A previously uncharacterized Factor Associated with Metabolism and Energy (FAME/C14orf105/CCDC198/1700011H14Rik) is related to evolutionary adaptation, energy balance, and kidney physiology

A icle h ps://doi.o g/10.1038/s41467-023-38663-7 A p e iously uncha ac e ized Fac o Associa ed wi h Me abolism and Ene gy (FAME/C14o 105/CCDC198/1700011H14Rik) is ela ed o e olu iona y adap a ion, ene gy balance, and kidney physiology A lis o au ho s and hei a filia ions appea s a he end o he pape In his s udy we use compa a i e genomics o unco e a gene wi h uncha - ac e ized unc ion (1700011H14Rik/C14o 105/CCDC198), which we he eby name FAME (Fac o Associa ed wi h Me abolism and Ene gy). We obse e ha FAME shows an unusually high e olu iona y di e gence in bi ds and mammals. Th ough he compa ison o single nucleo ide polymo phisms, we iden i y gene flow o FAME om Neande als in o mode n humans. We conduc knockou expe imen s on animals and obse e al e ed body weigh and dec eased ene gy expendi u e in Fame knockou animals, co esponding o genome-wide associa ion s udies linking FAME wi h highe body mass index in humans. Gene exp ession and subcellula localiza ion analyses e eal ha FAME is a memb ane-bound p o ein en iched in he kidneys. Al hough he gene knockou esul s in s uc u ally no mal kidneys, we de ec highe albu- min in u ine and lowe ed e i in in he blood. Th ough expe imen al alida- ion, we confi m in e ac ions be ween FAME and e i in and show co- localiza ion in esicula and plasma memb anes. Th ough na u al selec ion, majo animal g oups ha e de eloped unique mechanisms o adap a ions o he en i onmen . The gene ic landscape co esponds o de elopmen al, mo phological, and phy- siological adap a ions1. Gene egula o y egions unde go apid e olu- iona y change o une he p oduc ion o mRNAs encoding he ac ual e ec o s o adap a ion, he p o eins. The gene p oduc s also unde go na u al selec ion, which shapes hem acco ding o a ious benefi s de i ed om hei unc ions2. Impo an ly, no all such p oduc s, including mainly p o eins, a e essen ial o basic emb yonic de elop- men and he me e su i al o animals. Nume ous gene knockou expe imen s in mice ha e highligh ed a coho o p o eins ha a e unc ionally impo an o some ex en , and ye he animals can li e pe ec ly wi hou hem unde beneficial ci cums ances3. These non- essen ial p o eins con ey an adap i e ad an age o hei hos s when he animals a e exposed o he di e si y o challenging na u al en i onmen s4. Fu he mo e, hese seemingly non-essen ial genes migh become a basis o di e se pa hologies o loss o fi ness5,6. Finally, simila o essen ial genes, non-essen ial genes migh be a pe ec subs a e o e olu ion, especially o uning me abolic, s ess- and ene gy- ela ed ea u es. Being non-essen ial, such genes can e ol e much as e o p o ide he necessa y e ol abili y unde in ense selec i e p essu e. In ex eme cases, he e olu ion o p o eins (especially when i comes o a non-essen ial g oup) migh esul in a comple e change o unc ion7,8 o a pseudogeniza ion, which occu s in genes in ol ed in den al gen- esis in bi ds, u les, and oo hless mammals9. The compa a i e genomics app oach10 is pe ec ly designed o elucida e genomic p o ein-coding and non-coding egions esponsible Recei ed: 24 Sep embe 2021 Accep ed: 11 May 2023 Check o upda es e-mail: julian.pe e [email protected];i[email p o ec ed] Na u e Communica ions | (2023) 14:3092 1 1234567890():,; 1234567890():,; o di e en animal g oups’di e gence and adap i e adia ion11,12.Fo ins ance, such analyses b ough o wa d he genomic changes asso- cia ed wi h “bi dness”o “mammalness”.Manyiden ified egions appea ed o be in ol ed in he e olu ion o egg p oduc ion, placen al de elopmen , o geni al shaping11,12.Indeed, hedi e genceo majo e eb a e g oups such as bi ds, ep iles, and mammals esul ed in hea y modifica ions o me abolism13,14, ep oduc ion15, and exc e ion16, oge he wi h associa ed genomic changes and adap ing p o ein s uc u es. Fo example, he p ocesses o wa e and nu ien e- abso p ion in bi ds and mammals di e d ama ically a s uc u al, cellula , and molecula le els, including gene ics. Bi ds and ep iles p edominan ly exc e e u ic acid ins ead o he u ea used by mammals and, hus, ely on negligible amoun s o wa e o ni ogen exc e ion17. Al hough many o he compa a i e s udies pinpoin ed well- cha ac e ized genes ha can be analyzed in he con ex o unc ional ne wo ks in ol ed in he di e ging o gan sys ems and physiological unc ions, he uncha ac e ized genes emained enigma ic in his e o- lu iona y pa adigm. Al hough he human and mouse genomes con ain a ound 20.000 p o ein-coding genes, no all o hese a e iden ified, anno a ed, and cha ac e ized in e ms o hei exp ession and biolo- gical unc ion18–20. Cha ac e izing such genes unc ionally and in es i- ga ing hei e olu iona y oles a e essen ial o comple e he holis ic pic u e o genome ans o ma ions h ough ime. He e we unco e ed an uncha ac e ized p o ein-coding 1700011H14Rik/C14o 105/CCDC198 gene he eby named FAME (Fac- o Associa ed wi h Me abolism and Ene gy), ha e ol es a an excep- ional a e in bi ds and mammals. Specific alleles o FAME flow om Neande als in o mode n humans, highligh ing i s in ol emen in ou fi ness. We add essed he exp ession, subcellula localiza ion, mole- cula s uc u e, unc ional oles, and po en ial disease associa ion o FAME. Ou esul s es ablish FAME as a as -e ol ing gene modula ing i on exchange, exc e ion, ene gy expendi u e, and p ocesses po en- ially associa ed wi h cance p og ession. Resul s FAME sequence e ol es a an ex a-high a e du ing he di e gence o ep iles, bi ds, and mammals To iden i y p e iously iden ified genes wi h uncha ac e ized unc ion ha could ensu e di e ging adap a ions in majo amnio e g oups, we ook ad an age o he compa a i e genomics app oach. E olu iona y p essu e on p o eins a e o en quan ified by he a io o subs i u ion a es a non-synonymous and synonymous si es. To elucida e p o eins co-e ol ing wi h majo e eb a e g oups, we compa ed he a io o he numbe o non-synonymous subs i u ions pe non-synonymous si e (dN) o he numbe o synonymous subs i u ions pe synonymous si e (dS) (dN/dS signa u e) o 27, 16, and 28 pai s o genomes o bi ds, ep iles and mammals, espec i ely (Supplemen a y Da a 1 and 2). We s a ed wi h 20 pai s o mammals and bi ds and added 8 mo e pai s o mammals and 7 mo e pai s o bi ds o inc ease di e si y. We had di - ficul y finding pai s o ep iles because o ewe a ailable genomes and ewe b anches on hei e olu iona y ee, bu included se e al selec ed pai s om di e en clades o ep iles. We iden ified 312 p o eins, which showed significan ly di e en dN/dS a ios be ween ep iles and bi ds (FDR < 0.001) (Supplemen a y Da a 3). Among hem, 129 genes had significan ly highe dN/dS in ep iles, and 183 p o eins had significan ly highe dN/dS in bi ds. In e es ingly, when pe o ming bidi ec ional compa isons o iden i y p o eins wi h he mos flexible sequence in ep iles, only h ee p o- eins (STOX1, CEP126, and CCDC198) had a mean o iden i y o less han 60% (Fig. 1a, b and Supplemen a y Da a 4). Two o hem ha e been p e iously desc ibed. STOX1 is a p o ein in ol ed in ee adical equilib ium and mi ochond ial unc ion21,whe easCEP126isacen- osomal p o ein in ol ed in p ima y cilium o ma ion22.The hi d p o ein, CCDC198/C14o 105 in humans o 1700011H14Rik in mice (he ea e called FAME, Fac o Associa ed wi h Me abolism and Ene gy), howe e , has no been cha ac e ized ye . FAME demons a ed a highe a e age o dN/dS in mammals and bi ds han in ep iles: 0.3912, 0.4235, and 0.2779, espec i ely, which indica es a high e o- lu iona y a e in mammals and bi ds (Fig. 1b and Supplemen- a y Da a 4). Despi e he high a e o e olu iona y changes, his p o ein did no unde go pseudogeniza ion in any o he s udied clades (Supplemen- a y Fig. 1a). Nex , we es ed how his gene’s e olu ion coincided wi h he animals’li es yles. Fo his, we c ea ed he ma ix o li es yles based on he The a-base da abase (h ps://esapubs.o g/a chi e/ecol/E090/ 184/me ada a.h m (“PanTHERIA_1-0_WR93_Aug2008. x ”)) wi h 42 eco ded pa ame e s o species wi h known and published genomes. Then we es ed co ela ions be ween hese pa ame e s and p o ein alignmen s, bu also wi h he selec ion o specific egions in he alignmen s (Fig. 1c, Supplemen a y Fig. 2a). The esul sugges ed ha specific po ions o FAME co-e ol ed wi h me abolic and exc e ion ai s (Fig. 1d, Supplemen a y Fig. 2b). This indica ed ha FAME migh be in ol ed in he con ol o wa e and nu ien exchange. To es his, we c ea ed ou own ma ix based on he animals’habi a s, including dese -li ing species and wa e -dwelling mammals such as whales (Supplemen a y Da a 5). In e es ingly, scanning o he highes co ela ion o in acellula loca ion sugges ed he impo ance o he N- e minal egion wi hin he p o ein. A mo e in-dep h sequence analysis e ealed a possible N-my is oyla ion si e and se e al phospho yla ion si es (Supplemen- a y Fig. 3a–d, Supplemen a y Da a 6). This analysis also showed co - ela ions o he FAME s uc u e wi h he habi a (Supplemen a y Fig. 3e). Fu he mo e, no homologs in any o he explo ed genomes we e e iden , whe eas a domain wi h unknown unc ion DUF4619 was p esen . To es i e olu iona y changes o FAME also occu ed in humans, we nex analyzed he e olu iona y modifica ions in FAME since he spli be ween mode n humans and Neande hals. Al hough some con ac may ha e occu ed23, Neande hals and mode n humans e ol ed independen ly be o e he majo ou -o -A ica dispe sal ~70,000 yea s ago24,25. Du ing ~0.5 million yea s, hese wo lineages accumula ed mu a ions ha eached fixa ion, o nea fixa ion in hei genome, in bo h g oups. To explo e he ecen e olu ion in FAME we in es iga ed all single-nucleo ide polymo phisms (SNPs) o which 108 Nige ian Yo ubans ( ep esen ing he mode n human)26 ca y one allele in a homozygous o m and h ee high-co e age Neande hals27–29 homozygously ca y a di e en allele. In his egion encompassing FAME and 50 kb ups eam and downs eam (ch 14:57886018- 58010575, hg19), which we compa ed wi h he ances al allele, we ound 23 such SNPs (Fig. 1e, , Supplemen a y Fig. 1b, c). Fo 20 o hese SNPs, mode n humans ca y he ances al allele, and o h ee SNPs Neande hals ca y he ances al allele. Tha mo e alleles a e de i ed om he Neande hal lineage is compa ible wi h he lowe e ec i e popula ion size o Neande hals30,and he ac ha weonlyha e h ee high-co e age Neande hals and 108 Yo ubans bias us o de ec alleles ha a e de i ed on he Neande hal lineage. O e all, hese esul s indica e e idence o gene flow om Neande hals a he locus encompassing FAME. Howe e , he e a e wo al e na i e scena ios as o why he Neande hal-like alleles in Fig. 1 could be ound among p esen -day people. These alleles we e p esen in he sha ed ances al popula ion be ween Neande hals and mode n humans, o hese alleles we e in oduced by gene flow om Nean- de hals when hese wo g oups me 31. In he geneflow scena io, we expec o find hese alleles on long Neande hal-like haplo ypes because meio ic ecombina ion has no had ime o b eak hese DNA segmen s down o sho e pieces du ing he ime since he gene flow ook place (~50,000 yea s). To explo e hese wo scena ios, we