Spatial and temporal characterization of the rich fraction of plastid DNA present in the nuclear genome of Moringa oleifera reveals unanticipated complexity in NUPTs' formation

Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 h ps://doi.o g/10.1186/s12864‑024‑09979‑5 RESEARCH Spa ial and empo al cha ac e iza ion o  he ich ac ion o plas id DNA p esen in henuclea genome o Mo inga olei e a e eals unan icipa ed complexi y inNUPTs´ o ma ion Juan Pablo Ma czuk‑Rojas1,2 , Angélica Ma ía Álamo‑Sie a1,2, An onio Salme ón3 , Al edo Alcayde4 , Vik o Isanbae 4 and Lo enzo Ca e e o‑Paule 1,2* Abs ac Backg ound Beyond he massi e amoun s o DNA and genes ans e ed om he p o oo ganelle genome o he nucleus du ing he endosymbio ic e en ha ga e ise o he plas ids, s e ches o plas id DNA o a ying size a e s ill being copied and eloca ed o he nuclea genome in a p ocess ha is ongoing and does no esul in he concomi an sh inking o he plas id genome. As a esul , plan nuclea genomes ea u e small, bu a iable, ac ion o hei genomes o plas id o igin, he so‑called nuclea plas id DNA sequences (NUPTs). Howe e , he mech‑ anisms unde lying he o igin and ixa ion o NUPTs a e no ye ully elucida ed and esea ch on he opic has been mos ly ocused on a limi ed numbe o species and o plas id DNA. Resul s He e, we le e aged a ch omosome‑scale e sion o he genome o he o phan c op Mo inga olei e a, which ea u es he la ges ac ion o plas id DNA in any plan nuclea genome known so a , o gain insigh s in o he mech‑ anisms o o igin o NUPTs. Fo his pu pose, we examined he ch omosomal dis ibu ion and a angemen o NUPTs, we explici ly modeled and es ed he co ela ion be ween hei age and size dis ibu ion, we cha ac e ized hei si es o o igin a he chlo oplas genome and hei si es o inse ion a he nuclea one, as well as we in es iga ed hei a angemen in clus e s. We ound a bimodal dis ibu ion o NUPT ela i e ages, which implies NUPTs in mo inga we e o med h ough wo sepa a e e en s. Fu he mo e, NUPTs om e e y e en showed ma kedly dis inc i e ea u es, sugges ing hey o igina ed h ough dis inc mechanisms. Conclusions Ou esul s e eal an unan icipa ed complexi y o he mechanisms a he o igin o NUPTs and o he e olu iona y o ces behind hei ixa ion and highligh mo inga species as an excep ional model o assess he impac o plas id DNA in he e olu ion o he a chi ec u e and unc ion o plan nuclea genomes. Keywo ds Mo inga, NUPTs, Plas id DNA, Chlo oplas , Genome E olu ion Open Access © The Au ho (s) 2024. Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecom‑ mons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed in a c edi line o he da a. BMC Genomics *Co espondence: Lo enzo Ca e e o‑Paule [email p o ec ed] Full lis o au ho in o ma ion is a ailable a he end o he a icle Page 2 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 Backg ound Nea ly all plan s con ain a small, bu signi ican , ac ion o hei nuclea genomes composed o DNA sequences de i ed om hei chlo oplas s [1]; hese nuclea in e- g an s o plas id DNA a e commonly known as nuclea plas id DNA sequences (NUPTs) [2] The p ocess o NUPTs´ o ma ion has been commonly associa ed o he p ocess by which mos genes p esen in he bac e- ial ances o o plas ids we e ans e ed o he nuclea genome and hei p oduc s e en ually e a ge ed o hei ances al compa men a e he endosymbio ic e en ha ga e ise o he chlo oplas o ganelle. How- e e , whe eas he la e en ails he loss o as amoun s o DNA wi h he subsequen educ ion o i s size and he ans e o mos o he genes o iginally p esen in he p o oo ganelle o ganism o he nuclea genome [3, 4], he o me in ol es he copy o s e ches o DNA om he chlo oplas genome. E en hough mos NUPTs a e less han 1 kb in leng h, NUPTs o ecen o igin spanning he whole chlo oplas ch omosome ha e been de ec ed in O yza sa i a ( ice) and Populus ichoca pa [5, 6], and did no esul in he sh inking o he plas id genome. Al hough he p ocess o NUPTs’ o ma ion is s ill poo ly unde s ood, i is expec ed o in ol e he ollowing sequence o e en s. Fi s , he duplica ion o a s e ch o DNA p esen in he chlo oplas genome. Second, he lysis o chlo oplas o ganelle memb anes o allow he leak- age o duplica ed plas id DNA. Thi d, he impo o he nucleus o he leaked plas id DNA. Fou h, he in eg a- ion o plas id DNA in o he nuclea