Inhibitory mechanisms of docosahexaenoic acid on carbachol-, angiotensin II-, and bradykinin-induced contractions in guinea pig gastric fundus smooth muscle

We studied the inhibitory actions of docosahexaenoic acid (DHA) on the contractions induced by carbachol (CCh), angiotensin II (Ang II), and bradykinin (BK) in guinea pig (GP) gastric fundus smooth muscle (GFSM), particularly focusing on the possible inhibition of store-operated Ca2+ channels (SOCCs). DHA significantly suppressed the contractions induced by CCh, Ang II, and BK; the inhibition of BK-induced contractions was the strongest. Although all contractions were greatly dependent on external Ca2+, more than 80% of BK-induced contractions remained even in the presence of verapamil, a voltage-dependent Ca2+ channel inhibitor. BK-induced contractions in the presence of verapamil were not suppressed by LOE-908 (a receptor-operated Ca2+ channel (ROCC) inhibitor) but were suppressed by SKF-96365 (an SOCC and ROCC inhibitor). BK-induced contractions in the presence of verapamil plus LOE-908 were strongly inhibited by DHA. Furthermore, DHA inhibited GFSM contractions induced by cyclopiazonic acid (CPA) in the presence of verapamil plus LOE-908 and inhibited the intracellular Ca2+ increase due to Ca2+ addition in CPA-treated 293T cells. These findings indicate that Ca2+ influx through SOCCs plays a crucial role in BK-induced contraction in GP GFSM and that this inhibition by DHA is a new mechanism by which this fatty acid inhibits GFSM contractions.

DHA versus U46619 was calculated to be 5.13 17 , which was equivalent to the value of 5.16 obtained in porcine coronary artery 10 .Functional inhibition of voltage-dependent Ca 2+ channels (VDCCs) was also assumed to be partly responsible for the DHA-induced suppression of GFSM contractions by U46619 and prostanoids 17 .Furthermore, the functional inhibition of VDCCs was shown to be involved in the DHA inhibitory actions on GP ileal/colonic longitudinal SM contractions induced by some prostanoids 18 .
In addition to prostanoids, non-prostanoid substances including acetylcholine, angiotensin II (Ang II), and bradykinin (BK) can be contractile regulators of GFSM [19][20][21] .Acetylcholine is involved in physiological contractions of GFSM, and its release from parasympathetic nerve endings causes gastric peristalsis 22 .Although fluctuations in plasma concentrations of Ang II in the physiological range are suggested not to play an important role in the normal regulation of gastric motility, angiotensin-positive neurons have also been shown to exist in the stomach and may play some roles locally 23 .Furthermore, in streptozotocin (STZ)-induced diabetic mice, expression of the Ang II AT 1 receptor in GFSM and that of angiotensin converting enzyme (ACE) in gastric mucosa are reported to increase, and Ang II-induced GFSM contractions are greater in the diabetic mice than in normal mice 24 .Additionally, BK-induced GFSM contractions are reported to be enhanced in STZ-induced diabetic mice, which may be involved in the development or maintenance of the diabetic lesions 21 .
Regarding contractions induced by non-prostanoid biological substances such as acetylcholine and histamine, we previously showed that DHA partly inhibited their contractile effects on GP ileal/colonic longitudinal SMs by functionally inhibiting VDCCs 18 .We assumed that DHA would exhibit the same inhibitory actions against non-prostanoid substance-induced contractions even in GFSM.Therefore, this study was performed to test this hypothesis.Herein, we show the inhibitory actions of DHA against GP GFSM contractions induced by non-prostanoid substances, focusing on carbachol (CCh, a choline ester), Ang II, and BK, which were found to produce detectable and quantifiable contractions.In addition, we report evidence to support that store-operated Ca 2+ channels (SOCCs) are also a target of DHA especially in its inhibitory actions against BK-induced contraction of GP GFSM.

Inhibitory actions of DHA on GFSM contractions induced by CCh, Ang II, and BK
In the present study, we examined whether DHA inhibited contractions induced by non-prostanoid chemical stimulants.We focused on CCh, Ang II, and BK because, among the 22 chemicals tested, (1) sufficiently detectable and quantifiable contractile effects were observed with CCh, Ang II, and BK; (2) strong contractions were also induced by neurokinin A but could not be quantified; and (3) the other 18 biological substances (histamine, substance P, serotonin, melatonin, dopamine, glucagon-like peptide-1, urotensin II, atrial natriuretic peptide, neuromedin B, neuromedin C, neuromedin U, orexin A, motilin, galanin, guanosine, guanosine triphosphate, adenosine, and adenosine triphosphate) did not induce substantial contractions (data not shown).

