Dried SM was purchased from Boncho Co. Tanshinone I (T), tanshinone IIA (TA), and cryptotanshinone (CT) were purchased from Sigma-Aldrich; Merck KGaA. The plant sample was identified by Dr Yun Tai Kim according to a previous study (23) and a voucher specimen (#NP-0103) was deposited with the Research Group of Innovative Functional Foods, Korea Food Research Institute. Dried SM (600 g) was extracted with 70% ethanol (6,000 ml) for 4 h at 80°C. The extract was filtered through a membrane filter (0.45 µm; EMD Millipore). After removing the solvents via rotary evaporation, the remaining extracts were freeze-dried, yielding 33.6% of the dried weight (w/w). The freeze-dried extract powder (100 mg) was dissolved in 5 ml of methanol:dimethyl sulfoxide (DMSO; 1:1, v/v), before it was filtered through a 0.45-µm regenerated cellulose membrane filter (Sartorius Stedim) and diluted in methanol/DMSO (1:1, v/v) to a final concentration of 100 µg/ml prior to the injection of 10 µl of the solution for HPLC. Analytical HPLC was performed using a Jasco HPLC system (Jasco), comprising a PU-980 pump, an AS-950-10 autosampler, and an MD-2010 Plus multi-wavelength detector. Chromatographic separation was conducted at 30°C using a Waters Symmetry^® C18 (4.6×250 mm, particle size 5 µm) column with gradient elution using a mobile phase composed of 100% methanol (mobile phase A) and water containing 0.5% (v/v) glacial acetic acid (mobile phase B). The mobile phase change was carried out with a linear gradient system from 20:80 (mobile phase A:mobile phase B, v/v) to 80:20 (mobile phase A:mobile phase B, v/v) over 30 min with a 1 ml/min flow rate, before the samples were detected at 254 nm. Quantitative analysis was replicated four times. The regression equation and correlation coefficient (r²) of each standard curve were automatically determined using the Jasco HPLC system. The regression equations for T, TA, and CT were y=70277.0913x + 103899.0348 (R² was 0.99847), y=65183.0492× + 76181.5418 (R² was 0.99937), and y=85154.2548x + 84310.5739 (R² was 0.99981), respectively, indicating that a high linear correlation was achieved for all standard curves. The concentrations of T, TA, and CT were determined to be 2.61±0.131, 30.84±0.324, and 2.92±0.096 mg/g, respectively, using the peak area in the chromatogram and regression equation (Fig. 1).

A total of 82 mice (40 male and 42 female; 4-8 days old; weighing 2.0-2.3 g) of the Institute of Cancer Research (ICR) mice from the Samtako Bio Korea Co., Ltd. (Osan) were used for the ICCs experiments, 32 mice (male; 5-6 weeks old; weighing 20-25 g) for the intestinal transit rate (ITR) experiments, and 21 mice (male; 5-6 weeks old; weighing 20-25 g) for the intestinal hormones and protein expression experiments. ICCs experiments were completed within 12 h after culture and ITR experiments within 1 h. In the hormone measurement experiments, it took about 1 min to draw blood from the tail vein after immobilizing the mouse using a holder. Also, in the protein expression experiments, the mice were anesthetized and then sacrificed. Subsequently, the small intestine was removed. The whole process took ~3 min. All mice were housed in a specific pathogen-free laboratory environment under a controlled temperature (21–23°C) and humidity (50–60%) with day and night cycles (light on at 7:00 a.m. and light off at 7:00 p.m) and ad libitum access to normal diet and autoclaved water. During the study, indicators of the general condition of the mice were observed daily, such as fur brightness, food and water intake, defecation and behavior. Furthermore, body weight was measured every day. According to the cellular survival status and experimental repetitions, 135 ICR mice were sacrificed in this research. There were no other causes of mortality of mice other than execution for the experiment. Before execution, their skin condition and autonomous movements were monitored, and a warm environment was required. Euthanasia of mice was performed by decapitation after anesthesia. Mice were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg). Brain palsy was considered effective when spontaneous or stimulating movements and squeaks were not detected, and then, the mice were decapitated. The Institutional Animal Care and Use Committee at Pusan National University (approval no. PNU-2019-2462) approved all animal care and experiments (Busan, Republic of Korea). Additionally, animals were treated in accordance with the Guide for the Care and Use of Laboratory Animals (24).

The small intestine was removed, and the luminal contents were removed using Krebs-Ringer bicarbonate solution. As reported previously, mucosae were removed by sharp dissection and small tissue strips of intestinal muscle were equilibrated for 30 min in Ca²⁺-free Hank's solution. Then, cells were dispersed in an enzyme solution and cultured at 37°C in a 95% O₂−5% CO₂ incubator in a smooth muscle growth medium (Clonetics). Finally, ICCs were identified (25). The patch-clamp technique was used on ICCs that showed the network-like structures in culture.

