Variant-specific pathophysiological mechanisms ofAFF3differently influence transcriptome profiles

mesomelic dysplasi a, ho r seshoe kidne y , in t el lectu a l d isability , t r ans c r ipt ome, e x ome . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Abstr act Backgr ound W e pr ev iou s ly de s cribed t he K I NSS HIP sy nd rome, an autosomal dominant disorder as s o ciated with int ellectual d is ability (ID ), mes om eli c dysplasia a n d h orses hoe kid ney, cau s ed by de nov o variants in the degron o f AFF3. Mou s e kno ck-in s and overexpression in zebrafish prov ided evidenc e f or a dom i n an t-negative (D N) mode-o f -action, wherein an i nc r eas ed level of A FF 3 resulted in pathologi cal effects. Met hods Evolutionary c onstr aints s u gge st t hat ot her mo d e-of -inh e ritance could be a t play. We c h a llenged this hypothesi s b y s creening ID c oh or ts for i n div i du a ls with pr edic t ed- to -b e delete r i o us variant s i n AFF3 . W e u s ed both animal and c ellular mode ls to asse ss the delet erio usne ss of the ident ifie d variants.

W e identified an i n div idua l w it h a KI NS S H IP-like phenot ype ca r ryin g a de novo p a r tial duplicati o n of AF F 3 furt her str engthening t he hypot hes is that an increa sed level of A FF3 i s pathological. W e als o d e t ected sev ent een individuals displayi n g a m i lder s ynd rome with ei t her het e rozygou s LoF or biallel i c missen s e variants in AFF3. C on s istent with s emi- dominance, we dis covered t hree patien t s with homozyg o u s L o F and o ne compoun d het e r ozy got e fo r a LoF and a miss en se var i ant , who presented mor e s evere ph enot ypes than th ei r he tero zygou s par ents. M atching zebr afis h knoc k d own s exhibit neu rological defects that could be r escued b y expres sing human AF F3 m R N A , confirming t hei r as so ciation with t he ablation o f aff3. Conve r s ely, s o m e o f the hu man A FF3 mRNAs carr yi n g mis s en se v ar ia n ts identi f i ed in af fected ind ividuals did n ot c om plemen t . Ov er ex p ression o f mut ated AFF3 mRNAs in zeb rafish embr y o s p roduced a significant increa se of abnorm al lar v a e compar ed to wild -type overexpr es sion fur ther d em ons t ra ting deleteriousness. To fu rther ass es s the ef fect of AFF3 variation , we pr o fil ed th e tr ans cript ome of fibr ob las t s from affected ind ividuals a n d engi n ee red i s o ge n ic cells harb oring +/ +, D N /D N , LoF/ +, LoF/ LoF or D N /L o F AFF3 genotype s. The expre s si o n of mor e than a third of th e AFF3 bound loci is modified i n eithe r the DN/ D N or t he LoF/ LoF lines. W h ile th e same pat hwa ys are affected, o nly about o ne- thir d o f the di ffer en tial ly expr es sed gene s a re c o mmon t o these hom ozy gote datasets, indicating tha t . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint AFF3 LoF and DN variant s lar gely modu l at e tr a n s cript omes differentl y, e. g. the D N A repair pathway dis p l a yed oppo s ite mo dulati on. Conclu si o ns Our r esult s and the high pleiotr opy sho w n b y v aria tion at this l o c us s u g g e s t th at minut e chang es in AFF3 function ar e d e le t erious. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint

AFF3 enc odes t he ALF T r ans cription Elong a tion F act or 3 (MIM *601464), a member o f a gen e f a mily with f our par alogs (AFF1-4 ) in ma mmals. These nucl ear pr ot eins f un ction as tr ans cription a l activa t or s, pr o moting R N A elong a tion 1-3 . They shar e con ser v ed N -t ermin al ( N H D) and C- t erm i n a l homology dom a in s (C HD) 4 , an AF4- L A F 4- FMR2 (A L F) domain 2,3,5 , which c on t ains the degr on moti f , a Serine- r ich tr ans activ a tion domain ( T A D) 6 , and a nuclear/nuc leolar loc alization se q uence (NL S) (Figur e 1A). A FF pr ot eins ar e in t egr a l c omponen ts o f tr ans cr i p tional supe r elong at ion c omple x es (SE C s) t hat include pos it i v e tr anscription elong a ti on f ac t or (P-TEFb) 2,3 . S E C s a r e m a d e o f a n A F F f amily member as s c af f old, YEA TS d omain-c ont aining M L L T pr o t eins ( m y el oid/ lymphoid or m i x ed - lineag e leuk em i a; tr ansloc at ed t o) , a nd an ELL (Elong a tion F act or f or RN A P olymer ase II) pr ot ein 2 . B y ph os pho ryla ting t he C- t erm inal d omain of R N A po lymer ase II, these c omp l e x es r egula t e th e RNA tr ans cript ion elong a tion pr oces s 3,7 . D i s t i n c t c o m b i n a t i o n s o f c o m p o n e n t s y i e l d d i f f e r e n t S E C s pr oviding g ene t ar g et specific it y 2,3 . A FF3 r egulat es t he e xpr es s ion of g enes in v olved i n mesoder m and ect oderm devel op ment, as w ell a s mes en ch ymal cel l pr ol if era tion, cell adhes ion , an g iog enesi s , c a r tilag e and lens develop men t and imm unoglobulin class swit ch r ecom bina tion 8,9 . It w as r ec en t ly link ed with t he es t ablishmen t of bi ologic al r h ythms, e.g. somit og enesis pr ogr ession and nic he swit ching 10,11 . The SE C- L 3 com ple x , whi c h inc orpor at es AFF3, is enr ic hed a t im pri n ted l oci, f or e x ample a t th e lncRN A XI S T locu s tha t initiat e X chr omosome i n a ctiv a tion 12,13 . AFF3 binds bo th silen t and ac t ive chr oma tin r egions t o modu lat e e xpre s sion of imp r i n t ed r egions. F or e x ample, within the D lk1-Dio 3 in t er v a l, it is r ecruit ed by ZFP 281 t o th e Meg3 enhancer r eg ion t o maintain an ac t ive c hr omati n st a t e th r o ugh H3K27ac modific a tion a nd an allele -spec ific e xpr es s ion 8,14 . W e pr ev iou s ly r epor t ed th e a sso c ia t i o n of AFF3 alt er a ti o ns with K IN S SHIP s yndr ome 11 . T w enty-on e af f ect ed indiv iduals allow ed del inea t ion o f i t s car d i n a l c har ac t erist ic s. Su ch individuals pr esent e d with dev elopme n t a l dela y/ in t ell ectual dis ability ( DD / ID ), br a in a tr oph y , e pilept i c enc eph a lop at h y , f ailur e t o thr i v e, hor seshoe kidney , a s p e cific mes om el ic dy s pla sia , fibular h ypoplasia, scoliosi s , h ypert richos i s, dy smor phic f a cial f eat ur es, g as tr oin t est i n a l and pulm on a r y s ympt om s . Thi s aut osomal dominant dis ease i s as so cia t ed w it h de novo g erml ine missen se v arian ts and delet ions , as w ell as mosaic v a r ian ts, in t he c o nser v ed degr on moti f o f AFF3 11,15 ( Figur e 1A ) . Bot h m ous e knoc k - ins and ov er expr es sion in zeb r afis h su g g est ed a do minan t -neg a tive (DN) mode of action , wher ein an incr ease d lev el o f AFF3 r e sult ed in t he p a thol ogic al e ff ec t s . . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Acc or ding t o popu l a tion met r ics pr e sen t ed in G no mAD v2.1.1 16 , AFF3 is under c ons tr ain t with a pLI=1 and a pLOEU F=0.22 1, whic h su g g e st s t ha t AFF3 haploinsuf ficiency c ould also be de let erious . Consi st en t with this h ypot hes i s, mo s aic C GG t rinucleotide-r epea t e xpansi ons in t he pr omoter o f AFF3 th a t r esult in i ts h yperm eth yla tion and s ilencing , w er e as so ci a t ed with mi ld ID , speech and mot or dela y s, s ei zur es , beha vior al di stu rbance s , g ener alized h y p ot onia, dy smorp hic f ea tur es an d c ong enit al a n omal ies 17,18 . Her e we des cribe nov el AFF3 g enetic alt er a t ions as s o cia t ed with an over e x p ress ion disea s e mechanis m , a s w ell as t he ef f ect of decr ease d AFF3 function thr ough haploinsuf ficiency , homo zy gou s tr unc a tion and aut osomal r ecessiv e i n heri t ance. The af f e ct ed individuals pr esen t s ympt oms tha t part ial l y ov erlap t hos e of K IN SSHI P . Mat erials and Methods Samples and variant s iden ti f ic at i on Inf or med c ons en t f or ms we r e ob t ain ed f or a ll a f f ec t e d ind ivi d uals or thei r guar dians part ic ipa ting in this stu dy . The curr en t st udy w as a ppr ov ed by the CER ( “ C ommiss ion d'é thiqu e d e l a r ec her c h e ”) of the c ant on of V aud ( Pr ot o col n u mber : CER- VD 2021-01400) . Thi s r es ear ch c omplies with the principles of t he Declar a ti o n o f H els inki. The ID s of aff ect ed i n div idu a ls a r e unknown to anyone outside the research group . Aff ect e d individuals unde r wen t g enet i c cou ns elling and clinic a l e xam i n at ion f oll o wed by e x ome s equ e n c ing a s de scribed 19,20 and/ or arra y c omp a r a tiv e g eno m e h ybridiza tion, made e x ception of proband B1 who w as s equenced with the Illumina T ruSigh t O ne Expanded panel cov ering about o ne thir d o f the e x ome. G enom e sequenci n g w a s p erf or med o n the D U P1 individual t o char act er i ze the br ea kpo i n t s . Prot ei n model Alignment of mult iple AFF3 or thologous sequence s w as p erf or med with the Clust al Ome g a t ool 21,22 . 3D m odell ing f or A FF 3 (UniPr ot: P5182 6) and SIAH 1 (UniPr ot :Q8IU Q4) in t er action w as built usi n g t he S w is s - P db View er 23 as pr e vi ously de scribed 11 . Z ebrafish hus ba ndry Ze b ra f i s h ( Danio rer io, Or egon AB ) w er e maint ai n ed a t 28.5 ° C o n a 1 4:10 h lig h t / dar k cy cl e. Z ebr afish ar e s t ag ed b y hour s ( h) or da y s (d) pos t f ert ili za tion (p f ). A du lt zebr afish w er e ho us ed in Activ e Blue r ack s (T ecniplast , Bugu ggiat e, It a ly) with a ma xi mum of 20 fis h per t ank. A l l p roc e d ure s . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint c omplied wi t h the Eur opean Con v e n tion fo r th e Pr ot ection of A nima l s f o r Experimen t a l and Sc ien t ific Purpose s (ET S number 123) and t he Na tional Ins titut es of Heal t h guide f or the c are and use of Labor a t o ry animals . Housing and e xperimen t s wer e appr ov ed b y the V aud c an t on a l author ity (autho r iza t i o n VD-H21). Z ebrafish CRISPR-Cas 9 model W e g ener at ed f ounder F0 mut an t zeb r af i sh deplet ed f or aff 3 by CRIS PR/Ca s 9 g enome ed i t ing. T wo s ingle s y n thetic guide RN A s ( sgR NA s) t ar g eting th e c oding sequence in aff3 e x on 6 of bo th isof or ms annot at e d by Ensembl (Z ebr a f is h G R Cz11) ( sgRNA_r 2 5 ’ - T CC AAAG C AG T A C CCA G CCA AGG -3’; sgRNA_r 19 5 ’ - G C A C C T G A G A AT ATATA C C T T G G -3’) w er e designed wit h the CH O P C HO P t ool 24,25 a nd or der ed fr o m S yn thego , R edw ood City , CA, U SA. A t o t al of 1n l of a c ockt ail co n t ain ing 50 ng /μl of gRNA_r2, 5 0 ng /μl o f gR NA_r19 and 200 ng / μl of T ru e C ut TM Cas 9 v2 ( In vi t r og en) w as inject ed in t o one t o t wo -c ell s t ag e e mbry os. In moc k-injec t ed lar v ae, the Cas9 w a s r eplaced b y t he s ame v olum e of w a t e r . K C l (200 mM ) w a s added t o inc r eas e e f f ic ienc y . T o det erm i n e t he C RISP R-Cas 9 t ar g et i n g e fficiency of ea ch sgR NA in 5dp f f ound er (F0) mut an t s , a mi s m at c h de t ec t ion a s sa y u sing T7 endonucleas e 1 ( Ne w En gland Biolabs, Ip swich , M A, Unit ed S t a tes) w as perf or med. B r iefly , DNA w as e xtr ac t ed, and PC R amplifie d w it h pri m er s flanking the s gRNAs t ar get si t e (5’- TCC A AA G CAG T A CCC AG CC A A GG T A T A T A T TCT C AG GT GC - 3’). P CR pr oducts w er e d ena tur ed, r eannealed, and incuba t ed with T7 f or 15 minut es a t 37° C. The r eaction w a s s t opp e d by add i n g 1 .5 μl of 0.25 M ED T A . The pr oduct s w er e t hen separ a ted on 2% a g a r o s e gel t o det e rmi n e r earr ang ements a t the t ar g et ed s it e. Locomot io n as say s A t 72 hp f , the esc ape r e sponse t est w a s p erf or med t o ev al u at e the s wimm i n g ability of the fis h upon a sligh t t ouch st imul a tion. The motion o f ev e ry lar v a w as e xamined and scor ed as « norm a l swimm i n g », « pause », « l o oping sw imming », « pinwheel swimming » o r « mo tio nless » due t o malf orm ations. A t 5 dp f , we analyzed spon t aneous z ebr afis h m otili ty usi n g t he Zeb r abo x ® r ec o r ding s y s t em (Viewpoin t, L issie u , Fr an ce) equipped with infr ared i llumi n ation f or im a ging in t he dar k . Locom otion w as r ec or ded f or each la r v a on a 96-w ell plat e f o r 30 minu t es ( 15-minu t e adapt atio n phase in th e lig h t f ollow ed by a 15-minut e phase in th e dark) and pr esen t ed as slow (3- 6 mm/ s ) and high v elocities (>6 mm / s) 26 . The v el o c it y of th e fis h w as t r a ck ed wit h the V iew p oin t softw ar e and e xperim en t s w e r e pe r f orm ed a t m inimu m th r ee t i mes . Th e r es u l t ing dat a w e r e p oole d . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint t og ether f or st a tistic al analy sis. F i sher ’ s e x act t est or one- wa y A N OV A a naly s i s w er e perf ormed bas ed o n the da t a with Prism10. Immunof l uo res cen ce PT U ( 1-phen yl 2-thiour ea - 75μM) tr e a tm ent w as used on 24hp f zebr afi sh t o pr ev e n t p i gm ent ation . A t appr opri a t e dev e lopment al s t ag es , em bry os wer e dechorionat ed and e uthanized with 0.0168 % tr i c aine ( S igma- A ldrich) a n d immedi at ely fix ed in 4% PF A f or 1 h a t R T . P er meabiliza t ion of larvae w as a c h i ev ed with 1X p hos ph at e saline b u ff er (PBS), 0.5% T r it on X-100, f or 90 min a t R T an d s u b s equen t ly in 1X PBS, 1%T rit on X- 100, f or 2h a t R T on a s low s h ak er . Embr yos w er e th e n incuba t ed i n bl ocking buf f er (1% BSA i n 1X PBS) fo r 1h a t R T and incuba te d in prima ry an t ib odies , mouse an ti-s y n ap t ot agmin 2 (Znp-1 , dilut ed 1:100 in blo ckin g solution – DSH B , Iow a Cit y , IA, U n i t ed S t a tes) or mous e an t i -is let 1 a nd 2 (39.4D5, d i lut e d 1:100 in b lockin g s olut i o n – DS HB , Io w a City , IA, U nit ed St a t es), ov ernigh t at 4°C on a slow s ha k er . Af t er 3 w as hes in 1X P BS, the emb ry os w er e incuba t ed with a sec ondary an tibody , Ale x a Fluor™ 48 8 c onju g a t ed (dilut ed 1:500 in bloc ki n g s o l u tion, In vitr og en), ov ern i gh t at 4 °C. Nu c lei w er e st ai n ed wit h D API (dilut ed 1: 8000, Sigma- Aldr i ch) f or 5/10 min a t R T . A f ter w ashing in PBS, zebr afis h l arv a e w er e mo un t ed o nt o m icr osc opi c s lides with Mowiol 4-8 8 (Sigma-Aldr i c h) . Ima ging w a s perf or med usin g LS M880 air ys c an c o nf oc al micr oscop e (Car l Z eiss). E v alua tion of m otor neur on s ’ str uctur e and hind br ain spinal c or d pr ojecting neur ons’ d evelopm en t w as per f ormed 27 . Mor pholo gic al an alyse s Imag es of 5dp f zebr af ish w er e acq uir ed with a L e ic a mic r os c ope (M 1 65 F C) and Leic a C M OS c a m er a ( IC9 0E, Leic a Camer a A G, W etz-lar , German y) f or mor phologic a l ins p ec t ion. In t er-ocula r dist anc e and head widt h w er e qu a n t ified using th e Fij i s of tw ar e 27 . Fis her ’ s ex a ct t est or o ne-w a y ANOV A analy si s w er e per f orm e d w it h Prism10. St ain ing of cartil agin ous s tructur e A t 5dp f , em b ry os wer e eu thanized with 0.0168% tr icaine (Sigma-Aldrich) a nd fix ed ov ern igh t in 4% PF A a t R T . Fix ed embr y os w er e w a s hed f o ur times with 1X PB S and 0.1 % T w een-20 ( PB ST) and bleac h ed with 30% h y dr ogen per o xid e f or 2h a t R T . A f t er th r ee wash cy c les with P BST , specimen s w er e t r ans f er r ed int o an A lc ian Blue solut i o n (1% co n c en t r a t ed h y dr ochloric acid, 70% ethanol, 0.1% A lc ian blue) and st ained ov ernigh t a t 4 °C. Embry o s w er e rinsed a f ew times with a cidic . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint ethanol (5% c oncen tr a t ed h y dr ochloric ac id, 70 % et hanol, HCl-E t O H ) and incuba t ed in acidic ethanol f or 20m in at R T on a slow sh a k er . Specimens wer e then r e - h yd r a ted as f ollow s : (i) 5/ 10min a t R T i n 1 m L o f 7 5 % H C l - E t O H / 2 5 % H 2 O d ; (ii) 5/ 10 min a t R T in 1m L of 50% HCl-E t O H / 50 % H 2 O d ; (iii) 5/ 10 min at R T in 1mL o f 25% HCl- E t O H / 75% H 2 O d and (iv) 5/ 10m i n a t R T i n 1 mL of 1 00% H 2 O d . Speci m ens w e r e s t or ed in 1mL of 5 0% G ly ce r ol and 50% (1%) K O H or k ept in 100% Gly cer ol f or e xt ended s t or ag e. St aine d emb r y os w er e p os itioned in 50% G ly cer ol and 50% (1%) K OH solution i n a P et ri dis h. The h ea d w a s phot ogr aphed i n a ve n t r al–dor sal and a l at er al view u s ing a s t er e o micr osco pe (Mot i c SM Z-171 ) wi t h the Mo tic Im age Plus s o ftw ar e (v er sion 3 .0). O ver expr es si o n a nal ysis in z ebrafish T ag g ed hum an AFF3 wi ld- type mRNA (GenBank: NM_002 285. 3) w as c loned int o pEZ-M13 v ect or 11 . The v arian t s of i n t e r es t, i . e . the tw o KINSSH IP v arian ts V al 2 35 G l y and Ala233Thr and the thr e e newly-iden tif ied mis sense v arian ts G ln179Glu, L y s5 28Ar g and Thr594 Ser , w er e enginee r ed usi n g the QuikChang e II XL Sit e- Di r ect ed M ut ag enesis Kit f o l lowing t he ma nuf actur er ’ s ins tr uc t ions (Agilent T ec h nologies ). P ositiv e clones w er e c onfirmed by Sanger seque ncing. AFF 3 m R N A w a s tr an scribed fr om th e l inea r ized v ect or pE Z-M13+AFF3-FLA G W t 11 or co n t a i ning eac h of the s tudied v a r i a n t s u s i n g t h e m M E S S A G E m M A C H I N E T 7 t r a n s c r i p t i o n k i t ( A m b i o n ) a n d p u r i f i e d u s i n g R N e a s y Mini Kit (Qiagen) fo l lowing the man uf actu r er s’ ins tr uction s. The injection mix c ons is t ed of th e mRNAs a t th r ee dif f er en t c oncen tr a ti on s (18 0 n g , 360 n g and 720 ng) d i lut ed in RNAse-fr ee w a t er . 1nl of each diluted mRNA w as injected inside t he y olk, below t he cell, in A B wild- t y pe zebr afis h embry os a t the on e t o tw o-cell s t ag e. D istille d wat er w as inject ed as v e hicle c on tr ol in a simi la r v olume. Dep end i n g on R N A am ounts, e x per ime n ts w er e r ep eat ed t wi ce or thr ee ti m es . Phenotypic R es cue in zebr afis h W e en g ineer ed F0 zebr afish deplet ed f o r af f 3 by CRISPR /Ca s 9 g enome editing and e x pr es sing the human AFF3 mRNA o f i n ter es t. The r escue e xperime n t w as co nduct ed by evalua ti n g th e s p on t aneous zebr afi s h mot ility in t he dark u sing the Z ebr abo x ® r ec or ding s y st em (Viewpoin t, Li s sie u , Fr an ce). The inje ction mix con s is t ed of sgR NAs/Ca s9 c omple x and hum an AFF3 m R N A W t (f or the pheno typic r es cue) or th e h uman AFF3 m RN A c a r rying each v ari an t o f i n t e r est . Di f f er e nt c onc en t r a tions of hum an AFF3 mRNA W t (25 n g , 5 0 n g , 7 5 ng, 10 0 n g , 15 0 ng , 200 n g) w er e t es t e d t o r ea ch t he phenotypic r escue. Phen otypic r es c u e by t he v a r ia n ts was ass es s ed by injecting 1nl o f mix c on t aining sgR N A s/Ca s 9 com ple x + 75n g (75 pg /μl) of each mR N A in t o on e or tw o-cell st age . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint embry os. The s gR N A s w er e inject ed a lone in m oc k-inject ed l arv a e wit h Cas9 r eplac ed b y t he same v olume o f w at e r . The Gr aph P ad Pri sm s of tw ar e (v er si o n 10 .0) wa s u s e d t o per f orm s t atis tica l analysi s o f the da t a. A one-w a y A N OV A t est was adopt ed t o det e rmi ne diff e r ences betw een e xperime n t al gr ou ps . Experimen t s w er e per f or m ed s ev en tim es. HEK293 T i sogenic cell line s HEK293 T c ells w er e used t o engineer AF F 3 knock-ins (K INSSHI P) and knoc k outs (LoF) cell lines by CRISPR/ Ca s 9 g enome editing. F our guides w er e used t o cr eat e th e LoF l ines: one t ar g et i n g th e c oding sequen ce of e x on 6, desi gn e d with t he Ther mo Fisher Scien ti fic t ool, and th r ee t ar g etin g e x on 5 w it h th e Gene K nock out Kit v 2 of Syn t hego . T o engineer th e KIN S SH IP lines, on e s gR NA t ar g et i n g the c oding sequence in e x on 6 desig n ed u sing t he Ther mo F is her Sci en t ific t ool, w as c ombined with a D NA dono r t empl a t e t o knock-in the A la233Thr v arian t . The sg R N A s, D N A dono r t empl a t e and c or r esponding sequen c ing prim er pai r s w er e or der ed a t I n vitr og e n , S yn thego , or Sigma-Aldr i ch. The c oc k t ail t o indu ce AFF3 k n o ck-ou t co n t ained 7.2μg of th e f our c ombined sgRNA s and 36.2 μ g o f T r ueCut TM Cas 9 v2 pr ot e in ( In vi t r og en). In the KINS S H IP model, 7.2μ g of th e sgRNA by Thermo Fi sher Sc ien t i f ic , comb i n ed wi th 36.2 μg of T r ueCut TM Cas9 v2 pr ot ein and 14.5 μg d s D N A dono r , w ere u s ed. E ach mix w as tr ans f ect ed using t he Lipof e ct amine™ CR ISPRMAX ™ Cas9 T r an sf ection R eag en t Kit (In vi tr o gen ) on 10 c m HEK293T c ell pla t e s acc or ding t o th e manuf actur er's pr ot ocol. 48h aft er tr ansf ec t i o n, cell s w er e collect ed, r esus pend ed pos t-c oun ting , and diluted a t a dens it y of 8 cell s/ml. 100 μl of t his r es uspen sion w as tr an sf er r ed t o ea ch w el l of a 96/i1 well pla t e. A t the desir ed cell conf luenc y , clones w er e s cr eened with t he Q IA pr ep& CRISPR kit ( Q IA st ock , Q IAG E N , A G). V arian t s wer e c onfirm ed by Sang er s equ enc i ng. Het er o zy got es w er e fur ther c onfirmed b y cloning and s equencing of both alleles . W e engi neer ed fi v e biall e lic L o F HEK293 T li n es (LoF /L oF) w it h dif f e r en t c omb ina t ions of var iants (line s N o.20 and 9 8: s t op- g ain/ s t op-g ain; No .15: s t op-g ain/20b p deletion ; N o .4: 4bp deletion / 114b p del et ion; No.216: 94bp deletion/94bp deletion ) , o ne het e r o zy gous LoF st o p-g ain/ + line (No.1), tw o homo zy gous Ala233Thr / Ala233Thr KI NS S H IP / KI N S SHIP lines (N o .54 and 90) and tw o co mpound het e r o zy gou s KI NS SH I P and LoF li n e s (No.5 1 and 86: A la233Thr/ st o p-g ai n ). Thes e t en li nes and thr ee unm u t a ted HEK293 T lines w er e gr own simult aneously i n b i o l o g ic al tripli c at e b e f or e RNA e x t r action wit h RNeas y Mini Kit (QIAs t ock, QIA GEN A G). The nomencla tur e of t he engineered v ariants is: A. Stop g ain ( t hrou gh A insertion) : G RC h37:2:10 06232 65:A :AT, N M _002 285.3 : c.70 1dup, NP _00227 6.2:p.