ZmSWEET Sucrose transporters expressed in the endosperm adjacent to the maize embryo are necessary for carbon partitioning and embryo growth

S eed s pl ay a cen tral rol e in the gl obal f ood system , servin g a s a major sourc e o f calo rie s through both dir ect h uman c on s umption and indi rec t u s e i n anima l f eed . Thi s e s s e n tial c o nt ribu tio n to g lobal e n ergy int ak e ste m s from t he fact t ha t s e e d s s to re r e serve s, p rimarily c arb ohydrat e s, p rot ein s, a nd lipids . S e ed a r e comp l ex b iologic al sy st em forme d of th re e ma in t issue s : the emb ryo, t h e endo spe rm and m atern al ti s s ue s. The embryo i s fully su r round e d by th e e ndo spe rm, and bo t h of t h e m a re fur the r e nc lo sed w ithin t he ma terna l ti s s u e s (L afon - Pla ce tt e and Kö h ler , 2014; Widie z et a l ., 2017 ; D oll and Ingram, 2022). Pl a n t spec ie s exhibi t sub s tan tial v ariation in bo t h see d s i ze and the c omposi tion of th es e storage pr o d uct s (S re eni va s ulu a nd Wobu s , 2013 ). I n m aize f or exampl e , th e end o sperm r e ma ins p er si s t ent and s e rve s a s a sta r c h re servoi r, while the embryo a cc umulate s high con cen tra tion s o f oil ( Doch ler t , 19 90 ). Thus, se ed s re pr e sent a he tero tr o p hic “sink ” ti ssu e .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint a nd r ely on a n impor tan t orga nic ca r b on sup ply from t h e ph oto auto tr op hi c sour ce ti s s ue s, e sp ec ially in maiz e (P alm er et a l. , 1973). Suc ro s e i s the pr i mary photos y n tha te t ra ns por ted predominan tly f rom matu re p hoto s yn t h e t i c l e ave s v ia t h e v a s c ula r sys tem to non -pho to s y nth etic s ink ti s sues , where i t s u ppo r t s th eir g r o wth and develo pm ent (G i fford a nd Ev a ns, 1981; Ko ch, 20 04; Ruan , 2014) . Upon arriva l a t t h e p r ox imity of d ev elo ping se ed s, su cro s e e xit s the v a s c ular s y s t em at t he phlo e m termin u s throug h a pr o c e ss kn own a s phl oem unl oading (Th o r ne , 1985; Pat rick and Offl e r, 2 001; Br aun, 2022 ) . Du e t o the organiz ation o f th e se ed, th e mat ernal ti ssue s a re s y mpla s t i call y isol ate d fr o m the e nd o s pe r m cell s, m eaning tha t th ei r c ytopla sm s a re not c onne ct ed vi a pla smode sma tal por e (S t a dler et a l . , 20 05; Widi ez et a l . , 2 017). C on sequ ently , suga r c oming f ro m t h e mat ern al vasc ul ar t is su e s mu st g o thr o u gh the apopla s t and be l oaded w ithin t h e e ndo spe r m by crossin g pl asma membr ane s ( Pa tr i ck an d Of fler , 2001; Po vilu s and G ehring , 2 022). In m aize , the b a sal e ndo s p e r m tr a ns fer lay e r (BE TL), c on tain s ce ll typ es sp ecia lized in nu trien t tr an sport from th e moth er pla n t to the en do s pe r m, de s ig n ed f or nutrie n t up t a k e, and ca n be v iew ed toge the r with th e p lace nt o - c hal a z al c ells a s an impor tant i nt e rfac e r e gul ating the suga r en tr a nce from the spor o p hyte t o the se ed (Cho urey an d H u e r o s, 20 17; Pov ilu s and G ehring, 2022). Th is ma te r na l -t o-fi li al nu t ri tion tr an sf er i s ce ntr al to g r a in dev el opment a nd yield, a s i llu strate d by the sever e s hrinkag e o f maize k e r ne l s w hen nu tri tion -rel ate d g e n e s a r e m u ta te d (K a n g et al . , 2009 ; S osso et a l. , 2015; Ya ng et al., 2022 ). S i milar apopl a stic sug a r loa di ng should occ ur a t t h e endo spe rm- embryo inter fac e (Doll a n d Ingram, 2 022) . The nutr iti ona l import anc e o f maiz e e ndo s p e r m-e mb r y o inte r f a ce s is in direc tly highl ighted by the expre s sion o f sug ar tr a nspor ter s a s well a s a n i nverta se in hibi tor in bo th t he embryo- sur r ou nding r e gion of the end os perm ( ES R, 4 - 1 0 DA P ) a nd l at er i n d evelo pmen t (9 -20 DA P) in the endo sp e rm adja cent to s c utell um (E A S ) in te rf a c e (Ba te et al . , 2004; Doll et al. , 2 020; Shen et al. , 2022 ). S o fa r, our un der s ta nd ing o f th e mec ha nism s un derly ing ap opla smic pa thw ays fo r c arbohyd r a te s t ran sport a cro ss ma ize f i li al i nt e rfa c es (embryo /end o sperm ) remai ns par tial . E fficie nt c arb ohydra te pa r tit ioni ng require s th e sp ecific ac t i vity of me mb r a n e suga r tra n s por ter s. Suga r tran spo rte rs a re e s s e n t i al com p onen t s o f bo th in tr ace llula r t ra ffick ing a nd a p opl as tic t ran spo r t pathw ay s and form thr ee k ey f amilie s i n plan ts : MS Ts (mono sac chari de tr a n s p ort er s), SU T s ( sucro se t ran sporte r s) an d S WEETs (Suga rs Wi ll Ev entu ally be Ex ported Tra n s po rter s) (E o m et al. , 201 5; Kell e r a nd N e uhau s, 2025 ). MST s a nd SUTs b e long to the major faci li tat or supe rf amily ( MF S) , have t w elve tran sm embrane d oma in s, a nd mo st c harac t e r i z ed memb ers func tion as acti v e t ran sp ort sy st em s u sing th e dr ivin g fo r c e o f the H + AT Pa se pump (L alonde et al. , 2 004 ; Niño - Gonz á l ez et al . , 201 9; Geige r, 20 20) . SWEET s ar e a cla ss o f unipo rt er s c har ac t e r ize d by 7 t ra ns m e m b r a ne dom a i ns t h at tr a ns por t h e x os e a n d s u c ros e a n d me d i at e s u g a r i nf l u x or ef f l u x al o n g a co n ce n t r at i on g r a d i e nt ( C hen et a l. , 2 015 a ; Anja li et al. , 2020). In Ar a bid op sis t h alia na, a c a s c a d e o f seque nt i ally expr es sed S WEET s ucro se tr ans p o r t er s i n th e s eed c oat and end o s p e rm provi de s n utrition f or th e em b r y o . The at s w ee t11;12 ;15 tr i pl e mut ant ha s a wrinkl ed s eed phen o t y pe, w i th acc umulatio n o f starc h in th e .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint se ed coa t a n d del ay ed embryon ic de vel o pmen t (Ch e n et a l . , 2 015 b ) . Cruci al role of S W E ETs in seed phy s i olo gy h as be en demon s t ra ted in sev eral c ro ps . I n soybe a n ( Glyc ine max ), G mS WE ET15a and GmSWEET15 b are e ss ential for early s tag e s o f se ed dev elo pme nt by media t in g s uc ro se e xport fr o m the e ndo s p e r m tow ar ds th e e mbryo. Knoc k-o ut o f bo th gene s ca u se s r eta rded embryo g r ow th, endo spe rm per s i ste nce and se ed abo rtio n (Wang et a l . , 201 9 ). In t o ma to ( Solan um lyco pers i cu m) , kn ock- out of Sl SWEET15 c aus ed de fec t s in s eed fill ing a nd embryo d e velop men t, s ug ge st i ng th at SlS W E ET15 i s re quire d fo r sucro s e unl oadin g fro m t h e s ee d coa t to the g r ow ing embryo (Ko et al . , 2 0 21) . In r i ce ( Oryza sativ a ) , O sS WEET11 , O sS WEET14 a nd O sSWEET15 ha v e b een shown to m e diat e sucro se e f flux i n devel oping c a ryops e s and pla y ke y r ole s in grain filling. o s swe e t11;14 an d o sswe et11;15 do ubl e mut ant s b ot h e xhibited redu c ed grai n weig ht a nd a bn ormal st arch ac cumul a tion i n the p e r i car p (Yang et a l. , 2018 ; F ei et al. , 202 1; H u et a l. , 2023 ). In mai ze, the B ET L ex pr e ssed ZmS W EE T4c i s re spon sible fo r t ran sfe rring h exo se s ( g lu cose an d fr uc t o s e ) o f de velo pin g kern els from the ma te rnal ti ssu e s t o t h e e nd os pe r m . K no ck - o ut m u t at i o n of Zm SWEET4c g e n e r a t e s a d r a m a t i c r e d u c t i o n o f e n d o s p e r m s i z e a n d k ernel w eight (So s so et al ., 2015 ). In a dditi on to ZmS WEET t ran spo rter, mu t a t i on o f mai ze suga r NP F tran spo rte r gene S ucr o se and G luc o se Ca rrier 1 (ZmS UGC AR1/ZmN PF7.9 ), whic h is exp r e s s ed in BE TL i nt e rfa ce , a l s o c a u s e i m p a i r m e n t o f e n d o s p e r m e n l a r g e m e n t r e s u l t i n g i n p o o r l y f i l l e d k e r n e l s w i t h e m p t y p e r i c ar p ( Y a n g et al . , 2022) . T he se da ta on m aize ill u s tr ate th e key imp orta nce of BET L regio n f or sugar en try in to the maiz e se ed ( D ol l et al. , 2017 ). Our p r e viou s w ork id enti fi ed a novel e nd os p erm ce l l t y pe adjac en t to the em bryo, te r me d th e Endo sp erm Adj acen t to S cu tellu m (EAS ) ( Doll et a l., 2020 ). EAS c ell laye rs rep re sen t t he l a r g e s t end os perm sur f a ce in di r ec t c on t ac t w i t h t h e e m br y o a n d a r e p r o g r es s i ve l y el i m i n a te d t h r ou gh c e ll d e a t h , t he r e b y c r e at i n g s pac e f or e mbryo growth ( Doll et a l . , 202 5). Nota bly, th e E A S cel l s e xhibi t e nr i che d ex pressi on of thr ee S WEET tran spo rte rs (ZmSWEE T14a, Z mSWEET1 4b, and ZmS WEET15a ) (Doll et a l. , 2020 ), provi ding a n oppo r tuni ty t o i nvest igate th e r ole of suga r tran sp or t a t t h i s en d osperm –e mbr y o i nte rf ace . In th e pr e sen t s tu dy, we d emon st r ate tha t t h e se S WEETs pl ay a c ritical rol e in kern el de velopm e nt, embryo g r o wth and ca rbo n p art i tion ing w ithin th e ma iz e ke rnel . Re su lts T hr e e plasma - memb rane loc aliz e d cla d e III SWEET sucrose t ranspor t e r s s ho w pr e f e r entia l express ion in the E A S endosperm s ub-doma in. P r o tei n sequenc es of th e S WEET fam ily t ran spor t e rs fr om maize, ric e and A rabi do ps i s w ere u s e d to con stru ct a p hyloge netic tre e ( Figu re 1A ). The thr ee Z mSWEET enco ding g ene s ( Z m SW EET 1 4 a = Zm00 001eb11 3080; Z mSWEET1 4b = Zm000 01eb170 150; Z mSWEET15 a = Zm000 01eb180 830), pr e v iously i den ti fied a s s how ing e n r i c h e d e x p r e s s i o n i n t h e E A S a t 1 3 D A P ( D o l l et a l . , 2020) we r e f o und to bel on g to c lade III of SWEET fami ly .