Reduced adiponectin expression after high-fat diet is associated with selective up-regulation of ALDH1A1 and further retinoic acid receptor signaling in adipose tissue

THE JOURNAL •RESEARCH •www. asebj.o g Reduced adiponec in exp ession a e high- a die is associa ed wi h selec i e up- egula ion o ALDH1A1 and u he e inoic acid ecep o signaling in adipose issue Jean-F ancois Land ie ,* ,†,‡ Elnaz Kasi i, §,{ Esma Ka keni,* ,†,‡ Johanna Mih´ aly, § Gab iella B´ eke, § Ka h in Weiss, § Rena a Lucas, § Gamze Aydemi , § J´ e ome Salles, k S ´ ephane Wal and, k Angel R. de Le a, # and Ralph R ¨ uhl §,{, ** ,1 *Ins i u Na ional de la Reche che Ag onomique, Uni ´ es Mix es de Reche che 1260, Ma seille, F ance; † INSERM, Uni ´ es Mix es de Reche che 1062, Nu i ion, Ob´ esi ´ e e Risque Th ombo ique, Ma seille, F ance; ‡ Aix-Ma seille Uni e si ´ e, Facul ´ edeM ´ edecine, Ma seille, F ance; § Depa men o Biochemis y and Molecula Biology and { MTA-DE Public Heal h Resea ch G oup, Hunga ian Academy o Sciences, Facul y o Public Heal h, Uni e si y o Deb ecen, Deb ecen, Hunga y; k Uni ´ es Mix es de Reche che, Ins i u Na ional de la Reche che Ag onomique (INRA) 1019 Uni ´ e de Nu i ion Humaine, Cen e de Reche ches INRA de Cle mon -Fe and/Theix, Sain -Gen` es-Champanelle, F ance; # Depa amen o de Qu´ ımica O g´ anica, Uni e sidade de Vigo, Facul ad de Qu´ ımica, Cen o de In es igaciones Biom´ edicas and Ins i u o de In es igaci´ on Biom´ edica de Vigo, Vigo, Spain; and **Pap ika Bioanaly ics BT, Deb ecen, Hunga y ABSTRACT: Adiponec in is an adipocy e-de i ed adipokine wi h po en an idiabe ic, an i-in lamma o y, and an i- a he ogenic ac i i y. Long- e m, high- a die esul s in gain o body weigh , adiposi y, u he in lamma o y-based ca dio ascula diseases, and educed adiponec in sec e ion. Vi amin A de i a i es/ e inoids a e in ol ed in se e al o hese p ocesses, which mainly ake place in whi e adipose issue (WAT). In his s udy, we examined adiponec in exp ession as a unc ion o high die a y a and high i amin A condi ions in mice. A dec ease o adiponec in exp ession in addi ion o an up- egula ion o aldehyde dehyd ogenase A1 (ALDH1A1), e inoid signaling, and e inoic acid esponse elemen signaling was selec i ely obse ed in WAT o no mal i amin A–and high- a die – edmice.Reducedadiponec inexp essioninWATwasalsoobse edinhigh i aminAdie – ed mice. Adipocy e cell cul u e e ealed ha endogenous and syn he ic e inoic acid ecep o (RAR)a-andRARg-selec i e agonis s, as well as a syn he ic e inoid X ecep o agonis , e icien ly educed adiponec in exp ession, whe eas ALDH1A1 exp ession only inc eased wi h RAR agonis s. We conclude ha educed adiponec in exp ession unde high- a die a y condi ions is dependen on i) inc eased ALDH1A1 exp ession in adipocy es, which does no inc ease all- ans- e inoic acid le els; ii) u he RAR ligand–induced, WAT-selec i e, inc eased e inoic acid e- sponse elemen –media ed signaling; and iii)RARligand–dependen educ ion o adiponec in exp ession.— Land ie , J.-F., Kasi i, E., Ka keni, E., Mih´ aly, J., B´ eke, G., Weiss, K., Lucas, R., Aydemi , G., Salles, J., Wal and, S., de Le a,A.R.,R¨ uhl, R. Reduced adiponec in exp ession a e high- a die is associa ed wi h selec i e up- egula ion o ALDH1A1and u he e inoicacid ecep o signalinginadipose issue.FASEBJ.31,000–000 (2017).www. asebj.o g KEY WORDS: i amin A •nuclea ho mone ecep o •obesi y •diabe es • e inaldehyde dehyd ogenase Obesi y is conside ed o be one o he mos common nu- i ional diso de s o Wes e n socie y and is cha ac e ized by a disp opo iona e expansion o body a mass [ e iewed in Gasba ini and Piscaglia (1)]. In addi ion o being an ene gy s o age si e, whi e adipose issue (WAT) also unc ions as a highly ac i e me abolic egula o and ABBREVIATIONS: ALDH1A1, aldehyde dehyd ogenase 1A1; ATRA, all- ans- e inoic acid; CTRL, con ol; FABP4, a y acid binding p o ein 4; HF, high a ; HODE, hyd oxyoc adecadienoic acid; LF, low a ; LXR, li e X ecep o ; NF, no mal a ; PPAR, pe oxisome p oli e a o -ac i a ed ecep o ; RALDH, e inaldehyde dehyd ogenase; RAR, e inoic acid ecep o ; RARE, e inoic acid esponse elemen ; RE, e inol equi alen s; RETSAT, all- ans- e inol 13,14- educ ase; RXR, e inoid X ecep o ; TG2, ansglu aminase 2; VDR, i amin D ecep o ; WAT, whi e adipose issue 1 Co espondence: Depa men o P e en i e Medicine, Facul y o Public Heal h, Uni e si y o Deb ecen, Kassai u. 26/b, H-4028 Deb ecen, Hunga y. E-mail: [email p o ec ed] This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial 4.0 In e na ional (CC BY-NC 4.0) (h p://c ea i ecommons.o g/licenses/by-nc/4.0/) which pe mi s noncomme cial use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. doi: 10.1096/ j.201600263RR 0892-6638/17/0031-0001 © The Au ho (s) 1 The FASEB Jou nal a icle j.201600263RR. Published online Oc obe 11, 2016. Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om majo endoc ine o gan ha sec e es a ious adipokines (2–4). Adiponec in is a majo adipokine wi h s ong an i- diabe ic, an i-in lamma o y, and an ia he ogenic ac i i y, and i s exp ession is dec eased in WAT unde high die a y a condi ions [ e iewed in Ouchi e al. (5)]. Wi h he ex- cep ion o he pu a i e ole o in lamma ion (6–8), he p ecise mechanisms ha media e his down- egula ion emain o be elucida ed. Re inoids a e impo an egula o s o adipogenesis. Die s ha a e high in i amin A (9), excess o e inoic acid (10, 11), and die high in b-ca o ene (12–14), esul in inc eased adipocy e apop osis and inhibi ion o adipogenesis, while low concen a ions o e inoic acid we e desc ibed o be p oadipogenic ( e iewed in e s. 9, 11, 15, 16). Re inoids [i.e., na u ally occu ing and syn he ic e inol analogues ( e iewed in e s. 17, 18)], a e esponsible o ac i a ion o speci ic nuclea e- cep o s: he e inoic acid ecep o (RAR) and he e - inoid X ecep o (RXR). Bioac i e e inoic acids a e o med om p ecu so e inaldehydes by he ac ion o e inaldehyde dehyd ogenase enzymes (RALDHs/ ALDH1A) (19). RALDH1/ALDH1A–null adul mice ha e been shown o be esis an o high- a die –induced weigh gain (20, 21), which sugges ed ha ALDH1A1 and i s me abolic p oduc s a e necessa y o high- a die – induced obesi y (22–24). ALDH1A1 can syn hesize e i- noic acids (25), such as all- ans- e inoic acid (ATRA), 9-cis- e inoic acid, and, p esumably, he newly ound endogenous RXR ligand, 9-cis-13,14-dihyd o e inoic acid (26). The las 2 a e ligands o bo h RXRs and RARs, whe eas ATRA only binds RAR. Un o una ely, only a ew s udies ha e de ec ed e inoic acids in low concen a ions in adipose issue (27, 28), bu no s udy has add essed he p esence o e inoic acids in WAT