( Tyr2 34Ter ) N C _00 00 02.11: g.10 0623 266dup . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint B. 4bp del : G RC h37:2: 1006 23262 :CA CAT :C, NM_002 285.3 :c .701_70 4del, NP _00227 6.2:p.( Tyr2 34Ter ) N C _00 00 02.11: g.10 0623 265_100 62326 8del C. 20bp del : G R Ch37: 2:100 62324 7: GC C GT CCAT T GGC CT CA CATA:G, N M _002 285.3 :c .700_7 19del, N P _00 2 276.2:p.( Tyr2 34Pr of sTer6) N C _0 0000 2.11: g.1 00623 249_100 623 268del D. 94bp del : G R Ch3 7:2:1 00623 727:A C G A G G GCT G G T TC T G GGC TC TT GAATC T GCAAC AAA A TG T TCA T C G A TC TT GTTCA CA G G AG TC T GAGGAAC CC CAG G T TT G GGAAC T CCAAC G :A, N M _002 285.3 :c .276_3 69del, N P _00 2 276.2:p.(Val9 3LeufsTer97) N C _0 0000 2.11: g.1 00623 728_100 623 821del E. 114bp del: G R Ch3 7:2:1 00623 258: G CC TCA CATAC GC G GTC G GTT TC T GC T GGACCA GG CT GG G T TT T GAAGC T AGGGAT G G A GG A AAGT TC T G A ACAC A G T G TC C GC T G C T GC T G T G CT TG GCC G C CATG GC AGG T GG C: G, NM _002 285.3 :c.5 94_70 8del, NP _002276.2 :p .(Arg198Ser f s Ter 16) N C _0 0000 2.11: g.1 00623 260_100 623 374del F. Ala233Thr K IN SSHIP variant : G R Ch37: 2:10062 3268: C GC :G GT, N M _002 285.3 :c .697_6 99delinsACC , N P_00227 6.2:p.(A l a233 Th r) N C _0 0000 2.11: g.1 00623 268_100 623 270delins GGT Fibroblas ts Fibr oblas t cell s fr om tw o pa ti en ts ’ ski n biopsies and t hr ee health y ag e-ma t c hed con t r ol indiv iduals (2-16 year s of age) w er e gr own simult aneo us ly . A t the desir ed cell c onfluency , R N A was e xtr a ct ed with RNeas y Mini Kit ( Q IA st ock , QIA GEN A G). T r an scr i p t ome pr ofil ing RNA q uali t y w as as s e ss ed on a Fr ag men t A nalyzer ( Agilen t T ec hn ologi es ). The R N As had R QNs betw een 9.0 and 10.0. R NA- s eq libr ar ies w er e pr epar ed fr om 500 ng of t otal RNA with the Illum ina T ruSeq Str and ed mRN A reag en ts (Illu mina) using a u niqu e dual inde x ing s tr a t egy , and f ollowing t h e of ficial pr o t o col au t omat ed on a Scic lon e liquid h a n dli n g r obot ( P erki nElmer). Libr ar ies w ere quan tif ied by a fluor ometr ic method (QubIT , Lif e T e chnologies) and their quality asse ssed on a Fr agmen t An a ly zer (Agilen t T ec h nologies ). Sequencing w a s perf or med on a n Illumina Nov a Seq 6000 f or 100 cy c les singl e r ead. S equencing da t a w e r e demul tiple x e d using the bc l2f a s t q2 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Con v er sion Soft war e ( v er sion 2.20, Illumina). W e pr ofiled tr ans cript omes wi t h a mini mum of 17.9 and 52.9 million mapped r eads f o r HEK 293 T a n d fibr oblas ts, r e spec t ively . The HEK293 T and fibr ob l as t r eads ar e depo s it ed in G E O under a c ce s s ion G SE 2416 21 (t o k e n ot e zqc sslbctb yx ) and G S E246 554 (t ok en wx ibc y qa t dyphkl), r espec t ively Ra w r eads w er e aligned t o th e human (hg38) g enome us ing S T A R (2 .7. 10b), the e x act par amet er s ar e: ST AR 28 --runMode alignR eads - - tw opass M ode Basic -- outSAM type BAM Sor t edB y C oor dina t e --ou tSA Ma t tribut es A ll -- r eadFiles C ommand "g zip -dc" - -quan tM ode GeneCoun ts. Gene c oun ts w er e g ener a t e d us ing F eat ur eCou n ts 29 and diff er en t ial e xp r ession analy s i s w a s perf o rmed w it h th e DE Seq2 (v .1.3 6. 0) 30 pac kag e fr om Bioc onduct or (v3.15) 31 . G enes w er e c ons ider ed diff er e n t ia ll y e xpr es sed bas ed on an adjust ed p-v alue c u t o f f of <0.05. P a th w a y en richmen t analy s i s w as c arried o ut using clus t erP r ofile r (v .4.4.4) 32,33 fr om Bioc onduct or using the en r ic h er function . G SE A 34 analy sis w as carr ied out us in g the G SE A function in Clust erPr o file r , and t he f ollowing ann ot at ed g ene s et s fr om MSigDB v6.2 35 : the H all mark g ene s et 36 . F or com pari son with ChI P- s eq s tudies in human H EK293T 13 and E S mou s e cel l lines 14 , e x t ernal sequenci n g da t a in bigW ig f or ma t w er e acquir e d fr om GE O . UCS C b i g Wig files w er e cr ea t ed a t 1bp r esolution and normalized t o t o t al al ignable r eads (r eads-per -mi llion ). P eak det ection w as perf or med wi th M ACS v3.0 37,38 us ing t he bdgpeak call functi on ( wi t h cut-o f f s 0.4 and 0.6 r es p ec t i v ely). The A F F3 peak r e gions in m ic e w er e l i f ted ov er t o t h e h g 3 8 hum a n g enom e assembly . The pea k r egions w er e an not at ed in R us ing t he ChIP seek er 38 pack ag e, in par ticula r the “ ann ot at e P eak” f unc t ion.

KI NS SH IP probands T h ro u g h d ata ag g regat i o n , we i d e nt i f i e d t h re e m o re K I N SS HI P i n d i v i d u a l s ( K22-K24), t w o of whom (K22-K 23) harb or pr ev iou s ly unr eport ed de novo v a r i a n t s ( Figur e 1A ; T able S1). T w o m o re individuals (K25- K26) car rying pr eviously det ec t ed v arian t s 11 a re d e s c ri be d i n re f e re n c e 39 . T h e p.(Pr o231Ser), p.( Ala233Ser), p .(Ala233Thr) , and p.(M et238 Thr) var ian ts of K22-K 26 f all wi t hin th e nine amino acids long 2 30-K P T A YVR PM-238 de gr on motif and f urth er expand the number of i t s r esidues who s e mod i f ica tion i s as s o cia t ed with K IN S SHIP ( i.e. Pr o231 , Ala233, V al235 and Met238 ; number ing acco r ding t o NM_00228 5 .3 thr oughout ) ( Figur e 1A ). P a t hogenic it y of the pr ev iously undes c r ibed mis sen se v arian t s is suppor t ed by the 3D r epr es en t a tion of the enc oded degr on . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint peptide ( Figur e S1 ). Wher eas chang e s at Pr o231 w er e p r evi ous ly su g g e st ed t o aff ect the b a ckbone kink c onf err ed by this c ons erv ed r esidue 11 , t he M et238 sidechain is point ing out war d, f orw ar d - f acing the Ser15 4 sidechain of a S IAH ubiquitin l ig ase loop. M odeling s u g g es ts tha t v arian ts a t t his position s ho uld sligh tly a lt er binding , pr edicting a les s sever e p henoty pe. Con sist en t w it h t his h ypothesi s , pr oband K 22 pr e sen t ed wi t h mild DD , mi ld speec h impairme n t , f acial dy s mo rphism s , sk elet al malf o rmations, mild h y p ertr i c ho sis , and mild h ypot onia, a phenot ype milder than t hat o f typic al KI NSSHI P individuals with var ian ts of r es idues tha t dock in the ubiquit in lig ase binding pock et 11 . Duplicati on pr ob and Da t a ag gr eg a tion also enab l ed as c er t ainme n t o f an indi vidual wi t h a KIN SSHIP- l ik e phenotype ca rr y i n g a de novo partial dup lic a ti on of AFF3 furt her s tr eng t hening t he h ypot hes i s tha t an incr eas ed level of AFF3 is pat hological . Th i s D U P 1 p r oband pr es en t ed with sev e r e f ailur e t o thriv e with post na t al o ns et , sev er e D D with poor ey e c on t a ct , poor h ea d c on tr ol, inabili ty t o s it and s p eak, epileps y , c orp us callosum h ypo plas ia, f acial d ysmor phis m , h ypert richos i s, h ypot onia, hip, knee, ankle and wris t f l e xion c ontr actur es and s ev er e sco l iosis. W ho le g eno me sequenci n g r ev ealed a t ande m dupl ica t ion of t he int erval enc om passi n g e x on 10 t o e xo n 24 of AF F 3 en coding part o f t h e ALF domai n, t he T AD , NLS and CH D do m ai n s and e x on 1 t o 3 (up to i n tr on 3-4) o f th e s am e orien t at i o n ubiqu itous REV 1 (chr2 :g. 100, 077, 649_10 0,3 59 ,928dup (h g19), N C _0 0000 2.11: g.1 00,0 77,6 49_100 ,35 9,928du p) ( Figure 1A-B ). The e xpres si o n of t he par tia lly duplic at ed c opy of AFF3 is then u nde r th e con t r ol o f the REV1 pr omot er , whic h could r e s u l t in th e e xpr ess ion of a degr o n-l es s AFF3 , a h ypothesi s t hat w e c ould no t fu rt her t est dir ec t ly due t o lac k of a vailable sample. Het er o z ygous LoF and bial lelic pro ban ds T o furt her challe n ge t he h ypothesi s t ha t dimin ished e x pr ession of AF F 3 i s delet er ious, w e s ear ched f or individuals with los s - of-fun c t ion ( L oF) v arian t s in AFF3. U s ing da t a aggr eg a tion of multipl e labor a t o r ies and clinical c en t er s , e. g. GeneMa t cher 40 a n d D E C I P H E R 41 , w e iden tif ied t en aff ect ed individuals wi t h m on o-all elic (indi vid uals L1-L9 and L14) and thr ee (L11- L13) with bia lle lic AF F 3 tr unca t ing v aria n ts, as w ell as a pr ob a n d c ompound het er o zy gous f or a LoF and a r ar e m is sen s e v arian t (L 10 ; Figure 1A and 1C ; T able S2) . Of not e, one of th e af f ec t ed i ndividuals desc r ibed in re f e r e n c e 18 w as a c ompound h et ero zy g o t e f or a C G G e xpansi o n and a d el et ion of the AF F 3 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint pr om oter . The id en tified trun cating v arian ts ar e no t desc r ibed in Gnom AD (v2.1.1) ( Ta b l e S 2 ). Consi st en t with the dele t er ious n es s of d i m i nis hed or absen ce of AF F 3 e x pr es sion, t he s e f o urt een individuals s h a r ed c omm on p henoty pes su ch a s global DD / ID ( 11 out o f 11) , abnor mal c or pus c a llosum (4/6), spee ch impairm en t (10/ 11) , muscle disor der s/h y p ot onia (7/9), f acial dy s m orphism s (6 / 7) , mild cr anial d y s m orphism s (3 / 8) and s k elet al d ef ec t s ( 4/ 7). T o r e s pect medrxiv guideli nes and k eep af f ect e d individuals unident ifiab le t he lis t of all s ympt oms i s not incl u ded in t his p r epri n t , bu t c an is a vail able upo n r eques t t o t he corr esponding author . This gr oup c oun ts 10 male s and 4 f emales sug g e st ive of a h igher “ male suscept i b i lit y ” ( p =0.08 9) 42 . The s ibling s L12 a n d L13, who ar e h omozy gou s f or a tru nca t i n g v ar ian t, and L10, w ho is c ompound het er o zy got e f or a LoF var ia n t and miss en se p.