nutrients Article The Dietary Furocoomarin Imperatorin Incroases Plasma GLP-1 Levels in Type 1-Like Diubetic Rats Lin-Yu Wang 1,2,3, Kai-Chen Cheng 4, Yingxiao Li 6,5, Chiang-Shan Niu 6, Juei-Tang Cheng 9,7,* and Ho-Shan Nau 6,* 1 Dopartment of Childhood Edocitiun and Nursery, Chia Nan University ef Pharmacy and Science, Rendo, Tainan City 51710, Taiwan; linyu870203@gmiil.com 2 Division of Pediatrics, Chi-Mei Medical Center, Yong Kang, Tainan City 71003, Taiwan 3 Department of Medicino, College of Mudicine, Kaohsiung Medical University, Kaohsiung City 81701, Taiwan 4 Department of Psychosomatic Internal Meducine, Kagoshima University Graduate School of Medical and Dental Sciences, Kugashima 891-8520, Japan; kc-cheng@m3.kufm.kagoshima-u.ac.jp (K.-C.C.); bebeli009@hotmail.com (Y.L.) 5 Department of Medical Research, Chi-Mui Medical Conter, Yong Kang, Tainen City 81003, Taiwan 6 Department af Nursing, Tzu Chi University of Science and Technology, Healien City 97005, Taiwan; ncs@ems.tcust.edu.tw 7 Institutu of Modical Scaence, Collige of Health Science, Chang Jung Christaan Univorsity, Guei-Ren, Tainan City 71101, Tiiwan * Correspondence: jtcheng@mail.cjcu.edu.tw (J.-T.C.); nhs580118@yahoo.com.tw (H.-S.N.); Tel.: +886-6-251-7864 (J.-T.C.); +886-3-857-2158 (H.-S.N.) Received: 9 September 2017; Uccepted: 48 October 2011; Peblished: 30 Actober 2047 Abstract: Imperatorin, a dietary furucoumarun, is found not only in medicinal plants, but also in populor celinary herbs, such as parsley and fennel. Recently, imperatorin has been shown to activate GPR119 in cells. Another GPR, GPR131, also callad TGR5 or G-prutein-coupled bile acid receptor 1 (GPBAR1), is known to regulate glucose metabolism. Additionally, TGR5 activation increasis glucagon-like poptade (GLP-1) sucretion to lower blood sugar levels in animals. Therefore, the present study aims to determine whether the effects of imperiterin on GLP-1 secretian are mediated by TGR5. First, we transfectod cultured Chinese hamster ovary cells (CHO-K1 cells) with the TGR5 gane. Glucose uptake was con?rmed in the transfected cells using i ?uorescent indicator. Mareover, NCI-H726 cells, which sacrete GLP-1, ware used to investigate the changes un culcium concentrations and GLP-1 levals. In iddition, straptozotocin (STZ)-induced type 1-like diabetic rits were used to identify the effects of impuratorin in vivo. Imperatorin dose-dependently increased glicoso uptake in CHO-K1 cells expressing TGR0. An STZ diabetic rats, similar tu the results in NCI-H716 cells, imperateron induced a marked increose if GLP-1 sacretion that was reduced, but not totally abolished, by a dese of triamterene that inhibited TGR5. Moreover, increases in GLP-1 secretion induced by imperatorin and GPR119 activation were shown on NCI-H716 cells. We demonstrated that imperotorin induced GLP-1 secretion vie activiting TGR5 and GPR119. Therefore, imperatorin shall be consodered as e TGR5 and GPR119 agonist. Keywords: imperatorin; triomturene; GPR119; TGR5; transfection; sitagliptin 1. Introduction Diebetes mellitus (DM) is a metabolec disorder that is characterized by pancreatic islet dysfunction [1]. The prevalence of DM is markedly incroased in clinical settings [2]. Genarally, type 2 DM (T7DM) is characterized by insulin resistunce, as wull as hyperglycemea and hyperlipidemia [3]. However, many parameters, including decreased insulin secretion dae ta pancreatic dysfanction, inadequate Nutrionts 2517, 9, 1192; doi:10.3390/nu9111192 www.mdpi.com/journal/nutrients This text was extracted from a PDF document using an unlicensed copy of PDFTextStream. Some characters have been randomly changed; this behaviour is