genome. A p esen , no mechanism has been o mally p oposed o explain he ecu en duplica ion o s e ches o plas id DNA o a ying sizes ha a e a he o igin o NUPTs. The biologi- cal mechanisms in ol ed in he leakage o plas id DNA o he cy oplasm and i s subsequen impo by he nucleus a e no ye comple ely elucida ed ei he , al hough game- ogenesis and cell s ess (especially pollen de elopmen and mild hea s ess, espec i ely) ha e been epo ed o induce he dis up ion o chlo oplas o ganelle mem- b anes [2, 7–10]. I has been also sugges ed ha ce ain kinds o s esses, such as ionizing adia ion and pa hogen in ec ions, may, no only igge he leakage o plas id DNA o he nucleocy osolic compa men , bu also a o i s in eg a ion in o he nuclea genome [11]. The molecu- la mechanisms o NUPTs´ in eg a ion in o he nuclea genome a e no ully desc ibed ei he , bu hey a e p ob- ably di e se and gene ally in ol e double-s anded b eaks (DSBs) and DNA damage and hus a e po en ially mu agenic. Fo example, i has been hypo hesized ha NUPTs´ in eg a ion is media ed by non-homologous end joining (NHEJ) du ing DSB epai e en s [12–14], mos NUPTs a e expec ed o be apidly agmen ed and shu led away h ough ansposi ions and genome a angemen s and, e en ually, pu ged om he nuclea genome [15–17]. As a consequence, he dis ibu ion o NUPTs by age should ollow an exponen ial dis ibu- ion, indica ing a con inuous a e o NUPTs’ o ma ion and decay h oughou ime [15]. Al hough such a pa - e n has been sugges ed o ice, Medicago unca ula, P. ichoca pa and Zea mays [15, 17, 18], di e en pa e ns ha e been obse ed in o he species such as A abidopsis, Ca ica papaya, F aga ia esca, Mo inga olei e a (mo - inga) and Vi is ini e a [17–19]. A second consequence is he expec ed posi i e co ela ion be ween NUPTs’ size and age, an obse a ion ha has been sugges ed o se - e al species, despi e no being explici ly es ed s a is i- cally [7, 16, 17, 20]. Indeed, he ac ion o nuclea genomes occupied by NUPTs a ies eno mously among species and e en wi hin di e en popula ions o he same species [5, 21, 22]. Mos species showed a ound 0.1% o plas id DNA in hei nuclea genome, wi h e y ew showing mo e han 1% [1] These la ge a ia ions in he ac ion o nuclea genomes occupied by NUPTs aise he ques ion o wha e olu iona y o ces may lie behind he ixa ion o a i- able ac ions o plas id DNA in plan nuclea genomes. Howe e , p e ious s udies on he mechanisms o o igin and e olu iona y a e o NUPTs we e mos ly ocused on a limi ed numbe o species and in ol ed a educed num- be o NUPTs. A mo e de ailed pic u e will ce ainly ben- e i om a la ge numbe o NUPTs and a highe ac ion o he nuclea genome occupied by plas id DNA. So a , he la ges ac ion o DNA o plas id o igin ound in any plan nuclea genome (4.71%) has been de ec ed in he o phan c op mo inga [19]. In he p esen s udy, we le e aged a ecen ch omosome-scale e sion o he mo inga genome [23] o examine he spa ial dis i- bu ion and a angemen in clus e s o NUPTs, o explic- i ly model and es he co ela ion be ween hei age and size dis ibu ion, o cha ac e ize hei o igin wi hin he chlo oplas genome and hei si es o inse ion a he nuclea one, as well as o in es iga e hei a angemen in clus e s. Ou esul s e eal an unan icipa ed complexi y o he mechanisms a he o igin o NUPTs as well as o he e olu iona y o ces behind hei ixa ion. Resul s Widesp ead dis ibu ion o NUPTs in hemo inga nuclea genome In o de o de ec NUPTs p esen in he mo inga nuclea genome, a ch omosome-scale assembly o he mo inga genome, AOCC 2 [23], was scanned using BLASTN and he mo inga chlo oplas genome sequence (NCBI Re Seq numbe : NC_041432.1) [24] as que y, esul ing in 13,901 o al alignmen s. We isually inspec ed he alignmen s and de ec ed a signi ican ac ion o hem Page 3 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 (8657; 62.28%) a ising om wo speci ic egions o he chlo oplas genome. Those wo egions we e 200 bp and 350 bp in leng h and we e essen ially composed by As and Ts (Addi ionalFile1), hus likely co esponding o low complexi y egions, which a e known o esul in spu ious alignmen s no e lec ing ue homology bu a i ac s. Indeed, BLASTN sea ches on NCBI da abases using hose wo egions as que ies esul ed in ma ches o seemingly un ela ed genomes wi h high pe cen o iden i y, indica ing hey p obably co espond o a i ac s ( esul s no shown). The e o e, we e an BLASTN wi h he -dus op ion u ned on in o de o mask alignmen s esul ing om low complexi y egions. 