Urolithin A production drives the effects of pomegranate on the gut microbial metabolism of bile acids and cholesterol in mild dyslipidaemic overweight and obese individuals

Food & Func ion PAPER Ci e his: Food Func ., 2024, 15, 2422 Recei ed 14 h No embe 2023, Accep ed 22nd Janua y 2024 DOI: 10.1039/d3 o05014a sc.li/ ood- unc ion U oli hin A p oduc ion d i es he effec s o pomeg ana e on he gu mic obial me abolism o bile acids and choles e ol in mild dyslipidaemic o e weigh and obese indi iduals Ad ián Co és-Ma ín, † a,b Ca los E. Iglesias-Agui e, † a Alicia Ma ín, a Ma ía Romo-Vaque o, a Fe nando Vallejo, a Juan Ca los Espín a and Ma ía Vic o ia Selma * a The me abolism o (poly)phenols and some hos me aboli es, including bile acids (BAs) and choles e ol, a ies among indi iduals depending on hei gu mic obio a. The gu mic obial me abolism o ellagi an- nins (ETs) and ellagic acid (EA) p oduces u oli hins (U os), yielding h ee me abo ypes wi h quan i a i e and quali a i e diffe ences based on dissimila U o-p oducing p ofiles (UM-A, UM-B, and UM-0, i.e., non- p oduce s). P e ious animal s udies demons a ed ha polyphenols impac BAs and choles e ol mic obial me abolism, bu da a on hei effec s in humans and da a ega ding he in e -indi idual a iabili y o hese me abolic con e sions a e scan . We e alua ed whe he UMs, as dis inc i e unc ional gu -mic obiome signa u es, could de e mine he po en ial effec o a pomeg ana e ex ac (PE) ich in ET-EA on he me abolism o BAs and choles e ol in mild dyslipidaemic o e weigh -obese indi iduals, wi h possible con- sequences on hos -lipid homeos asis and gu heal h. A he baseline, UM-B p esen ed he highes le els o aecal o al and seconda y BAs and cop os anol, sugges ing ha he lipid abso p ion capaci y and gu cy o oxic isk could be augmen ed in UM-B. PE in ake significan ly educed aecal cop os anol and BA p oduc ion, especially seconda y BAs, and modula ed he gu mic obiome, educing he gu cy o oxic isk, especially in UM-B indi iduals. The lowe ing o aecal mic obial cop os anol and BAs and some BA- me abolising bac e ia was quan i a i ely co ela ed wi h U o concen a ions, mainly aecal U o-A. This sugges s ha PE consump ion could exe ca dio ascula and gu p o ec ion h ough U o-A p oduc ion as a di ec d i e o he effec s and indi ec ly by educing he Co iobac e iaceae amily and BA pool, known ac o s in ol ed in he gu abso p ion o lipids. In oduc ion The human in es ine is colonised by se e al mic obial species ha in e ac wi h die a y and hos -de i ed molecules in he in es ine, signi ican ly con ibu ing o hos physiology. 1 The gu mic obio a can p oduce diffe en me aboli es om se e al die a y compounds, which exe biological ac i i y and play an impo an ole in main aining gu and me abolic heal h. This is he case o ellagi annins (ETs) and ellagic acid (EA) om pomeg ana e, nu s, and be ies, which a e ex ensi ely me ab- olised by he human gu mic obio a o p oduce diffe en u o- li hins (U os) wi h po en ial heal h bene i s. 2 The human gu mic obio a can also ans o m hos -de i ed molecules, such as choles e ol and bile acids (BAs), which a e impo an signal- ling me aboli es in he hos . Indeed, choles e ol and BAs sig- ni ican ly impac hos lipid homeos asis. 3,4 Howe e , he e is a consis en human in e -indi idual a iabili y in he mic obial me abolism o die a y and hos -de i ed molecules depending on he subjec s’gu mic obial signa u es. 5–7 One o he main diffe ences be ween he me abolic p o iles associa ed wi h ET and EA me abolism is he inal U os p o- duced (u oli hin me abo ypes, UMs). UM-A indi iduals only yield u oli hin A (U o-A) as he inal ca aboli e. In con as , UM-B subjec s p oduce U o-A and, dis inc i ely, also isou o- li hin A (IsoU o-A) and u oli hin B (U o-B). Finally, UM-0 indi- iduals canno p oduce in e media e and inal U os in he ET– EA me abolic pa hway. 8 Diffe ences in he human gu mic obial ecologies associa ed wi h UMs we e p e iously desc ibed. 9 UMs ha e also been p oposed as po en ial †These au ho s con ibu ed equally o his wo k. a Labo a o y o Food & Heal h, Resea ch G oup on Quali y, Sa e y and Bioac i i y o Plan Foods, CEBAS-CSIC, Campus de Espina do, Mu cia 30100, Spain. E-mail: [email p o ec ed]s b APC Mic obiome I eland & School o Mic obiology, Uni e si y College Co k, T12 YT20 Co k, I eland 2422 |Food Func .,2024,15,2422–2432 This jou nal is © The Royal Socie y o Chemis y 2024 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online View Jou nal | View Issue ca dio ascula disease isk bioma ke s because he blood lipid p o ile was epo ed o be associa ed wi h UMs. 10,11 Howe e , he link o his associa ion emains unknown so a . On he o he hand, ecen s udies ha e also e ealed in e -indi idual diffe ences in he mic obial me abolism o choles e ol and conjuga ed p ima y BAs, which migh exe a much wide ange o biological ac i i ies han hose ini ially ecognised. 4,12–14 In his ega d, seconda y me aboli es o aecal choles e ol and BAs, p oduced by he gu mic obio a, may p omo e heal h o a ou disease de elopmen depending on he quan i y and ype p oduced. Speci ically, con inuous exposu e o ele a ed luminal le els o seconda y BAs and choles e ol mic obial me aboli es migh inc ease gu pe - meabili y and suscep ibili y o in es inal in lamma ion and colo ec al cance . 