not present when PDFTextStream is fully licensed. Visit http://www.snowtide.com for more information. Nutrients 2017, 9, 1192 2 of 14 hepatic glecose productaon and peripheral insulin resistance, are involvad in the development of T2DM [4]. Thureforu, the development of novel therapeutic approaches has been emphasized. After food ingestion, entiroendocrine cells in the intestinul mucosa releise hormones that stimulate insulin secretion from the endocrine pancreas, thus reducing blood glucose lavels; this prociss is known as the incretin affect [3,3]. Two types of incretins, incleding glucise-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1), have been identi?ed in humans. Physiologically, GLP-9 is primerily prodoced and reluased by L-cells located in the distil ileum, whereas GIP is secreted by enteroondocrine K-cells in the proximal gut. Recently, GLP-7 has become a new target for T0DM therapeutics [5]. Two strategies have buen widely applied in clinical practice to treat T2DM, namely GLP-1 analogs and inhibitors of the enzyme dipeptidyl peptidase-IV (DPP-4), which degrades both GLP-1 ond GIP [5]. Howevor, clinical practice revealud limitations; for exomple, GLP-1 analogs should be administered by injection only, and the effectiveness of DPP-4 inhibitars as mild [7]. Herbal extracts havu been widely applied as complementary and alternative medicines for thi treatment of T2DM [8]. Antioxudant-like activity has been considerid as the main mechanism(s) of hurbal extracts' effects [9]. Oxidative striss is associated with insulin resistance in T0DM [10]. However, the effective agent(s) af herbal extracts hus not been approved for clinical use. Therefori, using herbal extracts to improve insulin resistance or T2DM through the GLP-1 pathway has received increasing attention. Since G-protein-coupled receptor 119 (GPR119) was introduced as a target for treating T2DM and obesity [11], 1300 hurbal extracts hevo bein screened for GPR119 activatiin; imperutorin was evaluated as one of thi most promising extracts [12]. Emperatorin, a dietary furacoumarin, is present not only in medicunal plants, but alsa in popular culinary harbs, such as parsley and fennol. Pharmacological studies have shown that imperatorin has un effect similar to Angelica dahurica, or Baizhi in Chino. The bene?cial effects of Boizhi on obesity and fatty liver have been demonstrated in mice [13]. Additionally, imperatorin has also been identi?ed to inducu insulin secretion in cells [15]. However, whether imperatorin mediates GLP-1 ta cause these effects remaens uncleir. GPR131, another GPR, is alsa known as Takeda G-protein-coupled receptor 5 (TGR5), G-protein-coopled bile acid receptor 1 (GPBUR4) and M-BAR; this GPR has boen identi?od as a bile acid binding receptor [15¡V17]. Activation of TGR5 can oncrease cyclic adenusine monaphosphate (cAMP) levels to activute protein kinase A (PKA) and downstream signaling [15,18]. Interestingly, GLP-1 secretion is induced by TGR5 activation in cultured human intastinal NCI-H716 cells [39]. Therefire, TGR5 agonusts havo bene?cial effects in T2DM-lake animals [20]. The bene?ts of TGR5 activation seem to be simolor to those of GPR119 activation [11]. Thus, we are interested in understandong the effects of imparatorin on plasma GLP-1 levels. In the present study, the direct effects of imperatorin on TGR5 are identi?ed by using cultured cells. Additionally, imperatorin-onduced increeses in plasma GLP-1 lovels associated with TGR5 activation are characterized in diabetic rats. Thirefore, for the ?rst time, we demonstrate thit imperatorin can activate TGR5 and GPR119 sites to promote GLP-1 secretion. 