Ursodeoxycholic acid and its taurine/glycine conjugated species reduce colitogenic dysbiosis and equally suppress experimental colitis in mice.

1 U sodeoxycholic acid and i s au ine/glycine conjuga ed species educe coli ogenic dysbiosis and 1 equally supp ess expe imen al coli is in mice 2 3 4 Lien Van den Bossche1, Pie e Hind yckx1, Lindsey De issche 1, Sa ah De iese1, Sophie Van 5 Welden1, Tom Hol oe 1, Rami o Vilchez-Va gas2, Ma ius Vi al3, Die ma H. Piepe 3, Julie 6 Vanden Bussche4, Lynn Vanhaecke4, Tom Van de Wiele2, Ma ine De Vos1 and Debby 7 Laukens1# 8 9 1Depa men o Gas oen e ology, Ghen Uni e si y, Ghen , Belgium; 2Cen e o Mic obial 10 Ecology and Technology, Ghen Uni e si y, Ghen , Belgium; 3Mic obial In e ac ions and 11 P ocesses Resea ch G oup, Depa men o Medical Mic obiology, Helmhol z Cen e o In ec ion 12 Resea ch (HZI), B aunschweig, Ge many; 4Labo a o y o Chemical Analysis, Depa men o 13 Ve e ina y Public Heal h and Food Sa e y, Facul y o Ve e ina y Medicine, Ghen Uni e si y, 14 Ghen , Belgium 15 16 17 Running i le: UDCA and i s conjuga es educe coli ogenic dysbiosis 18 19 20 #Add ess co espondence o: Debby Laukens, debby.laukens@ugen .be. 21 22 AEM Accep ed Manusc ip Pos ed Online 23 Janua y 2017 Appl. En i on. Mic obiol. doi:10.1128/AEM.02766-16 Copy igh © 2017 Ame ican Socie y o Mic obiology. All Righ s Rese ed. on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 2 ABSTRACT 23 The p omising esul s wi h seconda y bile acids in expe imen al coli is sugges ha hey may 24 ep esen an a ac i e and sa e class o d ugs o he ea men o in lamma o y bowel diseases 25 (IBD). Howe e , he exac mechanism by which bile acid he apy con e s p o ec ion om 26 coli ogenesis is cu en ly unknown. Since he gu mic obio a plays a c ucial ole in he 27 pa hogenesis o IBD, and exogenous bile acid adminis a ion may a ec he communi y s uc u e 28 o he mic obio a, we examined he impac o he seconda y bile acid u sodeoxycholic acid 29 (UDCA) and i s au ine/glycine conjuga es on he ecal mic obial communi y s uc u e du ing 30 expe imen al coli is. Daily o al adminis a ion o UDCA, au ou sodeoxycholic acid (TUDCA) o 31 glycou sodeoxycholic acid (GUDCA) equally lowe ed he se e i y o dex an sodium sul a e-32 induced coli is in mice, as e idenced by educed body weigh loss, colonic sho ening and 33 exp ession o in lamma o y cy okines. Illumina sequencing demons a ed ha bile acid he apy 34 du ing coli is did no es o e ecal bac e ial ichness and di e si y. Howe e , bile acid he apy 35 no malized he coli is-associa ed inc eased a io o Fi micu es o Bac e oide es. In e es ingly, 36 adminis a ion o bile acids p e en ed he loss o Clos idium clus e XIVa and inc eased he 37 abundance o Akke mansia muciniphila, bac e ial species known o be pa icula ly dec eased in 38 IBD pa ien s. We conclude ha UDCA, which is an FDA-app o ed d ug o choles a ic li e 39 diso de s, could be an a ac i e ea men op ion o educe dysbiosis and imp o e in lamma ion 40 in human IBD. 41 42 IMPORTANCE 43 Seconda y bile acids a e eme ging as a ac i e candida es o he ea men o in lamma o y 44 bowel disease. Al hough bile acids may a ec he in es inal mic obial communi y s uc u e, 45 which signi ican ly con ibu es o he cou se o hese in lamma o y diso de s, he impac o bile 46 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 3 acid he apy on he ecal mic obio a du ing coli is has no ye been conside ed. He e, we s udied 47 he al e a ions in he ecal mic obial abundance in coli ic mice ollowing he adminis a ion o 48 seconda y bile acids. Ou esul s show ha seconda y bile acids educe he se e i y o coli is and 49 imp o e coli is-associa ed ecal dysbiosis a he phylum le el. This s udy indica es ha seconda y 50 bile acids migh ac as a sa e and e ec i e d ug o in lamma o y bowel disease. 51 52 INTRODUCTION 53 In lamma o y bowel diseases (IBD) a e ch onic in lamma o y diso de s o he gas oin es inal 54 ac cha ac e ized by in es inal dysbiosis. Res ic ed bac e ial di e si y and unde ep esen a ion 55 o an i-in lamma o y mic oo ganisms such as Clos idium clus e XIVa and Akke mansia 56 muciniphila ep esen ypical dysbio ic ea u es in IBD (1–4). Since he in es inal mic obial 57 communi y pe o ms a wide ange o bile acid modi ica ions including deconjuga ion, 58 dehyd oxyla ion, oxida ion and epime iza ion (5), shi s in he composi ion o he gu mic obio a 59 a e associa ed wi h pe u ba ions o he ecal bile acid p o ile (6, 7). O pa icula in e es , Duboc 60 and colleagues demons a ed ha he con e sion o p ima y bile acids (syn hesized in he li e 61 om choles e ol) o seconda y bile acids (gene a ed by bac e ial modi ica ions) is impai ed in 62 IBD pa ien s (7). Because seconda y bile acids exhibi immunomodula o y unc ions (7–10), 63 inc easing seconda y bile acid le els in he in es inal lumen could be an e icien he apeu ic 64 app oach o IBD. In line wi h his hypo hesis, he adminis a ion o he seconda y hyd ophilic 65 bile acid u sodeoxycholic acid (UDCA) amelio a es expe imen al coli is bu he exac mechanism 66 p o ec ing om coli ogenesis is no ully unde s ood (11). 