Herbal components of Japanese Kampo medicines exert laxative actions in colonic epithelium cells via activation of BK and CFTR channels

Japanese Kampo medicines Junchoto and Mashiningan are mixtures of numerous herbal plant extracts and empirically known to exert laxative actions by stimulating fluid secretion in the colonic epithelium. However, it is unknown which and how the herbal components of these crude Kampo drugs are effective to stimulate ion effluxes causing fluid secretion. Here, we selected four herbal components of Junchoto and Mashiningan, Mashinin (MSN), Kyonin (KYN), Tonin (TON), and Daio (DIO), which are putatively laxatives, and examined their effects on the ion channel activity of human colonic epithelial Caco-2 cells. Patch clamp analyses revealed that MSN activated whole-cell current characteristics of the cystic fibrosis transmembrane conductance regulator (CFTR) channel, whereas KYN, TON, and DIO activated the large-conductance and voltage-activated K+ (BK) channel. Furthermore, electronic cell sizing showed that MSN induced secretory volume decrease (SVD) sensitivity to a CFTR blocker, whereas TON, KYN, and DIO induced SVD sensitivity to a K+ channel blocker. In conclusion, MSN and TON, KYN, and DIO promote fluid secretion from colonic epithelial cells by activating CFTR and BK channels. Thus, Japanese Kampo medicines, Junchoto and Mashiningan, exert anti-constipation actions by inducing KCl efflux through the combined actions of CFTR- and BK-stimulating herbal components.


Results
Four herbal components of Japanese Kampo laxatives stimulate fluid secretion by activating whole-cell ionic conductance in Caco-2 cells. Solute secretion from cells drives the transfer of osmotically obliged water, resulting in a cell volume reduction, which is called a secretory volume decrease (SVD) 10,11 . We therefore investigated herbal components in two types of Japanese Kampo medicines, JCT and MSG, which are widely used as laxatives, by measuring the SVD in human colonic epithelial Caco-2 cells. As shown in Fig. 1, exposure of Caco-2 cells to either MSN or TON resulted in a marked SVD by ~17% of the control cell volume at 30 min after administration. In contrast, applications of KYN and DIO induced smaller, but significant, SVD events by ~4% and ~3% of the control cell volume at 30 min after administration, respectively.
Because isosmotic fluid secretion associated with SVD must be driven by ion fluxes, we next examined the effects of these four herbal components on whole-cell membrane conductance in Caco-2 cells under a nystatin-perforated patch-clamp. As shown in Fig. 2a-d (top panels) and summarized in Fig. 2e, all of these herbal components gradually increased the membrane conductance, and the effect was partially reversible within 15 min after washout of herbal components (Fig. 2f). The current-voltage (I-V) relationships measured by a ramp clamp before and after exposure to herbal components were found to intersect at −70 to −75 mV for KYN, TON, and DIO (Fig. 2a,b,d, bottom panels), whereas the I-V curves intersected at around −40 mV for MSN (Fig. 2c, bottom panel). Because the reversal potentials of K + and Cl − are −82 and −41 mV, respectively, these results strongly suggest that MSN predominantly activates Cl − conductance, whereas KYN, TON, and DIO mainly activate K + conductance in Caco-2 cells.

Mashinin activates CFTR channels endogenously expressed in Caco-2 cells and heterologously expressed in HEK293T cells. Previous studies performed by short-circuit current measurements with
Ussing chambers suggested that MSG and JCT produce membrane currents sensitive to a CFTR blocker, CFTR inhibitor-172 (CFTR-inh) 12,13 , across the polarized human bronchial epithelial cell layer 9,14 . In addition, we previously showed that JCT facilitates intestinal Cl − secretion by activation of CFTR 10 . We therefore next investigated the effects of these four herbal components on CFTR activity in HEK/CFTR cells by conventional whole-cell recordings. As shown in Fig. 3, MSN activated Cl − currents (a, left panel) exhibiting a linear I-V relationship (a, right panel) under symmetrical Cl − conditions, whereas KYO, TON, and DIO did not activate whole-cell currents in HEK/CFTR cells (b). Similar Cl − currents were also activated by MSN in Caco-2 cells (Fig. 4a). The    Fig. 5h, KYN, TON, and DIO activated outwardly rectifying currents (curve b) with superimposing on background basal currents (curve a). The activated currents again exhibited voltage-dependent (depolarization-induced) activation and the reversal potential of around -70 to -80 mV. A well-known K + channel blocker, tetraethylammonium (TEA), was found to abolish activation of KYN-, TON-and DIO-induced whole-cell currents in Caco-2 cells at 5 mM (curve c). Under intracellular Ca 2+ chelation attained by introduction of BAPTA (5 mM) into the cells, KYN and TON were still effective to activate outwardly rectifying currents (Fig. 5e,f, bottom panels), but DIO did not activate the current (Fig. 5g, bottom panel), as summarized in Fig. 5i. These data indicate that KYN, TON, and DIO activate large-conductance and voltage-activated K + (BK) channels in Caco-2 cells, and that DIO and KYN or TON induce activation of the channels in a manner directly dependent on and independent of intracellular Ca 2+ , respectively.