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint ( Fi g u r e 1A ) , as pr e v i o us l y r ep or t ed ( W a ng et al . , 2024). Compa r a t i ve ex pressio n pro file s reve aled tha t Z mSWEET 15a is mo re stro ngly expre s s e d in t he E AS c ompa red to Z mSWEET14 a and Zm SWEET14b ( D o ll et al . , 2 020) ( Supplemental figure S1A). Ex pr e ssion pr ofil e s a cro ss dif f eren t ma iz e ti ssue s als o show expre s s ion o f Z mSWEET 14a a nd Zm SWEET14b i n pr i m ary roo t s du ring germin ation , a nd e xpres s ion o f Zm SWEET15a i n m eiotic t as s el s a nd l ea f ba s e s (Hoop e s et a l . , 201 9; Zhu et a l. , 2022 ) ( S upplemen tal figure S 1B ) . S i n c e c l a d e I I I o f S WEET fa mily a r e know n t o be inv olve d in sucro se t r a n s po rt (Eom et a l . , 2015 ), a ye a st h e t e r ol ogou s s y s t em w as u sed t o a s se ss w he the r Z mSWEET14 a, Z mSWEET1 4b a nd Z mSWEET1 5a are a ble to t r a n s po rt suc r o s e. E xpres s ion o f Z mSWEET1 4b or ZmSWEE T 15a al lows the yea s t mut ant st r ain SU S Y7, w hich en ables pheno typ ic rec ognition o f a s uc r o s e c a rr ie r a c tivity ( Ri es m e ier et a l. , 1992), to grow mo r e e f f i cient ly u s in g s uc ro se a s th e sol e ca rbon s o u r c e, in dica ting t hat Z mSWEET14 b and ZmSWEE T15a ca n t r a n s po rt suc ro se in thi s s y s te m ( Fi g u r e 1B ). C on si s ten t w i t h a r o l e in int erce llula r sucro se t ran sp ort , s ubc ellul a r l ocali z atio n an aly se s in m aiz e m es op h yll protopl a sts indic a ted th at Z mSWEET15 a:mCitri ne co -loc alize s with the pl a s ma m embran e marker p r ote in LTI6 b:mTurquoi se ( Fig ur e 1C ). E mbr y os o f t he zmsweet14a/14b/15a tr iple mutant are s maller and acc umulate less oil at matu rity T o inve s tiga t e the functio n of ZmSWE E T14a, Z mSWEET1 4b and ZmSWEET 15 a in kernel d evelop men t, a C RIS PR /Ca s9 -medi ate d g enome editi ng a pproach was u sed t o ind uce m ut a t i ons. Two sing le gui de RNA s (sgR NA s) w ere de signed t o targe t ful ly ide ntical s e qu ence s in e xon 4 o f Z mSWEET 14a a n d ZmSWEET 14b , w hil e tw o a ddi tional s g R N A s w e re de s ig ned t o ta rget Z mSWEET1 5a ( Fi g . 2 A -B ) . F o l l o w i n g m a i z e t r a n s f o r m a t i o n , g ene- sp eci f i c prime r pai r s ( Fi g . 2 A ) we r e u sed to d e t e c t s e quence a l tera tion s a t th e CR IS PR /Ca s9 targe t si t e s throug h S ang er s e qu enci ng. Thi s a n alys is r ev ealed sh ort i ns er t i on s and d ele t i o ns (indel s ), re sulting i n t hre e di s ti n ct m ut an t a l l e le s f o r b o t h ZmS W E ET14a a nd ZmS W E ET14b, a nd o ne mu tan t a llel e for ZmS WEET15a ( Fig . 2B ) . A ll i de n t if i e d a lle l es , e x ce pt z ms w eet 14 b _ 2 , are fram es hif t mut a ti on s tha t disru pt th e s econ d c on serve d MtN3 _slv ( PFA M ac ce s sion numbe r: PF0 3 083 ) family pr o tei n domai n an d cre ate prematu re st o p c odon s ( Fig . 2C ) . C r o s s e s w e r e p e r f o r m e d t o r e m o v e t h e C R I S P R / C a s 9 T - D N A c a s s e t t e a n d g e n e r a t e a zmswee t 1 4a _1/zmswe e t14b _3/ z m s weet15 a_1 t ripl e mu tan t su bse que ntly r e fe r r ed to a s zmswee t 1 4a / 1 4b / 15 a fo r simpli city. T o inve stiga te mo r phol ogic al ch a r a c teri st i c s of th e embryo and t o v isualize oil di s tr i bu tion in m atur e ke rne ls, n uc lear magne t i c re s o n ance i maging ( MRI ) w a s em ployed a s n on - d estruc tiv e and non -inva siv e t ec hniqu e to acq uire th ree -dime nsion al imag e s of intac t emb ryos f rom zmswee t 1 4a / 1 4b / 15 a muta n ts a nd WT ( Fig. 3). This an a lysi s reve a le d t ha t embryo s of t he zmswee t 1 4a / 1 4b / 15 a mu t a n t w ere sign i f i cantly (~ 16 % ) smal le r in v olume comp ared to WT, al thoug h thei r o verall s h ap e r em ain ed s im ilar ( Fi g . 3 A - D , I ). M ea s uri ng the ke rnel weig h t a lso rev eal e d a r educe d se ed we i g ht f o r z m s w eet 14 a/ 14 b/ 1 5a mu t a n t as comp a r e d to wi ld- t y pe ( Supplemental figu r e S2 - A ). St orage oi l s (tri ac ylgl yce r id e s ) c on s t itu te th e pr i ma ry dry ma s s com pone nt of the embry o and c an be visuali zed v ia MRI . Oi l d is t ribut ion s ho w ed s pa t i al het er o g enei ty in bo th WT and zm s w ee t14a /14b /15 a e m b r y o s ( Fi g . 3 E - H). The .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint hi g h e st l i p i d c on c e nt r at i o ns w er e c ons ist e n t l y o bs e r v e d i n t h e s c ut e l lu m , w i t h lo we r le v e ls i n th e r a d i cl e a nd p lumule region s . This pat tern is t y pica l for mai ze embryo s a nd was fou nd in b oth g eno type s. H o w eve r , the total oil c on t e n t p er embryo wa s sig ni f ican t ly red uc ed (by ~21% ) in t he m u t a nt c ompar ed t o WT ( fig . 3J ). Whe n c omp aring the ave r a ge lipid sign al pe r uni t volum e (th e rela tiv e conc e n tr a tio n o f oil wi t h in th e se ed v olume), no stat is tica lly s i gni fic ant di ff e renc e was fou nd be t w een geno typ e s ( S uppleme nt a l figure S2B, C ), w hic h c an be expla ined by a pr op o r tio na l redu c tion in b o t h emb ryo si z e an d total oil ac cumula t i on i n th e m ut a nt. Thu s, thi s indic a te s tha t th e re d uced oil c on t e n t in the zm s w eet14 a/14 b /15a m u t a nt embryo i s mai nly d ue to it s s ma ll er siz e r a t he r tha n a meta b olic def ect in oil ac cum ulati on in t h e e mbryo. Th e t r ip l e zmsweet14a/14b/15a mut a nt i s impa ired in ge rmina t ion and seedling es tablishment T o as se ss t h e germina tion c apac ity a nd seed ling d ev elopme n t o f th e zm s w e e t 1 4a/ 1 4b/ 1 5a tr ipl e muta nt, two d iffe re nt a s says were condu c ted. Fir s t , the rolled tow e l m e t h od (G onzal e s et al . , 2024) wa s u se d by plac in g k ernel s in rolle d-up g e r m inati on pap e r . P r im ary roo t l ength wa s mea s u red a t 6 a nd 9 day s af te r so wing (DAS ) ( Fi g . 4 A - B ). The zm s w eet14 a /14b /15a mutan t ex hibi ted a sig ni fican t reduc ti on in primary root length a t bot h time poi n t s c omp ared to WT ( Fig . 4B) . S econdl y, a hy dr o ponic a s s ay w a s pe r formed to eva lu ate mul tiple a s p e ct s of th e root sys tem arc hi tectu r e, a s well a s sho ot and r oo t f r e sh w eight ( Fi g . 4C -G ). At 11 DAS , zmswee t 1 4a / 1 4b / 15 a m ut a nt s s ho we d a s i g n if i ca n t r e du ct i o n i n bot h s h oo t ( ~2 7 %) a nd r o ot f r es h we i g ht (1 4%) as co mpa red to WT ( Fig . 4D). Addi tionall y, th e le ngth o f primary root s ( ~16 % i nhibition) a nd th e number o f seminal root s (~19% inhibi tion) , but n ot thei r averag e l ength , were s i gnific antly re duce d in t he zmswee t 1 4a / 1 4b / 15 a m u t a n t c o m p a r e d t o W T ( Fi g . 4 E- F). Tak en t oge t h e r , t he s e da ta sugge st th at th e zmswee t 1 4a / 1 4b / 15 a mut ation signi fi ca ntly impairs se edling ro ot devel o pment . C ar b on part i tioning is altered in zmsweet14a/14b/15a maize kernels Be cau se Z mSWEET1 4a , Zm SWEET14b an d Z mSWEET15 a enc ode sucro s e tra n s por ter s with enric he d ex pre ssio n i n maize k ernel s , th e in flue nc e o f their c o mbined lo s s o f func tion wa s us ed to in vestiga te th ei r r ole i n carb on p a r t i t i o n i n g d u r i n g k e r n e l d e v e l o p m e n t . T o t h i s e n d , t h e l a r g e s i z e o f t h e m a i z e k e r n e l s w a s l e v e r a g e d t o e xami ne the sp atia l di stribu tion of sucro se by applyi ng the F our i e r-tran s form in fra r e d (FTI R) pro toc ol ( G u end el e t a l., 2018 ). Sucro se di str ib u t i on wa s s u c ce ss fully v is ua lized in c r y o- sec tion ed ke r ne l s a t the e arly grain- fil lin g st age (13 da ys a f t e r poll ina tion, DA P ) a nd late r a t 17 DAP ( Fi g . 5 ). Analysi s of w ild-type (WT) k erne l s ec t io n s rev eale d a n apic al –ba sal sucro s e gradien t , w ith higher s u cro se a ccu mulati on in the bas al end o s pe rm c ompared to th e uppe r end o sperm ( F i g . 5 C , E ). S trong s uc ro se ac cum ula tion w as al s o ob s e r v ed in mat ernal tis su e s , no t a bly i n t h e pedic el r egio n wh ere v a s c ula r stra nd s termina t e and as s i milate unl oading occ urs , a t th e a pica l t i p o f the p eric arp , and in a nar ro w s t rip o f pe ricarp adja c ent to the e m br y o ( F i g . 5 C , E ). Compa r i s on b etwee n WT and zms w ee t14a /14b /15a mutant ke r n e l s sh owed tha t the ove rall sucro se di stribut ion pa tte rn .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint rema ined largel y uncha ng ed. How eve r , a tre nd t owa r d r educ ed s uc ro se acc u mula t io n w a s d etec t e d i n th e b asa l endo spe rm at both 13 and 1 7 DAP , and in the emb r y o at 13 DA P in the mu tant bac kground ( F i g . 5 D , F ). Giv en th e s e mi -quan tita tive na ture of F TIR ima ging, a s ta ble i s o tope trac i ng as s a y wa s dev el oped t o c omplem ent th e se ob se rvation s . This in volv ed feeding 1 3- D A P ke r ne l s w it h un i fo r ml y labele d ¹³C -suc ro se, in w hic h all c arbon atoms in th e s u cr o s e b a ckb one are 13 carbon ( 13 C) ( Fig. 6 A ). T he 13 C a ccu mulation of dif fer ent t i s s u e s w a s m e a s u r e d u s i n g i s o t o p e - r a t i o m a s s s p e c t r o m e t r y ( I R M S ) i n d i s s e c t e d k e r n e l c o m p a r t m e n t s : t h e e mbryo, and the ba s a l a nd up pe r part s of the e ndo s p e r m. The two endo s pe r m r e gions w ere de fine d by a h or i z on tal lin e d raw n a t the top of th e embryo: endo sp erm ti s s u e be low thi s line w a s c on s i d ered a s b a sal e ndo s p e r m, and ti s s u e ab ove a s upper e ndosp erm ( Fig . 6A ). Fig ur e 6 B s h ows 13 C qu an t if i c at io n i n t h e d if fe r e nt c o m p a r t m e n t s a f t e r 2 4 h , 4 8 h a n d 7 2 h o f f e e d i n g w i t h t h e ¹ ³ C - s u c r o s e s o l u t i o n . R e g a r d l e s s o f t h e g e n o t y p e , l onger incuba t i on time s with the ¹³ C - s u c rose soluti on led to gre at er 13 C acc u mula t io n in all c ompartm ent s an a l yz e d ( Fig . 6B ). Among th e c ompar t ment s , th e basa l e ndo s p e r m ac cumul at ed the hig he st leve l s of 13 C, fol lowed by the u p per endo sp erm and the emb ryo ( Fig . 6B). Th is o bs er v a t i on i s c ons ist e nt wi t h th e F T IR i maging , although the IRMS me thod me asu re s t otal ¹ ³C , wh ich c ould be pre s ent as ¹ ³C - sucro se but c o uld al so h ave be en me taboli zed in to v ario us c o mpound such a s s tarch, ami no aci ds or l ipid s fo r exa mple . Fo cu sing o n the c ompa ri son b etwe en th e WT and t he z ms w ee t 1 4a / 1 4b/ 1 5 a m u t a n t , t h e t r i p l e m u t a n t a c c u m u l a t e d l e s s 13 C, w ith diff ere nce s tha t bec ame signi fi cant af ter 72h o f incuba t i on with the ¹ ³C - sucro s e s ol u t i on . Impairmen t of 13 C acc umulatio n w a s found in all th ree comp a r tm en ts of th e zm s w ee t14a / 14b/15a mutan t, indic ating a g lobal disrupti on i n carb on a lloc a tion du e to the l o ss o f t he se s u cr ose tran spo rte r s in b ot h t he endo spe rm an d the embryo .