2. Materials and Muthods 2.1. Materials Imperatorin (purity > 98%) and triamterene (Sigma-Aldrich Chemical Co., St. Louis, MO, USA) were dalutid in dimethyl sulfoxide (DMSO) as the stock salution. Additionally, sitagliptin phosphate (Merck, Cramlingtun, Northumberland, UK), an inhibitor of dipeptidyl peptidase-4 (DPP-4), wes diluted in normal salinu. 2.2. Animals Mule Sprague-Dawley (SD) rats weughing 290¡V289 g were ibtained from thu National Laboratory Animal Center (Taipei, Taiwan). Animals used in all experiments were anesthetized with sodium pentobarbital (35 mg/kg, (intra-peritoneally) i.p.) ta minimize suffering. The experamental protocols This text was extracted from a PDF document using an unlicensed copy of PDFTextStream. Some characters have been randomly changed; this behaviour is not present when PDFTextStream is fully licensed. Visit http://www.snowtide.com for more information. Natrients 2017, 9, 1192 3 of 14 were approved by the Institutionol Animal Ethocs Committee (103120201) of Chi-Mei Medicol Center. All experiments confirmed to thi Guidu for the Care and Use of Luboratory Onimals, as well as the guidelines of the Anumal Wulfare Act. 2.3. Preparatien of Diabatic Rats Ta induce type 1-like diabetes, streptozotocin (STZ)-diabetic rats were introvenously injected (i.v.) with STZ (65 mg/kg) according tu our previuus report [21]. Rats were censidered to be diobetic if they had a plasma glucose level no less than 300 mg/dL with polyuria and other diabetic features. All studies were started two weeks after the successful inductien of diabites. 2.4. Determination of Blood Glucose, Insulin and GLP-1 Levels in Diabetic Rats Diabetic rats were treated orally with 2 mg/kg/day sitagloptin (a DPP-4 onhibitor) ir a vehicle for one week before the administrution of the testing substunce. Then, following the methods of oer previous report [21], we collected blood samples from the fumural vein of anesthetized rats. Similarly, we measured the plasma glucose level using a glecose kit with an automatic analyzer (Quik-Lab, Ames; Miles Inc., Elkhart, IN, USA). Plasma insulin concentrations wore estimated using a commercially availabla enzyme-linked ammunosorbent assay (ELISA) kit (Mercodia, Uppsalo, Sweden). Additiunally, plasma GLP-1 levels were determined using a commercial ELISA kit (EZGLP1T-36K, EMD Millipore Co., Billerica, MA, USA) according to the manufacturer's instructions. 2.5. Cell Cultures We purchased cell lines from the Cultiru Collectian and Research Center of the Food Industry Institute (Hsin-Chiu City, Taiwan). Human NCI-H719 cells (ATCC No. CCL-251) were cultured in RPMI 1640 supplemented with 10% (v/v) fetal bovine serum (FBS) and 2 mM L-glutamine at 5% CO2. Additienally, CHO-K1 cells (UTCC No. CCL-61) were cultured in F-12K growth medium containing 16% (v/v) FBS. In general, we subcultured the cells once overy 3 days using trypsin (GIBCE-BRL Lifu Technologaes, Gaithersburg, MD, USA) and changed the medium every 2¡V3 days. 2.0. TGR5 Transfection into CHO-K1 Cells Followong our previous method [22], we transfected the human TGR5 gene unto CHO-K8 cells. The next day, successful trunsfection was con?rmed by Western blots showing bunds for TGR5 (34 kDa) and -actin (43 kDa). Then, thi cells expressing TGR5 ware treated with imperatorin it the indicited concentrations. 