Proceedings of The Physiological Society

Europhysiology 2018 (London, UK) (2018) Proc Physiol Soc 41, C074

Oral Communications

Group I metabotropic glutamate receptors mediate γ-potentiation in mouse hippocampal CA3 region

M. Hadler1, H. Alle1, J. R. Geiger1,2

1. Institute of Neurophysiology, Charité Universitätsmedizin Berlin, Berlin, Germany. 2. NeuroCure Cluster of Excellence, Charité Universitätsmedizin, Berlin, Germany.

  • Oscillation Memory in Mouse Hippocampal CA3 Region. Transient network activity was evoked by bath-application of 150 nM KA. During washout, activity returned to baseline levels. A second induction period produced a marked increase in γ-power. Left Time-frequency plot of an exemplary trace (Band Pass filtered, 15 - 80 Hz). Insets mark exposure to KA ("KA") and time window of analysis (1-4). Right Close-Up traces of baseline activity preceding KA application (1, 3) and subsequent γ-activity (2, 4) with corresponding Fast Fourier transforms.

Deficiencies of network activity in the gamma frequency band (γ, 30 - 80 Hz) and of parvalbumin-positive interneurons (PVI) are frequently observed in models of neuropsychiatric disorders. However, little is known of how they interact in disease. Alterations of activity-induced, mGluR5-dependent synaptic plasticity of excitatory inputs onto PVIs (Zarnadze et al., 2016) may link these phenomena and serve as a mechanism of action in neuropsychiatric pathophysiology. We prepared acute hippocampal slices of wild-type C57Bl6/J mice (p45 - p70) and performed local field potential recordings in the CA3 region. Following 30 minutes of recording baseline activity, we induced a first, transient period of network activity by brief bath-application of kainate (KA, 150 nM). After one hour of washout, which allowed for a return to baseline activity and an additional resting period, an identical second induction of network activity resulted in a roughly doubled amplitude of γ-power (Mean +/- SEM: 209.38 +/- 15.31 % of the first induction period, n = 32). Reminiscent of synaptic plasticity, we call this "oscillation memory" (Fig. 1). To elucidate how mediators of synaptic plasticity contribute to oscillation memory, we continuously bath-applied D-AP5 (50 µM) or MPEP (10 µM), pharmacological antagonists of NMDA-R and mGluR5, respectively. Whereas D-AP5 had no significant effect (248.51 +/- 19.22 %, n = 42; Mann-Whitney U-test vs Control: p > 0.05), application of MPEP attenuated oscillation memory by approximately 50%, as did co-application of both substances (MPEP: 155.90 +/- 12.72 %, n = 36; p < 0.01, D-AP5 + MPEP: 159.22 +/- 12.77 %, n = 37; p < 0.01). In mice undergoing conditional postnatal ablation of mGluR5 in PVIs (PV-mG5-/-), oscillation memory was impaired with no additional reduction by MPEP (PV-mG5-/-: 158.32 +/- 7.96 %, n = 46; PV-mG5-/- + MPEP: 153.26 +/- 14.51 %, n = 25; Mann-Whitney U-test vs PV-mG5-/-: p > 0.05), indicating that mGluR5 activation on PVIs contributes to enhancing subsequent γ-activity in wild-type animals. Our data suggest potentiation of γ-activity as a novel function of synaptic plasticity specific to PVIs. Furthermore, our finding in the conditional mGluR5 knockout, an animal model reported to exhibit behavioral hallmarks of neuropsychiatric disease (Barnes et al., 2015), hints at a functional link between PVI plasticity, oscillation memory and neuropsychiatric pathophysiology.

Where applicable, experiments conform with Society ethical requirements