Back

Calcium-permeable AMPAR in hippocampal parvalbumin-expressing interneurons protect against memory interference

Cooper, M.; Moniri, M.; Abuelem, M.; Bannerman, D. M.; Mann, E. O.

2025-04-30 neuroscience
10.1101/2025.04.29.651201 bioRxiv
Show abstract

Parvalbumin-expressing (PV+) interneurons exert exquisite control over spike output, and plasticity in these inhibitory circuits may be important for maintaining network stability in learning and memory. PV+ interneuron recruitment is primarily mediated by GluA2-lacking Ca2+-permeable AMPA receptors (CP-AMPAR), which support anti-Hebbian plasticity. However, the functional significance of CP-AMPAR-mediated plasticity remains unknown. Using a viral approach to artificially express the GluA2 subunit in hippocampal PV+ interneurons, we replaced CP-AMPAR with GluA2-containing receptors, and in doing so reduced synaptically-evoked Ca2+ transients and anti-Hebbian plasticity. Transfection of hippocampal PV+ interneurons with GluA2 resulted in delay-dependent spatial working memory deficits which increased across trials per session, and impaired reversal learning in the Morris water maze but not initial acquisition. Our data suggest that loss of CP-AMPAR-mediated plasticity in these cells leads to proactive interference, revealing a significant role for dynamic recruitment of PV+ interneurons in the segregation of memories and accurate memory retrieval. HighlightsO_LIViral expression of GluA2 in PV+ interneurons alters synaptic AMPA receptor profile C_LIO_LIGluA2 overexpression in PV+ cells reduces synaptic Ca2+ transients and plasticity C_LIO_LIUpregulating GluA2 in PV+ cells causes delay-dependent deficits in working memory C_LIO_LIAcquisition of reference memories is preserved, but reversal learning is impaired C_LI

Matching journals

The top 5 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.