Vm regulates mGlu5-NMDA receptors crosstalk in hippocampal neurons
The NMDA receptor conductance is well established to be regulated by type I mGlu receptors . Specifically, NMDA receptor potentiation by mGlu5 receptors involves the Gq-protein-coupled receptor (GPCR) pathway, which includes protein kinase C (PKC) and Src signaling, in various neuronal contexts . In this study, we aimed to investigate whether Vm could regulate the crosstalk between mGlu5 and NMDA receptors in primary cell cultures of hippocampal neurons (Figure 6 ). To avoid stimulation of mGlu1 receptors, we used the mGlu1-specific negative allosteric modulator (NAM), CPCCOEt (100 µM), while stimulating mGlu5 receptors with DHPG. We recorded NMDA-induced currents in the absence of magnesium (Mg2+) to prevent voltage-dependent blockade of NMDA receptors. When applied alone at a holding potential of -80 mV, DHPG (50 µM) had a negligible effect (1.046 ± 0.45 pA/pF). In contrast, NMDA (30 µM) induced an inward current of 23.4 ± 3.7 pA/pF (Figure 6A ), which was potentiated by 39.5 ± 6.5% by DHPG when applied at -80 mV (INMDA + DHPG/INMDA ratio measured just before and after DHPG application, Figure 6B, 6C left ). However, at a holding potential of -40 mV, DHPG only induced a 25.4 ± 5.2% potentiation of NMDA current density (Figure 6B, 6C right ). Paired measurements of INMDA + DHPG/INMDA performed subsequently at -80 mV or -40 mV on the same neuron in a random order confirmed a reduction of DHPG-induced NMDA receptor potentiation at -40 mV (+25 ± 4.8%) compared to -80 mV (+38.11 ± 8.2%) (Figure 6D ). These results demonstrate the crucial role of Vm in the control of mGlu5-NMDA crosstalk in neurons, and they corroborate our previous findings on the global voltage-dependence of mGlu5 receptor activity, which is inhibited by depolarized membrane potentials.
Therefore, the optimal functioning of the mGlu5 receptor, which enhances NMDA receptor activity in neurons, occurs at the resting potential of neurons. This finding appears to conflict with the fact that NMDA receptors are typically blocked by Mg2+ at resting membrane potential. Activation of the NMDA receptor indeed requires depolarization of the postsynaptic element to release magnesium from the channel pore, and simultaneous binding of glutamate released from the depolarized axon terminal. However, recent studies have shown that the physiological concentration of Mg2+ in the interstitial medium (0.7 mM, ) is lower than what is typically used experimentally in the ACSF composition to record ex vivo neuronal activity, suggesting that the NMDA receptor activity at resting potential in physiological concentrations of Mg2+ may have been underestimated . Furthermore, GluN2D-containing receptors may exhibit reduced Mg2+ sensitivity compared to GluN2A- or GluN2B-containing receptors . Therefore, we investigated the mGlu5-induced potentiation of NMDA receptors at resting potential in the presence of 0.7 mM Mg2+, by recording currents and calcium transients (Figure 7 ). Whole-cell current recordings revealed a residual NMDA-induced current of 1.80 + 0.30 pA/pF density, which was potentiated to 2.40 + 0.40 pA/pF by DHPG co-application (Figure 7A ). The full current potential relationships are displayed in Figure 7B .
In a second set of experiments, we measured calcium fluctuations, still in presence of physiological concentrations of Mg2+and following successive application of DHPG and AP5 (Suppl Video 3 ). GCaMP6s fluorescence fluctuations highlighted basal spontaneous Ca2+ transients, of various shape, size, and kinetics, as illustrated by a projection of all spontaneous Ca2+ transients recorded during 2 min and 15 sec (Figure 7C ). The variety of these Ca2+ events certainly relies on the nature of the receptors and channels involved in their triggering. To focus on synaptic events, we selected small and non-propagating calcium transients. The majority of these events were blocked by AP5 (50 µM) at the end of the experiment, revealing NMDA-dependent events (Figure 7D ). We could a posteriorfocused in these regions of interest to measure the influence of mGlu5 activation on NMDA function (Figure 7E ). DHPG increased the number of basal calcium transients (Figure 7D ) and their area under curve (AUC indeed significantly increased from 0.102 +0.001 to 0.154 + 0.002 following DHPG application, Figure 7F ). More importantly, DHPG changed the shape of the NMDA-dependent events inducing a more sustained calcium inflow over time (Figure 7E ). Taken together, our data show that the mGlu5 receptor potentiates NMDA receptor activity at resting membrane potential, increasing currents and particularly calcium influx, thus expanding the functional importance of NMDA receptors to resting neurons.