Garcinia buchananii stem bark extract and its bioactive constituents manniflavanone, GB-2 and buchananiflavanone attenuate intestinal inhibitory neuromuscular transmission

Abstract

Garcinia buchananii stem bark extract (GBB), commonly used for treating diarrhea in Africa, triggers ectopic aboral contractions, causing inhibition of propulsive motility in the colon ex vivo. To determine whether or not these effects were associated with decreased inhibitory neuromuscular transmission, the responsible constituent compounds, and mechanisms of action, we studied the effects of GBB and specific fractions and flavanones isolated from GBB on intestinal motility using pellet propulsion assays in guinea pig distal colons. In addition, microelectrode recordings were used to measure the effects on the inhibitory junction potentials (IJPs) in the porcine ileum and descending colon smooth muscle. Psychoactive Drug Screening Program secondary receptor functional assays were used to determine whether or not GBB and its constituent compounds act via purinergic (P2Y) and muscarinic receptors. GBB inhibited propulsive motility, but (2R,3S,2″R,3″R)-manniflavanone (MNF), (2R,3S,2″R,3″R)-GB-2 (GB-2) and (2R,3S,2″S)-buchananiflavanone (BNF), the main ingredients of GBB, did not affect motility. We discovered that, in the porcine descending colon, IJPs contained purinergic, nitrergic, and nonpurinergic nonnitrergic components. Furthermore, ileal IJPs were purely purinergic. GBB blocked all components of IJPs, while MNF and GB-2 inhibited purinergic IJPs only. BNF inhibited the purinergic and nonpurinergic components of IJPs. MRS2365, a Y1 (P2Y) agonist, did not evoke sustained membrane hyperpolarization in the presence of GBB. However, GBB, MNF, GB-2 and BNF did not affect P2Y or muscarinic receptors. In conclusion, inhibitory neuromuscular transmission in the porcine descending colon involves all components of IJPs. GBB decreases inhibitory neuromuscular transmission, likely by the actions of MNF, GB-2 and BNF. These effects do not involve P2Y or muscarinic receptors.

Introduction

Altered enteric nervous system (ENS) regulation plays a crucial role in the pathophysiology of gastrointestinal hypermotility and hypersecretion, which underlie the various forms of diarrheal illnesses and accompanying gastrointestinal pain (1, 2). Therefore, several investigators have proposed targeting the ENS to treat diarrhea and bowel pain (1, 2). Plant extracts widely used to treat diarrhea and gastrointestinal pain represent an important source of novel therapies targeting the ENS (3,4,5) and may be useful as alternatives to opiates, which mitigate diarrhea and gastrointestinal pain by inhibiting ENS synaptic neurotransmission (6). One such potential preparation is the extract from the stem bark of Garcinia buchananii Baker trees (GBB) used in sub-Saharan Africa to treat diarrhea and abdominal pain (4, 5).

We previously reported that GBB decreases intestinal motility by inhibiting synaptic transmission in the myenteric plexus, treats lactose-induced diarrhea and mitigates pain (5, 7, 8). However, how GBB inhibits gastrointestinal neurotransmission and propulsive motility and the bioactive compounds involved are unclear. Using bioactivity-guided fractionation, we showed that only two out of five GBB preparative thin-layer chromatography (PTLC) fractions (PTCL 1–5), PTLC1 and PTLC5, inhibited motility and mitigated lactose-induced diarrhea (7). We also showed that out of eight medium-pressure liquid chromatography (MPLC) GBB fractions (M1-M8) used in chemical isolation and identification, PTLC1 matched M4 and M5, and PTLC5 corresponded to M7. (2R,3S,2″ R,3″R)-manniflavanone (MNF) is the major component of M3 and GBB; (2R,3S,2″ R,3″R) GB-2 (GB-2) is the main ingredient of M4 and the second major component of GBB; and (2R,3S,2″S)-buchananiflavanone (BNF) is the main constituent of M5 (9,10,11). Furthermore, M3 and MNF do not inhibit motility but decrease calcium influx into smooth muscle cells by inhibiting L-type calcium channels (12). Taken together, these results suggest that the antimotility and antidiarrheal compounds in GBB are GB2, BNF and the unknown compounds in M7 (9,10,11). However, whether or not these flavanones inhibit gastrointestinal motility and neurotransmission is unclear.

In previous in vitro motility assays in the guinea pig (Cavia porcellus) distal colon, we observed that GBB causes contraction with notable ectopic contractions aboral to the pellets, suggesting that it blocks smooth muscle relaxation (Boakye and Balemba personal observations). Intestinal smooth muscle relaxation occurring aboral to the bolus is regulated by the inhibitory nonadrenergic noncholinergic neurotransmitters from the ENS (13,14,15,16). These neurotransmitters act by triggering membrane hyperpolarization and inhibitory junction potentials (IJPs) (13,14,15,16,17). This relaxation allows propulsive movements (15, 17, 18). Inhibitors of NO synthase, such as Nω-nitro-L-arginine methyl ester hydrochloride (L-NAME), inhibit nitrergic IJPs and increase muscular contractions (17, 19). In addition, it has been shown that daikenchuto, an aqueous extract of ginger, ginseng and zanthoxylum fruit, inhibits neuromuscular transmission in the guinea pig distal colon (20), suggesting that plant extracts may reduce intestinal motility by inhibiting neuromuscular transmission.

Therefore, we tested the hypothesis that GBB, its fractions M3, M4 and M5, and MNF, GB2 and BNF, the respective primary compounds in these fractions trigger ectopic aboral contractions, thereby causing decreased propulsive motility by decreasing inhibitory transmission to smooth muscle. We also examined the effects of GBB, M3, M4, M5, MNF, GB2 and BNF on various components of IJPs. We use secondary purinergic (P2Y) receptor functional assays to determine whether or not GBB, MNF, GB2 and BNF act via P2Y receptors. We also used porcine models (domestic pigs; Sus scrofa domesticus) to test our ideas, as the porcine gastrointestinal physiology closely matches that of humans (21). Previous studies have shown that, in the porcine intestine, IJPs are purely purinergic (13, 16, 22); subsequently, another goal was to characterize the IJPs in the porcine descending colon and elucidate potential sex differences. In testing these ideas, we discovered that GBB inhibits excitatory junction potentials (EJPs), which prompted us to use secondary muscarinic receptor functional assays to determine whether or not GBB, MNF, GB2 and BNF inhibit muscarinic receptors.

Methods

Animals, tissue acquisition and solutions

This study used the guinea pig distal colon for motility assays and the porcine (domestic pig; S. scrofa domesticus) ileum and descending colon to study IJPs because the porcine gastrointestinal physiology closely matches that of humans (21). In addition, it was cost-effective to use freely donated pig samples. The University of Idaho Animal Care and Use Committee approved the guinea pig studies under IACUC protocol numbers 2016-09 and 2017-38.

Sixteen 2- to 3-week-old male and female Hartley guinea pigs weighing 200–300 g were purchased from Elm Hill Labs, Chelmsford, MA, USA. They were housed in plastic cages with soft bedding. Animals were maintained at 23–24 °C on a 12-h light-dark cycle with access to food and water ad libitum as described previously (5, 23). The entire distal colon segment of approximately 55–60 cm in length was collected from each animal via midline laparotomy following isoflurane anesthesia and exsanguination. Each distal colon was cut into 3 equal segments approximately 12 cm long (oral, middle or aboral segments), which were kept in ice-chilled Kreb’s solution (mmol L⁻¹: NaCl, 121; KCl, 5.9; CaCl₂, 2.5; MgCl₂, 1.2; NaHCO₃, 25; NaH₂PO₄, 1.2; and glucose 8; all from Sigma, St. Louis, MO, USA; aerated with 95% O₂/5% CO₂) until use in motility assays.