( G ln1 020Ar g) pr esen t a m or e s ev er e phen otype than t hei r par e n ts who ar e h e t er o zy g o t es f or t he LoF var ia n t (f ami lies 6 and 8 ; Figure 1A and 1C ; T able S2). Our sear ch als o iden t ified four af f ect ed individua ls wi t h bialleli c hom o zy gou s (B1-B 3) or c ompound het er o zy gous (B7) mis sense v arian ts in AFF3 ( Figur e 1A and 1C ; T able S2 ). A c ons an guineou s f amily with t hr ee aff ect ed individuals was pr eviousl y de s c ribed in re f er ence 43 (B4- B6). These v arian ts ar e eit he r not d e s c r ibed in G n omAD and/or aff ect th e e x pr es sion of AFF3 (s ee below and T able S2) . The y pr e sen t ov erlapp i n g s ympt oms s u ch a s DD / ID (6/7) and ADH D (A t t e n tion D e f icit Hyper activity D is o r der) (2/3), epilept ic ence p halopa th y or abnor mal slee p EE G (Electr oencephalogr a p h y ) ( 3/ 4), shor t / no a t t en t ion s pan (3/4), spee c h impair me n t (3/4), heart de f ects (2/ 4 ), and vis ion i m pairmen t (2/ 4) , and other de f ects. Thes e t w o c oh orts show ed a milder phenot y p e t han KINSSH IP pr oband s, s u g g e s ting they migh t r epr esen t a new s yndr o me. In sili c o m ode ling of most of the ide n t if ied m iss en se v arian t s is hamper ed by t he lack of r el iab l e AFF s st ru ctur al i n f orma tion with t he e x c ept ion of the C HD tha t i s imp ort an t f or dimer iza tion and the ALF that c o n t ains the degr on a nd t he ELL-binding do mains (ELL bow , s ee below) 44,45 . T he p.(Gln1020Ar g ) , p .(V al10 36Ile), p.(Ar g 1 186 G ln) and p.( Gly1 215V al) var ian ts f all within th e CHD (Figure 1A ). A bulky side c hain a t po s ition 1 215 w ill c ol lide wit h Leu1063 and/o r Leu1192. L ik ewi s e, V al1036 is op timally s ur r ounded b y t he h y dr ophob i c sidechai n s of Leu106 8, Leu1071 , T yr1 072 and Met 1075, and canno t ac c omm oda t e the bu lkier p .(V al1036Ile) v a rian t without a f f ecting loc a l pac k ing. Gl n 10 20 and Ar g1186 ar e loc a t ed a t the doma in s u rf ace, wher e chang e s in the loc al char g e migh t af f ect bind ing spe ci f ic ity . . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint La s tly , it is pos s ible th at s o me mis sen se v a r ian ts outside of t he d egr on ar e link ed t o an au t os om a l dominan t dis ease, as w e iden t i f ie d an indi vidual c arr ying a de nov o p.(A la886Thr ) v arian t pr esen ting with D D , speech impairmen t and ASD ( Autism Spectr um Disor der ) sym pt oms and a s an individual with a de novo p.(L eu312P he) varia n t pr e sen ti n g wi t h DD w as descr i b ed in re f er ence 39 (M 1-M2, Figure 1A , T able S2). The p. (L eu3 12Phe) var i an t m a p s t o the EL Lbow and 3D m odels s u g g e s t th at f our of the f i v e possible Phe r ot am er s w ill sever ely c la sh with either th e AFF3 Phe329 or t he EL L2 His618 r esidue. Animal models T o furt her a sse ss if t he d imin ished e xpr ess ion of AFF3 w as delet er i o us t o or ganismal pheno types w e knoc k e d -down ( KD) aff3, the zebr afish o rtholog of AF F 3 , usi n g CRI SP R-C a s9 geno me editing. W e u s ed tw o single guide R N A s t ar g eting e x on 6 each pr ov iding mor e than 90% e f ficiency . A t 5 da y s pos t f er ti liza t i o n (dpf ) , we obs er ved malformat ions in 10 % o f KD larv ae, including inc omplete ey e pigmen t a ti on , alt e r ed h e ad st ructur e, la t er al b elly edema, peric a r dial ed ema an d sk elet om us cular dy s mo rphology ( Figure 2A ). St aining of th e c artilagi nous c r anial str uctur e r ev ealed m a lf ormat i o ns in 75% of KD larvae ( Figur e 2B ) . in t er -ocular dis t ance ( IOD ) a n d head width (H W ) w er e s ignific an tly decr eased in af f3 K D c o m p a r e d t o u n in j e ct ed ( Un ) (p =0.011 IOD and p =0.001 H W ) an d moc k (M) inject ed l arv a e ( p =0.041 IOD ; p =0.004 H W) (Figure 2C-E) . T h e e s c a p e r esponse t es t upon a t actile s ti mulu s per f ormed a t 3 dp f s ho wed tha t while none o f t he mo c k- inject ed zebr afish show ed pert urbed es c ape r es pon s e s , about a th ir d of the af f 3 K D l a rv a e w e re af f ect ed ( p <0.0001). Wher eas the major i t y pr es ent ed ei t her l o opin g ( 22.5%) or p i n w heel swimm i n g (4.7%) , beha vior s lin k ed t o neur ologi c al and m e c h a n o-s en sor y s ys t em impairme n t 26,46,47 and 5.6% w er e m ot ionless due t o ex t e n s iv e malf orma t io ns ( Figur e 2F ). A t 5dpf , the l o com o tion abi lit y w a s q uan tit a t i v ely ev alua t ed wi th an auto mat ed t rac ki n g devic e. The aff3 KD larvae show ed a st a t i s tic ally si g nifican t dec r eas e in global s wimm ing v elocity in the dar k c ompar ed t o Un ( p <0.0001) and M lar v ae ( p =0.0025) ( Figur e 2G ). A s such h y p o-loco motion is oft e n asso ciat ed with neur o m ot o r d e ficits and akines ia 46 w e i m munos t ain ed af f 3 KD larvae hindbr ai n and mo t o neur ons. Hind b r ain Mauth n er cell s 47 p r esen t ed a g en er al develop men t a l de la y , a n d w h i l e norm al in gr owth a n d ar chit ectu r e, m ot oneu r ons w er e dis or gan i zed i n de f or med larvae ( Figure S2). Si m il ar ly , w e pr evi o us l y s how e d tha t t he or tho logou s mou se kno c k outs , Aff3 +/- and Af f 3 -/- C57BL/6N, e xhibit s k elet al de f ects, an abnorm a l sk ull shape, kidne y d e f ects and n e u r ologic a l dy sf unc t ion 11,48 . Ho m o zy gou s Aff3 -/- e xhibited sig n i f ican t l y enla r ged lat er al vent ricles and . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint decr ease d c orpus c allos u m si ze, w h en c ompar ed with b oth wild-t ype and Aff3 +/- males 11,48 . Our zebr afis h and m ous e r es ult s support t he c on t e n tion o f causa tion f or AFF3 L oF v arian ts. As s e s s ing varian t s W e pr eviou s ly sh ow ed t ha t ov er e xpres sion in zebr afi sh embr y os o f huma n AFF3 l e a d s t o a d o s e - dependen t incr ease of develop m en tal anomalies 11 , a ph enotype tha t w as fu rther e x acerba t e d upon over e x pr es sion of the p.( Ala23 3Thr ) KI NSSHI P isof or m 49 . T o as se ss t he pa thog en icity of th e miss en se v arian t s iden t ifi ed in t he bia llel ic ind i viduals, w e injec t ed zebr afish w it h hu man AF F 3 mRNA wild-type ( W t) , tw o s elect ed m is sen s e v arian t s pr es en t in h omo zygou s s t a t e in pr oband s B1 and B2 and his af f ec t e d sist er B3 a nd mapping out s ide of c r y st alized do mains (L ys528Ar g and Thr594Ser ), tw o KI N SSHI P v arian ts ( Ala233Thr and V al235 Gl y ) and a s c on tr ol G ln179Glu (Chr2 (GR Ch37) g.100 62343 2: G>C , c.535 C >G ) , a v arian t n ot desc r ibed in G n om AD , w e ide n ti fied i n homo zy go s ity i n a health y ind i vidual. The r esulting 5dp f lar v ae w e r e c a t a l og ed as des cribed 49 : (i ) norm al phenotype, (ii ) Class 1 with sk el et omusc u l ar dy s mor phology and sm a ll d imension, (ii i) Class 2 with a mo r e sever e phenotype including a t leas t t hr ee of sk elet omuscul a r dy s m orphology , s m al l dimensions , head malf or m a tions, ey es’ alt er a t i o n, per i c ar dia l ed ema, an d la t er al belly ede m a o r (iv) deceas ed. Con si s t en t wi t h pr evi o usly publi s h ed obser v a tions , accumul a tion o f AFF3 W t m R N A s ignific an tly incr eased the n umber o f lar v ae with debil ita t ing tr ai ts ( p =0.00 02). Compar ed t o AF F 3 W t mRNA acc um ulat ion, both Ala233Thr and V al235Gl y isof or ms led t o a f ur ther s ignific ant incr eas e in t he nu mb er o f malformed lar v ae a n d mort ali ty at al l dos es ( p <0.0001) (Figure 2H ). Over e x pr es sion of the tw o m i ssen se v arian ts iden tified in pr oband B1 and B2 si m il ar ly c au s ed higher malf or m ations and m or t ali t y r a t es t han over e xpr es sion of AFF3 W t (L ys528Ar g p<0.000 1 and Thr594Ser p=0.0018) albeit no t a t the r a te o f th e KIN SSH IP v arian ts. O n the c on tr ar y , th e c on tr ol v ar ian t p .(G ln179 Glu) had an e ff ect similar t o t h a t o f Wt ov er e xp r e s s ion ( p=0.7; Figure 2H ). These r e s ult s su g g e st t ha t lik e t runc a ting v arian ts, a t least s om e of the mis s en se v arian ts iden tifi e d in the aff ect ed indiv iduals co uld be causa tiv e. T o challeng e this h ypothesi s fu rther , w e per f o rmed a phenot y p i c r es cue e x per imen t 50 . A s described abov e, dep let i o n of af f3 i n 5 dp f zebr afish larvae r esult ed in decr eased s wimming v elocity . That de cr eas e c ould be r es cued by co- injec t ion of human AFF3 W t m R N A d e m o n s t r a t i n g f i r s t t h a t h u m a n A F F 3 c a n c o m p e n s a t e f o r t h e l o s s o f i t s ze b r a f i s h ort holog and se c ond that abl a tion o f af f3 activi t y w as c aus a tiv e of t he p henotype ( Figure 2G ). Consi st en t with the det r ime n t al ef f ec t o f th e ov ere xpr essi o n of the KINSS H IP v arian ts, we ob s erved . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint an ev en lower a ver age v e locity upon c o-injection o f V al235G ly and A la233Thr mRNA s (both p<0.001 c ompar ed to inj ection o f Wt; bot h p <0.0001 c ompar ed t o M). W hile c o- injection o f L y s52 8Ar g mRNA did not r escue af f3 a b l a t ion, c o-injection o f Thr 594Ser mRNA r esult ed i n a n “ o ver -r escue” with inject ed larvae p r es en t ing a n i n c r ea sed v elocity com par ed t o m oc k ( Figur e 2G ). The c on tr o l var i a n t p.