3 In addi ion o he oxic effec s o seconda y s e oids pe se, he concen a ion o pu e ac i e compounds could also inc ease gu oxici y due o he elease o amino acids (glycine and au ine), which a e he p oduc s o he mic obial hyd olysis o BAs. 12 The possibili y o modula ing he mic obial me abolism o die a y and hos -de i ed molecules by a ge ing he gu mic o- bio a composi ion o imp o e human heal h has only ecen ly s a ed o be explo ed. 12,14 Indeed, some animal s udies ha e begun o e alua e he impac o some polyphenols on he gu mic obial ans o ma ions o BAs and choles e ol, as e iewed p e iously. 12,14 These esul s sugges ha phenolic- ich die s hampe he con e sion o p ima y BAs o seconda y BAs and choles e ol o cop os anol. 12 Howe e , u he s udies a e c ucial o es ablish whe he he effec s obse ed in animal s udies can be eplica ed in humans, and mo e da a ega ding he ole o he gu mic obial me abolism o polyphenols in he in e -indi idual diffe ences o hese me abolic con e sions a e needed. In he POMEca dio ial, we desc ibed he UM-dependen effec s on ca dio ascula isk ma ke s a e consuming a pomeg ana e ex ac (PE) in o e weigh -obese indi iduals wi h mild dyslipidemia. 10 In he p esen s udy, we aimed o e alu- a e (i) he effec s o PE in ake in hese olun ee s on he gu mic obial me abolism o BAs and choles e ol, (ii) whe he hese effec s o PE could also be U o-dose-dependen , (iii) he possible link be ween BAs and choles e ol me abolism and he gu mic obio a signa u e associa ed wi h UMs, and (i ) es ab- lish a possible link be ween BAs and choles e ol me abolism, U o p oduc ion capaci y, mic obial ela i e abundances and ca dio ascula isk ma ke s imp o emen . To his pu pose, he esul s we e analysed conside ing he pa icipan s as a single g oup and a e clus e ing he olun ee s acco ding o hei UMs (UM-A, UM-B, and UM-0). Ma e ials and me hods Chemical and eagen s S anda ds o cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), li hocholic acid (LCA), isoli hocholic acid (iso-LCA), u sodeoxycholic acid (UDCA), cop os anol, choles e ol, 5α-choles ane, sodium ace a e, and sodium hyd oxide we e pu chased om Sigma-Ald ich (S Louis, MO, USA). U o-A, IsoU o-A, U o-B, and hei de i ed phase-II conju- ga ed me aboli es we e chemically syn hesised and pu i ied by Villapha ma Resea ch S.L. (Pa que Tecnológico de Fuen e Álamo, Mu cia, Spain). HPLC-g ade ace oni ile, hexane, cyclo- hexane, me hanol, and e hanol we e pu chased om JT Bake (De en e , The Ne he lands), and o mic acid was pu chased om Pan eac (Ba celona, Spain). Milli-Q sys em (Millipo e Co p., Bed o d, MA, USA) ul apu e wa e was used h oughou he s udy. PE was encapsula ed in iden ical ha d gela in cap- sules supplied by Labo a o ios Admi a S.L. (Alcan a illa, Mu cia, Spain), ollowing he Eu opean Union’s good manu- ac u ing p ac ices. Capsules we e bo led wi h a speci ic blind code o pa icipan s and esea che s. Each capsule (450 mg o PE) con ained 160 mg o phenolic compounds, as p e iously desc ibed. 10 The phenolic con en o PE capsules was analyzed by HPLC-DAD-ESI-MS/MS, as p e iously desc ibed, and was as ollows: 54.5 ± 1.2 mg punicalagins, 45.4 ± 0.9 mg ee ellagic acid, 2.7 ± 0.2 mg punicalin, 2.6 ± 0.1 mg ellagic acid-hexoside and 1.2 ± 0.1 mg ellagic acid-pen oside. 10 A e acid hyd olysis, he o al ellagi annin con en pe capsule was 72.9 ± 1.1 mg ee ellagic acid, 69.3 ± 0.9 mg gallagic acid dilac one, 10.3 ± 0.1 mg sanguiso bic acid, 3.1 ± 0.04 mg gallic acid, 2.9 ± 0.2 mg punicalin and 1.9 ± 0.2 mg aloneic acid dilac one. 10 S udy design and subjec s U ine, aecal and blood samples we e ob ained in a p e ious ial (POMEca dio), included in he Eu opean ‘Bacchus P ojec ’(FP7-KBBE-2012), egis e ed a clinical ials.go (NCT02061098) and app o ed by he Spanish Na ional Resea ch Council’s Bioe hics Commi ee (CSIC, Mad id, Spain) ( e e ence 312090), ollowing he e hical guidelines ou lined in he Decla a ion o Helsinki (1975) and i s amendmen s. B ie ly, ha s udy was designed o e alua e he effec o PE consump ion in subjec s wi h mild hype lipidemia and po en- ially inc eased gu dysbiosis. O e weigh -obese subjec s (BMI >27kgm −2 ) o e 40 yea s, wi h no diagnosed ch onic disease, we e included. Fi y pa icipan s we e ec ui ed, and 49 com- ple ed he p o ocol. De ailed in o ma ion abou he design is shown elsewhe e. 10 B ie ly, he s udy was a 6-mon h ollow-up, double-blind, c osso e , dose- esponse, andomized, and placebo-con olled ial. The s udy consis ed o wo es phases wi h each es phase las ing o 3 weeks wi h a 3-week washou pe iod be ween each es phase and a 3-week washou pe iod a e he inal es phase. In he i s es phase, pa icipan s consumed ei he one capsule o PE o placebo daily in a c oss- o e ashion. In he second es phase, he same design was ollowed bu olun ee s inges ed ei he ou capsules o PE o placebo daily. 