5203 NUPTs we e now de ec ed, which we e con iden ly de ined as NUPTs in ou analysis (Supplemen al Table S1). 11 ou o he 14 ch omosomes hos ed mo e han 100 NUPTs ( ang- ing om 118 o 1072) and se en ch omosomes plus one sca old con ained NUPTs summing up abo e 160,600 bp (i.e., he size o he mo inga chlo oplas genome) (Sup- plemen al Table S1). The o al aligned egion be ween he chlo oplas genome and he nuclea genome, i.e., he o al egion o he nuclea genome occupied by NUPTs, summed up a o al o 9,781,275 bp, which ep esen s a 4.14% o he size o he nuclea genome assembly, close o es ima ions ob ained wi h p e ious e sions o he genome [25–27] (Table1). A e co ec ing o edundancy in BLASTN hi s esul ing om In e ed Repea (IR) egions o he mo inga chlo oplas genome (1272), he ac ion o he mo inga nuclea genome co esponding o NUPTs was o 3.29%, again p e y simila o es ima ions ob ained wi h he h ee o he e sions o he mo inga genome [25–27] (Table1), and u he suppo ing hese esul s we e no due o genome assembly e o s. Mos NUPTs inmo inga o igina ed h ough wo dis inc o ma ion episodes sepa a ed in ime In o de o gain insigh s on he iming o plas id DNA acquisi ion by he mo inga nuclea genome, we exam- ined he ela i e age dis ibu ion o NUPTs using he pe cen iden i y o he co esponding BLASTN hi s as a p oxy o e olu iona y ime. Assuming he mu a ion a e is p opo ional o e olu iona y ime, i. e., he molecula clock hypo hesis holds, he lowe he pe cen iden i y, he olde he NUPTs. Pe cen iden i y o BLASTN hi s anges be ween 72.37 and 100% and shows an appa en bimodal dis ibu ion (Fig. 1A). Indeed, when Gaussian mix u e models we e i ed o he co esponding den- si y cu es, wo clea peaks, cen e ed a ound 79.05 and 93.1%, espec i ely, we e de ec ed (Fig. 1A). Acco d- ing o he pos e io p obabili ies o assigning a NUPT o ei he one o ano he peak, using a h eshold o 95%, 776 NUPTs (14.91% o he o al) summing up a o al o 253,096 bp (2.59% o he o al) belonged o he olde peak ( om now on Episode I, o NUPTs-I), while 3855 NUPTs (74.09% o he o al) summing up a o al o 9,189,682 bp (93.95% o he o al) belonged o he younge peak ( om now on Episode II o NUPTs-II). The es o NUPTs (572, summing up a o al o 338,497 bp, i.e., 3.46% o he o al) we e no con iden ly assigned o ei he one o he o he peak. Taking as a whole, hese esul s suppo wo main episodic o ma ion e en s a he o igin o mos NUPTs. Nex , we examined he size dis ibu ion o NUPTs, pa - i ioned by each o he e ie ed episodes. While NUPTs- I anged in size om 69 o 3591 bp, NUPTs-II anged om 33 o 71,935 bp (Fig.1B). Bo h ollowed a non-no - mal igh -skewed unimodal dis ibu ion (Fig.1B), wi h a mean and a median size o 326.2 and 127 o 2384 and 778 bp o NUPTs-I and NUPTs-II, espec i ely. F om s udies in ice and o he plan species, i had been sugges ed an appa en posi i e co ela ion be ween size and sequence iden i y o NUPTs, i.e., la ge NUPTs end o be mo e conse ed a he sequence le el. This obse a ion can be in e p e ed as young, la ge con- se ed NUPTs declining and agmen ing o e ime, and e en ually being pu ged om he genome [7, 15–17, 20, 28]. To es whe he his obse a ion also applied o mo inga NUPTs, we s udied he co ela ion be ween size and sequence iden i y by means o wo di e en es s app op ia e o no -no mally dis ibu ed da a, again pa - i ioned by e e y episode de ec ed (Fig.1C and Table2). In e es ingly, while o younge NUPTs om episode II size nega i ely co ela ed wi h sequence iden i y in bo h es s (Table2), no signi ican co ela ion was ound o Table 1 Summa y o he mo inga nuclea genome e sions used in his s udy The ac ion o plas id DNA de ec ed in each e sion, be o e and a e co ec ing o edundan NUPTs, is also indica ed Nuclea genome e sion To al ac ion o plas id DNA (%) To al ac ion o plas id DNA a e emo ing edundan NUPTs (%) Re e ence AOCC 2 4.14 3.29 [23] Shyamli, e al., 2021 [27] 4.73 3.81 [27] AOCC 1 4.25 3.28 [26] Tian, e al., 2015 [25] 4.19 3.12 [25] Page 4 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 NUPTs-I (Table 2), sugges ing di e en mechanisms migh ha e been a he o igin o