Porcine ileum and descending colons were collected from butchers 5–15 min after animals were stunned by a gunshot to the head, followed by exsanguination. They were donated by C & L Lockers in Moscow, Idaho and Garfield Meat & Locker Plant, Garfield, Washington. Samples were transported (10–35 min) to the laboratory in ice-chilled HEPES (mmol L⁻¹: 134 NaCl, 6 KCl, 2.0 CaCl₂, 1.0 MgCl₂, 10 glucose, 10 HEPES; pH adjusted to 7.4 with NaOH).

The University of Idaho Animal Care and Use Committee approved the porcine studies; however, using donated samples does not require the IACUC protocol.

Propulsive motility assays in the guinea pig distal colon and intracellular microelectrode recording

The individual segments of the distal colon were pinned in a 50-ml silicone resin (Sylgard)-lined tissue bath, with continuous recycling of Kreb’s solution (cycling rate: 10 ml min⁻¹) maintained at 36.5 °C. Samples were equilibrated for 30–45 min. Next, following a 25-min baseline data recording, tissues were randomly treated with GBB or its fractions or compounds isolated from fractions for 30 min. Propulsive velocities were determined using a Gastrointestinal Motility Monitoring system (GIMM; Med-Associates Inc., Saint Albans, VT, USA) to film nail-polish-coated guinea pig pellets collected from animal cages a day before the experiment. Motility assays and pellet velocity calculations were performed using the GIMM software program (Med-Associates Inc.), as in previous studies (5, 23). Pellets were inserted in the oral end of the colon segments every five min throughout the 30-min equilibration period, the baseline recording period, and after drug application. GBB, its fractions and derivative compounds were delivered by Krebs and used to superfuse isolated colon segments in the organ bath. Treatment velocities were compared to baseline recordings. In the present study, pellet velocities were expressed as the distance the pellet was propelled through the segment and the amount of time taken, in millimeters per second. We randomized the treatments such that each drug was tested on the oral, middle and aboral segments.

Intracellular microelectrode recordings in porcine ileum and descending colon

Conventional microelectrode recordings (18, 24,25,26,27) were used to measure junction potentials in the porcine ileum and descending colon. Immediately after arriving in the lab, porcine ileum and descending colon samples (approximately 2 × 4 cm) were pinned while stretched in ice-chilled HEPES in a Petri dish line with Sylgard and dissected to remove the mucosa and submucosa and expose the inner circular muscle, as described previously (26). Muscularis externa (approximately 1.2 × 1.0 cm) preparations were then transferred and pinned while stretched in a recording chamber (3 × 2 cm) with the circular muscle layer up to record IJPs in circular smooth muscle cells, as described previously (13, 16, 26). To record IJPs in the longitudinal muscle layer of descending colon samples, tissues were pinned stretched in the recording chamber serosal surface up. Fat and the serosa were then gently teased off using fine scissors and forceps to reduce serosal thickness. Tissues were transferred onto a stage of an inverted Nikon Ti-S microscope (Nikon, Melville, NY, USA) and visualized using 10× and 20× objective lenses. Tissues were equilibrated at 36.5 °C by continuous superfusion with constantly aerated (95% O₂:5% CO₂) recirculating Krebs solution (approximately 10 ml/min) for 3 h.

Dissection and IJP recordings were performed in solutions containing nisoldipine and atropine (1.5 µM) to block muscle contractions. Steel stimulating electrodes were placed in the bathing solution, oral and aboral to the muscularis externa sample (distance between electrodes: 1.5 cm). IJPs were then evoked by electrical field stimulation (train duration, 500 ms; frequency, 10 Hz; pulse duration, 0.8 ms; and voltage, 100 V) to simulate the release of neurotransmitters from the myenteric plexus using established procedures (18, 24,25,26,27). Tissue stimulation was repeated at five-minute intervals to allow for re-equilibration. The membrane voltage changes were measured by impaling smooth muscle cells with fine glass electrodes (tip resistance 90 to 130 mΩ) filled to the shoulder with 1.0 mol L⁻¹ KCl and filling the rest of the glass electrode with 2.0 mol L⁻¹ KCl. A Narishige MX-1 Micromanipulator (Tokyo, Japan) was used for coarse and fine movements to impale circular smooth muscle cells. Stimulation was accomplished using a Grass S88 Dual Output Square Pulse Stimulator and GRASS S1U5 stimulus isolation unit (Grass Instruments Co., Quincy, MA, USA). Electrical signals were acquired and analyzed using a SIGLENT SDS 1202X-E digital storage oscilloscope (Siglent, Solon, OH, USA), an IX2-700 Dual Intracellular Preamplifier (Dagan Corporation, Minneapolis, MN, USA) and PowerLab 8/30 data acquisition device and the “LabChart” software program, version 5.01 (ADInstruments, Colorado Springs, CO, USA). We actively sought smooth muscle cells exhibiting a resting membrane potential of 47–55 mV. Intracellular microelectrode recording in circular smooth muscle cells in the muscularis externa from the porcine ileum produces rhythmic electrical membrane potential fluctuations called slow waves (12, 13, 16). Therefore, to record IJPs in the ileum, single-pulse transmural field electrical stimulations were triggered during the end of a slow wave, as the smooth muscle cell membrane potential returned to resting potential.

GBB fractions and derived biflavanones and control drugs

G. buchananii stem bark was collected in October 2006 and September 2009 and processed into powder, which was then used to prepare the extract, as described previously (5, 23). GBB was prepared by mixing 0.5 g stem bark powder in 100 ml Krebs, stirring for 30 min, and then filtering particles from the solution, as described previously (5, 23). GBB fractions M3 (41 mg), M4 (6 mg), M5 (10 mg) and M7 (8 mg) were studied based on the weight (in milligrams) each fraction contributes to 100 mg of M1-M8 combined (9, 28). The compounds isolated from GBB were MNF (from M3), GB-2 (from M4) and BNF (from M5). They were tested at concentrations ranging from 0.1 µM to 2,000 µM. Fractions and compounds were first dissolved in absolute ethanol to prepare stock solutions for mixing in 100 ml oxygenated Krebs bathing solution. The procedures for isolating the fractions and isolated compounds were previously published (9, 28).

The commercial compounds used in the control experiments were quercetin 3-O-β-D-glucoside and (±)-hesperetin (Sigma-Aldrich, Cleveland, OH, USA) (200 µM each). Compounds used to characterize IJPs or determine the mechanism of action were tetrodotoxin (TTX; 1 µM), Nω-nitro-L-arginine methyl ester hydrochloride (L-NAME; 200 µM), (±)-(E)-4-ethyl-2-[(Z)-hydroxyimino]-5-nitro-3-hexen-1-yl-nicotinamide (NOR-4; 20 µM), apamin (1 µM), α,β-methylene adenosine 5′-triphosphate lithium salt (α,β-methylene ATP; 10 µM), [[(1R,2R,3S,4R,5S)-4-[6-amino-2-(methylthio)-9H-purin-9-yl]-2,3-dihydroxybicyclo[3.1.0]hex-1-yl]methyl] diphosphoric acid mono ester trisodium salt (MRS2365; 1 µM) and 2′-deoxy-N6-methyladenosine 3′,5′-bisphosphate tetrasodium salt (MRS 2179; 3 µM). Other compounds were (1R,2S,4S,5S-4-[2-Iodo-6-(methylamino)-9H-purin-9-yl]-2-(phosphonooxy)bicyclo[3.1.0]hexane-1-methanol dihydrogen phosphate ester tetraammonium salt (MRS2500;1 µM), forskolin (10 µM), N-Cyclohexyl-N-[2-(3,5-dimethyl-pyrazol-1-yl)-6-methyl-4-pyrimidinamine (CyPPA) and [D-p-Cl-Phe6,Leu17]-VIP (1 µM).