67 When adminis e ed o ally, unconjuga ed UDCA is apidly conjuga ed wi h glycine in humans, 68 and o a lesse ex en wi h au ine, on i s i s pass h ough he li e (12, 13). Based on he 69 obse a ion ha ecal bile acid hyd ophobici y co ela es wi h he se e i y o coli is (14), i is 70 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 4 easonable o assume ha conjuga es o UDCA, which a e mo e hyd ophilic han unconjuga ed 71 UDCA, migh be mo e a o able he apeu ic agen s o in es inal in lamma ion. In his ega d, we 72 and o he s ha e shown ha au ou sodeoxycholic acid (TUDCA) alle ia es dex an sodium 73 sul a e (DSS)-induced coli is in mice (15, 16). The po en ial bene icial e ec o 74 glycou sodeoxycholic acid (GUDCA) in coli is, howe e , has no been add essed so a and 75 s udies compa ing he he apeu ic e ec i eness o hese di e en bile acid species a e lacking. 76 While he composi ion o he luminal bile acid pool is con olled by in es inal bac e ia, i is well 77 es ablished ha bile acids, in u n, also shape he gu mic obio a. Bile acids es ic bac e ial 78 p oli e a ion and o e g ow h di ec ly by causing memb ane damage, which is posi i ely 79 co ela ed wi h bile acid hyd ophobici y (17–19). Thus, he bac e icidal ac i i y o bile acids 80 dec eases wi h inc easing numbe s o hyd oxyl g oups and by conjuga ion o he bile acid side 81 chain wi h au ine o glycine (19). In addi ion o hei ole as an imic obial agen s, bile acids also 82 s imula e he g ow h o selec ed bac e ial species (5). Simila ly, hese p ope ies a e de e mined 83 bo h by he hyd oxyla ion pa e n and he conjuga ion s a us o he bile acid s e oid nucleus. Fo 84 example, inc eased in es inal le els o bile acids ca ying a hyd oxyl g oup a posi ion C7 o he 85 s e oid co e a o he g ow h o 7α-dehyd oxyla ing bac e ia, such as Clos idium clus e XIVa 86 membe s (20, 21). Fu he mo e, he amino acids in conjuga ed bile acids ac as mic obial 87 subs a es o dis inc bac e ial g oups; glycine is me abolized by Clos idium species (22, 23), 88 while au ine is a sou ce o sulphi e om which Bilophila wadswo hia de i es ene gy o i s 89 g ow h (24, 25). In e es ingly, a die high in sa u a ed a p omo es au ine-conjuga ion o hepa ic 90 bile acids, esul ing in he ou g ow h o B. wadswo hia and exace ba ion o coli is (26). 91 Conside ing ha he gu mic obial a chi ec u e and me abolism con ibu e o he cou se o IBD 92 (27, 28), we compa ed he he apeu ic e ec i eness o UDCA and i s au ine/glycine conjuga es 93 in DSS-induced coli is in mice and in es iga ed hei impac on he ecal mic obial communi y. 94 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 5 95 MATERIALS AND METHODS 96 Animals. Male 8-week-old C57Bl/6J mice we e ob ained om Ha lan (Ha lan Labo a o ies, 97 Ho s , The Ne he lands) and main ained unde s anda d labo a o y condi ions wi h ad libi um 98 access o ood (mice main enance chow, Ca il Labo ood, Pa an Se ice, Belgium) and wa e . 99 P io o he expe imen , mice we e co-housed o homogenize gu mic obio a be ween 100 expe imen al g oups. A e a one-week acclima iza ion, mice we e assigned o he ea men 101 g oups based on body weigh s. In o de o a oid bac e ial c oss-con amina ion be ween g oups, 102 mice o di e en ea men g oups we e housed in sepa a e cages. The s udy was app o ed by he 103 Ins i u ional Re iew Boa d o he Facul y o Medicine and Heal h Science o Ghen Uni e si y 104 (ECD 2014-25). 105 Bile acid ea men . Mice we e di ided in o i e g oups (n = 8 in each g oup). Th ee o hem 106 ecei ed bile acid ea men : UDCA (Tokyo Chemical Indus y Co. L d, Toshima-Ku, Tokyo, 107 Japan), TUDCA (Calbiochem, Da ms ad , Ge many) o GUDCA (Sigma-Ald ich, Diegem, 108 Belgium). Bile acids we e dissol ed in phospha e-bu e ed saline (PBS) o Lab a il® M1944 109 (Ga e osse, Sain -P ies Cedex, F ance) and adminis e ed daily by o al ga age (500 mg/kg/day). 110 T ea men s s a ed a day 0 o DSS exposu e. A non-DSS con ol g oup and DSS con ol g oup 111 ( e e ed o as placebo- ea ed g oup) ecei ed he ehicle (PBS o Lab a il®) alone. 112 Induc ion and assessmen o coli is. Acu e coli is was es ablished by adding 4% (w/ ) DSS 113 (molecula weigh 36,000–50,000; MP Biomedicals, Illki ch, F ance) o he d inking wa e o 7 114 days, ollowed by no mal wa e o 3 days. A non-DSS con ol g oup ecei ed no mal d inking 115 wa e h oughou he expe imen . Body weigh and disease ac i i y we e eco ded daily. A 116 disease ac i i y index (DAI) was calcula ed as he combined sco e o body weigh loss (0, none; 117 1, 0-10%; 2, 10-20%; 3, >20%), s ool consis ency (0, no mal d oppings; 1, loose d oppings; 2, 118 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 6 dia hea) and ecal blood loss (0, none; 1, hemoccul posi i e; 2, g oss bleeding). Occul blood 119 was de ec ed using he Colosc een Hemoccul ki (Helena Labo a o ies Inc., Beaumon , Texas, 120 USA). Ten days a e ini ia ion o he expe imen , mice we e anes he ized and blood was 121 collec ed om he e o-o bi al sinus. The mice we e hen sac ificed by ce ical disloca ion, he 122 colons we e emo ed and hei leng hs we e measu ed. Segmen s o dis al colon we e cu , insed 123 wi h PBS and ozen in liquid ni ogen. The blood was cen i uged (10.000 pm o 10 min a 124 4°C) and se um was collec ed. All samples we e s o ed a -80°C un il u he p ocessing. 