CaMKII binding to GluN2B is important for massed spatial learning in the Morris water maze.

Ivar S Stein,Michaela S Donaldson,Johannes W Hell

doi:10.12688/f1000research.4660.1

Ivar S Stein, Michaela S Donaldson + Show 1 more

Open Access

https://doi.org/10.12688/f1000research.4660.1

Copy DOI

Journal: F1000Research	Publication Date: Aug 12, 2014
Citations: 18	License type: CC BY 4.0

Affiliation: University of California, Davis

Abstract

Learning and memory as well as long-term potentiation (LTP) depend on Ca (2+) influx through the NMDA-type glutamate receptor (NMDAR) and the resulting activation of the Ca (2+) and calmodulin-dependent protein kinase (CaMKII). Ca (2+) influx via the NMDAR triggers CaMKII binding to the NMDAR for enhanced CaMKII accumulation at post-synaptic sites that experience heightened activity as occurring during LTP. Previously, we generated knock-in (KI) mice in which we replaced two residues in the NMDAR GluN2B subunit to impair CaMKII binding to GluN2B. Various forms of LTP at the Schaffer collateral synapses in CA1 are reduced by 50%. Nevertheless, working memory in the win-shift 8 arm maze and learning of the Morris water maze (MWM) task was normal in the KI mice although recall of the task was impaired in these mice during the period of early memory consolidation. We now show that massed training in the MWM task within a single day resulted in impaired learning. However, learning and recall of the Barnes maze task and contextual fear conditioning over one or multiple days were surprisingly unaffected. The differences observed in the MWM compared to the Barnes maze and contextual fear conditioning suggest a differential involvement of CaMKII and the specific interaction with GluN2B, probably depending on varying degrees of stress, cognitive demand or even potentially different plasticity mechanisms associated with the diverse tasks.

Highlights

The acquisition and storage of new tasks as well as the modification of already existing memories depend on the selective strengthening and weakening of synaptic interactions embedded in extensive networks of neurons (Kessels & Malinow, 2009; Martin et al., 2000; Morris, 2013; Neves et al, 2008)
In order to further dissect the memory deficit observed during the initial Morris water maze (MWM) experiments (Halt et al, 2012), two independent cohorts of GluN2B knock in (KI) mice and their WT littermate controls (10 mice of each genotype per cohort) underwent a massed training protocol on a single day in the MWM
In a number of studies CaMKIIα KO and mutant mice and WT mice (9–10 mice per genotype, previously tested in the elevated plus maze (EPM)) was fear conditioned with three shocks on a single day and a different cohort of naïve litter-matched GluN2B KI and WT mice (10 mice per genotype) was fear conditioned with four shocks over a period of 4 days

Summary

Introduction

The acquisition and storage of new tasks as well as the modification of already existing memories depend on the selective strengthening and weakening of synaptic interactions embedded in extensive networks of neurons (Kessels & Malinow, 2009; Martin et al., 2000; Morris, 2013; Neves et al, 2008). CaMKII binding to aa 1290–1309 on the NMDAR GluN2B subunit is required for this activity-dependent translocation (Halt et al, 2012; Leonard et al, 1999; Strack & Colbran, 1998; Strack et al, 2000) and is crucial for LTP (Barria & Malinow, 2005; Halt et al, 2012; Zhou et al, 2007). Stimulation of CaMKII results in its auto-phosphorylation on T286 causing a persistent Ca2+-independent activation of CaMKII (Lisman et al, 2002). This auto-phosphorylation is increased after LTP and spatial learning in the Morris water maze (MWM)

Methods

Results

Discussion

Conclusion