Ion dependence of mGluRs activity has been

Ion dependence of mGluRs activity has been previously reported, notably to Ca2+ and Cl− (Kuang and Hampson, 2006). While mGlu1 and mGlu3 receptors were demonstrated to be sensitive to Ca2+, Cl− modulation was reported for all mGluRs, with a lesser extent for the mGlu2 receptor (DiRaddo et al., 2014; Jiang et al., 2014; Tora et al., 2015; Vafabakhsh et al., 2015). Of note, the cooperativity between an amino Triptolide ligand and ions to stabilize the active conformation of the receptor is also observed in another receptor from the GPCR class C, the Ca2+ sensing receptor. This receptor is activated concomitantly both by Ca2+ ions and an l-amino acid, such as L-Phe, that binds in a pocket corresponding to the glutamate binding pocket in mGluRs (Zhang et al., 2014). Here, we showed that the mGlu3 receptor is particularly strongly potentiated by Cl− ions as compared to other mGluRs. This strong positive cooperativity between Cl− ions and glutamate occurs through a unique strengthened interaction network between the two lobes of the extracellular domain of the mGlu3 receptor. This network acts as a “Cl−-lock” that dramatically reinforces glutamate binding and function. Previously, both Ca2+ and Cl− ions were proposed to display agonist activity on the mGlu3 receptor (DiRaddo et al., 2015; Vafabakhsh et al., 2015). However, in our experiments, Cl−per se does not display direct agonism, as revealed by the glutamate insensitive mGlu3 mutant T174A, which is not activated by Cl−. We propose that what has been interpreted previously as Cl− direct agonism in very sensitive assays may correspond in fact to the potentiation by Cl− ions of the activity induced by residual ambient glutamate. Indeed, cells are constitutively releasing glutamate in extracellular media. We have recorded concentrations ranging from 0.6 to more than 5 μM. These glutamate concentrations are in line with those measured in another recent publication (Doornbos et al., 2018). Thus, Cl− ions are pure positive allosteric modulators of the mGlu3 receptor. It is unlikely to observe the occurrence of rapid Cl-concentration changes compatible with dynamic modulation of mGlu3 receptor activity. Short-term variations occur during transport or synaptic activities where rapid Cl− fluxes through GABA or glycine-gated ion channels modify Cl-concentrations (Staley et al., 1995). We have previously evaluated the discrete variations of Cl− concentrations in the synaptic cleft during GABAergic events in physiological or pathological conditions (Tora et al., 2015). We found that multiple synaptic events can modify synaptic Cl− concentrations up to 25 mM. Long-term changes in Cl− concentrations can be observed during maturation (Ben-Ari et al., 2007), modifications of extracellular matrix (ECM) and pathologies (De Koninck, 2007; Kaila et al., 2014). Interestingly, fast extracellular Cl-changes due to synaptic activity and slow changes could be additive. Thus, in particular conditions, such as anoxia where variations of Cl-exceeding 30 mM can be observed (Jiang et al., 1992), Cl-variations could reach up to 55 mM. However, given the potency of Cl− on the mGlu3 receptor (∼30 mM) as compared to the physiological extracellular Cl− concentration in mature CNS (∼120 mM), the potentiation of mGlu3 receptor activity by Cl− ions is probably mostly constitutive. Other members of the mGluR family could be dynamically regulated by Cl−, such as the presynaptic mGlu4 receptor (Tora et al., 2015). Indeed, unlike the mGlu3 receptor, mGlu4 presents characteristics that makes it more susceptible to be actively regulated by discrete Cl− variations. First, this receptor is expressed in GABAergic synapses where rapid Cl-variations occur following opening of GABAA receptors. Second, the potency of Cl− on mGlu4 receptoris close to the physiological Cl-concentration (∼80 mM) and presents a steep relationship between Cl− and glutamate-induced activity (with an NHill ∼6). This means that small changes in Cl− concentration yield to large changes in response to glutamate. On the other hand, dramatic changes in Cl− homoeostatis occur during CNS maturation (Ben-Ari et al., 2007). We can hypothesize that mGlu3 receptors should shift from a “dynamic” mode of glutamate response during the early stages to the highly-glutamate sensitive and sustained response mode described in the present study in mature stages. This should affect its biological functions in early stages, such as the crucial control of mGlu5-mediated neuroprotection described in Di Menna et al. (Di Menna et al., 2018). One can hypothesize that when extracellular Cl− concentrations are low in premature stages, mGlu3 receptor should be able to signal and regulate the release of neurotrophic factors from astrocytes, given its high Cl− sensitivity. However, in the absence of specific data this remains purely speculative and further investigations will be needed to evaluate the impact of Cl− on mGluR function in the premature brain. In the mature brain, while it is unlikely to observe the occurrence of Cl− concentrations changes compatible with a significant modulation of mGlu3 receptor activity, a Cl-mediated rapid adaptation of the regulatory role of some mGluRs, such as mGlu4 receptors present on GABAergic terminals, could be observed.