AbstractA subset of the GABAA receptors expressed in recombinant systems and neurons is known to exhibit both constitutive- and agonist-induced gating. Two such receptors are the δ-subunit containing GABAA receptors α4β1δ and α4β3δ, which are expressed in adult rodent hippocampal dentate gyrus granule cells (DGGCs). Here we show that the GABAA receptor mediated tonic current recorded in the presence of tetrodotoxin in adult rodent DGGCs is almost exclusively mediated by constitutively active δ-subunit containing GABAA receptors and that the constitutive current is absent in recordings at 24 °C or in recordings at 34 °C including an intracellular inhibitor of protein kinase C. These factors simultaneously govern the efficacy of an orthosteric agonist at α4β1/3δ receptors, Thio-THIP, in a reciprocal manner. In the absence of constitutive receptor activity, the efficacy of Thio-THIP was increased approximately four-fold relative to recording conditions that favors constitutive activity. Further, only under conditions of an absent constitutive current, the classified neutral antagonist gabazine (GBZ) alone, induced a tonic current in DGGCs (EC50 2.1 μM). This effect of GBZ was not seen in recording conditions of high constitutive activity, was inhibited by picrotoxin (PTX), potentiated by DS2, completely absent in δ-/- mice and reduced in β1-/- mice, but could not be replicated in human α4β1/3δ receptors expressed recombinantly in HEK cells. We hypothesize that specific intracellular components in neurons interact with receptors to determine constitutive gating and receptor responsiveness to orthosteric ligands.Significance statementThe presented data highlight how recording conditions for whole cell patch clamp analysis of α4β1/3δ GABAA receptors can mask important pharmacological effects. Specifically, orthosteric agonists appear with reduced efficacy, and other ligands, here exemplified with the well-known antagonist GBZ, are misinterpreted as being inactive/neutral, although they could have effect in constitutively silent receptors. Unmasking of potential hidden effects are easily done using recording conditions of reduced kinase activity in a relevant neuronal context. It follows that in pathologies with changes in phosphorylation level of δ-subunit containing GABAA receptors, the efficacy of an agonist of these receptors, measured by whole-cell recordings in vitro, will not match the efficacy of the same agonist in an unperturbed neuron in vivo.