when compa ing ALDH1A1 +/+ o ALDH1A1 2/2 mice. Whe he e inoic acids, and which e inoic acids, a e he majo me aboli es o ALDH1A1 in WAT is ye unknown. Mo eo e , ALDH1A1 exp ession has been shown o be egula ed by li e X ecep o (LXR) (29) as well as es ogen ecep o –media ed pa hways (30, 31). Va ious nuclea ho mone ecep o pa hways a e in- ol ed in adipokine sec e ion and adipocy e di e en ia- ion, p oli e a ion, and lipid accumula ion ( e iewed in e s. 7, 32). In pa icula , RXRs, he cen al he e odime - o ming pa ne s, play impo an oles du ing obesi y (33–35). RXRa-KOaswellasRXRg-KO mice and RXR- an agonis ea men induce esis ance o weigh gain a e high- a die and also p omo e a highe me abolic a e (36–38). RXRs can also in e ac wi h se e al nuclea ecep o s, such as RAR, LXR, pe oxisome p oli e a o - ac i a ed ecep o (PPAR), i amin D ecep o (VDR), o NR4A-o phan nuclea ecep o s (39, 40), and he ac i a ion o a ious so-called pe missi e he e odime s (RXR-PPAR, -LXR, -VDR, and -NR4A1/2) by an RXR ligand can ini ia e he e odime -media ed signaling (39–41). The aim o ou s udy was o ind ou how high- a die educes adiponec in exp ession in WAT, ocusing p i- ma ily on i amin A–media ed RAR- and RXR-dependen pa hways. MATERIALS AND METHODS Expe imen al die s Manually p epa ed die s we e made wi h whea s a ch (Weizens ¨ a ke, Foods a , Ge many; p o ided by K ¨ one -S ¨ a ke, Ibbenb¨ u en, Ge many), saccha ose (pu chased om a local supe ma ke in Hunga y), casein (Sigma-Ald ich, Budapes , Hunga y), cellulose (Vi apu ; JRS Pha ma GmbH; Rosenbe g, Ge many), i amin mix (Vi amin-Vo mischung C1000; Al omin GmbH, Lage, Ge many), mine al mix u e (Mine al- Spu enelemen e-Vo mischung C100; Al omin GmbH), and sun lowe oil (Hen y Lamo e, B emen, Ge many). Animal expe imen s Animal expe imen s we e pe o med in he Labo a o y Animal Co e Facili y o he Uni e si y o Deb ecen. Expe imen s we e pe o med acco ding o Hunga ian e hical guidelines. Expe imen wi h low, no mal, o high- a supplemen a ion die s A e he acclima iza ion pe iod, animals ecei ed a i amin A–de icien [0 e inol equi alen s (RE)/kg die ] die o 10 wk ha con ained 5% sun lowe oil as a die a y lipid, which ep esen ed a die wi h no mal a (NF) con en (42). Animals we e di ided in o di e en eeding g oups (n= 6 pe g oup) and we e ed o 4 wk wi h speci ic die s ha con ained di e en amoun s o die a y a and equal amoun s o i amin A (2500 RE/kg die , no mal i amin A). Sun lowe oil was added as die a y a , which con- ained ei he 2% [as weigh %; low- a (LF) die ], 5% (NF die ), o 25% [high- a (HF) die ]. The sou ce o he a was always sun- lowe oil in di e en p opo ions added o eed. On he basis o he analyzed eed o he NF die , i con ained 11.6% sa u a ed a s, 20% monounsa u a ed a y acids, and 68.4% polyunsa u a ed a y acids (Weiss e al., in p epa a ion). Fu he mo e, die a y composi ion was 180 g/kg casein, 10 g/kg i amin mix, 45 g/kg mine al mix, and 20 g/kg cellulose o all applied die s (42). As a esul o he inc eased amoun o a in he die , ca bohyd a e p opo ion was lowe ; he low LF con ained 29.5% suc ose and 43% s a ch, he NF die 28% suc ose and 41.5% s a ch, and he HF die con ained 17% suc ose and 32.5% s a ch (42). Expe imen wi h no mal o high i amin A supplemen a ion die s Fo i amin con en , die s we e supplemen ed wi h i amin mix (Vi amin-Vo mischung C1000) ha con ained ei he 2500 RE/kg as no mal i amin A die , o o high i amin A die s, an addi- ional e inyl-palmi a e (Re Pal) supplemen ( inal 326,500 RE/kg; Sigma-Ald ich) was added o he no mal i amin A die (42, 43). A e eu hanizing mice, blood collec ion was ca ied ou by ca diac punc u e. Blood was cen i uged o 20 min and plasma was s o ed a 280°C. Mice we e ana omized and WAT samples we e immedia ely ozen in liquid ni ogen a e dissec ion and la e s o ed a 280°C un il RNA ex ac ion. Bioimaging Re inoic acid esponse elemen (RARE)-Luc emale mice (n=6) we e ob ained om Cgene (Oslo, No way) and ecei ed LF, NF, o HF die s o 4 wk o he o al e inoid ea men s as desc ibed be o e (43, 44). We conduc ed ex i o o gan analysis by bioluminescence imaging. All animals we e ea ed wi h 120 mg/kg D-luci e in 2 Vol. 31 Janua y 2017 LANDRIER ET AL.The FASEB Jou nal xwww. asebj.o g Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om (Bioscience, Budapes , Hunga y) ia in ape i oneal injec ions 15 min be o e eu hanasia and u he o gan sc eening. Mice we e eu hanized by ce ical disloca ion. A e sac i ice, mouse li e , WAT, in es ine, and b ain we e collec ed o bioluminescence imaging. O gans we e analyzed o bioluminescence signal by using an Ando -Ixon CCD came a (Bel as , Uni ed Kingdom), and analysis was pe o med by Ando -IQ so wa e. A e im- aging, in eg a ed in ensi y/a ea was calcula ed o li e , WAT, in es ine, and b ain o each ea ed animal. Cell cul u e 3T3-L1 p eadipocy es (Ame ican Type Cul u e Collec ion, Manassas, VA, USA) we e seeded in 3.5-cm-diame e dishes a a densi y o 15 310 4 cells/well. Cells we e g own in DMEM ha was supplemen ed wi h10% FBS a 37°C in a 5% CO 2 humidi ied a mosphe e, as p e iously epo ed (45, 46). To induce di e en- ia ion, 2-d pos con luen 3T3-L1 p eadipocy es (day 0) we e s imula ed o 48 h wi h 0.5 mM isobu ylme hylxan hine, 0.25 mM dexame hasone, and 1 mg/ml insulin in DMEM ha was supplemen ed wi h 10% FBS. Cells we e hen main ained in DMEM ha was supplemen ed wi h 10% FBS and 1 mg/ml in- sulin (47). To examine he e ec s on gene exp ession o ATRA (a gi om BASF AG, Ludwigsha en, Ge many), an RARaagonis (BMS753), an RARgagonis (BMS189961; bo h we e p epa ed in ou labo a o ies as desc ibed in he o iginal pa en s (48, 49)], and an RXR agonis (LG268; gi om Ligand Pha maceu icals, San Diego, CA, USA), 3T3-L1 adipocy es we e incuba ed wi h 1 mM o hese molecules o 24 h, as p e iously epo ed (47). Da a p esen ed a e he mean o 3 independen expe imen s each pe - o med in iplica e. Human adipose biopsies Ele en lean (body mass index: 22.5 60.5 kg/m 2 ) and 14 obese (body mass index: 31.7 60.9 kg/m 2 )malepa icipan swe e ec ui ed,asp e iously epo ed(50). Leanandobese olun ee s we e age 44 67yand4465 y, espec i ely. Subcu aneous adipose issue biopsies we e pe o med be ween 6:30 AM and 7:30 AM a e an o e nigh as . Biopsies we e ob ained by needle aspi a ion in he pe iumbilical a ea unde local anes he- sia. Adipose issue samples we e insed in physiologic se um, immedia ely ozen in liquid ni ogen, and s o ed a 280°C un il RNA ex ac ion. The expe imen al p o ocol was pe o med in acco dance wi h he guidelines in he Decla a ion o Helsinki and was app o ed by he E hical Commi ee o he Au e gne Region (ag eemen