NUPTs om e e y epi- sode and / o , once in eg a ed, hey migh also ha e ol- lowed di e en e olu iona y ajec o ies. To p o ide u he suppo o he accu acy o he ob ained esul s and disca d hei o igin h ough genome assembly e o s, we epea ed all he analysis using he h ee p e iously published e sions o he mo inga nuclea genome assembly a ailable [25–27]. In each case, when i ing Gaussian mix u e models o each dis ibu- ion o pe cen iden i ies, he wo main peaks could be Fig. 1 Modeling he dis ibu ion o pe cen iden i y and size o mo inga NUPTs. A his og am o he dis ibu ion o NUPTs pe cen iden i y alues. The wo densi y plo s esul ing om i ing Gaussian mix u e models, pu a i ely co esponding o dis inc e en s o NUPTs´ o ma ion (I and II), a e shown. B His og am o he dis ibu ion o NUPTs size alues pa i ioned by o ma ion e en . C sca e plo o pe cen iden i y e sus sizes o mo inga NUPTs pa i ioned by o ma ion e en . Fo an easie isualiza ion, NUPT size alues ha e been log 10‑ ans o med Table 2 Co ela ion analysis be ween NUPTs’ sequence iden i y and size by NUPTs’ o ma ion e en Me hod Co ela ion coe icien PEpisode Kendall’s ank co ela ion au −0.05 0.06 I Spea man’s ank co ela ion ho −0.03 0.42 I Kendall’s ank co ela ion au −0.05 6.72 × 10−7 II Spea man’s ank co ela ion ho − 0.07 6.67 × 10−7 II Page 5 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 simila ly e ie ed (Supplemen al Fig. S1 and Supple- men al Table S2). Nega i e co ela ions be ween size and sequence iden i y we e also simila ly e ie ed o NUPTs-II (Supplemen al Table3), while no signi ican o only ma ginally signi ican posi i e co ela ion was ound o NUPTs-I. We ound 61 NUPTs, 51 o hem no edundan , span- ning a o al o 14,177 bp, showing 100% iden i y wi h he chlo oplas genome. These NUPTs migh no ep esen a eal biological phenomenon bu be he esul o a mis- assembly ha e oneously inco po a ed plas id egions in o he nuclea genome sequence. In o de o disca d his possibili y, we sampled he sequences om six ep- esen a i e NUPTs showing 100% iden i y and a ious sizes plus 100 bp o hei lanking egions in he nuclea genome and scanned o hei occu ence in he h ee addi ional e sions o he mo inga nuclea genome a ail- able. As e ealed by he co esponding mul iple sequence alignmen s, he six NUPTs plus lanking egions selec ed could be iden ically e ie ed in a leas one o he emaining h ee genome e sions (Addi ional iles2, 3, 4, 5, 6and7), u he alida ing ou indings. Cha ac e iza ion o  hedi e en ial dis ibu ion o NUPTs´ inse ion si es in hemo inga nuclea genome The dis ibu ion and equency o NUPTs ac oss he 14 ch omosomes con o ming he mo inga nuclea genome was ep esen ed in a Ci cos plo as independen densi y plo s o e e y episode (Fig.2). In con as o NUPTs-I, which showed an appa en homogenous dis ibu ion h oughou he mo inga nuclea genome, mos NUPTs- II appea ed o be highly concen a ed in some speci ic egions o ch omosomes one, ou , i e, six and 10, which showed p ominen peaks in he densi y plo s, likely co - esponding o ho spo s whe e NUPTs in eg a ion and / o subsequen ixa ion is a o ed (Fig.2). A ecen su ey in A ican and Asian ice epo ed a composi ional bias a he lanking egions o NUPTs’ Fig. 2 Ci cos plo ep esen a ion o NUPTs in he mo inga nuclea genome. Nuclea and chlo oplas ch omosomes a e ep esen ed as g ey and g een illed blocks, espec i ely, o ming a ci cum e ence. Resul s a e shown o he 14 nuclea ch omosomes, hos ing 4812 NUPTs (92.49% o he o al numbe ) spanning 8,928,478 bp (91.28% o he o al leng h). The block co esponding o he chlo oplas genome is loca ed a 12 o’clock, and he 14 nuclea ch omosomes a e a anged clockwise. Nuclea ch omosomes a e d awn o scale, wi h leng hs p opo ional o size and exp essed in Mb, while he chlo oplas genome has been upscaled o occupy a qua e o he image ci cum e ence; i s size uni was se o 10,000 bp. Line plo s ep esen ing he espec i e densi y dis ibu ions o NUPTs‑I ( ed) and NUPTs‑II (blue) a e displayed. Windows o 500,000 and 100 bp we e selec ed o he nuclea and chlo oplas ch omosomes, espec i ely. Local BLASTN sequence alignmen s be ween he chlo oplas and he nuclea genome co esponding o indi idual NUPTs a e ep esen ed as ibbons. Ribbons a e colo ed acco ding o he pe cen age o sequence iden i y o he local alignmen