calcula ed he linkage disequilib ium be ween he 23 SNPs in Fig. 1 . We ound ha 14 o hese alleles a e inhe i ed oge he ( 2 > 0.8) among he indi iduals (n= 2504) in he 1000 Genomes P ojec A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 2 Fig. 1 | FAME sequence changes du ing he e olu iona y di e gence o ep iles, bi ds, and mammals. a Iden ifica ion o FAME in a sc een o e olu iona y di e - ging p o eins co esponding o he spli o majo e eb a e g oups. bCompa ison o dN/dS a ios o unanno a ed p o eins o low p o ein iden i y be ween ep iles. Mean ± SEM and n(genome pai s): STOX1 bi ds 0.5297 ± 0.0120 n= 14 mammals 0.3733 ± 0.0095 n=26 ep iles0.3369±0.0071n=7.CEP126 bi ds 0.5982 ± 0.0126 n= 14 mammals 0.5026 ± 0.0159 n= 23 ep iles 0.4581 ± 0.0116 n=14.FAME bi ds 0.4235 ± 0.0150 n= 22 mammals 0.3912 ± 0.0176 n= 28 ep iles 0.2779 ± 0.0144 n= 14. Desc ip i e s a is ics can be accessed in Supplemen a y Da a 13. cCophe- ne ic co ela ion be ween dend og ams ob ained om amino acid sequence alignmen s and dend og ams based on PanTHERIA sco es o di e en ecological ac o s. The longes p o ein sequences we e ob ained o 68 mammalian species using biomaR (Ensembl). dco ela ion be ween 2-me alignmen egions o eco- logic (PanTHERIA) and phylogene ic (TimeT ee) dend og ams. The species coun wi h bo h ecologic da a and o hologue sequence a ailable is indica ed in he op igh co ne . Regions wi h ends ies a e ma ked wi h a ows. These egions a e likely connec ed wi h ecological ac o s. O he examples a e shown in Supple- men a y Fig. 2. No ably, he e a e di e en shapes o ends and posi ions o he ies. e, Single nucleo ide polymo phisms (SNPs) o which Neande hals (n=3) and Yo ubans (n= 108) homozygously ca y di e en alleles. Genomic coo dina es a e in hg19. The ances al alleles we e aken om Ensembl90 and he Neande hal alleles om p e iously published genomes27,28. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 3 (Supplemen a y Fig. 1b). This haplo ype, agged by s149643449-G, has a leng h o ~87 kb (ch 14:57958614-58046101) and includes he p omo e egion and he fi s wo exons o FAME (Supplemen a y Fig. 1c), as well as he fi s ou exons o he neighbo ing gene SLC35F4. We in es iga ed i a haplo ype his leng h could ha e su i ed since he ime o he common ances o as p e iously desc ibed32, i.e., using he equa ion 1-GammaCDF (m, shape = 2, a e = 1/L), whe e m is he measu ed haplo ype leng h and L he expec ed leng h gi en by he equa ion L = 1/( × ). He e is he ecombina ion a e pe gene a ion pe bp and is he leng h o he human and Neande hal b anches since di e gence. Fu he mo e, we used he local ecombina ion a e (1.47 cM pe Mb)33, and p e iously published es ima es o b anch leng hs and gene a ion ime32. Unde his assump ion, he p obabili y o a haplo ype o his leng h su i ing since he common ances o o mode n humans and Neande hals is low (p=3.7e–06). We hus con- clude ha his haplo ype has been in oduced in he gene pool o p esen -day people by gene flow om Neande hals. In he 1000 Genomes da a se 26, hese haplo ypes a e ound a low equencies in Asia, eaching a maximum allele equency o 1.0% among Han Chinese (n= 208) and in admixed Ame icans, whe e i eaches an allele e- quency o 0.8% in people o Mexican ances y (n=64). FAME is a memb ane-associa ed p o ein en iched in he kidney, panc eas, li e , and allopian ube. Using Geno ype-Tissue A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 4 Exp ession (GTEx) da a, we disco e ed ha he exp ession o FAME was pa icula ly high in he kidney and o a smalle ex en in he panc eas, li e , and allopian ube (Fig. 2a and Supplemen a y Fig. 4a). The analysis o publicly a ailable single-cell ansc ip omics da a o he mouse kidney34,35 u he confi med he specific exp ession in kidney epi helial cell ypes. This includes he loop o Henle and collec ing duc cells, p oximal ubules, and mino exp ession in in e cala ed A cells (Fig. 2b and Supplemen a y Fig. 4b). By u ilizing publicly a ailable mass spec ome y da a, we ound e idence o he p esence o FAME a he p o ein le el in di - e en issues and species. Fo ins ance, FAME p o ein is de ec ed in P o eomicsDB [h ps://www.p o eomicsdb.o g], Phosphomouse [h ps://phosphomouse.hms.ha a d.edu]andPep ideA las[h ps:// db.sys emsbiology.ne /sbeams/cgi/Pep ideA las/Sea ch] public mass spec ome y da abases36–39. S ong expe imen al e idence o FAME p o einp oduc ionexis sinbo h hehuman 40 and mouse kidney37. Fu he mo e, FAME p o ein was de ec ed in cul u ed mu ine collec - ing duc cells41, alida ing he p esence o FAME p o ein in a cell ype shown o p oduce i s mRNA in i o (Fig. 2b). The e o e, we nex ocused on he kidney and alida ed he p e- sence o FAME p o ein in he p oximal ubules by immunohis- ochemis y. This is suppo ed by he ac ha we did no de ec FAME in samples om knockou animals (Fig. 2c and Supplemen a y Fig. 5). Impo an ly, we ensu ed ha ou an ibody is unc ional and specific ia de ec ing FAME as a pa o FAME-EGFP usion in cul u ed cells ha do no p oduce FAME endogenously (Supplemen a y Figs. 5 and 6). Howe e , al hough we alida ed he unc ionali y o he an ibody, we mus also acknowledge i s limi a ions connec ed o po en ial low sensi i i y, which esul s in inabili y o de ec FAME in wes e n blo wi hou o e exp essing FAME, which we discussed in de ail in he me hod sec ion. Wi h he help o molecula cloning and o e exp ession in HEK293T cells, we ound ha FAME localizes o plasma memb anes as well as o small cy oplasmic esicles (Fig. 2d, e, and Supplemen a y Fig. 4c). To u he alida e he p edic ed my is oyla ion si e (Sup- plemen a y Fig. 3c), we ea ed HEK293T cells o e exp essing FAME-EGFP wi h IMP-1088 (an inhibi o o he human N-my is oyl ans e ases NMT1 and NMT2) and DDD85646 (an inhibi o o T. b ucei N-my is oyl ans e ase (TbNMT)), as well as, 2-B omohexadecanoic acid as a nega i e con ol (a non-selec i e inhibi o o lipid me abolism) (Fig. 2d, –i). These esul s show ha once HEK293T cells a e ea ed wi h my is oyla ion-specific inhibi o s, he localiza ion o FAME shi s om he plasma memb ane owa d he nucleus (Fig. 2 –i). DMSO ea men o he non-selec i e inhibi ion o lipid me abolism did no al e he localisa ion o FAME. These esul s suppo he exis ence o a my is oyla ion si e in FAME. In addi ion, mu a ing he amino- e minal glycine esidue (si e o my is oyla ion) o alaninealso esul edin henuclea localisa iono hep o ein(Fig.2e). Li e-cell imaging expe imen s e ealed as a ficking o FAME in he memb anes and exocy osis- ela ed o exosome anspo om ans- ec ed o non- ans ec ed cells, as well as memb ane sha ing (Sup- plemen a y Fig. 4c–e and Supplemen a y Mo ie 1). O e all, hese da a show a high exp ession and esicula na u e o FAME in he mouse kidney, sugges ing a link o cellula /memb ane anspo . Binding pa ne s o FAME sugges a ole in i on me abolism and cell cycle-associa ion. To elucida e he molecula binding pa ne s o FAME, we pe o med a yeas wo-hyb id sc een. Fo his, we used wo di e en lib a ies (mouse kidney emb yo om E18.5 and mouse b ain emb yo mix o E10.5 and E12.5) o co e mo e possible in e ac ion pa ne s (Fig. 3a). Bo h sc eens iden ified Fe i in Hea y Chain 1 (FTH1), Fo min Binding P o ein 1 Like (FNBP1L), as well as Tousled Like Kinase 1 (TLK1) as op hi s (Fig. 3b). F h1 encodes he hea y subuni o e i in, he majo in acellula i on s o age p o ein in p oka yo es and euka yo es.A main unc ion o e i inis he s o age o i on ina soluble non oxic s a e and u he i on up ake in capsule cells o he de el- oping kidney42. FNBP1L, on he o he hand, is equi ed o coo dina e memb ane ubula ion wi h he eo ganiza ion o he ac in cy oskele- on du ing endocy osis43,44. TLK1 has se e al di e se subs a es and is ac i e when phospho yla ed. Ac i a ion by phospho yla ion is cell cycle-dependen wi h i s peak ac i i y in S-phase. O e exp ession o TLK1 p o ein causes se e e g ow h de ec s and cell cycle a es in he G2/M phase wi h apop osis45. Ou yeas - wo-hyb id sc een also de ec ed FAME in e ac ions wi h ansc ip ion ac o s and p o eins wi h his one H4-specific ace yl ans e ase ac i i y, including CREB/ ATF BZIP T ansc ip ion Fac o (CREBZF) and Lysine Ace yl ans e ase 7 (MYST2). Bo h CREBZF and MYST2 a e associa ed wi h cell cycle p og ession and eplica ion. CREBZF a es s he g ow h o os eo- sa coma cells by displacing MDM2 and s abilizing p5346, whe eas MYST2 has c ucial unc ions in ansc ip ion, eplica ion, and DNA epai 47. Howe e , since FAME was no loca ed in he nucleus (Fig. 2d, e), he pu a i e Y2H in e ac ion pa ne s in he nucleus (Fig. 3b) mus be conside ed wi h cau ion. Ne e heless, we confi med he in e ac ion wi h FTH1 using mChe y-labelled FTH1 exp essed in HEK293T cells co- ans ec ed wi h GFP- agged FAME (Fig. 3c). To ob ain a be e pic u e o he in e ac ome o FAME we pe - o med a p oximi y-dependen bio in iden ifica ion (Bio-ID) expe i- men oge he wi h classical immunop ecipi a ion ollowed by mass Fig. 2 | FAME is highly exp essed in kidneys and localises o he cell memb anes and esicles. a Analysis o FAME exp ession in a ious human issues based on Geno ype-Tissue Exp ession da a. FAME exp ession is pa icula ly high in he kid- ney, panc eas, li e , and allopian ube. Mean ± SEM and n(GTEx samples): kidney (8.071 ± 0.1468 n= 32) panc eas (6.575 ± 0.0392 n= 171) li e (6.447 ± 0.06853 n= 119) allopian ube (4.702 ± 1.230 n= 5) bladde (1.414 ± 0.3387 n= 11) ce ix u e i (0.9330 ± 0.5028 n= 11.) bDe ailed exp ession analysis o Fame using Tabula Mu is, a single-cell a las o he mouse. The da a demons a e high exp ession in kidney epi helial cell ypes, he loop o Henle and collec ing duc cells, bu also in he p oximal ubules. Mean ± SEM and n(single mouse cells): loop o Henle (1.546 ± 0.03942 n= 471) collec ing duc (1.403 ± 0.04278 n= 443) p oximal ubule (0.6419 ± 0.02091 n= 1198) in e cala ed A-cells (0.07076 ± 0.03823 n=45). cImmunofluo escence o he adul wild ype and knockou mouse kidney s ained o FAME and imaged oge he wi h au o fluo escence. Rep esen a i e image om 3 di e en animals is shown. See Supplemen a y Fig. 5 o addi ional s ainings. dValida ion o he FAME N-my is oyla ion si e. Visualiza ion o o e exp essed fluo escen ly agged FAME-EGFP in HEK293T cells upon ea men wi h he N-my is oyla ion inhibi o s DDD85646 and IMP-1088 and he palmi oyla ion inhi- bi o 2-b omohexadecanoic acid (2-BP) as con ol. Da a om 4 independen expe imen s is shown. eO e exp ession o he usion p o ein in HEK293T cells shows he localisa ion o FAME in he plasma memb ane and in acellula esicles (whi e a ows). –iQuan ifica ion o localisa ion changes o o e exp essed FAME- GFP upon N-my is oyla ion inhibi ion in cellula compa men s. Violin plo s o he pe cen age o FAME-EGFP localised in he plasma memb ane upon ea men wi h he indica ed inhibi o s. Mean ± SEM and n: DMSO (53.78 ± 9.491 n=4)2-BP (73.71 ± 6.262 n= 4) DDD85646 (35.62 ± 7.446 n= 4) IMP-1088 (0.00 ± 0.00 n=4), p- alue DMSO s IMP-1088 = 0.0013, p- alue 2-BP s DDD85646 = 0.0078, p- alue DDD85646 s IMP-1088 = 0.0030. gViolin plo s o nuclea FAME localiza ion upon ea men . Mean ± SEM and n: DMSO (0.00 ± 0.00 n= 4) 2-BP (1.622 ± 1.029 n=4) DDD85646 (13.10 ± 13.10 n= 4) IMP-1088 (75.36 ± 2.151 n=4),p- alue DMSO s IMP- 1088 = 0.0001, p- alue 2-BP s IMP-1088 = 0.0001. hViolin plo s o cy oplasmic Fame localiza ion upon ea men . Mean ± SEM and n:DMSO(0.00±0.00n=4) 2-BP (4.882 ± 3.806 n= 4) DDD85646 (0.00 ± 0.00 n= 4) IMP-1088 (16.38 ± 1.882 n=4),p- alue DMSO s IMP-1088 = 0.0001, p- alue 2-BP s IMP-1088 = 0.0351, p- alue DDD85646 s IMP-1088 = 0.0001. iViolin plo s o memb anous and cy oplasmic/nuclea Fame localisa ion upon ea men . Mean ± SEM and n: DMSO (48.09 ± 8.612 n= 4) 2-BP (19.05 ± 5.971 n= 4) DDD85646 (51.34 ± 12.16 n=4)IMP- 1088 (8.255 ± 2.894 n=4),p- alue DMSO s IMP-1088 = 0.0046, p- alue DDD85646 s IMP-1088 = 0.0137. Sou ce da a a e p o ided as a Sou ce Da a file. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 5 spec ome y (Fig. 3d and Supplemen a y Da a 7, 8). The FAME-IP esul s a e isualized in Fig. 3d oge he wi h o e lapping hi s om he BioID expe imen . STRING analysis o hese o e lapping hi s e ealed a s ong associa ion o FAME wi h he ca aly ic complex, in acellula p o ein anspo , mi ochond ial inne memb ane, espi a o y elec- on anspo , and p o ein expo (Fig. 3e). These da a a e suppo ed by gene co ela ion da a om publicly a ailable single-cell ansc ip omics da a o a ious issues (Supplemen a y Fig. 7). F om hese posi i e co ela ions, we could show he co-localiza ion o genes specific o he mic o ubule, mi ochond ia, and PCP-pa hway asso- cia ion o he FAME p o ein (Fig. 3 –h). FAME con ols he exc e ion o nu ien s and i on. To unde s and he unc ional ole o FAME in gene al de elopmen , he mo phology A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 6 o o gans, and i s physiology, we gene a ed Fame knockou mice using an FVB/An gene ic backg ound. CRISPR/Cas9 was used o c ea e hese knockou mice by inse ing a double STOP codon downs eam o he ini ia o ATG (see Me hods sec ion). These mice appea ed iable wi hou majo de elopmen al, mo phological, o beha iou al de ec s. We could e ec i ely p opaga e he colony in he homozygous knockou s a e. In e es ingly, body weigh and mean lean body mass we e sig- nifican ly al e ed in FVB/An Fame knockou mice (Fig. 4a(i, ii)). When analysing se um e i in le els, we e ealed a significan dec ease in e i in in knockou animals compa ed o he con ol (Fig. 4a(iii)). Fu he mo e, we ound an unusually high amoun o albumin in he u ine o knockou animals (Fig. 4a(i )). To examine his pheno ype in mo e de ail, we ca e ully in es iga ed he kidneys o con ol and knockou animals. Analysis o he kidney olumes o adul , 12-week old mice and he amoun and size o he fil e ing glome uli did no show any di e ences (Supplemen a y Fig. 8a). His ological analyses o Pe iodic acid–Schi (PAS) sec ions showed no mal his omo phology in bo h wild ype and Fame knockou mice, wi hou any signs o pa hological al e a ions o glome uli, essels, o he ubuloin e s i ium (Supplemen a y Fig. 8b). As con- sequences o kidney mal unc ion, pa icula ly p o einu ia, can be a ec ed by changes only isible a he ul as uc u al le el, we also analysed he kidneys using ansmission elec on mic oscopy. The analysis showed an in ac and no mally de eloped fil a ion ba ie o he glome uli (Supplemen a y Fig. 8b), wi h egula ly shaped podocy e oo p ocesses, egula glome ula basal memb ane, and hin enes a ed endo helium o glome ula capilla ies. Also, he ubula cells showed a no mal ul as uc u al appea ance wi h a p ominen b ush bo de in p oximal ubules and high amoun s o mi ochond ia. No signs o me abolic s ess we e obse ed, includ- ing no signs o in acellula accumula ions o lipids o glycogen, inc eased esicles, o high lysosomal ac i i y. Because adul homozygous knockou animals did no show any s uc u al pheno ype a he le el o he kidney, we hypo hesized ha FAME migh con ey specific adap i e highly unable unc ions and is impo an o he compe i i eness o animals in di e en ecological niches. To add ess he ole o FAME in ad e se en i onmen al condi- ions and o challenge he de eloping sys ems o o gans, we es ed he e ec o a low i on die du ing emb yonic de elopmen o wild ype and knockou mice (FVB/An backg ound). Fo his, we kep emales on a low-i on o con ol die o ou weeks be o e ma ing. New-bo n pups we e hen analysed using mic oCT coupled wi h 3D segmen a ion. All Fame (FVB/An ) knockou mice on he con ol die showed a size educ ion o he kidneys, ad enal glands, in e scapula b own adipose issue, and li e compa ed o he wild ype mice on a con ol die . This pheno ype was compa able wi h con ol pups being on a low-i on die . Fame (FVB/An ) knockou mice on a low-i on die showed no u he changes in o gan size (Fig. 4b– ). These findings sugges ha FAME is impo an o scaling he inne o gans in esponse o ad e se condi- ions, including he ene gy s o agesobse edby he educedin e - scapula b own adipose issue. To dig deepe in o he unc ion o FAME, pa icula ly in he kid- ney, we pe o med single-cell RNA sequencing o wild ype and knockou kidney samples (FVB/An backg ound) (Fig. 5a–candSup- plemen a y Fig. 9). The da a confi med he p esence o wo con- secu i e s op codons in he p o ein-coding egion o Fame mRNA in indi idual cells om he knockou condi ion (Supplemen a y Fig. 10a). As a main esul , he single-cell ansc ip omics analysis confi med he p esence o all cell popula ions in knockou kidneys, wi h only mino composi ional changes (Fig. 5a, b). By analyzing he gene exp ession wi hin clus e s in mo e de ail, we iden ified only a ew significan di - e en ially exp essed genes (Fig. 5c and Supplemen a y Fig. 10b–e), including he p e iously iden ified FAME-binding, i on- anspo ing F h1 and Slc25a39—a mi ochond ial memb ane anspo e in ol ed in biosyn hesis and po en ially i on homeos asis48. This sugges s ha he knockou pheno ype has a molecula na u e wi hou no iceable de ec s a issue o ganiza ion le els. Nex , o in es iga e i such me abolism uning oles o FAME depend on he gene ic backg ound49,50 and addi ional modifica ions o expe imen al condi ions, we c ea ed and es ed a second mouse knockou model using a C57BL/6NC l backg ound. Al hough he exc e ion o albumin in he u ine and se um e i in le els we e no al e ed in hese mice wi h knocked ou Fame, he body weigh and lean body mass we e significan ly changed (Supplemen a y Fig. 11). These wo knockou models on di e en gene ic backg ounds e ealed ha depending on he exac gene ic backg ound, he body weigh and lean body mass show di e en significan al e a ions compa ed o he con ols o he same backg ound. The eason o hese di e ences can be mul i ace ed and migh include he complex and di e gen con ex o di e en ly exp essed in e ac ing molecules. Fo ins ance, he con- ols o di e en backg ounds showed di e ences in he exc e ion o albumin and se um le els a s eady s a e (compa e Fig. 4aandSup- plemen a y Fig. 11). Fu he mo e, he di e ences in fine mo emen and he changed me abolic s a us in bo h animal g oups can cause di e ences in body weigh and lean body mass. In e es ingly, despi e he di e ences in weigh changes in he wo knockou models, he le el o he ene gy balance influencing ho mones gh elin and lep in we e no significan ly changed in ei he mouse model (Supplemen a y Fig. 12). Fu he mo e, analysis o mul iple addi ional blood pa ame e s did no show significan di e ences, excep a dec eased pla ele coun in knockou emales wi h an FVB/An backg ound (Supplemen a y Fig. 13) and lowe eosinophil numbe in emales wi h a C57BL/6NC l backg ound (Supplemen a y Fig. 14). Knockou o Fame influences me abolic pa ame e s and ac i i y o he animals. To elucida e he me abolic pheno ype o he knockou animals, we pe o med me abolic pheno yping. Measu emen s we e pe o med on 75-day old male mice wi h an FVB/An backg ound, ou wild ype and ou Fame knockou mice, a e a h ee-day aining phase in specific me abolic cages. Fo he C57BL/6NC l expe imen al g oup o simila age, we had eigh wild ype and en knockou males and se en wild ype and fi e knockou emales, ha unde wen me abolic pheno yping. Food and wa e in ake, locomo o ac i i y, O 2 Fig. 3 | Binding pa ne s o FAME sugges a ole in i on me abolism. a G aphical ep esen a ion o he ULTIma e Y2H™sc een pe o med by Hyb igenics. Mouse FAME bai was cloned in o he pB27 (N-LexA-AKR2-C usion) ec o , and used o sc eening using mouse kidney emb yo_RP1 and mouse emb yo B ain_RP2 agmen lib a ies as p ey. The in e ac ion o wo p o eins econs i u es an ac i e an- sc ip ion ac o and enables yeas g ow h. bTop sco ing in e ac ion pa ne s o kidney emb yo and emb yo b ain lib a ies a e indica ed. In o al, 118 million in e ac ions we e es ed. cValida ion o FTH1 as op FAME in e ac ion pa ne om he ULTIma e Y2H™sc een. FAME-EGFP was o e exp essed oge he wi h FTH1- mChe y o isualize he hea y subuni o e i in in HEK293T cells. The whi e a ow poin s owa ds encapsula ed FTH1 by FAME. Rep esen a i e image o 3 indepen- den expe imen s. dVisualiza ion o FAME in e ac ion pa ne s iden ified by bo h immunop ecipi a ion (IP)/mass spec ome y and p oximi y-dependen bio in iden ifica ion analysis (BioID). eSTRING analysis om all o e lapping IP and BioID hi s. –hValida ion o op FAME in e ac ion pa ne s om he mass spec ome y analysis. Rep esen a i e image o 3 independen expe imen s. O e exp ession o FAME-EGFP oge he wi h α-Tubulin-mChe y in HEK293 cells o he isualiza ion o mic o ubules. Whi e a ows poin a FAME co-localising wi h α-Tubulin. gO e exp ession o FAME-EGFP oge he wi h MITO-7-mChe y o iden i y mi o- chond ia. Whi e a ows highligh he memb anous localisa ion o FAME in mi o- chond ia. hO e exp ession o FAME-EGFP and VANGL1-myc isualized wi h an an i- VANGL1 an ibody. Whi e a ows show co-localiza ion o bo h p o eins wi hin he plasma memb ane. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 7 consump ion, and CO 2 p oduc ion o he mice we e moni o ed o 48 h. Be o e and a e he measu emen , body composi ion analysis o he mice was ca ied ou by EchoMRI (Fig. 6a). As s a ed abo e, knockou animals wi h FVB/An gene ic backg ounds showed a sig- nifican ly highe a e age body weigh as compa ed o he wild ypes (Fig. 4a). In e es ingly, he e ec was e e sed in he C57BL/6NC l backg ound g oup (Supplemen a y Fig. 11). These di e ences in body weigh sugges ha Fame dele ion al e s ene gy homeos asis in a way depending on di e en gene ic backg ounds. When we in es iga ed ood in ake and no malized i o o al body weigh , only he ood in ake o he knockou mice wi h FVB/An backg ound was significan ly lowe du ing he day ime (Fig. 6b). In addi ion, he ene gy expendi u e o he FVB/An Fame knockou animals no malized o o al body weigh (Fig. 6b), as well as Z-ac i i y, A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 8 a measu e o explo a o y beha iou , o C57BL/6NC l knockou animals di e ed significan ly compa ed o wild ype mice o hei co e- sponding gene ic backg ound (Fig. 6c and Supplemen a y Fig. 15). O no e, we de ec ed a no iceable endency o changes in fine ac i i y in he FVB/An backg ound mice: he knockou animals showed ewe fine mo emen s a nigh , and hei a e age fine mo emen s we e ewe as compa ed o he wild ypes (Fig. 6b). Since bo h mouse models exhibi ed sligh ly di e en me abolic p ofiles, we challenged hem using di e en app oaches o u he alida e he p e iously obse ed pheno ypes. To exclude he e ec o ood on ene gy expendi u e and he po en ial impac o he di e ing day ime ood in ake be ween he geno ypes on a FVB/An backg ound, we epea ed he expe imen upon as ing. Fo his, we emo ed he ood om he cage, and s a ed he measu emen a e an 8-hou -long as ing pe iod wi hou he u he addi ion o ood. Du ing his se up, he same pa ame e s as in he expe imen s be o e we e moni o ed o 24 h. Simila o he expe imen be o e, we obse ed di e ences be ween he wo g oups o FVB/An mice in ac i i y a he beginning o he nigh - ime. Bo h fine and Z-ac i i y di e ed significan ly ea ly du ing he 12-hou da k pe iod (Supplemen a y Fig. 16a–d). In line wi h his, he ene gy expendi u e o he FVB/An knockou mice was sig- nifican ly less be ween 7 and 9 PM as compa ed o con ols o he same backg ound (Supplemen a y Fig. 16e, ). These da a indica e ha he ood-seeking ac i i y o he FVB/An knockou animals was less p o- nounced compa ed o wild ypes. These ime poin s ea ly du ing he da k pe iod o e lap wi h he in ensi e ood-seeking ac i i y o FVB/An wild ype mice, indica ing ha he significan ly highe ene gy expen- di u e o he FVB/An wild ype animals is due o he highe locomo o ac i i y. On he o he hand, i is wo h men ioning ha he ene gy expendi u e o he FVB/An knockou mice s ayed below he ene gy expendi u e o FVB/An wild ype mice du ing he whole measu emen . O e all, FVB/An knockou mice had highe body weigh wi h highe lean body mass. Fu he mo e, FVB/An knockou mice seemed o be less hung y a e 8-hou as ing pe iods esul ing in less in ense ood- seeking beha io . The ene gy expendi u e o he FVB/An knockou mice was significan ly lowe due o hei dec eased locomo o ac i i y. Nex , we challenged C57BL/6NC l mice by exposing hem o a fi s wa m and hen cold en i onmen o s udy e ec s on ene gy expen- di u e o knockou animals. While e ec s on ene gy expendi u e we e small, we did find ha depending on he sex o he animals, di e en measu ed pa ame e s, such as Z-ac i i y, ood in ake and o a low deg ee ene gy expendi u e, we e al e ed di e en ly be ween emales (Supplemen a y Fig. 17) and males (Supplemen a y Fig. 18). This sug- ges s ha FAME has a sex-specific ole. These esul s suppo he e olu iona y di e ging and plas ic ole o FAME in uning he ene gy expendi u e balance in a ious animal g oups and di e en gene ic backg ounds. In humans, he analysis o genome-wide associa ion s udies (GWAS) showed a co ela ion o mu a ions in FAME wi h highe body mass index and diabe es- ela ed pa hologies as well as macula degene a ion(Supplemen a yFig.19).These esul sa einlinewi hou p e ious findings, fi ing ha i on homeos asis and age a e ac o s ela ed o macula degene a ion51, oge he wi h he ole o i on in diabe es and body mass indexes52. To in es iga e his u he , we examined he pheno ypic e ec s o ecen e olu iona y changes in CCDC198 (FAME) on he mode n human lineage. O he h ee mu a ions de i ed om he mode n human lineage ( s143439954, s147526762, s148585073) (Fig. 1 ), only s147526762 has an ances al allele s ill p esen among Eu - opeans (allele equency ~0.1%). Gi en ha he as majo i y o gene ic associa ion s udies ha e been ca ied ou on Eu opeans, we explo ed whe he s147526762 had any pheno ypic e ec s ela ing o he pheno ypes discussed abo e (me abolic synd omes, kidney- ela ed diso de s, and macula degene a ion). We examined possi- ble associa ions using PhenoScanne 53, and al hough no associa ion passed co ec ion o mul iple compa isons, he hi wi h he lowes p- alue was an associa ion wi h high-densi y lipop o ein (p= 0.02, be a = 0.74, in e se no mally ans o med uni s) om he UK Household Longi udinal S udy54. Fo his associa ion, he ances al allele inc eased he HDL le els. We also in es iga ed any associa ion be ween s147526762 and 1,400 b oad pheno ypes among 400,000 B i ons (using UK BioBank da a and PheWeb (PheWeb, n.d.). The associa ion wi h he lowes p- alue was agains hype ensi e ch onic kidney disease, o which he ances al Neande hal-like allele inc eases he isk (p= 0.04, be a = 6.6, log odds uni s). Al hough hese associa ions did no pass co ec ion o mul iple compa isons, we no e ha he en a i e associa ions ma ch he he e sugges ed ole o CCDC198 (FAME). The low allele equency o he ances al allele makes pheno ypic analyses challenging. Fu u e, mo e ex en- si e s udies, pa icula ly among Sou he n Han Chinese people, whe e 3% ca y he ances al allele26, a e needed o co obo a e hese pu a i e associa ions. Co ela ion o FAME wi h cance . Due o he me abolic pheno ypes obse ed in FAME knockou mice, we became in e es ed in he ole o FAME in umo s, whe e ene gy expendi u e and me abolism a e al e ed. Indeed, he exp ession o FAME appea ed s ably main ained in all umo ypes de i ed om heal hy human FAME+ issues and cell ypes acco ding o ou analysis o public TCGA and GTEx da ase s (Fig. 7a, b and Supplemen a y Fig. 20a). The su i al p obabili y o di e en ypes o cance can be s a ified by low o high FAME exp ession (Supplemen a y Fig. 20b). To es his in i o, we o e - exp essed FAME in HEK293T and A549 cells (human adenoca cinoma om he lung), which led o a dec ease in p oli e a ion (Fig. 7c, d). The knockou o FAME in HEK293T cells by CRISPR-Cas9 genome edi ing did no lead o a change in p oli e a ion (da a no shown). This is likely because Fame is no endogenously exp essed in HEK293T cells, as confi med by qPCR (Supplemen a y Fig. 20c) and acco ding o human p o ein a las da a. Con e sely, he knockou o Fig. 4 | Fame is in ol ed in he exc e ion o p o eins and pa icipa es in he scaling o inne o gans unde ad e se condi ions such as nu ien deficiency. a(i)Violin plo s showing he di e ences in weigh be ween male wild ype and Fame knockou (KO) animals on an FVB/An backg ound. Mean ± SEM and n o each g oup: WT (27.26 ± 0.3204 n= 4), KO (25.24 ± 0.1775 n=4),p- alue WT s KO = 0.0015. ii Compa ison o mean lean body mass. Mean ± SEM and n:WT (20.75 ± 0.3303 n= 4) KO (22.50 ± 0.5986 n=4),p- alue WT s KO = 0.0436. (iii) Compa ison o se um e i in le els. Mean ± SEM and n:WT(3817±548.6n=4)KO (2238 ± 295.5 n=7),p- alue WT s KO = 0.0206. (i ) Compa ison o u ine albumin o c ea inine a io. Mean ± SEM and n: WT (0.07496 ± 0.007053 n=11)KO (0.1983 ± 0.04809 n= 11), p- alue Mann-Whi ney es WT s KO < 0.0001. bMic o- CT images o P0 pups wi h con ol and low i on die , con aining 178.58 mg i on/kg o 5.16 mg i on/kg, espec i ely. The kidney ( ed), in e scapula b own adipose issue (IBAT) (yellow), li e (g een) and ad enal glands (o ange) a e segmen ed using 3D Visualiza ion so wa e and supe imposed on o he pups. c– Violin plo s compa ing inne o gan scaling amongs P0 wild ype and knockou pups on di - e en die s. cKidney olume. Mean ± SEM and n: WT con ol (2.261 ± 0.2301, n=8), KO con ol (1.401 ± 0.1354, n=8),p- alue WT con ol s KO con ol = 0.0062, WT low i on (1.120 ± 0.0973, n= 8), KO low i on (1.642 ± 0.2669, n=6).dIBAT olume. Mean ± SEM and n: WT con ol (8.44 ± 1.299, n= 4), KO con ol (4.955± 0.1169, n=4),p- alue WT con ol s KO con ol = 0.037, WT low i on (4.210 ± 0.8292, n= 4), KO low i on (4.897 ± 0.9432, n=3).eLi e olume. Mean ± SEM and n:WT con ol (29.94 ± 2.959, n= 4), KO con ol (19.90 ± 1.988, n=4),p- alue WT con ol s KO con ol = 0.0304, WT low i on (17.72 ± 2.049, n= 4), KO low i on (16.72 ± 2.632, n=3). Ad enal gland olume. Mean ± SEM and n:WTcon ol (0.095 ± 0.005345, n= 8), KO con ol (0.06125 ± 0.002950, n=8),p- alue WT con ol s KO con ol < 0.0001, WT low i on (0.06250 ± 0.00491, n= 8), KO low i on (0.07667 ± 0.004944, n= 6). Sou ce da a a e p o ided as a Sou ce Da a file. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 9 (100 µM), IMP-1088 (100 nM) o DDD85646 (1 µM) 30 min be o e ans ec ion. The ans ec ion was pe o med wi h 250 ng o RIK- pEGFP-N1 using Lipo ec amine LTX® and Plus™Reagen (In i ogen). A e o e nigh incuba ion, he cells we e fixed using 4% PFA o 20 min a oom empe a u e, washed 3 imes wi h PBS and s ained wi h DAPI.Imagingandcoun ingwaspe o medusingaThunde Sys em (Leica). Violin cha s we e gene a ed using P ism So wa e 9.0 (G aphpad). In e ac ion pa ne s Yeas Two-Hyb id Y2H Sc eening. The ULTIma e Y2H sc eening was pe o med by Hyb igenics Se ices (Pa is, F ance; www.hyb igenics- se ices.com) ollowing p e iously desc ibed me hods73,74. The mouse 1700011H14Rik (aa 1-294) bai was PCR-amplified, sequenced, cloned in he pB27 (N-LexA-AKR2-C usion) ec o , and used o sc eening using mouse kidney emb yo_RP1 and mouse emb yo B ain_RP2 ag- men lib a ies as p ey. A o al o 174 and 94 p ey agmen s, espec- i ely,o heposi i eclones,we eamplified by PCR and sequenced a hei 50 and 30 junc ions. The esul ing sequences we e used o iden i y he co esponding in e ac ing p o eins in he GenBank da a- base (NCBI) using a ully au oma ed p ocedu e. Bio-ID. The P oximi y-dependen Bio in Iden ifica ion (BioID) expe i- men s we e pe o med using he PROFACGEN Se ice (www. p o acgen.com). The De ailed epo , as well as RAW da a, a e a ail- able unde [h ps://da ad yad.o g/s ash/sha e/oj XiYX S3yzg5SZwd XHogg geQBDaSgP hRBdU8Yw]. Immunop ecipi a ion (IP) and MS/MS analysis o pep ides. HEK293T cells we e ans ec ed wi h a plasmid encoding un agged mouse 1700011H14Rik/FAME. The day a e ans ec ion, cells we e washed wi h PBS, sc aped in ice-cold PBS, and pelle ed by 200 × g, 4 °C cen i uga ion. Cells we e lysed in 1 ml o cold lysis bu e (0.5 % NP40, 50 mM T is bu e , pH 7.4; 300 mM NaCl; 1 mM EDTA) supplemen ed wi h p o ease inhibi o s (Roche, 11836145001) and 0.1 mM di hio- h ei ol (Sigma, E3876). A e 15 min, he lysa e was clea ed by cen- i uga ion a 16000 × g o 15 min. 1 µg o mouse monoclonal 1700011H14Rik an ibody o no mal mouse IgG as nega i e con ol (Me ck, 12-371) was used pe sample and incuba ed o e nigh a 4 °C on a o a ing wheel. 