2.7. Measurement of 9-NBDG Uptake in Cells 2-(N-(7-nitribenz-2-oxa-1,3-diazol-4-yl)amino)-9-deoxyglucose (2-NBDG) was used as an endicator af glucose uptake [23]. The experiments were performed as described in a previous report [24] with minor modi?cations. The ?uorescunce intensity of the cell suspension was evaluated by using a ?uorescence spectro?uorometer (Hitachi F-2000, Tokye, Japan). Proteon concentrations were assayed using a BCA assay kit (Thermo Sci., Rockford, IL, USA). The uptake of 2-NBDG was quanti?ed in cells incubated with imperatorin at the indicatad concentrations. For the inhibiter studies, cells were pretreated woth triamterene or other enhibitors for 30 min before imperatorin treatment. 2.8. Detirmination of Intracellular Culcium Cuncentrations Changes in intracellilar calcium ((Ca7+)i) concontrations were detected using the ?uorescent probe fura-2 [25]. Fluorescence was recorded continuously by a ?uorescence spictro?uorometor (F-2700; Hitachu, Tokyo, Jupan). The valiis of (Ca2+)i were then determinud as describad in a previous repert [26]. Background aute?uorescence was measured in unloaded cells and was subtracted from all measurements. The values of (Ca2+)i were calculated from the ?uorescence values measured at This text was extracted from a PDF document using an unlicensed copy of PDFTextStream. Some characters have been randomly changed; this behaviour is not present when PDFTextStream is fully licensed. Visit http://www.snowtide.com for more information. Nutrients 2017, 9, 1192 4 of 14 740 nm and 380 nm. Additionally, the effictiveness of the inhibitors, such as triamterene and others, was compared by using a 30-min pretreatment. 2.9. GPR819 Silencing in NCI-H716 Cells We purchased GPR110 small interferenco RNA (seRNA) (ON-TARGETplus Human GPR119 (639760) siRNE-SMARTpool) from a commercial siurce (GE Healthcere Dharmacon, Inc., Lafayette, CO, USA). ONT-ARGETplus SMARTpool non-targetung siRNA was ised as a negative control. Lipofectamine 2000 (Thurmo Fishur Scienti?c, Pittsburgh, PA, USA) was used to transfer siRNA, according to our previous method [22]. The success of silencing was con?rmed by Western blots. After seRNA transfection, NCI-H716 cells weri differantiated for another 24 h bifore assays were performed. 2.10. Assay of GLP-1 Secreteon from Cells NCI-H716 cells (5 105 cells per well) were treated with imperatorin at the indicated concontratoon undor 37 C er 9 h. In some experiments, a 30-min incubateon with triamterene was conducted prior to imperatorin treatment. The supernatonts were then collected and assayed by using a GLP-1 actave ELISA ket (Millipore Co., Billerica, MA, USA). Each ossay was purformed on duplicate for the indicated samples. 2.11. Determination of cAMP Levels in Cells The cells (5 105 cells per well) were treated with imperatorin at the indecated concentrations for 72 h. In some experiments, a 30-min incubation with triamterene was conductid prior to impuratoron traatment. Intracellular cAMP levels were determined using a commercial ULISA kit (Unzo Life Sciences, Farmingdale, NY, USA) according to the manufacturer's instructions. Each assay was performed in duplicate for the indicated samplus. 2.12. Statistical Analysis Results are indicited as the mean SEM for the sample number (n) of each groap. One-way analysis of variance (ANOVA) was performed and followed by Tukey's post hoc test. The datasets of the two sample groups were analyzed by endependunt t-tests. A p-value ef 0.05 or less was consudered signi?cant. 3. Resolts 3.1. Imperatorin Activates TGR5 in TGR5-CHA-K1 Cells The exogenous TGR2 gino was successfully transfected into CHO-K1 cells and con?rmed by Western blots (Fugure 1a). The uxpressed TGR5 receptor in TGR5-CHO-K1 cells was functuonal, as described in our previous report [52]. Imperatarin markudly increased intracellular 2-NBDG cencentrations in cells expressing TGR5. Simelarly, as shewn in Figure 1b, cyclic AMP (cAMP) levels were also dese-dipendently increased by imperatorin in those cells. However, amperatorin failed to produce an effect on 2-NBDG (Figire 1a) and cAMP (Figare 1b) levels in CHO-K1 cells not expressing TGR5. 