Commercial drugs were solubilized in water, ethanol or dimethyl sulfoxide (DMSO) to prepare stock solutions according to the manufacturer’s recommendations. In previous studies (5, 12, 23), vehicles (ETOH and DMSO) were shown to have no effect on guinea pig propulsive motility at >1:1,000 dilution, which is the concentration used in the present study. All drugs were applied to tissues in the recording chamber using recirculating Krebs solution (100 ml). When an individual tissue sample was tested with multiple drugs, tests were separated by a one-hour washout with nonrecycling Krebs to restore IJPs to their original amplitude and duration. Tissues treated with apamin, GBB, L-NAME, MRS2500 and high concentrations of biflavanones (≥200 µM) were not washed for additional experiments. Information on drug sources can be found in the supplementary materials.

Psychoactive Drug Screening Program (PDSP) secondary P2Y receptor and muscarinic functional assays

The P2Y1 receptor is the main receptor subtype mediating nonadrenergic, noncholinergic inhibitory responses in the gut. However, P2Y2, P2Y6 and P2Y11 are expressed on smooth muscle cells and thought to have a role in smooth muscle relaxation (15, 17). Therefore, we determined whether or not GBB, MNF, GB-2, and BNF inhibit intestinal muscle relaxation by inhibiting P2Y1, P2Y2, P2Y4, P2Y6 and P2Y11 receptors using secondary P2Y receptor functional assays. In addition, cholinergic activation of muscarinic receptors—in particular M2 and M3—triggers smooth muscle depolarization and the discharge of EJPs (29,30,31). We therefore investigated whether or not GBB and isolated molecules affect smooth muscle muscarinic receptors by inhibiting muscarinic receptor subtypes M2 and M4.

Primary receptor binding assays were used to determine whether or not GBB and GB-2 act on M3 macrophages. This was done via the PDSP (Roth Laboratory at the University of North Carolina at Chapel Hill, Chapel Hill, NC, USA). The assays performed were the antagonist and agonist secondary P2Y receptor and muscarinic receptor functional assays at a final concentration of 10 µM using The National Institute of Mental Health (NIMH) PDSP protocols, described in the assay protocol book version III (32). (±)-Hesperetin and (±)-naringenin were used as control drugs. Data are represented as the mean % inhibition or agonist activity. The results were considered significant at 30% agonist activity, 50% antagonist activity or negative (-) below these values.

Data and statistical analyses

Pellet propulsion velocities were computed as described previously (5, 23). Statistical analyses were performed with a one-way analysis of variance and unpaired Student’s t-test using the GraphPad Prism 6.07 software program (Boston, MA, USA). Differences were considered statistically significant at a probability value of less than 0.05 (P<0.05), and levels are indicated by asterisks, where *P is *P <0.05, ** P is <0.01, *** P is <0.001, and **** P is <0.0001. All data are given as the means ± standard error of the mean (SEM).

Results

M4, M5 and M7 fractions of GBB inhibit colon motility

GBB, PTLC1 and PTLC5 have antidiarrheal and antimotility properties (7, 23). PTLC1 corresponds to MPLC GBB fractions M4 and M5, while PTLC5 matches M7 (9,10,11). We studied whether or not M4, M5 and M7 as well as MNF, GB2 and BNF, the main compounds found in M3, M4 and M5, respectively, have antimotility effects by analyzing pellet propulsion in isolated guinea pig distal colons (5, 23). The pellet propulsion velocity in the isolated guinea pig distal colon was 2.6 ± 0.5 mm sec^–1 (Fig. 1A). GBB reduced the pellet propulsion after 3 to 10 min. M4 reduced the pellet velocity to 0.86 ± 0.37 mm/sec (n=4; *P<0.001), M5 reduced the pellet velocity to 1.01 ± 0.14 mm/sec (n=4; *P<0.005), and M7 reduced the pellet velocity to 0.77 ± 0.03 mm/sec (n=3; *P<0.005) after 10 min. In contrast, GB2 and BNF did not affect the pellet propulsion (Fig. 1B). M3 did not affect the pellet velocity even after 30 min (2.708 ± 0.273 mm/sec, n=5). Similarly, MNF, the major constituent of M3, did not affect the pellet velocity even after 30 min (2.423 ± 0.5066 mm/sec, n=4). However, (±)-hesperetin inhibited pellet propulsion.

Fig. 1.

GBB and its three fractions M4, M5 and M7 inhibit pellet propulsion in isolated guinea pig distal colon. A. Summary, data showing that superfusing GBB (0.5 g), M4 (6.0 mg), M5 (10.0 mg) and M7 (8.0 mg) per 100 ml Krebs in the tissue bath reduced the pellet propulsion velocity in the guinea pig distal colon after 10 min, but M3 (41.0 mg) did not affect motility (n=4–8 animals). B. MNF, GB-2 and BNF, biflavanones isolated from M3, M4, and M5, respectively, did not reduce pellet propulsion velocity in the guinea pig distal colon after 20 min. In contrast, (±)-hesperetin inhibited pellet propulsion (n=3–4 animals). C–G, Pictures demonstrating that, compared to untreated control, 0.5 g GBB, 6.0 mg M4, 10.0 mg M5, 8.0 mg M7 and a mixture of 2.0 mg M4 + 3.3 mg M5 + 2.7 mg M7 triggered ectopic contractions (arrows) in guinea pig distal colon aboral to pellets. M4, M5 and M7 elicited similar contractions. Analyses were performed by a one-way analysis of variance. Tukey’s post hoc test was used to correct for multiple comparisons of data from more than two groups. ***P<0.001; ****P<0.0001. GBB: Garcinia buchananii stem bark extract; BNF: buchananiflavanone; MNH: manniflavanone.

GBB, M3-M5 and MNF, GB-2 and BNF—biflavanones isolated from M3-M5—diminish inhibitory neuromuscular transmission in porcine intestine

In this study, we observed that GBB, M4, M5 and M7 caused ectopic aboral contractions (Fig. 1C–G). Therefore, we studied whether or not GBB; M4, M5 and M7; GB-2 and BNF inhibit IJPs in the porcine ileum and descending colon.