No. AU 800, Ma ch 2010). Pa icipan s ga e hei w i en in o med consen o pa icipa e in he s udy. Analysis o mRNA exp ession Analysis o o al cellula RNA ex ac ed om 3T3-L1 cells was pe o med in F ance by using T izol eagen acco ding o man- u ac u e ins uc ions. Human adipose issue sample ex ac ion was also pe o med in F ance, whe eas WAT and li e sample analysis om mice was done in Hunga y. Fo he cell cul u e ma e ial in he F ench labo a o y, cDNA was syn hesized om 1 mgo o alRNAin20mlbyusing andom p ime s and Moloney mu ine leukemia i us e e se ansc ip- ase. Real- ime quan i a i e RT-PCR analyses o genes we e pe o med by using he Mx3005P Real-Time PCR Sys em (S a- agene, La Jolla, CA, USA) as p e iously desc ibed (51). Fo each sample, exp ession was quan i ied in duplica e and 18S RNA was used as he endogenous con ol in he compa a i e cycle h eshold (C T )me hod. Fo WAT and li e issue analysis o human and mu ine o igin, issues we e homogenized in T i eagen solu ion (The mo Fishe Scien i ic, Wal ham, MA, USA) and o al RNA was isola ed om issue acco ding o manu ac u e guidelines and as p e iously de- sc ibed (52). Concen a ion and pu i y o RNA was measu ed by using he NanoD op spec opho ome e (The mo Fishe Scien i ic). Fo eal- ime quan i a i e PCR, o al RNA was e e se an- sc ibed in o cDNA by using he Supe Sc ip II Fi s -S and Syn hesis Sys em (The mo Fishe Scien i ic). Quan i a i e eal- ime PCR was ca ied ou in iplica e using p edesigned MGB assays (The mo Fishe Scien i ic) on an ABI P ism 7900 (Applied Biosys ems, Villebon-su -Y e e, F ance). Rela i e mRNA le els we e calcula ed by using he C me hod and we e no malized o cyclophilin A mRNA. Sequence De ec o So wa e ( . 2.1; Applied Biosys ems) was used o da a analysis. Analy ical p ocedu es WAT samples we e collec ed and s o ed in da k ials a 280°C un il analysis. Sample p epa a ion was pe o med as p e iously desc ibed o e inoid (53) and eicosanoid/docosanoid (54) analysis. HPLC– andem mass spec ome y analyses o e - inoids as well as eicosanoids and docosanoids, which ocused on eicosanoids wi h known PPAR ac i a ion po en ial, we e also pe o med as p e iously explained (53, 54). ELISA assays To examine he e ec o e inoids on adiponec in sec e ion, 3T3-L1 adipocy es we e incuba ed wi h 1 mM o he e inoids (ATRA, RARa,RARg, o RXR ligand) o 48 h. Adiponec in quan i ica ion was ealized on he cul u e supe na an by using adiponec in ELISA assay acco ding o manu ac u e p o ocol (Quan ikine ELISA; R&D Sys ems, Lille, F ance). S a is ics Da a a e exp essed as means 6SEM. Signi ican di e ences be- ween con ol and ea ed cells/g oups we e de e mined by S uden ’s es using S a iew so wa e (SAS Ins i u e, Ca y, NC, USA). Values o P,0.05 we e conside ed signi ican . RESULTS E ec s o HF die on body weigh gain Body weigh gain was obse ed in animals a e 4 wk o HF die compa ed wi h LF o NF die supplemen a- ion (LF: 1.07 60.08 g; NF: 0.92 60.11 g; HF: 3.24 60.37 g; LF-HF P=0.03andNF-HFP=0.04).Foodin akesligh ly dec eased in he HF die g oup (LF: 2.93 g/d/animal; NF: 2.70 g/d/animal; HF: 2.25 g/d/animal). Supplemen a ion o HF die esul s in up- egula ion o ALDH1A1 and down- egula ion o adiponec in exp ession ALDH1A1 was signi ican ly inc eased only in WAT (LF: 160.80; NF: 1.83 60.36; HF: 5.26 60.23) o HF die – supplemen ed mice compa ed wi h LF die –and NF die – ed mice and issue selec i e o WAT (Table 1,WAT) compa ed wi h unchanged exp ession in he li e (Table 1, li e ). ALDH1A2 exp ession emained unchanged in li e and WAT (Table 1). ALDH1A3 was also signi ican ly ADIPONECTIN IS REGULATED DEPENDING ON RAR SIGNALING 3 Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om inc eased in adipose issue o HF die –and NF die – ed animals compa ed wi h LF (Table 1, WAT).In addi ion, exp ession o RAR pa hway a ge genes, such as CYP26A1 and CYP26B1, emained unchanged (Table 2), whe eas exp ession o he highly sensi i e common RAR/RXRpa hway a ge gene ansglu aminase2(TG2) was s ongly inc eased (Table 1; LF: 1 60.36; NF: 9.56 6 0.13; HF: 15.36 60.17) in HF die –supplemen ed mice. In addi ion o inc eased e inoid signaling, exp ession o adiponec in was educed in HF die – ed mice (LF: 1 6 0.31; NF: 0.86 60.12; HF: 0.15 60.37). Inc eased ALDH1A1 and educed adiponec in exp ession in obese olun ee s Expe imen s using adipose issue biopsies om no mal weigh and obese human olun ee s con i med inc eased ALDH1A1 (heal hy olun ee s we e se as 1; 1.20 60.07) and educed adiponec in (0.85 60.04) exp ession in he obese olun ee s (Table 3). High i amin A–supplemen ed die esul s in inc eased exp ession o ALDH1A1 and educed adiponec in exp ession Exp essiono ALDH1A1inc eased(NF, no mal i aminA was se as 1: 1 60.19; NF, high i amin A: 2.32 60.47) in he WAT o high i amin A–and NF die –supplemen ed mice wi h NF con en , whe eas adiponec in exp ession (NF, no mal i amin A was se as 1: 1 60.47; NF, high i amin A: 0.37 60.21) was dec eased in WAT (Table 4). Dec eased e inoic acid concen a ions p esen in WAT o HF die –supplemen ed animals do no co espond o inc eased RARE- media ed signaling in RARE-Luc mice, and PPARgligands emain mainly unchanged Re inol le els emained s able in he WAT o LF-, NF-, and HF-supplemen ed animals, whe eas ATRA (LF: 2.2 60.1 ng/g; NF: 1.7 60.2 ng/g; HF: 0.6 60.1 ng/g) le els we e lowe in he WAT o HF die –supplemen ed animals (Table 5). Inc eased e inoid signaling was con i med in RARE- Luc mice, wi h inc eased RARE-media ed signaling de- ec ed speci ically in adipose issue o HF die –compa ed wi h LF and NF die –supplemen ed animals, whe eas in li e , in es ine, and b ain, no inc eased RARE-media ed signaling was obse ed (Fig. 1). Endogenous PPAR ligands [9-hyd oxyoc adecadienoic acid (HODE), 13-HODE, 13-ke o-oc adecadienoic acid, 12-ke o-eicosa e aenoic acid, PgJ2 and d15d12PgJ2] we e mainly unchanged, excep he adipose issue–speci ic PPARgligand, hepoxilin B3, which is inc eased in adipose issue o HF die –supplemen ed animals (Table 5). TABLE 1. Rela i e adiponec in and ALDH1A1 mRNA exp ession Gene Fold ac i a ion Significance LF NF HF LF:NF NF:HF LF:HF WAT ALDH1A1 1 60.80 1.83 60.36 5.26 60.23 0.46 0.05 0.01 ALDH1A2 1 60.11 1.05 60.14 1.09 60.19 0.77 0.87 0.69 ALDH1A3 1 60.17 1.65 60.08 2.59 60.24 0.03 0.12 0.02 Adiponec in 1 60.31 0.86 60.12 0.15 60.37 0.72 ,0.01 0.04 Li e ALDH1A1 1 60.10 1.10 60.11 1.39 60.13 0.53 0.64 0.09 ALDH1A2 1 60.12 0.79 60.09 0.74 60.10 0.17 0.64 0.09 Exp ession shown in WAT and li e o LF (se as 1), NF, and HF die – ed mice wi h a no mal con en o i amin A in he die . Gene exp ession (all n= 6) o adiponec in and e inoic acid syn hesizing enzymes (ALDH1A1, ALDH1A2, ALDH1A3). Significan alues s. LF a e in i alics. TABLE 2. Rela i e gene exp ession o genes in ol ed in RAR and PPAR signaling in mouse WAT Gene Fold ac i a ion Significance LF NF HF LF:NF NF:HF LF:HF RAR pa hway CYP26A1 1 60.50 0.16 60.41 0.43 60.49 0.13 0.26 0.33 CYP26B1 1 60.63 0.58 60.52 0.93 60.66 0.60 0.65 0.94 TG2 1 60.36 9.56 60.13 15.36 60.17 ,0.01 0.08 ,0.01 PPAR pa hway PPARg160.12 1.23 60.09 0.93 60.09 0.18 0.07 0.67 RETSAT 1 60.21 1.96 60.23 1.38 60.26 0.09 0.37 0.37 FABP4 1 60.03 1.00 60.10 1.08 60.10 0.99 0.57 0.47 FADS2 1 60.43 1.31 60.41 1.46 60.48 0.71 0.88 0.64 Exp ession in WAT o LF (se as 1), NF, and HF die – ed mice wi h a no mal con en o i amin A in die (all n= 6). Significan alues s. LF a e in i alics. 