Primate cerebellar granule cells exhibit a tonic GABA_AR conductance that is not affected by alcohol: A possible cellular substrate of the low level of response phenotype

In many rodent brain regions, alcohol increases vesicular release of GABA, resulting in an increase in the frequency of spontaneous inhibitory postsynaptic currents (sIPSCs) and the magnitude of tonic GABA_A receptor (GABA_AR) currents. A neglected issue in translating the rodent literature to humans is the possibility that phylogenetic differences alter the actions of alcohol. To address this issue we made voltage-clamp recordings from granule cells (GCs) in cerebellar slices from the non-human primate (NHP), Macaca fascicularis. We found that similar to Sprague Dawley rats (SDRs), NHP GCs exhibit a tonic conductance generated by α6δ subunit containing GABA_ARs, as evidenced by its blockade by the broad spectrum GABA_AR antagonist, GABAzine (10 μM), inhibition by α6 selective antagonist, furosemide (100 μM), and enhancement by THDOC (10-20 nM) and THIP (500 nM). In contrast to SDR GCs, in most NHP GCs (~60%), application of EtOH (25-105 mM) did not increase sIPSC frequency or the tonic GABA_AR current. In a minority of cells (~40%), EtOH did increase sIPSC frequency and the tonic current. The relative lack of response to EtOH was associated with reduced expression of neuronal nitric oxide synthase (nNOS), which we recently reported mediates EtOH-induced enhancement of vesicular GABA release in rats. The EtOH-induced increase in tonic GABA_AR current was significantly smaller in NHPs than in SDRs, presumably due to less GABA release, because there were no obvious differences in the density of GABA_ARs or GABA transporters between SDR and NHP GCs. Thus, EtOH does not directly modulate α6δ subunit GABA_ARs in NHPs. Instead, EtOH enhanced GABAergic transmission is mediated by enhanced GABA release. Further, SDR GC responses to alcohol are only representative of a subpopulation of NHP GCs. This suggests that the impact of EtOH on NHP cerebellar physiology will be reduced compared to SDRs, and will likely have different computational and behavioral consequences.