125 Luminex. Colonic issues we e homogenized in PBS con aining p o ease and phospha ase 126 inhibi o s and o al p o ein concen a ion was measu ed using he B ad o d me hod (Bio-Rad, 127 Naza e h, Belgium). P o ein le els o chemokine (C-X-C mo i ) ligand 1 (CXCL1), g anulocy e 128 colony-s imula ing ac o (G-CSF) and in e leukin (IL)-6 we e de e mined in colon homogena es 129 and se um using he Bio-Plex P o Mouse Cy okine G oup I mul iplex ki (Bio-Rad), acco ding o 130 he manu ac u e ’s ins uc ions. Measu emen s we e pe o med wi h he Bio-Plex MAGPIX 131 Mul iplex Reade and da a we e analyzed using he Bio-Plex Manage 6.1 so wa e (Bio-Rad). 132 DNA ex ac ion om ecal samples. F esh ecal pelle s we e collec ed a day 9 o coli is and 133 immedia ely s o ed a -80°C. To al DNA was ex ac ed om he ecal samples using he QIAamp 134 DNA S ool Mini Ki (Qiagen Benelux, Venlo, The Ne he lands). Fi s , 180 o 220 mg o s ool 135 was esuspended in 1.4 ml bu e ASL. Then, 0.5 g 0.1 mm Zi conia beads (Biospec P oduc s, 136 Ba les ille, Oklahoma) and 4 glass beads (Biospec P oduc s) we e added and samples we e 137 homogenized by o exing. The suspension was hen hea ed a 95°C o 15 min and he 138 manu ac u e ’s ins uc ions we e ollowed. 139 Illumina sequencing. The V1-2 egion o he 16S RNA gene was ampli ied as p e iously 140 desc ibed (29). B ie ly, in a i s 20 cycle polyme ase chain eac ion (PCR) eac ion, he 16S 141 RNA gene a ge was en iched using he well-documen ed 27F and 338R p ime s (30, 31) as 142 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 7 p e iously speci ied (32). This eac ion mix u e was used as empla e in a second 15 cycle PCR 143 eac ion wi h p ime s comp ising sequences complemen a y o he Illumina speci ic adap o s o 144 he 5’-ends (29). The la e eac ion mix u e was hen used as empla e in a hi d 10 cycle PCR 145 eac ion wi h p ime s designed o in eg a e bo h he sequence o he speci ic Illumina 146 mul iplexing sequencing p ime s and he index p ime s. Lib a ies p epa ed by pooling equimola 147 a ios o amplicons we e inally sequenced on a MiSeq (Illumina, Haywa d, CA, USA). 148 A e wa ds, eads we e anno a ed as desc ibed by Ve s aelen e al. (33). 149 Illumina da a analysis. Da a-analysis was pe o med as p e iously desc ibed (33). A e 150 esampling o he minimum sequencing dep h using he phyloseq package (34) om he R 151 p og am (35), a o al o 8,911 eads we e ob ained. Ra e ac ion cu es we e gene a ed using he 152 egan package om R (36). All phylo ypes we e assigned a axonomic a ilia ion based on he 153 nai e Bayesian classi ica ion (RDP classi ie ) (37) wi h a h eshold o 80%. Rela i e abundances 154 o all phylo ypes we e hen compa ed be ween di e en expe imen al g oups. 155 Quan i a i e eal- ime PCR (qRT-PCR). To al ecal DNA was dilu ed 1:2 in wa e and 3 µl 156 was used in qRT-PCR wi h SYBR G een (SensiMix™ SYBR No-ROX Ki , Bioline Reagen s, 157 UK) and 250 nM o each p ime (BioLegio, Nijmegen, The Ne he lands). P ime sequences used 158 o ampli ica ion o A. muciniphila we e 5’-CAGCACGTGAAGGTGGGGAC-3’ and 5’-159 CCTTGCGGTTGGCTTCAGAT-3’ (38). A wo-s ep p og am was pe o med on he Ligh Cycle 160 480 (Roche). Cycling condi ions we e 95°C o 10 min, 45 cycles o 95°C o 10 s and 60°C o 1 161 min. The amoun o A. muciniphila 16S RNA gene in each sample was no malized o he o al 162 amoun o bac e ial 16S RNA gene. Fo he quan i ica ion o o al 16S RNA gene copies, ecal 163 DNA was dilu ed 1:10 in wa e and he uni e sal bac e ial 16S RNA gene p ime s PRBA338 164 5’-ACTCCTACGGGAGGCAGCAG-3’ and PRUN518 5’-ATTACCGCGGCTGCTGG-3’ we e 165 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 8 used (39). To al bac e ial load was calcula ed using he o mula 2del aC / [ o al DNA 166 concen a ion]. 167 Bile acid quan i ica ion. 168 Sample p epa a ion. F esh ecal pelle s we e collec ed a day 4 o coli is and immedia ely s o ed 169 a -80°C. Be o e bile acids we e ex ac ed om ecal samples, 20 µl o an in e nal s anda d 170 solu ion (TUDCA-d5 (San a-C uz Bio echnology, Heidelbe g, Ge many) a 25 ng/µl in me hanol 171 (VWR In e na ional, Me ck Millipo e, Da ms ad , Ge many)) was added o 25 mg eces. The 172 ex ac ion p o ocol s a ed wi h he addi ion o 5 ml ice-cold ace oni ile (VWR In e na ional) 173 con aining 5% ammonium hyd oxide (Me ck Millipo e, Da ms ad , Ge many). The solu ion was 174 homogenized wi h an Ul a-Tu ax homogenize , ho oughly mixed by o exing o 1 min and 175 hen placed in an ul asonic ba h o 30 min. The esul an mix u e was cen i uged a 9,000 x g 176 o 10 min and supe na an was collec ed. The ex ac ion p ocedu e was epea ed once mo e and 177 he combined supe na an s we e subsequen