Apopla s tic s uga r t ran sport a nd develo p mental outcomes at the em br y o–endosp er m int e rfac e During s eed deve lopm ent, nut r i en t flux es mus t trave rse t w o s ym pla st i c barr i ers: be tween ma terna l ti s s ue s a nd th e e ndo s pe r m, and be t w een th e endo sp erm and t h e emb ryo (P a tr ic k a nd O ffl er , 2001) . In maize a popla stic po s t - phl oem t ran spo rt t o th e endo spe rm BETL ce lls i s fa cili tat ed b y a s ui t e o f tran spo rte r s and e nz y mes ( Chou rey a nd H u e r o s, 2017 ). B oth sucr os e and h exo se tran sp or t er s (S osso et al . , 2015 ; Ya ng et a l . , 2 022) and c ell wa ll invertas e s c onve r tin g suc r o se to hexo se s (Kang et al . , 2 0 0 9 ) a r e c r i t i c a l f o r s u g a r t r a n s f e r from ma te r n a l ti ssue s to th e end o spe rm a t thi s in te rfac e. By con tra st, mu ch le ss i s know n ab out th e m echa nis m s gove rnin g suga r tra n s po rt at the endo sp erm–emb ry o in ter fac e . Our work focu se s on th e l es s- c harac t e r i z ed maize endo sp erm- embry o inte rfa ce, ac r o s s w hich apopl a s tic su ga r tra n s por t al s o oc cur s ; t he .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint E ndosp er m Adjace n t t o t h e emb ryo S cut ell um (EAS ). P rev iou s tran sc r ip tomi c ana l yse s iden tifi ed ZmSWEET 14a, Z mSWEET 14b and Z mSWEET1 5a a s b e i n g s p e c i f i c a l l y e x p r e s s e d a t t h i s i n t e r f a c e ( D o l l et al. , 2020; Sh e n et al. , 2 022) providin g a s t rong r ation al e to ta r ge t t hem by gen ome e diting . W e s ho w that muta tion of t h e s e thr ee E AS-expr e sse d ZmSWEE T s u cro se tran sp orte rs a ff ec t s ma iz e k ernel an d embryo s iz e . A comp a r a ble ex pre ssio n p atte r n in the endo spe rm s urr o un ding th e e mbryo ha s b ee n r e po rt e d fo r t h e orth ol ogous rice O sSWEET15 and soyb e an GmSWEE T15 ge ne s , with mu tati ons le ading t o ret arded embryo de velop me nt, s ug ge st i ng a con se rve d f u nc t i on f or s u g a r t r ans f er fr o m t he e nd o s pe r m t o e mb r yo in d iv e rs e s pe ci es (W a n g et a l . , 20 19 ; H u et a l., 2 023). Soybe a n g m s w eet15 mu tant s ee ds a l s o di spl ay def e ctive end o sperm f u n ction, cha ra cte riz e d by th e p ersi ste nce o f e ndo s p e r m tha t i s norm ally deg r a d ed i n wi ld- type soy bean see ds (Wa ng et a l. , 2019 ). Thi s e ndo s p e r m pe rsi stenc e in so ybean s ug ges ts t h a t SWEET - d e pend ent suga r exp ort i s r e qu ir e d no t o nly fo r e mbryo growth b u t a l s o f o r the p r ope r d e velopm en tal t ran s i t i on o f th e endo sp er m during se ed ma tura tion. In a ddition to havin g s ma lle r emb ryos , the z ms w e e t 1 4a/ 1 4b/ 1 5a m uta nt a ccu mul ate s l e ss oil in the embryo at m atur i t y t h a n wild-type ( Fi g u r e 3 ) . A l t h o u g h S W E E T s a r e n o t d i r e c t l y i n v o l v e d i n l i p i d o r o i l b i o s y n t h e s i s , t h e tran spo rte d s uga rs like ly provide c arbo n s ke le ton s ne ce s s a ry for fat t y a cid syn the si s (Raw s t horn e, 2002 ), a s i llus t ra ted in soybean (M i a o et al . , 2020; W a ng et al ., 2020, 2025; Z hang et al ., 2 0 20). Z mSWEET -depe ndent carbon partitioning and t ranscriptional reg ulation by DED1 at the emb ryo– endos per m in t e r f a ce I n a ddi tion to sm all er e mbr y o s , th e zm sw eet14a /14b /15a mu tan t al so s how s re duc ed kern el we igh t, as ha s b een r epor ted for t he s in gle zm s we et 1 5a m ut ant ( Wa n g et al ., 20 24) , a s well a s d ef ect s in ca r bo n par t i t io ni ng b etwee n the embry o a nd en do sperm ( Fi g u r e 5 ). One plau sib le ex plana t i on i s th at the mut ation s redu ce sink st rength at the emb r y o -endo sp er m in te rfac e , th ere by reducin g t h e ov er all ca r b on imp or t in to t h e tw o fil i al tis su e s . Alte rna tiv ely, we ak ex pre s si on of ZmSWEET 15a in the a leur one la y er (Doll et a l. , 2 020) could pl ay a role i n suga r imp ort from mat ernal ti ss ues (nu cell u s or i ntegum ent ) to t he e nd os p e r m, thu s direc tly l imitin g suga r ent ry i nt o th e end osp er m . The o ve r e xpre ssion of Z mSWEET15 a w a s sho wn t o inc r e a s e the s uc r o s e c onten t in k ernel , i n a gre emen t with a role of ZmS W E ET15a i n main tainin g s uc r o se h omeo s t as is in the see d (L iu et al. , 2022). Int ere s tingly, th e R2R3-MYB tran scri ption fact or DOSAGE -EF FECT DEFE CTI VE1 (DE D1/ZmMYB73 ) ha s be en i denti fi ed as a direc t r e gula tor o f the se thre e EA S-expre s se d Z mSWEET g e n e s ta r g e t e d i n our st u dy (Da i et al. , 2022) . DED1 a c t s a s a pa tern ally de rived r eg ula tor o f k ernel d evelo pm ent and size in ma ize ( D a i et al . , 2 022 ). Tog e the r with our findin g s , the s e ob se rva t io n s su gge s t tha t modul a tion of E AS- e xpre ssed Zm SWEET tr a n s po rte r s by DED1 repre s en t s o ne molecu la r me chani sm by whi ch DED1 i nflue nce s k ernel a nd e mbr y o si ze, li k ely thr ough the co ntr ol o f sucro s e flux e s at the e mbryo-end o sperm i n t e r f ac e. How ever, the mo re s e ve re ke rnel de vel opme nt a l de fect s r e p or t ed in the de d1 mutant comp ared w ith tho se ob s e r v ed i n t h e zm s w ee t14a /14b /15a mutant in dica te th at DE D1-m edia te d r e gul ation o f k ernel dev elopm en t .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint d oe s not re ly exc lusively o n Z mSWEET- dep end ent s uga r tran sp or t , a s sup por te d by the fa c t t ha t add it i onal d ir e ct t arge t s have bee n foun d for D E D 1 (D a i et a l. , 2022). Does sugar recy cling in EAS c ells link p r o g r a m med cel l death wi t h e mbr y o and e ndo s perm gro wth? Ano t h e r int r ig uing aspe ct o f E A S biol o g y is the dynami c fa te o f th e c ell l ay er s adja cen t to th e embryo sc ute llum, whic h are progr e ss iv ely elim in ated a s th e em br y o g r ow s thr ou gh a tightly regul at ed c ell deat h p r oc e s s (D oll et al . , 2020, 20 25). Thi s re gulated c ell e limina tion i s a s socia ted wit h the remobili z atio n of sta rch re serve s (in the fo rm of gran ule s ) acc um ulated be f o re sta rchy endo sp erm ce ll s a cquire E AS i denti ty( Doll et a l. , 2 025). I t is rea s on a ble to hypoth e siz e t h a t th e de grad atio n produc t s of s t arch i n EAS cel ls mu st b e e ffici ently e xported an d rec yc led be fore the c ell s di e, a proce s s tha t l ikely r e quir es suc r o s e ef flux media t e d by ZmS WEET tran spo rte rs. Con si ste nt with t his hypoth esi s, th e al te red carb on p artiti oni ng ob se rved in bo th th e emb r y o an d e ndo s p e r m of the zm swee t14a / 14b /15 a m u t a n t ( Fig u r e 5 ) s ugges t s t h a t s ug ars rel e a s e d from EAS cell s c ontribu te t o fu eling t w o adja cen t si nk tis s u es ; th e rapidl y growi ng embr yo a nd the adja c ent sta r c hy e ndo s p e r m, whic h remai n s a me tabol i c sink during maize k erne l fillin g. Thu s, Z mSWEET-me di at ed s uga r rec yc ling fr om EAS cel l s may r e p r e s e nt a k ey coo rdina tion mec h ani sm f or c a rbon r edi st ributio n be t w ee n c ompeting si nk ti s s ue s. I n t e r e s tingly , ma iz e k er n e l s lac king an embryo (du e to em br y o de fe ctive em b 8522 m ut a t i on ) show a s tr ong e ctopic ac tivat ion of ZmS W E ET15a ge ne expre s s io n i n the scu tella r al eu rone laye r (S AL), w hich i s u suall y si tua ted be twe en the embryo and th e a leur one ( Doll et al. , 2020). This s ug ge s ts that t he p r e senc e o f e mbr y o i t sel f ca n modula t e S WEET gene expr es s i on in th e n eighbo r in g e ndo spe rm ti s s u e s, a lthough th e mecha ni s m unde rlyin g thi s embryo- endo spe rm c ommunic ation r em ain s to be ide nti fied . ZmSWEET functions extend bey o nd s e e d filling t o s eedli ng establishment I n add it i on to s eed dev el opment ph en otype s, t h e zms w ee t14a /14b /15a mut a nt al so dis p l ays pro nounc ed p ost-g ermina t i ve d e fec t s, inc ludi ng red uce d shoo t an d ro ot fr e s h wei gh t and impai r e d roo t d ev elopmen t d ur i ng ea rly see dling g row th ( Figure 4 ). T he se phe notype s indi cat e t hat l o s s o f ZmSWEE T14a /14b /15a fun ctio n impac ts no t only kernel fil li ng but also s eedl ing e sta bli shmen t. T wo non-mutuall y exc lus i ve h ypothe se s m ay ac coun t f or the s e ob ser vation s. Fir st, t he re duced e mbryo siz e c ombine d with al t e red carb on a cc umulation duri ng k ernel de vel opm e nt may re s ul t in emb r y o s with di mini sh ed re se rve a vail abili ty and m etabolic c apac i ty. The se “w eak e r ” embryos c ould mobil i z e f ew er r e sourc es d ur i ng germinati on, ul t im a tel y l eading t o c omp romi sed s e e dling v igor compa r e d with wi ld -t y p e plan ts . Al tern at ively , ZmSWEE T t ran s p ort er s m ay p lay a more direct role du ring g erminati on and e a r l y s eedli ng grow t h phas e s . In thi s scen a r i o, t he a ctiva tion o f SWEET -me dia ted sugar tra ns p o r t af te r imbibi t io n woul d b e re quir e d to s u stain c arb on all oc ati on to rap idly g r ow ing ti s s u e s. T hi s hypoth es is i s suppor ted by s tu die s re por ting the t ran scripti onal activ a tion o f a t l ea st 17 ZmS W E ET ge ne s in the p rimary r oot du r ing germin a t i on , inc luding the th r e e ZmSWEE T gene s .