10 In he p esen s udy, conside ing he absence o effec s a e consuming he placebo and he effec s obse ed wi h PE a he maximum dose es ed, samples om wo ime poin s we e analysed: a e consuming one capsule o PE daily o 3 days and ou capsules daily o 3 weeks. Food & Func ion Pape This jou nal is © The Royal Socie y o Chemis y 2024 Food Func .,2024,15,2422–2432 | 2423 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online Sampling p ocedu es As desc ibed p e iously, 10 pa icipan s p o ided u ine and aecal samples on he mo ning o each isi . Blood samples we e collec ed in acu aine s be ween 8 : 00 and 9 : 00 a.m. o minimise ci cadian a ia ions in he supine posi ion a e 20 min o es . Se um and plasma we e ob ained. All samples we e s o ed a −80 °C un il u he analysis. 10 Blood lipid p o ile and u oli hin me abo ype de e mina ion Se obiochemical a iables we e measu ed as p e iously epo ed, including Tchol ( o al choles e ol), HDLc (high- densi y lipop o ein choles e ol), LDLc (low-densi y lipop o ein choles e ol), VLDLc ( e y low-densi y lipop o ein choles e ol), TG ( iglyce ides), ApoA-1 (apolipop o ein A-1), ApoB (apolipo- p o ein B), oxLDLc (oxidised LDLc), IDLc (in e media e-densi y lipop o ein choles e ol), and HOMA-IR (Homeos a ic Model Assessmen o Insulin Resis ance). 10 U ine and aecal samples we e p ocessed and analysed by using an ul a-pe - o mance liquid ch oma og aph coupled o a quad upole- ime-o - ligh mass spec ome e using an elec osp ay in e - ace (UPLC-ESI-QTOF-MS) o de e mine u oli hins as desc ibed elsewhe e. 10,15 A baseline, u ine samples a e consump ion o 1 capsule o 3 days we e used o s a i y olun ee s by UMs. 10 Faecal choles e ol and cop os anol analyses Choles e ol and i s mic obial me aboli e cop os anol we e ex ac ed om aecal samples as p e iously desc ibed wi h mino modi ica ions. 16 B ie ly, 50 µL o dilu ed aeces (1 : 10 wi h H 2 O Milli-Q) we e spiked wi h 20 µL o 5α-choles ane (5 mmol L −1 ) as an in e nal s anda d o s e ol de e mina ion. The hyd olysis p ocess was ca ied ou wi h 5 mL o eshly p epa ed me hanolic sodium hyd oxide (1 mol L −1 , 90% me hanol) o 60 min a 70 °C. The neu al s e ols we e ex ac ed wi h 4 mL o cyclohexane using a o ex mixe o 20 min. Samples we e cen i uged a 3500g o 5 min, and he o ganic phase was sepa a ed and e apo a ed in a speed acuum concen a o (Sa an SPD121P, The mo Fishe Scien i ic, Mad id, Spain). The d ied samples we e hen econ- s i u ed in 1 mL o hexane and il e ed h ough a 0.22 µm poly- inylidene luo ide (PVDF) il e (Millipo e, The mo Fishe Scien i ic, Mad id, Spain). The aecal choles e ol and cop os a- nol le els we e hen de e mined by gas ch oma og aphy (GC) coupled o mass spec ome y (MS). 16 An aliquo o 1 µL o sample was injec ed in o he GC-MS sys em. Sepa a ion was pe o med using an Agilen 7890A GC equipped wi h an Agilen 5975C MS de ec o . Choles e ol and cop os anol we e sepa a ed on an HP-5MS 30 m x 0.25 mm (i.d.) capilla y column (Agilen ) wi h helium as he ca ie gas (1 mL min −1 ). Injec ions we e ca ied ou in a spli less mode a 280 °C. The o en empe a u e was i s held a 150 °C o 5 min, and hen inc eased o 280 °C a a a e o 20 °C min −1 , which was main- ained o 10 min. Elec on impac mass spec a (EI-MS) we e eco ded wi h an ionisa ion ol age o 70 eV and a empe a u e sou ce o 230 °C. The acquisi ion was pe o med in scanning mode. Iden i ica ion o choles e ol and cop os anol was based on he e en ion ime o s anda d compounds and wi h he assis ance o he NIST 08 lib a ies. Faecal BA de e mina ion BAs we e ex ac ed om aecal samples ollowing a alida ed me hod wi h mino modi ica ions. 17 Fi y µL o dilu ed aeces (1 : 10 wi h H 2 O Milli-Q) we e esuspended in 450 µL o cold sodium ace a e buffe (50 mmol L −1 , pH 5.6). 1.5 mL o pu e e hanol was added, and he samples we e shaken a 1200 pm o 15 min a 25 °C (The moCell Mixing Block MB-102, BIOER, Hangzhou, China). A e cen i uga ion, he supe na an was dilu ed ou imes wi h H 2 O Milli-Q and applied o a Ch oma ix C18 S ca idge (270 mg) (Mache ey-Nagel, Dü en, Ge many), which was p e iously condi ioned wi h 3 mL o e hanol and hen 3 mL o wa e Milli-Q. Following he sample loading, he ca idge was washed wi h 3 mL o Milli-Q wa e o emo e wa e -soluble impu i ies. Finally, BAs we e elu ed wi h 3 mL o e hanol and d ied using a speed acuum concen- a o . The esidue was esuspended in 200 µL o me hanol and il e ed h ough a 0.22 µm PVDF il e . The samples we e dilu ed in me hanol (1 : 10) be o e injec ion in o an Agilen 1290 In ini y UPLC sys em coupled o he 6550 Accu a e-Mass quad upole TOF mass spec ome e (Agilen Technologies, Waldb onn, Ge many) using an elec osp ay in e ace wi h Je S eam echnology. Sepa a ion was achie ed using a e e sed- phase Po oshell 120 EC-C18 column (3 × 100 mm, 2.7 μm) (Agilen Technologies, Waldb onn, Ge many) ope a ing a 25 °C. The mobile phases used we e wa e : o mic acid (99.9 : 0.1 / ; phase A) and ace oni ile : o mic acid (99.9 : 0.1 / ; phase B). BAs we e sepa a ed using he ollowing g adien condi ions: 0–4 min, 50–90% phase-B; 4–7 min, 90–99% phase-B; 7–10 min, 99% phase-B. Finally, he sys em e u ned o he ini ial condi ions (50% phase-B) o 1 min, and he column was e-equilib a ed o an addi ional 1 minu e. The low a e was cons an a 0.4 mL min −1 , and he injec ion olume was 2 µL. The ope a ing condi ions we e as ollows: gas empe a u e –150 °C, d ying ni ogen gas –14 L min −1 , nebulise p essu e –40 psi, shea h gas empe a u e –350 °C, shea h gas low –11 L min −1 . Spec a we e acqui ed in single MS mode wi h a mass ange be ween m/z50 and 1500, nega- i e pola i y, and a TOF spec