After a 2 hr incubation, we observed significant increases in amounts of dCREB2-b compared to amounts in brains cultured in 20-mM-Mg2+ medium (Figure 7E). The increase in

dCREB2-b in Mg2+-free medium is strongly suppressed by the NMDAR antagonist MK801 and by dNR1 mutation, suggesting that in the absence of Mg2+ block, Ca2+ entry through dNMDARs increases dCREB2-b protein levels. We next wanted to determine whether the increase click here in dCREB2-b expression observed in elav/dNR1(N631Q) flies is sufficient to inhibit LTM formation. Thus, we plotted one-day memory scores after spaced training as a function of dCREB2-b protein levels in wild-type and hs-dCREB2-b flies heat shocked for various durations. As seen in Figure 8A, defects in LTM formation were highly correlated with dCREB2-b protein expression. Heat-shocked wild-type flies and non-heat-shocked hs-dCREB2-b flies had normal one-day memory and similar dCREB2-b expression. On the other hand,

hs-dCREB2-b flies expressed increasing Kinase Inhibitor Library cell line amounts of dCREB2-b protein upon increasing heat-shock duration. This increase in dCREB repressor expression was correlated with a decrease in one-day memory in a linear fashion in the range tested. Significantly, we found that data from elav/dNR1(wt) and elav/dNR1(N631Q) flies plotted on the same graph fit the same regression line as our hs-dCREB2-b data; elav/dNR1(wt) flies expressed wild-type levels of dCREB2-b and had one-day memory scores comparable to wild-type and non-heat-shocked dCREB2-b flies, while elav/dNR1(N631Q) flies expressed similar amounts of dCREB2-b and had similar however memory to hs-dCREB-2b flies heat shocked for 30 min. In contrast, dNR1EP3511 flies did not show any increases in dCREB2-b mRNA or protein but had poor LTM scores (see also Figures 7B and 7C). These results indicate that the increase in expression of the dCREB2-b repressor in Mg2+ block mutants is

correlated with and sufficient to cause the decrease in LTM observed in these mutants, while memory defects in dNR1 hypomorphs likely occurs through a different mechanism. Although the mechanism through which Mg2+ block restricts NMDAR activity is well known, the cellular and behavioral functions of Mg2+ block have not been extensively studied. In this study, we used transgenic flies expressing dNR1(N631Q) to show that Mg2+ block is important for formation of LTM. Previous studies of hypomorphic mutants have shown that NMDARs are required for both learning and LTM. In contrast, our Mg2+ block mutants do not have learning defects. This suggests that although Ca2+ influx through NMDARs is important for learning, inhibition of influx during uncorrelated activity is not. Notably, elav/dNR1(N631Q) flies have slightly enhanced learning. Consistent with this result, NMDAR-dependent induction of hippocampal LTP is enhanced in the absence of external Mg2+ ( Mizuno et al.