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint c harac t e r i z ed in t h i s s tudy (H oope s et a l . , 2019 ; L óp ez - Cori a et al. , 2019; Zhu et a l. , 2 022) (S upplemental fig ur e S1B ) . Tog e ther , th e s e da ta sup por t a dua l rol e f or ZmS W E ET14a / 14b /15a i n both s eed d evelop men t an d e arly s eedl ing grow th. S ucrose gradients and ZmS W E ET func t ion at t he embryo–endosperm inter face T hanks to t he d e velop men t o f F TI R ima gi ng applie d t o ma ize k ern el s e c tion s, thi s s t udy r e por t s , for the fir st time , t he spa tial imaging o f sucro se di st ri bution acr o ss th e maize ke rnel . A cle ar suc rose gradi ent w a s d et ecte d w ithin t he endo s p erm, with highe r s uc r os e abund an ce in the ba s al endo sp er m c ompared with the upp er e ndo s p e r m. Ou r imagi ng dat a ar e co n s i st ent wi th th e re cently me t a bolic c omp artmen ta tion ob s erved in th e m aize kernel (Ch e n et al ., 20 26) . Suc r o s e w as found to be hig hly enric he d in the p lace nt o -c hal a z al r e gio n. Thi s d is t ribut ion i s c on s i ste nt wi t h t he curr en t model o f nut r i en t tra n s f er int o the mai ze kernel , in w hich a ssimila t e s a r e del ivered p rimarily via phlo em un l o ading at the p edic el. N u trien t s rel ea sed into t he pl a cent o-ch a lazal reg ion ar e th en ta ken -up i n t o end o sper m via the BETL ce ll (F el ker and Shan non, 1980; P a tr ic k a nd Offle r , 2001 ; Of fler e t a l. , 2003). I n t e re sti ngly , sucro s e imagi ng o f ma ize ke rnel s revea led addi t i onal sucro s e ac cumul a tion i n m atern al ti s s ue s at s i te s o the r than th e ge neral build up o f s uc r o se in the pedic e l . Indeed , both t h e t i p s of th e k ernel and th e ma t e r n a l t i s sue s in c ont act wi t h the de velopi ng embryo ( Fi gu r e 5 ) a cc umulate sucro s e . It w ill b e inter e st i ng to inve s tiga t e th e propo rt i on of suc ro s e s uppi ed t o the end os perm and t h e embryo th rough t hi s route . T aken t o ge ther , our r e s ul ts supp ort a m odel i n whic h EAS- ex pre ssed Z mSWEET t ran spor t e rs pl a y a c entr al rol e i n c oordinati ng c arbon tra n s f er, re cyc ling , a nd alloc ation a t th e em br y o– endo sp erm inte rf a c e. By regul a ting suc r o s e flux e s d ur i ng s e ed deve lopm e nt, th es e tra n s por te rs i n flu ence embr yo grow t h, kernel siz e and u ltimat ely s e e dling vigor. B eyond t h e ir loca l tran s p o r t func ti on, th e r eg ul atio n o f ZmSWEE T14a/14 b/15 a e xpre ss i on requi re s int eg r a t i on o f deve l opmental p rogram s (via DE D1 sig nali n g) as w ell a s ti ssue - to -ti ssue c ommunic ation, highlig hting t he E A S a s a k ey reg ulatory hub for c arb on par titio ni ng in the ma iz e kern el.

P lant ma terial and gro wth conditions T he maize (Z ea may s) in bre d li ne A18 8 a nd de r iv ed edi ted plan t s w er e grow n und er the French S2 sa fety st anda rds for t h e cultur e of t ran sgenic p lant s, in grow t h c hambe rs or gre en hou s e. Th e photo peri od c ons i sted o f 16 h light and 8 h da r k ness in a 24 h diurna l c yc le. Temperatu re wa s s e t to 25°C /18° C (day /night ). The r e l a t i v e h u m i d i t y w a s c o n t r o l l e d a t 5 5 % ( d a y ) a n d 6 5 % ( n i g h t ) . S e e d s w e r e g e r m i n a t e d i n 0 . 2 L o f F a v o r i t M P Gode t s s ub s tra te (Er i t e rre , Sa int -A n d r é - de -Co r c y) and tran sfe rred a ft er 2 wee ks to 8 L of F avorit Ar g ile TM + .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint 2 0% p er l ite sub s t rat e (Eri te rre, S aint -A ndré -de -C orcy) suppl em ent ed w ith liqu i d fer tilize r. All pl ant s wer e p r opa gat ed by hand pollin a t i on . S W E ET phyl ogeny T he 24 mai z e SWEE T pro t e in sequenc es we r e r et rieve d at from Gr amene ( h ttp: / /ww w.g r a men e.o rg/ ) ( Gupt a et a l. , 201 6 ), and t he 1 7 Ar a bi dop si s a nd 21 ric e S WEET pr o tei n sequenc e s w ere downloa ded fr om Phy tozom e ( ht tp s: / /phyt ozome .jgi .doe .gov / p z/porta l .html ) . Con s erved si te s were s e l ecte d using G-BL OCKS wi t h all the thre e opt i o ns a vail able t o mi nimize strin ge ncy. The phy logeny was bui lt u s ing Ph yML , based on the amino a ci d a lignme nt, a s de sc ribed pr ev iou sly (Guin don et al . , 20 09). T r a ns ient protopla st t r ansfo r m a t ion f or subce llular loca liz a t io n Pl as mid D N A Pr e p a r at ion P la smid s a re pu rifi ed u sing t he N u cl eoBo nd Xtr a Midi /M a xi k it (Ma cher ey -Nag el) . Bac te ria from gly cer ol st ock s a r e pl ated on LB me di um with 50 µg/m L spec tino myci n. A singl e colony i s u se d to s e ed 5 mL o f LB with the same a ntibio tic an d incuba t e d a t 37 °C w ith agitati on (250 rpm ) fo r 8 h. This p r e-c ul ture i s t hen ex pand ed to 4 00 mL of LB c ontaining s pe ctin omyc in an d incu bate d ove r ni ght at 37 °C, 250 rpm . O ptic al de n s ity a t 600 nm i s m easu r e d and adj u s t ed to 4 b e fore c ent rifuga tion a nd removal o f t h e s upe r na ta nt . Pla smid ex t r ac tion follow s the manu fac t u r e r’ s pr otocol : cell ly si s , n eutra lizati on, c olumn loa ding, wa shing, e lution, p rec ipi tati on, w a shing , d r y ing, a nd r e s u spen sio n i n s t eril e wat er . Pla s mid quan tity i s a s se ss ed w ith s p e ctr ophotom etr y a nd N ano Drop , a nd qual ity is con firmed by re stric t i o n digesti on an d agaro s e ge l elec t r oph o r e si s, guide d by S napG en e so ftwar e p red iction s . Pr o to pla s t Is o la t i on a nd t ra ns fo r m at i on P r o topla st I sol ation pro tocol are c lo s e t o previ ou s ly publi sh ed (Fie rlej et a l. , 20 22). A188 wi ld- type se edling s a r e grown in a gre enhou se un til eme rge nc e (~0 .5– 1 cm) an d tran s fer r e d to dark nes s u n t i l th e thre e -le af sta ge (~ 12 d ay s a ft er sowing ) . L ea f sec t i on s (~ 1 mm) a re cut fr om th e c entra l p orti on, av oiding th e f i rst le af a nd lea f tip s. S ec tion s a re i ncuba ted in 2 0 mL o f enzyme solu tion (1.5% Cel l ula se R10, 0 .75% Macer ozyme R10 , 0 .1% P ecto ly ase Y23, 0 .6 M mann i t o l, 10 mM MES pH 5.7, 10 mM Ca Cl ₂, 0.1% BSA ) un der va cuum ( -50 0 mbar) for 30 m in in the dark. Dige s tion c o ntinu e s 2 h a t 25 °C wi th g en tle ag i t a ti on (30 rpm) . T he enzyme -pr otopl a st mixtur e i s fil ter ed thro ugh a 40 µm c ell st rai ne r, c entrifug ed at 100 g f o r 3 min (slow a cce lera tion /dece l era tion ), and w a s h e d t w i c e w i t h c o o l e d 0 . 6 M D - m a n n i t o l . P r o t o p l a s t s a r e f i n a l l y r e s u s p e n d e d i n 2 m L o f M M G Z m b u f f e r (0 .6 M manni tol, 15 mM MgCl ₂, 4 mM M ES , pH 5 .7) and count ed w ith a Mala ss ez chambe r. The co ncen tra tion i s adj u s t ed to 1 × 10^6 p rot opla st s/mL a nd kept on ic e in the dark . For pr ot o pl as t tra ns fo r ma t i o n, 100 µL o f p r otopl a s t s a r e mi xed with 110 µL o f fr eshly prep are d PEG sol ut i o n (40 % PE G 4000, 0.2 M mann itol, 0.1 M C aCl ₂ ) a nd 20 µL o f D NA (10 µg pl a s mi d) in MM GZm. The mix t u r e i s incub a te d 10 min in t he da r k a t ro om tem pera tu r e , an d the r eac ti on is stopp e d with 80 0 µL of W 5 buf f e r (15 4 mM Na Cl, 125 mM Ca Cl₂, 5 mM KCl, 2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint m M MES, pH 5.7 ). Pro t o pla s ts a re ce n t rif uge d a t 100 g fo r 3 min, re sus pend ed in 1 mL W5 b uff e r , an d tran s fer r e d to 24 -w ell plate s . Cult ure s a r e inc ubat ed in the dark a t 25° C wi t h g entle agit atio n (25 rpm) until f u rt h e r an al y s is . C onf ocal microsco py P r o topla sts ar e image d u s in g a n inv erte d c onf oca l mic r o scop e ( Z eis s L MS710) at 20 hou rs and 44 ho u r s po s t- tran s forma tion . A 4 0X wa t e r-imme rsi on o bjectiv e ( Pla n - Ap oc hroma t 40× /1.0 water ; Z eis s) i s us ed . F luoropho re s a re e xci t e d with an a rgon la ser a t 458 nm for mTurqu oi se and 51 4 nm for mCi trin e, whi le 561 n m ex citation i s u s e d to de t e ct auto fl uore sce nc e. Image s a r e ac qui red seq u entia lly to se par at e c hannel s a cc or d ing to thei r ex cita tion and emi ssio n (Gill e s et a l . , 2021) . D e tec tion set t i ng s are: 519 -58 0 nm for mCit rine , 4 63-500 nm for mTurquoi s e and 6 00-795 nm for aut oflu ore s c e nc e. Y eas t he t erolog ous c omplementation assay T o a s s e s s t he suga r t r a n s po rt ac t i vity o f the can dida te gene s , th eir full -le ng th c oding seque nce s wer e c lon ed i nto t h e yea st c DNA c loni ng vec t or p DR 196, mo dified to be c ompa tible w it h gat ew ay tec hnology pD R1 96 - G W (Wip f et a l. , 2003; Koeg el et