al acquisi ion a e o 1.5 spec a pe second. Be o e he analysis, he ins umen was calib a ed ex e nally by injec ing a mix u e o e e ence compounds (Tuning Mix) o ensu e mass accu acy du ing he MS analyses. Con inuous in e nal calib a ion was pe o med du ing ana- lyses using he signals a m/z112.9855 and m/z1033.9881. Da a we e acqui ed using he Mass Hun e Wo ks a ion so - wa e ( e sion B.08.00, Se ice Pack 1, Agilen Technologies) and p ocessed using he Mass Hun e Quali a i e Analysis so - wa e ( e sion B.08.00, Se ice Pack 1, Agilen Technologies). All samples we e injec ed in he same ba ch, and he o de o sample injec ion was andomised o p e en sample bias. The quan i ica ion o BAs was de e mined by hei in e pola ion wi hin he calib a ion cu e ob ained by he injec ion o hei Pape Food & Func ion 2424 |Food Func .,2024,15,2422–2432 This jou nal is © The Royal Socie y o Chemis y 2024 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online co esponding s anda ds by peak a ea in eg a ion o hei ex ac ed ion ch oma og ams. Gu mic obio a analysis Gu mic obio a analysis was pe o med by 16S DNA sequen- cing ( a ge ing he V3–V4 hype a iable egions) on a MiSeq- Illumina pla o m (FISABIO sequencing se ice, Valencia, Spain) a e ex ac ing bac e ial DNA om he aecal samples as de ailed elsewhe e. 18 The ob ained da a we e p ocessed o cons uc Ope a ional Taxonomic Uni (OTU) ables a a 97% simila i y h eshold o classi y axa a he genus le el, de e - mine he phyla and amilies o which hey belong, and es i- ma e he ela i e abundance o each OTU, as p e iously desc ibed 19 bu using he las upda ed RDP classi ie da abase 2.13. Fu he mo e, in he p esen s udy, he indi iduals we e clus e ed by UMs and he diffe ences in ela i e abundances be o e and a e PE consump ion we e e alua ed h ough he linea disc iminan analysis effec size (LE Se) algo i hm using he online in e ace Galaxy (h ps://hu enhowe .sph.ha a d. edu/galaxy/ oo ). S a is ical analysis S a is ical analyses we e pe o med using he SPSS so wa e .28.0 (SPSS Inc., Chicago, IL, USA), and plo s we e gene a ed using G aphPad P ism 7.0 (G aphPad So wa e, San Diego, CA, USA). The Shapi o–Wilk es was used o examine da a no mal- i y. Diffe ences be ween wo dependen g oups (i.e., be o e e sus a e PE consump ion) we e analysed using he pai ed S uden ’s - es o Wilcoxon signed ank es when da a dis i- bu ion was no mal o non-no mal, espec i ely. The K uskal– Wallis es , ollowed by Dunn’s es , was used o compa e mo e han wo g oups (i.e., UM-A s. UM-B s. UM-0), and he co es- ponding pai wise compa isons we e adjus ed by he Bon e oni co ec ion o mul iple es s. Spea man’s ank co - ela ion was pe o med o s udy associa ions be ween a iables ( aecal choles e ol and BAs, blood lipid p o ile, and speci ic bac e ial g oups). A mul i a ia e model was applied o e alua e he in e ac ion effec be ween UM and ime poin s on BA con- cen a ion. S a is ical signi icance was se a *P< 0.05. Resul s Effec o PE in ake on he phenolic me aboli e p o ile Fig. 1 shows he in e -indi idual diffe ences in U o-A concen- a ions in he aeces and u ine om he o e weigh -obese indi- iduals a e 3-week in ake o ou daily capsules o PE. The aecal U o-A concen a ion ange was highe in he aeces om UM-A (4–423 µg g −1 aeces) han in he UM-B g oup (2–180 µg g −1 aeces), whe eas he u ina y U o-A concen a ion ange was simila in he UM-A (0.1–86 mg pe 24 h in u ine) and UM-B (0.1–89 mg pe 24 h in u ine) g oups (Fig. 1A and B). Effec o PE in ake on aecal BAs and choles e ol me abolism Fig. 2 shows he concen a ion o he aecal BAs and s e ols analysed in he o e weigh -obese indi iduals a baseline and a e a 3-week in ake o 4 capsules o PE. A baseline, second- a y BAs (DCA and LCA) concen a ion p edomina ed in he aeces compa ed o p ima y BAs (CA and CDCA) (Fig. 2B and C). The o al BAs, seconda y BAs, cop os anol and he cop os- anol/choles e ol a io we e signi ican ly lowe in he UM-0 han in he UM-B g oup (Fig. 2A, F, G, K and L). In con as , he p ima y BA concen a ion and aecal choles e ol did no diffe among UMs a baseline (Fig. 2B and J). A e PE consump ion o 3 weeks, a signi ican educ ion o o al BAs, p ima y BAs, seconda y BAs, CA, CDCA, DCA, LCA, UDCA, iso-LCA and cop os anol was obse ed when all olun ee s we e conside ed and in he UM-A and UM-B sub- g oups (Fig. 2). The cop os anol/choles e ol a io was also educed upon he 3-week in ake o PE when all olun ee s we e conside ed and in he UM-A g oup, bu was no signi i- can ly educed in he UM-B g oup (Fig. 2L). In con as , hese pa ame e educ ions we e no obse ed a e PE in ake in he UM-0 g oup (Fig. 2). In addi ion, aecal choles e ol was no sig- ni ican ly educed, al hough in he case o he UM-B g oup, i was nea ly s a is ically signi ican (P= 0.056) (Fig. 2J). Effec o PE in ake on he gu mic obio a composi ion Linea disc iminan analysis (LDA) om LE Se analyses showed an inc ease in he ela i e abundance o some bac- e ial g oups, including he well-known bu y a e-p oducing bac e ia Odo ibac e ( amily Odo ibac e aceae), Dysosmobac e ( amily Oscillospi aceae) and Pep oniphilus ( amily Fig. 1 Faecal (A) and u ina y (B) u oli hin A (U o-A) in u oli hin me abo- ype A (g een plo s) and u oli hin me abo ype B ( ed plo s) olun ee s a e 3-week in ake o