Long term potentiation (LTP) is thought to be the cellular equivalent of learning and memory. The cellular changes occurring during this long-lasting strengthening of synaptic transmission are best studied at glutamatergic synapses in the CA1 region of the hippocampus. This type of LTP is, like spatial learning and memory, dependent on the N-methyl-D-aspartate-type glutamate receptor (NMDAR) as well as the calcium/calmodulin-dependent protein kinase II (CaMKII).
Using GluN2B KI mutant mice, the importance of the CaMKII/NMDAR interaction for spatial learning and memory formation was investigated in this study. GluN2B KI mice carry two point mutations that specifically disrupt the critical interaction between CaMKII and the GluN2B subunit of the NMDAR. The findings presented here provide a specific requirement for the activitydependent interaction of CaMKII with the NMDAR during the acquisition of elaborate spatial learning tasks under aversive, stressful conditions. Specifically, learning of the Morris Water Maze (MWM) by GluN2B KI mice is impaired in a single day massed training paradigm. This learning impairment is not observed in spaced MWM training distributed over 6 days, but this paradigm reveals a subsequent consolidation deficit (Halt et al., 2012). The presented results argue not only for a role for CaMKII binding to GluN2B during learning in emotional situations but also consolidation that becomes obvious under the more demanding MWM conditions (Halt et al., 2012) but not the less stressful Barnes maze, as presented here. Contextual fear conditioning is a strong and aversive associative learning paradigm and is not affected in the GluN2B KI mice. This discrepancy between spatial and contextual learning might be due to an unaffected formation of a spatial representation before shock delivery and increased stress levels or slight regional and signaling differences between the two forms of learning.
The activity-induced clustering of CaMKII at postsynaptic sites and the underlying interaction with the NMDAR contribute to synapse specificity of LTP. As part of this thesis it was shown in hippocampal cultures that the activation, diffusion and postsynaptic trapping of endogenous CaMKII takes about 3 min. The clustering of the bulk of the translocated kinase is, except for a small stable pool, reversed within 5 min. In addition, the activity-driven translocation of the kinase is dependent on postsynaptic anchoring of apocalmodulin
under basal conditions by neurogranin, but the CaMKII interaction with the GluN1 subunit of the NMDAR is not required.
The calcium influx through the NMDAR is important for the modulation of synaptic plasticity and is regulated through PKA phosphorylation. In this thesis the novel regulatory phosphorylation site S1166 on the GluN2B subunit of the NMDAR was biochemically characterized. It was shown that S1166 phosphorylation is controlled by norepinephrine and dopamine in acute forebrain slices as well as in vivo, at least under basal conditions. S1166 phosphorylation is upregulated through forced swim-induced stress and might be important for plasticity and learning under emotionally charged conditions.