Activation of CFTR and BK channels is responsible for fluid secretion in Caco-2 cells induced by herbal components of Japanese Kampo laxatives.
Finally, we determined whether induction of SVD by the herbal components is caused by their activating actions on CFTR and BK channels. As shown in Fig. 6, KYN-, TON-and DIO-induced SVD events in Caco-2 cells were virtually eliminated by TEA (a, b and d, +TEA), whereas the MSN-induced SVD was prominently inhibited by SQ and abolished by CFTR-inh (c, +SQ and +CFTR-inh). Intracellular Ca 2+ chelation completely inhibited DIO-induced SVD, but did not affect KYN-induced SVD and only partially suppressed TON-induced SVD (Fig. 6, +BAPTA).
Taken together, fluid secretion from Caco-2 cells in response to MSN is induced by activation of CFTR channels, and that in response to KYN, TON, or DIO is induced by BK channel activation. Thus, KCl efflux attained by parallel activation of BK and CFTR channels is responsible for fluid secretion induced by these herbal components of Japanese Kampo laxatives, Junchoto and Mashiningan.

Discussion
In response to stimulation by a variety of secretagogues, cell shrinkage is induced in association with fluid secretion in a number of secretory epithelial cell types (see Review 15 ), which is called the secretory volume decrease (SVD) 11 . In the present study, four herbal plant extracts, hemp fruit MSN, apricot kernel KYN, peach kernel TON, and rhubarb DIO, which are putatively known to exert laxative effects, provoked the SVD response in human www.nature.com/scientificreports www.nature.com/scientificreports/ colonic epithelial Caco-2 cells (Fig. 1). The MSN-induced SVD in Caco-2 cells was suppressed by a CFTR blocker and adenylate cyclase inhibitor, whereas the SVD responses to KYN, TON, and DIO were inhibited by a K + channel blocker (Fig. 6). In fact, patch-clamp experiments showed that MSN induced activation of CFTR-mediated anion currents in Caco-2 and CFTR-expressing HEK293T (HEK/CFTR) cells, and that KYN, TON, and DIO did not activate CFTR currents in both cell types but activated large-conductance and voltage-activated K + (BK) channel currents in Caco-2 cells (Figs 2-5). Taken together, MSN induces fluid secretion by stimulating CFTR anion channels through promotion of cAMP production, whereas KYN, TON, and DIO cause fluid secretion by activating BK channels in human colonic epithelial Caco-2 cells. In addition, among the four herbal extracts, MSN and TON exhibited much higher fluid-secreting activity estimated by the SVD extent than KYN and DIO. Intracellular Ca 2+ chelation by BAPTA (5 mM) introduced into the cells abolished DIO-induced activation of BK channels (Fig. 5g, bottom panel and i). In contrast, BK channels activated by KYN and TON were unaffected www.nature.com/scientificreports www.nature.com/scientificreports/ by this treatment (Fig. 5e,f, bottom panels and i). The BK channel is a homotetramer of the pore-forming α subunit and modulatory auxiliary subunits including four types of β subunits and four types of γ subunits (see Review 16 ). Recently, a Ca 2+ -independent activation mechanism of BK channels was found to be mediated by their γ-subunits including LRRD26, LRRD52, LRRD55, and LRRD38 (see Reviews 16,17 ). Thus, there is the possibility that KYN-and TON-induced activation is mediated by γ subunits of BK channels.
MSN is derived from hemp that has long been used as the raw material for fibres, foods, oils, and medicines. Hemp seed soft capsule (HSSC) ameliorates constipation by increasing the wet weight and water content of stool of loperamide-induced constipation model rats. This effect was suggested to be induced by activating CFTR, Ca 2+ -activated Cl − channels, Na + -K + -2Cl − (NKCC) cotransporters, and Cl − /HCO 3 − exchangers in colonic epithelia, based on pharmacological sensitivity of HSCC-induced short-circuited currents in rat colonic mucosa 18 . Thus, our present study suggests that the CFTR-activating effect of HSSC is, at least in part, caused by MSN. However, there are no studies of the effect on the intestinal transport of the peach kernel, from which TON is derived, although its extract exhibits an anti-cancer action in human colonic cancer HT-29 cells 19 .