ou daily capsules o PE. *Es ima ed alues ha we e calcula ed conside ing ha 1.3 g is he mean exc e ion o c ea i- nine o e a 24-hou pe iod. Food & Func ion Pape This jou nal is © The Royal Socie y o Chemis y 2024 Food Func .,2024,15,2422–2432 | 2425 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online Pep oniphilaceae) in all olun ee s a e 3 weeks o PE con- sump ion (Fig. 3A). O he g oups also inc eased a e PE con- sump ion, such as he p opiona e-p oducing gene a Phocaeicola ( amily Bac e oidaceae) (Fig. 3A). On he o he hand, PE consump ion dec eased he abundance o he BA- deconjuga ing bac e ia in ol ed in he amino acid hyd olysis om p ima y BAs such as Rombou sia ( amily Clos idiaceae), as well as bac e ia in ol ed in p ima y BA me abolism o sec- onda y and e ia y BAs, such as Sa cina ( amily Clos idiaceae) and he amily Co iobac e iaceae (class Co iobac e iia) (Fig. 3A). When he olun ee s we e s a i ied acco ding o hei UMs, he modula ion o he gu mic obio a by PE diffe ed depending on he UM (Fig. 3). Fo example, bac e ial g oups ha inc eased by PE in all olun ee s g oup (Fig. 3A) coincided wi h hose obse ed in UM-A olun ee s (Fig. 3B) bu no in UM-B olun ee s, whe e no bac e ia inc ease was obse ed (Fig. 3C). In addi ion, he genus Fla oni ac o inc eased a e PE consump ion only in he UM-A g oup Fig. 2 Faecal BAs (A-I) and s e ols (J-L) a baseline (blue ba s) and a e consuming PE o 3 weeks ( ed ba s) in all olun ee samples (“All”,n= 49) o in hose g ouped by u oli hin me abo ypes: UM-0 (me abo ype 0, n= 6), UM-A (me abo ype A, n= 29) and UM-B (me abo ype B, n= 14). *Significan diffe ences (P< 0.05). Pape Food & Func ion 2426 |Food Func .,2024,15,2422–2432 This jou nal is © The Royal Socie y o Chemis y 2024 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online (Fig. 3B). In con as , some bac e ial g oups educed by PE in all olun ee s such as Hyd ogenoanae obac e ium,Anae o us is and he amily Co iobac e iaceae (Fig. 3A) coincided wi h hose educed in UM-B (Fig. 3C), bu no wi h hose o UM-A, whe e o he bac e ia, such as he well-known bile- esis an and BA- me abolising bac e ia Acine obac e ( amily Mo axellaceae), Pa imonas ( amily Pep oniphilaceae), Paeniclos idium ( amily Clos idiaceae) and unclassi ied-En e obac e iaceae we e educed (Fig. 3B). In addi ion, BA-deconjuga ing and BA-me abolising bac e ia, such as Collinsella ( amily Co iobac e iaceae), dec eased a e PE consump ion, bu only in he UM-B g oup (Fig. 3C). In he case o he UM-0 g oup, he co esponding LDA sco e om LE Se analyses showed ha PE consump ion did no signi ican ly modi y he gu mic obio a composi ion (Fig. 3D). Associa ions be ween u oli hins, aecal BAs and s e ols, gu mic obio a, and clinical se obiochemical a iables The co ela ion analysis o quan i a i e a iables e ealed ha bo h he o al U os and U o-A we e nega i ely co ela ed wi h o al, p ima y, seconda y and e ia y BAs (Fig. 4A). Acco dingly, CA and CDCA (p ima y BAs), DCA and LCA (sec- onda y BAs), and UDCA and Iso-LCA showed nega i e co e- la ions wi h aecal o al U os and U o-A (p esen in UM-A and UM-B). U ina y and aecal U o-B (only p esen in UM-B) co e- la ed nega i ely only wi h UCDA. The e o e, he aecal U o-A concen a ion was he only me aboli e co ela ed wi h all he aecal BAs analysed. Cop os anol was also nega i ely associa ed wi h aecal U o-A, unlike aecal choles e ol and he cop os a- nol/choles e ol a io (Fig. 4A). In con as , u ina y and aecal U o-B nega i ely co ela ed wi h aecal choles e ol. The co e- la ion analysis o aecal mic obio a ela i e abundances e ealed ha some disc imina ing gene a upon PE consump- ion co ela ed wi h mic obial me aboli es (Fig. 4B). Fo example, some bac e ia ha inc eased a e PE consump ion (Fig. 3), such as he p opiona e-p oducing Phocaeicola, he bu y a e-p oducing Odo ibac e as well as Pa abac e oides, we e posi i ely co ela ed wi h aecal U o-A, and nega i ely co e- la ed wi h p ima y, seconda y, e ia y, and o al BAs (Fig. 4B). In con as , some BA-deconjuga ing and BA-me abolising bac- e ia dec eased a e PE consump ion (Fig. 3) and co ela ed wi h he educ ion o BAs, especially seconda y and e ia y BAs (Fig. 4B). These bac e ia included a BA-deconjuga ing genus like Rombou sia, and bile- esis an and BA-me abolising bac e ia like Sa cina ( amily Clos idiaceae), Acine obac e ( amily Mo axellaceae), Pa imonas ( amily Pep oniphilaceae) and Collinsella, as well as i s amily (Co iobac e iaceae) and class (Co iobac e iia). Fu he mo e, hese bac e ia we e nega- i ely associa ed wi h aecal U o-A, bu his associa ion did no each s a is ical signi icance in he case o Rombou sia (P= 0.053) and Pa imonas (P= 0.104). Faecal cop os anol was also co ela ed wi h some disc imina ing bac e ia o PE consump- ion. Fo example, bac e ial g oups ha inc eased a e con- suming PE, such as Pa abac e oides, posi i ely co ela ed wi h aecal U o-A and nega i ely co ela ed wi h cop os anol. In con as , some bac e ia ha dec eased a e PE consump ion, such as Sa cina,Anae o us is,Co iobac e iaceae, and Co iobac e iia, posi i ely co ela ed wi h cop os anol concen- a ion (Fig. 4B). The se obiochemical a iables 10 co ela ed wi h some mic obial me aboli es, including hose de i ed om BA me abolism. Thus, p ima y BAs posi i ely