40 μl o p o ein G-Sepha ose beads slu y (GE Heal hca e, 17-0618-05) was washed in 1 ml o lysis bu e and 200 × g cen i uga ion. Equilib a ed beads we e added o he lysa es wi h an ibodies, and incuba ed a 4 °C on a o a ing wheel o 4 h. The beads we e washed 6 imes in lysis bu e by cen i uga ion. The las wo washes we e done in bu e wi hou de e gen . Following IP washes, bead bound p o ein complexes we e diges ed di ec ly on he beads by adding o 1 µg ypsin (P omega, sequencing g ade) in 50 mM NaHCO 3 bu e . Beads we e mixed and incuba ed a 37 °C wi h mild agi a ion o wo hou s. Beads we e o exed and emo ed, while he eleased p o ein complexes we e u he incuba ed a 37 °C o e nigh (16 h) wi hou agi a ion. The esul ing pep ides we e ex ac ed in LC-MS ials by 2.5% o mic acid (FA) in 50% ace oni ile (ACN) and 100% ACN wi h he addi ion o polye hylene glycol (PEG-20.000; final concen a ion 0.001%) and concen a ed in a SpeedVac concen a o (The moFishe ) o a final olume o 15 µl. Six independen eplica es we e analyzed by mass spec ome y (Supplemen a y Figs. 23 and 24). LC-MS/MS analyses o pep ide mix u es we e done using an Ul i- ma e 3000 RSLCnano sys em connec ed o an O bi ap Eli e hyb id spec ome e (The mo Fishe Scien ific). P io o LC sepa a ion, yp ic diges s we e online concen a ed and desal ed using a apping col- umn (100 μm×30mm)filled wi h 3.5-μm X-B idge BEH 130 C18 so ben (Wa e s). A e washing o he apping column wi h 0.1% FA, he pep ides we e elu ed (flow a e 300 nl/min) om he apping column on o an analy ical column (Acclaim Pepmap100 C18, 3 µm pa icles, 75 μm × 500 mm; The mo Fishe Scien ific) by a 100 min nonlinea g adien p og am (1–56% o mobile phase B; mobile phase A: 0.1% FA in wa e ; mobile phase B: 0.1% FA in 80% ACN). Equilib a ion o he apping column and analy ical column was done p io o sample injec ion o sample loop. The analy ical column ou le was di ec ly connec ed o he Digi al PicoView 550 (New Objec i e) ion sou ce wi h a PicoTip emi e SilicaTip (New Objec i e, FS360-20-15-N-20-C12). ABIRD (Ac i e Backg ound Ion Reduc ion De ice) was ins alled. MS da a we e acqui ed in a da a-dependen s a egy selec ing up o op 10 p ecu so s based on p ecu so abundance in he su ey scan (m/z 350-2000). The esolu ion o he su ey scan was 60 000 (m/z 400) wi h a a ge alue o 1 × 106ions, one mic oscan and maximum injec ion ime o 200 ms. HCD MS/MS spec a we e acqui ed wi h a a ge alue o 50 000 and esolu ion o 15 000 (m/z 400). The max- imum injec ion ime o MS/MS was 500 ms. Dynamic exclusion was enabled o 45 s a e one MS/MS spec a acquisi ion and ea ly expi a ion was disabled. The isola ion window o MS/MS agmen a- ion was se o 2 m/z. Fo da a e alua ion, he MaxQuan so wa e (2.0.1.0)75 wi h inbuild And omeda sea ch engine was used using de aul se ings unless o he wise no ed. Sea ch was done agains UniP o KB p o eome da abase o Homo sapiens (downloaded om [h ps:// p.unip o .o g/ pub/da abases/unip o /cu en _ elease/knowledgebase/ e e ence_p o eomes/Euka yo a/UP000005640/UP000005640_9606. as a.gz], e sion om 2021-06-16, 20,600 p o ein sequences), a sepa a e as a file con aining he mouse C14o 105/FAME Q9CPZ1 (CN105_MOUSE) sequence and he cRAP con aminan s da abase (112 sequences, e sion om 2018-11-22, downloaded om [h p://www. hegpm.o g/c ap]. Modifica ions we e se as ollows o he da abase sea ch: oxida ion (M), deamida ion (N, Q), and ace yla ion (P o ein N- e m) as a iable modifica ions, wi h ca bamidome hyla ion (C) as a fixed modifica ion. Enzyme specifici y was yp ic/P wi h wo pe missible misclea ages. Second pep ides and ma ch be ween uns (MBR) ea u es we e enabled. Only pep ides and p o eins wi h alse disco e y a e h esh- old unde 0.01 we e conside ed. Da a a e publicly a ailable in he PRIDE da abase wi h he iden ifie PXD039259. The p o einG oups. x file, he esul ing ou pu om MaxQuan , was u he p ocessed in R, . 4.1.1. using he Di e en ial En ichmen o P o eomics Da a (DEP) R package76.In hewo kflow, fi s ly con- aminan hi s we e fil e ed ou and p o ein in ensi ies we e log2 ans o med. Only p o eins wi h in ensi y > 0 in mo e han 4/6 samples o a leas one condi ion we e e ained. In ensi ies we e no malized using LoessF no maliza ion, and missing alues we e impu ed using minimal alue. Finally, limma es wi h Benjamini-Hochbe g adjus men o mul iple compa ison was used o es o he di e en ially exp es- sed p o eins. P o eins we e deno ed as up egula ed when passing he h eshold o log2 old change > 1 and adjus ed p- alue < 0.05. Co e- sponding cellula localiza ions o up egula ed p o eins we e isualized using he Human Cell Map esou ce77 (Supplemen a y Fig. 23c). Mouse models The o iginal FVB/An colony was a kind gi om he Uni e si y o An we p, Belgium. The colony was e eshed om he Jackson Labo a o y [h ps://www.jax.o g/s ain/004828] se e al imes. All expe imen s we e pe o med in acco dance wi h he Ins i u ional E hical Codex, Hunga ian Ac o Animal Ca e and Expe imen a ion (2013, 40/2013) and he Eu opean Union guidelines (Di ec i e 2010/ 63/EU), and wi h he app o al o he Ins i u ional Animal Ca e and Use Commi ee o he Ins i u e o Expe imen al Medicine. Mice we e main ained on a 12-hou ligh /da k cycle, and ood and wa e we e p o ided ad libi um behind a SPF ba ie acco ding o FELASA ecommenda ion. All mice we e heal hy wi h no ob ious beha io al pheno ypes. Low i on (C1038) and con ol die (C1000) we e ob ained om Al omin. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 16 All C57Bl/6NC l animals we eb ed and housed a he Czech Cen e o Phenogenomics, Ves ec, which is acc edi ed by he Minis y o Ag icul u e o he Czech Republic. Mice we e housed unde s anda d condi ions (12:12 - ligh :da k cycle) in he indi idually en ila ed cages (Tecniplas g een line) and had ee access o s anda d chow (Al omin 1310) and pu ified chlo ina ed wa e . Labo a o y animal acili y en i- la ion is se o op imal ai quali y (HEPA fi a ion) and quan i y o animal and wo king s a com o and s able empe a u e (in ange 19-21 °C) and humidi y (45–65%). All animal expe imen s we e app o ed by he Animal Ca e and Use Commi ee o he Ins i u e o Molecula Gene ics o he Czech Academy o Sciences, P ague, in acco dance wi h guidelines and p ac ices es ablished by he Di ec i e 2010/63/EU o he Eu opean Pa liamen on he P o ec ion o Animals Used o Scien ific Pu poses. Gene a ion o Fame KO mice.Fo FVB/An ansgenic mice gene a- ion, we co-injec ed Cas9 p o ein (30 ng/µl), he gRNA (GGAACACA GGGCCAGTTGA(GGG)) (50 ng/µl), and ssODN dono (CAGTCTCGTGA ATGAGCTTTCTTCTTCCAGGTTCCGATTCAATGCAAAGAACACAGGG CCTCACTAGGGTGTCTCTGTTTCTTGGCTTTGTAAAGGTG) (15 ng/µl) in o FVB/An e ilized eggs. Injec ed emb yos we e ansplan ed in o he o iduc o B6CBAF1 pseudop egnan emales. The geno yping o he new-bo n mice was ca ied ou by a PCR-based s a egy, whe e we could de ec he co ec modifica ion by using specific p ime s o he modified sequence. Gene a ion o he Ccdc198/Fame knockou model.TheCcdc198/ Fame KO mouse (C57BL/6NC l- Ccdc198 em1(IMPC)Ccpcz) was gene - a ed on a C57BL/6NCRL backg ound (Cha les Ri e Labo a o ies) by dele ion o a c i ical exon, specifically exon 3 o he Ccdc198/Fame gene (ENSMUSG00000021850) by using he CRISPR/Cas9 sys em. The guide RNAs (gRNAs) o highes sco e and specifici y we e designed using CRISPOR. The ollowing guide RNAs we e selec ed o gene a e he exon dele ion: gRNAs, 5′- AAGGACCTGAATCTAGCACT-3′and 5′- CATTTCCAGTACAGACTAGT-3′. The gRNAs we e assembled in o a ibonucleop o ein (RNP) complex wi h Cas9 p o ein (In eg a ed DNA Technologies, 1081058, 1072532), elec opo a ed in o 1-cell zygo es, and ans e ed in o pseudop egnan os e emales (C l:CD1(ICR)). Pu a i e ounde s we e analyzed by PCR and sequencing. A ounde ha bo ing a 661 bp dele ion o e lapping he en i e exon 3 o he Ccdc198/Fame gene was chosen o subsequen b eeding. Geno yping was pe o med by PCR wi h he o wa d p ime 5’- GCTGAACTGT GGAGCAGGTA-3′and e e se p ime 5′-CAATCCACCCCCAATACC CC-3′. Tissue con as ing and X- ay compu ed mic o omog aphy mea- su emen s. To inc ease he con as o so issues o X- ay com- pu ed mic o omog aphy (mic oCT), he samples we e s ained wi h 1% iodine. A e emb yo dissec ion in ice-cold PBS, he samples we e fixed in 4% o maldehyde in PBS solu ion o 24 h a 4 °C wi h slow o a ion. Subsequen ly, samples we e dehyd a ed in inc emen ally inc easing e hanol concen a ions (30%, 50%, 70%), 1 day in each concen a ion. Samples we e ans e ed in o 1% iodine in 90% me hanol o issue con as ing. The iodine-me hanol solu ion was changed a e 3 days. P0 pups we e s ained o 7 days. The con- as ing p ocedu e was ollowed by ehyd a ion o he samples by incuba ion in e hanol se ies (90%, 70%, 50%, and 30%). Then, he ehyd a ed emb yos we e embedded in 1% aga ose gel (Sigma- Ald ich, A5304) and placed in polyp opylene conical ubes o a oid mo ion a i ac s du ing mic oCT scanning. The mic oCT measu e- men s we e conduc ed using he sys em GE phoenix | ome|x L 240 (GE Sensing and Inspec ion Technologies) wi h a 180 kV/15 W max- imum powe nano ocus X- ay ube and fla panel dynamic 41 | 100: 4000 × 4000 px, wi h pixel size 100 × 100 μm. The exposu e ime was 900 ms and 2000 p ojec ions we e aken o e 360°. Th ee p ojec ions in e e y posi ion we e a e aged o educ ion o he noise in he da a. The u ilized accele a ion ol age and cu en we e 60 kV and 200 µA. X- ay spec um was fil e ed by 0.2 mm o alumi- num pla e. The oxel size o he econs uc ed da a was 12 µm o P0 pups, 5.8 µm o he whole kidneys, and 1.3 µm o sec ions o he kidneys. The omog aphic econs uc ions we e pe o med using GE phoenix da os|x 2.0 3D compu ed omog aphy so wa e (GE Sensing and Inspec ion Technologies). Fo 3D isualiza ion, he segmen a- ion was done by an ope a o using a combina ion o so wa e A izo (The moFishe Scien ific) and VG S udio MAX (Volume G aphics). U ine and blood analyses. U ine was collec ed o each mouse a he same ime in he mo ning. To collec u ine, he mouse was held o e a Pe i dish. The u ine was ans e ed in o a collec ion ube and s o ed a 4 °C o di ec ly used o u he analysis. Blood collec ion was pe o med om he ail ein o he mice. Subsequen ly, he se um was isola ed by cen i uga ion o 20 min a 2500 pm. The se um was hen ans e ed in o a clean ube and used o immedia e analysis. The ollowing ki s we e used acco ding o he manu ac u e ’s ins uc ion. Mouse C ea inine Assay Ki (C ys alchem, 80350), Mouse Albumin ELISA Ki (C ys alchem, 80630), Mouse Fe - i in ELISA Ki (C ys alchem, 80636), Sodium Assay Ki (C ys alchem, 80179), Po assium Assay Ki (C ys alchem, 80169). T ansmission elec on mic oscopy. Kidney issue was cu in small pieces and fixed in 3% glu a aldehyde in PBS. Samples we e washed in 0.1 M Soe ensen’s phospha e bu e (Me ck), pos -fixed in 1% OsO 4 (Ro h, Ka ls uhe, Ge many) in 25 mM suc ose bu e (Me ck) and dehyd a ed by ascending e hanol se ies (30, 50, 70, 90 and 100%) o 10 min each. The las s ep was epea ed 3 imes. Dehy- d a ed specimens we e incuba ed in p opylene oxide (Se a) o 30 min, in a mix u e o Epon esin (Se a) and p opylene oxide (1:1) o 1 h, and finally in pu e Epon o 1 h. Samples we e embedded in pu e Epon. Epon polyme iza ion was pe o med a 90 °C o 2 h. Ul a hin sec ions (70–100 nm) we e cu by ul amic o ome (Reiche Ul acu S, Leica) wi h a diamond kni e (Leica) and picked up on Cu/Rh g ids (HR23 Max a o m, Plano). The con as was enhanced by s aining wi h 0.5% u anyl ace a e and 1% lead ci a e (bo h EMS). Samples we e examined using a Zeiss Leo 906 ans- mission elec on mic oscope (Ca l Zeiss) ope a ing a an accele a- ion ol age o 60 kV. Single-cell p epa a ion om adul kidneys. Con ol and mu an li - e ma e mice (male and emale) we e used o single-cell an- sc ip omic and mo phological analysis o he adul kidney. Mice we e anes he ized wi h isoflu ane o e dose and ansca dially pe used wi h PBS. Kidneys we e dissec ed. One kidney pe mouse was fixed in 4% PFA a 4 °C o e nigh o be used o mo phological analysis. The sec- ond kidney was chopped in small pieces and diges ed wi h 2 mg/ml Collagenase P o 10 min a 37 °C, an equal olume o 0.10% ypsin/ EDTA was added, and issue was diges ed o an addi ional 10 min a 37 °C. Following enzyma ic diges ion, issue was i u a ed using a wide-bo e pipe e ip, and clumps we e mechanically dissocia ed using a100µm mesh and a sy inge plunge . Cell suspensions we e washed wice wi h PBS/10% FBS and esuspended a a concen a ion o 1,000 cells/µl o p ocessing wi h a 10x con olle (10x Genomics). Lib a y p epa a ion and sequencing. Lib a y p epa a ion was pe - o med using a 10x con olle (10x Genomics) wi h he Single Cell 3’ 3 chemis