In the ileum, IJPs were triggered during the intervals between slow waves (Fig. 2A, B). Tetrodotoxin (TTX) was used to determine whether or not IJPs are mediated by neurotransmitters (18, 19). TTX (1.0 µM) abolished IJPs after 3 to 5 min (Fig. 2C, D). IJPs are biphasic electrical events consisting of a fast component triggered by purines—mainly ATP acting via P2Y1 receptors—and a slow component primarily triggered by nitric oxide (17). Vasoactive intestinal peptide (VIP) and pituitary adenylate cyclase-activating polypeptide (PACAP) play a smaller role in triggering slow IJPs (15, 18). In the porcine intestine, IJPs have been studied in the jejunum (13), ileum (16) and ascending colon (22) and were found to be purinergic. To test whether or not purines stimulate IJPs in the porcine ileum by activating P2Y1 receptors, we used MRS 2179, a P2Y1 receptor inhibitor (15, 17). MRS 2179 (3.0 µM) inhibited IJPs after 3 to 7 min. P2Y1 receptor activation elicits the opening of apamin-sensitive small-conductance Ca²⁺-activated K⁺ channels (SK channels) (14, 16). We studied the effect of 250 nM apamin and observed that it irreversibly blocked IJPs after 5 min (Fig. 2E–I).

Fig. 2.

Traces of membrane potential recordings obtained by intracellular microelectrodes from circular smooth muscle cells of the porcine ileum and summary data showing that in the circular muscle of the porcine ileum, IJPs are purely purinergic. A, B. Rhythmic slow waves and an IJP elicited in the interval between slow waves under basal conditions. C, D. Compared with baseline IJPs, 1 µM TTX blocked IJPs after 3–5 min. E–H. IJPs were blocked by the P2Y1 receptor antagonist MRS2179 (3 µM; G-F) and the Ca²⁺-activated small conductance K⁺ channel selective blocker apamin (250 nM; G, H). I. Summary data showing that IJP amplitudes were significantly reduced by TTX, apamin and MRS2179 but were not affected by 200 µM L-NAME or 10 µM NORI-4 (n=5–6 animals; one-way analysis of variance). Furthermore, L-NAME did not affect the duration of IJPs (J; baseline: 1.125 ± 0.091 sec vs. 200 µM L-NAME: 1.081 ± 0.075 sec, @ n=4 animals; unpaired Student’s t-test, P=0.717). Summary data were generated from IJPs evoked by 100 V stimuli. ***P<0.001; ****P<0.0001. IJP: inhibitory junction potential; TTX: tetrodotoxin.

Next, we evaluated the role of nitric oxide (NO) using L-NAME, a nitric oxide synthase inhibitor (15, 17). L-NAME (200 µM) did not affect IJPs in the ileum. Similarly, the NO donor (20 µM NOR-4) did not affect IJPs (Fig. 2I, J). We previously found that female mice had larger IJP amplitudes in the distal colon than male mice, suggesting the presence of sex differences in inhibitory neurotransmission (26). To examine whether or not IJPs were influenced by sex, we compared IJPs elicited in castrated male to female pigs. Circular smooth muscle cells of the ileum from neutered male and female pigs had similar resting membrane potentials and IJPs amplitudes and durations (Supplementary Fig. 1A–C).

An analysis of the effects of GBB, M4, M5, and M7 on the discharge of IJPs in the circular muscle of the porcine ileum revealed that GBB (0.5 g/100 ml Krebs) significantly reduced IJP amplitudes after 3 to 5 min and abolished IJPs after 10 to 20 min. Both M4 (6.0 mg/100 ml Krebs) and M5 (10.0 mg/100 ml Krebs) significantly inhibited the discharge of IJPs after 10 to 20 min. M7 (8.0 mg/100 ml Krebs) did not affect IJPs (Fig. 3 A–I). Therefore, we investigated the effects of GB-2 and BNF (200 µM each), the most abundant compounds in M4 and M5, respectively (9,10,11), on ileal IJPs. MNF, a biflavanone isolated from M3 (9); (±)-hesperetin, a flavanone monomer; and quercetin 3-O-β-D-glucoside, a flavonoid (19) (200 µM each), were used in control experiments. GB-2 significantly reduced the IJP amplitude after 10 min, and its effect on IJPs increased with exposure time. Similarly, BNF inhibited IJPs after 10 min, and the effect increased in the same manner (Fig. 4A, B). Of note, GB-2 and BNF appeared to have additive effects (Fig. 4C–E). To our surprise, at increased concentrations, MNF inhibited IJPs after 10 to 20 min. Similar to GB-2 and BNF, the effect of MNF-3 on fast IJPs was concentration-dependent (Fig. 4F). (±)-Hesperetin did not affect IJPs after 20 min.

Fig. 3.

Traces of slow waves and IJPs from circular smooth muscle cells in porcine ileum and summary data showing that GBB blocked IJPs, while its fractions M4 and M5 reduced the amplitudes of ileal IJPs. A, B. Demonstration of slow waves and a baseline IJP between slow waves and blockade of IJPs by GBB (0.5 g/100 ml Krebs) after 10 min. C–E. Compared to baseline IJPs, M4 (6 mg/100 ml Krebs) and M5 (10 mg/100 ml Krebs) reduced IJPs but did not block IJPs after 20–30 min. G, H. M7 (8 mg/100 ml Krebs) did not affect IJPs. E. Summary data showing that 1 µM TTX and GBB blocked IJPs after 5–10 min. M4 and M5, the GBB fractions with antimotility effects, significantly reduced IJPs but did not block the discharge of IJPs after 10 or even 20 min (not shown), but M7 had no effect on IJPs. n=3–5 animals; one-way analysis of variance. *P<0.05, ***P<0.001; ****P<0.0001. GBB: Garcinia buchananii stem bark extract; TTX: tetrodotoxin.

Fig. 4.

Traces of slow waves and IJPs from circular smooth muscle cells and summary data demonstrating that MNF, GB-2 and BNF, the major biflavanone compounds of GBB, inhibit IJPs in the porcine ileum. A–D. Demonstration of slow waves, baseline IJP between slow waves (A&C), and inhibition of IJPs by BNF (B) and a mixture of GB-2 + BNF (200 µM, each; D) after 25 min. E. Summary data showing that, compared with baseline IJPs, MNF, GB-2, BNF and GB-2 + BNF significantly inhibited IJPs after 20 min. GB-2 + BNF appeared to have additive effects but still had much less potency than GBB. Treatment with 200 µM (±)-hesperetin did not affect IJPs (n=4–7; unpaired Student’s t-test). F. shows the concentration–response relationship curves for MNF, GB-2 and BNF. *P<0.05, **P<0.01; ***P<0.001; ****P<0.0001; ns, not significant. IJP: inhibitory junction potential; BNF: buchananiflavanone; MNF: manniflavanone.

Taken together, our results support previous findings (16) that, in porcine ileum, IJPs elicited by single pulse stimuli are triggered by the release of purines from the ENS, acting via P2Y1 receptors, and activation of apamin-sensitive Ca²⁺-dependent K⁺ channels.

In summary, GBB inhibits purinergic IJPs, probably by the combined actions of MNF, GB-2 and BNF, even though MNF, GB-2 and BNF each have weaker inhibitory effects than GBB. (±)-Hesperetin does not affect intestinal purinergic IJPs.

