Reduced phosphorylation of GluA1 subunits relates to anxiety-like behaviours in mice

Abstract

Anxiety and depression are highly prevalent and frequently co-morbid conditions. The ionotropic glutamate receptors N-methyl-D-aspartate and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) mediate actions of monoaminergic antidepressants and have been directly targeted by novel fast-acting antidepressants. Less is known about the role of these receptors in anxiety-like states. Here we investigate how two distinct anxiolytic agents, buspirone, a partial 5-HT(1A) agonist, and diazepam, a benzodiazepine, influence phosphorylation of GluA1 subunits of AMPA receptors at the potentiating residue Ser(845) and Ser(831) in corticolimbic regions. To test the functional relevance of these changes, phosphomutant GluA1 mice lacking phosphorylatable Ser(845) and Ser(831) were examined in relevant behavioural paradigms. These mutant mice exhibited a reduced anxiety-like phenotype in the light/dark exploration task and elevated plus maze, but not in the novelty induced hypophagia paradigm. These data indicate that reduced potentiation of the AMPA receptor signalling, via decreased GluA1 phoshorylation, is specifically involved in approach-avoidance based paradigms relevant for anxiety-like behaviours.

Fig. 1

GluA1 double phosphomutant (MT) mice show alterations in anxiety- and depression-related behaviours. Genotypes showed similar locomotor activity in a novel open field test (a), but MT mice spent significantly more time in exploratory rearing than wild type (WT; b). Genotypes did not differ in anxiety-related measures of thigmotaxis (c) or centre time (d ; n=12). In the light/dark exploration task, MT spent significantly more time in (e) and made more entries into (f) the light compartment than WT (n=16). In the elevated plus maze, MT spent more time in (g) and made more entries to the open arms (h), and made significantly more entries to the closed arms (i), than WT (n=19–20). (j) Genotypes did not differ in sucrose preference (n=13–16). In the novelty-induced hypophagia test (k), there was a significant decrease in the amount of milk and an increased latency to drink milk (l) in a novel cage relative to the home cage, but no difference between genotypes (n=15–16). (m) Serum corticosterone levels after forced swim, as compared to home cage, but genotypes did not differ (n=9–11). Data are means ± S.E.M. * p<0.05 ; ** p<0.01. PA, Peripheral activity ; HA, horizontal activity ; FST, forced swim test.

Fig. 2

Regulation of GluA1 phosphorylation by the anxiolytics buspirone and diazepam in C57BL/6J mice. (a) Immunoblot found a significant decrease in Ser⁸⁴⁵ phosphorylation after 0.25 mg/kg diazepam treatment, relative to saline, in the dorsal hippocampus (DH) but not basolateral amygdala (BLA) or prefrontal cortex (PFC). (b) There was a significant decrease in Ser⁸⁴⁵-GluA1 after 1.0 mg/kg buspirone treatment, relative to saline, in DH and BLA but not PFC. No changes were observed in Ser⁸³¹ phosphorylation after diazepam (c) or buspirone (d). n=9–12/group. Data are means ± S.E.M. * p<0.05.