4 Vol. 31 Janua y 2017 LANDRIER ET AL.The FASEB Jou nal xwww. asebj.o g Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om NF and HF die supplemen a ion does no esul in al e ed PPARg-media ed signaling Exp ession o PPARgand PPARg a ge genes e i- nol sa u ase (RETSAT)/ a y acid binding p o ein 4 (FABP4)/FADS2 emaineduna ec edbyNF-andHF- supplemen ed die compa ed wi h LF-supplemen ed die in mouse WAT (Table 2). Adiponec in exp ession is educed by RAR and RXR agonis s using 3T3-L1 adipocy es cell cul u e T ea men o cul u ed adipocy es wi h syn he ic RARa- selec i e ligands [con ol (CTRL) se as 1; adiponec in: 0.22 60.01 and ALDH1A1 2.37 60.04], RARg-selec i e ligands (CTRL se as 1; adiponec in: 0.22 60.01 and ALDH1A1 2.64 60.01), and he na u al RAR ligand ATRA (CTRL se as 1; adiponec in: 0.25 60.03 and ALDH1A1 3.19 60.07), in addi ion o a syn he ic RXR agonis (LG268; CTRL se as 1; adiponec in: 0.51 60.04 and ALDH1A1 1.04 60.04), esul ed in inc eased ALDH1A1 exp ession o RAR agonis s, whe eas adipo- nec in exp ession was educed o all RAR and RXR li- gands. In addi ion, hese esul s we e con i med a he p o ein le el in cell cul u e supe na an s whe e adipo- nec in sec e ion was educed o all adminis e ed RAR and RXR ligands, excep o he RARg-selec i e ligand, which displayed a nonsigni ican dec ease (Table 6). DISCUSSION Obesi y is classically associa ed wi h a dec ease o adi- ponec in plasma le el in humans and oden s, as well as a dec eased exp ession in adipose issue (5). This e- la ionship be ween obesi y and dec eased adiponec in exp ession is suspec ed o be linked o he inc eased in- lamma o y s a us o adipose issue, as TNF-a, one o he main in lamma o y ma ke s p oduced by adipose issue (55), is known o educe adiponec in exp ession (56). Howe e , his mechanism is p obably no exclusi e, and o he pa hways—RAR signaling among hem—could be in ol ed in his egula ion. In his s udy, we epo ed ha , in mice, educed adipo- nec in exp ession in WAT a e HF die supplemen a ion was associa ed wi h an inc ease o ALDH1A1 exp ession. Simila esul s we e also ob ained by compa ing lean s. obese WAT biopsies. Su p isingly, inc eased ALDH1A1 exp ession in mice does no esul in inc eased ATRA le els in WAT. ALDH1A1, he majo enzyme o e inoic acid syn hesis using e inaldehyde as a subs a e, is highly likely o play an impo an ole in he ela ionship be ween e inoid signaling and obesi y (20, 21, 57). Indeed, i s exp ession is inc eased in WAT du ing HF-induced obesi y (58). In ALDH1A1 2/2 - de icien adipocy esaswellasinALDH1A1 2/2 mice, adipogenesis is impai ed and mice a e esis an o HF die –induced obesi y (20), which is sugges ed o be ela ed o al e ed e inoid signaling [(25) and e iewed in e s. 9, 11, 15]. Re inoic acids, he p oduc s om ALDH1A1 me abolism, a e he endogenous ac i a o s o RARs and RXRs. Reduced e inaldehyde and e inol le els we e measu ed in adipose issue o HF die –supplemen ed animals, and ATRA le els we e specula ed o be inc eased upon ALDH1A1 ac i i y (20). Howe e , he de ec ion and quan i ica ion o e inoic acid le els in adipose issue ha e been sca cely examined (27, 28) and, un o una ely, he connec ion o e inoic acids in esponse o ALDH1A1 exp ession in adipose issue has no been s udied be o e. In he p esen s udy, we epo ha inc eased ALDH1A1 exp ession in mice does no esul in inc eased ATRA le els in WAT. On he con a y, ATRA le els we e e en lowe in WAT o HF s. LF o NF die – supplemen ed animals. Simila educed le els o ATRA we e con i med in se um and adipose issue o obese olun ee s compa ed wi h obese olun ee s a e a weigh loss die (unpublished da a), which indica es ha obesi y is ela ed o educed local and sys emic e inoid le els in humans. These indings o educed local e inoid le els in adipose issue o obese animals i well wi h p e ious s udies on i amin A–de icien die – ed animals, which we e ound o become obese upon educed ATRA syn- hesis, le els,and ATRA-media ed signaling[ e iewed in Bone e al. (11)]. In addi ion, i is well es ablished ha e inoids, andespeciallyATRA, as signaling ligands, ha e he abili y o inhibi p oli e a ion o adipocy es; enhance up- egula ion o genes in ol ed in lipid oxida ion, en- e gy dissipa ion, and insulin esponse; and he eby p e en obesi y and insulin esis ance [ e iewed in Bone e al. (11)], p obably by a ge ing adipocy e oxida i e phos- pho yla ion and mi ochond iobiogenesis (59). As a esul o his unclea e idence and inconclusi e de e mina ion o e inoic acids le els in WAT, we op ed, like o he s [(21, 57) plus ollow-up e iews (15, 22)], o an indi ec me hod o de ec ion o e inoid signaling by using a RARE- epo e mouse model (44) and we con i med inc eased WAT-selec i e, RARE-media ed signaling in he WAT o HF die – s.LFdie – ed mice (57). P e ious ex- pe imen al s udies claimed, wi hou any analy ical p oo , TABLE 4. Rela i e adiponec in and ALDH1A1 mRNA exp ession le els in WAT depending on i amin A Gene Fold ac i a ion Significance No mal i amin A High i amin A ALDH1A1 1 60.19 2.32 60.47 0.05 Adiponec in 1 60.47 0.37 60.21 0.02 Exp ession in WAT o no mal i amin A and high i amin A die – ed mice wi h a NF die (se as 1; n= 6). Significan alues s. NF, no mal i amin A a e in i alics. TABLE 3. Rela i e exp ession o human adiponec in and ALDH1A1 in WAT in humans Gene Fold ac i a ion SignificanceNV (n= 20) OB (n= 26) ALDH1A1 1.00 60.08 1.20 60.07 0.03 Adiponec in 1.00 60.04 0.85 60.04 0.01 Exp ession in WAT o obese (OB) and no mal olun ee s (NV). Significan alues s. NV a e in i alics. NV was calcula ed o be se as 1. ADIPONECTIN IS REGULATED DEPENDING ON RAR SIGNALING 5 Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om he in ol emen o ALDH1A1-syn hesized ATRA in adi- pose issue and, only on he basis o inc eased RARE sig- naling (21, 57), ha ATRA is he me aboli e o ALDH1A1 in adipose issue and ha he desc ibed e ec s o ALDH1A1,by consequence, a e media ed by ATRA-RAR signaling. Fu he mo e, hey claimed ha he ALDH1A1 p oduc ATRA mus be in ol ed in he ALDH1A1- media ed inc ease o adipose issue expansion and die - induced obesi y. Ou solid da a, gene a ed by using HPLC– andem mass spec ome y quan i ica ion o ATRA in adipose issue, con adic s hese claims and wa ns abou he common ob ainmen o alse-posi i e da a om RARE- Luc ac i a ion models (60). We concluded ha ei he a s ill-uncha ac e ized endog- enous RAR ligand mus be syn hesized in he WAT o HF die –supplemen ed mice o induce WAT-selec i e, RARE- media ed signaling o al e na i e mechanisms ha possibly in ol e anspo e p o ein–media ed signaling [ e iewed in (61) and specula ed by (62)] o pos - ansla ional modi- ica ions in adipocy es [(21) and e iewed in e . 