Converging evidence suggests a link between stress, depression and excitatory amino acids in the prefrontal cortex (Moghaddam , 1993; Jedema and Moghaddam, 1994). We have shown that chronic treatment with the tricyclic antidepressant imipramine reduces glutamate release in rat prefrontal cortex and increases the expression of the GluR2 subunit of the AMPA (-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid) receptor (Michael-Titus et al., 2000; Nazir et al., 2002). In this study we have investigated changes in GluR2 expression in a stress model in tree shrews. In this model we have also tested the effects of the new compound SONU20176289 (7-{4-[3-(5-fluoro-1H-indol-3-yl) - propyl] - piperazin-1-yl} -3H-benzooxazol-2-one). This compound shows in vitro and in vivo a combined dopaminergic and serotonergic profile (SSRI activity combined with partial 5-HT_1A agonist and partial dopamine D₂ agonist activity). Male adult tree shrews were submitted to chronic psychosocial stress as previously described (Fuchs et al., 1996). SONU20176289 (or the vehicle) was administered orally at 6 mg/kg/day for 4 weeks. At the end of the experiments animals were anaesthetised and perfused with 4% paraformaldehyde and brain sections (30 microns) were stained with a GluR2 antibody (concentration 1:500, Chemicon, UK) with the diaminobenzidine method. The staining intensity was analysed in two non-overlapping fields in the prefrontal region. The labelling of cells was quantified by densitometry in one section per animal. The average cell signal minus the surrounding background signal was calculated for 180-200 cells in each section, and was expressed in arbitrary units. Means ± SEM obtained with 3-5 animals per group were compared with two-way ANOVA followed by Bonferroni's t test. Stress induced a significant decrease in GluR2 staining. Thus, in field 1 the mean ± SEM staining in controls was 68.25 ± 4.01 vs. 40.2 ±3.66 in stressed animals (P<0.05), i.e. a stress-associated decrease in signal of 41% compared to controls. SONU20176289 had no intrinsic effect on GluR2, but fully reversed the effects of stress (58.17 ± 4.72 and 63 ± 4.09, respectively). The analysis showed a significant interaction between stress and treatment (P<0.01). Similar results were obtained in field 2: the staining in controls was 62.75 ± 2.83 vs. 39.4 ±2.53 in stressed animals (P<0.05), and this was reversed in treated animals (58.25 ± 2.83). These observations show that stress can trigger significant changes in the GluR2 subunit. This subunit confers low calcium permeability to AMPA receptors (Hollmann et al., 1991) and its decrease may lead to an increased risk of intracellular calcium overload. Interestingly, a decrease in GluR2 transcripts in the dorsolateral prefrontal cortex in major depression has recently been reported (Beneyto and Meador-Woodruff, 2002). Treatment with compounds with antidepressant potential may have a long-term neuroprotective role, as postulated previously (Michael-Titus et al., 2000).

Beneyto, M. and Meador-Woodruff, H. (2002) Soc. Neurosci. Abstr. 497.16.
Fuchs, E. et al. (1996) Pharmacol. Biochem. Behav., 54, 219-228.
Hollmann, M. et al (1991) Science, 252, 851-853.
Jedema, H.P.and Moghaddam, B (1994) J. Neurochem., 63, 785-788.
Michael-Titus, A. T. et al. (2000) Neuroscience , 100, 681-684.
Moghaddam, B. (1993) J. Neurochem., 60, 785-788.
Nazir, S. et al. (2002) Br. J. Pharmac., 135 (Proc. Suppl), 370P.