Effects of GBB on P2Y1 receptor activation by exogenous agonists

Given that purines elicit fast IJPs by activating P2Y1 receptors (18), the P2Y1 receptor agonist MRS2365 (2 µM) (27) was used to determine whether or not GBB affects P2Y1 receptor-mediated signaling in the porcine ileum. MRS2365 induced TTX-insensitive membrane hyperpolarization of circular smooth muscle cells. Pretreating tissues with GBB for 5 min reduced the effect of MRS2365, suggesting that GBB has compounds that antagonize the actions of MRS2365 (Fig. 5A; Supplementary Fig. 2A, B). To test this idea further, α,β-meATP, a stable analog of ATP (14, 33), was applied to tissues pretreated with GBB for five min. α,β-meATP (10 µM) caused TTX-insensitive, transient membrane depolarization (Fig. 5B). The effects of α,β-meATP were reduced by GBB. GBB itself caused membrane hyperpolarization (−5.6 ± 1.6 mV; n=5). These results suggest that GBB has molecules that can inhibit P2Y1 receptors, apamin-sensitive Ca²⁺-dependent K⁺ channels or other critical steps in P2Y1 signaling, such as cAMP-dependent activation of Ca²⁺ release from intracellular ryanodine-dependent stores (33).

Fig. 5.

GBB reduces MRS 2365- and alpha methylene ATP-induced membrane depolarization, suggesting that it inhibits purinergic IJPs via direct effects on ileal smooth muscle by inhibiting the P2Y1 receptor signaling pathway. Bar (A) and line (B) graphs demonstrate that GBB hyperpolarizes smooth muscle cells. In addition, GBB decreases the level of membrane hyperpolarization elicited by MRS 2365 (B; asterisk) and alpha methylene ATP as well as the transient depolarization caused by alpha methylene ATP in the porcine ileum. The number of animals (n)=4–7. Analyses were performed by a one-way analysis of variance and unpaired Student’s t-test. *P<0.05. GBB: Garcinia buchananii stem bark extract.

Smooth muscle cells in the circular muscle of the porcine descending colon generate IJPs with typical purinergic, nitrergic and nonnitrergic, nonpurinergic components

The findings that GBB and derived compounds inhibit purinergic IJPs in porcine ileum prompted us to determine whether or not they inhibit all components of IJPs. In the human colon, electrical field stimulation causes an IJP with two phases: a purine-dependent, apamin-sensitive fast component followed by an L-NAME-sensitive sustained component (27). We hypothesized that the porcine descending colon has these mixed IJPs and is thus appropriate to test whether or not GBB and derivative molecules affect all components of mixed IJPs.

Therefore, we first analyzed the characteristics of IJPs in porcine descending colon circular muscle to identify whether or not it had fast purinergic IJPs and slow nitrergic and nonnitrergic nonpurinergic IJPs (18, 27). Circular smooth muscle cells in the descending colon discharged spontaneous membrane depolarization and hyperpolarization but lacked slow waves. They generated IJPs with a prolonged duration, which were abolished by TTX after 3 to 5 min (Fig. 6A, B). To isolate the purinergic and nitrergic components of these IJPs, we measured the amplitudes and duration of IJPs in the presence of MRS2179 (3 µM), L-NAME (200 µM) or a mixture of MRS2179 + L-NAME. MRS2500 (1 µM) and apamin (250 nM) were also used to characterize the purinergic IJPs, while the VIP antagonist [D-p-Cl-Phe⁶,Leu¹⁷]-VIP (1 µM) was used to determine the contribution of VIP (13, 18, 27). MRS 2179 and MRS2500 reduced the amplitudes of the IJPs but did not block the IJP or reduce the IJP durations (Fig. 6C, D). L-NAME reduced the duration of IJPs without affecting the amplitude. A mixture of L-NAME + MRS 2179 reduced the amplitude and duration of the IJPs but did not block the IJPs (Fig. 6E, F). A mixture of L-NAME + MRS 2179 + [D-p-Cl-Phe6,Leu17]-VIP reduced the amplitude and duration of IJPs, almost blocking them. The resting membrane potentials of smooth muscle cells in the circular muscle layer of the descending colon of neutered male and female pigs were similar. IJPs of both male and female descending colons had similar amplitudes and durations (Supplementary Fig. 3A–C). Descending colon IJPs had greater amplitudes and durations than ileal IJPs (Supplementary Fig. 4A, B).

Fig. 6.

Traces and summary data demonstrating that circular smooth muscle cells of the porcine descending colon generate mixed IJPs with purinergic, nitrergic and nonpurinergic nonnitrergic components. A. Representative recordings of the resting membrane potential (green) and an IJP (red) obtained under basal conditions. B. Compared with baseline IJP (red), 1 µM TTX (violet, B) blocked IJPs after 2–5 min. C, D. The nonpurinergic components were pharmacologically isolated by inhibiting the purinergic IJP with 3 µM MRS2179 or 1.0 µM MRS2500 (blue traces) for 10 min. E, F. The nitrergic component was defined using 200 µM L-NAME to block nitric oxide synthase in the absence (E, yellowish green trace) or presence of 3 µM MRS2179 (F, cyan). G. Isolation of the nonpurinergic and nonnitrergic (cyan trace) IJPs using a mixture of 200 µM L-NAME and 3 µM MRS2179 for 20 min followed by application of the VIP antagonist [D-p-Cl-Phe6,Leu17]-VIP (1 µM) for 20 min revealed the nonpurinergic nonnitrergic component mediated by VIP. This mixture of drugs did not block IJPs entirely, suggesting that other mediators of nonpurinergic nonnitrergic smooth muscle relaxation, such as pituitary adenylate cyclase activating peptide, may have a role in eliciting IJPs in the porcine descending colon. H. Summary data showing that 1 µM TTX blocked IJPs, MRS2179, MRS2500 and MRS 2179 + L-NAME mixture significantly reduced the amplitudes of IJPs after 10 min. L-NAME did not affect IJP amplitudes (n=4–7 animals; one-way analysis of variance). I, The L-NAME and MRS 2179 + L-NAME mixture significantly decreased the duration of IJPs in samples from males and females, but MRS2179 and MRS2500 did not affect the duration of IJPs (n=5–7 animals; one-way analysis of variance). **P<0.01; ***P<0.001. IJP: inhibitory junction potential; TTX: tetrodotoxin.

In summary, our results show for the first time that circular smooth muscle cells of the porcine descending colon produce mixed IJPs. These IJPs have larger amplitudes and longer durations than ileal IJPs.

GBB and BNF inhibit the nitrergic, nonnitrergic, nonpurinergic components of IJP in the circular muscle of the porcine descending colon

We determined whether or not GBB inhibits all components of an IJP in the porcine descending colon and observed that GBB rapidly reduced the amplitude of IJPs and blocked IJPs after 20 min (Fig. 7A–D). To identify what compounds in GBB inhibit the nitrergic and nonnitrergic nonpurinergic components of IJP, we studied the effects of MNF, GB-2 and BNF on the components of IJPs in the descending colon.

Fig. 7.

GBB inhibits all components of IJPs in the circular muscle of the porcine descending colon. A, B. Traces showing that, compared to baseline (red), GBB (0.5 g/100 ml Krebs) inhibits the fast purinergic and nonpurinergic components of IJPs in circular muscle cells of the porcine descending colon (black). B. Application of the purinergic IJP inhibitor MRS2179 (3 µM) for 20 min (blue) was used to block the purinergic IJPs (compared with baseline; red). GBB was then applied in the presence of 3 µM MRS2179 to confirm that it inhibited the discharge of the nitrergic and nonpurinergic nonnitrergic components of IJPs (black trace). Asterisks show that GBB inhibited excitatory junction potentials. C, D. Summary data showing that GBB significantly reduced the amplitude of IJPs much more rapidly (C) than the IJP durations (D) and blocked IJPs after 20 min (n=7 animals; analysis by a one-way analysis of variance). *P<0.05, **P<0.01; ***P<0.001; ****P<0.0001. GBB: Garcinia buchananii stem bark extract.