(Gln179Gl u ) ha d an in t e r medi a te eff ect h al f wa y bet ween W t and L y s528Ar g mRNA injections. T og et her these results s u g g es t tha t t hes e mi ssen s e v ariants impact the activi ty of AFF3 and th at bial lelic AFF3 v a r ian t s c ould be as s o cia t ed with an aut o s om a l r ecess iv e disea s e. Consi st en t with the la tt er h ypot hes is neith er hom o zy gous nor c ompound het er o z y gous c la s s ified as ’ w ea k missense var ian t or w or s e’ , i.e. wi t h a MAF ≤1% and REV EL sc or e ≥0.644 51 wer e identif ie d in G no mAD v2.1.1. T rans c r ipt ome prof il i ng T o c ompar e t h e t r ansc r iptional c onsequences of A F F3 los s and over e x p ression, we us ed CRISPR- Cas9 g enome editing t o engineer mult ip le D N and LoF v arian ts in an isogen i c cel l model, the human embry onic k idn ey 2 93 T l ine. H EK293T w a s chosen (i) as KI NSSH IP individuals oft en presen t with a hor se shoe kidne y 50 , (ii) a s AF F3 i s e xpr essed i n this cell line, and (iii) as in th i s line both tr an script o me pr ofil es of AF Fs shRNAs knoc k do w n s 3 and (iv) ChiP-seq of AFF3 ha v e been published 13 . W e engineered f iv e bial l eli c LoF H EK293T lines (L o F /L o F ) with di f f er ent c ombi na t i ons of var ian ts (lines No.2 0 and 98: s t op-g ain/s t op -g a in; No.1 5: s t op- g a in/2 0bp deletion; N o .4: 4bp deletion/114bp deletion; N o.216: 94 b p deletion/94bp delet ion), on e hete r o zy gous LoF st op-g ain/ + line (No.1), tw o homo zy gou s Ala233Thr/ Ala233Thr K I NS SH IP /KI NS S H IP l i nes ( N o.54 and 90) and tw o com pound he t er o zy gou s KI NS SH IP and LoF li n es (No.51 an d 86: Ala233Thr / s t op-g ain). W e pr of i led the tr ans cript omes of thr ee biological r epl ic a t es of each of th es e lines by R N A -sequenci n g and c ompar ed them t o those of thr e e b i o l o gi c al r eplic a tes of t hr ee wild type lines (Wt1, Wt2 and W t 4), f or a t ot al of 39 profi les. While AF F 3 mRNA l evels ar e sig n i f ican tl y decr ea s ed in the fi ve biallel i c LoF / LoF lines ( padj =3.5 E-53), t he Ala233Thr / Ala233Thr KI NSSH IP /KI NSSHI P and the LoF / + lines pr esen t AFF3 t r ansc r ipt l ev els c ompar able and int er mediat e ( padj = 0.011 ) t o tha t f ou nd i n c on tr ol W t lines , r es pectiv ely ( Figur e S3 ). The similar le v el of e x p r es sion of AFF3 i n t h e w i l d t y p e and the Ala233Thr/ Ala233Thr lines su g ge s t t ha t ther e is n o neg a tive f eed back loo p t o balance th e amoun t o f AFF3. W e fir s t c ompar e d the tr anscript om e of h o mo zygou s LoF / LoF and Ala233Thr / Ala233Thr lines t o tha t of +/+ l i n es and i d e n t ified 3,553 and 4,177 dif f er e n t ia ll y e xpr ess ed g ene s (D E G ) a t an a d just ed p-v alue th r es h old o f 0.05, r espectiv ely ( Figure 3A-B , Ta b l e . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint S3-S4 ). W e ob ser v ed an ov erlap of 2 3% of DE G s wi t h pr evious tr a n s cript ome pr of i l ing o f HEK293T cel ls in whic h A FF3 w a s k no ck ed dow n with shRNA s 3 . The LoF /L oF and KI N SSHIP /K I N SSHI P (D N /D N) lines pr esen t wi t h signific a n t r epress ion of g ene s in v ol ved in the G2-M t r ansition, o xid a tive phosphor y la tion , t ar g ets of E2F and MY C-r el a t ed g enes , mos t mar k edly in th e DN model. Both lines show ed an upr eg u l a t ion of the inflamm a t o r y r espons e (e. g TN F A -s ignaling via N FKB) and pa t hw a y s impor t an t f or m y og enesis a n d apical ju nc t ion ( Figure 3B-C) . O n l y abou t a thi r d o f t he DE Gs a r e c ommon t o both d a t as et s su g g est ing tha t AF F 3 L oF and DN v arian ts lar g ely modu l a t e tr anscript om es dif f e r ently ( Figur e 3A ) . F o r e x a m p l e , p a t h w a y s in v ol ved in t he epi theliu m- mes ench y me tr ans it i o n, early r esponse t o est r og en, h ypo xia, x enobiot ic met abolism and apop t osis , as well as g enes tha t ar e downr egula t ed by K RAS ar e s p e c i f i c a l l y upr egula t ed in the KIN SSH IP /KI N S SH I P lines ( Figure 3B-C ). Within the set of 1272 c ommo n DE Gs , 121 g enes pr es en t opposit e mir r or ef f ects in both s tr ai n s , i.e. they ar e up-r egulat ed in one g enotype and downr egula ted in the ot her ( Figur e 3A ) . They ar e enr ic hed f or D N A r epair g enes , a p a t h w a y a ct i v a t e d in Lo F / Lo F b u t n ot i n K I N S S H I P / K I NS S H I P ce l l s ( Figure 3C). A cor e s et of 20 DE Gs ar e simila rly modif ied upon AFF2, AF F3 and AFF4 knock-d own 3 o r w h e n AFF3 harbor s homo zy gou s DN or LoF v a r ian ts s u g g e s ting tha t they ar e s en sitiv e t o an y SE C s ‘ pert urba tion. W e then ass e s s ed if D EG s w er e di r ect or ind i r ect t a r g e t s of AFF3. While i n e x ce ss of 3 ,500 g enes ar e dy sr egula t ed in each g enotype, only 226 genes pr es en t ed with a ne i ghbor ing AFF3 ChIP - s eq peak u sing a FD R of 0. 05% 13 ( Ta b l e S 5 ), s u gg es ting t ha t man y of the obse rv ed t r ans c r ipt ome chang es ar e downs tr eam e ff ects. H owev er , 32% (74 out o f 226) of the b o und l oci w ere dy s r egulat ed in ei t her th e LoF /L oF and /or the DN/ D N lines ( Figure 3D ). The binding s it es of the ort hologous mou s e Af f3 w e r e de t ermined i n E S c ells b y ChiP- seq 14 . Upon lif ting A f f3 ChIP- s eq peak s t o the hum an g enome, w e si m il ar ly f ound th at 42% of g enes with a binding sit e (374 out of 881) w er e DE G s in either the L oF /LoF and/ or t he D N /D N lines ( Figure 3D ; T able S6). While su c h in t e r- c lade binding c ompar i sons ha v e c a vea t s , our H EK29 3T and E S r esults su g ge s t tha t a s u b st an t ial p r oportion of bou nd loc i ar e dysr egula t ed u pon change s in the e xpr ession level of AF F 3 a n d / or s t a bi l i ty o f AF F 3 ( Figur e 3D-E ; Figur e S4 , T able S7). Thes e dy s r egula t ed di r ect t ar get s incl u de g enes as so c ia t ed wit h tr ai t s pr es en t in AFF3 v ariants c arrier s su ch as n e u r odev elopm ent a l dis o r der s (e.g. A GO1, AR V 1, CDK 5 RAP2, C N NM2 , CR ADD , D PP 9, EDE M3, G A N, H N RN P A2B1, HNRN PU , IRX 5 , M G R N 1, M TRR, PR EP L , SO X 4 , SR CAP , T OR1A TUBB , VP S35L ) and aut i sm (i.e. par t o f SF A RI g ene l is t, e. g. CD K5RAP2, CT NNA3, D A GL A , D L X3 , LDB1, MYH10 , PREX1, PRKA CA , SETDB1, SR CAP and ST 7 ) , ossi f ica tion and l i mb def ects (DL X 3 , DPP 9 , D Y N C 2I2 , FN 1, IR X5 , , RPS7SR CAP , . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint VPS35L), pilosity abnor malities (DLX3, GAN , JUP , N R 3C 1), r enal d i seases ( CNNM2 , R PS7, Z N F42 3), c a rd i a c d i s o rd e r s ( CT N N A3 , JU P , KI F2 0A, L MN A , VPS35L ) and dy smor phisms ( LEMD 2, IRX5, LM N A, RPS7, SO X4) . Th ey al so com prise k e y g enes im pli c a t ed in a x on guidance, c e ll migr a tion a n d c ell f a te (e. g. D D X17 , EF NA5, FZ D 7 , G A N , I S L 2 , J U N , M D K , M Y O 1 D , N A V 1 , N O T C H 2 , S N A I 1 and SP1). Impor t antly , some dir ec t t ar get s ar e up r egula t ed, while oth er s ar e downr e gu l a t ed. F or e xample, DD X17 is do wnr egula t ed in bo th LoF /L oF and DN/ D N lines, wher eas CT N N A 3 and NR3C1 a r e o n l y downr egula t ed in LoF / LoF lines and CDK 5RAP2 o nly in D N /D N lines. On t he c on tr a r y , CNNM 2 i s upr egula t ed in D N /DN lines ( Figur e 3E , Figure S4 ). As m a n y of our af f ect ed indiv id ua l s pr esen t het e r o zy gou s LoF v arian ts, w e then c ompared th e tr an script o me pr ofiles of th e LoF /+ li nes with tha t of t he LoF /LoF lines and obser v ed tha t, w h i l e the s ame p a t hw a y s ar e aff ect ed ( Figur e S5 ), o nly 22% of t he D E G s of th e hom o zy gous line w ere als o dysr e gula t ed in the he t er o zy gou s l ine, s ugg es ting a do s e depen den t m odificat ion ( T able S8). W e similarly co mpar ed the t r an sc r ipt omes of th e comp ound het er o zy got e D N /L o F lines t o th os e of both the LoF / LoF and DN/ D N lines. While 37% of t he DE G s co mmon t o LoF /LoF and D N /D N lines ar e modifi ed in th e D N/ LoF l ines, w e a lso ob ser v e in this het er o zy go us line modific at ions in e xpr ess ion levels tha t ov erlap with D E Gs spe cifically modified i n e ither h om o zygou s l ine c orr esponding t o 22 and 27% of t he DE G s specifically modif i ed in eac h gr oup of lin es, r espectiv ely (T able S9). This su g g es t s a c o-d omi nance of the Ala233Thr and s t op - gain v aria n ts wher e th e incr eas ed st abil it y of t he fir st al lel e does part ial ly c omp ens a t e the d e cr e as ed expr ession lev el o f the seco nd al lele in some inst anc es and ov er-c ompensa te in o th er s. In par allel, w e com par ed the t r anscri pt ome o f pr ima r y f ibr oblas ts f r om t w o pr obands with bia llel ic miss en se alt er a ti ons of AFF3 (B1: homo zygou s p.(L y s52 8Ar g) ; B7: c ompound he t er o z y got e p.(V al1036Ile)/p.(Ar g1186 G ln ( Figur e 1A ; T able S2) t o those of thr ee h ealth y c on tr ols by R NA- s eq uenci n g. W e f ound 142 DE Gs a t an adjp- value t h r eshold of 0.05 ( Ta b l e S 1 0 ). AFF3 mRNA lev el s ar e significan t ly dec r eas ed in both probands (p< 0.0002; T able S10) and a comp ar able nu mber o f dist inc t ive r eads c orr esponding t o bot h allel es o f t he com pound het e r o zy got e w er e ide n ti fie d (T able S11), which i s c onsis t en t w it h t he not i on th a t the thr ee AFF3 m i ss e n s e v a r i an t s h ar b o r e d b y these pr obands ar e delet er ious. W hil e o nly ninet een and s ixt een per cen t of the fibroblas ts DE G s a r e a l s o D E G s i n t h e L o F / L o F a n d D N / D N H E K 2 9 3 T l i n e s , r e s p e c t i v e l y , t h e s a m e h a l l m a r k p a t h w a y s ar e dysr egula t ed . F or e xam ple, g enes in v olved in the G2-M t r ans it ion, t ar g ets of E2 F and M Y C- r elat ed g enes and in t er f er on alpha -r espons e ar e enr ic h e d with i n t heir list of r es pect i v e DE G s . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint (T able S11, Figure S5 ). The se r es ult s s ug g e st t ha t s imilar pa thol ogic al mechanisms a r e a t pla y when AFF3 is haploinsuf ficien t and when it har bor s bialle lic miss en se v a r ian ts.