63] mus be aken in o conside a ion. Wi h ega d o ligands o he han ATRA, i is s ill unknown which RAR- and/o RXR-ac i a ing ligands could be syn hesized upon ALDH1A1 exp ession in WAT, and we could no conclusi ely sugges a possible s uc u e using ou cu en analy ical expe ise (26, 53). Howe e , se e al known and unknown candida es, including 9-cis-andall- ans-13,14- dihyd o e inoic acid, e inal, apo-lycopenoic acids, apo-139- ca o enone, apo-109-ca o enoic acid, and apo-149-ca o enoic acid (20, 26, 44, 47, 64–72) we e ecen ly iden i ied and could cons i u e po en ial endogenous e inoids. To u he exclude he in ol emen o PPARg, he key egula o o adipogenesis (7, 73), as a majo nuclea ecep o esponsible o adiponec in educ ion a e a HF-supplemen ed die , endogenous PPAR ligands we e de e mined. Le els in adipose issue we e mainly un- al e ed a e LF, NF, o HF die supplemen a ion (74–77). Only he le els o he endogenous and adipose issue– speci ic PPARgligand, hepoxilin B3, (78, 79) we e signi - ican ly inc eased in HF die – s. LF die –supplemen ed animals. In addi ion, ou da a show no inc eased exp es- sion o PPARgand known PPARg a ge genes, RETSAT, FABP4,andFADS2, in heWATo HFdie –supplemen ed mice, which, in pa , con as s wi h p e ious s udies. In gene al, inc eased PPARgexp ession in adipose issue a e HF die is mainly ela ed o omen al and no subcu aneous a in humans, as e iewed in (80). In mice, inc eased PPARg exp ession is obse able jus a e die s wi h ex eme Figu e 1. In eg a ed in ensi y a eas o bioluminescence imaging o a ious o gans o RARE-LUC mice (n= 6) ha we e ed wi h LF, NF, and HF die s, wi h no mal i amin A con en in he die . The line o e he ba s indica es s a is ical significance. TABLE 5. HPLC– andem mass spec ome y analysis o e inoids and eicosanoids in WAT Compound Le els (ng/g) Significance LF NF HF LF:NF NF:HF LF:HF Re inoid ATRA 2.2 60.1 1.7 60.2 0.6 60.1 0.19 0.01 ,0.01 ROL 1461 697 1591 694 1520 650 0.35 0.38 0.41 Eicosanoid 13-HODE 557 646 605 679 803 683 0.40 0.21 0.11 9-HODE 186 617 157 618 211 627 0.29 0.22 0.35 13-KODE 228 618 674 6266 433 695 0.22 0.34 0.16 12-KETE 10.3 62.4 7.6 61.1 19.6 64.6 0.31 0.12 0.20 PgJ2 0.2 60.0 0.4 60.0 0.2 60.2 0.08 0.09 0.44 d15d12PgJ2 UQL UQL UQL HXB3 0.5 60.2 2.2 60.5 5.3 61.1 0.08 0.13 0.03 Analysis o e inoids, ATRA and e inol, as well as he endogenous ele an PPAR ligands, 13-HODE, 9-HODE, 13-ke o-oc adecadienoic acid (KODE), 12-ke o-eicosa e aenoic acid (KETE), PgJ2, d15d12PgJ2, and hepoxilin B3 (HXB3); all in ng/g 6SEM o WAT samples om LF, NF, and HF die – ed mice wi h a no mal con en o i amin A in die (all n= 4). Significan alues s. LF a e in i alics. ROL, e inol; UQL, unde he quan ifica ion limi . 6 Vol. 31 Janua y 2017 LANDRIER ET AL.The FASEB Jou nal xwww. asebj.o g Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om HF condi ions, s ong weigh gain, and a e a long ime o HF die supplemen a ion (,8 wk) (81–84). Finally, i is well es ablished ha PPARgsignaling ac i a ion in- c eases sec e ion o adiponec in a he han dec eases i (85). In summa y, all hese da a s ongly imply ha PPARg-media ed signaling in adipose issue o HF die – supplemen ed animals is unlikely o be o majo impo - ance o u he educed adiponec in exp ession. To e alua e RAR- and RXR-media ed signaling pa h- ways in adipocy es di ec ly, 3T3-L1 adipocy e cell cul u e models we e used, and we de e mined ha ALDH1A1 was inc eased a e adminis a ion o ATRA and syn he ic RARa-andRARg-selec i e RAR ligands, and no by a syn he ic RXR ligand, whe eas adiponec in exp ession and sec e ion in he cell supe na an we e dec eased a e adminis a ion o RAR o RXR agonis s. We conclude, he e o e, ha his di ec down- egula ion o adiponec in is an RAR- o RXR-media ed pa hway and ha ALDH1A1 exp ession is egula ed by an RAR ligand. In summa y (Fig. 2), we ound ha educed adiponec in exp ession in he WAT o mice is unde he con ol o e inoid-media ed signaling, mainly ia RAR-media ed signaling pa hways. We sugges ha al e ed e inoid sig- naling in adipose issue is an impo an mechanism o HF die –induced obesi y. In pa icula , ALDH1A1 seems o be hekeyenzyme ha is esponsible o hesyn hesiso al- e na i e endogenous RAR ligands selec i ely in WAT. This inc eased ALDH1A1 and educed adiponec in exp ession was also con i med o occu in adipose issue om obese human olun ee s. Endogenous as well as syn he ic RAR ligands we e shown o u he di ec ly inhibi adiponec in exp ession in cul u ed adipocy es. The na u e o he en- dogenous RAR/RXR agonis s o an agonis s syn hesized by ALDH1A1 in WAT emains elusi e and is he opic o u u e s udies. Cha ac e iza ion o hese no el endogenous e inoids wi h mainly RAR, as well as po en ial RXR, ligand ac i a ion po en ial and hei me abolic pa hways can help cla i y he con o e sy o he al e ed e inoid signaling in adipose issue. On he basis o hese da a, no el s a egies can be de eloped o selec i ely inhibi dis inc e inoid sig- naling, especially ha which in ol es ALDH1A1 p oduc s unde HF die , ocused on adipose issue o enable su icien bene icial adiponec in exp ession. AUTHOR CONTRIBUTIONS R. R¨uhl and J.