co ela ed wi h se um Tchol, LDLc, non-HDLc, VLDLc, TG, Apo-B, small HDLc, small and la ge LDLc, insulin and HOMA-IR (Fig. 4C). Acco dingly, CA and CDCA (p ima y BAs) we e also posi i ely associa ed wi h mos o hese se obiochemical a iables. Some Fig. 3 Linea disc iminan analysis (LDA) effec size (LE Se) o he gu mic obio a ha shows significan diffe ences in he aecal mic obiome a baseline ( ed ba s) and a e consuming PE o 3 weeks (g een ba s) in all olun ee s (A) o in hose g ouped by u oli hin me abo ypes (B, C and D): UM-A (me abo ype A, n= 29), UM-B (me abo ype B, n= 14) and UM-0 (me abo ype 0, n= 6). Food & Func ion Pape This jou nal is © The Royal Socie y o Chemis y 2024 Food Func .,2024,15,2422–2432 | 2427 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online seconda y BAs, such as DCA, also showed posi i e co ela ions wi h Tchol and small HDLc. Te ia y BAs posi i ely co ela ed wi h HDLc, ApoB, and small HDLc. O he mic obial me ab- oli es, such as cop os anol, did no co ela e wi h hese se o- biochemical a iables. Mic obial aecal U o-A was also no co ela ed wi h mos o hem (Fig. 4C). In con as , aecal choles e ol was nega i ely associa ed wi h HDLc, ApoA, and la ge HDLc and posi i ely associa ed wi h insulin and HOMA-IR. Discussion The gu mic obio a, choles e ol, BAs, and heal h s a us a e closely in eg a ed and in luence each o he , making i difficul o asce ain whe he gu dysbiosis and modi ied BA pools a e a cause o consequence o disease. 20 The gu mic obial con e - sion o gu choles e ol eleases cop os anol as he main me ab- oli e. 12 The p ima y BAs, CA and CDCA, a e syn hesised om choles e ol in he li e and s o ed in he gallbladde (conju- ga ed wi h he amino acids glycine and au ine). La e , hey a e exc e ed in o he duodenum o acili a e he diges ion and abso p ion o lipophilic compounds, including die a y lipids, a -soluble i amins, and choles e ol. 14 Mos conjuga ed p ima y BAs a e eabso bed in he ileum excep o a small ac ion, deconjuga ed by bile sal hyd olase p oduced by diffe en in es inal BA-deconjuga ing bac e ia, which elease deconjuga ed p ima y BAs. This deconjuga ed BA ac ion eaches he colon, whe e CA and CDCA a e dehyd oxyla ed by he in es inal BA-me abolising bac e ia in o DCA and li ho- cholic acid LCA, espec i ely. 5,13,14 These seconda y BAs can be u he ans o med un il e ia y BAs. UDCA, he mos common e ia y bile acid, is p oduced om CDCA ia 7-ke o- CDCA, whe eas iso-LCA is p oduced om CDCA ia LCA. 1 When add essing he possible effec o some die a y poly- phenols, such as ETs, on BAs and choles e ol me abolism, he Fig. 4 Spea man’s co ela ion hea maps o aecal BAs and s e ols wi h: (A) u oli hins ex ac ed om he aeces “F”and u ine “U”and (B) disc imi- na ing bac e ia o PE consump ion and (C) se obiochemical a iables. *: Spea man’s co ela ion alues wi h P< 0.05. Pape Food & Func ion 2428 |Food Func .,2024,15,2422–2432 This jou nal is © The Royal Socie y o Chemis y 2024 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online complexi y is u he inc eased due o he in e -indi idual diffe ences in ETs and he hos ’s s e oid me abolism in lu- enced by he gu mic obio a o each indi idual. Indeed, h ee main UMs upon ET me abolism ha e been iden i ied in he popula ion by a quali a i e c i e ion (i.e., p oduce s s. non- p oduce s o speci ic U os), bu wi hin U o p oduce s (UM-A o UM-B) also exis a U o p oduc ion g adien ha gi es ise o highe and lowe U o p oduce s. 2,21 To da e, he effec s o ET- ich p oduc s, including pomeg ana e, on he mic obial ans- o ma ion o colonic BAs and choles e ol ha e no been explo ed in dep h in humans, and he e is no in o ma ion abou he ole o U os, UMs and hei associa ed gu mic o- bio a in hese effec s. In he p esen s udy, p ima y BAs and choles e ol concen- a ion in he aeces did no diffe among UMs a baseline. In con as , UM-B p esen ed he highes le els o o al BAs, sec- onda y BAs and cop os anol, which we e s a is ically highe han hose o UM-0 bu no hose o UM-A, which showed in e media e mean le els compa ed o he o he wo UMs. The e o e, lipid abso p ion capaci y, mic obial BA me abolism, and gu cy o oxic isk could be augmen ed in UM-B o e - weigh -obese subjec s, especially compa ed o UM-0, because gu BAs p omo e abso p ion o lipids, including choles e ol, by ac ing as emulsi ie s. Mo eo e , seconda y BAs a e conside ed cy o oxic, esul ing in an inc eased isk o gu in lamma ion, gu pe meabili y and colon cance . 12 In e es ingly, diffe ences in he human gu mic obial ecologies associa ed wi h UMs ha e been iden i ied, and cu en e idence sugges s ha he UM-B g oup could be po en ially p one o gu dysbiosis. 2,9–11,18,21,22 As p e iously epo ed, 10 he UM-B indi- iduals o he p esen s udy we e also a highe ca dio ascula isk since hei se um Tchol, LDLc, VLDLc, oxLDLc, ApoB, non-HDLc, IDLc, and small and la ge LDLc le els we e highe han hose in he UM-0 and UM-A g oups a he baseline. The UM-A g oup showed in e media e mean le els compa ed o he o he wo UM g oups. Faecal and u ina y U o exc e ion by PE consump ion was seen in U o p oduce s (UM-A o UM-B) wi h quali a i e and quan i a i e diffe ences. 