y. Sequencing was pe o med using a HiSeq 3000 (Illumina) a he Biomedical Scien ific Facili y (BSF), Vienna, Aus ia. Single-cell RNA sequencing o he fi e samples (KO emale, KO male, KO mix, WT emale, WT mix) esul ed in 103,554,203 eads (86.30% and 67.90% o hem we e confiden ly mapped o he genome A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 17 and he ansc ip ome, espec i ely) o KO emale; 58,088,304 eads (87.30% and 69.10%) o KO male; 73,512,818 (90.30% and 75.00%); 75,266,219 eads (86.40% and 65.40%) o WT emale; and 58,636,386 eads (92.50% and 77.90%) o WT mix. The inse size o each sequencing was 350 bp. To check whe he DE genes be ween wild ype and knockou a e no sex-specific, we compa ed hem wi h he lis o he co esponding DE genes be ween emale and male p oximal ubule (PT) samples om78. The genes whose adjus ed p- alues a e less han 0.01 we e excluded om he compa a i e analysis. Bo h lis s ha e a li le in e - sec ion; mo eo e , in he case o hese ew in e sec ed genes, he mos significan DE genes be ween wild ype and KO we e no e en in he op 40 sex-specific genes. The excep ions a e Inm (male-specific) and Spp1 ( emale-specific) genes. The mos in iguing findings a e ha he FAME-binding i on- anspo ing F h1 and po en ially i on homeos asis associa ed Slc25a39 gene seem o be gende neu al. These Da a a e now ou lined in Supplemen a y Da a 10. Single-cell RNAseq analysis.Rawfiles we e p ocessed, mapped, and coun ed o he Cell Range mm10-2020-A genome and i s co e- sponding anno a ion by Cell Range e sion 4.0.0. The ou pu coun ma ices o eachsamplewe e u he p ocessedwi h heSeu a package pipeline ( .3.2.2.9001)79. Each da ase was fil e ed o emo e genes exp essed in less han h ee cells and o emo e cells ha had ewe han 1000 mRNA coun s pe cell and mo e han 50 % o mi o- chond ial coun s (accoun ing o he high me abolic a e obse ed in he kidney)35. Pu a i e double cells we e p edic ed by Sc uble be o e he fil a ion and log-no maliza ion s eps80.Fo u he compa a i e analysis, fi e da ase s we e in eg a ed in o one join da ase . To imp o e downs eam dimensionali y educ ion and clus e ing, he mi ochond ial gene con en was eg essed ou by using he ScaleDa a unc ion om he Seu a package. The fi s 40 p incipal componen s and 30 nea es neighbo s we e used o g aph-based clus e ing and u he isualiza ion by UMAP. Each clus e was assigned a cell ype based on he ma ke genes epo ed in p e ious s udies35,78,81–83 and Supplemen a y Fig. 9. To es ima e he di e ence in geno ype com- posi ional con en , he exac Fishe es was applied. Iden ifica ion o di e en ially exp essed genes in wild ype and mu an geno ypes in each clus e was done by he FindMa ke s unc ion ha implemen ed Wilcoxon ank-sum es and used a cu o o minimum log FC di e - ence (0.25). Mo e de ails can be ound online h ps://gi hub.com/ ipo e ennaya/RIK_pape . Me abolic cage expe imen s (FVB/An ). To examine he e ec o he absence o 1700011H14Rik/Fame gene on he me abolism o adul mice, me abolic pheno yping was ca ied ou by using he PhenoMas e Sys em (TSE Sys ems). Mice we e placed indi idually in o me abolic aining cages o habi ua e o he new en i onmen o 3 days. A e ha , he body composi ion, including o al body weigh , o al and ee wa e con en , a and lean body mass o he animals we e de e mined by an EchoMRI whole-body magne ic esonance analyze (Zinsse Analy ic). Du ing pheno yping, he wa e and ood in ake (g), he locomo o ac i i y (coun s/hou ), and he calo ime ic pa ame e s, like O 2 consump ion (VO 2 ), CO 2 p o- duc ion (VCO 2 ) o he animals we e con inuously moni o ed o 24 o 48 h. The ene gy expendi u e (kcal/h) was calcula ed acco ding o he Wei equa ion84. The es ing ene gy expendi u e was es i- ma ed om he ene gy expendi u e da a in specific ime poin s when a mouse indi idually mo ed less han 1% o i s maximum ambula o y alue o he las 30 min and a e less han 0.1 g o he las 1 h85. The espi a o y exchange a io is he a io o CO 2 p o- duced and O 2 used (VCO 2 /VO 2 ). A he end o he me abolic mea- su emen , he body composi ion analysis was epea ed. All RAW Da a om hese measu emen s can be ound in Supplemen a y Da a 11. Me abolic cage expe imen s (C57BL/6NC l).APhenoMas e (TSE Sys ems) sys em was used o he indi ec calo ime y. The so wa e used in he PhenoMas e PC was TSE PhenoMas e .7.1.2. Be o e he s a o he indi ec calo ime y measu emen s, a comple e calib a ion p o ocol o he gas analyze s was pe o med acco ding o he man- u ac u e ’s ecommenda ions using no mal ai -comp essed, CO 2 1% and N2 100%. The mice we e indi idually housed in a mul iplex sys em wi h 8 cages plus a e e ence cage. The sampling equency o measu e he CO 2 and O 2 gas measu emen s we e e e y 15 min. All measu e- men s we e ini ia ed in he mo ning be ween 9:00 and 11:00. We p o ided e e y cage ad libi um access o wa e and ood, a s anda d chow die (Al omin, 1314). Wooden chips bedding olume was limi ed o app oxima ely 125 ml du ing indi ec calo ime y mea- su emen s o p ope ly de ec he locomo o ac i i y o he mice by an in a ed beam b eak ame su ounding he cage in he ho izon al plane. The en i onmen al condi ions inside he clima ic chambe we e 55% ela i e humidi y and a ligh cycle o 12 h o ligh (6:00 o 18:00) and 12 h o da kness (18: o 6:00) synch onized wi h he animal acili y whe e he mice we e p e iously housed. Fo he indi ec calo ime y measu emen , he empe a u e was se up as ollows: a 23 °C o 48 h, and a e ha , we kep he mice in he moneu ali y (30 °C) o app oxima ely 24 h, in o al 36 h. Fo indi ec calo ime y using a cold challenge, he empe a u e was se up as ollows: We s a ed he indi ec calo ime y measu e- men using 23 °C o 7 h. A 17:00 he clima e chambe was wa ming up he en i onmen o 30 °C o 7 h. A 00:00 he empe a u e s a ed o dec ease g adually o each 4 °C. A app ox. 8:00, we kep 4 °C o 4 h, and a 12:00 he empe a u e was inc eased back o 23 °C and emained a 23 °C o app ox. 22 h when he calo ime y finished, a e a o al o 48 h. This allowed us o e alua e whe he he cold challenge p oduced a me abolic ca y o e e ec . The indi ec calo ime y eco ded he CO 2 p oduc ion and O 2 consump ion. F om hese alues, he Ene gy Expendi u e (EE) and Respi a o y exchange a io (RER) was calcula ed. Mo eo e , mea- su emen s o locomo o ac i i y and ood and wa e in ake we e eco ded e e y 15 min. When he indi ec calo ime y p o ocol ended, he expe imen was s opped, and he mice we e weighed and placed in o hei o iginal cages. All RAW Da a om hese measu emen s can be ound in Supplemen a y Da a 12. Exp ession analysis— umo da a The Cance Genome A las (TCGA) gene exp ession da a we e down- loaded om he Genomic Da a Commons (GDC) po al in Decembe 2018 [h ps://po al.gdc.cance .go /]. Replica es and samples flagged by he Pan-Cance A las ini ia i e (PanCanA las; gdc.cance .go /abou - da a/publica ions/pancana las) we e emo ed, yielding 9,510 umo s and 713 ma ched no mal issue samples om 31 di e en issue si es. Gene exp ession alues ep esen uppe quan ile no malized HTSeq- acqui ed F agmen s Pe Kilobase pe Million eads mapped (FPKM) p ocessed by he TCGA IlluminaHiSeq_RNASeqV2 pla o m. Pa hologic umo s age da a o kidney umo s we e ob ained om he PanCa- nA las p ojec si e. Gene a ion o FAME KO cell lines CRISPR-Cas9 based gene edi ing was used o c ea e FAME KO lines in A549 cells. Single Guide RNAs (sgRNAs) we e designed using CRISPOR86 and cloned in o len iCRISPR 2 (Addgene #52961) flowing he Zheng lab p o ocol87 (sgRNA: AAGTCCACACGGCCAGCCGA). Plasmids we e alida ed by sequencing. Len i i al pa icles we e p o- duced in HEK293T cells by co- ans ec ion o 2.4 µg o psPAX2 (Addgene #12260), 1.8 µg MD2.G (Addgene #12259) and 3.6 µglen i- CRISPR 2-sgRNA cons uc using polye hyleneimine (PEI, 25 K, Poly- sciences) a a 1:3 DNA:PEI (1 µg/µl) a io. Supe na an s we e ha es ed 48 h pos -in ec ion. Fo knocking ou FAME,1×10 5cells we e seeded A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 18 in o 6-well pla es, i us-pa icle con aining supe na an was added, and p o amine sul a e (Me ck) a 8 µg/mL was added. Cells we e ansduced by spin ec ion a 800 x g o 60 min. Selec ion o in ec ed cells was s a ed 24 h pos - ansduc ion using 2 µg/mL pu omycin (Cayman Chemical). Single-cell clones we e ob ained by limi ing dilu- ion. Genomic DNA was ex ac ed using he Mona ch gDNA Pu ifica- ion Ki (New England Biolabs) acco ding o manu ac u e ins uc ions. FAME KO was alida ed by sequencing o he FAME locus wi h he sequencing p ime s (Mic osyn h): sgRNA KO locus, wd 5’GCCAT- GAAGGAAATGACTGCT 3’, s5’TCAAAACCACAAAGTCTGGTGC 3’. The alida ion o hese knockou s can be ound in Supplemen- a y Fig. 22. P oli e a ion analysis o FAME KO cell lines. Fo p oli e a ion assays, 5×10 4A549 o HEK293T cells we e seeded in o 12-well pla es. The nex day, cells we e ans ec ed wi h 1 µg o pEGFP-N1 as con ol o pEGFP- N1-FAME using Lipo ec amine 3000 (In i ogen) and placed in o an Incucy e S3 Li e-Cell Analysis Ins umen (Sa o ius). Images o ans- mi ed ligh and g een fluo escence we e aken e e y 2 h. A e ~95 h, he image acquisi ion was s opped. G een a ea confluence was ana- lyzed using he Incucy e Base Analysis So wa e(Sa o ius) and P ism 9 (G aphpad). KO alida ion. RNA isola ion was pe o med using he Nucleospin RNA II ki (Mache ey-Nagel). The concen a ion o RNA was measu ed using a NanoD op ND-1000 sys em (The mo Fishe Scien ific). RNA quali y and in eg i y we e assessed by he TapeS a ion 2200 sys em (Agilen Technologies). Fo qRT–PCR exp ession analysis, he RNA was e e se ansc ibed using he Ve so cDNA Syn hesis Ki (The mo Fishe Scien ific) acco ding o he manu ac u e ’s ins uc ions. Quan i a i e exp ession analysis was pe o med using he Quan S udio 5 Real-Time PCR ins umen and p edesigned TaqMan gene exp ession assays (The mo Fishe Scien ific): C14ORF105 (Taqman p obe Hs00216847_m1). GAPDH exp ession was used as an in e nal con ol. Rela i e quan ifica ion was pe o med acco ding o he ΔΔCT me hod88. S a is ics and ep oducibili y S a is ical analysis was pe o med using app op ia e so wa e (G aphpad P ism 9). Desc ip i e s a is ics a e displayed as mean ± s anda d e o o he mean (SEM). All da a ollowing Gaussian dis- ibu ion we e analyzed by unpai ed - es s ( wo- ailed). p- alues smalle han 0.05 we e conside ed s a is ically significan (*p<0.05, **p< 0.01, ***p< 0.001). All p- alues wi hou addi ional indica ion s em om - es s. Nonpa ame ic da a we e anlyzed by Mann-Whi ney es and specifically indica ed in he figu e legends. Repo ing summa y Fu he in o ma ion on esea ch design is a ailable in he Na u e Po olio Repo ing Summa y linked o his a icle. Da a a ailabili y Two knockou mice s ains we e gene a ed o his manusc ip . Mice on aC57BL/6NC l backg ound will be a ailable h ough he In e na ional Mouse Pheno yping Conso ium [h ps://www.mousepheno ype.o g/ da a/sea ch? e m=CCDC198& ype=gene]whe easFVB/An a e depos- i ed o he Jackson Labo a o y (JAX S ock No. 038293). All he in o - ma ion necessa y o ep oduce he single cell analysis is p esen ed in he Me hods Sec ion. All o he ele an da a suppo ing he key findings o his s udy a e a ailable wi hin he a icle and i s Supplemen a y In o ma ion files o om he co esponding au