-F. Land ie designed he expe imen s; E.Kasi i,E.Ka keni,J.Mih ´ aly,G.B ´ eke, K. Weiss, R. Lucas, G. Aydemi , J. Salles, and S. Wal and pe o med he expe imen s; E. Ka keni, J. Mih´ aly, G.B´ eke, and G.Aydemi analysed he da a; and A. R. de Le a p o ided eagen s. REFERENCES 1. Gasba ini, A., and Piscaglia, A. C. (2005) A na u al die e sus mode n Wes e n die s? A new app oach o p e en “well-being syn- d omes”.Dig. Dis. Sci. 50,1–6 2. Ahima, R. S. (2006) Adipose issue as an endoc ine o gan. Obesi y (Sil e Sp ing) 14, 242S–249S 3. Saku ai, T., Ogasawa a, J., Kizaki, T., Ishibashi, Y., Sumi ani, Y., Takahashi, K., Ishida, H., Miyazaki, H., Sai oh, D., Haga, S., Izawa, T., and Ohno, H. (2012) P e en i e and imp o emen e ec so exe cise aining and supplemen in ake in whi e adipose issues on obesi y and li es yle- ela ed diseases. En i on. Heal h P e . Med. 17,348–356 4. Mau y, E., and B icha d, S. M. (2010) Adipokine dys egula ion, adipose issue inflamma ion and me abolic synd ome. Mol. Cell. Endoc inol. 314,1–16 Figu e 2. Simplified scheme showing how HF die induces ALDH1A1 exp ession, inc eased RAR ligand (RAR-LIG), and educed adiponec in exp ession selec i ely in WAT. TABLE 6. Rela i e adiponec in concen a ions and ela i e adiponec in and ALDH1A1 mRNA exp ession in 3T3-adipocy es Adiponec in ALDH1A1 Adiponec in ALDH1A1 Re inoid (ELISA) ( ela i e exp ession) ( ela i e exp ession) (ELISA) ( ela i e exp ession) ( ela i e exp ession) ATRA 0.57 60.14 0.25 60.03 3.19 60.07 0.02 ,0.01 ,0.01 RARa-LIG 0.67 60.04 0.22 60.01 2.37 60.04 0.05 ,0.01 0.01 RARg-LIG 0.81 60.04 0.22 60.01 2.64 60.01 0.20 ,0.01 ,0.01 RXR-LIG 0.64 60.04 0.51 60.04 1.04 60.04 0.05 0.03 0.14 Exp ession a e 24 h in cul u ed 3T3-L1 adipocy es wi h ATRA (1 mM), an RARa-specific agonis BMS753/RARa-LIG (1 mM), an RARg- specific agonis BMS189961/RARg-LIG (1 mM), and an RXR ligand RXR-LIG/LG268 (1 mM) calcula ed wi h CTRL ea men s se as 1. Significance and SEM a e based on n= 6 pa allel ea men s. Significan alues s. CTRL a e in i alics. LIG, ligand. ADIPONECTIN IS REGULATED DEPENDING ON RAR SIGNALING 7 Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om 5. Ouchi, N., Pa ke , J. L., Lugus, J. J., and Walsh, K. (2011) Adipokines in inflamma ion and me abolic disease. Na . Re . Immunol. 11,85–97 6. Ka ki,S.,Chak aba i,P.,Huang,G.,Wang,H.,Fa me ,S.R.,and Kand o , K. V. (2011) The mul i-le el ac ion o a y acids on adipo- nec in p oduc ion by a cells. PLoS One 6, e28146 7. Iwaki, M., Ma suda, M., Maeda, N., Funahashi, T., Ma suzawa, Y., Makishima, M., and Shimomu a, I. (2003) Induc ion o adiponec in, a a -de i ed an idiabe ic and an ia he ogenic ac o , by nuclea e- cep o s. Diabe es 52, 1655–1663 8. Lagishe y, V., Nandiwada, V. B., Kalashikam, R. R., and Manchala, R. (2007) E ec o ma e nal i amin and mine al es ic ions on he body a con en and adipocy okinele els o WNIN a o sp ing.Nu . Me ab. (Lond.) 4,21 9. Bone , M. L., Ribo , J., Felipe, F., and Palou, A. (2003) Vi amin A and he egula ion o a ese es. Cell. Mol. Li e Sci. 60, 1311–1321 10. Bone , M. L., Puigse e , P., Se a, F., Ribo , J., V´azquez, F., Pico, C., and Palou, A. (1997) Re inoic acid modula es e inoid X ecep o alpha and e inoic acid ecep o alpha le els o cul u ed b own adipocy es. FEBS Le . 406, 196–200 11. Bone , M. L., Ribo , J., and Palou, A. (2012) Lipid me abolism in mammalian issues and i s con ol by e inoic acid. Biochim. Biophys. Ac a 1821,177–189 12. Lobo, G. P., Amengual, J., Li, H. N., Golczak, M., Bone , M. L., Palczewski, K., and on Lin ig, J. (2010) Be a,be a-ca o ene dec eases pe oxisome p oli e a o ecep o gamma ac i i y and educes lipid s o age capaci y o adipocy es in a be a,be a-ca o ene oxygenase 1-dependen manne . J. Biol. Chem. 285, 27891–27899 13. Canas, J. A., Damaso, L., Al oma e, A., Killen, K., Hossain, J., and Balagopal, P. B. (2012) Insulin esis ance and adiposi y in ela ion o se um b-ca o ene le els. J. Pedia . 161,58–64.e1–2 14. Amengual, J., Gou an on, E., an Helden, Y. G., Hessel, S., Ribo , J., K ame , E., Kiec-Wilk, B., Razny, U., Lie z, G., Wyss, A., Dembinska-Kiec, A., Palou, A., Keije , J., Land ie , J. F., Bone , M. L., and on Lin ig, J. (2011) Be a-ca o ene educes body adiposi y o mice ia BCMO1. PLoS One 6, e20644 15. Yasmeen, R., Jeyakuma , S. M., Reiche , B., Yang, F., and Ziouzenko a, O. (2012) The con ibu ion o i amin A o au oc ine egula ion o a depo s. Biochim. Biophys. Ac a 1821, 190–197 16. Ma co o chino, J., Tou niai e, F., and Land ie , J. F. (2013) Vi amin D, adipose issue, and obesi y. Ho m. Mol. Biol. Clin. In es ig. 15, 123–128 17. R¨uhl, R. (2007) E ec s o die a y e inoids and ca o enoids on immune de elopmen . P oc. Nu . Soc. 66,458–469 18. Blomho , R., and Blomho , H. K. (2006) O e iew o e inoid me abolism and unc ion. J. Neu obiol. 66,606–630 19. Napoli, J. L. (1999) In e ac ions o e inoid binding p o eins and enzymes in e inoid me abolism. Biochim. Biophys. Ac a 1440,139–162 20. Ziouzenko a, O., O asanu, G., Sha lach, M., Akiyama, T. E., Be ge , J.P.,Vie eck,J.,Hamil on,J.A.,Tang,G.,Dolnikowski,G.G.,Vogel, S., Dues e , G., and Plu zky, J. (2007) Re inaldehyde ep esses adipogenesis and die -induced obesi y. Na . Med. 13,695–702 21. Reiche , B., Yasmeen, R., Jeyakuma , S. M., Yang, F., Thomou, T., Alde , H., Dues e , G., Maiseyeu, A., Mihai, G., Ha ison, E. H., Rajagopalan, S., Ki kland, J. L., and Ziouzenko a, O. (2011) Conce ed ac ion o aldehyde dehyd ogenases influences depo - specific a o ma ion. Mol. Endoc inol. 25,799–809 22. Pe osino, J. M., Disil es o, D., and Ziouzenko a, O. (2014) Aldehyde dehyd ogenase 1A1: iend o oe o emale me abolism? Nu ien s 6, 950–973 23. Mcil oy,G.D.,Delibego ic,M., Owen,C., S oney, P.N.,Shea e ,K.D., McCa e y, P.J., and Mody,N. (2013) Fen e inide ea men p e en s die -induced obesi y in associa ion wi h majo al e a ions in e inoid homeos a ic gene exp ession in adipose, li e , and hypo halamus. Diabe es 62,825–836 24. Zhang,M.,Liu,C.,Hu,M.Y.,Zhang,J.,Xu,P.,Li,F.,Zhong,Z.Y.,Liu, L., and Liu, X. D. (2015) High- a die enhanced e inal de- hyd ogenase ac i i y, bu supp essed e inol dehyd ogenase ac i i y in li e o a s. J. Pha macol. Sci. 127,430–438 25. Gagnon, I., Dues e , G., and Bha , P. V. (2003) Enzyma ic cha ac e iza ion o ecombinan mouse e inal dehyd ogenase ype 1. Biochem. Pha macol. 65,1685–1690 26. R¨uhl, R., K zy˙zosiak,A.,Niewiadomska-Cimicka,A.,Rochel,N., Szeles, L., Vaz, B., Wie zych-Schindle , M., ´ Al a ez, S., Szklena , M.,Nagy,L.,deLe a,A.R.,andK e ˛˙zel, W. (2015) 9-cis-13,14- dihyd o e inoic acid is an endogenous e inoid ac ing as RXR ligand in mice. PLoS Gene . 11, e1005213 27. Ob och a, K. M., Kane, M. A., and Napoli, J. L. (2014) E ec s o die and s ain on mouse se um and issue e inoid concen a ions. PLoS One 9,e99435 28. Kane,M.A.,Folias,A.E.,Wang,C.,andNapoli,J.L.(2008) Quan i a i e p ofiling o endogenous e inoic acid in i o and in i o by andem mass spec ome y. Anal. Chem. 80, 1702–1708 29. Huq,M.D.,Tsai,N.P.,Gup a,P.,andWei,L.N.(2006)Regula iono e inal dehyd ogenases and e inoic acid syn hesis by choles e ol me aboli es. EMBO J. 25,3203–3213 30. R¨uhl, R., F i zsche, B., Ve mo , J., Niede ei he , K., Neumann, U., Schmid , A., Schweige , F. J., and Doll´e, P. (2006) Regula ion o exp ession o he e inoic acid-syn hesising enzymes e inaldehyde dehyd ogenases in he u e i o o a iec omised mice a e ea men wi h oes ogen, ges agen and hei combina ion. Rep od. Fe il. De . 