Discussion
We have proposed TP receptor antagonism and functional inhibition of VDCCs as the mechanisms by which DHA produces immediate inhibitory activity against SM contractions induced by a TP receptor agonist/prostanoids and non-prostanoid agonists [8][9][10][11]17,18 . In adition, we now show that inhibition of SOCC-mediated Ca 2+ influx is a new mechanism responsible for the DHA-induced inhibition of SM contractions (Supplementary Fig. 7).We previously reported that DHA suppressed GP GFSM contractions induced by U46619 (a TXA 2 mimetic) and prostanoids 17 .In the present study, we showed that DHA significantly suppressed the maximum contractions induced by CCh/Ang II/BK, although this unsaturated fatty acid showed a significant inhibition only against BK when contractions were assessed by AUC (Fig. 1).Contractions induced by CCh/Ang II/BK were shown to be mediated through their corresponding receptor subtypes (acetylcholine M 3 , Ang II AT 1 , and BK B 2 , respectively) since they were almost abolished by their corresponding selective receptor antagonists (solifenacin 30 , losartan 31 ,  and icatibant 32 , respectively) (Supplementary Fig. 2, Fig. 7AB).In contrast, DHA did not inhibit intracellular Ca 2+ increases in Ca 2+ -free solution containing EGTA and Gd 3+ produced by these chemical receptor stimulants in cells expressing acetylcholine M 3 , Ang II AT 1 , or BK B 2 receptors, respectively (Supplementary Fig. 3, Fig. 7C).In Ca 2+ -free solution, these chemical receptor stimulants increased intracellular Ca 2+ concentrations due to Ca 2+ release from intracellular Ca 2+ stores, but not due to extracellular Ca 2+ influx.Therefore, potential inhibition of these chemical receptors and Ca 2+ release from intracellular Ca 2+ stores were judged to be excluded as mechanisms by which DHA inhibits non-prostanoid stimulation-induced contractions of GFSM.
Other than chemical receptors, DHA's targets for its inhibitory actions against non-prostanoid-induced contractions would include pathways of Ca 2+ influx from extracellular spaces 18 .Indeed, we previously showed that functional suppression of VDCCs could partly account for the inhibitory actions of DHA versus acetylcholine-/ histamine-induced contractions in ileal and colonic longitudinal SMs 18 .Therefore, we next examined the extracellular Ca 2+ dependency and extent of inhibition by verapamil (a VDCC inhibitor) 33 of GFSM contractions induced by CCh/Ang II/BK.All contractions induced by these chemicals were almost completely abolished by extracellular Ca 2+ removal, indicating that these contractions almost completely depend on extracellular Ca 2+ influx and not Ca 2+ release from intracellular Ca 2+ stores (Fig. 2A).However, the extent of inhibition by verapamil differed among the contractions (Fig. 2B); CCh-induced contractions were almost completely suppressed by verapamil, whereas those induced by Ang II and BK were suppressed by 52% and 19%, respectively.These results indicate that the extent of functional contribution of VDCCs is 100%, ~ 50%, and ~ 20% for GP GFSM contractions induced by CCh, Ang II, and BK, respectively.In addition to these findings, we previously showed that DHA significantly but partly inhibits high KCl-induced contractions of GP GFSM 17 .Therefore, functional inhibition of VDCCs may be partly but substantially responsible for the inhibitory actions of DHA on these non-prostanoid-induced contractions.This interpretation supporting the involvement of functional VDCC inhibition is consistent with our previously proposed mechanism by which DHA inhibited contractions induced by U46619 and various prostanoids in GP GFSM 17 .
In contrast to CCh-induced contractions, more than 50% of Ang II-and BK-induced contractions remained in the presence of verapamil (Ang II: 48%, BK: 81%).This finding suggests that Ca 2+ influxes through non-VDCC pathways play a sub-principal (Ang II) or principal (BK) role in inducing GFSM contractions and that such Ca 2+ influx pathways are inhibited by DHA.Especially, the finding that ~ 80% of BK-induced contraction remained in the presence of verapamil implies that suppression of BK-induced contractions by DHA was mediated by inhibition of Ca 2+ influxes through non-VDCC pathways.Potential candidates for the non-VDCC Ca 2+ influx pathways responsible for SM contractions are ROCCs and SOCCs 34 .Therefore, we next examined whether these Ca 2+ channels are responsible for BK-induced GFSM contraction using their corresponding inhibitors.The results clearly showed that BK-induced contractions in the presence of verapamil were not inhibited by LOE-908 (an ROCC inhibitor) 35 (Fig. 3A) but were strongly inhibited by SKF-96365 (an ROCC/SOCC inhibitor) 36 (Fig. 3B,  C ).Therefore, SOCCs may play an important role as the primary extracellular Ca 2+ influx pathway to generate BK-induced contractions in GP GFSM.Furthermore, the finding that DHA (3 × 10 −5 M, 10 −4 M) significantly suppressed BK-induced contractions in the presence of verapamil plus LOE-908 (Fig. 4) suggests that SOCCs are a target of DHA to produce its inhibitory action against BK-induced contractions.