ly e apo a ed unde ni ogen a 40°C. Each d y 178 ex ac was hen esuspended in 200 µl o a 40:60 mix u e o sol en A (7.5 mM ammonium 179 ace a e (Me ck Millipo e) in ul apu e wa e , pH 4.0) and sol en B (5% ace oni ile in 180 me hanol), cen i uged a 9,000 x g o 10 min and supe na an was collec ed. To compensa e o 181 ma ix e ec s, he s anda d addi ion me hod was applied o bile acid quan i ica ion (40). B ie ly, 182 supe na an was di ided in o wo equal aliquo s and ans e ed o liquid ch oma og aphy-mass 183 spec ome y ials. One aliquo was spiked wi h 20 µl o he 40:60 mix u e o sol en A and B. 184 The o he aliquo was spiked wi h 20 µl o bile acid solu ion (a 40:60 mix u e o sol en A and B, 185 supplemen ed wi h 16.5 ng/µl li hocholic acid (LCA; Sigma-Ald ich), 89.4 ng/µl UDCA (Simga-186 Ald ich), 0.75 ng/µl TUDCA (Calbiochem) and 0.75 ng/µl GUDCA (Sigma-Ald ich)). A 10 μl 187 aliquo o each sample was injec ed in o he ul a-high pe o mance liquid ch oma og aphy wi h 188 high esolu ion mass spec ome y (UHPLC-HRMS) sys em. 189 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 9 UHPLC-HRMS analysis. Ch oma og aphic sepa a ion o bile acids was ca ied ou on an Accela 190 UHPLC sys em o The mo Fishe Scien i ic (San José, CA, USA), wi h an Acqui y UPLC HSS 191 C18 column (1.8 μm, 50 mm × 2.1 mm, Wa e s). The bina y sol en sys em consis ing o wo 192 sol en s A and B was se a a cons an low a e o 300 µl/min a 35°C. Fo elu ion, a g adien 193 p o ile was applied wi h he ollowing p opo ions ( / ) o sol en A: 0 – 1.0 min a 40%, 1.0 – 194 6.0 min om 40% o 1%, 6.0 – 8.0 min a 1%, 8.0 – 8.1 min om 1% o 40%, ollowed by 3.9 195 min o e-equilib a ion. 196 HRMS analysis was pe o med on an Exac i eTM s and-alone bench op mass spec ome e 197 (The mo Fishe Scien i ic), equipped wi h a hea ed elec osp ay ioniza ion sou ce (HESI-II), 198 ope a ing in he nega i e ioniza ion mode. Ioniza ion sou ce wo king pa ame e s we e op imized 199 and we e se o a shea h, auxilia y and sweep gas o 40, 5 and 1 a bi a y uni s (au), espec i ely, 200 hea e and capilla y empe a u e o 120°C and 375°C and ube lens, skimme , capilla y and sp ay 201 ol age o 123 V, 22 V, 43.5 V and 4 kV (+/-), espec i ely. A scan ange o m/z 300-550 was 202 selec ed and he esolu ion was se a 100,000 FWHM a 1 Hz (1 scan pe second). The au oma ic 203 gain con ol (AGC) a ge was se a high dynamic ange (3 × E6 ions) and he maximum 204 injec ion ime was 100 ms. 205 Da a p ocessing. HRMS da a p ocessing was pe o med wi h Xcalibu ™ 3.0 (The mo Fishe 206 Scien i ic). The concen a ion o a selec ed bile acid was calcula ed using he ollowing o mula 207 (40): 208 Cunk= CSA × ARunk ARSA – ARunk wi h Cunk being he unknown concen a ion o he bile acid in he o iginal ecal sample, CSA being 209 he spiked concen a ion o he bile acid in he ecal sample a e s anda d addi ion, ARunk being 210 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 16 deg ee o dysbiosis in IBD (55–57). O no e, he phylum-le el popula ion shi s om 354 Bac e oide es o Fi micu es induced by DSS in ou s udy esemble hose obse ed in obese 355 indi iduals and in animals on a high- a die , and ha e been associa ed wi h low-g ade in es inal 356 and sys emic in lamma ion in obesi y (58, 59). In his con ex , ecal calp o ec in and plasma C-357 eac i e p o ein le els showed a posi i e co ela ion wi h bac e ia belonging o he Fi micu es, 358 whe eas a nega i e co ela ion was ound be ween C- eac i e p o ein le els and speci ic g oups 359 wi hin he Bac e oide es (58). I is he e o e likely ha bile acid he apy coun e ac s he 360 de elopmen o a “p o-in lamma o y” mic obio a du ing coli is. This is specula i e since i 361 emains unknown i he changes seen in he Fi micu es/Bac e oide es a io ollowing bile acid 362 ea men a e a cause, a he han a consequence, o he bile acid an i-in lamma o y e ec . 363 Howe e , bile acid he apy did no p e en he DSS-induced dec ease in unclassi ied membe s o 364 he phylum Bac e oide es bu inc eased he ela i e abundance o Bac e oidia, which was no 365 a ec ed by DSS. Thus, we can specula e ha bile acid he apy di ec ly in e e es wi h an 366 imbalanced mic obial en i onmen . 367 We demons a ed ha Clos idium clus e XIVa species we e signi ican ly unde ep esen ed upon 368 DSS challenge, con i ming p e ious obse a ions in bo h human and expe imen al IBD (2, 60–369 62). Howe e , o al adminis a ion o UDCA o i s au ine/glycine conjuga es was able o p o oke 370 an en ichmen o hese species compa ed wi h placebo- ea ed mice. I has been shown ha 371 selec ed membe s wi hin he clos idial clus e XIVa possess 7α-dehyd oxyla ion ac i i y (20), 372 which is in ol ed in a mul is ep biochemical pa hway con e ing UDCA o LCA (63, 64). In ou 373 expe imen , o ally adminis e ed TUDCA and GUDCA we e apidly deconjuga ed o UDCA, so 374 ei he bile acid ea men c ea ed a subs a e- ich en i onmen o hese species. This may explain 375 he bloom o Clos idium clus e XIVa ha was obse ed in coli ic mice ha we e ea ed wi h 376 bile acids. Clos