18, 339–345 31. Ve mo , J., F aulob, V., Doll´ e, P., and Niede ei he , K. (2000) Exp ession o enzymes syn hesizing (aldehyde dehyd ogenase 1 and einaldehyde dehyd ogenase 2) and me abolizaing (Cyp26) e inoic acid in he mouse emale ep oduc i e sys em. Endoc inology 141, 3638–3645 32. Masoodi, M., Kuda, O., Rossmeisl, M., Flachs, P., and Kopecky, J. (2015) Lipid signaling in adipose issue: connec ing inflamma ion & me abolism. Biochim. Biophys. Ac a 1851,503–518 33. Des e gne, B. (2007) RXR: om pa ne ship o leade ship in me abolic egula ions. Vi am. Ho m. 75,1–32 34. Szan o, A., Na ka , V., Shen, Q., U ay, I. P., Da ies, P. J., and Nagy, L. (2004) Re inoid X ecep o s: X-plo ing hei (pa ho)physiological unc ions. Cell Dea h Di e . 11,S126–S143 35. Shulman, A. I., and Mangelsdo , D. J. (2005) Re inoid X ecep o he e odime s in he me abolic synd ome. N. Engl. J. Med. 353, 604–615 36. Yamauchi,T.,Waki,H.,Kamon,J.,Mu akami,K.,Mo ojima,K., Komeda, K., Miki, H., Kubo a, N., Te auchi, Y., Tsuchida, A., Tsuboyama-Kasaoka, N., Yamauchi, N., Ide, T., Ho i, W., Ka o, S., Fukayama, M., Akanuma, Y., Ezaki, O., I ai, A., Nagai, R., Kimu a, S., Tobe, K., Kagechika, H., Shudo, K., and Kadowaki, T. (2001) Inhibi ion o RXR and PPARgamma amelio a es die -induced obe- si y and ype 2 diabe es. J. Clin. In es . 108, 1001–1013 37. Imai, T., Jiang, M., Chambon, P., and Me zge , D. (2001) Impai ed adipogenesis and lipolysis in he mouse upon selec i e abla ion o he e inoid X ecep o alpha media ed by a amoxi en-inducible chi- me icC e ecombinase (C e-ERT2)inadipocy es.P oc.Na l.Acad.Sci. USA 98,224–228 38. Me zge , D., Imai, T., Jiang, M., Takukawa, R., Des e gne, B., Wahli, W., and Chambon, P. (2005) Func ional ole o RXRs and PPARgamma in ma u e adipocy es. P os aglandins Leuko . Essen . Fa y Acids 73,51–58 39. Mangelsdo , D. J., Thummel, C., Bea o, M., He lich, P., Sch¨u z, G., Umesono,K.,Blumbe g,B.,Kas ne ,P.,Ma k,M.,Chambon,P.,and E ans, R. M. (1995) The nuclea ecep o supe amily: he second decade. Cell 83,835–839 40. Mangelsdo , D. J., and E ans, R. M. (1995) The RXR he e odime s and o phan ecep o s. Cell 83,841–850 41. Pe lmann, T., and Jansson, L. (1995) A no el pa hway o i amin A signaling media ed by RXR he e odime iza ion wi h NGFI-B and NURR1. Genes De . 9,769–782 42. Weiss, K., Mih´ aly, J., Liebisch, G., Ma os ¨ olgyi, T., Ga cia, A. L., Schmi z, G., Decsi, T., and R¨uhl, R. (2014) E ec o high e sus low doses o a and i amin A die a y supplemen a ion on a y acid composi ion o phospholipids in mice. Genes Nu . 9,368 43. Mih´ aly, J., Ge icke, J., Aydemi , G., Weiss, K., Ca lsen, H., Blomho , R., Ga cia, J., and R¨uhl, R. (2012) Reduced e inoid signaling in he skin a e sys emic e inoid-X ecep o ligand ea men in mice wi h po en ial ele ance o skin diso de s. De ma ology (Basel) 225,304–311 44. Aydemi , G., Ca lsen, H., Blomho , R., andR¨uhl, R. (2012) Lycopene induces e inoic acid ecep o ansc ip ional ac i a ion in mice. Mol. Nu . Food Res. 56,702–712 45. Land ie , J. F., Gou an on, E., El Yazidi, C., Maleze , C., Balague , P., Bo el, P., and Amio , M. J. (2009) Adiponec in exp ession is induced by i amin E ia a pe oxisome p oli e a o -ac i a ed ecep o gamma- dependen mechanism. Endoc inology 150, 5318–5325 46. Ma co o chino,J., Gou an on,E.,Romie ,B.,Tou niai e,F.,As ie ,J., Maleze , C., Amio , M. J., and Land ie , J. F. (2012) Vi amin D educes he inflamma o y esponseand es o es glucose up ake in adipocy es. Mol.Nu .FoodRes.56, 1771–1782 47. Gou an on, E., Aydemi , G., Reynaud, E., Ma co o chino, J., Maleze , C., Ca is-Vey a , C., Blomho , R., Land ie , J. F., and R¨uhl, R. (2011) Apo-109-lycopenoic acid impac s adipose issue biology ia he e inoic acid ecep o s. Biochim. Biophys. Ac a 1811, 1105–1114 8 Vol. 31 Janua y 2017 LANDRIER ET AL.The FASEB Jou nal xwww. asebj.o g Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om 48. Swann, R. T., Smi h, D. E., T amposch, K. M., and Zusi, F. C. (1996) P epa a ion and RARg-specific e inoic ecep o ansaci a ion o e inobenzoic acid de i a i es. U.S. Pa en : 5624957 A. Washing on, DC,Ap il29,1997 49. Zusi, F. C., Reczek, P. R., and Os owski, J. (1998) P epa a ion o 5- subs i u ed-1,1,3,3- ame hyl-2-ke oindanes as e inoid-like compounds. Eu opean Pa en 98912117.3. Munich, Ge many, Janua y 19, 2000 50. Tou niai e, F., Romie -C ouze , B., Lee, J. H., Ma co o chino, J., Gou an on,E.,Salles,J.,Maleze ,C.,As ie ,J.,Da mon,P.,Blouin,E., Wal and, S., Ye, J., and Land ie , J. F. (2013) Chemokine exp ession in inflamed adipose issue is mainly media ed by NF-kB. PLoS One 8, e66515 51. Land ie , J. F., Maleze -Desmoulins, C., Reboul, E., Ma ie Lo ec, A., Josephe Amio , M., and Bo el, P. (2008) Compa ison o di e en ehicles o s udy he e ec o ocophe ols on gene exp ession in in es inal cells. F ee Radic. Res. 42,523–530 52. Ka keni, E., Ma co o chino, J., Tou niai e, F., As ie , J., Pei e i, F., Da mon, P., and Land ie , J. F. (2015) Vi amin D limi s chemokine exp ession in adipocy es and mac ophage mig a ion in i o and in male mice. Endoc inology 156, 1782–1793 53. R¨uhl, R. (2006) Me hod o de e mine 4-oxo- e inoic acids, e inoic acids and e inol in se um and cell ex ac s by liquid ch oma og aphy/ diode-a ay de ec ion a mosphe ic p essu e chemical ionisa ion andem mass spec ome y. Rapid Commun. Mass Spec om. 20, 2497–2504 54. Szklena , M., Kalkowski, J., S angl, V., Lo enz, M., and R¨uhl, R. (2013) Eicosanoids and docosanoids in plasma and ao a o heal hy and a he oscle o ic abbi s. J. Vasc. Res. 50,372–382 55. G ego , M. F., and Ho amisligil, G. S. (2011) Inflamma o y mechanisms in obesi y. Annu. Re . Immunol. 29,415–445 56. Fasshaue , M., Klein, J., Neumann, S., Eszlinge , M., and Paschke, R. (2002) Ho monal egula ion o adiponec in gene exp ession in 3T3- L1 adipocy es. Biochem. Biophys. Res. Commun. 