ho upon eques . SNP da a o ances al alleles (108 Nige ian Yo ubans, YRI) was aken om he 1000 Genomes P ojec deposi ed in Ensembl [h ps://www. ensembl.o g/in o/genome/ a ia ion/species/popula ions.h ml]. The Neande hal genomes we e published p e iously27:[h p:// p.e a.mpg. de/neande al/Chagy skaya/VCF/]28, Eu opean Nucleo ide A chi e (ENA) PRJEB21157 [h ps://bioin .e a.mpg.de/jb owse]29,2014ENA ERP002097 [h p://cdna.e a.mpg.de/neande al/al ai/]. The RAW mass spec ome y can be accessed h ough PRIDE wi h he iden ifie PXD039259. RAW Single cell sequencing files can be downloaded om GEO wi h he accession numbe GSE206860. RAW Yea -Two-Hyb id da a can be accessed h ough [h ps://da ad yad.o g/s ash/sha e/ oj XiYX S3yzg5SZwd XHogg geQBDaSgP hRBdU8Yw]. P o ein and CDS sequences om di e en amnio a o ganisms we e ob ained om NCBI da abase (Supplemen a y Fig. 3). The Cance Genome A las P o- g am (TCGA, Decembe 2018, [h ps://po al.gdc.cance .go /]and Geno ype-Tissue Exp ession (GTEx) p ojec da ase s ( elease V6p) [h ps://www.g expo al.o g/home/]we eused ocollec da a o Fig. 7a, b and Supplemen a y Fig. 4a. Tabula mu is [h ps:// abula-mu is. ds.czbiohub.o g/] was used o ob ain da a o Fig. 2aandSupplemen- a y Fig. 4b and downloaded om [h ps://figsha e.com/p ojec s/ Tabula_Mu is_T ansc ip omic_cha ac e iza ion_o _20_o gans_and_ issu es_ om_Mus_musculus_a _single_cell_ esolu ion/27733]. GWAS89 was used o gene a e da a om Supplemen a y Fig. 19. Sou ce da a a e p o ided wi h his pape . Code a ailabili y All cus om-made sc ip s used in he analysis a e a ailable a : [h ps:// gi hub.com/ipo e ennaya/RIK_pape ]. Re e ences 1. Jones, F. C. e al. The genomic basis o adap i e e olu ion in h eespine s icklebacks. Na u e 484,55–61 (2012). 2. Bus aman e, C. D. e al. Na u al selec ion on p o ein-coding genes in he human genome. Na u e 437, 1153–1157 (2005). 3. Wang, T. e al. Iden ifica ion and cha ac e iza ion o essen ial genes in he human genome. Science 350,1096–1101 (2015). 4. Lieben, L. E olu ion: Redefining gene essen iali y. Na . Re . Gene 17, 66 (2016). 5. Manuylo , N. L., Manuylo a, E., A doshina, V. & Te osian, S. Se - din1/L c10 is dispensable o mouse de elopmen . Genesis 46, 441–446 (2008). 6. B ody, M. J. & Lee, Y. The Role o Leucine-Rich Repea Con aining P o ein 10 (LRRC10) in Dila ed Ca diomyopa hy. F on Physiol. 7, 337 (2016). 7. S asse , B. e al. E olu iona y o igin and di e sifica ion o epi- de mal ba ie p o eins in amnio es. Mol. Biol. E ol. 31, 3194–3205 (2014). 8. Kawasaki,K.,La on ,A.G.&Si e,J.Y.Thee olu iono milkcasein genes om oo h genes be o e he o igin o mammals. Mol. Biol. E ol. 28,2053–2061 (2011). 9. Me edi h,R.W.,Ga esy,J.&Sp inge ,M.S.Molecula decayo enamel ma ix p o ein genes in u les and o he eden ulous amnio es. BMC E ol. Biol. 13,20(2013). 10. Elleg en, H. Compa a i e genomics and he s udy o e olu ion by na u al selec ion. Mol. Ecol. 17,4586–4596 (2008). 11. Zhang, G. e al. Compa a i e genomics e eals insigh s in o a ian genome e olu ion and adap a ion. Science 346,1311–1320 (2014). 12. Ja is, E. D. e al. Whole-genome analyses esol e ea ly b anches in he eeo li eo mode nbi ds.Science 346,1320–1331 (2014). 13. Seebache , F. The e olu ion o me abolic egula ion in animals. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 224,195–203 (2018). 14. Hed ick, M. S. & Hillman, S. S. Wha d o e he e olu ion o endo- he my? J. Exp. Biol. 219,300–301 (2016). 15. Robe s, R. M., G een, J. A. & Schulz, L. C. The e olu ion o he placen a. Rep oduc ion 152,R179–R189 (2016). 16. Vize, P. D. & Smi h, H. W. A Home ic iew o kidney e olu ion: A ep in o H.W. Smi h’s classic essay wi h a new in oduc ion. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 19 E olu ion o he kidney. 1943. Ana . Rec. A Disco. Mol. Cell E ol. Biol. 277,344–354 (2004). 17. Poulson, T. L., McNabb, F. M. & Folk, R. L. U ic acid: he main ni ogenous exc e o y p oduc o bi ds. Science 170,98–99 (1970). 18. Galpe in, M. Y. & Koonin, E. V. F om comple e genome sequence o ‘comple e’unde s anding? T ends Bio echnol. 28,398–406 (2010). 19. Galpe in, M. Y. & Koonin, E. V. ‘Conse ed hypo he ical’p o eins: p io i iza ion o a ge s o expe imen al s udy. Nucleic Acids Res 32,5452–5463 (2004). 20. Pawlowski, K. Uncha ac e ized/hypo he ical p o eins in biomedical ‘omics’expe imen s: is no el y being swep unde he ca pe ? B ie . Func . Genom. P o eomic 7,283–290 (2008). 21. Do ido , L. e al. Ni oso- edox balance and mi ochond ial home- os asis a e egula ed by STOX1, a p e-eclampsia-associa ed gene. An ioxid. Redox Signal 21,819–834 (2014). 22. Bona i a, R. e al. Cep126 is equi ed o pe icen iola sa elli e localisa ion o he cen osome and o p ima y cilium o ma ion. Biol. Cell 106,254–267 (2014). 23. Chen, L., Wol , A. B., Fu, W., Li, L. & Akey, J. M. Iden i ying and In e p e ing Appa en Neande hal Ances y in A ican Indi iduals. Cell 180,677–687.e616 (2020). 24. Pos h, C. e al. Pleis ocene Mi ochond ial Genomes Sugges a Single Majo Dispe sal o Non-A icans and a La e Glacial Popula ion Tu no e in Eu ope. Cu . Biol. 26,827–833 (2016). 25. Ri o, T. e al. A dispe sal o Homo sapiens om sou he n o eas e n A ica immedia ely p eceded he ou -o -A ica mig a ion. Sci. Rep. 9,4728(2019). 26. Genomes P ojec , C. e al. A global e e ence o human gene ic a ia ion. Na u e 526,68–74 (2015). 27. Ma essoni, F. e al. A high-co e age Neande al genome om Chagy skaya Ca e. P oc. Na l Acad. Sci. USA 117, 15132–15136 (2020). 28. P u e , K. e al. A high-co e age Neande al genome om Vindija Ca e in C oa ia. Science 358,655–658 (2017). 29. P u e , K. e al. The comple e genome sequence o a Neande hal om he Al ai Moun ains. Na u e 505,43–49 (2014). 30. Ma essoni, F. & P u e , K. Be e suppo o a small e ec i e popula ion size o Neande als and a long sha ed his o y o Nean- de als and Deniso ans. P oc.Na lAcad.Sci.USA114, E10256–E10257 (2017). 31. G een, R. E. e al. A d a sequence o he Neande al genome. Science 328,710–722 (2010). 32. Hue a-Sanchez, E. e al. Al i ude adap a ion in Tibe ans caused by in og ession o Deniso an-like DNA. Na u e 512,194–197 (2014). 33. Kong, A. e al. Fine-scale ecombina ion a e di e ences be ween sexes, popula ions and indi iduals. Na u e 467,1099–1103 (2010). 34. Tabula Mu is, C. e al. Single-cell ansc ip omics o 20 mouse o gans c ea es a Tabula Mu is. Na u e 562,367–372 (2018). 35. Pa k, J. e al. Single-cell ansc ip omics o he mouse kidney e eals po en ial cellula a ge s o kidney disease. Science 360, 758–763 (2018). 36. Deu sch,E.W.ThePep ideA lasP ojec .Me hods Mol. Biol. 604, 285–296 (2010). 37. Hu lin, E. L. e al. A issue-specific a las o mouse p o ein phos- pho yla ion and exp ession. Cell 143, 1174–1189 (2010). 38. Schmid , T. e al. P o eomicsDB. Nucleic Acids Res 46, D1271–D1281 (2018). 39. Wang, D. e al. A deep p o eome and ansc ip ome abundance a las o 29 heal hy human issues. Mol. Sys . Biol. 15, e8503 (2019). 40. Li, S. e al. Digging Mo e Missing P o eins Using an En ichmen App oach wi h P o eoMine . J. P o eome Res 16, 4330–4339 (2017). 41. Rinschen, M. M. e al. Quan i a i e phosphop o eomic analysis e eals asop essin V2- ecep o -dependen signaling pa hways in enal collec ing duc cells. P oc. Na l Acad. Sci. USA 107, 3882–3887 (2010). 42. MacKenzie, E. L., Iwasaki, K. & Tsuji, Y. In acellula i on anspo and s o age: om molecula mechanisms o heal h implica ions. An ioxid. Redox Signal 10, 997–1030 (2008). 43. Ho, H. Y. e al. Toca-1 media es Cdc42-dependen ac in nuclea ion by ac i a ing he N-WASP-WIP complex. Cell 118,203–216 (2004). 44. Kakimo o, T., Ka oh, H. & Negishi, M. Regula ion o neu onal mo - phology by Toca-1, an F-BAR/EFC p o ein ha induces plasma memb ane in agina ion. J. Biol. Chem. 281, 29042–29053 (2006). 45. Lee, J., Kim, M. S., Pa k, S. H. & Jang, Y. K. Tousled-like kinase 1 is a nega i e egula o o co e ansc ip ion ac o s in mu ine emb yo- nic s em cells. Sci. Rep. 8, 334 (2018). 46. Zhang, R., Thamm, D. H. & Mis a, V. The e ec o Zhang ei/CREBZF on cell g ow h, di e en ia ion, apop osis, mig a ion, and he un olded p o ein esponse in se e al canine os eosa coma cell lines. BMC Ve . Res 11,22(2015). 47. Pa do, M. e al. Mys 2/Ka 7 his one ace yl ans e ase in e ac ion p o eomics e eals umou -supp esso Niam as a no el binding pa ne in emb yonic s em cells. Sci. Rep. 7, 8157 (2017). 48. Nilsson, R. e al. Disco e y o genes essen ial o heme biosyn hesis h ough la ge-scale gene exp ession analysis. Cell Me ab. 10, 119–130 (2009). 49. Tschop, M. H. e al. A guide o analysis o mouse ene gy me abo- lism. Na . Me hods 9,57–63 (2011). 50. Wes , D. B., Booze , C. N., Moody, D. L. & A kinson, R. L. Die a y obesi y in nine inb ed mouse s ains. Am. J. Physiol. 262, R1025–R1032 (1992). 51. Dunaie , J. L. I on induced oxida i e damage as a po en ial ac o in age- ela ed macula degene a ion: he Cogan Lec u e. In es Oph halmol. Vis. Sci. 47, 4660–4664 (2006). 52. Sypes, E. E. e al. Highe Body Mass Index Is Associa ed wi h I on Deficiency in Child en 1 o 3 Yea s o Age. J. Pedia . 207,198–204 e191 (2019). 53. Kama , M. A. e al. PhenoScanne V2: an expanded ool o sea ching human geno ype-pheno ype associa ions. Bioin o ma ics 35,4851–4853 (2019). 54. P ins, B. P. e al. Genome-wide analysis o heal h- ela ed bioma ke s in he UK Household Longi udinal S udy e eals no el associa ions. Sci. Rep. 7, 11008 (2017). 55. Gopal, S. K. e al. YBX1/YB-1 induces pa ial EMT and umou - igenici y h ough sec e ion o angiogenic ac o s in o he ex a- cellula mic oen i onmen . Onco a ge 6,13718–13730 (2015). 56. Tang, J. e al. Knockdown o TPT1-AS1 inhibi s cell p oli e a ion, cell cycle G1/S ansi ion, and epi helial-mesenchymal ansi ion in gas ic cance . Bosn. J. Basic Med. Sci. 21,39–46 (2021). 57. Ye, Z. e al. ODC1 p omo es p oli e a ion and mobili y ia he AKT/ GSK3be a/be a-ca enin pa hway and modula ion o acido ic mic oen i onmen in human hepa ocellula ca cinoma. Onco Ta - ge s The . 12,4081–4092 (2019). 58. Meng, Q. e al. Ab oga ion o glu a hione pe oxidase-1 d i es EMT and chemo esis ance in panc ea ic cance by ac i a ing ROS- media ed Ak /GSK3be a/Snail signaling. Oncogene 37, 5843–5857 (2018). 59. Xu, C. e al. SPP1, analyzed by bioin o ma ics me hods, p omo es he me as asis in colo ec al cance by ac i a ing EMT pa hway. Biomed. Pha maco he . 91, 1167–1177 (2017). 60. Solda o , R. e al. Spa io empo al s uc u e o cell a e decisions in mu ine neu al c es . Science 364, eaas9536 (2019). 61. Guan, C., Ye, C., Yang, X. & Gao, J. A e iew o cu en la ge-scale mouse knockou e o s. Genesis 48,73–85 (2010). 62. Wenge ,M.J.,DellaValle,D.M.,Mu ay-Kolb,L.E.&Haas,J.D. E ec o i on deficiency on simul aneous measu es o beha io , b ain ac i i y, and ene gy expendi u e in he pe o mance o a cogni i e ask. Nu . Neu osci. 22,196–206 (2019). 63. Blankenhaus, B. e al. Fe i in egula es o ganismal ene gy balance and he mogenesis. Mol. Me ab. 24,64–79 (2019). A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 20 64. Bogdan, A. R., Miyazawa, M., Hashimo o, K. & Tsuji, Y. Regula o s o I on Homeos asis: New Playe s in Me abolism, Cell Dea h, and Disease. T ends Biochem Sci. 41,274–286 (2016). 