In contrast, Ang II-induced contractions in the presence of verapamil were significantly inhibited by both LOE-908 and SKF-96365 (Supplementary Fig. 8).Thus, unlike BK-induced contractions, Ang II-induced contractions may be elicited by extracellular Ca 2+ influx through ROCCs in addition to VDCCs/SOCCs.
To obtain more direct evidence to support the hypothesis that DHA exerts an immediate inhibitory effect on Ca 2+ influx through SOCCs, we subsequently conducted two types of experiments with GP GFSM tissues and cultured 293T cells and obtained the following results.(1) SOCCs are the extracellular Ca 2+ influx pathways, whose activation is triggered by Ca 2+ depletion in the SER 39 .Therefore, chemical depletors such as SERCA inhibitors, including CPA, increase intracellular Ca 2+ levels following activation of extracellular Ca 2+ influx through SOCCs 40,41 , thus generating SM contractions 42 .In GP GFSM tissues, CPA induced contractions in the presence of verapamil and LOE-908 (VDCC and ROCC inhibitors, respectively).Since CPA-induced contractions were almost completely inhibited by SKF-96365, such contractions were shown to be caused by extracellular Ca 2+ influx through SOCCs (Fig. 5B).The finding that the SKF-96365-inhibitable contractions induced by CPA were strongly attenuated by DHA indicated that DHA was able to inhibit the extracellular Ca 2+ influx through SOCCs (Fig. 5A).(2) More direct evidence was obtained by Ca 2+ measurements in 293T cells.In these cells, an SKF-96365-inhibitable increase in Ca 2+ concentration was generated by Ca 2+ addition following CPA treatment in Ca 2+ -free medium.This indicated that this intracellular Ca 2+ increase was induced by Ca 2+ influx through SOCCs, the activation of which was triggered by CPA-induced depletion of SER Ca 2+ (Fig. 6B).DHA again strongly suppressed this intracellular Ca 2+ increase, indicating that DHA could inhibit SOCC-mediated extracellular Ca 2+ influx (Fig. 6A).Note that (i) this intracellular Ca 2+ increase was unaffected by LOE-908 43 , and (ii) in a separate series of experiments, the CPA-induced intracellular Ca 2+ increase in Ca 2+ -containing medium was unaffected by verapamil plus LOE-908 (Supplementary Fig. 4).These results exclude the potential contribution of VDCCs and ROCCs.To the best of our knowledge, this is the first report indicating the potential inhibitory action of DHA against SOCCs that are responsible for contraction of SM tissues.To confirm that DHA more directly inhibits drug receptor-stimulated Ca 2+ entry through SOCCs, the effects of DHA on intracellular Ca 2+ increases following depletion of Ca 2+ stores after repeated stimulation of the drug receptor should be examined in the future to complement the use of CPA.www.nature.com/scientificreports/One type of SOCC is composed of Orai channels, which are the main Ca 2+ entry channels, and stromal interaction molecules (STIMs), which are SER Ca 2+ sensors that control Orai activation 34 .There are three isoforms of Orai channels (Orai1-3) and two isoforms of STIMs (STIM1, 2).In this study, we measured the mRNA expression of these isoforms in GFSM and found relatively high expression of Orai1, Orai3, and Stim2 (Fig. 8A).Transient receptor potential canonical (TRPC) channels represent another type of SOCC 44 .We also measured the mRNA expression of TRPC channel isoforms (Trpc1, Trpc3-7) in GFSM and found relatively high expression in the following order: Trpc3 > Trpc6 > Trpc4 > Trpc1 (Fig. 8B).To estimate the functional SOCC molecules in GP GFSM, we investigated the effects of five types of Orai/TRPC channel inhibitors on the CPA-induced contractions.CPA-induced contractions in the presence of verapamil and LOE-908 were significantly suppressed by 2-APB (a non-selective Orai/TRPC channel inhibitor) and almost completely suppressed by Synta66 (a selective Orai1 inhibitor), but not Pyr10 (a selective TRPC3 channel inhibitor), ML204 (a selective TRPC4 channel inhibitor), and SAR7334 (a selective TRPC6 channel inhibitor) (Supplementary Fig. 6).Since LOE-908 has been reported to inhibit TRPC1 45 , the involvement of TRPC1 can be excluded under these conditions.The involvement of Orai3 could not be investigated in this study because a selective inhibitor is not available.However, since the CPA-induced contractions were almost completely abolished by the selective Orai1 inhibitor (Synta66), the contribution of Orai3 should be negligible.Therefore, in GP GFSM, the depletion of SER Ca 2+ may activate STIM2, followed by the activation of Orai1.Although the detailed mechanisms by which DHA suppresses SOCCs are still unknown, two possibilities exist: (1) DHA directly suppresses Orai channels (Orai1) or (2) DHA suppresses the activation of Orai channels (Orai1) by STIM (STIM2).More detailed information can be obtained with SOCCrelated gene knockout animals or knockdown GFSM cells.
In conclusion, to the best of our knowledge, we are the first to find that DHA can inhibit GP GFSM contractions mediated through non-prostanoid receptor agonists.The mechanisms by which DHA inhibits these contractions involve inhibition of Ca 2+ influx through SOCCs, which may be largely responsible for the inhibition of BK-induced contractions.

