Figure 9.

PSD-95 acts downstream of synaptic activity and other signaling pathways involved in LTP. A–D, AMPA-EPSCs measured in paired recordings from CA1 neurons held at –60 mV to compare amplitudes in uninfected and PSD-95–GFP-expressing neurons. Slices had been subjected to different pharmacological treatments (10 mm Mg ²⁺, 20 μm KN-93, 20 μm PD98059, 20 μm SB203580). Calibration: 30 pA, 40 msec. E, Summary and comparison of changes in AMPA-EPSCs. *Significant potentiation within one experimental group. In high Mg ²⁺, PSD-95–GFP increased AMPA currents from 100 ± 17.2 to 204.7 ± 28.3% (n = 22; p = 0.003), in KN-93 from 100 ± 12.0 to 184.2 ± 15.1% (n = 27; p = 0.001), in PD98059 from 100 ± 10.1 to 227.7 ± 16.1% (n = 39; p < 0.0005), and in SB203580 from 100 ± 15.5 to 196.1 ± 11.6% (n = 40; p < 0.0005). Comparison between different treatments was done for AMPA_Inf/AMPA_Uninf ratios from each group using KS tests (see Materials and Methods; p values above bars); no significant differences were detected. F, Model for the role of PSD-95 in activity-dependent AMPAR delivery to synapses. Strong synaptic activity leads to signaling events that trigger changes in several proteins, including PSD-95. Synaptic accumulation, multimerization and formation of PDZ interactions by PSD-95 result in the recruitment of extrasynaptic GluR1-containing receptors to synapses, possibly in a complex with PSD-95 and other proteins. Interestingly, an increased level of PSD-95 can drive this process, whereas an increased level of GluR1 does not.