idium spp. belonging o clus e XIVa a e impo an induce s o egula o y T 377 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 17 cells in he colon (65). In addi ion, 80% o he bu y a e-p oducing s ains isola ed om human 378 ecal samples belong o he clos idial clus e XIVa (66). Bu y a e is a sho -chain a y acid wi h 379 dis inc i e an i-in lamma o y p ope ies ha has al eady p o en i s e icacy in C ohn’s disease 380 (67). Howe e , bu y a e-p oducing bac e ia a e deple ed in he ecal mic obio a o IBD pa ien s 381 (61, 62, 68). Thus, ou obse a ion ha UDCA o i s au ine/glycine conjuga es inc eased he 382 abundance o Clos idium clus e XIVa du ing colonic in lamma ion is o pa icula in e es and 383 may sugges an immunomodula o y ole o hese bile acids. 384 Ano he inding o his s udy was he o e ep esen a ion o Bac e oidaceae, P e o ellaceae and 385 Akke mansia in ecal samples o bile acid- ea ed mice ollowing DSS exposu e. These esul s 386 migh be ela ed o he s imula o y e ec o bile acids on mucin sec e ion as a de ense 387 mechanism o p o ec he gas oin es inal epi helium agains po en ial bile acid oxici y (69–71). 388 Bac e ial species belonging o he gene a Bac e oides, P e o ella and Akke mansia p oduce one 389 o mo e enzymes equi ed o mucin deg ada ion (72), which is enhanced du ing he acu e phase 390 o DSS-induced coli is (46). The e o e, i is easonable o assume ha hese bac e ia can g ow 391 be e in an en i onmen ha is, esul ing om exogenous bile acid adminis a ion, en iched wi h 392 mucins. A. muciniphila is a commensal bac e ium esiding in he mucus laye o he in es inal 393 ac and has been shown o be educed in IBD pa ien s (3, 4). Al hough con lic ing esul s we e 394 ob ained in s udies assessing he ole o Akke mansia in colonic in lamma ion, hese species a e 395 hough o play a key ole in he egula ion o gu ba ie unc ion and mucosal immune esponses 396 owa d he commensal mic obio a (73, 74). 397 The molecula s uc u e o a bile acid de e mines i s me abolism, physicochemical p ope ies and 398 biological e ec s (75). In he p esen s udy, we used h ee bile acids sha ing he same s e oidal 399 hyd oxyla ion pa e n bu di e ing in hei amino acid conjuga ion pa e n. Nei he bile acid 400 species es ed p o ed o be mo e o less e icacious han he o he in educing colonic 401 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 18 in lamma ion. Likewise, adminis a ion o UDCA induced simila changes in he bac e ial 402 communi y compa ed wi h i s au ine/glycine conjuga ed species. These obse a ions can be 403 explained by he apid in i o bio ans o ma ion o o ally adminis e ed bile acids by he li e and 404 by he in es inal mic obio a. Wi h he excep ion o ecal GUDCA concen a ions, which we e 405 only inc eased ollowing GUDCA he apy, he e we e no di e ences in ecal concen a ions o 406 UDCA, TUDCA o LCA be ween mice ha we e adminis e ed UDCA o i s conjuga es. This is 407 in con as wi h da a om p e ious s udies in pa ien s wi h p ima y bilia y ci hosis (12) and a s 408 (43) showing ha , compa ed wi h UDCA, o ally adminis e ed TUDCA unde goes educed 7-409 dehyd oxyla ion o LCA. I is concei able ha in e species di e ences in in es inal mic obio a 410 accoun o hese disc epancies. Fo example, deconjuga ion o TUDCA o GUDCA is a 411 p e equisi e o u he 7-dehyd oxyla ion and is ca alyzed by bile sal hyd olases (43). Because 412 Lac obacilli, which exp ess bile sal hyd olases, a e mo e abundan in he mouse gu mic obio a 413 as compa ed o he human gu mic obio a (76), i is likely ha bio ans o ma ion o hese 414 conjuga ed bile acids occu s o a la ge ex en in mice. 415 In summa y, we epo ha UDCA and i s au ine/glycine conjuga ed species amelio a e colonic 416 in lamma ion in mice wi hou di e ing in he apeu ic e ec i eness, and educe DSS-induced 417 ecal dysbiosis a he phylum le el, i espec i e o he bile acid conjuga ion s a us. As we 418 demons a ed no ad an age o using ei he he au ine o glycine conjuga e o UDCA, we sugges 419 ha UDCA could be a sa e and eadily a ailable ea men op ion o IBD. This conclusion is 420 u he suppo ed by he cu en he apeu ic use o UDCA in choles a ic pa ien s (77) and by i s 421 p e en i e e ec s on IBD-associa ed colo ec al ca cinogenesis (78–80). 422 423 ACKNOWLEDGEMENTS 424 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 19 The au ho s a e g a e ul o D . Falk Pha maceu icals, who kindly p o ided TUDCA o esea ch 425 pu poses. We would also like o hank Hilde De lies, G ie D iesschae and Pe a Van 426 Wassenho e o p o iding echnical assis ance. The e a e no con lic s o in e es o decla e. 427 428 FUNDING INFORMATION 429 This wo k was suppo ed by esea ch g an s om he Resea ch Founda ion Flande s (FWO; 430 11J9915N and 1298213N), a conce ed ac ion g an (GOA) om he Special Resea ch Fund 431 (BOF; GOA 2012/01G00812) o Ghen Uni e si y, and a g an om he Belgian ounda ion o 432 C ohn’s disease and ulce a i e coli is pa ien s (CCV zw, esea ch g an 2014). 