W e pr esen t evidenc e su g ge s ting tha t mult iple AFF3 var i an t-spec if i c mec hanis ms ar e a s s o c ia t ed with c ognitive i m pai r m en t. W h i le s om at ic varian ts in t he pr o mo t er of th is g ene w er e pr ev iously link ed t o mild ID 17,18 , w e show tha t DN 11 , du pli c a tion, tr unc at ion, delet ion, ab s ence as well as b i a l l e lic v a r ia n t s i n AFF3 ar e as so cia t ed with ID . Sug g es tiv e of s emi- domi nance, homo zy gou s L o F (L12 and L13) and co mpound heter o zy gous LoF /mis s en se ( L 10) individu als pr esen t more s ev ere phenot y p e s t han their h e t er o zy g o u s par en ts ( Figure 1C) . The h ypothesis tha t non - degr on de no v o miss en se v arian ts ar e also link ed t o D D/ID war r an ts fur ther in ves tig a ti on a n d the i den tificat ion o f mor e a f f ect ed indi viduals. Comm o n v ariants in th is locus w e r e simila r ly GW AS- o r M T A G - asso ciat ed (mult i-tr ai t analy sis of G W AS) with cognit ion pr o x ies suc h as fluid i n tellig e n c e , educ a tional at t ain men t , and m at h e ma tic al abil it y , or with c orr ela t ed t r aits s u ch a s h ou s ehold inc ome, occupa tional at t a inment , an d br ain morphol o g y 52 . C ons is t en t wi th these findings, AF F 3 and its mac a q ue, mouse, r a t, r abbit a nd chick en ort hologs ar e e x pr es sed dur ing t he ear ly st ag e s of br ain and cer e b e ll u m dev el opment i n particular in la t e n eur on s 53,54 , whe r e it pla y s a d i r ec t r ole in the migr a tion of c ortic al neur on s 55 . L i k e w i s e , c o m m o n v a r i a n t s i n t h i s l o c u s a r e a s s o c i a t e d w i t h s c oliosi s , a n t hr opo metric t r ai ts (BM I, heigh t) and pu lm ona ry in v ol v eme n t (vi t al c apac ity , as thma, chr onic obs tru ctiv e pul monar y d i seas e), thr ee c ar dinal f e a tur es of KIN SSH IP s yndr om e. AFF3 i s a l so GW AS-/MT A G-a ssociat ed wi t h diabetes (type 1, t ype 2, diabetic nephr o pa t h y and HD L choles t er ol), addictions (smo king init i a t ion, alc oh ol c onsumpt ion, c annabis dependen ce, t elev ision wa t ching), aut oim m unity (lupus, celiac diseas e, r heuma t oid and ju v eni le idiopa thic ar thri t is ), se xu a l dev elopme n t and di mor phis m (ag e a t m enar che, endomet r iosis , mammogr aphic density , m al e baldness, biologi c al se x), blood m ea sur emen t s (e. g. hema t oc r i t, h emogl obin measur em e n t), eye dis ease s ( e.g. as tigma t i sm, in tr aocula r pr ess u r e, c orneal measur emen ts), and insomnia 52 . This high pleiotr opy is c on si st en t with t he lar ge and div er sified t r ansc r iptional r ole of AFF3. It s u g g es t s tha t an y pert urba t i o n of i t s e x pr es sion lev el m igh t be d eleter ious . Consist en t with this h ypothesis we iden t ify m ult iple mod es-of- ac t ion an d obser v e v arian t -s pecific/e xpr es sion level m odu la t ion of th e phenot y p e . Fir st ly , un timely (ov er )e x pr ess ion of KI NS SHIP v arian t s tha t a r e less s en sitiv e to SIA H r egula tion leads t o extr emely s ev er e phenot ypes in human, zebr af ish and r oden ts, e.g. homo zy gou s Ala233Thr kno ck-in l eads t o mouse lethality 11 . Sec ondly , C5 7BL/ 6N and CD1 gen eti c . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint bac k gr ound s mod ulat e the phenotype s pr esen t ed by Af f 3 mouse k noc k o uts 11,56,57 . Thir d ly , knoc k do w n and ov er e xpr ess ion of mouse A ff3 in d e r mal cells impair niche s wit chin g , whic h is r equir ed f or hair r ec onst itu tion 10 . F our thly , AFF3 ove r e x p r es sion in H e L a cells p e r tur bed th e dynamic s of the nu clear spe c k les 1 . O ur RN A -s eq e x per imen ts fur ther d e monst r a t e t ha t chang es in the a m ou n t a n d/or function of A FF 3 dr ama tic al ly alt e r tr anscript om e p r o f il es. W e show tha t th e e xpr ess ion of abou t one- thi r d of the A FF 3 t ar g ets ( bound l o c i) ar e dif f eren tial ly e xpr essed upon AFF3 modific a t ion and o bs erve a pr o gr es s ion in the e xten t of tr anscript om e al t era tions with t h os e link ed t o haplo insuf ficiency bei n g l ess dr as tic than tha t of h omo zy gou s L oF cells , whi ch in tu r n are les s impa ct ed than cells har boring homo zy gou s D N v arian t s .

In c onclu s ion, w e ar e adding t o a gr owing lis t of v ar i an t -spec ific neur odev el o pment al mechanisms and th ei r ass o cia t ed g enot y p e - phen o t y pe cor r elations 58-60,61 . W e demo nstr a t e t hat beside degr on v arian ts tha t impair t he degr ada t i on of t he enc oded pr otein 11 and downr egula tion d ue to pr om oter h yperm eth yla tion 17,18 , dy sr egulat ion o f AFF3 t hr ough g ene d uplic at ion, heter o zy gou s and bialleli c tr unc a ting var ian ts, bia llel ic missen se var ia n t s and c ompou nd het er o zy gou s tr unca t ing / missens e v arian ts ar e asso cia t ed with c ogniti v e i m pa irme n t . List of abbreviations ID = in tellectual d isabili t ies DD = development al dela y DN = dominan t- neg at i ve LoF = loss-of-fu nc t ion + = w ild-type allel e N H D = N-t er minal ho mology domain CHD = C-t erminal ho mology domains ALF = A F4-L AF4-FMR2 do main T AD = tr ans activ a tion dom a in N LS = nu clea r / n uc leolar loc aliza t i o n s equence SE C s = tr anscript ional s uper elong at io n c omple x es P-TE F = posi t ive t r ansc r iption elong a ti on f act or MLL T = m y eloid/l ymp hoid or mix ed-l i neag e leuk emia; t r ansloca t ed t o ELL = Elong a tion F ac t or f or RN A P olym er ase II ELL bow = ELL - binding domain pLI = p r obabili ty of being loss -o f-function i n t o ler an t pL OEUF = los s-of-fu nc t ion obser v ed/e xpect ed upper bound fr a ction PT U = 1-p hen yl 2- thiour ea K22-K26 = KI N SSHIP pa tien t s 22 t o 26 DUP = pa tien t wi th par t ial dupl ic at ion of AFF3 L1-L 1 4 = pa t ie n ts 1 t o 14 w it h los s-of-function v ariants . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint B1-B7 = pa tie n ts 1 t o 7 w it h bial l elic miss en se var ian ts AD HD = a t t ention d e ficit h yper a ct i vit y disor der EE G = Electr oencephalogr a p h y ASD = aut i sm spectrum d i sor der KD = knoc k-down Dp f = da y s post f ertil iza tion IOD = in t er-oc u l ar dist a n c e HW = head wi d th Un = U n t r e at e d M = mock W t = w ild- type MA F = minor alle l e f requency DE G = dif f eren ti ally e xpr ess ed g enes MT A G-asso cia t ed = as so ciat ed by multi-tr ai t analy s is of G W AS BMI = bod y mas s ind ex Ethics approval, consent to participate and consent for publication Inf or med co ns en t f orms w er e obt aine d f or all aff ec t ed i n div idua ls or t heir g uar dians participa ting in this stu dy . The curr en t st udy was a ppr ov ed by the CER ( “ C ommission d'é thiqu e de la r e cher c he”) of the c ant on of V aud (Pr ot oc ol n umb er: CER -VD 2 021-0140 0). Thi s r es ear ch c omplies with the principles o f the Declar a tion of Helsinki. Availability of data and materials The HEK293 T and fibr oblas t RNA-seq r eads ar e deposit ed in GE O un der ac ce s s ion GSE2 41621 (t ok en ot ezqc sslb ctby x) and GSE24 65 54 (t ok en wxibc y qat dyphkl). Conflicts of In terests Annabelle T utt le, Hou da Zghal Elloumi and C haof an Zhang ar e em ploy ees of G eneD x and D es ir ee DeM ill e w o rk s f or ARUP Labor a t o rie s . Jame s R. L up ski ha s st oc k owner s hip in 23and Me and is a paid c onsult an t f or G enom e In ter na t ional. C la u dia M.B. Car v alho p r ov ide s c on s ult ing ser vi c e f or Ionis P harma ceutic als . The ot her author s ha v e no c ompeting in ter est s t o d eclar e. Authors contributions SB , JC, N V , B Y en g ineer ed and phenot y ped anim a l and cell mod els. GA, F S and CI analy zed tr an script o mes pr ofiles . NG 3D-mod el ed mis sense v ar i an t s . A B , FS, L T , S S, R A J , J -U S, D D , PB- T , GR N, KN W , L D , MM, C G , LEL M V , R P , R K , HY , GÅM H, CJ , M F S, K MB , M JL , C MB C, C Z, JR L , LP , LF- G, RM- T , F P , A T , HZE, LM, MK, OK, JH , M S , MI, F O , FZ , K W , AM , M KH, PP and HA c ol le c t ed cli n i c al inf or m a tio n . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint and g enom i c D NAs , s equen c ed and analy zed e x ome s and/ or g enomes . AR and SB c on ceiv ed the st udy and wr ot e th e m anus cript . All o ther author s c omment ed o n the man us cript. Acknowledgmen ts and Fundings W e thank J a c qu es B e c kmann for comments. Th is wor k wa s s upp orted b y grant s from the Sw i s s N at ional Sc ien ce Foundation (31003A_182632 and I ZST Z0_21 6615 t o AR), t he Leje u ne Foundati on (#1838-201 9A to AR) , t h e Blac ks wan Foun dation (t o A R ), a P RIN 20 20 gr an t f r om the It al ia n Minis try of U niv er sities and R e s ear ch (20203P 8C3 X t o AB ), and the US National Institu tes o f Health (NS1 05078 and H G01175 8 t o J RL ) . This s t ud y m akes us e of data gene rat e d by the D ECI PHER community. Fund i n g f or the D EC IPH ER project was p rovided by the Well co me Tr us t [grant number W T2 237 1 8/Z/21 / Z]. The fu nders had no ro l e in stu dy des i gn, data collection and analy si s , de cis ion to publish, or pr eparat ion o f the man us cript. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Bibliogr aphy 1. Melk o , M . e t a l . Funct i o nal c har ac t eriza t ion o f the AFF (AF4/ FM R2) f amily of RNA-bindin g pr ot eins : ins igh t s in t o the m olec ular pa th ology of FRAXE i n t ellec t ual disabil it y . Hum Mol Ge n e t 20 , 187 3-85 (2011). 2 . G uo, C . e t a l . The super elong a tion c omple x (SE C) m ediat es ph as e tr an s it i o n of SPT 5 during tr an script i o nal pause r elease. EMBO R ep 24, e55699 (2 023). 3. Luo , Z . e t a l . The super elong a tion com ple x f a m il y of RNA polymer ase II el ong a tion f act or s : g ene t ar g et specific it y and t r ans c r iptional outp ut. Mo l Cell B iol 32, 2608-17 (20 12). 4. Chen, Y . & Cr amer , P . St ructur e of the super -elong a tion c omple x subunit AFF4 C -t erm i n a l homology domain r eveals r equir em e n ts f or AFF hom o- and het er odimerizat ion. J Biol Che m 294, 10 663-1067 3 (20 19). 5. N ilson , I. e t a l . Ex o n/ in tr on s truct ur e of the hum a n AF-4 g ene, a m ember of the A F - 4/ L AF- 4/ FM R-2 g ene f amily cod i n g f or a nuclea r pr ot ei n wit h str uctur al alte r a tions in acut e leuk aemia. B r J H ae ma t ol 98, 157-69 (199 7). 