Each compound was tested in the presence of 3 µM MRS2179 or a mixture of 3 µM MRS2179 + 200 µM L-NAME to isolate the nitrergic and nonpurinergic nonnitrergic components, respectively (15, 27). (±)-Hesperetin and quercetin 3-O-β-D-glucoside were used as controls. Application of the purinergic IJP inhibitor MRS2179 (3 µM) for 20 min (blue) was used to block the purinergic IJPs (compared with baseline: red). GBB was then applied in the presence of 3 µM MRS2179 to confirm that it inhibited the discharge of the nitrergic and nonpurinergic nonnitrergic components of IJP (black trace). GBB significantly reduced the amplitude of the IJPs after 5 min and blocked the IJPs after 20 to 25 min. Asterisks show that GBB inhibits EJPs (Fig. 7B). MNF and GB-2 (at 500 µM) did not affect the nitrergic and nonpurinergic components after 30 min. BNF (500 µM) significantly reduced the amplitude of the nonpurinergic components of IJPs after 20 to 25 min (Fig. 8A–E). (±)-Hesperetin and quercetin 3-O-β-D-glucoside (not shown) did not affect the amplitude of the nonpurinergic components of IJPs after 20 min. However, (±)-hesperetin increased the duration of IJPs.

Fig. 8.

GB-2 and MNF did not affect the nonpurinergic IJPs, but BNF reduced the amplitude of nonpurinergic IJPs. A–C. Traces showing the effect of 500 µM @ GB-2 (A; cyan), BNF (violet) and MNF (yellowish green) on the nonpurinergic IJPs isolated from baseline IJPs (red) by application of 3 µM MRS2179 (blue traces) for 20 min. D, E. Summary data demonstrating that BNF decreased the nonpurinergic IJPs, while MNF, GB-2 and 500 µM (±)-hesperetin had no significant effects. In contrast, only (±)-hesperetin significantly increased the duration of IJPs (n=5–7 animals; one-way analysis of variance). **P<0.01; ***P<0.001. BNF: buchananiflavanone; MNF: manniflavanone.

Taken together, our results suggest that GBB inhibits purinergic, nitrergic and nonpurinergic nonnitrergic IJPs. BNF inhibits the nonpurinergic components of IJP as well, albeit with significantly less potency than GBB. In contrast, by prolonging the duration of IJPs, (±)-hesperetin was able to augment the nitrergic and nonnitrergic nonpurinergic components of IJPs in the porcine descending colon.

GBB decreases the effect of exogenous P2Y1 receptor agonists in the circular muscle of the porcine descending colon

The activation of P2Y1 receptors by 1 µM MRS2365 and 10 µM α,β-meATP was performed in the presence of GBB as described above. GBB alone hyperpolarized the resting membrane potential of smooth muscle cells. MRS2365 induced rapid membrane hyperpolarization of circular smooth muscle cells. Pretreating tissues with GBB for five min significantly attenuated the hyperpolarization effect of MRS2365 (Fig. 9A, D). Unlike in the ileum, α,β-meATP caused membrane hyperpolarization without prior membrane depolarization (compare Fig. 5B with 9B). The α,β-meATP membrane-hyperpolarizing effect was reduced by GBB (Fig. 9B, D; Supplementary Fig. 5).

Fig. 9.

GBB decreases MRS 2365-, α,β-methylene ATP- and forskolin-induced smooth muscle membrane hyperpolarization in the descending colon. A, B. Traces demonstrating that, compared with 1 µM MRS 2365 (A) and 10 µM alpha methylene ATP (B) alone, applying each compound in the presence of GBB diminished the membrane hyperpolarization response. C. GBB decreased the rate of membrane hyperpolarization elicited by the adenylyl cyclase activator forskolin (10 µM), although it did not change the magnitude of membrane hyperpolarization. D. A line graph showing that GBB significantly reduced the hyperpolarization triggered by MRS 2365 (MRS2365 vs. MRS2365 + GBB) and α,β-meATP vs. α,β-meATP + GBB). GBB decreased the rate of membrane hyperpolarization elicited by forskolin (Forsk. vs. Forsk.+GBB) but did not affect the final magnitude of membrane hyperpolarization (after 10 min). Alpha methylene ATP did not cause the transient membrane depolarization observed in the ileum (n=4–8 animals; one-way analysis of variance). There were significant differences after 3 min: ***MRS2365 vs. MRS2365+GBB, **α,β-meATP vs. α,β-meATP + GBB +GBB and *Forsk. vs. Forsk.+GBB; 5 min: **MRS2365 vs. MRS2365+GBB, ** α,β-meATP vs α,β-meATP + GBB, * Forsk vs. Forsk+GBB; and 10 min: *** MRS2365 vs. MRS2365+GBB, **α-β Met vs. α-β Met +GBB. After 10 min, Forsk vs. Forsk+GBB were not significantly different. *P<0.05, ***P<0.001; ****P<0.0001. GBB: Garcinia buchananii stem bark extract.

We next investigated whether or not GBB inhibited SK⁺ channels by applying a selective SK3 and SK2 modulator, CyPPA (34), after pretreatment with GBB for approximately 5 min. GBB did not affect CyPPA-induced hyperpolarization of smooth muscle cells (Supplementary Fig. 6C, D). Therefore, we determined whether or not GBB inhibits IJPs via direct actions on smooth muscle by testing whether or not GBB could modulate the effects of the adenylyl cyclase activator forskolin (33). Forskolin (10 µM) induced TTX-insensitive membrane potential hyperpolarization and thus reduced the amplitudes of the IJPs (Fig. 9C, D; Supplementary Fig. 6A, B). GBB did not affect the amplitude of hyperpolarization induced by forskolin but did reduce the rate of forskolin-induced membrane hyperpolarization, suggesting that the active ingredients of GBB affect cAMP signaling.

These results further support observations in the ileum suggesting that GBB contains molecules that inhibit the P2Y1 receptor signaling pathway but do not affect SK⁺ channels.

GBB inhibits IJPs and excitatory IJPs in longitudinal smooth muscle

We wished to determine whether or not GBB inhibited IJPs in the longitudinal muscle layer of the descending colon. We found that almost every IJP was associated with an EJP (asterisks, Fig. 10A, B). GBB rapidly inhibited the fast component of the IJP, significantly reducing the amplitude of the IJPs after three to five min. In contrast, during the first 10 min, GBB appeared to enhance the amplitude of EJPs. However, it abolished EJPs after 10 to 15 min and completely blocked the slow component of IJPs after 15 to 25 min (Fig. 10A–D). In addition, GBB appeared to prolong the duration of the slow component of IJPs, although this effect was not statistically significant.

Fig. 10.

Traces of membrane potential recordings and summary data showing the effects of GBB IJPs in longitudinal smooth muscle cells of the porcine descending colon. A–C. Compared with baseline IJPs (black traces), GBB at 0.5 g/100 ml rapidly inhibited the fast IJPs but enhanced the amplitude of EJPs. C, D. GBB abolished EJPs after 10 to 15 min and blocked slow IJPs after 15 to 25 min. *P<0.05, ***P<0.001; ****P<0.0001. GBB: Garcinia buchananii stem bark extract.