433 434 REFERENCES 435 1. Nagao-Ki amo o H, Ki amo o S, Ku a P, Kamada N. 2016. Pa hogenic ole o he gu 436 mic obio a in gas oin es inal diseases. In es Res 14:127–138. 437 2. Andoh A, Imaeda H, Aoma su T, Ina omi O, Bamba S, Sasaki M, Sai o Y, Tsujikawa 438 T, Fujiyama Y. 2011. Compa ison o he ecal mic obio a p o iles be ween ulce a i e 439 coli is and C ohn’s disease using e minal es ic ion agmen leng h polymo phism 440 analysis. J Gas oen e ol 46:479–486. 441 3. Png CW, Lindén SK, Gilshenan KS, Zoe endal EG, McSweeney CS, Sly LI, McGuckin 442 MA, Flo in THJ. 2010. Mucoly ic bac e ia wi h inc eased p e alence in IBD mucosa 443 augmen in i o u iliza ion o mucin by o he bac e ia. Am J Gas oen e ol 105:2420–2428. 444 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 20 4. Vigsnæs LK, B ynsko J, S eenhold C, Wilcks A, Lich TR. 2012. G am-nega i e 445 bac e ia accoun o main di e ences be ween aecal mic obio a om pa ien s wi h 446 ulce a i e coli is and heal hy con ols. Bene Mic obes 3:287–297. 447 5. Wahls öm A, Sayin SI, Ma schall H-U, Bäckhed F. 2016. In es inal C oss alk be ween 448 Bile Acids and Mic obio a and I s Impac on Hos Me abolism. Cell Me ab 24:41–50. 449 6. Kakiyama G, Pandak WM, Gille e PM, Hylemon PB, Heuman DM, Dai a K, Takei 450 H, Mu o A, Ni ono H, Ridlon JM, Whi e MB, Noble NA, Mon ei h P, Fuchs M, 451 Thacke LR, Sika oodi M, Bajaj JS. 2013. Modula ion o he ecal bile acid p o ile by gu 452 mic obio a in ci hosis. J Hepa ol 58:949–955. 453 7. Duboc H, Rajca S, Rain eau D, Bena ous D, Maube M-A, Que ain E, Thomas G, 454 Ba bu V, Humbe L, Desp as G, B idonneau C, Dume z F, G ill J-P, Masliah J, 455 Beauge ie L, Cosnes J, Chazouillè es O, Poupon R, Wol C, Malle J-M, Langella P, 456 T ugnan G, Sokol H, Seksik P. 2013. Connec ing dysbiosis, bile-acid dysme abolism and 457 gu in lamma ion in in lamma o y bowel diseases. Gu 62:531–539. 458 8. Calmus Y, Weill B, Ozie Y, Ché eau C, Houssin D, Poupon R. 1992. 459 Immunosupp essi e p ope ies o chenodeoxycholic and u sodeoxycholic acids in he 460 mouse. Gas oen e ology 103:617–621. 461 9. G e e JW, Gouma DJ, Buu man WA. 1989. Bile acids inhibi endo oxin-induced elease 462 o umo nec osis ac o by monocy es: an in i o s udy. Hepa ol Bal im Md 10:454–458. 463 10. Yoneno K, Hisama su T, Shimamu a K, Kamada N, Ichikawa R, Ki azume MT, Mo i 464 M, Uo M, Namikawa Y, Ma suoka K, Sa o T, Koganei K, Sugi a A, Kanai T, Hibi T. 465 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 21 2013. TGR5 signalling inhibi s he p oduc ion o p o-in lamma o y cy okines by in i o 466 di e en ia ed in lamma o y and in es inal mac ophages in C ohn’s disease. Immunology 467 139:19–29. 468 11. Ma ínez-Moya P, Rome o-Cal o I, Requena P, He nández-Chi laque C, A anda CJ, 469 González R, Za zuelo A, Suá ez MD, Ma ínez-Augus in O, Ma ín JJG, de Medina 470 FS. 2013. Dose-dependen an iin lamma o y e ec o u sodeoxycholic acid in expe imen al 471 coli is. In Immunopha macol 15:372–380. 472 12. In e nizzi P, Se chell KD, C osignani A, Ba ezza i PM, La ghi A, O’Connell NC, 473 Podda M. 1999. Di e ences in he me abolism and disposi ion o u sodeoxycholic acid and 474 o i s au ine-conjuga ed species in pa ien s wi h p ima y bilia y ci hosis. Hepa ol Bal im 475 Md 29:320–327. 476 13. Rudolph G, Kloe e s-Plachky P, Saue P, S iehl A. 2002. In es inal abso p ion and bilia y 477 sec e ion o u sodeoxycholic acid and i s au ine conjuga e. Eu J Clin In es 32:575–580. 478 14. S enman LK, Holma R, Fo sgå d R, Gylling H, Ko pela R. 2013. Highe ecal bile acid 479 hyd ophobici y is associa ed wi h exace ba ion o dex an sodium sul a e coli is in mice. J 480 Nu 143:1691–1697. 481 15. Cao SS, Zimme mann EM, Chuang B-M, Song B, Nwokoye A, Wilkinson JE, Ea on 482 KA, Kau man RJ. 2013. The un olded p o ein esponse and chemical chape ones educe 483 p o ein mis olding and coli is in mice. Gas oen e ology 144:989–1000.e6. 484 16. Laukens D, De issche L, Van den Bossche L, Hind yckx P, Vandenb oucke RE, 485 Vandewynckel Y-P, Cu elie C, B inkman BM, Libe C, Vandenabeele P, De Vos M. 486 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 22 2014. Tau ou sodeoxycholic acid inhibi s expe imen al coli is by p e en ing ea ly in es inal 487 epi helial cell dea h. Lab In es ig J Tech Me hods Pa hol 94:1419–1430. 488 17. Ku di P, Kawanishi K, Mizu ani K, Yoko a A. 2006. Mechanism o G ow h Inhibi ion by 489 F ee Bile Acids in Lac obacilli and Bi idobac e ia. J Bac e iol 188:1979–1986. 490 18. Ta an o MP, Pe ez-Ma inez G, Fon de Valdez G. 2006. E ec o bile acid on he cell 491 memb ane unc ionali y o lac ic acid bac e ia o o al adminis a ion. Res Mic obiol 492 157:720–725. 493 19. Sonomo o K, Yoko a A. 2011. Lac ic Acid Bac e ia and Bi idobac e ia: Cu en P og ess 494 in Ad anced Resea ch. Ho izon Scien i ic P ess. 495 20. Ridlon JM, Al es JM, Hylemon PB, Bajaj JS. 2013. Ci hosis, bile acids and gu 496 mic obio a. Gu Mic obes 4:382–387. 497 21. Islam KBMS, Fukiya S, Hagio M, Fujii N, Ishizuka S, Ooka T, Ogu a Y, Hayashi T, 498 Yoko a A. 2011. Bile acid is a hos ac o ha egula es he composi ion o he cecal 499 mic obio a in a s. Gas oen e ology 141:1773–1781. 