Extrasynaptic localization is essential for α5GABAA receptor modulation of dopamine system function

Abstract

Dopamine system dysfunction, observed in animal models with psychosis-like symptomatology, can be restored by targeting Gamma-Aminobutyric Acid type A receptors (GABA_AR) containing the α5, but not α1, subunit in the ventral hippocampus (vHipp). The reason for this discrepancy in efficacy remains elusive; however, one key difference is that α1GABA_ARs are primarily located in the synapse, whereas α5GABA_ARs are mostly extrasynaptic. To test whether receptor location is responsible for this difference in efficacy, we injected a small interfering ribonucleic acid (siRNA) into the vHipp to knock down radixin, a scaffolding protein that holds α5GABA_ARs in the extrasynaptic space. We then administered GL-II-73, a positive allosteric modulator of α5GABA_ARs (α5-PAM) known to reverse shock-induced deficits in dopamine system function, to determine if shifting α5GABA_ARs from the extrasynaptic space to the synapse would prevent the effects of α5-PAM on dopamine system function. As expected, knockdown of radixin significantly decreased radixin-associated α5GABA_ARs and increased the proportion of synaptic α5GABA_ARs, without changing the overall expression of α5GABA_ARs. Importantly, GL-II-73 was no longer able to modulate dopamine neuron activity in radixin-knockdown rats, indicating that the extrasynaptic localization of α5GABA_ARs is critical for hippocampal modulation of the dopamine system. These results may have important implications for clinical use of GL-II-73, as periods of high hippocampal activity appear to favor synaptic α5GABA_ARs, thus efficacy may be diminished in conditions where aberrant hippocampal activity is present.Significance Statement Currently available treatments for psychosis, a debilitating symptom linked with several brain disorders, are inadequate. While they can help manage symptoms in some patients, they do so imperfectly. They are also associated with severe side effects that can cause discontinuation of medication. This study provides preclinical evidence that the drug, GL-II-73, possesses the ability to modulate dopamine activity, a key player in psychosis symptoms, and further provides some mechanistic details regarding these effects. Overall, this work contributes to the growing body of literature suggesting that GL-II-73 and similar compounds may possess antipsychotic efficacy.

Figure 1.

Schematic of radixin knockdown. Diagram of an inhibitory synapse and surrounding extrasynaptic area under (A) baseline conditions and (B) when radixin is knocked down. Figure made using BioRender.

Figure 2.

GL-II-73 was unable to restore dopamine system function when radixin was knocked down. In vivo extracellular electrophysiology was used to measure dopamine cell activity in the VTA. A, left, IS exposure significantly increased the number of spontaneously active cells/track (population activity), which was reversed by intra-ventral hippocampus injection of GL-II-73 (100 ng/μl; 0.75 μl), but not when radixin was knocked down. Right, Representative brain slice with electrode placement in the VTA (black arrow) and cannula placement for drug administration in the vHipp (white arrow), with the corresponding schematics of the brain section (−5.40 mm posterior to bregma) with the box indicating the area in which tracks were found. Neither (B) firing rate nor (C) burst firing was affected by siRNA, shock, or drug treatment. D, Representative traces from control (left) and shocked (right) rats. n = 6–8/group, males and females represented as circles and squares, respectively, ***p = 0.0001, ****p < 0.0001; RDX, radixin.

Figure 3.

Radixin knockdown does not alter PPI. Two days of IS had a significant main effect on PPI as did treatment with GL-II-73 (p = 0.0042); however, post hoc analysis revealed no relevant group differences. n = 9–11/group, males and females represented as circles and squares, respectively. **P < 0.005; RDX, radixin.

Figure 4.

Radixin knockdown increases synaptic α5, without changing surface or total α5 expression. A, Coimmunoprecipitation of α5 and radixin revealed a significant decrease in α5-associated radixin in rats that received radixin-targeted siRNA compared with those that received scrambled siRNA. Representative image of bands below. n = 10/group. B, Conversely, coimmunoprecipitation of α5 and gephyrin revealed a significant increase in gephyrin levels in the radixin knockdown group. Representative image of bands below. n = 9–11/group. C, Treatment with the cross-linking agent caused a significant decrease in the optical density of α5 immunoreactive bands, but no differences in total α5 (homogenate) or surface (cross-linked) were observed between rats that received scrambled siRNA or radixin-targeted siRNA. Representative image of bands below graphs. n = 5/group, males and females represented as circles and squares, respectively. *p < 0.05, **p < 0.01 n = 10; RDX, radixin; OD, optical density.