65. Wasse man, D. H., O’Dohe y, R. M. & Zinke , B. A. Role o he endoc ine panc eas in con ol o uel me abolism by he li e du - ing exe cise. In J. Obes. Rela . Me ab. Diso d. 19,S22–S30 (1995). 66. Bha acha ya, D., Azambuja, A. P. & Simoes-Cos a, M. Me abolic Rep og amming P omo es Neu al C es Mig a ion ia Yap/Tead Signaling. De . Cell 53,199–211.e196 (2020). 67. Manz,D.H.,Blanche e,N.L.,Paul,B.T.,To i,F.M.&To i,S.V.I on and cance : ecen insigh s. Ann. N. Y Acad. Sci. 1368, 149–161 (2016). 68. Yuan, M. e al. N-my is oyla ion: om cell biology o ansla ional medicine. Ac a Pha m. Sin. 41,1005–1015 (2020). 69. Lin, C. Y. e al. Memb ane p o ein- egula ed ne wo ks ac oss human cance s. Na . Commun. 10, 3131 (2019). 70. Edga ,R.C.MUSCLE:mul iplesequencealignmen wi hhigh accu acy and high h oughpu . Nucleic Acids Res 32, 1792–1797 (2004). 71. Ca i he s, L. J. e al. A No el App oach o High-Quali y Pos mo em Tissue P ocu emen : The GTEx P ojec . Biop ese Biobank 13, 311–319 (2015). 72. Bu le , A., Ho man, P., Smibe , P., Papalexi, E. & Sa ija, R. In e- g a ing single-cell ansc ip omic da a ac oss di e en condi ions, echnologies, and species. Na . Bio echnol. 36,411–420 (2018). 73. Fo ms eche , E. e al. P o ein in e ac ion mapping: a D osophila case s udy. Genome Res. 15,376–384 (2005). 74. Rain, J. C. e al. The p o ein-p o ein in e ac ion map o Helicobac e pylo i. Na u e 409,211–215 (2001). 75. Cox, J. & Mann, M. MaxQuan enables high pep ide iden ifica ion a es, indi idualized p.p.b.- ange mass accu acies and p o eome- wide p o ein quan ifica ion. Na . Bio echnol. 26,1367–1372 (2008). 76. Zhang, X. e al. P o eome-wide iden ifica ion o ubiqui in in e ac- ions using UbIA-MS. Na . P o oc. 13,530–550 (2018). 77. Go, C. D. e al. A p oximi y-dependen bio inyla ion map o a human cell. Na u e 595,120–124 (2021). 78. Ransick, A. e al. Single-Cell P ofiling Re eals Sex, Lineage, and Regional Di e si y in he Mouse Kidney. De . Cell 51,399–413 e397 (2019). 79. S ua , T. e al. Comp ehensi e In eg a ion o Single-Cell Da a. Cell 177, 1888–1902.e1821 (2019). 80. Wolock, S. L., Lopez, R. & Klein, A. M. Sc uble : Compu a ional Iden ifica ion o Cell Double s in Single-Cell T ansc ip omic Da a. Cell Sys . 8,281–291.e289 (2019). 81. Wilson, P. C. e al. The single-cell ansc ip omic landscape o ea ly human diabe ic neph opa hy. P oc. Na l Acad. Sci. USA 116, 19619–19625 (2019). 82. Chung, J. J. e al. Single-Cell T ansc ip ome P ofiling o he Kidney Glome ulus Iden ifies Key Cell Types and Reac ions o Inju y. J. Am. Soc. Neph ol. 31,2341–2354 (2020). 83. Sch oede , A. W. e al. NOVEL HUMAN KIDNEY CELL SUBSETS IDENTIFIED BY MUX-SEQ. T ansplan a ion 104,S85 (2020). 84. Wei , J. B. New me hods o calcula ing me abolic a e wi h special e e ence o p o ein me abolism. J. Physiol. 109,1–9 (1949). 85. E en, P. C., Mokh a ian, A. & Pele, A. P ac ical aspec s o indi ec calo ime yinlabo a o yanimals.Neu osci. Biobeha Re . 18, 435–447 (1994). 86. Conco de , J. P. & Haeussle , M. CRISPOR: in ui i e guide selec ion o CRISPR/Cas9 genome edi ing expe imen s and sc eens. Nucleic Acids Res 46,W242–W245 (2018). 87. Shalem, O. e al. Genome-scale CRISPR-Cas9 knockou sc eening in human cells. Science 343,84–87 (2014). 88. Li ak,K.J.&Schmi gen,T.D.Analysiso ela i egeneexp ession da a using eal- ime quan i a i e PCR and he 2(-Del a Del a C(T)) Me hod. Me hods 25,402 –408 (2001). 89. Beck, T., Rowlands, T., Sho e , T. & B ookes, A. J. GWAS Cen al: an expanding esou ce o finding and isualising geno ype and phe- no ype da a om genome-wide associa ion s udies. Nucleic Acids Res 51,D986–D993 (2023). 90. Howe, K. L. e al. Ensembl 2021. Nucleic Acids Res 49, D884–D891 (2021). Acknowledgemen s We would like o hank Vendula No osado a (Ins i u e o Molecula Gene ics o he Czech Academy o Sciences) and Roldan Medina De Guia (Ins i u e o Molecula Gene ics o he Czech Academy o Sciences) o eedback on he s a is ics as well as clinical chemis y. M.T., M.K., T.Z., and J.K. acknowledge CzechNanoLab Resea ch In as uc u e suppo edbyMEYSCR(LM2018110).M.T.acknowledges heB noCi y Municipali y as a B no Ph.D. Talen Schola ship Holde and Ma ina Roeselo a Memo ial Fellowship. R.S. was suppo ed by he Czech Academy o Sciences RVO 68378050 and by LM2018126, Czech Cen e o Phenogenomics, p o ided by he Minis y o Educa ion, You h and Spo s o he Czech Republic. CIISB, Ins uc -CZ Cen e o Ins uc -ERIC EU conso ium, unded by MEYS CR in as uc u e p ojec LM2023042 and Eu opean Regional De elopmen Fund-P ojec „UP CIISB“(No. CZ.02.1.01/0.0/0.0/18_046/0015974), is g a e ully acknowledged o he financial suppo o he measu emen s a he CEITEC P o eomics Co e Facili y. Compu a ional esou ces o he IP LC-MS/MS da a p o- cessing we e p o ided by he e-INFRA CZ p ojec (ID:90140), suppo ed by he Minis y o Educa ion, You h and Spo s o he Czech Republic. I.A. was suppo ed by Be il Halls en Resea ch Founda ion, Medical Uni e si y o Vienna and Gus a sson’s Founda ion P ize 2021, T.B. was suppo ed by a g an om he Aus ian Science Fund M2688-B28, J.K. was suppo ed by he G an Agency o Masa yk Uni e si y (MUNI/H/ 1615/2018). I.P. was suppo ed by he Eu opean Union’s Ho izon 2020 Resea ch and Inno a ion P og am unde Ma ie Sklodowska-Cu ie (g an ag eemen No. 860635, ITN NEUc es ), P.B. was suppo ed by he Ge man Resea ch Founda ion (DFG, P ojec IDs 322900939, 454024652, 432698239 & 445703531), Eu opean Resea ch Council (ERC Consolida o G an No 101001791), and he Fede al Minis y o Educa ion and Resea ch (BMBF, STOP-FSGS-01GM2202C). The wo k o J.K. was suppo ed by Czech Science Founda ion (22-02794 S). Pa s o Figs. 1,3,6and Supplemen a y Figs 6, 11 –18 we e c ea ed wi h BioR- ende .com. OG and RD we e suppo ed by Minis y o Science and Highe Educa ion o he Russian Fede a ion g an 075-15-2021-1344 and Japan Socie y o he P omo ion o Science (JSPS) KAKENHI JP23H02226. Au ho con ibu ions J.Pe., L.E., A.A., I.P., R.M., T.B., M.T., R.D., A.S.S., E.E.A., D.P.R., H.Z., M.K., M.E.K., J.K ., T.R., K.G., S.K., D.P., Z.Z., R.S.G., A.G., M.E.B., M.iK., H.A., and D.L. acqui ed all biological da a and pe o med he ele an analysis. R.K. and C.K. p o ided human samples and ga e eedback on expe i- men al aspec s. J.Pe., I.A., T.Z., F.E., Z.M., G.S., T.K., V.B., T.H., K.F., J.Ka., P.B., C.F., J.R., P.K., J.P.R., R.S. and O.G. ga e eedback on expe imen al aspec s, supe ised expe imen al app oaches, and implemen ed he da a in e p e a ion. J.Pe., L.E. and I.A. made all figu es con aining da a and esul ing analysis. J.Pe. and I.A. designed he s udy, o ganized he expe imen al wo k, and w o e he manusc ip . All au ho s p o ided eedback on figu es, manusc ip composi ion, and s uc u e. Funding Open access unding p o ided by Ka olinska Ins i u e. A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 21 Compe ing in e es s The au ho s decla e no compe ing in e es s. Addi ional in o ma ion Supplemen a y in o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.o g/10.1038/s41467-023-38663-7. Co espondence and eques s o ma e ials should be add essed o Julian Pe e sen o Igo Adameyko. Pee e iew in o ma ion Na u e Communica ions hanks E ich Ja is and he o he , anonymous, e iewe (s) o hei con ibu ion o he pee e iew o his wo k. A pee e iew file is a ailable. Rep in s and pe missions in o ma ion is a ailable a h p://www.na u e.com/ ep in s Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju - isdic ional claims in published maps and ins i u ional a filia ions. Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons license, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons license and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his license, isi h p://c ea i ecommons.o g/ licenses/by/4.0/. © The Au ho (s) 2023, co ec ed publica ion 2023 Julian Pe e sen 1,25 ,LukasEnglmaie 2,3,25 ,A emV.A emo 4 , I ina Po e ennaya 4 ,RubaMahmoud 1 , Thibaul Boude lique 4 , Ma ke a Tesa o a 5 , Ruslan De ia iia o 6,7 ,Ane Szil ásy-Szabó 8 ,E genyE.Akku a o 9,10 , Da id Pajuelo Regue a 11 , Hugo Zebe g 12,13 , Ma ke a Kaucka 14 , Ma ia Eleni Kas i i 4,13 ,JanK i anek 15 , Tomasz Radaszkiewicz 16 , K is ína Gömö yo á 16 , Sa ah Knau h 1 , Da id Po esil 17 , Zbynek Zd ahal 17 ,RanjaniS iGanji 17 , Anna G abowski 4 , Mi iam E. Buhl 18 , Tomas Zikmund 5 , Michaela Ka ko a 5,15 ,HåkanAxelson 19 ,Da idLindg en 19 , Ra ael K amann 20 , Ch is oph Kuppe 20 , Fe enc E délyi 21 , Zol án Má é 21 , Gábo Szabó 21 , Till Koehne 1 , Tibo Ha kany 22 , Kaj F ied 13 , Joze Kaise 5 , Pe e Boo 18 ,CsabaFeke e 8 , Jan Rozman 11,23 ,Pe Kaspa ek 11 , Jan P ochazka 11 , Radisla Sedlacek 11 ,Vi ezsla B yja 16 ,OlegGuse 6,24 & Igo Adameyko 4,13 1 Depa men o O hodon ics, Uni e si y Leipzig Medical Cen e , Leipzig, Ge many. 2 CeMM Resea ch Cen e o Molecula Medicine o he Aus ian Academy o Sciences, 1090 Vienna, Aus ia. 3 Ludwig Bol zmann Ins i u e o Ra e and Undiagnosed Diseases, 1090 Vienna, Aus ia. 4 Depa men o Neu oimmunology, Cen e o B ain Resea ch, Medical Uni e si y Vienna, Vienna, Aus ia. 5 Cen al Eu opean Ins i u e o Technology, B no Uni e si y o Technology, B no, Czech Republic. 6 Regula o y Genomics Resea ch Cen e , Ins i u e o Fundamen al Medicine and Biology, Kazan Fede al Uni e si y, Kazan, Russia. 7 Endoc inology Resea ch Cen e , Moscow, Russia. 8 Labo a o yo In eg a i e Neu oendoc inology, Ins i u e o Expe imen al Medicine, 1083 Budapes ,Hunga y. 9 Depa men o Applied Physics, Royal Ins i u e o Technology, Science o Li e Labo a o y, 171 65, S ockholm, Sweden. 10 Uni e si y o Ox o d, MRC Wea he all Ins i u e o Molecula Medicine, Radcli e Depa men o Medicine, Ox o d OX3 9DS, UK. 11 Ins i u e o Molecula Gene ics o he Czech Academy o Science, Czech Cen e o Phenogenomics, Ves ec, Czech Republic. 12 Depa men o Neu oscience, Ka olinska Ins i u e , S ockholm, Sweden. 13 Depa men o Physiology and Pha macology, Ka olinska Ins i u e , S ockholm, Sweden. 14 Max Planck Ins i u e o E olu iona y Biology, Plön 24306, Ge many. 15 Depa men o His ology and Emb yology, Facul y o Medicine, Masa yk Uni e si y, B no, Czech Republic. 16 Ins i u e o Expe imen al Biology, Facul y o Science, Masa yk Uni e si y, B no, Czech Republic. 17 Cen al Eu opean Ins i u e o Technology, Masa yk Uni e si y, B no, Czech Republic. 18 Ins i u e o Pa hology & Elec on Mic oscopy Facili y, RWTH Aachen Uni e si y Hospi al, Aachen, Ge many. 19 T ansla ional Cance Resea ch, Depa men o Labo a o y Medicine, Lund Uni e si y, Medicon Village, Scheele ägen 2, Lund, Sweden. 20 Ins i u e o Expe imen al Medicine and Sys ems Biology, RWTH Aachen Uni e si y, Aachen, Ge many. 21 Medical Gene Technology Uni , Ins i u e o Expe imen al Medicine, Budapes , Hunga y. 22 Depa men o Molecula Neu osciences, Cen e o B ain Resea ch, Medical Uni e si y Vienna, Vienna, Aus ia. 23 Luxembou g Cen e o Sys ems Biomedicine, Uni e si y o Luxembou g, 6, a enue du Swing, 4367 Bel aux, Luxembou g. 24 In ac able Disease Resea ch Cen e , G adua e School o Medicine, Jun endo Uni e si y, Tokyo, Japan. 25 These au ho s con ibu ed equally: Julian Pe e sen, Lukas Englmaie . e-mail: julian.pe e [email protected];igo [email protected] A icle h ps://doi.o g/10.1038/s41467-023-38663-7 Na u e Communica ions | (2023) 14:3092 22