a l. , 2 013 ), and con st r uc t s we re ve rifi e d by sequ e nc ing. Rec ombina nt p DR196 - Z mSWEETs pla s mid s, toge the r w it h the t wo type o f e mp ty vec tor p D R19 6 and p DR196 -GW u se d a s a nega t i ve c ontrol, w ere i ntr oduce d in to t he ye a st m ut a n t s t rain s S U S Y7 /ur a 3 ( M at α le u2-3 112 u ra3 -52 t r p1 mal 0 suc 2::UR A 3 ur a 3 LEU2: :PA DH1 - StSUSY1 ), w hich i s impa ire d s uc r o s e up ta ke. SUSY 7/ ura3 c anno t utiliz e e xtracel lul ar s ucro s e unl es s a functio nal s uc ro se tran sp orte r i s pre s ent a t the pla sma me mbran e (Ri e smeie r et al . , 1992 ). T ran sfo r me d y ea st ce ll s we r e c ult ur e d on synth etic de fic ien t medium lac king ur ac il and suppl e ment ed w it h eit her 2% gluc o se o r 2 % filte red suc ro se (Si gma Ul tra, 99.5 % GC re f: S79 03 -250) a s th e sol e c arbon source . Y ea st c ul ture s w er e adju st ed to an OD ₆₀ ₀ o f 0.1, seri ally diluted , an d s po tt ed ont o s olid me dia . Growth wa s monitor ed af t e r inc uba tion at 3 0 °C for 2 day s f or gl ucos e contai ni ng m edium and 3 days for suc r o s e con taini ng med ium. F uncti onal com plemen tati on o f s ucro s e upt ake wa s evalu at ed ba sed on th e a bility of t h e t r a n sformed s train s to grow be t t er a s c ompar ed to emp t y vec tors on suc r o se c on taining medium . sgRNA design and C R I SP R–C as9 vec tor const ruc t ion S ingle guide RN A s w er e de s i gned usi ng the CRIS PO R w eb to ol ( Concor det and H aeu s sler , 2018) ba s ed on th e B7 3 mai z e r e fe r e nc e genome and dou ble c hec k for c ompatibil i t y wi t h A188 g enome. Two 20-nuc leo tide ta r g e ts s e qu ence w ere sel ec ted f or eac h gene (Sup p lemen tal Ta ble S1 ). Targe t si t e s w ere ch o s en ba sed on hi g h p r e d i c te d o n- t ar g e t a ct iv i t y s c o r es a nd t h e a bs e n ce o f pr e d ic te d of f- t a r g e t s i tes w it h fe we r t h a n t h r ee m is ma t c h e s i n the ma iz e ge nome . B ec a us e C RI S POR prio ritiz es gu ide spe cif ic ity at t he in dividua l gene lev el , r e l a t i v e l y l o w s p e c i f i c it y s c o r e s w e r e e x p e c t e d f o r t a r g e t s s h a r e d a m o n g c l o s e l y r e l a t e d g e n e f a m i l y m e m b e r s . Al l g enome edi t i ng -re lat ed cl oning s t e p s were perf ormed ex actly f ol lowing the s tr a t e gy descr ibed in ( Doll et al . , 2019) . Brie fly, the fir s t sgRNA wa s cl oned under t h e c ont rol of ric e U 3 pro moter in b inary vec tor L 1609 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint c ontaini ng a r i ce codo n op tim iz ed Ca s9 driven by c onst i tutiv e maize ubiquitin promot er. T he s ec on d s g R N A w as clon ed un de r th e whe a t U6 prom o ter i n L 1611 v ect or. The U 6 - d riv en gRN A ta rge t c a sse tte from L 1611 w as s ub s equen tly exc i s e d with re s t rictio n e nz ym e s an d cl oned i nt o th e L 1609 d eriv ative, dow n s t ream of th e U3-driv en ta r g e t c ass et te, yiel ding a du al- sgRN A C RI SPR con s tr u ct ta r g e t i ng r e g ions o f the ge n e s of in ter e s t ( Fi g u r e 2 ). The two final bina ry vec tors obta ined (L198 6 ta rgeting ZmS WEET14a an d ZmS WEET14b; L1 987 ta r g e t i ng ZmSWEE T15a) we r e v erif ied b y restric t i on a naly s i s and S an ger sequ e n cing prior to t r an sfo rmati on. T he val idated co n s tr uct s w ere intro duce d into Agroba ct erium tume faci en s st r a i n LBA4404 (pSB1) by bac teri al c onjug ation and u s ed for maiz e tra n s fo r mation. Mai z e tran sforma tion a nd s c reen for edited plants A. t um e f acie n s st r a in LBA44 04 h arbori ng the rec ombina t i on pr oduct o f pS B1 and the c on s t ruc t o f in te r e s t wa s u s e d t o tran s form 12 DA P i mmat ur e e mbry os of ma iz e inbr ed l ine A188 ac cording to a stan dard pro t o co l (I shida et a l . , 19 96; Fie rle j et al ., 20 22). S c r e ening fo r ge nome edi ting even ts wa s perf ormed on g DNA f rom T0 l eave s by s pe cific P CR ampl ifica t io n ( Supple men tal Ta ble S1 for p r ime r se quen ce s) f ollowe d by S anger seque ncing . Chr omatog r a ms f r om wi ld - t y pe and t ran sgene - b e aring plant s we re th en compa r e d to i den tify p ut a t i ve targe ted mu tagen e s i s eve nt s . MRI a nd TD NMR me as ure ments Non-i nv asive 3 D mea surem ent s o f maize (Zea may s) s eed s w ere pe rformed a t 11.7 T Ava nce Neo 500 MHz S uper W id e Bo re u sing an NM R s pect r omete r (Br uk er BioS pin ) equipp ed w ith a 1H quadrature pr obe he ad w ith i nner dia me ter o f 6 6mm fo r m ult i- s ee d mea s u remen t s ( Plu ten ko et a l . , 2025). The se ed hol de r wa s pr i nt ed w it h a r es olu t io n of 5 0 μ m o n a F o r m 2 3D p r i n te r ( F o r m l a bs, U S A) u s in g hi g h t em p er at u r e n o n- m agnetic re sin (t y pe flh tam01 and flht am02). The M RI pa rame ter s for mul tipl e s eed s c re ening we re set a s f o l l o w s : f i e l d o f v i e w ( F O V ) 4 0 x 4 0 x 6 0 m m , r e p e t i t i o n t i m e ( T R ) 7 5 0 m s ; e c h o t i m e ( T E ) 3 m s ; a v e r a g e t i m e 1 8 7 m in and r es oluti on 400 μm. F o r gene ra t ing the w a ter image s , the pul se s wer e a pp lied r e sonan t on the wate r f r e qu e n c y , a n d w i t h a n o f f s e t o f − 1 4 0 0 H z f o r t h e l i p i d i m a g e s ( L a n g e r et a l. , 202 3 ). M a nual ima g e seg men ta tion wa s pe r f ormed u sing Am ir a s o ftwa re (FEI Vi suali z a tion Scie nc es G r o up, Fr anc e ). The lipid c onten t wa s me a s ure d u sing T D - NMR ( M Q60 i n s tr ument , B ruker, G ermany) a s detai l ed earli e r (Roll et s c hek et al . , 201 5) . S ucrose mapping us ing high resolution F T IR Qua nti tativ e mappi ng o f sucr os e di s tr ibu t io n in the ma iz e caryop s i s wa s d on e usi ng Fo ur i e r Tr a n sform In fra re d m icrosp ectro s c opy (FT IR ). Im aging and data p roc e ssing w ere done as e ssen tial l y describ ed i n ( Gu end el et a l. , 2 018). In s h o r t , intac t c aryop s e s w ere h arves t e d a t 13 DAP , ins tantly f roz e n in l iquid nitrogen , embe dde d in T issue -Te k c ryomold s a t −20 ° C, a nd cr os s sec tion ed (6 µm) with a cryo tome (Cr y oS tar NX7, ThermoF i s h e r .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint S cie nt i f i c). Ti ssue sec tion s we re lyophil ized and s t o red in darkn e s s at ro om temp erat ure un til a naly si s. Imag ing w as pe rfo r me d u s ing a Hyperio n 300 0 FTIR mi cros c op e (B ruker Opt i c s) co upled t o a Te ns or 27 FTI R spec tr ome t e r (B ruker Opt ics ) with an int ernal mid -in frar ed s ou rc e. FT I R image s were r ec orde d in th e spec tra l ran ge of 39 00 c m −1 to 800 cm −1 at a spatial r eso lutio n o f 11 µm and a s p ec tra l re soluti on of 6 c m −1 . D ata p r oc e s s ing and sucro se fing er p r intin g wa s don e using so ft w a re Ma tLa b (The Ma t h Work s) ex actly a s de scrib ed. Ger mination i n hydroponic as say Ma iz e s e ed s were su rfac e -s te rilized i n 1.4% (v/v) bl each , 1 ‰ (v/v ) Tw een - 20, for 15 m in und er gent le a g i t a t i o n . T h e s e e d s w e r e t h e n t r e a t e d w i t h H 2 O 2 ( 3 5 % ; v / v ) f o r 2 m i n , r i n s e d w i t h 7 0 % ( v / v ) e t h a n o l , a n d w a s h e d 6 t i m e s w i t h s t e r i l i z e d w at er . T h e s e e d s w e r e o v er l a i d w i t h we t c l a y b e a d s i n a p l a s t i c b o x , w h i c h w a s i t s elf c overed by a tra nspa re nt pl as t i c fi l m. S eed s we re ge r mi n at ed and fur ther g rown i n a growth c h amber a t 6 5% relativ e humidi ty, w ith 15 h / 9 h li g ht/ d a r k cyc les (1 50 μE m-2 s-1) a t 22 °C (light) /20 °C ( d a r k ). At 5 DAS , se ed ling s w ere t r an sf erre d to hydropon ic contain e rs (61.6 X 35 .8 X 1 3.6 cm; 2 0 s e e dling s / c on tain er ) f i ll ed w ith 24 L of a me dium c on t a ini ng 1 .25m M KNO 3, 0 .75mM MgS O4, 1 .5mM Ca (N O 3)2, 0 .5mM KH2P O4, 0 .1mM Mg Cl2, 0.05mM Fe -EDTA, 0.05mM H 3B O3,0.0 12mM MnS O 4 , 0.7mM CuS O4, 0 .001mM ZnSO4, 24.10-5m M MoO4N a2,1. 10 -5 mM Co Cl2, 0 .1mM N a 2SiO3, and 1 mM M E S. Pla nts we re g r o w n in th es e solu tion fo r 6 day s. At 11 DA S , roo t s w ere ex ci sed an d image d for th eir r oot s y stem a r c hitec ture . Roo t a nd s hoot s w e re wei ghed . I s otopic mas s spectromet r y 13 D A P K e r ne ls of zm s w ee t14a /14b /15a a n d w i l d t y p e ar e co m pa r e d f or t he i r a cc u m u la t i o n 13 C- l a be kl l ed suc r o s e on all carb on s (CL M-7757 -PK, CAS # : 41055-68 - 9 , C a mbr i dge I sotop e L aborat orie s , Inc . ( USA )) . The k ernel s a re sep ara ted f rom the e ar u si ng a s c alp el bla d e, t ak ing ca r e n ot t o damag e th e m and ke eping a p edice l l ength of appr oxim ately 2 mm . The g lume s a re r e moved and the ke r n e l s are rin sed 3-4 time s f or 30 s e c o n d s e a c h i n 1 X P B S , t h e n t h e e n d o f t h e p e d i c e l i s t r i m m e d b y a p p r o x i m a t e l y 0 . 