XB-ART-42214Neuropharmacology ; 60 (2-3): 488-95.
Ginkgolide B and bilobalide block the pore of the 5-HT₃receptor at a location that overlaps the picrotoxin binding site.
Extracts from the Ginkgo biloba tree are widely used as herbal medicines, and include bilobalide (BB) and ginkgolides A and B (GA and GB). Here we examine their effects on human 5-HT(3)A and 5-HT(3)AB receptors, and compare these to the effects of the structurally related compounds picrotin (PTN) and picrotoxinin (PXN), the two components of picrotoxin (PTX), a known channel blocker of 5-HT(3), nACh and GABA(A) receptors. The compounds inhibited 5-HT-induced responses of 5-HT(3) receptors expressed in Xenopus oocytes, with IC(50) values of 470 μM (BB), 730 μM (GB), 470 μM (PTN), 11 μM (PXN) and >1mM (GA) in 5-HT(3)A receptors, and 3.1mM (BB), 3.9 mM (GB), 2.7 mM (PTN), 62 μM (PXN) and >1mM (GA) in 5-HT(3)AB receptors. Radioligand binding on receptors expressed in HEK 293 cells showed none of the compounds displaced the specific 5-HT(3) receptor antagonist [(3)H]granisetron, confirming that they do not act at the agonist binding site. Inhibition by GB at 5-HT(3)A receptors is weakly use-dependent, and recovery is activity dependent, indicating channel block. To further probe their site of action at 5-HT(3)A receptors, BB and GB were applied alone or in combination with PXN, and the results fitted to a mathematical model; the data revealed partially overlapping sites of action. We conclude that BB and GB block the channel of the 5-HT(3)A receptor. Thus these compounds have comparable, although less potent, behaviour than at some other Cys-loop receptors, demonstrating their actions are conserved across the family.
PubMed ID: 21059362
PMC ID: PMC3070799
Article link: Neuropharmacology
Grant support: 081925 Wellcome Trust , G0601378 Medical Research Council , MRC_G0601378 Medical Research Council , WT081925 Wellcome Trust
Genes referenced: gabarap ptn ptx pxn
Article Images: [+] show captions
|Fig. 1. Structures of the compounds used in this study (Panel A), and an alignment of human Cys-loop receptor M2 regions (Panel B). The alignment includes the human 5-HT3 receptor A and B subunits in addition to subunits from other Cys-loop receptors that are inhibited by BB, GB and PTX. M2 residues are usually referred to using prime (′) notation, which is shown below the alignment. Accession numbers for the alignment are: 5-HT3A (P46098), 5-HT3B (O95264), glycine α1 (P23415), glycine β2 (P48167), GABAA α1 (P14867), GABAA β1 (P18505), GABAA γ2 (Q8N1C3).|
|Fig. 2. BB (panel A), GB (panel B), PXN (panel C) and PTN (panel D) inhibition of the 5-HT3A (filled squares, solid line) and 5-HT3AB (open squares, dotted line) receptor EC50 responses. A inset shows the current rebound sometimes observed following the removal of high concentrations of BB. Panel E shows current-voltage relationships for 5-HT3A receptor responses in the presence of BB, GB, PXN or PTN. The curves have been normalised to the current at measured at −90 mV, and the inset shows the ratios of current amplitudes recorded at +40 mV relative to those at −40 mV. The ratios in the absence of antagonist were unaltered in the presence of BB, GB, PTN or PXN. GA inhibition is not shown as only minimal inhibition was seen at 1 mM in the 5-HT3A receptor only.|
|Fig. 3. Evidence for channel block. A: The use-dependency observed for GB was absent for BB and PXN. For GB, the percentage inhibition measure at peak (31 ± 3.1%) was less than that measured during the first application (38 ± 0.8%). For GB, full recovery was not seen until a second application of GB was applied (inset); at 0.3 mM GB, the first 5-HT response was 91 ± 2% of the second 5-HT response. B–D: The inhibition observed following the co-application (i) of each compound was not seen if the same compound was pre-applied before 5-HT (ii and iii). When compounds were pre-applied immediately before 5-HT (iii), no reduction in the 5-HT response was noted. 5-HT = 2 μM. BB = 1 mM, GB = 300 μM, PXN = 30 μM. For each compound, traces are representative of >5 experiments on ≥2 batches of oocytes.|
|Fig. 4. Concentration–response curves in the absence and presence of increasing concentrations of BB and PXN. Increasing the concentration of PXN (Panel A), caused a rightward shift of the concentration-response curves and a change in maximal current that was consistent with non-competitive antagonism. Because of the low potency and poor compound solubility at high concentrations, it was not possible to perform experiments at higher concentrations and observe shifts for BB (Panel B) or GB (data not shown). Because of these limitations, inhibition of the 5-HT3 receptor response was recorded at single compound concentrations, both alone and in combination (seeTable 3).|