For airway and intestine epithelial Cl − transport mediated by CFTR, TMEM16A (ANO1) is required for proper expression of CFTR in the plasma membrane 20 . TMEM16A is expressed in colonic Caco-2 cells and involved in glucose-induced enhancement of intestinal Cl − secretion caused by rotavirus infection 21 . However, under the present experimental conditions, Ca 2+ -activated Cl − currents were not observed. Thus, the activity of TMEM16A does not appear to be directly implicated in SVD stimulated by the herbal components, although there remains an indirect role of TMEM16A in plasmalemmal CFTR expression.
MSN consists of numerous chemical substances including fatty acids, such as linoleic acid, palmitic acid, oleic acid, and linolenic acid, as well as cannabicin A-G and vitamin E (https://www.genome.jp/kegg/). TON contains a much greater variety of chemical substances including fatty acids, such as palmitic acid, oleic acid, linolenic acid, and stearic acid, as well as steroids and amygdalin 22 . Thus, fatty acids, such as palmitic acid, oleic acid, and linolenic acid, which are commonly contained in MSN and TON, could be ruled out as CFTR-and BK-activating agents from the effective chemical components of MSN and TON, respectively. In fact, when we tested the change in CFTR currents in response to administration of 0.  Electrophysiology. Membrane currents of the cell were recorded at room temperature (22-27 °C) using the conventional and nystatin-perforated whole-cell modes of the patch-clamp technique with an Axopatch 200B patch-clamp amplifier (Axon Instruments/Molecular Devices, Union City, CA, USA). For whole-cell recordings, patch electrodes were prepared from borosilicate glass capillaries with an input resistance of 3-5 MΩ. Current signals were filtered at 5 kHz with a four-pole Bessel filter and digitized at 20 kHz. pCLAMP software (version 10.5.1.0; Axon Instruments/Molecular Devices) was used for command pulse control, data acquisition, and analysis. Data were also analysed using Origin software (OriginLab Corp., Northampton, MA, USA). For conventional whole-cell recordings, series resistance was compensated (to 70-80%) to minimize voltage errors. For CFTR current recordings, conventional whole-cell recordings were performed using the external solution containing (in mM) 110 CsCl, 2 CaCl 2 , 1 MgCl 2 , 5 glucose, and 10 HEPES (pH 7.4 adjusted by CsOH and osmolality adjusted to 310 mosmol/kg-H 2 O with D-mannitol), and the pipette solution containing (in mM) 110 CsCl, 2 MgSO 4 , 1 EGTA, 10 HEPES, 1 Na 2 ATP, and 15 Na-HEPES (pH 7.4 adjusted by CsOH, and osmolality adjusted to 300 mosmol/kg-H 2 O with D-mannitol). For whole-cell BK current recordings under intracellular Ca 2+ -chelating conditions, conventional whole-cell recordings were performed using isotonic Tyrode solution for external and pipette solutions, which contained (in mM) 55 K 2 SO 4 , 20 KCl, 1 MgCl 2 , 5 BAPTA, and 5 HEPES (pH adjusted to 7.4 by KOH and osmolality adjusted to 300 mosmol/kg-H 2 O with D-mannitol). For BK single-channel recordings, cell-attached recordings were obtained using high K + Tyrode solution for the external solution, which contained (in mM) 5 NaCl, 140 KCl, 1 MgCl 2 , 2 CaCl 2 , 10 D-glucose, and 10 HEPES (pH 7.4 adjusted by KOH and osmolality adjusted to 310 mosmol/kg-H 2 O with D-mannitol), and the pipette solution that contained (in mM) 55 K 2 SO 4 , 20 KCl, 1 MgCl 2 , and 5 HEPES (pH adjusted to 7.4 by KOH and osmolality adjusted to 300 mosmol/kg-H 2 O with D-mannitol). For nystatin-perforated whole-cell current measurements, cells were exposed to isotonic Tyrode solution and dialyzed against the pipette solution containing (in mM) 55 K 2 SO 4 , 20 KCl, 5 MgCl 2 , 0.2 EGTA, and 5 HEPES (pH adjusted to 7.4 by KOH and osmolality adjusted to 300 mosmol/kg-H 2 O with D-mannitol) supplemented with 200 μg/ml nystatin. Statistical evaluation. All data are expressed as means ± S.E.M. Data for each condition were acquired from at least three independent experiments. Statistical analyses were performed using the Student's t-test. P < 0.05 was considered as significant.

Data availability
All relevant data are included within the paper.