GBB, GB-2 and BNF do not affect P2Y1, P2Y2, P2Y4, P2Y6 or P2Y11 receptors or muscarinic receptors

To test the effect of GBB and the isolated compound on P2Y1 receptors, we performed secondary P2Y receptor functional assays. The results showed that neither GBB nor its major constituent compounds of MNF and BNF or (±)-hesperetin or (±)-naringenin, which were used as controls, affected P2Y1, P2Y2, P2Y4, P2Y6 and P2Y11 receptors as antagonists or agonists (Supplementary Table 1). GB-2 was not tested. Based on the results suggesting that GBB transiently augments and then inhibits EJPs, we studied whether or not GBB inhibited smooth muscle contraction induced by muscarinic receptors and the subsequent depolarizing stimuli (29,30,31, 35) via muscarinic receptor secondary binding assays. As shown in Supplementary Table 2, GBB and the derived flavones MNF, GB-2 and BNF did not affect the M2 or M4 receptors. Furthermore, (±)-hesperetin and (±)-naringenin, used as controls, did not affect the M2 or M4 receptors. In primary receptor binding assays (not shown), GB-2 did not affect M1, M2, M3, M4 or M5 muscarinic receptors. Given the lack of potential hits, GBB and flavanones isolated from it were not further analyzed.

Taken together, these results suggest that GBB, MNF, GB-2, BNF, (±)-hesperetin and (±)-naringenin do not inhibit muscarinic receptors.

Discussion

The goals of this study were 1) to test the hypothesis that GBB inhibits intestinal propulsive motility and increases the overall gastrointestinal transit time by decreasing aboral smooth muscle relaxation through inhibition of neuromuscular transmission; 2) to identify the compounds responsible for triggering these effects and how they act; and 3) to identify the type of IJP in the porcine descending colon and potential sex differences.

Overall, our study generated several new findings. Unlike in the porcine jejunum (13), ileum (16) and ascending colon (22), in which electrical field stimulation causes purely purinergic IJPs, stimulating descending colon preparations generated mixed IJPs with purinergic, nitrergic and nonnitrergic nonpurinergic components. Thus, we discovered that neuromuscular transmission in the porcine descending colon is mediated by purines, nitric oxide and other inhibitory neurotransmitters, such as VIP and PACAP (13, 15, 17, 33). The IJPs of neutered male and female pigs had similar amplitudes and durations in both the ileum and colon. In testing GBB, its fractions and isolated compounds, we found that three (M4, M5, M7) out of eight previously reported GBB fractions (11) inhibited intestinal propulsive motility and generated ectopic aboral contractions, suggesting that they blocked inhibitory neuromuscular transmission. GBB completely blocked IJPs in the ileum and descending colon, suggesting that it inhibits all components of IJPs, which is similar to TTX. M4 and M5 as well as GB-2 and BNF, the molecules purified from M4 and M5, respectively, inhibited purinergic IJPs. In addition, BNF appeared to be a weak inhibitor of nonpurinergic IJPs. We thus discovered that biflavanones and other unknown compounds in GBB could decrease intestinal motility by reducing inhibitory neuromuscular transmission in the inhibitory branch of the peristaltic reflex. This finding is consistent with our hypothesis that GBB has compounds that inhibit aboral smooth muscle relaxation by diminishing IJPs, causing ectopic aboral contractions. We also discovered that GBB transiently augments EJPs in the longitudinal muscle, suggesting that it could trigger transient contraction before inhibiting both IJPs and EJPs.

New findings also emerged from studies aimed at identifying the mechanisms by which GBB and its constituents reduced IJPs. We found that GBB significantly reduced membrane hyperpolarization elicited by MRS2365- and α,β-methylene ATP. However, receptor-binding assays showed that GBB and its derived compounds (MNF, GB-2 and BNF) were neither P2Y1, P2Y2, P2Y4, P2Y6 or P2Y11 antagonists nor agonists. GBB reduced the rate by which forskolin hyperpolarized the cell membrane but did not affect CyPPA-induced membrane hyperpolarization. Our results suggest that GBB and its active compounds do not act via the P2Y receptor and small conductance Ca²⁺-activated K⁺ channels, SK2 and SK3, to reduce inhibitory myenteric neuromuscular transmission. These results indicate that the inhibitory effects of GBB and likely its bioactive constituents on neurotransmission (5) underlie the disruption of inhibitory neuromuscular communication. Nevertheless, our results show that GBB and its bioactive constituents directly affect smooth muscle cells, supporting previous findings that MNF inhibits the excitability of intestinal smooth muscle (12).

In the present study, we observed that the MPLC fractions M4, M5, and M7 are the components of GBB that significantly inhibit motility, suggesting that the anti-motility compounds within GBB reside in these fractions. Regulation of gastrointestinal propulsive motility requires complex integration of neurochemical signaling within the ENS as well as ENS signaling to smooth muscle to coordinate rhythmic contraction orally and relaxation aborally (15, 17, 18). To exert antimotility effects, a molecule must reduce neurotransmission in ENS signaling, specifically synaptic transmission and/or neuromuscular transmission, or inhibit smooth muscle contraction or relaxation (5, 36). Our results support findings that GBB can be separated into components that retain antimotility and antidiarrheal effects (7, 23). In addition, these results strongly suggest that M4, M5, and M7 contain the compounds causing GBB’s antidiarrheal effects by reducing hypermotility associated with diarrheal illnesses. The finding that GBB, M4, M5 and M7 elicit ectopic aboral contractions supports the view that bioactive molecules in these preparations decrease IJPs, which underlie the inhibitory branch of the peristaltic reflex.

Gastrointestinal smooth muscle relaxation is triggered by the release of several inhibitory neurotransmitters from myenteric neurons, which elicit IJPs in smooth muscle cells. IJPs of various gastrointestinal organs have been identified using pharmacologic agents to isolate two main components, i.e. fast, purine-mediated IJPs and slow, nitric oxide-mediated IJPs (15, 16, 18). A third nonpurinergic nonnitrergic component has also been reported (13, 15, 16). These previous studies highlighted differences between species and the various organs in the gut and purine-mediated IJPs in the porcine ascending colon (22) and small intestine (16). We discovered that, in the porcine descending colon, IJPs have the fast component, which is sensitive to the Ca²⁺-activated SK channel blocker apamin, and the P2Y1 receptor antagonists MRS2179 and MRS2500, and the slow component which is sensitive to the nitric oxide synthase inhibitor L-NAME. In addition, they also have L-NAME + P2Y1 receptor antagonist-insensitive components, suggesting that the porcine descending colon generates mixed IJPs in response to purines, NO and VIP neurotransmitters from myenteric neurons.

Among the important new findings is the observation that GBB, M4 and M5 inhibited motility by significantly inhibiting purinergic IJPs and that MNF, GB-2 and BNF, the main neuroactive compounds within M3, M4 and M5, respectively, also inhibited purinergic IJPs. Our results support the hypothesis that GBB, M4 and M5 inhibit the aboral smooth muscle relaxation required for creating the peristaltic pressure gradient between the oral and aboral ends by diminishing IJPs, thereby causing ectopic aboral contractions. In support of this view, inhibitors of NO synthase, such as NG-nitro-L-arginine (L-NNA) or L-NAME, which inhibit nitrergic IJPs, increase muscular contractions (17, 19). Notably, GBB inhibited IJPs more dramatically than M4 and M5, as well as GB-2 and BNF and the GB-2 + BNF mixture. The main reasons for this difference are not currently apparent. MNF and GB-2 significantly inhibited purinergic IJPs and had additive effects, suggesting that they are the main neuroactive compounds of M4 and M5 and therefore GBB. However, unlike GBB, MNF, BNF, and GB-2 had minimal to no effect on the nonpurinergic IJPs in the descending colon. M7 inhibited motility but did not affect IJPs. M7 bioactive molecules are currently being investigated. Based on the current understanding of GBB, it is likely that the active molecules in M7 inhibit synaptic transmission or muscle contraction via mechanisms other than IJPs. Taken together, these results suggest that we have not yet identified all neuroactive molecules in GBB.