We used the whole-cell patch-clamp method to record the effects of SM on the pacemaker potentials of ICCs. For the bath solution, 5 mM KCl, 135 mM NaCl, 2 mM CaCl₂, 10 mM glucose, 1.2 mM MgCl₂, and 10 mM HEPES (pH 7.4) were used and, for the pipette solution, 140 mM KCl, 5 mM MgCl₂, 2.7 mM K₂ATP, 0.1 mM NaGTP, 2.5 mM creatine phosphate disodium, 5 mM HEPES, and 0.1 mM EGTA (pH 7.2) were used. Patch-clamping was conducted using Axopatch I-D and Axopatch 200B amplifiers (Axon Instruments). Command pulses were applied using an IBM-compatible personal computer and pClamp software (version 6.1 and version 10.0; Axon Instruments). All experiments were performed at 30-31°C. In case of Ca²⁺ free experiments, the experiment was conducted after removing 2 mM of Ca²⁺ in bath solution. In case of Na⁺ 5 mM experiments, the experiment was carried out by changing 130 mM Na⁺ to 130 mM N-Methyl-D-glucamine (NMDG).

1,3-Dihydro-3,3-bis(4-hydroxyphenyl)-7-methyl-2H-indol-2-one (BHPI; selective ERα antagonist) and MA2029 (MTL receptor antagonist) were purchased from Tocris Bioscience and fulvestrant (ERα and ERβ antagonist), 4-[2-phenyl-5,7-bis(trifluoromethyl)pyrazolo[1,5-a]pyrimidin-3-yl]phenol (PHTPP; selective ERβ antagonist), guanosine 5′-O-(2-thiodiphosphate) (GDP-β-S; for the inactivation of G protein-binding proteins) and all other drugs were obtained from Sigma-Aldrich; Merck KGaA.

Evans blue (5%, w/v; 0.1 ml/kg) was administered through an orogastric tube 30 min after the intragastric administration of SM to ICR mice. After 30 min of Evans blue administration, the ITR was measured (the distance that Evans blue traveled from the pylorus to the most distal point).

After SM (0.5 g/kg) was fed once a day for 5 days, serum levels of gut hormones, such as MTL, substance P (SP), somatostatin (SS), and vasoactive intestinal polypeptide (VIP), were detected by radioimmunoassay using commercial kits purchased from Abbkine Scientific Co., Ltd.

After feeding SM (0.5 g/kg) for 5 days, small intestine samples were collected. The samples were prepared by incubation with in RIPA buffer containing protease and phosphatase inhibitor cocktail (Calbiochem). The total protein extracted from the samples was quantified using the Bradford method (Bio-Rad Laboratories, Inc.). An equal amount of protein (35 µg per lane) from the samples was resolved using 8% SDS-PAGE and then transferred to polyvinylidene difluoride membranes. The membranes were blocked with 5% skim milk in TBS + 0.1% Tween-20 for 1 h at room temperature and probed with the indicated antibodies. Anti-transmembrane protein 16A (TMEM16A; Abcam), anti-c-kit (Cell Signaling Technology, Inc.), anti-transient receptor potential melastatin 7 (TRPM7; Abcam), and anti-β-actin (Santa Cruz Biotechnology, Inc.) antibodies were used. An enhanced chemiluminescence reagent kit (Advansta) was used for detection. All other procedures were carried out as previously described (26).

Results are expressed as mean ± SEM. For multiple comparison analysis, we used one-way analysis of variance (ANOVA) with Bonferroni's post hoc comparison and when only two groups were compared, Student's t-test for unpaired data was used. For statistical analyses, we used Prism 6.0 (GraphPad Software Inc.) and Origin version 8.0 (OriginLab Corporation). P<0.05 was considered to indicate a statistically significant difference.

The presence of T, TA, and CT in SM was established by HPLC and their levels were quantified using calibration curves obtained from purchased standards (Fig. 1). Validation of the method used confirmed its reliability and stability.

ICC clusters generated spontaneous rhythmic contractions (21,22,27). Under the current clamp mode (I=0), ICCs generated pacemaker potentials with a mean resting membrane potential of −57.6±1.4 mV and mean amplitude of 24.6±1.1 mV (Fig. 2). SM (0.5-10 mg/ml) depolarized pacemaker potentials and decreased their amplitudes in a concentration-dependent manner (Fig. 2A-D). The mean degrees of depolarization by SM were 9.8±1.5 mV (P<0.01) at 0.5 mg/ml, 12.6±1.9 mV (P<0.01) at 1 mg/ml, 24.6±1.8 mV (P<0.01) at 5 mg/ml, and 42.2±1.9 mV (P<0.01) at 10 mg/ml (Fig. 2E, n=13) and the mean amplitudes were 13.3±0.8 mV (P<0.01) at 0.5 mg/ml, 6.9±0.7 mV (P<0.01) at 1 mg/ml, 2.9±0.9 mV (P<0.01) at 5 mg/ml, and 1.6±0.7 mV (P<0.01) at 10 mg/ml (Fig. 2F, n=13). These results show that SM dose-dependently depolarizes ICC pacemaker potentials.

