Psoralea corylifolia extract induces vasodilation in rat arteries through both endothelium-dependent and -independent mechanisms involving inhibition of TRPC3 channel activity and elaboration of prostaglandin

Abstract

Context: Fructus Psoralea, Psoralea corylifolia L. (Leguminosae), has been widely used in traditional medicines for the treatment of dermatitis, leukoderma, asthma and osteoporosis.

Objectives: In this study, we sought to study mechanisms underlying the vasoactive properties of Psoralea corylifolia extract (PCE) and its active ingredients.

Materials and methods: To study mechanisms underlying the vasoactive properties of PCE prepared by extracting dried seeds of Psoralea corylifolia with 70% ethanol, isometric tension recordings of rat aortic rings and the ionic currents through TRPC3 (transient receptor potential canonical 3) channels were measured with the cumulative concentration (10-600 μg/mL) of PCE or its constituents.

Results: Cumulative treatment with PCE caused the relaxation of pre-contracted aortic rings in the presence and absence of endothelium with EC₅₀ values of 61.27 ± 3.11 and 211.13 ± 18.74 μg/mL, respectively. Pretreatment with inhibitors of nitric oxide (NO) synthase, guanylate cyclase, or cyclooxygenase and pyrazole 3, a selective TRPC3 channel blocker, significantly decreased PCE-induced vasorelaxation (p < 0.01). The PCE constituents, bakuchiol, isobavachalcone, isopsoralen and psoralen, inhibited hTRPC3 currents (inhibited by 40.6 ± 2.7, 27.1 ± 7.9, 35.1 ± 4.8 and 47.4 ± 3.9%, respectively). Furthermore, these constituents significantly relaxed pre-contracted aortic rings (EC₅₀ 128.9, 4.5, 32.1 and 114.9 μg/mL, respectively).

Discussion and conclusions: Taken together, our data indicate that the vasodilatory actions of PCE are dependent on endothelial NO/cGMP and also involved in prostaglandin production. PCE and its active constituents, bakuchiol, isobavachalcone, isopsoralen and psoralen, caused dose-dependent inhibition of TRPC3 channels, indicating that those ingredients attenuate Phe-induced vasoconstriction.

Keywords: COX; Ethanol extract; hypotensive; nitric oxide; non-selective cation channel.

Figure 1.

(A, B) HPLC chromatograms of the standard solution (A) and 70% ethanol extract of P. corylifolia (B). (C) Chemical structure of the eight marker compounds of PCE: psoralen (1), isopsoralen (2), bavachin (3), corylin (4), psoralidin (5), isobavachalcone (6), bavachinin (7) and bakuchiol (8).

Figure 2.

Endothelium dependence of PCE-induced vasorelaxation of Phe-contracted, isolated rat aortic rings. (A, B) Different concentrations of PCE were added to endothelium-intact (A) or -denuded (B) aortic rings 30 min prior to performing cumulative Phe concentration–response studies. (C) PCE concentration dependent relaxation was measured in endothelium-intact and -denuded rat aortic rings pre-contracted with 10 µM Phe. (D) Effects of eNOS and GC inhibition on PCE-induced vasorelaxation of Phe-pre-contracted rat aortic rings. Rings were pretreated for 30 min with the non-specific NOS inhibitor, l-NAME (50 µM) or the GC inhibitor ODQ (20 µM) prior to pre-contracting with Phe. Data are means ± SEM of the relaxing effect, expressed as a percentage of the maximum Phe contraction (n = 6–8; *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 3.

The role of K ⁺ channels in PCE actions. (A) Effect of PCE on high K⁺ (60 mM)-pre-contracted, isolated rat aortic rings. (B) Effects of inhibitors of K_Ca channels (TEA) and KATP channels (glibenclamide) on PCE-induced relaxation of isolated aortic rings. Cumulative concentration–response curves for PCE following a 30-min incubation with vehicle (n = 4), 1 mM TEA (n = 6) or glibenclamide (n = 6) in aortic rings precontracted with 10 µM Phe are shown. Data are means ± SEM of the relaxing effect, expressed as a percentage of the maximum Phe contraction.

Figure 4.

Effects of PCE on Ca²⁺ entry. The effects of nifedipine (A), ML-204 (B), Pyr3 (C) and indomethacin (D) were tested. Aortic rings were pre-incubated with each chemicals for 30 min, and relaxation of Phe (10 µM)-pre-contracted rings by different concentrations of PCE was measured. Data are means ± SEM of the relaxing effect, expressed as a percentage of the maximum Phe contraction (n = 5–6; *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 5.

Effects of PCE on TRPC3 currents in HEK293 cells overexpressing M3 muscarinic receptors and human TRPC3 channels. Transiently transfected HEK293 cells were stimulated with 100 µM carbachol (A, B) or 100 µM OAG (C, D), followed by addition of 200 µg/mL or 50 µg/mL PCE, respectively. Representative traces showing the inhibitory effects of 200 or 50 µg/mL PCE on carbachol-stimulated (A) and OAG stimulated hTRPC3 (C) currents, respectively. Effects of carbachol and OAG are summarized in B and D, respectively. Bars represent the mean values of inhibition ± SEM (**p < 0.01, ***p < 0.001).

Figure 6.

Effects of constituents of PCE on TRPC3 currents in HEK293 cells. (A) HEK293 cells transiently transfected with TRPC3 were stimulated with 100 µM carbachol followed by addition of 200 µg/mL or 150 µg/mL of the indicated chemical constituent. (B) The concentration-dependent inhibition of TRPC3 currents by bakuchiol was studied. (C–E) Representative traces for treatment with vehicle (C), bakuchiol (D) and isobavachalcone (E) are shown. Bars represent the mean values of inhibition ± SEM (**p < 0.01, ***p < 0.001).

Figure 7.

Effects of constituents of PCE on Phe-induced contraction of endothelium-intact aortic rings. Each constituent of PCE (final concentration, 50 µM) was added to endothelium-intact aortic rings 30 min prior to performing cumulative Phe concentration–response studies. (A–E) Representative concentration–response curve of isobavachalcone (A), isopsoralen (B), bakuchiol (C), psoralen (D) and corylin (E) are shown. (F) Effects of accumulative treatment with constituents on Phe (1 µM)-induced aortic contraction. Bars represent the mean values of inhibition ± SEM (n = 5). *p < 0.05, **p < 0.01 versus control.