6. Hou s e, C .M. e t a l . A binding motif f o r Siah ubiquiti n l ig a se. Pr oc N at l Acad Sci U S A 100, 3101-6 (2003 ). 7. Jonk er s, I. & Li s , J . T . Get ting u p t o sp eed with tr ans cription elong at ion by RNA p oly m er a s e II. N a t Re v Mo l C e ll Bi o l 16, 167- 77 (2015). 8. W an g , Y . e t a l . A per mis sive chr oma tin st a t e r eg u l a t ed by ZF P281- AFF3 i n c on tr oll ing the impri n t ed M eg3 poly cis tr on. N ucleic A c ids R es 45 , 11 77- 1185 (20 17). 9. T s u k umo , S.I. e t a l . AFF3 , a s u s cepti bility f ac t or f or au t oim mune dis ease s , is a molecular f acil it at or of im mun oglobulin clas s sw it ch r ec ombination . Sc i A d v 8 , eabq0008 (2022) . 10. T ak eo , M . e t a l . C yclic a l d ermal micr o- nic he swit chin g gov erns the m orph ologic al infr adian rh yth m of mouse zigz a g hair . Nat Co mm un 14, 447 8 (2023). 11. V oisi n , N. e t a l . V aria n t s i n the d egr on of A FF 3 ar e a sso c ia t e d with i n t e l lec t ual disabili t y , mesomelic dy splasia, ho r seshoe kid ney , and e p i lept ic encephalopa th y . A m J H um G enet 108, 85 7-873 (2021 ). 12. V eitia, R. A . AFF3: a new pla y er in main t ain ing XIST mo noall elic e xpr essi o n. J Mo l Cell Bi o l 11, 723-72 4 (2019 ). 13. Zhan g , Y . e t a l . AFF3-D N A meth yla tion in t erp la y in main t aining t he mono-al lelic e xpr ess ion pa t t e r n of XIST in t erm i n a lly diff er en t i a t ed cells . J M o l C e l l B io l 11, 761- 769 (2019). 14. Luo , Z. e t a l . R egula tion of t he im pr i n t ed Dlk1-D io3 locus by allele-spec if i c enhancer activi t y . Gene s D ev 30, 92-101 ( 201 6). 15. Inoue, Y . e t a l . Thr ee KI N SSHI P s yndr o me pa ti en ts with m os aic and g ermline A FF 3 var i an t s . Clin G enet 103 , 5 90- 595 (202 3). 16. K ar c zew s ki , K . J . e t a l . The mut ation al con str ain t s pect rum quant i fied from v ari a tion i n 141,45 6 humans. Na t u r e 581 , 434-44 3 ( 2020). 17. Jadha v , B . e t a l . A GCC r epeat e x pansion in AFF3 is a s ignific an t c au se of in t e l lectu a l dis abilit y . medRxiv (2023). 18. Met s u, S . e t a l . F R A2A i s a C GG r epea t e x p ans ion asso ci a t ed with silenci ng of AFF3. PLoS Ge n e t 10 , e 1 004 242 (201 4). 19. Ma t t io li, F . e t a l . Bi alle lic tr unc ati on v aria n ts in A T P 9A ar e a ssoci a t ed with a nove l aut osomal r eces siv e neur odev elopm en t al di sor d er . NP J Genom Med 6 , 9 4 (2021). 20. Alf aiz, A. A. e t a l . TBC1D 7 mut a tion s ar e a s soc ia t ed wi t h in t el lectual disabil i t y , m acr o c r ania , pa t el lar disloca t i on, and c eliac dis ea s e. Hum Mutat 35, 447- 51 (2014). . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint 21. G o ujon, M. e t a l . A new b ioinf ormat i cs analy s i s t ools fr amew ork at EM BL- EBI. Nucl eic Acids Re s 38 , W695-9 (201 0) . 22. Siever s , F . e t a l . F a st , s c alable g e ner a tion of high- qual ity pr o t ein m ult iple s equ enc e alig n men ts us ing Clu s t al Omeg a. Mol Sys t Biol 7 , 539 (2011). 23. G u ex, N . & P eits c h, M . C . SWI SS-M ODEL and the S wi s s-P dbView er: an en vir onmen t f o r c ompar a tiv e p r ot ein mod eling. Elec t rophor e sis 18, 27 1 4-23 (1997). 24. La b un, K. e t a l . CH OP CH OP v3 : e x p an ding the CRISPR w eb t oolbo x bey ond g enome edit ing. N u c leic Ac ids R e s 47 , W171-W 174 (20 19). 25. Mon t ague, T . G ., Cru z, J .M., Ga gno n, J . A., Chur ch, G .M . & V alen, E. CH O P CH OP: a CRISPR/ Ca s 9 and T ALE N w eb t ool f or g enome editing. Nucleic Acids R e s 42 , W 4 01-7 (201 4). 26. Bassani, S. e t a l . V arian t s in USP4 8 enco ding ubiquitin h y dr olase ar e as s ocia t ed wit h aut osomal domina n t non - syndr omic her ed i t ary hearing l o s s. Hum Mol Ge net 30 , 178 5- 1796 (2021). 27. Sc h i n deli n , J . e t a l . Fiji: an open-sour c e pla tf or m f o r biological-i m a g e analy sis. Nat Met hods 9 , 676-82 (20 12) . 28. Dob i n , A . et al. ST AR : ultr af as t univ er sal R N A -s eq aligner . Bioi nf or ma tics 29, 15- 21 (2013). 29. Li ao , Y ., Sm yth, G. K. & Shi, W . f ea t u r eCoun t s : an e f fici en t g ener al purpose pr ogr am f or assignin g s equen ce r eads t o g enomic f ea t ur es . Bioin f or mat ics 30 , 923-30 (2 014). 30. Love, M .I., Hu ber , W . & Ander s , S. M oder a t ed est i m ation of f old chang e a nd disper si o n f or RNA-seq da t a with DE Seq2. Genome Biol 15, 550 (2014). 31. Hub e r , W . e t a l . Or che s tr a ting high-thr oughput g enomic an a ly si s with Biocon duct or . Nat Met hods 12, 1 15-21 ( 2015). 32. W u, T . e t a l . clu st erPr o file r 4.0: A univ er sal enr i chmen t t oo l f or int erp r eting om i cs da t a . Innovation (Camb) 2 , 100141 (2021). 33. Y u, G . , W an g , L . G ., H an, Y . & H e, Q . Y . clust erPr of il er : an R p a c k ag e f or com paring biologic a l themes among g ene clust er s . OMICS 16, 284-7 (2 0 12). 34. Subr amanian, A. e t a l . G ene set en richmen t analy s is: a kno wledg e-bas ed appr oach f or in t e rp r et ing g enome-wide e xpr ess ion pr o files . Proc N a tl A c ad Sci U S A 102, 15 545-50 (2005). 35. Li b erzo n, A. e t a l . Molecular signa tures da t abase (M S igD B) 3.0. Bio inf ormat i cs 27, 1739-40 (2011). 36. Li b erzo n, A. e t a l . The M olec ular Signa tur es Da t abase (M Si gD B) hal lm a r k g ene set c oll ection . Cel l S yst 1 , 417-425 (201 5). 37. Zhan g , Y . e t a l . M odel-bas ed analy si s of ChI P-Seq (MA CS). G enome Biol 9 , R137 (20 08). 38. Y u, G . , W an g , L . G. & He, Q . Y . C hI P s eek er: a n R/ B i o condu c t or pack a g e f or ChIP pea k annot a ti o n, com paris o n a n d v isualiza tion . B i oi nf o r ma t i cs 31, 238 2-3 (2015 ) . 39. Deciphering D ev el o pment al Disor der s , S. Pr evalence and ar chit ectur e o f d e nov o mut at ions in dev el o p ment al disor der s. Nature 542 , 433-4 38 ( 2017). 40. Sobr eir a, N., S chiet t ec a t t e, F ., V all e, D . & H a m os h, A. G eneM a t cher: a ma t c hing t ool f o r c onnecting in ves tig a t or s with an in t e r est in t he s ame g ene. Hu m M u ta t 36, 928-30 (2015) . 41. Firth, H . V . e t a l . DE CI PH ER: Da t abase of Chr omosomal Imbalance and P hen otype in Hu mans U sing En s em bl R es our c e s. Am J H um Gene t 84 , 524-33 (2 009). 42. Jacquemon t , S. e t a l . A hig h er mut at i onal bur den in f ema les s upp orts a "f emale pr o t ec t i v e model" in neurodev el opment al disord er s. A m J Hum G enet 94, 4 15-2 5 ( 20 14). 43. Har ripaul, R. e t a l . Mapping aut o s omal r ece ss iv e in t ellectual dis ability: c ombine d micr oarr a y and e x ome sequencin g iden tifi es 26 nov el c a n didat e g enes in 192 c ons an guineou s f amilies. Mol P sychia try 23, 973-9 84 ( 20 18). . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint 4 4 . T an g, D . e t a l . S t ructur al and functi on a l insi gh t i n t o t h e ef f ec t o f A FF 4 dimeriza tion on activa t ion o f HIV -1 pr ovir a l tr ans cr i p tion. Cell Discov 6 , 7 (20 20). 45. Qi , S. e t a l . Structur al ba s i s f or ELL2 a nd AFF4 activa t ion of H IV-1 pr ov i r al t r ans cription. Nat Commun 8 , 14076 ( 2 017). 46. K alue f f , A. V . e t a l . T ow ar d s a c ompr ehensi v e c a t alog of zebr af i sh beh avior 1.0 and bey ond. Z ebrafish 10, 7 0-86 (2013) . 47. Sill ar , K. T . Mauthn er cel ls. Curr Biol 19, R353-5 (2009) . 48. Sk a r ne s , W . C. e t a l . A c onditional kno ck out r esour ce f or the g enom e-wide st udy of m ous e g ene function. Natur e 474 , 33 7 -42 (2011). 49. Khan, H . e t a l . A nov e l v a r ia n t in A FF 3 underlying isolat ed synd act yl y . C l i n G en et 103, 341- 345 (2023). 50. N ieder rit e r , A. R. e t a l . In viv o model i ng of th e morbid human g enome us i ng Danio r erio. J Vis Ex p , e50338 (2 0 13). 51. Ioannidis, N .M . e t a l . REV EL : A n En s e m ble Metho d f or P r edicting the P a t h og enic it y of Rar e Mis s ense V arian ts. Am J Hum Genet 99 , 877-885 (2 016) . 52. Sollis , E . e t a l . The NH GRI-EBI GW A S Ca t alog: kno w ledg eba se a n d depo s ition r es our c e. N u c leic Ac ids R e s 51 , D 977-D 98 5 (202 3). 53. T elley , L . e t a l . Se q uen tial tr anscrip tional w a v es dir ec t th e di f f e r entiati on of newbor n neur on s in the mo us e neoc ort e x . Scienc e 351 , 1443- 6 (20 16). 54. Car doso-M or eir a, M . e t a l . G en e expr ession acr os s mamm a lian or g a n deve lopm ent. Nature 571, 50 5-509 (2019 ). 55. Moor e, J .M. et a l . L af 4/ Af f 3, a gen e i n volv ed in int ell ectual disability , i s r e quir ed f or cellula r migr a tion in t he mouse cer ebr al cor tex. PLoS One 9 , e10 5933 (20 14). 56. K r aft , K . e t a l . D eletions , In v er si o n s , Duplic a tions: Engineering of Str uctur al V ar ian ts usi n g CRISPR/ Ca s in M ic e. Cell R ep 10, 8 33-839 (2015 ). 57. Birling , M . C. e t a l . A r es o ur ce of t ar g et ed m ut ant mouse lines f or 5,0 61 g enes . Nat G enet 53, 416-41 9 (2021 ). 58. den H oed, J . e t a l . M ut a tion - s peci fic pa thoph y siological m e c hanism s de fine di f f er e nt neur od evelop m ent al disor der s assoc i a t ed with SA TB1 dy s function. A m J H u m G e n e t 108, 346-356 (20 21). 59. O' D onnell-Luria, A. H . e t a l . H et e r o zy gous V arian t s in KM T 2E Cau se a Spectrum of N eu r odev elopment al D i sor der s and Ep i leps y . A m J Hum G enet 104 , 1210- 1 222 (2019). 60. Haijes, H .A. e t a l . De N ov o Het er o zy gou s POLR2A V arian ts C a u s e a Neur odev elopm en t a l S yndr ome with Pr of ound Inf a n t ile-O ns et H ypot onia. Am J H um Genet 105 , 283-301 ( 2019). 61. P aul, M . S . e t a l . Rar e EIF4A2 v a r ian t s ar e as soc ia t ed with a neur odev el o pmen t a l dis o r der char a ct erized by in t ellectual d isabilit y , h ypot onia, and epi leps y . A m J H u m G e n e t 110 , 548 (2023). . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Caption Figure 1. AFF3 va r i a nts . (A) Schema tic pr ot ei n s tru c t ur e of AFF3 ( N M _0022 85.3) wi t h it s N-t er m inal h om ology domai n (NHD , c y an) , t he AF4-LAF4-FM R2 (A L F , pink) dom a in 2,3,5 c on t a ining the degr on mot if , a Serine-r i ch tr an s a ctiv a tion do main (T A D , y el low) 6 , a bipart i t e nuc lear / nucleolar loc a l iza tion s eq uenc e (NL S , gr een), a n d the C-t erm ina l homol ogy dom ai n (CH D , blue) 4 showing po si t ioning of all AF F 3 c o di ng v arian ts mention ed in th e t e x t. M is s ense and tr unc at ing v ariants ar e shown abo ve, while e xten t s of m icr odeletions and m icr od uplica t ion s ar e dep i ct ed below the str uctu r e with con tinuous and das h ed lines, r e s pectiv ely . The v arian ts ar e c olor coded: lo ss-of- function (trunc a tion and deletion) in r ed, bial lelic mis sense out s ide the degr on in blue and KIN SSH IP-a ssoci at ed missen s e var ia n t s , deletion and dup lica t i on in pur p le. The de novo missen s e iden t if ied in in dividuals M1 and M2 ar e s h own in gr een. While t he p.(A r g947Pr o ) sho wn in black w a s sh own t o seg r e g at e with i sola t e d s ynda ctyl y 49 ., we ha v e als o ident if ied it in indi viduals with n o d i git a b norm alities. (B) UCSC g enom e br ow ser sna p s ho t of th e C hr2 99.5 t o 101.4 Mb r egion showin g the g enes mapping t o t his in terv a l. The e xt en t o f th e d uplica ti o n identif i ed in the D U P1 ind i vidual t ha t enc o mpas s es e x on 10 t o e x on 24 of AFF3 and e x on 1 t o 3 of REV 1 is indica t ed by the blac k b a r and t he ligh t b l u e s h a d ow . (C) Ex a m ples of pedigr ees of t r ansmit ting af f ect ed f ami lies. Figure 2. aff3 knock ed-down zebr afish larvae display alter ed behavior and morphological anomalies. The c ondit ions analyzed ar e t he f ol l owing: U n t r ea ted (Un), M ock-injec t ed ( M), and af f 3 Kn oc k - Down (KD ). (A) Pr opor tions of normal and developm e n t al ly de f ective 5 dp f emb ry os. In 10% of aff3 KD zebr afish we iden t ifi ed sever al mor p hologic al ano m a lies such as : head malf ormat ions , belly and hear t ed e m a, sk elet o n-mu s c u l ar dy smorpho l o gi es, and alt er a ti on of ey e pig m ent ation . (B) A lc ian-b l ue s t aining a t 5 dp f r evealed jaw malf ormat ion in 57% of aff3 KD zebr afish . (C) Visuali za tion of m o rphologic al Int er- O cula r D ist ance (IO D ) an d Head W idt h (H W ) m easur emen t s fr om dor sov en tr al zebr afis h imag e. Quan tific ation o f IOD (D ) and H W (E) indic a t es a si gnifica t iv e decr ease i n I O D and HW in aff3 K D la r v ae; p *<0.04; p ** < 0.004 9. (F) T ou ch t est r espons e as s ay a t 3 dp f . Upon a t ouch s timulus, w e cl a s s if i ed the lar v ae sw imm ing beha v ior i n: «nor mal s wimming», «pause», «looping swi m ming», «pin wheel swi m ming» or « motionl ess » d ue t o m a lf ormation s . (G ) S wi m ming gl o bal v e locity analysi s a t 5dp f in the dar k of U n, M , and af f 3 K D and aff3 KD c o-inject e d with hu man AFF3 ( hAFF3 ) mRNA w il d- type (Wt) or h a rbor i n g the ind i c ated mis sen se v arian t, i.e . . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint the KINS SHIP v arian ts Ala233Thr and V al235G l y or t he biall e lic v ariants iden tif ied in t his r e p or t in a health y (Gln1 79 G lu) o r af f ect ed in dividuals (L y s52 8Ar g and Thr 592Ser ); p **<0.0025; p ** ** < 0.0001. (H) P r opor tions of nor mal an d d e v elopmen t al ly de f ective 5 dp f e mbry os uninject ed ( U n), inject ed with w at er as c on tr ol ( H 2 0) or with 360ng of h AFF3 m R N A W t o r t h e i n d i c a t e d m i s s e n s e v arian t. Lar v ae w ere c a t alog ed as de s c r ibed: (i) normal ph eno t ype, ( i i) Class 1 with sk elet om us cular dy s mo rphology an d /or small dimension, (iii) Clas s 2 with a mor e s ev ere phenot y p e including a t least thr ee of the f ollowing char act eris tics : s k elet o muscular dy s m orphology , small dimensions, h ead malf or m at i ons, ey e s ’ alt er a tion, peric ar dial edema, an d la t er al b el ly edema or (iv ) dead. Inje ctions o f 180ng and 720n g of h AF F 3 mRNA show ed s imila r r esults. Figure 3. T ranscriptome profiles of engineered isog enic HEK293T cells (A) F our -w a y V enn diagr am of diff eren t i al ly e x pr es sed gen es (DE G s ) in bial lel ic l o s s-of function (LoF / LoF) AFF3 lines and biallelic dominan t n ega t i v e (D N /D N) AFF3 KI NSSHI P lines upon c omparison with unmu t at ed wildtype lines . D E G s c oun ts ar e st r a tifi ed in g e nes up- (UP) and down- r egula t ed ( DO WN). (B) V olc ano Plot s of DE Gs i n bial lelic loss -o f function ( LoF /LoF) AFF3 lines (le f t panel) and biallel ic do minan t neg a tive (DN/D N) AFF 3 KI NSSH IP l ines (r ig h t panel) upon c ompar is on with unm uta ted wild type lines. The t op 30 mos t significan t DE Gs in L oF /L oF t ha t ar e dy sr egula t e d in an oppo s it e manner in DN/D N ar e ind ic a t ed, t og ether w i t h some of the mos t d iff eren tiall y e xpr ess ed g enes (-log10( P adj)>20 a nd ab s(log2F oldChang e) > 0 .5 ). (C) G ene S et Enrichmen t Analy s is ( GSEA ) f or h al lmar k pa thw a ys of D E Gs in biall elic loss -o f function ( LoF /L oF) AF F 3 l i n e s ( l e ft panel) and biallel ic do minan t neg a tive (DN/D N) AFF 3 KI NSSH IP l ines (r i gh t panel) upon c ompar is on with unmut at ed wild type lines . (D ) F our -w a y V enn diag r am of di f f e r e n tial ly e x pr es sed g enes (D EG s) in bial lel ic loss-of function (L oF /LoF) AFF3 lines and biallelic dom inan t neg a ti v e (D N/ D N ) AFF3 K INSSHI P lines upon co mparis on with unmut at ed w ild t ype lines and AFF3 Ch IP- s eq peak s iden t ified in HEK293T cell s (HEK29 3T ) and in Mus mu sculus E S cells (mmE S). D E G s bound b y A FF 3 dis cu s s ed in t he t e xt ar e i n dica t e d . (E) E x a m p l e s o f D E G s NR C31 (t op) and D D X17 (bot t om) l oc i bound by AF F3. UCS C g eno me br o w ser s nap shot s ho w ing fr om t o t op to bo t t o m AFF3 ChI P -s eq HEK293 T r e sults, U CS C and REF Seq c u r a t ed gen e s truct ur e and v er t ebr at e Ph yloP c on ser v a tio n s c or es (le ft panels ). Expr ess ion level of N R C31 (t op) and DD X17 (b ott om) in +/ + (blue), L oF /L o F (y ellow) and DN/ D N ( g r een) HEK2 93T engineer ed li nes (r igh t panels ) . . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Supplementary Figures Supplementary Figure S1. 3D protein modelling of the human AFF3 degron region bound to SIAH ubiquitin ligase. The A FF 3 degr on ch ain is shown a s a whit e stick st ructur e, wi t h r esidues mut a t e d in KINSSH IP af f ec t ed indiv idu a ls highligh t ed in whit e s p a ce-f i ll, fr om top t o b o t tom : Met238, V al235, Ala233, and Pr o231. The SIA H ubiquit in lig as e is pr e sen t ed as an or ang e ribb on embedd e d in its gr een tr anspar en t surf ace. Am i no acids in t er acting with the degr on r esidues si d ec hain s ar e cl u st er ed in t wo s pa t i al r egions al o n g bet a s tr ands and r epr esen t ed spa c e-fi lled as f ollow: Ser15 4 y ell o w , Thr15 6 b ur gund y Leu15 8 salmon and A s p177 r ed, V al179 blue, Met180 c y an. Supplementary Figure S2. Immuno staining of hindbrain neurons and motoneur ons in 3dpf zebr afish. M a x im um pr ojections of c onf oc al imag es r eg ar ding hindbr ain neur onal st ructur es (A) and mot o neu r ons (B) in 3dpf Mock-inject ed (Mock) and aff 3 KD lar v ae. Supplementary Figure S3. AFF3 e xpression levels in engineer ed isogenic HEK293T cells. D o t b l o t o f t h e V S T - n o r m a l i ze d ( r - l o g ) c o u n t s o f t h e AF F 3 g ene acr o ss all HEK293T en gineer ed s am ples . W e engineer ed fiv e biall el ic LoF H EK2 93T lines (LoF /Lo F) with dif f er ent co mbina t ions of v arian ts (lines No.2 0 and 9 8: st op - g ain/ s t op-g ain; No.15: s t op-g ain/20 bp deletion; No.4: 4bp deletion/114bp deletion; N o.216: 94 b p deletion/94bp delet ion), o ne hete r o zy gous LoF st op-g ain/ + line (No.1), tw o homo zy gou s Ala233Thr/ Ala233Thr K I NS SH IP /KI NS S H IP l i nes ( N o.54 and 90) and tw o com pound he t er o zy gous KI NSSH IP a n d LoF lines ( N o.51 and 86: A la233Thr / st o p-g ai n ). The v arian t no mencla tur e of the engineer ed var ia n ts is s pecified in the m a t er ial s and m ethods s e ction. Supplementary Figure S4. Examples of DE Gs loci bound by AFF3 U C SC gen ome br ow ser sna p s ho t of th e CDK5R AP2 , C N N M 2, C T NNA3 and D Y N C2I 2 loci a bound b y AFF3 showing fr om t o top t o bo t t om AFF3 ChI P-seq H E K293 T r esult s, U CSC and REF Seq cur a t e d g ene s tr uc t ur e and ver t ebr at e Ph yloP c on s erva t ion sc or es (le ft panels). Expr es s ion level of th e c orr esponding D E Gs in +/+ (blue ), L o F /L oF ( yellow) a n d DN/ D N (gr een) H EK293 T engineer ed lines (righ t panels ). . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Supplementary Figure S5. Gene Set E nrichmen t A nalysis G en e Set Enri chmen t An a ly sis (G SE A) f or hallmark pa thw a y s of DE Gs (A) in biallel i c los s - of function (LoF / LoF) AFF3 lines (le f t panel) and het er o zy got e loss-of fu nc t ion (Lof/+) AFF3 l i n e s ( r i g h t p a n e l ) upon co mparison with unmu t a t ed w ildt ype lines; (B) in biallel ic l o ss-of function (LoF / LoF) A FF3 lines (le f t panel ) and f ibr ob l as ts of pr ob a n ds (f i b r oblast s ; r igh t panel ) upon c ompar i son with unmut at ed wild type lines and fibr obl ast s of co n tr ols, r espe ctiv ely ; and (C) in het er o zy gous lo ss-of function (L o F /+) AF F 3 l ines (lef t pane l) an d fib r ob l as ts of pr obands (f i b r ob las ts ; righ t panel) upo n c omparison with unmut at ed wild type lines and fibr oblast s of c on t r o ls, r es p ectiv ely . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figures Figure 1 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figure 2 Inter-Ocular Distance Head Width . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figure 3 1 mm . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Supplementary Figures Figure S1 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figure S2 A B Figure S3 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint 7.5 8.0 8.5 9.0 9.5 Wt1 Wt2 Wt4 1 4 15 20 98 216 54 90 51 86 Clones VST−normalized counts +/+ DN/DN DN/LoF LoF/+ LoF/LoF AFF3 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figure S4 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint Figure S5 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.14.24301100doi: medRxiv preprint