290, 1084–1089 57. Yasmeen, R., Reiche , B., Deiuliis, J., Yang, F., Lynch, A., Meye s, J., Sha lach,M.,Shin,S.,Volz,K.S.,G een,K.B.,Lee,K.,Alde ,H.,Dues e , G., Zechne , R., Rajagopalan, S., and Ziouzenko a, O. (2013) Au oc ine unc ion o aldehyde dehyd ogenase 1 as a de e minan o die - and sex- specific di e ences in isce al adiposi y. Diabe es 62,124–136 58. Kie e , F. W., Ve noche , C., O’B ien,P.,Spoe l,S.,B own,J.D., Nallamshe y, S., Zeyda, M., S ulnig, T. M., Cohen, D. E., Kahn, C. R., and Plu zky, J. (2012) Re inaldehyde dehyd ogenase 1 egula es a he mogenic p og am in whi e adipose issue. Na . Med. 18,918–925 59. Tou niai e, F., Musino ic, H., Gou an on, E., As ie , J., Ma co o chino, J., A eguin, A., Be no , D., Palou, A., Bone , M. L., Ribo , J., and Land ie , J. F. (2015) All- ans e inoic acid induces oxida i e phospho yla ion and mi ochond ia biogenesis in adipocy es. J. Lipid Res. 56, 1100–1109 60. Napoli, J. L. (2012) Physiological insigh s in o all- ans- e inoic acid biosyn hesis. Biochim. Biophys. Ac a 1821,152–167 61. F ey, S. K., and Vogel, S. (2011) Vi amin A me abolism and adipose issue biology. Nu ien s 3,27–39 62. Noy, N. (2013) The one- wo punch: Re inoic acid supp esses obesi y bo h by p omo ing ene gy expendi u e and by inhibi ing adipo- genesis. Adipocy e 2, 184–187 63. Ahmadian, M., Suh, J. M., Hah, N., Liddle, C., A kins, A. R., Downes, M., and E ans, R. M. (2013) PPARgsignaling and me abolism: he good, he bad and he u u e. Na . Med. 19,557–566 64. Aydemi , G., Kasi i, Y., Ba ´ok,E.M.,Bi a,E.,F ¨ohlich, K., B¨ohm, V., Mihaly, J., and R¨uhl, R. (2016) Lycopene supplemen a ion es o es i amin A deficiency in mice and possesses he eby pa ial p o- i amin A ac i i y ansmi ed ia RAR signaling. [E-pub ahead o p in ] Mol. Nu .FoodRes.10.1002/mn .201600031 65. Moise, A. R., Kuksa, V., Blane , W. S., Baeh , W., and Palczewski, K. (2005) Me abolism and ansac i a ion ac i i y o 13,14- dihyd o e inoic acid. J. Biol. Chem. 280, 27815–27825 66. Sun, J., Na ayanasamy, S., Cu ley, R. W., J ., andHa ison, E. H.(2014) b-Apo-13-ca o enone egula es e inoid X ecep o ansc ip ional ac i i y h ough e ame iza ion o he ecep o . J. Biol. Chem. 289, 33118–33124 67. Wang, C. X., Jiang, H., Yuen, J. J., Lee, S. A., Na ayanasamy, S., Cu ley, R. W., J ., Ha ison, E. H., and Blane , W. S. (2015) Ac ions o b-apo- ca o enoids in di e en ia ing cells: di e en ial e ec s in P19cells and 3T3-L1 adipocy es. A ch. Biochem. Biophys. 572,2–10 68. E oglu, A., H uszkewycz, D.P., dela Sena, C., Na ayanasamy, S., Riedl, K.M.,Kopec,R.E.,Schwa z,S.J.,Cu ley,R.W.,J .,andHa ison, E. H. (2012) Na u ally occu ing eccen ic clea age p oduc s o p o i amin A b-ca o ene unc ion as an agonis s o e inoic acid ecep o s. J. Biol. Chem. 287, 15886–15895 69. Bone , M. L., Canas, J. A., Ribo , J., and Palou, A. (2015) Ca o enoids and hei con e sion p oduc s in he con ol o adipocy e unc ion, adiposi y and obesi y. A ch.Biochem.Biophys.572, 112–125 70. Sima, A., Manolescu, D. C., and Bha , P. (2011) Re inoids and e inoid-me abolic geneexp essioninmouseadipose issues. Biochem. Cell Biol. 89, 578–584 71. Aydemi , G., Kasi i, Y., Bi a, E., B´ eke, G., Ga cia, A. L., Ba ´ ok,E.M., and R¨uhl, R. (2013) Lycopene-de i ed bioac i e e inoic acid ecep o s/ e inoid-X ecep o s-ac i a ing me aboli es may be ele an o lycopene’s an i-cance po en ial. Mol. Nu . Food Res. 57, 739–747 72. de Le a, A. R., K ezel, W., and R¨uhl, R. (2016) An endogenous mammalian e inoid X ecep o ligand, a las ! ChemMedChem 11, 1027–1037 73. Tishinsky, J. M., Ma, D. W., and Robinson, L. E. (2011) Eicosapen aenoic acid and osigli azone inc ease adiponec in in an addi i e and PPARg-dependen manne in human adipocy es. Obesi y (Sil e Sp ing) 19, 262–268 74. Dozsa, A., Mihaly, J., Dezso, B., Csizmadia, E., Ke esz essy, T., Ma ko, L., R¨uhl,R.,Remenyik,E.,andNagy,L.(2016)Dec eased pe oxisome p oli e a o -ac i a ed ecep o gle el and signalling in sebaceous glands o pa ien s wi h acne ulga is. Clin. Exp. De ma ol. 41,547–551 75. Dozsa, A., Dezso, B., To h, B. I., Bacsi, A., Poliska, S., Came a, E., Pica do, M., Zouboulis, C. C., B´ı ´o, T., Schmi z, G., Liebisch, G., R¨uhl, R., Remenyik, E., and Nagy, L. (2014) PPARg-media ed and a achidonic acid-dependen signaling is in ol ed in di e en ia ion and lipid p oduc ion o human sebocy es. J. In es . De ma ol. 134, 910–920 76. Nagy,L.,Ton onoz,P.,Al a ez,J.G.,Chen,H.,andE ans,R.M. (1998) Oxidized LDL egula es mac ophage gene exp ession h ough ligand ac i a ion o PPARgamma. Cell 93,229–240 77. Flachs, P., Rossmeisl, M., B yhn, M., and Kopecky, J. (2009) Cellula and molecula e ec s o n-3 polyunsa u a ed a y acids on adipose issue biology and me abolism. Clin. Sci. 116,1–16 78. Hallenbo g, P., Jø gensen, C., Pe e sen, R. K., Fedde sen, S., A aujo, P., Ma k , P., Lange , T., Fu s enbe ge , G., K ieg, P., Koppen, A., Kalkho en, E., Madsen, L., and K is iansen, K. (2010) Epide mis- ype lipoxygenase 3 egula es adipocy e di e en ia ion and pe oxisome p oli e a o -ac i a ed ecep o gamma ac i i y. Mol. Cell. Biol. 30, 4077–4091 79. Hallenbo g, P., Pe e sen, R. K., Kouskoum ekaki, I., Newman, J. W., Madsen, L., and K is iansen, K. (2016) The elusi e endogenous adipogenic PPARgagonis s: lining up he suspec s. P og. Lipid Res. 61, 149–162 80. La sen,T.M., Toub o,S.,andAs up,A.(2003)PPARgammaagonis s in he ea men o ype II diabe es: is inc eased a ness commensu a e wi h long- e m e ficacy? In . J. Obes. Rela . Me ab. Dis- o d. 27, 147–161 81. Jones,J.R.,Ba ick,C., Kim,K. A., Lindne ,J.,Blondeau,B.,Fujimo o, Y.,Shio a,M.,Kes e son,R.A.,Kahn,B.B.,andMagnuson,M.A. (2005) Dele ion o PPARgamma in adipose issues o mice p o ec s agains high a die -induced obesi y and insulin esis ance. P oc. Na l. Acad. Sci. USA 102,6207–6212 82. Inoue, M., Oh ake, T., Mo omu a, W., Takahashi, N., Hosoki, Y., Miyoshi, S., Suzuki, Y., Sai o, H., Kohgo, Y., and Okumu a, T. (2005) Inc eased exp ession o PPARgamma in high a die - induced li e s ea osis in mice. Biochem. Biophys. Res. Commun. 336, 215–222 83. Gao, M., Ma, Y., and Liu, D. (2015) High- a die -induced adiposi y, adipose inflamma ion, hepa ic s ea osis and hype insulinemia in ou b ed CD-1 mice. PLoS One 10, e0119784 84. Kubo a, N., Te auchi, Y., Miki, H., Tamemo o, H., Yamauchi, T., Komeda, K., Sa oh, S., Nakano, R., Ishii, C., Sugiyama, T., E o, K., Tsubamo o, Y., Okuno, A., Mu akami, K., Sekiha a, H., Hasegawa, G., Nai o, M., Toyoshima, Y., Tanaka, S., Shio a, K., Ki amu a, T., Fuji a, T., Ezaki, O., Aizawa, S., and Kadowaki, T., e al. (1999) PPAR gamma media es high- a die -induced adipocy e hype ophy and insulin esis ance. Mol. Cell 4,597–609 85. Maeda, N., Takahashi, M., Funahashi, T., Kiha a, S., Nishizawa, H., Kishida, K., Naga e ani, H., Ma suda, M., Komu o, R., Ouchi, N., Ku iyama, H., Ho a, K., Nakamu a, T., Shimomu a, I., and Ma suzawa, Y. (2001) PPARgamma ligands inc ease exp ession and plasma concen a ions o adiponec in, an adipose-de i ed p o ein. Diabe es 50, 2094–2099 Recei ed o publica ion Feb ua y 12, 2016. Accep ed o publica ion Sep embe 22, 2016. ADIPONECTIN IS REGULATED DEPENDING ON RAR SIGNALING 9 Vol., No. , pp:, Oc obe , 2016The FASEB Jou nal. 193.6.136.39 o IP www. asebj.o gDownloaded om