01, ANOVA, n = 11–14 per group). Post hoc analysis showed that repeated stress caused a substantial downregulation of eEPSC amplitude in saline-injected animals (AMPA: 50%–59% decrease; NMDA: 44%–52% decrease, p < 0.01) but had little effect on MG132-injected animals (AMPA: 3%–7% decrease; NMDA: 2%–5% decrease, p > 0.05). Injection of MG132, but not saline, also blocked the reducing effect of repeated stress on mEPSC amplitude and frequency in PFC slices (Figures

6C and 6D, MG132: 14.0 pA ± 0.5 pA, 3.2 Hz ± 0.4 Hz, n = 8; MG132+stress: 15.0 pA ± 0.5 pA, 3.6 Hz ± 0.5 Hz, n = 10, p > 0.05). In vitro studies further confirmed that the proteasome-mediated degradation of glutamate receptors may underlie the reduction of mEPSC by long-term CORT treatment. As shown in Figure 6E, CORT (100 nM, 7 day) significantly decreased mEPSC in vehicle-treated Quisinostat neurons (control: 37.1 pA ± 2.9 pA, 12.1 Hz ± 1.8 Hz, n = 9; CORT: 23.3 pA ± 2.9 pA, 7.1 Hz ± 1.2 Hz, n = 7, p < 0.05) but failed

to do so in MG132-treated (1 μM) neurons (MG132: 36.8 pA ± 3.2 pA, 11.5 Hz ± 2.3 Hz, n = 11; MG132+CORT: 35.4 pA ± 2.8 pA, 10.4 Hz ± 1.9 Hz, n = 7, p > 0.05). Another proteasome inhibitor lactacystin (1 μM) gave similar blockade (lact: 34.5 pA ± 3.0 pA, 10.5 Hz ± 2.0 Hz, n = 8; lact+CORT: 33.9 pA ± 1.8 pA, 9.2 Hz ± 1.1 Hz, n = 8, p > 0.05). However, the find protocol reducing effect of CORT was insensitive to the general lysosomal enzyme inhibitor chloroquine (200 μM, Chlq: 36.2 pA ± 3.9 pA, 9.4 Hz ± 1.4 Hz, n = 6; Chlq+CORT: 22.4 pA ± 1.2 pA, 5.0 Hz ± 0.8 Hz, n = 6, p < 0.05), the lysosomal protease inhibitor these leupeptin (200 μM, leu: 35.9 pA ± 2.4 pA, 12.2 Hz ± 0.9 Hz, n = 8; leu+CORT: 22.3 pA ± 1.3 pA, 5.6 Hz ± 1.4 Hz, n = 8, p < 0.05),

or the membrane-permeable calpain protease inhibitory peptide 11R-CS (2 μM, Wu et al., 2005; 11R-CS: 34.9 pA ± 3.9 pA, 9.8 Hz ± 1.2 Hz, n = 7; 11R-CS+CORT: 21.0 pA ± 1.9 pA, 5.2 Hz ± 0.3 Hz, n = 5, p < 0.05). Biochemical measurement of glutamate receptor subunits in PFC slices (Figures 6F and 6G) indicated that MG132-injected rats exhibited the normal level of GluR1 and NR1 after being exposed to 7 day restraint stress (GluR1: 6.6% ± 10.7% decrease; NR1: 10.5% ± 12.8% decrease, n = 4 pairs, p > 0.05), which was in sharp contrast to the reduced expression of GluR1 and NR1 in saline-injected rats after repeated stress (GluR1: 48.3% ± 10.1% decrease; NR1: 59.7% ± 11.9% decrease, n = 4 pairs, p < 0.01). In addition, the CORT-induced (100 nM, 7 day) decrease of GluR1 expression (49.0% ± 1.4% decrease, n = 6, p < 0.01) was abolished by proteasome inhibitors (Figure 6H, MG132: 8.2% ± 11.7% decrease; lactacystin: 7.9% ± 11.2% decrease, n = 4, p > 0.05). Taken together, these results suggest that repeated behavioral stress or long-term CORT treatment induces the ubiquitin/proteasome-dependent degradation of GluR1 and NR1, leading to the depression of glutamatergic transmission in PFC.