be implicated in age-associated plastic changes at synapses that might contribute to memory loss in Alzheimer's disease. As APP has previously been reported to have multiple functions during normal development, we have employed a one-trial passive avoidance task in day-old chicks to study its role in the process of memory formation. Administration of anti-APP antibodies, injected 30 min pretraining, prevented memory for a one-trial passive avoidance task in day-old chicks without effects on general behaviour or initial acquisition. Amnesia was apparent by 30 min post-training and lasted for at least 24 h. The same result was obtained by down-regulation of APP expression by APP-antisense, injected 8-12 h pretraining. However, injections of anti-APP antibodies or APP antisense at later post-training time did not cause amnesia for the task. Unlike antibodies and antisense, injection of the APP328-332 pentapeptide, in either orientation, 30 min pretraining, rescued the memory and prevented antisense-induced amnesia. The post-training time within which the antibody- and antisense-induced amnesia, and within which the APP peptides prevent amnesia, correspond to that during which memory formation is vulnerable to disruption of the putative signal transduction functions of APP. These results suggest that: (i) APP is required during an early phase of memory formation, and (ii) the memory enhancing effect of secretory APP is localized within a 5-mer sequence of growth-promoting domain.

No abstract