Epipregnanolone as a Positive Modulator of GABAA Receptor in Rat Cerebellar and Hippocampus Neurons

Direct measurements of neurosteroid binding to specific sites on GABAA receptors

BACKGROUND AND PURPOSE

Neurosteroids are allosteric modulators of GABAA currents, acting through several functional binding sites although their affinity and specificity for each site are unknown. The goal of this study was to measure steady-state binding affinities of various neurosteroids for specific sites on the GABAA receptor.

EXPERIMENTAL APPROACH

Two methods were developed to measure neurosteroid binding affinity: (1) quenching of specific tryptophan residues in neurosteroid binding sites by the neurosteroid 17-methylketone group, and (2) FRET between MQ290 (an intrinsically fluorescent neurosteroid) and tryptophan residues in the binding sites. The assays were developed using ELIC-α1GABAAR, a chimeric receptor containing transmembrane domains of the α1-GABAA receptor. Tryptophan mutagenesis was used to identify specific interactions.

KEY RESULTS

Allopregnanolone (3α-OH neurosteroid) was shown to bind at intersubunit and intrasubunit sites with equal affinity, whereas epi-allopregnanolone (3β-OH neurosteroid) binds at the intrasubunit site. MQ290 formed a strong FRET pair with W246, acting as a site-specific probe for the intersubunit site. The affinity and site-specificity of several neurosteroid agonists and inverse agonists was measured using the MQ290 binding assay. The FRET assay distinguishes between competitive and allosteric inhibition of MQ290 binding and demonstrated an allosteric interaction between the two neurosteroid binding sites.

CONCLUSIONS AND IMPLICATIONS

The affinity and specificity of neurosteroid binding to two sites in the ELIC-α1GABAAR were directly measured and an allosteric interaction between the sites was revealed. Adaptation of the MQ290 FRET assay to a plate-reader format will enable screening for high affinity agonists and antagonists for neurosteroid binding sites.

Neurosteroids Mediate Neuroprotection in an In Vitro Model of Hypoxic/Hypoglycaemic Excitotoxicity via δ-GABAA Receptors without Affecting Synaptic Plasticity

Neurosteroids and benzodiazepines are modulators of the GABA_A receptors, thereby causing anxiolysis. Furthermore, benzodiazepines such as midazolam are known to cause adverse side-effects on cognition upon administration. We previously found that midazolam at nanomolar concentrations (10 nM) blocked long-term potentiation (LTP). Here, we aim to study the effect of neurosteroids and their synthesis using XBD173, which is a synthetic compound that promotes neurosteroidogenesis by binding to the translocator protein 18 kDa (TSPO), since they might provide anxiolytic activity with a favourable side-effect profile. By means of electrophysiological measurements and the use of mice with targeted genetic mutations, we revealed that XBD173, a selective ligand of the translocator protein 18 kDa (TSPO), induced neurosteroidogenesis. In addition, the exogenous application of potentially synthesised neurosteroids (THDOC and allopregnanolone) did not depress hippocampal CA1-LTP, the cellular correlate of learning and memory. This phenomenon was observed at the same concentrations that neurosteroids conferred neuroprotection in a model of ischaemia-induced hippocampal excitotoxicity. In conclusion, our results indicate that TSPO ligands are promising candidates for post-ischaemic recovery exerting neuroprotection, in contrast to midazolam, without detrimental effects on synaptic plasticity.

Positive allosteric modulators of GABAA receptor restore chloride current from blockade by competitive antagonists in a ligand-dependent manner

γ-Aminobutyric acid (GABA) is the main inhibitory neurotransmitter. GABA receptor type A (GABAAR) possesses binding sites for a large group of pharmacological agents which are supposed to interact allosterically with each other. The aim of this work was to study the interaction between the positive allosteric modulators (PAMs) and the competitive antagonists of GABAARs. The GABA-induced chloride current (IGABA) was measured in isolated Purkinje cells of rat cerebellum using the patch-clamp technique. PAMs, neurosteroid allopregnanolone (Allo) and zolpidem (Zolp), a drug that positively modulates the GABAAR through interaction with the benzodiazepine (BDZ) site, doubled the IGABA amplitude in the control solution. Competitive antagonist of GABAARs, bicuculline (Bic, 5 µM) blocked the IGABA by 90%. The addition of 1 μM Allo or 0.5 µM Zolp to the Bic solution caused an unblocking effect, so that the IGABA amplitude increased 10 and 4 times from control value, correspondingly. This unblocking effect developed slowly, as evidenced by a threefold increase in the current rise time. Competitive antagonist of GABAARs, gabazine (GBZ, 0.5 µM) blocked the IGABA by 87%. The addition of 1 μM Allo to the GBZ solution caused an unblocking effect, so that the IGABA amplitude increased 7-fold. However, the addition of 0.5 µM Zolp to the GBZ solution did not cause an unblocking effect. So, Allo appeared to have a stronger unblocking potential than Zolp, and Bic binding site showed a higher sensitivity to the action of unblocking PAMs than GBZ binding site. The results indicate for the first time the existence of an allosteric relationship between the sites binding PAMs and the competitive antagonists of GABAAR.

