Back

Activation of CP-AMPARs is required for homosynaptic and heterosynaptic structural LTP in the hippocampus

Koek, L. A.; Sanderson, T. M.; Bond, G.; Georgiou, J.; Scholl, B.; Collingridge, G. L.

2025-10-16 neuroscience
10.1101/2025.10.16.682923 bioRxiv
Show abstract

Long-term potentiation (LTP) involves alterations in synaptic structure that are believed to underlie the persistent increase in synaptic efficacy. Here we compared structural LTP (sLTP) in EGFP-labelled spines with functional LTP, using field potential recording, at CA3-CA1 synapses in mouse hippocampal slices for [~] 2 h following theta-burst stimulation (TBS). Activity-dependent labelling with FM4-64 allowed us to compare activated (FM+) and non-activated synapses and thereby compare homo- and hetero-synaptic sLTP. In addition, we related spine volume changes according to the probability of release, P(r), of activated synapses. At homosynaptic sites there was the expected NMDA receptor (NMDAR)-dependent potentiation of spine volume that persisted throughout the recording period. We found that this sLTP also required the synaptic activation of CP-AMPARs. There was also sLTP at heterosynaptic sites that, surprisingly, developed more quickly than the associated homosynaptic sLTP. This heterosynaptic sLTP was also dependent on the synaptic activation of both NMDARs and CP-AMPARs. Additionally, we observed a trans- and hetero-synaptic interaction, whereby the heterosynaptic spines grew according to the P(r) of the neighbouring active (homosynaptic) synapse. These observations have therefore advanced our understanding of sLTP in several ways; the first demonstration of the absolute requirement for the synaptic activation of CP-AMPARs for sLTP; the magnitude of heterosynaptic sLTP relative to homosynaptic sLTP and the hitherto unexpected combination of trans- and hetero-synaptic interactions.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.