Animals
We used male GPs (weight: 310-690 g, age: 4-16 weeks; Kyudo Co., Ltd., Saga, Japan), which were housed under a fixed 12/12 h light/dark cycle (08:00-20:00) and controlled conditions (20-22 °C, relative air humidity: 50 ± 5%) with food and water available ad libitum.This study was carried out in compliance with ARRIVE guidelines and the guidelines of the Laboratory Animal Center of Faculty of Pharmaceutical Sciences, Toho University.This study was approved by the Toho University Animal Care and User Committee (approval number: 20-444).

GFSM strips
GFSM strips were prepared and tension changes were recorded according to our previous report 17 .GPs were anesthetized with isoflurane (inhalation) and exsanguinated from the carotid artery.The isolated stomach was denuded of its connective and adipose tissues, and then separated into the GF and gastric body in Locke-Ringer solution containing (in mM) NaCl, 154; KCl, 5.6; CaCl 2 , 2.2; MgCl 2 , 2.1; NaHCO 3 , 5.9; and d-(+)-glucose, 2.8.The GF interior was irrigated with the solution.The GF was further cut along the longitudinal axis into 2-4 segments, and the epithelium of the segments was removed gently.The GFSM strips were approximately 5-20 mm in length and 2-3 mm in width.
The GFSM strips were suspended under a 1.0-g resting tension with clips and cotton thread in a 20-ml organ bath filled with Locke-Ringer solution.The solution was maintained at 32 ± 1 °C and oxygenated with 95% O 2 and 5% CO 2 .The suspended strips were allowed to equilibrate for 60 min.Tension changes of GFSM strips were isometrically recorded with PowerLab™ and LabChart™ (Version 7) software (ADInstruments Pty. Ltd., Bella Vista, NSW, Australia) using force-displacement transducers (TB-612T, Nihon Kohden, Tokyo, Japan; FORT 25, World Precision Instruments, Sarasota, FL, USA) and carrier amplifiers (MSC-2 Signal Conditioner, Labo Support Co., Osaka, Japan; AP-621G, Nihon Kohden; TBM4M, World Precision Instruments).After a 60-min incubation, the strips were contracted by CCh (10 −5 M) at least three times with a 10-min interval.Thereafter, the strips were contracted by CCh (6 × 10 −8 M), Ang II (10 −7 M), and BK (10 −6 M) for 10 min at least twice with an interval of 30 min (CCh/BK) or 120 min (Ang II) until stable contractions were obtained.To prevent the potential action of endogenous prostaglandins, all tension recordings were carried out in the presence of indomethacin (3 × 10 −6 M).

