As expected, the average NMDAR-EPSC decay times (τw) recorded from Cre-expressing cells from ΔGluN2A mice were significantly slower than cells from ΔGluN2B mice and control cells. Importantly, decay rate was not affected by the amplitude of the NMDAR-EPSC, indicating effective space clamp (Figure S1B). Normalizing and aligning the traces at the stimulus onset (Figure 1D) shows that NMDAR-EPSCs from ΔGluN2B cells have a significantly www.selleckchem.com/products/BAY-73-4506.html faster

rise than ΔGluN2A cells, with control cells intermediate. These results are consistent with rise times and decays previously described for diheteromeric GluN1/GluN2A and GluN1/GluN2B receptors in heterologous systems (Vicini et al., 1998) and suggest that the Cre-expressing cells have pure diheteromeric populations of synaptic NMDARs. Since GluN2C and GluN2D subunits have lower sensitivity to Mg2+ blockade compared with GluN2A and GluN2B subunits (Monyer et al., 1992), we examined the voltage-dependent Mg2+ sensitivity of the NMDAR-EPSCs in Cre-expressing ΔGluN2A and ΔGluN2B cells. As shown in Figure 1E, there is a high level of voltage-dependent Mg2+ block in Cre-expressing ΔGluN2A and ΔGluN2B cells that was indistinguishable from control

cells, further excluding a measurable contribution of diheteromeric GluN2C- or GluN2D-containing NMDARs. Previous studies have shown that the decay rate of NMDAR-EPSCs is voltage-dependent in the absence of Mg2+ (Hestrin, 1992 and Konnerth et al., 1990) and that early in development (<5 weeks) the decay is slower at positive potentials while CHIR-99021 chemical structure in older mice the decay is faster at positive potentials (Kirson and Yaari, 1996). This developmental switch in the direction of voltage-dependent decay rate may be related

to GluN2 subunit composition. However, as shown in Figure 1F, NMDAR-EPSC decay kinetics are slower at positive holding potentials regardless of subunit composition. Studies in heterologous systems have suggested that the probability of NMDAR opening in response to glutamate is dependent on the GluN2 subunit composition, with GluN2A imparting a higher open probability (PO) than GluN2B (Chen et al., 1999 and Erreger et al., 2005). However, the differential effect of GluN2 subunits on NMDAR open probability has been challenged by previous work in neurons (Chavis and Westbrook, 2001 and Prybylowski whatever et al., 2002). Using the pure diheteromeric GluN1/GluN2A and GluN1/GluN2B synaptic populations, we assessed NMDAR open probability using MK801, an open channel blocker that is effectively irreversible and has been used to estimate PO (Huettner and Bean, 1988 and Jahr, 1992). For each recording, a stable NMDAR-EPSC was obtained, stimulation was stopped for 10 min as 40 μM MK801 was perfused onto the slice, and then stimulation was restarted (Figure 2A). A greater rate of MK801 block was seen with ΔGluN2B than with ΔGluN2A (Figure 2B), suggesting a higher PO in the absence of differences in the presynaptic release probability (see Figure 5D).