The Inhibition of TREK-1 K+ Channels via Multiple Compounds Contained in the Six Kamikihito Components, Potentially Stimulating Oxytocin Neuron Pathways

Abstract

Oxytocin, a significant pleiotropic neuropeptide, regulates psychological stress adaptation and social communication, as well as peripheral actions, such as uterine contraction and milk ejection. Recently, a Japanese Kampo medicine called Kamikihito (KKT) has been reported to stimulate oxytocin neurons to induce oxytocin secretion. Two-pore-domain potassium channels (K2P) regulate the resting potential of excitable cells, and their inhibition results in accelerated depolarization that elicits neuronal and endocrine cell activation. We assessed the effects of KKT and 14 of its components on a specific K2P, the potassium channel subfamily K member 2 (TREK-1), which is predominantly expressed in oxytocin neurons in the central nervous system (CNS). KKT inhibited the activity of TREK-1 induced via the channel activator ML335. Six of the 14 components of KKT inhibited TREK-1 activity. Additionally, we identified that 22 of the 41 compounds in the six components exhibited TREK-1 inhibitory effects. In summary, several compounds included in KKT partially activated oxytocin neurons by inhibiting TREK-1. The pharmacological effects of KKT, including antistress effects, may be partially mediated through the oxytocin pathway.

Keywords: KCNK2; Kampo; TREK-1; herbal medicine; kamikihito; oxytocin; paraventricular nucleus; supraoptic nucleus.

Figure 1

Effects of Kamikihito (KKT) on the TREK-1 activity in human TREK-1-expressing human embryonic kidney (HEK293) cells. The activation of human TREK-1 was measured using a FluxOR^TM potassium ion channel assay. The cells were pretreated with the vehicle or 1–300 µg/mL of KKT for 2 min and subsequently treated with a TREK-1 agonist, ML335. (A) Representative tracing of the mean of fluorescence intensity in randomly selected TREK-1-expressing cells (n = 15–30). (B) The activity of TREK-1 was estimated using the time to peak fluorescence following ML335 treatment (n = 45–90). The data are expressed as means ± standard errors of the means (SEMs). *** indicates p < 0.001 compared to the vehicle; Bonferroni’s multiple comparisons test followed a one-way ANOVA.

Figure 2

Effects of 14 crude herbal components of KKT on TREK-1 activity induced via ML335 in human TREK-1-expressing HEK293 cells. Each herbal component of KKT (10 μg/mL) or its vehicle was pretreated to the HEK293 cells, respectively. The cells were subsequently treated with the selective TREK-1 agonist, ML335 (10⁻⁵ M) (n = 15–287). The activity of TREK-1 was estimated using the time to peak fluorescence following ML335 treatment. Data are expressed as means ± standard errors of the means (SEMs). ** and *** indicate p < 0.01 and p < 0.001, respectively, compared to the vehicle; Bonferroni’s multiple comparisons test followed a one-way ANOVA.

Figure 3

Among the 14 components in KKT, Poria, Longan arillus, Bupleuri radix, Ziziphi semen Zingiberis rhizoma, and Angelicae acutilobae radix inhibited ML335-induced TREK-1 activation in human TREK-1-expressing HEK293 cells. The TREK-1 expressing cells were pretreated with 1, 10, or 30 µg/mL of Poria ((A) n = 30–165), Longan arillus ((B) n = 30–240), Bupleuri radix ((C) n = 30–240), Ziziphi semen ((D) n = 30–180), Zingiberis rhizoma ((E) n = 30–225), or Angelicae acutilobae radix ((F) n = 30–180), and subsequently treated with ML335 (10⁻⁵ M). The activity of TREK-1 was estimated using the time to peak fluorescence following ML335 treatment. The data are expressed as means ± standard errors of the means (SEMs). ** and *** indicate p < 0.01 and p < 0.001, respectively, compared to the vehicle; Bonferroni’s multiple comparisons test followed a one-way ANOVA.

Figure 4

The effects of six identified active herbal components of KKT on ML335-induced TREK-1 activity in human TREK-1-expressing HEK293 cells induced via numerous compounds in components that were effective in the channels. The cells stably expressing TREK-1 were pretreated with 10⁻⁵ M of each component of Poria (pachymic acid, ergosterol, and adenosine; (A) n = 59–150), Longan arillus (myristic acid, myristicin, tartaric acid, corilagin, and gallic acid; (B) n = 43–180), Bupleuri radix (saikogenin A, saikosaponin b2, saikogenin D, saikosaponin C, and α-spinasterol; (C) n = 45–195), Ziziphi semen (betulin, spinosine, magnoflorine, puerarin, swertisin, jujuboside A, jujuboside B, and betulinic acid; (D) n = 40–180), Zingiberis rhizoma (6-gingerol, 8-gingerol, 10-gingerol, 6-shogaol, 8-shogaol, 10-shogaol, zingerone, citral, paradol, and borneol; (E) n = 45–210), or Angelicae acutilobae radix (palmitic acid, linoleic acid, nicotinic acid, senkyunolide, senkyunolide H, levistolide A, bergapten, umbelliferone, feruloyltyramine, and xanthotoxin; (F) n = 44–165). They were subsequently treated with ML335 (10⁻⁵ M), and the activity of TREK-1 was estimated using the time to peak fluorescence following ML335 treatment. The data are expressed as means ± standard errors of the means (SEMs). *, **, and ***, respectively, indicate p < 0.05, p < 0.01, and p < 0.001 compared to the vehicle; Bonferroni’s multiple comparisons test followed a one-way ANOVA.