Previous studies have shown that GBB exerts gastrointestinal antimotility effects through the inhibition of interneuronal synaptic transmission in the myenteric ganglia (5). GBB inhibits mechanosensory neurons and nociceptive signaling in mesenteric afferents (8). The current study revealed that GBB acts by completely inhibiting intestinal inhibitory neuromuscular transmission. Although GBB appeared to inhibit EJPs, we also noted that GBB and BNF did not block one type of residual EJP we consider to be mediated by tachykinins and/or acetylcholine. This suggests that the specific molecular mechanisms utilized by antimotility molecules in GBB do not involve generalized blocking of neurotransmitter release from myenteric neurons. Our findings suggest that GBB, M4 and M5 inhibit neuromuscular transmission and that MNF, GB-2 and BNF are likely neuroactive molecules in GBB. This notion needs to be confirmed by studying their effects on neurotransmission in the myenteric plexus.

Both nitric oxide and purines act as the major inhibitory neurotransmitters to intestinal smooth muscle cells in humans and in some animal species (13, 15, 16). In porcine ileum smooth muscle cells, however, purines, which act via the P2Y1 receptor, are the predominant inhibitory neurotransmitters (13, 16, 22). Our results show that GBB, M4, and M5 and higher concentrations of MNF, GB-2 and BNF inhibit purinergic IJPs, suggesting that GBB-derived fractions and MNF, GB-2 and BNF inhibit purinergic neurotransmitter release or possibly inhibit a key step in P2Y1 receptor signaling. This is supported by the antagonistic action of GBB on MRS2365- and α,β-methylene ATP-induced membrane hyperpolarization in circular smooth muscle cells in both the ileum and descending colon. However, the observation that IJPs were blocked by TTX suggests that GBB-derived fractions and active compounds act via neuromechanisms. While GBB may block inhibitory neuromuscular transmission by inhibiting neurotransmitter release from inhibitory motor neurons, it is also possible that GBB acts directly on smooth muscle cells. The antagonistic effects of GBB on MRS2365- and α, β-methylene ATP and the reduction in the rate of forskolin-induced membrane hyperpolarization support this idea. However, our results suggest that GBB, MNF, GB-2, BNF, hesperetin and naringenin (controls) do not inhibit P2Y1 receptors or Ca²⁺-activated SK, highlighting the need for further studies to define the mechanisms and active molecules in future studies.

Our results showed for the first time that MNF, GB-2 and BNF significantly inhibited purinergic IJPs in the porcine ileum. MNF is the primary component (by concentration) of GBB and is the main component of M3 (9,10,11). In motility studies, M3 did not affect motility, so we assumed that MNF would not affect IJPs and thus used it as a negative control. MNF unexpectedly inhibited purinergic IJPs after prolonged exposure (20–30 min), similar to GB-2 and BNF. These results suggest that biflavanones (MNF, GB-2, BNF) and compounds with related chemical structures may reduce intestinal motility by decreasing neuromuscular transmission. In the present study, (±)-hesperetin inhibited pellet propulsion in the guinea pig colon, indicating that it inhibits intestinal motility. However, in contrast to MNF, GB-2 and BNF, (±)-hesperetin did not inhibit IJPs in the porcine intestine; instead, it increased the duration of nonpurinergic IJPs in the descending colon. These results suggest that (±)-hesperetin modulates neurotransmission in the myenteric plexus, including the release of nitric oxide. Our results suggest the need to determine how (±)-hesperetin affects excitatory neuromuscular transmission as well as synaptic transmission, as these are possible mechanisms underlying its antimotility effects. Although the flavonoid monomer quercetin has been shown to inhibit synaptic transmission (37), quercetin 3-O-β-D-glucoside did not inhibit IJPs.

GBB caused membrane hyperpolarization and significantly reduced membrane hyperpolarization elicited by MRS2365- and α,β-methylene ATP-induced membrane hyperpolarization. However, receptor-binding assays showed that GBB and the derived compounds tested are neither P2Y1 or other P2Y receptor antagonists nor agonists. It has been reported that intramuscular platelet-derived growth factor receptor alpha-positive cells mediate purinergic hyperpolarization in mouse colonic muscle. Hyperpolarization spreads via gap junctions to smooth muscle cells, resulting in fast IJPs, the hallmark of purinergic inhibitory neurotransmission in gastrointestinal muscles (35). It is possible that GBB and its derived compounds decrease IJPs by inhibiting Gq protein or downstream signaling mechanism or Ca²⁺ influx (12) and Ca²⁺ mobilization from the ER, and K⁺ handling mechanisms in smooth muscle cells, or in both intramuscular platelet-derived growth factor receptor alpha-positive cells and smooth muscle cells. GBB reduced the rate by which forskolin hyperpolarized the cell membrane but did not affect CyPPA-induced membrane hyperpolarization. Therefore, while it is possible that GBB and its active compounds inhibit intestinal relaxation by direct action on the smooth muscle, they do not inhibit P2Y receptors or SK2 or SK3 channels.

Given that GBB inhibits both IJPs and EJPS, we postulated that GBB inhibits smooth muscle contraction induced by muscarinic receptor activation and subsequent smooth muscle depolarization (29,30,31). Our results conflict with this idea, since GBB, MNF, GB-2 and BNF did not affect M2 muscarinic receptors, and primary receptor binding assays suggest that GBB and GB-2 do not affect M3 muscarinic receptors.

There is strong evidence for structural and functional gender differences in the gastrointestinal tract of humans and other mammals. Our study in the mouse colon suggested that females might have greater inhibitory neuromuscular transmission than males (26). It is believed that sex hormones play critical roles in causing these differences (26, 38). In the present study, the IJPs of neutered male and female pigs had similar amplitudes and durations in both the ileum and colon, suggesting that there were no marked sex differences. We speculate that castration-induced hormonal changes in male animals used in this study may at least partially explain the lack of difference between neutered male and female animals reported herein.

Conclusion

In conclusion, the porcine descending colon inhibitory limb of the reflex is mediated by purines, nitric oxide and nonnitrergic nonpurinergic inhibitory neurotransmitters, such as VIP and PACAP. GBB blocks all components of intestinal inhibitory neuromuscular transmission. MNF, GB-2 and BNF contribute to GBB actions mainly by inhibiting purinergic IJPs. Further studies are needed to determine the neuroactive molecules of GBB that inhibit neurotransmission and their mechanisms of action. We showed that flavanones and likely flavonoids commonly used as ingredients in food supplements due to their antioxidative effects (10, 39) were able to decrease intestinal motility by reducing intestinal inhibitory neuromuscular transmission.

Funding

This work was supported by the University of Idaho—Dyess Faculty Fellowship, and Institutional Development Awards (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health grant number P20 RR016454 and The National Center for Research Resources grant number P20 GM103408 through IDAHO INBRE.

Conflict of Interest

All authors wish to confirm that they have no competing interests financial or otherwise and have nothing to disclose that could have influenced its outcome.