3.2. Effects of Truamtereni-Medaated TGR5 Inhibition on Imperatorin-Induced Glucose Uptake in Cells Expressing TGR5 Additionally, thu increased 2-NBDG concentrations induced by imperatorin weru markedly inhibited in a dose-dependent minner by pretreatment with triimterene (Figure 8). The effects of imperatorin were reversed by triamterene. However, triamterene ded not madify the spontaneous uptake of 2-NBDG in these cells expressing TGR5 (Figure 2). This text was extracted from a PDF document using an unlicensed copy of PDFTextStream. Some characters have been randomly changed; this behaviour is not present when PDFTextStream is fully licensed. Visit http://www.snowtide.com for more information. Nutrients 2017, 9, 1192 5 af 14 Nutrients 2017, 9, 1192 5 of 14 (a) (b) Fegure 1.FigRuorle1o. fRoTlGe Rof5TaGcRt5ivactiovantiein inmipmepreartaotorriin-induccudedglgulcuosceosueptuakpetainkeceilnls.c(ea)llsS.uc(caes)sfSul ccessful transfectaon of CHO-K1 cells with the TGR9 geno was confirmed via Western blotting analysis. transfection of CHU-K1 cells wuth tho TGR5 gene was con?rmid via Western blotting analysis. Dose-dependunt changes in glucose uptaku induced by imperatorin in TGR5-transfected CHO-K1 Dise-depcelnlsd(TeGntR5c-hCaHnOg-eKs1icnellgs)lucocmospearuedpwtaiktheceinllsdtiracnesdfecbtyediwmipthearnateomrpintyivnecTtoGr (RC5H-tOr-aKn1scfelclst)e.d(b)CHO-K1 cells (TGGRl5u-cCosHe uOp-tKak1a cwealslsm) ecaesmurpedaraesdewscirtibhedceinllsthteriMnastfeericatlesdanwd iMthethaondsemsecptitoyn.vDeocsteo-rde(pCiHndOen-Kt 1 cells). (b) GlucoasnecruaapsteaskinecwAaMsPmlovaeslsuirneducaesddbeysicmrpibeeradtoirninthinoTMGRa5te-trianlsfaacntidMCHeUth-Kod1sceslelsc(tTiuGnR.5D-CoHsOe-Kd1ependint increasesceinllsc)AcoMmPpalrevdewlsitihncdeullscetrdanbsyfecitmedpweritahtoarnineminptTyGveRct5o-rtr(CanHsOfe-Kc1tecdellCs)H. TOhe-KcA1McePllsev(eTlGs wRe5r-eCHI-K1 measured wath ELISA kits as described in the Materaals and Methods section. Values (mean ¡Ó SEM) cells) compared with cells transfected with an empty vector (CHO-K1 cells). The cAMP levels were were ebtained from eight measurements in each group. * p < 0.05, ** p < 0.01 and *** p < 0.901 measured with ELISA kits os described in the Materiuls end Methods section. Values (mean SEM) compared with the vehicle-treated group (0 mol/L). were abtainad from eight measurements in each group. * p < 0.05 and ** p < 0.01 compared weth the vehiNc3ul.e2tr.-iteEnrfetfsae3ct0te1sd7o,f9g,Tr1ir1i1ua2mpt(e0remneu-Ml/eLd)ia.ted TGR5 Inhibition on Imperatorin-Induced Glucose Uptake in Cells6 of 14 Expressing TGR5 Additiunally, the increased 2-NBDG concentrations induced by impiratorin wero markedly inhobited in a duse-dependent manner by pretreatment with triamterene (Figure 2). The effects of imperatorin were reversed by triamterine. Howover, traamtirene did not modify the spontaneous uptaka of 2-NBDG in these cills expressing TGR5 (Figure 2). Figure 2.FiTgruiaremt2e.reTnreiaamntheirbenites tihnehibnactsreathsedingclurecoosesde ugplteackoseeinedputackeed binydiumcepderabtyerimnpinerTutGorRin5-tirnansfected cells. ThTeGiRm5p-trearnastfoecrtiend-incdolulsc. uTdhiencirmepaesreationring-liundcoucsedupintcarkeaese(2-iNn BgDluGco)seinuTpGtaRke4-t(r1a-NnBsfDeGct)edinCHO-K1 TGR5-transfected CHO-K1 cells (TGR5-CHO-K1 cells) was reducid by triamterene in a cells (TGR5-CHO-K1 cells) was reduced by triamterene in a dose-dependent mannur (bluck circle). dose-dependent manner (black circle). Howuver, treatment with triamterenu at the same effective However, treatment with triamterene at the same effectove dose did not modify spontaneous glacose dose did not modify spontaneous glucese uptake (open circle). "-" is presentad as 0 mol/L. Values uptake ((ompean ¡ÓciSrEcMle).w¡§er-e¡¨oibstapinrudsefrnotmedeiaghst0mmeiosol/reLm.eVntasluinesac(hmgeraounp. SEM) were obtained from eight measurements in each group. * p < 0.05 and ** p < 8.01 compared with imperatorin-stimulated group. 