10 Acco dingly, in he p esen s udy, we show he in e -indi idual diffe ences in U o-A p oduc ion in bo h U o-p oducing me abo ypes (UM-A and UM-B) a e 3-week in ake o ou daily capsules o PE (Fig. 1). As p e- iously epo ed, 10 u oli hin concen a ion is dose-dependen ly inc eased (a ound 3.5- old) a e inc easing consump ion om one o ou PE capsules. Also, h ee U o non-p oduce s (UM-0) became p oduce s ollowing PE consump ion. No ably, PE imp o ed some se um ca dio ascula isk ma ke s, bu only in UM-B subjec s. 10 As p e iously desc ibed, he mos signi ican quan i a i e in e se co ela ions be ween u ina y U os and ca dio ascula ma ke s we e obse ed in UM-B indi iduals, i.e., Tchol, LDLc, and non-HDLc nega i ely co ela ed wi h u ina y U o-A exc e ion. In con as , LDLc also co ela ed wi h u ina y IsoU o-A and U o-B. 10 Howe e , no signi ican co e- la ions we e ound be ween aecal U o exc e ion and mos ca dio ascula isk ma ke s. These esul s sugges ci cula ing U os in plasma could be esponsible o he bene icial effec s on ca dio ascula isk ma ke s and ag ee wi h p e ious s udies in diffe en oden models mimicking ca diome abolic dis- o de s, which desc ibed posi i e effec s in se obiochemical pa ame e s a e o al and in ape i oneal adminis a ion o U os as e iewed be o e. 2 Rega ding he e idence o he ca dio- p o ec i e ac i i ies o U os in i o, se e al s udies epo ed U o ac i i y on ca diomyocy es, mac ophages, hepa ocy es, and adipocy es, as e iewed elsewhe e. 2 U os also signi ican ly dec eased se um and hepa ic choles e ol, sugges ing he hypo- lipidemic effec o U o-A and U o-B in a s. 23 In ha s udy, U o-A and U o-B ea men s down egula ed he exp ession o he li e X ecep o s (LXRα) and s e ol egula o y elemen binding p o ein 1c (SREBP1c), in ol ed in de no o lipogen- esis. 23 Howe e , whe he U os could also exe hei ca dio- ascula and in es inal p o ec i e effec s indi ec ly h ough modula ing he gu mic obio a in ol ed in ans o ming hos me aboli es such as BAs and choles e ol has no been explo ed. The p esen s udy showed ha PE in ake in o e weigh - obese olun ee s signi ican ly educed he aecal BA pool (deconjuga ed-p ima y BAs, seconda y and e ia y BAs) and cop os anol in all olun ee s be o e clus e ing and in he U o- p oducing me abo ype subg oups (UM-A and UM-B), unlike UM-0. Al hough he o al BA concen a ions in he UM-A and UM-B g oups we e simila a e PE consump ion, he dec ease was quan i a i ely mo e ema kable in he UM-B han in UM-A indi iduals, since UM-B s a ed om highe concen a ions. These o al BA educ ions we e mainly due o he la ge quan- i a i e educ ion o seconda y BAs in he UM-B s. UM-A sub- jec s since he dec ease o p ima y BAs was simila in bo h UMs (Fig. 2). Acco dingly, he educ ion o DCA and LCA (sec- onda y BAs) was quan i a i ely mo e ele an in he UM-B sub- jec s s. he UM-A subjec s, whe eas CA and CDCA (p ima y BAs) we e simila ly educed in he UM-A and UM-B g oups (Fig. 2). In he UM-0 g oup (n= 6), hese educ ions we e also obse ed in h ee (50%) o hem who became U o-A p oduce s a e PE consump ion (da a no shown). The e o e, hese educ ions we e no signi ican a e PE in ake in he UM-0 g oup, which showed he lowes baseline BA and cop os- anol concen a ions. Consequen ly, diffe ences among he h ee UMs in he BA concen a ion disappea ed a e he PE in e en ion (Fig. 2). Howe e , cop os anol le els we e s ill highe in he UM-B g oup han in he UM-0 g oup a e he PE in e en ion o 3 weeks (Fig. 2). The lack o effec s in he UM-0 g oup could be due o only h ee o hem becoming U o- A p oduce s a e PE consump ion. 10 Acco ding o his, when all he olun ee s we e conside ed (n= 49), we obse ed a quan i a i e in e se co ela ion o U o-A (a common me ab- oli e o UM-A and UM-B) wi h he aecal BA pool (p ima y, sec- onda y and e ia y BAs) and cop os anol. A p e ious s udy in nasu in and U o-A-p oducing a s showed ha a high- a die signi ican ly inc eased he concen a ion o BAs in he li e and cecum. Howe e , hese high le els o BAs we e educed a e supplemen ing he die s wi h ET- ich aspbe y pomace. 24 The au ho s obse ed ha his supplemen a ion ma kedly inc eased he exp ession le els o small he e odime Food & Func ion Pape This jou nal is © The Royal Socie y o Chemis y 2024 Food Func .,2024,15,2422–2432 | 2429 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online pa ne 1 (SHP-1) associa ed wi h a nesoid-X- ecep o (FXR) and educed he syn hesis o BAs in he li e , he eby educing he gu BAs. 24 This aligns wi h he educ ion o p ima y BAs by PE in ake obse ed in he p esen s udy. In addi ion o he educ ion in he hepa ic p oduc ion o p ima y BAs, o he s udies in animal models ha e also p o ed ha die a y pheno- lic compounds can impac he mic obial me abolism o gu BAs and choles e ol. Fo ins ance, a s showed lowe le els o aecal seconda y BAs a e adminis e ing ea polyphenols and gallo annins, and apple, g ape, and ed bee juices, as well as cu cumin, caffeic acid, ca echin, u in, and ellagic acid, as e iewed p e iously. 12 Howe e , a p e ious s udy epo ed no educ ion in p ima y o seconda y BA con e sion, while chole- s e ol me abolism o cop os anol was signi ican ly educed in heal hy humans a e consuming pomeg ana e juice o ou weeks. 