Effects of extracellular Ca 2+ removal on GFSM contractions
After stable contractions were obtained, the strips were incubated for 20 min (CCh/BK) or 110 min (Ang II).After the 20-or 110-min incubation, the bath solution was replaced with Ca 2+ -free solution containing (in mM)

Measurement of intracellular Ca 2+ changes
The measurement of intracellular Ca 2+ changes was performed as previously described 10 .Briefly, the day before the measurement, 293T and B 2 -293 T cells were seeded at ~ 90% confluence and cultured overnight.The next day, these cells were incubated in the presence of Fura-2 AM for 60 min.The cells were then rinsed with the medium.
In the experiment shown in Fig. 6, the medium was replaced with Ca 2+ -free medium, and in the experiment shown in Fig. 7B, the medium was replaced with Ca 2+ -free medium containing EGTA (2 × 10 −4 M)-and Gd 3+ (10 −5 M).After this procedure, the fluorescence intensity at 510 nm emission generated by 340 nm and 380 nm excitation was measured using microplate readers (Nivo, PerkinElmer Inc., Waltham, MA, USA; Infinite F200 Pro, Tecan Group Ltd., Mӓnnedorf, Switzerland).We assumed that the changes in the ratio of fluorescence intensities at 510 nm emission generated by 340 nm and 380 nm excitation (F340/380) reflected the relative changes in intracellular Ca 2+ concentration.After a 10-min incubation in the presence of DHA (3 × 10 −5 M), icatibant (3 × 10 −5 M), or DHA/icatibant vehicle (0.5% EtOH/H 2 O), BK (10 −6 M) or CPA (10 −5 M) was added via the injector module, and the fluorescence intensity was measured for 5 min.After CPA application, Ca 2+ (1.8 mM) was added via the injector module, and the fluorescence intensity was measured for 5 min.At the end of the experiment, to determine background fluorescence, ionomycin (5 × 10 −6 M) and Mn 2+ (5 × 10 −2 M) were added.This background fluorescence was subtracted from the fluorescence intensities of all measurements.

RT-qPCR of SOCC-related mRNA expression
RT-qPCR was performed as previously described 17 .Briefly, total RNA was isolated from GFSM.First-strand cDNA was synthesized using ReverTra Ace® qPCR RT Master Mix with gDNA Remover (TOYOBO Co. Ltd., Osaka, Japan).RT-qPCR was performed on a CronoSTAR™ 96 Real-Time PCR System (Takara Bio Inc., Shiga, Japan) using Taq Pro Universal SYBR® qPCR Master Mix (NIPPON Genetics Co. Ltd., Tokyo, Japan).Supplementary Table 1 shows the primers used in this study.The thermal cycler parameters were set to 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s.The fluorescence intensities were measured at each 60 °C step to confirm DNA amplification.We used CronoSTAR™ 96 Software (Takara Bio Inc.) to analyze the mRNA expression level of each gene.The mRNA expression levels were normalized to that of glyceraldehyde-3-phosphate dehydrogenase (Gapdh), which was set to 1.Samples that did not produce a Ct value after 40 cycles were considered to have no expression.The extent of relaxation induced by DHA/SKF-96365/2-APB/Synta66/Pyr10/ML204/SAR7334 on CPAinduced contractions was calculated relative to the steady-state tension level prior to the application of DHA/ SKF-96365/2-APB/Synta66/Pyr10/ML204/SAR7334 (0% relaxation) and to the tension level after the application of 10 −4 M PPV (100% relaxation).
Data are expressed as means ± standard error of the mean (SEM), where n refers to the number of experiments.Statistical analyses were performed with paired t-tests, unpaired t-tests with Welch's correction if necessary, or one-way ANOVA followed by Dunnett's test, using GraphPad Prism™ (Version 6.0) (GraphPad Software, Inc., San Diego, CA, USA).Statistical significance was set at P < 0.05.