500 22. Dü e P, And eesen JR. 1983. Pu ine and glycine me abolism by pu inoly ic clos idia. J 501 Bac e iol 154:192–199. 502 23. Lebe z H, And eesen JR. 1988. Glycine e men a ion by Clos idium his oly icum. A ch 503 Mic obiol 150:11–14. 504 24. Laue H, Denge K, Cook AM. 1997. Tau ine educ ion in anae obic espi a ion o 505 Bilophila wadswo hia RZATAU. Appl En i on Mic obiol 63:2016–2021. 506 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 23 25. Laue H, Smi s THM, Schumache UK, Cla os MC, Ha emink R, Cook AM. 2006. 507 Iden i ica ion o Bilophila wadswo hia by speci ic PCR which a ge s he au ine:py u a e 508 amino ans e ase gene. FEMS Mic obiol Le 261:74–79. 509 26. De ko a S, Wang Y, Musch MW, Leone V, Fehlne -Peach H, Nadimpalli A, 510 An onopoulos DA, Jab i B, Chang EB. 2012. Die a y- a -induced au ocholic acid 511 p omo es pa hobion expansion and coli is in Il10-/- mice. Na u e 487:104–108. 512 27. Wloda ska M, Kos ic AD, Xa ie RJ. 2015. An in eg a i e iew o mic obiome-hos 513 in e ac ions in in lamma o y bowel diseases. Cell Hos Mic obe 17:577–591. 514 28. B inkman BM, Becke A, Ayiseh RB, Hildeb and F, Raes J, Huys G, Vandenabeele P. 515 2013. Gu mic obio a a ec s sensi i i y o acu e DSS-induced coli is independen ly o hos 516 geno ype. In lamm Bowel Dis 19:2560–2567. 517 29. Cama inha-Sil a A, Jáu egui R, Cha es-Mo eno D, Oxley APA, Schaumbu g F, 518 Becke K, Wos-Oxley ML, Piepe DH. 2014. Compa ing he an e io na e bac e ial 519 communi y o wo disc e e human popula ions using Illumina amplicon sequencing. 520 En i on Mic obiol 16:2939–2952. 521 30. LANE DJ. 1991. 16S/23S RNA sequencing. Nucleic Acid Tech Bac Sys 125–175. 522 31. E chebehe e C, Tiedje J. 2005. P esence o Two Di e en Ac i e ni S Ni i e Reduc ase 523 Genes in a Deni i ying Thaue a sp. om a High-Ni a e-Remo al-Ra e Reac o . Appl 524 En i on Mic obiol 71:5642–5645. 525 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 24 32. Cha es-Mo eno D, Plumeie I, Kahl S, K isme B, Peschel A, Oxley APA, Jau egui R, 526 Piepe DH. 2015. The mic obial communi y s uc u e o he co on a nose. En i on 527 Mic obiol Rep 7:929–935. 528 33. Ve s aelen H, Vilchez-Va gas R, Desimpel F, Jau egui R, Vankei sbilck N, Weye s S, 529 Ve hels R, De Su e P, Piepe DH, Van De Wiele T. 2016. Cha ac e isa ion o he 530 human u e ine mic obiome in non-p egnan women h ough deep sequencing o he V1-2 531 egion o he 16S RNA gene. Pee J 4:e1602. 532 34. McMu die PJ, Holmes S. 2013. phyloseq: An R Package o Rep oducible In e ac i e 533 Analysis and G aphics o Mic obiome Census Da a. PLOS ONE 8:e61217. 534 35. R Co e Team. 2012. R: A language and en i onmen o s a is ical compu ing. R 535 Founda ion o S a is ical Compu ing, Vienna, Aus ia. 536 36. Oksanen J, Blanche FG, F iendly M, Kind R, Legend e P, McGlinn D, Minchin PR, 537 O’Ha a RB, Simpson GL, Solymos P, S e ens MHH, Szoecs E, Wagne H. 2016. egan: 538 Communi y Ecology Package. 539 37. Wang Q, Ga i y GM, Tiedje JM, Cole JR. 2007. Naï e Bayesian Classi ie o Rapid 540 Assignmen o RNA Sequences in o he New Bac e ial Taxonomy. Appl En i on 541 Mic obiol 73:5261–5267. 542 38. Schneebe ge M, E e a d A, Gómez-Valadés AG, Ma amo os S, Ramí ez S, Delzenne 543 NM, Gomis R, Cla e M, Cani PD. 2015. Akke mansia muciniphila in e sely co ela es 544 wi h he onse o in lamma ion, al e ed adipose issue me abolism and me abolic diso de s 545 du ing obesi y in mice. Sci Rep 5. 546 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 25 39. O eås L, Lj F, Daae FL, Vl T. 1997. Dis ibu ion o bac e ioplank on in Me omic ic lake 547 Sælen anne , as de e mined by dena u ing g adien gel elec opho esis o PCR-ampli ied 548 gene agmen s coding o 16S RNA. Appl En i on Mic obiol 63:3367–73. 549 40. Cai X, Liu Y, Zhou X, Na anee han U, Shen B, Guo B. 2012. An LC-ESI-MS me hod o 550 he quan i a i e analysis o bile acids composi ion in ecal ma e ials. Biomed Ch oma og 551 BMC 26:101–108. 552 41. Alex P, Zachos NC, Nguyen T, Gonzales L, Chen T-E, Conklin LS, Cen ola M, Li X. 553 2009. Dis inc cy okine pa e ns iden i ied om mul iplex p o iles o mu ine DSS and 554 TNBS-induced coli is. In lamm Bowel Dis 15:341–352. 555 42. Pe e MR, Je kic M, So o V, Douda DN, A delean DS, Ghamami N, Laksche i z F, 556 Khan MA, Robe son SJ, Glogaue M, Philpo DJ, Palaniya N, Le a e M. 2014. 557 Impai ed Resolu ion o In lamma ion in he Endoglin He e ozygous Mouse Model o 558 Ch onic Coli is. Media o s In lamm 2014. 559 43. Rod igues CM, K en BT, S ee CJ, Se chell KD. 1995. Tau ou sodeoxychola e inc eases 560 a li e u sodeoxychola e le els and limi s li hochola e o ma ion be e han 561 u sodeoxychola e. Gas oen e ology 109:564–572. 562 44. Alnou i Y, Csanaky IL, Klaassen CD. 2008. Quan i a i e-P o iling o Bile Acids and hei 563 Conjuga es in Mouse Li e , Bile, Plasma, and U ine Using LC-MS/MS. J Ch oma og B 564 Analy Technol Biomed Li e Sci 873:209–217. 