Rational Search for Betaine/GABA Transporter 1 Inhibitors─ In Vitro Evaluation of Selected Hit Compound

Abstract

Inhibitory neurotransmission mediated by γ-aminobutyric acid (GABA) plays an important role in maintaining body homeostasis. Disturbances in GABA signaling are implicated in a multitude of neurologic and psychiatric conditions, including epilepsy, ischemia, anxiety, depression, insomnia, and mood disorders. Clinically relevant increases in GABA neurotransmitter level can be achieved by inhibition of its uptake into presynaptic neurons and surrounding glial cells, driven by GABA transporters (GAT1, BGT1, GAT2, and GAT3). Herein, we focused on the search for inhibitors of the BGT1 transporter which is understudied and for which the therapeutic potential of its inhibition is partly unknown. We applied multilevel virtual screening to identify compounds with inhibitory properties. Among selected hits, compound 9 was shown to be a preferential inhibitor of BGT1 (IC₅₀ 13.9 μM). The compound also revealed some inhibitory activity against GAT3 (4x lower) while showing no or low activity (IC₅₀ > 100 μM) toward GAT1 and GAT2, respectively. The predicted binding mode of compound 9 was confirmed by mutagenesis studies on E52A, E52Y, Q299L, and E52A+Q299L human BGT1 mutants. Subsequent evaluation showed that the selected hit displayed no affinity toward major GABA_A receptor subtypes. Moreover, it was nontoxic when tested on normal human astrocytes and even showed some neuroprotective activity in SH-SY5Y cells. Compound 9 is considered a promising candidate for further evaluation of the therapeutic potential of BGT1 transporter inhibition and the development of novel inhibitors.

Keywords: betaine/GABA transporter 1; biological evaluation; inhibitor; rigid GABA analogue; virtual screening.

Figure 1

Virtual screening of ZINC15 and ENAMINE databases in the search of novel BGT1 inhibitors.

Figure 2

Compounds were selected from virtual screening.

Figure 3

In vitro screening results for selected compounds are presented as a percentage of total [³H]GABA uptake by hBGT1 stably expressed in CHO cells (A). Compounds were tested at 10 and 100 μM concentrations. Data shown are pooled experiments (n = 2) with three technical replicates normalized to 100% [³H]GABA uptake (mean ± SEM). IC₅₀ determination for the most active compound 9 (B). Data represent one of three independent experiments, each performed in three technical replicates. Activity of compound 9 (10 and 100 μM) at hGAT1 (C), hGAT2 (D) and hGAT3 (E) transporters stably expressed in CHO cells, given as a percentage of total [³H]GABA uptake. Data shown are pooled experiments (n = 2) with three technical replicates normalized to 100% [³H]GABA uptake (mean ± SEM).

Figure 4

Binding mode of compound 9 within the hBGT1 transporter. The figure represents the state of ligand-transporter complex after 200 ns molecular dynamics simulation (A, B). RMSD (root-mean-square deviation) changes for compound 9 (ligand) and for residues within 7 Å of compound 9 in the course of molecular dynamics (C). Distance changes between amino acids and ligand’s functional groups involved in key interactions during MD simulation (D).

Figure 5

Concentration–response curves for compound 9 at the wild type (wt) of hBGT1 and mutants E52A, E52Y, Q299L, and E52A+Q299L transiently expressed in tsA201 cells. Curves were normalized to the maximal [³H]GABA uptake. The figure shows one representative curve for each transporter. Three to four independent experiments, performed with three technical replicates for each, gave similar results. The determined IC₅₀ values are reported in Table 1.

Figure 6

Effect of compound 9 on normal human astrocytes. (A) Viability and proliferation of NHA after the 24 or 48 h incubation with compound 9 at the concentration of 1–100 μM (solutions in culture medium) were analyzed by the WST-1 assay. Graphs represent mean ± SD of three independent experiments performed in triplicate (n = 3). (B) Representative images of NHA treated with 500 μM compound 9 (solution in culture medium) for 72 h. The cells were imaged by a phase-contrast microscope; scale bar = 200 μm.

Figure 7

Morphological changes of astrocytes induced by compound 9 and reversal of the effect by GABA. After the 72 h incubation of NHA in PBS (control), 500 μM GABA in PBS, 500 μM compound 9 in PBS, or a solution of GABA (500 μM) and compound 9 (500 μM) in PBS, the cells were fixed and stained with Phalloidin-Atto 565. The samples were scanned using a laser confocal microscope; scale bar = 200 μm.

Figure 8

Effect of compound 9 at 10 or 50 μM and/or GABA at 10 μM on SH-SY5Y neuroblastoma cell viability damaged by 1 mM MPP⁺ after 24 h of incubation. Graphs represent mean ± SD of two independent experiments performed in quadruplicate. Statistical significance was set at ***p < 0.001, **p < 0.01, *p < 0.05 by GraphPad Prism 8 software using one-way ANOVA and Bonferroni’s post hoc test in comparison with the positive control MPP⁺ (1000 μM).