s (NUPTs) g ouped by qua iles (wi h yellow, ligh o ange, o ange, and ed co esponding o he i s , second, hi d and ou h qua iles, espec i ely). LSC, La ge Single Copy; IRA, In e ed Repea A; IRB, In e ed Repea B; SSC, Small Single Copy Page 6 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 inse ion si es [22]. Simila ly, we examined whe he he 100 bp egions lanking egions o NUPTs in mo inga also showed any composi ional bias. While he 100 bp lanking egions o NUPTs-I we e ea u ed by a g ea e GC con en on a e age (36.4%) han he es o he genome a e excluding NUPT sequences (35.72%), he opposi e end was obse ed o NUPTs-II, which dis- played a lowe GC con en on a e age (32.3%) wi h di - e ences being signi ican acco ding o Mann-Whi ney U- es s (P = 2.07 × 10−14; P = 2.99 × 10−103, espec i ely). Mo eo e , p e ious analysis on NUPTs om A abi- dopsis and ice iden i ied hei endency o g oup in clus e s, de ined as a g oup o wo o mo e non-o e - lapping NUPTs whe e he dis ance be ween wo con- secu i e in eg an s was less han 5 kb [7]. We ied o de e mine whe he NUPTs in mo inga we e also o m- ing clus e s. 880 NUPTs (16.91% o he o al) summing up a o al o 1,232,888 bp (12.6% o he o al) we e ound g ouping in o 282 clus e s, which we e de ec ed in he 14 ch omosomes plus nine sca olds, and whose sizes anged om 122 o 46,929 bp (Supplemen al Table S4). Then we examined sepa a ely clus e s g ouping NUPTs om e e y episode. 56 NUPTs-I (i.e., 7.22%) summing up a o al o 18,145 bp (i.e., 7.17%) we e ound o ming 24 clus e s which hos ed up o i e in eg an s (Fig.3) (Supplemen al Table S4), whe eas 476 NUPTs-II (i.e., 12.35%) summing up a o al o 976,761 bp (i.e., 10.63%) we e ound inside 150 clus e s which hos ed up o 11 in eg an s (Fig.3) (Supplemen al Table S4). The es o he clus e s (108) hos ed 380 NUPTs om ei he one o bo h episodes and / o unclassi ied NUPTs (Supplemen- al Table S4). We u he checked whe he he o de ing o NUPTs wi hin indi idual clus e s we e a anged collinea ly wi h espec o he chlo oplas genome o we e a he shu led in some way. Fo his pu pose, we g aphically ep esen ed he en la ges clus e s in e ms o numbe o in eg an s om e e y episode and he co esponding dono egions in he chlo oplas genome (Fig.4). While clus e s o med by NUPTs-I showed a endency o be a anged collinea ly wi h he chlo oplas genome (Fig.4A), no such collinea - i y could be obse ed o clus e s o NUPTs-II (Fig.4B). The g ouping in o clus e s o NUPTs a speci ic posi- ions migh be e lec ing ei he la ge NUPTs agmen ing o e ime a e hei in eg a ion in o he nuclea genome o ch omosomal ho spo s. I he o me we e he case, he sequence iden i y o NUPTs should co ela e wi h hei endency o g oup in o clus e s. To es his hypo h- esis, we examined he co ela ion be ween he a e age sequence iden i y o he NUPTs in e e y clus e and he numbe o in eg an s. The es s we e pe o med inde- penden ly on clus e s o med exclusi ely by NUPTs-I and NUPTs-II. No signi ican co ela ion was ound o NUPTs om ei he episode (Supplemen al Table S5). Biased dis ibu ion o NUPTs‑I in hemo inga chlo oplas genome Finally, we s udied he dis ibu ion o NUPTs ac oss he mo inga chlo oplas genome. Fo his pu pose, we di ided he co esponding DNA sequence in o 100 bp egions and ep esen ed he equency o occu ence o NUPTs as densi y plo s (Fig.2). We pe o med he analy- sis conside ing sepa a ely NUPTs-I and NUPTs-II. F om he densi y plo s o NUPTs-I, ou peaks we e appa - en , which accoun ed o 354 NUPTs-I, i.e., 45.61% o Fig. 3 Dis ibu ion o he numbe o in eg an s o NUPT clus e s Page 7 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 he o al. Two o he peaks, designa ed 1 and 2 (Fig.2), spanned 200 bp each and we e loca ed in almos con- secu i e egions o he La ge Single Copy (LSC) egion o he chlo oplas genome. The emaining wo, designa ed 3 and 4 (Fig.2), we e o 3800–3900 bp in size and co - esponded o edundan sequences om he IR egions o he chlo oplas genome. In con as , NUPTs-II we e ound o be almos uni o mly dis ibu ed ac oss he chlo oplas genome, excep o he IR egions, whe e, as expec ed, a ound wice he numbe o NUPTS-II could be obse ed (Fig.2). Discussion By le e aging a ecen ly ob ained high-quali y long- ead ch omosome-scale assembly o he nuclea genome o mo inga (i.e., AOCC 2) [23], we gained a ine cha - ac e iza ion o he ich ac ion o plas id DNA o igi- nally de ec ed in an olde , less con iguous, e sion (i.e., AOCC 1) [26], he highes epo ed o any plan species so a [19]. While he o al ac ion o plas id DNA was simila using bo h e sions o he genome, di e ences we e obse ed ega ding he e en s unde lying such en ichmen . Ou p e ious epo [19], using he dis i- bu ion o synonymous subs i u ions a es as a p oxy o e olu iona y ime, a ibu ed such en ichmen in plas id DNA o a ecen single bu s o plas id gene duplica es eloca ing o he mo inga nuclea genome. He e, in u n, by i ing Gaussian mix u e models o he dis ibu ions o sequence iden i y o NUPTs ( aken ins ead as a p oxy o e olu iona y ime), wo dis inc main episodic e en s o NUPTs’ o ma ion could be de ec ed, namely NUPTs- I and NUPTs-II. The eason o his disc epancy likely esides in e o s in he anno a ion o he AOCC 1 mo - inga nuclea genome, ea u ed by an o e ep esen a ion o small genes anno a ed wi h chlo oplas and pho osyn- he ic unc ions. While 656 and 114 genes we e anno- a ed wi h he e ms “chlo oplas ” o “pho osyn hesis”, espec i ely, in he AOCC 1 mo inga genome, only 378 and 51 genes we e anno a ed wi h such e ms in AOCC 2 [23]. Fo example, while 45 agmen ed nuclea genes we e anno a ed as encoding o he plas id-encoded la ge subuni o ibulose-1,5-bisphospha e ca boxylase/oxyge- nase (RBCL) in AOCC 1, only h ee we e anno a ed as such in AOCC 2, al hough all o hem could be mapped o speci ic genomic egions in AOCC 2. Al oge he sug- ges s he p e ious en ichmen in chlo oplas ela ed unc ions obse ed among nuclea genes was likely due o agmen ed DNA o plas id o igin, i.e., NUPTs, encompassing coding egions, w ongly anno a ed as gene coding models. Hi he o, ela i e ages o NUPTs’ o ma ion in di e en plan species had been epo ed o be ea u ed by ei he exponen ially dec easing o uni o mly cons an dis ibu- ions [15, 17, 18], which i , espec i ely, in o wo di e - en modes o NUPTs´ o ma ion, i.e., single e en s and ho spo s [7, 28]. The single e en mode commonly esul s Fig. 4 G aphical ep esen a ion o he en la ges clus e s o NUPTs o e e y episode in e ms o numbe o in eg an s and he co esponding dono egions in he chlo oplas genome (A) NUPTs‑I. B NUPTs‑II. Fo e e y clus e , dono egions in he chlo oplas genome a e shown as g een blocks, while NUPTs‑I and NUPTs‑II a e depic ed as ed and blue blocks, espec i ely. Fo e e y NUPT, he co esponding BLASTN sequence alignmen be ween he chlo oplas and he nuclea genome is ep esen ed as a ibbon. Ribbons a e colo ed acco ding o he pe cen age o sequence iden i y o he unde lying alignmen g ouped by qua iles (wi h yellow, ligh o ange, o ange, and ed co esponding o he i s , second, hi d and ou h qua iles, espec i ely). The di e en elemen s in he diag am a e d awn o scale, wi h he chlo oplas genome and i s ou canonical egions (LSC, La ge Single Copy; IRA, In e ed Repea A; IRB, In e ed Repea B; SSC, Small Single Copy) displayed on op as a e e ence o size Page 8 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 in long con inuous NUPTs collinea wi h speci ic egions o he chlo oplas genome, which a e concen a ed in speci ic egions o he nuclea genome, e.g., (pe i)cen o- me ic egions [7, 15, 16, 28], and a e expec ed o decay in o smalle agmen s and eloca e as a consequence o ch omosomal ea angemen s and eshu ling in ol ing ansposable elemen ac i i y [16]. In con as , ho spo s esul in he concomi an in eg a ion o mul iple sho NUPTs om di e en o igins a anged as a mosaic in speci ic loci o he nuclea genome [28, 29]. To he bes o ou knowledge, no p e ious s udies ha e epo ed he bimodal dis ibu ion o NUPT ela i e ages obse ed he e o mo inga. The obse ed bimodal dis i- bu ion implies NUPTs in mo inga we e o med h ough wo e en s sepa a ed in ime. Fu he mo e, NUPTs om e e y e en showed ma kedly dis inc i e ea u es, sug- ges ing hey o igina ed h ough dis inc mechanisms. Fo example, acco ding o he ela i e dis ibu ion o sizes, younge NUPTs om episode II showed seemingly andom o igins h oughou he chlo oplas genome and we e ea u ed by a wide ange o sizes, hei p e e en- ial loca ion in ho spo s ac oss he nuclea genome and nega i e co ela ion be ween sequence iden i y and size. Howe e , al hough