5 m m w i t h a f i n e r a z o r b lade. Ke r n el s a re th en inc uba ted i n a 96 - w ell p la te co n t a ining 2% (w/v) lab ele d suc rose, dilu ted in 1X PB S. Th e 13 C- sucro se s o l ut i on s a r e p r e pa red as foll ow s : a final 2% sucro se s olu tio n cont ain i ng 10% 13 C s uc r o se di lu ted i n 1 X PBS . The s oluti on i s t he n di sp en sed at 1 60 µl pe r we ll . I n ord er to preve n t ev apora tion , the p la te s a r e p lace d a t r oom tempe r a t ure , a nd 48 ke rnel s pe r re plic at e ar e coll ec ted an d d i ssec ted a t t he in dica ted time s. During s ampl ing, the ke r n e l s a r e rin sed i n two suc ces s i ve ba th s of ste rile 1X PBS (~10 ml), follo wed by a final rins e w ith 1X PBS to r emove n on -acc um ulated 13 C. To di ss ect the di f fer ent par ts of the k e r n el , th e pe rica r p i s i ncised on ei ther side o f the embryo an d th en r emov ed ( D o ll et al. , 2020 ). T h e embryo i s r e moved, rin s ed i n st eril e 1X P BS, and b rie fl y blott ed on ab sor bent p aper for 5 s e co nd s, th en pla ce d in a n aluminu m cupule . Th e e ndo s p e r m is c ut tran sve rsely just ab ove the sp ac e lef t by the e mbryo, r e sulting i n a n upper and a lower pa rt. T hes e pa rt s a re plac ed in sep ara te tin cup ule s. Fi nally , the samp le s a re d ehy drat ed by plac ing t he m i n a v entilat ed ov en fo r 4-5 day s w i t h th e f ollow ing se tt i ngs: Te mper atu re: 60 ° C; F an: 70% ; Va lve: 50% . Sample s .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint ar e an a l y z e d b y Is ot o pi c ma ss s pe c tr om e t r y ( Va r i o- P YRO cu b e e l e me nt al ana l y z e r c o up l ed wit h an Is oP r i me P r e ci sion m a ss sp ec t r omet er, Elemen t a r , U K ) by the A QuI pl a te for m (Mo ntp e llier, Franc e) . The dried and g r o und s a mpl e s are plac ed in a tin c upule a nd injec t e d int o an ele men t al ana lyzer (Va r i o -PY RO c ub e, El e m e n t ar , U K) . A ft e r c o m b us t i on at 9 20 ° C i n t he p r es e n c e of o xy g en a n d C uO , t h e g a s m o l e cu les f r o m t h e s a m p l e ( m a i n ly H 2 O, CO 2 , N 2 , a nd N 2 O ) a re t ran sported via a car rier ga s flow (ult ra -pure he lium) to a r e ducti on furna c e w he re the nit rogen oxide s a re re duc ed t o N 2 in the pre s enc e of co ppe r a t 600° C (Duma s re ac t io n ) . The H 2 O produc ed i s trapp ed by SI CA PENT column s (Me r c k ). T he N 2 a n d C O 2 are t he n sep ara ted . The C O 2 i s trapp e d a t room t e mpe rat ur e in a TP D (Temper atu re Programmabl e De sorp tio n) c olumn and su bse que ntly rel ea s ed a t 100 °C . A the rmal c onduc t i vi ty det ec t o r (T CD ) quan ti fie s tota l N and C. N 2 a n d t h e n C O 2 a r e t h e n a naly z ed by ma s s s p ect rome tr y ( Is oP ri me Pr ecisi on, Eleme n ta r , UK) to d ete r mine t he i s otopic a bun danc e s 13 C/ 12 C. Th e δ 13 C is ca lcul at ed u sing the fol lowing formula a nd h a s no unit : E qua tion 1 . Th e r e fe r en c e i s Pee D ee B e l em n it e , t h e s h e l l of a f oss i l b e lem n i te f r om t he Cr e t ac e o us p e r i od . T h is r e f er en ce h as an abno r ma lly high 13 C/ 12 C r a ti o (0 . 0 112372), mea ning that mo st na tura l s ampl e s hav e a n egativ e δ 13 C (S lat er et a l . , 20 01) . S ta ti stic al and gra phic al anal y s e s w ere produc e d u sing R s tu dio so ft w a re. The p rimary o bjectiv e was to es t a bli sh a r elati on ship betw ee n the mea s u red par ame te r , δ 13 C, a nd t h e fac tor s te s ted , suc h a s geno type, compa r tmen t, a nd so aking time. A ckn ow ledgem ents T he auth or s a r e t ha nk ful to Camil le Kna upp, J u s tin B erge r, Pat ric e B olland , I s ab e lle De sbouch a ge s, a nd He rvé L eyral f or tec hnic al a ssi stanc e in ma ize cultur e and media pre para tion ; Ci ndy Via l, L aure en Gr a ngi er, N e lly C a m i l l e r i , a n d J u l i e P r a t a f o r a d m i n i s t r a t i v e a s s i s t a n c e . W e a l s o t h a n k I . P l u t e n k o , S . W a g n e r a n d S . O r t l e b f o r e xperimen ta l s up p or t in s e ed ph eno typing and MR I. We a ckn owledg e Joa n D oid y for p r ov iding the S US7 ye a st st rain and th e assoc ia ted yea s t prot oc o ls toge the r with Gael Yve r t . HR a nd L B ack nowled ge funding by the F eder al Mini s tr y o f Educ ation and R es earch (M AGDI - Projec t; B MBF g rant 031B14 51A) , the De ut sch e F ors c h ungsgem ei n s c ha ft (g ra nt no . RO 2411/7-2 ), th e Europea n Reg ional D eve l opment Fund (ER DF ) and th e I nves tmen t Ba nk o f S ax ony-Anh al t (F KZ: ZS/2019 / 09 /1014 44). YF w a s s uppor ted by a CIF RE fellow ship of th e ANRT (gra nt N °2018 /0480 ) . L G wa s s up p or te d by a Ph. D. fell owship from the Minis tère de l’E n sei gnemen t .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint S upérie ur e t d e la Rec h erche . TW and BL ac knowl edge supp ort f rom th e Pla n t Sci e nce a nd Br eeding Divi sion o f the In s t i t u t N a t i on al d e la R ec herch e en Ag r ic ultu re e t Alime nta tion e t Env ir onnemen t (BAP , INR AE) . We a ckn owledg e the c ontri bution o f SF R Bi osci enc e s (U ni ver s i t e Cl au de Bern ard Ly on 1, CNRS UAR3444 , I n s e rm US8 , ENS d e L yon) for s up p ort in i magini ng and th e he lp of the s t a f f of L yMIC -PL AT IM. W e h onor t h e m emory o f P et er Rogow sky , an ir repl acea bl e c olle ag ue, w ithou t whom thi s “ SWEET m anus cript ” w ould no t e xist .

Anja li A, Fatima U , Ma nu MS, Ra masa m y S , Senthil-Kuma r M . 2 020. S truc t ur e an d regula tio n of S WEE T tran spo rte rs in plant s: An upda te . P lan t Physiol og y an d Bioch emi stry 156, 1– 6. Ba t e N J, Niu X, Wang Y, Reimann KS, Helentjaris T G . 2004. An Inve r t a se Inhi bito r from Ma ize Loc alize s to t he E mbryo Su r roundin g Regi on du ring E arly Ke r ne l D evelo pmen t. Plan t P hy siology 134, 2 46–254 . Braun D M . 20 22. Phl o em Loa ding a nd U nloadin g of Su cr o s e: Wha t a L ong, S trang e Trip from S ou r c e to S ink . Ann ual Rev iew o f Plan t Bio logy 73 , 553– 584. C hen L-Q, Cheung LS, Feng L, Ta nner W , From me r W B . 2015a . Tr a n s po rt o f S ugar s. A nnual Re view o f Bi ochem i s try 84 , 865–8 94. C hen X-M, Lia ng X - G, Chen Z-Y, et al. 202 6. Met aboli c c ompar tment ation o f ma ize endo spe rm fo r ef ficie n t synth e sis and stor age o f carbon a nd ni tr ogen a ssimil a t e s. New P hytolog i s t 24 9, 2 52– 269. C hen L-Q, Li n IW , Q u X - Q, Soss o D , Mc Fa rlane H E, Londoño A, S amuels AL, F r om me r WB . 20 15 b . A Ca s c a de o f S equen tia lly E xpres sed Sucro s e Tran spor ters in t h e See d Coa t and Endo sp erm Pr o vides Nu tr i tio n f or t h e Arabi dop s i s Em br y o. Th e Plan t C el l 27 , 6 07–619. C hour ey PS, Hueros G . 2017. The b as al e ndosp erm t r a n s f er la ye r (BETL ): ga tew ay to the mai ze ke r n el . CAB I Bo oks . Maiz e kernel d eve lopmen t. 56 – 67. C oncordet J-P, Haeussler M . 201 8. CR IS P OR: in t u i t iv e gui de sel ec t i on for CRIS PR / Ca s9 gen ome e diting e xperimen ts a nd s c r e en s. N uc leic Acid s R es earc h 46 , W242 – W 2 45. Dai D , M udunk ot h ge JS, Galli M, et al. 2 02 2. Pat ern al imprinting o f d o s ag e -e f fect de f e c tive 1 contri bute s to se ed weig ht xeni a in ma ize . N a t u re Com munic ation s 13 , 536 6. Doc hlert D C. 1990 . Di st r i but i on of enzyme activ itie s within the d evelo ping ma iz e (Z ea ma ys) ke rnel in r elati on to starch , oil and pro t e in ac cumula t i on . P hysiol ogia P lant arum 78 , 560–567 . Doll N M, Depège -Fa rgeix N , R og owsky P M, Widiez T. 2017. S ignal ing in Early M a i z e Kernel Dev elopm ent . Mol ecula r Plan t 10 , 375–38 8. Doll N M, F ier l ej Y, Ee khout T, et al. 2025 . KIL tran scrip tion f a cto rs fac ili tat e e mbry o growth i n maize by p r omo t in g end o sperm el imin ation v ia l yt ic c ell d eath . The Plan t C el l 37 , k oaf162 . Doll N M, Gilles L M, Gé rentes M-F, et al. 2 019. S ingle and mul t ip l e gen e knoc kout s by CRI SPR –C as9 i n maiz e. P lant Cell R epor t s 38 , 487–5 01. Doll N M, Ing r a m G C . 2022. E mbr y o–End osp erm In terac tion s . A n nual R ev iew of Pl ant Bi ology 73 , an nure v- a r p lan t-102820 -0 91838. .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint Doll N M, Just J, B runaud V , et al. 202 0. Tran script omic s a t M a ize Embryo/Endo sp erm Int er face s Id e nti fie s a T r a ns c ri ption ally Di s tinc t Endo sperm Sub do main Adja cent to th e E mbryo Sc utellu m. The Pl an t Cell 32 , 833 – 8 52. E om J-S, Chen L- Q , Sos so D, Julius BT, L in I, Qu X- Q , B raun DM, F r om me r WB . 20 1 5. SWEET s , tran spo rt ers fo r in t r ac e l lu la r a n d int e r ce l l u l a r s u gar t ra ns lo ca t i o n. C u rr e nt O p i n i o n in P l a nt B i o lo g y 25 , 53–6 2. F ei H, Yang Z, L u Q, W e n X, Zhang Y, Zhang A, L u C . 2021 . O sSWEE T14 c ooper ate s with Os S W E ET11 t o c ontribu te t o grain filling in ric e . Pl ant S ci en ce 30 6, 1108 51. F elker F C, Shannon J C . 