565 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 32 eu hanasia on day 10. (D,E) Cy okine le els o CXCL1, G-CSF and IL-6 in colonic issue (D) 689 and se um (E) collec ed on day 10. Da a a e ep esen ed as he mean ± SEM (n = 8 in each 690 g oup). *P<0.05, **P<0.01. C = Con ol; P = Placebo; T = TUDCA; U = UDCA; G = GUDCA. 691 Figu e 2. O al adminis a ion o UDCA, TUDCA and GUDCA du ing DSS-induced coli is 692 p e en s coli is-associa ed dysbiosis a he phylum le el. Fecal samples we e collec ed on day 693 9 o coli is and mic obio a p o iles we e cha ac e ized by 16S RNA Illumina MiSeq sequencing. 694 (A-C) Es ima ion o (A) species ichness (i.e., o al numbe o ope a ional axonomic uni s), (B) 695 species di e si y (i.e., Shannon index) and (C) bac e ial load in he ecal mic obio a. Bac e ial 696 load was calcula ed as 2del aC / [ o al DNA concen a ion]. (D) Ra io o he pe cen age o 16S 697 RNA gene sequences belonging o Fi micu es and Bac e oide es. (E) Composi ion o he ecal 698 mic obial communi y a he phylum le el. Da a a e ep esen ed as he mean ± SEM (n ≥ 6 in each 699 g oup). *P<0.05, **P<0.01. C = Con ol; P = Placebo; T = TUDCA; U = UDCA; G = GUDCA. 700 Figu e 3. O al adminis a ion o UDCA, TUDCA and GUDCA du ing DSS-induced coli is 701 al e s he ecal mic obio a a lowe axonomic le els. Fecal samples we e collec ed on day 9 o 702 coli is and mic obio a p o iles we e cha ac e ized by 16S RNA Illumina MiSeq sequencing. (A) 703 Composi ion o he ecal mic obial communi y a he class le el. (B-F) Pe cen age o 16S RNA 704 gene sequences belonging o (B) Bac e oidaceae, (C) Po phy omonadaceae, (D) P e o ellaceae, 705 (E) Clos idium clus e XIVa and (F) Akke mansia. (G) qRT-PCR esul s o A. muciniphila. 706 Copy numbe s we e no malized o he 16S RNA gene copy numbe in each sample. Da a a e 707 ep esen ed as he mean ± SEM (n ≥ 6 in each g oup). *P<0.05, **P<0.01. C = Con ol; P = 708 Placebo; T = TUDCA; U = UDCA; G = GUDCA. 709 Figu e 4. O ally adminis e ed UDCA, TUDCA and GUDCA unde go ex ensi e 710 bio ans o ma ion. Fecal samples we e collec ed a day 4 o coli is and bile acids we e 711 quan i ied using UHPLC-HRMS. (A) UDCA, (B) TUDCA, (C) LCA and (D) GUDCA. Da a a e 712 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 33 ep esen ed as he mean ± SEM (n ≥ 7 in each g oup). *P<0.05, **P<0.01. C = Con ol; P = 713 Placebo; T = TUDCA; U = UDCA; G = GUDCA. 714 on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om 0123456789 0 1 2 3 4 5 6 Days DAI sco e ** * CPTUG 0 100 200 300 400 500 ColonCXCL1 (pg/mg o alp o ein) ** ** DSSNo DSS D. Figu e 1 A. E. CPTUG 0 1000 2000 3000 ColonG-CSF (pg/mg o alp o ein) ** ** DSSNo DSS CPTUG 0 50 100 150 Se umIL-6(pg/ml) ** DSSNo DSS CPTUG 0 500 1000 1500 Se umCXCL1 (pg/ml) ** * DSSNo DSS CPTUG 0 10000 20000 30000 Se umG-CSF (pg/ml) ** ** * DSSNo DSS CPTUG 0 100 200 300 400 ColonIL-6 (pg/mg o alp o ein) ** ** DSSNo DSS CPTUG 0 2 4 6 8 10 Colonleng h(cm) ** * p=0.08 DSSNo DSS B. C. 12345678910 -30 -25 -20 -15 -10 -5 0 5 Days %bodyweigh change * DSS + GUDCA DSS + UDCA DSS + TUDCA DSS + Placebo No DSS con ol on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om Rela i e abundance(%) CPTUG 0 10 20 30 40 50 60 70 80 90 100 DSS No DSS Species ichness CPTUG 0 200 400 600 800 1000 DSSNo DSS ** p=0.086 Speciesdi e si y CPTUG 0 1 2 3 4 5 6 DSSNo DSS ** * Bac e ialload CPTUG 0 10 20 30 40 50 60 DSSNo DSS ** F/B a io CPTUG 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 DSSNo DSS ** * Figu e 2 A. B. C. D. E. CPTUG 80 85 90 95 100 DSS No DSS Bac e oide es Candida us Saccha ibac e ia De e ibac e es Fi micu es P o eobac e ia Unclassi ied Bac e ia on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om Rela i e abundance(%) CPTUG 0 10 20 30 40 50 60 70 80 90 100 DSSNo DSS Figu e 3 A. B. C. D. E. Bac e oidaceae Rela i e abundance(%) CPTUG 0 10 20 30 40 50 DSSNo DSS ** * ** Po phy omonadaceae Rela i e abundance(%) CPTUG 0 2 4 6 8*** DSSNo DSS P e o ellaceae Rela i e abundance(%) CPTUG 0 2 4 6 8 DSSNo DSS ** ** Akke mansia Rela i e abundance(%) CPTUG 0.00 0.05 0.10 0.15 0.20 0.25 DSSNo DSS * p=0.06 A. muciniphila No malized copy numbe CPTUG 0 500 1000 1500 2000 2500 3000 DSSNo DSS ** p=0.086 * Clos idium clus e XIVa Rela i e abundance(%) CPTUG 0.0 0.1 0.2 0.3 0.4 0.5 DSSNo DSS ** * F. G. CPTUG 70 75 80 85 90 95 100 DSS No DSS Bac e oidia Unclassi ied Bac e oide es Saccha ibac e ia_gene a_ince ae_sedis De e ibac e es Bacilli Clos idia E ysipelo ichia Unclassi ied Fi micu es Be ap o eobac e ia Del ap o eobac e ia Gammap o eobac e ia Unclassi ied P o eobac e ia Ve ucomic obiae Unclassi ied Bac e ia on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om D. Figu e 4 A. B. C. UDCA ng /mg eces CPTUG 0 500 1000 1500 2000 2500 3000 ** ** DSS No DSS TUDCA ng /mg eces CPTUG 0 50 100 150 200 * ** DSS No DSS LCA ng /mg eces CPTUG 0 200 400 600 800 1000 1200 DSS No DSS ** ** GUDCA ng /mg eces CPTUG 0 50 100 150 200 250 ** DSS No DSS on Feb ua y 20, 2017 by Helmhol z-Zen um ue In ek ions o schung - BIBLIOTHEK-h p://aem.asm.o g/Downloaded om