Pregnane neurosteroids exert opposite effects on GABA and glycine-induced chloride current in isolated rat neurons

The ability of endogenous neurosteroids (NSs) with pregnane skeleton modified at positions C-3 and C-5 to modulate the functional activity of inhibitory glycine receptors (GlyR) and ionotropic ɣ-aminobutyric acid receptors (GABA_A R) was estimated. The glycine and GABA-induced chloride current (I_Gly and I_GABA ) were measured in isolated pyramidal neurons of the rat hippocampus and in isolated rat cerebellar Purkinje cells, respectively. Our experiments demonstrated that pregnane NSs affected I_GABA and I_Gly in a different manner. At low concentrations (up to 5 μM), tested pregnane NSs increased or did not change the peak amplitude of the I_GABA , but reduced the I_Gly by decreasing the peak amplitude and/or accelerating desensitization. Namely, allopregnanolone (ALLO), epipregnanolone (EPI), pregnanolone (PA), pregnanolone sulfate (PAS) and 5β-dihydroprogesterone (5β-DHP) enhanced the I_GABA in Purkinje cells. Dose-response curves plotted in the concentration range from 1 nM to 100 μM were smooth for EPI and 5β-DHP, but bell-shaped for ALLO, PA and PAS. The peak amplitude of the I_Gly was reduced by PA, PAS, and 5α- and 5β-DHP. In contrast, ALLO, ISO and EPI did not modulate it. Dose-response curves for the inhibition of the I_Gly peak amplitude were smooth for all active compounds. All NSs accelerated desensitization of the I_Gly . The dose-response relationship for this effect was smooth for ALLO, PA, PAS and 5β-DHP, but it was U-shaped for EPI, 5α-DHP and ISO. These results, together with our previous results on NSs with androstane skeleton, offer comprehensive overview for understanding the mechanisms of effects of NSs on I_Gly and I_GABA .

Neuroactive steroids and depression in early pregnancy

Progesterone (P4) can be metabolized to two general classes of neuroactive steroids (NAS) -those like allopregnanolone (ALLO) and pregnanolone (PA) which are positive allosteric modulators of the Gamma Aminobutyric Acid type A (GABAA) receptor and those like isoallopregnanolone (ISOALLO) and epipregnanolone (EPI) which are negative allosteric modulators of the GABAA receptor. While exogenous administration of ALLO is effective in treating postpartum depression, knowledge gaps remain in the dynamic interplay of NAS across the perinatal period. In particular little is known about ALLO and PA in relation to depression earlier in pregnancy, and the role of ISOALLO and EPI in relation to depression at any point in the perinatal period. In a prospective, nested case/control study in low-income women of color, we compared the metabolism of P4 to four NAS (i.e., ratios ALLO:P4, PA:P4, ISOALLO:P4, EPI:P4) in pregnant women with depression at either or both of the first and second trimesters (cases) and women without depression at either time point (controls). Fifty women (36% depressed, 56% Black, 28% Latina) completed depression screening using a computerized adaptive test of mental health (CAT-MH™) and provided blood serum samples in both trimesters. In longitudinal mixed effects models of both trimesters, PND cases showed higher ratios of ALLO:P4 (p = .002) and PA:P4 (p = .03) compared to controls. In regression models of only first trimester data, there was no significant difference in NAS ratios between cases and controls (p > .05). Conversely, in models of the second trimester, ratios of PA:P4 (p = .002) and ISOALLO:P4 (p = .01) were significantly higher in cases compared to controls, and ratios of ALLO:P4 (p = .08) and EPI:P4 (p = .1) also trended higher in cases. The most severe cases, those with depression at both trimesters, showed an increase in ALLO:P4 (p = .06) and EPI:P4 (p < .001) ratios from the first to the second trimester, whereas controls showed a decrease in these ratios. Secondary analyses confirmed higher levels of ALLO (p = .04) and PA (p = .07) overall in cases compared to controls, along with higher levels of PA (p = .005) and ISOALLO (p = .02) in the second trimester alone. This work suggests a dynamic relationship between NAS and PND; whereas low ALLO levels have been previously associated with postpartum depression, earlier in pregnancy a higher metabolism of P4 to ALLO (and higher ALLO levels) is associated with depression. Some women may show a hormone-sensitive depressive response to acute increases in NAS metabolism in early pregnancy.