1980. M o vement of 1 4C -lab el ed A ssimil a te s i n t o Kerne ls o f Zea mays L . P lan t Ph y s i o lo g y 65 , 86 4–870. F ierlej Y, Jac qui er NMA , G uil le L, et al. 2022 . Eva luation o f genome an d ba se e di ti ng tool s in ma iz e p r otopl a s t s. F ronti er s in Pl an t Scie nc e 13 , 1010 030. Geig er D . 2020. P l ant glu co se tr a n s po rt e r struc t u re and func t i o n. P fluge rs Archiv 472, 1 111–112 8. Gif ford R M, Evans L T. 198 1. Pho to synth esi s , Car bon P arti tioning , an d Y ield. Ann u a l Rev iew o f Pl ant Phy s i ol ogy 32 , 485– 509. Gille s L M, Calhau A R M, L a Padul a V, Jacquie r N MA, Lion net C, Mart ina nt J - P, Rog owsky P M, Widiez T . 2021. L ipid anchoring and e lec tro s t a t i c in ter ac ti ons ta r g e t N OT -LI KE- D A D to pol len end o -pl a s ma membr ane . Journa l o f C ell Bi ology 220, e2 0201007 7. Gonzales CF, Co wling CL, Kel ley DR . 202 4 . The Rolled T owel Me t ho d for Hormon e Re s po n s e As sa ys in Mai ze. C old Spring Ha r b or P r o toc ol s, pro t oc ol s; pdb.pro t108623 v1. Guende l A, Rollets c hek H, Wagn er S, Mus zynska A, Bori sjuk L . 2018. Micro Ima g ing Displ ay s the Suc r o s e L andsca p e wi t hi n and a lo ng It s Alloc a tio n Pa t hw ay s1. Pl ant Phys iolog y 178 , 1448– 1460. Guindon S, Delsuc F, D ufay ard J-F, G ascuel O . 200 9. Estima ting Ma ximum Like lihood P hyloge ni e s with Ph yML. I n: Po s ada D , e d. M eth od s in Molec ula r Biolo gy. Bi oin forma tics for DNA S equ enc e Analy s is . Humana Pre ss, 1 13–137 . Gupta P, N ai t hani S, Te llo-Ruiz MK, et al. 2016. G ram e ne da tab a se: Na viga t in g plan t compara tiv e gen omic s re source s . Cu rren t Pl a nt Biol ogy 7–8 , 1 0–15. Hoopes GM, H amilton J P, Wood JC, Est e ban E, Pa s ha A , V a illanc our t B, Prov ar t NJ, Buell CR . 2019. An u pdate d ge ne a tla s fo r mai z e r ev eal s org an- spec i fic a nd s t re s s-induc ed g en e s. Th e P lan t Jou rnal 97 , 1154– 1 167. Hu Z , Tang Z, Yang J, et al. 20 23. Knoc ko ut of OsS WEET15 Impair s Rice Embryo F ormation a nd Seed -S ettin g. P lant and C ell Ph y s i ology 64 , 258 –268. I s hida Y, Saito H, Ohta S, Hiei Y, Komari T, Kumas hi ro T . 1996. H i gh e f fic iency tr a nsforma tion o f maize (Zea m ays L.) medi ated b y Agro bac te r i um tu mefac ien s. N a t ur e Biot echnol ogy 14 , 745– 750. Ka ng B- H, Xiong Y , W i lliams D S , Pozueta -R o m er o D , C hour e y PS . 2 009. Mini a ture 1 -Encode d Cell Wa ll I nvert ase I s Ess enti al fo r As sembly and F un ction o f Wal l -in -G r ow t h in th e Maize E ndos p e r m Tran sfer Ce ll . Pla n t Ph y s i o lo g y 151, 1 366–137 6. Ke ller I, Neuhaus HE . 2025 . Humbold t re vie w: Fu nction a nd cha ra cte riz a t i on o f su gar tran sp or t a cr o ss pla nt m embrane s: Hi story a nd pe r s p ec tive s. Jo urnal of Plan t P hy siology 314, 1 54600. .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint Ko H -Y, H o L - H, N euha us HE, G uo W -J . 2 021 . Trans por t e r S lS WEET 15 unloa d s suc r os e f rom phl oem and s e ed c oat fo r f ruit and s eed devel o pment in to mato. P lant Phy s i ol ogy 187 , 2230 –2245. Ko ch K . 2 004. S ucro s e me taboli sm: r eg ul atory mec hani sm s and piv ot al role s in su ga r s e nsi ng and pla nt d evelo pmen t. C urre nt O p in ion in P l ant Bi olog y 7 , 235–246 . Ko egel S , Ait Lahmidi N , Arnould C, Chat agnier O, Walder F, Ineiche n K, Boller T, W ipf D, Wiem k en A, Court y P- E . 2013 . The fam ily o f a mmonium tran spo rte rs (A MT) in S orghum bic olor : two AMT member s ar e induc ed l ocal ly, but n ot sy st emic ally in root s c olo nized by arbu scula r mycorrhi zal fu ngi . N ew Phy tologi st 198, 853–865 . L afon-Plac ett e C, Köhler C . 2 014. Emb r y o and e ndo s p e r m, par tne rs in see d deve l opment . Cur ren t Opinion i n P lant Bi olo gy 17 , 64–69 . L alonde S, Wipf D, From mer WB . 2004. TRAN S P O RT MECH ANIS M S FOR ORGAN I C FORMS OF C ARBO N AND NITRO GEN BETWEEN S OURCE AN D S INK. Annual Review o f Pla n t Biology 55 , 3 41– 372. L ange r M, Hilo A , G uan J - C, et al. 202 3. Cau se s and c on se quenc e s of e ndogeno u s hy poxia on grow t h and m etaboli sm o f dev elopin g maiz e k erne ls. Plant Phy s i o logy 192, 12 68–1288 . L iu M, L iu T, Lu J, Zhou Y, Li u S, Jiao P, Li u S , Q u J , G ua n S, Ma Y . 2022. C harac teri zation an d Func t i on al A n al y s is of Z m S W E ET 1 5 a in M a i z e . D NA a n d C e ll B io l o gy 41 , 564–57 4. L ópez-Cor i a M, Sánc hez -S ánchez T, M a r tínez-Ma rcelo VH, Aguile r a-Alv a r a do G P , Flores-Ba r re ra M, King - Día z B, Sánc hez -Nieto S . 2 019. S W E ET Transp orte rs fo r the Nouri shme nt o f Embryon ic T is s ue s during Ma ize Germina t i on . Ge ne s 10 , 78 0. Mia o L, Ya ng S, Zhan g K, He J, Wu C, Ren Y, G a i J, Li Y . 2020. N a t ural vari atio n and sel ec tion in GmS WEE T39 a f fect soybe a n se ed oil con ten t. New P h ytologi st 225, 1651–1 666. Niño- Gonz á lez M , N ovo- Uzal E, Ric hardson D N, Bar ros P M, Duque P . 2019. More Transpo r t er s, Mo re S ubs t ra te s: The Ar a b i d o p s i s Ma jo r Fac ilit ator Sup e rfami ly Rev isi t e d . Mol ec ular Pl a nt 12 , 1182 –1202. Offler CE, Mc Cur dy DW , P a t ric k J W, Talbot MJ . 2003 . Tran s fe r Cell s : Cell s S p ecia li z e d fo r a Sp ecia l Purpo se . Ann ual Rev iew o f Plan t Bio logy 54 , 431– 454. P alme r AFE, Heic hel GH, Mus g rav e RB . 1 973 . Pat te rn s o f Tra ns l ocati on, Re s p i r a to ry Loss, and Re di stribu tion o f 1 4c i n Maize La beled a f t e r Flow e r in g. Cr op Sc ienc e 13 , crop sci197 3.001118 3X00 1300 030025x . P atric k JW, O ffle r CE. 2001. C ompa rtme ntati on of tran sp or t a n d tr a nsfe r ev en ts i n dev eloping see d s. J ou rna l o f Experime n t a l Bo tany 52 , 55 1–564. P lut e nko I, Radc huk V, May er S, Keil P, O rtleb S , Wagner S, Lehmann V , Rollets c hek H, Borisjuk L . 2025. M R I - S eed -Wiza rd: comb ining de ep lea r ni ng al g orithm s w ith mag netic r e sonanc e ima gi ng e nable s adv anc ed se ed p henotyping . Journ al of E xperim enta l Bo tany 76 , 393– 410. P ovilus RA, G e hr ing M . 2022. Ma tern al -f ilial tr a n sfe r s t ruc tu r e s in en do sperm : A nexu s o f nu t ri tion a l dyna mics a nd s eed de vel opme n t . Cu r re nt Opi nio n in Pla nt Bi ology 65 , 1 02121. Ra wsthorne S . 2002. Ca rbon flux and fa t ty a cid s yn t h e s is in p l ant s . Pr o gre ss in L ipid Re sea r c h 41 , 182–196 . Ri es mei er J W, Will mitze r L, F r om me r W B . 19 92. I sola tion a n d ch aract eriz ation o f a s uc r o s e c arri er cDN A f r om spina ch by fu nctiona l ex pr e ssion in yea s t . The EMBO J ournal 11 , 47 05 –4713 . Rol lets c hek H, Fuc hs J , Friedel S , Börner A , Todt H , J a kob PM, Borisjuk L . 2015. A novel noninva s iv e p roce du r e for hi gh -th r ou ghput scr eening o f major s eed tr ait s . Pla n t B io t e chnol ogy Jo urnal 13 , 18 8–199. .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint Rua n Y-L . 2014. S ucro se M et aboli sm: G a tewa y to Div e rse C a r bo n Us e an d Suga r S igna ling. A n nual R ev iew of P lant Bi olo gy 65 , 33–67 . S hen S, M a S, Chen X-M , Yi F, L i B-B, L iang X- G, L iao S-J, G ao L -H , Zhou S-L , Ruan Y-L . 2022 . A tra n s c r i pt i onal l andsca pe und erlyi ng suga r import for gr ain set in maize . The Pl ant J ournal 110, 2 2 8–242. S later C, P rest on T , Weaver L T . 200 1. S ta ble iso t op e s a nd th e inte rna tiona l sys t e m of unit s. R apid C ommunic atio ns in M a s s S pect romet ry 15 , 1270–1 273. S os so D, Luo D , Li Q-B, et al. 20 15. S e ed filli ng in d ome s tic a ted ma ize and ric e d e pend s on SWEET-medi at ed he x o s e t r a ns po r t. N at u re G en e t i c s 47 , 14 89– 1493. S r e enivasulu N, Wobus U . 2 013. S e ed- D evelop men t Progr ams : A Sys tem s Biol og y–Bas ed C omp ari s on B etwe en Dic ot s a nd M onoc ots . Annu a l Revi ew of Pl ant Bi olog y 64 , 189–21 7. S t a dler R, La ut e r bac h C, S auer N . 2005. Cell -to - Cell Mov em ent o f G reen F lu ore s c ent P rot ein Rev e als Po st - P hloem T r a n s p ort in the O u te r Int egum e nt and Id e nti fie s Sympl a s tic Domain s in Arabid op s i s S e ed s and E mbryos. Plant Ph y s i ology 139, 70 1–712. T hor ne JH . 1985. Ph loem Unl oadi ng o f C a nd N A ssimila te s in D e vel opi ng Se ed s. A nnual Re view of Plan t Bi ology 36 , 31 7–343. Wang S, L iu S, W ang J, et al. 20 20. Si multaneo us c ha nge s in see d s ize , o il co nte nt and prot ein c ont ent driven b y selec t i on o f SWEET homolog ue s durin g soybea n dom es tica tion . N a ti onal S cien c e Re view 7 , 1776 –1786. Wang Y, Wu F, Z ou R, Xu M, S ha n H, Che ng B, Li X . 2024. The ma iz e sug ar t r a n s po rter s Z mSWEET1 5a and Z mSWEET1 5b positivel y regula t e s a l t tol erance in pla nts. Pl an t Phy siolog y and Bi oche mistry 213 , 108 845. Wang S, Yok os ho K, G uo R, W he lan J, Ru an Y - L , Ma JF, Shou H . 2019 . The Soybe a n Sug ar Tran sporter GmS W E ET15 Med iat e s Sucro se Ex por t f r om E ndosp er m to Early Embryo. Pl ant Ph ysiolog y 180 , 2133– 2141. Wang J, Zhang L, Wang S, et al. 2025. AlphaFol d -G uid ed Be sp oke G e ne Editing E nhanc e s Fiel d - G rown Soybe an Oil Con ten ts. Adv ance d Sc ience 12 , 25 00 290. Widie z T, Ingram G C, Gut i e r rez -Ma r c os J . 