SM is involved in the activation of ERs (12). Various ER agonists and antagonists were used to confirm the relevance of the ER in the response of ICCs to SM. Daidzein (ER agonist) depolarizes pacemaking activity (25). Pretreatment with fulvestrant (both an ERα and ERβ antagonist) blocked SM-induced effects (Fig. 3A). However, pretreatment with BHPI (ERα antagonist) did not (Fig. 3B). Additionally, pretreatment with PHTPP (ERβ antagonist) blocked SM-induced effects (Fig. 3C). The mean degrees of depolarization by SM were 4.7±0.7 mV (P<0.01) with fulvestrant, 23.6±1.7 mV with BHPI, and 4.1±0.8 mV (P<0.01) with PHTPP (Fig. 3D) and the mean amplitudes were 22.7±0.8 mV (P<0.01) with fulvestrant, 3.7±0.8 mV with BHPI, and 23.5±1.1 mV (P<0.01) with PHTPP (Fig. 3E). The results show that SM affects ICC pacemaker potentials via ERβ.

GDP-β-S was used to identify the relevance of the G protein in the response of ICCs to SM. GDP-β-S disables G protein-binding proteins (28,29). When GDP-β-S (1 mM) was injected intracellularly, SM had a slight pacemaker potential depolarization reaction (Fig. 4A). The mean degree of depolarization by SM was 4.3±0.8 mV (P<0.01) with GDP-β-S (Fig. 4B) and the mean amplitude was 21.6±1.3 mV (P<0.01) with GDP-β-S (Fig. 4C). Extracellular Ca²⁺ and Na⁺ have important roles in GI motility (30,31). To investigate the involvement of extracellular Ca²⁺ and Na⁺ in the SM-induced response in ICCs, we conducted experiments with no extracellular Ca²⁺ or Na⁺. Pre-treatment with no extracellular Ca²⁺ solution (Fig. 4D) or no extracellular Na⁺ (Fig. 4E) abolished the pacemaker potentials and inhibited the SM-induced response in ICCs. The mean degree of depolarization by SM was 1.1±0.7 mV (P<0.01) with no extracellular Ca²⁺ and 0.8±0.7 mV (P<0.01) with no extracellular Na⁺ (Fig. 4F). The results suggest that G protein and extracellular Ca²⁺ and Na⁺ influx are involved in the SM-induced response in ICCs.

ITR was measured 30 min after the administration of Evans blue in normal mice. In normal mice, the average ITR was 34.5±1.5% (Fig. 5A). After the administration of SM, ITR was 37.1±2.3% at 0.01 g/kg, 54.4±2.5% (P<0.01) at 0.1 g/kg, and 48.9±3.8% (P<0.01) at 1 g/kg (Fig. 5A). In comparison, Poncirus trifoliata Raf., which is another herbal medicine commonly used in GI motility studies, showed an average ITR of 55.7±2.2% (P<0.01), similar to the results of a past study (32). GI hormone levels in the mouse serum were evaluated by radioimmunoassay. The level of MTL in the GI was significantly elevated (Fig. 5B) but the levels of SP (Fig. 5C), SS (Fig. 5D), and VIP (Fig. 5E) showed no significant changes after SM administration. Furthermore, feeding SM (0.5 g/kg) and MA2029, MTL receptor antagonist together for 5 days reduced ITR (Fig. 5F). These results suggest that the SM increases ITR through an increase in MTL.

The TMEM16A (33,34) and TRPM7 (21) channels are involved in ICC activity. Furthermore, c-Kit is a transmembrane protein associated with the density of ICCs (35). Therefore, TMEM16A, TRPM7, and c-Kit may play a role in the treatment of GI motility disorders. After treatment with SM, the expression of TMEM16A, TRPM7, and c-Kit was evaluated using western blotting. The expression of TMEM16A and c-kit increased considerably after SM treatment (Fig. 6A). TMEM16A and c-kit expression significantly increased by 46.7 and 15.3%, respectively (P<0.01) after SM treatment (Fig. 6B and C). The expression of TRPM7 was almost unchanged (Fig. 6D). The results suggest that the SM-induced ITR increase is mediated through an increase in TMEM16A and c-kit expression.