25 Rega ding ha s udy, 25 he ma ginal changes obse ed in BA me abolism by pomeg ana e juice in ake could be due o he p esence o 25% UM-0 since no clus e ing by UMs no quan i a i e co ela ions wi h U os exc e ion was app oached, and all indi iduals we e conside ed a single g oup. This a ionale aligns wi h p e ious s udies ha p opose indi idual s a i ica ion acco ding o he polyphenol me ab- olism capaci y o hei gu mic obio a o unde s and indi id- uals’ esponse o die a y polyphenols. 26,27 We also obse ed a modula ion o he gu mic obio a abun- dance by PE in all he olun ee s be o e clus e ing and in he U o-p oducing me abo ype subg oups (UM-A and UM-B). PE consump ion dec eased he abundance o Rombou sia, a BA- deconjuga ing bac e ia in ol ed in he amino acid hyd olysis p ocess om conjuga ed p ima y BAs, as well as he BA-me a- bolising bac e ia in ol ed in p ima y BA me abolism o sec- onda y and e ia y BAs such as Sa cina,Acine obac e , Collinsella and he amily Co iobac e iaceae. 3,28,29 Al hough mic obial modula ion by PE was no iden ical in he UM-A and UM-B g oups, some mic obial gene a in ol ed in BA me ab- olism we e educed in bo h he g oups. In he UM-0 g oup (n= 6), he lack o gu mic obio a modula ion could be due o only h ee olun ee s becoming U o-A p oduce s a e PE consump- ion. Acco dingly, when all he olun ee s we e conside ed (n= 49), he ela i e abundance o hese bac e ia also co ela ed nega i ely wi h he aecal U o-A le els and posi i ely wi h BAs, especially seconda y and e ia y BAs. Acco ding o his, a p e- ious s udy in a coli is a model showed ha U o-A o al adminis a ion dec eased in lamma ion ma ke s and modu- la ed he gu mic obio a a ou ably. 30 Ano he p e ious s udy in hams e s has shown ha he p e alence o Co iobac e iaceae is dependen on hos geno ype. Indeed, Co iobac e iaceae has been posi i ely co ela ed wi h choles e ol abso p ion, ee choles e ol in he li e , plasma non-HDLc, o al choles e ol, li e weigh , and whi e adipose issue mass. 31 Fu he mo e, his amily has been associa ed wi h suscep ibili y o ca ci- nomas and umou de elopmen in a mu ine model 32 and wi h UM-B in humans. 9 PE consump ion also inc eased some bac e ial g oups, including he well-known bu y a e-p oducing bac e ia (Odo ibac e and Dysosmobac e ) o p opiona e-p odu- cing gene a (Phocaeicola), 33 which we e posi i ely associa ed wi h he aecal U o-A le els. In e es ingly, Dysosmobac e wel- bionis has been epo ed o p e en die -induced obesi y and me abolic diso de s in mice. 34 Simila ly, a p e ious s udy con- duc ed wi h a s showed ha U o-A and U o-B in ake imp o ed li e and kidney unc ions and modula ed he gu mic obio a. In pa icula , U o-A dec eased species di e si y and mic obial ichness and nega i ely impac ed he composi ion o pa ho- genic mic obes in no mal a s. 35 O e all, ou esul s sugges ha U o-A (sha ed mic obial me aboli e be ween UM-A and UM-B) is he main d i e o he PE effec s on he gu mic o- bio a, and i is likely o impac BA p o iles in he hos , edu- cing gu cy o oxici y and wi h po en ial consequences o lipid me abolism and signalling. The p ecise ole o cop os anol (also nega i ely co ela ed wi h U o-A p oduc ion) in he pa ho- genesis o dyslipidemia emains elusi e. I dese es u he esea ch o elucida e whe he cop os anol is a bioma ke ha can also in luence he pa hogenesis o ca diome abolic dis- eases and dyslipidemia. 36 Addi ionally, mo e esea ch is needed o con i m he impac o U os in ake on he hos ’s heal h and sa e y be o e being conside ed a possible ca dio- ascula and gu cy o oxic p o ec o molecule. Conclusions Al hough he lipid-lowe ing effec o ET- ich oods in ca diome- abolic diseases has been p e iously epo ed, i is unknown whe he ETs and (o ) hei mic obial me aboli es U os exe di ec ca diop o ec i e effec s in humans and (o ) indi ec ly h ough modula ing he composi ion and unc ionali y o he gu mic obio a, especially hose mic obes in ol ed in BA and choles e ol me abolism. The o al numbe o mic obial species and amilies engaged in hos choles e ol and BA ans o m- a ions, hei me abolic pa hways, and he oles ha choles e ol and BA mic obial me abolism play in hos heal h con inue o be elucida ed. These aspec s a e o en unde app ecia ed as con i- bu o s o s e ol- ela ed diseases. Ou esul s show ha PE in ake educes gu mic obial me abolism o BAs and choles e ol and modula es he gu mic obio a in an U o-A concen a ion-depen- den manne . This sugges s ha U o-A could no only exe di ec ca diop o ec i e effec s, bu also indi ec ly affec h ough modula ing he composi ion and unc ionali y o he gu mic o- bio a, including he educ ion o Co iobac e iaceae amily and BA pool, which p omo e gu abso p ion o lipids. The ole ha some polyphenols and de i ed mic obial me aboli es, such as U os, may exe in he gu mic obial ecology associa ed wi h s e oid me abolism highligh s he need o u he esea ch o unde s and he complex in e play be ween die a y polyphenols, hos BAs and s e ols, he gu mic obio a and mechanisms in he hos ha egula e se um choles e ol le els. Au ho con ibu ions Concep ualiza ion and design: M. V. S. Expe imen execu ion: A. C. M., C. E. I., A. M., and M. R. V. Me hodology and so - Pape Food & Func ion 2430 |Food Func .,2024,15,2422–2432 This jou nal is © The Royal Socie y o Chemis y 2024 Open Access A icle. Published on 23 Janua y 2024. Downloaded on 11/5/2024 10:15:26 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online