some NUPTs-II may ha e o igina ed as long agmen s subsequen ly b eaking in o smalle pieces a anged collinea ly as clus e s h oughou he nuclea genome, in acco dance wi h he single e en mode [28], no co ela ion was obse ed be ween he numbe o NUPTs-II g ouping in clus e s and sequence iden i y. This lack o co ela ion sugges s a leas some NUPTs-II may ha e also o igina ed as smalle agmen s landing in speci ic landma ks o he nuclea genome, i.e., ch omosomal ho spo s, e en ually u he dispe s- ing ough di e en kinds o genome ea angemen s. This was also in ag eemen wi h he obse a ion ha NUPTs-II g ouped in clus e s ended o be ound shu - led in some way a he han a anged collinea ly wi h he chlo oplas genome. Al oge he suppo s he o igin o NUPTs-II h ough bo h single e en s and ho spo s modes o o igin. In u n, olde NUPTs om episode I, ea u ed by a na owe dis ibu ion o sizes, no co ela ion be ween sequence iden i y and size and a endency o be a anged colinea ly wi h he chlo oplas genome when ound g ouped in clus e s, do no seem o i in o any o he wo modes o NUPTs’ o ma ion p e iously desc ibed. Mo eo e , almos hal o he NUPTs om episode I o igina ed om ou speci ic egions in he chlo oplas genome, an obse a ion only epo ed p e iously o Aspa agus o icialis [20] and in con as o p e ious s ud- ies in A abidopsis, ice and o he species, which showed a homogenous dis ibu ion o NUPTs h oughou he chlo oplas genome [15, 17]. We he e o e p opose he e a hi d mode o NUPTs’ o ma ion h ough small-scale ecu en e en s. Once indi idual NUPTs a e o med, wo scena ios a e plausible i) mul iple copies o NUPTs i s ly o ming in he chlo oplas and la e eloca ing o he nucleus, o ii) indi idual NUPTs ecu en ly duplica - ing once in eg a ed in o he nuclea genome. In espec o he possible e olu iona y o ces unde - lying he leakages and subsequen ixa ion o a iable amoun s o plas id DNA in plan nuclea genomes, hese migh be ela ed o he di e en s ess ul condi ions o which e e y species would ha e been subjec ed h ough- ou hei ecen e olu iona y his o y; di e en s esses ha e been shown o p omo e DNA mig a ion om chlo- oplas s o he nucleus [10, 30]. The massi e amoun s o plas id DNA ound in he mo inga nuclea genome migh be well ela ed o he exposu e o s ess ul condi ions du ing i s ecen e olu iona y his o y [31, 32]. Indeed, domes ica ion o mo inga om he sub-Himalayan low- lands in NW India, i s pu a i e loca ion o o igin whe e mean annual p ecipi a ions exceed 1100 mm, o opical and sub- opical a eas a ound he wo ld whe e i s cul u e has sp ead [31] likely in ol ed he selec ion o a ie ies be e adap ed o d ie and ho e en i onmen s [32, 33]. Fu he mo e, mo inga shows a g ea adap i e po en ial o success ully cope wi h mul iple s esses, pa icula ly wa e de ici and UVB adia ion [34]. A his espec , i has been no ed ha he 11 gian NUPTs ound in Asian ice ended o dis ibu e in na u al popula ions om highe la i ude egions ea u ed by lowe empe a- u es and ligh in ensi ies [22]. This obse a ion led he au ho s o a ibu e NUPTs a po en ial ole in enhancing en i onmen al adap a ion by inc easing he numbe o chlo oplas -de i ed genes which migh , in u n, imp o e pho osyn hesis [22]. Howe e , we belie e his adap i e- o-s ess hypo hesis seems unlikely gi en ha “ ecen ” plas id- o-nuclea gene ans e s a e exceedingly a e, especially o pho osyn he ic genes, wi h he genes mos equen ly ans e ed in ex an lineages being ibosomal p o eins [35]. Wha e e he speci ic o ces ha a e a he o igin o he ixa ion o he massi e amoun s o plas id DNA ound in he mo inga nuclea genome, hey appea o be o a di e en na u e o e e y independen e en o NUPTs o ma ion de ec ed he e. Conclusions Resul s p esen ed he e e eal an unan icipa ed complex- i y o he mechanisms a he o igin o NUPTs and o he e olu iona y o ces behind hei ixa ion. Compa a i e genomics o domes ica ed mo inga oge he wi h ha o he 12 wild Mo inga species ha make up he axonomic amily Mo ingaceae wi hin he B assicales o de [36], eme ges as an excellen model o econs uc ing he Page 9 o 11 Ma czuk‑Rojase al. BMC Genomics (2024) 25:60 mechanisms o o igin and e olu iona y ixa ion o plas id DNA in he nuclea genome. Me hods De ec ion andanalysis o plas id DNA in henuclea genome NUPTs in he published e sions o he mo inga nuclea