Luteolin inhibits GABAA receptors in HEK cells and brain slices

Abstract

Modulation of the A type γ-aminobutyric acid receptors (GABAAR) is one of the major drug targets for neurological and psychological diseases. The natural flavonoid compound luteolin (2-(3,4-Dihydroxyphenyl)- 5,7-dihydroxy-4-chromenone) has been reported to have antidepressant, antinociceptive, and anxiolytic-like effects, which possibly involve the mechanisms of modulating GABA signaling. However, as yet detailed studies of the pharmacological effects of luteolin are still lacking, we investigated the effects of luteolin on recombinant and endogenous GABAAR-mediated current responses by electrophysiological approaches. Our results showed that luteolin inhibited GABA-mediated currents and slowed the activation kinetics of recombinant α1β2, α1β2γ2, α5β2, and α5β2γ2 receptors with different degrees of potency and efficacy. The modulatory effect of luteolin was likely dependent on the subunit composition of the receptor complex: the αβ receptors were more sensitive than the αβγ receptors. In hippocampal pyramidal neurons, luteolin significantly reduced the amplitude and slowed the rise time of miniature inhibitory postsynaptic currents (mIPSCs). However, GABAAR-mediated tonic currents were not significantly influenced by luteolin. These data suggested that luteolin has negative modulatory effects on both recombinant and endogenous GABAARs and inhibits phasic rather than tonic inhibition in hippocampus.

Figure 1. Inhibition of GABA-activated currents by luteolin in the recombinant GABA_ARs.

Recombinant GABAARs composed by α1β2 (A), α1β2γ2 (B), α5β2 (C), and α5β2γ2 (D) were first activated by GABA in the absence of luteolin (circles). Dose-response relationships were calculated from the value of peak whole-cell current amplitudes induced by varying GABA concentrations normalized to the value of I_max that were activated by maximum GABA concentration. In the presence of 50 μM of luteolin (square), the amplitudes of GABA-activated currents were also normalized to the value of I_max of the same cell. All data points and bars represent mean values ± s.e.m. More detailed data for the dose-response curves of GABAARs are presented in Table 1.

Figure 2. Inhibition curves of luteolin in the recombinant GABA_ARs.

GABA currents were activated by 3 μM of GABA in α1β2 (n = 8) (A) and α1β2γ2 (n = 10) (B) receptors, and by 2 μM of GABA in α5β2 (n = 10) (C) and α5β2γ2 receptors (n = 10) (D). Inhibition curves were calculated by normalizing values of the relative currents obtained following application of varying concentrations of luteolin to the values obtained in the absence of luteolin. All data points and bars represent mean values ± s.e.m.

Figure 3. Inhibition effect of luteolin on medium versus high dose of GABA-activated current responses in recombinant GABA_ARs.

Representative current traces showed medium or high doses of GABA-activated current responses in α1β2 (A), α1β2γ2 (C), α5β2 (E), and α5β2γ2 receptors (G) before and after 50 μM of luteolin. The quantitative results of luteolin inhibition were calculated from the value of GABA currents in the presence of luteolin normalized to the value before luteolin in α1β2 (n = 9) (B), α1β2γ2 (n = 9) (D), α5β2 (n = 10) (F), and α5β2γ2 receptors (n = 10) (H). All data points and bars represent mean values ± s.e.m. *P < 0.05; **P < 0.01; ***P < 0.001; n.s., no significance using student’s t-test.

Figure 4. Luteolin slowed the activation of recombinant GABA_ARs.

Representative current traces (superimposed and scaled) illustrating the effect of 0.1 and 100 μM of luteolin on current activation of α1β2 (A), α1β2γ2 (C), α5β2 (E), and α5β2γ2 receptors (G). The quantitative results summarized the 10–90% activation time calculated for control currents or after 0.1 and 100 μM of luteolin in α1β2 (n = 15) (B), α1β2γ2 (n = 10) (D), α5β2 (n = 10) (F), and α5β2γ2 receptors (n = 10). All data points and bars represent mean values ± s.e.m. **P < 0.01compared with control using one-way ANOVA.

Figure 5. The effect of luteolin on mIPSCs in hippocampal slices.

(A) Representative traces showed mIPSCs recorded before (top trace) or after 100 μM of luteolin treatment (middle trace) in a hippocampal CA1 pyramidal neuron. 10 μM of bicuculline was applied at the end of the experiment to eliminate mIPSCs (bottom trace). (B) Cumulative probability plots of mIPSC amplitudes (left) and inter-event intervals (right) from the recorded neuron shown in (A). (C–E) Summary of changes in mean mIPSC amplitudes (C), mean frequencies (D), and mean rise time (E) after 0.1 (n = 7) or 100 μM (n = 9) luteolin treatments. All data points and bars represent mean values ± s.e.m. *P < 0.05, **P < 0.01compared with control using one-way ANOVA.

Figure 6. The effect of luteolin on tonic inhibitory currents in hippocampal slices.

(A) The representative recording showed the tonic currents before and after 100 μM of luteolin treatments in a CA1 pyramidal neuron. The tonic current was revealed by 10 μM bicuculline at the end of the experiment. (B,C) Whisker plots (boxes, 25–75%, whiskers, Min-Max; lines, median; + , mean) showed that 0.1 μM (n = 6) (B) or 100 μM of luteolin (n = 10) (C) had no significant effects on tonic inhibition by using student’s t-test.