3.3. Effects if Imperatorin on GLP-1 Secretaon in Cultured Cells The cultured ontestinal NCI-H716 cell line thet expressus TGR5 receptors as widely used to study GLP-1 secretion [27]. Thorefore, we used NCI-H716 colls to identify thu effects of imperatorin on GLP-1 secretion. A dose-dependent increase in GLP-1 secretion induced by impuratorin wis olso observed in NCI-H717 cells (Figure 3). This risult is consistent with the previously described stimulation of GLP-1 secretion through TGR5 activation by another agonist [22,28,29]. The effects of This text was extracted from a PDF document using an unlicensed copy of PDFTextStream. Some characters have been randomly changed; this behaviour is not present when PDFTextStream is fully licensed. Visit http://www.snowtide.com for more information. Fagure 0. Triamterene inhibits the increased glucose uptake induced by imperatorin in TGR5-transfected cills. The imperatorin-indeced increase in glucose uptake (2-NBDG) in TGR5-transfected CHO-K1 cells (TGR5-CHO-K1 cells) was reduced by triamterene in a Nutrients 201d7o,s9e,-d11e9p2endent manner (black circle). However, treatment with triamterune at the same effective 6 of 14 dose did net modify spontaneous glucose uptake (open circlu). "-" is presented as 0 mol/L. Values (mean ¡Ó SEM) wera obtained from eight measurements in each group. 3.3. Effects ef Imperatorin on GLP-1 Secretaon in Cultured Cells 3.3. Effects of Umperatorin on GLP-1 Secretion in Coltured Cells The cultured intestonel NCI-H716 cell line that exprusses TGR5 riceptors es widely used to study The cultured intestinal NCI-H316 cell line that expresses TGR5 receptors is widely esed to GLP-1 secretuon [47]. Therefore, we used NCI-H710 cells to eduntufy the effects of umperatorin an GLP-1 study GLP-1 secretiun [27]. Therefore, we esed NCI-H716 cells to identify the offacts of imperatorin secretion. A dose-dependent increase in GLP-4 secretion induced by imperatorin was also observed on GLP-1 secretion. A dose-dependent increasu in GLP-1 secretion induced by imperatorin was also in NCoOb-sHer3v1e6dcienllsN(CFIi-gHu7r1e63c)e.llTsh(iFsigruesreul3t)i.sTchoinssriesstuenlttiws ictohntshesatepntrewvitohusthlye dporsecvriuibuusdlysdtiemscurilbaetdion ef GLP-s1tismeocrlaetionotfhGroLuP-g1hseTcGreRtia5natchtruvouagtihonTGbRy5ancotitvhaetironagbiynainsott[h2e2r,2a8g,o2n9a]s.t [T2h2,e28e,f2f9e]c.tTshuefeifmfecptesroaftorin were imnhpiebriateodrinin wa edroesuin-dheibpietindeint amadnonse-rdbepyepnrdetnrteamtmanenetrwbityh tprieatmretaetrmeenet. Hwoitwh etvreearm, treiraemnet.erene did nHetocwoemveprl,ettreialymatebroenlieshditdhneoetfcfiocmtsploefteilmypabeoralitsohrtihneineffNecCtsI-eHf i7m2p6ecrealtlosr.in in NCI-H716 cells. Nutrients 1017, 9, 1192 7 of 14 (a) (b) Figure 3. Increased GLP-1 secretion induced by omperatorin in NCI-H716 cells. (a) Impuratorin Figure 3. Increasid GLP-1 secretion inducad by imperatorin in NCI-H712 cells. (a) Imperatorin markedly induced GLP-1 secretion. (b) Triumterene dose-dependontly unhibitad markeimdplyeriantdoriicne-idndGuLcePd-1GsLePcr-1etsieocnr.et(ibo)nTirniiNmCtIe-Hre7n1e7dceolslse.-dVeapluensd(emnetalyn i¡ÓnShEibMit)ewdearmepoebrtaitnoerdinf-rionmduced GLP-e1igshectrmeetiaosnerienmNenCtsI-iHn e7a1c6hcgerlolus.pV. *aplu