genome [25–27] we e de ec ed using he BLASTN local alignmen ool om he BLAST+ p og am pack- age 2.12.0+ [37]. The chlo oplas genome sequence o mo inga [24] (Table1) was used as que y and he pub- lished e sions o i s nuclea genome sequence (Table1) as da abases. The pa ame e s we e as ollows: -e alue 1e-5 -wo d_size 9 -penal y − 2 -show_gis -dus no -num_ h eads 8. In o de o deal wi h low complexi y egions pu a i ely p esen in he chlo o- plas genome ha migh esul in spu ious alignmen s w ongly de ec ed as homologous egions, he analyses we e epea ed by u ning on he -dus se ing (−dus yes). Resul s in e ms o sequence iden i y and densi y o NUPTs we e ep esen ed as ci cula plo s, cons uc ed using Ci cos e sion 0.69–8 [38]. In o de o co ec o edundancy o NUPTs esul ing om he IR egion o he chlo oplas genome, BLASTN hi s in ol ing IR egions we e coun ed only once. In o de o de ec NUPTs showing 100% iden i y wi h he chlo oplas genome plus hei 100 bp lanking egions in he p e ious published e sions o he mo inga nuclea genome, BLASTN alignmen s we e i s ly pe o med using he whole se o 100% iden i y NUPTs as que y and he genome sequence o each e sion as da abase. NUPTs and hei bes sco ing hi s de ec ed in each e sion o he genome we e hen aligned using he MUSCLE algo- i hm [39] h ough he SeaView 5.0.5 p og am [40]. The esul ing mul iple sequence alignmen s we e edi ed using GeneDoc 2.7 [41]. In o de o examine whe he NUPTs in clus e s we e a anged collinea ly wi h he dono egions o he chlo- oplas genome o shu led in some way, he co espond- ing BLASTN alignmen s we e isualized h ough he R genoPlo R 0.8.11 package [42]. Gaussian mix u e modeling o NUPTs’ pe cen iden i y dis ibu ion In o de o de ec peaks in he dis ibu ion o pe cen iden i y alues pu a i ely co esponding o episodic e en s o NUPTs in eg a ion in he nuclea genome, Gaussian mix u e models we e i ed o he co e- sponding dis ibu ion by employing he Expec a ion- Maximiza ion (EM) algo i hm o mix u es o no mal dis ibu ions. We i s de e mined he op imal numbe o Gaussian componen s (k) using he boo .comp() unc ion om he R mix ools 1.2 package [43], which pe o ms a pa ame ic boo s ap by p oducing B boo - s ap ealiza ions ( eplica es) o he likelihood a io s a- is ic o es ing he null hypo hesis o a k-componen i e sus he al e na i e hypo hesis o a (k + 1)-com- ponen i o a ious mix u e models. Fo his s ep, we used 1000 eplica es, a signi icance le el o 0.01, and se he maximum numbe o componen s o nine. The numbe o componen s de e mined in he p e ious s ep was hen used o i a mix u e o Gaussian models o he dis ibu ion o pe cen iden i y alues, u ilizing he no - malmixEM() unc ion om he same package and he ollowing pa ame e s: maxi = 1e-30, max es a s = 1e− 3, epsilon = 1e− 10. Each peak was cha ac e ized by an age (exp essed in pe cen iden i y alues) ha co esponded o he mean o he Gaussian mix u e componen . Se - e al o he pa ame e s we e es ima ed om each o he models, including he s anda d de ia ion o each compo- nen , as well as he mixing p obabili ies o each NUPT o belonging o each e ie ed peak. Abb e ia ions NUPT nuclea plas id DNA sequence DSB double‑s anded b eak NHEJ non‑homologous end joining SSA single s and annealing Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps:// doi. o g/ 10. 1186/ s12864‑ 024‑ 09979‑5. Addi ional ile1. Addi ional ile2. Addi ional ile3. Addi ional ile4. Addi ional ile5. Addi ional ile6. Addi ional ile7. Supplemen al Tables. Supplemen al Figu es. Acknowledgemen s No applicable. Au ho s’ con ibu ions LC‑P concei ed and designed he p ojec and all esea ch ac i i ies. JPM‑R pe o med all he analyses, wi h con ibu ions om AMA‑S. AS con ibu ed o he s a is ical analysis implemen ed in he pape . VI and AA con ibu ed o coding sc ip s used in he pape and p o ided compu a ional suppo . All au ho s con ibu ed o da a analysis and in e p e a ion. LC‑P w o e and edi ed he manusc ip wi h subs an ial con ibu ions om JPM‑R. All au ho s e iewed he manusc ip . Funding This wo k was suppo ed by a “P oyec os I+D Gene ación de Conocimien o” g an om he Spanish Minis y o Science and Inno a ion (g an code: PID2020‑113277GB‑I00) o LC‑P, and by unds ecei ed by he “Sis ema de In o mación Cien í ica de Andalucía” Resea ch G oup id BIO359 o LC‑P.