2017. E mbryo-Endo sp erm- Sporop hyte I ntera ction i n Mai ze Se e d s . E mbryo-Endo spe rm-Sp orophy te In tera ct ion i n Maize Se e d s . B oston, MA: B rian A. Lark ins , 95–107 . Wipf D, Benjdi a M, Rikirsch E, Zim mer m ann S , Tegeder M , From mer WB . 2003. An expre s s io n cD NA libr ary for suppr e s s ion cloni ng in y ea st mu tant s, c omple menta tion o f a yea st hi s4 mut ant , a nd EST analy s i s f rom th e symb iotic ba s idi omyc e t e Heb eloma cyl in dro s porum. G en o me 46 , 1 77–181 . Y ang J, L uo D, Yang B , Frommer WB, Eom J-S . 20 18. SWEE T11 a nd 15 a s k ey pl ayer s in s eed fil ling i n rice. N ew P hytologi st 218, 6 0 4– 6 15 . Y ang B, Wang J , Yu M, et al. 2 022. Th e sugar tran s p o r t er ZmS U G CAR1 o f t h e nitra t e t r a n s po rte r 1/p eptid e tran spo rte r fa mily is critica l for maiz e gra in filling . The P l ant Cell, koac 256 . Z hang H, Goettel W, Song Q , Jia ng H, Hu Z, Wang M L , An YC . 2 020. S elec t i on o f G mSWEET39 for oil an d p r ote in improveme n t in s oybea n. PLOS G ene t i c s 16 , e1 009114 . Z hu J, Zhou L, L i T, Ruan Y, Zhang A , Don g X, Z hu Y, L i C, Fa n J . 2022. G e nome -Wi de Inv e s tig a tion an d C har acte rizati on of S W E ET Gene F amily with F ocus on T hei r Evol ution and Expre s s io n du ring Hormone a nd A b io t ic St r ess R es p on se in M a i ze . Ge n es 13 . .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint Figure 1. Three ZmSWEET 14a/14b/15a are sucrose transporter which are expressed in the EAS region and localize to the plasma membrane in protoplast. (A) Phylogenetic analysis of SWEET transporters from Zea mays (Zm, blue), Oryza sativa (Os, green), and Arabidopsis thaliana (At, magenta). Three ZmSWEET genes from Clade III, showing preferential expression in the endosperm adjacent to the scutellum (EAS), are highlighted with a red underline. ( B) Heterologous expression of ZmSWEET14a; ZmSWEET14b and ZmSWEET15a in the SUSY7 yeast strain which allows the phenotypic recognition of a sucrose carrier activity. Glucose containing medium is used as control . (C) Subcellular localization of ZmSWEET15a:mCitrine in maize protoplasts. The LTI6b:mTurquoise was used as plasma membrane marker . Fluorescent signals from mCitrine, mTurquoise, and chloroplast autofluorescence were visualized by confocal laser scanning microscopy. A B Yeast SUSY7 without vector Empty vector 1 (pDR196-GW) Empty vector 2 (pDR196) ZmSWEET15a ZmSWEET14a ZmSWEET14b Yeast SUSY7 without vector Empty vector 1 (pDR196-GW) Empty vector 2 (pDR196) ZmSWEET15a ZmSWEET14a ZmSWEET14b - uracil with 2% sucrose - uracil with 2% glucose C ZmSWEET15a:mCitrine LTI6b:mTurquoise Chloroplast autofluorescence Merged ZmSWEET16 OsSWEET16 ZmSWEET17a ZmSWEET17b AtSWEET16 AtSWEET17 OsSWEET15 ZmSWEET15a ZmSWEET15b ZmSWEET11b ZmSWEET11a OsSWEET11 OsSWEET14 ZmSWEET14a ZmSWEET14b OsSWEET13 ZmSWEET13c ZmSWEET13a ZmSWEET13b AtSWEET9 OsSWEET12 ZmSWEET12a ZmSWEET12b AtSWEET13 AtSWEET14 AtSWEET11 AtSWEET12 AtSWEET15 AtSWEET10 ZmSWEET2 OsSWEET2b AtSWEET2 OsSWEET2a AtSWEET1 ZmSWEET1a OsSWEET1a ZmSWEET1b OsSWEET1b AtSWEET3 ZmSWEET3b OsSWEET3b ZmSWEET3a OsSWEET3a AtSWEET8 AtSWEET6 AtSWEET7 OsSWEET4 ZmSWEET4a ZmSWEET4c ZmSWEET4b OsSWEET5 OsSWEET7b OsSWEET7e OsSWEET7a OsSWEET7c OsSWEET6a OsSWEET6b ZmSWEET6b ZmSWEET6a AtSWEET4 AtSWEET5 OsSWEET_not_numbered 0.5 Clade II Clade I Clade III Clade IV .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint zmsweet14b_1 zmsweet14b_2 zmsweet14b_3 WT zmsweet14a_1 zmsweet14a_2 zmsweet14a_3 WT zmsweet15a_1 WT A C Figure 2. Generation of knock out mutants in ZmSWEET14a, ZmSWEET14b and ZmSWEET15a by genome editing. (A) Schematic representation of the gene models for ZmSWEET14a, ZmSWEET14b, and ZmSWEET15a. Light blue boxes indicate untranslated regions (UTRs), blue boxes represent exons, and thin black lines indicate introns. Green and red triangles mark the positions of sgRNA1 and sgRNA2, respectively, used for CRISPR-Cas9 genome editing. Black arrows indicate the locations of primers used for genotyping and mutation characterization. ( B) Overview of the mutant alleles obtained. DNA sequences targeted by sgRNA are underlined. PAM sequences are shown in blue. Letters in red and red hyphens indicate insertions and deletions respectively. Alleles highlighted in bold were selected to generate the triple mutant line. (C) Predicted consequences of the mutations at the protein level. Purple boxes indicate conserved MtN3/SWEET domains. Black boxes represent in-frame amino acids, while grey boxes indicate out-of-frame sequences. Red STOP symbols denote the positions of premature stop codons. B Alelle ID Sequence ZmSWEET14a (WT) CTGGGTCTGCGTCGCCTTCT-CCGTCAGCGTCTTCGTCGCGCCGC zmsweet14a _1 (+T) CTGGGTCTGCGTCGCCTTCTTCCGTCAGCGTCTTCGTCGCGCCGC zmsweet14a _2 (+A) CTGGGTCTGCGTCGCCTTCTACCGTCAGCGTCTTCGTCGCGCCGC zmsweet14a _3 (-C) CTGGGTCTGCGTCGCCTTCT--CGTCAGCGTCTTCGTCGCGCCGC ZmSWEET14b (WT) CTGGGTCTGCGTCGCCTTCT-CCGTCAGCGTCTTCGTCGCGCCGC zmsweet14b _1 (-CC) CTGGGTCTGCGTCGCCTTCT---GTCAGCGTCTTCGTCGCGCCGC zmsweet14b _2 (-24bp) CTGGGTCTGCGTCGC-------------------------GCCGC zmsweet14b _3 (-C) CTGGGTCTGCGTCGCCTTCT--CGTCAGCGTCTTCGTCGCGCCGC ZmSWEET15a (WT) TGGCCATGTACCTGGTGTACGCGCCCAAGGCCGCCCGGGTGCTGG zmsweet15a _1 (-23bp)TGGCCATGTACCTGG-----------------------GTGCTGG .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint * * I J WT zmsweet 14a/14b/15a WT zmsweet 14a/14b/15a Figure 3. Comparative magnetic resonance imaging of WT and mutant zmsweet14a/14b/15a kernels at maturity. (A,B) Representative MRI image showing three-dimensional wild-type (WT) maize kernels (A) and individual intact embryos (sagital and axial view) (B); Images shown are representative of five biological replicates. All images were digitally extracted for comparison and are displayed at the same scale. (C,D) Equivalent images for zmsweet14a/14b/15a mutant kernels (C) and embryos (D). (E,F) Distribution of lipid-specific signal in embryonic tissues of WT embryos (E): virtual sections (axial) through the embryos (F); signal intensity is colour -coded in relative units. ( G,H) Same for zmsweet14a/14b/15a mutant embryos. (I) Quantification of embryo volume in WT and zmsweet14a/14b/15a mutant embryos. (J) Quantification of total lipid signal per embryo in WT vs. zmsweet14a/14b/15a mutant. Data are given in relative units. Asterisks in panels (I) and (J) indicate statistically significant differences (p<0.05, n=40, T-test). Abbreviations: en, endosperm; em, embryo. D F H E C G 1mm 1mm 1mm 1mm em en em en em em en en α α WT zmsweet14a/14b/15a WT zmsweet14a/14b/15a B A .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint A Figure 4. Triple mutant zmsweet14a/14b/15a are impaired in germinations (A) Illustration of the germination assay on paper . (B) Quantification of primary root length at 6 and 9 days after sowing (DAS). Values represent the mean of 80–117 seedlings; error bars indicate standard error . (C) Illustration of experiment set-up for hydroponic assay. ( D) Shoot and root fresh weigh, (E) primary root length, (F) seminal roots number and (G) average seminal root length of 11 DAS WT and triple zmsweet14a/14b/15a grown under hydroponic conditions . Values represent the mean of 13-20 seedlings ; error bars indicate standard error . Asterisks denote statistical significance based on t-test: **** = p < 0.0001; *** = p < 0.001; * = p < 0.05. WT zmsweet14a/14b/15a 6 DAS 6 DAS WT zmsweet 14a/14b/15a WT zmsweet 14a/14b/15a B DC E F G * * **** *** C .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint em A B pe pd em pe pd en en Figure 5. Imaging of sucrose distribution in the maize kernel of WT and zmsweet14a/14b/15a using FTIR microspectroscopy. (A, B) Bright-field images of cryo-sectioned kernels at 13 days after pollination (DAP) from wild-type (WT) (A) and zmsweet14a/14b/15a triple mutant (B). (C, D) Color map showing sucrose distribution across the same sections of WT (C) and mutant (D) kernels, as detected by FTIR microspectroscopy. (E,F) FTIR microspectroscopy on 17 DAP wild-type (WT) (E) and zmsweet14a/14b/15a triple mutant (F) kernel. Abbreviations: en – endosperm; em – embryo; pe – pericarp. Bar: 2 cm. D 2cm 2cm C WT zmsweet14a/14b/15a WT zmsweet14a/14b/15a WT zmsweet14a/14b/15a FE .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint Figure 6. ¹³C-sucrose feeding assay reveals altered carbon partitioning in the zmsweet14a/14b/15a. (A) Schematic representation of the ¹³C-sucrose feeding assay. Kernels at 13 days after pollination (DAP) from wild-type (WT) and zmsweet14a/14b/15a mutant plants were detached from the ear and immersed in a sucrose solution containing ¹³C-labeled sucrose for 24, 48, or 72 hours . After incubation, embryos and endosperms were dissected. The endosperm was further separated into basal and upper parts. (B) The ¹³C content derived from the labeled sucrose was quantified in each tissue using isotope ratio mass spectrometry in WT (green) and mutant (orange) kernels. Values represent the mean of 13-20 kernels. A B .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707659doi: bioRxiv preprint